Intellectual Property Arrangements Productivity Commission Inquiry Report No. 78, 23 September 2016  Commonwealth of Australia 2016 ISSN ISSN ISBN ISBN 1447-1329 (print) 1447-1337 (online) 978-1-74037-592-4 (PDF) 978-1-74037-593-1 (print) Except for the Commonwealth Coat of Arms and content supplied by third parties, this copyright work is licensed under a Creative Commons Attribution 3.0 Australia licence. To view a copy of this licence, visit http://creativecommons.org/licenses/by/3.0/au. In essence, you are free to copy, communicate and adapt the work, as long as you attribute the work to the Productivity Commission (but not in any way that suggests the Commission endorses you or your use) and abide by the other licence terms. Use of the Commonwealth Coat of Arms For terms of use of the Coat of Arms visit the ‘It’s an Honour’ website: http://www.itsanhonour.gov.au Third party copyright Wherever a third party holds copyright in this material, the copyright remains with that party. Their permission may be required to use the material, please contact them directly. Attribution This work should be attributed as follows, Source: Productivity Commission, Intellectual Property Arrangements, Inquiry Report. If you have adapted, modified or transformed this work in anyway, please use the following, Source: based on Productivity Commission data, Intellectual Property Arrangements, Inquiry Report. An appropriate reference for this publication is: Productivity Commission 2016, Intellectual Property Arrangements, Inquiry Report No. 78, Canberra. Publications enquiries Media and Publications, phone: (03) 9653 2244 or email: maps@pc.gov.au The Productivity Commission The Productivity Commission is the Australian Government’s independent research and advisory body on a range of economic, social and environmental issues affecting the welfare of Australians. Its role, expressed most simply, is to help governments make better policies, in the long term interest of the Australian community. The Commission’s independence is underpinned by an Act of Parliament. Its processes and outputs are open to public scrutiny and are driven by concern for the wellbeing of the community as a whole. Further information on the Productivity Commission can be obtained from the Commission’s website (www.pc.gov.au). Canberra Office Level 2, 15 Moore Street Canberra City ACT 2600 GPO Box 1428 Canberra City ACT 2601 Telephone Facsimile 02 6240 3200 02 6240 3399 Melbourne Office Telephone 23 September 2016 03 9653 2100 www.pc.gov.au The Hon Scott Morrison MP Treasurer Parliament House CANBERRA ACT 2600 Dear Treasurer In accordance with Section 11 of the Productivity Commission Act 1998, we have pleasure in submitting to you the Commission’s final report into Intellectual Property Arrangements. Yours sincerely, Jonathan Coppel Commissioner Karen Chester Commissioner & Deputy Chair Terms of reference INQUIRY INTO AUSTRALIA’S INTELLECTUAL PROPERTY ARRANGEMENTS I, Joseph Benedict Hockey, Treasurer, pursuant to Parts 2 and 3 of the Productivity Commission Act 1998, hereby request that the Productivity Commission undertake an inquiry into Australia's intellectual property arrangements, including their effect on investment, competition, trade, innovation and consumer welfare. Background Australia provides statutory protection for intellectual property through patents, trade marks, geographical indications, registered designs, plant breeders' rights, copyright, moral rights, performers' rights and circuit layout rights. Current laws are consistent with treaties under the auspices of the World Trade Organization, the World Intellectual Property Organization and the World Health Organization to which Australia has acceded, as well as bilateral and regional trade agreements. The global economy and technology are changing and there have been increases in the scope and duration of intellectual property protection. The Australian Government seeks to ensure that the appropriate balance exists between incentives for innovation and investment and the interests of both individuals and businesses, including small businesses, in accessing ideas and products. Scope of the inquiry The Australian Government wishes to ensure that the intellectual property system provides appropriate incentives for innovation, investment and the production of creative works while ensuring it does not unreasonably impede further innovation, competition, investment and access to goods and services. In undertaking the inquiry, the Commission should: 1. examine the effect of the scope and duration of protection afforded by Australia's intellectual property system on: (a) research and innovation, including freedom to build on existing innovation (b) access to and cost of goods and services (c) competition, trade and investment. iv INTELLECTUAL PROPERTY ARRANGEMENTS 2. recommend changes to the current system that would improve the overall wellbeing of Australian society, which take account of Australia's international trade obligations, including changes that would: (a) encourage creativity, investment and new innovation by individuals, businesses and through collaboration while not unduly restricting access to technologies and creative works (b) allow access to an increased range of quality and value goods and services (c) provide greater certainty to individuals and businesses as to whether they are likely to infringe the intellectual property rights of others (d) reduce the compliance and administrative costs associated with intellectual property rules. 3. in undertaking the inquiry and proposing changes, the Commission is to have regard to: (a) Australia's international arrangements, including obligations accepted under bilateral, multilateral and regional trade agreements to which Australia is a party (b) the IP arrangements of Australia's top intellectual property trading partners and the experiences of these and other advanced economies in reforming their IP systems to ensure those systems meet the needs of the modern economy (c) the relative contribution of imported and domestically produced intellectual property to the Australian economy, for example to Australia's terms of trade and other economic impacts of IP protection, including on inward investment (d) the Government's desire to retain appropriate incentives for innovation and investment, including innovation that builds on existing work, and production of creative works (e) the economy-wide and distributional consequences of recommendations on changes to the existing intellectual property system, including on trade and competition (f) ensuring the intellectual property system will be efficient, effective and robust through time, in light of economic and technological changes (g) how proposed changes fit with, or may require changes to, other existing regulation or forms of assistance (such as research subsidies) currently providing incentives for the development of intellectual property (h) the findings and recommendations of the Harper Competition Policy Review in the context of the Australian Government's response, including recommendations related to parallel import restrictions in the Copyright Act 1968 and the parallel importation defence under the Trade Marks Act 1995 (i) the findings and recommendations of the Advisory Council on Intellectual Property's Review of the Innovation Patent System the Senate Economics References Committee's inquiry into Australia's innovation system the Australian Law Reform Commission's Copyright and the Digital Economy report. TERMS OF REFERENCE v Process The Commission is to undertake an appropriate public consultation process, inviting public submissions and releasing a draft report to the public. The Final report is to be provided to the Government within 12 months of receipt of this Terms of Reference. J B Hockey Treasurer [Received 18 August 2015] Note that the Australian Government approved a revised delivery date of the final report from August to September 2016. (The inquiry timetable was impacted by the caretaker period associated with the Federal Election in 2016, which precluded full engagement with government departments and agencies.) vi INTELLECTUAL PROPERTY ARRANGEMENTS Contents Terms of reference iv Acknowledgments xi Abbreviations xii Overview 1 Recommendations and findings 31 1 About this inquiry 43 1.1 Background to this inquiry 43 1.2 What has the Commission been asked to do? 45 1.3 Scope of the inquiry 45 1.4 Conduct of this inquiry 48 1.5 A guide to the remainder of this report 48 An analytical framework for assessing the IP system 51 2.1 How do IP arrangements affect wellbeing? 52 2.2 What should Australia’s IP system achieve? 54 2.3 What principles are important for the IP system? 60 How does the system fare? 77 3.1 Is the IP system effective? 78 3.2 Is the IP system efficient? 87 3.3 Is the IP system adaptable? 92 3.4 Are IP arrangements accountable? 96 3.5 Are IP arrangements suited to Australia’s circumstances? 98 2 3 4 Copyright term and scope 103 4.1 Australia’s copyright arrangements and the creative sector 104 4.2 Applying the Commission’s framework 110 4.3 The scope of copyright protection 112 4.4 The term of copyright protection 127 4.5 Towards more globally efficient copyright arrangements 133 CONTENTS vii 5 6 7 8 9 Copyright use and licensing 135 5.1 Consumer access to content 136 5.2 Geoblocking 142 5.3 Parallel import restrictions on books 145 5.4 Collective licensing and collecting societies 153 5.5 Statutory licensing 162 Fair use or fair dealing — what is fair for Australia? 165 6.1 Copyright exceptions — why do they matter? 166 6.2 Fair dealing and fair use — how do they compare? 168 6.3 The merits of a fair use exception 170 6.4 What would a fair use exception look like? 185 6.5 Fair use, orphan and unavailable works 193 The patent system — getting the fundamentals right 199 7.1 How well is the patent system performing? 200 7.2 A strategy for reforming the patent system 215 7.3 Introducing an objects clause 217 7.4 Reforming the inventive step 220 7.5 Improved decision making for granting patents 229 7.6 Making better use of patent fees 232 The innovation patent system 239 8.1 The innovation patent system: a primer 240 8.2 How well is the IPS performing? 245 8.3 The IPS — where to from here? 258 Business method patents and software patents 263 9.1 Business methods 264 9.2 Patents for software 265 9.3 Assessing the effectiveness and efficiency of software patents 272 9.4 Where to from here for software? 281 10 Pharmaceuticals – getting the right policy prescription viii 285 10.1 Why focus on pharmaceuticals? 286 10.2 Extensions of pharmaceutical patent term 292 INTELLECTUAL PROPERTY ARRANGEMENTS 10.3 Manufacture for export 310 10.4 Data protection 312 10.5 Strategic behaviour 319 11 Registered designs 331 11.1 Design rights in Australia 332 11.2 Concerns about Australia’s design rights system 336 11.3 Reform options 346 12 Trade marks and geographical indications 371 12.1 What are trade marks and GIs? 372 12.2 Improving effectiveness by reducing the scope of cluttering 377 12.3 Improving effectiveness by reducing confusion among users of trade marks 389 12.4 Improving efficiency by reforming parallel imports 393 12.5 Challenges for trade marks in the digital age 397 12.6 Geographical indications 399 13 Plant Breeder’s Rights 407 13.1 Plant Breeder’s Rights: A primer 408 13.2 PBR in practice 411 13.3 Are reforms to PBR required? 416 14 Circuit layout rights 431 14.1 What are circuit layouts and how are they protected? 431 14.2 The role of CLRs today 433 14.3 Where to from here — policy lessons rather than policy change 439 15 Intellectual property rights and competition law 443 15.1 The exemption for licensing or assignment of IP 445 15.2 Compulsory access to intellectual property 457 16 IP and public institutions 461 16.1 IP rights and public funding of research 463 16.2 Should IP rights be afforded for the results of publicly-funded research? 465 CONTENTS ix 16.3 Who should own IP arising from publicly-funded research? 474 16.4 Improving access to publicly-funded research 478 17 Intellectual property’s institutional arrangements 485 17.1 The institutional landscape 486 17.2 Participant concerns about current arrangements 490 17.3 Reform options — domestic context 504 17.4 Options for reform — international context 520 18 International cooperation in IP 527 18.1 What motivates international cooperation? 528 18.2 Approaches to international cooperation 534 18.3 The outcomes of Australia’s IP cooperation 540 18.4 Where to from here? 544 19 Compliance and enforcement of IP rights 551 19.1 The extent and impact of IP infringement 552 19.2 Addressing infringement — how do we fare? 556 19.3 A case to do more on copyright infringement? 562 19.4 Improving dispute resolution 570 19.5 Better enforcement of rights abroad 581 A Conduct of the inquiry 583 B International Commitments 605 C The IP-intensity of Australia’s trade 615 D Evidence on patents: social value, additionality and thickets 631 E Copyright evidence 643 F Registered rights in Australia: a primer 669 G Additional material about trade marks and geographical indications 683 H Economic impacts of recommendations 691 References x INTELLECTUAL PROPERTY ARRANGEMENTS 701 Acknowledgments The Commission is grateful to the many individuals and organisations who have taken the time to contribute to this inquiry, including those who participated in visits, roundtables, public hearings and provided submissions. The Commission is also grateful for the assistance it received from IP Australia in preparing this report. Staff from IP Australia facilitated access to the Intellectual Property Government Open Data (IPGOD) database and provided helpful responses to many questions about existing arrangements. ACKNOWLEDGMENTS xi Abbreviations ABS Australian Bureau of Statistics ACCC Australian Competition and Consumer Commission ACIP Advisory Council on Intellectual Property (now defunct) ALRC Australian Law Reform Commission API Active Pharmaceutical Ingredient ARC Australian Research Council ASEAN Association of South East Asian Nations ASIC Australian Securities and Investments Commission AUSFTA Australia-United States Free Trade Agreement CCA Competition and Consumer Act 2010 (Cth) CLR Circuit Layout Right CSIRO Commonwealth Scientific and Industrial Research Organisation DFAT Department of Foreign Affairs and Trade DIBP Department of Immigration and Border Protection DIIS Department of Industry, Innovation and Science EC European Commission EPC European Patent Convention EPO European Patent Office EU European Union FCA Federal Court of Australia FTA Free Trade Agreement GDP Gross Domestic Product GI Geographical Indication IP Intellectual Property IPCRC Intellectual Property and Competition Review Committee IPEC United Kingdom Intellectual Property Enterprise Court IPGOD Intellectual Property Government Open Data xii INTELLECTUAL PROPERTY ARRANGEMENTS IPR Intellectual Property Rights IPS Innovation Patent System IPTA Institute of Patent and Trade Mark Attorneys of Australia OECD Organisation for Economic Cooperation and Development PBR Plant Breeder’s Rights PC Productivity Commission PCT Patent Cooperation Treaty SITC Standard International Trade Classification SMEs Small– and medium–sized enterprises TPP Trans-Pacific Partnership TRIPS Agreement on Trade Related Aspects of Intellectual Property Rights UPOV International Convention for the Protection of New Varieties of Plants 1961 USPTO United States Patent and Trademark Office WIPO World Intellectual Property Organization WTO World Trade Organization ABBREVIATIONS xiii Key points • Australia’s intellectual property (IP) arrangements fall short in many ways and improvement is needed across the spectrum of IP rights. • IP arrangements need to ensure that creators and inventors are rewarded for their efforts, but in doing so they must: − foster creative endeavour and investment in IP that would not otherwise occur − only provide the incentive needed to induce that additional investment or endeavour − resist impeding follow–on innovation, competition and access to goods and services. • Australia’s patent system grants exclusivity too readily, allowing a proliferation of low-quality patents, frustrating follow–on innovators and stymieing competition. − To raise patent quality, the Australian Government should increase the degree of invention required to receive a patent, abolish the failed innovation patent, reconfigure costly extensions of term for pharmaceutical patents, and better structure patent fees. • Copyright is broader in scope and longer in duration than needed — innovative firms, universities and schools, and consumers bear the cost. − Introducing a system of user rights, including the (well-established) principles–based fair use exception, would go some way to redress this imbalance. • Timely and cost effective access to copyright content is the best way to reduce infringement. The Australian Government should make it easier for users to access legitimate content by: − clarifying the law on geoblocking − repealing parallel import restrictions on books. New analysis reveals that Australian readers still pay more than those in the UK for a significant share of books. • Commercial transactions involving IP rights should be subject to competition law. The current exemption under the Competition and Consumer Act is based on outdated views and should be repealed. • While Australia’s enforcement system works relatively well, reform is needed to improve access, especially for small– and medium–sized enterprises. − Introducing (and resourcing) a specialist IP list within the Federal Circuit Court (akin to the UK model) would provide a timely and low cost option for resolving IP disputes. • The absence of an overarching objective, policy framework and reform champion has contributed to Australia losing its way on IP policy. − Better governance arrangements are needed for a more coherent and balanced approach to IP policy development and implementation. • International commitments substantially constrain Australia’s IP policy flexibility. − The Australian Government should focus its international IP engagement on reducing transaction costs for parties using IP rights in multiple jurisdictions and encouraging more balanced policy arrangements for patents and copyright. − An overdue review of TRIPS by the WTO would be a helpful first step. • 2 Reform efforts have more often than not succumbed to misinformation and scare campaigns. Steely resolve will be needed to pursue better balanced IP arrangements. INTELLECTUAL PROPERTY ARRANGEMENTS Overview 1 The task at hand Intellectual property arrangements are important Intellectual property (IP) arrangements offer opportunities to creators of new and valuable knowledge to secure sufficient returns to motivate their initial endeavour or investment. In this respect, they are akin to the property rights that apply to ownership of physical goods. But ideas are not like physical goods in other key respects. As observed by Thomas Jefferson more than 200 years ago, the use of an idea by one party does not reduce its capacity for use by another: He who receives an idea from me, receives instruction himself without lessening mine; as he who lights his taper at mine, receives light without darkening me. (Jefferson 1813) Ideas also provide economic and social value as others draw on them and extend the frontiers of knowledge. For these reasons, property rights over ideas and their expression are not granted in perpetuity and limitations are placed on their application. IP rights take a variety of forms. The most familiar are patents, copyright and trade marks, but there are others, including rights over performances, designs, plant varieties and circuit layouts. A single product can — and often does — embody many IP rights (figure 1). IP arrangements form part of the broader innovation system. The role they play differs depending on the right afforded. Patents and copyright seek to promote product innovation and the creation of new works. Design rights seek to encourage improvements in the look and feel of consumer products. Trade marks differ again, providing consumer information and protecting brand reputation. OVERVIEW 3 Figure 1 IP phone Today’s smartphones are protected by over 1000 patents, including for their semiconductors, cameras, screens, batteries and calendars. Copyright protects the artwork and software code within smartphones. Design rights protect the aesthetics, and the placement of cameras, buttons and screens. Circuit layout rights protect the electrical integrated circuits. Brands, logos and other distinctive marks such as ‘iPhone’ are protected by trade marks. But IP rights can lead to IP wrongs Because IP rights give their holders the ability to prevent others from using that IP, there is a risk parties will unduly exercise market power. As noted by the Harper Competition Policy Review, this may allow owners of IP rights to extract excessive royalties from IP licences or place anticompetitive restrictions on knowledge dissemination, with adverse knock-on effects for innovation and ultimately consumers. When innovation is cumulative, IP rights can reduce the flow of benefits from new ideas and processes. This is particularly harmful for Australian firms, who tend to ‘adopt and adapt’ innovations, building on the knowledge of others. Overly strong restrictions on diffusion can be so detrimental to innovation as to undo the benefits of the IP system in the first place: … a poorly designed intellectual property regime — one that creates excessively “strong” intellectual property rights — can actually impede innovation. … Knowledge is the most important input into the production of knowledge. Intellectual property restricts this input; indeed, it works by limiting access to knowledge. (Stiglitz 2008, pp. 1694, 1710) And while patents and other IP rights can encourage innovation, they are not always necessary for it (figure 2). For example, in industries where the speed of technological change is fast moving, innovators tend to rely more on market-based arrangements, such as first-mover advantage. Similarly, IP rights are less important where innovations are difficult to copy or only entail minor development costs. Poorly designed IP rights impose costs irrespective of whether countries are net importers or exporters of IP. However, Australia is overwhelmingly a net importer of IP, and the gap between IP imports and exports is growing rapidly. This means that the costs to consumers and follow-on innovators from higher prices and restricted availability are not offset by increases in Australian producer profits. 4 INTELLECTUAL PROPERTY ARRANGEMENTS Figure 2 30 IP rights are used alongside other mechanismsa Innovation-active businesses Non innovation-active businesses per cent of businesses 25 20 15 10 5 0 Patents Registration of Copyright or Secrecy or Complexity of At least one design trade mark confidentiality product method used design to protect IP arrangements a Businesses can nominate more than one type of protection. Advancing reform in a constrained environment There have been many recent reviews into IP, such that some inquiry participants have understandably questioned the need for yet another. However, previous reviews have focused on specific areas of IP, such as innovation patents, pharmaceutical patents, design protection, and copyright, and so lacked a consistent and coherent approach across Australia’s IP arrangements — a point highlighted by the Harper Competition Policy Review. The Commission has taken a more holistic perspective to identify ways that the IP system could be improved. The goal of IP policy should be to achieve a balance between the incentive to create and the risk of damaging the productive use of new ideas through over-protection, while also recognising that Australia’s IP arrangements form part of a global system. With the overarching objective of maximising community wellbeing, the Commission has identified four guiding principles that the IP system should embody — it should be effective, efficient, adaptable and accountable (figure 3). In applying these principles, the Commission has considered each aspect of the IP system — how rights are assigned, used and enforced. The Commission has also examined the governance and institutional arrangements underlying IP policy development, decision-making and implementation. OVERVIEW 5 Figure 3 The Commission’s approach Overarching objective: to maximise wellbeing of Australians Goal: That the IP system provides appropriate incentives for innovation, investment and the production of creative works while ensuring it does not unreasonably impede further innovation, competition, investment and access to goods and services. Principles to apply to the IP system to achieve this goal: Effective Efficient Adaptable Accountable The system should be effective in encouraging additional ideas and in providing incentives that ensure knowledge is disseminated through the economy and community. The system should provide incentives for IP to be created at the lowest cost to society. This principle includes considering whether IP rights generate returns that are sufficient to encourage new ideas, the relative merits of public and private IP generation, and the longer-term effects on competition and innovation from granting IP rights. The system should adapt to changes in economic conditions, technology, markets and costs of innovating. The policies and institutions that govern the system, and the way that changes are made to them, need to be evidence-based, transparent, and reflect community values. But IP arrangements are not a blank slate. Many aspects of Australia’s IP arrangements have come about, or been strengthened, to give effect to commitments in international agreements. These agreements often contain prescriptive obligations relating to key policy levers such as the duration and scope of protection, and significantly curtail Australia’s capacity to change domestic policy arrangements. There are also practical constraints to independent IP policy-making. IP is a globally tradeable asset and Australia is a relatively small market. Significantly departing from the IP arrangements in other countries could frustrate Australia’s access to overseas innovations. 6 INTELLECTUAL PROPERTY ARRANGEMENTS While these constraints may see Australia fall short of achieving a balance across all aspects of IP arrangements, the Commission has identified much we can do to progress reform. Doing so necessitates an approach that: • examines reform opportunities within the limits imposed by our international obligations • embeds institutional and governance arrangements that promote transparent, informed and coherent policy outcomes • advocates for multilateral change where the stakes are sufficiently high. It also requires a dedicated reform champion with resolve to pursue change in the face of strong vested interests. 2 Copy(not)right — looking at the evidence Copyright protects the material expression of literary, dramatic, artistic and musical works, as well as books, photographs, sound recordings, films and broadcasts. In addition to being instrumental in rewarding creative and artistic endeavour, many creators value the recognition that the copyright system provides. It does so by granting creators the exclusive right to reproduce or adapt their work in material form, as well as to publish, perform, and communicate their work to the public. Exercise of these rights is commonly licensed to intermediaries, such as publishers, record companies, film studios, broadcasters, and copyright collecting societies. However, copyright protection in Australia suffers from a number of shortcomings. It is overly broad, applying equally to: commercial and non-commercial works; works with very low levels of creative input; works that are no longer being supplied to the market; and works where ownership can no longer be identified. Further, copyright does not target those works where ‘freeriding’ by users would undermine incentives to create new works. As such, Australia’s copyright arrangements are skewed too far in favour of copyright owners to the detriment of consumers and intermediate users. Despite many claims to the contrary, the Commission is not recommending any changes to the length of copyright term — doing so would require amendments to international agreements such as the Berne Convention, TRIPS and AUSFTA. But even within the limits of these agreements, there is scope to do more. Overly long term reduces community access to valuable works Copyright protects literary, musical, dramatic and artistic works for the duration of the creator’s life plus 70 years, sound recordings and films for 70 years, television and sound broadcasts for 50 years, and published editions for 25 years. To provide a concrete OVERVIEW 7 example, a new work produced in 2016 by a 35 year old author who lives until 85 years of age will be protected until 2136. Evidence (and logic) suggests copyright protection lasts far longer than is needed. Few, if any, creators are motivated by the promise of financial returns long after death, particularly when the commercial life of most copyright material is less than 5 years. Studies have found that a term of around 25 years enables rights holders to generate revenue comparable to what they would receive in perpetuity (in present value terms). Of course, some very successful works have commercial lives well beyond a few years, as repeatedly cited by inquiry participants in submissions and public hearings. But it remains the case that these are exceptions to the norm. While some copyright holders claim that there are few, if any, costs associated with excessive term, this has not been borne out in practice. Many works become commercially unavailable during their period of copyright protection. Overly long copyright term perversely increases the likelihood and duration for which works are unavailable. Demand for works that have been created, but are not being supplied while under copyright protection, reduces community welfare and returns to original rights holders and potential new providers. Nothing better exemplifies the costs of excessive copyright term than the fact that once copyright expires and works enter the public domain, many become commercially available again. Long periods of copyright protection, coupled with automatic application and no registration requirements, also result in ‘orphan works’ — works protected by copyright but unusable by consumers, libraries, and archives because the rights holder cannot be identified. The existence of orphan works has become a greater issue as libraries and archives have sought to make their collections available online. The Australian National Film and Sound Archive estimated as much as 20 per cent of its collection is orphaned or abandoned and highlighted examples of projects that have been shelved, and opportunities to celebrate Australia’s heritage foregone, due to the time and expense of identifying the relevant rights holders. Governments and academics, here and overseas, continue to explore innovative options for promoting a better balance on copyright term. In the United States (US), for example, the Register of Copyrights has publicly discussed the idea of requiring registration for rights holders to benefit from copyright term in excess of life plus 50 years (registration is already required to bring a court action for infringement in the US). Such arrangements would underscore the notion that rights holders should face obligations in order to benefit from protections. A fairer system of user rights Australia’s current limited exceptions, fair dealing being the most well-known, do little to restore the copyright balance. 8 INTELLECTUAL PROPERTY ARRANGEMENTS Australia’s exceptions are too narrow and prescriptive, do not reflect the way people today consume and use content, and do not readily accommodate new legitimate uses of copyright material. Legislative change is required to expand the categories of use deemed to be fair. Even when this occurs, changes have simply ‘caught up’ with existing community practice — Australia did not legalise the wide-spread practice of home VCR recording until as late as 2006, by which time most VCRs were household relics. Universities Australia summarised the extent of the problem: After 20 years of reviews that have considered this question, the evidence is in: Australia’s existing inflexible, purpose-based copyright exceptions are no longer fit for purpose. They are holding Australia back, not just in our universities and schools, but also in our digital industries. Innovative and useful technologies, and new ways of using content in socially beneficial ways, automatically infringe copyright in Australia unless their use falls within one of the existing narrow, purpose-based exceptions. (sub. DR453, p. 1) Australia’s narrow purpose-based exceptions should be replaced with a principles-based, fair use exception, similar to the well-established system operating in the US and other countries. As part of modernising its copyright arrangements, Israel recently adopted fair use to enable better access to copyright material ‘for the advancement of culture and knowledge’. Fair use would similarly allow Australia’s copyright arrangements to adapt to new circumstances, technologies, and uses over time. Some inquiry participants suggested that the benefits from fair use are largely academic because, although current exceptions do not reflect how people use copyright material in the digital age, rights holders do not pursue infringements for ‘ordinary’ uses. The example of teenagers sampling music and videos to make mash ups was raised more than once. But the opportunities Australian businesses and consumers forego because of the current inflexible exceptions are much more extensive. Participants argued that Australia’s current exceptions frustrate the efforts of online businesses seeking to provide cloud computing solutions, prevent medical and scientific researchers from taking full advantage of text and data mining, and limit universities from offering flexible Massive Open Online Courses. The education sector has also indicated that fair use would avoid the current perverse situation where Australian schools pay millions of dollars each year to use materials that are freely available online. Recent analysis undertaken by EY for the Australian Government assessed the benefits and costs of introducing a broad US-style fair use exception, and concluded that adoption of fair use in Australia would be a net benefit to the Australian community. While intrinsically difficult to assess, the analysis (unlike others commissioned by inquiry participants) examined the impact of fair use on Australian consumers and the broader community, users of copyright material such as schools and libraries, and rights holders. Some aspects of fair use offer larger gains, including education and government use, and improved community access to orphan works. Other changes reduce uncertainty for consumers and businesses, improving Australia’s innovation environment. OVERVIEW 9 Rights holders have argued against the adoption of fair use in Australia. They claim that by design, fair use is imprecise and would create significant legal uncertainty for both rights holders and users. Initial uncertainty is not a compelling reason to eschew a fair use exception, especially if it serves to preserve poor policy outcomes. Australia’s current exceptions are themselves subject to legal uncertainty, and evidence suggests that fair use cases, as shown in the US, are more predictable than rights holders argue. Moreover, courts routinely apply principles-based law to new cases, such as in consumer and employment law, updating case law when the circumstances warrant doing so. And over time, both rights holders and users will become increasingly comfortable with making judgements about when uses of copyright material are likely to be fair. Where the courts are called on to determine whether a new use is fair, legislation would require that they be guided by four fairness factors: • the purpose and character of the use • the nature of the copyright material • the amount and substantiality of the part used • the effect of the use upon the potential market for, or value of, the copyright material. Rights holders also argued fair use would significantly reduce their incentives to create and invest in new works, holding up Canada as an example. Some have proclaimed that fair use will equate with ‘free use’, particularly by the education sector. But these concerns are ill-founded and premised on flawed (and self-interested) assumptions. Changes in Canada’s publishing industry had little to do with copyright exceptions (where fair dealing still prevails) and more to do with other market factors. Notably, the Australian education sector has repeatedly made clear that fair use would coexist with the current education statutory licence scheme. Indeed, rather than ignore the interests of rights holders, under fair use the effect on the rights holder is one of the factors to be considered. Where a use of copyright material harms a rights holder, the use is less likely to be considered fair. In the US, where fair use is long established, creative industries thrive. In addition to the fairness factors above, uncertainty would be further limited by including a non-exhaustive list of illustrative fair uses to guide rights holders and users. By drafting the fairness factors to closely follow the wording of Australia’s existing fair dealing exceptions, as well as the wording of fair use overseas, existing Australian and foreign case law (particularly from the US where fair use has operated for some time) would provide an additional source of guidance. The use of foreign case law to reduce uncertainty was a key factor in Israel’s successful implementation and transition to a fair use regime. Among heavy users of copyright material, such as education and government users, as well as those in the creative sector, the Commission notes the abundance of guidelines developed collaboratively to further assist users in how to make judgements. 10 INTELLECTUAL PROPERTY ARRANGEMENTS Making it easier for users to access legitimate content Rights holders and consumer organisations raised concerns about online copyright infringement. Some see Australia’s efforts to curb unauthorised downloading as woefully inadequate; others consider existing steps as overreach. Arguments made in submissions reflect the polarised stance on this issue. Research consistently demonstrates that timely and cost effective access to copyright-protected works is the best way for industry to reduce online copyright infringement. Therefore, in addition to implementing a new exception for fair use, the Commission is recommending making it easier for users to access legitimate copyright-protected content. Geoblocking Geoblocking restricts a consumer’s access to digital products, enabling rights holders and intermediaries to segment the Internet into different markets and charge different prices (or offer different services) to consumers depending on their location. The use of geoblocking technology is pervasive, and frequently results in Australian consumers being offered a lower level of digital service (such as a more limited music or TV streaming catalogue) at a higher price than in overseas markets. Studies show Australian consumers systematically pay higher prices for professional software, music, games and e-books than consumers in comparable overseas markets. While some digital savvy consumers are able to avoid these costs (such as through the use of proxy servers and Virtual Private Networks), most pay inflated prices for lower standard services and some will ultimately infringe. The Australian Government should make clear that it is not an infringement of Australia’s copyright system for consumers to circumvent geoblocking technology and should avoid international obligations that would preclude such practices. Parallel importation of books Parallel import restrictions (PIRs) on books are the physical equivalent of geoblocking. Except in limited cases, Australian booksellers are prevented from purchasing stock from lower priced suppliers overseas, but must purchase from an Australian publisher regardless of the price. This restriction applies to booksellers only — Australian consumers can purchase books themselves from overseas online retailers. The restrictions can put Australian booksellers at a competitive disadvantage, and result in those Australians unable to purchase online paying higher prices. No fewer than eight past reviews, including by the Commission, and most recently by the Harper Competition Policy Review, have recommended that prohibitions on parallel imports be repealed. The Australian Government supports the removal of the restrictions OVERVIEW 11 and agreed to progress this reform subject to the recommendations of this inquiry regarding transitional issues. In responding to a range of false claims and flawed analyses made by participants, the Commission has undertaken a comprehensive analysis of book prices, comparing the price of over 1000 like-for-like titles sold in Australia and the UK (and 400 in Australia and the US). Over three quarters of the books in the sample were more expensive in Australia than the UK, with Australian prices around 20 per cent higher. Under reasonable assumptions regarding discounting and freight costs, the Commission estimates the benefits to Australians from repealing the restrictions could be around $25 million per year. The publishing industry has stridently opposed the removal of the restrictions. In doing so it has put forth a number of (often contradictory) arguments, including that the: • restrictions do not raise the price of books in Australia, but at the same time are crucial to supporting the production of Australian literature (which would require a premium on Australian book prices) • price of Australian books is competitive with those in the US and UK, yet removal of PIRs would result in importation of cheaper books and the demise of local publishing • removal of the restrictions would unduly harm local authors. Yet the Commission has found the benefits of the restrictions are overwhelmingly enjoyed by global publishers and offshore authors. The Commission found arguments about the role of publishers in supporting local authors particularly unconvincing. In order for this to occur, publishers would need to charge higher prices (which they deny) and channel the revenues from these higher prices back to Australian authors. During public hearings the Commission sought (but did not receive) evidence from publishers on the quantum of support they provide to Australian writers and how their support differs from that provided by publishers in other jurisdictions where PIRs do not apply, such as the US. The Commission recognises the cultural and educational value of books is significant. While most of these benefits are captured in the price readers are willing to pay, some are not. However, these broader benefits are best targeted by direct public support — as is already provided by Australian Governments (of around $40 million each year for Australian books and authors) — rather than through the ill-targeted PIRs. Publishers also expressed concern that removing PIRs will harm Australian booksellers. Yet the Commission received evidence that Australian publishers act as the local supplier when individuals import books from foreign online retailers. In this way, publishers appear less concerned about Australian consumers accessing books at lower prices than they are about ensuring their continued primacy in the local supply chain. Dymocks highlighted how PIRs unequally discriminate against Australian booksellers: … when an Australian customer makes a purchase from UK based Book Depository the order is fulfilled through a local Australian publisher rather than being sent from the UK. Australian 12 INTELLECTUAL PROPERTY ARRANGEMENTS booksellers — unable to source supply from overseas — are not given the same freedom. (sub. DR613, p. 1) And concern that overstocked books in foreign markets (remainders) would harm Australian publishers ignores the fact that, for the majority of books, the same publishing house holds both the Australian and foreign rights. For example, the Commission matched 1126 book titles across the Australian and UK market and found that 95 per cent were published in both markets by the same publisher or an owned subsidiary. Claims that lower priced books from overseas — especially those of Australian authors — will be ‘dumped’ in Australia are unsubstantiated and misleading, and may reflect a desire by some publishers to continue price discriminating against Australian readers. In short, no new evidence was presented in this inquiry that overturns the existing case for removing the restrictions. The Australian Government should proceed with its announced plans to repeal parallel import restrictions, with effect no later than the end of 2017. Additional transitional arrangements are not needed given the positive confluence of efficiencies made by the Australian publishing industry and broader economic circumstances. 3 Patents — getting the fundamentals right Patents can advance human knowledge by encouraging socially valuable innovation that would not have otherwise occurred. However, if poorly calibrated, they also impose net costs on the community. By design, patent protection inhibits competitors from freely using an inventor’s technology, but over-protection can stifle competition more broadly, leading to reduced innovation and excessive prices. Moreover, by blocking subsequent innovators, patent protection can perversely inhibit the advancement of knowledge through ‘follow-on’ innovation. Notwithstanding reforms introduced under the 2012 Raising the Bar initiative, Australia’s patent system remains tipped in favour of rights holders and against the interests of the broader community. • A multitude of low-value patents make it harder for innovators to signal the value of their inventions to investors, and also frustrate follow-on innovators and researchers who are forced to invest in costly workarounds. Costs are ultimately borne by the users of technology. • Australia provides stronger patent rights than most other advanced economies. As a net importer of patented technology, the strength of rights is particularly costly for Australia. As in other areas of IP, reform options are restricted by Australia’s international obligations. However, within these constraints, the Commission has identified a package of reforms that would go some way to striking a better balance. OVERVIEW 13 Making clear what Australians want from their patent system Consistent with the absence of overarching principles to guide IP policy, the Patents Act does not have an objects clause to guide legal interpretation. Many participants supported the principle of introducing an objects clause to provide greater guidance to decision makers involved in the design and application of the Act. An objects clause would provide a number of benefits. Greater guidance would play an important role given the scope for administrative and judicial interpretation to diverge over time from policy intent. Setting out broad objectives would also help the Act remain adaptable and fit for purpose as technologies emerge and economies and business models evolve. An objects clause should make clear that the principal purpose of the patent system is to enhance the wellbeing of Australians by promoting technological innovation, and by promoting the transfer and dissemination of technology. In so doing, the patent system should balance the interests of producers, owners and users of technology. Reforming the inventive step An invention must satisfy five criteria to qualify for patent protection, including that it involve a non-obvious ‘inventive step’ (box 1). The test for inventive step is particularly important because it provides the closest proxy for an invention’s technological advance. A high inventive step means that only significant improvements on existing inventions achieve patent protection, while a low inventive step means that incremental advances can secure the same term and scope of protection. Box 1 What are the criteria for granting a patent? IP Australia grants patents to inventions that meet the criteria outlined in the Patents Act. To satisfy the criteria for a standard patent, inventions must: • be a ‘manner of manufacture’ — described by the courts to be an invention that involves human intervention to achieve an end result, and has an economic use • be novel — the invention must be novel in light of ‘prior art information’ (information about the current state of technology) • involve an inventive step compared with the prior art base — the invention must not be obvious to a person skilled in the relevant art in light of ‘common general knowledge’ (knowledge of a worker in the field) • be useful — there must be a specific, substantial and credible use for the disclosed invention • have not been secretly used — the invention cannot be used before the priority date (the date from which a patent application is assessed against the patent criteria — typically the date when a party first files an application). 14 INTELLECTUAL PROPERTY ARRANGEMENTS In assessing whether an application has an inventive step, IP Australia must consider a number of factors, including the: • definition of the invention • ‘prior art’, or current state of technology • minimum advance over the prior art required to meet the test, or ‘obviousness test’ • ‘person skilled in the art’, who is assumed to have common general knowledge. The inventive step has been subject to ongoing reform. Most recently, the Raising the Bar initiative increased the inventive step threshold by reforming the definitions of prior art and common general knowledge. This has moved Australia’s requirements closer to the thresholds applied in the US and the European Union (EU). These reforms have moved the inventive step in the right direction, but there are grounds to go further. Measures of patent quality suggest that thresholds in the US and EU fall short of the ideal, and so are not sufficiently high benchmarks. And post-Raising the Bar patent outcomes (analysed by the Commission) indicate IP Australia still has a greater propensity to grant patent applications that have been rejected by the European Patent Office (EPO). Ongoing disparities between outcomes in Australia and the EU are not surprising, as Australia still applies a less rigorous test for obviousness. In particular, the required minimum advance over the prior art in Australia is a mere ‘scintilla of invention’, which is highlighted by some patent attorneys and referenced in IP Australia’s Patent Examiners Manual. Evidence also suggests that the inventive step is not always effective in filtering out patents that fail to advance technology. A robust case exists for raising the inventive step further to reduce the proliferation of low-value patents. Raising the threshold would also help to address specific concerns with pharmaceutical and software patents (see below). To raise the threshold, the required advance over the prior art should be increased and efforts should be taken to better ensure only technological inventions pass the inventive step. Given the weight of evidence that patent systems are out of balance, these unilateral reform options would leave the inventive step below the optimal level. Going further and significantly raising the threshold above the level applied in other countries would, however, entail risks. Such endeavour is best pursued in collaboration with like-minded countries. Improving the evidence base for granting patents As patents may impose costs on the community, judgements about whether or not to grant a patent must be well informed. OVERVIEW 15 Patent examiners draw on a significant amount of information when deciding whether to grant a patent, including on the current state of technology. In many cases an applicant will have better access to such information than patent examiners. In Europe, a patent applicant must identify the technical features of the invention in their set of claims. This enables the patent office to better target genuine advances in technology, establishes a clearer link between the prior art and the market protection being sought, and allows follow-on innovators to identify the core technical element of a patent claim. Given applicants are best placed to identify the technical features of their invention, requiring them to do so as part of an application for an Australian patent would impose minimal burden while helping to ensure only technological inventions are granted patent protection. Making better use of patent fees The structure and level of patent fees is another policy lever for improving the patent system. The Australian Government should set patent fees to promote broader IP policy objectives, rather than the current primary objective of achieving cost recovery. Renewal fees influence decisions about whether to maintain a patent. As such, they can help achieve a number of policy aims, including reducing economic rents that arise from patent holders exercising market power, limiting the risk that patents are used strategically, and ensuring only valuable patents are held in force. As a policy lever, renewal fees are underutilised. Many patented inventions require less than 20 years protection. Yet renewal fees only increase in three stages across the life of a standard patent. The structure of renewal fees in Australia should be reformed to increase more steeply with patent age, akin to the approach in the UK. Claim fees, in combination with effective rules on how claims are constructed, can decrease the scope of claims, and in so doing the breadth of market protection. Fewer claims also decrease the time taken to review applications. The structure of claim fees in Australia suggests they can be better deployed to discourage rights holders casting claims too widely and from using the system strategically. Currently, applicants only pay a flat fee for each claim in excess of 20 claims. Australia should adopt a similar approach to Japan, South Korea and Europe by lowering the initial threshold for claim fees, and applying much higher fees for applications with a large number of claims. 16 INTELLECTUAL PROPERTY ARRANGEMENTS 4 Other patent system improvements The ‘second-tier’ patent experiment has failed In addition to standard patents, Australia has a (second-tier) innovation patent system (IPS). The system’s objective (and that of comparable systems overseas) is to promote innovation by small- and medium-sized enterprises (SMEs). Compared to the standard patent system, the IPS provides more contained rights — innovation patents are limited to five claims and the maximum duration of protection is eight years. Australia’s IPS is little used. In 2015, innovation patents made up fewer than 5 per cent of patents in force. The IPS was introduced in response to concerns that the previous petty patent system was not meeting the needs of firms (especially SMEs) that invested in ‘incremental innovations’. Reflecting this, the ‘innovative step’ required to receive an innovation patent is lower than the inventive step for standard patents; even where innovation patents apply to obvious contributions, they have been found valid by the courts. The low innovative threshold has proven more harmful than helpful, including (perversely) for SMEs. It has encouraged a multitude of low value patents, covering everything from a pet bed to a pizza box that converts to a bib. This, in turn, has reduced the credibility that patents provide for attracting finance for commercialisation, and created uncertainty for other innovators who are unsure whether they are infringing on another party’s patent. Patent attorneys openly advertise ways in which users can game the system, including to improve their bargaining position in patent disputes and to frustrate entry by competitors. Some participants have called for the IPS to be abolished; others have called for its reform. Were the IPS to be reformed, there would be strong grounds to exclude obvious inventions by setting the innovative step at the same level as the inventive step for standard patents. It would also be necessary to address strategic behaviour, most likely by reintroducing a mandatory examination process, and limiting the period in which damages could apply. However, reforming the IPS along these lines would see innovation patents resemble petty patents, and so represent a return to an approach already found to be lacking — tantamount to a policy ‘Groundhog Day’. The community’s interests, and the interests of SMEs, would be better served by abolishing innovation patents and directly tackling the IP issues of greatest concern to SMEs, such as patent infringement and enforcement costs. Software patents — staying on track The rise of the digital economy means that software is now a part of many everyday goods and services, and is a vital building block for new ideas and technologies. But while software represents the future, the legal constructs of software patents are stuck in the past — using concepts that stem from England’s 1624 Statute of Monopolies. Unsurprisingly, the use of a four century old definition has proven challenging to apply to contemporary innovations. OVERVIEW 17 Software innovations are also increasingly at odds with the economic underpinnings of the patent system. Software development typically occurs rapidly, builds sequentially on existing ideas, and is getting cheaper. In contrast, patents provide a long period of protection and can frustrate follow-on innovation. Over the last decade, there has been growing concern that software patents are being used to protect simple or straightforward ideas, and to gain exclusivity over existing business processes that are merely automated using a computer, rather than being particularly novel. Recent court decisions have helped to narrow the circumstances where computer-implemented innovations can gain patent protection. Business methods are no longer patentable, and other software innovations must now embody some technical contribution in order to qualify for patent protection. The Commission’s proposed patent reforms would assist further in limiting low-value software patents. The patentability of software merits close and ongoing scrutiny given its importance to the modern economy, and to ensure that the effect of recent legal decisions has been in the best interests of the community. 5 Pharmaceuticals — a better policy prescription The pharmaceutical sector relies on IP protection more than most, since many pharmaceutical advances require large upfront investment in research and development and are easy to copy. In addition to the standard suite of IP protections, the pharmaceutical sector benefits from bespoke IP arrangements. Extensions of term Further to the 20-year term applying to all patents, pharmaceutical patents can qualify for an additional five years of protection. Extensions of term (EoT) are capped at an effective market life of 15 years. These bespoke arrangements were intended to attract pharmaceutical research and development investment to Australia and to improve incentives for innovation by providing an effective market life for pharmaceuticals more in line with other technologies. However, Australia’s EoT scheme has had little effect on investment and innovation; Australia represents a meagre 0.3 per cent of global spending on pharmaceutical research and development. As pharmaceutical companies have acknowledged, the prospect of future returns in such a small market (accounting for only 2 per cent of global pharmaceutical revenues) provides little in the way of additional incentive. Moreover, the benefits sought from EoT arrangements have proven largely illusory, resulting in a costly policy placebo. Poor targeting means that more than half of new chemical entities approved for sale in Australia enjoy an extension in patent term, and consumers and governments face higher prices for medicines. 18 INTELLECTUAL PROPERTY ARRANGEMENTS Rather than compensating firms for being slow to introduce drugs to the Australian market, extensions should only be allowed where the actions of the regulator result in an unreasonable delay. Timeframes (of around one year) set by Government for the Therapeutic Goods Administration (TGA) provide a ready benchmark for determining what constitutes a reasonable processing period. EoT should only be granted where the time taken by the regulator exceeds this period. The Commission estimates that this approach would lower the cost of pharmaceuticals in Australia and save consumers and taxpayers more than $250 million per year. Sharing rather than protecting data The confidential data submitted in support of regulatory approval processes are also protected for a period of five years. During this period, manufacturers of generic pharmaceuticals must independently prove that their products are safe and effective, even though they are chemically identical to already approved drugs. Pharmaceutical companies have pressed the Australian Government to extend the duration of data protection. They view data protection as an insurance policy to guard against what they see as inadequate patent protection. Most recently, negotiations for the Trans-Pacific Partnership Agreement saw (unsuccessful) calls to extend data protection for biologics from 5 to 12 years. Despite decade-long claims of inadequate patent protection, there is little evidence of a problem. Even if isolated cases were verified as genuine, extending protection to a broad class of products to address exceptional cases would represent a blunt and costly response. And using data protection as a proxy for patent protection has drawbacks. Beyond the obvious absence of disclosure of information to promote further innovation, data protection lacks other important balances that apply to patents. Data protection arises automatically and cannot be challenged in court. As well as there being strong grounds for resisting further calls to extend the period of data protection, there is a case for making data more widely available. At present, not only are follow-on manufacturers prevented from relying on clinical data for a period of five years, the data is kept confidential indefinitely. Allowing researchers access to this data could provide substantial public health benefits. But doing so unilaterally would have some downsides. Companies may respond by delaying the release of medicines in the Australian market. Accordingly, any moves to publish the relevant data need to be internationally coordinated. Reducing the scope for strategic behaviour The ability of companies to leverage their IP rights to forestall entry by generics — effectively extending the term of exclusivity — can have a significant negative impact on consumers and (through the Pharmaceutical Benefits Scheme (PBS)) on taxpayers. OVERVIEW 19 Firms can use a variety of strategies to further extend the commercial life of their products including (so-called) evergreening and pay-for-delay. Evergreening refers to the strategy of obtaining multiple patents that cover different aspects of the same product, typically on improved versions of existing products. Some of these patents relate to genuine improvements that increase consumer wellbeing, such as significantly reducing side effects of certain medications. However, some ‘improvements’ may involve a slightly different chemical combination or process of production, which show no appreciable difference to the user. An additional benefit of changing the inventive step is it would reduce the scope for the latter type of behaviour — by granting new patents only for genuinely inventive products. Pay-for-delay refers to the practice whereby patent holders pay generic manufacturers, as part of a settlement for a patent infringement case, to keep their products off the market beyond the scope of a patent. Delays of this kind limit competition by restricting the number of products on the market and any subsequent price reductions, including those triggered under the PBS. In contrast to the US and Europe, which have arrangements to detect suspect agreements, Australia has taken a ‘see no evil’ approach to pay-for-delay settlements. A transparent reporting and monitoring system should be put in place to detect pay-for-delay settlements. This would require reporting to the Australian Competition and Consumer Commission (ACCC) settlement arrangements between originator and follower pharmaceutical companies that affect the timing of market entry for a generic version of a product into the Australian market. To minimise compliance and transition costs, monitoring arrangements should be based on those employed by the US Federal Trade Commission. 6 Other IP rights Australia’s IP arrangements encompass other protections. Protections are available for the physical features of products (designs) and their branding or styling (trade marks). Sui generis rights are intended to fill apparent gaps in established IP protection, such as in plant varieties and circuit layouts. Registered designs Registered design rights serve a niche yet important role in Australia’s IP rights system — protecting the appearance of products that have an industrial or commercial use. Inquiry participants expressed concerns about Australia’s design rights system, including the low uptake of design rights due to the cost of registration and enforcement, and a poor understanding of design law, which can lead to designers inadvertently losing their rights or failing to seek protection in the first place. 20 INTELLECTUAL PROPERTY ARRANGEMENTS The Australian Government has committed to making changes that would partly address these issues. Following a review by the Advisory Council on Intellectual Property, the Government has agreed, among other things, to the introduction of a grace period for filing registered design applications. This will help ensure designers do not inadvertently lose eligibility for design protection and allow them to undertake some market testing prior to incurring the cost of filing. The Commission is also recommending some general measures to improve dispute resolution processes, discussed below. These reforms would go some way to addressing concerns among designers about enforcement costs and access to dispute resolution options. Many participants see joining the Hague Agreement as offering the potential for lowering the costs of registration. Under Hague, Australian designers would be able to seek protection in multiple countries through a single international application. But the benefits to Australian firms, and in particular SMEs, are likely to be much smaller than some anticipate. Filing for protection under the Hague Agreement is not necessarily cheaper than directly filing for protection, particularly where firms seek protection in a limited number of countries. More importantly, joining the Hague Agreement would involve extending the maximum term of protection for registered designs from 10 to 15 years. The Australian Government has already agreed to further investigate the costs and benefits before making a decision to sign on to Hague. Consistent with the approach taken by the Commission in this inquiry, such a process should ensure the gains from ‘harmonisation’ outweigh the costs of extending term, and that the interests of Australian consumers are adequately considered. Trade marks Trade marks help consumers to identify goods and services and provide a means for businesses to build and maintain a positive reputation. But when trade marks are granted too broadly or in too great a number, they can inhibit new market entrants by making branding difficult — an outcome known as ‘cluttering’. These difficulties have been exacerbated by legislative change, which has broadened the ‘presumption of registrability’, resulting in protection being sought and granted more often. While legislative change has made it easier to achieve trade mark protection, there has been less effort to ensure unused marks — such as those held by defunct firms — are removed quickly from the trade mark register. Requiring trade mark applicants to nominate whether they are using the mark applied for, and if not, to later provide evidence of use in order to retain trade mark rights would remedy this problem. OVERVIEW 21 The protection and information that trade marks convey is also causing confusion for consumers. Marks are being used to convey an ‘impression’ of provenance or quality. For example, there have been recent cases where goods have been marked with terms or logos to indicate they are handmade in picturesque locales like the Barossa Valley or Byron Bay, when in fact they are factory produced in industrial centres. Strengthening the existing requirements for marks not to be misleading or confusing would address this issue. Firms also find the trade mark regime confusing, often conflating the protection afforded by a trade mark with that of registering a business name. This confusion can result in firms undertaking costly rebranding after unintentionally infringing on a trade mark. Linking the trade mark and business name registers would reduce this confusion. The law that governs the importation of legitimately trade marked goods produced in other countries also needs reforming. While the Trade Marks Act contains provisions about when parallel imports may be allowed, recent legal cases have ‘muddied the waters’ to the point where firms are unsure if they are able to import marked goods legally. Amending the Act to make clear that parallel imports are allowed, would resolve the uncertainty and ultimately benefit the community. Plant breeder’s rights Plant breeder’s rights (PBRs) provide their holders with exclusive, time-limited control over the sale and propagation of registered plant varieties. PBR protection is less extensive than patent protection because of the breeder’s exception, which recognises the incremental and long-term nature of conventional plant breeding, and allows new plant varieties to be used in further breeding programs. PBRs have helped transform agricultural plant breeding in Australia by introducing competition and price signals to a market that was previously characterised by a high degree of state provision. Growers pay directly for access to new plant varieties, and their willingness to pay rewards successful breeders. Notwithstanding the success of the regime in encouraging greater private sector activity, plant breeders and other stakeholders have expressed concern that the scope of protection provided by PBRs is being undermined by technological changes. This may have opened the door to greater free-riding on protected varieties. Currently, so long as they do not register copied varieties with IP Australia, breeders are potentially able to copy and sell PBR-protected varieties with only minor variations, undermining the protection afforded by the right. Amending the Act would address this. Misrepresentation of varieties and refusal to pay royalties remain concerns, particularly for breeders of pasture crops. Improving compliance with PBR and licensing agreements is best achieved through closer cooperation and consultation, with industry groups best placed to lead these efforts. 22 INTELLECTUAL PROPERTY ARRANGEMENTS Circuit layout rights Circuit layout rights (CLRs) protect the layout designs (three-dimensional topography) of integrated circuits. The rights granted to circuit designers are narrow, and rapid change in the industry has brought the need for CLRs into question. Most circuits are custom designed for specific purposes and not generally adaptable for other uses. Australia’s adoption of CLRs is illustrative of the ‘protect first, assess later’ way IP rights have been expanded in the past. While the legislative protection for circuit layouts was premature, given international obligations, the removal of such rights would cause more problems than solutions. Retaining CLRs remains the ‘least worst’ option. 7 Improving the broader landscape Improving interactions between IP rights and competition policy IP rights holders currently enjoy an exemption from aspects of Australia’s competition law. But the rationale for the exemption has largely fallen away. IP rights and competition are no longer thought to be in ‘fundamental conflict’. IP rights do not, in and of themselves, have significant competition implications. Recognising that competition and IP policy are not at odds, a better approach would allow the ACCC to address any anticompetitive conduct, while minimising uncertainty for rights holders and licensees. Repealing the exemption, combined with ACCC guidance on the application of competition law to IP transactions, would achieve this outcome. No less than seven reviews have recommended repealing the exemption. The only remaining obstacle to doing so will be removed when recommendations of the Harper Competition Policy Review, to limit the scope of ‘per se’ prohibitions on anticompetitive conduct, are given effect. Commercialisation of publicly-funded research IP arrangements can facilitate commercialisation of publicly-funded research by allowing exclusivity over certain inventions created with the benefit of public funding. Where IP rights are used in combination with broader innovation policies, such as direct funding for research, it is important that the neutrality of public sector funding allocation is not compromised. The current policy settings for publicly-funded research, whereby recipients of funding own any resultant IP, and specialised technology transfer offices facilitate the dissemination of research results, are generally sound. OVERVIEW 23 However, copyright restrictions on access to publicly-funded research publications limit the dissemination of knowledge, and digitisation has significantly diminished the rationale for limiting access in this way. Publicly-funded research publications should be available to the public under open access arrangements after a 12 month embargo period. Suggestions for a ‘use it or lose it’ approach to university-owned IP are not supported by the available evidence, and may impose a higher barrier to access than existing compulsory licensing arrangements. Recent concerns around low rates of research collaboration have prompted government, academic and industry-led initiatives to improve the commercialisation of publicly-funded research. These initiatives should be given time to work before any further interventionist approaches are considered. Making it easier to resolve IP disputes While large, well-resourced firms are able to satisfactorily resolve their IP disputes, SMEs are often deterred from doing so due to the high costs and risks involved. Participants pointed to the UK’s Intellectual Property Enterprise Court (IPEC) as one model for addressing these concerns. The Commission has examined this model and the evidence suggests that the IPEC has improved access to justice for SMEs, who now have an avenue for timely and low cost dispute resolution. The Federal Court has already initiated reforms to improve the efficiency of IP litigation in Australia. While welcome, these reforms are unlikely to provide the savings to litigants afforded by the IPEC model. The benefits of the IPEC derive from its ability to minimise parties’ court appearances and the limits on claimable damages and costs. Some see the specialist nature of the court as further contributing to its success. The Federal Circuit Court was established to be a lower cost court with less formal rules. Consistent with this approach, the Federal Circuit Court routinely refers IP cases to mediation prior to litigation. Its ‘low-cost DNA’ and informal approach makes it well-placed to play a greater role in resolving lower value IP disputes. The Commission recommends the Federal Circuit Court introduce a specialist IP list, with procedural rules similar to the IPEC. The Court’s jurisdiction should be expanded to cover the full range of IP matters, mandatory caps should apply to cost and damages awards, and strict case management adopted to minimise court events. A separate small claims track suitable for self-represented litigants should provide an informal forum for low-value cases. The Commission anticipates that these reforms will result in some additional demand for the Court’s services. The Court should be adequately resourced to ensure that any increase in its workload does not result in longer resolution times. 24 INTELLECTUAL PROPERTY ARRANGEMENTS 8 Charting a new course in IP policy Strengthening domestic governance arrangements Australia has strayed on IP policy for a number of reasons. The absence of an overarching objective, policy framework and reform champion have collectively contributed to poor policy outcomes. To promote a more coherent, economywide perspective, there would be value in specifying the overarching objectives of the IP system to inform the broader community and guide agencies and departments involved in IP policy development and administration. A common framework for formulating IP policy would also assist; the four principles employed by the Commission throughout this report provide a ready starting point (figure 3). Responsibility for policy development and advice being shared across multiple agencies has further contributed to poor policy outcomes. The Department of Industry, Innovation and Science (DIIS) has kept a low profile in IP policy debates and has afforded few resources to this responsibility. IP Australia has played a more active policy role, but in doing so has blurred the line between policy development and administration. To help clarify the respective roles of the IP administrator and the department, and to increase transparency, the Minister responsible for IP should outline the functions and responsibilities for IP Australia through a public statement of expectations. The statement could cover issues such as the Government’s overall objectives for the IP system (mentioned above) and how IP Australia should contribute to IP policy development. The Commission also considered whether consolidating responsibility for IP policy (including for copyright) into a single department would promote a more coherent approach. While such an approach has merit, on balance the Commission considers that the Government should instead introduce an interdepartmental IP Policy Group that is responsible for overseeing IP policy development. Doing so would provide many of the same benefits of policy consolidation, but with relatively low costs and disruption to the system. This should be complemented with formal arrangements specifying how agencies and departments will work together to achieve the objectives of the IP system and adhere to the common policy framework. Good governance is equally important for private sector intermediaries. In Australia, as well as overseas, copyright collecting societies issue collective licences, collect payments from users, and distribute royalties to their rights holder members. Collective licensing has merit to the extent that it can help reduce transaction costs, particularly for high volume, low-value transactions. But the ability to collectively license IP rights can also give rise to market power. OVERVIEW 25 It is for this reason that Australia’s collecting societies are governed by a voluntary code of conduct and (while lesser known) subject to ACCC scrutiny. However, participants raised concerns about the efficacy of the current code of conduct and the extent to which it constrains the behaviour of collecting societies. There are grounds for bolstering these arrangements. The code is voluntary and does not appear to be as robust as those operating in other jurisdictions, such as Europe. The ACCC should review the guidelines to ensure that they not only reflect contemporary international best practice, but are being followed. This review would also inform whether the guidelines are made mandatory. Better understanding and pursuing our international interests A ‘more is better’ mindset, and poor consultation and transparency, have proven problematic in Australia’s international IP dealings. International agreements that commit Australia to implement specific IP provisions — such as the duration of patent or copyright protection — have worked against Australia’s interests. These agreements typically involve trade-offs, and keen to cut a deal, Australia has capitulated too readily. Australia’s cooperation with other countries on IP arrangements should focus on minimising the transaction costs associated with assigning, using and enforcing IP rights, and encouraging more balanced policy arrangements for patents and copyright. Supporting global cooperation among international patent offices through the World Intellectual Property Organization is a good example. Good policy outcomes also depend on a high-quality information and evidence base, underpinned by transparent policy development. Many inquiry participants expressed concerns with Australia’s approach to negotiating IP provisions in international agreements, and the absence of meaningful stakeholder consultation. As international treaties strongly influence Australia’s IP settings, and are difficult to reverse, transparency and substantive public consultation processes are critical. As the Commission and others have previously recommended, greater use of independent and public reviews, and more effective consultation, would improve treaty-making processes. These recommendations are equally applicable to agreements dealing with Australia’s IP arrangements. There is also scope to better identify and articulate defensive and offensive interests. Some examples could include maintaining the right to draft exceptions and limitations (such as in public health) and identifying ‘no go’ outcomes (such as retrospective extensions of IP rights). Finally, the Commission has identified specific reforms that Australia should pursue with like-minded countries in the ‘long game’ of achieving more balanced IP settings. These include introducing formalities for copyright, improving the quality of patents, and 26 INTELLECTUAL PROPERTY ARRANGEMENTS allowing manufacture of pharmaceuticals for export, as well as the publication of clinical trial data. This should not be seen as an exercise in horse-trading or cajoling. Many of the issues are equally problematic in other countries. An overdue review of the TRIPS Agreement by the World Trade Organization would be a helpful first step. 9 An improved IP system has broad benefits International agreements significantly constrain Australia’s flexibility for IP policy reform. Nonetheless, the Commission has identified improvements to better target IP protection while not unduly disadvantaging rights holders. The package of reforms is expected to improve community wellbeing. • Consumers would benefit from access to new and cheaper goods and services, and more easily avoid unintentional infringement. • Government and ultimately taxpayers would benefit from a substantial reduction in health costs through a more efficient PBS. • Rather than hindering innovation and creativity as claimed by some participants, IP reform would also invigorate innovation as: – Australian firms will be able to take full advantage of opportunities in cloud computing solutions – medical and scientific researchers will be able to better utilise text and data mining – universities will have the flexibility to offer Massive Open Online Courses – the education sector will avoid paying millions of dollars each year to use materials that are freely available online – innovative SMEs will be able to innovate without fear of infringing frivolous or strategic patents and be better able to enforce legitimate rights through low-cost dispute resolution mechanisms. Table 1 summarises the anticipated benefits from pursuing the Commission’s recommendations. But achieving reform will not be easy. Some vocal interest groups have long shaped Australia’s IP arrangements to advance their own interests. And in the past, reform efforts have more often than not succumbed to misinformation and scare campaigns. The same tactic has been deployed here, with some parties publishing more fiction than fact about the Commission’s draft report. Government will need to show steely resolve to pursue a better balanced IP system in the face of strong vested interests. OVERVIEW 27 Table 1: Summary of reforms and their expected benefits Proposed reform Expected benefits PATENTS Raise the inventive step for patent eligibility, add an objects clause to the Patents Act, improve patent filing processes, restructure patent fees and abolish the innovation patent system (7.1, 7.2, 7.3, 7.4 and 8.1). Elevate patent quality over time to improve the signal value of patents, reducing thickets, limiting strategic misuse and shortening pendency, stimulating innovation and business activity. Reform extensions of term for pharmaceutical patents (10.1). Restructuring renewal fees will reduce the risk that poor quality patents remain entrenched. Reforming extensions of term will lower the cost of pharmaceuticals, benefiting consumers and saving the government an estimated $258 million each year. Additional public health benefits will arise from improved access to affordable medicines. Improve monitoring of settlements between originator and generic drug companies (10.2). Reducing opportunities for pay-for-delay settlements will ensure timely access to affordable medicines and improve competition in the pharmaceuticals market for the benefit of consumers. COPYRIGHT Australia’s copyright system will better adapt to Replace Australia’s existing fair dealing exceptions in the Copyright Act with a broad technological change and new uses of copyright material, without compromising incentives to create. and open-ended fair use exception (6.1). Improved access to copyright works would increase economic activity and community welfare. Material gains include: • In the case of orphan works, flexible exceptions that improve access are conservatively estimated to generate new economic activity worth between $10 million and $20 million per year. • Consumers would enjoy better access to archived, commercially-unavailable, or otherwise hard-to-access works. • Fair use would end the practice where education and government users pay statutory licence fees for freely available online material, saving taxpayers an estimated $18 million per annum. Repeal parallel import restrictions for books (5.3). Australian consumers will be able to directly access competitively priced books in Australian bookstores. Compared to average selling prices in the UK, prices in Australia are higher by an average of 20 per cent. This will benefit consumers (especially students), Australian bookstores, and overall community welfare. Strengthen the Copyright Act to make clear Consumers of software, TV shows, movies, music and circumventing geoblocking technology is not games gain from better access and more competitive prices. Greater consumer certainty will drive competition and a copyright infringement (5.2). reduce price differentials between Australian and overseas markets — which were about 49 per cent in professional software, 67 per cent in music, and 61 per cent in games in 2013. 28 INTELLECTUAL PROPERTY ARRANGEMENTS Proposed reform Expected benefits ACCC review to ensure best practice in governance, reporting and transparency arrangements for collecting societies (5.4). Best practice governance and transparency will improve the efficiency of collecting societies and their distribution practices, and facilitate fair negotiations between users and rights holders. Separate accounting of statutory and voluntary licence revenue will ensure taxpayer funds achieve value for money. OTHER IP RIGHTS Trade marks Expedite the removal of unused marks, and make it harder to register misleading marks (12.1). Fewer but more accurate trade marks will enhance their value to businesses and consumers. Link the business name and trade mark registers, and allow the importation of legitimately marked goods (12.1). Linking the trade mark and business name registers will reduce renaming and rebranding costs caused by unintentional infringement, while allowing legitimate imports will lead to lower prices and greater choice for consumers. Plant Breeders’ Rights Enable IP Australia to make essentially derived variety declarations in respect of any new plant variety (13.1). Improved enforcement and compliance will increase incentives to invest in pasture and fodder crop breeding, contributing to genetic gain increases and boosting livestock farming productivity and profitability. ENFORCEMENT AND GOVERNANCE Enhance the role of the Federal Circuit Court by introducing a dedicated IP list with caps on claimable costs and damages (19.2). Individuals and SMEs would face lower costs to resolve IP disputes through the court system. Lower risks and costs provide rights holders with greater certainty while improving access to enforcement and justice. Expand the safe harbour scheme to cover all online service providers (19.1). Online service providers, such as cloud computing firms, would face fewer impediments to establish operations in Australia. The copyright system will be more adaptable as new services and technologies are developed, facilitating greater innovation. Aligning with international systems further reduces business uncertainty. Implement an open access policy for publicly-funded research (16.1). Publicly-funded research publications will be cheaper to access, facilitating faster and wider dissemination of the knowledge and ideas contained within them. Identify overarching objectives and a common framework for IP policy development, and establish an interdepartmental policy group and other formal working arrangements between agencies (17.1). Adherence to a whole-of-government policy framework will promote a more balanced and integrated approach to IP policy and its development. Develop best practice guidance for developing IP provisions in international treaties (17.2). Work with like-minded countries through multilateral forums to achieve more balanced IP settings and to reduce transaction costs (18.1). More independent input and transparency in trade negotiations involving IP will promote public confidence and help ensure any changes to IP laws are in Australia’s interests. Greater balance in IP arrangements will facilitate the production of creative works and innovation (including follow-on innovation), boosting productivity. Reducing the risks and costs of seeking protection abroad will facilitate the flow of IP and capital across borders. OVERVIEW 29 Recommendations and findings Chapter 2: An analytical framework for assessing the IP system RECOMMENDATION 2.1 In formulating intellectual property policy, the Australian Government should be informed by a robust evidence base and be guided by the principles of: • effectiveness, which balances providing protection to encourage additional innovation (which would not have otherwise occurred) and allowing ideas to be disseminated widely • efficiency, which balances returns to innovators and to the wider community • adaptability, which balances providing policy certainty and having a system that is agile in response to change • accountability, which balances the cost of collecting and analysing policy–relevant information against the benefits of having transparent and evidence–based policy that considers community wellbeing. Chapter 4: Copyright term and scope FINDING 4.1 The scope and term of copyright protection in Australia has expanded over time, often with no transparent evidence-based analysis, and is now skewed too far in favour of copyright holders. While a single optimal copyright term is arguably elusive, it is likely to be considerably less than 70 years after death. RECOMMENDATIONS AND FINDINGS 31 Chapter 5: Copyright use and licensing RECOMMENDATION 5.1 The Australian Government should amend the Copyright Act 1968 (Cth) to: • make unenforceable any part of an agreement restricting or preventing a use of copyright material that is permitted by a copyright exception • permit consumers to circumvent technological protection measures for legitimate uses of copyright material. RECOMMENDATION 5.2 The Australian Government should: • amend the Copyright Act 1968 (Cth) to make clear that it is not an infringement for consumers to circumvent geoblocking technology, as recommended in the House of Representatives Standing Committee on Infrastructure and Communications’ report At What Cost? IT pricing and the Australia tax • avoid any international agreements that would prevent or ban consumers from circumventing geoblocking technology. RECOMMENDATION 5.3 The Australian Government should proceed to repeal parallel import restrictions for books to take effect no later than the end of 2017. RECOMMENDATION 5.4 The Australian Government should strengthen the governance and transparency arrangements for collecting societies. In particular: • The Australian Competition and Consumer Commission should undertake a review of the current code, assessing its efficacy in balancing the interests of copyright collecting societies and licensees. • The review should consider whether the current voluntary code: represents best practice, contains sufficient monitoring and review mechanisms, and if the code should be mandatory for all collecting societies. 32 INTELLECTUAL PROPERTY ARRANGEMENTS Chapter 6: Fair use or fair dealing — what is fair for Australia? RECOMMENDATION 6.1 The Australian Government should accept and implement the Australian Law Reform Commission’s final recommendations regarding a fair use exception in Australia. RECOMMENDATION 6.2 The Australian Government should enact the Australian Law Reform Commission recommendations to limit liability for the use of orphan works, where a user has undertaken a diligent search to locate the relevant rights holder. Chapter 7: The patent system — getting the fundamentals right RECOMMENDATION 7.1 The Australian Government should incorporate an objects clause into the Patents Act 1990 (Cth). The objects clause should describe the purpose of the legislation as enhancing the wellbeing of Australians by promoting technological innovation and the transfer and dissemination of technology. In so doing, the patent system should balance over time the interests of producers, owners and users of technology. FINDING 7.1 The Raising the Bar initiative moved the inventive step and other elements of patent law in the right direction by raising the threshold for granting a patent. There is a strong case, however, for further raising the threshold. RECOMMENDATIONS AND FINDINGS 33 RECOMMENDATION 7.2 The Australian Government should amend ss. 7(2) and 7(3) of the Patents Act 1990 (Cth) such that an invention is taken to involve an inventive step if, having regard to the prior art base, it is not obvious to a person skilled in the relevant art. The Explanatory Memorandum should state: • a ‘scintilla’ of invention, or a scenario where the skilled person would not ‘directly be led as a matter of course’, are insufficient thresholds for meeting the inventive step • the ‘obvious to try’ test applied in Europe would in some instances be a suitable test. IP Australia should update the Australian Patent Office Manual of Practice and Procedure such that it will consider the technical features of an invention for the purpose of the inventive step and novelty tests. RECOMMENDATION 7.3 IP Australia should reform its patent filing processes to require applicants to identify the technical features of the invention in the set of claims. RECOMMENDATION 7.4 The Australian Government and IP Australia should set patent fees to promote broader intellectual property policy objectives, rather than the current primary objective of achieving cost recovery. To this end, the Australian Government, with input from IP Australia, should: • restructure patent renewal fees such that they rise each year at an increasing rate (including years in which patents receive an extension of term) — fees later in the life of a patent would well exceed current levels • reduce the initial threshold for claim fees, and increase claim fees for applications with a large number of claims. Chapter 8: The innovation patent system RECOMMENDATION 8.1 The Australian Government should abolish the innovation patent system. 34 INTELLECTUAL PROPERTY ARRANGEMENTS Chapter 9: Business method patents and software patents FINDING 9.1 Raising the inventive step, requiring technical features in patent claims, and the inclusion of an objects clause would better balance the patent rights of software innovators and users. Chapter 10: Pharmaceuticals - getting the right policy prescription RECOMMENDATION 10.1 The Australian Government should pharmaceuticals such that they are only: reform extensions of patent term for (i) available for patents covering an active pharmaceutical ingredient, and (ii) calculated based on the time taken by the Therapeutic Goods Administration for regulatory approval over and above 255 working days (one year). The Australian Government should reform s. 76A of the Patents Act 1990 (Cth) to improve data collection requirements for extensions of term, drawing on the model applied in Canada. Thereafter no extensions of term should be granted until data is received in a satisfactory form. FINDING 10.1 There are no grounds to extend the period of data protection for any pharmaceutical products, including biologics. RECOMMENDATION 10.2 The Australian Government should introduce a system for transparent reporting and monitoring of settlements between originator and generic pharmaceutical companies to detect potential pay-for-delay agreements. This system should be based on the model used in the United States, administered by the Australian Competition and Consumer Commission, and include guidelines on the approach to monitoring as part of the broader guidance on the application of the Competition and Consumer Act 2010 (Cth) to intellectual property (recommendation 15.1). The monitoring should operate for a period of five years. Following this period, the Australian Government should review the regulation of pay-for-delay agreements (and other potentially anticompetitive arrangements specific to the pharmaceutical sector). RECOMMENDATIONS AND FINDINGS 35 Chapter 11: Registered designs FINDING 11.1 The Australian Government has committed to implement many of the recommendations made by the Advisory Council on Intellectual Property in its recent review of Australia’s designs system. These measures will help address participant concerns about the cost of acquiring registered design rights, and the lack of understanding of design law. Recommendation 19.2 provides for a low-cost avenue for IP enforcement currently sought by designers. Chapter 12: Trade marks and geographical indications RECOMMENDATION 12.1 The Australian Government should amend the Trade Marks Act 1995 (Cth) to: • reduce the grace period from 5 years to 3 years before new registrations can be challenged for non-use • remove the presumption of registrability in assessing whether a mark could be misleading or confusing at application • ensure that parallel imports of marked goods do not infringe an Australian registered trade mark when the marked good has been brought to market elsewhere by the owner of the mark or its licensee. Section 97A of the Trade Marks Act 2002 (New Zealand) could serve as a model clause in this regard. IP Australia should: • require those seeking trade mark protection to state whether they are using the mark or ‘intending to use’ the mark at application, registration and renewal, and record this on the Australian Trade Mark On-line Search System (ATMOSS). It should also seek confirmation from trade mark holders that register with an ‘intent to use’ that their mark is actually in use following the grace period, with this information also recorded on the ATMOSS • require the Trade Marks Office to return to its previous practice of routinely challenging trade mark applications that contain contemporary geographical references (under s. 43 of the Trade Marks Act) • in conjunction with the Australian Securities and Investment Commission, link the ATMOSS database with the business registration portal, including to ensure a warning if a business registration may infringe an existing trade mark. 36 INTELLECTUAL PROPERTY ARRANGEMENTS RECOMMENDATION 12.2 The Australian Government should amend the Australian Grape and Wine Authority Act 2013 (Cth) and associated regulations to allow the Geographical Indications (GIs) Committee to amend or omit existing GIs in a manner similar to existing arrangements for the determination of a GI (including preserving the avenues of appeal to the Administrative Appeals Tribunal). Any omissions or amendments to GIs determined in such a manner should only take effect after a ‘grace period’ determined by the GI Committee on a case-by-case basis. Chapter 13: Plant Breeder’s Rights RECOMMENDATION 13.1 The Australian Government should proceed to implement the Advisory Council on Intellectual Property’s 2010 recommendation to amend the Plant Breeder’s Rights Act 1994 (Cth) to enable essentially derived variety (EDV) declarations to be made in respect of any variety. Chapter 14: Circuit layout rights FINDING 14.1 Dedicated intellectual property protection for circuit layouts is not ideal and seldom used, but given Australia’s international commitment to protect circuit layouts and no superior alternatives, the best policy option is to maintain the status quo. Chapter 15: Intellectual property rights and competition law RECOMMENDATION 15.1 The Australian Government should repeal s. 51(3) of the Competition and Consumer Act 2010 (Cth) (Competition and Consumer Act) at the same time as giving effect to recommendations of the (Harper) Competition Policy Review on the per se prohibitions. The Australian Competition and Consumer Commission should issue guidance on the application of part IV of the Competition and Consumer Act to intellectual property. RECOMMENDATIONS AND FINDINGS 37 Chapter 16: IP and public institutions RECOMMENDATION 16.1 The Australian, and State and Territory governments should implement an open access policy for publicly-funded research. The policy should provide free and open access arrangements for all publications funded by governments, directly or through university funding, within 12 months of publication. The policy should minimise exemptions. The Australian Government should seek to establish the same policy for international agencies to which it is a contributory funder, but which still charge for their publications, such as the Organisation for Economic Cooperation and Development. FINDING 16.1 The adoption of an additional ‘use it or lose it’ provision for patents owned by publicly-funded organisations is not warranted. Chapter 17: Intellectual property’s institutional arrangements RECOMMENDATION 17.1 The Australian Government should promote a coherent and integrated approach to IP policy by: • establishing and maintaining greater IP policy expertise in the Department of Industry, Innovation and Science • ensuring the allocation of functions to IP Australia has regard to conflicts arising from IP Australia’s role as IP rights administrator and involvement in policy development and advice • establishing a standing (interdepartmental) IP Policy Group and formal working arrangements to ensure agencies work together within the policy framework outlined in this report. The Group would comprise those departments with responsibility for industrial and creative IP rights, the Treasury, and others as needed, including IP Australia. 38 INTELLECTUAL PROPERTY ARRANGEMENTS FINDING 17.1 Australia’s approach to negotiating IP provisions in international treaties could be improved through greater use of independent impact assessment and more meaningful stakeholder consultation. RECOMMENDATION 17.2 The Australian Government should charge the interdepartmental IP Policy Group (recommendation 17.1) and the Department of Foreign Affairs and Trade with the task of developing guidance for IP provisions in international treaties. This guidance should incorporate the following principles: • avoiding the inclusion of IP provisions in bilateral and regional trade agreements and leaving negotiations on IP standards to multilateral fora • protecting flexibility to achieve policy goals, such as by reserving the right to draft exceptions and limitations • explicitly considering the long-term consequences for the public interest and the domestic IP system in cases where IP demands of other countries are accepted in exchange for obtaining other benefits • identifying no go areas that are likely to be seldom or never in Australia’s interests, such as retrospective extensions of IP rights • conducting negotiations, as far as their nature makes it possible, in an open and transparent manner and ensuring that rights holders and industry groups do not enjoy preferential treatment over other stakeholders. Chapter 18: International cooperation in IP RECOMMENDATION 18.1 The Australian Government should: • pursue international collaborative efforts to streamline IP administrative and licensing processes separately from efforts to align standards of IP protection. In so doing, it should consider a range of cooperative mechanisms, such as mutual recognition • use multilateral forums when seeking to align standards of protection. RECOMMENDATIONS AND FINDINGS 39 RECOMMENDATION 18.2 The Australian Government should play a more active role in international forums on intellectual property policy — areas to pursue include: • calling for a review of the TRIPS Agreement (under Article 71.1) by the WTO • exploring opportunities to further raise the threshold for inventive step for patents • pursuing the steps needed to explicitly allow the manufacture for export of pharmaceuticals in their patent extension period • working towards a system of eventual publication of clinical trial data for pharmaceuticals in exchange for statutory data protection • identifying and progressing reforms that would strike a better balance in respect of copyright scope and term. Chapter 19: Compliance and enforcement of IP rights RECOMMENDATION 19.1 The Australian Government should expand the safe harbour scheme to cover not just carriage service providers, but all providers of online services. FINDING 19.1 Timely and competitively-priced access to copyright-protected works is the most efficient and effective way to reduce online copyright infringement. 40 INTELLECTUAL PROPERTY ARRANGEMENTS RECOMMENDATION 19.2 The Australian Government should introduce a specialist IP list in the Federal Circuit Court, encompassing features similar to those of the United Kingdom Intellectual Property Enterprise Court, including limiting trials to two days, caps on costs and damages, and a small claims procedure. The jurisdiction of the Federal Circuit Court should be expanded so it can hear all IP matters. This would complement current reforms by the Federal Court for management of IP cases within the National Court Framework, which are likely to benefit parties involved in high value IP disputes. The Federal Circuit Court should be adequately resourced to ensure that any increase in its workload arising from these reforms does not result in longer resolution times. The Australian Government should assess the costs and benefits of these reforms five years after implementation, also taking into account the progress of the Federal Court’s proposed reforms to IP case management. RECOMMENDATIONS AND FINDINGS 41 1 About this inquiry In August 2015, the Australian Government asked the Productivity Commission to undertake an inquiry into Australia’s intellectual property (IP) arrangements. The terms of reference for this inquiry are set out at the beginning of this report. 1.1 Background to this inquiry Intellectual property refers to ‘ … creations of the mind, such as inventions; literary and artistic works; symbols; names and images used in commerce.’ (WIPO 2011, p. 2) IP arrangements offer opportunities to creators of new and valuable knowledge to secure sufficient returns to motivate their initial endeavours or investment. In this respect, they are not different from the property rights that apply to ownership of physical goods. However, unlike physical goods, IP rights are not granted in perpetuity and there are limitations on their application. These limits recognise that the use of an idea by one party does not reduce its capacity for use by another, and that ideas provide economic and social value, as others draw on new knowledge to create their own. Since new ideas are a major source of economic growth, any defects in arrangements that encourage their creation and diffusion can be very costly. IP rights take a variety of forms. The most familiar are patents, copyright and trade marks, but there are quite a few more, including rights over performances, designs, plant varieties and circuit layouts (box 1.1). Impetus for this inquiry As noted in the terms of reference, there have been increases in the scope and duration of IP protection. At the same time, the global economy and technology have been changing. The Australian Government is reviewing IP arrangements to ensure that the appropriate balance exists between incentives for innovation and investment, and the interests of individuals and businesses in accessing ideas and products. Although there has been a number of reviews of IP in Australia in recent years, they have focused on specific areas of IP, such as innovation patents, pharmaceutical patents, design protection, and copyright (ACIP 2015a; ALRC 2013; Harris, Nicol and Gruen 2013; Johnson et al. 2015) (figure 1.1). ABOUT THIS INQUIRY 43 Box 1.1 IP rights Patents protect products or processes. In exchange for the exclusive rights provided, patent owners must make technical information about their invention publicly available. Copyright protects the original expression of literary, musical, artistic and dramatic works, as well as their industrial form, such as books, sound recordings, films and broadcasts. Music, paintings, sculptures, computer programs, databases, advertisements, maps and technical drawings can all qualify for copyright protection. Plant breeder’s rights provide protection for new plant varieties. Circuit layout rights protect the layout design (three-dimensional topography) of integrated circuits (commonly known as semi–conductor chips). Trade marks distinguish the goods or services of one firm from those of other enterprises. A trade mark can be a letter, number, word, phrase, smell, shape, logo, picture, or aspect of packaging. Geographical indications identify goods as originating in a specific territory, region or locality where a particular quality, reputation or other characteristic is essentially attributable to its geographical origin. Registered designs protect the appearance of a product, such as its shape, configuration, pattern and ornamentation. Figure 1.1 Australia’s IP policies have been frequently revieweda Competition policy Copyright Enforcement Governance International cooperation Patents Plant Breeder's Rights Public Research Registered designs Trade marks Whole System 1999 2001 2003 2005 2007 Year 2009 2011 2013 2015 a Size of mark indicates number of reviews released in that year. A small circle signifies one review and a large circle represents more than one review. While the Commission has considered the recommendations and findings of these reviews, it has taken a more holistic perspective to identify ways that the IP system as a whole could be improved. A key benefit of this broader perspective is that it can facilitate a more coherent approach across the different IP rights. As noted by the recent Competition Policy Review (Harper Review), there has been ‘no overarching IP policy framework or objective guiding changes to IP protection’ (Harper et al. 2015, p. 104). 44 INTELLECTUAL PROPERTY ARRANGEMENTS 1.2 What has the Commission been asked to do? The terms of reference direct the Commission to consider whether current arrangements provide an appropriate balance. In recommending changes to the current system to improve the overall wellbeing of Australian society, the Commission is to have regard to: • incentives for innovation and investment, including freedom to build on existing innovation • Australia’s international obligations • the relative contribution of IP to the Australian economy • the economywide and distributional consequences of recommendations, including their impacts on trade, investment and competition • ensuring the IP system will be efficient and robust through time, in light of economic changes • how proposed changes fit with, or may require changes to, other existing regulation or forms of assistance • the relevant findings and recommendations of recently completed reviews. On 1 February 2016, the Minister for Education and Training wrote to the Treasurer requesting that the Commission have regard to a finding of the recent Review of Research Policy and Funding Arrangements. In particular, the review concluded that the Commission would be better placed to give further consideration to ‘use it or lose it’ arrangements for IP arising from public funding, as part of this inquiry. 1.3 Scope of the inquiry For the purposes of this inquiry, the Commission has taken Australia’s IP arrangements to include: • formal IP rights, including patents, trade marks, geographical indications, registered designs, plant breeder’s rights, copyright, and circuit layout rights • the governance and institutional settings for IP, including the processes through which IP rights are established, used and enforced (figure 1.2 and box 1.2) • Australia’s involvement in IP internationally through international agreements and institutions (such as the World Intellectual Property Organization and other multilateral frameworks) • IP–related provisions in other laws (such as the Competition and Consumer Act 2010 (Cth)) • the intersection between IP rights and public funding for research agencies and universities. ABOUT THIS INQUIRY 45 Figure 1.2 Intellectual property rights sit within a broader landscapea a The Department of Communications and the Arts administers copyright and circuit layout rights. b Department of Foreign Affairs and Trade (DFAT); Department of Industry, Innovation and Science (DIIS); Department of Communications and the Arts (DCA). c Australian Competition and Consumer Commission (ACCC). 46 INTELLECTUAL PROPERTY ARRANGEMENTS Box 1.2 Establishing, using and enforcing IP rights Establishing IP rights IP rights establish ownership and exclusive rights. IP rights vary in their scope, duration and safeguards. While much focus is placed on the rights themselves, IP arrangements also include the mechanisms and institutions that define IP rights as well as the way that rights are used and enforced. The use of IP rights Owners of IP rights have a variety of ways to realise the value of their IP. They can elect to keep, licence, or sell their rights to other parties. On rare occasions they may be required to licence them (compulsory licensing). The way in which rights holders elect to realise the value of their IP is influenced by a range of factors including whether: • it is economical to produce the IP themselves • they can generate more economic rent by licensing their rights • they are seeking to expand their product line by licensing their rights in exchange for a license back of other IP rights (cross–licensing) • they are seeking to acquire IP for the sole purpose of enforcing rights to generate a profit (assertion entities) • they are acquiring IP rights to avoid litigation (defensive buyers). Enforcement of IP rights The value of IP rights to creators — and the value to users — depends on the ability of both sides to enforce their rights. ‘Enforcement’ means different things to rights holders and users, but can include the ability of a patent holder to stop an alleged infringer from producing copies of their goods, a public interest organisation challenging the validity of a patent claim, a firm challenging the use of a trade mark or geographical indication, or an artist seeking to prevent infringing copies of their work being shared online. A broad assessment of Australia’s innovation system is outside the scope of this inquiry. However, the ability for people to use informal IP protections (such as trade secrets) and access other incentives for innovation (such as government funding or tax incentives) are both relevant for assessing possible changes to Australia’s IP arrangements and understanding the role of IP in the broader innovation system. ABOUT THIS INQUIRY 47 1.4 Conduct of this inquiry The Commission has consulted widely with stakeholders, drawing on input from participants through meetings, visits, roundtable discussions, written submissions and public hearings (appendix A). • The Treasurer issued the terms of reference for this inquiry on 18 August 2015. The Commission subsequently released an issues paper on 7 October 2015, inviting public submissions. • The Commission held informal consultations and roundtable discussions with governments, regulatory bodies, peak industry groups, as well as a number of private and government organisations. • The Commission released a draft report on 29 April 2016. The draft report was informed by 148 initial submissions and included the Commission’s analysis, findings and draft recommendations at that time, as well as requests for feedback on particular issues. Stakeholders were invited to provide formal public submissions on the draft report. • In mid-May 2016, the Commission met with IP experts and institutions in France, Germany, Switzerland and the United Kingdom to better understand international approaches to IP and inform its analysis for the final report. • During June 2016, the Commission held public hearings in Brisbane, Canberra, Sydney, and Melbourne to allow participants to respond to the proposals in the draft report. There were 121 presenters over 6 hearing days. The Commission published transcripts of proceedings on the inquiry website. • The Commission also held roundtable discussions with key participants on the topics of fair use (15 June 2016) and pharmaceutical patents (17 June 2016). The presentations from these roundtables are also available on the inquiry website. • The Commission received 620 submissions (148 pre-draft and 472 post-draft). Non-confidential submissions are available on the inquiry website. 1.5 A guide to the remainder of this report This report consists of an overview, 19 chapters and 8 appendixes. The Commission has structured the chapters into three parts (figure 1.3). • Chapters 2 and 3 outline the framework for assessing Australia’s IP arrangements and consider how the system is faring overall. • Chapters 4 to 14 examine specific forms of IP rights and options for their reform. • Chapters 15 to 19 examine cross–cutting issues. 48 INTELLECTUAL PROPERTY ARRANGEMENTS Figure 1.3 Framework Chapter 1 About this inquiry Chapter 2 An analytical framework for assessing the IP system Chapter 3 How does the system fare? Structure of the report Specific forms of IP rights Chapter 4 Copyright term and scope Chapter 5 Copyright use and licensing Chapter 6 Fair use or fair dealing what is fair for Australia? Chapter 7 The patent system getting the fundamentals right Chapter 8 The innovation patent system Chapter 9 Business methods patents and software patents Chapter 10 Pharmaceuticals getting the right policy prescription Chapter 11 Registered designs Chapter 12 Trade marks and geographical indications Chapter 13 Plant breeder?s rights Chapter 14 Circuit layout rights Cross-cutting issues Chapter 15 Intellectual property rights and competition law Chapter 16 IP and public institutions Chapter 1 7 Intellectual property?s institutional arrangements Chapter 18 International cooperation in IP Chapter 19 Compliance and enforcement of IP rights ABOUT THIS INQUIRY 49 2 An analytical framework for assessing the IP system Key points • The economic characteristics of knowledge are such that there is likely to be an under provision of ideas and innovation without some form of intervention by government. • Intellectual property (IP) rights are intended to encourage creative and innovative activity by providing a range of exclusive rights to stop others from using the expression of ideas without permission or payment. • IP arrangements can correct the under provision of new innovative products and processes to the benefit of society, but they also generate costs. ‘Stronger’ IP rights are not necessarily ‘better’ from a community wide perspective. • The reach of IP and IP policy has expanded over recent decades, is now far-ranging, and is still growing. The evolution of how IP rights are assigned, used and enforced has advanced without a clearly articulated policy objective for the IP system. The lack of a policy objective has hindered past IP reform efforts, and has the potential to stymie future reform. • The objective of the IP system should be to maximise the wellbeing of all Australians. To this end, the Commission has adopted four principles by which to assess and frame a well-functioning IP system. These principles are: – effectiveness — that the IP system encourages the creation and dissemination of valuable ideas that would not have occurred in the absence of the system – efficiency — that ideas are generated by the most efficient, lowest-cost creators, traded freely, and do not unduly impede competition – adaptability — that the IP system adapts to changes in technology, markets and economic conditions – accountability — that changes to the IP system are transparent, evidence-based and reflect community values. • The Commission has adopted an economic framework to assess the different dimensions of the IP system. An economic approach is most appropriate to achieve the objective of maximising wellbeing as it considers the effects of current arrangements and potential reforms on all parties, and only pursues change where the benefits to the Australian community are likely to exceed the costs. The analytical framework outlined in this chapter is the lens the Commission uses to assess and make recommendations to improve the intellectual property (IP) system. The framework adopted recognises the need for a balance of incentives between creators, rights holders and users of IP and the broader costs and benefits that IP rights can impose on the AN ANALYTICAL FRAMEWORK FOR ASSESSING THE IP SYSTEM 51 rest of society. It also underscores the importance of an IP system that is adaptable and accountable through time. 2.1 How do IP arrangements affect wellbeing? Firms, households and governments all produce and use ideas, and do so in different ways. Businesses develop new products and processes, which are often built upon previous IP and are themselves used as inputs by other businesses. Households purchase goods and services that embody or are produced using IP. Governments fund the creation of IP directly through public institutions like the CSIRO and universities. The reach of IP — and thus IP policy — is broad and growing: IP’s overall role in economies has evolved from a policy area that was mainly relevant to a handful of industries to a force that influences a wide swath of demand and sectors. Consequently, IP policy has become a more influential framework condition that affects not only innovation, but trade, competition, taxes, consumer protection, and other areas. (OECD 2015a, p. 6) The incentives created by the IP system to develop new expressions of ideas 1 — in the form of inventions or creative works — are part of the innovation system. This system encompasses a broader set of activities, institutions and linkages between sectors that allow for the creation and use of innovations (chapter 1). The IP system defines how rights are: • assigned — including the way legislation is used to satisfy the goals of IP policy and the effects that international obligations have on society • used — how IP influences markets, including the commercialisation, licensing and trade of IP • enforced — how enforcement agencies and the courts adjudicate the rights to IP and the resulting transactions costs. In doing so, the IP system affects the incentives to innovate and use innovations. Australia’s IP arrangements affect community welfare through their impact on productivity growth and national income. Having a clear policy objective for the IP system is therefore important. The IP system is aimed at addressing impediments to the creation and use of ideas Unlike physical capital goods such as machinery or equipment, knowledge and ideas are not tangible or necessarily finite. The use of an idea does not consume it — it is still 1 IP provides a means to protect the expression of ideas and knowledge, such as in the form of inventions or creative works, rather than the underlying ideas themselves. For convenience, the ‘expression of ideas’ is simply referred to as ‘ideas’ in the rest of this chapter and where relevant, in the rest of this report. 52 INTELLECTUAL PROPERTY ARRANGEMENTS available for others to use and share. It is these inherent characteristics of knowledge that can lead to an undesirable outcome, both in terms of the creation and in the use of new knowledge (box 2.1). As put by Ergas in the context of creative works: The main feature that flows from these characteristics is that, left to their own devices, it will be difficult if not impossible for investors in creative effort to align the rewards from that investment with its costs. They will not, in other words, be able to bargain their way around the externalities associated with creative effort. The result will be under-investment and distorted investment in creative effort and in activities complementary to creative effort. At the same time, control over the outputs of creative effort will not be transferred to those who can put these outputs to most productive use. (2002, p. 11) Box 2.1 Characteristics of knowledge and ideas Consumers demand inventive and creative output, and in a well-functioning market, creators would invest resources into developing new knowledge to satisfy this demand. However, a number of attributes of knowledge and ideas mean that creators and innovators may not produce as much as consumers would be willing to buy. Knowledge can be: • ‘non–rivalrous’ — when someone uses an idea, it does not stop others from using that idea • ‘non–excludable’ — it can be difficult to prevent other people from using someone’s inventive or creative ideas • cumulative — new ideas often build upon old ideas • subject to ‘network effects’ — in some cases, it can be easier to generate IP in a cooperative, rather than competitive environment. This has implications for legal frameworks about collusion and the overriding general desire for competition in markets. The costs borne by creators in developing new ideas is often considerably higher than the cost of reproducing that knowledge many times over — innovation is sometimes characterised by large upfront costs (‘sunk costs’) and low costs for use (‘marginal costs’). Coupled with the non-excludable nature of ideas, creators who bear the cost of developing new works may be unable to compel others who reproduce their work to contribute to the original development costs. Unable to earn a sufficient return on their investment, creators might instead opt not to produce that new knowledge. Sources: Adapted from IPCRC (2000); OECD (2016c). The principal justification for a system of IP rights is to overcome this ‘public good problem’ by providing a way to exclude the use of ideas without consent (or payment) of the owner. But while this provides a broad justification for an IP rights regime, it says little about the scope, duration and form that such rights should take, or even whether IP rights are the best means to solve the problem in the first place. A well-defined objective for the IP system — what the system is meant and not meant to achieve — is necessary in order to assess and determine the most appropriate policy settings. AN ANALYTICAL FRAMEWORK FOR ASSESSING THE IP SYSTEM 53 2.2 What should Australia’s IP system achieve? Despite the broad reach of the IP system, it has no clearly articulated objective. As noted by the Harper competition review: The Panel is concerned that Australia has no overarching IP policy framework or objectives guiding changes to IP protection or approaches to IP rights in the context of negotiations for international trade agreements. (Harper et al. 2015, p. 41) Clear articulation of a policy objective would help to ensure that all elements of the IP system are consistent and ‘pulling in the same direction’, while providing regulators, government and the judiciary a common understanding of what the IP system is meant to achieve. But the Commission has found little consensus as to what the objective of the IP system should be, beyond some broad themes (table 2.1). These include that it should: • provide some incentive for the creation of ideas and innovations that would not otherwise occur • benefit both creators and the public • not unduly hinder competition. For the purposes of this inquiry the Commission’s view is that the IP system’s overarching objective should be to recognise and encourage the creation of new and valuable ideas and innovations in a way that maximises the wellbeing of all Australians. To implement this objective the Commission has identified and applied a set of principles that also meet the specific goals of the IP system outlined in the terms of reference for the inquiry. 54 INTELLECTUAL PROPERTY ARRANGEMENTS Reviews The general objective of the system of intellectual property law in Australia is utilitarian, and more specifically economic, rather than moral in character. It serves to (and, in the Committee’s view, should aim to) maximise the difference between the social value of intellectual property created and used, and the social cost of its creation, including the cost of administering the system. More specifically, the intellectual property laws ought to provide incentives for efficient investment in innovation. (IPCRC 2000, p. 24) Where innovation is difficult to copy, or there are large rewards for being first to market with a product, the competitive spur to innovation is effective. In other circumstances, these risks and costs are a disincentive to innovate. That is why we need intellectual property rights. (Hargreaves 2011, p. 11) Government Appropriate intellectual property settings and enforcement mechanisms are important in positioning Australia in the global context as having a policy and economic environment that is conducive to innovation and open to trade, investment and capital movement. (Department of Foreign Affairs and Trade, sub. 65, pp. 3–4) Australian IP law is designed to encourage innovation and protect businesses that develop original IP in order to have a competitive advantage. (Austrade 2015) The objective of the intellectual property (IP) rights system is to support innovation by encouraging investment in research and technology in Australia and by helping Australian businesses benefit from their good ideas. (Carr 2011, p. 8) IP rights allow owners to temporarily exclude others from using their invention or creation. The existence of IP rights encourages innovation and prevents competitors from imitating products and services. The promotion and protection of IP rights enhances economic growth, while creating new industries and jobs. An efficient IP system benefits innovators and the public. (IP Australia 2015b, p. 5) [The Department] considers the policy direction for reforms should focus on an IP system optimised to promote future opportunities for economic growth and job creation. At the same time, it should be mindful of the need to balance the broader social requirement for dissemination of knowledge and competitive access to new technology and creative content. (Department of Industry, Innovation and Science, sub. DR615, pp. 2–3) Business Perspectives on what the IP system is meant to achieve A sound IP framework is an important facilitator of the innovation that Australia needs to build a more diverse and resilient economy. A key to a sound IP system lies in striking the appropriate balance between rewarding past efforts and encouraging new ideas. Innovative ideas will always be undersupplied in competitive markets with no IP regime because innovators have no way to recoup costs if their ideas can be quickly copied. (Australian Industry Group, sub. 60, p. 5) … in a “knowledge economy”, intellectual property should encourage creativity and investment by providing incentives that take into account the potential for bountiful rewards when a creator creates something that society values. (Australian Institute of Professional Photography, sub. DR387, p. 2) International Organisations Table 2.1 IP is protected in law … which enable people to earn recognition or financial benefit from what they invent or create. By striking the right balance between the interests of innovators and the wider public interest, the IP system aims to foster an environment in which creativity and innovation can flourish. (WIPO nd) In the age of the knowledge economy, the efficient and creative use of knowledge is a key determinant of international competitiveness, wealth creation and improved social welfare. An effective [IP] system embedded within a national strategy which anchors IP considerations firmly within the policy-making process will help a nation to promote and protect its intellectual assets, thereby driving economic growth and wealth creation. (Idris in Jaiya 2007) IP rights support innovation by making it a more worthwhile investment and encouraging knowledge diffusion. The economic rationale for IP rights is that it is in everyone’s long-term interest for people and businesses that create knowledge to have well-defined, enforceable rights to exclude third parties from appropriating their ideas, or the expression of their ideas, without permission. Failing to put restrictions on appropriating others’ inventions and creations would dilute the rewards for investing in innovation, thereby reducing the incentives for making such investments. (OECD 2015a, p. 5) AN ANALYTICAL FRAMEWORK FOR ASSESSING THE IP SYSTEM 55 A principles–based approach to the IP system … The needs and wants of the different users of the IP system are complex, interconnected and sometimes at odds with one another. Determining ‘prescriptive rules’ that satisfy all users of the system and are in the long-term interests of the Australian community is difficult. The Commission has therefore applied principles in its assessment of the IP system and as a basis to recommend welfare-enhancing reforms. The merit of a principles-based approach is that it is flexible and can adapt more readily to change, including to changes in the cost of innovation, markets and economic conditions. This is in contrast to the present system, which is often left trying to ‘catch up’: … the current framework of regulation based on specific forms of intellectual property has been left behind by the rapid change and convergence of technologies. Previous and ongoing reviews of specific legislation, such as those covering copyright, designs, patents and trademarks, have not examined the overall framework for intellectual property protection. In addition, changes to intellectual property regulation have occurred without careful economic analysis. As a consequence, producer interests dominate over interests of users. (PC 1996, p. 154) … with an economic framework at its heart The nature of the tradeoffs between users of the IP system, and the impacts IP rights have on the rest of the economy, mean any principles need to be based on a framework that considers both those that stand to benefit and lose from policy change. A number of frameworks can be applied in this regard, some of which are summarised in table 2.2. The pros and cons of each are subjective, but implicitly many of them are aimed at promoting the interests of segments of the community. Because the reach of the IP system is so broad, its settings should take into account all users. This includes the creators of ideas, those wishing to subsequently develop those ideas further, and the broader Australian community that use goods and services, which embody or are created with IP. The settings of the IP system should consider the needs of the community today, but also the wellbeing of the community in years to come. 2 The costs and returns to some inventions and creations can occur over years, if not decades. Thus, when designing the goals of an IP system, the longer-term incentives to supply and use creative ideas need to be considered, rather than just making sure the institutional settings are appropriate for the ‘here and now’. 2 Wellbeing is a complex concept that is central in assessing policy changes affecting people’s lives. Both the OECD (2011a) and the Treasury (2011) have frameworks for understanding and measuring wellbeing. 56 INTELLECTUAL PROPERTY ARRANGEMENTS Table 2.2 Stylised examples of approaches to the IP system Approach Description Potential advantages and disadvantages ‘Legal’ A prescriptive regime governed by statute and regulation, which affects the evolution of case law. Can provide a clear picture of ownership of rights, but the evolution of case law can introduce complexity into who owns which ideas, and encourage costly litigation. ‘Natural justice’ The ownership and rights to some IP is always inextricably linked, in part or in full, to the original creator. Provides a clear picture of ownership, but can act as a restraint of trade in rights and lead to frequent disputes over who the original creators may be (especially in the case of creator-centric and derivative works). ‘Command’ That the public good aspects of IP are so strong that rights should be curtailed and IP left largely to the state to provide. Could make for easier cumulative innovation, but reduces the incentive for new innovations. It may lead to an overreliance on the public sector to direct resources to innovation. ‘Internationalist’ That trade, and the regulation of trade, requires compromises between nations, which include the harmonisation of IP rights. Can lead to a system of harmonised rights across countries that facilitate trade in IP. But can also reduce the ability of nations to adjust their IP system as needs and circumstances change. ‘Mercantilist’ That the rights of domestic IP holders should be assigned (and enforced) more enthusiastically than that of IP from overseas. Encourages the domestic production of IP, but can discourage trade in IP and increase the costs of IP. Accordingly, the Commission has chosen to employ an economic framework to assess the IP system. Such an approach puts the welfare of the whole community as the overarching objective and recognises that changes should only be made to the IP system so that those that gain could compensate those that lose, without being any worse off themselves. Such an approach is also consistent with the last broad based review into IP arrangements in Australia, and with other reviews of IP undertaken in the UK (IP Australia, sub. 23). Some participants in the inquiry expressed concerns that an economic approach may not account for all the effects on welfare that could stem from changes to the IP system. For example: We note the possibility that a community may accept a level of less than perfect economic wellbeing in order to achieve other desirable goals by which it judges its overall general wellbeing. We recommend that the policy analysis should enable those other factors to be considered. (CSIRO, sub. 126, p. 3) The proposed framework appears to only take account of values that can be measured and calculated to a financial value, and does not take account of cultural, personal or social values that are inherent of the creation of artwork. In particular, the proposed framework fails to take account of the differences between utilitarian applications of intellectual property and cultural and aesthetic applications of intellectual property. (Arts Law Centre of Australia, sub. 117, p. 3) However, an economic approach does attempt to account for all welfare changes, including those that can be difficult to ‘monetise’. What can be contentious is finding consensus on AN ANALYTICAL FRAMEWORK FOR ASSESSING THE IP SYSTEM 57 how to measure and value such welfare changes when setting the parameters of an IP system. This is likely to be more difficult for some IP rights than others — where relevant, these considerations are addressed in the chapters examining specific IP rights. An economic framework can be helpful to examine traditional knowledge Another concern raised by some participants in the inquiry was whether an economic framework was the appropriate lens to assess the way that IP rights apply to Indigenous cultural goods or traditional knowledge (Australia Council for the Arts, sub. DR553; Australian Institute of Aboriginal and Torres Strait Islander Studies, sub. DR583). This concern relates to the broader issue of how Indigenous Cultural and Intellectual Property (ICIP) is used and accessed — especially the difficulties that can arise when IP is created using the traditional knowledge of communities: The indigenous people would like to see a stronger regime that actually protects their traditional cultural expressions and their traditional knowledge, which are embodied in work that they create. … it comes at intellectual property from sort of a different perspective [which is] very communal in its nature and it has been passed down from generation to generations over thousands of years. So it’s not all about individual rights for individual rights holders or creators. (Arts Law Centre of Australia, trans., p. 137) ICIP encompasses a wide range of material such as genetic resources, traditional knowledge and cultural expression, which often has profound importance or significance to Aboriginal and Torres Strait Islander Australians. 3 Accordingly, the protection of ICIP is affected by more than the IP system alone (figure 2.1). While numerous reviews have previously examined the protection of ICIP from a variety of perspectives, few have tackled the issue with a holistic approach (box 2.2). This was noted in the Australian Law Reform Commission’s recent review of the Native Title Act 1993 (Cth): The ALRC considers that the question of how cultural knowledge may be protected and any potential rights to its exercise and economic utilisation governed by the Australian legal system would be best addressed by a separate review. An independent inquiry could bring to fruition the wide-ranging and valuable work that has already been undertaken but which still incompletely addresses the protection of Aboriginal and Torres Strait Islander peoples’ cultural knowledge. (ALRC 2015, pp. 263–264) The Commission considers that the proposed analytical framework is helpful to examine IP issues relating to ICIP. Accordingly, where the Commission has considered the protection of ICIP, it has done so from the perspective of maximising the welfare of the Australian community through the assignment, use and enforcement of all IP. Given the wide-ranging 3 As acknowledged by the National Congress of Australia’s First Peoples, ICIP of Aboriginal and Torres Strait Islander Peoples can extend to literary, performing and artistic works (including songs, music, dances, stories, ceremonies, symbols, languages and designs); scientific, agricultural, technical and ecological knowledge; all items of movable cultural property; knowledge about culture, roles and relationships; human remains and tissues; immovable cultural property (including sacred sites, historically significant sites and burial grounds); and documentation of Indigenous Peoples’ heritage in archives, film, photographs, videotape or audiotape and all forms of media (NCAFP 2013, p. 5). 58 INTELLECTUAL PROPERTY ARRANGEMENTS nature of factors that affect ICIP, the Commission’s recommendations about improved governance arrangements (chapter 17) will further ensure a broad based examination of IP issues confronting Aboriginal and Torres Strait Islander Australians in the future. Figure 2.1 ICIP is affected by more than IP Laws Box 2.2 Indigenous Cultural and Intellectual Property (ICIP) The scale of what ICIP covers and how it should be protected is reflected in the numerous reviews that have touched on ICIP in the past (Davis 1996; Janke 2012; Stoianoff and Roy 2015). Within Australia, these include: • Report of the Working Party on the Protection of Aboriginal Folklore (1981) • Report on the Recognition of Aboriginal Customary Law (1986) • Report of the Review Committee on the Aboriginal Arts and Crafts Industry (1989) • Report of the Royal Commission into Aboriginal Deaths in Custody (1991) • Creative Nation (1994) • the issues paper ‘Stopping the Rip-Offs: Intellectual Property Protection for Aboriginal and Torres Strait Islander Peoples’ (1994) • the Social Justice Reports (1995) • the Australian Copyright Council Discussion Paper Indigenous Intellectual Property Rights: A Copyright Perspective (1997) • ‘Our Culture Our Future’ report (1997) • the Law Reform Commission of Western Australia report ‘Aboriginal Customary Laws: The Interaction of Western Australian Law with Aboriginal Law and Culture’ (2006) • the Australian Law Reform Commission’s Review of the Native Title Act (2015). (continued) AN ANALYTICAL FRAMEWORK FOR ASSESSING THE IP SYSTEM 59 Box 2.2 (continued) However, despite extensive review, there has been little legislative change in response, with many recommendations stemming from these reviews not adopted by governments. (Janke 2012). Consultation is still ongoing, with IP Australia presently seeking further information about how Indigenous Knowledge and the IP system intersect, with the goal of facilitating further public discussion on the topic (IP Australia 2016h). An alternative to a legislated approach has been the use of protocols, which are a voluntary means to use ICIP in a respectful way that still benefits all parties. For example, the Australia Council has published free protocol guides for working with Indigenous artists (Australia Council, sub. DR553). The advantage of protocols is that they may be a more flexible means of establishing protection that can be adapted to particular subject matter (Janke 2012). But adoption of protocols is not mandatory, and some groups have indicated that the incidents of goodwill and cooperation are outweighed by circumstances where protocols ‘have been ignored and cultural sensitivities trampled upon’ (Arts Law Centre of Australia 2016b, p. 5). The Victorian Government has recently put in place legislation to establish a scheme for some elements of ICIP in the Aboriginal Heritage Amendment Act 2016 (Victoria). Features of the scheme include the establishment of a group of traditional owners to adjudicate on when ICIP may be used, in what form, and when royalties are payable. 2.3 What principles are important for the IP system? For this inquiry, the Commission has identified four principles that, when applied, should lead to a balanced and well-functioning IP system. These principles are ‘effectiveness’, ‘efficiency’, ‘adaptability’, and ‘accountability’. Taken collectively, they form the means to achieve the goal and objective of the IP system (figure 2.2). Each of these principles is discussed in detail, below. 60 INTELLECTUAL PROPERTY ARRANGEMENTS Figure 2.2 The Commission’s approach Overarching objective: to maximise wellbeing of Australians Goal: That the IP system provides appropriate incentives for innovation, investment and the production of creative works while ensuring it does not unreasonably impede further innovation, competition, investment and access to goods and services. Principles to apply to the IP system to achieve this goal: Effective Efficient Adaptable Accountable The system should be effective in encouraging additional ideas and in providing incentives that ensure knowledge is disseminated through the economy and community. The system should provide incentives for IP to be created at the lowest cost to society. This principle includes considering whether IP rights generate returns that are sufficient to encourage new ideas, the relative merits of public and private IP generation, and the longer-term effects on competition and innovation from granting IP rights. The system should adapt to changes in economic conditions, technology, markets and costs of innovating. The policies and institutions that govern the system, and the way that changes are made to them, need to be evidence-based, transparent, and reflect community values. Effectiveness: targeting additional innovation and creative output An effective IP system is one that rectifies the inherent problems in the supply and use of ideas. It must provide incentives that help to overcome the potential under provision of ideas that stem from the public–good nature of knowledge, as well as make sure that the ideas created are disseminated widely to allow for further innovation to occur. Where it is not possible for the system to correct the under provision entirely, it should still seek to address the most onerous problems that inhibit the creation and dissemination of new ideas. The creation of additional and novel IP needs to be encouraged An IP system is effective if it promotes the creation of genuinely new and valuable ideas, which would not have occurred in the absence of IP rights. This ‘additionality’ of ideas is important given that the objective of the IP system is to improve wellbeing by correcting an under provision of knowledge and innovation that may exist (figure 2.3). While the IP system can provide a monetary incentive to correct this under provision of new ideas, it is also the case that many creators see value in the system because it gives them recognition of their works. AN ANALYTICAL FRAMEWORK FOR ASSESSING THE IP SYSTEM 61 Figure 2.3 Conceptualising additionality The key concept of additionality can be summed up as a question: ‘Would the ideas have happened anyway?’ In many cases, the answer would be ‘yes’ even in the absence of IP rights. Competitive pressures may provide sufficient incentive for creators to produce and commercialise new ideas and innovations. The return they receive between when their product is brought to market and when others start to use the same ideas can be enough to merit the innovation in the first place. Such an outcome is not rare — a relatively large proportion of innovation-active firms do not use a form of IP protection — around 70 per cent of all firms, and 30 per cent of firms with more than 200 employees (ABS 2014). Other measures also compensate for the costs and risks associated with innovating, like R&D assistance, taxation incentives, and a legal system to pursue actions of passing off and breach of contract. In other cases, some parties create works for pleasure, which improves their welfare. In most cases, however, it may not be possible to separate the IP rights from the other incentives provided. For example, a firm may use an R&D incentive to undertake innovation only if there is an IP right. This demonstrates there is a likely tradeoff between IP rights and other measures to encourage the creation of ideas. However, whether IP rights, as presently framed, are effective in encouraging the creation of new, additional ideas and innovations is contentious (for example, Boldrin and Levine 2008; Moir 2013; Stiglitz 2008). As put by the Australian delegate at an OECD roundtable on competition policy and IP rights: Some have argued that little, if any, additional investment in intellectual property is generated by IP [rights]. Certainly there is no consensus concerning how much additional creative and inventive activity is induced by IP [rights] which would not otherwise occur. (OECD 1997, p. 73) 62 INTELLECTUAL PROPERTY ARRANGEMENTS The Centre for Law and Genetics made a similar point, as well as stressing the difficulties associated with measuring additionality and assessing what innovations might occur in the absence of IP rights: There exists a large body of economic research that has attempted to analyse whether or not intellectual property (IP) encourages innovation, and to provide an analytical framework, including appropriate measurements on which to base this analysis. However, we are not aware of any comprehensive framework for addressing the question of whether IP rights encourage genuinely innovative and creative output. Nor are we aware of the existence of an appropriate measure of the overall impact of IP to determine whether creative or innovative outputs would nevertheless have occurred in its absence. (sub. 61, p. 4) From a policy perspective, then, if IP rights are serving only to reallocate rents within an economy, and not encouraging the creation of new and valuable ideas, then the IP system is not achieving its objective. Dissemination is also essential for an effective IP system While an IP system must provide incentives necessary for new, additional ideas to be created, it is also important for the system to encourage the dissemination of those ideas for further innovation. The need for dissemination is explicit in some IP rights, 4 but the way rights are assigned, enforced and used all affect how ideas are diffused and used in further innovation. Greater dissemination of ideas, however, may reduce the returns to some innovations and can reduce the incentive to undertake new research in the first place. Thus there is a ‘balancing act’ between allowing for the greatest dissemination and the greatest incentive to innovate. As put by Nordhaus (1969), the optimal degree of IP rights balances the marginal benefit from increased innovation and the marginal cost from reduced diffusion. And as put by FICPI Australia: … it is most important to strike the right balance that both supports the originators and innovators of IP, and permits improvement in competition in the marketplace by follow-on innovators. If the pendulum is swung too much in favour of the latter, then real innovation in Australia will be stifled, allowing overseas corporates and multinationals to use their financial resources to further entrench marketplace dominance and prevent local start-ups competing in Australia and the world stage, whereas swinging the pendulum too much in favour of the former may indeed restrict greater engagement of local industry in a marketplace dominated by IP savvy overseas corporates and multinationals, to participate in it with socially valuable products and services. (FICPI Australia, sub. DR581, p. 3) Even in cases where an innovation may no longer have a commercial market, having it locked away behind an IP right is still detrimental to the community, as it can hinder follow-on innovation from occurring. Overly strong restrictions on the diffusion of ideas 4 For example, in the disclosure provisions of patents, fair dealing provisions of copyright, and exception of rights to ‘reverse engineer’ in circuit layout rights. AN ANALYTICAL FRAMEWORK FOR ASSESSING THE IP SYSTEM 63 can be so detrimental to innovation that it can undermine the additionality of the IP right in the first place: … a poorly designed intellectual property regime — one that creates excessively ‘strong’ intellectual property rights — can actually impede innovation. … Knowledge is the most important input into the production of knowledge. Intellectual property restricts this input; indeed, it works by limiting access to knowledge. (Stiglitz 2008, pp. 1694, 1710) The IP rights system also plays a role in disseminating innovations from overseas by facilitating the transfer of IP through licensing, trade and investment between countries. While the transfer of IP is also affected by other factors — including search and negotiation costs, and the willingness to pay licensing fees between different jurisdictions — rights holders may feel more comfortable bringing their inventions and creations to countries where there is greater certainty that national IP laws will afford and enforce their rights. International IP agreements help to provide a consistent set of rules for the dissemination of ideas protected by IP. However, the standards of protection defined in international agreements may not always be in the interest of all countries. Ideas that are protected by existing IP settings may already have received sufficient protection to be commercialised, yet can receive further benefits if IP rights are strengthened. Thus, stronger standards stemming from international agreements can unduly favour rights holders, while increasing the costs of IP protection for users and follow-on innovators (chapter 18). An ideal IP system would consider the scope of protection more broadly in other jurisdictions before adopting stronger domestic standards of IP protection. Efficiency: getting the balance right The way IP rights are balanced influences innovative behaviour through the returns to creators and to users of ideas protected by IP. On the one hand, it may be necessary to provide incentives to innovators to create by restricting access to those ideas and thereby raising the private return to that IP. On the other hand, the effect of these restrictions can impose a cost to the community in the form of higher prices and restricted access. The Commission has previously noted the range of factors that affect this balance: … IP protections that are either too strong or too weak can have adverse economic effects. For individual countries, the optimum design and level of IP rights also depends on the extent to which they are net importers or exporters of different forms of IP material and other considerations, such as their level of economic development and the nature of their legal system. (PC 2010, pp. 257–258) How to strike the balance is one of the key questions the Commission has been asked to examine in this inquiry. The principle of efficiency addresses this by finding the appropriate degree of IP protection that balances: • 64 the incentives for creators to produce new and additional works that IP rights and protections provide; against INTELLECTUAL PROPERTY ARRANGEMENTS • the costs to users that IP rights cause, by extending market power and restricting innovation elsewhere in the supply chain. As Tabarrok (2013) suggests, the net benefits that accrue from the IP system are in the shape of an ‘inverted-U’ as the ‘strength’ of IP arrangements increases (figure 2.4). Such a relationship has also been suggested specifically for copyrighted material (Molloy 2016). The shape of the curve reflects the greater benefits of providing an incentive to create ideas, and then the greater costs of protecting those ideas with overly strict IP arrangements. The difficulty for policymakers is determining where policy settings currently sit, as this will determine whether strengthening or weakening of IP rights increase net benefits to society. Figure 2.4 The stylised relationship between the strength of IP rights against net benefit to society Net benefit to society Point at which IP rights yield the maximum benefit As IP rights are strengthened, the benefit to the rightsholders exceeds the cost to the public, and so net benefit increases… … but if IP rights are strengthened further, the cost to the public exceeds the gains to right holders, and so net benefit declines. 'Strength' of IP rights Source: Adapted from Tabarrok (2013). Ensuring IP arrangements are efficient requires that: • ideas are generated at the lowest cost to society • ideas are traded and made use of by those that can generate the greatest value from them • the effects of granting IP rights on competition and innovation are balanced against their impacts on growth and wellbeing. The distributional impacts on creators, consumers and governments should also be considered. AN ANALYTICAL FRAMEWORK FOR ASSESSING THE IP SYSTEM 65 Efficient systems should generate ideas at the lowest cost to society An efficient IP system should only reward IP rights holders by the amount needed to induce them to generate additional ideas. Returns in excess are windfall gains to the IP rights holder, and windfall ‘losses’ to the rest of society. This is true in all economies, but has particular implications for national welfare depending on whether a country is a net importer or exporter of IP. Not transferring enough of the benefits of the idea is also problematic as insufficient rewards may thwart the innovation and result in the loss of the social benefits associated with it. A balanced IP system is one that reaches the best possible compromise between the two — where the costs of innovating and the costs to the rest of society for that innovation are as small as possible (box 2.3). As the Australian Competition and Consumer Commission (ACCC) put it: Competitive forces are optimised where the appropriate balance is struck in the IP system between creating and maintaining incentives for the creation of IP, and maintaining incentives for its efficient use. The ACCC recognises that it is difficult to precisely define this balance, however the guiding principle in assessing the extent of IP protections is that they should not extend beyond the point where the costs of protection start to exceed the benefits. That is, they should be determined within a cost–benefit framework. (sub. 35, p. 10) The efficiency of IP rights should be considered through a ‘wide lens’ of all the rights that may apply and other policies that also stimulate innovation, such as government funds for research and development. Equally, reform of the IP system should account for how substitution between IP rights and other policy instruments may occur. The distributional impacts on creators, consumers and governments should also be considered before changes are made to the IP system. IP rights may not be the most cost-effective way to secure and encourage new ideas IP rights are not the only way to secure property or to ensure a return from ideas. Alternatives (including trade secrets, complexity of design and common law agreements) exist and are frequently used by Australian businesses (figure 3.6) (Cohen and Walsh 1998; Hall et al. 2014; Levin et al. 1987). Rewards or ‘bounties’ for innovations can also provide the return necessary to create ideas. And, in some cases, competitive markets alone mean a ‘first-mover advantage’ may be sufficient to provide an incentive to innovate without further intervention. As put by López: [Intellectual Property Rights], including patents, copyright, trademarks, industrial designs, utility models and plant breeders’ rights, are some of the appropriability mechanisms that may be used by innovators. However, as is well known, there are other available mechanisms, including the exploitation of lead time, moving rapidly down the learning curve, the use of complementary manufacturing capacities and secrecy … Since labor mobility is also a form of technology imitation, labor legislation, contracts and human resource management practices are also very relevant appropriability mechanisms … There are also a number of practical and 66 INTELLECTUAL PROPERTY ARRANGEMENTS technical means of protection, such as passwords, digital signatures, copy prevention mechanisms, etc. which are used in some industries. (2009, p. 3, 21) Box 2.3 IP rights, returns and transfers: a stylised example A simpler, more stylistic way to think about the incentives facing innovators and society is to assume that intellectual property (IP) rights reallocate some of the broader public good benefits that stem from innovation back to the firm or individual that generated the idea. Suppose an individual or a firm has two choices: to generate IP or to use their resources to do something else. The net present value of each choice will determine the decision that the firm makes, but the firm may not take into account the net present value of other returns that may accrue to the community as a result of the innovation as part of its decision making. In the presence of IP rights, some of that value is transferred to the individual or firm, which makes it more attractive to undertake the IP activity. But the strictness of the IP system not only transfers the benefit to the individual or firm, it also leads to a loss based on the transaction costs of undertaking such a transfer, and in lost dissemination and use of the IP as the rights holder is granted more power. The following figure demonstrates this relationship. The individual or firm can choose the non-IP activity and receive the associated return, and depending on the extent of IP rights its choice between generating IP or not will change. At the same time, a greater extent of the IP rights could be expected to diminish the returns to the rest of society. The optimal return in the stylistic example is to make the private return just high enough for the individual or firm to innovate, but not extend IP rights beyond this to minimise the loss of benefits to the rest of community. Private return Non-IP activity Other returns to the community IP Activity with no IP rights IP Activity with weak IP rights Return to non-IP activity IP Activity with optimal IP rights IP Activity with excessive IP rights Some of these alternatives have limitations. For example, ideas protected by trade secrets can make it difficult for cumulative innovation to occur. However, these limitations should be considered against those of other IP rights in determining which is most efficient. Indeed, other systems of reward for innovations in some circumstances may be more efficient than IP rights (Shavell and van Ypersele 2001; Stiglitz 2008), and in those markets where competition spurs innovation, IP rights that hamper competition may have a detrimental effect on encouraging new ideas. An efficient IP system encourages the use of alternatives to IP rights where appropriate and should not default to a reflexive use of IP rights. The costs and benefits of creating ideas vary between the public, private and not-for-profit sectors. The creative activity in one sector can also serve as a complement or substitute to AN ANALYTICAL FRAMEWORK FOR ASSESSING THE IP SYSTEM 67 the activity of another sector. But there appears to have been little consideration as to how the IP system should be tailored to take account of the different incentives, costs and goals that each sector has, or the best way to make use of the relative merits of each. As put by Carroll: There is no generally accepted framework for assessing the tradeoffs between granting intellectual property rights, investing public funds directly in innovation through grants or prizes or indirectly through tax policy, or some combination of these to encourage desired levels of inventive and creative activity. (2011, p. 15) An IP system designed to protect IP generated in the private sector can lead to distortions in research choice in other sectors. For example, Jensen and Webster (2014) examine how academic scientists choose their research projects based on the presence of patents. An efficient IP system ensures that the interplay of innovation between the public and private sectors is not disrupted, so that the comparative advantages of each can be best exploited. The role of IP rights in publicly funded innovation is discussed in more detail in chapter 16. The costs of securing and enforcing IP rights also affect efficiency Securing an IP right can be costly, and so affects the incentives for innovators to apply for an IP right, use another form of IP protection, or to innovate in the first place. The costs of applying for an IP right can include the application cost and the costs of using any necessary IP professionals or intermediaries. The nature of the IP system affects these costs directly and indirectly. Costs of some rights are set in regulation, and choices around the complexity of the right affects the need, demand, and ultimately price of IP professionals. An economic approach to the optimal enforcement of rights is different to that of a legal approach. An economic approach considers the costs and benefits of enforcing and defending rights, especially the social costs that include the cost of a court system. An IP system that enforces IP just for those with the deepest pockets is unlikely to be efficient, equitable or in the interests of the community. That said, enforcement of rights also needs to be considered in a broader context than costs to users of the IP system. Persistent infringement of IP rights could be suggestive of a poorly functioning IP system, as it may indicate a failure to set appropriate licensing fees that reflect the balance between private and social benefits of innovation, a lack of clarity about who owns the ideas and what has been licensed, or a combination of the two. It may be preferable to examine and adjust the settings of the IP rights themselves, rather than providing for stricter enforcement of rights that could be deficient. There is also a broader question about the efficiency of the international enforcement provisions to which Australia is a signatory. This creates an extra dimension to determining what an efficient IP system looks like, as the balance between the creators and users of IP is considered simultaneously with the balance between the welfare of 68 INTELLECTUAL PROPERTY ARRANGEMENTS Australians and those overseas. The design, practice and reform of IP enforcement are discussed in chapter 19. An efficient IP system encourages the transfer of IP, trade and investment The seamless exchange and licensing of IP ensures that those who can make the best use of ideas have an opportunity to do so. An efficient IP system therefore should not impede the transfer of IP rights or trade in IP either domestically or internationally. The benefits of trade and transfer in IP rights extend beyond dissemination; it allows businesses to license products and processes for more efficient operations and to produce higher quality goods and services for consumers. How IP rights allow for such use of ideas is therefore relevant when it comes to making sure that the most efficient users of IP can gain access to it. As put by the ACCC: The potential for IP rights to address the market failure arising from the potential for ‘free riding’ rests on the assumption that transaction costs are low and the negotiating parties have roughly equal bargaining power. (sub. 35, p. 7) An efficient system needs to allow for international trade in IP, rather than create barriers to technology transfer and follow-on innovation between jurisdictions. Such barriers reduce the wellbeing of the Australian community by constraining where IP can be sourced, increasing the costs of its creation, and potentially hindering economic growth. Given that IP rights are allocated on a jurisdictional basis (system of territorial rights) international collaboration is necessary to remove barriers to the transfer of IP rights and trade in goods and services that embody IP. Australia’s approach to international IP collaboration is discussed in chapter 18. The longer-term effects of IP rights should be accounted for While a well-functioning IP system ensures those who can make best use of the IP rights can do so, it must also consider the longer-term effects that exclusivity of inventions can create. This balance has been described as how IP rights ‘generate monopoly positions that reduce current consumer welfare in return for providing adequate payoffs to innovation, which then raises future consumer welfare’ (Maskus 2000, p. 29). But when the balance is tilted too far, there are dangers that the IP system can hinder competitive outcomes in the longer term: IP rights can help to break down barriers to entry but, when applied inappropriately, can also reduce exposure to competition and erect long–lasting barriers to entry that fail to serve Australia’s interests over the longer term. This risk is especially prevalent in commitments entered into as part of international trade agreements. (Harper et al. 2015, p. 41) Where IP licensing or assignments are used to restrict or deter competition, for example by collective or crosslicensing, or other practices designed to exclude competition or leverage market power, a conflict may arise with the promotion of competition and efficiency. (ACCC, sub. 35, p. 10) AN ANALYTICAL FRAMEWORK FOR ASSESSING THE IP SYSTEM 69 IP rights are designed to recompense a creator, but not to materially distort markets over the longer term in order to do so. Understanding how competition and innovation interact over the longer term is therefore necessary in determining the settings of the IP system, as is the role that competition policy plays more broadly (chapter 15). Adaptability: making sure IP rights are apt for the future Given that IP rights can affect society considerably over a period of time, the IP system should be adaptable to change. IP arrangements and stakeholders have been, and continue to be, affected by a number of developments, including the rise of cloud computing, the Internet, digitisation, and globalisation including the increasingly specialised nature of production chains (OECD 2015a). The clear boundaries around physical goods that once made it easy to define IP protection are now becoming increasingly blurred. As the Ai Group put it: Business investment in innovation will be aided if businesses are confident that new innovations will be protected. While Australia’s IP system rates very highly relative to many other OECD countries, our members report that changes in technology often move faster than legislative change and, as such, the relevance and effectiveness of Australia’s copyright laws is being challenged. Ai Group believes it is critical that Australia’s intellectual property system accurately reflects changes in technology and the nature of assets, and that it is clear, coherent and robust. (2014, p. 46) While these developments have given rise to new challenges, they have also given rise to new opportunities for diffusion and commercialisation. New business models and research tools — such as those based on text and data analytics and open access models — have the capacity to promote inventions and creativity and provide for greater access to information and creative works. By being technology neutral — that is, not creating incentives that encourage particular types of ideas over others — an IP system better enables society to take advantage of inventions and creations. The IP system also needs to be adaptable to changing economic and market structures. For example, structural change in the Australian economy has seen changes in the composition of the size of businesses, as well as the shares of inputs and outputs from different industries. Thus the IP system has to adapt to the changing needs (and abilities) of new and different innovators who access and use the IP system. An adaptable system is one that reflects changes in the underlying costs of innovating as well. In practice, the adaptability of the system will depend on the markets where IP is used, and be highly specific to the nature of those markets and their level of competition, innovation and openness to trade. The degree of adaptability, and the form it should take, may therefore need to be tailored specifically for different IP rights (Schmidt 2011), and include a strong governance framework to help keep the ‘rights right’. There are, however, impediments to adaptability — the most prominent being the international obligations that Australia adheres to as part of bilateral and multilateral trade, 70 INTELLECTUAL PROPERTY ARRANGEMENTS investment and IP commitments. 5 The outcomes of international agreements can be difficult to reverse, as noted in the UK Gowers Review: … as a result of the international treaties to which the UK is a signatory and the responsibility of the European Commission for this area of policy, the UK can make no unilateral change to the length of copyright and is bound by strong minimum standards. (Gowers 2006, p. 39) As such, in an international context, transparent and evidence-based decisions are even more important to ensuring that Australia only agrees to policies with a clear net benefit. Australia’s IP institutional and governance arrangements are discussed in chapter 17. Accountability: a transparent, evidence-based system The institutions that bear on the IP system need to be accountable for their conduct both in administering existing policies and in developing and implementing policy changes. As Bovens outlined: Accountability is not only useful as a check, it also leads to prevention. Accountability forces administrators to trace connections between past, present and future … An administrator who is called to account is confronted with his policy failures and he is aware that, in the future, he can be called upon again, even more pitilessly, to render account. (2005, p. 26) Public policy accountability requires three key ingredients — that decisions are informed by articulated policy objectives, evidence, and that they are reached transparently. Decision making needs to be visible and should provide adequate opportunity for public consultation in order to be transparent. This is a key principle across all types of public policy, but is especially relevant in an area like IP, where the impacts of policy changes are likely to be distributed widely across the economy and society. As noted by Banks: The wider the impacts of a policy proposal, the wider the consultation should be. Not just with experts, but also with the people who are likely to be affected by the policy, whose reactions and feedback provide insights into the likely impacts and help avoid unintended consequences. Such feedback in itself constitutes a useful form of evidence. (2009, p. 14) As well as being a mechanism for eliciting evidence from stakeholders, consultation can help build consensus and promote stakeholder ‘buy in’ to the ultimate policy direction. This can help avoid a cleverly designed regime failing to reflect the interests of the broader community or meet its objectives because of a lack of user understanding. Transparency requires public articulation of policy objectives and a comparison of alternative mechanisms for achieving them. For example, as outlined above, a goal of the IP system should be to encourage the creation of ideas that would not have occurred otherwise. Once objectives like this are made clear, policies can be tested against them as 5 The Commission has had regard to the need for Australia to meet its existing international agreements and obligations, as per the Productivity Commission Act 1998 (Cth). AN ANALYTICAL FRAMEWORK FOR ASSESSING THE IP SYSTEM 71 they are being developed (ex ante) and to evaluate their success (ex post). 6 Such testing requires an evidence base. A well-targeted evidence base can also support learning and improvement of policy over time when used by policy makers and other analysts to: • assess the contribution of IP to social and economic objectives • understand the barriers to innovation and the creation of IP • test the impacts of policies in addressing problems and evaluate the effectiveness of existing policies compared with alternatives • improve the design and delivery of programs • stimulate stakeholder engagement and debate (OECD 2015c). What should be done when there is a lack of evidence? Some participants in this inquiry have highlighted the poor state of evidence and data used to justify previous IP policy decisions, while others have criticised the Commission for the quality of evidence used to support its draft recommendations. 7 This, in part, reflects differences of opinion about the role and effectiveness of the IP system. It also reflects a lack of policy-focused research and difficulties in genuinely testing the value of IP policies for rights holders and the community at large. While a nascent evidence base for some IP rights is already developing (such as IP Australia’s Intellectual Property Government Open Data project and the ABS’ Expanded Analytical Business Longitudinal Database), they are not yet sufficiently well-developed to answer some of the vexatious questions about the effectiveness of IP rights and impact of IP policy. Calls for a better, targeted evidence base — especially one of a quality commensurate with the impacts of the IP system on the community — to inform IP policymaking are reasonable. As put by Kalanje: … differences of opinion persist amongst economists and policymakers about the exact role of intellectual property (IP) in relation to innovation. On the one hand, in theory, the IP system is considered to be absolutely necessary “to encourage creative intellectual endeavor in the public interest,” and on the other, some observers believe that, in practice, the IP system hinders competition to the extent that it is often seen to be playing a negative role in innovation. Hence the need for a systematic and periodic study and review of the actual use by businesses of the tools of the IP system so that economists are able to provide empirical, evidence-based guidance to policymakers to adapt the IP system so that it continues to serve the conflicting 6 IP Australia suggested that an additional ‘strategic’ principle — one that ‘[sets] out the reason as to why and how the Australian Government goes about making changes’ — would also be desirable (IP Australia, sub. 23, p. 7). The Commission considers that the principles of adaptability and accountability would provide for such an outcome. 7 For example: the Australian Copyright Council (sub. DR543), AIPPI (sub. DR551), Clode (sub. DR215) and IPTA (sub. DR562). 72 INTELLECTUAL PROPERTY ARRANGEMENTS private and public interest in spurring further innovation and its wide diffusion in the shortest possible time. (2005, pp. 1–2) Importantly, sound public policymaking is not well served when stakeholders that benefit from current IP settings — especially those based on limited or dubious evidence — demand an excessively high burden of proof to justify changes to policy. Such demands can result in policy paralysis and a rigid IP system that cannot adapt to change. Where the Commission has proposed reforms, it has made use of available evidence, and considered not only the effect of changes to the IP system, but the potential costs of maintaining the status-quo. More broadly, evaluation may still be possible through the use of qualitative approaches, even where quantitative evidence is mixed or lacking. Such approaches include: • examining the history and factors that have contributed to the current IP settings, and determining whether previous policy decisions are consistent with the principles of the framework • returning to ‘first principles’ by considering the problem from a community-wide perspective, rather than focussing on industries or rights holders that may be directly affected • considering the experiences of other jurisdictions, with policy interpretation supported by a thorough and robust assessment of the relevant differences between those jurisdictions and Australia • identifying the evidence needed to make such a decision, including specifics on what sort of information is necessary, how it should be collected, and who should assess it. The nature of costs and benefits — and how they change with ‘greater’ or ‘lesser rights’ — should also be considered when setting IP policy when evidence is mixed. The net effects of setting IP rights ‘too leniently’ may not be the same as setting them ‘too strictly’. The balance between creators and consumers of IP-intensive goods should also be kept in mind — an extension of rights without an extension of exceptions is difficult to justify except in extreme cases where IP rights are manifestly ineffective. Despite making the best use of available evidence, there is still likely to be uncertainty around what future innovations may occur. But policymakers may find themselves obliged to ‘lock-in’ IP policy settings as part of an international agreement and without sufficient time or information to conduct a cost-benefit analysis or qualitative evaluation of what such parameters may mean for competition, innovation and overall wellbeing. Recent experience would also tend to suggest that it is easier to extend IP rights than narrow them, especially where international agreements are concerned. Given the asymmetric nature of how policy can be changed, the Commission considers it is appropriate to ‘err on the side of caution’ where there is imperfect information, and consciously set weaker parameters in the way that rights are assigned, used or enforced. Extending rights should only occur after careful consideration of how such a change might AN ANALYTICAL FRAMEWORK FOR ASSESSING THE IP SYSTEM 73 affect future innovations, whether IP rights are the best way to drive the desired outcome, and how it might affect the greater number of consumers relative to producers of IP. Bringing it all together The principles discussed above highlight the different tradeoffs to be considered when framing a well–functioning IP system. This inquiry applies an economic analytical framework as, in the Commission’s view, it is the best approach to assess and address these tradeoffs. In doing so, it considers the principles against each of the elements of the IP system — how rights are assigned, used and enforced — in recognition that optimising over only part of the system may not deliver the best outcome. Such an approach is needed to make sure that the ultimate goal of improving the wellbeing of Australians — by having a well–functioning IP system — is achieved (figure 2.5). Differences in IP rights mean that in practice the principles embodied in the Commission’s framework have to be applied holistically and objectively to each part of the IP system. Chapters 4–14 apply the framework to each of the IP rights to identify specific reforms that would benefit the Australian community. The following chapter assesses how the IP system fares against these principles in general. RECOMMENDATION 2.1 In formulating intellectual property policy, the Australian Government should be informed by a robust evidence base and be guided by the principles of: • effectiveness, which balances providing protection to encourage additional innovation (which would not have otherwise occurred) and allowing ideas to be disseminated widely • efficiency, which balances returns to innovators and to the wider community • adaptability, which balances providing policy certainty and having a system that is agile in response to change • accountability, which balances the cost of collecting and analysing policy–relevant information against the benefits of having transparent and evidence–based policy that considers community wellbeing. 74 INTELLECTUAL PROPERTY ARRANGEMENTS Figure 2.5 How the principles fulfil the objective Apply the principles Outcome Effective: Does the IP system lead to additional IP being generated? Is the IP system effective in disseminating IP? Efficient: Does the IP system fulfil the objective of ultimately improving community wellbeing? Is the IP system getting the right balance between encouraging IP creation and costs that rights can cause? Is the IP system ensuring IP is being generated at the lowest cost? If all principles are satisfied… the IP system is well placed to meet the objective Is the IP system ensuring that IP is traded so that those that can use it most efficiently can do so? Is the IP system appropriately balancing the long-term costs and benefits that stem from the system’s effects on competition and innovation? Adaptable: Does the IP system adapt as the nature of innovation, competition and broader economic conditions change? Accountable: Are the policies and changes made to the IP system evidence based, transparent, and do they reflect community values? If some or all principles are not satisfied… the IP system is unlikely to meet the objective at present or in the future AN ANALYTICAL FRAMEWORK FOR ASSESSING THE IP SYSTEM 75 3 How does the system fare? Key points • A fundamental tenet of intellectual property (IP) arrangements is that they need to ensure that creators and inventors are rewarded for their efforts. But the level and duration of that protection must not be so great so as to impede the benefits to consumers and other producers from the adoption and dissemination of new ideas. • This balance has not been achieved. Australia’s IP arrangements are not as effective as they could be. – Arrangements do not always result in socially valuable innovation or creative works. – In some cases, arrangements hamper, rather than encourage the use of creative and innovative works. This is particularly harmful for Australian firms, who tend to ‘adopt and adapt’ innovations, building on the knowledge of others. • Australia’s IP arrangements fail to strike an efficient balance between incentives for creators and costs to users. – IP policies do not account for overlapping rights. Protection can be sought under more than one right and firms are supported by a number of innovation policies. – For some rights, the length of protection is too long, often years or even decades longer than the commercial life of products. In the case of pharmaceuticals, extended patent terms impose significant costs on taxpayers and consumers with no evidence of an offsetting increase in investment or innovation. – Australians pay more than their overseas counterparts due to IP based market segmentation, including through parallel import restrictions and geoblocking. • Multilateral and bilateral trade agreements substantially constrain the adaptability of Australia’s IP arrangements. • Transparent and evidence-based policy helps ensure the public can hold the Australian Government and its agencies to account for policy decisions and use of public resources. – Australia has various checks and balances to ensure consistent application of IP regulations. – But there are clear areas for improvement. Responsibility for IP policy and administration is fragmented, and in some cases, policy development has not been afforded sufficient priority. Processes for including IP provisions in trade agreements, in particular, have suffered from a lack of transparency and a weak evidence base. • Poorly balanced IP settings harm all countries, but in Australia’s case, it is particularly costly. Australia is a net importer of IP intensive goods and services — the costs associated with any deficiencies in IP arrangements are borne by Australian consumers, largely to the benefit of overseas rights holders. HOW DOES THE SYSTEM FARE? 77 A dilemma lies at the heart of good intellectual property (IP) policy arrangements. On the one hand, a fundamental tenet of IP arrangements is the need to ensure that creators and inventors are rewarded for their inventiveness so that they have the motivation and incentive to create. On the other hand, the level and duration of protection must not be so great as to impede the benefits to consumers and other producers from the adoption and dissemination of new ideas. The previous chapter set out an analytical framework for assessing the IP system. This chapter employs that framework in order to assess how the current system is performing, and whether it strikes an appropriate balance. Its focus is consciously on the problems or limitations with the current system. Each of the principles underpinning the framework — effectiveness, efficiency, adaptability and accountability — are considered in turn (sections 3.1 through 3.4, respectively). The chapter concludes by considering the deficiencies with current IP arrangements in the context of Australia overwhelmingly being a net importer of IP-intensive goods and services (section 3.5). 3.1 Is the IP system effective? Ideally, Australia’s IP system should promote the creation and dissemination of genuinely new and valuable innovation and creative works, which in the absence of such a system would not have occurred. However, there are questions about the extent to which current arrangements are delivering these outcomes. IP arrangements do not always result in additional innovation … The main premise of IP arrangements is to ensure that creators of new and valuable knowledge are able to appropriate sufficient returns to motivate their initial investment (chapter 2). Rights holders, IP professionals and some research organisations promote the IP system as an effective tool for supporting innovation and creativity. Typical of these submissions, David Webber stated: It is clear private entities would never invest in developing IP at the level required to enhance the well being of Australians without the limited exclusivity afforded by IP rights. Reduction in protection gives rise to a corresponding reduction in investment. Reduced investment by private actors would need to be replaced by Government investment to achieve the same outcomes. (sub. 40, p. 1) Another rationale put forward by some participants is that the IP system supports trade and foreign direct investment, by encouraging overseas-based innovators to bring their innovations to Australia (DFAT, sub. 65; IPTA, sub. DR562) However, in many ways, these are examples of ‘retrofitted’ rationales. IP rights evolved for a number of other reasons, including as a way of raising revenue and censoring materials (box 3.1). 78 INTELLECTUAL PROPERTY ARRANGEMENTS Box 3.1 Genesis of some IP rights th Patent law in England began in the 14 century to attract skilled craftsmen from abroad, where monopoly rights were exchanged for technology-related products. Among such foreigners was John Kemp, the Fleming who gained letters of protection in 1331 for weaving, fulling and th dyeing. England later adopted the practice of patents for inventions in the 16 century, as a way to raise revenue to finance the military and elite. Copyright law was first introduced by churches and the state, in Europe, following the th introduction of printing presses in the mid–15 century, as a way to prevent the dissemination of ideas. This gave Stationers’ Company (a publisher) monopolistic rights over the publishing industry, and the ability to screen all works and limit the spread of any protestant reform movements at the time. Once a book was registered by a member of the guild of printers and booksellers, the right to print that book was perpetual and exclusive. In 1710, the Statute of Anne attempted to lift the monopoly by vesting authors, not the guild, with original copyright and limiting copyright to 14 years (with the possibility of a one-time renewal for an additional 14 years). Trade mark law originates from the middle ages, where marks were used to indicate the maker of specific products. By the 19th century these marks were then considered to be a form of property, and so trade mark laws were introduced to allow action to be taken at court against infringements. Source: Dourado and Tabarrok (2014). Looking beyond their genesis and focusing on their contemporary purpose, some IP rights — such as trade marks and geographical indications — have objectives unrelated to spurring innovation and creativity, and relate more to the provision of information and the protection of brands (chapter 12). Even where IP arrangements target innovation and creative works, it has been argued that they do so only loosely. As Tabarrok noted: … the influence of the patent system on innovation should not be exaggerated. The vast majority of innovations in most industries would occur without the existence of patents … It takes time for new ideas to diffuse and being the first to market, learning by doing, capturing market share when consumers face switching costs, secrecy, and other factors are in practice more important sources of competitive advantage than patents for most firms most of the time. … we should not be overly concerned that a weakening of patents will result in underinvestment. (2002, p. 16) Indeed, international studies reveal that, in all but a few industries, patents are seldom the most important means for appropriating returns to innovations. Lead-time and superior sales and service have been generally nominated as the most important appropriation mechanisms for product innovations (table 3.1). In Australia, a survey of managers of large firms between 2001–2006 found patents were, on average, considered the least effective appropriation mechanism for both product and process innovations (Jensen and Webster 2009). HOW DOES THE SYSTEM FARE? 79 Table 3.1 Appropriating the returns to product innovationsa Relative importance by means Survey (year) Country Ranking of mean importance 1 2 3 4 Yale (1982) United States sales/service lead time patents secrecy Harabi (1988) Switzerland sales/service lead time secrecy patents Dutch CIS (1992) Netherlands lead time retaining employees secrecy patents Carnegie-Mellon (1993) United States lead time secrecy/ sales/service complementary manufacturing Japan C-M Japan lead time patents sales/service secrecy complementary manufacturing SESSI/INSEE EFA (1993) France lead time patents secrecy complexity StatCan Innovation (1999) Canada confidentiality agreement trademarks patents secrecy CIS 3 2000 (2000) EU12 lead time secrecy trademarks complexity Melbourne Institute (2001–2006) Australia know-how brand name lead-time secrecy Gonzalez-Alvarez and Nieto-Antolin (2007) lead time complexity secrecy patents Spain patents a There are differences in the wording of questions across surveys. For example, in some surveys the question is phrased as what share of innovations are protected by the various appropriation mechanisms. Sources: Hall (2009); Levin et al. (1987); Cohen et al. (2000); Jensen and Webster (2009). The importance of patents in appropriating innovations varies across industries. Collectively, the evidence suggests that patents are only important for products that are relatively easy to imitate and entail large sunk research and development (R&D) costs. Surveys have found that patents are more important in the pharmaceutical industry, followed by specialised machinery and instruments and other chemicals (table 3.2). These findings support earlier survey-based research that suggests innovation in the pharmaceutical and chemical industries is more reliant on patent protection than innovation in other industries (Mansfield 1986; Mansfield, Schwartz and Wagner 1981). While in the services sector patents appear less important overall, there is evidence that the business service, telecommunications and media service industries are more reliant on patents than others (Baldwin et al. 1998; Blind et al. 2003; Hipp and Herstatt 2006). 80 INTELLECTUAL PROPERTY ARRANGEMENTS Table 3.2 Appropriating the returns to product innovations Relative importance by industry Survey (year) Country Industry preferences for patents in descending order Yale (1982) US pharmaceuticals, plastics, chemicals, steel, oil Harabi (1988) Switzerland research labs, machinery, chemicals, watches, paper Dutch CIS (1992) Netherlands pharmaceuticals, chemicals, instruments, rubber and plastics, oil Carnegie-Mellon (1993) US pharmaceuticals, medical instruments, special machinery, computers, chemicals SESSI/INSEE EFA (1993) France pharmaceuticals, instruments, transport, chemicals, machinery, paper StatCan Innovation (1999) Canada machinery, electronics, pharmaceuticals, communications, instruments, chemicals, motor vehicles CIS 3 2000 (2000)a EU12 transport, instruments, chemicals a Pharmaceuticals and chemicals are combined. Source: Hall (2009). … or creative works Concerns about poor targeting and a lack of additionality apply equally in copyright. Many types of works are protected by copyright, even though the case for protection is weak. Protection extends to non–commercial works that would be created in the absence of copyright protection. There are a number of examples of works that would have been created irrespective of the degree of copyright protection, including photos and videos of children taken by parents, blog posts, fan fiction, and letters written by a law firm to a client. As noted by Australian Digital Alliance (ADA): … copyright applies automatically to all creative works, without the need for registration or other administrative steps, and with very low originality and effort requirements. This is the case even where these works are not intended for commercial or even public use: for example, a doodle or text message receives the same protection as an oil painting. (sub. 108, p. 6) Grey literature is another area where copyright applies but is not needed to encourage creation. Grey literature encompasses research literature created and disseminated directly by organisations including academic centres and government departments outside of the commercial or scholarly publishing system. 1 As Australian Policy Online note, most creators of grey literature are not motivated by commercial considerations and in many cases works are paid for through public funds: Financial gain is not an important consideration for most organisations surveyed, even for those in the commercial sector. It is probable that most of the material produced by government, NGOs and education organisations is paid for through public funds. (Lawrence et al. 2014, p. 3) 1 It includes reports, discussion papers, briefings, case studies, literature reviews, fact sheets, evaluations, submissions, working papers, conference papers, data, technical reports and specifications, policies, strategic plans, and infographics. HOW DOES THE SYSTEM FARE? 81 Shavell (2010) and Posner (2012) similarly question the role that copyright plays in the academic domain. They note that the primary spur for academic writing is not profit but rather scholarly esteem and professional advancement. Not all innovation is socially valuable Even where IP arrangements provide an incentive, it is not clear that all additional innovation and creative works are socially valuable. In respect of patents, empirical studies find a skewed distribution of patent values, with value disproportionately concentrated within a small proportion of patents (Dahlin and Behrens 2005; Hall, Jaffe and Trajtenberg 2005; Schankerman and Pakes 1986). These studies are consistent with the Commission’s own analysis. While the patent system may play an important role in promoting some socially valuable inventions, many patented inventions are of little social value. As outlined in appendix D, the Commission constructed a number of proxies for a patent’s social value. While no single measure provides definitive evidence, as a collective they suggest that a significant number of patents in Australia are of low value (figure 3.1). Figure 3.1 The bulk of Australian patents are of relatively low value Distribution of composite patent value indexa a The index accounts for the following proxies for patent value: forward citations, a ‘generality’ index, a ‘radicalness’ index, citations to non-patent literature and patent family size. The higher the value the greater the social value of the patent. These measures are defined and reported separately to the composite index in appendix D. Source: Commission estimates based on IP Government Open Data (IPGOD) (2016 edition). 82 INTELLECTUAL PROPERTY ARRANGEMENTS Australia is not alone in granting patents of questionable quality. The Commission also assessed patent quality for the United States and Europe, and found that there is also a large share of low quality patents in these jurisdictions (chapter 7). This accords with the findings of other researchers who find evidence that the quality of patents in developed countries is low, and in some cases getting lower (Bakels and Hugenholtz 2002; Graham and Harhoff 2014; Hargreaves 2011; OECD 2015a, 2015a; van Pottelsberghe and van Zeebroeck 2008). The OECD (2011b) for example, has estimated patent quality indicators across countries and over time, and found evidence that average patent quality has been steadily declining. Problems around low social value are most pronounced in the innovation patent system (IPS) — a second tier patent system that targets ‘incremental innovations’. Reflecting this, the ‘innovative step’ required to receive an innovation patent is lower than the inventive step for standard patents. The courts have found that even obvious inventions can qualify for an innovation patent. The Commission has drawn on the results from a survey of users of the IPS to estimate the distribution of private value of innovation patents. The data suggest that a large proportion of innovation patents are of relatively low value — 40 per cent of innovation patents together account for around 3.6 per cent of the total reported value from the survey (figure 3.2). HOW DOES THE SYSTEM FARE? 83 Figure 3.2 Estimated distribution of private innovation patent valuea Per cent of total reported value 100 80 60 40 20 0 0 20 40 60 80 100 Per cent of survey respondents a Reported value was calculated by multiplying the number of respondents in the range by the midpoint of the range. For the ‘more than $1m’ range, the number of respondents was multiplied by $1 000 001. The cumulative reported value percentage was then plotted against the cumulative fraction of survey respondents for each value range with a straight line used to interpolate between points. This straight line interpolation implicitly assumes that the distribution of respondents within a value range is positively skewed with a mean equal to the midpoint of the value range. If the assumption that patents in the more than $1 million range are worth $1 000 001 was relaxed to account for patents with a higher value the distribution curve would shift inwards (except at the end-points) — in other words, a given fraction of survey respondents would account for a lower percentage of total reported value (so 40 per cent of innovation patents would account for less than 3.6 per cent of total value). Source: Commission estimates based on Verve Economics survey responses (2013). In some cases IP arrangements make it harder for innovators … As Australian firms tend to ‘adopt and adapt’ innovations (figure 3.3), arrangements for disseminating and building on the knowledge of others is key. 84 INTELLECTUAL PROPERTY ARRANGEMENTS Figure 3.3 Novelty of introduced innovations in Australia New or significantly improved goods or services or operational processes 100 Goods or services Operational processes 90 80 Per cent 70 60 50 40 30 20 10 0 New to the world New to Australia New to the industry within Australia only New to the business only Source: ABS (Innovation in Australian Business, 2014-15, Cat. no. 8158.0). However, IP rights can reduce the flow of benefits from new ideas and processes. Indeed, overly strong restrictions on diffusion can be so detrimental to innovation that it can ‘undo’ the benefits of the IP system in the first place: … a poorly designed intellectual property regime — one that creates excessively ‘strong’ intellectual property rights — can actually impede innovation. … Knowledge is the most important input into the production of knowledge. Intellectual property restricts this input; indeed, it works by limiting access to knowledge. (Stiglitz 2008, pp. 1696, 1698) Another way in which IP rights can impede innovation is through strategic anticompetitive behaviour. European studies reveal that low-value patents are deliberately used as a strategic tool for stalling or excluding market entry (EC 2012). Closer to home, local patent attorneys openly advertise strategies for raising the entry costs of competitors: By filing a series of innovation patents surrounding a product of a competitor, it is relatively easy and cost-effective to form a patent thicket around the product which subsequently makes it increasingly difficult and costly for the competitor to maintain freedom-to-operate. A variant on [the above strategy] is the filing of multiple divisional innovation patents claiming priority from a standard patent application, each for a minor variant of the invention. This makes opposing the invention a difficult and costly process as the opponent must then challenge each of the innovation patents, as well as the standard patent. Such a patent thicket is also very useful for an application as it creates a great deal of uncertainty for potential competitors as to whether they are likely to infringe one or more of the patents if they decide to enter the sector of the market to which the patents relate. (Baxter IP 2016) HOW DOES THE SYSTEM FARE? 85 In some areas of technology in Australia, patent thickets have grown dense. For example, in the area of mobile devices and networking, a dense thicket has developed within and around a set of patents held by firms including Sony, Ericsson and Samsung (figure 3.4; chapter 7; appendix D). Figure 3.4 Schema of an Australian patent thicketa a The firms on either side of an interconnecting line cite each other’s patents — that is, each firm pair represents a bilateral patent relationship. The thicket is initially identified by the interrelationships between firms that are part of ‘triples’ — three firms that each hold patents that cite patents held by the other two firms (appendix D). The dark green circles denote firms that are either part of a triple relationship or a broader relationship that involves more than three firms. The light green circles denote bilateral patent relationships. The red interconnecting line indicates that the bilateral patent relationship includes at least one innovation patent. Source: Commission estimates using IPGOD (2016 edition) and unpublished IP Australia citations data. Low–value patents can also impede innovation by contributing to ‘noise’ in the system. With more patents, it is harder for a follow-on innovator to identify and build on true advances in human knowledge and be sure that it is not infringing a patent. … and also harder for users Australian users of the copyright system have embraced digital distribution of creative content and have access to a vast amount of works that they may have had more difficulty finding in the past. However, in some cases, IP arrangements are hampering rather than encouraging the use of creative works. In the case of copyright, long periods of protection coupled with a lack of registration results in knowledge of who owns what rights being lost. This causes works to be ‘orphaned’, making it difficult or costly to obtain permission to use the work. Surveys 86 INTELLECTUAL PROPERTY ARRANGEMENTS reveal that the collections of Australia’s National and State Libraries contain significant amounts of unpublished orphan works — depending on the collection, orphan works make up anywhere between 10 and 70 per cent of their holdings (ADA, sub. 108). The National Film and Sound Archive (2010) estimated that around 20 per cent of its national audiovisual collection is orphaned or abandoned. The time required to diligently seek out copyright owners and licence those works is prohibitive, which prevents efficient access to those works. In the case of unpublished works, which remain in copyright in perpetuity, copyright owners must be identified and located potentially hundreds of years after the work was created. In a survey conducted in 2015, 14 universities (covering 20 collections), reported that their collections included over 12.9 km of unpublished works, or approximately 103 904 000 pages (ADA, sub. 108). While trade marks and geographical indications are intended to provide consumers with better information, they can be more harmful than helpful in some cases. There is evidence that trade marks are increasing consumer confusion (Greenhalgh and Webster 2015). 3.2 Is the IP system efficient? In addition to targeting additional and socially valuable innovations, Australia’s IP system would ideally balance the incentives for creators and the costs to users (arising from higher prices and restricted market access). However, as discussed below, the evidence suggests that IP arrangements fail to strike this balance and as a result, greater costs are being born by Australian society than is necessary. As the Australian Competition and Consumer Commission (ACCC) put it: Competitive forces are optimised where the appropriate balance is struck in the IP system between creating and maintaining incentives for the creation of IP, and maintaining incentives for its efficient use. The ACCC recognises that it is difficult to precisely define this balance, however the guiding principle in assessing the extent of IP protections is that they should not extend beyond the point where the costs of protection start to exceed the benefits. That is, they should be determined within a cost-benefit framework. This aligns with the PC’s proposed principles in ensuring that the extent of IP protections is both effective and efficient. … The ACCC is concerned that, in the granting of IP rights, Australia’s current IP system may not be striking the right balance between the extent of property rights and the efficient use of IP. The ACCC is concerned that the extent of current IP protections may, in some instances, go beyond what is needed to resolve the ‘free rider’ problem and incentivise innovation. (sub. 35, p. 10) HOW DOES THE SYSTEM FARE? 87 IP policies fail to account for overlapping rights Each of the IP rights — be they copyright, patents, trademarks or circuit layout rights — are intended to promote innovation and creative works. However, in practice, some innovations are covered by multiple rights (box 3.2). And in addition to relying on IP rights, parties can also rely on market-based protections (figure 3.5) as well as other policies. For example, a significant number of companies that access the R&D Tax Incentive also use IP rights, such as patents and copyright, to protect their R&D outcomes (Department of Industry, Innovation and Science, pers. comm., 13 April 2016). Box 3.2 Multiple rights or multiple rents? Australia’s IP rights have been progressively expanded to cover both new subject matter, and to grant new exclusive rights. As the products demanded by a modern economy have grown increasingly complex, businesses are increasingly deploying multiple IP protections over the same good as a source of competitive advantage (Beckerman-Rodau 2010). The application of multiple IP rights raises questions for the efficiency and effectiveness of the system. Using multiple rights can be appropriate if the ability to earn a return from one kind of right is ‘unlocked’ through the use of another. For example, the use of a trade mark in conjunction with a patent might ensure the benefits of the patented invention are not ‘crowded out’ in the market through consumer confusion. In 2015, the European Commission and the Organisation for Economic Development (OECD) reported that the top global corporate R&D investors generally used patents and trade marks in a complementary fashion rather than as substitutes (Dernis et al. 2015). But unless Australia’s IP arrangements explicitly take account of (and adjust for) the propensity to use multiple rights, the risk of providing excessive protection to rights holders is significant. In some cases, Australia’s IP arrangements do take account of the overlap between rights and the law attempts to ensure only one right applies in a particular situation. Examples include the overlap between copyright and designs for artistic works, the application of trade marks to copyright- and patent-protected goods, and the interplay between trade marks and the names of plant varieties. However, many potential overlaps remain (Chandler and Golder 2012). The shape of a product can be protected with both a registered design and a ‘shape’ trade mark. Software is automatically protected by copyright but can also be protected by a patent. In infringement cases, trade mark owners commonly allege both trade mark infringement and a breach for the tort of ‘passing off’. And prior to an amendment to the Copyright Act 1968, pharmaceuticals were protected through patents, exclusivity over the test data proving drug efficacy and copyright over the product information safety document. 88 INTELLECTUAL PROPERTY ARRANGEMENTS Figure 3.5 30 IP rights are used alongside other mechanismsa Innovation-active businesses Non innovation-active businesses per cent of businesses 25 20 15 10 5 0 Patents Secrecy or Complexity of At least one Registration of Copyright or design trade mark confidentiality product method used arrangements design to protect IP a Businesses can nominate more than one type of protection. Source: ABS (Selected Characteristics of Australian Business, 2014-15, Cat. no. 8167.0). There are concerns that IP is protected for too long … The duration of copyright protection is a striking example of inefficient levels of protection. As the Law Council of Australia noted: The term afforded copyright (the life of the author plus 70 years) is on any rational basis too long in terms of providing an incentive for the creation, development or marketing of works. (sub. 64, Part A, p. 8) Those who stand to gain from longer copyright protection agree that recent extensions to term have done little, if anything, to incentivise additional creative works: We have never asserted that there was any enormous benefit to Australian writers and we have certainly never asserted that the additional term would incentivise writers to create more works. I, personally, don’t believe for one minute that the extension of term is an incentive to create a new work. (APRA and AMCOSS, trans., p. 317) Even within countries that are significant producers of copyright material, there is scepticism about the benefits of extensions of term in copyright: Comparing the main economic benefits and costs of the CTEA [The Copyright Term Extension Act of 1998], it is difficult to understand term extension for both existing and new works as an efficiency-enhancing measure. Term extension in existing works provides no additional incentive to create new works and imposes several kinds of additional costs. Term extension for new works induces new costs and benefits that are too small in present-value terms to have HOW DOES THE SYSTEM FARE? 89 much economic effect. As a policy to promote consumer welfare, the CTEA fares even worse, given the large transfer of resources from consumers to copyright holders. (Akerlof et al. 2002, p. 15) (United States) Copyright term is a global issue and any discussion of U.S. term therefore should acknowledge international norms. Nonetheless, the current term — in most circumstances, the life of the author plus seventy years — is long, and the length has consequences. … The question now is how to make the long term more functional. … Perhaps the next great copyright act could take a new approach to term, not for the purpose of amending it downward, but for the purpose of injecting some balance into the equation. (Pallante 2013, pp. 336–337) In conclusion, the Review finds the arguments in favour of term extension unconvincing. The evidence suggests that extending the term of protection for sound recordings or performers’ rights prospectively would not increase the incentives to invest, would not increase the number of works created or made available, and would negatively impact upon consumers and industry. Furthermore, by increasing the period of protection, future creators would have to wait an additional length of time to build upon past works to create new products and those wishing to revive protected but forgotten material would be unable to do so for a longer period of time. (Gowers 2006, p. 56) (United Kingdom) Some participants argued that if a work has a short commercial life, the length of copyright is immaterial: If it is the case that a copyright has a limited commercial applicability, then it doesn’t matter if it’s tied up in perpetuity … because nobody’s going to want to use it after the first five years. (Australian Copyright Council, trans., p. 785) But this argument equates the commercial value to the rights holder, or their ability to extract that value, with the economic value a work may have to the community. Even in cases where a work may no longer have a commercial market, having it locked away behind an IP right is still detrimental to the community. Many works are unavailable for long periods due to copyright protection and only become available to the community once copyright term ends and they enter the public domain (chapter 4). While concerns about excessive term are most pronounced in respect of copyright they are not restricted to this IP right. There is also evidence that investment in some patented inventions requires less than 20 years protection. Many product life-cycles are shorter than 20 years (Bilir 2014). Drawing on the results from a theoretical model and simulation analysis, Cornelli and Schankerman (1999) concluded that the range of optimal patent duration is between 8–15 years. Analysis of patent administrative data shows that only around 15 per cent of standard patents reach their full term (figure 3.6). These data also show that the duration of protection varies across technologies — in the biotechnology, medical technology and pharmaceutical sectors more than 20 per cent of patents last the full term, while in the electrical machinery, macromolecular chemistry and transport sectors the equivalent figure is around 10 per cent or less (chapter 7). 90 INTELLECTUAL PROPERTY ARRANGEMENTS In the case of pharmaceuticals, an additional 5 years of protection is available under extension of term (EOT) provisions. These arrangements delay the entry of generics into the market, resulting in higher prices for consumers and government (through the Pharmaceutical Benefits Scheme (PBS)). The Commission estimates that the annual cost to the Australian Government (and ultimately taxpayers) of EOTs is in the order of $260 million per year. Figure 3.6 Share of Australian standard patents by patent lengtha a Standard patents granted between 1980–1995. Most standard patents have a maximum term of 20 years, so 1995 was used as a cut–off point to avoid truncation. The small number of patents that lasted longer than 20 years (due to receiving a pharmaceutical extension) are not included (chapter 10). Source: IPGOD (2016 edition). … and that users are paying too much Australian consumers pay higher prices than in overseas markets (colloquially known as the ‘Australia tax’) due to business models that rely on market segmentation supported by IP arrangements. Parallel import restrictions and geoblocks enable these business practices. Although hampered by a lack of comprehensive data, survey analysis shows systemic price discrimination against Australian consumers across a range of copyright-protected categories. • Professional software: A comparison of more than 150 products showed an average price difference of 50 per cent, with a median price difference of 49 per cent. • Music: Across more than 70 products, Australian prices were, on average, 52 per cent more expensive, while the median difference was 67 per cent. HOW DOES THE SYSTEM FARE? 91 • Games: The average price difference across more than 70 games was 84 per cent, while the median difference was 61 per cent. • E-books: Price comparisons of more than 120 e-books books sold in Australia and the United States (US) revealed average price differences of 16 per cent, while the median difference was 13 per cent (House of Representatives Standing Committee on Infrastructure and Communications 2013, p. 18). The Commission’s report on Restrictions on the Parallel Importation of Books similarly concluded that Australians were paying more for books than their overseas counterparts. The Commission examined book prices over a two year period, matching over 900 titles across Australia, the United States and the United Kingdom. While price comparisons differed across titles and were influenced by the exchange rate, it concluded that, but for the parallel import restrictions, Australian booksellers could have shipped many titles to Australia for significantly less than prices currently charged by Australian publishers (PC 2009, p. XVIII). As part of this inquiry, the Commission updated the 2009 work. A ‘like-with-like’ comparison of book titles sold in Australia and other markets was made, which resulted in over 1000 matches with the United Kingdom and nearly 500 title matches with the United States. The results paint a different picture depending on whether comparisons between countries are made using average sales price (ASP) data or recommended retail prices (RRPs). The results suggest that RRPs are similar in the Australian and UK and US markets. However, when using average sales price data (reflecting the actual price paid by consumers), the average Australian price exceeded that of the UK by about 20 per cent (there were no ASP data for the US). Under reasonable assumptions regarding discounting and freight costs, the Commission estimates the benefits to Australians from repealing the restrictions could be around $25 million per year. 3.3 Is the IP system adaptable? Given that Australia’s IP arrangements can affect society over long periods of time, ideally they would be adaptable to changing conditions. However, technological and legal lock-in are substantial obstacles to ensuring that IP rights are apt for the future. Keeping pace with technological change Technological changes have drastically modified the ways consumers access products and services, and in some cases, the nature of innovation itself. IP arrangements and stakeholders have been, and continue to be, affected by a number of developments, including the rise of cloud computing, the growth of the Internet, digitisation and globalisation (OECD 2015a). The clear boundaries around physical goods that once made it easy to define IP protection are now becoming increasingly blurred. 92 INTELLECTUAL PROPERTY ARRANGEMENTS Technological developments have given rise to new challenges, such as the ease of piracy, but have also given rise to new opportunities for diffusion and commercialisation. New business models and research tools, such as those based on text and data mining, and open access, have the capacity to promote inventions and creativity, and provide for greater access to information and creative works. Australia’s IP system has not always clearly accommodated these new business models and research tools, arguably creating a less conducive environment for innovation. As Google remarked: Australia’s copyright system arguably prohibits critical technologies and innovative activities from being conducted in Australia, such as: • basic Internet functions such as system level caching to provide a search engine; • cloud computing; • creative and transformative works, such as mashups; • medical and scientific research, such as text and data mining; and • various common consumer uses of copyright materials. (sub. 102, p. 2) The Australian Chamber of Commerce and Industry also expressed concerns about foregone opportunities based on big data: The Australian Chamber is concerned that changing technology means that existing fair dealing exemptions may not be enough to prevent copyright laws from restricting legitimate business activity. For example, big data analysis is a major emerging field, but it is subject to legal uncertainty in relation to copyright. This legal uncertainty means Australian businesses may be reluctant to use big data solutions, and Australian innovators may be reluctant to develop products based on data mining. (sub. 70, p. 8). Universities Australia suggested that in addition to missed opportunities in areas such as data and text mining, the lack of adaptability of IP laws also limited opportunities in new areas of services delivery, such as Massive Open Online Courses: Yet Australia’s inflexible and unbalanced copyright laws are blocking Australian universities from making full use of cutting edge digital technologies such as data and text mining in research. They limit the ways in which Australian universities can deliver innovative content via Massive Open Online Courses (MOOCs). They prevent Australian academics from fully engaging with colleagues and the broader community. They stand in the way of Australian universities collaborating with business and industry. (sub. DR453, p. 1) Technological developments can also affect the cost of innovating. In terms of inputs such as R&D, the changes are not uniform (box 3.3). But tailoring IP is constrained by a ‘one size fits all’ approach, restricting the ability to apply flexible approaches based on costs. In respect of the patent system, Tabarrok (2002, p. 16) remarked, ‘it would be remarkable if 20 years were the optimal duration regardless of the size of sunk costs.’ HOW DOES THE SYSTEM FARE? 93 What are the costs of innovating? Box 3.3 There is imperfect data on the cost of creating new ideas, and comparisons between different fields of innovation can be difficult. Data collected by the Australian Bureau of Statistics (ABS) on expenditures on R&D indicate that both the capital (such as land and machinery) and current costs (such as labour and intermediate inputs) of R&D have increased in nominal terms over the last couple of decades (figure (i)). Data on the prices of investing in capital goods, including R&D, artistic originals and computer software — the ‘intellectual property products’ defined in the national accounts — indicate that, for R&D and artistic originals, the price of investing in such innovation has grown faster than that of other investments in general. However, the cost of investing in computer software appears to have declined (figure (ii)). While it may not be appropriate to generalise from R&D to all inventions and creations that may be covered by an IP right, what data is available tends to suggest that, net of computer software, the costs of innovation are increasing. Expenditure on R&Da (i) 20 18 16 (ii) 3.2 Current expenditure Annual growth in price (%) 3.2 1.5 $ million 14 12 10 8 6 4 2 0 1992-93 Capital expenditure 2013-14 R&D -4.5 Artistic All capital Computer Originals Software a Data for expenditure on R&D is not available for 2012-13. The markers denote expenditure for 2013-14. Sources: OECD (2016c); Commission estimates based on ABS (Research and Experimental Development, 2013-14, Cat. No. 8104.0; Australian System of National Accounts, 2014-15, Cat. No. 5204.0). Binding international rules limit the adaptability of IP arrangements Many aspects of Australia’s IP arrangements are embodied in international agreements that set out minimum IP protections and contain obligations relating to key policy levers such as the duration and scope of protection (box 3.4; appendix B). These obligations limit Australia’s capacity to tailor rights to suit local circumstances and to accommodate change. Weatherall, Alexander and Handler commented on the implications of international obligations for system flexibility: New international rules have also closed off various sources of flexibility Australia would otherwise have had to reform domestic IP law, and as a result have created real barriers to 94 INTELLECTUAL PROPERTY ARRANGEMENTS reform of Australian IP law in ways that would make domestic law more effective, efficient, and adaptable. (sub. 99, p. 11) Box 3.4 Key International Obligations The Agreement on Trade Related Aspects of Intellectual Property Rights (TRIPS) establishes minimum standards of protection, which in some cases have been raised in other agreements. Copyright • Copyright in works must last 70 years from the date of first publication (or 70 years following the death of the author), 50 years for radio and television broadcasts and 25 years for published editions of a work. • The duration was extended from 50 years in TRIPS to 70 years in the Australia United States Free Trade Agreement (AUSFTA). • A web of treaties governs who and what is protected by copyright. • The use of formalities (such as a requirement to register copyright) is not allowed as a condition of protection. Patents • Minimum patent term is set at 20 years. Australia agreed in the AUSFTA to ‘adjust’ the term of pharmaceutical patents beyond 20 years in certain circumstances. • Patents must be available for any inventions, whether products or processes, in all fields of technology, provided that they are new, involve an inventive step and are useful. Industrial designs • The minimum term of protection for industrial designs is 10 years. The Hague Agreement (of which Australia is not a member) extends this to 15 years. Trade marks and geographical indications (GIs) • Countries must provide for the registration of signs capable of distinguishing the source of goods or services. Additional protection is required for well-known marks. • Countries must provide legal means to prevent use of GIs which mislead the public or which constitutes an act of unfair competition. • For GIs identifying wines and spirits, a higher level of protection applies. Circuit layout rights • Integrated circuit layout designs must be protected for at least 10 years. Undisclosed information • Countries must have a legal system for protecting trade secrets from unfair disclosure. • Undisclosed test data submitted for regulatory approval of pharmaceuticals or agricultural chemical products must be protected against unfair commercial use. The AUSFTA extended this obligation to a requirement to provide data protection for at least five years. Enforcement • Countries must provide effective enforcement of IP, including civil enforcement measures. For copyright piracy and trade mark counterfeiting, border measures and criminal enforcement measures are required. HOW DOES THE SYSTEM FARE? 95 While these international commitments may see Australia fall short of achieving a balance across all aspects of IP arrangements, there is still much that can be done to progress IP reform. Some adaptability can be achieved by adjusting the policy settings for one or more of the components that make an IP right (such as duration, scope of rights, or exceptions and limitations). Alternatively, changes can be made to other laws that affect the ways that the IP rights themselves are used and enforced. For example: Exceptions to IP are of fundamental importance to achieving the policy objectives that justify the grant of IP rights. To the extent to which an IP right is considered to be ‘too strong’, the way to ‘weaken’ (or, to better ‘balance’) is through the use of an exception to it. (Christie 2011, p. 121) 3.4 Are IP arrangements accountable? Transparent and evidence-based policy helps ensure the public can hold the Australian Government and its agencies to account for policy decisions and use of public resources. In many respects, Australia has relatively open and transparent processes for IP policy development and various checks and balances exist to ensure consistent application of IP regulations (including external scrutiny by independent review bodies and the courts). Further, there is evidence that the Australian Government and IP Australia have been responsive to concerns raised by stakeholders in the past. For example, IP Australia has strengthened its capacity to provide evidence-based policy advice and introduced a new quality assurance system for granting IP rights. However, there are clear areas for improvement. Responsibility for IP policy design and administration is shared across a number of portfolio departments and agencies, frustrating whole-of-government perspectives on IP (figure 3.7). The resources dedicated to policy development outside of IP Australia are also quite limited, particularly in the Department of Industry, Innovation and Science which only has a small team working on IP policy. 96 INTELLECTUAL PROPERTY ARRANGEMENTS Figure 3.7 Main public institutions responsible for IP rights policy and administrationa Minister for Agriculture and Water Resources Minister for Industry, Innovation and Science Minister for Communications Minister for the Arts Dept. of Industry, Innovation and Science Dept. of Communications and the Arts Dept. of Agriculture and Water Resources IP Australia Copyright Copyright Act 1968 Circuit layouts Circuit Layouts Act 1989 Patents Patents Act 1990 Designs Designs Act 2003 Wine Australia Trade marks Trade Marks Act 1995 Policy advice and input Registered IP right IP rights allocation and administration Unregistered IP right Plant breeder’s rights Geographical Indications Plant Breeder’s Rights Act 1994 Australian Grape and Wine Authority Act 2013 a Grape-based wine and spirit geographical indications are administered by Wine Australia, while geographical indications for all other goods are administered through the certification trade mark system. In some areas, government policy development has suffered from both a lack of transparency and a weak evidence base, especially for IP arrangements in trade agreements (chapter 17). Inquiry participants raised concerns about Australia’s processes for agreeing international trade agreements incorporating IP provisions. 2 The lack of evidenced-based policy results from several factors, including policy responsibility fragmentation, practical challenges obtaining data (particularly for copyright, which is an unregistered right) and quantifying the effects of IP policy. And with the recent abolition of the Advisory Council on Intellectual Property (ACIP), there will no longer be a standing body to provide independent advice on domestic IP policy. Inquiry participants have also raised questions about IP Australia’s dual role as both policy adviser and rights administrator given the potential risks from blurring policy and regulatory functions, although views about the extent of these risks vary (chapter 17). 2 National Tertiary Education Union, sub. 24; Choice, sub. 26; Business Council of Australia, sub. 59; Australian Industry Group, sub. 60; The Institute of Patent and Trade Mark Attorneys, sub. 73; Weatherall, Alexander and Handler, sub. 99; Alexander et al., sub. DR505; Gleeson, sub. 128; Moir, sub. 137; Lateral Economics, sub. DR187. HOW DOES THE SYSTEM FARE? 97 3.5 Are IP arrangements suited to Australia’s circumstances? Australia is a significant net importer of IP-intensive goods and services Poorly designed IP rights impose costs, irrespective of whether countries are net importers or exporters of IP. Hence, all countries have an interest in ensuring that IP rights are optimally set (chapter 18). However, Australia is a large net importer of IP-protected goods and services, and the gap between IP imports and exports is growing rapidly (figure 3.8). This status as a net importer significantly influences the impact Australia’s IP arrangements — and any change in those arrangement over time — have on community welfare. Figure 3.8 Trade in intellectual property $ billion (2014-15) (i) 5.0 4.5 4.0 Charges for the use of IP Imports Exports 3.5 (ii) IP-intensive goods 80 70 Imports 60 Exports 3.0 50 2.5 40 2.0 30 1.5 1.0 20 0.5 10 0.0 0 Sources: ABS Balance of Payments and International Investment Position, Australia, Cat. no. 5302; appendix C. 98 INTELLECTUAL PROPERTY ARRANGEMENTS Productivity improvements depend on both investment in locally produced IP, as well as the knowledge gained (transferred by way of import) from abroad. Productivity growth attributable to technological innovation is an outcome of local and overseas R&D efforts. In a small country like Australia, the bulk of technological productivity improvements are sourced from abroad. This is illustrated by the fact that over 90 per cent of patents in Australia are foreign owned (chapter 7). Imported IP can also provide a source of investment in the creative arts and enrich cultural life through access to foreign music, literature, and films. As the Department of Foreign Affairs and Trade submitted: Imports of intellectual property can be valuable to the economy. Only a small number of countries export more than they import. Importing can be an efficient way of utilising intellectual property. Mining is a good example, where imported intellectual property helps support our biggest export industry; transforming imports into exports. (sub. 65, p. 7) Similarly, IP Australia reported: It is worth noting that being a net importer of IP does not necessarily have adverse economic implications. As long as imported knowledge and technology translates into improved domestic productivity, there is scope for significant economic benefits. (2013, p. 22) But inefficient levels of IP protection result in Australia paying too much for imported knowledge. By international standards, Australia has adopted strong IP settings (box 3.5) and the level of protection afforded is likely to be beyond the point where it promotes significant increases in technology transfer (figure 3.9). This imposes additional costs with diminishing benefits. These costs are ultimately borne by Australian consumers, and where imports are transformed into exports, our exports become less competitive. Conversely, protection above minimum international standards does not benefit our exporters without reciprocal protection in our trading partners. Discussions about access to foreign technology and copyright material should therefore be framed in terms of balancing rather than maximising IP rights. HOW DOES THE SYSTEM FARE? 99 Box 3.5 Australia provides relatively strong IP rights Researchers have developed indexes of the strength of countries’ IP rights. Such indexes are often used in studies that examine the relationship between the strength of IP rights and technology transfer. The Commission has data for the following six IP and patent-related indexes. • The US International Trade Commission (Riker 2014) infers the strength of IP rights based on data on US cross-border receipts of international royalties and license fees. • The Taylor Wessing Global IP Index is based on various legal measures, published empirical data such as royalty fee payments and a worldwide survey of IP owners and users. • The World Economic Forum’s Global Competitiveness Report measures market participants’ perceptions about the effectiveness of different countries’ IP policies. • The Ginarte Park index (Ginarte and Park 1997; Park 2008) averages countries’ scores on five patent-related dimensions: strength of patent coverage; membership in international treaties; duration of patent coverage; enforcement mechanisms; and restrictions. • Papageorgiadis, Cross and Alexiou (2015) measure enforcement practices and the administrative functioning of national patent systems as perceived by managers. • The US Chamber of Commerce Global IP Center index is based on countries’ scores on 30 IP-related indicators, including enforcement and ratification of international treaties. Australia’s score exceeds the average for all countries and high income countries in every index. Australia’s score in IP strength indexes relative to other countriesa,b Australia High income countries All countries USITC (2012, IP) Taylor Wessing (2015, IP) WEF (2015-16, IP) Ginarte Park (2010, patents) Papageorgiadis (2014-15, patents) USCC (2016, IP) 0.00 0.25 0.50 0.75 1.00 a The chart shows the most recent year in which data are available. This year, along with the focus of the index (whether IP in general or patents), is reported in brackets. b To better compare scores index values are converted to the range 0–1. 100 INTELLECTUAL PROPERTY ARRANGEMENTS Figure 3.9 Trade in IP and strength of IP arrangementsa As measured by the Global Intellectual Property Index (GIPI)b 670 AUS 650 NZL AUT 640 GIPI score GBR DEU 660 SWE NLD CHE USA CAN IRL 630 HUN 620 KOR 610 POL CZE ITA JPN ESP ISR SVK 600 590 580 MEX -80 -60 -40 -20 0 20 Trade balance (per cent) 40 60 80 a Trade balance = (receipts-payments)/total trade. Trade data are for 2014, except Canada, New Zealand, Korea, Mexico (2013) and Slovak Republic (2012). b The vertical line indicates the average GIPI score of the countries listed for 2016. A higher GIPI corresponds to stronger IP arrangements. Sources: OECD Main Science and Technology Indicators database; Taylor Wessing (2016). Several participants cautioned against taking a static view of Australia’s position as a net IP importer, with some misinterpreting the Commission’s position as mercantilist. These participants maintained that reforms should be directed at improving incentives for creativity and innovation that could improve Australia’s international competitiveness. 3 However, those who produce IP are often also users of IP. Inefficiencies in IP arrangements can increase transaction costs for such producers. This is particularly important not only for sequential innovation, but also in view of the rise of global value chains, where intermediate goods, services and IP are sourced from many countries and at different stages of production. As figure 3.8 above makes clear, current IP arrangements are doing little to transition Australia from being a net importer of IP to a net exporter. 3 WeCreate, sub. DR238; WiseTech Global, sub. DR274; Australian Society of Authors, sub. DR343; APRA AMCOS, sub. DR404; igea, sub. DR437; Screenrights, sub. DR454; Music Council of Australia, sub. DR498; Australian Music Publishers Association, sub. DR535; Australian Copyright Council, sub. DR543; University of Technology Sydney, sub. DR564; IP Australia, sub. DR612. HOW DOES THE SYSTEM FARE? 101 Increases in the protection of IP rights may have done little to promote technology imports Some participants emphasised the role stronger IP rights can play in promoting imports of technology (‘technology transfer’) (DFAT, sub. 65; FICPI, sub. DR581; IPTA, sub. DR562). In this context, the Department of Foreign Affairs and Trade said that international IP agreements are an important means to facilitate inward investment. IP rights promote technology transfer through various channels, including trade, licensing and foreign direct investment (FDI). For countries with relatively strong IP rights such as Australia, there is evidence that licensing is the most important mechanism for importing technology (Breitwieser and Foster 2012). However, it does not follow from the available literature and evidence base that recent increases in the protection of IP rights in Australia would significantly increase technology transfer (box 3.6). Box 3.6 The Goldilocks Principle: IP and Technology Transfer Too much, too little or just right? Similar to its effects on domestic innovation, the level of IP protection can affect both access to and the cost of technological knowledge developed overseas. Too little protection may lead to access to technology being withheld or delayed. Too much protection may impede technology transfer, by limiting diffusion of technology or raising input costs to Australian innovation. While international empirical literature finds an association between stronger IP rights and technology transfer, this is mainly in the area of patents — copyright and trade mark rights are less important in promoting technology transfer. More importantly, most data sets used in the literature relate to middle-income countries, which typically provide relatively weak IP rights. The association between stronger IP rights and technology transfer is thus likely driven by these countries increasing the level of IP protection from a weak base. For countries, such as Australia, that already have relatively strong IP rights, increases in the strength of rights may not promote technology transfer. While some studies find that licensing payments to foreign rights holders increase with stronger IP rights, this may reflect higher prices due to the rights holder’s greater bargaining strength rather than increased technology transfer — empirical evidence suggests that stronger IP rights increase the share of a technology’s value that the licensor can appropriate. Technology transfer relies on numerous country-specific factors, of which IP rights are just one. Recent analyses of international technology licensing conclude that barriers to licensing mainly arise due to informational constraints, such as difficulties finding licensing partners. Adopting or locking-in higher standards of IP protection is therefore likely to be targeting the wrong area to effectively increase investment and technology transfer. Sources: Branstetter et al. (2006); Wakasugi (2007); Park and Lippoldt (2005); United States International Trade Commission (2016); Hall (2014); Thursby and Thursby (2006); Zuniga and Guellec (2009); Kani and Motohashi (2012). The remainder of this report examines Australia’s IP system in further detail, and makes recommendations to improve its operation. 102 INTELLECTUAL PROPERTY ARRANGEMENTS 4 Copyright term and scope Key points • Copyright protects the material expression of literary, dramatic, artistic and musical works, as well as books, photographs, sound recordings, films and broadcasts. In addition to being instrumental in rewarding creative and artistic endeavour, many creators value the recognition that the copyright system provides. • But an effective and efficient copyright system needs to balance the cost of creating new works (including the incentives necessary for that creation) against the community benefit from the use of creative works. Australia’s arrangements fail to strike this balance. • Australia’s copyright arrangements have expanded over time and new rights have been granted to rights holders. In some instances, expansion has been justified, as for much online material. • In other cases, copyright extension has swung too far in favour of rights holders, often with no transparent evidence-based analysis. Retrospective extension of term from 50 to 70 years after death is a prime case. • Australia’s copyright arrangements lack balance and have been slow to adapt to technological change, imposing costs on the broader community. – Overly broad and long copyright protection means Australians pay more or have difficulty accessing copyright material. – Excessively long copyright protection increases the likelihood works will become commercially unavailable or orphaned (where rights holders can no longer be identified). Copyright works that have already been produced, but cannot be accessed and used by consumers, benefit no one. • Technological change and accelerating digital disruption underscore the need for an adaptable copyright system. Research suggests today’s ‘representative consumer’ infringes the copyright of non–commercial and commercial works over 80 times each day. • While the options for reform are limited by Australia’s international intellectual property agreements, there is scope to redress the imbalance. – The proposal of the Australian Government in late 2015 to abolish perpetual copyright protection for unpublished works should be implemented without delay. – The Australian Government should work with like-minded international partners to achieve greater balance in the way copyright operates internationally. – Targeted reforms should be implemented to improve access to copyright material, including through changes to copyright exceptions and transparent statutory licensing arrangements (covered in the following chapters). COPYRIGHT TERM AND SCOPE 103 Australia’s copyright arrangements protect the material expression of original literary, musical, artistic and dramatic works, as well as published editions, sound recordings, films, and television and radio broadcasts. Some exceptions aside, copyright owners have the exclusive rights to prevent their creative expressions from being copied, performed, published, communicated or adapted without their consent. The advent of the digital age has given rise to disparate views about whether copyright remains ‘fit–for–purpose’. Copyright material is created and used in ways that would be unfathomable to the early developers of copyright. Well-functioning copyright arrangements need to adapt to technological shifts and resulting changes in user behaviour. A pragmatic policy response to these challenges must also recognise Australia’s policy choices are constrained by international agreements. This chapter forms part of a broader analysis of copyright arrangements. Chapter 5 examines how creators, intermediaries and users engage with copyright-protected material through licensing arrangements, while chapter 6 explores the role and use of copyright exceptions. 4.1 Australia’s copyright arrangements and the creative sector The Copyright Act 1968 (Cth) (Copyright Act) protects the material expression of an original idea. To be original, a work must not be a copy of an existing work and must have a human author. Copyright can apply to four broad categories of works — literary, dramatic, musical and artistic — and four categories of subject matter other than works — published editions, films, sound recordings and broadcasts. Key features of the copyright system include: • scope — copyright only applies to the expression of an idea and not to the idea itself. For instance, copyright does not protect scientific formulas, short phrases or slogans • formalities — unlike other IP rights, registration or notices of copyright are not required for creators to benefit from copyright; protection is triggered when an original work is reduced to a material form • exclusive rights — generally, rights holders have the exclusive right to reproduce their work in material form as well as publish, copy or adapt their work, perform their work in public or communicate their work to the public. Authors also have moral rights, requiring their work to be properly attributed and preventing a work from being damaged, destroyed or altered in a way prejudicial to the author’s reputation • term — copyright generally protects literary, musical, dramatic and artistic works for the duration of the creator’s life plus 70 years. Following publication, sound recordings and films are protected for 70 years, television and sound broadcasts for 50 years, and 104 INTELLECTUAL PROPERTY ARRANGEMENTS published editions for 25 years. Unpublished works currently have an unlimited period of protection • use — rights holders commonly licence third parties to exercise their exclusive rights, often in exchange for the payment of royalties. Intermediaries, including literary publishers, record companies, film studios and broadcasters play a significant role in enabling rights holders to commercialise their work. Collecting societies can also work on behalf of rights holders to license low value/high volume uses of copyright material, collecting and distributing royalties. Some intermediaries will adapt or transform a work for further sale, such as a musician sampling music in a new composition, or film studios adapting a book • exceptions — certain uses of copyrighted material are allowed without the authorisation of rights holders. Australia’s copyright system includes an exception for ‘fair dealing’ for research or study, criticism or review, parody or satire, reporting the news, judicial proceedings and professional advice. Exceptions also allow for temporary reproductions to be made in the course of communicating a work, and for recording a television show on a video tape for private viewing, or copying music to an mp3 player • unauthorised use of copyright material — unauthorised use generally constitutes a civil infringement, requiring rights holders to enforce their rights, usually in the Federal Court of Australia. Commercial-scale infringements of copyright are a criminal offence and prosecuted by the Commonwealth Director of Public Prosecutions. Rights holders are able to seek an order requiring an Internet Service Provider to block access to an overseas website that facilitates online copyright infringement and the Australian Border Force has a role in detecting and seizing potentially infringing copyright-protected goods at the border (chapter 19) • international obligations — although copyright law is implemented on a domestic basis, the minimum coverage and term of protection has long been governed by international treaties. The Berne Convention, signed in 1886, was the earliest multilateral copyright treaty. While further multilateral agreements have increased standards of protection, IP standards are governed increasingly by bilateral and regional trade agreements to which Australia is a party. The Australian creative sector produces many works … One indicator of the scope of Australia’s copyright arrangements is the value of copyright material. The most direct and robust measure of the market value of copyright material is the amount spent creating it. In Australia, the capital expenditure on ‘artistic originals’ (the category of goods covered by copyright) was estimated at $2.7 billion in the year ending June 2015 — this equates to about 0.16 per cent of Australia’s gross domestic product (ABS 2015a). COPYRIGHT TERM AND SCOPE 105 While a range of other estimates have been advanced, many are overstated and have little economic meaning. For example, some value the contribution of the copyright industries at more than seven per cent of gross domestic product per year (Australian Copyright Council, sub. 36, p. 3). Such valuations capture more than the value of copyright material, and include the costs of labour and other capital in activities such as advertising, distribution and collecting royalties. They also include the contributions of industries that are related to copyright or use materials that may be subject to copyright, including the manufacturing, wholesaling, retailing, renting and leasing of televisions, computers, tablets and smartphones, musical instruments, photographic and cinematographic devices, photocopiers, blank recording materials and paper (WIPO 2015a). Debate on the precise value of copyright industries is ultimately not that helpful. The market value of copyright material is not necessarily equal to its economic value. For example, free works are potentially very valuable, yet are not captured in calculations of employment, industry value added or gross domestic product. Nor does the size of an industry per se indicate whether resources are allocated to activities most valued by consumers. … even so, Australians overwhelmingly consume foreign works Notwithstanding the value of locally produced copyright content, Australia is a significant net importer of copyright material, importing much more copyright material than it exports (box 4.1) This holds true across the spectrum of works. Commission analysis reveals that of the top 5000 book titles sold in Australia in the 12 months to 31 May 2016, less than one third were written by an Australian author. Similarly, of the top 100 albums sold in Australia in 2015, 35 were recorded by Australian musicians or bands, and only 4 of the top 20 albums were by Australians (ARIA 2015). With respect to films screened in Australian cinemas, just 6 per cent were Australian and 7.2 per cent of Australian box office takings in 2015 were from films under Australian or shared creative control (Screen Australia 2016). These findings are not surprising given the relative size of the Australian population coupled with the modern day ease of international trade in copyright goods and services, especially with digital data and advances in digital compression. 106 INTELLECTUAL PROPERTY ARRANGEMENTS Box 4.1 Australia’s trade in copyright goods and services Australia is a net importer of goods and services that embody copyright material, which include printed matter, sound recordings, video games, software, computer and information services, and audiovisual products. Australia now imports close to $9 billion in copyright-protected goods and services — around 2.4 times more than it exports. Imports of copyright protected material have grown much faster than Australia’s exports. Between 1998-99 and 2014-15 imports grew by over 120 per cent compared to export growth of about 40 per cent. Australia’s copyright trade has shifted away from physical goods towards services with the emergence of digital distribution and consumption. Australia’s imports of copyright-protected goods have increased only slightly (about 20 per cent) in real terms since 1998-99, while service imports have more than tripled. Copyright services now account for around three quarters of Australia’s copyright imports. 10 9 8 $ billion 7 The spike in exports for 2000-01 reflects a one–off increase due to the broadcasting rights for the Sydney Olympic Games. Imports 6 5 4 3 2 Exports 1 0 Sources: Commission estimates based on ABS Balance of Payments and International Investment Position, Australia, cat. no. 5302.0; International Trade in Goods and Services, Australia, cat. no. 5368.0; unpublished data. Some participants questioned the policy relevance of Australia’s net importer status. Some even criticised the Commission’s draft report for framing the analysis of copyright in terms of ‘imports versus exports’ (for example, the Australian Copyright Council (sub. DR543), the Interactive Games and Entertainment Association (sub. DR437), and APRA AMCOS (sub. DR404)). As discussed in chapter 3, poorly designed IP rights impose costs, irrespective of whether countries are net importers or exporters of IP. However, Australia’s status as a copyright net importer amplifies the impact copyright arrangements — and any change in those arrangements — have on community welfare. COPYRIGHT TERM AND SCOPE 107 Digital disruptions to the copyright system Digital technologies are disrupting both the supply and demand side of the copyright ‘coin’: • In some cases, the cost for creators to produce new works and for intermediaries to bring works to market has declined (box 4.2), but cost reductions also threaten to ‘disintermediate’ many businesses within the copyright value chain by enabling many artists to market and sell their works to consumers directly. The creation and dissemination of copyright material by altruistic providers at low or no cost is greater than ever before. • Consumers have access to a wider array of copyright–protected works than ever before, in a variety of formats, with 24–hour access and purchasing the new norm. But such access also enables greater scope for infringement. Multiple views exist on the direction copyright policy has taken in responding to these challenges. Some argued that Australia’s copyright arrangements were working well in the digital era. For example, the Copyright Agency stated: The objectives underlying the copyright system continue to be sound, and the system has adapted better than is sometimes acknowledged. There have been amendments to the legislation in response to technological and other developments … but many technological developments have been accommodated without legislative change. (sub. 47, p. 14) Others questioned whether the system was keeping pace with technological change. For example, in its submission the New South Wales Department of Justice stated: Copyright law has failed to adapt to the changes resulting from digital technology that have happened over the past 20 years and continuing. In an array of cases courts and tribunals have attempted, with varying levels of success, to understand digital processes and apply copyright law to them … (sub. 39, p. 1) Electronic Frontiers Australia similarly described Australia’s copyright system: The complexity of the present regime, and references to out-dated technologies, increases disregard for copyright law as being ‘out of touch’ with current realities. (sub. 114, p. 2) And the Australian Digital Alliance argued: … copyright has faced significant challenges in adapting to the digital era. The technologies and markets used to create and deliver copyright works have changed significantly, yet Australian copyright law has not moved sufficiently to accommodate these changes. This has resulted in a system that, when taken in its whole, is neither efficient nor effective. It is inflexible and slow to adapt to new technologies and markets, and there is little transparency in or accountability about how changes to the system are determined. (sub. 108, p. 1) 108 INTELLECTUAL PROPERTY ARRANGEMENTS Box 4.2 The changing costs of producing and marketing music Digital technologies continue to reduce the cost of producing, distributing and marketing music. In the 1990s, producing an album on a major label cost over $250 000 (in today’s dollars). It required expensive equipment, personnel and studio time, but now it is possible to create and record commercial-quality work with equipment costing only a few thousand dollars (Albini 1993; Waldfogel 2014). To distribute music before the Internet, music needed to be produced in a physical form (for example, CDs, cassettes, vinyl) and shipped to stores. This was especially expensive for independent musicians (Albini 2014). In comparison, it now costs around $10 to make a song available on iTunes (Waldfogel 2014) and it is free for artists to share their music on websites such as Bandcampa and SoundCloud.b The Internet has also reduced the bottlenecks of traditional music marketing. In the 1980s, promoting a music single cost $150 000 (over $400 000 in today’s prices) and required getting a song on the radio — major radio stations added around 200 new songs to their playlist every year at a time when industry released over 100 000 songs each year (Waldfogel 2014). Now, digital review outlets such as Pitchfork and Stereogum review hundreds of new albums each year at much lower cost than traditional media outlets — the fixed costs of operating a website are much lower than running a newspaper or magazine (Waldfogel 2014). Creators also use social media to communicate directly with fans to announce tours and album releases (Waldfogel 2012). The Internet also assists creators seeking funding. Creators can use crowd-funding websites like Kickstarter to access capital. As of August 2015, Kickstarter campaigns had raised $153 million for music-related projects (Johnson 2015). a Bandcamp takes a 15 per cent share of music sales, but when the artist’s total revenue surpasses $5000 in a year, Bandcamp’s cut falls to 10 per cent. Bandcamp also provides a premium service for $10 a month and a service for music labels, which costs $20 a month (Bandcamp 2015). b SoundCloud’s free account option offers artists three hours of upload quota. SoundCloud’s paid service costs $145 per year, wherein artists receive an unlimited upload quota and more extensive statistics (SoundCloud 2015). The transition to the digital era is yet to fully play out. While the music industry has been leading the way on digital media, further improvements in compression technology have made new copyright distribution platforms viable, threatening existing distribution, retail and broadcasting business models in other sectors. The speed at which businesses are disrupted has accelerated, underscoring the need for an adaptable copyright system. In this environment, the role for government should be to ensure a flexible and adaptive copyright system that balances these competing and evolving forces. As noted by Pollock: Much of the motivation for strengthening copyright in recent years … has been based on the implicit assumption that the move to a digital environment necessitated an increase in the strength of copyright because technological change made unauthorised copying (‘piracy’) easier. But focusing only on the reduction in the costs of unauthorised copies ignores the impact of technology on authorised production and distribution. … such an approach omits a major part of the overall picture and may lead to erroneous conclusions regarding both the necessity and direction of policy changes. (2007, p. 12) COPYRIGHT TERM AND SCOPE 109 4.2 Applying the Commission’s framework In chapter two, the Commission laid out a framework for how copyright policy should be considered and applied in Australia. In particular, Australia’s copyright system should be: • effective, encouraging the creation and dissemination of creative works that would not have occurred in the absence of copyright • efficient, ensuring new works are generated by the most efficient creators at the least cost to society, traded efficiently, and do not impede competition • adaptable, adapting to changes in technology, competition and general economic conditions (particularly given the advent of the digital era) • accountable, where Australia’s domestic copyright policy settings demonstrate considered analysis that is transparent, evidence–based and reflects community values. As with other IP, the prime economic aim of the system is to create sufficiently strong property rights to ensure adequate incentives for the creation and dissemination of works, while not permitting the use of excessive market power. In both cases, the focus is on the consumer, not the creator. A consumer does not benefit from a product that is not available or when prices for goods are unnecessarily high. Many submissions supported the use of the Commission’s framework to assess the copyright system, such as the Australian Digital Alliance (sub. 108), the Australian Competition and Consumer Commission (ACCC, sub. 35), and the NSW Department of Justice (sub. 39). However, other participants questioned whether the Commission’s approach could fully encompass the wide range of interests and outputs of Australia’s creative sector. For example, in the Commission’s public hearings, the Arts Law Centre of Australia expressed its concern that an: … economic framework might not take into account all the effects on welfare that could stem from changes in the IP system, including the cultural, personal or social values inherent in creation of creative work. (Arts Law Centre, trans., p. 131) The Australasian Music Publishers’ Association Limited (sub. 34) and the International Confederation of Music Publishers both highlighted the role copyright plays in generating an income for creators: Strong copyright laws are needed to provide a positive environment that rewards authors and composers. Without an effective copyright and [intellectual property rights] enforcement framework, the ability of authors and composers to receive proper remuneration from their works — and thus to make a living — would be diminished. Their incentive to create new works and invest in innovation would be removed entirely. (sub. 32, p. 4) 110 INTELLECTUAL PROPERTY ARRANGEMENTS And the Australian Publishers Association argued that copyright is quite distinct from other forms of IP: The Commission should modify its framework so that its assessment of Australia’s intellectual property arrangements takes into account the fundamental differences between copyright and other forms of intellectual property. It should ensure that its consideration of the Australia’s copyright arrangements includes their effect on matters overlooked by the predominance of debates over balance between incentives and access, including their contribution to the operation of Australia’s public domain and within that the integrity of an author’s choice of whether and how to publish. (sub. 48, p. 5) Of course, the notion that an author must get a sufficient expected reward is integral to the Commission’s framework as this is a key driver of the incentives for creation. However, the words ‘sufficient’ and ‘expected’ are critical. • First, from the consumer perspective, the return needed to elicit supply is the relevant measure of ‘reward’, not a separate normative valuation of the worth of the activity. • Second, as in all entrepreneurial endeavours, a return cannot be guaranteed. The rates of return in the creative industries are highly skewed, with relatively few creators having large earnings and many making low returns. The difference between expected returns and realised ones is not inherently a symptom of a system that undervalues creation or creators. In some instances, when creators refer to the concept of a just return they are pointing to the relative gains from the sale of copyright by creators and intermediaries, including concerns about the potential exercise of market power by intermediaries. Evidence suggests much of the returns from copyright material are earned by intermediaries, rather than authors, musicians and the like (box 4.3). While the bargaining power between the original artist and the intermediary may be one-sided, intermediaries are similar to financial managers of risky equity portfolios. They assume risks across individual creators with the goal of securing enough returns from some big successes to balance the losses or low returns on many. That model means that intermediaries must have a significant share of the returns from successful works. To the extent that intermediaries exert market power, this is a matter best addressed by competition policy. It is hard to see a model of copyright that determined the specific terms and conditions of the commercial exchange of rights between parties. COPYRIGHT TERM AND SCOPE 111 Box 4.3 Does copyright provide a just reward for creators? Some participants (for example, the Australasian Music Publishers’ Association Limited, sub. 34 and the International Confederation of Music Publishers, sub. 32) argued that the copyright system was aimed at providing creators with a ‘just reward’ or a ‘living income’. Evidence suggests much of the returns from copyright material are earned by intermediaries, rather than authors, musicians and the like. The stereotype of the ‘struggling artist’ has some degree of truth to it, and evidence suggests many involved in creative endeavours work multiple jobs and receive financial support from their families (Throsby and Zednik 2010). However, the distribution of revenues along the value chain reflects the risks borne in bringing works to market. Creators commonly licence their copyrights to a publisher (Giblin 2015) and according to some, creators have little choice but to accept the terms presented by a publisher. Individuals such as myself are powerless to resist the egregious terms offered by publishers and others. We have little ability to negotiate. The intellectual property laws do not provide a level playing field. … For me, the loss of my ability to claim payments [from] statutory licences because a publisher insists that I sign a clause assigning such payments to it is difficult to resist. The publisher has all the power. (Fisher, sub. 18, p. 1) Others have argued that a ‘just rewards’ framework overly relies on emotion to obscure the realities of commercial arrangements. Giblin notes: Such [just rewards] arguments are highly effective because they appeal to our inclinations to reward authors for their creative contributions. In practice however, relatively few of copyright’s rewards find their way to those creators. Indeed, such a huge proportion of the benefits of increased protection are captured by other cogs in the cultural production chain that authors are sometimes viewed as a mere ‘stalking horse’ masking the economic interests of others. In the case of the US term extension for example, the beneficiary of the unbargained-for windfall from the US term extension was the rightholder at the time it was granted; very little of it accrued to the original author or their family if it had previously been transferred. (2015, p. 16) While some point out that the role of intermediaries has always been central to copyright. Beyond the absence of data, the biggest problem in discussions of copyright policy is the failure to recognise the centrality of distributors to copyright policy design. Most copyright policy discussion is founded on the myth that copyright is designed to meet the needs of authors. Yet the history of copyright policy shows clearly that copyright was an exchange of censorship services for monopoly privileges for publishers. (Moir, sub. 137, p. 2) Other intermediaries, such as copyright collecting societies, also play a role in bringing copyright-protected works to market, often charging rights holders a fee for doing so. Chapter 5 discusses the role of collecting societies further. 4.3 The scope of copyright protection The scope of copyright protection has a significant influence on the strength of the system and encompasses the nature of the works protected — be they music, movies, books, software or databases — and the rights enjoyed by rights holders. Reflecting the important role copyright can play in encouraging and protecting commercial works, many associate copyright with works intended for the market. However, as noted above, copyright protection applies freely and automatically at the time an original work is 112 INTELLECTUAL PROPERTY ARRANGEMENTS reduced to a material form. Unlike other forms of IP rights, registration of copyright is not a requirement for protection, nor is it permissible in Australia. 1 The lack of a screening mechanism means that copyright protects a wide array of material that do not need protection, were produced with no intention or desire for commercial return, or would be produced even in the absence of copyright protection. In short, the copyright system cannot differentiate between the million dollar movie and the family home video. The Australian Digital Alliance contrasted the reach of copyright with other forms of IP. Evidence for the extent of this overreach can be seen by comparing copyright to other intellectual property rights. It by far exceeds them in relation to the: • breadth of material captured — copyright applies automatically to all creative works, without the need for registration or other administrative steps, and with very low originality and effort requirements. This is the case even where these works are not intended for commercial or even public use: for example, a doodle or text message receives the same protection as an oil painting … • scope of uses prohibited — all acts of reproduction, communication and performance are prohibited, even where they are non-commercial; private (at least in the case of reproductions); or (in many cases) temporary. The rise of digital technology and its reliance on reproduction and communication as part of technical processes, coupled with decisions by governments worldwide to count each of these acts as separate copyright uses, has significantly increased the reach of copyright law. (sub. 108, p. 6) Much copyright protection is unwarranted … Copyright’s overly broad reach, in conjunction with the advent and rapid spread of digital technologies, means that every day actions undertaken by most Australians are now subject to copyright. Most emails, social media posts and photos taken on smartphones are copyright protected. The copyright system was not envisaged to cover such a large volume of works. Nor is it needed, as these creations would still be created if they were not protected given they are made for personal reasons and are non-commercial in nature. Whether the protection of these works in Australia really poses infringement risks for users depends on enforcement and the willingness of a court to impose penalties. One researcher observed that — assuming full enforcement and a strict reading of case law — a ‘representative user’ infringes the copyright of non–commercial and commercial works over 80 times in one day2 — potentially facing a liability of $12.45 million (Tehranian 2007). 1 As recently discussed by the Register of the US Copyright Office, while registration is not possible for the initial period of life plus 50 years under the Berne Convention, some formalities are possible for the period beyond life plus 50 years, up to the current term of life plus 70 years (Pallante 2013). 2 Copyright infringements include replying to twenty emails (including the original email in the reply or forwarding an email constitutes an infringement), doodling a copy of a famous building, showing a Hanna–Barbera tattoo in public, and publicly–reciting a poem. COPYRIGHT TERM AND SCOPE 113 Such claims are more hypothetical than substantive — there has been no flood of cases involving breaches of copyright involving non–commercial works. But as some inquiry participants pointed out, laws should reflect community values and it should not be up to the discretion of rights holders as to whether acts — which the community considers appropriate — are the subject of consent and potential legal action. Further, some participants have suggested there can be real costs to the community associated with the application of copyright laws for non–commercial works. For example, Australian Policy Online (APO) highlighted the contributions of one category of non-commercial works protected by copyright known as ‘grey literature’, which includes: … reports, discussion papers, briefings, case studies, literature reviews, fact sheets, evaluations, submissions, working papers, conference papers, data, technical reports and specifications, policies, strategic plans, infographics and much more. Known collectively by the term ‘grey literature’, it makes a substantial contribution to public policy, education, commercial innovation and social development. (sub. DR444, p. 2) APO estimates over 4000 organisations in Australia produce and use grey literature: around one third are universities, one third government departments and agencies, and one third non-government organisations, ‘think tanks’ and consultants. APO’s current collection of grey literature extends to over 30 000 items, the vast majority of which is originally ‘… made available to the public, or to a limited audience, for free, disseminated in print or digitally online, however a portion is also sold’ (sub. DR444, p. 2). Although difficult to measure, researchers have estimated a lower-bound value of better access to grey literature in Australia at around $2.4 billion annually (Lawrence et al. 2014). The key point is that much of this work is either commissioned and paid for upfront, is produced for the intention of influencing government decision making, or is undertaken in the course of research. The provision of copyright protection is irrelevant in the creation of much of this work, yet as APO highlights, the exclusive rights embodied in copyright represent a major impediment to libraries, archives and other repositories curating and making this work available for use. In a similar vein, the Commission heard that schools, libraries and government users are particularly risk–averse, and often avoid engaging in activities even where the risk of being held liable for an infringement is negligible. Poorly drafted exceptions can further contribute to user uncertainty about what uses of copyright material are permissible. While international obligations mean Australia cannot change copyright arrangements to only apply to commercial works, adoption of broader copyright exceptions can improve access to non-commercial works. Chapter 6 discusses these options in detail. Orphan and commercially-unavailable works Copyright also applies to ‘orphan works’ — works where the copyright owner cannot be identified. All types of works can be orphaned, including books, sound recordings, 114 INTELLECTUAL PROPERTY ARRANGEMENTS photographs, diaries, maps and films. Libraries and archives tend to be the common repositories where orphan works are found. Works can be orphaned by circumstance (such as the long passage of time), because identifying information is missing or deliberately removed (such as if metadata is stripped from an image used online), or because the potential user lacks the skills or tools to correctly identify the rights holder. Across the world, orphan works are becoming increasingly problematic, especially for libraries and archives as they have sought to digitise, and make available online, their collections. The time required to diligently seek out copyright owners and licence those works is prohibitive. The Copyright Act provides no specific exception or defence for the use of orphan works, other than for those purposes covered by the narrow fair dealing exception. Given the lack of a registration system for copyright, estimating the number and value of orphan works is difficult. Estimates are scarce and often based on small samples of library holdings. For example, the British Library has estimated 40 per cent of all copyright material is orphaned, and that it holds around a million hours of broadcasts in its archives that cannot be used because no one knows who holds the rights (Dawes 2010; Menand 2014). That a vast existing cultural patrimony, already paid for and amortized, sits locked behind legal walls, hostage to outmoded notions of property, when at the flick of a switch it could belong to all humanity — that is little short of grotesque. (Baldwin 2014, p. 409) The Australian National Films and Sound Archive (2010) estimates around 20 per cent of its collection is orphan works. The Archive highlights how the presence of orphan works can frustrate the use and dissemination of audio visual material that documents Australia’s heritage. The NFSA sought to use a radio serial from the mid 1940s on SoundCloud (an online distribution platform that allows NFSA to share rare interviews and unique recordings from the mid 1940s). While the broadcast rights have expired, the music and script were still in copyright. The NFSA approached who they believed held the underlying copyright and despite being unaware they held the copyright they granted permission for two episodes to be uploaded. In the process of researching the copyright status of more serials, the NFSA discovered that it was more likely that a second party held the rights to the copyright initially cleared. Faced with competing claims to copyright ownership, the NFSA made a business decision to stall the project, assessing that it would be too time-consuming and costly to negotiate with both parties, particularly given the extensive research and efforts made to date to clear copyright with the first claimant. As a result the NFSA, the industry and the general public lost the opportunity to easily access a unique part of Australia’s audiovisual cultural heritage. (quoted in Australian Digital Alliance, sub. 108, p. 9) The issue of ‘orphans’ is not confined to older works and can affect relatively modern works (box 4.4). COPYRIGHT TERM AND SCOPE 115 Box 4.4 Multiplayer mode — copyright challenges in videogames System Shock 2, a classic videogame from the late 1990s, illustrates the challenges of complex intellectual property arrangements and reveals how issues can emerge many decades before the copyright expires. System Shock 2 was commercially unavailable for nearly 15 years despite it being the most requested game on a digital distribution website and more than 34 000 people registered their interest in playing the game. A game developer and digital distributor were keen to update and sell the game but they could not track down the rights holders. The rights were a ‘tangled mess’ — one of the original game developers, Looking Glass Studios, closed a year after the game was released and the copyright then transferred to Meadowbrook Insurance Group, which acquired Looking Glass’s assets. But even though Meadowbrook wanted to sell the rights they could not because the game’s publisher, Electronic Arts, held a trade mark. Both parties needed to agree before the rights could be licensed or a sequel made. It took 4 to 5 years to track down publishers and developers, but only six months to update the game and host it on the digital distribution website. Questions still remain around who now holds the rights. Star Insurance, an affiliate of the Meadowbrook Insurance Group, claims to hold the copyright and the trade mark. But Irrational Games, which co–created the game, may also have copyright and Electronic Arts may still own the trade mark. Night Dive, the company that re–released System Shock 2, is said to have also acquired the rights, but cannot proceed with a sequel without the permission of the publisher. Sources: Halfacree (2013); Newman (2011, 2013); Smith (2013). The application of copyright to out-of-commerce, or works that are no longer available, is equally problematic. Unlike orphan works, where the rights holder is unknown, for unavailable works the rights holder is usually known, but is choosing not to supply the market. While works often become orphaned due to the passage of time, copyright-protected works can be unavailable commercially quite soon after their original supply. As noted above, the issue of unavailable works stems from the exclusive rights copyright grants creators, and the lack of any obligation on rights holders to disseminate or make their work available. An infringement of a creator’s exclusive right occurs if someone else exercises those rights without licence or agreement. As the Australian Publisher’s Association noted, in practice this means: The author has the right to control the publication of her work, including the right not to publish it. (sub. 48, p. 6) Rights holders might not supply the market for a variety of reasons. Following the initial sales period, ongoing demand may be insufficient to justify the costs incurred in continuing production. A work might depend on particular technology for its use, such as a video game released for a particular console. Should that technology no longer be available, the rights holder might not want to update the work to operate with the new technology. A 116 INTELLECTUAL PROPERTY ARRANGEMENTS rights holder may also no longer want a work to be available, such as if their style has changed over time, or for personal reasons. While Australia’s copyright system currently has no ‘use it or lose it’ provisions, both the trade mark and patent systems have mechanisms to allow third party access where a rights holder is not exercising, or refuses to exercise, their exclusive rights: • The Trade Marks Act 1995 provides a mechanism for removing trade marks from the register if they are not being used by the rights holder. A three year period of non-use permits another potential user to apply for removal of the mark from the register. • The Patents Act 1990 allows a potential user to apply for a compulsory licence if, after a period of negotiating with the rights holder, the patented invention is not being (but is capable of being) worked in Australia on a commercial scale. Similar to the case of orphan works, estimating the number of commercially unavailable works — that is, existing books, music, television shows, movies, and video games that consumers wish to use but are not available to buy — is difficult. However, some research does show that many works, particularly books, are commercially unavailable for long periods during their period of copyright protection, only to become commercially available once they enter the public domain (when the copyright term ends) and can be reproduced and sold by anyone. For example, Heald (2014) took a random sample of 2000 books carried for sale by Amazon in its warehouse, and worked out the original publication date for each book. As shown in figure 4.1, recently published books (for example, published in the 2000s) were some of the more common books available for sale, featuring prominently in the sample. But the percentage share of titles in the sample published prior to the 2000s was low, except for titles now in the public domain (those published in 1923 and earlier and whose copyright term had lapsed), when availability spiked again. Part of the reason for the lack of availability is the expense in warehousing physical books — Heald’s research compared the markets for books and music, finding old commercial music (particularly in digital format) had much better availability. For example, well over 90 per cent of music ‘hits’ from the period 1923 to 1932 were available for purchase on iTunes in 2014, although the situation in the book market may be a ‘legacy’ issue that will change over time as more titles are released (both originally, and rereleased) as e-books. COPYRIGHT TERM AND SCOPE 117 Figure 4.1 Commercial availability of books Distribution of titles by decade of publication Percentage of titles in sample 20% Spike in availability when books enter public domain 18% 16% 14% 12% 10% 8% Books protected by copyright but unavailable 6% 4% 2% 0% 2000s 1990s 1980s 1970s 1960s 1950s 1940s 1930s 1920s 1910s 1900s 1890s 1880s Decade of publication Source: Heald (2014). The lack of any requirement for rights holders to actively supply the Australian market reduces the efficiency of Australia’s copyright regime. As with non-commercial works, reforms to copyright exceptions and limitations on remedies offer the potential to expand consumer access to orphan and unavailable works in ways that improve community welfare. These options are examined in chapter 6. … and rights have expanded over time Copyright has its genesis in protecting the publication and distribution of books. But as new technologies developed to produce and transmit creative works, new rights were extended to copyright holders. Progressively, copyright has expanded to cover a wider range of activities — some only loosely considered ‘creative’. New exclusive rights have also been granted to rights holders, including controls over importation of goods, moral and performers rights, and rights to control communication of a work to the public (table 4.1). 118 INTELLECTUAL PROPERTY ARRANGEMENTS Table 4.1 Selected changes to Australia’s copyright system Year Coverage 1905 1912 Copyright including translation rights recognised in books. Introduction of import controls on books. Copyright over mechanical reproductions recognised. Subject matter other than works (broadcasts, recordings and mechanical performances) granted copyright. Copyright extended to computer programs. 1968 1984 1986 2000 Copyright extended to satellite broadcasts. New criminal penalties for copyright infringement. Moral rights and right to communicate to the public introduced. Prohibitions against circumventing technological protection measures. 2004 Terms of copyright protection expanded (with retrospective coverage of existing works). Moral and performers’ rights Moral rights and performers’ rights were introduced in Australia in 2000 and 2004 respectively. Moral rights ensure a work is accurately attributed to an author and that the work is not treated in a way that harms an author’s reputation. Performers’ rights require the consent of a third party to record a performance, to broadcast or re–broadcast it, and to distribute any such recordings. 3 Performers can seek remuneration in exchange for such consent. A significant distinction exists between the two rights. Performers’ rights have an economic purpose, while moral rights are largely an expression of what is regarded as ‘fair’. That distinction is not absolute. For example, some participants argued moral rights assist creators in building a reputation, which in turn can underpin income generation from their work (Copyright Agency Ltd, sub. 47; International Confederation of Music Publishers, sub. 32; National Tertiary Education Union, sub. 24). 3 In 2005, Australia introduced copyright for some sound recordings of live performances. The owner of the recording medium (typically a record label or producer) and the performer(s) contributing the sounds to the performance are granted copyright, unless the performance was commissioned or the performers were provided services under an employment contract. Performers who hold copyright share the exclusive rights to make copies of the recording, to publicly communicate the recording, and to enter into commercial rental arrangements in respect of the recording. Performers were also granted copyright over sound recordings made before 2005, but these rights are more limited. COPYRIGHT TERM AND SCOPE 119 Many consider the protections under moral rights to go beyond those afforded by defamation laws, the tort of ‘passing off’, and misleading and deceptive conduct. For example, moral right protections cover actions of individuals in non–commercial settings, whereas previous protections only covered corporate and commercial dealings (CLRC 1988). APRA AMCOS explained the additional protections afforded by moral rights. There may be some crossover with other areas of law, such as defamation, passing off or, as suggested, misleading and deceptive conduct. But the grant of moral rights provides something beyond those other rights. Defamation is of no assistance where the author is unidentified, or where the work is changed in a way that does not injure the author’s reputation. Likewise, passing off can be a useful substitute for instances where an infringer claims the work to have been authored by him or her, and not by the real author. But this is but one aspect of moral rights, and passing off will be of limited use even in this scenario where the author lacks a protectable goodwill or reputation, for example where the author is only known outside the jurisdiction. (sub. 113, p. 8) To the extent that moral rights confer additional protection, they only do so in select circumstances. And it is not clear that the limited gaps in protection that may have existed prior to moral rights were in any way problematic. Indeed, the Copyright Law Review Committee found that (along with the case for performers’ rights) the case for moral rights was weak as: … there is some protection available under Australian law for the moral rights of authors and artists. Indeed, as is mentioned further below, this protection is apparently sufficient to satisfy Australia’s obligations under the Berne Convention. However, the European countries have for many years specifically protected moral rights, and the United Kingdom and Canada have recently indicated an intention to adopt legislation to like effect. The Committee notes this legislation and the reasons for its adoption. Nonetheless, a majority remains unconvinced that Australia should follow suit at least at this time. The view of the majority, that it is inappropriate to introduce legislation to protect moral rights at this time, is based on the following matters: • The practical problems associated with the introduction of protection of moral rights are not satisfactorily dealt with by those supporting adoption of such rights. • The theoretical basis for moral rights protection in a common law based system has not been identified. • There is insufficient indication of support for the introduction of moral rights of authors of copyright works. • It does not appear that violations of moral rights occur with such frequency or in such number as to warrant legislative intervention. • The Australian community is unlikely to endorse laws providing for moral rights protection. (1987, sec. 117) While the policy rationale for the introduction of moral and performers’ rights has been questioned, having now been established, the evidence these rights have adverse effects is not strong. While some legal cases have been brought in Australia and the United 120 INTELLECTUAL PROPERTY ARRANGEMENTS Kingdom against creators who ‘remix’ copyright material (on the basis these new works ‘debase’ or diminish the value of the original (Rimmer 2005)), a few examples are not enough to either revoke the current arrangements or to amend them as they stand. Right of communication to the public When some rights holders argued that traditional copyright laws would not protect the use of copyright material online, many countries introduced a ‘right of communication to the public’, to extend copyright protection to online activities. Introduced in Australia in 2000, the right of communication to the public entitles rights holders to make their work available online and to electronically transmit their work (Christie and Dias 2005). The right is technologically neutral and broad, covering any electronic means of transmitting a work including wireless transmission and any way of providing works online. Some see this as an adverse outcome, arguing that the right of communication further tilts the copyright balance in favour of rights holders as the added exceptions are relatively limited compared to the breadth of the right (Christie and Dias 2005; Suzor 2006). They consider that broad–based copyright over online presentation and transmission of creative works unduly restrict the flow of digital information and offers a higher level of protection than is granted to physical products, despite the introduction of an exception for technical copies (O’Shea 1995; Suzor 2006). Further, with the rise of digital technology, and its reliance on reproduction and communication as part of technical processes, watching a movie at home — an act which in the analogue age would not require any copying by users — now potentially results in multiple separate copies being made of copyright-protected material. This, coupled with decisions by governments worldwide to count each of these acts as separate copyright uses, means that each instance of copying remains subject to control by the copyright owner and damages if it is found to be unauthorised (Australian Digital Alliance, sub. 108, p. 6). But these perspectives paint an overly bleak view of the implications of the digital world. Prima facie, the rationale for the extension of rights is economically sound and, were it not present, creators would have few incentives to produce and publish works online, to the detriment of consumers. The Australasian Screen Association also argued the introduction of third party exceptions alongside the right of communication ensured that the copyright system maintained its original balance. The Copyright Act already strikes a balance between the interests of creators and consumers, including in the digital era. When the act was revised in 2000 as part of the Digital Agenda amendments, the introduction of the new ‘right of communication to the public’ for copyright owners was balanced by the introduction of exceptions available to users who provided facilities for communications (under 39B and 112E). The scope of the exceptions has been confirmed by the Courts. There is no evidence that this balance has impeded the ability of consumers to access legitimate copyright material or that business have been unable to develop new models of distribution and exploitation in Australia. Quite to the contrary, there has been COPYRIGHT TERM AND SCOPE 121 an explosion in availability of copyright-protected content, distribution models, and price points for consumers. (sub. 43, p. 14) Indeed, new technology has changed the economic calculus in reaching judgments about the strength and nature of online protections. For physical forms of copyrighted material — such as a book or DVD — a consumer can freely pass on or sell the material to a third user without any further return to the original seller. If nothing else, this recognises that any alternative is not only hard to enforce, but that the damage associated with such transfers must be small since only the original copy can be passed on. However, in the online environment, the free transfer of digital material could encompass the whole market for a product, and so additional protection is reasonable. Moreover, while the availability of digital dissemination has led to new forms of protection, they have arguably not further restricted consumer access. In fact, features of the digital age have allowed far more efficient models of the distribution of content. Unlike physical forms of copyright material, the marginal costs of digital distribution are effectively zero for the distributor. In contrast, any new physical version of a book or recording involves relatively high costs (printing, wholesaling, transport) and the consumption of the product is rivalrous. The implication is the optimum price of digital information will be zero when the fixed costs of creation and distribution can be recovered through another charge. Subscription services, like Netflix and Spotify, follow this model. Copyright owners receive returns from the net subscription revenues, but for any consumer, the marginal direct price of consuming an additional movie, song or television program is zero — an efficient outcome. The concern with extending or granting new rights (such as the right to communicate to the public) is not in the extension per se, but ensuring there remains an ongoing balance between creators and users. This issue of balance is discussed in further detail in chapters 5 and 6. Restrictions on transformative uses of existing material Transformative works — works that use existing material to create new original works — have existed for centuries. For example, Shakespeare’s plays were derived from Greek tragedies, and many Disney films are based on The Brothers Grimm fairy tales. The rise of the Internet and digital technologies in particular have led to the creation of an unprecedented volume of transformative works, by making it easier to source, combine and loop existing materials. Many songs and YouTube videos sample existing creative works, and Google publicly hosts images online (Suzor 2006). But copyright law was written with a particular form of industry in mind. The flourishing of information technology gives amateurs and home-recording artists powerful tools to build and share interesting, transformative, and socially valuable art drawn from pieces of popular 122 INTELLECTUAL PROPERTY ARRANGEMENTS culture. There’s no place to plug such an important cultural sea change into the current legal regime. (Werde 2004) Modern transformations often use copyright material, with some transformative works relying almost entirely on the electronic use of other works (box 4.4). While many transformative works are created on a commercial basis (such as Google presenting images on its website), it is becoming more common for users to create non-commercial transformative works. For example, home videos that use part of a copyrighted song are frequently uploaded to YouTube. Rights holders claim that unauthorised use of copyright material is costly as it displaces sales and damages artists’ reputations. As a result, rights holders and governments have increased efforts to remove infringing content from the Internet and to discourage further infringements. The degree to which they should, and decisions about infringement penalties need to be mindful of the consequences for users since transformative works are rarely reproductions. … mainstream copyright discourse has focused, to a large extent, on the risks that technological change pose to copyright owners. As a result, the benefits that these same technological changes can potentially bring to consumers, creators, and society, have been under-explored. … The reuse of copyright expression [cannot] be simply dismissed as ‘piracy’ or ‘free-riding’. There are significant benefits arising from transformative use, including the enhanced availability of diverse and decentralised speech and the freedom of individuals to express themselves, but also including the social benefits that come from deconstructing the media saturated environment we inhabit, and the benefits of not having such a large portion of that environment off-limits to creative expression. (Suzor 2006, pp. 7, 24) The key question is whether transformative works harm the market for the original work — do they have an appreciable impact on the demand for, and creation of, original material? If they do not, then there are few grounds for extending the bounds of copyright to transformative works (and regarding the use of the original material as an infringement). Mashups are an essentially harmless form of cultural expression. It is almost unimaginable that a potential consumer of an original recording would be satisfied with a mashup album to the exclusion of the original. … there is little chance that the majority of remix artists would be able to obtain a copyright licence, even if they could afford it. There is accordingly no financial loss to the copyright owners. Further, there is no strong argument that loss to reputation or other losses could arise from mashups, primarily because they are easily distinguishable from the original sound recordings. It is accordingly hard to see why this form of creative expression is not permissible. (Suzor 2006, p. 29) In that case at least, it does not appear to be in the public interest to curtail the transformative works, and in some cases, transformative works may increase the commercial value of original material because it brings that material to a new audience. Australia currently permits transformative uses of copyright material only to the extent it is covered by one of the purposes in the current fair dealing exception, or the use is licensed. COPYRIGHT TERM AND SCOPE 123 Box 4.5 Bring the noise and pay for it — the perils of music sampling Music ‘sampling’ refers to the practice of taking parts of recorded sounds from songs, videogames, radio plays, interviews and films, and combining them with instrumentation or other samples to create new music. Samples can be sourced directly from vinyl records, cassette tapes or digital recordings. In some cases, the sample is used ‘as is’, but in other cases changes are made to the samples’ pitch, speed or sound. Music producers can create songs ‘that are completely new, unexpected, and radically different from one another even if they derive from a common source’ (Sewell 2014, p. 45). Myself and Kanye West and Just Blaze can have the same sample, the same record, and you might hear it three different ways. It all depends on the ear of the sampler, which part of the song we want to take. (9th Wonder, quoted in Sewell 2014, p. 46) Sampling features prominently in hip–hop and electronic music and can introduce music–lovers to obscure and long forgotten works. For example, the drum solo in James Brown’s 1970 ‘Funky Drummer’ is now one of the most used samples — it is in nearly 1200 songs — but it was relatively unknown for the first decade after its release (Hein 2009; WhoSampled 2016). The music genre Chiptunes revolves around sampling sounds and music from 1980s videogames and computer programs, many of which are commercially unavailable. Sampling in the 1980s existed in a legal ‘grey area’ — artists asserted that they were creating new works using fragments of recorded music and rights holders argued that unauthorised sampling infringed their IP. Most samples were used without authorisation and any disputes were settled out of court. In the early 1990s, a couple of high profile court cases changed the sampling landscape. Biz Markie was ordered to pay $250 000 in damages and referred to the criminal court for copyright infringement on his 1991 song ‘Alone Again’ (Wang 2013). Australian courts typically take a firm view against unauthorised music sampling, particularly if the remixes are sold (Rimmer 2005). The shift to court–based dispute resolution prompted artists and labels to invest more in the sample clearance process or forgo releasing music when samples could not be cleared. Licensing costs can be significant as songs use multiple samples and separate licences may be needed to cover the music, lyrics and the recording (McLeod and DiCola 2011). Estimates of the licensing costs for sample–heavy albums produced in the 1980s suggest that the Beastie Boys’ Paul’s Boutique would have cost nearly $20 million and Public Enemy’s Fear of a Black Planet would have cost over $6 million (McLeod and DiCola 2011). Darren Seltmann from the Australian band The Avalanches has discussed the time and financial costs involved in clearing their critically–acclaimed album Since I Left You, which used more than 3500 samples. Some of the legal hurdles in clearing an album filled with so many samples can be ‘more trouble than it’s almost worth’, [producer Darren] Seltmann says. ‘We had one track that might have had three or four songwriting samples credited, and they each requested 50 percent [of the royalties]’, he says. ‘We gave up on the prospect of making money from this album before it was even released’. (Klein 2002) Sampling is now less frequently used in commercial music even though the technology has made it easier and cheaper (Mazzone 2012). Unauthorised sampling remains popular in music freely released to the public (mixtapes), but artists still risk being sued. A high profile example is the $10 million lawsuit filed by Lord Finesse against Mac Miller, which was settled for an undisclosed amount (Kaufman 2013). 124 INTELLECTUAL PROPERTY ARRANGEMENTS On the other hand, ‘transformation’ can be an elusive concept since there is a continuum of variations in a work, and no clear boundary about when the use of the original material is genuinely transformative. For example, additions of atypical instrumentation, a prelude with dialogue, and changes in tempo certainly may vary an original score, but not be seen as genuinely ‘transformative’ — so mere claims that a work is transformative is not sufficient to waive copyright protection. There is already a policy solution to this quandary. Fair use provisions (chapter 6) provide a potentially proportionate measure to allow transformative uses of copyright material. Calls to extend copyright further Some participants in this inquiry highlighted the absence of copyright for film directors and the protection of Indigenous culture, and have called for the scope of copyright to be expanded. Is there a case for extending copyright to film directors? The Australian Directors Guild (sub. 10, sub. DR185) noted Australia’s copyright arrangements consider the producer of a film to be its ‘maker’, and thus copyright vests with producers and not directors. The Guild argued Australia’s position was anomalous compared to arrangements in Europe, South America and Hong Kong, stating: In a nutshell, and as stated previously Australian screen directors are denied any ongoing return in the films and television they make because of an out dated and unfair interpretation of Australia’s copyright laws. For close to 50 years, the directors of Australian film and television have been denied any meaningful ‘ownership’ of the films they make. (sub. DR185. p. 4) The Guild made three principal arguments for extending copyright to directors: overseas directors have copyright so Australian directors should too; directors do not earn very much (with half earning less than $25,000 per year); and films are an important part of Australia’s creative sector (sub. DR185, p. 5). While the Commission recognises that directors play an important role in film-making, applying the Commission’s framework, there is not a clear case for extending copyright protection to this group. As set out in chapter 2, Australia’s IP arrangements, including copyright, are designed to correct for any under provision of inventive or creative works that would occur in the absence of such rights being afforded. The Commission has not received any evidence that a lack of copyright protection results in an under-provision of directing services in Australia. Further, there is an important distinction between the investment made by a director and that of the producer. Directors (as distinct from director producers) do not appear to invest in a film in the same way as a producer, or the way an author does in the preparation of a script. Notwithstanding the critical and creative role directors play in film making, their economic relationship is more akin to employees in the film making process, providing a COPYRIGHT TERM AND SCOPE 125 service that is remunerated through contractual agreements with producers or film studios. This is not dissimilar to the relationship between an academic employed by a university, whereby the copyright or patent rights afforded their written work or inventions reside with the employer. Copyright and indigenous cultural protection Select participants pointed to what they consider deficiencies in IP arrangements for protecting Indigenous cultural and intellectual property (ICIP) (National Tertiary Education Union (sub. 24), the Australia Council for the Arts (sub. DR553), and the Australian Institute of Aboriginal and Torres Strait Islander Studies (sub. DR583)). Advocates for stronger protection of ICIP contend that the current requirements are inconsistent with the obligations, kinship requirements, and cultural norms and practices that characterise many Indigenous communities. For example, the Australia Council for the Arts stated: Currently there is no legal right for Indigenous communities to control, maintain and protect their traditional cultural expression, which is primarily communal. This includes songs, stories, dances and cultural knowledge, which is passed down through the generations, but may fall outside the current intellectual property framework as it is not in material form, and may be considered to be in the public domain. Indigenous artists and communities need their cultural and intellectual property rights to be upheld so they can continue and maintain the integrity of their culture and protect it from misappropriation. (sub. DR553, p. 14) Many advocates for stronger protection of ICIP argue such concerns should be addressed by amending copyright laws or providing new sui generis rights. Australia is not the only country where Indigenous communities have raised concerns over the protection of ICIP. Others include New Zealand, Canada, the United States, and countries in Africa, South America and the Pacific. At a high level, Aboriginal and Torres Strait Islander Australians and their creative output have the full protection of Australian copyright law. Copyright protects literary, musical, artistic and dramatic works produced by Indigenous and non-Indigenous Australians alike. However, this belies the deep complexity of this issue. As noted by WIPO, traditional knowledge and traditional cultural expressions are: … generally regarded by conventional IP systems as being in the public domain and so free for anyone to use. Indigenous peoples, local communities and many States argue this opens up traditional knowledge and traditional cultural expressions to unwanted misappropriation and misuse. (WIPO 2016i) The challenge lies in the inherent conflict between copyright law and traditional knowledge and culture. Much traditional knowledge and culture fails to meet the requirements for copyright protection, including the requirement that it have an identifiable 126 INTELLECTUAL PROPERTY ARRANGEMENTS author and be reduced to material form. Moreover, copyright protects only the expression of ideas (such as a particular dot painting), rather than ideas themselves (the use of dots as a traditional painting style), and thus does not prevent someone from producing works in a style others might wish to prevent. And the time-limited nature of copyright, which sees works become public domain eventually, conflicts with the long-term nature of cultural expression. In the Commission’s view, expanding Australia’s IP arrangements to protect traditional knowledge and culture would not be consistent with the framework established in chapter 2. Doing so would depart from the fundamental ‘ideas/expression’ dichotomy at the heart of copyright, and would see perpetual protection granted over artistic styles and ideas. Any new sui generis rights would likely lead to similar outcomes. That is not to say traditional knowledge does not warrant preservation and respect in its handling. As discussed in chapter 2, reforms to IP arrangements alone are unlikely to be the most appropriate way to meet the broader goals of Indigenous cultural preservation raised in the inquiry. 4.4 The term of copyright protection Copyright term is the period over which rights holders control how their work is used. Once copyright expires, others can copy and use creative works without permission or compensating rights holders. • As in other aspects of IP arrangements, copyright term must strike a balance. Ideally, term would be set at a level that encourages creation without unduly constraining access to creative works. If the term of copyright is too short, creators and rights holders will have difficulties recouping their development costs and this may reduce their incentive to create works. However, an excessively long period of protection has the potential to harm consumers because the marginal costs of reproducing the content are zero (or close to it). • Even many years into the future, some consumers value the output above zero. • After a relatively short period of time, further returns make little or no difference to the incentives to create. Consequently, after a certain period, the benefits of positive prices in creating incentives to supply are less than the benefits to consumers. The degree to which this is a problem depends on conduct by the rights holder. • They may reduce prices in recognition of the lower inherent value of older works, which will then commensurately reduce the consumer costs of extended exclusivity rights. • As discussed above, they may no longer supply copyright–protected works on the basis that old material is a substitute for new material and that this may somewhat curtail COPYRIGHT TERM AND SCOPE 127 revenues from new works (or simply withhold works because not much is at stake for them in making them available). This induces potentially significant welfare losses for those consumers who do not consider the material to have a close substitute (such as a vintage computer game enthusiast). Some participants argued the Commission ‘can’t have it both ways’, and that if copyright material has a short commercial life, then there are no negative consequences from excessively long protection. For example, the Australian Copyright Council argued: If it is the case that a copyright has a limited commercial applicability, then it doesn’t matter if it’s tied up in perpetuity … because nobody’s going to want to use it after the first five years. (Australian Copyright Council, trans., p.785) But this argument mistakenly equates the commercial value to the rights holder (their ability to earn an income) with the economic value a work may have to the community. As this chapter has demonstrated, even in cases where a work no longer has a commercial market, there may still be users who wish to access and use those works. Non-commercial, orphan and commercially unavailable works are all examples where community use is foregone because of copyright protection. Australia’s copyright term is excessive Australia’s copyright term provides protection for the author’s life plus 70 years. To provide a concrete example, a new work produced in 2016 by a 35 year old author who lives until 85 years will be subject to protection until 2136. Some rights holders claim the existing copyright term is necessary as ‘anything shorter than [author’s life plus 70 years] has often proven to be too short to recoup on those investments’ (International Confederation of Music Publishers, sub. 32, p. 9). Some have gone further and advocated term be further extended (Bernaski 2014). However, providing financial incentives so far into the future has little influence on today’s decision to create or produce. For example, the addition of twenty years of protection many years in the future, such as occurred when Australia increased term from life plus 50 years to life plus 70 years (a requirement introduced with the Australia-United States Free Trade Agreement) only increases average revenue by 0.33 per cent (Akerlof et al. 2002). Such a small increase in revenue ‘offers at most a very small additional incentive for an economically minded author of a new work’ (Akerlof et al. 2002, p. 2). But it brings a long tail of costly access to commercially unavailable or orphan works for users. Even those who stand to gain from longer copyright protection agree that the extension of copyright from life plus 50 to life plus 70 has done little, if anything, to incentivise additional creative works. We have never asserted that there was any enormous benefit to Australian writers and we have certainly never asserted that the additional term would incentivise writers to create more works. 128 INTELLECTUAL PROPERTY ARRANGEMENTS I, personally, don’t believe for one minute that the extension of term is an incentive to create a new work. (APRA AMCOSS, trans., p. 317) And among countries that are significant producers of copyright material, there is scepticism about the term of copyright applying 70 years posthumously. Comparing the main economic benefits and costs of the CTEA [The Copyright Term Extension Act of 1998], it is difficult to understand term extension for both existing and new works as an efficiency-enhancing measure. Term extension in existing works provides no additional incentive to create new works and imposes several kinds of additional costs. Term extension for new works induces new costs and benefits that are too small in present-value terms to have much economic effect. As a policy to promote consumer welfare, the CTEA fares even worse, given the large transfer of resources from consumers to copyright holders. (Akerlof et al. 2002, p. 15) (United States) In conclusion, the [Gowers] Review finds the arguments in favour of term extension unconvincing. The evidence suggests that extending the term of protection for sound recordings or performers’ rights prospectively would not increase the incentives to invest, would not increase the number of works created or made available, and would negatively impact upon consumers and industry. Furthermore, by increasing the period of protection, future creators would have to wait an additional length of time to build upon past works to create new products and those wishing to revive protected but forgotten material would be unable to do so for a longer period of time. (Gowers 2006, p. 56) (United Kingdom) Extensions of term where works are still commercially available means consumers can expect to pay higher prices for longer. For example, one study suggests major publishers charge around 40 per cent more for books in copyright than books (out of copyright) in the public domain (Heald 2008). When considered from a community–wide perspective, these costs can be significant (Concept Economics 2009; Dee 2004). For example, it is estimated that the obligations in the Australia-United States Free Trade Agreement to extend copyright from life plus 50 years to life plus 70 years resulted in Australian users paying an additional $88 million per year — after accounting for the extra revenue accruing to domestic rights holders (Dee 2004). A similar obligation on New Zealand as a result of the Trans–Pacific Partnership was estimated to cost $55 million per year (Concept Economics 2009). … [the] public is not getting a good deal in return for its generous grant of above-incentive rights. Current terms are neither optimized to maximize continued investment in existing works nor to recognise and reward creators, and they cause knowledge and culture to languish underused, or even vanish altogether. (Giblin 2015, p. 21) How long is long enough? While it is increasingly accepted that the optimal, or ideal copyright term has been exceeded, there is far more contention about where the optimal term might lie. Evidence COPYRIGHT TERM AND SCOPE 129 suggests that the vast majority of works do not make commercial returns from copyright beyond their first couple of years on the market. 4 For example, the ABS estimates: • the average commercial life of music is between 2 and 5 years. Around 70 per cent of musical originals provide a return in the first 2 years, with the remaining 30 per cent providing a return fairly evenly over the next 3 years • most original visual artistic works no longer yield an income within 2 years, but the distribution is highly skewed and a small proportion of works receive income over a longer period. These works represent the majority of income received • the average commercial life of a film is between 3.3 and 6 years. This estimate is benchmarked against similar international markets including the United Kingdom, Europe and the United States. Very few films make money in their sixth year • literary works provide returns for between 1.4 and 5 years on average. Three quarters of original titles are retired after one year and by 2 years, 90 per cent of originals are out of print (ABS 2015b, pp. 374–376). This latter result is consistent with the Commission’s own analysis of the top 5000 books sold in Australia in the 12 months up to 31 May 2016 (detailed in appendix E). The results show that less than 2 per cent of the titles sold were published prior to the year 2000, and only 12 per cent were published more than 5 years ago. A commercial life of a couple of years suggests most works are granted protection for decades longer than necessary. Some submissions (including ACCC, sub. 35; NSW Department of Justice, sub. 39; Telstra, sub. 76; Electronic Frontiers Australia, sub. 114; Moir, sub. 137) echoed this point. For example, the ACCC stated it: … is concerned that the extent of current copyright protections may go beyond what is necessary to provide an incentive to create and disseminate original copyright materials, and that they may be providing excessive protections to holders of IP rights. For example, the extension of copyright protections from 50 to 70 years (following the death of the author or first date of performance/publication) are unlikely to have produced a commensurate incremental value of increased copyright works and may have deterred valuable use of older copyright works. (sub. 35, p. 11) While, for the overwhelming majority of creative works, a far shorter term of copyright would more than cover the term of their commercial life, the Commission recognises that a select few very successful works have commercial lives well beyond a few years — a point also made by participants. But it is not possible to define terms specific to each given work and thereby capturing the few outliers, an ‘optimal’ term is a period that, on average, creates reasonable incentives for creation, while avoiding the consumer losses associated with overly lengthy or permanent exclusivity. The situation is conceptually similar to that applying to patents. 4 Some authors also received payments from the Education and Public Lending Rights schemes over a longer period of time. These payments are made in recognition that income could be lost through the use of their books in public and educational libraries. 130 INTELLECTUAL PROPERTY ARRANGEMENTS Numerous studies have attempted to estimate the ‘optimal’ duration of copyright protection. Landes and Posner (2002) argue a term of around 25 years enables rights holders to generate revenue comparable to what they would receive in perpetuity (in present value terms), without imposing onerous costs on consumers and suggests that a term of around 25 years is sufficient to incentivise creative effort. 5 In addition, any estimate of optimal term duration must make assumptions about the pattern of demand for the works over time — a difficult task. The truly ‘optimal’ period may accordingly be more or less than 25 years after creation, but 70 years after death is completely implausible. Pollock (2007) uses an alternative, more comprehensive, methodology to estimate the optimal length of copyright protection. Using data from the US copyright register, Pollock estimates the additional number of creative works produced when copyright protection is increased and the cost that is paid by the community. His work suggests a copyright term around 15 years after creation balances the benefits and costs of the system. 6 Australia has no unilateral capacity to alter copyright terms — reflecting this, the Commission has made no recommendation to reduce term. However, as discussed in the final section of this chapter, the Australian Government should work with other governments to advance a broader set of multilateral reforms. FINDING 4.1 The scope and term of copyright protection in Australia has expanded over time, often with no transparent evidence-based analysis, and is now skewed too far in favour of copyright holders. While a single optimal copyright term is arguably elusive, it is likely to be considerably less than 70 years after death. Grandfathering extensions is costly Previous extensions to copyright term have been applied retrospectively to all existing copyright-protected works, rather than only to works produced after the extension. While such an approach may be administratively simple, retrospective extensions to the term of 5 Landes and Posner (2002) use a discount rate of 10 per cent and show that the present value of returns lasting 25 years is 90 per cent of the present value of infinite returns. As the discount rate varies, so too does the calculation of an optimal term. 6 Pollock undertook sensitivity analysis to test the robustness of his results. A probability distribution indicates that the modal value for term is 11 years, the median is 15 years and the 99th percentile is 38 years (Pollock 2007). COPYRIGHT TERM AND SCOPE 131 protection cannot, by definition, increase the incentive to create works already in existence. As Buccafusco and Heald remarked: The incentive-to-create rationale fails entirely, however, in the case of extending the copyright term for already existing books, music, and movies. The extension of protection for The Sun Also Rises does not increase the incentives for Hemingway to produce more or better work. He is, after all, dead. (2012, p. 3) Instead, such extensions are a windfall gain to creators and rights holders at the expense of consumers and the broader society (Buccafusco and Heald 2012). For example, the major beneficiaries of previous retrospective term extensions were creators’ estates and corporate rights holders. A Swedish study found that the share of royalties flowing to composers’ estates increased — from 2.4 per cent in 1995 to 14.1 per cent in 2006 — as a result of a 20 year extension in copyright term (Bently et al. 2008). These windfall gains do not have any positive flow–on effects for the wider community in terms of the creation of new works. The grandfathering arrangements cannot readily be addressed now, but provide a lesson about the need for careful analysis of changes to copyright law and the costs they may impose on consumers. As discussed in chapter 17, the Commission had identified retrospective extensions of term as a ‘no-go’ area that should be avoided in international agreements. Perpetual term — the case of unpublished works Unpublished materials — including diaries, letters, journals, recipes and sketches — make up a large part of libraries’ and archives’ collections. The Australian Digital Alliance highlighted the amount of unpublished works. … in August 2015, the ALCC [Australian Libraries Copyright Committee] conducted an informal survey of 14 Australian universities (over 20 collections covering roughly 1/3 of the university sector) to establish the incidence of unpublished works in their collections. Cumulatively, the universities surveyed reported that their collections included over 12.9 km of unpublished works, or approximately 103,904,000 pages. (sub. 108, p. 8) But much of this material cannot easily be preserved or presented to the public. Unlike published works, Australia’s copyright system protects many unpublished works in perpetuity, an anomaly compared to the UK, Canada, New Zealand, the US and much of the European Union (Professional Historians of NSW & ACT, sub. 3). The author or their heirs hold copyright in unpublished works and their permission is generally required before the unpublished material can be digitised or used in public exhibitions. Many submissions (including Prof. Kathy Bowrey, sub. 86; Professional Historians of NSW & ACT, sub. 3; the Australian Digital Alliance, sub. 108; the Australian Library Copyright Committee, sub. 125) noted that tracking down rights holders of unpublished 132 INTELLECTUAL PROPERTY ARRANGEMENTS works is a complex and costly barrier to displaying, digitising and publishing historical materials and conducting research. The Copyright Act makes provisions for libraries and archives to use unpublished works in limited circumstances. 7 However, such provisions do not appear to provide sufficient scope to enable full use of unpublished works. In its current state Australian copyright law places significant restrictions and compliance costs on libraries and archives. The current exceptions available to libraries and archives are too rigid, complex and difficult to apply, creating significant inefficiencies for the sector and presenting a barrier to the dissemination of knowledge. To add to the complexity, these exceptions are often excluded by contracts and technologies. As a result, many of the works in our national collections remain locked behind complex laws, unable to be used. (Australian Libraries Copyright Committee, sub. 125, pp. 2–3) The NSW Department of Justice highlighted the problem that (particularly local) governments face in making information public when it includes unpublished reports, databases or other copyright–protected material (sub. 39). Major repositories for unpublished works, such as the Australian War Memorial, the National Library, the National Archives and the Art Gallery of New South Wales also stressed the importance of making unpublished items available to the public. In December 2015, the Australian Government released a proposal to introduce a time-limited period of protection for unpublished works, essentially harmonising the treatment of published and unpublished works. Unpublished works with a known author would be protected for the author’s life plus 70 years. Where an author is not known, copyright would last 70 years from the year the work was made. This proposal is similar to solutions put forward by participants (Australian Digital Alliance, sub. 108). No case exists for unlimited copyright protection for unpublished works and the appropriate term is almost certainly less than the current term of protection for published works. But given Australia’s inability to unilaterally decrease term for all copyright material, the Australian Government’s proposal represents a reasonable compromise. 4.5 Towards more globally efficient copyright arrangements As the Commission has noted in this chapter and elsewhere in the report, Australia’s international commitments substantially constrain the scope to change the copyright system. International agreements drafted in the 1880s are not fit to regulate how 7 Under s. 51(1) of the Copyright Act, libraries can photocopy unpublished material held in their collections for research purposes without authorisation if the author died more than fifty years ago. If this condition is not met, libraries need the permission of the rights holder before using the unpublished work. COPYRIGHT TERM AND SCOPE 133 governments manage their copyright arrangements in the 21st Century. To restore balance to the system Australia will need to work with other countries. The United States is one country Australia should look to work with in achieving better global copyright outcomes. The US Register of Copyrights has recently highlighted a range of issues the US Congress needs to wrestle with, including the scope and duration of the exclusive rights, licensing schemes, enforcement mechanisms, exceptions and limitations, and registration (Pallante 2013). These are also issues facing Australia and many other countries. One idea put forward is for the last 20 years of copyright protection (the period from life plus 50 years to life plus 70 years) to be contingent on registration. … perhaps the law could shift the burden of the last twenty years from the user to the copyright owner, so that at least in some instances, copyright owners would have to assert their continued interest in exploiting the work by registering with the Copyright Office in a timely manner. And if they did not, the works would enter the public domain. (Pallante 2013, p. 23) Participants in this inquiry have similarly put forward suggestions for ways copyright can be made more workable, while retaining its intended purpose. For example, Giblin and Weatherall (sub. DR524, p. 405) raise the issue of copyright reverting to authors, after having been licenced to an intermediary, after a period of time, such as 15 to 25 years. The constraints of international agreements are not unique to copyright. Chapter 18 discusses the broader IP multilateral agenda. 134 INTELLECTUAL PROPERTY ARRANGEMENTS 5 Copyright use and licensing Key points • Copyright licensing is common in Australia. It enhances the copyright system’s efficiency, facilitates creative works being brought to market and improves consumer access. • The modern era of digital consumption and access to copyright-protected works is changing how consumers access copyright material. – While consumers can easily and inexpensively access a vast amount of works, contracts and licences can be used to erode user rights. – Licence conditions that remove user rights to exceptions should be unenforceable. • Geoblocking results in Australians paying higher prices (often for a lesser or later service) than consumers overseas. – Consumer rights to circumvent geoblocks should be enshrined in the Copyright Act. • The Commission re-examined the price effects of parallel importation restrictions on books. – While price differences have narrowed, opportunities still exist for bookstores to import equivalent books from overseas at materially lower prices. – The Australian Government should immediately repeal the remaining parallel import restrictions on copyright-protected works. Additional transitional arrangements are not needed given the positive confluence of efficiencies made by the Australian publishing industry and broader economic circumstances. • Copyright collecting societies issue collective licences, collect payments from users, and distribute royalties to rights holders. Such collective licensing has merit to the extent it reduces transaction costs, especially for high volume, low value transactions. • But the ability to collectively license can also give rise to market power. Australia’s collecting societies are therefore governed by a voluntary code of conduct and (while lesser known) subject to Australian Competition and Consumer Commission (ACCC) scrutiny. Participants raised concerns about the efficiency and effectiveness of the current code of conduct, which does not appear as robust as international comparators. • The ACCC should review the code of conduct to ensure the code reflects contemporary international best practice, is being followed and whether it should be made mandatory. • Recent proposals to improve the administration of the education statutory licensing scheme are likely to result in benefits to rights holders and users. As discussed throughout this report, owners of intellectual property (IP) rights often do not bring a product or idea to market themselves, but rely on others to do so on their behalf. Creators can permanently assign their rights to a third party, but more commonly license their rights to third parties, such as publishers, record companies, movie studios and dedicated collection agencies. COPYRIGHT USE AND LICENSING 135 Licensing allows intermediaries to exercise some or all of the exclusive rights held by a copyright owner. Licences may limit: • the exclusive rights that the licensee may exercise. For example, an author may grant a limited licence for someone to translate their work into another language, or to adapt the work into a film screenplay • the geographic location that the licensee may operate in. For example, the publisher of a literary work in one country may licence a publisher in another country to produce and sell a version of the work only in that country. A licensee will generally pay a fee or royalty in return for exercising a rights holder’s exclusive rights. A royalty may be a single payment, or an ongoing proportion of the profits earned through the use of the work, or a combination of both. Licensing has the potential to improve the efficiency of the copyright system by allowing greater specialisation within the IP supply chain. This chapter examines five specific issues that affect the use and licensing of copyright material. • Consumer access to content, including the issue of contracting out of copyright exceptions and the use of technological protection measures (TPMs) (section 5.1). • Geoblocking (section 5.2). • Restrictions on the parallel importation of books (section 5.3). • Collective licensing and collecting societies (section 5.4). • Australia’s statutory licensing schemes (section 5.5). 5.1 Consumer access to content The way consumers access copyright content is changing Wide-scale access to the Internet has fundamentally changed the distribution model for copyright-protected works, with implications for both traditional intermediaries and consumer interests. Prior to the advent of the Internet, copyright works were generally licenced from creators — such as authors, musicians and artists — to intermediaries — such as publishers, record companies and film studios. Creative works were then embedded in physical products, including books and magazines; records, cassettes and CDs; and VHS tapes and DVDs. Once purchased, consumers were able to watch, listen or read them as often and when they wished. The Internet has led to the development of new business models such as online music and television streaming services. Consumers can easily and inexpensively access a vast 136 INTELLECTUAL PROPERTY ARRANGEMENTS amount of copyright material with more choice in how they do so. They can purchase physical copies, obtain licences for digital goods, or subscribe to streaming services. Digital distribution allows for easier unbundling of goods — for example, consumers can purchase individual songs instead of an entire album. Cloud and subscription services allow consumers to pay an annual or monthly fee to access a vast catalogue of works, rather than purchase individual works. Australian consumers have largely embraced digital access to creative content. For example, Australian consumers are now more likely to digitally access music than purchase physical copies (Bowker 2012; Hopewell 2015) (figure 5.1). Figure 5.1 Wholesale sales of recorded music in Australia 160 Physical Digital 140 120 millions 100 80 60 40 20 0 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 Source: ARIA (2014). The transition to a digital marketplace has raised problems While digital markets have provided many benefits to consumers, they are also accompanied by some restrictions not present in the physical transactions of past. Given the nature of licences, when consumers purchase physical copies of works, the rights of creators over the physical copy are ‘exhausted’ under the ‘first sale doctrine’, and consumers are able to freely deal with their physical goods as they see fit — to lend or sell the work to others, and to keep their copy as long as they want. But when consumers enter a licensing arrangement to access digital content, some of the flexibility that consumers have over the control of physical copies is forgone. COPYRIGHT USE AND LICENSING 137 Some have raised concerns with the limits that apply under digital access arrangements. One concern is that the absence of second hand markets for digital versions of works reduces consumer options and limits competition (Suzor and Dootson 2013). However, there are two main reasons why this may not be a substantial issue. First, consumers can still purchase many works in physical forms — with the additional rights that this bestows. Second, it is difficult to see how second hand digital markets would function without replicating and thereby eroding the first sale market. In the digital market, a second hand copy — were one to exist — would be identical to the original. But in the case of physical books (and other media), the purchaser in the second hand market is not receiving a pristine copy, and accordingly, the prices are usually less than that in the original transaction. In effect, the wear and tear of physical copies of works creates multiple new product lines, differentiated by the degree of wear, thus allowing multiple prices. The digital world disrupts the differentiation created through second hand markets. Another concern technology imposes is the potential, over time, for consumers to lose access to works for which they have obtained digital licences. Under copyright law, the underlying work remains the property of the rights holder and they can choose to remove the work at any time. For example, Amazon removed e-books written by George Orwell from its catalogue (and consumers’ e-book readers) because of a dispute with a rights holder. Similarly, in 2014 musician Taylor Swift removed her music from online music streaming service Spotify (Engel 2014; Stone 2009). While Amazon compensated consumers, a lack of clarity about consumer rights led some to believe they had purchased the books. These examples highlight that consumers are not always well-informed about the nature of the agreements they enter into when consuming digital copyright products (especially under ongoing subscription payment models). In its submission, CHOICE (sub. 26) suggested: Where applicable, retailers must make it clear to consumers that they are not purchasing products, but rather rights to access. This includes removing ‘buy’ or ‘purchase’ buttons from online stores. (p. 14) However, a regulatory requirement for this would be both impractical (given the global nature of online firms) and premature. Any such requirement would likely be unenforceable for foreign sites. Moreover, apart from the few notable exceptions above, few cases exist where content has been withdrawn, and there is little evidence consumers are unaware of the consequences of purchasing digital rather than physical goods. To the extent any residual problems exist, there may be superior non-regulatory approaches. For example, a recent review by a US Government taskforce into digital consumption of copyright works recommended rights holders and consumer groups work together to develop better information for consumers on the difference between purchasing content to own, and accessing content under a different licensing model (Department of Commerce Internet Policy Task Force 2016, p. 69). The taskforce also recommended against extending the first sale doctrine to digital transmissions of copyright works, 138 INTELLECTUAL PROPERTY ARRANGEMENTS concluding it was too early to warrant government intervention as markets and licensing arrangements are still evolving. The Commission agrees. Given the changing nature of digital markets, the Australian Competition and Consumer Commission (ACCC) should continue to monitor developments for any policy-relevant effects on consumers. Contracting out and TPMs can harm consumers Statutory exceptions play a valuable role in the copyright system — they help balance incentives for creators to invest in new works with the benefits to users from accessing and using those works (chapter 6). Increasingly, however, licences include conditions that prevent users from benefiting from the current statutory exceptions, such as the ability to make back-up copies of works they own. Licence conditions preventing the use of exceptions may also be supplemented with TPMs, a broad term characterising any technology controlling access to, or preventing certain uses of, copyright-protected works. Such technology can prevent a user from doing something that would ordinarily be permitted by an exception. Contracting out can reduce consumer rights … Several inquiry participants commented on the practice of contracting out. SBS (sub. 112) noted that in practice, it was difficult or impossible to negotiate contracts with suppliers to enable the use of fair dealing provisions, particularly with major international subscribers who are less accustomed to Australian law. The Australian Libraries Copyright Committee highlighted that the ‘contracting out’ of copyright exceptions significantly impacted libraries’ usual operations. … the National Library of Australia noted that 79% of digital products (e-books, databases, aggregator licences) purchased by the Library prohibited document supply. The law in Australia is currently unclear as to whether copyright or contract takes precedence in such cases. Due to this uncertainty, lack of expertise among library and archives officers and the need to maintain relationships with publishers, industry practice is to follow the licences. This is an endemic problem within the library sector, and significantly undermines the usability of electronic materials just as publishers and libraries are seeking to move towards digital. (sub. 125, p. 15) The restrictions faced by consumers from contracting out arrangements can be accentuated by TPMs, as the Australian Digital Alliance observed: The problems created for consumers by the combined use of restrictive licensing and TPMs [technological protection measures] are demonstrated well by the example of the “closed environment” created by most digital music services. When consumers purchase, or rather license, music from an online music service they are routinely restricted (by both license and TPM) from transferring that music to devices that are not licensed for that particular service (eg non-Apple products for iTunes). This is in direct contravention of the individual’s rights under s.109A of the Copyright Act to transfer legally acquired music to other devices for private and domestic use. Similarly, few music services permit remixing or public performance of their COPYRIGHT USE AND LICENSING 139 products, even where it would be permitted under a fair dealing or fair use exception. (sub. DR578, p. 16) The Australian Law Reform Commission (ALRC) examined contracting out in its copyright inquiry, finding evidence that digital licences can conflict with copyright exceptions, and recommended amendments preventing the fair dealing exceptions from being overridden by contracts (ALRC 2014). Other reviews have made similar recommendations, including the Copyright Law Review Committee (CLRC 2002) and the United Kingdom’s Hargreaves Review (Hargreaves 2011). The Commission considers that there are good grounds for preventing copyright exceptions from being overridden by contracts. The use of contracts to override exceptions effectively enables the rights holder to rewrite the limits that the law has set on the extent of the right conferred by copyright. 1 Copyright exceptions are an essential component of the balance between incentives to create new works, and the benefits to consumers from those works. The use of contracts and TPMs to prevent access to works has the potential to restrict uses that have been expressly permitted by parliament, reduce competition and efficiency, and increase the return to creators over and above what is necessary to incentivise their creation. Limitations on contracting occur in other areas of Australian law. Numerous cases exist where the Australian Parliament has determined that, notwithstanding whatever benefits might be available to individuals if they do give up rights, such agreements will not be permissible. Such arrangements are particularly prevalent in circumstances where consumers may not have the bargaining power to efficiently negotiate contracts, or where contracts are offered to consumers unilaterally. For example, the Australian Consumer Law, the National Employment Standards and residential tenancy conditions are all examples where consumer, employee and tenant protections cannot be overridden by contract. Some participants acknowledged the need to prevent contracting out, but advocated for limited application. For example, the Coalition of Major Professional & Participation Sports (COMPPS) stated it did: … not oppose limitations on contracting out being applied to the exceptions for libraries and archives. … However, COMPPS does not support the introduction of any limitations on contracting out of the fair dealing exceptions in a broader context. (sub. DR312, p. 17) Similarly, the University of Melbourne (sub. 100) recommended changes to the Copyright Act 1968 (Cth) (Copyright Act) to restrict contracting out as it relates to specific exceptions for libraries and archives. However, the Copyright Advisory Group to the COAG Education Council (sub. 97) highlighted the problem with creating a ‘hierarchy’ of exceptions, some of which might be protected from contracting out, and others not. In any event, it is not clear to the 1 The Hargreaves Review reached a similar conclusion (Hargreaves 2011, p. 51). 140 INTELLECTUAL PROPERTY ARRANGEMENTS Commission why only some exceptions are meritorious of protection, when exceptions are designed to ensure the copyright system is balanced between all users and producers. Other participants shared this view. If contracts are permitted to over-ride statutory exceptions (as is currently the case) this will erode the benefits gains from such a flexible exception. To ensure that “fair use” and/or “fair dealing” operates in the manner in which it is intended we recommend that all voluntary licences are negotiated or entered into in the context of what is “fair” ie: no restrictions placed on fair use/fair dealing. (The University of Sydney sub. 104, p. 12) These exceptions serve an important social purpose in promoting freedom of expression and information. Contractual relationships should not be able to interfere with rights that are essential in order to better protect the public interest. (FreeTV Australia, sub. 129, p. 8) Given the careful balance copyright arrangements seek to strike, and the Commission’s view that these arrangements in Australia remain tilted in favour of rights holders, contract arrangements should not be able to override the limited protections offered by copyright exceptions. The Commission does not see any reason why this should only apply to libraries and archives, and considers that a broader prohibition on contracting out is warranted. … and TPMs are poorly targeted Concerns are arguably more pronounced in relation to TPMs than contracting out. As the Australian Digital Alliance (sub. 108) highlighted, there is currently no requirement that TPMs provide protection against copyright infringement for protection under the Copyright Act to apply. Therefore not only can TPMs be used to prevent uses permitted under copyright exceptions, they can also be used to prevent use of public domain material and (arguably in Australia at least) to limit products which have little or no link to copyright. While circumvention of a TPM is permitted under the Copyright Act, it is still illegal to supply technologies or services to enable such uses. This leads to the unusual situation where individuals and institutions wishing to legitimately crack a TPM (such as a library wishing to make a preservation copy) will have to create the technologies required to do so themselves. As the Australian Digital Alliance observed: This is clearly implausible, and creates an environment in which users are required to access illegally created products to conduct legal acts. (sub. DR578, p. 15) In its report, the ALRC underscored the importance of concurrent reform of contracting out arrangements and TPMs. If limitations on contracting out are implemented, consistent amendments to TPM provisions may be justified. That is, there may be little point in restricting contracting out of exceptions if TPMs can be used unilaterally by copyright owners to achieve the same effect. (2014a, p. 457) COPYRIGHT USE AND LICENSING 141 RECOMMENDATION 5.1 The Australian Government should amend the Copyright Act 1968 (Cth) to: • make unenforceable any part of an agreement restricting or preventing a use of copyright material that is permitted by a copyright exception • permit consumers to circumvent technological protection measures for legitimate uses of copyright material. 5.2 Geoblocking What is geoblocking and why does it matter? Geoblocking is the practice of restricting a consumer’s access to websites and digital goods and services within their ‘home market’. Geoblocking can be implemented via a range of technologies including Internet addresses, credit card numbers and other means of electronic identification (Alexander et al., sub. DR505, p. 6). Geoblocking enables rights holders and intermediaries to segment the Internet into different markets and charge different prices (or offer different services) to consumers based on their location. This facilitates geography based price discrimination. While the original purpose of copyright was to prevent copying, geoblocking allows rights holders to control copying and the distribution of copyright material. Copyright, exclusive licensing and geoblocking can work together to further strengthen the ability of rights holders and their intermediaries to control distribution and thereby price discriminate. The evidence supports the finding that the use of geoblocking technology is widely imposed on Australian consumers and it frequently results in Australian consumers being offered a lower level of digital service (such as a more limited music or TV streaming catalogue) at a higher price than in overseas markets. In its submission, CHOICE commented on the prevalence of what has become known as the ‘Australia Tax’. Australians have long been subject to the ‘Australia Tax’ — international price discrimination that leads to higher prices for a variety of goods and services in Australia, disadvantaging Australian consumers and businesses. Often this price discrimination is supported through the use of geoblocks, digital barriers that prevent consumers from particular geographic regions from transacting with online sellers based in different regions. In 2012, CHOICE research found that on average Australians paid 50% more for a variety of digital goods in comparison with consumers in the United States. (sub. 26, p. 17) Overseas Internet intermediaries — websites selling physical goods and firms selling digital downloads or streaming of copyright protected material — are often required, as a condition of their copyright licences, to restrict sales to their own market, and not sell to 142 INTELLECTUAL PROPERTY ARRANGEMENTS foreign consumers. For example, an online streaming service might be required to present Australian consumers a smaller catalogue of content than consumers in other markets. The Australian Parliament’s House of Representatives Standing Committee on Infrastructure and Communications (2013) examined the extent and impact of geoblocking in Australia in its report At What Cost? IT pricing and the Australia tax. Although hampered by a lack of comprehensive data, survey analysis submitted by participants to the inquiry showed systemic price discrimination against Australian consumers across a range of copyright-protected categories. • Professional software: A comparison of more than 150 products showed an average price difference of 50 per cent, with a median price difference of 49 per cent. • Music: Across more than 70 products Australian prices were, on average, 52 per cent more expensive, while the median difference was 67 per cent. • Games: The average price difference across more than 70 games was 84 per cent, while the median difference was 61 per cent. • E-books: Price comparisons of more than 120 e-books books sold in Australia and the United States revealed average price differences of 16 per cent, while the median difference was 13 per cent. Geoblocks impose costs on consumers Many of the arguments that arose in the IT pricing and the Australia tax inquiry were also raised during the course of this inquiry. Consumers and consumer representative groups considered that geoblocking and price discrimination, while legal, were not legitimate business practices, and that the price differentials between markets went well beyond the costs of providing local services and warranties (House of Representatives Standing Committee on Infrastructure and Communications 2013). Participants from the creative sector largely argued geoblocking was a tool to manage markets globally and that it underpinned common business practice. For example, Foxtel stated: Geoblocking and technological protection measures are of critical importance to the sustainability of the Australian media industry. They enable creators to obtain a fair return on their investment. (sub. DR550, p. 2) Some participants from the creative sector went further, and suggested that any attempt to assist or encourage Australians to circumvent geoblocks would irreparably damage local creative industries. The Australian domestic market will be completely undercut by local consumers accessing content offshore. The long term effects will be the weakening of Australian business models. This will have an adverse impact on investment and innovation as investors and content COPYRIGHT USE AND LICENSING 143 creators will not have an incentive to produce and invest in Australian content and business models. This will damage the competitiveness of Australia in the digital economy. (The Australian Recording Industry Association, sub. DR499, p. 7) … it is unlikely that Australian content businesses will have the resources to acquire global distribution rights to international content, as Australia is a relatively small player in the international market and there are significantly larger distribution companies internationally that operate on either a global or a multi-territory basis. This may lead to the elimination of local services and in turn, a dramatic reduction in the production of local Australian television content, given that such content is unlikely to be commissioned on a regular basis by international distributors. (Foxtel, sub. DR550, p. 5) Australia’s free to air networks rely on, and pay for Australian rights for the major US TV shows to generate 25 viewers and advertising. This, in turn, places a requirement on the network to invest in local productions under government licence models. If Australians access all their content offshore, then the local networks will surely wither and die. (The Australian Home Entertainment Distributors Association, trans., p. 290) The Commission notes the arguments made by those supportive of geoblocking are substantially the same as those made in favour of parallel importation restrictions more generally (see below, and chapter 12). That is, in order to ‘support’ production of Australian content, those buying the distribution rights for foreign content must be able to charge higher prices, protected by laws preventing consumers or other businesses from importing content at lower prices. Such arguments typically ignore other factors driving the demand for, and supply of, Australian content. In the audio-visual sector, these include the significant local content requirements imposed on holders of broadcast licenses. Arguments about local content aside, market segmentation that raises prices in Australia is unlikely to be welfare improving from the perspective of the Australian community (the perspective adopted by this inquiry, see chapter 2). This is because Australia is a small country with little impact on the decision to produce content and the income from higher prices typically flows offshore. Consumers would benefit from clarity in the law Given the prevalence of geoblocks, it is unsurprising consumers seek to circumvent them, often through the use of virtual private network services to mask their online location, making it appear as if they are located overseas. As CHOICE identified in its submission, a link exists between the price and availability of digital goods and services in Australia, and copyright infringement (sub. 26, p. 17) (see chapter 19 for issues related to compliance and enforcement). 144 INTELLECTUAL PROPERTY ARRANGEMENTS At present, it is not clear under Australian law whether consumer actions to circumvent geoblocks risk breaching the Copyright Act’s provisions on technological protection measures and access control measures. In its review, the Standing Committee noted ongoing uncertainty about the legal status of circumventing geoblocking technology, recommending: … the Australian Government amend the Copyright Act’s section 10(1) anti-circumvention provisions to clarify and secure consumers’ rights to circumvent technological protection measures that control geographic market segmentation. (House of Representatives Standing Committee on Infrastructure and Communications 2013, p. 108) The Commission considers that there is a strong case for resolving the ambiguity in favour of consumers. The ACCC and Digital Rights Watch both pointed to the benefits for competition with such an approach. Making clear that it is not an infringement under Australian law to circumvent geoblocking technology would remove a significant impediment for consumers to access legitimate content, and thereby promote competition. The ACCC notes, however, that other restrictions, such as contractual terms and conditions that content providers offer to consumers and/or statutory provisions in jurisdictions outside Australia, would likely continue. As a result, broader market effects would likely take time. (ACCC, sub. DR603, p. 7) Enabling parallel importation by clearly permitting circumvention of geoblocking is likely to encourage rights holders to adopt distribution models that adequately serve the Australian market. Left unchecked, rightsholders are likely to continue with current profit maximisation strategies that restrict competition in distribution channels. (Digital Rights Watch, sub. DR414, p. 10) In the Commission’s view, as a minimum step, the Australian Government should prevent the future possibility that rights holders seek to use ambiguity in the Australian copyright system to prevent consumers’ circumvention of geoblocks. RECOMMENDATION 5.2 The Australian Government should: • amend the Copyright Act 1968 (Cth) to make clear that it is not an infringement for consumers to circumvent geoblocking technology, as recommended in the House of Representatives Standing Committee on Infrastructure and Communications’ report At What Cost? IT pricing and the Australia tax • avoid any international agreements that would prevent or ban consumers from circumventing geoblocking technology. 5.3 Parallel import restrictions on books ‘Parallel imports’ are goods protected by IP rights and produced with the permission of the rights holder overseas, but imported into Australia without permission of the domestic COPYRIGHT USE AND LICENSING 145 rights holder. Contrary to some misunderstanding, ‘parallel imports’ refer to genuine products produced with the authorisation of the relevant rights holder overseas, and are not counterfeit or infringing products. While there are no restrictions on parallel importation of sound recordings, computer software and goods embodying electronic literary or music items, Australia retains parallel import restrictions (PIRs) on books. The Copyright Act grants IP rights holders the ability to control importation of their works, unless an exception applies. Individuals may parallel import books for personal consumption. But only limited exceptions apply for commercial parallel importation of books — when a book first published overseas is not subsequently published in Australia within 30 days, or if an Australian publisher cannot supply a book within 90 days. Prohibiting parallel imports enables IP rights holders to engage in geographic price discrimination — charging different prices for the same good in different countries, or varying the quality of goods supplied in different countries. Price discrimination is only possible if rights holders can prevent consumers, retailers or importers from purchasing those goods from overseas markets directly. Rights holders price discriminate to increase the total returns they earn from their works globally, by charging prices that reflect consumers’ willingness to pay in each country. By raising book prices, PIRs adversely affect Australian consumers with little or no change in the incentives for producing works by authors, notwithstanding claims to the contrary. Numerous reviews over 20 years, including by the Commission, the Australian Senate, the ACCC, and most recently the Harper review of Competition Policy, have recommended that PIRs be repealed (box 5.1). In 2015 the Australian Government (2015a) announced the removal of the restrictions and agreed to progress reform subject to the findings of this inquiry. In submissions to this inquiry, rights holders typically argued the remaining PIRs were not inconsistent with competition policy, because consumers could circumvent the restrictions and parallel import for personal use. The fact that individuals, and not book sellers and book shops in Australia, can parallel import, allows only a modicum of the competitive pressure that would come with the removal of PIRs. In fact, the ACCC, among others, argued there were likely to be benefits from removing the broader restrictions, and consumers likely faced higher search, transaction and delay costs under the current exception for personal importation. 146 INTELLECTUAL PROPERTY ARRANGEMENTS Box 5.1 Previous reviews of parallel import restrictions of books Australia’s parallel import restrictions have been reviewed many times in the past 20 years and most have recommended their removal. In 1989, the Prices Surveillance Authority (PSA) surveyed the recommended retail prices of books in the Australian, UK and US markets. It found Australian prices to be higher on average, and recommended the restrictions be removed, with some limited protection for Australian authors for a period of 10 years. The PSA updated its report in 1995 and concluded that while the 1991 reforms had improved the timeliness of book releases in Australia, prices continued to be higher in Australia than in the UK and US. The ACCC issued reports in 1999 and 2001 that extended the PSA’s price comparisons, and concluded the price of bestselling books was generally higher in Australia than in the US and on par with prices in the UK over a six-and-a-half year period, although prices differed over time. The 2001 report noted that removal of the PIRs on books in New Zealand in 1998 had not caused the collapse in the publishing industry that many had predicted. Both ACCC reports recommended the restrictions be repealed. In 2000, the Intellectual Property and Competition Review Committee examined the restrictions, and began with the presumption that restrictions on competition should be removed unless they result in a net benefit for the community and there are no better alternatives. It concluded the costs of removing the restrictions were likely to be small relative to the gains to Australia, noting the benefits of the restrictions largely flowed to overseas rights holders. The Senate Legal and Constitutional Legislation Committee reviewed the available evidence on the effects of the PIRs in 2001. The majority report recommended legislation to remove the restrictions proceed to the Parliament. The Productivity Commission examined the restrictions in 2009. Price comparison analysis found that, in 2007-08, a selection of around 350 trade books sold in Australia were on average 35 per cent more expensive than in the US. In many cases, the price difference was greater than 50 per cent. The Commission recommended the restrictions be repealed. A 2012 study by Deloitte Access Economics showed a NZ$3.06 price difference in a 100 book title-for-title comparison between New Zealand and Australia, and NZ$3.20 difference in a truncated sample (removing the top and bottom 25 per cent of price differences). Deloitte concluded parallel import restrictions on books in Australia are, on average, resulting in prices between NZ$3.06 and NZ$3.20 higher — or around 10 per cent — than they otherwise would be. The Australian Government’s 2015 Competition Policy Review (Harper Review) also considered the costs and benefits of retaining parallel importation restrictions on books. The Harper Review placed the repeal of parallel import restrictions in its ‘areas for immediate reform’. Sources: ACCC (1999, 2001); Deloitte Access Economics (2012); Harper et al. (2015); IPCRC (2000); PC (2009); SLCLC (2001). Hundreds of authors, publishers and booksellers made submissions, and many appeared at public hearings for this inquiry, arguing the Australian Government should reverse its COPYRIGHT USE AND LICENSING 147 decision to remove PIRs in response to the Harper Review to repeal the PIRs. 2 They claimed that removing the restrictions would see cheap books, especially of Australian authors, ‘flood’ the market to the detriment of Australian publishers, authors and booksellers. Many participants made similar arguments in favour of retaining the restrictions, and pointed to recent competitive advances made by local publishers. Several participants argued: • the average selling price of a book in Australia had fallen in recent years. For example, the Australian Publishers Association argued the average selling price of books in Australia had ‘fallen in real terms by over a third in the last decade’ (sub. DR435, p. 3) • the number of individual titles sold in Australia had increased by 120 per cent, from 252 000 in 2003 to around 558 000 in 2014 • under a voluntary agreement between the Australian Publishers Association and the Australian Booksellers Association, books were available twice as fast as legislatively required under the PIRs, with most major titles available simultaneously in Australia and foreign markets • removal of the provisions would take away protections enjoyed by UK and US authors. The Commission has previously heard and considered arguments of this nature. Similar arguments have been made by the publishing industry in almost all previous reports on PIRs (box 5.1), and neglect the evidence on which the recommendation of the Commission and others were based. In particular: • Comparing the price of books in the Australian market over time, in and of itself, is not relevant to assessing the impacts of PIRs, which are about the price of books in Australia compared to overseas prices at a point in time. • While several participants submitted price comparisons to the Commission in this inquiry, these suffer from a range of methodological flaws — for example, they compare only a limited number of books, or compare titles which are not randomly selected. The Commission has updated its 2009 analysis with a comprehensive comparison of prices (appendix E). While price differences have narrowed, opportunities still exist for bookstores to import the same titles from overseas at materially lower prices. • Submitters pointed to circumstances in New Zealand (NZ) following the removal of PIRs as evidence of what could happen in Australia, arguing that while the average 2 These included Black Inc (sub. 9), Spinifex Press (subs. 19, DR397), UNSW Press Ltd (sub. 27), the Australian Copyright Council (subs. 36, DR543), Mem Fox (sub. DR413), Hachette Australia (subs. 41, DR393), Harlequin Enterprises Australia (sub. 45), Griffin Press (sub. 49), Andy Griffith (sub. DR456), HarperCollins Publishers Australia (subs. 56, DR418), Pearson (subs. 63, DR398), Garth Nix (DR405) Eleanor Curtain Publishing (sub. 69), the Australian Literary Agents Association (subs. 74, DR427), Working Title Press (sub. 81), Romance Writers of Australia (sub. DR413) and the Australian Society of Authors (subs. 121, DR343). 148 INTELLECTUAL PROPERTY ARRANGEMENTS selling price in NZ had fallen by 14 per cent between 2008 and 2015, the range of titles had fallen by 34.5 per cent and sales volumes fallen by 15.7 per cent over the same period (Australian Publishers Association, sub. DR435, att. 2). Appendix E contains a further examination of these claims. But NZ removed its PIRs in 1998, well over a decade before the outcomes now attributed to their removal. And a review of NZ’s arrangements 14 years after PIRs were removed found: Overall, the 1998 changes in New Zealand do not appear to have had significant negative effects on domestic creative effort in the book publishing industry. The number of new NZ book titles that published annually has remained fairly steady between 2005 and 2008. Data on the number of authors shows that following the changes the share of authors in overall employment has increased in New Zealand. (Deloitte Access Economics 2012, p. 6) • While publishers argue PIRs allow them to support Australian authors whose books may not be profitable (through sales of Australian editions of foreign authored books), publishing is not a charitable venture, and publishers select authors they believe will be viable, even knowing that a proportion of their publishing list will not be. Repealing PIRs does not change the fundamental economics of publishing, and publishers will still publish those authors from whom it expects to earn a return. While the Commission sought information from publishers during the public hearings about the extent of this support, none was forthcoming (trans., p. 825). • Notwithstanding claims that income is used to support local authors, most of the additional income from higher book prices goes to overseas authors and publishers whose works are released in Australia. In its 2009 analysis, the Commission estimated the additional income flowing overseas is around 1.5 times that retained by local copyright holders (PC 2009). In effect, PIRs impose an implicit tax on Australian consumers that largely benefits foreign copyright holders. Indeed, none of the authors of the top ten titles in the sample provided by HarperCollins are Australian. • The UK and US arrangements are not directly comparable to Australia. The US Supreme Court 3 has held that US authors do not have the right to prevent importation of commercial quantities of their books from overseas. And authors in the UK have limited protections, and are unable to prevent imports of their books from elsewhere in the EU. • Publishers expressed concern that removing PIRs will harm Australian booksellers. Yet the Commission received evidence from Dymocks (sub. DR613) that Australian publishers act as the local supplier when individuals import books from foreign online retailers. In this way, publishers appear less concerned about Australian consumers accessing books at lower prices than they are about ensuring their continued primacy in the local supply chain. • Claims that local author’s books published overseas will be ‘dumped’ (remaindered) in Australia are unsubstantiated and misleading, and reflect the desire by some to continue price discriminating against Australian readers. Notably, for the majority of books, the same publishing house holds both the Australian and foreign rights. For example, in the 3 See the 2013 US Supreme Court decision in Kirtsaeng v John Wiley & Sons, Inc. COPYRIGHT USE AND LICENSING 149 Commission’s price comparison analysis of the Australian and UK markets, 95 per cent of matched titles were published by the same publisher or a subsidiary. PIRs reduce incentives for the local book industry to operate efficiently, and distort the allocation of resources from their highest value uses. Whether foreign markets retain PIRs is not relevant for Australia’s policy settings — as with trade barriers in other industries, the costs to Australia of retaining PIRs does not depend on whether other countries also have protected markets. • Dymocks summarised the arguments in favour of repealing PIRs. Removing these restrictions would create an open market for books: enabling Australian booksellers the freedom to choose from local and overseas suppliers. This would lead to globally competitive prices and quicker availability on book titles. It would encourage greater efficiency from both retailers and publishers/suppliers in the Australian market and ultimately support a vibrant and flourishing publishing and book retail industry. This will benefit our customers and all Australian readers as well as the broader Australian economy. (sub. DR613, p. 1) PIRs impact the price paid for books in Australia, which costs readers more and disadvantages local bookshops. The Commission recognises that the cultural and educational value of books are significant. But PIRs have intrinsic limitations for targeting such benefits, not least the substantial leakage of PIR assistance offshore. The Commission remains of the view that these benefits are best targeted through direct support to local authors — as is already provided by Australian, state and territory governments (of around $40 million each year) — rather than through PIRs. Many participants raised concerns that governments had substantially cut that funding, but an examination of recent funding suggests this concern is not well founded (table 5.1). Table 5.1 Government funding for literature — 2010-11 to 2014-15a 2010-11 Public Lending Rights ($m) 2011-12 2012-13 2013-14 2014-15 8.7 9.2 9.4 9.5 9.7 11.0 11.0 11.5 11.3 11.8 Australia Council – Literature fundingb ($m) 6.6 5.9 Prime Minister’s Literary Award ($m) 0.4 0.4 7.1 0.6 9.0 0.6 6.3 0.6 Educational Lending Rights ($m) Total Commonwealth funding ($m) 26.7 26.5 28.6 30.4 28.4 a States and territories also provide approximately $12 million in annual funding for literature. b Australia Council – Literature funding includes Get Reading! expenditure (excluding some supplier costs relating to market development in 2010-11 and 2011-12). The fluctuations in funding over time reflect these changes to how Get Reading! funding was accounted for in annual reporting and changes to government policy. In 2014-15 some funding was redirected to the newly established National Programme for Excellence in the Arts. The first grants under this programme were awarded after the 2014-15 reporting period. Sources: ABS (Cultural Funding By Government, Australia, Cat no. 4183.0); Public Lending Rights Committee Annual Reports, 2012/13, 2013/14 and 2014/15; Australia Council Annual Report 2014/15; Australian Government, Budget Paper No. 2, 2009/10 to 2014/15. 150 INTELLECTUAL PROPERTY ARRANGEMENTS Re-examining the evidence on PIRs The Commission’s 2009 analysis of the price impacts of PIRs was based on sales prices over two years (2007-08 and 2008-09) — the result of matching over 900 titles across Australia, the US and the UK. In that study, the Commission concluded that while price comparisons differed across titles and were influenced by the exchange rate at any point in time, Australian booksellers could have obtained and shipped many trade titles to Australia for significantly less than what they were charged for by Australian publishers (PC 2009, p. XVIII). At the time, average selling prices in Australia were on average 35 per cent more expensive than like editions sold in the US, and in many cases the price difference exceeded 50 per cent. In responding to a range of misleading claims and flawed analyses made by participants, the Commission has undertaken a comprehensive analysis of book prices, comparing the price of over 1000 like-for-like titles sold in Australia and the UK (and 400 in Australia and the US). The Commission, in using the same methodology as in 2009, analysed Neilsen Bookscan price data on the top 5000 selling titles in Australia, the UK and the US, for the 12 months ending 28 May 2016. Matching books by International Standard Book Number gave approximately 900 equivalent titles between Australia and the UK, and 109 titles with the US. A broader comparison of titles allowed over 1000 title matches between the Australian and UK, and Australian and US, markets. The Australian average selling price (ASP) excluding GST was 20 per cent higher than in the UK at the average exchange rate, and 33 per cent at the highest exchange rate. For the US, a lack of data precluded a comparison of ASPs such that comparisons could only be made on RRPs which were higher in Australia by 1 per cent at the average exchange rate, and 8 per cent higher at the highest exchange rate (figure 5.2). These results show smaller price differences than the Commission’s analysis in 2009. Nonetheless, the Commission’s analysis accords with its 2009 results, and shows that for a significant number of titles, opportunities exist for bookstores to import titles from overseas at materially lower prices than offered in Australia. Importantly, average price differences obscure more significant price differences at the individual book level — a substantial number of books are cheaper overseas. For example, over three-quarters of books are more expensive in Australia than in the UK using ASP. For these books, the average price differential is $4.20 (around 30 per cent of the average ASP in the UK). Appendix E contains further detail on the price comparisons submitted to the Commission by participants, as well as the Commission’s methodology, sensitivity testing and results. In short, no new evidence was presented in this inquiry that overturns the existing case for removing PIRs on books. COPYRIGHT USE AND LICENSING 151 Figure 5.2 The extent to which Australian prices exceed UK/US pricesa Per cent 40 35 2007-08 2008-09 2015-16 30 25 20 15 10 5 0 US - RRP UK - RRP UK - ASP a 2015-16 year is for end of May 2015 to end of May 2016. Sources: Commission estimates based on the Nielsen BookScan data; PC (2009); RBA (2016). Transitional issues In responding to the Harper Review, the Australian Government asked the Commission to give regard to transitional issues in phasing out PIRs. In its 2009 study, the Commission recommended that the repeal of the PIRs should take effect three years from the date of announcement (PC 2009). Additional transitional arrangements are not needed given the positive confluence of efficiencies made by the Australian publishing industry and broader economic circumstances. The Australian Government should proceed with its announced plans to repeal parallel import restrictions, with effect no later than the end of 2017. RECOMMENDATION 5.3 The Australian Government should proceed to repeal parallel import restrictions for books to take effect no later than the end of 2017. 152 INTELLECTUAL PROPERTY ARRANGEMENTS 5.4 Collective licensing and collecting societies Collective licensing helps reduce transactions costs … Collective licensing allows copyright users to access a range of copyright material in a ‘class’ through a single licence. In Australia, as well as overseas, copyright collecting societies issue collective licences, collect payments from users, and distribute royalties to their rights holder members. Rights holders who are members of a collecting society may grant a broad authorisation for the collecting society to issue licences on their behalf, or may only allow licences covering a narrower set of exclusive rights (such as the public performance right). Australia has copyright collecting societies covering the full range of copyright protected works and subject matter (box 5.2). As noted by the ACCC, collective licensing can help avoid transactions costs, particularly for high volume, low-value transactions. … collective licensing provides a particularly efficient way to overcome the high transaction costs of licensing copyright in markets where the value of individual rights may be low relative to transaction costs and it may be difficult or impossible to predict in advance precisely which rights may be required. (sub. 35, p. 8) The ability to collectively licence IP rights is particularly important for institutions such as restaurants or gymnasiums where negotiating with individual rights holders to play recorded music would otherwise be prohibitive. In the absence of collective licences, many transactions beneficial to both users and rights holders would not occur. … but can also give rise to market power While collective licencing can improve efficiency in licensing, it can also impose costs. Several participants highlighted the potential for collecting societies to exert monopoly power over both rights holders and users. As noted by the ACCC: … collective licensing can also raise concerns under competition laws as it brings together copyright owners who would otherwise be in competition with one another and may enable the collecting society to exercise market power in the setting of licence fees and/or non-price licensing conditions. (sub. 35, p. 8) It is for this reason that the operation of copyright collecting societies is subject to ongoing scrutiny by the ACCC and the Copyright Tribunal — the latter plays a role in the determination of royalty rates under collective licensing arrangements. COPYRIGHT USE AND LICENSING 153 Box 5.2 Australia’s copyright collecting societies Australia’s major copyright collecting societies represent rights holders producing copyright-protected material, with several societies representing smaller, more distinct sections of the creative industries. Australia’s major societies include: • The Australasian Performing Right Association (APRA) administers the public performance and communication rights for musical compositions and lyrics on behalf of authors, composers, music publishers and other music copyright owners. • The Australasian Mechanical Copyright Owners’ Society (AMCOS) collects, administers and distributes royalties collected for the use of mechanical and synchronisation rights for composers and music publishers. • The Phonographic Performance Company of Australia (PPCA) issues licences for the broadcast, communication and public playing of recorded music (sound recordings) and music videos. Its members are Australian recording artists and record labels. • The Copyright Agency Limited (CAL) represents authors, journalists, illustrators, visual artists, photographers and newspaper, magazine and book publishers. It licenses the right to reproduce (such as by photocopying or uploading online) and communicate published print materials (books, newspapers, magazines and online content). • Viscopy represents visual artists, and handles voluntary (non-statutory) copyright licences to reproduce, publish and communicate their artistic works. • Screenrights represents the owners of rights in film and television, including broadcasters, producers, distributors, and rightsholders in scripts, musical works, sound recordings and artistic works. Other collecting societies represent the interests of writers, screen directors and Aboriginal and Torres Strait Islander artists. The Commission heard three main concerns about collecting societies. First, participants raised concerns about the licence fees charged to users by collecting societies, and in particular, increases in the charges for recorded music. The Association of Liquor Licensees Melbourne Inc. submitted: Our industry is required to pay copyright fees to two collection agencies, one for the copyright in the publishing (Australasian Performance Rights Association Limited – APRA), and the other the copyright in the recording (Phonographic Performance Company Australia Pty Ltd – PPCA). Their fee structures and methodologies are calculated separately and independently. Copyright fees increased significantly in Australia in 2008 following an application by PPCA to the Copyright Tribunal of Australia to increase fees. As a result, Australian businesses commenced paying significantly more, up to 10 times more, for copyright than similar businesses in USA, UK, Canada, and European countries. This is not in the best long term interest of the Australian Community. (sub. 62, p. 1) 154 INTELLECTUAL PROPERTY ARRANGEMENTS Determining whether prices are set competitively is inherently difficult. 4 In the case referred to by the Association of Liquor Licences Melbourne Inc, fees were increased following a determination by the Copyright Tribunal. In its decision to increase the licence fees for recorded music, the Tribunal noted a number of approaches exist for determining licences fees, including: • the relevant market rate (if any exists) • a ‘notional bargaining’ rate • comparable rates for other works • judicial estimation • previous agreements between parties • comparison with other jurisdictions • value of the copyright material • capacity to pay • the general interests of the public and consumers. While economic modelling was presented by the collecting society as evidence in support of the licence fee increase, the Tribunal stated ‘ … it is usually not possible to calculate mathematically the correct licence fee in any particular case’. Further, collecting societies argued that to the extent any bargaining power does exist, it is offset by the bargaining power of licensees. As Screenrights remarked: To the very limited extent that the collecting societies operate as monopolies (ie that they are declared to cover all works within a class), then they are dealing with monopsonies (ie single buyers) with equal or greater bargaining power. Educational institutions, government jurisdictions and even private company retransmitters routinely negotiate collective agreements in order to maximize their buying power. (sub. DR454, p. 16) Screenrights went on to note that its inability to refuse licences further curtailed any market power that it may have. The power of the declared collecting societies such as Screenrights, is fundamentally limited by their inability to refuse a licence. Unlike an ordinary copyright owner, or indeed the owners of other property rights, Screenrights is never able to withhold its service. To use broadcasts for the purposes prescribed in the statutory licences, the licensees merely send Screenrights a notice. From the date of the notice they are able to use the copyright material, even absent agreement as to how much they must pay. (sub. DR454, p. 16) 4 A fact acknowledged by the Tribunal itself in Phonographic Performance Company of Australia Limited under section 154(1) of the Copyright Act 1968 (Cth) [2007] ACopyT 1. COPYRIGHT USE AND LICENSING 155 Second, concerns were raised that collecting societies charge excessive fees to rights holders for managing their rights. Collecting societies are funded out of the licence fees charged for licences, and governance arrangements are required to ensure that expenses paid from licensing fees are appropriate and reasonable (table 5.2). However, a single year examination suggests that the Australian collecting societies with the lowest expense ratios (the proportion of licence fees collecting societies spend on expenses) is not significantly different to collecting societies overseas. For example, in 2014-15 the UK Copyright Licensing Agency collected £76.3 million in licence fees with £8.3 million in administrative expenses (10.8 per cent) (Copyright Licensing Agency (UK) 2015, p. 21). In 2014-15, the UK musical collecting society PRS for Music had a total revenue of £539.5 million and an expense ratio of 13 per cent (PRS for Music 2015). Table 5.2 Revenue and expenses for Australia’s major copyright collecting societies 2014-15 Financial Year APRA AMCOS CALa Screenrights PPCA Viscopy Revenue $300.8m $135.1m $45.9m $42.8m $2.2m Payments to rights holders Calculated expense ratio $262.7m 12.4% $136.0m 13.6% $38.6m 14.2% $31.3m 26.8% $1.8m 18.9% a Rights holder payments exceeded revenue due to timing of receipts and payments — expense ratio detailed in annual report. Sources: APRA AMCOS (2015), Viscopy (2015). Copyright Agency (2015), PPCA (2014), Screenrights (2015), Finally, many participants — including from within government — consistently raised concerns about the lack of transparency of collecting societies, and in particular, the lack of information and disclosure around the distribution of funds to rights holders. There are also few requirements to be transparent about how remuneration that is collected is to be distributed to members and non-members of collecting societies. (Department of Communications and the Arts, sub. DR154, p. 2) … there are few legal requirements on collecting societies to act transparently - or to act fairly - in relation to licensees. The Act does not give the Minister any power to direct collecting societies to act in any particular way in relation to licensing or distribution of royalties, or in their general obligations to members or licensees. (COAG Education Council, sub. DR429, p. 32) … copyright collecting societies exercise substantial power in a monopoly situation with little oversight. Because of their monopoly position there is often no obvious point of comparison, and it is not easy for users to determine whether or not it is necessary or beneficial for them to enter a collecting society licence agreement. Terms and conditions of licence agreements are not always publicly available, and it can be difficult or impossible for potential licensees to find out whether the material they use is within the collecting society’s repertoire. This is of 156 INTELLECTUAL PROPERTY ARRANGEMENTS particular concern because of the methods used by some collecting societies to recruit new licensees. (The NSW Department of Justice, sub. 39, p. 8) The current code of conduct does little to lessen concerns Australia’s copyright collecting societies operate under a voluntary code of conduct that establishes minimum principles of transparency, consultation, governance, accountability and dispute handling (box 5.3). In addition, collecting societies with responsibility for administering rights under the statutory licence schemes (section 5.5) must be ‘declared’ by the relevant Minister under the Copyright Act, with the Minister able to revoke such a declaration in limited circumstances. The Copyright Agency Ltd and Screenrights are the declared (and therefore single) collecting society for the purposes of statutory licences. The Copyright Advisory Group (CAG) to the COAG Education Council highlighted what it considered deficiencies in the code. CAG submits that a major failing of the code in its current form is the lack of precision in the scope of collecting society obligations, and the absence of any significant independent oversight or capacity for meaningful enforcement to resolve identified disputes. (sub. DR429, p. 33) CAG also noted that the threat of revocation for declared collecting societies was unlikely to ever operate as a genuine constraint, saying: … the result of revocation would have such a devastating effect on copyright owners and licensees it is extremely unlikely this power would be exercised by a Minister other than in the most extreme circumstances. Other than this power to revoke declaration, the Minister does not otherwise have a general power to direct the collecting societies in relation to their activities in relation to the administration of the licensees, or their conduct towards members or statutory licensees. (sub. DR429, p. 32) CAG also pointed to features of the UK arrangements it argued were missing from Australia’s code. For example, the UK Copyright (Regulation of Relevant Licensing Bodies) Regulations 2014 requires collecting societies to enact a code of practice that includes the following obligations: • ensure that its dealings with licensees or potential licensees are transparent; • consult and negotiate fairly, reasonably and proportionately in relation to the terms and conditions of a new or significantly amended licensing scheme; and • provide to licensees, and to any potential licensees who have requested it, information about its licensing schemes, their terms and conditions and how it collects royalties. (sub. DR429, p. 33) The specific issue of how collecting societies calculate the remuneration payable to rights holders generated significant debate. Information of this nature is important because it enables licensees to better bargain with individual rights holders. While the Copyright Agency publicly disclosed its revenues and expenses and breaks down its payments into COPYRIGHT USE AND LICENSING 157 categories of works, such as books, journals, magazines, websites, newspapers and survey plans, some collecting societies overseas provide a greater degree of detail and transparency. For example, Access Copyright, the Canadian equivalent of the Copyright Agency, provides detail on how payments are split between authors and publishers (Access Copyright 2016, p. 8). Box 5.3 Australia’s collecting society code of conduct Since 2002, Australia’s copyright collecting societies have been subject to a voluntary code of conduct, aimed at ensuring their operations meet minimum standards of accountability and transparency. The Code sets out the obligations of collecting societies in dealing with their rights holder members and licensees, including: • fairness in their dealings with rights holder members and licensees • making available information on their Constitution, licences and licence schemes, both to members and licensees (and potential members and licensees) • drafting licences that are clear and easily understood • consulting rights holder members and licensees on the terms and conditions of licences • fair and reasonable licence fees, giving consideration to matters such as the value of the copyright material and the purpose and context of how the material will be used. In addition, each society must have a policy (available to rights holder members on request) setting out how rights holder entitlements and payments from remuneration and/or licence fees will be calculated, the manner and frequency of payments to members and the amount to be deducted for expenses before distribution. These deductions include the management and operating expenses and other amounts authorised by the constitution of the collecting society. The societies also have minimum governance and accountability obligations. • Each society’s Board of Directors must be representative of, and accountable to, its rights holder members. • The Code establishes financial reporting requirements regarding the content and annual auditing of financial reports. Each society’s annual report must detail total revenue for the reporting period, total and nature of expenses, and allocation and distribution of payments to rights holder members. • Collecting Societies must provide rights holder members with information about their payment entitlements upon request. An independent report is produced annually (currently by a retired Federal Court judge) on each society’s compliance with the code, and the code itself is reviewed every three years. The latest compliance review code concluded that, notwithstanding some minor issues with the provision of information to the review, societies generally complied with the requirements of the code. Sources: Arts Law Centre of Australia (2003), Lindgren (2015a). 158 INTELLECTUAL PROPERTY ARRANGEMENTS Commenting on the various disputes about how it distributes its licence revenue, the Copyright Agency went on to explain: Recently, the schools sector, through the Copyright Advisory Group to the COAG Education Council (CAG), requested additional information that is largely derived from our research rather than from data provided by teachers: the name of the publisher for works recorded as used in schools. We agreed to provide that information. CAG sought the information to assist it to contact publishers with requests for permission to use their content (rather than relying on the statutory licence scheme). Data about usage of content in the government sector has been limited in recent years. We have therefore distributed licence fees in accordance with content likely to have been used in the sector, based on information from a variety of sources. Information about how we distribute licence fees is available on our website. The NSW Government has sought information about payments to individual recipients. Our position is that, consistently with obligations to members in the Code of Conduct, we should not do this without the consent of the recipients concerned. The NSW Government and CAG also requested additional information about distributions of licence fees in our annual report, which we provided in our recent annual report (and will continue to include): • a more detailed breakdown of primary recipients of licence fees (at page 26); • a breakdown of recipients of licence fees paid by governments (on page 28); and • a breakdown of funds held in trust, by licence fee source (pages 35 and 36). (sub. DR510, pp. 19–20) Concerns about how collecting societies calculate the remuneration payable to rights holders was raised, by both the NSW Government and CAG, as a significant issue during the last review of the collecting society code of conduct (Lindgren 2015b). While a number of reforms were put forth, the review did not ultimately recommend that additional obligations be imposed on the declared collecting societies, on the basis the issue was a larger one than could be accommodated within the three-yearly code review (Lindgren 2015b). In commenting on the current reporting obligations, the Copyright Agency (a declared collecting society) noted that it provided: … our annual reports to the Minister, who tables them in Parliament. We have from time to time been asked to include additional information in our annual reports by the Minister and the Department to address requests for information about our activities, with which we have always complied. (sub. DR510, p. 1) Australia is far from alone in seeking to ensure copyright collecting societies are efficiently managed and work in the best interests of rights holders and users. In 2014, the European Parliament passed a directive outlining the regulation of copyright collective management organisations (collecting societies) in the EU (EC 2014). Unlike Australia’s voluntary code, the EU framework will be mandatory for all collecting societies, and contains COPYRIGHT USE AND LICENSING 159 detailed and prescriptive requirements regarding the annual reporting of how payments are made to, and between, various classes of rights holders. Such a framework, while still being implemented by EU members, does provide a benchmark for how governance and transparency requirements could potentially be strengthened in Australia. Reforming the code would help increase transparency Ensuring collecting societies operate efficiently is important to rights holders and consumers as it impacts on the operation of Australia’s copyright arrangements. Over time, technological developments will likely continue to improve the ways consumers can access copyright material, encouraging further efficiencies in collecting societies, and possibly even their need (box 5.4). However, while collecting societies continue to operate, the case for reforms to strengthen the governance and oversight arrangements, and in particular for declared collecting societies, is strong. The Commission is particularly concerned by the submissions of statutory licence users that they are not able to access the information needed to allow them to effectively negotiate directly with rights holders. The purpose of the statutory licence systems is to ensure licensing solutions are available in cases where transaction costs would prevent ordinary bargaining from occurring. The fact that users have indicated their desire to negotiate directly with rights holders, but are unable to do so, suggests arrangements are not supporting efficient and fair market outcomes. In the Commission’s view, greater information about the size of payments to classes of rights holders should be made available in collecting society annual reports. And consistent with the requirement for governments to spend taxpayer funds efficiently, there is also a case for licence fees paid under the statutory licences to be held, and accounted for, separately by the declared collecting societies. Any funds that cannot be paid to rights holders (for example, because the works are orphaned) should be returned to government, rather than distributed to other rights holders who have no connection with the work used. The arrangements for reviewing and amending the code are also deficient, particularly given the recent supplementary review found that the grievances of licensees could not be accommodated within the existing system. Given its experience managing similar arrangements in industries where parties have potential market power, in the Commission’s view the ACCC should undertake a comprehensive review of Australia’s collecting societies and the current code of conduct. 160 INTELLECTUAL PROPERTY ARRANGEMENTS Box 5.4 Technological developments can further improve consumer access Over time, technological developments will likely reduce the costs of collecting societies, and possibly even their need. For instance, Spotify, a music streaming service, is now offering commercial use services in Sweden, and aims to extend this to other countries. Developments such as weCreate, Linked Content Coalition and Arrow also represent technological developments that change the business models and costs of transactions between creators and users. A key example where technological developments are providing alternatives for accessing copyright content is the UK Copyright Hub, which stemmed from a recommendation of the Hargreaves Review of IP. The Copyright Hub is working to facilitate development of a digital copyright exchange, to improve secondary licensing of existing copyright material between rights holders and users. In addition to its roles in facilitating engagement between rights holders and users and educating the public about copyright, the Copyright Hub Foundation describes the Hub as: The idea is simple. A copyright work has or is assigned a unique identifier which resolves back, machine to machine, to the rights owner/the creator. The prospective reuser/licensee can do something as simple as right-clicking on the work, currently using a plug-in but soon we hope to be incorporated in browsers, and find out what licences are available to reuse the work and/or be directed back to the rights owner’s website. The transaction cost is zero. The licence itself may of course cost something. That decision, whether to charge for a reuse licence, is up to the rights owner, nothing to do with Copyright Hub. Many rights owners/creators are happy to have their work reused for free as long as they can be properly acknowledged and have their name spelt properly! (sub. 6, pp. 5–6) The technology-based approach seeks to overcome rights holder concerns about consumer reuse of copyright-protected material online. Consumers often reuse protected works, such as text, artwork, photographs, sounds, music and film clips as part of using the Internet, including through blog posts, memes and other social media activities. From a consumer perspective, accessing information about relevant rights holders, or the options for purchasing licences, is difficult, if not impossible. The Hub is testing technology solutions to improve licensing, but is yet to rollout a fully operational product. The Copyright Hub is supported by rights holders and collecting societies in the UK, US and recently Australia, and received initial funding from the UK Government. The UK Government is considering the case for additional funding, following a recent review by EY. Source: The Copyright Hub Foundation London (sub. 6). RECOMMENDATION 5.4 The Australian Government should strengthen the governance and transparency arrangements for collecting societies. In particular: • The Australian Competition and Consumer Commission should undertake a review of the current code, assessing its efficacy in balancing the interests of copyright collecting societies and licensees. • The review should consider whether the current voluntary code: represents best practice, contains sufficient monitoring and review mechanisms, and if the code should be mandatory for all collecting societies. COPYRIGHT USE AND LICENSING 161 5.5 Statutory licensing Reform or removal — where to for statutory licensing? The Copyright Act contains two statutory licence schemes that allow education institutions and government to use copyright material and payment of a prescribed fee (box 5.5). The Copyright Tribunal has jurisdiction to hear and settle disputes arising under the operation of the schemes, including determining adequate remuneration. Both schemes require users to undertake administrative steps, such as recording the use of copyright material for determining remuneration payable. Box 5.5 Australia’s statutory licensing schemes Australia has two statutory licensing schemes. One, designed for educational and other institutions, covers: • copying and communication of copyright material transmitted via a broadcast (Part VA) • copying and communication of text and images, including digital content (Part VB). Part VA licences permit educational institutions to copy and use television and radio broadcasts for educational purposes, as well as podcasts and webcasts that originated on free-to-air television and radio services. Screenrights administers the Part VA licence. Part VB licences permit copying and communication of literary, dramatic, artistic and musical works. The Copyright Agency Ltd administers Part VB licences. The second is designed for government use. Under the Copyright Act the Australian, State and Territory Governments are able to use copyright material ‘for the services of the Commonwealth’. Rather than negotiate with individual rights holders, the respective schemes allow education institutions and governments to access copyright material without seeking the agreement of rights holders. The National Copyright Unit to the COAG Education Council submitted that in 2015, Australian schools paid: … over $90 million in licensing fees to copyright collecting societies for the use of copyright materials in schools under collective licensing arrangements. This is made up of: • Approximately $62 million for the use of literary, dramatic, artistic and musical works under statutory licence; • Approximately $22 million for the use of broadcasts under statutory licence; and • Approximately $7.5 million for the use of music under voluntary licensing arrangements. (sub. 97, p. 4) The operation of the statutory licensing schemes — and in particular education licensing arrangements covered by parts VA and VB of the Copyright Act — was reviewed by the ALRC in its copyright report. In its Discussion Paper (draft report), the ALRC recommended Australia’s statutory educational licensing arrangements should be repealed, with rights holders and users free to negotiate voluntary collective licences, as happens in other cases where the statutory regime does not apply. 162 INTELLECTUAL PROPERTY ARRANGEMENTS In support of its recommendation to repeal the licence arrangements, the ALRC argued that statutory (non-voluntary) licences: • conflicted with the exclusive rights granted to rights holders, were inefficient and not suited to the digital age • risked under-rewarding rights holders for the value of their copyright protected works • reduced the incentives for rights holders to develop new approaches to licensing • for educational institutions, which the scheme was established to assist: – already used voluntary collective licences for copyright-protected works not covered by the statutory scheme (recorded music, for example) – specifically sought repeal of the arrangements (ALRC 2013). Rights holders and collecting societies objected strongly to the ALRC’s draft recommendations, largely over concerns that educational institutions may not purchase voluntary licences were the scheme abolished. In its final report, the ALRC recommended the statutory regime be retained, but reformed to make the system work better (ALRC 2014). The ALRC noted that many of the criticisms with the statutory regime, particularly what is counted and paid for, could be addressed through adopting its recommended fair use exception (discussed in the following chapter). Specifically, the ALRC recommended: The statutory licences in pts VA, VB and VII div 2 of the Copyright Act should be made less prescriptive. Detailed provisions concerning the setting of equitable remuneration, remuneration notices, records notices, sampling notices, and record keeping should be removed. The Act should not require sampling surveys to be conducted. Instead, the Act should simply provide that the amount of equitable remuneration and other terms of the licences should be agreed between the relevant parties, or failing agreement, determined by the Copyright Tribunal. (2013, p. 208) Agreed reforms represent the most practical option In December 2015 the Australian Government proposed reforms to the education statutory licensing scheme, which follow the ALRC recommendations. Notably, this decision was absent the complementary recommendation to move to a fair use system of copyright exceptions. Importantly, the proposed amendments to the Act make it clear that the statutory licence regime is compulsory for rights holders and not users, with voluntary licensing permissible if such an approach is more efficient and effective for rights holders and users. The proposed reforms were jointly proposed and negotiated by the collecting societies and education users, representing a consensus position (CAL, sub. 47). Given this, the proposed reforms may be a reasonable approach to simplifying the administrative arrangements under which the education statutory licence operates. COPYRIGHT USE AND LICENSING 163 6 Fair use or fair dealing — what is fair for Australia? Key points • Australia’s current exceptions for fair dealing are too narrow and prescriptive, do not reflect the way people today consume and use content in the digital world, and do not accommodate new legitimate uses of copyright material. • Instead of targeting the effect of any potential use on the incentives to create new works, fair dealing is unduly focused on the type or purpose of use. • Introducing the principles–based fair use exception as Australia’s system of user rights, would go some way to redress the imbalance between copyright holders, consumers and intermediate users. – In the US, where fair use is well established, creative industries thrive, while Israel recently adopted fair use to enable better access to copyright material ‘for the advancement of culture and knowledge’. • Previous inquiries have recommended fair use in Australia, most recently by the Australian Law Reform Commission in its 2014 report Copyright and the Digital Economy. – Independent analysis commissioned by the Australian Government shows adopting fair use would deliver net benefits for the Australian community. • Fair use does not (as some suggest) mean free use, and nor has it in any of the jurisdictions in which it operates as the exception specifically considers the impact of the use on the rights holder. – Four ‘fairness factors’ would be applied when determining if a use of copyright material is fair, as well as a number of illustrative uses to guide use of the exception. – Importantly, the exception would operate alongside Australia’s existing arrangements for voluntary and statutory licensing. • Many argue that the move to fair use would herald an unacceptable level of uncertainty. – As a principles and context–based exception, fair use is adaptive (as distinct from uncertain) over time. – Evidence suggests the certainty of Australia’s existing arrangements is overstated, with a lack of adaptability resulting in uncertainty. – The fairness factors, illustrative uses, Australian and overseas case law, and industry-led guidance would all help to provide greater certainty and so ease the transition to a fair use regime. – The Government commissioned independent analysis found that beyond this transition there is no evidence to suggest fair use is inherently less certain than the status quo. • Looking forward in a digital age, the independent analysis also found that prescription (fair dealing) can be more of a liability and flexibility (fair use) an asset. FAIR USE OR FAIR DEALING — WHAT IS FAIR FOR AUSTRALIA? 165 Australia’s copyright laws contain a number of exceptions to the exclusive rights granted to creators; an essential complement to a broad system of unregistered rights. Exceptions operate as a defence for acts that would otherwise infringe a creator’s exclusive rights. Australia’s most commonly known exceptions are for ‘fair dealing’ in copyright material; for time- and format-shifting of copyright material; to allow libraries, archives and other cultural institutions to preserve and disseminate works, particularly in the digital era; and to allow the operation of some technology processes. The chapter focuses on whether there is a case for replacing Australia’s fair dealing exceptions with a broad fair use exception. It begins by outlining the importance of exceptions in the digital age, and setting out the main differences between fair use and fair dealing. The chapter then examines the relative merits of the two approaches, and considers how to facilitate the transition to fair use, with a particular focus on minimising transition costs arising from initial uncertainty. In examining these issues, the Commission has drawn on evidence provided by participants, its own research, as well as economic principles. The Commission has also drawn on earlier reports relating to copyright, most notably the Australian Law Reform Commission’s (ALRC) 2014 report on copyright and the digital economy, which recommended Australia adopt a fair use exception. 6.1 Copyright exceptions — why do they matter? As set out in chapter 4, while copyright encourages investment in creative works by allowing creators and rights holders to exploit their value, it is poorly targeted. Unlike other IP rights, copyright makes little attempt to focus on those works where ‘free riding’ by users would undermine the incentives to create. Instead, copyright is expansive and ‘all encompassing’, providing the same levels of protection to commercial and non-commercial works, to works with essentially no degree of creativity, to works that are no longer being supplied to the market, and to works where ownership can no longer be identified. A lack of registration required for copyright protection to apply exacerbates the problem, leading to beneficial uses of material, such as the use of orphan or unavailable works, going unrealised. Exceptions allow limited use of copyright material without the authority of, and with or without payment to, the rights holder. In an efficient system, copyright exceptions and limitations would balance the incentive to create new works with the benefits to users from dissemination and consumption, and would allow for new and valuable uses of copyright material. The ALRC highlighted the critical role exceptions play in ensuring balance in Australia’s copyright system: The ALRC considers that the property rights granted to creators and rights holders are important and may be necessary to provide an incentive to create, publish and distribute copyright material. But this should not be extended further than necessary. Rights holders should not be entitled to all conceivable value that might be taken from their material. The incentive to create will not be undermined by the unlicensed use of copyright material for entirely different purposes from the purpose for which copyright material was created, and in 166 INTELLECTUAL PROPERTY ARRANGEMENTS markets that do not compete with rights holders. Rather, such uses will stimulate further creativity, and increase competition. (ALRC 2014, p. 132) While a range of other rationales have been advanced (box 6.1), the Commission considers that this balancing role is among the most important functions of copyright exceptions. The ALRC put it eloquently when they noted, ‘copyright must leave ‘breathing room’ for new materials and productive uses that make use of other copyright material’ (p. 23). Box 6.1 Why have copyright exceptions? Exceptions to the exclusive rights granted by copyright have existed as long as the copyright system itself, and the potential for nations to implement exceptions was recognised in the Berne Convention. A number of theories have developed to explain the underlying economic basis for the exceptions commonly seen in copyright arrangements around the world. Landes and Posner (2003) note that the presence of high transaction costs will, in many cases, be a barrier to users efficiently bargaining with rights holders. In cases where the use of copyright material is relatively minor or trivial, the willingness of consumers to pay for small amounts of content will likely be low. If there are many users, the transaction costs of bargaining will be high, and as a result, many transactions that might otherwise be profitable do not occur, reducing the wellbeing of both rights holders and consumers. Efficiency is also reduced in cases where a rights holder refuses to licence a work at any price. This can be the case with parodies and satires that draw on previous works. Both Yen (1991) and Landes and Posner (2003) note that while society as a whole values the production of new satires and parodies based on existing works, given a parody makes fun of an original work, it is unlikely many rights holders would licence their works for such a purpose. An exception overcomes this ‘hold up’ problem, with the benefits from allowing parodies and satires likely to exceed the costs to rights holders. A similar case exists for the time-sensitive task of reporting news. Although rights holders and news outlets could strike an agreement to use copyright protected material, the commercial imperative to be the first to report a news event likely makes negotiating with rights holders impractical, if not impossible. Again, given society’s demand for news reporting, a broad exception likely has net benefits, even if it reduces the returns from intellectual property a rights holder might otherwise have earned. Well-functioning exceptions are becoming more important The importance of this balancing role is increasing. In order to realise the benefits afforded by the digital economy, Australia’s copyright arrangements must facilitate (or at a minimum not discourage) transformative, innovative and collaborative use of copyright materials. Digital technology, including search functions, cloud-based solutions and other digital platforms, provides opportunities to create and deliver new and valuable products and services, as well as productivity gains for individuals, businesses and governments. Participation in the digital economy will be a critical source of innovation for Australian firms and consumers. FAIR USE OR FAIR DEALING — WHAT IS FAIR FOR AUSTRALIA? 167 6.2 Fair dealing and fair use — how do they compare? There are two main approaches for providing exceptions to the exclusive rights granted to creators — fair use, which relies on principles and context to determine if use of copyright material is ‘fair’, and fair dealing, which prescribes a confined list of purposes for which exceptions can be granted. Due to its prescription of included matters, fair dealing arrangements are typically described as ‘closed’. In Australia, copyright exceptions are provided by way of fair dealing (box 6.2) — similar to the approach taken by other Commonwealth countries such as the UK and New Zealand. The United States, the Philippines, South Korea, Singapore and Israel all rely on fair use. Box 6.2 Australia’s fair dealing arrangements Fair dealing is a defence under the Copyright Act 1968 (Cth) for acts that would otherwise infringe the exclusive rights granted by copyright. They allow copyright material to be used, without the authorisation of the rights holder. In considering whether the defence applies, a court must be satisfied of two things. First, the dealing must be for one of the following prescribed purposes: • research or study • criticism or review • parody or satire • reporting the news • giving legal advice. Second, the dealing must be considered ‘fair’. Whether a dealing is fair depends on the facts of each individual case. In the case of research and study, and indirect recording of performances, the Copyright Act also sets out matters for the court to consider when deciding if a dealing is fair. The exception does not apply to all forms of copyright protected material, but to literary, dramatic, musical or artistic works; adaptations of literary, dramatic or artistic works; and ‘audio-visual items’, including sound recordings, films and broadcasts. One of the key differences between fair dealing and fair use is where responsibility lies for determining the ‘fairness’ of new uses of copyright material. In Australia (and other Commonwealth countries), legislative change is required to expand the categories of use deemed to be fair. In contrast, in countries such as the US, courts have the latitude to determine if, on the facts, a new use of copyright material is fair (box 6.3). The US courts have developed considerable case law on the scope of fair use of copyright-protected material, and a range of guidance tools exist to assist rights holders and users to determine what uses of copyright material the courts might consider fair. 168 INTELLECTUAL PROPERTY ARRANGEMENTS Box 6.3 How do fair dealing and fair use differ? In Australia, if a defendant claims their dealing in a copyright work is ‘fair’, the courts must first determine whether the dealing is covered by one of the specified fair dealing exceptions. If a dealing is not for the purposes of research or study; criticism or review; parody or satire; reporting the news; or for giving legal advice; then the dealing cannot be considered fair. While similar to Australia’s exception for fair dealing, the United States’ fair use exception embodies some important differences. Section 107 of the US Copyright Act provides: … In determining whether the use made of a work in any particular case is a fair use the factors to be considered shall include: • the purpose and character of the use, including whether such use is of a commercial nature or is for non-profit educational purposes; • the nature of the copyrighted work; • the amount and substantiality of the portion used in relation to the copyrighted work as a whole; and • the effect of the use upon the potential market for or value of the copyrighted work. These are known as the ‘fairness factors’. Under US law, courts determine whether a use of copyright material is fair only by reference to the fairness factors. In this way, new uses of copyright material can be considered fair if the facts of the case meet the test. As a result, many have argued fair use is a more adaptable exception than fair dealing, with the courts able to take into account contemporary values and attitudes when ruling on fair use. Only Australia’s exception for fair dealing for the purpose of research or study (s. 40) contains ‘fairness factors’ the court should take account of. Interestingly, these fairness factors contain a factor not found in the US laws — s. 40(2)(c) directs the court to also consider ‘ … the possibility of obtaining the work or adaptation within a reasonable time at an ordinary commercial price’. Importantly, US-style fair use is not about private use per se. While some private uses of copyright material, such as time shifting, have been found to be non-infringing under the fair use defence, not all private use will be considered fair. And some public and even commercial uses of copyright material have been found ‘fair’ by the courts. Many activities permissible in the US under the fair use exception would likely be considered an infringement in Australia in the absence of a voluntary or statutory licence (table 6.1). FAIR USE OR FAIR DEALING — WHAT IS FAIR FOR AUSTRALIA? 169 Table 6.1 Illustrative US fair uses of copyright works that require a licence in Australia Illustrative scenario Australian fair dealinga US fair useb An Internet search engine publishes thumbnail images of websites in its search results.   An author quotes a number of unpublished letters and journal entries in a biography. An artist creates a collage using images from a photography book. A database of TV clips enables users to search broadcasts using keywords, and then view a clip containing the keywords.       Scenes from a film are used in a subsequent biographical film about the lead actor.   An election advertisement uses a sample of a song used in an opponent’s advertisement.   A rap song pays homage to another well-known song by using the opening lyrics. Researchers access a database for text and data mining.     A teacher wants to record a specific TV or radio news program for use in class. A teacher copies a single chapter of a book for inclusion in a set of class materials (30 copies). A teacher scans pages from textbooks to use in their lessons via an interactive whiteboard. A school library copies thumbnail images of books from the Internet for use in an online library catalogue.  Potentially fair use  Potentially fair use  Potentially fair use  Potentially fair use a Activity not covered by fair use and is remunerable, although a licence may be granted without payment. b Based on US case law and guidelines; dependent on application of fairness factors. Sources: Google Australia (sub. 102); National Copyright Unit, COAG Education Council (sub. 97); Stanford University (2015). 6.3 The merits of a fair use exception Participants offered different perspectives on the relative merits of fair use and fair dealing (table 6.2). Individuals and industries currently benefiting from copyright protection universally opposed the adoption of fair use and argued that fair dealing was sufficiently adaptable in the digital age. Those who would gain from more flexible exceptions, as well as a number of academics, supported the adoption of fair use and argued that it builds on Australia’s current fair dealing exceptions, retaining the focus on fairness, but removing unnecessary limitations to specific types of use. The Commission has examined each of these arguments in turn. 170 INTELLECTUAL PROPERTY ARRANGEMENTS Table 6.2 The varied perspectives on the merits of adopting fair use Opponents of fair use Proponents of fair use Australia and other countries have considered and rejected fair use A number of countries have successfully made the transition to a fair use regime and its adoption has been recommended in Australia Fair dealing is sufficiently responsive and has been effective in retaining the balance between the creators and users of copyright material Fair dealing has not kept pace with change and the balance of rights is increasingly out of kilter Fair dealing is sufficiently flexible Fair use allows for new, valuable uses of copyright material and for the emergence of new industries Fair use is free use, and will significantly reduce the revenue earned by rights holders Recompense will continue under fair use, however users will no longer pay collecting societies for freely available material Fair use will significantly undermine creative industries The purported impacts of fair use on the creative industries are ‘overblown’ Fair use would give rise to legal uncertainty due to a lack of legislative clarity Fair use is sufficiently certain and predictable and no more uncertain than current fair dealing exceptions While copyright law may not align with community expectations, rights holders do not, as a matter of course, enforce their rights Fair use better aligns with community expectations Fair use is incompatible with the three step test under international copyright law Fair use is compatible with the three step test under international copyright law Source: Based on submissions and public hearings. More countries have embraced fair use than rejected it A number of participants argued that many countries have examined the case for fair use, and rejected it, and that this was grounds for the Commission to reject its adoption (see, for example, sub. 181). The example most commonly raised was the Hargreaves Review of Intellectual Property in the United Kingdom. The Hargreaves Review noted ‘[t]he copyright regime cannot be considered fit for the digital age when millions of citizens are in daily breach of copyright, simply for shifting a piece of music or video from one device to another’ (Hargreaves 2011, p. 5). However, the review ultimately recommended an expanded approach to fair dealing, primarily given the uncertainty and potential difficulties of adopting fair use within the European Union framework (Australia is not constrained by the EU framework). Others have argued Hargreaves’ approach to fair use was more strategic, using it as a ‘punching bag’ for rights holder concerns, and allowing the review to make more extensive recommendations on fair dealing (Boyle 2015). Notwithstanding the position taken by the Hargreaves Review, countries other than the UK have considered or adopted fair use. For example, the Philippines adopted a US-style exception in 1998, Singapore in 2006 and the Republic of Korea in 2012. And in 2007, Israel modernised its copyright arrangements with a new Copyright Act, replacing its FAIR USE OR FAIR DEALING — WHAT IS FAIR FOR AUSTRALIA? 171 previous exception for fair dealing with a broader fair use exception. The preamble to the new Copyright Act states: The balance required is mainly between the need to provide a sufficient incentive to create, which is in the form of granting general financial rights in the creations, and between the need to enable the public to use the creations for the advancement of culture and knowledge. (Nair 2012) While the new Copyright Act gives examples of some permitted fair use purposes, the Israeli approach is more permissive again, allowing the Minister of Justice to prescribe additional fair uses in regulation. Closer to home, several participants also noted the report by the Australian Intellectual Property and Competition Review Committee in 2000, did not recommend Australia adopt fair use. At the time, the Committee did not consider that there was a need to alter the balance between owners and users of copyright material. However, the Committee went on to note, that as technology changes, the balance struck in the grant and limitation of rights also changes. The Committee suggested that, as part of a three-year review program, the Government examine whether the development of technology and new markets had altered the balance to the detriment of rights holders. Almost a decade and a half later — and following further significant technological change — the ALRC examined Australia’s copyright arrangements in 2014 and recommended a suite of changes to ensure Australia’s copyright arrangements are contemporary and adaptable in the digital age. The ALRC’s key recommendation was that Australia’s current fair dealing exceptions be repealed and replaced with a fair use exception. In making its recommendation, the ALRC argued adopting fair use in Australia would restore balance to the copyright system, particularly since the Australia-United States Free Trade Agreement came into force and strongly favoured rights holders. Moreover, fair use would better align the law with modern expectations about how copyright material could be used, transformed and moved between platforms and devices. Fair use would provide an important counterbalance As discussed above, the Commission considers that one of the most important functions of copyright exceptions is to balance the interests of rights holders and users. A number of participants argued that Australia’s copyright regime is already sufficiently responsive to shifts in the balance, pointing to the previous amendments to introduce new exceptions. For example, the Australian Screen Association, commenting on changes to Australia’s copyright system noted: The changes have enhanced the protections for copyright owners and introduced a range of new exceptions to balance the interests with those of copyright users. (sub. 43, p. 13) 172 INTELLECTUAL PROPERTY ARRANGEMENTS However, amendments to the Copyright Act to introduce new exceptions have largely sought to codify existing practices (such as ‘allowing’ the existing widespread practices of private time- and format-shifting), or to remove unintended overreach from the copyright system (such as allowing temporary copies and website caching). To the Commission’s knowledge, copyright exceptions have never been expanded to counterbalance the increase in the scope or duration of protection for rights holders. Indeed, the Joint Standing Committee on Treaties report into the Australia-US Free Trade Agreement specifically recommended the adoption of a US-style fair use exception because of concerns about adopting the US level of copyright protection without the US counterbalances: The Committee recommends that the changes being made in respect of the Copyright Act 1968 replace the Australian doctrine of fair dealing for a doctrine that resembles the United States’ open-ended defence of fair use, to counter the effects of the extension of copyright protection and to correct the legal anomaly of time shifting and space shifting that is currently absent. (Joint Standing Committee on Treaties 2004) Similarly, the ALRC recognised that fair use can counterbalance what has been described as an ‘unreasonably broad grant of rights’ and an ‘unduly narrow set of negotiated exceptions and limitations’. Fair use allows for new, valuable uses of copyright material While proponents of fair dealing argue it is sufficiently flexible to accommodate new and valuable uses of copyright material, other participants have argued that Australia’s fair dealing exception limits innovative activities. As noted by Universities Australia: Australia’s existing inflexible, purpose-based copyright exceptions are no longer fit for purpose. They are holding Australia back, not just in our universities and schools, but also in our digital industries. Innovative and useful technologies, and new ways of using content in socially beneficial ways, automatically infringe copyright in Australia unless their use falls within one of the existing narrow, purpose-based exceptions. (sub. DR453, p. 1) Google Australia highlighted some of the foregone opportunities: Innovation is dynamic, not static. In contrast, Australia’s copyright exceptions are ‘static’ — confined to specific purposes and technologies, and not capable of adapting to changes in technologies, consumer uses or business practices. Australia’s copyright system arguably prohibits critical technologies and innovative activities from being conducted in Australia, such as: • basic Internet functions such as system level caching to provide a search engine • cloud computing • creative and transformative works, such as mashups • medical and scientific research, such as text and data mining • various common consumer uses of copyright materials. (sub. 102, p. 2) FAIR USE OR FAIR DEALING — WHAT IS FAIR FOR AUSTRALIA? 173 In contrast, fair use affords greater flexibility including to new and innovative copyright-dependent industries, provided those uses meet the fairness factors. Courts can determine if new uses of copyright material are fair, avoiding the need for industries to wait for legislative change. The United States, where fair use has operated for decades, has long been recognised as a leading innovator, and many innovations Australians use and rely on today, such as Internet search engines, were developed under a fair use exception. Universities Australia highlighted some of the innovative opportunities they could exploit under fair use: Fair use would put Australian universities on a level playing field with universities in comparative jurisdictions such as the USA, Singapore, Israel, and South Korea. It would mean: • Australian academics being able to take full advantage of innovative new technologies such as data mining and text mining • Australian universities having greater flexibility when creating MOOCs [Massive Open Online Course] • Australian universities no longer having to cut third party content from student theses before making these publicly accessible online • Australian academics being free to include small amounts of third party content in conference papers • Australian universities not being blocked by copyright when engaging in collaborative projects with business and industry. (sub. DR453, p. 2) Following the final ALRC report Copyright and the Digital Economy, the Australian Government commissioned a cost-benefit analysis of the recommendation to adopt fair use in Australia (EY 2016). Assessing the impacts of fair use is intrinsically difficult. However, the EY analysis (unlike others commissioned by inquiry participants) provides a sound assessment of the impact of fair use on Australian consumers and the broader community. EY’s ‘bottoms-up’ analysis sought to calculate, in dollar terms where possible, the change in welfare to the Australian community from adopting additional fair dealing exceptions for quotation, non-commercial private use, incidental or technical use, library and archive use, education, and access for people with a disability; and then the benefit of shifting to an open-ended fair use exception (where these uses would become illustrative fair uses). Importantly, the cost-benefit analysis includes the impact of changes on consumers and end users, intermediate users such as schools and libraries, and original creators. The analysis sought to measure the change in both short-term welfare and long-term incentives to create works. The report concluded that improving education uses and access to orphan works offered the largest benefits from adopting fair use in Australia. Other changes reduce uncertainty for consumers and businesses, improving Australia’s innovation environment. Reflecting that the ALRC recommendations incorporated some of Australia’s existing exceptions. 174 INTELLECTUAL PROPERTY ARRANGEMENTS In summarising their findings, the report stated: Overall, our analysis of new fair dealing suggests that the ALRC’s proposed recommendations should be beneficial, albeit not substantially in some areas. From the standpoint of an ‘openended’ (fair use) or ‘closed-ended’ (fair dealing) system of exceptions, the former is likely to have the largest net benefit. In a rapidly changing digital environment, the costs and benefits associated with copyright will change over time. In this context, prescription can be a liability and flexibility an asset. Beyond a transition period, there is no evidence to suggest that fair use is inherently less certain than the status quo. Where it does differ is flexibility; it would be difficult to argue that fair use is less flexible than the status quo. (EY 2016, p. viii) While fair use would clearly benefit copyright-reliant sectors and users, quantifying the impacts of a broad, open-ended exception are compromised by measurement challenges. Appendix E contains a more detailed examination of the fair use analyses. Fair use is not free use A number of publishers argued it was not necessary to allow more uses of copyright material to be covered by fair use, because such uses could be licenced by rights holders (and thus remunerated). Participants also argued fair use would become synonymous with ‘free use’: Our primary concern is that the introduction of a ‘fair use’ scheme will be interpreted as allowing free use of material where that use will be (and should be) paid for. (R.I.C. Publications, sub. 12, p. 7) However, fair use does not equate with free use in any of the jurisdictions in which it operates, nor would it in Australia. Indeed, the exception specifically considers the impact of the proposed use on the market for a work. The creative sector was particularly concerned that fair use would significantly curtail, if not eliminate, payments for educational uses under the statutory licence. As discussed in chapter 5, statutory licences allow education institutions, and the Australian, state and territory governments, to use copyright-protected works upon payment of ‘equitable remuneration’. FAIR USE OR FAIR DEALING — WHAT IS FAIR FOR AUSTRALIA? 175 Representative of the concerns raised, the Australian Publishers Association stated that to replace Australia’s current fair dealing arrangements with a US‐style ‘fair use’ exception that includes the use of copyright material for education: … will effectively eliminate statutory licensing fees for the education sector. This will impact on the sustainability of educational publishing in Australia and will lead to: • less income for creators, including writers, illustrators and designers • less investment in Australian‐specific educational content • greater legal risks to users of copyright material • more litigation around copyright • loss of jobs and skills in educational publishing • less Australian educational content for Australian students. (sub. 435, p. 9) In contrast, the Copyright Advisory Group to the COAG Education Council argued fair use would avoid the current situation where Australian schools pay for freely available material and that concerns about educational licensing where overblown: Fair use would fix the current situation, where Australian schools pay millions of dollars each year in licensing fees to use freely available internet materials for students or orphan works for which no author can be identified. Fair use would also enable sensible digital uses of copyright materials in Australian schools, in situations where there is no harm to copyright owners’ markets from the use … Fair use will not destroy educational licensing in Australia. The claim that it would was specifically tested - and rejected - by the Australian Law Reform Commission. The education sector at the highest levels has given repeated assurances over a number of years that the sector would continue to enter into collective licensing arrangements if a fair use exception were to be introduced. (sub. DR429, p. 1) A key issue here appears to be about what is, and is not, paid for under the education statutory licence, and whether the licence should cover all copying by schools (but with a zero-rate charged for some activities), or whether the licence should only cover certain remunerable activities. Both the Copyright Agency Ltd, and the Copyright Advisory Unit for the COAG Education Council, representing the school sector, presented conflicting evidence to the Commission (box 6.4). 176 INTELLECTUAL PROPERTY ARRANGEMENTS Box 6.4 Coverage of the education statutory licence Copyright Agency Ltd (CAL) provided information on the operation and coverage of the education statutory licence. They note that their estimates are based on surveys of activity in statistical samples of schools and universities, carried out by an independent research company. According to CAL, there are two points at which uses are excluded from estimates of overall usage: • uses that are not recorded in the surveys in the first place; and • uses that are recorded in the surveys, but excluded from fee negotiations in accordance with protocols agreed between Copyright Agency and education sector representatives. In addition, overall discounts are negotiated for uses such as: • ‘small portions’; • copying from ‘blackline masters’; and • use of content that may lack sufficient ‘originality’ to be protected by copyright. (sub. DR510, Att. 1, pp. 1–2) However, the Copyright Advisory Group (CAG) to the COAG Education Council maintained that it continued to pay for uses of copyright material that should not be remunerated: Our schools are paying millions of dollars a year for uses that do not attract payment anywhere else in the world, and which do not cause any harm to rights holders. CAG estimates that payments for works in this category may represent approximately $9.16 million of total amounts that schools paid to Copyright Agency licence payments in 2014. (sub. DR429, p. 11) In the Commission’s view, adoption of a fair use exception appropriately recognises that some uses of copyright material, whether by the general community or the education sector (such as the examples provided above by the Copyright Advisory Group and in box 6.5), should not be remunerated. Future negotiations around the remuneration payable under the statutory licences would be informed by the fair use exception, recognising complete coverage of school copying is neither efficient, nor the purpose of the statutory licence scheme. Even so, fair use will not allow unrestricted and unremunerated copying by the education sector. The claims made suggesting fair use will eliminate the bulk of, or all, statutory licence fee revenue are implausible. Moreover, the purported impact on the education publishing sector ignores evidence from those jurisdictions where fair use and education publishing already coexist, or the fact that statutory licence fees make up only a small component of the sector’s revenue, with direct licensing and purchasing of content a substantial proportion. As the Copyright Advisory Group to the COAG Education Council submitted: Australian schools spend upwards of $700 million per annum in purchasing educational content for students. In addition to this, the sector pays approximately $90 million each year on collectively negotiated copyright licences. As repeatedly guaranteed by the sector, these licences will continue to exist in a fair use environment. Schools do not shy away from the fact however, that fair use would correct the current untenable position where millions of FAIR USE OR FAIR DEALING — WHAT IS FAIR FOR AUSTRALIA? 177 dollars of public funds are spent each year on public interest educational uses of orphan works, freely available internet materials, or non-harmful uses such as placing thumbnail images of book covers on a school intranet to show students what books are available in the school library. (sub. DR429, p. 2) Tales of the widespread demise of industries are just that Many rights holders made dire predictions about the impact of transitioning to fair use. In making their case, many pointed to the Canadian experience. For example, several publishers and other representative bodies argued Canada provided a case study on the impact that adopting fair use in Australia would have on educational publishing, including (among others) McGraw-Hill Education (Australia) (sub. 14), Cambridge University Press (sub. 22), UNSW Press Ltd (sub. 27), the Australian Copyright Council (sub. 36), Hachette Australia (sub. 41), the Australian Publishers Association (sub. 48) and the International Publishers Association (sub. 57). In part, this fear stems from the inclusion of education as an illustrative use in the ALRC’s recommended fair use exception (section 6.4 discusses illustrative use further). Typical of the observations made, Oxford University Press (which also has a Canadian presence) argued that a range of impacts flowed from Canada’s broader fair dealing exception, in particular: • A reduction in royalties paid by the education sector, directly impacting the financial viability of education publishing in Canada. • Closure, sale and bankruptcy of publishers, and relocation or rationalisation of distribution activities to the United States. (sub. 8, p. 1) Oxford University Press went on to argue: If similar revisions are made to the definition and terms of ‘fair dealing’ or ‘fair use’ provisions in Australia, Oxford University Press in this country will be exposed to huge financial risk, and its authors and creators will be significantly impacted. (sub. 8, p. 2) APRA AMCOS, Copyright Agency, Foxtel, News Corp Australia, PPCA and Screenrights commissioned an economic assessment of the potential costs and benefits of introducing a fair use provision in Australia, based on the Canadian experience. The Commission shares many of the concerns raised with the assumptions underpinning the PwC study (box 6.5 and appendix E). 178 INTELLECTUAL PROPERTY ARRANGEMENTS Box 6.5 Industry assessment of the costs and benefits of fair use APRA AMCOS, Copyright Agency, Foxtel, News Corp Australia, PPCA and Screenrights commissioned PricewaterhouseCoopers (PwC) to undertake an assessment of the ALRC’s proposed fair use exception. Specifically, the PwC report finds: … an increase in litigation costs, a reduction in Australian publishing and increased transaction costs for licensing. PwC’s findings are based, in part, on the experience in two countries where changes similar to those recommended by the ALRC were introduced: Singapore and Canada. In Canada the effects for Canadian educational [publishers] were severe: a 98% reduction in copyright fees for Canadian authors and publishers and the closure by a major publisher of its Canadian K to 12 publishing operations. PwC also finds no evidence to support offsetting benefits and no evidence that copyright impedes innovation. On the other hand, there is evidence that firms that use intellectual property are more successful and attract more investment than those that don’t, and that innovation is driven by certainty of regulation. (sub. 133, p. 1) The analysis has several significant shortcomings: • First, the report assumes the current copyright settings are optimal, and the balance between the incentives to creators and the costs to users is correct. However, the Commission’s analysis in the previous chapter shows that copyright is excessively long in duration and broad in its coverage. • Second, the report conflates fair use and third party use. While in Australia the ALRC has proposed that education be added to the list of illustrative fair use purposes, not all education purposes will be considered fair, and Australian courts will make judgements based on the facts of each case. • Third, significant contextual differences exist between the Canadian and Australian publishing industries, and it cannot be assumed that the market situation in Canada would be replicated in Australia. In particular, Australia is not bordered by the US, which houses the world’s largest English-speaking publishing industry, and was presumably able and willing to supply the Canadian market following changes in Canadian copyright laws. • Fourth, there is debate about the extent to which all of the declines in the Canadian publishing sector can be ascribed to changes in Canadian copyright law. • And finally, the cost-benefit benefit analysis was methodologically flawed. For example, it concentrated on potential impacts on publishing, ignoring the fact that fair use would apply to all of the copyright industries. The cost benefit analysis also implicitly assumed a closed economy model where transfers represent a redistribution of welfare between consumers and producers without a change in overall welfare. However, as a large net importer of copyright material, transfers from Australian consumers to foreign producers do affect community welfare. Sources: APRA AMCOS, Copyright Screenrights (sub. 133, att. 1). Agency, Foxtel, News Corp Australia, PPCA and Is Canada a good example of the impact of fair use? The Australian Digital Alliance responded to the analysis commissioned by APRA AMCOS, Copyright Agency, Foxtel, News Corp Australia, PPCA and Screenrights, FAIR USE OR FAIR DEALING — WHAT IS FAIR FOR AUSTRALIA? 179 questioning the reliance on Canada as the base case for expected changes, and outlining a number of other concerns with the findings, including: • the assumption fair use tilts copyright ‘away from creators’, when evidence suggests most fair use cases in the US are about subsequent (often transformative) uses of copyright material by follow-on creators, rather than copying of original works • the estimates of transaction litigation costs are based on faulty assumptions, including the assumption that collecting societies would cease to exist under fair use • the analysis is at best a partial assessment of changes occurring in the production, transformation and distribution of copyright material, and takes no account of dynamic, long-term outcomes from digital innovation (sub. 141). Other participants argued the Canadian case was not applicable to Australia, and the Canadian educational books sector was already struggling by 2012, and a range of other factors have contributed to the decline in revenue for the sector, limiting the lessons that can be drawn from the Canadian experience. In its submission, the National Copyright Unit, COAG Education Council addressed the points made by Oxford University Press, and highlighted some of the other factors that explained the changing fortunes of the Canadian education book sector. • In the Oxford University Press (OUP) 2013-14 Annual Report, copyright reform is not mentioned, however it does state that OUP’s decision to wind back its schools division in Canada followed ‘a decade-long decline in the Canadian market for educational resources during which purchases of materials have fallen by nearly 50 per cent.’ OUP added that the decision to wind back in the schools market does not affect the company’s other activities in Canada ‘including our market-leading Higher Education and ELT programmes.’ • OUP also asserts that the 2012 copyright reforms were the reason that Canadian educational publisher Nelson Education Ltd failed. And yet, an affidavit filed by Nelson’s chief executive officer in what were effectively bankruptcy proceedings, the company lists reduced spending on new curriculum by Canadian schools, increasing use of open education resources, the use of used textbooks, and the transition from traditional print books to digital products (which is said to be ‘having a transformative effect on the business’) as matters that adversely affected the company’s profitability. (sub. 97, pp. 41– 42) More critical again, academics from the Program on Information Justice and Intellectual Property at the American University Washington College of Law stated: Ultimately, evaluating the impacts of fair use, or any specific policy change, is hard work. The diffuse and forward-looking benefits of open exceptions like fair use may be hard to measure, but they are no less real. The PWC’s evaluation of the costs and benefits of fair use are not real. It is full of imagined horror stories that are unlikely to take place in fact and should be disregarded in their entirety. (sub. DR149, p. 3) In its analysis, EY considered the Canadian example, noting: Nevertheless, the loss of licensing income has been attributed in a recent PwC report as the trigger for the closure of a number of Canadian publishers and expected detrimental longer 180 INTELLECTUAL PROPERTY ARRANGEMENTS term impacts on incentives to create new content. Citing evidence from the Copyright Board of Canada, it is not clear that the introduction of a fair dealing exception in Canada has led to significant and widespread impacts of the sort suggested by PwC. (EY 2016, p. 48) PwC concluded that if a loss of licensing revenue and economies of scale, caused by the introduction of fair use, resulted in copyright collecting societies ceasing to operate, rights holders and users would face significantly higher administrative costs (sub. 133, att. 1, p. 24). However, EY’s analysis showed that for this to occur: A significant proportion (>86 per cent) of the relevant uses of copyright material managed by copyright Agency and Screenrights would need to be found to be ‘fair’ in order for the ‘education licensing income’ of both entities to fall below their ‘education operating costs’ … If this did occur, and the education statutory licences ceased to operate, the additional administrative costs borne by rights holders and users (i.e. relative to the status quo) would likely be significantly less than that implied by PwC. (EY 2016, p. 58) As the Commission noted in box 6.5, it is implausible that copyright collecting societies will cease to exist as a result of fair use, given they coexist with fair use in other countries, such as the United States. Fair use is no more uncertain than fair dealing Many participants suggested that by design, fair use is imprecise about the permissible uses of copyright material, and its adoption would create significant legal uncertainty for both rights holders and users. Putting the decision about which uses are fair in the hands of the court system necessitates litigation to determine the scope of infringements. Given the time and cost such court action entails, both rights holders and users might face some, at least initial uncertainties about the degree of protection afforded new uses. For example, the Copyright Agency stated: One of the consequences of ‘open-ended’ exceptions like the US ‘fair use’ exception is reduced certainty and predictability. The filing of copyright cases in the US is vastly greater (per capita) than that in Australia, and the fair use exception is raised in a significant (and growing) proportion of them. An analysis of copyright cases filed in 2014 showed that a defence of fair use was raised in 43% of the defended cases. (sub. 47, p. 3) Similarly, Harper Collins Australia commented on the potential need for litigation to establish the boundaries of any new exception: In brief, any such moves would introduce an unacceptably high degree of uncertainty and unpredictability to a currently stable and flourishing publishing landscape. In particular, such a move would likely lead to a sustained period of litigation while the contours of the new exception/s were established, providing both an unnecessary and expensive distraction from the real work of publishers — looking for new and innovative ways of creating, distributing and licensing creative material. (sub. 56, p. 5) Others argue Australia’s current copyright arrangements are far from certain. Weatherall, Alexander and Handler (sub. 99) highlighted a range of inconsistencies giving rise to legal FAIR USE OR FAIR DEALING — WHAT IS FAIR FOR AUSTRALIA? 181 uncertainty including differences in the way works and subject matter other than works are treated, issues with moral rights, needlessly complicated parallel importation restrictions and a lack of clarity around protection for performers. Australia’s current approach to exceptions embed a degree of uncertainty about whether new uses of copyright material are permissible, and mean the Act requires ongoing amendment to remain contemporary. Associate Professor of Law at the University of Sydney, Ms Kimberlee Weatherall, summed up the trade-off between certainty and uncertainty: … the application of legal principles to specific facts is always by definition uncertain. Our goal, in my view, should be that we are looking for a predictable legal framework not a certain one. As human beings we don’t demand certainty as we move through life. And as people and actors in the market and in public life we stake our fortunes daily on our ability to predict outcomes with some level of accuracy. A driver in a car, a doctor performing surgery, can't be certain about how the principle of negligence will apply, nor can a business be certain whether the ACCC will think their merger substantially lessens competition … In all of these various legal contexts where principles are applied we seem to be able to live and work with laws that provide a sufficiently predictable framework so that we can decide how to [be]have. And I don’t think copyright should be any different. (trans., p. 196) In any event, fair use is likely to be more certain than claimed. In its submission, the National Copyright Unit, COAG Education Council highlighted the similarities between the US’ fair use ‘fairness factors’ and the fairness factors contained within Australia’s exception for fair dealing for research or study (sub. 97. pp. 43–45) (see also table 6.3) suggesting Australian courts already apply principles to those contained within fair use. And in their submission to the ALRC inquiry, Hinze, Jaszi and Sag (ALRC 2014, sub. DR483) noted the application of fair use in the US was less uncertain than other participants had alleged: At a system level, the last 30 years of case law have generated a fairly coherent set of principles that lend themselves to forward-looking application. At the level of individual cases, it is true that no copyright expert agrees with every court decision on fair use, but we are not aware that such consensus exists in any other significant area of the law. (p. 3) In considering the issue of legal uncertainty in its analysis, EY noted the potential for some short-term uncertainty as industry and users adjusted to a more flexible environment, but concluded: Beyond these short-term impacts, it is unlikely that the fair use exception, by its inherent nature, would reduce certainty for rights holders and users compared to the status quo. This judgment is based on the observations that: (1) the status quo is relatively uncertain; (2) fair use, as a concept, is likely to be relatively predictable; and (3) the available evidence does not suggest that the relationship between fair use and increased enforcement costs is as strong as commonly assumed. (EY 2016, p. x) In the Commission’s view, legal uncertainty is not a compelling reason to eschew a fair use exception in Australia, nor is legal certainty desirable in and of itself. Courts interpret the 182 INTELLECTUAL PROPERTY ARRANGEMENTS application of legislative principles to new cases all the time, updating case law when the circumstances warrant it. To say otherwise would be to argue all laws should be prescriptive, which itself would be inconsistent with many laws across all social and economic arenas. In addition, even under a fair use regime, it is possible to specify a non-exhaustive list of illustrative uses which provides strong guidance to parties. As noted by the ALRC, the fairness factors contained within the exception, existing Australian case law, and other jurisdictions’ case law, would all provide guidance on the uses that might be considered fair, helping to reduce uncertainty. Fair use would better align the law with community expectations Several participants pointed to the disconnect between Australia’s copyright arrangements and community expectations and understanding about permissible uses of copyright material. The most notable divergence between the law and community understanding occurred prior to the 2006 addition of new exceptions for private time and format shifting, where the widespread practice of consumers recording television shows to a video recorder, or copying music from a CD to an iPod, was an infringement. And while the amendments purportedly meant the law ‘caught up’ with community expectations, in practice technology had already ‘leapt ahead’ of the new exception. Consumer association CHOICE highlighted a range of common practices that remain copyright infringements: CHOICE conducted a nationally representative survey in 2013, and found 8% of Australians were likely to have breached s109A of the Copyright Act by copying a CD or audio file that they own onto more than one personally owned device (e.g. a computer and an iPod). Among those who regularly use legal digital content, the number was 20%. 9% of consumers were breaching the law by copying a DVD or video file to at least one personally owned device. Section 10AA of the Copyright Act provides an exception for copying “cinematograph film” for private use. This narrow, technical exception refers to “videotapes” and consequently does not apply to DVDs or digital video files. 22% of consumers were using cloud storage services in 2013 to store copyrighted music and films, another act that breaches current copyright laws. These are common uses of copyrighted material, and are perceived by the public to be legal. 60% of Australians surveyed by CHOICE agreed that they should be able to transfer devices to as many devices as they own, with only 5% disagreeing. (sub. 26, p. 10) Copyright holders rightly point out that they do not pursue enforcement actions where parties make copies of DVDs and produce mash-ups in their garage. But this benign approach undermines the system by normalising infringement and adding to confusion — and ultimately runs counter to rights holders interests. More fundamentally, it should not FAIR USE OR FAIR DEALING — WHAT IS FAIR FOR AUSTRALIA? 183 be left to the discretion of rights holders as to whether or not to give effect to laws so out of kilter with community expectations. Fair use is more likely to ensure the law remains contemporary with community expectations. As noted by CHOICE: Under a fair use system, this behaviour and the acts described above would likely not breach the law. Instead of implementing yet more piecemeal reforms intended to address individual, specific problems like the “videotapes” issue, the introduction of a fair use exception would be a more flexible option. (sub. 26, p. 10) Fair use complies with international copyright law Despite the fact that a number of countries have introduced a fair use exception, including in the United States where it has been in operation for almost four decades, an argument often raised against the adoption of fair use in Australia is that it may not comply with the three-step test under international copyright law. The three-step test under the Berne Convention has become the international standard for assessing the permissibility of copyright exceptions generally (ALRC 2014). Under the test, limitations and exceptions must be confined to: 1. ‘certain special cases’ 2. which do ‘not conflict with a normal exploitation’ of the copyright material 3. do ‘not unreasonably prejudice the legitimate interests’ of the rights holder. As WIPO observed, terms such as ‘special’, ‘normal’ and ‘unreasonable’ are all open to interpretation rather than being absolute in meaning. The ALRC, as part of its review into Copyright and the Digital economy examined whether fair use was consistent with the three step test. They formed the judgment that it was consistent, based on the history of the test, an analysis of the words of the test itself, and on the absence of any challenge to the US and other countries that have introduced fair use or extended fair dealing exceptions. DFAT, in a submission to this inquiry similarly observed: While the specific details of any fair use exception would need to be reviewed in light of these commitments, DFAT’s view is that any fair use exception developed in Australia is likely to meet the ‘three-step test’. (sub. 144, p. 2) The Commission’s assessment The Commission considers there are firm grounds now, and even stronger grounds looking to the future, for amending the Copyright Act to replace Australia’s current exceptions 184 INTELLECTUAL PROPERTY ARRANGEMENTS with a broader fair use exception. The key policy question for Government should be how to design exceptions that maximise the net benefit to the community. Importantly, fair use would not replace payment for copyright works that are commercially available to users, but reinforces that user interests should also be recognised by Australia’s copyright system. Adopting fair use would benefit follow on creators and innovators, Australian consumers, schools, other education institutions, libraries and archives (appendix E). 6.4 What would a fair use exception look like? With respect to Australia’s prescriptive fair dealing exception, the Copyright Act makes clear: • which exclusive rights are covered by the exception • who can access the exception; such as whether it is open to all potential users or just certain users, such as libraries and archives • whether an exception requires the uses of copyright protected material to be for non-commercial purposes. Under fair use, these prescriptive requirements take a ‘back seat’ to the broader issue of whether a use is (or is not) considered fair. In designing a fair use exception the relevant questions become: • what factors are to be considered in determining whether use of copyright material is fair • whether there would be value in a list of illustrative uses that may qualify as fair use • whether third parties can rely on the exception. The fairness factors Australia’s exception for fair use should allow uses of copyright material that do not materially reduce a rights holder’s commercial exploitation of their works. This is in keeping with the underlying reason for copyright — to prevent copying of those works where free-riding reduces the incentive to invest. It is important for the fairness factors — the factors to be considered in determining whether use of that copyright material is fair — to reflect this underlying premise. Such an approach targets the effect of any use, rather than arbitrary categories such as the nature of the user or use. Consistent with this, the Commission’s draft report proposed four fairness factors the courts would consider when testing whether a use of copyright material interferes with the normal exploitation of the work. FAIR USE OR FAIR DEALING — WHAT IS FAIR FOR AUSTRALIA? 185 These were the: • effect of the use on the market or value of the copyright-protected work at the time of the use • amount, substantiality or proportion of the work used, and the degree of transformation applied to the work • existing commercial availability of the work • purpose and character of the use, including whether the use is commercial or private. Some participants commented on the specific formulation the Commission used in its draft recommendation on fair use. For example, several Australian copyright academics argued that the fairness factors proposed by the ALRC in its report would achieve the same ends: In our view the fairness factors identified by the ALRC allow all relevant considerations to be taken into account (including those of concern to both right holders and the Commission). In particular, issues of availability would clearly be of relevance when considering ‘the effect of the use upon the potential market for, or value of, the copyright material’. For example, a use is more likely to be fair if it involves use of an out-of-commerce work. Consideration of market harm and potential market harm appropriately protects the genuine economic interests of right holders, but does not require that any potential for a licence precludes fair use. (sub. DR505, p. 14) The ALRC, in formulating its fairness factors, deliberately based them on factors that are common to both the US fair use provision and the existing Australian provisions for fair dealing for the purpose of research or study (table 6.3). Hence, in addition to achieving the outcome the Commission had intended, the ALRC’s fairness factors have the added benefit of drawing on established jurisprudence and so reducing uncertainty. As a number of Australian copyright academics went on to observe: We prefer the drafting proposed by the ALRC (including, notably, of the fairness factors) which both addresses the Commission’s concerns and draws on Australia’s own wellestablished jurisprudence and that of other countries, and hence will contribute to ensuring sufficient guidance regarding the application of the exception. (sub. DR505, p. 13) 186 INTELLECTUAL PROPERTY ARRANGEMENTS Table 6.3 Fairness factors Reducing uncertainty by drawing on well understood concepts Proposed ALRC fairness factor Existing fairness factors for the purpose of research or studya Parallel under US Copyright law the purpose and character of the use the purpose and character of the dealing the purpose and character of the use, including whether such use is of a commercial nature or is for nonprofit educational purposes the nature of the copyright material the effect of the use upon the potential market for, or value of, the copyright material the amount and substantiality of the part used the nature of the work or adaptation the nature of the copyrighted work the effect of the dealing upon the potential market for, or value of, the work or adaptation in a case where part only of the work or adaptation is reproduced, the amount and substantiality of the part copied in relation to the whole work or adaptation the effect of the use upon the potential market for or value of the copyrighted work the amount and substantiality of the portion used in relation to the copyrighted work as a whole a An additional factor also provides for the possibility of obtaining the work or adaptation within a reasonable time at an ordinary commercial price. The Commission considers that the factors outlined by the ALRC are preferable. Like the ALRC before it, the Commission is not persuaded that because the factors are broad and principles-based they are inherently uncertain. By adopting a principles based approach the fairness factors are both inclusive and forward looking (aspects lacking in the current fair dealing exception). And, by drawing on established jurisprudence, they allow for consistency in treating analogous cases alike. How would the fairness factors be applied? In determining whether a use was fair, all four fairness factors would need to be considered and balanced. As in the US, no one factor would be more important than another. The factors would be, in a legal sense, rebuttable presumptions: default positions that could be overturned, depending on the evidence in a particular case. For example: • if the rights holder or licensee is exploiting the work in the Australian market, then subsequent uses are less likely to be fair • the greater the amount of a work that is used, the more likely the use will not be fair • the greater the degree of transformation, the more likely the use is in a different market to the original work and not interfering with the rights holder’s ability to exploit the work, and the more likely it will be fair • if a work is not being commercially exploited or is unavailable to consumers, or the rights holder cannot be identified, then use of the work is more likely to be fair. FAIR USE OR FAIR DEALING — WHAT IS FAIR FOR AUSTRALIA? 187 Illustrative uses While the Commission considers that fairness factors that draw on existing jurisprudence should go some way to reducing perceived uncertainty, a list of illustrative uses would also be helpful. Such lists are not intended to be exhaustive, or to single out any particular use as presumptively ‘fair’. Instead, whether an illustrative use is fair would be considered on a case-by-case basis, having regard to the fairness factors. As part of its proposed reforms, the ALRC recommended a number of new illustrative uses be added to the fair use exception. The proposed list of eleven illustrative uses is both comprehensive and consistent with comparative law in other jurisdictions. As noted by academics at the time of the review, the list is ‘very much in the tradition of s. 107 of the US Copyright Act: it tries to map the contours of fair use, without attempting to set its future boundaries’ (ALRC 2014, p. 150). While some of these illustrative uses cover some existing exceptions within the Copyright Act (such as non-commercial private use and incidental or technical use), others enable Australia to comply with its international obligations (such as access for people with a disability). The ALRC recommended the following illustrative uses as part of its fair use exception: • research or study • criticism or review • parody or satire • reporting news • professional advice • quotation • non-commercial private use • incidental or technical use • library or archive use • education • access for people with disability. Some participants highlighted the certainty a list of illustrative uses would provide for them. We are particularly supportive of the Commission’s endorsement of the ALRC’s further recommendation that library and archive uses be explicitly recognised in the non-exhaustive list of illustrative purposes for the exception. This will significantly increase legal certainty for libraries and archives wishing to make innovative and socially beneficial use of their collections. (sub. DR602, p. 2). 188 INTELLECTUAL PROPERTY ARRANGEMENTS The Commission considers that a new fair use exception should include the ALRC’s recommended non-exhaustive list of illustrative uses. Government use and public administration In its final report, the ALRC recommended a number of new exceptions for the purposes of government and public administration (box 6.6). Current arrangements for government use of copyright material are similar to those applying to education providers. As discussed in chapter 5, a statutory licence arrangement exists for government copying of copyright material, provided the copying is done for the services of the Commonwealth or State. When a government uses copyright material, it must notify the copyright owner (or its representative declared collecting society), and pay ‘equitable remuneration’. Governments currently pay an annual per-employee fee for the statutory licences. Australia also has a limited fair dealing exception for judicial proceedings, and for parliamentary libraries to copy material for members of parliament. Box 6.6 The ALRC’s recommendations on government use In its final report into Copyright and the Digital Economy, the ALRC recommended a number of new or expanded exceptions for government and public administration purposes: • Recommendation 15–1 The parliamentary libraries exceptions in ss. 48A, 50(1)(aa) and 104 of the Copyright Act should be extended to apply to all types of copyright material and all exclusive rights. • Recommendation 15–2 The Copyright Act should provide for a new exception for the purpose of the proceedings of a tribunal, or for reporting those proceedings. • Recommendation 15–3 The Copyright Act should provide for a new exception for the purpose of the proceedings of a royal commission or a statutory inquiry, or for reporting those proceedings. • Recommendation 15–4 The Copyright Act should provide for a new exception for uses where statutes require local, state or Commonwealth governments to provide public access to copyright material. • Recommendation 15–5 The Copyright Act should provide for a new exception for use of correspondence and other material sent to government. This exception should not extend to uses that make previously published material publicly available. Source: ARLC (2014a). A number of governments argued to both the ALRC and this inquiry that, similar to the case of education use, the current statutory licence arrangements result in governments paying for uses of copyright material that should not be remunerable (under current fair dealing arrangements). The NSW Department of Justice highlighted two such examples of how the current arrangements result in additional costs for taxpayers: display of copyright material (architectural plans, planning and engineering reports) as part of making FAIR USE OR FAIR DEALING — WHAT IS FAIR FOR AUSTRALIA? 189 development applications available online, and government datasets with embedded copyright material made available under open data policies. In commenting on these examples, the NSW Department of Justice stated: In both of these examples the use by government of the relevant copyright material is of considerable public benefit and would have little or no impact in terms of incentivising creation of new work. The vast bulk of government use of copyright material has similarly little or no impact on creation of new works. (sub. DR610, p. 2) The Department argued an Australian fair use exception should include government and public administration as an illustrative use: Intellectual property arrangements must be tailored in their application to government to ensure that governments are unconstrained in their ability to govern effectively in the modern electronic environment and that members of the public are able to effectively engage in government processes. It is essential that the copyright system facilitate adoption by governments of technological advances to optimise the efficient performance of their service provision and regulatory functions, and to promote government accountability and transparency. (sub. DR610, p. 1) In its report, the ALRC noted the way the Australian, state and territory governments access and deal with copyright material today is very different to the time when the statutory licence regimes were designed: Government use of copyright material has changed significantly in response to the emergence of digital technologies. Governments are much less likely to subscribe to hardcopy newspapers, books, journals and looseleaf services, and government officers are less likely to photocopy these items. Instead, governments subscribe to online libraries and media portals. Governments now receive large amounts of copyright material via email and online, scan and digitally store documents sent to them and email documents internally. Legislation and policy related to open government principles … means they are now more likely to publish material on external websites. The effect of these changes is that government use of commercially available material is more likely to be under direct licence. An increased amount of material is being used under the statutory licence, but most of it is not commercially available. (ALRC 2013, p. 331) The ALRC’s recommendations, similar to the inclusion of education as an illustrative use, sought to ensure Australia’s copyright arrangements would remain contemporary and efficient. Other countries have copyright exceptions for government purposes or public administration. The UK Copyright, Designs and Patents Act 1988 contains exceptions for parliamentary and judicial proceedings, royal commissions and statutory inquiries, material open to public inspection or on an official register, material communicated to the Crown in the course of public business, and for maintaining public records. The New Zealand Copyright Act 1994 contains very similar exceptions, drawing on the UK arrangements. 190 INTELLECTUAL PROPERTY ARRANGEMENTS Moreover, US courts have considered the application of fair use to government use of copyright material. In Williams & Wilkins Co. v. United States,1 the US Federal Court of Claims found government uses could be considered fair, but no automatic right exists for governments to use copyright material, and the fairness factors must be satisfied as in any other case. In the Commission’s view, fair use is likely to have similar impacts for government users as it will for education users, and the statutory licence regime will exist alongside (and be informed by) fair use. While some government uses of copyright material may be considered fair (and thus not remunerable under the statutory licence), others will not, and the statutory licence will apply in those circumstances. Even so, the Commission considers there are good grounds for also including ‘government and public administration’ as an illustrative use akin to the treatment of education users, to provide further guidance. Third party use Australia’s current exceptions can generally be relied upon by individuals, but not by other parties exercising an exception on behalf of an individual. For example, a student is able to copy copyright-protected material under the fair dealing exception for research and study. A teacher is able to direct students to go and make a copy for their research and study purposes. But a teacher is not able to rely on the fair dealing exception to make copies on behalf of students — such copies are covered by the education statutory licence scheme. The Copyright Advisory Unit to the COAG Education Council highlighted this issue at the Commission’s public hearings: So even though a student might be permitted under fair dealing to do a lot of the uses that we’re talking about here, because fair dealing is a purpose based exception it just doesn’t apply to anything the school does. (trans., p. 809) Similarly, individuals are able to record free-to-air television broadcasts for their private use at a later time. But in the 2012 Optus TV Now case 2, the Full Court of the Federal Court of Australia held that an online service that allowed consumers to record a free-to-air television broadcast for later streaming to a personal device could not rely on the exception for time shifting, because the exception only protected individuals undertaking the recording. Notwithstanding that these outcomes all resulted in the same things — consumers getting access to content — they were treated differently under the law. An exceptions regime that makes artificial distinctions between users and uses of material reduces the overall efficiency of the copyright system. In the Commission’s view, it makes little sense to legislatively preclude all third party uses of copyright material, even when such uses would meet the fairness factors. The key advantage of shifting to fair use is its 1 Williams & Wilkins Co. v. United States, 487 F.2d 1345 (Ct. Cl. 1973) 2 National Rugby League Investments Pty Limited v Singtel Optus Pty Ltd [2012] FCAFC 59 FAIR USE OR FAIR DEALING — WHAT IS FAIR FOR AUSTRALIA? 191 focus on the use of copyright material. Building in artificial distinctions runs contrary to the flexible nature of fair use. In its final report, the ALRC recommended that its proposed fair use exception not preclude applications to third parties, provided the general fairness factors are met: The ALRC considers that fair use is a suitable exception to apply to determine whether a third party use of copyright material infringes copyright. These third parties should not be precluded from relying on fair use. (ALRC 2014, p. 172) The Commission agrees with the ALRC approach to third party reliance on fair use. Scope for further reducing uncertainty The role of foreign jurisprudence In giving effect to any new fair use provision, Australian courts would be able to take account of existing Australian case law on fair dealing, as well as draw on foreign jurisprudence where doing so would assist judicial decision-making. While US court decisions would not be binding on Australian courts, the Commission sees no reason why Australian courts would not draw on the principles laid out in US decisions as a starting point. Indeed, this appears to be the approach taken in other jurisdictions, such as Israel, that have adopted fair use. Such an approach was explicitly considered by the Australian Government as part of the Raising the Bar reforms to patent law, where the explanatory memorandum to the legislative amendments stated that some of the concepts were adopted from, and were to be interpreted according to, UK and US developments (ALRC 2014, p. 155). Industry guidance Regulatory guidance and best practice information provides a further mechanism for reducing uncertainty. For example, the Copyright Advisory Office of Columbia University has developed a fair use checklist that guides users in considering how the fairness factors apply to their proposed use of copyright-protected works (Copyright Advisory Office, Columbia University 2016). The fair use checklist is now used widely across US academic institutions. Although not definitive, compliance with such guidance assists users in arguing to the courts that their use of copyright-protected works was in good faith. Similar guidelines have been introduced in Canada to inform education users on the application of fair dealing (Council of Ministers of Education, Canada nd). 192 INTELLECTUAL PROPERTY ARRANGEMENTS RECOMMENDATION 6.1 The Australian Government should accept and implement the Australian Law Reform Commission’s final recommendations regarding a fair use exception in Australia. 6.5 Fair use, orphan and unavailable works Orphan works are those works protected by copyright, but where the copyright owner cannot be identified. All types of works can be orphaned, including books, sound recordings, photographs, diaries, maps, and films. Libraries and archives tend to be the repositories where orphan works are found. Works can become an orphan by circumstance (such as the long passage of time), because identifying information is missing or deliberately removed (such as if metadata is stripped from an image used online), or because the potential user lacks the skills or tools to correctly identify the rights holder. Unlike orphan works, where the rights holder is unknown, for unavailable works the rights holder is usually known, but is choosing not to supply the market. While works often become orphaned due to the passage of time, copyright-protected works can be unavailable commercially quite soon after their original supply As the Commission discussed in chapter 4, the inability for users to access and use orphan and unavailable works reduces the efficiency of Australia’s copyright arrangements. How are other countries dealing with these issues? The Commission is not aware of any country that has fully resolved the issue of accessing and using orphan and unavailable works. Three broad approaches have been suggested and debated. Given libraries are the common repository for orphaned literary works, most approaches have been designed with books in mind (although where libraries hold other works such as photographs and sound recordings, these have been included). Ideally, any solution would be applicable to all copyright-protected works, since it is not only books that can become orphaned. The first approach is to permit use via a statutory licence, an approach adopted in several countries (box 6.7). Upfront licensing makes it clear to users which exclusive rights they can exercise with the orphaned works, and the financial cost of doing so (with fees typically in trust for any rights holders subsequently identified). FAIR USE OR FAIR DEALING — WHAT IS FAIR FOR AUSTRALIA? 193 Box 6.7 Licensing orphan works Several countries have attempted to tackle the problem of orphan works through licensing their use. Canada established a system in 1998 to allow those wishing to use an orphan work to apply to the Copyright Board of Canada for a non-exclusive licence, following ‘reasonable efforts’ to locate the rights holder. Licences are only available to published works. Following an application, the Board and the Canadian Copyright Licensing Agency set the royalty fee, terms and conditions of the licence, with fees held in trust for five years after the expiration of the licence. If the royalty is not collected by the rights holder at that time, the Licensing Agency may distribute the fee to other rights holders. In 2014, the UK Intellectual Property Office launched its Orphan Works Licence, allowing commercial and non-commercial users to apply for a non-exclusive licence to use an orphan work, following a diligent search for the rights holder. A fee is payable for a seven year licence, which can be renewed, and fees are held in trust should the rights holder be subsequently identified. To assist identification, a public register of all licensed orphan works is searchable online. The UK Intellectual Property Office has published guidance on what constitutes a ‘diligent search’ for a rights holder. Source: ALRC (2014). However, as the Australian Digital Alliance observed, international experience suggests that licensing schemes can be ineffective and inefficient: … statutory licensing and similar schemes that have been introduced in relation to orphan works by other countries have proven to be costly, inefficient and ineffective. Problems that have been experienced with such schemes include administrative costs that are higher than payments to authors; fees sitting unused because authors do not come forward; and comments from administrators that show the schemes are designed to tax users rather than to benefit creators. (sub. DR578, p. 8) The second approach for handling orphan works is to create an exception for their use, similar to the existing exceptions allowing the use of copyright-protected material. Creating a new exception for the use of orphan works allows the particular uses of the material to be prescribed (for example, only allowing non-commercial uses), or only allowing particular entities to benefit from the exception (such as libraries, archives and public broadcasters). In 2012, the EU issued a directive on how orphan works could be used by member states, with national legislation to be implemented by 2014. In practice, the directive creates an exception for public cultural institutions, including libraries, museums, public broadcasters and public archives to copy and communicate orphan works, if the institution undertakes a diligent search in good faith. Similar to the UK model, orphan works used in this way are registered on a public register, and rights holders can come forward and re-establish their rights. Compensation may be payable, but rules are left at the discretion of each country. 194 INTELLECTUAL PROPERTY ARRANGEMENTS The US Copyright Office has proposed a third approach to deal with orphan works (US Copyright Office 2015). The proposed approach limits the damages and remedies available where an infringement involves an orphan work — damages vary depending on the use and user of the work (box 6.8). APRA AMCOS argued that by their nature, a user of orphan works was unlikely to face a claim for infringement, and: Even if one were to bring such a claim, the person alleged to have infringed the rights in the work could avail him or herself of the protection in section 115(3) of the Act that states that where a defendant is not aware, and had no reasonable grounds for suspecting, that the act constituting the infringement was an infringement of the copyright, the plaintiff is not entitled to damages (but is entitled to an account of profits). Moreover, retroactive licences are available, such that a person could agree to a licence for past, unauthorised, uses once a rightsholder appears. (sub. DR404, p. 9) Box 6.8 Limited damages for use of orphan works The US Copyright Office has recommended several times a system of limited damages to address the issue of orphan works. In 2008, legislation for an orphan works scheme was introduced in the Congress, but was not successful. In 2015, the US Copyright Office put forward a revised model that limits liability for users of orphan works with the following features: • Application to all categories of works. • Application to all uses and users. Rather than being constrained to libraries, or only digital reproductions, or only for non-commercial purposes, the US Copyright Office’s proposal would allow anyone to exercise the exclusive rights in copyright on orphaned work, for both non-commercial and commercial purposes. • Requirements for limitation of liability. The proposed approach requires certain conditions be fulfilled for someone to benefit from the limitation of liability, including a good faith diligent search and registration of a ‘notice of use’ of the orphan work. Source: United States Copyright Office (2015). Very few countries have tried to solve the problem of out-of-commerce works, despite the benefits to would-be users. France established a compulsory collective licensing scheme for out-of-commerce books held by the French National Library in 2012. Relevant books are listed on a register managed by the Library. If an author or publisher does not lodge an objection within six months of a book being registered, a designated collecting society is able to licence the digital rights to the work to third parties. Registration of an objection triggers an obligation on the original publisher to bring the book back to market within two years. The register currently includes approximately 99 000 out-of-commerce works, with objections for around 2500 titles (US Copyright Office 2015, pp. 26–27). However, as a solution it only seeks to address those works held in the French National Library, rather than any work that is commercially unavailable, so is only a partial solution at best. FAIR USE OR FAIR DEALING — WHAT IS FAIR FOR AUSTRALIA? 195 Unlocking the value of orphan and commercially unavailable works In the case of orphan and out-of-commerce works, creators are not actively exploiting their creation in order to generate an economic return. Proposals to create licensing schemes, whereby consumers can pay to access such works, is one approach to unlocking their value, but is not preferred by the Commission. A licensing scheme would ultimately result in a windfall gain to rights holders (the original rights holder, or other rights holders when the original remains unknown), and would raise the costs to users, reducing the benefits from enabling access. The Commission considers it unlikely that a creator, prior to investing the time and effort in a new work, does so on the basis that their work will have an initial commercial life, a period ‘out of the market’, and a subsequent revival perhaps decades down the track. While this does occur for some successful works, it is largely by happenstance rather than design. Several participants argued fair use is an appropriate mechanism for addressing access to orphan and commercially unavailable works, and that the factors contained within a fair use exception were well adapted to considering the benefits and costs of access on a caseby-case basis. For example, the Australian Libraries Copyright Committee stated: The explicit inclusion of orphan works in the fair use exception would be of great benefit to libraries and archives, as it would allow them to make available to the Australian community thousands of collections items which are currently only available physically at the institution. However, it would be even more valuable to Australian society as a whole in that it would extend the ability to use these materials beyond the holding institution itself to their collaborators, their clients and the population at large, including private individuals, artists and businesses. It would truly enable these moribund works to find new life. (sub. DR602, p. 3) The National Archives of Australia also commented on the benefits of applying fair use to orphan and unavailable works: Fair use will address concerns around the availability of works after their commercial lifetime, whilst also maintaining the rights of those authors whose works do retain value. It will also allow all authors to retain rights to stop “unfair” uses of their work during their lifetime, even once the commercial value has diminished or disappeared. … We also strongly support the Commission’s recommendation that fair use be extended to orphan works … Making material available as a digital copy online is one thing, but if clients or users are unable to make further use of the material, its usefulness is limited (e.g. by family historians or genealogists). Fair use would provide a seamless experience: archives and libraries are able to provide access to orphan works → users are able to make use of them = benefits for all. (sub. DR525, pp. 1–2) In its final report, the ALRC noted that where a proposed fair use of copyright material involved an orphan work, that use was more likely to be fair: Some uses of orphan works can be expected to constitute fair use. Where use of an orphan work is for an illustrative purpose such as ‘quotation’, ‘research and study’, ‘reporting the 196 INTELLECTUAL PROPERTY ARRANGEMENTS news’, ‘criticism and review’ and ‘libraries and archives’, it is more likely to be fair. (ALRC 2014, p. 300) Moreover, the ALRC considered fair use to be particularly suited to addressing the concerns raised by participants such as the National Library of Australia: The ALRC expects that fair use would be particularly helpful to cultural institutions that are digitising or making available access to orphan works for non-commercial purposes, such as research or study. Cultural institutions suggested that they would be more confident relying on a fair use exception, rather than the exception under s 200AB when using orphan works. For example, the NLA considered that fair use ‘will provide workable solutions to many issues of providing access to orphan works’. (ALRC 2013, p. 300) The Commission concurs with the ALRC and considers fair use is well placed to deal with commercially unavailable works. The ALRC went on to note that, in the case of orphan works, a users’ liability should be limited if they undertook a ‘diligent search’ for the relevant rights holder. Such an approach is similar to that recommended by the US Copyright Office, and is in recognition of the fact that the use of an orphan work has minimal, if any, impact on the returns earned by rights holders. While Australia does not have a Government register of copyright material that could be searched, users seeking to benefit from the limited liability would need to demonstrate to the courts that the steps they had taken to try and identify the relevant rights holder were reasonable. This could involve searching for a rights holder online, searching copyright collecting societies or commercial databases. Such a requirement would also incentivise rights holders to improve the general availability of information covering the rights they hold. The Commission considers that adopting the ALRC’s additional recommendation would further improve access to orphan works without distorting incentives or harming rights holders. RECOMMENDATION 6.2 The Australian Government should enact the Australian Law Reform Commission recommendations to limit liability for the use of orphan works, where a user has undertaken a diligent search to locate the relevant rights holder. FAIR USE OR FAIR DEALING — WHAT IS FAIR FOR AUSTRALIA? 197 7 The patent system – getting the fundamentals right Key points • Patents can help advance human knowledge by encouraging socially valuable innovation that would not have otherwise occurred. This is especially so where the costs of innovation are high, but the costs of imitation are low. Patents can also underpin the development, commercialisation and wider diffusion of innovations. • Australia’s patent system fails the principles of a well-functioning intellectual property system. – Australia provides stronger patent rights than most other advanced countries. As a large net importer of patented technology, overcompensation is particularly costly for Australia. – The rules and processes for granting patents lead to a multitude of patented innovations of little or no community benefit, some of which are used for strategic purposes. • This frustrates follow–on innovators and researchers, raising the costs of innovation and imposing costs on technology consumers and the community. • Recent reforms improved the balance of the patent system, but further reform is needed. • While achieving a more effective and efficient patent system is constrained by international agreements, reform can be achieved. There is no policy silver bullet however: a package of reforms is needed. – Introducing an objects clause to the Patents Act would improve the likelihood that decisions on the application and design of the Act promote the public interest, and would ensure the system remains adaptable and fit-for-purpose. – Raising the inventive step for patents would ensure patented innovations are more likely to be socially valuable, and would help to address specific concerns in the areas of software and pharmaceutical patents. – Requiring applicants to identify the technical features of their innovation would improve patent decision making and make clearer the nature of a patented innovation for other innovators, helping them to undertake follow-on innovation and avoid infringement. – Raising and restructuring patent renewal and claim fees would reduce strategic use of patents, and better ensure only valuable patents are held in force. Application fees should remain unchanged to avoid adverse impacts on SMEs and individual inventors. A patent provides its owner with an up to 20 year exclusive right to commercially exploit an innovation. 1 In exchange the owner discloses the nature of the innovation. By preventing third parties from ‘free riding’ on innovative efforts, patents increase incentives 1 While the Patents Act 1990 (Cth) refers to ‘inventions’, in many places the Commission uses the broader concept ‘innovation’. THE PATENT SYSTEM — GETTING THE FUNDAMENTALS RIGHT 199 for firms and individuals to develop and commercialise innovations. Greater innovation benefits the community where it leads to the development of new goods and services and more efficient production methods. These benefits are greatest where the knowledge embodied in innovation disseminates to other parts of the economy (knowledge ‘spillovers’). Patents can also facilitate technology exchange, enabling firms to specialise in creating, developing and commercialising IP. This lowers barriers to market entry, enabling firms that possess only intangible assets to enter different parts of the innovation chain, improving efficiency and incentives to innovate, and increasing the salvage value of failed firms. Yet with the benefits of patent protection also come costs. The costs from administering a patent system are readily apparent. Less obvious are costs incurred when, by preventing a competitor from using a technology, patent protection increases market power and lessens competition. Where innovation builds on previous innovations, patent protection can also inhibit the advancement of knowledge through ‘follow-on’ innovation. The challenge for policy makers is designing a system that strikes the right balance between incentives to innovate and the costs from patent protection. This chapter assesses how well Australia’s patent system achieves the principles that underpin a well-functioning intellectual property (IP) system (chapter 2). It focuses on the allocation and definition of rights (section 7.1). These are fundamental dimensions of the patent system that determine its scope and strength and have an important bearing on the effectiveness and efficiency of patents. The chapter then outlines a strategy for reforming the patent system (section 7.2) and the specific measures recommended by the Commission to achieve a better balanced patent system (sections 7.3 to 7.6). The Commission has separately assessed the innovation patent system (chapter 8), pharmaceutical sector patents (chapter 10) and software and business methods (chapter 9). Enforcement of rights applies to all forms of IP, and is considered in chapter 19. A patent primer, which outlines the rights patents confer, what qualifies for protection and who uses the patent system, is set out in appendix F. 7.1 How well is the patent system performing? Allocating patent rights: is the system effective? In keeping with the broad principles that underpin a well–functioning IP system, the patent system should only grant protection to innovations that are: • socially valuable — innovations are of greater social value 2 where they result in new goods and services (in the case of product innovations) or production methods (process innovations) that improve the allocation of society’s limited resources. Social value is 2 Social value is equivalent to the total value generated from an innovation — the sum of private value that accrues to the innovator and any external value captured by others. 200 INTELLECTUAL PROPERTY ARRANGEMENTS higher where innovations advance human knowledge and create knowledge spillovers into other areas of the economy • additional — innovations are additional where they would not have been developed or commercialised absent patent protection (or would have been developed or commercialised at a later date). Patents for innovations that are not additional impose net costs on the community regardless of whether the innovation is socially valuable. Some inquiry participants recognised additionality as an important principle for granting IP protection (BCA, sub. 59; Moir, sub. 137; NSW Department of Justice, sub. 39; OSIA, sub. 21). Targeting socially valuable and additional innovations avoids the two potential problems in allocating patents — ‘false positives’ (cases where patents are granted where it is not in the public interest) and ‘false negatives’ (cases where patents are not granted, when to do so would be in the public interest). Commentary from participants and empirical evidence suggests that instances of false negatives are relatively rare. That is, to the extent that well-drafted patent applications for socially valuable and additional innovations are made, they are likely to be accepted and receive protection. Examples of significant patented innovations in Australia include: • Cochlear implant (Australian application number 1978041061) — an implant to assist hearing-impaired and deaf people. • Inkjet printer engine (201220372) — technology to enable high speed inkjet printing. • Bionic eye (2008232233) — technology to help restore some sight to people with particular forms of vision impairment. By encouraging socially valuable innovations that would not have otherwise occurred, the patent system plays an important role in advancing human knowledge. This is especially the case for technologies where the costs of innovation are high, but the costs of imitation are low. However, while the system captures the above sorts of innovations, there is evidence that it also admits a multitude of patents that are against the public interest; the system captures not just the ‘innovative wheat’, but also the ‘low-quality chaff’. The system fails to target socially valuable innovations While there are no ‘bright lines’ when it comes to identifying whether an innovation provides sufficient social value to justify patent protection, a range of proxy measures are available (OECD 2015a). The Commission constructed a number of such proxies (appendix D). Collectively, they suggest a significant percentage of Australian patents are of relatively low value (figure 7.1). THE PATENT SYSTEM — GETTING THE FUNDAMENTALS RIGHT 201 Figure 7.1 The bulk of Australian patents are of relatively low valuea Distribution of composite patent value index a The index accounts for the following proxies for patent value: forward citations, a ‘generality’ index, a ‘radicalness’ index, citations to non–patent literature and patent family size. These measures are defined and reported separately to the composite index in appendix D. The higher the value of the index the higher the social value. Source: IPGOD (2016 edition). This evidence is consistent with examples of patents accepted or granted in Australia (box 7.1). While some of the innovations in box 7.1 may be successful in the marketplace, this alone is unlikely to justify patent protection. To ensure the benefits of patent protection outweigh the costs, patented innovations must at least embody some advance in human knowledge, and ideally go further and create spillovers into other areas of the economy. The need to advance human knowledge is reflected in the TRIPS objectives, which are focused on advancing technology, rather than products per se. Further, from a practical viewpoint it is difficult to predict commercial success at the time an application is made. On the other hand, advances in human knowledge can, to some extent, be identified at the time an application is made by comparing the innovation with the prior art. Evidence on low-value patents is also consistent with the findings of Moir (2013), who assessed 72 business method patents granted in Australia and concluded that many of the factors integral to a patent being granted were trivial in nature, and that it was difficult to discern any new knowledge in any of the patented innovations (appendix D). While business method patents can no longer be patented in Australia (chapter 9), these findings are indicative of a system that admits patents of low social value. 202 INTELLECTUAL PROPERTY ARRANGEMENTS Box 7.1 Some Australian standard patents Some innovations involve advances on the prior art that seem trivial or aesthetic in nature, and thus are likely to do little to advance human knowledge or create knowledge spillovers. • ‘Patterned candle wick’ (Australian application number 2012351979) — a candle wick formed with a three dimensional pattern. • ‘Headband with 360-degree glitter pattern’ (2012286942) — a headband covered in glitter with non-slip and non-snag properties using a specific sewing method. • ‘Heating utensil comprising a non-stick coating with a three-dimensional pattern’ (2012216875) — a heating utensil with a layer comprising particles oriented to form a three-dimensional pattern. • ‘Pizza box with removable portions used as reward tokens’ (2008255242) — a pizza box with outer and inner wall panels, with perforated removable portions in the outer wall. • ‘A container’ (1997038350) — another pizza box, this one with a cardboard insert that enables the storing of two pizzas. • ‘Plant pots’ (2003202654) — a set of plant pots that lock together. • ‘Endoskeletal pet toy’ (2010202886) — a pet toy with an internal skeletal member and sound-emitting squeaker device. Other innovations appear to mainly involve combinations of pre-existing objects. • ‘Towel’ (2007282257) — a towel in a pouch, which attaches to the user’s clothes or belt. • ‘Air freshener with picture frame’ (2007235565) — a picture frame with a vapour permeable membrane that stores an air freshener. • ‘Bottle opener’ (2014301962) — a bottle cap that when used triggers a promotional or entertainment action such as a sound recording. • ‘Interactive toothbrush and removeable audio output module’ (2014200662) — a toothbrush that plays music for a period of time. • ‘Streamer ball’ (2002347430) — a ball with a stem connected to streamers. Source: AusPat. Australia is not alone in granting low-value patents. Proxies for the social value of innovations for the United States and Europe indicate a significant share of patents in these jurisdictions are of relatively low value (figure 7.2). This is consistent with the views of a number of researchers in other countries, who have concluded that the quality of patents is low, and in some cases getting lower (Bakels and Hugenholtz 2002; Graham and Harhoff 2014; Hargreaves 2011; OECD 2011b, 2015a; van Pottelsberghe and van Zeebroeck 2008). The OECD (2011b), for example, has estimated similar patent quality indicators across countries and over time, and found evidence that average patent quality has been steadily declining. THE PATENT SYSTEM — GETTING THE FUNDAMENTALS RIGHT 203 Figure 7.2 Europe and the United States also have a problem with low-value patentsa,b European Patent Office (EPO) US Patent and Trademark Office (USPTO) Composite patent value index a The index accounts for the following proxies for patent value: forward citations, a ‘generality’ index, a ‘radicalness’ index, citations to non–patent literature and patent family size. The higher the value of the index the higher the social value. These measures are defined in appendix D. b The EPO and USPTO patent value indexes include granted patents filed between 2000 and 2005. Data on the quality of USPTO patents is sourced from the OECD Patent Quality Database. The EPO patent value index is calculated by matching EPO patents to USPTO patents by PCT number and using OECD data on the quality of the corresponding USPTO patents. This approach is necessitated by the EPO and USPTO having different processes for calculating forward citations. Source: OECD Patent Quality Database (2016a). Nor does the system effectively target additional innovations An abundance of survey evidence shows patents are often not important for promoting innovation (appendix D). This evidence is supported by the results from empirical models of the relationship between patenting and R&D (a proxy for innovation). Collectively the evidence suggests that, at best, patents are only important for promoting innovation in products that entail large sunk R&D costs and are relatively easy to imitate, such as pharmaceuticals, biotechnology and machinery. It is impossible to directly incorporate an additionality test in the Patents Act, since additionality also depends on contextual factors (box 7.2). It is unsurprising therefore that the patent system fails to target innovations that would not have otherwise occurred in its absence. An innovation that passes the usefulness test, or is found to involve an inventive step, may still have been developed and commercialised in the absence of patent protection. 204 INTELLECTUAL PROPERTY ARRANGEMENTS Box 7.2 Additionality depends on the context A range of factors bear on whether there are sufficient anticipated returns to develop and commercialise an innovation in the absence of patent protection. • Technological factors — easily imitated technology increases the scope for free riding, reducing expected returns and the likelihood of investment. Larger upfront sunk costs from innovating increase investment risks in the absence of patent protection. • Market factors — market characteristics influence whether innovating firms can earn a sufficient return on investment in the absence of patent protection. Firms may earn a sufficient return without protection using alternative appropriation mechanisms such as secrecy, first-mover advantages, network effects and leveraging the technology with a complementary asset (such as a distribution network or brand–name reputation). Some market factors may increase innovation costs, for example if there is significant uncertainty about consumer demand. – In some markets firms may have an incentive to release information protected by IP. Making information freely available can prompt other (non–rival) firms to build on innovation, and can help to develop other revenue streams. • Other incentives — patent rights interact with other government policies aimed at increasing innovation. For example, patent protection may not be needed to promote innovation where R&D tax incentives are available. And where IP rights overlap, other forms of protection may be sufficient. Where innovation is publicly funded the main role of the patent system is to promote the transfer and commercialisation of IP rights, rather than induce innovation. The importance of these factors changes over time. In some sectors, such as biopharmaceuticals, the costs of innovation have increased in recent decades (Scherer 2011). The costs of imitation are likewise subject to change. Imitation costs, for example, are expected to fall from greater use of 3D printers (Harper et al. 2015; WIPO 2015h). In some industries lower imitation costs could reduce the effectiveness of lead time as an appropriation strategy. A poorly targeted patent system imposes substantial costs A multitude of low-value patents imposes substantial costs on the community. Low-value patents impede innovation by frustrating the efforts of follow–on innovators and researchers, who may be forced to invest in costly workarounds. In some cases, low-value patents are deliberately used as a strategic tool for stalling or excluding market entry (EC 2012). Consumers incur much of the costs from a multitude of low-value patents, either because new products and services are not developed, or because the higher costs of innovation are passed on. IPTA (sub. DR562) disputed many of the costs associated with low-value patents, arguing that innovators can simply innovate around them. IPTA’s view is at odds however with the evidence that patents inhibit follow-on innovation, especially in industries where innovation builds on previous innovations (box 7.3). In this context, low-value patents are not always necessarily found to be invalid (Lemley and Shapiro 2007), and even if they THE PATENT SYSTEM — GETTING THE FUNDAMENTALS RIGHT 205 would be found invalid if challenged, there are weak incentives to challenge a patent (Farrell and Shapiro 2008).3 Box 7.3 Inhibiting follow–on innovation: economic evidence Patent protection over a given technology can frustrate the efforts of subsequent innovators and researchers looking to build on the technology (follow–on innovation). Researchers have sought to derive causal estimates of the effects of patents on follow–on innovation. The evidence suggests patents are more likely to inhibit follow–on innovation in industries where innovation builds on previous innovations in an iterative and cumulative fashion. • Galasso and Schankerman (2015) study the effect that court invalidation of a patent has on subsequent innovation, as measured by citations of invalidated patents. Invalidation leads to a 50 per cent increase in citations on average. The evidence suggests that patent rights block innovation in computers, electronics and medical instruments, but not in drugs, chemicals or mechanical technologies. Using clinical drug trial data Sampat and Williams (2015) find no evidence that human gene patents affect follow–on innovation. In an Australian study, Christie et al. (2013) identify substantial investment in follow–on innovation by competitors in the pharmaceutical sector. • A related strand of literature examines the effect of patent thickets on market entry (which could have an indirect effect on follow–on innovation). In general these studies find that denser thickets inhibit entry, especially for SMEs (Hall, Helmers and Graevenitz 2015; IPO 2013). Cockburn and MacGarvie (2009, 2011) found evidence of a negative and significant relationship between thickets and entry in the software industry. Nor is it possible in many cases to innovate around clusters of patents, commonly known as ‘patent thickets’. Low-value patents can contribute to the development of thickets, which potential market entrants must ‘hack’ their way through to compete in a particular technology space. There is evidence that thickets inhibit market entry (box 7.3). Hall, Helmers and Graevenitz (2015), for example, find that increases in the size of thickets can decrease the likelihood of market entry by up to 20 per cent. In this way, patents can impede the competitive process, which is itself a driver of innovation. In the context of innovation patents, one patent attorney firm advises potential clients: By filing a series of innovation patents surrounding a product of a competitor, it is relatively easy and cost-effective to form a patent thicket around the product which subsequently makes it increasingly difficult and costly for the competitor to maintain freedom-to-operate. (Baxter IP 2016, emphasis in original) Overcoming patent thickets can be especially difficult for SMEs and potential market entrants (Cockburn and MacGarvie 2009, 2011; Hall, Helmers and Graevenitz 2015; IPO 2013). These firms may not have sufficiently sized patent portfolios (or the resources needed to acquire such portfolios) to enter so called ‘cross-licensing agreements’, which firms can use to mitigate thickets by licensing large parts of their patent portfolios to each other. 3 Incentives to challenge a patent are weaker where the potential challenger would incur all the costs from challenging, but would share the benefits with other firms in the relevant market. 206 INTELLECTUAL PROPERTY ARRANGEMENTS In some areas of technology in Australia patent thickets have grown dense. For example, in the area of mobile devices and networking, a dense thicket has developed within and around a set of patents held by firms including Sony, Ericsson and Samsung (figure 7.3). The Commission identified most patent thickets in the digital communication and basic materials chemistry technology fields. The full results from the Commission’s analysis of patent thickets, including the assumptions made, are outlined in appendix D. Figure 7.3 Schema of an Australian patent thicketa a The firms on either side of an interconnecting line cite each other’s patents — that is, each firm pair represents a bilateral patent relationship. The thicket is initially identified by the interrelationships between firms that are part of ‘triples’ — three firms that each hold patents that cite patents held by the other two firms (appendix D). The dark green circles denote firms that are either part of a triple relationship or a broader relationship that involves more than three firms. The light green circles denote bilateral patent relationships. The red interconnecting line indicates that the bilateral patent relationship includes at least one innovation patent. Sources: Commission estimates based on IPGOD (2016 edition) and unpublished IP Australia patent citation data. There are other costs from low-value patents that are incurred regardless of whether it is possible to ultimately innovate around them. First, low–value patents can impede innovation by contributing to ‘noise’ in the system. With more patents, it is more difficult for a follow–on innovator to be sure it is not infringing a patent, and to identify and build on true advances in human knowledge. The noise from low–value patents can also weaken the credibility signal in patents, increasing the rate of return required by financiers and making it harder for firms to leverage their patents to acquire capital at lowest cost. 4 This information problem can hinder access to finance for those innovators who have valuable patents (Fabrizi et al. 2013). An accurate signal value in patents can be particularly important for start-ups 4 Patents can be an effective instrument for reducing information asymmetries between patenting firms and outside investors (Baum and Silverman 2004; Conti, Thursby and Rothaermel 2013; Haeussler, Harhoff and Mueller 2009; Hsu and Ziedonis 2013; MacMillan, Siegel and Subbanarasimha 1985; Mann and Sager 2007). THE PATENT SYSTEM — GETTING THE FUNDAMENTALS RIGHT 207 and SMEs (Farre-Mensa, Hegde and Ljungqvist 2016; Greenberg 2013; Hottenrott, Hall and Czarnitzki 2016; Kou, Rey and Wang 2013). Atal and Bar put it thus: If a patent system allows many bad patents, the perceived quality of patents is low and so is the value of holding patents. This limits the ability of the patent system to reward true inventors. (2014, p. 504) Second, a poorly targeted patent system imposes costs at the system level, a view shared by the ACCC (sub. 35). The increased number of patents results in more inadvertent infringement, infringement studies, validity investigations and consultations with patent attorneys. With more spurious patents, disputes may be harder and more costly to resolve because courts have more difficulty determining which patents are justified (Jensen and Webster 2004). Defining patent rights: is the system efficient? The above discussion is about the binary choice of whether to grant a patent or not. Another important aspect of patenting is the strength of the rights it bestows — and the effects of those rights on the costs and benefits of patent protection. By increasing the profits a rights holder can earn from patent protection, stronger patent rights can potentially benefit the community by increasing incentives to develop and commercialise innovations. Stronger rights may also play a role in facilitating the diffusion of technologies within and across borders. On the other hand, stronger rights may hinder follow-on innovation and allow patent holders to exploit market power. Ideally, the patent system would set rights so as to balance the costs and benefits of protection. A number of policy levers (table 7.1) and market features (box 7.4) can affect this balance. Australia provides strong patent rights The general view among participants is that the standard patent system provides strong rights to patent holders. In addition to the 20 years maximum duration, a patent applicant can make an unlimited number of claims, providing greater scope to extend the boundaries of market protection. 5 Further tipping the balance in favour of patent holders is that compulsory licensing provisions are rarely invoked. Patent rights may also be strengthened by weak incentives to challenge a patent (Alphapharm, trans., p. 212; Farrell and Shapiro 2008). Patent rights are stronger than other forms of IP rights by some measures, such as the protection against independent discovery (unavailable to copyright holders). 5 Raising the Bar reforms may have reduced the scope to extend protection using claims. The use of omnibus claims (which refer directly to the description and/or drawings of an application) has been curtailed, and claims must now be fully supported by the description of the invention. 208 INTELLECTUAL PROPERTY ARRANGEMENTS Table 7.1 Elements of patent strength — standard patents Duration Breadth Legal Protection Usage rules How it affects strength Rights are stronger the longer the maximum duration of protection Rights are stronger the broader the scope of protections (often proxied by the number of claims, but also influenced by specification rules) Rights are stronger the more easily they can be legally enforced and the harder they are to challenge Rights are stronger the less scope there is for authorities to mandate exceptions for third party use of patented inventions Settings in Australia Maximum 20 years. Extensions of 5 years are available for some pharmaceutical patents Can include any Considered difficult to Compulsory licences number of claims. challenge due to court (CL) can apply in Applications with more costs. A granted limited circumstances. than 20 claims incur patent is not assumed ‘Crown use’ can be additional fees. Use of valid under the invoked for ‘omnibus claims’ are Patents Act.b government use. Use-it-or-lose-it rules limiteda Protection against independent discovery apply in limited circumstances.c Examples of constraints Minimum 20 years (TRIPS). Extensions for pharmaceuticals (AUSFTA) No restrictions on Minimum requirements on number of claims. damages (AUSFTA) Some standards on relationship between claims and description Can only revoke if CL would not remedy non-use (Paris Convention).d Licence rules (TRIPS).e Limits on grounds for CL and protection of undisclosed information (AUSFTA) Where considered Section 7.7 and chapter 10 Section 7.7 Chapter 15 Chapter 19 a Omnibus claims refer directly to the description and/or drawings of an application. b When deciding whether to grant interim injunctions courts have taken the view that because a patent has been granted following examination it is prima facie valid (Apple Inc. v Samsung Electronics Co. Limited [2011] FCA 1164 [at 28]). This is consistent with the AUSFTA requirement that Parties provide a rebuttable presumption that the patent is valid in ‘proceedings concerning the grant of provisional measures’. It is not clear if there is a presumption of validity in the substantive case — courts tend to re-examine patents de novo. In the United States the presumption of validity has a statutory basis. c A patent can be revoked under s. 134 of the Patents Act if it is not being exploited, but only after a compulsory licence has been in place for two years. d The Paris Convention is open to interpretation as to whether a compulsory licence must have been issued, or whether a country can revoke immediately if they consider a compulsory licence would be insufficient. e For example, licences must be nonexclusive and the patent holder must be paid adequate remuneration. THE PATENT SYSTEM — GETTING THE FUNDAMENTALS RIGHT 209 Box 7.4 The exchange of patent rights and market frictions A patented invention must be put to some use if it is to benefit the community. In some cases the party best placed to further develop or commercialise an invention is not its creator. The exchange of rights enables specialisation in different activities involved in creating, developing and commercialising IP. Such ‘vertical disintegration’ facilitates the entry of new firms that possess only intangible assets into different parts of the innovation chain, increases the salvage value of failed firms, and improves opportunities for firms to participate in global value chains. In some cases, the further development and commercialisation of technology may be the main or only role of the patent system. Where technology arises through direct public funding, patent licences can help to ensure the community receives the full benefits of the technology. The efficient exchange of patented inventions may be particularly important for SMEs and market entrants. Australian survey data suggests many SMEs rely on licensing their technologies to other parties for further development and commercialisation (Jensen and Webster 2006). The costs of trading patented inventions may be prohibitive. Transaction costs include the costs of searching for potential trading partners, negotiating licences, monitoring compliance with agreements, and taking infringement action if necessary. A number of public and private initiatives seek to reduce transaction costs by creating a platform for patent holders and potential users to interact, including patent pools and IP clearinghouses. Source IP is an Australian digital marketplace launched in 2015 aimed at making it easier for small businesses to access public sector innovation (IP Australia 2015j). The Society for University Lawyers (sub. 98) emphasised that universities are taking advantage of some of the above initiatives. According to various indexes of patent strength, Australia provides stronger rights than most other advanced economies (figure 7.4). Although these indexes do not account for every factor that influences patent strength, they do account for important factors such as the term of protection and standards for enforcement. Increases in the strength of patent rights in Australia have proceeded without robust evidence that they were in the community’s interests. Lawson argued that increases in the strength of rights have failed to comply with the Competition Principles Agreement.6 … legislated patent privileges, and in particular ‘stronger’ patent privileges … have failed to … [demonstrate] that the benefits of restricting competition to the community as a whole outweigh the costs, and that the objectives of the patent privileges can only be achieved by restricting competition. This demonstration is the founding principle articulated [by] the … Hilmer Committee and the subsequent codification of this principle in the Competition Principles Agreement. (2005, p. 8) 6 The Competition Principles Agreement establishes that the guiding principle in reviewing legislation that restricts competition is that such legislation should, among other things, not restrict competition unless it can be demonstrated that the benefits of the restriction to the community outweigh the costs. 210 INTELLECTUAL PROPERTY ARRANGEMENTS Figure 7.4 Australia provides relatively strong patent rightsa Taylor Wessing (2015) USCC (2016) Ginarte Park (2010) Papageorgiadis (2014-15) 0.00 Australia 0.25 0.50 High income countries 0.75 1.00 All countries a To better enable comparison, indexes are converted to the 0–1 range. High income countries are identified using World Bank categories. Sources: Papageorgiadis et al. (2015); Park (2010a); Taylor Wessing (2013); US Chamber of Commerce (2015). Long periods of protection are not always required As noted in table 7.1, one feature that affects the strength of patent rights is the duration of protection. TRIPS mandates that standard patent duration be at least 20 years. This provision required Australia to increase the maximum duration of protection from 16 years, including retrospectively for existing patents. 7 However, many patented inventions require less than 20 years protection. • Many product life cycles are shorter than 20 years (Bilir 2014). Drawing on patent citation data, Jaffe and Trajtenberg (1996) provide evidence that technological obsolescence is more rapid in electronic technologies than in chemical and mechanical technologies. Where innovators expect the economic use of their inventions will expire within 20 years, a maximum duration of 20 years is unnecessary to induce investment. • Survey evidence shows significant variation in revenue profiles for different pharmaceutical products, implying that the term of protection needed to induce investment can vary even within a sector (Harris, Nicol and Gruen 2013). • Some research suggests optimal patent duration is less than 20 years. Drawing on the results from a theoretical model and simulation analysis, Cornelli and Schankerman (1999) conclude that the range of optimal patent duration is between 8–15 years. 7 Analysis of the extension of patent term from 16 to 20 years under TRIPS concluded that the extension of rights to existing patents could result in a large net cost to Australia (IC 1996). Over half of the costs from extending term was attributed to the retrospective extension of existing patents. THE PATENT SYSTEM — GETTING THE FUNDAMENTALS RIGHT 211 About 15 per cent of standard patents reach their full term (figure 7.5). In the biotechnology, medical technology and pharmaceutical sectors more than 20 per cent of patents last the full term, while in the transport, macromolecular chemistry and environmental technology sectors the figure is about 10 per cent or less. 8 While there are limits to what policy conclusions can be drawn from these data (box 7.5), they do show renewal rates vary. Differences in the length of protection sought across technologies suggest there are benefits from a system that is flexible in the strength of rights provided. Figure 7.5 Share of Australian standard patents by patent lengtha a Standard patents granted between 1980–1995. Because most standard patents have a maximum term of 20 years, 1995 was used as a cut–off point to avoid truncation. The small number of patents that lasted longer than 20 years (due to receiving a pharmaceutical extension) are not included (chapter 10). Source: IPGOD (2016 edition). Rules governing claims allow for strategic use Another factor influencing the strength of patent rights is the breadth of claims (table 7.1). There is evidence some patent holders draft claims for strategic purposes. A sharp jump in the distribution of claims from 19 to 20 claims suggests some applicants seek to include as many claims in their applications as possible before having to pay higher fees, regardless of their merit (figure 7.6). In a guide on drafting US patent claims (which has higher fees for claims in excess of 20), Rosenberg (2014) notes: Given that 20 claims are allowed under the basic filing fee for a patent application, it is tempting to draft dependent claims that are not really necessary just to meet the twenty claim limit … these claims are worthless, since just the broad ‘seat’ in the independent claim already covers all of them. (p. 257, cited in sub. DR521) 8 Commission estimates based on IPGOD (2016 edition). 212 INTELLECTUAL PROPERTY ARRANGEMENTS Box 7.5 Interpreting patent renewal data Patent holders must decide each year if the benefits (or ‘option value’) from continuing to hold their patent exceeds the costs of renewal. Renewal data can therefore provide insights into the private value of patents, which can help inform discussions about the optimal maximum duration of protection. However, renewal data should be interpreted carefully. First, just because a small portion of patents reach the end of their maximum term does not mean the length of protection is too long. • The inherent risks from innovating mean most innovation efforts fail. It should thus not be surprising that most patents lapse within 20 years — this simply reflects the nature of innovation. Even if a patent lapses early, a 20 year term may still have been required to induce investment. For example, consider a firm looking to develop a new diagnostic tool for detecting cancer. The firm expects the tool’s economic life would exceed 20 years, and that about 20 years patent protection would be required to recover R&D costs. As it turns out, the tool is superseded in the marketplace after 10 years, and the patent is allowed to lapse. While the patent only lasted 10 years, the prospect of a 20 year term was required to induce investment. Second, it does not necessarily follow that a patent held in force provides net benefits to the community. The option value of holding a patent may include the potential to use the patent for strategic purposes, such as resurrecting and reinterpreting the patent to cover a technology subsequently developed by a competitor. It may also be unclear if the profits associated with a patent arise purely from the exclusion of competitors from the market, or because the technology has created value for consumers. With the patent criteria poorly targeted, there is a greater risk of profits arising purely from the exclusion of competitors than there would otherwise be. • Claims are more likely to be used strategically in some industries than others. The Commission compared the percentage point difference in the number of patents with 19 and 20 claims for different technologies (figure 7.7). On average, firms in IT–related industries (electrical engineering) have the biggest jump from 19 to 20 claims, and so appear more likely to use claims strategically. Firms in the chemistry industries appear less likely. Figure 7.6 Number of claims per applicationa a The chart displays the percentage of all patents granted between 2005–2015 by number of accepted claims. Applications with greater than 50 claims are excluded. Source: IPGOD (2016 edition). THE PATENT SYSTEM — GETTING THE FUNDAMENTALS RIGHT 213 Figure 7.7 IT industries are more likely to use claims strategicallya a The percentage point difference in the proportion of standard patents with 19 and 20 accepted claims for the period 2010–2015. The category ‘Other fields’ is excluded. Source: IPGOD (2016 edition). There is a strong case for reform While the patent system has a role to play in promoting socially valuable and additional innovations — especially in highly codified technologies that involve large sunk costs such as machinery and pharmaceuticals — it is clear that the system is poorly targeted and in many cases provides excessively strong patent rights. In an environment where the patent system is not effectively targeting innovations that provide net benefits to the community, overly-strong patent rights compound the costs associated with false positives. Longer rights increase the scope for unused patents to be opportunistically revived to capture infringing firms, which imposes costs without a resulting increase in innovation. As a large net importer of patented technology (appendix C), overcompensation in the strength of rights is particularly costly for Australia. And as a relatively small consumer market for technology, stronger rights in Australia do little to promote innovation by global firms. This is supported by evidence that increases in the strength of patent rights in other 214 INTELLECTUAL PROPERTY ARRANGEMENTS countries have failed to promote innovation Branstetter 2001). As Shelston IP observed: (Lerner 2000; Sakakibara and … numerous significant inventions of the last 100 years would have occurred whether or not Australia had a patent system. For example, many block buster drugs, the transistor, integrated circuit, radio, television, fertilizers etc, were invented by foreign research groups. In their commercialization of their products, Australia would have represented about 1-2% of their potential market. It is likely they would have been invented, with or without Australia having a patent system. (sub. DR483, p. 2) While the patent system has a role to play in facilitating the diffusion of innovation within and across borders, it does not follow from the available literature and evidence base that recent increases in the strength of IP rights in Australia would significantly increase technology transfer (chapter 3). The Commission’s overall judgment is that the community is not getting a reasonable return from the protection granted to patent holders. There is a strong case for reforming the patent system to better promote community interests. 7.2 A strategy for reforming the patent system On the basis of the evidence outlined above, the patent system should be reformed so as to improve the system’s: • effectiveness by better targeting socially valuable and additional innovations. This would help to separate the innovative wheat from the low-quality chaff. Doing so would free up space for firms to create and launch new products and services without concern for patents that entail little innovation • efficiency by introducing greater limits on and flexibility in the strength of rights. This would help to ensure the community does not incur costs from patent protection for no resulting benefits. No silver bullet: a package of reforms is needed There is no single policy reform that would fully achieve the Commission’s proposed reform objectives, and as outlined in table 7.1, international agreements restrict the scope of reform options. • Targeting the patent system at socially valuable and additional innovations relies on identifying practical and effective allocation mechanisms. – In principle, socially valuable innovations could be targeted by adapting the three criteria that a patentable invention must be novel, useful and involve an inventive THE PATENT SYSTEM — GETTING THE FUNDAMENTALS RIGHT 215 step (perhaps in tandem). 9 Reconsideration of the test for inventive step holds most promise given that passing the test requires some advance over the prior art — that is, some contribution to knowledge in the relevant field of endeavour. – Additionality is even more difficult to target. A specific test for additionality would be unworkable, and may contravene TRIPS, which mandates novelty, usefulness and the inventive step as the only criteria (other than subject matter eligibility) for granting a patent. Nor is there likely to be a single policy to effectively limit and tailor the strength of patent rights. The various factors influencing patent strength interact with each other in different ways, and are constrained by international agreements to different extents. • A solution to these challenges and constraints is to identify a package of reforms that collectively improve the effectiveness and efficiency of the patent system. The Commission assessed four fundamental areas of patent policy with scope to advance unilateral reform (table 7.2). Australia’s interaction with patent systems in other countries, however, means achieving optimal settings also requires international collaboration (chapter 18). Table 7.2 Policy tools for reforming the patent systema Policy tool Nature of the tool Main principle/s targeted Considered in this chapter Objects clause Overarching guide Effectiveness, efficiency, adaptability Patent criteria Screening mechanism Effectiveness Examination practices Screening mechanism Effectiveness Patent fees Upfront fees Screening mechanism Effectiveness Renewal fees Revelation mechanism Efficiency Claim fees, rules Screening mechanism Effectiveness, efficiency Where considered Section 7.3 Section 7.4 Section 7.5 Section 7.6 Considered elsewhere Compulsory licensing Revelation mechanism Efficiency, adaptability Legal mechanisms Screening mechanism Effectiveness Chapter 15, PC (2013a) Chapter 19 a Screening mechanisms are filters to patentability, and operate either ex ante (patent criteria, examination practices, upfront and claim fees) or ex post (legal mechanisms). Revelation mechanisms elicit decisions from either patent holders (renewal fees) or competitors (compulsory licences) that influence whether patented inventions are released into the public domain. 9 The manner of manufacture test can also be important as it is central to the patentability of software and business method patents. The manner of manufacture test was recently considered by the High Court in D’Arcy v Myriad Genetics Inc [2015] HCA 35, which ruled that gene patents over isolated DNA sequences did not meet the test. In a separate but concurring judgment, Justices Gageler and Nettle emphasised that, to qualify as a manner of manufacture, an invention must be something more than a mere discovery. Chapter 9 addresses the patentability of software and business methods. 216 INTELLECTUAL PROPERTY ARRANGEMENTS 7.3 Introducing an objects clause Consistent with the absence of overarching principles to guide IP policy, the Patents Act does not have an objects clause to guide legal interpretation. Many participants supported the principle of introducing an objects clause to provide greater guidance to decision makers involved in the design and application of the Act. 10 However, others raised concerns. 11 The idea of including an objects clause in the Patents Act is not new. In 2010, ACIP argued the Act should contain objectives outlining its purpose, and recommended including an objects clause to assist with the test for patentable subject matter (ACIP 2010b). The Australian Government accepted ACIP’s recommendation, but as yet has not implemented it. In 2013, IP Australia (2013c) sought interested parties’ views on the wording of an objects clause. Other countries including Japan, Korea, China and New Zealand have an objects clause or equivalent in their patents legislation. The New Zealand objects clause emphasises the Act should provide for an efficient and effective system that balances the interests of patent owners and society. The clause also emphasises the importance of ensuring patents are only granted in appropriate circumstances. The Japanese and Korean objects clauses (combined with legislative definitions), emphasise targeting the highly advanced creation of technical ideas, which — on the surface at least — appears to set a relatively high inventive hurdle. Would an objects clause improve the patent system? An objects clause would provide a number of benefits. It would help to ensure that decisions in the application and design of the Patents Act are consistent over time with a well–functioning IP system. The Commission (2013a) previously found an objects clause would help clarify the context for compulsory licensing, and the considerations that should guide a court. An objects clause could also be useful in underpinning decisions on whether to grant a patent. This is important given the scope for administrative and judicial interpretation to diverge over time from the intent of policy. As noted by IP Australia: The purpose of an objects clause is to set out the underlying purpose of a piece of legislation. As such, an objects clause could provide guidance to the community on the purposes of the Patents Act and assist the Courts in interpreting the legislation. (sub. DR612, p. 7) An objects clause would influence the granting of patents through the interpretation of the patent criteria, including the manner of manufacture test. Given it is impossible to construct direct tests for important concepts such as additionality, an objects clause would not be a 10 Alphapharm, sub. DR584 and trans., p. 210; Christie, trans., p. 450; Dent, sub. DR286; Lawson, sub. 7; Department of Health, sub. 84; Moir, sub. DR295; IP Australia, sub. DR612; Law Institute of Victoria, sub. DR558; Swinburne University, sub. DR557; Telstra, sub. DR316; University of Tasmania, sub. 61; Walter and Eliza Hall Institute of Medical Research, sub. DR571. 11 AIPPI, sub. DR551; Ausbiotech, sub. DR419; Medicines Australia, sub. DR529; IPTA, sub. DR562; Law Council of Australia, sub. DR490. THE PATENT SYSTEM — GETTING THE FUNDAMENTALS RIGHT 217 panacea for achieving effectiveness in the granting of patents. Nonetheless, at the margin, an objects clause would help to improve the likelihood that decisions align with policy objectives. In a recent case the High Court showed that it was willing to account for policy factors when considering the patentability of an invention. 12 The broad guiding principles in an objects clause would also help ensure the system remains adaptable and fit-for-purpose as technologies emerge and economies and business models evolve, and would help frame policy debates and reform. By enshrining the core economic principles that underpin a well-functioning IP system, an objects clause could help shield the system against further expansion in the scope and strength of rights, and guide disputes over the intent of future legislative change. While IPTA (sub. DR562) and the Law Council of Australia (sub. DR490) were generally opposed to introducing an objects clause on the basis that it would create confusion and scope for dispute, others considered that an objects clause would result in more, rather than less, certainty. The Law Institute of Victoria (LIV) argued: The LIV is in favour of incorporating an objects clause into the Patents Act. The LIV believes that having an objects clause in the Patents Act would encourage greater certainty and clarity in decision making regarding patent matters. (sub. DR558, p. 2) Even if there were some short–term uncertainty from introducing an objects clause, this would be preferable to the alternative scenario where outcomes are certain but impose net costs on the community. Overall, the Commission considers that the benefits from introducing an objects clause would exceed the costs. The wording of an objects clause In its 2013 consultation, IP Australia (2013c) outlined two options for an objects clause: • a proposal by ACIP (2010b) — ‘provide an environment that promotes Australia’s national interest and enhances the wellbeing of Australians by balancing the competing interests of patent rights holders, the users of technology, and Australian society as a whole’ • its own proposal — ‘provide an environment that enhances the wellbeing of Australians by promoting innovation and the dissemination of technology and by balancing the competing interests of patent applicants and patent owners, the users of technology and Australian society as a whole’. In the draft report, the Commission recommended that an objects clause should describe the purpose of the Patents Act as enhancing the wellbeing of Australians by providing patent protection to socially valuable innovations that would not have otherwise occurred. This prompted substantial commentary from participants. 12 D’Arcy v Myriad Genetics Inc [2015] HCA 35 [at 28]. 218 INTELLECTUAL PROPERTY ARRANGEMENTS • Several argued for amendments to the proposal, including emphasising that the system is intended to protect technological innovations (Moir, sub. DR295), mentioning inventors (Walter and Eliza Hall Institute of Medical Technology, sub. DR571), and recognising the role of the patent system in facilitating the further development and commercialisation of inventions (Dent, sub. DR286). • Some were concerned the proposal would require an assessment of social value and that this would make the objects clause unworkable, especially with respect to legal interpretation by the courts. 13 IPTA said it could live with an objects clause if it were ‘plain vanilla’ (trans., p. 675). Targeting socially valuable and additional innovations is a necessary condition for the patent system to meet the principles of a well-functioning IP system. At the same time, an objects clause should be workable and articulated as simply as possible. These principles could be reflected in an objects clause by drawing on the simpler wording in the TRIPS objectives. Some interested parties have argued that an objects clause should be modelled on the TRIPS objectives, or reflect its principles (ACIP 2010b; AusBiotech 2013; Generic Medicines Industry Association and GMIA 2013; IPTA 2013). In responding to IP Australia’s options for an objects clause, IPTA (2013, p. 2) argued that the ‘interests of intellectual property right holders, users of technology and society as a whole are more correctly and clearly articulated in the TRIPS statement of objectives’. Following further analysis and input from participants, the Commission has refined the proposed objects clause to more closely reflect the TRIPS objectives. To better ensure the patent system meets the principles of a well-functioning IP system, the objects clause should make clear that the principal purpose of the patent system is to enhance the wellbeing of Australians by promoting technological innovation and the transfer and dissemination of technology. In so doing, the patent system should balance the interests of producers, owners and users of technology. RECOMMENDATION 7.1 The Australian Government should incorporate an objects clause into the Patents Act 1990 (Cth). The objects clause should describe the purpose of the legislation as enhancing the wellbeing of Australians by promoting technological innovation and the transfer and dissemination of technology. In so doing, the patent system should balance over time the interests of producers, owners and users of technology. 13 Australian Industry Group, sub. DR582; CSIRO, sub. DR575; FICPI Australia, sub. DR581; IP Australia, sub. DR612; Law Council of Australia, sub. DR490; Qualcomm, sub. DR345; Telstra, sub. DR316; University of Sydney, sub. DR566; University of Technology Sydney, sub. DR564; Webber, sub. DR447. THE PATENT SYSTEM — GETTING THE FUNDAMENTALS RIGHT 219 7.4 Reforming the inventive step To satisfy the criteria for a standard patent an invention must be a ‘manner of manufacture’, novel, useful, and involve an inventive step (appendix F). The inventive step plays a critical role in ensuring the patent system targets socially valuable innovations. The greater the advance on the prior art, the more likely an innovation advances human knowledge and promotes knowledge spillovers. Ideally, the inventive step would help distinguish highly valuable from less valuable innovations. There are four key elements to the inventive step that influence how difficult it is to pass the test: the definition of the invention itself, the prior art, the ‘person skilled in the art’, and the test for obviousness (figure 7.8). Figure 7.8 The inventive step has four key elementsa a The full wording of the test is outlined in box 7.6. Sources: Patents Act and Australian Patent Office Manual of Practice and Procedure. 220 INTELLECTUAL PROPERTY ARRANGEMENTS Recent reform raised the threshold closer to other countries The first three elements of inventive step outlined above were reformed as part of the Raising the Bar initiative. These reforms: • broadened the allowable prior art considered as part of the test. Before the reforms, prior art documents were limited to only those that would have been ‘ascertained, understood and regarded as relevant’ by the skilled person • expanded the assumed background knowledge of the skilled person against which the prior art is assessed. Previously the skilled person was assumed to reside in Australia • required the specification of the invention be clear and complete enough for it to be performed by the skilled person, and that the claims be fully supported by the description. Application outcomes following Raising the Bar suggest the reforms have gone some way to narrowing the grant rate between IP Australia and the European Patent Office (EPO) (table 7.3). There are good reasons for looking to the approach in Europe, with numerous studies showing the EPO is more effective at filtering out low-value patents than patent offices in other large markets for technology (Graham and Harhoff 2014; van Pottelsberghe 2010; de Rassenfosse, Jaffe and Webster 2016; Scellato et al. 2011). While the evidence in table 7.3 may overstate the impact of the reforms, 14 Raising the Bar was clearly significant, and moved the inventive step (and other elements of patent law) in the right direction. Table 7.3 Raising the Bar reforms show encouraging early signsa Applications that received a first report and were deemed resolved by both IP Australia and the EPO Outcome Pre Raising the Bar Post Raising the Bar IP Australia EPO IP Australia EPO Withdrawnb 6 579 0 2 178 5 856 307 2 594 953 0 145 907 13 178 Implied grant rate (per cent)c 75.1 66.9 86.8 82.6 Granted Rejected Grant rate differential 8.2 4.2 a Raising the Bar reforms apply to applications that file an examination request on or after 15 April 2013. Applications subject to the reforms and filed through the PCT were matched with the corresponding EPO application. The status of matched EPO patents was sourced from the European Patent Register. This table only includes applications where the application was deemed ‘resolved’ (granted, refused or withdrawn) by both offices, and a first examination report was received from both offices. b ‘Lapsed’ patents in IPGOD have been categorised as ‘Withdrawn’. c Number of granted patents divided by the sum of granted, rejected and withdrawn patents. Grant rates are higher post Raising the Bar because insufficient time has passed for some of the more difficult applications to be resolved. As more time passes, the grant rates will decline. Sources: Commission analysis drawing on IPGOD (2016 edition) and European Patent Register. 14 While post Raising the Bar applications that clearly meet the criteria are resolved quickly and thus have likely been granted in both offices, there has unlikely been enough time for more ‘line ball’ applications to be resolved in both offices. To the extent Australia is still more likely to accept such applications compared to Europe the post Raising the Bar grant rate differential is expected to increase over time. THE PATENT SYSTEM — GETTING THE FUNDAMENTALS RIGHT 221 Other countries still fall well short of the ideal threshold … While Raising the Bar edged Australia closer to the thresholds applied in other countries, some participants argued insufficient time has passed to assess Raising the Bar’s impact, and that the patent system should not undergo further reform at this time (Davies Collison Cave, sub. DR417; IP Australia, sub. DR612; IPTA, sub. DR562; Law Council of Australia, sub. DR490; Webber, sub. DR447). However, even if IP Australia’s claim that its practices would ultimately align with that of Europe held true, there would be a case for further raising the inventive threshold. This is because the thresholds applied in other countries still fall short of the ideal. Since the draft report, the Commission estimated patent value indicators for the United States and Europe (figure 7.2). These indicators show that the bulk of patents in the United States and Europe have a low value. This evidence accords with the literature. Researchers in other developed countries have found evidence that the quality of patents in their systems is low, and in some cases it is getting lower (Bakels and Hugenholtz 2002; Bessen and Meurer 2008; Graham and Harhoff 2014; Hargreaves 2011; OECD 2011b, 2015a; van Pottelsberghe and van Zeebroeck 2008). Indeed, the problems associated with a low inventive step appear to have led the US Supreme Court to broaden the circumstances under which an application can be rejected on the basis of the test.15 In the Commission’s public hearings, Gruen argued: … the net exporter, the clear net exporter of IP, the United States, is not pursuing its own interest in any coherent transparent sense. It is pursuing the interests of certain rights holders and it is entirely happy to pursue the interests of those rights holders at the expense of the country itself, quite apart from the expense of us. (trans., p. 723) Delaying further reform until Raising the Bar changes fully play out is both unnecessary and costly. … and Australia still fails to meet the standard in Europe Despite the Raising the Bar reforms, Australia still has a lower threshold for inventive step compared to Europe, which differs in a number of important ways. The minimum advance on the prior art is lower In Australia, the High Court has described the minimum advance on the prior art required to meet the obviousness test as ‘a scintilla of invention’. 16 The scintilla standard is highlighted by some patent attorneys and referenced in IP Australia’s Manual of Practice and Procedure. In giving effect to the test, the High Court has held that the inventive step is 15 KSR v Teleflex 550 U.S. 398 (2007). 16 Lockwood Security Products Pty Ltd v Doric Products Pty Ltd (No 2) [2007] HCA 21 [at 52]; Meyers Taylor Pty Ltd v Vicarr Industries Ltd [1977] HCA 19 [at 54]. 222 INTELLECTUAL PROPERTY ARRANGEMENTS not met if the skilled person would be led directly as a matter of course to try a particular approach. 17 The High Court linked this approach to the ‘taking of routine steps’. 18 During its consultations the Commission heard that by abandoning the ‘obvious to try’ 19 approach to assessing inventive step, the High Court’s decision in Aktiebolaget Hassle v Alphapharm put Australia behind the level of the inventive step applied in Europe. A number of legal authorities have highlighted the lower threshold in Australia created by the scintilla standard. The High Court noted that the UK may require more than a scintilla. 20 A former Australian High Court Justice pointed to the unique scintilla standard in Australia (Crennan 2007). And the Royal Court of Justice in the UK commented on the lower obviousness threshold in Australia. 21 It is perhaps noteworthy that currently Australian courts seem to be taking a very pro-patent view of obviousness and that patents are being upheld there which are not upheld elsewhere. IPTA (sub. DR562, p. 3), on the other hand, downplayed the role of the scintilla standard, arguing that once a claimed invention meets the threshold ‘there is nothing to be gained by discussing how much the threshold is exceeded’. IPTA drew an analogy to a pole vaulter, saying it does not matter how much the bar is cleared, just that it is cleared. However, in keeping with IPTA’s analogy, it is important to recognise that a scintilla does not describe the amount by which a pole vaulter must clear the bar. Rather, a scintilla sets the bar — and does so at a level that even pole vaulters of questionable ability can clear. The inventive step fails to target technical innovations Another distinguishing feature is that in Australia, inventions may pass the inventive step (and novelty test) on the basis of ‘non-technical features’ in the claims. This risk arises because the invention as a whole is assessed for the purposes of inventive step and novelty. By contrast, in Europe these tests are based only on the technical features in the claims. Assessing the invention as a whole makes it harder to reject applications where the invention is based on a mix of technical and non-technical features. Indeed, it appears that in Australia patents can be granted to inventions that fail to embody a non-obvious advance in technology (box 7.1). Non-technical inventions are less likely to advance human knowledge or generate spillovers into other areas of the economy. It could be argued that some or all of the patents in box 7.1 embody a non-obvious product. This, however, runs counter to the role of the patent system, as reflected in the TRIPS objectives. The protection and enforcement of IP rights should contribute to the promotion of technological 17 Aktiebolaget Hassle v Alphapharm Pty Ltd [2002] HCA 59. 18 Aktiebolaget Hassle v Alphapharm Pty Ltd [2002] HCA 59 [at 58]. 19 The obvious to try approach asks whether a course of action required to arrive at the invention or solution to the problem would have been obvious for a person skilled in the art to try with a reasonable expectation of success. 20 Aktiebolaget Hassle v Alphapharm Pty Ltd [2002] HCA 59 [at 48]. 21 Angiotech Pharmaceuticals & Anr. and Conor Medsystems Inc [2007] Court of Appeal A3/2006/0657 [at 43]. THE PATENT SYSTEM — GETTING THE FUNDAMENTALS RIGHT 223 innovation and to the transfer and dissemination of technology, to the mutual advantage of producers and users of technological knowledge and in a manner conducive to social and economic welfare, and to a balance of rights and obligations. (Article 7, emphasis added) Australia’s wording is more complex While not directly relevant to the inventive step threshold, the legislative wording of the inventive step in Australia is more complex than analogous provisions in other jurisdictions (box 7.6). The Australian Industry Group (sub. DR582, p. 4) said the test in Australia is ‘overly complicated and has been described by members as a ‘nightmare’ to interpret and apply in practice’. While other patent offices, such as the EPO, have equally complex provisions in their examiners manuals, Australia appears to be alone in enshrining such complexity in its legislation, making it more difficult to adapt the application of its law as knowledge, technologies and markets evolve. A gap remains Consistent with the different thresholds applied in Australia and the EU, a gap remains. Australia still has a materially greater propensity to grant patents when the EPO does not (table 7.4). It is also unclear if the Raising the Bar reforms addressed the tendency for IP Australia to grant broader claims (and thus allow for broader market protection) than the EPO (Christie, Dent and Lim 2013; Christie, sub. DR580; trans., p. 439).22 Table 7.4 A closer look at application outcomes suggests further reform is needed Applications that received a first report and were deemed resolved by both IP Australia and the EPO, and where there is a different outcome across the offices Pre Raising the Bara Post Raising the Bar per cent per cent IP Australia grants, but EPO does not 67 65 EPO grants, but IP Australia does not 33 35 Application outcome a Applications where an examination was requested between 15 April 2011 and 30 December 2012. 22 In the Commission’s public hearings Christie (trans., p. 449) argued that Raising the Bar was unlikely to have addressed the difference in the scope of claims between Australia and Europe. 224 INTELLECTUAL PROPERTY ARRANGEMENTS Box 7.6 Comparing legislative tests for inventive stepa Australia An invention is to be taken to involve an inventive step when compared with the prior art base unless the invention would have been obvious to a person skilled in the relevant art in the light of the common general knowledge as it existed (whether in or out of the patent area) before the priority date of the relevant claim, whether that knowledge is considered separately or together with the following information: (a) any single piece of prior art information; or (b) combination of any two or more pieces of prior art information that the skilled person could, before the priority date of the relevant claim, be reasonably expected to have combined. Canada The subject–matter defined by a claim in an application for a patent in Canada must be subject-matter that would not have been obvious on the claim date to a person skilled in the art or science to which it pertains, having regard to prior art information. European Patent Office An invention shall be considered as involving an inventive step if, having regard to the state of the art, it is not obvious to a person skilled in the art. Japan A patent shall not be granted for an invention where a person ordinarily skilled in the art of the invention would have been easily able to make the invention based on the prior art. New Zealand An invention involves an inventive step if it is not obvious to a person skilled in the art, having regard to any matter which forms part of the prior art base. United Kingdom An invention shall be taken to involve an inventive step if it is not obvious to a person skilled in the art, having regard to all matter in the prior art base. United States A patent for a claimed invention may not be obtained if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. a These are not exact excerpts from the relevant acts. For clarity and brevity they have been modified by omitting definitions of the prior art and combining articles and provisions. Sources: Convention on Grant of European Patents; Leahy–Smith America Invents Act 2011 (US); Patents Act 1977 (UK); Japan Patent Office English interpretation of Examination Guidelines for Patent and Utility Model in Japan; Patents Act 2013 (New Zealand); Patents Act 1990 (Cth); Patent Act 1985 (Canada). The overall threshold should be raised There is a robust case for raising the inventive step threshold. Patent systems in other jurisdictions admit a multitude of low-value patents, and thus are insufficient benchmarks for policy reform. This suggests a strong case for setting the threshold above the benchmarks set by other countries. The inventive step threshold — at a minimum — should meet the highest threshold set by any country with which Australia conducts THE PATENT SYSTEM — GETTING THE FUNDAMENTALS RIGHT 225 substantial technology trade; a principle highlighted in numerous reviews (Harris, Nicol and Gruen 2013; IP Australia 2009; IPCRC 2000). Measures for raising the inventive step To raise the inventive step, all policy options within the constraints of international agreements should be considered. Raise the minimum advance on the prior art Australia should align the obviousness test with the approach in Europe, where a similar ‘problem-solution’ approach to assessing inventive step is used,23 but a higher threshold applies. Many participants supported setting the inventive step at or above the level applied in Europe.24 The Commission is not swayed by arguments that adopting the European approach would fail to materially raise Australia’s inventive step, or that there are risks in seeking to replicate the EPO on one aspect of law (Law Council of Australia, sub. DR490). • Adopting the European approach would clearly raise Australia’s inventive step. Australia still grants patents for applications rejected by the EPO. As a key remaining difference in Australia’s application of inventive step, the obviousness test threshold plays a part in this. As noted, the UK Royal Court of Justice highlighted the role the obviousness test plays in patents being granted in Australia, but not elsewhere. One provider of attorney services said the European test is ‘the most difficult obviousness test that I experience as a day-to-day practitioner’ (Peter Treloar, trans., p. 764). And the Law Council of Australia (trans., p. 652) itself acknowledged differences in the application of the law across the two jurisdictions. • The Commission agrees that European law cannot be simply replicated and imported into Australian law. But this can be addressed by modifying the EPO wording to suit the definitions and provisions in the Australian Patents Act. Key concepts (‘invention’, ‘inventive step’, ‘prior art base’, ‘not obvious’ and ‘person skilled in the art’) are all either defined or understood in the Act or in case law. To raise the threshold, the Patents Act should be amended by borrowing from the simpler European wording, such that an invention is taken to involve an inventive step if, having regard to the prior art base, it is not obvious to a person skilled in the relevant art. The Explanatory Memorandum should emphasise that the obvious to try test applied in Europe 23 The problem-solution approach is based on the question of whether the claimed invention would have been obvious to a person skilled in the art when faced with a particular problem. 24 Australian Industry Group, sub. DR582; Alexander et al., sub. DR505; Alphapharm, sub. DR584; DIGI, sub. DR528; GBMA, sub. DR396; Moir, sub. DR295; IGBMA, sub. DR526; Law Institute of Victoria, sub. DR558; Microsoft, sub. DR420; Swinburne University, sub. DR557; Telstra, sub. DR316; University of Melbourne, sub. DR560. 226 INTELLECTUAL PROPERTY ARRANGEMENTS would in some instances be a suitable test.25 FICPI Australia (sub. DR581) suggested one way to ensure judicial and administrative bodies apply legislative intent to raise the threshold would be to identify case law that is no longer to be applied in determining inventive step. To this end, the Explanatory Memorandum should also state that a ‘scintilla’ of invention, or a scenario where the skilled person would not ‘directly be led as a matter of course’, are insufficient thresholds for meeting the inventive step. Target technical features The inventive step and novelty tests should better target an innovation’s technical features. This would help ensure the patent system targets technological innovations, especially where innovations are based on a mix of technical and non-technical features, or mainly involve obvious and/or trivial combinations of pre-existing objects. Better targeting the technical features of an innovation aligns with judicial decisions. In assessing the manner of manufacture test, the Full Federal Court considered whether an invention made a technical contribution (chapter 9).26 And in assessing inventive step, the same court considered whether a variation on the prior art amounts to more than ‘mere mechanical equivalents or workshop variations’,27 and said there must be ‘some difficulty overcome, some barrier crossed’. 28 Where a variation on the prior art solves no problem, overcomes no difficulty or serves no functional purpose, these tests are not satisfied. Targeting technical features would also align with the Commission’s proposed objects clause, which emphasises the purpose of the patent system is to promote technological innovations. The Australian Patent Office Manual of Practice and Procedure should be updated to emphasise that examiners will consider the technical features of an invention for the purposes of inventive step and novelty. Work with other countries to achieve the ideal threshold Introducing the above reforms would likely exhaust options for raising the inventive step unilaterally. However, given the weight of evidence that patent systems in Australia and overseas are out of balance, these reforms would leave the inventive step below the ideal level. Most compelling in this respect is the evidence that Europe still has an issue with low-value patents, a view shared by the former chief economist of the EPO (van Pottelsberghe and van Zeebroeck 2008). What would a higher threshold for inventive step entail? One option canvassed by participants is requiring a significant advance on the prior art (along the lines in the 25 This is the approach taken by the Boards of Appeal of the EPO in Retinoids v Kligman (T 149/93) and Refrigerants v EI du Pont (T 1877/08). 26 Research Affiliates LLC v Commissioner of Patents [2014] FCAFC 150 [at 114]. 27 RD Werner & Co Inc v Bailey Aluminium Products Pty Ltd [1989] FCA 37 [at 76]. 28 RD Werner & Co Inc v Bailey Aluminium Products Pty Ltd [1989] FCA 37 [at 39]. THE PATENT SYSTEM — GETTING THE FUNDAMENTALS RIGHT 227 Explanatory Memorandum for Raising the Bar). This sort of approach is not without precedent. China requires an advantageous technical effect compared to the prior art, while courts in Switzerland and Germany in the past adopted a similar approach (Romandini 2016). Brazil has a Bill pending that would reform its inventive step to require a significant technical advance. Pursuing unilateral reform of this nature would entail risks. There are mixed views on whether such a test would contravene TRIPS (Global Academics 2013; Romandini 2016) or a ratified TPP. 29 Such reforms could increase transaction costs for international patent applicants in Australia, since it would apply an inventive step substantially different to that applied elsewhere. Collaboration with like-minded countries would mitigate these problems. The Australian Government, with input from IP Australia, should work with other countries to examine ways to further raise the overall threshold for the inventive step. Chapter 18 considers avenues and mechanisms for pursuing international cooperation on IP policy. What impact would raising the threshold have? Raising the threshold would have some transitional costs. There would be some adjustment to a new legal provision, and a potentially greater risk of ‘hindsight bias’ (where a decision maker decides an invention is not sufficiently inventive because it is obvious in hindsight). However, if there is a risk of hindsight bias it would be incumbent on patent applicants to prove otherwise — there should be a high burden of proof on firms and individuals requesting patent protection. The benefits from raising the threshold would be significant. It is estimated that raising the threshold would result in about 700–800 fewer low-value patents granted each year (equivalent to about 4.5 per cent of annual patents granted). 30 Over time, this would materially reduce the patent stock, as existing low-value patents expire. Given current expiry rates, after 10 years the patent stock would be reduced by about 6700 patents. Rejecting less valuable innovations would better ensure the benefits of patent protection outweigh the costs and, by improving the signal value in patents, would help reduce capital costs for genuine innovations. The overall environment for innovation would be improved, making the Australian economy more innovative and productive. Raising the threshold would also help to address specific concerns with software and pharmaceutical patents (chapters 9 and 10). 29 TRIPS says that Member States may deem ‘inventive step’ to be synonymous with ‘non-obvious’. The TPP text says the same, but also goes on to say ‘In determinations regarding inventive step (or nonobviousness), each Party shall consider whether the claimed invention would have been obvious to a person skilled or having ordinary skill in the art having regard to the prior art’. 30 These estimates account for Raising the Bar reforms. The Commission’s full workings and assumptions are outlined in appendix H. 228 INTELLECTUAL PROPERTY ARRANGEMENTS FINDING 7.1 The Raising the Bar initiative moved the inventive step and other elements of patent law in the right direction by raising the threshold for granting a patent. There is a strong case, however, for further raising the threshold. RECOMMENDATION 7.2 The Australian Government should amend ss. 7(2) and 7(3) of the Patents Act 1990 (Cth) such that an invention is taken to involve an inventive step if, having regard to the prior art base, it is not obvious to a person skilled in the relevant art. The Explanatory Memorandum should state: • a ‘scintilla’ of invention, or a scenario where the skilled person would not ‘directly be led as a matter of course’, are insufficient thresholds for meeting the inventive step • the ‘obvious to try’ test applied in Europe would in some instances be a suitable test. IP Australia should update the Australian Patent Office Manual of Practice and Procedure such that it will consider the technical features of an invention for the purpose of the inventive step and novelty tests. 7.5 Improved decision making for granting patents In addition to the role of legislation and case law, assessment procedures have an important bearing on patent allocation. A number of studies find that patent offices in other countries often grant patents that should have been rejected under the relevant country’s patent law (GAO 2013; Palangkaraya, Webster and Jensen 2011; de Rassenfosse, Jaffe and Webster 2016). Decisions on whether to reject or grant a patent rely on a significant amount of information. Decisions on inventive step (and to some extent novelty) are particularly information-intensive, requiring the examiner to: • determine the relevant prior art, which can include documents and information publicly available through doing an act (this is integral to novelty) • form a view of the common general knowledge and skills held by the hypothetical person skilled in the art • identify and disregard any inessential features of the invention and, where relevant, identify the particular problem that the invention solves. While the information available to examiners has improved with the digitisation of patent records and greater collaboration between patent offices, in many cases a patent applicant will have a better understanding of the above factors than the patent examiner. Overcoming THE PATENT SYSTEM — GETTING THE FUNDAMENTALS RIGHT 229 the information asymmetry between applicants and patent offices remains a key challenge in achieving improved decision making. It is therefore important to identify ways of improving the information available to examiners while minimising the costs incurred by patent offices and applicants. Options for eliciting better information from applicants Rules regarding the construction of claims International experience provides ready options for improving the information available to examiners. The EPO for example, can request applicants to: • specify their claims in two parts: (1) the prior art relevant to the specific claim; and (2) the features of the invention that add to the prior art. This is intended to allow the patent examiner to clearly see which features of the invention, and in particular the individual claims, are part of the prior art • identify in their set of claims the technical features of the invention. This enables the EPO to more easily ensure that patent protection is only made on the basis of technical features, and allows follow-on innovators to identify the core technical element of the patent claim. The EPO also requires patent applicants to describe their invention in a way that makes clear the technical problem that the invention is intended to solve, and in this context, state any advantages of the invention over the prior art. Requiring applicants to explain why their invention is non-obvious A third option for improving the information available to examiners is to require that patent applicants provide information on what the inventive step is, and in so doing explain the problem that their invention solves and why it is non-obvious (Lawson, sub. 7; 2008b). It is argued that this requirement would help to better identify the inventive concept for which the applicant is claiming protection, and would provide insight into what someone in the field of technology thought at the time of the application (Lawson 2008b). There was mixed feedback on the merits of these options Some participants supported the principle of requiring patent applicants to provide more information. Moir (sub. DR295) argued the onus should be on the applicant to clearly articulate the core inventive contribution of their invention. In expressing its support for the above measures, OSIA (sub. DR486) suggested that if an applicant does not know the prior art, they should not be applying for a patent. Barr (sub. DR521) said the additional burdens involved do not seem onerous. Other participants were opposed (IPTA, trans., pp. 684–685; Qualcomm, sub. DR345; University of Sydney, sub. DR566). Qualcomm claimed there would be a burden from requiring applicants to specify their claims in two-parts. And IPTA said it is impossible for 230 INTELLECTUAL PROPERTY ARRANGEMENTS an applicant to know what prior art its application will be assessed for non-obviousness against, and highlighted the burden this would impose. Which option is best? All three of the above options would help clarify the link between the prior art and market protection being sought, and the potential basis for legal challenge and evidence for parties in any court proceedings. It could be argued that requiring an applicant to explain why its invention is non-obvious goes further than the other two options, and thus provides greater benefits. However, these benefits would be limited since in responding to an application IP Australia can already ask an applicant to explain why its invention is non-obvious (IP Australia, sub. DR612). Relying on the applicant to explain non-obviousness in the first instance could also lead to ‘benchmarking bias’, where the applicant’s explanation conditions the subsequent search behaviour of the patent office, as well as the behaviour of follow-on innovators. In this context, the usefulness of this approach relies on accurate disclosure by applicants and their knowledge of the prior art. While all three options would largely achieve the same outcomes, they would impose very different informational burdens. Requiring applicants to identify the technical features of their invention would impose minimal burden. This is because the applicant can easily observe the technical features of their invention (to the extent there are any). In contrast, the remaining two options impose much higher burdens, particularly for start-ups and SMEs. • As participants highlighted, not all applicants have a detailed understanding of the prior art — this will depend on the industry and the sophistication and resources of the applicant — and so requiring claims in the two-part form would increase costs for some. • Requiring applicants to articulate why their invention is non-obvious could impose an even greater cost burden, given it could require input from patent attorneys. Given the scope for dispute, this could also delay patent proceedings. Since this requirement does not exist in other countries (WIPO 2015d), it could also raise the costs of applying for a patent in Australia relative to other countries. Not only is requiring applicants to identify the technical features of their invention the least costly option, it also better aligns with the Commission’s broader reform agenda. As concluded in section 7.4, the Commission considers it important that the inventive step and novelty tests be based on the technical features of an invention. Patent applicants should therefore be required to identify the technical features of their invention. THE PATENT SYSTEM — GETTING THE FUNDAMENTALS RIGHT 231 RECOMMENDATION 7.3 IP Australia should reform its patent filing processes to require applicants to identify the technical features of the invention in the set of claims. 7.6 Making better use of patent fees Another policy lever for improving the patent system is the structure and level of patent fees. Making better use of patent fees would be consistent with Australia’s international agreements and, because fees are sector and technology neutral, would not compromise the adaptability of the patent system. Economic research points to the significant role that patent fees can play in achieving policy objectives. • Renewal fees that increase more than proportionally with patent age help to ensure only valuable patents are held in force, reduce economic rents, and limit the risk of patents being resurrected and reinterpreted to cover technology that was not originally contemplated (de Rassenfosse and Jaffe 2015; de Rassenfosse and van Pottelsberghe 2010, 2013). – While patent trolls are not a current concern in Australia, making better use of fees could help the system to mitigate these sorts of issues. There is evidence that such strategic uses of patents are more likely to occur late in the patent term (Love 2013). • Claim fees (and associated reforms to rules concerning the specification of claims) can help reduce the number and scope of claims, decreasing the time taken to review applications and limiting the breadth of market protection (Harhoff 2016; van Pottelsberghe and Francois 2009; de Rassenfosse and van Pottelsberghe 2013). • Other upfront fees (such as application and examination fees) can, in combination with other instruments such as examination intensity, act as a filter for excluding low-value inventions from the patent system (de Rassenfosse and van Pottelsberghe 2010, 2013; Schankerman and Schuett 2016). There is scope to better use patent fees as a policy tool The structure of patent fees in Australia suggests greater scope to use fees to promote efficiency. • 232 While IP Australia seeks to structure upfront and annual renewal fees to promote innovation (IP Australia 2012a), renewal fees only increase in three stages across the life of a standard patent. Other jurisdictions such as the United Kingdom have a more steeply increasing schedule of renewal fees (figure 7.9). Australia had a similar INTELLECTUAL PROPERTY ARRANGEMENTS escalating renewal fee structure up until 2006. A banded structure was adopted to make it easier for applicants to plan and budget for renewal fee payments (IP Australia, pers. comm., 17 March 2016). • Applications with more than 20 claims incur a flat $110 fee for each additional claim. This differs from the approach in Japan and South Korea, which charge an additional fee for every claim (IIP 2014). As well as having additional fees for more than 20 claims, the United States charges an additional fee for each independent claim (claims that do not reference other claims) after the number of independent claims surpasses three (USPTO 2016). The EPO charges claim fees over 15 claims, with a higher rate for additional claims over 50. Renewal fees by year patent in force, selected jurisdictions Figure 7.9 Expressed as a percentage of highest renewal fee (i) Australiaa (ii) European Patent Office 100 100 80 80 60 60 40 40 20 20 0 2 4 6 8 10 12 14 16 18 20 0 2 (iii) United Kingdom 100 80 80 60 60 40 40 20 20 2 4 6 6 8 10 12 14 16 18 20 (iv) United States 100 0 4 8 10 12 14 16 18 20 0 2 4 6 8 10 12 14 16 18 20 a On 10 October 2016 Australia’s renewal fees are planned to increase from $500 to $550 in years 10–14, $1120 to $1250 for years 15–19, and $2300 to $2550 for years 20–24 (IP Australia 2016d). Sources: Websites of the relevant patent offices. THE PATENT SYSTEM — GETTING THE FUNDAMENTALS RIGHT 233 How should patent fees be structured? Consistent with research findings and practice in other jurisdictions, renewal fees in Australia should increase more than proportionally with patent age. Renewal fees should be kept low in the early years of a patent while there is still likely to be a high level of uncertainty regarding the commercial value of the patent (Lanjouw 1998; Pakes 1986; Schankerman and Pakes 1986). The structure of claim fees should also be reformed. This could be advanced by reducing the initial threshold for claim fees from 20 to 15, and introducing a much higher claim fee for claims above 50. It is important to guard against the risk that patent applicants respond to a lower threshold for claim fees by drafting longer and more complicated claims (IP Australia, sub. DR612). However, new restrictions on omnibus claims and requirements that the claims be fully supported by the description of the invention (both introduced under Raising the Bar), limit the ability for applicants to broaden the scope of protection within fewer claims. Even so, it would be sensible to monitor outcomes. Other upfront fees such as application and examination fees should not be raised to provide an additional screening mechanism. Part of the motivation for raising upfront fees in other jurisdictions is to address patent backlogs, which Australia does not have. And some researchers highlight that SMEs and start-ups may have less capacity to absorb higher patent fees than larger firms (Hargreaves 2011; de Rassenfosse and van Pottelsberghe 2013). What impact would restructuring fees have? While some participants argued that making better use of fees would have limited effect on the number of patent applications, others contradicted this view. FICPI Australia (sub. DR581) recognised the role that renewal fees play in deterring patents only valued by the patent holder. And in the Commission’s public hearings, Qualcomm said higher patent fees in Australia would prompt it to reassess any low-value patents: … if the fees were raised that would just mean that we would probably file less patent applications in Australia. We already have a pretty rigorous program to try to identify the value of each individual invention and corresponding patents. So we would probably file those patents and we’d probably discontinue maintaining some of the lesser value ones in Australia. (trans., p. 669) The impact of higher fees on patent numbers can be estimated by using measures of the price elasticity of demand. 31 The Commission considered the impact of three alternative renewal fee structures on the stock of patents (table 7.5). 31 An ‘elasticity’ is an estimate of the percentage change in a variable that arises from a one per cent increase in another variable. In this context, an elasticity indicates the percentage change in demand for patents that arises from a one per cent increase in fees. 234 INTELLECTUAL PROPERTY ARRANGEMENTS Table 7.5 The impact of higher fees depends on the assumptionsa Current and potential renewal fee schedules Description % change to patents in stock Elasticity scheduleb Low Medium High IP Australia’s planned increasec –1.17 –1.56 –1.94 Fees steadily increase from year 11, reaching $1 800 in year 20 –2.18 –2.91 –3.64 Fees steadily increase from year 5, reaching $2 200 in year 20 –3.21 –4.28 –5.35 2200 Current Potential 1650 1100 550 Renewal fees, $AUD 0 5 10 15 20 2200 1650 1100 550 0 5 10 15 20 2200 1650 1100 550 0 5 10 15 20 Patent year a The Commission took the current stock of patents, distributed across years in force, and applied elasticity estimates for each corresponding year. The analysis thus only provides a long-term perspective of potential changes in fees, nor does it account for changes in future applications due to economic conditions or historic trends. b The medium elasticity schedule is from de Rassenfosse and van Pottelsberghe (2013). Elasticities increase for each year a patent is in force, ranging from -0.01 in year 1 to -0.8 in year 20. The low and high elasticity schedules are 75 per cent and 125 per cent of the medium elasticity schedule. c Planned to take effect 10 October 2016. Fees will increase from $500 to $550 in years 10–14, and from $1120 to $1250 in years 15–19. Sources: Commission estimates using information in IPGOD (2016 edition) and de Rassenfosse and van Pottelsberghe (2013). While far from precise, the evidence suggests patents could be decreased in the long term by about 2–5 per cent under scenarios where most of the increase in fees occurs in the later years of a patent. Importantly, these relatively modest increases in fees, concentrated towards the end of the patent term, would not be expected to materially impact innovation incentives. Short-lived technologies would bear a much smaller burden in fees relative to longer lived technologies, as would Australian patent holders, who are less likely than THE PATENT SYSTEM — GETTING THE FUNDAMENTALS RIGHT 235 international patent holders to hold patents for the full 20 year term (IP Australia, sub. DR612). Limiting the increases in renewal fees to later in the term of patents would also help to shield SMEs from increases in fees. The Commission also estimated the impact of lowering the threshold for claim fees from 20 claims to 15 claims, and doubling the claim fee for claims above 50. The impact of these changes would depend on to what extent patent applicants reduce the number of claims. The Commission considered two scenarios. • Only Australian applicants reduce their claims — Australian applicants (about 8 per cent of total successful applicants) with between 16 to 20 claims reduce their number of claims to 15, and the number of Australian claims over 50 reduces by 10 per cent. 32 Under this scenario the total number of claims requiring assessment decreases by about 3 per cent.33 • Australian and international applicants reduce their claims — Some international and Australian applicants with between 16 to 20 claims reduce their number of claims to 15, and the total number of claims over 50 reduce by 10 per cent. 34 Under this scenario the total number of claims requiring assessment decreases by about 7 per cent. Ensuring fees achieve broader policy objectives A potential barrier to achieving the above reforms is IP Australia’s status as a cost recovery agency. While the Australian Government’s cost recovery guidelines provide some flexibility to set fees to achieve broader policy objectives, increasing fees well above current levels ‘would be outside the cost recovery framework and may have policy and legal implications’ (IP Australia, sub. DR612, p. 11). Given the relatively inelastic nature of demand and significant body of low-value patents, it might be necessary to raise fees well above current levels. Setting fees primarily to achieve cost recovery squanders an important policy tool. Fees should instead be set to promote broader IP policy objectives. The Australian Government should take the lead responsibility for developing proposals for patent fees that aim to achieve policy objectives, with IP Australia and other experts providing input into this process. 32 The rate of decrease in the distribution of claims then mirrors the rate of decrease in the original distribution of claims until the 35th claim, from which point the number of claims does not differ from the first scenario until the 51st claim. 33 This is based on the USPTO’s (2013) lowest elasticity estimate for claim fees, which is -0.1 per cent. This implies an increase in fees by 100 per cent corresponds to a 10 per cent decrease in claims. 34 The spike in applications with 20 claims (that is, the number of applications exceeding the average number of claims at 19 and 21 claims) is assumed to shift to 15 claims. 236 INTELLECTUAL PROPERTY ARRANGEMENTS RECOMMENDATION 7.4 The Australian Government and IP Australia should set patent fees to promote broader intellectual property policy objectives, rather than the current primary objective of achieving cost recovery. To this end, the Australian Government, with input from IP Australia, should: • restructure patent renewal fees such that they rise each year at an increasing rate (including years in which patents receive an extension of term) — fees later in the life of a patent would well exceed current levels • reduce the initial threshold for claim fees, and increase claim fees for applications with a large number of claims. THE PATENT SYSTEM — GETTING THE FUNDAMENTALS RIGHT 237 8 The innovation patent system Key points • The innovation patent system (IPS) is intended to promote innovation by Australian small and medium sized enterprises (SMEs). While the Commission has been mindful of this policy objective, in assessing the IPS it has considered the welfare of the whole community. • There are very few innovation patents in force. In 2015 there were around 6500 active innovation patents in Australia, compared to over 130 000 standard patents. • Despite being little used, the IPS gives rise to a number of problems. – The innovative step is lower than the threshold for standard patents, which itself is too low. This contributes to a multitude of low value innovation patents, creating uncertainty for other innovators and financiers, and increasing the likelihood that patent thickets will develop. – Innovation patents are used strategically, either to target alleged infringers of standard patents or to increase uncertainty over the scope of rights for competitors. • Perversely, these outcomes have proven more harmful than helpful for innovative SMEs. Reforms would be required should the IPS be retained. – There would be strong grounds for setting the innovative threshold at the same level as the inventive threshold under the standard patent system. This would help to exclude obvious innovations from patent protection. – Strategic behaviour would need to be addressed. Options for doing so include reintroducing mandatory examination, and limiting the period in which damages can apply. • But addressing the flaws of the IPS would see innovation patents resemble their predecessor, petty patents, and represent a return to an approach already found to be lacking — tantamount to a policy ‘Groundhog day’. • The Commission’s view, and that of other experts, is that the IPS should be abolished. – While a reformed IPS would offer marginal improvements, there would be greater benefits for the community if the IPS were abolished. – Abolishing the IPS would simplify the overall patent system, reduce administrative and transaction costs, and remove the ability for patent holders to use innovation patents for strategic purposes. – SMEs would likely benefit from abolishing the IPS, with improved patent integrity and financier confidence. – Broader concerns about the costs to SMEs of securing IP rights and enforcement access and costs can be addressed more directly. IP Australia and the Australian courts are already exploring ways to reduce costs and timing delays, some of which are targeted at SMEs. The Commission has also recommended a low cost avenue for IP enforcement. THE INNOVATION PATENT SYSTEM 239 The innovation patent system (IPS) has been the subject of recent policy debate and review. Participants hold widely differing views on the system’s merits, and on whether it should be abolished or reformed. The Commission has examined the IPS by drawing together and assessing the relevant evidence using its economic framework. 8.1 The innovation patent system: a primer History and policy objectives Between 1979 and 2001 Australia had a ‘petty patent’ system — a second-tier patent system that operated in addition to the standard patent system. The petty patent system limited applicants to three ‘claims’ (which determine the scope of exclusive rights claimed by an applicant) and six years maximum protection. To gain protection, claimed inventions needed to meet the same inventive threshold as standard patents and undergo ‘substantive examination’ (which included an assessment against the patent criteria) prior to grant (ACIP 1995; Moritz and Christie 2006). The petty patent system was little used, and in 1995 the Advisory Council on Intellectual Property (ACIP) was asked to assess the effectiveness and efficiency of the system. ACIP (1995, p. 5) identified what it saw to be a ‘gap’, arguing that the system did not protect incremental innovations. It recommended the petty patent system be reformed to provide ‘fast, limited monopoly protection for lower level or incremental inventions’. In 2001, the petty patent system was replaced by the IPS. At the time, the Australian Government emphasised that the objective of the IPS was to promote innovation in Australian small and medium sized enterprises (SMEs) (Entsch 2000). In this context, it was noted that the petty patent system was unattractive to Australian SMEs. 1 The Government also highlighted the role second-tier patent systems play in other countries. The Government has … devised a ‘second tier’ patent system to better address the needs of business, particularly small to medium enterprises. The innovation patent will be relatively inexpensive, quick and easy to obtain. It will provide the same scope of protection as the standard patent, however it will require a lower inventive threshold than that required for a standard or a petty patent … Over forty-eight other industrialised countries, including Japan and Germany, have already introduced second-tier patent systems. Overseas experience suggests that the innovation patent should provide better access to intellectual property rights and foster innovation by local enterprises. (Entsch 2000, p. 18583) 1 Patents Amendment (Innovation Patents) Bill 2000, Revised Explanatory Memorandum, Regulation Impact Statement. 240 INTELLECTUAL PROPERTY ARRANGEMENTS Key features of the IPS Two key features of the IPS relative to the standard patent system are intended to help it achieve its objective of promoting innovation among SMEs. • The threshold of inventiveness, referred to as an ‘innovative step’, targets ‘lower level inventions’ (ACIP 2015b). 2 The Patents Act 1990 (Cth) (Patents Act) specifies that an innovative step can be met where an invention makes a substantial contribution to the working of the invention. • Lower fees and quicker administrative procedures are intended to reduce the compliance burden on applicants. Innovation patents are usually granted within one month, while standard patents can take six months to several years. This is largely because innovation patents do not undergo substantive examination before being granted. However, innovation patents must be substantively examined and certified before they can be enforced (figure 8.1). Figure 8.1 The innovation patent application processa Application • The initial application provides a priority date from which the claimed invention is assessed for novelty and innovative step Checked • Application checked it satisfies formalities Granted and published • Published in Official Journal of Patents • Usually takes about 1 month • Can sue for infringement from this date if a patent is eventually granted Examination • If requested by applicant or someone else • Examination fee Adverse report • Applicant can respond and resubmit • This process can continue for 6 months from Accepted • Certified in official journal first adverse report Opposition Objections not overcome Objections overcome • Either by amending the application, a successful hearing in IP Australia or appeal to the Federal Court • Other parties can file an opposition • Hearing held in IP Australia, can appeal to Federal Court Opposition upheld • Patent amended or revoked Opposition dismissed • Renewal fees paid a Lighter boxes indicate a step that relies on a decision by the applicant or a third party. Source: IP Australia (2016p). 2 The threshold under the standard patent system is referred to as the ‘inventive step’ (chapter 7). THE INNOVATION PATENT SYSTEM 241 The tradeoff for holders of innovation patents is that, relative to standard patents, innovation patents provide a shorter term and fewer claims (table 8.1). Innovation patents are limited to five claims (compared to unlimited claims for a standard patent) and the maximum duration of protection is 8 years (compared to 20 years). However, once certified, an innovation patent provides the same infringement remedies as a standard patent. 3 Australia is not alone in having a second-tier patent system. Second-tier systems operate in around 60 countries (IP Australia 2015k), and in many cases their purpose is similar. In Japan for example, the second-tier system was established to protect minor inventions and encourage the development of domestic industries (JPO 2012b). The Spanish patent office states their second-tier system is particularly suited for SMEs that adopt or improve existing products (SPTO 2015). In summarising second-tier patent systems, WIPO (nd) said such systems are considered particularly suited for SMEs that make minor improvements to, and adaptations of, existing products. Table 8.1 Key differences between standard and innovation patents Standard Patents Innovation Patents Main patentability criteria Must be useful, novel and involve an inventive step Must be useful, novel and involve an innovative step Inventive threshold Inventive step applies where an invention would not have been obvious to a person skilled in the relevant art 20 yearsa Innovative step applies where an invention makes a substantial contribution to the working of the invention Maximum duration Maximum number of claims Pre-grant opposition Application processing time Application and renewal feesb Unlimited Available 6 months to several years (depending on circumstances) Application $370 Renewal $300 to $2 550 8 years Five Not available 1 month for grant Application $180 Renewal $110 to $220 a Some pharmaceutical patents can be extended by 5 years. b Renewal fees increase over the life of a patent. Source: IP Australia (2016p). Countries vary in their approach to the design of second-tier patents. While the term of second-tier patents is typically shorter than for standard patents, not all jurisdictions adopt a lower inventiveness threshold or bypass examination processes (table 8.2). The German system, for example, has evolved from a regime tailored to cover a gap between design and patent protection towards a system closely resembling patent law (EC 2015c). Differences in the design of second-tier patents are possible because international agreements do not set minimum standards of protection for second-tier patent systems. 4 3 Remedies include: injunctions preventing the alleged infringer from undertaking further infringing conduct; the payment of damages to compensate the patent holder for losses suffered due to infringement; and accounts of profits, where infringers pay the patent holder a sum equivalent to the profits made from infringement. 4 Policy makers may not have complete discretion over the design of second-tier patent systems. The Paris Convention applies to utility models and requires national treatment. The Agreement on Trade Related Aspects of Intellectual Property Rights gives effect to the Convention. In the event the IPS were classed as a utility model, national treatment would need to apply. 242 INTELLECTUAL PROPERTY ARRANGEMENTS Table 8.2 International comparison of selected second-tier patent systemsa Country Maximum duration Australia Austria 8 years 10 years     Plants, animals, biological processes None China 10 years   Processes, products change by mere substitute of material Czech Republic 10 years  Processes, methods, microorganisms Denmark 10 years    Methods and war material Finland Germany 10 years 10 years     Processes Methods, processes, biotechnological inventions Hong Kong Lower Substantive inventiveness examination at grant thresholdb c Excluded subject matter 8 years   None Indonesia 10 years   Methods, processes or uses Italy 10 years   Japan 10 years   Methods, powders, liquids, chemicals or pharmaceutical compositions Methods, software, chemical compositions, plants, animals Republic of 10 years   Methods, processes, software, chemical Korea compositions Poland 10 years   Methods and processes Spain 10 years   Methods and processes a The United States, United Kingdom, Singapore and New Zealand, among other countries, do not have a second-tier patent system. b Relative to standard patents. c The wording of the legal provision sets a lower threshold than a standard patent, but in practice the same threshold is applied. Sources: ACIP (2015b); European Commission (2015c). Who uses the IPS? In 2015, there were around 6500 active innovation patents in Australia (compared to over 130 000 standard patents). Between 2010 and 2015 the number of innovation patents granted each year ranged between 1300 and 1800 (compared to around 18 000 per year for standard patents). Parties that use the IPS typically do so only once (IP Australia 2015k). Most active innovation patents in 2015 were in the fields of civil engineering, furniture and games, IT methods for management, and electrical machinery, apparatus, and energy. Use of innovation patents relative to standard patents varies across technology fields (figure 8.2). Relative use of innovation patents is highest in the IT methods for management technology field, where they make up around 20 per cent of total patents. At the other end of the scale, innovation patents comprise less than 1 per cent of total patents granted in chemistry-related technologies such as organic fine chemistry and biotechnology. THE INNOVATION PATENT SYSTEM 243 Figure 8.2 Use of innovation patents relative to standard patents across technologies, 2015a Percentage of total patents in each technology field that are innovation patents a Technology fields are based on International Patent Classification codes. Source: Commission estimates based on IPGOD (2016 edition). There are some other differences between users of the IPS and the standard patent system. • Consistent with the relative popularity of the standard patent system, SMEs hold more standard patents than innovation patents. At the end of 2015, 75 per cent of patents held by Australian SMEs were standard patents (6600 standard patents compared to 2200 innovation patents).5 • Of the patents held by Australian residents at the end of 2015, SMEs held 55 per cent of innovation patents and 68 per cent of standard patents. (For large firms the corresponding figures were 4 per cent and 16 per cent, and for private applicants (mostly individuals), 41 per cent and 16 per cent.) • Unlike for standard patents, most innovation patents are held by Australian residents. In 2015, 63 per cent of innovation patents were held by Australian residents. China is the largest non-residential filer, followed by the United States, Taiwan and Macao. Use of the 5 These figures do not account for any transfer of patents after they are granted. 244 INTELLECTUAL PROPERTY ARRANGEMENTS IPS by non-residents has increased over time — from 2010–2014 the stock of innovation patents held by non-residents increased by 106 per cent, compared to 11 per cent for Australian residents. Large multinational firms hold the largest portfolios of certified innovation patents (ACIP 2015b). 8.2 How well is the IPS performing? The IPS has been the subject of recent attention Concerns with the IPS have prompted recent reviews into its effectiveness and operation. • ACIP (2015b) found the IPS was failing to achieve its objectives and was resulting in a number of unintended consequences. While ACIP could not find sufficient evidence on whether to recommend abolishing or retaining the IPS, it did recommend raising the threshold for the innovative step and requiring substantive examination. • IP Australia (2015k) subsequently examined the IPS by drawing on patent administration data that was unavailable when ACIP completed its review. IP Australia concluded that the IPS fails to encourage research and development (R&D) (a proxy for innovation) that would not have otherwise occurred, particularly by SMEs. IP Australia suggested the IPS is unlikely to provide net benefits to the community. • ACIP (2015b, p. 1) then issued a corrigendum to its review, noting that in light of the evidence in IP Australia’s report, the IPS is likely to ‘result in a net cost to society’. ACIP concluded that the Australian Government should consider abolishing the IPS. It is important to assess the IPS from an economywide perspective While the Commission has considered the findings of the recent reviews of the IPS, it has undertaken its own assessment. In doing so, it has looked beyond the immediate effect of the IPS on SMEs. This is because, although the stated objective of the IPS is to promote innovation by SMEs, its impacts are more wide ranging. The IPS affects consumers and follow-on innovators as well as other users of the IPS — as noted above, larger firms and individuals also use the system. The Commission’s economic framework provides an appropriate lens through which to assess the IPS since it accounts for the welfare of the whole community rather than specific segments. In keeping with the principles outlined in chapter 2, the IPS should ideally only grant protection to socially valuable innovations that would not have been developed or commercialised absent protection (additionality). In these circumstances, patent protection is more likely to provide net benefits to the community. THE INNOVATION PATENT SYSTEM 245 Does the IPS target innovations of social value? Innovations are of greater social value where they advance technology and human knowledge, and generate knowledge spillovers in other areas of the economy. In practice, the IPS rewards innovations embodying varying levels of technology and knowledge advancement. A wide array of innovations have passed substantive examination under the IPS, from a heart rate monitor through to a pizza box bib (box 8.1). Box 8.1 Examples of certified innovation patents Title (application number) Extract from abstract/claimsa Heart rate path optimiser (2015101130) A device for determining the heart rate of a user comprising, among other things, a sensor configured for providing heart rate signals. A Bed for a Pet (2013100250) A mat and a frame that elevates the mat. Pizza Box Bib (2015100884) A pizza box where the lid includes a removable section, which when removed forms a recess that accommodates the neck of a user. Extract from drawing a For the purposes of clarity and brevity these are not exact excerpts from the relevant abstract/claims. Source: AusPat. 246 INTELLECTUAL PROPERTY ARRANGEMENTS The innovative step is little more than a test for novelty The key policy lever affecting whether the IPS targets socially valuable innovations is the innovative step. All else equal, the lower the innovative step threshold, the more likely it is the IPS will grant protection to innovations that are of low social value. The innovative step is lower than the threshold applying for standard patents, which the Commission has found is itself too low (chapter 7). Decisions by the Federal Court have further lowered the threshold. While the Patents Act specifies an innovative step can be met where an invention makes a substantial contribution to the working of the invention, in the Delnorth case the Federal Court held that an innovation patent is valid where it makes a substantial contribution to the working of the invention, even if it is obvious. 6 In the 2008 proceedings Justice Gyles observed: The phrase ‘no substantial contribution to the working of the invention’ involves quite a different kind of judgment from that involved in determining whether there is an inventive step. Obviousness does not come into the issue … where the point of differentiation does contribute to the working of the invention, then it is entitled to protection, whether or not (even if), it is obvious. [at 53] Many inquiry participants agreed that the innovative step threshold is too low.7 Illustrating just how low the threshold is, Aristocrat (sub. 139) said proving an innovation patent is invalid requires there to be a previous product that is nearly identical. Similarly, in the Commission’s public hearings a patent attorney observed that the innovative step is not much different to the test for novelty (Anthony Alder, trans., p. 246). Another provider of patent attorney services says as much on its website. Because it is so easy to meet the requirements for innovative step, in practice, the only way an innovation patent can be knocked out is with a ‘direct hit’ novelty destroying piece of prior art. (Patentec Patent Attorneys 2016) The low innovative step increases low value patents The social value of patented innovations can be estimated using proxy measures. Some researchers proxy social value by private value (the value of the innovation to the patent holder) for which there is more data. While in many cases the private and social value of innovations will diverge, the Organisation for Economic Co-operation and Development (OECD 2009a) argued that the revenue generated from a patented innovation should be commensurate with the innovation’s technological contribution to society, and therefore that the private and social value of innovations are closely related. 6 Dura-Post (Australia) Pty Ltd v Delnorth Pty Ltd [2009] FCAFC 81; Delnorth Pty Ltd v Dura-Post (Aust) Pty Ltd [2008] FCA 1225. 7 AIPPI, sub. DR551; Alphapharm, sub. DR584; Ausbiotech, sub. DR419; Webber, sub. DR447; Digital Industry Group Incorporated (DIGI), sub. 111; The Institute of Patents and Trade Mark Attorneys (IPTA), sub. 73; Law Council of Australia, sub. DR490; OSIA, sub. DR486; Shelston IP, sub. DR483; Telstra, sub. 76; USCC, sub. DR415. THE INNOVATION PATENT SYSTEM 247 The Commission drew on the results from the Verve Economics survey of users of the IPS to estimate the distribution of private value of innovation patents. These data suggest that a large proportion of innovation patents are of relatively low value — 40 per cent of innovation patents together account for a meagre 3.6 per cent of the total reported value from the survey (figure 8.3). IP Australia (2015k) also sought to estimate the private value of innovation patents. They took a number of approaches, deriving direct estimates from the Verve Economics survey results (box 8.2) and indirect evidence using renewal and certification rates (box 8.3). Figure 8.3 Estimated distribution of private innovation patent valuea Per cent of total reported value 100 80 60 40 20 0 0 20 40 60 80 100 Per cent of survey respondents a Reported value was calculated by multiplying the number of respondents in the range by the midpoint of the range (box 8.2). For the ‘more than $1m’ range, the number of respondents was multiplied by $1 000 001. The cumulative reported value percentage was then plotted against the cumulative fraction of survey respondents for each value range with a straight line used to interpolate between points. This interpolation implicitly assumes the distribution of respondents within a value range is positively skewed with a mean equal to the midpoint of the value range. If the assumption that patents in the more than $1 million range are worth $1 000 001 was relaxed to account for patents with a higher value the distribution curve would shift outwards (except at the end-points) — in other words, a given fraction of survey respondents would account for a lower percentage of total reported value (so 40 per cent of innovation patents would account for even less than 3.6 per cent of total value). Source: Commission estimates based on Verve Economics survey responses. IPTA and the Australian Federation of IP Attorneys (FICPI Australia) (2015) argued that IP Australia’s direct estimates underestimated the private value of innovation patents. They noted that the upper bound that respondents to the Verve Economics survey could value their innovation patent was ‘more than $1 million’ and that IP Australia was incorrect to assume that all patented inventions in this category were valued at $1 million. Some survey respondents that selected the ‘more than $1 million’ category noted in their survey 248 INTELLECTUAL PROPERTY ARRANGEMENTS response how much they considered their patented invention to be worth, with one indicating a value of $3–4 million and another $10 million. Box 8.2 IP Australia’s estimates of the private value-add of innovation patents IP Australia estimated that the total private value of innovation patents ranges from $10 million to $40 million per annum. These estimates are of the additional value an innovation patent adds to its invention: the private ‘value-add’ of innovation patents. In arriving at these estimates, IP Australia drew on two main information sources. • Reported values of innovation patents from the Verve Economics (2013) survey. The survey asked respondents to estimate the value of their patent within ranges. The highest range was ‘more than $1 million’. Reported values were combined to form a value distribution. • Upper and lower bound estimates of the premium that patent protection adds to the underlying invention (Arora, Ceccagnoli and Cohen 2008). These patent premia were estimated using data on standard patents from the United States in the early 1990s. IP Australia assumed innovation patents have patent premia comparable to standard patents in the United States, and they assumed the ‘more than $1 million’ patents were worth $1 million. IP Australia made further assumptions to develop upper and lower bound estimates. • An upper bound estimate was derived by assuming the highest patent premium estimate by Arora, Ceccagnoli and Cohen (2008) and assuming that the data on reported value collected by Verve Economics was representative of the total population of innovation patents. • A lower bound estimate was derived by assuming the lowest patent premium estimate and assuming that the data on reported value collected by Verve Economics overestimated the value of innovation patents. Full details of IP Australia’s approach are included in appendix 4.2 of the IP Australia report. Source: IP Australia (2015k). While there might be some debate about the exact value of high value innovation patents, it is clear that a significant portion of innovation patents are low value. This in itself is not necessarily indicative of a failure of the IPS. The inherent risks from innovating mean most innovation efforts fail; a high number of low value innovation patents simply reflects the nature of innovation. From a policy perspective however, it is evident that the design features of the IPS, and in particular the low threshold for the innovative step, contributes to a multitude of low value patents, which can impose costs on the community (see below). THE INNOVATION PATENT SYSTEM 249 Box 8.3 Indirect measures of innovation patent value by IP Australia IP Australia (2015k) provided indirect evidence on the private value of innovation patents. This analysis focused on the value of innovation patents to SMEs using two observable rates. • Renewal rates — IP Australia observed that while 57 per cent of innovation patents held by large firms lasted eight years, only 34 per cent held by SMEs and 14 per cent held by private inventors lasted as long (around 30 per cent of total patents last eight years). IP Australia also found evidence that the decision to lapse innovation patents was unaffected by firm failure. It identified less than 100 instances where the choice to lapse occurred within a year of de-registration from either the Australian Business Register or from paying GST. • Certification rates — IP Australia observed that few holders of innovation patents seek to certify their patent and thus obtain enforceable rights. IP Australia (p. 24) suggested this means innovation patents create little value, arguing patent value ‘comes from having a legally enforceable monopoly over an invention for the term of the patent’. In a separate analysis IP Australia (2015k) did not find evidence that certified innovation patents help firms stay in business. The Commission considers that renewal and certification rates are of limited use in judging the total private value of innovation patents. While IP Australia presents evidence suggesting renewal rates are unrelated to firm failure, lapses in innovation patents may nonetheless be related to innovation failure. That said, a high number of lapsed patents may indicate the system is poorly targeted. With respect to certification rates, the Patents Act allows, in some circumstances, amendment of claims prior to, and during, examination. Delaying certification could preserve the option to amend claims, such that uncertified innovation patents could be more valuable. The value from delaying certification is evident under the standard patent system, where examination is requested voluntarily for only around 20 per cent of applications (IP Australia 2015i). Does the IPS target additional innovations? Compared to the standard patent system, the risk of granting protection to non–additional innovations is greater under the IPS due to the low innovative step threshold. This is because innovations with a lower level of inventiveness will, on average, involve less R&D and so lower upfront costs. All else equal, the lower are upfront costs, the more likely it is an innovation would have been developed and commercialised in the absence of patent protection. What is the evidence on additionality? The way the IPS affects innovation is complex. The height of the innovative threshold, and the maximum duration of protection, impact on innovation in different ways. How the IPS interacts with other parts of the IP system such as the standard patent system and protection for registered designs, as well as other incentives to encourage innovation like the R&D tax incentive (which provides extra concessions to SMEs), are also relevant. 250 INTELLECTUAL PROPERTY ARRANGEMENTS IPTA (sub. DR562) noted the IPS is mainly used by Australian applicants, and argued this strongly suggests the system provides incentives for Australian SMEs to innovate. However, usage of the system (in isolation) says little about whether it promotes innovation that would not have otherwise occurred. Firms use innovation patents for a variety of reasons, including strategic purposes (see below). Even for genuine innovations, it does not necessarily follow that innovation would not have occurred in the absence of the IPS. In many cases, such innovation would have occurred anyway and the costs of granting patent protection are incurred for no resulting benefits. In drawing on its experience with second-tier patent systems in a number of countries, Intel (sub. 66) said these systems are seldom successful in promoting innovation. IP Australia (2015k) attempted to assess additionality by estimating the relationship between innovation patents and innovation, with the latter proxied by R&D expenditure (an input into innovation). This revealed an association between patenting and R&D in the manufacturing sector, though not in other sectors. While it is difficult to draw firm conclusions from IP Australia’s research, evidence from other countries suggests the role of second-tier patenting systems in encouraging additional innovation is likely to be limited. • Kim et al. (2012) drew on a panel dataset of over 70 countries to conduct a cross-country econometric analysis of second-tier patent systems. They concluded the presence of a second-tier system is more likely to promote R&D in developing countries. In a firm-level study, Kim, Lee and Choo (2010) concluded second-tier patent innovations only promote growth in firms that are technologically lagging. • Maskus and McDaniel (1999) found that the second-tier patent system in Japan encouraged technology diffusion post war. They examined the role of the second-tier system on total factor productivity growth from 1960–1993, controlling for technology creation and diffusion. The authors noted their results were consistent with Japan being in a ‘technological catch-up phase’ in which diffusion and imitation were more important than pure invention. • Heikkila (2014) analysed the 2008 abolition of the Dutch second-tier patent system. He found there was only a temporary decrease in patent applications when the system was abolished. While filing activity is a poor proxy for innovation, Heikkila concluded this may indicate the abolition of the second-tier system led to an increase in applications for standard patents and that domestic innovative activity was not adversely affected. There are also claims that the abolition of the Belgian second-tier patent system in 2009 had little influence on standard patent filings in that country (Prud’homme 2014). • In a European Commission (2015c) survey of users of the German second-tier patent system, 88 per cent of respondents said they would have proceeded with their invention in exactly the same manner in the absence of the system. The study concluded that the role of the system in encouraging additional innovation is low. THE INNOVATION PATENT SYSTEM 251 Innovation patents have some unintended consequences By failing to target socially valuable, additional innovations, the IPS can have unintended consequences. Some participants linked the low innovative threshold to a proliferation of obvious patents, legal uncertainties and patent thickets, which in turn raise barriers to entering markets and blunt competitive pressure, itself a driver of innovation. Low value patents create uncertainty, increase noise and promote thickets A multitude of low value patents make it harder for an innovative firm to be sure it is not infringing someone else’s patent. This increases uncertainty for the innovating firm, and may raise its costs of innovation. Inquiry participants raised similar concerns. Drawing on its experience with second-tier patent systems in other countries, Intel (sub. 66, p. 3) said these systems ‘encourage the proliferation of weak and trifling patents, and create uncertainty and distortion in markets’. DIGI (sub. 111, p. 9) argued innovation patents that are ‘awarded for trivial differences over the prior art can harm innovation and economic growth by increasing the uncertainty, risk and cost of creating and providing new services and products’. The impact of low value innovation patents on legal uncertainty is compounded by innovation patents not requiring substantive examination. This is because there is an extra element of risk about whether innovation patents will become enforceable. FICPI Australia (sub. DR581, p. 38) said ‘it is not easily apparent to a member of the public as to whether an innovation patent has been certified or not, or being aware that an innovation cannot be enforced until it has been certified’. OSIA (sub. 21) argued that uncertainty about the enforceability of innovation patents — even if it is evident that the claims would not pass examination — creates extra risks for follow-on innovators and competitors. In this context, Ausbiotech (sub. 37) argued some firms may find themselves constrained by innovation patents they suspect are invalid, but do not have the resources to challenge. Low value patents are especially problematic where they contribute to ‘patent thickets’. Thickets are ‘a dense web of overlapping intellectual property rights that a company must hack its way through in order to actually commercialise new technology’ (Shapiro 2004, p. 120). This can lead to an ‘anti-commons’, where dispersed ownership of patents in a technology field increases the costs from negotiating access to technologies (Heller and Eisenberg 1998). The risk of patent thickets developing is greater for innovation patents due to the lower innovative step threshold. Indeed, in some cases, it would seem innovation patents are deliberately used for such purposes. As one patent attorney firm advises potential clients: By filing a series of innovation patents surrounding a product of a competitor, it is relatively easy and cost-effective to form a patent thicket around the product which subsequently makes it increasingly difficult and costly for the competitor to maintain freedom-to-operate. (Baxter IP 2016, emphasis in original) Overcoming patent thickets to enter and compete in a technology space can be especially difficult for SMEs and potential market entrants. Many of these firms may not have sufficiently large patent portfolios (or the resources for acquiring such portfolios) to enter 252 INTELLECTUAL PROPERTY ARRANGEMENTS cross-licensing agreements, in which firms license large parts of their patent portfolios to each other. A number of studies find that patent thickets inhibit market entry (Cockburn and MacGarvie 2009, 2011; Hall, Helmers and Graevenitz 2015; IPO 2013). The Commission identified a number of patent thickets using Australian patent administrative data, some of which include innovation patents (appendix D). For example, one of the thickets identified in the area of domestic appliances centres on an innovation patent held by Electrolux. The thicket (illustrated in appendix D), which also consists of standard patents held by LG, Bosch, Samsung and Mitsubishi, shows that innovation patents can add to complex webs of overlapping patent rights. Low value patents can also make the commercial value of innovation patents in financial markets uncertain. This uncertainty arises from the extra ‘noise’ from low value patents, dampening the credibility signal in patents (a concern that also holds for the standard patent system). This can increase the rate of return required by financiers and make it more difficult for firms to leverage their patents to acquire capital. 8 Noise in financial markets may be especially problematic for SMEs. Drawing on firm-level data from Belgium, Hottentrott, Hall and Czarnitzki (2016) estimate an econometric model of R&D investment that suggests small firms are the main beneficiaries from patents lowering financing constraints. Potentially reflecting the uncertainty over the value of innovation patents in financial markets, in the Commission’s public hearings AusBiotech argued that unexamined innovation patents are worth little to investors. … I act for a number of venture capital investors and other investors and do due diligence on our key portfolios. An unexamined innovation patent would be properly regarded as not worth very much at all by an investor. (trans., p. 147) By contributing to thickets and decreasing the credibility signal in patents, the IPS may make it harder to meet its stated objective of promoting innovation by SMEs. The Gowers (2006) review of IP in the United Kingdom similarly argued that a second-tier patent system would increase costs for other users and stunt innovation. That review recommended against introducing a second-tier system in the United Kingdom — a recommendation that was accepted by the UK Government. Low value patents can be used strategically Innovation patents can also be used for strategic purposes (ACIP 2015b; DIGI, subs. DR528 and 111; Generic and Biosimilar Medicines Association (GBMA), sub. 67; Intel, sub. 66; Society of University Lawyers (SOUL), sub. 98). A specific concern is that innovation patents are used to target alleged infringers of standard patents. The strategy involves carving a ‘divisional’ innovation patent out of an earlier standard 8 A body of literature finds that patents can be an effective instrument for reducing information asymmetries between patenting firms and outside investors (Baum and Silverman 2004; Haeussler, Harhoff and Mueller 2009; Hsu and Ziedonis 2013; MacMillan, Siegel and Subbanarasimha 1985; Mann and Sager 2007). THE INNOVATION PATENT SYSTEM 253 patent.9 It is argued that the claims for some divisional innovation patents are deliberately constructed to ensure competitors are found to infringe the earlier standard patent (ACIP 2015b; SOUL, sub. 98). ACIP said the legal dispute between Britax Childcare and Infa-Secure demonstrates this strategy (box 8.4).10 In public hearings IPTA (trans., p. 171) highlighted the role of innovation patents as a litigation tool. Box 8.4 Britax Childcare v Infa-Secure The Britax Childcare v Infa-Secure case relates to nine innovation patents and one standard patent owned by Britax concerning child safety seats. Each of the innovation patents is divided out of Britax’s earlier standard patent (the parent application). Britax submitted that Infa had made various modifications to its products in an endeavour to escape from the rights permitted to Britax based on its standard application. In a judgment focused on the construction of the relevant claims, the Federal Court [at 27] said that some of the patents and claims within those innovation patents were specifically drafted to catch alleged infringing Infa products and to bring those products before the Court. Source: Britax Childcare Pty Ltd v Infa-Secure Pty Ltd [2012] FCA 467 (9 May 2012). This strategy is widely promoted by providers of patent attorney services (Baxter IP 2016; Cotters 2016; FPA Patent Attorneys 2016; Patentec Patent Attorneys 2016). The strategy most commonly promoted is capturing the product of a competitor by tailoring the claims. A provider of patent attorney services described their approach to this strategy as follows: In this way, always recommend as a strategy the filing of a standard patent application, which you keep pending during the commercialisation stages of your idea is a so-called “mothership”. However, if you ever need to enforce your patent rights, we simply file a divisional innovation patent application from the mothership standard patent, leaving the mothership intact and using the divisional innovation patent as the weapon. In this process, we often times tailor the claims of the patent specifically around the activities of the alleged infringer so as to make a greater case for infringement. (Patentec Patent Attorneys 2016) Drawing on patent administration data, the Commission found that divisional patents grew from 5 per cent of innovation patents in 2001 to around 18 per cent in 2014. These findings are consistent with the Verve Economics survey, where around 15 per cent of respondents identified their patent as a divisional patent. Verve Economics further noted that divisional innovation patents can contribute to patent thickets. There is the possibility that divisional patents were used by inventors for strategic reasons … some 19 inventors indicated divisional patents were used for protection around intellectual property in a patent other than the parent patent. Such protection is consistent with the divisional patent being part of a patent thicket. (2013, pp. 42–43) 9 Divisional patents divide a previous ‘parent’ patent application into two or more applications. One of the intended purposes of divisional patents is to enable firms to file a separate application in cases where IP Australia assesses that an application contains more than one invention. 10 Reforms under the Intellectual Property Laws Amendment (Raising the Bar) Act 2012 (Cth) have reduced the scope for this sort of behaviour. 254 INTELLECTUAL PROPERTY ARRANGEMENTS Incentives for strategic behaviour are enhanced by divisional innovation patents having the same priority date as the original standard patent (parent application) (ACIP 2015b). This enables patent holders to claim infringement back to the filing date of the original standard patent (figure 8.4). That is, patent holders can claim damages for periods prior to the standard patent being published and the filing date of the innovation patent (ACIP 2015b). As noted above, the same remedies for infringement are available for innovation patents as standard patents. Under the standard patent system relief from infringement is unavailable for any period prior to the publication of claims. Another concern is that granted, but non-certified, innovation patents are used to increase uncertainty for competitors (DIGI, sub. 111; IP Australia 2015k; OSIA, sub. DR486). This uncertainty arises because the holders of innovation patents are not obliged to request substantive examination. 11 DIGI (sub. 111, p. 10) observed that uncertified patents create a difficult choice between ‘moving forward in the face of infringement risk, avoiding the field, or taking on the significant expense of assessing whether an uncertified innovation patent is sufficiently inventive to be enforced’. The University of Technology Sydney (sub. DR564) said ‘worthless innovation patents’ are currently creating uncertainty. IP Australia similarly pointed to the problems that can arise for competitors: … the applicant is better placed to know if the patent would pass examination than third parties at the point of application. Third parties can read patent applications that may impact their business to make their own determination of whether the patent would pass examination, but this comes at an additional cost to the third party. This creates additional costs in the marketplace, where uncertified applications may be filed to deter competitors. (2015k, p. 17) Figure 8.4 Timeline of potential strategic use of divisional innovation patent Source: Based on ACIP (2015b). 11 While competitors can request an examination, they are liable for half of the costs. THE INNOVATION PATENT SYSTEM 255 Uncertainty due to a lack of substantive examination was a factor that led to the abolition of second-tier patent systems in the Netherlands and Belgium in 2008 and 2009 (Prud’homme 2014). The evidence that innovation patents are filed for strategic purposes is consistent with a recent report on second-tier patent systems commissioned by the European Commission. The report concludes that these systems are mostly used by IP attorneys for strategic purposes. … [second-tier patent] systems in Europe have lost much of their supposed ability to protect ‘minor inventions by small inventors’. Rather, the [second-tier patent] has turned into an auxiliary tool of savvy IP professionals, who use it in specific national contexts to overcome shortcomings of the patent system. Overall awareness of [second-tier patents] outside this group of IP professionals was low. (EC 2015c, abstract) The benefits of the IPS may be illusory or not as large as thought … One of the key mechanisms intended to help achieve the objectives of the IPS is a cheaper and quicker patent procedure. Submissions to this inquiry make clear that the cost of securing and holding a patent is an issue for some (ANSTO, sub. 17; BCA, sub. 59; University of Wollongong, sub. 54). The University of Wollongong noted that costs can be particularly restrictive for SMEs. Patent costs include the upfront costs of securing a patent, such as application and examination fees, the charges from engaging a patent attorney, and the costs to patent applicants from complying with application procedures. The costs of securing a patent may also include the costs from engaging in post-certification and post–grant opposition proceedings. There are also costs of maintaining a patent. These include renewal fees and legal costs if the validity of the patent is challenged in court. The upfront costs of securing a patent are a large part of the total costs. Park (2010b) estimated the upfront fees of securing a standard patent in Australia represented around 40 per cent of the total patent cost (estimated to be about US$20 000). IP Australia estimated that the cost of applying for a standard patent using a patent attorney is about $8000, while the University of Wollongong (sub. 54) referred to estimates that suggest the cost can be around $10 000. By contrast, total renewal fees for a 20-year standard patent are about $11 000. The above figures underestimate the relative importance of upfront costs. Many patents fail to last their full term, meaning the actual costs incurred in maintaining a patent are lower than the figures suggest. It is also unlikely that the estimates of upfront costs account for the full costs that applicants incur from engaging in the application process (such as the time taken to assist their agents to prepare applications). A recent analysis of the regulatory burden on patent applicants suggests that for every $1 charged by a patent attorney for services involved in filing an application, the applicant incurs around $0.50 in costs from spending time with their attorney to conduct regulatory activities (KPMG 2014). 256 INTELLECTUAL PROPERTY ARRANGEMENTS For a patent of given complexity and number of claims, the upfront costs of a patent attorney are likely to be similar for innovation and standard patents (IP Australia 2015e). In this context, a provider of patent attorney services noted the same amount of work is required for examined innovation and standard patents (Anthony Alder, trans., p. 247). The IPS may therefore do little to reduce upfront costs. Indeed, the cost advantage that comes from the IPS mainly stems from the fact that applications need not undergo substantive examination. However, as discussed above, there are questions about the value of an unexamined innovation patent. … and there may be other ways to achieve IPS policy goals A number of initiatives are being progressed to reduce the costs that SMEs and other firms face in using the patent system. • IP Australia (2015i) recently developed proposals to streamline processes for obtaining, maintaining and challenging the different forms of IP rights, and to assist small businesses using the IP system. The Government is currently considering the proposals (IP Australia 2016m). • IP Australia is also trialling a new pre-examination process that aims to speed up the patent application process (2016o), and is examining ways to reduce regulatory burden as part of the Australian Government’s commitment to reduce red tape (2016l). • The Australian Federal Court has recently released draft reforms outlining ways it intends to improve the case management of IP disputes. Suggestions include more active case management by judges, greater use of written submissions and improvements to expert witness processes. The Commission supports these efforts to directly reduce the costs and delays of securing IP rights, while not compromising the efficacy of the patent process. However, it notes that the specific proposals to streamline IP rights mainly entail reforms to trade marks and plant breeder’s rights. There would be benefits from IP Australia extending these efforts and identifying any unnecessary costs and delays in the patent application process. The Commission is also recommending making greater use of the Federal Circuit Court (rather than the Federal Court) to reduce costs in less complex patent disputes and improve access to enforcement, which could be particularly beneficial for SMEs (chapter 19). But with many of the costs arising from attorney fees, measures by IP Australia can only form part of the solution. Park (2010b) provides evidence that the upfront costs of securing a standard patent in Australia are higher than in a number of other jurisdictions, but points out that around 90 per cent of upfront costs are agent fees (figure 8.5). The University of Melbourne also highlighted that most of the costs from obtaining a patent arise from engaging an adviser (primarily legal): The costs of obtaining patent protection are considered to be high; however, the direct government fees and charges are a relatively small portion of the overall costs. The remainder of the costs relate to obtaining sophisticated advice, along with the costs of language THE INNOVATION PATENT SYSTEM 257 translations of technical documents. Patenting is an area in which highly specialised advice is required in order to secure effective rights. (sub. 100, p. 4) Figure 8.5 The upfront costs of securing a standard patent are higher in Australia than a number of other countriesa 16 Official fees Agent fees US $ thousandsb 12 8 4 0 a Estimates of patent fees are for a patent of 25 pages in length, with five pages of drawings and 15 claims. Official fees include filing, examination, granting, and prosecution fees. Agent fees include attorney fees and in-house and miscellaneous charges. b The fees are in real 2005 US dollars and are for August 2010. Source: Park (2010b). 8.3 The IPS — where to from here? There are clearly flaws in the current design of the IPS. Perversely, legal uncertainty and strategic use of innovation patents could be particularly harmful for SMEs. With this in mind, the Commission has considered whether outcomes for the community could be improved by reforming or abolishing the IPS. Reforms would be required if the IPS were retained Supporters of the IPS recognise that were the IPS retained, reforms would be required to improve its operation (ACCI, sub. DR569; Ausbiotech, sub. DR419; Law Council of Australia, sub. DR490; Office of the Australian Small Business Commissioner (OASBC), sub. 101). Some participants proposed raising the innovative step and addressing strategic behaviour. Other proposals included examining the merits of restricting use of the IPS to small firms (ACCI, sub. 70; Australian Small Business and Family Enterprise Ombudsman, 258 INTELLECTUAL PROPERTY ARRANGEMENTS sub. DR403) and reducing the term of protection (Australian Information Industry Association (AIIA), sub. 89). Raising the innovative step Many participants supported raising the threshold for meeting the innovative step under a reformed IPS (AIPPI, sub. DR551; Ausbiotech, sub. DR419; Webber, sub. DR447; Intel, sub. 66; IPTA, sub. DR562; Law Council of Australia, sub. DR490; USCC, sub. DR415). Raising the threshold would provide a number of benefits. By helping to ensure only valuable innovations receive patent protection, a higher threshold would reduce the likelihood of granting costly patents and improve their credibility signal, increasing efficiency in technology and capital markets. 12 At the same time, raising the threshold is unlikely to come at the expense of innovations that provide net benefits to the community, a point also made by Mark Summerfield: … it appears that most of the value in the innovation patent system is extracted from the ‘top’ 20-30 per cent of applications/patents, while the cost of the system is assumed to be uniformly distributed across all applicants. So if, for example, the ‘bottom’ 50 per cent of applications were never filed, the regulatory cost of the system would be halved, while the total benefit would be substantially unchanged. (2015, p. 7) Interested parties that favour increasing the innovative step have different views on where to set the threshold. Some favour raising the threshold to the same level as the standard patent system (BSA The Software Alliance 2015; Intel sub. 66; ResMed 2015; USCC, sub. DR415). Aligning the threshold across the two systems was proposed by the Australian Government in IP Australia’s (2012b) Raising the Step consultation paper. The Government went as far as to provide drafting instructions for raising the threshold, but did not proceed with the reform. In these instructions it noted: … amendments are necessary to address an emerging problem with tactical use of Innovation Patents, which creates uncertainty in the market place and blocks follow-on innovation. We seek to amend the inventiveness test for Innovation Patents to replace the existing ‘innovative step’ test with the same ‘inventive step’ required for Standard Patents. This would align the Innovation Patent requirements with the well-known and legally-settled test for inventiveness that applies to Standard Patents. Raising the inventiveness requirement for Innovation Patents will address community concerns that the Innovation Patent system is being abused, particularly in the information technology industry. (IP Australia 2012b, p. 6) Others favour a threshold level below the standard patent level, but above the current innovative threshold. ACIP (2015b), for instance, supported increasing the threshold to a modified version of the ‘test of inventiveness’ previously used by the High Court.13 This option, also supported by Medicines Australia (2015) and Pfizer Australia (2015), would 12 Filtering out costly patents would have the added benefit of ‘freeing up’ the resources used in the application process (such as legal and administrative services) to more productive uses. 13 Minnesota Mining and Manufacturing Co v Beiersdorf (Australia) Ltd (1980) 144 CLR 253. THE INNOVATION PATENT SYSTEM 259 require an innovation to be non-obvious with respect to common general knowledge in the relevant field of technology. Other options for raising the threshold have been proposed. • Shelston IP (sub. DR483), Summerfield (2015) and Ausbiotech (2015) recommended setting the threshold to the same level that applied under the Patents Act 1990 as originally enacted. This would require an inventive contribution to the working of the invention, with inventiveness assessed against no more than one source of prior public information viewed in light of common general knowledge. • The Law Institute of Victoria (2015) and Telstra (sub. 76) proposed assessing whether the innovation represents a substantial contribution to the working of the ‘prior art’ (knowledge in the relevant field). • IPTA and FICPI Australia (2015) considered two potential thresholds: inventive in the light of common general knowledge in Australia; and a contribution not simply to the working of the invention, but by reference to the prior art. If the IPS were retained the Commission considers that there would be strong grounds for setting the innovative threshold at the same level as the inventive threshold under the standard patent system. This would help to preclude patent protection for innovations that contribute little social value, and would simplify the overall patent system. It is also consistent with second-tier patent systems in countries such as Germany and Japan. In favouring the same threshold as the standard patent, the Commission is mindful of its recommendation to raise the inventive step for standard patents to a level that is more commensurate with the threshold applied in other countries (recommendation 7.2, chapter 7). Addressing strategic use Raising the threshold for the innovative step would also help to address some of the issues around strategic use. A higher threshold would make it harder to certify claims that do not support a genuine innovation, but instead are drafted for strategic purposes. 14 However, raising the innovative step would not be a panacea for addressing strategic use. The patent criteria only apply at examination, which, under the IPS, is not mandatory. Moreover, users of the IPS still have the incentive and ability to file innovation patents to generate uncertainty for competitors. This uncertainty can last the full eight years of an innovation patent due to the absence of a requirement to request substantive examination. And in cases where innovation patents are also divisional patents, users can still seek 14 The Commission notes that some Raising the Bar reforms already seek to address some strategic uses of innovation patents. The reforms limit the opportunity to file a divisional innovation patent out of a standard patent to within three months after the advertisements of acceptance of the original patent application. Raising the Bar also sought to improve the clarity as to the exact scope of claims in a divisional innovation patent application by requiring such applications to include details about the original patent application, and by limiting the use of ‘omnibus claims’. 260 INTELLECTUAL PROPERTY ARRANGEMENTS compensation for infringement back to the filing date of the original application (ACIP 2015b). There are a number of options for addressing strategic use. One possibility would be to reintroduce arrangements for mandatory examination processes (AusBiotech 2015; CropLife 2015; Eliades, sub. DR579; DEDJTR 2015b; Nufarm 2015; ResMed 2015; Summerfield 2015; University of Technology Sydney, sub. DR564). Examination, for example, could occur on the third anniversary of an innovation patent being granted (ACIP 2015b; AusBiotech 2015; CropLife 2015; Eliades, sub. DR579). Research on renewal behaviour reveals that firms often do not know the value of the innovation protected by a patent until at least three years from grant, suggesting examination should not occur until this time (Lanjouw 1993; Pakes 1986; Schankerman and Pakes 1986). Another (complementary) option would be to limit the period in which damages could apply until after the official publication of the claims that have allegedly been infringed. The Law Council of Australia (sub. DR490) endorsed this proposal, and further suggested there would be benefits from limiting the number of divisional patents. FICPI Australia (sub. DR581) also suggested reforms are needed to divisional arrangements. Summerfield (2015) suggested limiting the period in which damages could apply would encourage patentees to settle on final claims early in the application process. This reform would also reduce incentives to use innovation patents as a strategic enforcement tool. A third, but less desirable, option would be to limit remedies for infringement, especially injunctive relief, and so make innovation patents less attractive as strategic enforcement tools (AIIA, sub. 89; BSA The Software Alliance 2015; Intel, sub. 66; Summerfield 2015). Intel (sub. 66) argued that having a lower innovative step threshold but the same remedies for infringement creates an imbalance between the innovative contribution by innovations patented under the IPS and the strength of enforcement measures. However, raising the threshold for the innovative step would remove this imbalance. And requiring substantive examination and limiting the period in which patent holders can claim infringement would help reduce the costs from strategic enforcement of patents. Abolishing the IPS remains the best policy option The above reform options would help to limit the costs imposed by the IPS and in so doing marginally improve the welfare of the community. However, the Commission, like ACIP, is of the view that the benefits from retaining an amended version of the IPS would be limited. Abolishing the IPS would deliver greater benefits for the community. It would simplify the overall patent system, reduce administrative and transaction costs, and remove the ability for patent holders to use the system strategically. To the extent that innovative and socially valuable ideas currently receive protection under the IPS, these would be expected to receive protection under the standard patent system. Innovations that would not pass the inventive step threshold under the standard patent system are more likely to impose net costs on the community, and should not receive protection. THE INNOVATION PATENT SYSTEM 261 Many inquiry participants favoured abolishing the IPS (Alphapharm, sub. DR584; DIGI, sub. DR528; GBMA, sub. DR396; iSignthis, sub. DR443; Microsoft, sub. DR420; OSIA, sub. DR486; Swinburne University of Technology, sub. DR557; USCC, sub. DR415). Yet at the same time some participants expressed concern that abolishing the IPS would work against the interests of SMEs (AIIA, sub. DR379; Australian Small Business and Family Enterprise Ombudsman, sub. DR403; FICPI Australia, sub. DR581; University of Technology Sydney, sub. DR564). However, as the Commission has set out above, the cost advantage from using the IPS is likely to be smaller than many participants consider and would be further reduced were the innovative threshold raised. Moreover, the Commission expects SMEs would also enjoy some benefits were the IPS abolished. In particular, SMEs would gain from a more accurate signal value in patents and reduced uncertainty about whether their business practices are likely to infringe a patent. RECOMMENDATION 8.1 The Australian Government should abolish the innovation patent system. 262 INTELLECTUAL PROPERTY ARRANGEMENTS 9 Business method patents and software patents Key points • Business methods span activities ranging from the trivial to the complex. Some business methods are specific to certain industries while others have economy wide application. With the rise of the digital economy, many business methods are implemented by software — testing the four century old legal construct underpinning the patent system. − Notably, recent Australian court decisions have made clear that business methods, whether implemented in software or not, are not patentable subject matter. • The term ‘software’ is also broad — capturing everything from simple procedures and mathematical formulas to more complicated processes with industrial applications — and so determining when the implementation of an idea using software is deserving of a patent has proved equally contentious. • Patent protection may not be the most efficient and effective way of encouraging innovation through software. − Software development typically builds sequentially on existing ideas, and is becoming cheaper. In some cases, development cycles are short, though where patents are granted, software tends to be longer-lived. − In contrast, patents provide long term protection and can make follow-on innovation more difficult. • Software patents can include overly broad claims and be used for anticompetitive purposes, making it difficult for new entrants and follow on innovation. − While there are examples of strategic use in other jurisdictions, there is little evidence of such behaviour in Australia. • Recent court decisions have narrowed the circumstances where software can gain patent protection, addressing some residual concerns about the effectiveness and efficiency of software patents. • Future grants of software patents should be monitored by IP Australia, with the data collected used to assess whether further software patent reform is needed. The digital economy has seen a shift in the way that goods and services are produced and used. Online shopping, cloud sharing and business processes automated via software all represent changes in the way that individuals and businesses operate. However, patent protection reflects the thinking of the steam age more than the streaming age. As a result, the role of patents in encouraging innovation in business methods and in software is contentious, both in Australia and abroad. BUSINESS METHOD PATENTS AND SOFTWARE PATENTS 263 In light of the growing importance of computer technology, and recent legal decisions, this chapter examines the patentability of business methods and software. The chapter begins by outlining what is meant by business methods and the intellectual property (IP) protection they are afforded (section 9.1). Section 9.2 discusses the nature of software innovation and the IP protections they are afforded. The following section applies the Commission’s analytical framework to assess the merits of affording software patent protection (section 9.3). Section 9.4 concludes by mapping out policy action for the future. 9.1 Business methods What are business methods and how are they used? Traditionally business methods have been defined as a method of operating any aspect of an economic enterprise, including ‘trading, transacting, finance, resource management, marketing and customer service’ (ACIP 2003a, p. 1). They include activities, ranging from the trivial to the complex and can apply across entire economies or only be relevant to narrow industry sectors. More recently, business methods have been incorporated into computer technology, whereby new and existing business methods are programed using a computer. For example, software has been used to implement business methods that make it faster and easier to purchase online goods as well as to calculate and evaluate different investment portfolios based on measures of risk and return. What IP protections are afforded to business methods? Although there are no legislative exclusions for business methods, and the term is not referenced in the Patents Act 1990 (Cth), it is generally accepted that business methods per se are not patentable. As noted by IP Australia: A scheme or plan (including a business scheme or method), by itself, is not suitable for a patent. … This is because it does not specifically involve any artificial application to implement the scheme. (IP Australia 2015g) In Australia, a patentable invention must be a ‘manner of manufacture’ — a concept that stems from England’s 1624 Statute of Monopolies. The implementation of business methods in computers has tested this long-held legal construct. Until recently, a method implemented through a computer could attract patent protection on the grounds that it was implemented in a manner that involved an artificial application — satisfying the manner of manufacture test. Owners of computer implemented business methods claimed protection for both the business method and the software as a single invention. 264 INTELLECTUAL PROPERTY ARRANGEMENTS Several recent court rulings, however, have clarified the patentability of business methods. • In Grant v Commissioner of Patents [2006] (‘Grant’), the Federal Court clarified ‘manner of manufacture’ with the introduction of a physical effects test: A physical effect in the sense of a concrete effect or phenomenon or manifestation or transformation is required. … It is necessary that there be some “useful product”, some physical phenomenon or effect resulting from the working of a method for it to be properly the subject of letters patent. 1 • In Commissioner of Patents v RPL Central Pty Ltd [2015] (‘RPL Central’), the Full Federal Court clarified that a business method implemented using a computer, where the computer was used for well–known and understood functions, is not patentable. As several participants in this inquiry noted, the subsequent decision by the High Court to deny an application to appeal the Full Federal Court’s RPL Central decision has made clear that computer implemented business methods are not patentable subject matter (BCA, sub. DR587; IPTA, sub. DR562; Summerfield, sub. DR388). This judicial interpretation is now reflected in IP Australia’s Patent Examiner’s Manual: … the Full Federal Court considered computer implemented business methods and found that the presence of computing hardware or processing steps within a claimed method or system was insufficient to confer patentability. The Court identified a distinction “between the employment of an abstract idea or law of nature and the idea or law itself” and “between technological innovation which is patentable and a business innovation which is not”. (IP Australia 2016j, sec. 2.9.2.7) The Commission considers that this result improves the effectiveness of the patent system in Australia. A method of doing things, which is otherwise ineligible for patent protection, should not receive protection by virtue of its implementation on a computer alone. But this still leaves open questions about whether the IP embedded in other software should qualify for patent protection. 9.2 Patents for software Computer technology exists in components of many everyday goods and services. Within computers is a set of instructions, known as computer programs or software, which enable the computer to function. Software (and the inventions it can implement) can encompass everything from simple mathematical calculations through to complicated processes, such as a self-driving car. Over the past few decades, there has been a shift from implementing innovations in hardware to implementing them in software. For example, in the Commission’s public hearings Qualcomm noted that they implement their telecommunication technology using software (trans., p. 662). The implementation of inventions in software has advantages in 1 Grant v Commissioner of Patents [2006] FCAFC 120, [32, 47]. BUSINESS METHOD PATENTS AND SOFTWARE PATENTS 265 that it allows for greater efficiency and cost-effectiveness, the incorporation of new functionalities, and greater flexibility in testing and correcting errors (Old, sub. DR161; Summerfield, sub. DR388; Qualcomm, trans., p. 662). Inquiry participants argued that whether an invention’s functionality is implemented through software or hardware is a distinction without difference (BCA, sub. DR587; Goetz, sub. DR160; Old, sub. DR161; Summerfield, sub. DR388). They argue that if a piece of hardware is patentable, then implementing the same procedure via software should also be patentable. However, comparing the protection afforded to software versus hardware is not helpful. Some have argued that unlike hardware, which includes defined property rights, the patent system can afford software an unprecedented breadth of protection, as it potentially allows the protection of abstract ideas, methods or processes to complete particular tasks (Bessen and Meurer 2008; Cohen and Lemley 2001; Rai 2013). The rise of computer implemented inventions has also led to debate about how widespread software patents have become, and whether traditional patentability criteria should apply to software. How widespread are software patents? Determining which patents apply specifically to software is not straightforward, as software can be used in a range of applications and inventions. One approach to identify software patents is to use the International Patent Classification (IPC) (ACIP 2003a; McEneiry 2012) as an imperfect proxy. Using this approach for Australian patent data (figure 9.1) reveals that: • most applications for software patents are filed from the United States • most applications filed by Australian firms are used in the computer system design and scientific and manufacturing industries • software patent application growth was strong through the 2010s, but has since been volatile, which is likely to be in response to various legal rulings abroad. 266 INTELLECTUAL PROPERTY ARRANGEMENTS Figure 9.1 Software patent applications in Australia i. 2000 1800 1600 number 1400 Origin of software patent applications in Australiaa Australia United States Rest of world ...but applications have been more volatile since recent court decisions. A majority of software patents are lodged from the US and overseas... ...growth in the 2010s was driven mainly by US applicants... 1200 1000 800 600 400 200 0 ii. 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 Software and patent applications made by Australian firms, by industryb Other Tertiary Education Administrative Services Other Store-Based Retailing Machinery and Equipment Wholesaling Internet Related Services Unknown Auxiliary Finance and Insurance Services Finance Machinery and Equipment Manufacturing Professional, Scientific and Technical Services Computer System Design and Related Services 0 200 400 600 Number of patent applications 800 a Based on IPC codes G06F and G06Q b By Australian New Zealand Standard Industry Classification (ANZSIC). Sources: Commission estimates based on Intellectual Property Government Open Data (IPGOD) (2015 and 2016 editions) and unpublished data from IP Australia. BUSINESS METHOD PATENTS AND SOFTWARE PATENTS 267 What IP protection is afforded to software? Software can attract copyright and patent protection, but the elements that these IP rights protect are different. Software code — the ‘step by step’ instructions to perform a task, process or method is protected by copyright. However, copyright does not protect the underlying ideas or particular methods used in the software code (IP Australia 2016k). Some of these particular methods can qualify for patent protection, but not all. Software that is simply a procedure for solving a mathematical problem is not patentable, nor are mathematical algorithms and abstract intellectual concepts (IP Australia 2016k). However, if the software can be industrially applied, such as allowing a machine to perform a task, then it may be patentable. Another contentious issue is determining whether the implementation of an idea, using software, is patentable. But even here, recent court decisions in Australia are providing greater clarity. In Research Affiliates, 2 the Full Federal Court judged that an invention implemented using a computer was not sufficient to achieve patent protection, but rather it needed to contribute to an improvement in computer technology. The method of the invention is not one that has any artificial or patentable effect other than the implementation of a scheme, which happens to use a computer to effect that implementation. There is no technical contribution to the invention or artificial effect of the invention by reason of the intervention of the inventors … [114] The claimed method in this case clearly involves what may well be an inventive idea, but it is an abstract idea. The specification makes it apparent that any inventive step arises in the creation of the index as information and as a scheme. There is no suggestion in the specification or the claims that any part of the inventive step lies in the computer implementation. Rather, it is apparent that the scheme is merely implemented in a computer and a standard computer at that. It is no part of the claimed method that there is an improvement in what might broadly be called “computer technology”. [118] In attempting to determine patentability, Australian courts have looked to the approaches taken in other jurisdictions, which have tended to favour patents involving physical effects and genuine improvements, rather than just ideas or their computerisation (box 9.1). Case law now requires that a technical contribution must exist for software to be patentable, and to reflect this IP Australia has updated their patent examiner’s manual (box 9.2). 2 Research Affiliates LLC v Commissioner of Patents [2014] FCAFC 150 268 INTELLECTUAL PROPERTY ARRANGEMENTS Box 9.1 Software patentability in other jurisdictions Australia has not been alone in clarifying the extent to which software can be patented. Some jurisdictions, including Germany, the United Kingdom, Canada, Thailand and New Zealand have all narrowed the definition of patentable subject matter in their patent legislation. Others, including Japan, Bangladesh and the United States have had the definitions of what constitutes a patentable ‘invention’ narrowed through judicial interpretation in the course of legal action (Sherman 2015). Of particular interest is how jurisdictions with significant software markets have changed their patent rules. In the European Union, a technical effect is required for patentability and article 52 of the European Patent Convention specifically excludes the patenting of methods of doing business and programs for computers. The provisions of paragraph 2 shall exclude patentability of the subject-matter or activities referred to in that provision only to the extent to which a European patent application or European patent relates to such subject-matter or activities as such. (EPO 2016a, sec. 52(3)) The wording ‘as such’ has given rise to discussions about what constitutes a computer program and to what extent the exclusion applies. It is accepted that a standalone computer program is excluded from the subject matter of patentability. But, the ‘as such’ means that a computer implemented invention, which incorporates a ‘technical effect’ and ‘technical contribution’, is not excluded from patentable subject matter. While a general practice is emerging, the wording has led to numerous court cases, and clear guidance is still lacking as to what should or should not be considered patentable material. In the United States, the legal finding of requiring a ‘machine or transformation test’ to be patentable has been further refined such that stating an idea is to be implemented in a computer does not make it patentable, as the subject matter is the abstract idea. The Federal Court took the view in Research Affiliates LLC v Commissioner of Patents [2014] FCAFC 150, that the Australian approach to patentability of software was consistent with that taken in the United States and United Kingdom. This view was affirmed by some participants in the inquiry including Summerfield (trans., p. 571) and Wilkinson: … Research Associates and RPL Central have also effectively narrowed the patentability of software…the result of Alice decision is largely similar … and thus there is now a high degree of similarity between the AU, US and UK approaches to patenting of software. (sub. DR359, p. 2) (continued) BUSINESS METHOD PATENTS AND SOFTWARE PATENTS 269 Box 9.1 (continued) Comparing jurisdictions: is software patentable? Is software broadly patentable What is the source of the caveat? What additional tests are used? Australia Yes, in certain circumstances United States Yes, in certain circumstances Interpretation of ‘manner of manufacture’, and improvement in computer technology as opposed to mere computer implementation The difference between improving an existing technological process and ‘generic computer implementation’ A ‘physical effect’ leading to a ‘useful product other than the implementation of a scheme’ A ‘machine or transformation’ test European Union No, subject to exceptions To what extent ‘as such’ in article 52(3) limits the exclusion (and thus makes more inventions patentable) A ‘technical effect’ or ‘technical contribution’ test Sources: Grant v Commissioner of Patents [2006] 154 FCR 62; Research Affiliates LLC v Commissioner of Patents [2014] FCAFC 150; Commissioner of Patents v RPL Central Pty Ltd [2015] FCAFC 177 in Australia; Alice Corporation Pty Ltd v CLS Bank International 134 SCt 2347 (2014) in the United States; European Patent Convention provisions in the United Kingdom and European Union. Box 9.2 What constitutes a technical contribution? In Research Affiliates and RPL Central the Full Federal Court considered computer implemented inventions and found that the implementation of a claimed method in computer hardware was not sufficient to confer patentability. However, being a method or scheme does not exclude it from patentability altogether, rather the invention in its entirety must afford a technical contribution. As noted in the RPL Central judgment: The fact that it is a scheme or business method does not exclude it from properly being the subject of letters patent, but it must be more than that. There must be more than an abstract idea; it must involve the creation of an artificial state of affairs where the computer is integral to the invention, rather than a mere tool in which the invention is performed. Where the claimed invention is to a computerised business method, the invention must lie in that computerisation. It is not a patentable invention simply to “put” a business method “into” a computer to implement the business method using the computer for its well- known and understood functions. [96] From these decisions, the following considerations are applied to determine whether an invention is in substance a manner of manufacture. Factors that work against a finding of a ‘technical contribution’ and being a patentable subject matter include: • the claimed method merely requires generic computer implementation • the computer is merely an intermediary or tool performing the method while adding nothing of substance to the idea • the method does not include any steps that are outside the normal use of a computer • statements in the specification that steps in the process can, in alternative embodiments, be implemented manually or offline. (continued) 270 INTELLECTUAL PROPERTY ARRANGEMENTS Box 9.2 (continued) Factors that help support a finding of a ‘technical contribution’ include: • the contribution of the invention is technical in nature • the invention solves a technical problem, either within or outside the computer • the invention results in an improvement in the functioning of the computer • the solution is solved irrespective of the data being processed • the solution provided by the computer invention involves more than the utilisation of the well-known function of a computer • the ingenuity of the invention is in a physical phenomenon in which the effect may be observed rather than in a scheme • the alleged invention lies in the way the method or scheme is carried out in a computer • the alleged invention lies in more than the generation, presentation or arrangement of intellectual information. Sources: Harrap and Irvine (2015); IP Australia (2016j). Case law, and subsequent guidance issued by IP Australia, are already affecting the proportion of software patent applications that are successful. Around 45 per cent of software patent matters (using the G06Q IPC category) in 2012 included an objection raised by a patent examiner for failing to meet the manner of manufacture test. By 2016, this proportion had risen to 67 per cent (IP Australia, pers. comm., 27 July 2016). The legal lens through which software patentability has been determined has answered the question of what is patentable in Australia (figure 9.2). But the broader, policy-relevant question as to whether software should be patentable, and whether the settings derived from recent case law are appropriate, remain unaddressed. BUSINESS METHOD PATENTS AND SOFTWARE PATENTS 271 Figure 9.2 9.3 To patent or not to patent: technical contribution is the question Assessing the effectiveness and efficiency of software patents Applying the Commission’s analytical framework to evaluate the effectiveness and efficiency of software patents raises a number of questions. These include: • do software patents provide incentives for inventions that would not have occurred in the absence of patent protection? • how do software patents impact follow-on inventions and the dissemination of knowledge? • are software patents the best means to provide an incentive for innovation in computer-enabled fields of innovation? 272 INTELLECTUAL PROPERTY ARRANGEMENTS Are software patents needed to encourage additional innovation? A threshold issue in assessing the effectiveness of software patents is whether innovation would occur in the absence of patent protection. Some have suggested that it would: Firms in the computer software and financial services industries were innovating rapidly long before it was thought possible to patent their innovations, yet they found effective ways to exploit their innovations without patents. (Hunt 2001, p. 11) This notion appears to be borne out in Australia’s case. Few innovators use patents in the fields most commonly associated with software, and many opt not to protect their innovation at all, or use other forms of IP protection (figure 9.3). Figure 9.3 IP protection methods in selected industriesa Per cent of firms 2014-15 80 Information Media and Telecommunications 70 Professional, Scientific and Technical Services Financial and Insurance Services 60 50 40 30 20 Few firms use patent protection 10 0 Patents Registration of Copyright or design trademark Secrecy or Complexity of confidentiality product design arrangements No IP protection methods Method used to protect IP a These ANZSIC divisions are selected as they include the most software-patent intensive industries, identified in figure 9.1. Note that the information media and telecommunications division includes industries like publishing, which may skew the figures towards copyright. Source: Commission estimates based on ABS (Selected Characteristics of Australian Business, 2014-15, Cat. no. 8167.0). One factor that may explain why patents are not widely used in software development is the short life span of software products, which is often less than the term of protection that patents afford. For example: • much software has moved away from discrete product life cycles and towards a system of continuous release, to the point where many companies release new iterations of software multiple times a day (Bosch 2014) BUSINESS METHOD PATENTS AND SOFTWARE PATENTS 273 • a study of German firms found that more than two thirds replace their software with improved products on an annual basis, and that around 40 per cent of enterprises’ product development takes less than six months (Blind, Edler and Friedewald 2005) • in Australia, the tax office allows firms to depreciate software developed in house over four years (ATO 2015), while the ABS uses an economic life of six years for software developed in house (ABS 2015b). Given it can take up to five years to finalise a patent in Australia, it is reasonable to presume that, on average, patent protection is unnecessary to incentivise software development. The short innovation cycles mean that inventors can recoup their investment from first-mover advantage alone (Lieberman and Montgomery 1988). While this is likely to be the case for most software, it is not necessarily the case for all software, or for the subset of software for which patent protection is sought. Bakels and Hugenholtz (2002) examined patented software in the European Union, and found instances where programs took much longer than the average to develop. Participants in this inquiry also noted that short development cycles do not necessarily rule out longer time periods to develop software infrastructure. For example, Qualcomm noted that the core technology currently used in their 3G devices was originally granted patent protection in the 1980s (trans., p. 665). The pattern of renewal for Australian software patents also suggests that software granted patent protection has a longer life than software in general. And ‘survival’ analysis of software patents lodged between 1990 and 1995 — as proxied using IPC codes — suggest that most are renewed up to 10 years, compared to 8 years for all patents over the same period (IP Australia, sub. DR612). The nature and magnitude of innovation costs also influences whether software needs patent protection. All else being equal, higher costs — especially higher upfront costs — would strengthen the case for patent protection as a means to recoup investment. In Australia, the costs of investing in a unit of software have fallen relative to the costs of research and development (R&D) in general, as well as relative to the costs of all capital goods (figure 9.4). The decline in the cost of investing in software indicates that the cost of development has fallen as well. What is unclear, however, is whether the costs of developing patentable software are materially different from that of all software. The Commission has not been able to identify any evidence to assess this in Australia. 274 INTELLECTUAL PROPERTY ARRANGEMENTS Figure 9.4 Prices deflators for investmenta Prices relative to 1990-91 250 Research and development All capital goods The cost of investing in R&D in general is rising... 200 Price deflator (Index 1990-91 = 100) Software 150 100 ... while the cost of investing in software is falling. 50 0 1990-91 1994-95 1998-99 2002-03 2006-07 2010-11 2014-15 a Based on gross fixed capital formation data. Source: Commission estimates based on ABS (Australian System of National Accounts, 2014-15, Cat. no. 5204.0). The act of holding a patent can also provide signals to banks and potential investors, and so play a role in securing capital to develop software (Mann 2005; Mann and Sager 2007). As put by one software developer: Without the availability of patent protection for the inventive features of our software we simply could not commit considerable capital to develop our product and our business. It would also be much harder to source and secure venture capital, unless our invention was protected. (Dalgleish, sub. DR201) Although patents may help provide a signal to financiers of value and potential appropriation, the signal is only as good as the quality of patents granted. A better quality patent system serves to provide a better signal, whereas weak patents can muddy the signal. As put by Wilkinson: Patents act as a key quality signal to the market place and assisted in securing funding as well as increasing the value of the company in the eyes of investors. (sub. DR359, p. 1) Do software patents help or hinder follow-on innovation? Software tends to be cumulative in nature. It can improve on or make use of existing products and processes, or involve a combination of both. New forms of software are typically enhancements of prior generations of innovations. For example, the recently released Pokémon Go is built on Niantic Labs’ original game Ingress (figure 9.5) (Allegra 2016); the Lotus 1-2-3 spreadsheet built on VisiCalc; and Microsoft’s Excel built on Lotus (Bessen and Maskin 2009). BUSINESS METHOD PATENTS AND SOFTWARE PATENTS 275 Figure 9.5 From Ingress to Pokémon Go The evolution of Niantic Labs Ingress Pokémon Go Follow-on innovation Ingress is a location based augmented reality game, where players can capture in game portals based on their location and real landmarks. The user-generated content from Ingress (popular landmarks), and the game’s concept (overlaying the virtual world in parallel to reality) helped create Pokémon Go. Niantic Labs used over 20 US-registered patents in developing the games (Hakes 2016). Source: Screenshots taken from Ingress and Pokémon Go. Granting a patent may serve to hinder the development of new software where innovation is cumulative. The behaviour of patent holders could turn key building blocks of development into barriers to follow on innovation: Specifically, in the software industry, progress is highly sequential: progress is typically made through a large number of small steps, each building on the previous ones. If one of those steps is patentable, then the patent holder can effectively block (or at least slow down) subsequent progress by setting high license fees. … Thus, in an industry with highly sequential innovation, it may be better for society to scrap patents altogether than try to tighten them. (Maskin 2012) Participants made similar observations during the course of ACIP’s review of innovation patents. For example, they noted that the software industry operates differently to other industries and benefited from ‘mixing and matching ideas’ from different projects and that the large number of software patents made it ‘near impossible’ for developers to search the existing stock of protected ideas in order to be sure that they were not infringing (OSIA 2013). 276 INTELLECTUAL PROPERTY ARRANGEMENTS Intel Corporation (sub. 66) notes a number of pitfalls that can arise from the granting of patent protection to software. One is the ‘hold–up’ problem (discussed in greater detail in chapter 15), where a patent holder can block incremental innovation, even if their patent only covers a small part of the total innovation upon which cumulative innovation is occurring. As noted by OSIA: In the modern world, no computer program exists in a vacuum. To be considered useful, a program must interact with other programs, written by other programmers … [This] can be done without infringing the other program’s copyright. But if the other program is patented, many of those things can be exceedingly difficult, or outright impossible, to accomplish without infringing on its patent. (2013, p. 5) Although patents may hinder follow-on innovation, preventing software from being patentable could also stifle dissemination. As put by the BCA: Unlike trade secrets, patents require that an invention be published openly, meaning that others are able to use that knowledge to generate follow-on innovation (under commercial arrangements). One response from software companies that would no longer be able to access patents would be increased reliance on trade secrecy or less public sharing of information publicly. (sub. DR587, p. 9) That said, given that few Australian firms use software patents, concerns about a loss of disclosure may be overstated. Indeed, the larger gains from disclosure of ideas are likely to be made in other jurisdictions that make greater use of software patents. Are patents the best means to encourage software innovation? Open source should be seen as a complement not a substitute for patent protection An emerging model of software development is one where code and ideas are not protected by choice. Rather than seek protection, some developers share their code, encourage third parties to copy, and contribute to the development of their software. This approach, referred to as open source, has also been adopted by larger companies including Google, IBM, and Sony (Google 2016; IBM 2016; Sony 2016). Open source approaches offer benefits for both the contributing software developers and innovating firms. Developers benefit from making software, as it allows them to learn new skills from collaborators (Lamine and Goran 2013; Sauer 2007). Firms gain as open source approaches are typically more adaptable, software can be built to accommodate follow-on innovations without the need for proprietary products, allowing for fast and dynamic improvements of their software product. BUSINESS METHOD PATENTS AND SOFTWARE PATENTS 277 The presence of open source has led some to question the need for software patents: … there is insufficient evidence that software patents result in increased innovation. … It is difficult to justify protection for a field of endeavour in which a significant amount of innovation comes from developers who have no interest in obtaining and seeking patents. One should be wary of those who argue that the industry will collapse without patents; just using any open source software product will prove them wrong. (Gonzalez 2006, p. 206) Using open source software is not without costs — a firm forgoes potential sales revenue when adopting the open source platform. As theorised by Haruvy et al. (2008) and Caulkins et al. (2013), when ongoing R&D costs are high, it may be worthwhile to release the software as open source and benefit from collaborative effort. However, where R&D costs are low and the software is of high value, a firm is better off retaining the software in house. Open source is not necessarily a good fit for all types of software. Some suggest that open source approaches lend themselves to more modular, self-contained and self-explanatory projects (Bollinger et al. 1999) and that proprietary software development models are more appropriate for larger projects with complicated objectives, which would benefit from the oversight of a core software developer (Stamelos et al. 2002). In practice, open source software and proprietary software work hand in hand. Open source allows for greater flexibility in innovation, whereas proprietary software potentially allows the firm to generate an income and be economically sustainable. As put by a former General Counsel for IP Licensing at Microsoft: The truth is that today we’re all mixed source companies. Every company that traditionally comes from an open source background has over time moved to the middle after realizing that in addition to the open source foundation, they also need proprietary offerings that will differentiate their services from others and therefore will enable them to build a viable business. (Guitierrez in Krill 2008) Thus, open source software development can be viewed to be working as both a substitute and a complement to proprietary software. Open source can encourage the creation of additional software, but not necessarily all. As put by Gutierrez: … striking a balance between [embracing open source software and brandishing patents] is one of the key things every commercial technology company must do in order to compete effectively. (Gutierrez in Letzing 2009) Patents can be a source of anticompetitive behaviour Where software patents afford broader protection than necessary, it can give rise to opportunistic litigation and create considerable uncertainty among inventors or those seeking to commercialise inventions (Bessen and Meurer 2008). In these circumstances, technological alternatives may not be commercialised and businesses and consumers suffer. As echoed by Bakels and Hugenholtz (2002, p. 25), ‘developers of Open Source 278 INTELLECTUAL PROPERTY ARRANGEMENTS Software are relatively vulnerable to patent infringement claims, particularly in respect of ‘trivial’ patents’. Even where parties do not avoid entry, the presence of many broad or low–quality patents may increase the time it takes for new entry to occur. In fast changing areas, such as software, a few months delay can relegate even a state of the art product to commercial failure (Hunt 2001; Hutchinson 2011). The way that patenting is used by firms can affect some types of competitors more than others. Patent thickets — networks of patents that make it difficult for new entrants to achieve protection without accidentally infringing on an existing claim — are likely to have a greater impact on smaller firms, as they do not have the means to ensure their products are not infringing (OSIA 2013). Whereas ‘standard essential patents’ tend to affect all firms — large and small (box 9.3). Box 9.3 Standard Essential Patents A standard essential patent (SEP) is a patent that must be used in order to be compliant with a technical standard. It is often claimed that SEPs discourage standard adoption because users fear being faced with litigation. Where there are a number of complementary patents owned by different firms, SEPs can also lead to patent thickets and hamper innovation. Galasso and Schankerman (2015) found that the negative effects of SEPs are more apparent in the field of information and communication technologies (ICT) — where software is ubiquitous. They argued that the negative effect is the result of highly fragmented patent ownership in this field, and that any transaction costs are exacerbated as users of SEPs need to negotiate with a larger number of rights holders. Intel Corporation highlighted the many SEPs in the ICT field and the costs associated with the use of such patents. By way of example, Intel noted that there are around 3000 SEPs used in WiFi technology. Of these SEPs, 32 patents attract royalties at a rate, which if applied to all the SEPs, would lead to a total royalty of around US$13. Such a royalty is around 13 times larger than the price of the cheapest WiFi chip on the market (Intel, sub. 66). Given software is highly used in the field of ICT, and the potential for unintended consequences, Australia should exercise caution when deciding whether to adopt new SEPs. These problems are not unique to software patents Some have raised concerns that the increased cost of market entry is not an unintended consequence of broad or low–level patents, but their raison d’être. There is some evidence from the United States that software patents are being used for strategic purposes (Bessen and Hunt 2007). And in Australia, patent attorneys advertise such possibilities for patents in general (chapter 8). However, it is not clear that problems associated with strategic use, or low-value patents more broadly, are any more pronounced for software than for other patentable inventions. BUSINESS METHOD PATENTS AND SOFTWARE PATENTS 279 The Commission has analysed the value of Australian patents based on different criteria. These include the number of times patents are cited, the number of technology classes they pertain to, whether they are filed in multiple countries, and the length of period that they are in force. These criteria are weighted into an index of relative patent values (appendix D). The Commission’s analysis reveals that there is a large tail of low-value software patents, but no more or less so than in other technology fields (figure 9.6). Figure 9.6 Software patents share a similar distribution to all other patentsa 2005–2010 15 All technology fields Per cent of patents Software (G06F & G06Q) 10 5 0 0 0.25 0.50 0.75 Index of patent scores (by technology field) a The composite patent value index is one that aggregates indexes for patent forward citations, scope, family size and renewal into a signal measure (these indexes are calculated by benchmarking against the 95th percentile of patents within the same technology field and with the same grant year). Because software patents do not appear to be used extensively in Australia, there is a small sample upon which to evaluate their value and the presence of strategic behaviour. Given this small sample size, any analysis should be treated with caution. Note that because the index of patent scores for software patents is calculated by comparing software patents to the ‘best’ patent in software, the distributions presented in this figure do not provide any information on the overall quality of software patents relative to other patents. Source: Commission estimates based on IPGOD 2016 edition (appendix D). Another indicator of strategic behaviour is the presence of patent thickets. The Commission has analysed the presence of patent thickets by identifying patent ‘triples’ — where three firms each hold patents that cite patents held by the other two (appendix D). While software patents do exhibit a higher ‘forward citation count’ relative to all patents, and this could be a precursor to thickets, the Commission has been unable to find any patent triples in the IPC codes used to proxy for software. Only when thickets are redefined as having ‘doubles’ — where two firms hold a patent that cite the other — is there 280 INTELLECTUAL PROPERTY ARRANGEMENTS evidence of possible thickets. Of the software patents examined by the Commission, only around one per cent were found to be part of a ‘double’. 9.4 Where to from here for software? While a number of participants have provided examples of software that clearly warrant patent protection, in general, patent protection is not critical for encouraging software innovation. Where patents are granted, they can limit, rather than encourage, some follow-on innovations. The policy issue is whether the patent system can be crafted so that only these worthy examples gain patent protection, but simpler computer-implemented innovations do not. Recent court decisions (and subsequent guidance issued by IP Australia) may enable the patent examination process to be a more effective filter in this regard. The Commission’s recommendations to raise the inventive step, require technical features in patent claims, and to introduce an objects clause (chapter 7), will also help to improve the quality of software patents. The question then is whether further measures are required, including the option of excluding software from being patentable subject matter — an option countenanced in other countries and raised by the Commission in its draft report. The High Court’s recent refusal to hear the RPL Central decision may have tipped the balance in favour of a wait and see approach. As the University of Melbourne remarked: Before any decision to exclude business methods and software as patentable subject matter, the University strongly supports lifting the test for inventive step and clarifying the manner of manufacture threshold of business methods and software. … IP Australia has taken measures to clarify the principles when examining computer implemented subject matter … In the medium term, this approach may address concerns relating to the harm claimed to be created by business methods and software patents, while allowing high quality innovation to be deserving of patent protection. (sub. DR560, p. 3) This approach allows for a degree of flexibility for patent examiners to determine what is worthy of a patent, and provides protection for inventions that are unanticipated at present, which a blanket exclusion would preclude (Dalgleish, sub. DR201). FINDING 9.1 Raising the inventive step, requiring technical features in patent claims, and the inclusion of an objects clause would better balance the patent rights of software innovators and users. BUSINESS METHOD PATENTS AND SOFTWARE PATENTS 281 More scrutiny of software patents is warranted in the future IP Australia is now giving effect to the decisions made in RPL Central and Research Affiliates in its examination of patents for computer implemented inventions. IP Australia has updated their patent examiners’ manual — which incorporates the requirement to establish a technical contribution in computer implemented inventions — and is considering patent applications under these new rules. Early evidence suggests that the new approach is narrowing the scope of patent protection (box 9.4). Box 9.4 Substance over style: studies of recent software patents Two recent decisions by the Australia Patent Office (APO) of IP Australia have provided new insights about the impacts that RPL Central and Research Affiliates have had on the patentability of computer implemented inventions. Both patent applications were for electronic slot machines that claimed to provide a new way to gamble, and both were initially rejected by patent examiners on the grounds they fail to meet the new criteria of manner of manufacture following the decisions in RPL Central and Research Affiliates. That is, the inventions: … are only patentable if what is claimed “as a matter of substance” meets the requirements for a manner of manufacture and in particular is not a mere scheme, abstract idea or mere information. (IP Australia 2016j, pt. 2.9.2.7) Konami sought a patent for gaming machines involving virtual spinning reels and the matching of Chinese Dominoes — dependent on the outcome free games are awarded to players. The firm’s attorneys suggested that because gaming machines are not business methods the two cases IP Australia referenced (Research Affiliates and RPL Central) did not apply. However, the hearing officer at the APO found that the substance of innovation lies within the new set of rules and gameplay (method or process), and are implemented using nothing more than a computer for its intended purpose. The hearing officer rejected Konami’s submissions and found that the claimed innovation was not a manner of manufacture. Aristocrat claimed a multi-game gaming machine, which allows a player to select a game from a range of options displayed simultaneously on a video screen, and to select a bet denomination listed under each game to gamble that amount in that game instantly. The firm’s attorneys argued the substance of the invention was within the interface with particular features; including the touch display that presented an option for selecting either a game or the denomination, and that the computer is not merely an incidental component of the invention. The hearing officer found that the evidence provided was irrelevant, but instead highlighted that the substance of the innovation is in the single action to select both the game and the bet denomination — making it easier to gamble. The hearing officer also stressed that Aristocrat had failed to direct any attention to this, nevertheless the innovation as a whole constitutes a technical contribution and met the manner of manufacture criteria. Both of these applications have been instructive in showing how IP Australia is applying the new manner of manufacture test for computer-implemented inventions, but are also suggestive that greater clarity of the rules could be beneficial for applicants. Sources: Aristocrat Technologies Australia Pty Limited [2016] APO 49; Konami Gaming, Inc. [2016] APO 46. What is less clear is where the standard of patentability of computer implemented inventions will end up. At face value, more of these inventions appear to be rejected than 282 INTELLECTUAL PROPERTY ARRANGEMENTS previously. However, inventors and their attorneys are still testing the boundaries of what the law will allow. The Konami and Aristocrat decisions indicate that there are still difficulties in determining ‘what’s in and what’s out’. A better evidence base is needed so users can better understand the system, and policymakers can fully understand the revised boundaries and how they compare to those applying in other jurisdictions. The community would also benefit from having more detailed commentary on the patent applications that are rejected based on the findings in RPL Central and Research Affiliates. To that end, the Commission considers that IP Australia should collect and publish information on: • patent applications where the new criteria are considered • what was granted and what was rejected, as it relates to the manner of manufacture test • how the decisions in Research Affiliates and RPL Central have affected IP Australia’s consideration, and the likelihood of patentability, of software. The growth of computer technology, and its pervasiveness in many innovations, means that the patentability of software will continue to take on greater importance. The Commission considers that closer scrutiny of software patents is warranted to ensure that they are working in the interests of innovators and the rest of the community. The Commission has recommended forming an IP policy group which would, amongst other things, identify and provide advice on how to remedy deficiencies of the IP system as they emerge (chapter 17). This forum provides the appropriate vehicle to use the evidence collected by IP Australia to address: • whether the scope of patentability of computer implemented inventions is too wide or narrow, in light of the RPL Central and Research Affiliates decisions • whether the manner of manufacture criteria is still relevant and useful for modern computer-implemented inventions • whether software should be patentable at all • how emerging technologies have been affected by software patentability • whether any further changes to patent law or examination practice are warranted in the public interest. BUSINESS METHOD PATENTS AND SOFTWARE PATENTS 283 10 Pharmaceuticals – getting the right policy prescription Key points • Pharmaceuticals are an archetypal patentable product with high research and development costs coupled with the ease of copying. Proponents reasonably argue that without patent protection many valuable medicines would not be developed. • In addition to ‘standard’ patent protection, the pharmaceutical sector benefits from extensions of term (EoTs), of up to five years. – EoTs prolong market exclusivity and impose considerable costs on consumers, government, and taxpayers (through the Pharmaceutical Benefits Scheme). The policy has been ineffectual in attracting R&D to Australia, and those firms that do invest spend little here. – EoTs therefore need to be carefully targeted to instances of unreasonable regulatory delay in the Therapeutic Goods Administration’s approval process. The Commission estimates this approach would save taxpayers $258 million per year. • Data protection is also afforded to the test data used for regulatory approval for a period of five years. Data protection is an automatic right, it is not assessed, cannot be challenged and can be detrimental to competition. Its duration should not be extended. – Instead, the Australian Government should work cooperatively with other countries for the publication of protected data. • Biologic drugs differ from standard pharmaceuticals. They are difficult, and costly, to copy and produce. Despite contrary claims, there is no substantive evidence that patents do not provide sufficient protection for biologics. – Extending data protection for biologics will likely overcompensate most products and increase consumer costs. • Some pharmaceutical companies use the patent system strategically to extend their period of market exclusivity. – Incremental patenting (or evergreening) is likely occurring to some extent in Australia and is best addressed through proposed changes to the inventive step for patents. – While there is no evidence of ‘pay-for-delay’ settlements in Australia, this may simply reflect the lack of monitoring arrangements, rather than the absence of such activity. In those jurisdictions where monitoring occurs, pay for delay is found to be costly. Australia should introduce similar monitoring arrangements to detect and deter such behaviour. The pharmaceutical sector is a prime user of the patent system (chapters 7 and 8) as well as the beneficiary of bespoke IP arrangements. PHARMACEUTICALS — GETTING THE RIGHT POLICY PRESCRIPTION 285 Consistent with the significant rents involved, and the implications that poorly designed policies have on the health and wellbeing of the community, arrangements covering pharmaceuticals are subject to periodic review. The last major review — the Pharmaceutical Patents Review (PPR) — was undertaken in 2013 (Harris, Nicol and Gruen 2013). The PPR made recommendations relating to extensions of term, data protection and trade negotiations, among other things. 1 Given the contemporary nature of that review, and the broad nature of this inquiry, the Commission has limited its focus to key aspects of intellectual property (IP) arrangements affecting pharmaceuticals. Within these key areas, the Commission has drawn upon the PPR for data and evidence. However, in forming its recommendations, the Commission has also drawn on additional information and conducted its own analysis. The chapter begins by briefly describing characteristics that distinguish the pharmaceutical sector from other sectors (section 10.1). It then considers four policy issues: • the appropriate arrangements for any extensions of pharmaceutical patent term (section 10.2) • arrangements governing manufacturing drugs for export purposes (section 10.3) • the role of data protection, including in relation to biologics — an emerging class of drugs (section 10.4) • the potential for strategic behaviour (section 10.5). 10.1 Why focus on pharmaceuticals? Pharmaceuticals are scientifically complex and costly to develop. While many other products share these attributes, few attract the degree of public policy attention afforded to pharmaceuticals. The attention is unsurprising given the impact pharmaceuticals have on the health of individuals — in some instances a matter of life or death. The importance of pharmaceuticals for health and wellbeing leads, in turn, to three main considerations: that there is sufficient investment in the creation of new drugs; that drugs available in Australia are safe; and that drugs are accessible to the general public (that is, they are affordable for those who need them, and available in a timely manner). Each of these considerations can be affected by IP arrangements and must be balanced to produce the best overall outcome for the Australian community. Notably, Australia is a net importer of pharmaceuticals — in 2012-13 pharmaceutical exports were $3.9 billion and imports were $10.5 billion (DIIS 2014). 1 Upon release of the PPR report, the Australian Government (2014a, p. 1) indicated that it did not intend to respond to it, but ‘may take information in the report into account when considering future policy’. 286 INTELLECTUAL PROPERTY ARRANGEMENTS The stakes are high for pharmaceutical companies As with any industry, pharmaceutical companies will only invest in the development of a new product if there is a likelihood that the market will allow them to secure a viable return on their investment (chapter 2). Development costs and times are significant The cost to a firm of developing pharmaceutical products is generally acknowledged to be large (measured in hundreds of millions of dollars). However, the precise extent of spending required in research and development (R&D) for an average pharmaceutical product is the subject of considerable debate, with significant variation between published estimates (box 10.1). Several factors explain the large costs of bringing a new drug to market. Pharmaceuticals go through a considerable development phase, although some costs, particularly those associated with early phase research, are not always borne solely by the private sector. As the McKeon Strategic Review of Health and Medical Research (2013, p. 60) noted, public funding (an estimated $2.9 billion in 2011-12) accounts for almost two-thirds of health and medical research expenditure in Australia, with the remainder ($1.7 billion) coming from business and private not-for-profit entities. 2 The bulk of the government spending relates to universities and funding through the National Health and Medical Research Council. Pharmaceutical research makes up approximately two-thirds of all health and medical research spending (Harris, Nicol and Gruen 2013, p. 31). Pharmaceuticals must also go through a regulatory approval process before coming to market. In Australia, a pharmaceutical product can only be supplied to the Australian market once it has been entered on the Australian Register of Therapeutic Goods (ARTG).3 The data and clinical trials that inform approval processes can be costly. The PPR noted that nearly half — $700 million of the estimated $1.5 billion — of R&D costs are spent on clinical trials. Regulatory approval costs are increasing due to the growing size (number of patients) and complexity of clinical trials (increasing the cost per patient) (OHE 2012). The multi-stage development and regulatory approval process (figure 10.1) also impacts on the time it takes to get pharmaceutical products to market. Medicines Australia (sub. 44, p. 5) submitted that the time to market is ‘between 10 and 15 years’. 2 The figures cited refer to overall health and medical research expenditure by destination sector. 3 For low risk medicines, regulation focuses on the safety of the product and the manufacturing process. For higher risk products, including all prescription medicines, the TGA also evaluates the product’s efficacy. The TGA also conducts ongoing monitoring of the safety, quality and efficacy of products on the market. PHARMACEUTICALS — GETTING THE RIGHT POLICY PRESCRIPTION 287 Box 10.1 Counting the cost: estimating pharmaceutical R&D Developing pharmaceuticals is an expensive (and time-consuming) process. But the question of just how expensive is a vexed one. This is due to a number of factors. First, estimates are complex — identifying and allocating the cost of many initial compounds to an eventually approved medicine involves considerable judgment. Second, given the considerable sums and lengths of time involved, small changes to variables (such as the discount rate applied) can have a large effect on the overall figures. Third, gaining adequate and reliable data can be difficult, as the holders of data are the pharmaceutical companies who have tended to only release incomplete information (on confidentiality grounds), or only release information to ‘select’ researchers. Submissions by Medicines Australia illustrate the degree of variance involved in such estimates. In a submission to the 2013 Pharmaceutical Patents Review (PPR), Medicines Australia quoted an average cost of US$1.5 billion to bring a new drug to market (Medicines Australia 2013). The estimate is drawn from an Office of Health Economics (OHE, an English research and consulting firm) study. In that study, the OHE (2012, p. v) noted that estimates were increasing: Published estimates of the mean (average) cost of researching and developing a successful new medicine suggest an increase in cost over the last decade — from the estimate of US$802 m by DiMasi et al (2003) at 2000 prices (US$1,031 m at 2011 prices) to the estimate by Paul et al (2010) of US$1,867 m at 2011 prices. More recently, in their submission to this inquiry, Medicines Australia cited an average cost of US$2.6 billion (sub. 44, p. 5). This estimate comes from a study by the Tufts Center for the Study of Drug Development. The study’s results were released in November 2014 and reportedly included cost estimates from 10 firms, which accounted for 35 per cent of the top 50 firm pharmaceutical sales. The study has been the subject of considerable criticism. Some noted the lack of transparency (results, but not the complete study, were released) and that 40 per cent of Tuft Center funding comes from the pharmaceutical industry (The Wall Street Journal 2014). Doctors without Borders (2014) pointed to substantially lower estimates (US$50 to US$186 million) and highlighted public sector contributions to R&D funding. The Union for Affordable Cancer Treatment (2015) also raised concerns relating to clinical trial sizes, specific cancer drugs, tax credits and public funding. Academics examining 13 previous pharmaceutical R&D cost studies also noted the substantial variance in estimates: Published estimates vary 4-fold even when restricted to studies published in the past 10 years (pertaining to drugs developed during the 1990s): with cash estimates ranging from Young’s 2001 estimate of USD$207 million to Paul’s 2010 estimate of USD$883.6 million, and capitalized estimates ranging from the Global Alliance for TB Drug Development’s 2001 estimate of $290.6 million to Paul’s 2010 estimate of $1.8 billion. (Morgan et al. 2011, p. 9) These Academics concluded that the generally confidential nature of the data used in the pharmaceutical cost estimates made it difficult to verify their quality: Despite three decades of research in this area, no published estimate of the cost of developing a new drug can be considered a gold standard. Existing studies vary in their methods, data sources, samples, and therefore estimates. While some methods are methodologically strong and some findings have been widely cited, the fact that the data and even the subjects of investigation are kept secret make it impossible to assess validity and reliability (Morgan et al. 2011, p. 11). 288 INTELLECTUAL PROPERTY ARRANGEMENTS Figure 10.1 The development cycle of a pharmaceutical Early Phase Research 5000 – 10 000 compounds 4 – 6 years Pre clinical testing 250 compounds 1 year Standard Patent Term 20 years Clinical trials 5 compounds 6 – 7 years Regulatory Approval 1 Compound 0.5 – 2 years Extension of Term Up to 5 years Sale and Post-Marketing Surveillance 1 Compound Continuous Source: Harris, Nicol and Gruen (2013). Many of the overall R&D costs are one-off. The Commission understands that, in most cases, where drugs are developed and approved for sale in larger overseas markets, the additional burden of obtaining regulatory approval in Australia (relying largely on data already submitted overseas) is not large and would not be material to the global returns available from a given drug. Global trials are largely conducted with a view to accessing much larger markets. It is also important to note that some of the regulatory compliance costs borne by pharmaceutical firms would have been incurred as part of prudent commercialisation (to ensure effective products and reduce the risk of negligence actions). Another factor that drives the cost of R&D is the risk involved in the development process — many drugs will fail at some stage of testing and not make it to market. Medicines Australia (sub. 44, p. 5) submitted that the level of risk was high and that ‘up to 93% of PHARMACEUTICALS — GETTING THE RIGHT POLICY PRESCRIPTION 289 potential therapeutic molecules do not make it beyond [the] clinical stage.’ This risk of failure is likely increasing as companies explore more experimental treatments and regulators respond with increased testing requirements (OHE 2012, p. viii). However, the new drug success rate varies between phases of development. A large number of molecules are filtered out in the early and pre-clinical phases and so not all incur the large expense of clinical trials. Where products do come to market, pharmaceutical companies reap rewards in a global market and the returns can be significant. Analysis of US pharmaceutical companies shows their profits were 3.2 times higher than non-pharmaceutical companies, and their return on assets was two to three times higher than the median for large companies (Harris, Nicol and Gruen 2013, pp. 33–34). Intellectual property arrangements bolster firms ability to earn returns The presence of significant R&D costs is only a barrier to investment where other parties can ‘free-ride’ on an innovating firm’s R&D efforts and so, make it hard for them to earn a return (chapter 2). In the absence of any policy intervention, free riding could be problematic in pharmaceuticals due to the ease of copying (small molecule) drugs, compounded by the (typically) low cost of manufacturing drugs (relative to the large cost of development). Patents seek to address concerns about free-riding. Patents can be granted not only to the medicine itself, but also for the way in which it is formulated, produced and used (Harris, Nicol and Gruen 2013, p. 25). The pharmaceutical industry is a prominent user of the patent system, ranking amongst the top patented technology areas in a number of jurisdictions, including Australia (figure 10.2). Reflecting the global nature of the industry, the vast majority of pharmaceutical patent applications in Australia are filed by foreign applicants. Only 4.3 per cent of applications filed between 2001 and 2014 were made by Australian residents — applicants from the United States and Switzerland accounted for around 47 and 7 per cent respectively, of foreign applications (Commission estimates based on Intellectual Property Government Open Data (IPGOD 2015 edition)). 290 INTELLECTUAL PROPERTY ARRANGEMENTS Figure 10.2 Pharmaceutical firms are prominent users of patentsa,b Switzerland 1st United Kingdom 1st Australia 3rd United States 3rd Canada 4th Germany 10th 0 2 4 6 8 10 Per cent of patent applications by residents, 2000-2014 12 14 a The ranking (1st, 3rd) refers to the position of pharmaceutical patents compared to other fields of technology within each country, between 2000 and 2014, and the axis denotes pharmaceutical patents as a percentage of total patent applications within each country in the same period. b This figure compares the number of pharmaceutical patent applications by the residence of the applicant. A similar comparison can be made using the number of patents granted by different Intellectual Property offices between 2000 and 2014. Pharmaceutical patents accounted for 12.1 per cent of patents granted in Australia (the highest ranked technology field), 7.0 per cent of patents granted in Canada (ranked second), 4.0 per cent of patents granted by the European Patent Office (ranked 9th) and 2.3 per cent of patents granted in the USA (ranked 15th). Source: WIPO (2015f). In addition to ‘standard’ patent protection, the pharmaceutical sector benefits from a number of other bespoke IP arrangements. • Pharmaceuticals can qualify for an additional five years of patent protection. • The data that are submitted in support of regulatory approval processes are also protected for a period of five years. Generic manufacturers seeking to enter the market during the period of data exclusivity must independently prove that their pharmaceuticals are safe and effective, even though their products are chemically identical to those of previously approved drugs. The stakes are high for the government and community The overarching, nationally agreed objective of Australia’s health system is to ‘improve the health of all Australians and to ensure equity of access and the sustainability of the Australian health system’ (SCRGSP 2016, p. E.6). The Department of Health, expanding on this objective, noted that it aims to accommodate: PHARMACEUTICALS — GETTING THE RIGHT POLICY PRESCRIPTION 291 … a balance of public and private interests to support the healthcare needs of the Australian community, and encourage commercial health sector interests to maintain investment in effective healthcare innovations to improve health outcomes. (sub. 84, p. 2) In this context, the regulatory settings (and IP settings) governing pharmaceuticals must provide a balance not only between ensuring that new drugs are developed and that they are safe and effective, but also ensuring that they are accessible and affordable. While the need to strike a balance is not unique to pharmaceuticals (chapter 2), the costs of getting it wrong are more readily apparent. These costs are borne in the first instance by the Australian Government (which subsidises the cost of medicines through the Pharmaceutical Benefits Scheme (PBS)) and are ultimately borne by taxpayers. Reflecting the price and volume of use of pharmaceuticals, the PBS is an expensive scheme. In the year ending 30 June 2015, government expenditure on the PBS was just over $9 billion. The community at large also bears costs — while Government expenditure accounts for just over 80 per cent of the total cost of PBS prescriptions, consumers contribute the remaining $1.5 billion (Department of Health 2016a). The entry of generic pharmaceuticals following the expiration of patent protection can drive significant savings for the government and broader community. Competition within the market for a given drug drives down the price. In Australia, market entry of the first competitor4 triggers an automatic statutory price reduction under the PBS, currently at 16 per cent. Price disclosure policies also cause reductions in prices under the PBS. These reductions are generally ongoing and can be substantial. In April 2016, price disclosure led to reductions ranging from 10 per cent to just over 87 per cent (Department of Health, pers. comm., 10 June 2016). Based on disclosure rounds for medicines between April 2011 and April 2016, the Commission estimates that the average overall price drop for a given drug 2 years after the entry of a competitor product is around 40 per cent. 5 10.2 Extensions of pharmaceutical patent term Reflecting the protracted development phase, patent applications for pharmaceuticals often occur well before the product is brought to market. These long lead times erode the effective patent life for pharmaceuticals. Recognising this, the Patents Act 1990 (Cth) provides for an EoT, so long as: • the patent (both the disclosure and the claim) is for a pharmaceutical substance per se, or pharmaceutical substances ‘when produced by a process that involves the use of 4 That is, a new brand or product with the same manner of administration that is bioequivalent or biosimilar. 5 This figure includes the 16 per cent statutory price drop and observed rounds of price disclosure from a sample of almost 1400 products. 292 INTELLECTUAL PROPERTY ARRANGEMENTS recombinant DNA’ (case law has expanded this to include compounds or mixtures of substances) • products containing, or made of, the substance in question are included on the ARTG • the time between the filing date of the patent and the ‘first regulatory approval date’ (typically listing on ARTG) is at least 5 years • the term of the patent has not been extended before. The duration of an extension is calculated by reference to the so-called ‘delay’, which is calculated as the length of time from the filing of the patent up until the date of marketing approval by the Therapeutic Goods Administration (TGA). An extension of term (EoT) is granted for a period of this delay, minus five years (so, if the delay is five years or less, there is no extension, and if the delay is eight years, there is an extension of three years). EoTs cannot be longer than five years. This allows a maximum patent life of 25 years and a maximum ‘effective market life’ (the period from market approval to patent expiry) of 15 years. EoTs are relatively common. Between 2003 and 2010, of all the new medicines (including formulations and combinations of existing chemical entities) approved by the TGA, the PPR estimated that 21–24 per cent would have received an EoT. Extensions of term are more common among new chemical entities, with around 58 per cent receiving an EoT over the same period (Harris, Nicol and Gruen 2013, p. 61). Applications for EoTs are typically accepted — from 1999 to 2014, nearly 95 per cent of the 697 applications made were accepted (IPGOD 2015 edition). More than half of all patents extended have received the maximum 15 years of effective patent life, and the median life has remained at or close to 15 years each year since the introduction of the current scheme (Harris, Nicol and Gruen 2013, pp. 203–204). What’s the case for extensions of term? At first glance, EoTs appear to have some policy appeal. As Pfizer put it: If there are delays in obtaining regulatory approval for new products, patent holders ought to be compensated. Without the modest and partial restorations of marketing exclusivity provided by patent term extensions, innovators would have less incentive and justification to make the substantial R&D investments needed to sustain the pharmaceutical innovation process. Patent term extension therefore represents an appropriate and necessary recognition by governments of the increasingly heavy burden of expense and risk incurred by innovators as a result of government requirements imposed during the R&D and regulatory review process. (sub. 83, pp. 4–5) However, closer inspection reveals some deficits with these arguments. PHARMACEUTICALS — GETTING THE RIGHT POLICY PRESCRIPTION 293 It is not clear that standard patent term is insufficient The policy history behind the current EoT scheme is telling. Prior to 1990, the EoT scheme applied to all patents, not just pharmaceuticals. That scheme allowed for extensions of the then 16 year standard patent in cases where there had been ‘inadequate remuneration’ (Lawson 2013a). A 1984 review of the scheme by the Industrial Property Advisory Committee 6 recommended its abolition, and noted the substantial likelihood of net social costs from extensions: In the view of the majority, in the absence of contrary empirical evidence, it strains credulity to contemplate that research or innovation investment decisions, made early in the life of the invention, could ever be materially influenced by the prospective availability of an extension after expiration of the initial 16 year term to compensate for inadequate remuneration, particularly when allowance is made for discounting. On the other hand, such extensions would increase social costs. (1984, p. 45) The Australian Government accepted this recommendation and repealed the general EoT scheme, but introduced a pharmaceutical-specific scheme intended to foster the pharmaceutical industry in Australia. Subsequently, the standard patent term was extended to 20 years by the Patents (World Intellectual Property Organization) Act 1994 (Cth) to give effect to TRIPS. Despite this across-the-board increase in patent life, the then Government reaffirmed its commitment to an effective 15 year patent life for pharmaceuticals on the grounds of regulatory delay (Harris, Nicol and Gruen 2013). Parity is not a convincing rationale The current scheme has not been justified on net economic benefit grounds. Parity has often been put forward as a rationale for the existence of the pharmaceutical-specific EoT scheme. The Explanatory Memorandum for the introduction of the current EoT scheme notes one of the scheme’s objectives is: … to provide an ‘effective patent life’ – or period after marketing approval is obtained, during which companies are earning a return on their investment – more in line with that available to inventions in other fields of technology. (Harris, Nicol and Gruen 2013, p. 63) In discussions of pharmaceutical EoTs, comparisons with other fields rarely centre on the sufficiency of profits earned across fields, but instead focus on requirements for regulatory approval. However, approvals are not unique to the pharmaceutical industry. Approvals are commonly required for a range of products and services for human health and safety, environmental and community wellbeing reasons. Industries with similarly pronounced 6 A minority of the Committee noted that delays caused by regulatory approval could truncate patent life for products such as pharmaceuticals, agricultural and veterinary chemicals. The minority recommended extensions equal to the regulatory delay, subject to a maximum of four years. The majority rejected this approach, noting that many products are subject to delay, and a range of factors affect the financial returns available from patents (IPAC 1984, p. 45). 294 INTELLECTUAL PROPERTY ARRANGEMENTS safety concerns, such as aircraft manufacture, do not benefit from extensions of patent term: … regulatory delay affects many innovations in many industries in many different ways. For example, automotive emissions, building and sanitary systems, telecommunications, human medicines and agrochemicals are all subject to regulation which may significantly delay the marketing of new products. (IPAC as quoted in Lawson 2013a, p. 389) Discussions of fairness focussing only on regulatory approvals ignore the reality of the returns that can be obtained in the market. As Bilir (2014) identified, many product life cycles in other industries are shorter than 20 years. 7 Accordingly, the effective market life of patents in those industries would also be shorter. And as chapter 7 outlines, the majority of non-pharmaceutical patents lapse well before 20 years. Aims to attract and stimulate investment have not been realised Another mooted objective of the scheme was to provide a system that was ‘competitive with other developed nations’, to counter perceptions of Australia as a ‘hostile environment’ (Lawson 2013a, p. 389), and so attract pharmaceutical R&D to Australia. However, the evidence suggests that EoT policies have been ineffectual in attracting R&D to Australia. There was no notable (above trend) increase in Australian investment following EoT changes (Harris, Nicol and Gruen 2013). Data provided to the Health Department as part of the EoT process adds further weight to this conclusion. While the data is incomplete due to compliance issues (section 10.6), of those companies who both received an EoT and provided data in their returns over the period 2009-10 to 2013-14: • only 39 per cent (36 of 92 returns) reported any R&D expenditure in Australia • the proportion of companies reporting R&D in Australia declined over the period, from a high of 58 per cent in 2010-11, to a low of 18 per cent in 2013-14 • those that did invest in R&D in Australia spent little here — average expenditure was only $3.9 million. 8 Further, participants in the Commission’s hearings conceded that EoTs in Australia were unlikely to lead to investment in new drugs, or to decisions to locate investment in Australia. For example, the Institute of Patent and Trade Mark Attorneys (IPTA) stated that: … I don’t think anyone could argue … that a Danish pharmaceutical company decided … we’re going to invest $5 billion and come out with a new diabetes drug because Australia has got a five 7 For example, electronics and computers have life cycles of approximately seven and eight years respectively, while shipping containers and cutlery, handtools and hardware have some of the longest identified life cycles (by patent citation lags) of just over 10 years (Bilir 2014). 8 There was significant variation around this average, with a standard deviation of $6.5 million. PHARMACEUTICALS — GETTING THE RIGHT POLICY PRESCRIPTION 295 year patent term extension that — no way. But along the way they will test where they’re at with patent term generally, including the Australian patent term, and they will decide whether or not to continue on based on the amount of patent term left internationally … (trans., p. 682) Similarly, Medicines Australia and the International Federation of Pharmaceutical Manufacturers and Associations acknowledged that, given the distance in time between the decision to begin development of a drug and the end of the 20-year patent term, EoTs were ‘more relevant when you’re thinking about your medicine launching [into the market in a given jurisdiction] than at the R&D phase’ (trans., p. 343). The lack of a nexus between Australia’s EoT scheme and pharmaceutical R&D is unsurprising: • Australia represents two per cent of global pharmaceutical revenues, and less than 0.3 per cent of pharmaceutical R&D (Harris, Nicol and Gruen 2013). The settings in the larger markets of the US and EU are far more determinative in firms’ decisions to invest in the creation of new drugs. • As the EoT is at the end of a patent term, the net present value of increased protection in a small market 20 years into the future would not be a substantial incentive for global companies contemplating the creation of new drugs. The PPR illustrated this by calculating the net present value of the incentive for a hypothetical drug (box 10.2). 296 INTELLECTUAL PROPERTY ARRANGEMENTS Box 10.2 The incentives and costs of extensions of term The PPR (Harris, Nicol and Gruen 2013) used Net Present Value (NPV) analysis to assess the magnitude of the R&D incentive effect of Extensions of Term (EoTs) relative to their cost for a hypothetical drug. They assumed that the drug earned $2.5 billion in revenues over its patented lifetime and that 70 per cent of this revenue was earned during the extension period. The PPR analysed PBS expenditure data for 2007–12 and found that 9 per cent of drugs fit this revenue profile. Assuming pharmaceutical firms apply a commercial discount rate of 13 per cent (reflecting the private cost of capital and a risk premium), the NPV of an extension for the drug, calculated at year 10, would be worth $370 million to the firm. This extra return should incentivise the firm to undertake additional R&D investment in the drug; however the PPR argued that this investment is unlikely to occur in Australia because Australia’s pharmaceutical industry is relatively small. The analysis in the PPR assumed that the additional R&D investment in the drug that occurs in Australia at year 10 would only be equal to 2 per cent of the NPV or $7.5 million (under a scenario which assumes a greater level of domestic R&D activity, this goes up to $93 million). In contrast, the PPR calculated that the net present cost to the Australian Government would be approximately $1.4 billion. This is calculated by discounting the revenues earned by the firm in the extension period (implicitly assuming that the Australian Government fully subsidises the drug) by a social discount rate of 3 per cent (reflecting the opportunity cost of capital to government). 9 This calculation does not factor in any subsidies that would be paid during years 21-25 of the patent if no extension were granted. This analysis demonstrates that the R&D incentive effect of EoTs is likely to be relatively small even in the most favourable of cases. The analysis does not include all the necessary components to be considered a definitive cost-benefit assessment of EoTs (for example, it is conducted for a single hypothetical drug rather than at the system-wide level and it does not consider the subsidies that would be paid if no extension were granted or the loss of benefits that consumers receive if the drug is not made available in Australia). That said it is likely that such an assessment would show that the policy represents a net cost to Australians. The costs of EoTs are significant Not only are EoTs unwarranted, they are also expensive. As EoTs exist over and above the standard patent system, they exacerbate the costs (and benefits) of patents themselves — returns to innovators come at the expense of competitors, consumers and government. The latter costs arise because EoTs impact on the entry onto the market of generic (competitor) products. In some cases, an EoT may mean that the remaining life of a product (before newer, better products are introduced to replace it) is not sufficient for generic companies to enter the market. In other cases, an EoT may delay generic entry by up to five years, leading to significant costs by delaying the market and statutory price reductions under the PBS. 9 Past Commission research suggested that 3 per cent is a lower bound for the social rate, with 8 per cent more appropriate (Harrison 2010). PHARMACEUTICALS — GETTING THE RIGHT POLICY PRESCRIPTION 297 The PPR (Harris, Nicol and Gruen 2013, pp. 75–78) identified that the annual cost to the Australian Government of EoTs was in the order of $244 million per annum. Based on more recent data and a slightly more nuanced approach to examining the price impacts of generic entry (appendix H), the Commission has estimated that the cost to the Australian Government (and ultimately taxpayers) of EoTs is $260 million per annum — a cost not dissimilar to that estimated by the PPR. This cost only accounts for government expenditure on the PBS, which makes up 80 per cent of the cost of all PBS medicines. Consumer costs make up the remaining 20 per cent. While the PBS co-payment system shields consumers from some price effects, Australian consumers would also benefit from reduced prices. 10 Further, for those medicines not on the PBS, consumers would receive the entirety of any price reductions from increased competition. Overall, the Commission considers that the policy case for EoTs was never made and that such provisions impose a net cost on the community. Industry has not provided evidence that EoTs in Australia result in new pharmaceuticals coming to market that would have not otherwise. As such, the preferred policy would be to dispense with such provisions altogether. But international commitments constrain Australia’s policy flexibility in relation to EoTs. the Australia-United States Free Trade Agreement (AUSFTA) requires that ‘compensation’ be offered for ‘unreasonable curtailment’ of the patent term. Article 18.48(2) of the Trans-Pacific Partnership (TPP) includes a similar provision: With respect to a pharmaceutical product that is subject to a patent, each Party shall make available an adjustment of the patent term to compensate the patent owner for unreasonable curtailment of the effective patent term as a result of the marketing approval process. At a minimum, the Commission considers that calls for further EoTs should be rejected and the EoT scheme be more carefully targeted. There are a number of elements in structuring an EoT scheme that can be used to improve targeting. These are discussed in turn below. Better targeting extensions of term Even if industry were to make a case that EoTs in Australia do result in pharmaceuticals coming to market that would not have otherwise, the case for better targeting extensions of term would still be strong. This is because currently EoTs in Australia are offered where the time between filing a patent and listing on the ARTG is at least five years. These arrangements represent a poor basis for measuring regulatory delay because they allow the applicant’s decisions, rather than the regulator’s actions, to start the clock in terms of 10 For example, based on previous (2010) reforms, Sweeny (2013) estimated that consumers would receive 6 per cent of the total benefits from reductions in the price of PBS medicines, with the remaining 94 per cent going to government. 298 INTELLECTUAL PROPERTY ARRANGEMENTS calculating delay. Regardless of the efficiency of the regulator or the standards they apply, where a patent holder delays filing for regulatory approval, a pharmaceutical product will be eligible for an EoT. There is evidence that applicants’ actions are contributing to the measured ‘delay’. As the Generic and Biosimilar Medicines Association (GBMA, sub. 67) pointed out, given the relatively small size of the market in Australia, it is common for pharmaceutical companies to only seek regulatory approval here after approval has been granted in the larger markets. The GBMA submitted analysis showing the delay in submission for regulatory approval compared to the United States between January 2010 and November 2012: It consider[ed] the average time elapsed from dossier submission in the US until dossier submission in Australia, by reference to products in respect of which an application for an EOT was submitted in Australia. … These results show that on average sponsors do submit dossiers to the TGA later than equivalent submissions to the FDA, the median delay being 297 days in 2012, 236 days in 2011 and 549 days in 2010. (GBMA, sub. 67, attachment, p. 20) Elements left in the applicants’ control can become part of an overall business strategy. For example, Australian legal advisers advertise their services to clients seeking to use technicalities to maximise their extensions of patent term in the United States (FlatteryO’Brien and Zammit 2016). In many respects, the current system for calculating EoT is a compromise that seeks to avoid the significant administrative costs associated with predecessor arrangements. Previously, EoTs were determined based on a case-by-case assessment of the sufficiency of the incentive to invest afforded by the standard patent and any extension. Australia is not alone in providing EoTs or in calculating ‘delay’ and periods of extension based on loose proxies. The PPR considered the EoT process in three other jurisdictions — the United States, the United Kingdom and Japan (figure 10.3). The Commission has expanded on this comparison by including the arrangements in Singapore (box 10.3). PHARMACEUTICALS — GETTING THE RIGHT POLICY PRESCRIPTION 299 Figure 10.3 Disparate ways of calculating EoTs by reference to the life of a pharmaceuticala,b a The figure refers to clinical trials commenced after the grant of the patent. In the US and Japan, EoTs are calculated with reference to the first clinical trials or the date of the patent grant, whichever is later. b Each letter refers to a period of time — for example ‘A’ refers to the period between patent application and grant. Refer to figure 10.1 for indicative timelines for each period. Source: Harris, Nicol and Gruen (2013). The PPR went on to compare effective patent lives and extension lengths in cases where extensions had been granted in Australia, the United Kingdom and the United States.11 In terms of effective market life, the median in Australia was the same as that in the United Kingdom, but was 12 months longer than the median life in the United States (where the maximum life is constrained at 14 years). 12 Further, the analysis showed that the median length of extensions in Australia was the same as the United Kingdom, but 8 months longer than the median in the United States (Harris, Nicol and Gruen 2013, pp. 212–215). A new approach is emerging for determining extensions of term Not all jurisdictions apply loose proxies when determining EoTs. Both Brunei and Singapore (box 10.3), only grant pharmaceutical term extensions where there has been delay due to the actions of the regulator. 11 Note that Canada was not included in the comparison as it previously did not have any EoT scheme, but under requirements from both the Canada – EU Trade Agreement and the TPP, it will introduce an extensions scheme (Geist 2015). 12 The PPR (Harris, Nicol and Gruen 2013) noted that, in the 47 per cent of cases where the 5 year limit of duration on extension is reached, the effective life in Australia is typically shorter than that in the US and the UK. 300 INTELLECTUAL PROPERTY ARRANGEMENTS Box 10.3 Singapore: strict extensions but a flourishing industry Extension of term (EoT) for pharmaceuticals are available under the Singapore Patents Act for a period of up to 5 years. These extensions are subject to several conditions that restrict their availability to certain products. Specifically, EoTs are only available where: • the substance is an active ingredient of a pharmaceutical product, and it is the first product to use the active ingredient to obtain marketing approval • the patent had not previously been granted an EoT • there was ‘unreasonable curtailment’ of the patent life during the marketing approval process — that is, the time between filing for and obtaining marketing approval from the Singaporean Health Services Authority was more than 2 years. If these conditions are all met, an extension will be granted for a period that is, in effect, the time taken to obtain marketing approval (less delays attributable to the applicant) that exceeds 2 years (with a maximum of 5 years). The strict requirements for accessing EoTs in Singapore mean that while they are available, they are far from common: In practice, MA [marketing approval] is normally based upon approval in the US or Europe, so will almost always be obtained within two years, meaning that it is unlikely that a [Patent Term Extension] will be granted in most cases. In fact, we understand that only one patent has been successfully extended under the PTE provisions in Singapore, and this required the innovator to seek its first marketing authorisation through Singapore’s Health Sciences Authority. (Kinnaird 2015) Some participants in this inquiry have argued that EoTs are important to securing outcomes for the pharmaceutical industry and healthcare more generally. However, Singapore’s less generous regime, of itself, does not seem to have harmed the industry nor healthcare outcomes. In terms of healthcare, the country’s system was ranked sixth best by the WHO (nd), and the pharmaceutical market is estimated to be worth US$1.1 billion in 2013, and is expected to reach US$1.7 billion in 2018 (Deloitte 2014). Singapore is also seen as a regional base for production and research — in 2013, international biomedical firms located in Singapore manufactured products worth US$25 billion and spent approximately US$500 million on research and development (Gross 2014). A component of the success of the pharmaceutical industry in Singapore can be attributed to the government’s decision to fund research institutes (through the Agency for Science, Technology and Research) and encourage public-private partnerships. However, the broader environment for commerce and investment also provides reasons for pharmaceutical companies to locate in, or enter, the market there: Singapore also brings together a well-developed infrastructure and logistics network, strong intellectual property (IP) laws, a record of safety, a good regulatory environment and active government support of the biomedical industry … [An Amgen executive commented that] Singapore’s rich talent pool and friendly business environment made it an ideal place to invest in a world-class manufacturing facility … (Gross 2014) Source: Jones and Lucas (2015). PHARMACEUTICALS — GETTING THE RIGHT POLICY PRESCRIPTION 301 Chile (another party to the TPP) awards ‘supplementary protection’ (extensions) that is calculated by reference to unjustified administrative delay, where such delays are greater than one year (Castro 2016). Israel and South Korea operate similar EoT systems. 13,14 New Zealand is in the process of implementing an EoT scheme following its accession to the TPP. 15 The New Zealand government has indicated it will focus on ‘unreasonable curtailment’ and proposes to grant extensions only where delay exceeds a benchmarked marketing approval period. The benchmark will differ for small molecule pharmaceuticals and biologics, but in both cases extensions will be capped at two years, and calculated as the shortest of two years, the marketing approval process, or the time between the grant of the patent and of marketing approval (MBIE 2016, p. 24). The New Zealand Government’s expectation is that very few unreasonable delays would occur (NZMFAT 2015). As such, amongst several of Australia’s major trading partners subject to comparable trade agreement obligations, most have introduced an EoT scheme based on unreasonable regulatory delay (or are in the process of doing so). Of these, most focus purely on the time taken in the marketing approval process. In the Commission’s view, Australia’s EoT scheme is a comparatively unfocused and overly generous approach. What should be included as regulatory delay? The Commission considers that, for the purposes of being eligible for an EoT, delay should be measured solely based on the actions of the regulator that are outside the control of the applicant. Specifically, delay should be measured with reference to the actions of the regulator once the marketing approval process has begun. Hence, further testing at the behest of the regulator during the marketing approval process is a particular burden for pharmaceutical companies and should be included. As representatives from Medicines Australia and IFPMA noted: … you cannot predict when you will get marketing approval from the regulator. They want to test your hypothesis, they may need you to conduct additional clinical trials in different patient 13 Israel calculates its extensions based on the duration of the marketing approval process, or for ‘related patents’ (those extended in the US and parts of Europe), the shortest extension period granted by another jurisdiction (The Law Library of Congress 2016). 14 In South Korea, the duration of an extension is based on the regulatory delay and the time taken for clinical trials that are performed in South Korea (Shin 2014). 15 New Zealand had patent term extensions until, as part of accession to TRIPS, it extended patent terms from 16 to 20 years. At the same time the New Zealand Government abolished extensions noting that the costs would likely exceed the benefits as any extension in New Zealand would be unlikely to lead to the creation of a new drug (Nowak and Doucas 2015). The New Zealand Government reconsidered, and again rejected, the case for EoTs in 2003 on the basis of the economic impact on consumers (Tansey and Dixon 2015). 302 INTELLECTUAL PROPERTY ARRANGEMENTS populations … . But it’s that lack of control and the significant period of time that makes agrichemicals and pharmaceuticals unique in that regard. (trans., p. 335) In contrast, general (pre-approval) clinical trial processes should not be included because the pace at which these occur is largely in the hands of pharmaceutical firms. Further, in a global market, the extent of any pre-approval trials that are required solely for the Australian market are likely to be limited. Medicines Australia submitted that in addition to the TGA’s process, the PBS processes should also be considered as a component of delay: … Australia operates in a publicly supported, universal health care system whereby medicines are additionally assessed for cost effectiveness before they are subsidised on the Pharmaceutical Benefits Scheme (PBS). Medicines are not widely available to patients until they are listed on the PBS. The evaluation and assessment process for listing on the PBS is complex and costly and frequently delays medicines’ availability by another year and sometimes several years. (sub. DR529, p. 12) Prima facie, such arguments appear to have merit — the PBS and its processes (box 10.4) can seem monolithic from the outside. However, there are several flaws in this argument. First, while the existence of subsidies makes listing on the PBS an attractive option, it is not a regulatory requirement in order to sell a pharmaceutical. The only legal requirement is listing on the ARTG. 16 More importantly, including PBS processes might give rise to a ‘moral hazard’. That is, if extensions are linked to delays in the PBS process, they would in effect act as a form of insurance for the sponsor companies ultimately engaged in a negotiation process with the government. This, in turn, would reduce any incentive for the company to do its part in reaching a speedy conclusion in pricing negotiations. 16 In relation to Australia’s international obligations, the term ‘marketing approval’ is taken by member states to refer to the approval process of bodies analogous to the TGA that allow a product to be sold at all, not subsidy schemes that determine the market price. PHARMACEUTICALS — GETTING THE RIGHT POLICY PRESCRIPTION 303 Box 10.4 The Pharmaceutical Benefits Scheme’s process The submission stage To list a drug on the Pharmaceutical Benefits Scheme (PBS), applicant companies (‘sponsors’) must first provide a submission to the independent Pharmaceutical Benefits Advisory Committee (PBAC). Major submissions can be accepted up to 17 weeks before the relevant PBAC meeting (11 weeks for minor submissions). During this time, the submission is extensively evaluated to examine patterns of likely use for the medicine and its cost-effectiveness (relative to current treatments). To enable this evaluation, sponsors must provide detailed information in support of the submission's claims, such as on the recommended treatment, data on comparative effectiveness, cost per patient, and the requested price of the medicine. PBAC then issues a recommendation to list the drug (‘positive’) or not. Following a positive recommendation Following a positive recommendation, sponsors must negotiate final arrangements with the Department of Health. These include pricing and any applicable prescribing restrictions (limits on the manner in which the medicine will be subsidised, such as for a particular type of treatment), as well as risk sharing agreements (for example, to reduce the price of a medicine once its use reaches a limit). These negotiations can be complex, and the time taken to complete them varies. Currently, the usual minimum for completion is around five months. The recommendation must then be approved by Government. For Medicines that are expected to cost less than $20 million a year, the approval can be granted by the Minister for Health. Medicines that cost more than $20 million in any one year of the Forward Estimates must be put to Cabinet for approval. Following approval by Government, six weeks are required to finalise the legal requirements, update information systems and implement any data requirements. While the timing varies, the average time to listing following a positive PBAC recommendation for cancer medicines is 6.1 months (7.3 months for major submissions, 4.9 for minor). Following a decision not to recommend After a deferral or decision not to recommend, it is up to the sponsor to resubmit with additional evidence or altered variables.17 They may then enter the next available 17-week cycle. Other mechanisms to expedite listing Reforms to PBS processes have sought to reduce the time to listing. From 1 January 2011, the Managed Entry Scheme was introduced which is a mechanism whereby the PBAC may recommend PBS coverage at a price justified by the existing evidence, pending submission of more conclusive evidence of cost-effectiveness to support listing of the drug at a higher price. Under the TGA-PBAC parallel processes, also introduced in 2011, a submission to the PBAC may be lodged at any time after the date of lodgement of a TGA registration dossier. Sources: Department of Health (2015b, 2016b). 17 Resubmissions ask PBAC to reconsider matters from relevant previous submissions. Even if such a submission is based entirely on new data, modifies the previously requested restriction or changes the comparator, it will be regarded as a resubmission. This is because the information in the resubmission will have to provide the basis for any change to PBAC’s earlier decision. 304 INTELLECTUAL PROPERTY ARRANGEMENTS While it might be argued that a moral hazard concern relates only to the post-recommendation negotiation phase, there are several elements within the control of companies that influence the timing of the listing process: • First, the company chooses when to submit to the PBS following listing on the ARTG. An analysis of new chemical entities and products for new indications registered by the TGA in 2004 found an average of 17 months (median 9 months) between TGA approval and first review by the Pharmaceutical Benefits Advisory Committee (PBAC) (Pearce et al. 2012). 18 • Second, sponsor companies can opt to use expedited regulatory approval and listing processes. In practice, take up of these options has been limited — for example, only 20 per cent of major submissions for cancer treatments have used parallel processing with the TGA and the PBS (Department of Health 2015b). • Third, in the course of the submission process, companies must provide an initial request for price. This is used by PBAC as an input to its cost-effectiveness analysis. Requesting a high price can, in effect, set a higher hurdle for the drug to obtain approval (but improvements in clinical effectiveness may still allow a costly drug to be recommended). As such, requesting too high a price can result in a negative recommendation from PBAC and a lengthy negotiation process can ensue. • Fourth, those that unsuccessfully sought a high price but wish to continue with PBS listing can choose to resubmit their application. The choice of when to do so is up to the sponsor. For some, the resubmission is immediate (within the next 17-week cycle), but this is not always the case. In many cases, a company’s decision to postpone resubmission is the cause of delays in achieving a positive recommendation. For instance, the Department of Health (2015b) noted examples of 6 cancer medicines that took an extended period to list on the PBS, primarily as a result of the sponsor deciding not to resubmit for a period of time (4 cycles, or over a year, in one instance). While companies have little, if any, scope to exert control over the final 6 week window prior to PBS listing, the time is used for setting in place legal agreements and arrangements for data and payment systems. It also allows for due diligence processes to be undertaken prior to the expenditure of significant amounts of taxpayer funding. In the Commission’s view, these periods do not constitute ‘unreasonable’ delay. Given the considerable scope for pharmaceutical companies to influence the timing of PBS listing processes, the Commission does not consider that any delays in the PBS process should form part of the calculation of EoTs. The sole focus in determining delay should be the TGA’s marketing approval process. 18 The data in this analysis covers PBAC meetings held between March 2004 and August 2010. Several procedural reforms have been instituted since that time (including parallel processing for TGA and PBAC submissions) that would likely reduce the overall regulatory approval period. PHARMACEUTICALS — GETTING THE RIGHT POLICY PRESCRIPTION 305 What length of delay is unreasonable? The second relevant policy consideration is what length of delay is unreasonable. The use of the term ‘unreasonable’ implies some measure of delay is reasonable. There are (at least) two options for determining what that benchmark might be. The first is the timeframes set for the TGA by the Australian government. Reflecting its own interests in seeing safe and efficacious drugs coming to market in a timely manner, the Australian government sets legislative timeframes of 255 working days. These timeframes do not count the time allocated to applicants to provide responses to information requests and ‘clock stop’ periods agreed with the applicant. The second option is to have regard to international norms for approval times. In practice, the two differ little. For the period from January to June 2015, the TGA’s approval time for new chemical entities ranged from 88 and 251 working days. 19 Excluding expedited approval pathways (as they are not offered in Australia), the TGA’s approval timeframes are on par with some of the larger jurisdictions — they are considerably faster than those in Europe, and only slightly lag behind those in the United States and Japan (Sansom, Delaat and Horvath 2015). A number of recommendations have been made in a recent review that, if adopted, would likely see Australian approval times fall — bringing them further in line with international norms (Sansom, Delaat and Horvath 2015). The Commission considers that a regulatory approval period of one year (255 working days) is well within the bounds of a ‘reasonable’ delay. Unreasonable delay (and any compensating EoT) should be limited to the time taken in excess of this, to a maximum of five years. In keeping with the TGA’s current time recording practises, the time should only be calculated on delays attributable to the TGA’s actions, not those of the applicants (for example where the applicant delays responding to TGA requests). The Commission estimates that, if adopted, this recommendation would significantly reduce the number of EoTs granted, leading to savings to government (and taxpayers) through reduced payments on the PBS of $258 million per year (appendix H). This compares to a saving of $260 million if there were no EoT system. Improved access to cheaper medicines can also have broader public health benefits, with high health costs disproportionately impacting on disadvantaged groups (Gleeson, Lopert and Reid 2013). 20 19 There are approximately 250 working days in a year. 20 Despite the presence of the PBS, affordability of medicines is still an issue for some parts of the Australian community — in 2014 15, 7.6 per cent of respondents to a national survey delayed or did not purchase prescribed medicines due to cost in the previous year (SCRGSP 2016, p. 10.27). 306 INTELLECTUAL PROPERTY ARRANGEMENTS Which products should EoTs apply to? A further policy consideration is whether EoTs should only apply to select products. Ideally, EoTs would apply to those drugs where the standard patent has not provided a pharmaceutical company with sufficient opportunity to recoup their investment. This depends on the costs of research and development, and the returns the pharmaceutical company is able to appropriate due to a period of market exclusivity. Allowing EoTs on a drug-by-drug cost basis would make the system more adaptable. However, as the failings of the previous EoT scheme highlight, utilising a case-by-case approach can be cumbersome and expensive. A simpler approach could be to use easily identifiable proxies, such as whether the patent is over an active pharmaceutical ingredient (API). New APIs are generally the most expensive form of drug to develop. They generally involve a higher risk and intensive development process than is required for follow-on products (section 10.5). New APIs also tend to be associated with step changes in innovation, rather than incremental improvements in the effectiveness of existing treatments. Restricting EoTs to new APIs would realign the scheme with its original objectives. While the inclusion of per se in s. 70(2) of the Patents Act 1990 (Cth) was originally intended to limit EoTs to new APIs, the boundaries of the definition have become blurred by developments in case law. The 2013 judgment in Spirit v Mundipharma 21 held that OxyContin, a controlled release formulation of the opioid oxycodone (which itself was first patented in Germany in 1916) was a different pharmaceutical substance to oxycodone itself, and that it was a pharmaceutical substance per se within the meaning of s. 70(2). This example highlights the potential for future cases to further expand the definition, allowing EoTs for progressively smaller advances. With this in mind, the Commission considers there is value in realigning the definition to restrict EoTs to APIs (as is the approach in Singapore).22 Assessing the ongoing necessity of extensions of term More systematic collection and analysis of data on EoTs and pharmaceutical R&D in Australia would aid policy makers in monitoring whether there is an ongoing case for the EoT scheme. The scope to collect data specific to EoTs already exists. At the time of the introduction of the current scheme, provision was made to collect data with each application for an extension. This was intended to assist in evaluating whether EoTs were achieving their stated objective, namely encouraging pharmaceutical R&D in Australia. Under s. 76A of 21 Spirit Pharmaceuticals Pty Ltd v Mundipharma Pty Ltd [2013] FCA 658. 22 In addition to changes to s. 70(2) to achieve this, there may be merit in adding per se (that is, limited to APIs) to the definition of ‘pharmaceutical substance’ in schedule 1 of the Patents Act 1990 (Cth). PHARMACEUTICALS — GETTING THE RIGHT POLICY PRESCRIPTION 307 the Patents Act 1990 (Cth) patent holders are required to lodge a return, regarding the drug whose patent is to be extended, with the Department of Health setting out: • the amount and source of Commonwealth funds spent on R&D for the drug • the name of any party which the applicant has contracted with and has received Commonwealth funds • the total amount spent on each type of R&D (including pre-clinical research and clinical trials) for the drug. The Department of Health (pers. comm., 10 March 2016) provided the Commission with a summary of the data collected from 2009-10 to 2013-14 (box 10.5). Box 10.5 Pharmaceutical R&D Data from EoT applications Currently, pharmaceutical companies provide data under s. 76A of the Patents Act 1990 (Cth) to the Department of Health on a commercial-in-confidence basis. To assist the Commission’s inquiry, the Department of Health provided the Commission with summary data drawn from the returns from 1 July 2009 to 30 June 2014. Generally, the data reveal little reliable or policy-relevant information. One reason for this is a general lack of compliance — of the 166 applications for extension of term of pharmaceutical patents in the period, 74 (45 per cent) did not submit a return under s. 76A. The 92 returns that were provided represented 57 separate pharmaceuticals. The bulk of those either reported that they did not spend any Commonwealth funds on R&D (91 per cent) or provided no information (2 per cent). Of the remaining 7 per cent who did report that they spent Commonwealth funds on R&D, the average (mean) expenditure was $2.3 million. In terms of overall spending on Australian R&D, 53 per cent of returns reported no R&D expenditure and 8 per cent provided no information. Of the remaining (39 per cent) who did report R&D expenditure within Australia, the average expenditure was $3.9 million, though there was considerable variability with returns ranging from $0.1 to $38.8 million. In line with the requirements of s. 76A, the data also records whether applicants had contractual agreements with third parties who were in receipt of Commonwealth funds. The bulk (79 per cent) of applicants reported that they had no contractual agreements. Of the remainder, 8 per cent provided no information and 13 per cent reported contractual agreements. Source: Department of Health (pers. comm., 10 March 2016). Implementation of these data requirements was lacking. As IPTA (2015, pp. 9–10) noted, the current drafting of s. 76A created uncertainty as to: the time period in which funds are spent; the drug in question; 23 the definition of Commonwealth funds; the types of R&D and which ones must be specified. IPTA advocated that the section be repealed as it introduced substantial compliance costs for little or no benefit. The PPR reached a similar conclusion, but also suggested that 23 Pharmaceutical patent extensions are related to a particular patent, and not tied to a particular drug (as noted above one drug can be associated with multiple patents). 308 INTELLECTUAL PROPERTY ARRANGEMENTS consideration be given to introducing a mechanism for reporting based on the system in Canada (Harris, Nicol and Gruen 2013). The Canadian approach requires pharmaceutical patentees to submit data to the Patented Medicine Prices Review Board (2015). An important feature of the Canadian system is it collects data on a patentee basis rather than on a ‘per drug’ basis as s. 76A currently requires. The Canadian system provides for comprehensive data on sales revenue and total R&D expenditure by firm, split by: type of research (chemical or biological); pre-clinical and clinical trial (multiple phases) stages; who conducted the research (company itself, outsourced, hospital, university); and source of funds (including government funds). The data collection system in Canada is also bolstered by clear compliance mechanisms, and a detailed ‘patentees guide to reporting’(PMPRB 2016). The guidelines include definitions of each data category. For example, the definition of R&D is linked to the definition of ‘scientific research and experimental development’ required to claim an investment tax credit under the Income Tax Act 1985 (Canada). Tying definitions to other government programs in this manner makes it easier for firms to comply. IP Australia suggested that, instead of s. 76A, it may be possible to rely on other, existing data collections by leveraging: … new and available data, including that collected by the Australian Bureau of Statistics, the Australian Institute of Health and Welfare, the Therapeutic Goods Administration and the Pharmaceutical Benefits Scheme. (sub. DR612, pp. 14–5) Better linkage of data is generally worth exploring (and is one element of the Commission’s current inquiry into Data Availability and Use). However, for the specific purpose of evaluating the effects of EoT, the Commission considers the broader data sets do not capture sufficient detail, and any changes to existing data would require consideration of broader issues. The Commission considers that data collection under s. 76A should be reformed, not abandoned. The template provided by the Canadian system shows that it is possible to collect standardised and worthwhile data. Although the Commission does not propose to dictate the exact format of data collection, there are some core requirements drawn from the contrasting experiences in Australia and Canada: • Data should be collected at the company (or business) level. • Definitions should be tied to other existing pieces of legislation — for example, in Australia R&D could be defined by reference to the term ‘core and supporting R&D activities’ as used in s. 355.25 of the Income Tax Assessment Act 1997 (Cth). • Data should be linked with existing collections where possible to minimise unnecessary duplication in the collection of information. PHARMACEUTICALS — GETTING THE RIGHT POLICY PRESCRIPTION 309 • There should be consultation in the development of the data requirements, including production of detailed guidelines and the development of a standardised pro forma to facilitate compliance. • Compliance in providing data should be a pre-requisite for any grants of an EoT. Some expressed concern about adopting the Canadian system. For example, Medicines Australia (sub. DR529) argued the changes were unnecessary and that the rationale for improvements was not ‘adequately supported’. However, the paucity of evidence to validate the purported policy objective at the time EoT were introduced underscores the importance of a transparent evidence base. Data are equally important when analysing and reviewing policies. In the absence of such evidence, there is a risk that assertions from vested interests may overwhelm genuine public interest in the policy formulation process. Accordingly, the Commission considers that improved data collection requirements should be put in place in concert with its proposed reforms to the EoT scheme. After the data have been collected for 5 years, they should be used to evaluate the effect of the Commission’s reforms, and to inform subsequent reviews into the ongoing necessity for EoTs. Under the auspices of the standing IP Policy Group (chapter 17), this review should also consider any further changes that are necessary to improve or expand the data collection system. RECOMMENDATION 10.1 The Australian Government should reform pharmaceuticals such that they are only: extensions of patent term for (i) available for patents covering an active pharmaceutical ingredient, and (ii) calculated based on the time taken by the Therapeutic Goods Administration for regulatory approval over and above 255 working days (one year). The Australian Government should reform s. 76A of the Patents Act 1990 (Cth) to improve data collection requirements for extensions of term, drawing on the model applied in Canada. Thereafter no extensions of term should be granted until data is received in a satisfactory form. 10.3 Manufacture for export At present, if a drug is under patent in Australia, Australian generic producers are unable to manufacture and export to countries where patents have expired (or indeed have never been sought). This reflects the prevailing interpretation of provisions in TRIPS (and other agreements) that patent holders enjoy exclusive rights to make, use, import and export the product. As discussed above, Australian patents often expire later than elsewhere (particularly where EoTs have been granted). This impedes the growth of generic 310 INTELLECTUAL PROPERTY ARRANGEMENTS manufacturers, as their competitiveness depends on reaching markets as soon as possible after the original patent expires. There are no contemporary estimates of the costs of precluding manufacture for export (MFE). At the time of a previous Commission study of pharmaceuticals (2003), the then Department of Industry, Tourism and Resources estimated that, over the period 2001 to 2009, export revenue of $2.2 billion could be lost if the existing system were maintained. Noting these losses, the Commission recommended in 2003 that MFE be allowed in the patent extension period. Since then, changes to the Patents Act 1990 (Cth) were made in 2015 to allow the export of patented drugs in limited circumstances in order to address public health problems in least developed countries in need. 24 Notwithstanding these changes, the Commission considers there are grounds for reforming arrangements for MFE, especially during the patent extension period. MFE would benefit domestic generic manufacturers (who might otherwise relocate their operations). As MFE only rules out that the product is manufactured in Australia, it is likely that a given overseas market will be supplied with a generic from another country. As such, allowing MFE would have little to no effect on Australian originators. The preferable reform option is to rely on sui generis rights, which would not be covered by TRIPS. This approach is in line with that taken in other jurisdictions and allows for MFE to other countries where patents are not in force. Indeed, the European Commission (EC) is reportedly considering a Supplementary Protection Certificate (SPC) manufacturing waiver to allow export to non-EU countries (IP Australia, sub. 23, p. 11). Using sui generis rights allows consideration of which subset of rights should be extended to pharmaceutical companies, providing additional flexibility. Further, by disconnecting the extended period from the standard patent system, domestic policy flexibility and adaptability are preserved into the future. The scope to use sui generis protection is clouded by ambiguous drafting within trade agreements, including AUSFTA and the TPP. Article 18.48 of the TPP states that parties must make available an ‘adjustment of the patent term’ as compensation for unreasonable curtailment, replicating text from AUSFTA. However, footnote 46 of the TPP states: For greater certainty, a Party may alternatively make available a period of additional sui generis protection to compensate for unreasonable curtailment of the effective patent term as a result of the marketing approval process. The sui generis protection shall confer the rights conferred by the patent, subject to any conditions and limitations pursuant to paragraph 3. 24 Compulsory licences for patented pharmaceutical inventions were implemented as part of a commitment to implement amendments to the TRIPS Protocol. They are available by application to the Federal Court, which must verify they are necessary to address a public health problem in a least developed country, made in good faith, cannot be diverted away from their purpose, and the patentee has had an opportunity to voluntarily authorise the use (IP Australia 2015c). PHARMACEUTICALS — GETTING THE RIGHT POLICY PRESCRIPTION 311 Other parties to the TPP consider that a sui generis approach and allowing MFE during the sui generis period would be consistent with obligations under the TPP. The Canadian Government has announced its intention to use a sui generis system with a maximum two years’ of protection when it introduces EoTs, and that it believes this is compliant with the TPP, and Canada’s agreement with the European Union (Government of Canada 2015). However, the TPP has yet to be signed. As such, the obligations in AUSFTA apply. A side letter (Vaile 2004) to that agreement appears to limit 25 the possibility of manufacture for export by stating that: … Australia may permit the export by a third party of a pharmaceutical product covered by that patent, only for the purposes of meeting the marketing approval requirements of Australia or another territory. (p. 1) Given the presence of this apparent constraint, in the absence of a later plurilateral agreement involving the United States superseding AUSFTA, 26 any attempts to allow manufacture for export could be subject to challenge. Specifically renegotiating AUSFTA to allow MFE (rather than relying on plurilateral agreements) may prove costly on other fronts. The Commission’s preferred approach is to rely on a more targeted EoT system to secure the majority of benefits. In the longer term, Australia’s negotiating approach to future agreements should include allowance for MFE during the patent EoT (recommendation 18.2). 10.4 Data protection A condition for registering pharmaceutical products is that sponsors are required to submit data relating to drugs’ quality, safety and efficacy (‘test data’). Whether and when that data should be able to be used for subsequent registration of products similar to those originally registered remains an ongoing issue. Under TRIPS, Australia is obliged to protect undisclosed test data on new chemical entities in two ways. First, the data must be protected against unfair commercial use. TRIPS does not define unfair commercial use, but under AUSFTA the regulator cannot rely on that data to approve a follow-on or generic product for a period of five years from approval of the original product. Second, Member States must protect the data against disclosure, except where necessary to protect the public or where steps are taken to ensure the data is protected against unfair commercial use. 25 Alternative constructions of the clause argue that the side letter’s restrictions are limited only to cases of springboarding (using a patented product to obtain regulatory approval of a generic). Under this interpretation, the side letter has no effect on more general manufacture for export during the extension period. 26 A side letter to the TPP suggests that agreement would in fact supersede the relevant AUSTFA side letter. 312 INTELLECTUAL PROPERTY ARRANGEMENTS Views about the desirability of data protection provisions are mixed. Some health specialists have argued against any concealment of data submitted for the approval of pharmaceuticals (Ollila and Hemminki 1996, p. 169). In their view, non-disclosure contradicts the right of the public to be informed about the efficacy and safety of approved pharmaceuticals. Other experts emphasise that health authorities should be able to use and rely on registration data submitted for similar products, or on the existence of a prior registration elsewhere. To do otherwise would require repetitive toxicological and clinical investigation (Dukes 1996, p. 146). There are also ethical concerns about repeating trials, which include untreated control groups, with a drug known to be efficacious. Experts also point to the consequences for competition of providing data protection. While data protection usually runs concurrently with patent protection, it can have different start and expiry dates. In most cases, patent life exceeds the period of data protection. But in those cases where it does not, data protection can become a means to block the timely entrance of generic competitors to off-patent drugs. This is because the cost of replicating clinical trials is likely to be sufficiently prohibitive to deter many competitors. In contrast, the pharmaceutical sector argues that the manufacturer has invested, often heavily, in conducting tests and deserves a return on that investment. They further argue that if governments do not protect data they risk foregoing access to pharmaceutical products: … equity demands that protection be provided for data, which can cost the original submitter several million dollars to produce. Disclosing this data to the public or allowing its use by another applicant unfairly denies the compiler of the data the value of its efforts and grants an economic advantage to later applicants for marketing approval, enabling them to avoid the cost of developing test data for their own products. Countries that allow such unfair advantages to later applicants discourage developers of new pharmaceuticals and agricultural chemicals from seeking to introduce their state-of-the-art products in the country’s market. (Priapanjta 2000, p. 4). Similarly, Amgen Australia submitted: Without the assurance provided by data protection that others may not free-ride (for a set period of time) on the clinical trials conducted, it is unlikely anyone would invest millions of dollars and many years to conduct the clinical trials that are necessary to prove a product is safe and effective. (sub. DR337, pp. 3–4) International use of data protection There is no international standard for the period of data protection. The period is longer in countries that are large net exporters of brand name pharmaceuticals such as the United States, the European Union and Japan. At the other extreme, some developing countries provide no period of exclusivity. The application of data protection and the conditions PHARMACEUTICALS — GETTING THE RIGHT POLICY PRESCRIPTION 313 under which it is granted also vary between countries, such that international comparisons are not straight forward (table 10.1). Table 10.1 Data protection periods Country New chemical entity Biologic Australia 5 5 New Zealand, Singapore, Chilea 5 5 Mexico, Peru 5 0 Israelb 6 0 Chinac 6 6 6+2 6+2 8 8 5 4+8 8+2 8+2 Canadad Japane United Statesf European Uniong a Chile provides five years data protection if the application for approval is filed within a year of the drug being approved in another country. b Israel provides 6 years of marketing exclusivity from the date of approval in Israel or 6.5 years from approval in another country for new chemical entities. c China provides 6 years marketing exclusivity where a new chemical ingredient is first approved in China. US industry has reported that drugs first approved outside China are not given protection. d Canada provides 6 years data protection plus 2 years marketing exclusivity. e Japan provides 8 years of exclusivity through a period of post-marketing examination to ensure efficacy and safety of the new drug. f The US provides 4 years data protection plus 8 years market exclusivity for biologics. g The EU provide 8 years data protection plus 2 years marketing exclusivity, plus an additional year for new indications. Sources: International Federation of Pharmaceutical Manufacturers Pharmaceutical Research and Manufacturers of America (2013). & Associations (2007); In many jurisdictions data protection is seen as non-binding due to longer patent terms. However, in the United States, data protection has been used as a form of industry policy. Data protection, or ‘regulatory exclusivities’ (a broader term used to refer to both data protection and marketing exclusivity), have been tailored to specific industries, products or policy goals, with different protections provided for new chemical entities, new clinical studies, biologics and for orphan drugs (drugs for rare conditions where the market reward is unlikely to justify the research cost), among other things. Such tailoring leads to complicated systems where the cumulative protection can approach, or even exceed, the effective market life afforded under a patent (Thomas 2013, p. 11). 27 The potential future evolution of regulatory exclusivities also poses significant policy issues. In the United States, the proposed Modernizing Our Drug & Diagnostics Evaluation and Regulatory Network Cures Act (MODDERN Cures Act) includes provision for 27 For example, a new chemical entity (five years protection) that was designated as qualified infectious disease product (additional five year protection) and which had studies of the effects on children (pediatric exclusivity, additional six months) could qualify for protection of ten years and six months. 314 INTELLECTUAL PROPERTY ARRANGEMENTS regulatory exclusivities that exceed the effective market life allowed under the United States’ pharmaceutical EoT system. Data protection should not be seen as a substitute for patents Some pharmaceutical companies simply prefer the automatic protection afforded by data protection over the expense of obtaining patents. IPTA stated that pharmaceutical companies rely on data protection when a patent application for the active agent has not been filed in Australia (sub. 73). As Eli Lilly submitted to a Senate committee: … there are more than 100 intellectual property regimes in the world, for smaller organisations developing medicines (eg biotechnology start-ups) – priority for patenting is given to larger markets – Australia is approximately 1% of the world market … (2014, p. 4) Others claimed that patents alone do not provide sufficient protection, and that data exclusivity is necessary to allow manufacturers the opportunity to recover costs associated with introducing a new drug onto the market (Amgen, sub. DR337; AusBiotech, subs. 37 and DR419; Medicines Australia, sub. 44; Pfizer Australia, sub. 83). Concerns about the sufficiency of patent protection were particularly pronounced for biologics (box 10.6), which now represent one of the main cost drivers of pharmaceutical expenditure around the world. Industry argues that biologics might not always be patentable or that patents may not be enforceable. Such concerns about the patentability of biologics arise because it is the method of manufacture that is the focus of IP protection rather than the molecule itself (AusBiotech, sub. 37). For example, Pfizer submitted: Due to the evolving nature of patent law surrounding biologic medicines, they are at a greater risk of imitation. Thus, data protection provides an important incentive for continued R&D. For example, a biosimilar may be analogous enough to rely on originator data (after the mandated period of data exclusivity expires) but different enough to not infringe a patent, leading to patent workarounds. In this case, data exclusivity would be necessary to protect intellectual property and the substantial investments of the company. (sub. 83, p. 5) PHARMACEUTICALS — GETTING THE RIGHT POLICY PRESCRIPTION 315 Box 10.6 What are biologics? Biological medicines (or biologics) are a type of medicine containing an active substance derived from living organisms, such as bacteria or cells. These active substances are typically larger and more complex than those of non-biological (small-molecule) medicines. While biologics are generally injected or infused, small-molecule medicines are typically administered in pill or capsule form. Examples of biologics include treatments for rheumatoid arthritis (Humira, Enbrel), diabetes (Lantus, a form of insulin), eye diseases (Lucentis) and osteoporosis and cancer (Prolia). Unlike chemical-based drugs that can be readily reverse-engineered, biologics enjoy a degree of natural protection — manufactured from living cells through biological processes, they cannot be exactly copied. As the name suggests, biosimilars are similar versions (rather than exact copies) of the original biologic. Some examples of biosimilars include treatments for cancer (Filgrastim) and rheumatoid arthritis and Crohn disease (Infliximab). The costs to develop and manufacture biosimilars are also high due to inherent complexities and competition from biosimilars is estimated to only reduce the market price of a drug by 20-25 per cent (compared with up to 70 per cent for generic chemical drugs). The market for biologics is also different to small-molecule drugs, as biologics are often targeted at specific conditions with smaller patient cohorts. Companies appear to be moving from the development of drugs targeted at larger populations to more specialised medicines that can command higher prices (Thomas 2014). However, some biologics, such as vaccines and treatment of conditions such as diabetes, are targeted more broadly. For example, the Human Papillomavirus (HPV) vaccine Gardasil is provided free in Australian schools to all males and females aged 12-13 years under the National HPV Vaccination Program. Sources: AusBiotech (2016); Department of Health (2015a); Hospira (2014). Similar arguments have been advanced in the academic literature (Manheim, Granahan and Dow 2006). The economist Henry Grabowski (2009) estimated that the breakeven lifetime for the mean biologic product was between 12.9 and 16.2 years, and thus data protection 316 INTELLECTUAL PROPERTY ARRANGEMENTS provides an ‘insurance policy’ to stimulate innovation in cases in which effective patent protection is limited. 28 Despite arguments of inadequate patent protection having been advanced for a decade, there is little evidence that a problem has manifested in practice. Medicines Australia (subs. 44 and DR529) submitted that it was aware of some instances in which biological medicines were not brought to the Australian market due to Australia’s short data protection period and pointed to the Lucrin example discussed in the PPR. However, there was no evidence presented to the Commission of systemic problems arising from Australia’s current data protection period. Extending protection to a broad class of products to address isolated cases is likely to be inefficient and overcompensate the majority of products. Not only is there a lack of evidence that patents are not doing the job, using data protection has drawbacks. While data protection has some characteristics that make it attractive to originator pharmaceutical manufacturers (for example, its non-contestable nature avoids expensive litigation) it lacks some of the features of patents designed to promote innovation. As noted by the Department of Health: In contrast to patents, data protection is an automatic right (i.e. no application is required or assessed), nor is the protection reviewable or contestable via administrative or judicial processes. In addition, whereas the grant of a patent requires full disclosure of the invention (as a measure to balance the monopoly rights against society’s desire to promote follow-on innovation), data protection requires that protected information be kept confidential for the duration of the protected period. (sub. 84, p. 7) More broadly, the policy rationale for data protection differs from that for patents. Data protection is intended to protect the investment in the test data that is required to demonstrate that pharmaceuticals are safe and effective. Arguments that data protection should serve as an ‘insurance policy’ where patent protection is limited, or as backup protection when companies make commercial decisions to not apply for patents, are misplaced. The patent system already accommodates some of the concerns relating to the patentability of processes for manufacture of biologics — for example, the Patents Act 1990 (Cth) allows extensions to patents for pharmaceutical substances when produced by recombinant DNA technology. The Commission considers that there are no grounds for extending the period of data protection. 28 The US Federal Trade Commission (FTC) was not persuaded by these arguments. The FTC (2009) concluded that extended data exclusivity for biologics was not warranted because the drug has already been incentivised through patent protection and market-based pricing. PHARMACEUTICALS — GETTING THE RIGHT POLICY PRESCRIPTION 317 Confidentiality of data In addition to follow-on manufacturers being prevented from free-riding on clinical data for a limited time, any such data is kept confidential indefinitely. The PPR considered that allowing researchers access to this data could provide substantial public health benefits. The European Medicines Agency (EMA) similarly argued that publishing clinical trial data can help avoid duplication of clinical trials, foster innovation and encourage development of new medicines, build trust in the regulators decision-making processes, and allow academics and researchers to reassess clinical data (EMA 2016). Several industry-led initiatives to increase access to clinical trial data already exist. The International Federation of Pharmaceutical Manufacturers and Associations said there was ‘a huge amount of information on clinical trials published online’ (trans., p. 341). Medicines Australia (sub. DR529) indicated that in many instances clinical trial data is already being made publicly available, including through industry developed Principles for Responsible Clinical Trial Data Sharing, and the Yale Open Data Access (YODA) project. However, the Commission notes that these initiatives do not represent full or systematic disclosure — it is up to the discretion of the contributing companies as to what is released. While Australia can do so, unilaterally publishing clinical trial data would not be without downsides. Such actions could have flow on effects in other markets — leading to a loss of protection. If faced with a loss of protection in larger markets, companies may instead choose to delay bringing products to Australia until protection in other jurisdictions has expired, leading to potentially valuable drugs being withheld. Any publication of data also requires steps to be taken to ensure the data is protected against unfair commercial use. The EMA completed extensive consultations on guidelines for publishing clinical trial data in 2014. In doing so, it had to address many of the same concerns that have been raised with the Commission. The guidelines seek to limit the downsides of data publication by encouraging only appropriate use of the data, and limiting unfair or damaging uses. For example, clinical reports published under the policy do not contain commercially confidential information, and cannot be used for commercial purposes. The Commission supports the eventual publication of data (after the patent and any extensions have expired) but considers that any moves to publish the data are best coordinated internationally (recommendation 18.2). FINDING 10.1 There are no grounds to extend the period of data protection for any pharmaceutical products, including biologics. 318 INTELLECTUAL PROPERTY ARRANGEMENTS 10.5 Strategic behaviour Pharmaceutical patents are valuable assets. The ability of companies to leverage their IP rights to forestall entry by generics has a direct and significant impact on their profitability. Reviews that have examined the use of IP rights by pharmaceutical manufacturers suggest that firms use a variety of instruments to further extend the commercial life of their products. Two such strategies are so-called ‘evergreening’ and pay-for-delay. Evergreening Although originator companies object to it, the term ‘evergreening’ has become commonplace to describe the strategic use of patents to maximise the exclusivity surrounding a pharmaceutical product. Medicines Australia argued that it: … should not be confused with patenting of incremental technological or other innovative advances. … no later granted patent can extend the term of an earlier one. … By definition, a second patent cannot be issued for the same invention. When patents on the original inventions expire, then imitators are free to copy the original from a patent system perspective. Subsequent patent applications will be for other innovations which build on the prior original invention and will usually be progressively narrower in scope. (sub. 44, p. 8) But evergreening is a broader concept relating to extending the protection of products, rather than extending the life of a given patent. It refers to the strategy of obtaining multiple patents that cover different aspects of the same product, typically by obtaining patents on improved versions of existing products (Thomas 2009). Some of these ‘improvements’ may be genuine innovations that improve consumer wellbeing — making dosages smaller or significantly reducing side effects. Equally, some ‘improvements’ may be technical changes with scientific impacts that are virtually indiscernible to the consumer (box 10.7). Critics of evergreening assert that the ability to obtain multiple patents on a product over a period of many years effectively extends the term of exclusivity that the patent holder obtains. The use of a series of patents around a single API can also create a ‘patent thicket’ (chapter 7, appendix D). Thickets obstruct generic entry (after the patent on the API itself has expired) by adding to uncertainty — requiring legal and chemical expertise to discern if any given form, delivery mechanism, combination or manner of manufacturing the API will or will not invalidate other patents (sometimes held by a variety of parties). PHARMACEUTICALS — GETTING THE RIGHT POLICY PRESCRIPTION 319 Box 10.7 Follow-on patents Evergreening is enabled by the nature of pharmaceutical products and how these characteristics interact with the patent system. Typically, an originator will seek various ‘follow-on’ patents surrounding a single pharmaceutical. These are further patents that relate to different embodiments, methods or uses of the drug. Examples include: • Enantiomers are a specific type of molecular structural configuration (‘isomer’) that are pairs of molecules that are non-superimposable mirror images (similar to a left hand and a right hand). A racemate is a mixture of equal parts of both enantiomers. Different mixes can have different effects — for example creating a more efficient treatment or having more or less side effects. Different mixtures that target the same ailment in broadly the same manner can be the subject of follow-on patents. • Formulations that include alterations to or particular compositions of the active ingredient to improve the delivery of the product. Examples include gel, tablet or capsule form of a drug, or intravenous injection. Similarly, delivery mechanisms and devices such as immediate, extended, delayed or sustained release can also be the subject of follow-on patents. • Combinations are a mixture of an API with other drugs. This includes synergistic combinations (those that produce a result greater than the sum of the parts), as well as patents where one component acts to enhance the other (reducing side-effects for example). However, if one component is clearly the active compound, and the other is auxiliary, the claim is classified as a formulation. • Second medical uses refer to the use of an existing pharmaceutical for the treatment of an entirely different disease or condition — a prominent example is that Viagra was originally developed to treat heart disease. A second medical use will also generate an additional period of data protection specific to the second use. Using such claims as the basis for follow-on patents only prevents generic entry in relation to the second use — though this can be hard to police as patients may receive a generic ostensibly for one purpose, but be informed by their doctor or pharmacist of the alternative use. • Methods or processes of production includes the process of synthesising the API (and any intermediate components used). These patents do not prevent a generic from producing an API in a different manner, as the method of production should not impact upon bioequivalence (and therefore on regulatory approval), except in the case of some biologics. Nonetheless, these patents may be sought to protect a particularly efficient method of production providing a market advantage to the patentee. Is there evidence of evergreening in Australia? Pharmaceutical companies in Australia are active users of follow-on patents. Christie et al. (2013) analysed 15 of the costliest drugs in Australia (that is, those with the highest total expenditure by government and consumers), and the patents associated with them. They found that the number of patents associated with each drug varied between 22 and 121 patents per drug, with a median of 45 patents (figure 10.4). Around one-quarter of follow-on patents were taken out by the firms that originated the API, with the remainder held by non-originators. 320 INTELLECTUAL PROPERTY ARRANGEMENTS Figure 10.4 Patent counts for high cost drugsa Simvastatin Omeprazole Salbutamol Ranitidine Enalapril Sertraline Ipratropium Bromide Felodipine Budesonide Captopril Fluoxetine Beclomethasone Glyceryl Trinitrate Famotidine Cimetidine 0 10 20 30 40 50 60 70 80 90 100 110 120 130 No. of patents (total=736) a Drugs arranged in descending order of total cumulative cost to the PBS over the period 1991–2008. Source: Christie et al. (2013). There are several indications that, amongst the body of follow-on patents, examples of evergreening exist in Australia. • Court cases such as Arrow v Merck 29 have identified instances of evergreening. In Arrow, the judge at first instance, Gyles J, found that the originator’s attempt to patent a different dosage amount of an existing drug amounted to ‘what would now colloquially be called an attempt to evergreen a pharmaceutical patent’. 30 • The operation of the PBS has identified likely instances of evergreening (box 10.8). Alphapharm (sub. DR584) submitted evidence that focused on the use of innovation patents in what it argued was evergreening surrounding citalopram and escitalopram. 29 Merck & Co Inc v Arrow Pharmaceuticals Ltd [2006] FCAFC 91 30 Arrow Pharmaceuticals Limited v Merck & Co., Inc. [2004] FCA 1282, [1]. PHARMACEUTICALS — GETTING THE RIGHT POLICY PRESCRIPTION 321 Box 10.8 Follow-on patents and the PBS Analysis of the top 50 drugs on the Pharmaceutical Benefits Scheme (PBS) indicates that more than 10 drugs are the result of follow-on patents. Total Government expenditure on the top 50 PBS drugs is in the order of $5 billion (in 2014-15). Accordingly, strategic behaviour by pharmaceutical companies, if unimpeded by the countervailing power of the Commonwealth, has the ability to cost taxpayers hundreds of millions of dollars per year. While it can be difficult to delineate between genuine follow-on innovations and pure strategic evergreening, within the top 50 drugs, there are: • three instances where the current drug involves a different delivery mechanism to an earlier version of the same compound • three instances where the current drug is an extended release version of the earlier drug • two instances where the current drug is a metabolite of the earlier drug (that is, the human body can make the current drug from the earlier version through the process of metabolism) • one instance where the current drug is an enantiomer of the earlier drug • one instance where the current drug is a biosimilar — from the same rodent antibody — to the earlier drug. The ability of the Pharmaceutical Benefits Advisory Committee (PBAC) to consider the cost effectiveness of a drug on the PBS allows it to largely counter the effects of evergreening, either by refusing to list a drug that is not cost effective compared to an earlier patented drug, or by recommending the formation of a therapeutic group. The formation of a therapeutic group (under s. 84AG of the National Health Act 1954 (Cth)) containing similar drugs means the PBS only pays the price of the cheapest drug in the group. A key consideration in forming a therapeutic group is whether drugs in the group are ‘interchangeable on an individual patient basis with another drug or medicinal preparation’. An example of the use of therapeutic groups to counter the effect of evergreening involves the depression drug desvenlafaxine, which is the major active metabolite of venlafaxine. Venlafaxine was first patented in 1983. A patent for an extended release version of venlafaxine was due to expire in 2023, but was found to be invalid in 2011 (following an injunction that prevented the entry of generic venlafaxine from June 2009 to November 2011). Patents for desvenlafaxine do not expire until 2023. The therapeutic group for venlafaxine and desvenlafaxine ensures that taxpayers do not pay a higher price for desvenlafaxine when it has the same effect as venlafaxine. Sources: Department of Health (2013, 2016a). While it is clear that the preconditions for evergreening are present and examples of the practice can be identified, it is difficult to be definitive about the extent and impact of the practice. This is because, as pharmaceutical companies argue and courts have found, a 322 INTELLECTUAL PROPERTY ARRANGEMENTS follow-on patent can in many cases represent genuine cumulative innovation and attempts to maximise returns (including through ‘life cycle management’): 31 • As Medicines Australia noted, incremental innovation can improve the properties or result it new physiological interactions of a medicine, improve its use in particular sub-groups (such as children), and can lead to a better understanding of the medicine or the condition it treats: … captopril was the first medicine to inhibit an enzyme, angiotensin converting enzyme or ‘ACE’, that was found to be linked to congestive heart failure. It was later discovered that captopril was accompanied by unpleasant side effects such as itching and headaches. Subsequent R&D to address the limitations of captopril not only eliminated unwanted effects, but also yielded a completely new understanding of the enzyme involved. (sub. DR529, att., p. 3) • The courts have also ruled that several follow-on patents have been genuinely inventive and novel and found them to be valid. Examples include Lundbeck 32 (where the Full Federal Court held that a follow-on patent for an isolated enantiomer of an antidepressant was novel and valid) and Aktiebolaget 33 (where the High Court overturned the Full Federal Court and found that a follow-on patent for a new formulation was inventive and thus valid). Despite the difficulty of quantifying the precise impact of evergreening, the combination of a profit incentive, a sophisticated industry and sufficient ambiguity in the law means that it would be surprising if some degree of evergreening did not occur. As Chalmers concluded when examining Australian trends in evergreening: … there is obviously a large financial incentive for originator drug companies to push the boundaries of protection systems. For these companies, a patent is another business tool to be exploited as part of their duty to maximise shareholder returns … Efforts to ‘evergreen’ … which might alternatively be views simply as an astute use of legal rules — will continue. (2007, p. 59) Policy responses lie in the broader framework Removing the availability of follow-on patents for pharmaceuticals would be a blunt response that could result in a loss of valuable medical advances. Better solutions to address any undue evergreening lie in broader policy settings, including a tighter focus on the additionality achieved by follow-on patents on a case-by-case basis. 31 Life-cycle management refers to a range of business and marketing strategies connected with the patent cycle. One example is prescription switching wherein prescribers are induced to switch from an old variety of the drug, to a newer variation protected by a follow-on patent with a later expiry date (Harris, Nicol and Gruen 2013, pp. 105–106). Other strategies could include trademarking the brand name of the drug to improve marketing. 32 H Lundbeck A/S v Alphapharm Pty Ltd [2009] FCAFC 70. 33 Aktiebolaget Hassle v Alphapharm Pty Ltd [2002] HCA 59; (2002) 194 ALR 485. PHARMACEUTICALS — GETTING THE RIGHT POLICY PRESCRIPTION 323 As discussed in chapter 7, the Commission considers that there are already grounds for raising the inventive step for patents (recommendation 7.2). The likelihood of evergreening practices further highlight the potential benefits of this reform. Addressing the level of inventiveness should go a large way to preventing evergreening on purely strategic grounds, while still rewarding innovations that are genuinely beneficial. Pay-for-delay The term ‘pay-for-delay agreements’ refer to patent holders (originators) paying generic manufacturers to keep the generic product off the market, beyond the scope of a patent (both in terms of a generic that may not breach the original patent, or preventing entry after the expiry date) as part of a settlement agreement to resolve a court action. Delays of this kind limit the number of products on the market and thereby any price reductions that may come with competition. Delayed entry also has the effect of postponing any regulatory price drops (such as the automatic price reductions or ongoing price disclosure savings under the PBS). Such anticompetitive behaviour benefits the firms to the detriment of consumers and the Australian Government. Evidence from overseas In the United States (US), the FTC has repeatedly raised concerns about the effects of pay-for-delay agreements on consumer prices. To illustrate the impact, the FTC (Leibowitz 2009) estimated that eliminating pay-for-delay agreements would save American consumers US$3.5 billion per annum, based on data relating to the US pharmaceutical market from 2004 to 2008. 34 Other estimates put the deadweight loss (measured as the amount by which the loss to consumers exceeds the gain to producers) at US$527 million over 5 years per challenged drug, or around US$21 billion in total over the next 25 years (Helland and Seabury 2015). Pay-for-delay settlements are well known within the US, where they had previously been held to be legitimate if used as part of a settlement of a patent infringement lawsuit. However, in June 2013, the US Supreme Court held in FTC v Actavis, Inc 35 that some payments by originators to generic competitors to settle patent litigation could violate the US antitrust (competition) laws. Combined with a transparency requirement that pay-for-delay agreements are filed with regulators, the effect of this judgment appears to be reducing pay-for-delay settlements (box 10.9). 34 These estimates are specific to the United States market. The size of any savings to Australia, if pay-fordelay was delaying generic entry here, would likely be several degrees of magnitude smaller. First, the sheer population difference would translate to a substantially smaller market in Australia. Second, as noted above, the price decrease due to generic entry appears to be lower in Australia than in the US. A further difference is the distribution of any savings — the presence of the PBS would likely mean that a greater portion of the savings would go to government than consumers. 35 133 S. Ct. 2223 (2013). 324 INTELLECTUAL PROPERTY ARRANGEMENTS In Europe, the EC has completed six monitoring exercises since 2010 aimed at determining if settlements have included detrimental pay-for-delay agreements. In its most recent monitoring report, the EC concluded that settlements that might attract competition law scrutiny — those involving restricted generic entry and value transfer from the originator to the generic company — have reduced since the time of the pharmaceutical sector inquiry (by 22 per cent of reported settlements from 2000 to 2008), and have ‘stabilized at a low level’ (12 per cent in 2014) (EC 2015a, p. 15). Nonetheless, high-profile instances of pay-for-delay agreements still arise in Europe. In February 2016, the United Kingdom’s Competition and Markets Authority (UK CMA 2016) announced a fine of over £37 million on the originator GlaxoSmithKline (and over £7 million on the generic pharmaceutical companies GUK and Alphapharm) for pay-for-delay agreements made as a settlement to a 2001 patent infringement case over paroxetine (an anti-depressant whose sales exceeded £90 million in 2001). When generic entry eventually occurred in 2003, the average price of paroxetine dropped by over 70 per cent over the course of two years. The CMA (2016) found that GlaxoSmithKline’s conduct infringed the competition law prohibitions on anticompetitive agreements and had abused its dominant position in the market. Another prominent example of enforcement activity saw fines totalling AU$209 million imposed on Lundbeck, a Danish pharmaceutical company (Ashurst Australia 2013). Canada has also raised concerns about the impact of pay-for-delay settlements. A report by Canada’s Competition Bureau (CBC 2014) observed that the absence of a formal notification system (as used in the United States) ‘could lead to potentially anticompetitive settlements evading review’ and argued that a settlement notification system may be necessary, despite differences in Canada’s regulatory regime. Subsequently, the Competition Bureau published its Intellectual Property Enforcement Guidelines (CBC 2016) which articulates how the Bureau intends to approach enforcement matters where competition policy and IP rights (including pay-for-delay agreements) intersect. The Law Council of Australia (sub. DR490, p. 23) submitted that it was ‘instructive to note that there is no suggestion in those Guidelines that a formal notification regime … be introduced’. However, the Guidelines are an enforcement document — they describe how the Bureau intends to examine and enforce particular competition issues in the IP sphere — not a policy document, and should not be interpreted as a change in policy position. PHARMACEUTICALS — GETTING THE RIGHT POLICY PRESCRIPTION 325 Box 10.9 Pay-for-delay: transparency and activity in the United States In the United States, since 2004, s. 1112 of the Medicare Prescription Drug, Improvement, and Modernization Act of 2003 has required that agreements between generic and originator companies that involve the manufacture, marketing or sale of drugs be filed with the FTC and the Assistant Attorney General (effectively, the authorities responsible for enforcement of competition law). Not all of these agreements are necessarily pay-for-delay settlements. Of the 160 agreements filed in the fiscal year 2014 (from October 2013 to September 2014), the FTC considered that 21 potentially involved pay-for-delay because those agreements contained: … both explicit compensation from a brand manufacturer to a generic manufacturer and a restriction on the generic manufacturer’s ability to market its product in competition with the branded product. (2016, p. 1) While the FTC has been able to observe these payments for over a decade, changes in case law had hamstrung its ability to enforce any penalties. FTC enforcement actions effectively deterred such arrangements until 2005, when the Eleventh Circuit held that pay-for-delay settlements were generally immune from antitrust scrutiny. As several other Circuits followed suit, the number of pay-for-delay settlements increased dramatically. (Fialkoff 2014, p. 4) The response from the pharmaceutical industry was rapid. The number of potential pay-for-delay settlements observed by the FTC (2016, p. 4) jumped from three in 2005, to 16 in 2008, to a high of 40 in 2012. In (US) fiscal year 2013, the year of the Actavis decision that allowed potential enforcement action against pay-for-delay settlements, the number fell to 29. The number fell again to 21 in fiscal year 2014 (the most recent year for which data was published). Although it is too early to reach definite conclusions, the combination of transparency and potential penalty appears to have reversed the increasing trend of pay-for-delay settlements in the United States. However, some have cautioned that further regulation may still be required to protect against anticompetitive outcomes. In particular, a subsequent (District Court) case36 held that the ruling in Actavis only applied to monetary payments, leaving other deals that result in delaying the generic (for example allowing the generic company to sell one product in exchange for not selling another) immune from antitrust scrutiny (Fialkoff 2014). Evidence from Australia While pay-for-delay settlements constitute an offence in Australia under the Competition and Consumer Act 2010 (Cth) (CCA), 37 Australia does not have any arrangements in place to detect such behaviour, and there is no incentive for either party to disclose the payment, 36 Lamictal Direct Purchaser Antitrust Litig. v. All Direct Purchaser Action (In re Lamictal Direct Purchaser Antitrust Litig) (2014) No. 12-cv-995. 37 Pay-for-delay agreements could breach the prohibition on anticompetitive contracts with the purpose of substantially lessening competition (s. 45), or the cartel prohibitions (Part IV, Division 1). In some instances, pay-for-delay may fall within the exception under s. 51(3). The Australian Competition and Consumer Commission (ACCC) could use the authorisation process (Part VII) to grant legal protection where it is satisfied that the public benefit from the proposed conduct outweighs the public detriment (ACCC, pers. comm., 22 July 2016). 326 INTELLECTUAL PROPERTY ARRANGEMENTS exacerbating the difficulty of regulatory detection. The Commission is unaware of any proven pay-for-delay cases in Australia to date. However, the unsuccessful action taken by the ACCC against Pfizer in 2014 38 highlights circumstances that could see pay-for-delay agreements arising in Australia in the near future. … these type of arrangements are often prompted by a similar issue to that which was being grappled with by Pfizer (i.e. the impending expiry of relevant patents). While key patents protecting most of the so-called ‘blockbuster’ drugs in Australia have expired in recent years, a significant number of high-volume drugs protected by patents remain on the PBS. As the terms of those patents approach expiry in Australia in the coming years, it may be that an Australian court will soon have the opportunity to consider whether pay-for-delay arrangements should be subject to scrutiny under the [Competition and Consumer Act 2010 (Cth)]. (Herbert Smith Freehills 2015) Following the draft report, some argued (for example, GBMA, trans., p. 350) that there were inherent differences between the legal framework in Australia and the United States which make it less likely that pay-for-delay cases will occur in Australia. However, the differences do not rule out the possibility of, or incentives for, pay-for-delay settlements in the Australian context. In particular, the potential for statutory price reductions through the PBS provides an additional incentive (at taxpayers’ expense) for any challenged originator to settle in a manner that delays entry. How can pay-for-delay be addressed? The Commission’s preferred option to manage pay-for-delay risks is to improve monitoring and transparency of settlement agreements to detect any pay-for-delay arrangements. Where this indicates further action is warranted, enforcement of existing competition law should be pursued, leaving the courts to determine the legality of any allegedly anticompetitive agreements. Monitoring would also improve the ‘credible threat’ of sanction under the existing regulation, providing a deterrent and potentially reducing the incidence of pay-for-delay agreements. Some participants (the Law Council of Australia, sub. DR490 and Medicines Australia, sub. DR529) argued that the ACCC’s existing powers (under s.155 of the CCA) were already sufficient to detect pay-for-delay agreements. However, this section requires that the ACCC first have formed the ‘reason to believe’ that there is a contravention. The ACCC highlighted the difficulty this can cause in the context of pay-for-delay agreements. 38 This case did not involve pay-for-delay. It related to Pfizer’s actions in the lead up to the expiry of its patent on atorvastatin (under the brand Lipitor), the then highest-selling medicine on the PBS with sales for 2011-12 of over $700 million (Herbert Smith Freehills 2015). The Federal Court found that it was not established that Pfizer’s actions had the required purpose of substantially lessening competition, and upon expiry, Pfizer was not in a position of market power. Pfizer had a settlement agreement with the generic company to sell the product before expiry (Medicines Australia, sub. DR529, p. 19). PHARMACEUTICALS — GETTING THE RIGHT POLICY PRESCRIPTION 327 Before the ACCC can investigate potentially anti-competitive pay-for-delay agreements we have to be aware that they exist in the first place. The delayed entry of a competitor is not something that is generally observable, so the ACCC would have no reason to commence an investigation. Some anti-competitive arrangements (for example cartels) are inherently secretive and thus difficult to detect. (pers. comm., 22 July 2016) This is the evidentiary gap that monitoring is intended to fill. While the potential benefits to the economy of avoiding anticompetitive agreements could be high, the Commission acknowledges that this would involve some minor compliance costs for Australian pharmaceutical companies. The approach used in the United States provides an example of a monitoring scheme with minimal compliance costs. There, pharmaceutical companies must lodge pre-existing settlement documents (drawn up for the purposes of the court action), and are not required to furnish any specific documents or data for the purposes of the regulator. The FTC then examines the documents for signs that the agreement may be a potential pay-for-delay agreement. Failure to file the documents with the FTC can lead to criminal or civil penalties. Following the draft report, Medicines Australia supported the concept of a transparent reporting system, but argued that compliance burdens could be minimised by relying on other data sources. In responding to a question from the Commission, Medicines Australia argued that data collected for multinational tax avoidance purposes could be used: The recently introduced multinational tax avoidance laws and country by country reporting will provide a range of new data to the ATO. It is suggested that the information reported to the ATO should be examined to determine if there is the possibility of detecting potential pay for delay arrangements. (sub. DR529, att., p. 3) The Commission has consulted with the Australian Taxation Office (pers. comm., 26 July 2016), and considers that the general financial statements, and financial records of transactions required under the Multinational Anti-Avoidance Law 39 (aimed at curbing international tax avoidance by multinational entities), would not be effective in detecting anti-competitive provisions in court settlements between unrelated companies. The Law Council of Australia (sub. DR490, p. 21) noted the important role that settlements can play in the court system and raised concerns that monitoring them could have a ‘substantial chilling effecting on the motivation of parties involved in pharmaceutical patent disputes to resolve such disputes prior to their adjudication …’. International experience suggests that such fears do no play out. • Monitoring commenced in 2010 in Europe and the overall number of settlements there continued to grow to 2012, before moderating in 2013 and 2014 (remaining at a higher level than the pre-monitoring phase). However, the proportion of settlements that were concerning from a competition perspective fell over the period. 39 Enacted by the Tax Laws Amendment (Combating Multinational Tax Avoidance) Act 2015 (Cth). 328 INTELLECTUAL PROPERTY ARRANGEMENTS • Since the Actavis decision, there has been an increase in the overall number of settlements in the United States, but a decrease in the number considered potential pay-for-delay settlements. To ensure that monitoring arrangements are well targeted, the ACCC should develop guidelines modelled on those used in the United States, Canada, and Europe but tailored to the conditions and legal framework in Australia. Such an approach was supported by the Law Council of Australia as a means to reduce uncertainty (sub. DR490, p. 22). As the ACCC (pers. comm., 22 July 2016) noted, these guidelines could usefully be incorporated into more general guidelines on the application of the CCA to IP (recommendation 15.1). To improve the effectiveness of monitoring arrangements, firms should be required to lodge any agreements with a direct impact on the entry of a generic (or competitor) product onto the Australian market, not just agreements made in Australia. Further, as in the United States, consideration should be given to the ability to levy civil penalties in order to ensure compliance with the regime (ACCC, sub. DR603). The Commission notes that failure to lodge a settlement (observable by examining court lists) may of itself present reason for further investigation by the ACCC. Bearing in mind the potential compliance burdens, the Commission recommends that the monitoring scheme be limited to an initial period of five years. Following this period there should be a review of the ongoing necessity of regulatory oversight of potentially anti-competitive behaviour in the pharmaceutical sector, including pay-for-delay. RECOMMENDATION 10.2 The Australian Government should introduce a system for transparent reporting and monitoring of settlements between originator and generic pharmaceutical companies to detect potential pay-for-delay agreements. This system should be based on the model used in the United States, administered by the Australian Competition and Consumer Commission, and include guidelines on the approach to monitoring as part of the broader guidance on the application of the Competition and Consumer Act 2010 (Cth) to intellectual property (recommendation 15.1). The monitoring should operate for a period of five years. Following this period, the Australian Government should review the regulation of pay-for-delay agreements (and other potentially anticompetitive arrangements specific to the pharmaceutical sector). PHARMACEUTICALS — GETTING THE RIGHT POLICY PRESCRIPTION 329 11 Registered designs Key points • Registered design rights serve a niche role in Australia’s intellectual property (IP) rights system; protecting the appearance of products that have an industrial or commercial use. – To enforce a design right in Australia, the design must be registered as well as examined and certified (to confirm the design is new and distinctive). • Participants raised concerns about Australia’s design rights system, including: – the cost of seeking protection in multiple jurisdictions – poor understanding of design law, which can lead to designers inadvertently losing their rights or failing to secure protection in the first place – unauthorised copying and the associated costs of enforcing design rights – lack of adaptability to changes in technology and industry practices. • The Commission has taken the Australian Government’s response to the Advisory Council on Intellectual Property review of the designs system (released in May 2016) into account when considering options to improve the designs system: – Australia should continue efforts to encourage international harmonisation and streamlining of registered designs systems in different jurisdictions, where it is in Australia’s interests. – The Government’s commitment to assess the costs and benefits of the Hague Agreement before joining is a positive step. – The introduction of a grace period for application filings, agreed to by Government, will help address concerns about inadvertent disclosure and allow designers to undertake some market testing prior to incurring filing costs. The design industry also has a role to play in advancing understanding of the designs system among designers. – Building the evidence base on virtual or non-physical designs will help to ensure Australia’s design law can adapt to take account of changing technology. • Broader proposed initiatives, such as reform to create low-cost avenues of enforcement, would help address design participant concerns about enforcement access and costs. In the past two decades, there have been two reviews of Australia’s registered design rights system: • The Australian Law Reform Commission (ALRC) undertook a three year review commencing in 1992. The review made recommendations to improve the registered design rights system through clearer definitions, stricter eligibility and infringement tests, a more streamlined registration system, and better enforcement and dispute REGISTERED DESIGNS 331 resolution procedures (ALRC 1995). The Designs Act 2003 (the Designs Act) is substantially based on the ALRC recommendations (ALRC 2010). • The Advisory Council on Intellectual Property (ACIP) released the final report for its review of the Australian designs system on 31 March 2015 (ACIP 2015a). The Australian Government responded to ACIP’s review on 6 May 2016, agreeing to the majority of ACIP’s 23 recommendations (Australian Government 2016b). This chapter examines opportunities for improving Australia’s designs system, drawing on ACIP’s review and taking account of the Government’s subsequent response. The chapter first provides background on Australia’s designs system (section 11.1). It then examines participant concerns about current arrangements (section 11.2) and options for improving the registered designs system (section 11.3). 11.1 Design rights in Australia Design rights protect visual appearance Registered design rights serve a niche role in Australia’s intellectual property (IP) rights system — protecting the appearance (rather than function) of products that have an industrial or commercial use. Australia’s current designs system has been in operation since 2004 with the commencement of the Designs Act. The Designs Act replaced the Designs Act 1906 (which also conferred design rights) (ACIP 2015a). Under Australian designs law, a design refers to the features of shape, configuration, pattern or ornamentation, which give a product a unique appearance (IP Australia 2015a). Registered design rights give a designer an exclusive right to exclude other parties from using or imitating their design for up to 10 years. The term of protection for registered designs is 15 years in the United States and up to 25 years in the European Union and United Kingdom (EUIPO 2016; UK Government 2016; USPTO 2015). Registering a design gives the designer protection for the visual appearance of the product but not the feel of the product, what it is made from, or how it works (ACIP 2013b). In Australia, businesses and individuals register designs across a range of categories (figures 11.1 and 11.2). The most common types of design applications include those relating to furniture and household goods, clothes, textiles and accessories, and tools and machines. The Organisation for Economic Co-operation and Development (OECD) has observed that during the period 2005 to 2013 the strongest acceleration in design applications in Australia and abroad has been for designs relating to information and communication technologies (ICT), such as screen displays and icons (OECD 2015b). 332 INTELLECTUAL PROPERTY ARRANGEMENTS Figure 11.1 Examples of designs registered in Australia Source: Australian Designs Register (Registration AU 362 158; Registration AU 314 267; Registration AU 350 133; Registration AU 349 632; Registration AU 359 971). Examples of Australian registered designs include the Albion Cricket Helmet (2004), the portable cooler (1987), Speedo’s Fastskin suit (2000) and the shape of the Holden Monaro (2003) (IP Australia 2015a). REGISTERED DESIGNS 333 Figure 11.2 Designs applications in Australia, by design category 2010 to 2015a Advertising Resident Agricultural products & food preparation Non-resident Health, pharma & cosmetics Leisure & education Electricity & lighting Transport ICT & audio-visual Packaging Construction Tools & machines Clothes, textiles & accessories Furniture & household goods 0 2 4 6 8 10 12 14 Share of total applications (per cent) 16 a The product categories are based on the OECD’s taxonomy for aggregating Locarno classes — an international classification used for the purposes of the registration of industrial designs. Sources: Intellectual Property Government Open Data database (2016 edition); OECD (2015b). The designs system is intended to encourage design effort The purpose of Australia’s industrial designs system is to encourage investment in designs ‘by giving designers the exclusive right to exploit their designs for a limited time and prevent competitors free riding on design innovations’ (Revised Explanatory Memorandum, Designs Bill 2003, p. 2). From an economic perspective, preventing free riding is desirable when it would otherwise have a significant detrimental effect on innovation. For example, in circumstances where the cost of developing a design is high and the costs of copying are low, there may be too little design creation in an unregulated environment. The right to exclude other parties from using a design provides opportunities for the owner to recoup their investment costs through exclusive exploitation of their design (ACIP 2013b). Design rights also seek to assist the dissemination of the creative ideas of designers, by providing an accessible register of existing designs. But this is a less convincing rationale and questions remain about the extent to which design protection helps disseminate creative ideas. As is the case with IP rights more broadly, providing exclusive design rights can create costs for the Australian community. These costs arise because design rights can prevent other 334 INTELLECTUAL PROPERTY ARRANGEMENTS businesses from offering products with a similar appearance, which can mean consumers face higher prices and more limited access to designs. Design law must therefore strike a balance. Obtaining a design right involves several steps A business or individual may seek protection for a new and distinctive design by lodging a design application with IP Australia. A fee of $250 applies to each design in an online application (IP Australia 2015a). If an application contains multiple designs, the applicant needs to pay for each design included. As is the case with other IP rights, the cost for an IP professional to prepare the relevant application often exceeds the application fee. Appendix F details the relevant criteria and application process. The vast majority of design applications proceed to registration IP Australia processes design applications and sends applicants a notice advising them of their design number and filing date. If an applicant does not request registration at the time of filing a design application, they have 6 months from the application priority date (usually the filing date) to decide whether to apply to have their design registered. Once an applicant requests design registration, IP Australia undertakes a formalities check to make sure the necessary information and representations are present and, if they are, IP Australia registers the design. The formalities check makes sure the application is in order for registration but does not check whether the design is new and distinctive. Over 90 per cent of design applications proceed to registration (ACIP 2013b). The initial period of design registration lasts for five years from the filing date. An applicant may choose to renew their design registration for a further five years, to a maximum of 10 years, by paying a $320 registration renewal fee (IP Australia 2015a). A small proportion of registered designs are examined and certified People who own registered designs may only enforce those rights after IP Australia has examined the registered design to ensure it is new and distinctive, and issued a certificate of examination. A registered design owner (or a third party) can request an examination of the registered design at any time after registration for a fee of $420 (if a third party requests examination, the third party and owner each pay half of the examination fee). The design owner therefore has a window of up to 10 years to exercise the option of enforcement by securing certification. If an examination results in an adverse report, IP Australia may revoke the design registration. IP Australia receives requests for examination for about one fifth of registrations, with most examinations resulting in certification. About 10 per cent of examination requests result in a deficiency notice because the applicant is unable to rectify problems with the application, or the applicant chooses not to proceed (ACIP 2013b, 2014). REGISTERED DESIGNS 335 IP Australia has suggested that the low level of certification relative to registrations might be partly because applicants avoid voluntary examination, and its associated fee, until there is a need to enforce design rights (IP Australia 2015b). Some designers might also consider that registration on its own provides sufficient incentive for others not to copy a design. Once IP Australia has examined a registered design and issued a certificate of examination, the applicant has the exclusive and legally enforceable right to use, license or commercialise their design. In a survey of designs system applicants undertaken for ACIP’s review, about 80 per cent of respondents indicated that their designs had been licensed, sold or otherwise commercialised (ACIP 2015a). Designers may also apply for design protection in other countries Australian designers can obtain design protection in other countries, typically through one of two methods: • making a new application in each foreign country (national applications) • making a new application in each foreign country within six months of their Australian application and specifying that they are claiming the date on which they lodged their Australian application as the priority date of their application (a convention application). Given Australia’s relative market size, Australian resident applications abroad typically outnumber resident applications in Australia (WIPO 2016d). 11.2 Concerns about Australia’s design rights system Stakeholders, including participants in this inquiry, have expressed various concerns about Australia’s design rights system, including low uptake of design rights (due to the cost of registration and enforcement, limits on the scope and duration of protection, and poor understanding of design law), the proliferation of low-cost imitations or ‘replica’ products and lack of adaptability to changes in technological and industry practices. In assessing the nature and extent of these concerns, the Commission has drawn on the framework set out in chapter 2 of this report. Low uptake of design rights in Australia Several commentators have suggested that the numbers of design applications and registrations each year are low given the economic significance of industrial design (Gilbert and Tobin, sub. 96), the vast array of industrial designs in day-to-day domestic and commercial life, and the wide range of products available in the Australian market (ALRC 1995). The Design Institute of Australia (sub. 131, p. 4) argued that ‘if the proof of 336 INTELLECTUAL PROPERTY ARRANGEMENTS the effectiveness of a registration system is its rate of utilisation, clearly Australia’s designs registration scheme is a failure’. ABS data on the use of formal IP protection by Australian businesses suggests use of design rights is not widespread. For example, less than four per cent of businesses in a 2012-13 survey used registered designs to protect their IP (figure 11.3). Figure 11.3 Businesses in Australia using formal IP protection 20 18 16 14 Per cent 12 10 8 6 4 2 0 2003 Patents 2008-09 Registration of design 2012-13 Copyright or trademark Source: ABS, Innovation in Australian Business, Cat. no. 8158.0. The number of design applications lodged each year is also typically much lower than the number of patent and trade mark applications received. For example, in 2015, IP Australia received approximately 7000 design applications (figure 11.4), 29 000 standard patent applications and 73 000 trade mark applications (IP Australia 2016a). ACIP (2015a) noted that, while there has been an upward trend in design applications over the past few decades, the number of design applications by Australian residents each year has been static (figure 11.4). Applications from non-residents (particularly companies) account for nearly all of the increase in total applications. Of the people who register a design, only a low proportion renew. As at 2013, less than 20 per cent of Designs Act 2003 applicants had renewed their registration at the 5 year threshold (ACIP 2014). However, there is no reason for there to be comparability across rates of IP registration. Nor is a low level of design registrations relative to trade marks and patents unique to Australia. For example, Ricketson and Suthersanen (2012, p. 182) find ‘for all the rich REGISTERED DESIGNS 337 diversity of national designs law options that may be accommodated within the TRIPS/Paris/Berne framework, it is surprising to see how little use is made … of registered design systems [relative to patents and trademarks]’. Moreover, it is important to recognise that design registration is not an end in itself. Rather, it is a means of encouraging socially valuable design activity that would have not otherwise occurred. Figure 11.4 Design applications in Australia, 1990 to 2015a Applications (000s) 8 7 Total Resident Non-resident 6 5 4 3 2 1 0 1990 1995 2000 2005 2010 2015 a Australia’s current designs system has been in operation since the commencement of the Designs Act 2003 on 17 June 2004. Source: Intellectual Property Government Open Data database (2016 edition). Designers protect their innovations using market strategies … One likely reason the take up of design rights is low is that designers are able to protect their innovations to some degree through market strategies. Researchers conducted interviews with managers and designers in Australian firms and found that buyers and sellers had adapted to imperfections in the market for design by employing multi-layered strategies to deal with concerns such as appropriation. One industrial designer noted ‘ … I’m much better off I think using my time to be there first, making it en masse at an achievable price-point, rather than trying to protect it’ (Lim, O’Sullivan and Falk 2014, p. 13). One participant noted that ‘designers in Australia these days are intentionally designing complex product so that they can’t be ripped off’ (Terri Winter, trans., p. 377). 338 INTELLECTUAL PROPERTY ARRANGEMENTS … or by using alternative forms of IP protection Low take up of design rights might also indicate that designers are opting to protect their innovations using other forms of IP protection. For example, a designer wishing to visually distinguish a product in the marketplace (such as a distinctively shaped perfume bottle) may opt for trade mark protection over design protection. 1 Similarly, a designer who wishes to protect a product for which there is a close relationship between function and appearance (such as an ergonomic chair) may opt for patent protection, provided the innovation is sufficiently inventive (figure 11.5). Figure 11.5 Overlaps in protection provided by design and other IP rightsa Trade marks Source identification Non-functional features that distinguish a trader’s goods or services (eg perfume bottles) Innovations for which function and form are closely related (eg ergonomic chair) Patents Product function Design rights Product appearance A literary or artistic work used as an ornamental feature of an industrially produced article (eg T-shirt print)b Copyright Original expressions a Figure does not show overlaps between other IP rights (such as patents and trade marks). b People creating ‘works of artistic craftsmanship’ can choose whether to register the design, but registering will generally result in a loss of copyright. 1 The Law Council of Australia (sub. DR490) questioned how effective market strategies and other forms of IP are in protecting a product’s design. For example, they noted in many if not most cases a shape trade mark registration will only be obtained on sufficient evidence of use of the shape as a trade mark. REGISTERED DESIGNS 339 Australia’s ‘copyright/design overlap provisions’ limit the scope for designers to use copyright instead of design protection. However, some overlaps exist. ACIP explained the copyright/design overlap provisions as follows: Where the copyright/design overlap provisions apply, a person wanting exclusive rights to make and sell three-dimensional products of some particular design cannot rely on copyright in any underlying drawings or models, but must register a design. The provisions do not eliminate copyright protection entirely: reproducing drawings in two-dimensional form (such as in a poster) without permission is an infringement of copyright. The policy is not meant to apply to the exploitation of an artistic work in two––dimensional form (for example, printed on T––shirts). The basis for this distinction is the view that both copyright and design protection should be available for an artistic work applied as a two-dimensional decorative design, since, used this way, an artistic work ‘retained its essential character as an artistic work’. Thus the copyright-design overlap provisions do not apply where an artistic work is applied as surface ornamentation. (2014, p. 30) Designers can also use different IP protections to protect different aspects of a product. For example, it is possible for a patent to apply to how a product works, for a registered trade mark to protect the product’s brand name, for a registered design to protect the product’s appearance and for copyright protection to apply to product manuals (IP Australia 2015a). ACIP found that registered design holders often use other forms of IP protection (patents, copyright and trade marks). For example, ABS data suggest 72 per cent of small to medium sized enterprises (SMEs) that used registered designs from 2005-06 to 2011-12, also used copyright or trade marks (ACIP 2014). In some cases, however, the marginal benefit of design protection to a designer’s ‘IP bundle’ might be insufficient to justify the associated cost. For example, Lim, O’Sullivan and Falk (2014, p. 13) noted one industrial designer’s strategy has been to take out trade mark protection on his name and brand, ‘eschewing options of patents and design registration for his products’. The extent to which particular designs lend themselves to protection through alternative forms of IP rights (and market strategies) will likely vary. In a survey of firms in the United Kingdom, for example, a number of design-intensive firms suggested that ‘brand is ultimately a more defendable asset, and a registered trade mark retains value even if the individual designs change frequently’ (Moultrie 2011, p. 17). Even if a business has not registered a design, they might still be able to take action to protect the goodwill and reputation they have acquired by the distinctive shape and configuration of a product under the tort of passing off or under the Australian Consumer Law (Davies Collison Cave 2015). Designers generally consider design rights as costly for what they offer Demand for design protection will depend on the benefits that it provides to rights holders (with respect to the term and scope of protection and enforceability of the right) relative to 340 INTELLECTUAL PROPERTY ARRANGEMENTS its costs (such as fees and administrative and legal costs). ACIP (2015a, p. 41) found that the designs system in Australia ‘is expensive for what it offers, and is, as a result, neglected by designers who find it does not offer the rights they need’. Several individual designers and design organisations that participated in this inquiry suggested that designers often elect not to register designs due to the expense. Specific concerns raised by designers regarding the value of registering designs include: • the expense of registering and certifying designs for products in industries with a high volume and turnover of designs (such as clothing and accessories) or which have multiple designs in a product range, is prohibitive (ACIP 2015a; Australian Design Alliance, sub. DR619; Dinosaur Designs 2013). One designer noted that the cost of design registration is ‘inhibitive’ and that ‘as such it’s something I can’t afford to do, because you’re working on a number of products and projects’(Fereday, trans., p. 390) • the current term of design protection is insufficient for some designers to recover a sufficient return to justify their investment, particularly in sectors where research and development is lengthy (ACIP 2015a; Australian Design Alliance, sub. DR619; Trubridge, sub. DR573; Terri Winter, sub. DR198) • the cost of enforcing design rights can be high relative to the revenue generated by a design (Fitzsimmons 2015). The Design Institute of Australia noted ‘our members tell us that essentially they need to spend a lot of money on lawyers to enforce that protection. Most design studios are small businesses, or very small businesses, being an individual and they simply don’t have the resources to enforce any of that’ (trans., p. 276) (chapter 19 outlines a proposal to enhance the role of the Federal Circuit Court, which should help address these concerns about the access to and cost of enforcing rights). Some participant concerns related to the lack of international harmonisation with other countries on matters such as administrative procedures, which they argued increased the legal and administrative costs incurred by Australians exporting to international markets (ACIP 2015a; Design Institute of Australia, sub. 131). For example, unlike in some jurisdictions, Australian applicants are unable to use the same application process for multiple international markets. There is little data on the extent of these costs to Australian designers. Efforts to reduce the cost of obtaining design protection also need to consider the trade-offs. Relaxing administrative requirements or reducing application fees can increase the risk that protection is afforded to products that have little social value, and may impede innovation (particularly as protection of product appearance can indirectly protect function). Reducing fees (as opposed to restructuring fees) for design applications means IP Australia must find other sources of revenue to fund its administrative activities, such as increasing other IP fees (the cost of administering registered design rights is already partly subsidised by fees for other IP rights). Further, it will not always be the case that designs will involve high upfront costs (for example clothing). REGISTERED DESIGNS 341 Designers’ understanding of the designs system is often low Some participants have suggested the lack of take up of design rights is partly because designers’ understanding of the design rights system is generally low (ACIP 2015a; Gilbert and Tobin sub. 96; IP Australia sub. 23). If designers do not understand the benefit of design protections (for example, because they falsely assume they have protection from other IP laws) or if they find the process of obtaining design protection daunting or confusing, they will be less likely to take up design rights. Law firm Gilbert and Tobin argued that even sophisticated businesses have a very poor understanding of the need to secure design rights: … it is our experience that most designers only commence registering designs (if at all) after a usually costly experience of third party copying of an unprotected design which the designer is powerless to prevent under Australian law. (sub. 96, p. 4) Participants suggested sources of confusion in Australian designs law include: • the multistep process for registration and certification, which can mean some applicants are unsure, or have a false impression, about the legal status of their design (ACIP 2015a; Design Institute of Australia, sub. 131) • the complexities of the copyright/design overlap provisions, which can lead to some designers inadvertently losing copyright protection (ACIP 2015a; Gilbert and Tobin, sub. 96; Institute of Patent and Trade Mark Attorneys of Australia, sub. 73). Some participants have called for stronger border protection measures Some participants have raised concerns about the importation of goods to Australia that may infringe upon Australian registered designs and the absence of border protection measures similar to those that apply for other IP rights (the Australian Border Force is empowered to seize goods that infringe trade marks, copyright and protected Olympic expressions) (ACIP 2015a; FICPI, sub. DR581; Gilbert and Tobin, sub. DR565). ACIP’s review recommended that the Australian Government consider introducing border protection measures that align with the Trade Marks and Copyright Acts to allow for the seizure by Customs of alleged design infringements which are identical to certified designs. However, the Government subsequently rejected the idea because it would pose a range of practical difficulties, and would be resource intensive for the Australian Border Force to implement. It further noted that the Designs Act currently provides rights holders with opportunities to take action domestically, such as by seeking injunctive relief and damages, against suspected infringements of certified designs (Australian Government 2016b). The Commission has elected to not re-prosecute the case for introducing border protection measures for registered designs in this report, in part because it has not received new evidence on the issue. However, it has made recommendations to create avenues to low-cost enforcement (including for designers) (chapter 19). 342 INTELLECTUAL PROPERTY ARRANGEMENTS Proliferation of low-cost imitations or replica products Several designers that participated in this inquiry were particularly concerned about the proliferation of low-cost imitation or ‘replica’ products, which they argued undermined incentives for Australian designers to create high-quality, original designs and have other undesirable effects. In particular, participants claimed that low-cost imitations or replica products: • devalue the public perception of the original design because poorer quality replicas are mistaken for the original, or because low-price replicas reduce the status of the original design (Design Institute of Australia, trans., p. 273; Swanton, sub. DR287) • lead to direct loss of sales for original designers (Trubridge, sub. DR573) • result in customer confusion and dissatisfaction (Swanton, sub. DR287; Terri Winter, sub. DR198) • contribute to a culture where design and designers are not valued (Kellock, trans., p. 535; Terri Winter, sub. DR198). In some cases, participants’ concerns about low-cost imitations were that other businesses were making unauthorised copies of registered designs, with little threat of penalty (as discussed above and in other chapters, several participants observed there is no low-cost avenue for enforcement, particularly for SMEs). In other cases, participants claimed that businesses were marketing reproductions of original designs in a manner that was misleading consumers as to the source, quality, and designer’s endorsement of the product. Several participants suggested that the ability for businesses to legally market items as ‘replicas’ was a problem (box 11.1). Concerns existed even when designers had not registered the original design or design registration had lapsed, such that the reproduction of a design was not infringing a registered design (Archer, trans., p. 402). Some participants objected to (legal) low-cost replica products on philosophical grounds: Understandably customers have different budgets. If I cannot afford a Ferrari I cannot buy one, I will buy the model car I can afford to own. The argument to allow others to create a cheaper copy of the Ferrari for someone who cannot afford the real one would make no sense – and yet this situation is the same argument for the allowance of a cheaper copy of an authentic piece of furniture. There are original products available for all budgets. Not having the money for an expensive chair should not give anyone the right to have a copy of it made legal (Terri Winter, sub. DR198, p. 3) … there’s no reason to allow a consumer to purchase a cheaper copy of a particular design of chair when there’s an enormous number of them that they can choose from that will do the same thing. (Schott, Design Institute of Australia, trans., p. 273) REGISTERED DESIGNS 343 Box 11.1 Replicas or rip-offs? Several participants raised concerns about the ability of businesses to market reproductions as a ‘replica’ of the original: At this stage replica is legal and whilst that exists that opens the door to copies. (Anne-Maree Sargeant, Authentic Design Alliance, trans., p. 270) The products that we import into the country, or manufacture ourselves, are able to be sold under the terminology “replica” as direct copies. Furthermore, the use of the designer’s name, the original name of the product, and the manufacturer, can also be used when replica products are sold. So what, I think, this demonstrates is that the replica work can actually utilise the owner and designer’s original ideas and actually trade off the back of them. (Jon Holland, Space Furniture, trans., p. 278) I think the concern about the “replica” word in itself is almost suggesting that it’s an authorised reproduction of the designer’s product. (Richard Munao, CULT, trans., p. 281) The way in which businesses market reproductions has been the subject of legal disputes. In 2011, Herman Miller filed a suit against Matt Blatt in the Federal Court of Australia, claiming that Matt Blatt had infringed its registered ‘Eames’ trade mark; had falsely represented that its products were genuine Eames products or were otherwise approved by Herman Miller, in contravention of the Competition and Consumer Act 2010; and had engaged in the common law tort of passing off. The case was settled with Matt Blatt undertaking to take steps to clearly identify its Eames copies as ‘replicas’ (Commonwealth Courts Portal 2011; McHugh 2011; Tresidder 2011). Where businesses are making unauthorised copies of registered designs, the ability of rights holders to enforce their rights materially impacts on the value of their IP. But the costs of pursuing action can be prohibitive and many SMEs in particular can be deterred from taking action. To directly address this problem, the Commission is recommending a timely and low-cost option for resolving IP disputes (chapter 19). Another variant of the copying problem occurs when designers seek to market test their designs so that they can make better informed decisions — including as to whether to incur the costs of design protection. Introducing a grace period (as discussed below) will provide designers with a window in which they can test their prototypes with potential customers without losing eligibility for protection. The case is less clear with respect to situations where businesses are making copies of designs where design rights have already expired. There is little empirical evidence on the extent replica products reduce the financial returns to original designers. In some cases, suggestions that consumers who purchase replica items (such as furniture) would have otherwise bought the original, or mistake a replica for the original, seem implausible. For example, there are examples of original furniture designs costing thousands of dollars, at four to five times the cost of their replicas (Richardson 2014). It appears consumers are often making informed choices, trading off price and quality, and that reproductions outside of the term of protection are facilitating this increased choice. 344 INTELLECTUAL PROPERTY ARRANGEMENTS Lack of adaptability to changes in technology and industry practices Like all laws, designs law must be adaptable to changing circumstances to stay relevant and effective. Some commentators point out that copying made possible as a result of 3D printing means that a far broader range of products may be copied, and that this will likely test various areas of IP law, including designs law (Lennon, Eade and Smyth 2013). Increasing implementation and customisation of products through software presents a further potential challenge for designs law, which governments established well before the digital age. Some participants have questioned whether Australia’s designs system is adaptable enough to accommodate these changes, noting: • 3D printing and copying may give rise to gaps in the legal protection provided to owners of registered designs (Design Institute of Australia sub. 131; Gilbert and Tobin, sub. 96) • designs law currently ties design protection to physical products and therefore does not encompass ‘virtual’ or software designs (ACIP 2014). With regard to 3D printing and design protection, ACIP acknowledged there were limitations in the Designs Act, which might give rise to gaps in the legal protection provided to owners of registered designs. However, ACIP (2015a) went on to note that the majority of stakeholders believed that to undertake any action in relation to 3D printing would be premature, in part reflecting a lack of evidence that potential gaps in protection are problematic. Participants to this inquiry reiterated the potential implications of 3D printing (such as gaps in legal protection) but did not provide evidence that problems have come to pass (Design Institute of Australia, sub. 131; Gilbert and Tobin, sub. 96; The Institute of Patent and Trade Mark Attorneys of Australia, sub. 73). Recent studies in the United Kingdom on the implications of 3D printing on IP similarly suggest that evidence of a problem is yet to be found. A legal and empirical study of 3D printing online platforms and an analysis of user behaviour found that while the number of IP issues in relation to 3D printing will likely grow given the rise in online platforms, ‘at the moment [the problem] is not widespread and as such does not give rise to major concern’ (Mendis and Secchi 2015, p. 43). A companion study, which used case studies to examine the impact of 3D printing within the industrial sector, similarly found that ‘there is no immediate concern posed by the growth of industrial [additive manufacturing] or consumer 3D printing in relation to intellectual property’ (Reeves and Mendis 2015, p. 68). The Australian Government (2016b, pp. 8–9) accepted ACIP’s recommendation that no change be made to the designs system at this time to respond to 3D printing and scanning technologies, noting ‘it would be premature to take legislative action when there is no evidence of any existing problem’. However, it agreed to monitor technological developments and their implications for the designs system. The Commission supports this position. REGISTERED DESIGNS 345 Where to from here? A key goal of the IP system is to encourage the creation and dissemination of valuable ideas that would not have occurred in the absence of the system. The Commission does not consider that the ‘low’ uptake of design rights is necessarily a problem to the extent designers are able to use other strategies to obtain a sufficient financial return on their innovations. Further, there are other ways of addressing participant concerns about the costs of securing and enforcing rights, other than through specific changes to the registered designs system or the Designs Act (such as general changes to reduce the costs of enforcement). While the Australian Government can influence the costs associated with protecting IP (such as through streamlined regulatory arrangements) a large portion of the costs associated with protecting IP (such as legal fees) arise in the private sector and are not under the Government’s direct control. The Government should nonetheless identify and act on opportunities to improve the operation of the registered designs system, such as by eliminating unnecessary regulatory or administrative barriers to obtaining and enforcing registered design rights both in Australia and internationally (detailed below). 11.3 Reform options There are a number of options for improving the existing designs system. Given the recent release of ACIP’s review of the designs system and the broad nature of this inquiry, the Commission has focused on a selection of material issues. These include: • measures to reduce the cost of design protection • measures to address poor understanding of design laws • making design law more adaptable. Box 11.2 details ACIP’s recommendations and indicates the Australian Government’s response. Some of ACIP’s recommendations accepted by Government should address many of the participant concerns outlined above. For example, the Government has agreed to address poor understanding of the design law by introducing a grace period for design application filings and clarifying terminology for a registered but uncertified design (Australian Government 2016b). In other areas, the Government has committed to building the evidence base before committing to changes in design law. 346 INTELLECTUAL PROPERTY ARRANGEMENTS Box 11.2 ACIP Review of the Designs System: recommendations and Government response ACIP made 23 recommendations as part of the Review of the Designs System, including that the Australian Government: • promote efforts to harmonise and streamline procedures for acquiring international registered design rights (2a) and work actively through the Designs Law Treaty process to promote harmonisation of filing requirements (2c) (accepted by the Australian Government) • commence an investigation into the implications of joining the Hague Agreement and monitor usage of the Hague system (2b) (accepted) • extend the maximum term of protection of designs to 15 years only if a decision is made to join the Hague Agreement (3) (accepted) • change the terminology for a registered but uncertified design to make it clear that the design does not, until certification, confer enforceable rights (4) (accepted) • remove the option of the publication regime from the designs process (5) (accepted) • require a request for examination by the first renewal deadline (6) and (if rec 6 is accepted) introducing a system of opposition following certification (7) (noted) • improve the process for multiple design applications by reducing the fees for each additional design added to the application (8) (noted) • make publication automatic at six months after the filing date (9) (noted) • retain the requirement of distinctiveness and s. 19 in its current form (10) (accepted) • allow amendment of statement of newness and distinctiveness up until certification (11) (accepted in principle) • introduce a grace period of six months before the filing date, with prior user defence (12) (accepted, with consultation on length of grace period) • retain the requirement that a design be registered for the whole product, while investigating further whether allowing partial product registrations would substantially advantage Australian applicants and not give rise to substantial practical or legal issues (13) (accepted) • reconsider the treatment of virtual or non-physical designs (14) (accepted) • take steps to make s. 18 of the Designs Act consistent with the overlap provisions of the Copyright Act 1968 (15) (accepted in principle) • consider introducing border protection measures that align with the Trade Marks and Copyright Acts (Notice of Objection Schemes) to allow for the seizure by Customs of alleged design infringements which are identical to certified designs (16) (not accepted) • retain section 71 of the Designs Act in its present form (17) (accepted) • amend the legislation to address various specific anomalies (18), in part to improve the operation and consistency of the Act and in some cases, consistency with other IP systems (accepted) • make no changes to the trade mark/design overlap (19), the repair defence (20) or to respond to 3D printing and scanning technologies (21) and not introduce an unregistered design right (22) (accepted). Specifically include the role of the designs system in any broader review of Australia’s IP framework (23) (accepted) (met by this inquiry). REGISTERED DESIGNS 347 Reducing the cost of design protection through increased international harmonisation The most common rationale for Australia moving toward international harmonisation is that it would reduce administrative costs for Australian design-intensive businesses that currently, or plan to, export. ACIP (2014) suggested the fact that a third of all SMEs using registered designs also export means that any change to reduce red tape associated with exporting will have substantial benefits to Australian business. ACIP went on to note that greater harmonisation might also benefit Australian consumers. They argued international harmonisation would likely make it easier for nonresident businesses to obtain protection in Australia, which could lead to increased imports of design-intensive products to Australia. Increased imports of design-intensive products could in turn strengthen competition, benefiting consumers, and incentives for increased productivity among Australia’s design-intensive businesses (ACIP 2014). However, ACIP did not present evidence regarding the extent of these purported benefits to consumers. Greater harmonisation would involve trade-offs. Acceding to some international agreements would impose obligations on Australia regarding minimum requirements for the designs system and reduce the degree of flexibility that Australia has to amend its own IP policies. There would also be costs associated with changing IP Australia’s existing administrative systems to accommodate new requirements (ACIP 2014). The two main multilateral instruments concerned with harmonising and streamlining application processes for design rights are the Hague Agreement Concerning the International Registration of Industrial Designs (‘Hague Agreement’) and the Draft Design Law Treaty (DLT). Joining the Hague Agreement would mean extending the term of design protection for potentially small benefits The Hague Agreement is an international registration system, which offers the possibility of obtaining protection for industrial designs in a number of States and/or intergovernmental organisations (‘contracting parties’) through a single international application filed with the International Bureau of the World Intellectual Property Organization (WIPO) (WIPO 2012). It allows applicants from member countries to file one international application, to comply with one set of formalities, in one language, with one set of fees paid to WIPO (ACIP 2014). Membership of the Hague Agreement requires a minimum level of harmonisation on certain features of the designs system. If Australia joined the Hague Agreement, Australia 348 INTELLECTUAL PROPERTY ARRANGEMENTS would need to make changes to Australian law, most notably, the extension of Australia’s maximum term of protection from 10 to 15 years (ACIP 2015a). 2 The Hague Agreement has 65 members, including many European countries, Singapore, South Korea, Japan, and the United States (the latter two countries joined on 13 May 2015) (WIPO 2016c). The United Kingdom’s membership is in progress (ACIP 2015a; IPO 2016d). New Zealand, China, and India are not signatories to the Hague Agreement (WIPO 2016c). Australia is already committed to making ‘best efforts’ to join the Hague Agreement system under the Australia-US Free Trade Agreement and the Singapore-Australia Free Trade Agreement, although there is no time limit for that to occur (ACIP 2015a). Despite many participants to the review of the designs system supporting Australia joining the Hague Agreement, ACIP recommended against joining the Hague Agreement immediately, due to questions about the benefits it would actually deliver and the requirements of the Hague Agreement membership (box 11.3). Specific issues included: • The Hague system was not much used (at the time ACIP cited that there were 3000 Hague system applications in 2013, covering about 13 000 designs. This represented 66 000 designs across multiple designations). 3 • Requirements for drawings and written descriptions vary considerably between Hague Agreement countries, making the process complicated to navigate (an application under the Hague system may, for example, have to include drawings suitable for every country where design protection is sought). ACIP (2015a, p. 17) noted the concern ‘that any gains arising from the single application process would be largely illusory for most applicants due to the need to comply with multiple different requirements’. • Transfer of ownership of designs registered under the Hague system is restricted to companies resident in participating member states. • The Hague Agreement would require system changes within IP Australia, the costs of which are presently unknown and cannot be ascertained without engaging in a detailed formal investigation with WIPO (ACIP 2015a). The Australian Federation of Intellectual Property Attorneys (FICPI, sub. DR581) noted that lack of harmonisation in filing requirements may also result in filing errors which cannot be corrected later, resulting in irrecoverable loss of rights. 2 It is possible for an applicant whose country is not a Hague member to file an application under the Hague system if they ‘have a real and effective industrial or commercial establishment in the jurisdiction’ (WIPO 2015b, p. 14). Very few Australians have filed applications through the Hague System (in 2014, there were two Hague system applications from Australia) (WIPO Statistical Country Profiles: Australia). 3 A designation is a request in an international application or registration for protection in a Hague member’s jurisdiction. REGISTERED DESIGNS 349 ACIP recommended closely monitoring usage of the Hague system, particularly with respect to recent members, and investigating the implications of joining the Hague Agreement more fully: A significant uplift in international usage would support Australia joining the Agreement. Additionally, a detailed investigation into the requirements for Hague membership should be undertaken. This would accurately inform the specific requirements, costs, limitations and opportunities for both applicants and IP Australia. It could also assist the preparation phase, thereby shortening the work needed to join at a later date. (2015a, p. 17) The Australian Government agreed to ACIP’s recommendation, noting IP Australia would be charged with these tasks (Australian Government 2016b). Box 11.3 Measuring the benefits to designers from joining the Hague Agreement The Hague Agreement is expected to provide a range of cost savings including: • Reduced filing fees: Under the Hague system, each international application attracts a fixed basic registration fee of 397 Swiss Francs, a publication fee for each reproduction to be published (17 Swiss francs), and a designation fee for each country (region) included in the application (which vary depending on the level of examination). Each additional design in the application attracts a fee of 19 Swiss francs. Filing applications through the Hague system generally becomes more cost-effective the greater the number of Hague member countries designated in a single application (as the average fixed fee per jurisdiction declines). • Reduced maintenance fees: The Hague system may reduce the cost of managing multiple design rights as it offers centralised payment of renewal fees. The UK IPO’s (2012) regulatory impact assessment of the United Kingdom joining the Hague system cites anecdotal evidence that national filing maintenance for six countries would be £1500 plus £500 for each national agent, adding up to £7500, but that the same set of registrations through the Hague would cost £500. • Reduced translation and other fees: the Hague system allows applicants to bypass each jurisdiction’s IP office to initiate the filing process and to avoid translation and other fees, which may result in costs savings. In locations where national or regional offices do not substantively examine applications, local counsel fees can be avoided entirely. Not all designers seeking protection overseas will realise these cost savings. Official fees under the Hague system are not necessarily less than the direct route, particularly if designers are seeking protection in a limited number of Hague member countries (Stockton 2015), which might be the case for many SMEs. In many cases, applicants may still incur local counsel fees as local counsel will be required to deal with local formality issues and to file in jurisdictions that have not joined the Hague Agreement. Rauch (2015) notes, while cost-savings may be realised from avoiding translation costs, unlike patents, design applications have little text to translate. Further, applicants that forgo local legal advice regarding local nuances may end up paying a much higher price in terms of lost protection. While an analysis of Australia joining the Hague Agreement would need to consider local circumstances and recent expansion of Hague members, the UK IPO (2012) estimate of the savings from the UK joining the Hague Agreement (which ranged from £0 to £250 000, with a best estimate of £90 000) provides some perspective to the limited scale of the benefits. 350 INTELLECTUAL PROPERTY ARRANGEMENTS A wait and be convinced approach to joining the Hague Agreement remains appropriate Since ACIP’s review, the United States and Japan have become signatories to the Hague Agreement and there has been an increase in filings under the Hague system. In 2015, there was approximately 4100 international design applications filed under the Hague system, covering about 16 400 designs and representing 74 200 designs across multiple designations (WIPO 2016j) (figure 11.6). Notwithstanding developments since the ACIP review, a ‘wait and be convinced approach’ to joining the Hague Agreement remains appropriate, particularly given the decision to join the Hague Agreement is not easily reversed and IP Australia’s investigation into the implications of joining the Hague Agreement has not yet been undertaken. Before signing up, a net benefit case would need to be transparently made The Designs Act provides protection of up to 10 years, which is the minimum requirement contained in the Agreement on Trade Related Aspects of Intellectual Property Rights (TRIPS). Australia would be obliged to provide a maximum term of at least 15 years if it joins the Hague Agreement. Participants have argued that extending the term of design protection from 10 to 15 years is desirable because it would promote consistency with international practices or because the longer protection would be attractive to some industries (ACIP 2014; FICPI, sub. DR581; Swanton, sub. DR287). Some have also suggested that the cost on consumers from extending protection from 10 to 15 years would likely be small because the scope of protection that design rights afford is narrow relative to patents, and only a small number of registered design owners renew their designs to receive the maximum possible term. IPTA (sub. DR562) noted that the term of protection of registered designs in Australia at only ten years is the shortest term of any of Australia’s major trading partners and considerably shorter than that of Europe which offers a 25-year term. FICPI (sub. DR581) argued that the system is self-regulating as most designs have a short product life, but provision needs to be made to protect highly valuable iconic designs with a much longer product life. In assessing the case for extending the term of design protection, ACIP (2015a, p. 18) stated that it did not receive ‘any economic or empirical evidence suggesting that such an extension is necessary’. ACIP noted that an extension of the maximum term should only be recommended if there is a reason to do so — either as a result of evidence that rights are needed and/or likely to act as an incentive, or as a result of international obligations undertaken to receive some other benefit. ACIP therefore recommended extension of the design term to 15 years only if Australia decides to join the Hague Agreement. The main objective of design protection is to encourage additional innovation that provides net benefits to Australia. The Commission, and ACIP before it, found no evidence for REGISTERED DESIGNS 351 extending the term of design protection as necessary to encourage innovation. 4 This indicates that an extension of exclusive rights is not warranted. Promoting international consistency in design protection is a poor reason for extending the term of design protection in Australia from 10 to 15 years. It assumes that other counties have struck the right balance between incentives for innovation and the interests of consumers and other businesses. The optimal IP policy setting will likely vary between countries according to their economic circumstances. Again, the aim is to provide only as much protection as is needed to encourage additional innovation. Figure 11.6 Design applications by filing route: direct and Hague system Design countsa 1400 1200 Thousands 1000 800 600 400 200 0 2004 2005 2006 2007 2008 Direct 2009 2010 2011 2012 2013 2014 2015 via Hague System a 2015 data not currently available for direct route. Source: WIPO statistical database. Moreover, the wider effects on the Australian community of joining the Hague Agreement have not been fully explored (box 11.4). Reversing such commitments, should they prove less beneficial than anticipated, presents a number of difficulties and would likely cause disruption to applicants. 4 Renewal rates for classes of designs with sometimes iconic designs (such as furniture) are not substantially higher than those for other classes (IPGOD, 2016 edition). 352 INTELLECTUAL PROPERTY ARRANGEMENTS Box 11.4 Measuring the costs to Australia from extending term There has been little public analysis of the impacts of joining the Hague Agreement on the wider community, including consumers. This partly reflects that some countries that have decided to join the Hague Agreement already had a longer maximum term of protection than the 15 years required under the Hague Agreement (for example, the United Kingdom has 25 years), and so did not need to consider the costs associated with extending the term of design protection as part of their assessment. Some Australian studies have estimated the effects of extending the term of protection for other forms of IP rights (such as copyright and patents) through changes in Australia’s balance of payments (Dee 2004; Gruen, Bruce and Prior 1996). An increase in the term of protection in Australia will tend to increase payments by non-resident IP users to Australian rights holders, while increasing payments by Australian IP users to non-residents who hold Australian IP rights. The effect on Australia as a whole will depend on the net effect of these income flows. An increase in the term of protection in Australia will also have distributional impacts, increasing payments from Australian IP users to Australian rights holders. Under the Designs Act, the Australian Government reduced the term of registered protection from 16 years to 10 years because ‘it would not be in Australia’s interest to provide a period of registration in excess of its international obligations as Australia is a net importer of intellectual property’. As noted in section 11.2, non-residents account for the majority of design applications. Hence, it is reasonable to expect Australia remains a net importer of design IP. Measuring the effects of extending the terms of design protection in Australia is likely to be challenging, given that relevant data (such as the income from licence fees) is more limited for registered designs than patents and copyright. For example, balance of payments data on charges for the use of IP does not separately report charges for the use of design rights. Moreover, collecting detailed data for this purpose may well prove costly and/or infeasible. Notwithstanding these practical challenges, the potential costs of extending the term of design protection should be assessed both qualitatively and quantitatively, albeit partially for the latter. Research to establish a range of plausible estimates of net income flows related to design rights might only need to put an order of magnitude on the costs of extending protection, to establish whether joining the Hague Agreement would be beneficial. Transparent cost-benefit analyses of extending term could also shed light on implementation issues. For example: • past studies show that applying an extension of term retroactively to existing rights (as opposed to only new rights) can substantially increase the costs of the policy (Gruen, Bruce and Prior 1996) • as noted by ACIP (2015a) and submissions to this inquiry, an extension of term could be accompanied by measures such as mandatory examination before the first renewal and/or an increase in the renewal fee at the 10 year stage with a view to providing an incentive to renew only those registrations having sufficient economic value (AIPPI, sub. DR551; Law Council of Australia, sub. DR490). IP Australia is still determining other effects, such as the cost of upgrading IT systems. REGISTERED DESIGNS 353 In the Draft Report, the Commission argued that before Australia commits to joining the Hague Agreement, the Australian Government should assess the costs and benefits of doing so, to ensure the interests of Australian as a whole (as opposed to only right holders) are adequately considered and to ensure that the gains from ‘harmonisation’ outweigh the costs of extending term. While some participants questioned the need for such an assessment (AIPPI, sub. DR551; Arts Law, sub. DR536), IP Australia has noted: IP Australia notes the draft recommendation that any decision to join the Hague Agreement should be based on a comprehensive cost-benefit analysis. ACIP recommended that the Government monitor and only consider joining the Hague Agreement after a comprehensive cost-benefit analysis in its Review of the Designs System. The Government has recently accepted this recommendation, and IP Australia has commenced research into the costs and benefits of joining this agreement. (sub. DR612, p. 15) IP Australia has advised that it has commenced research on the costs and benefits of joining the Hague Agreement and expects this research to be completed in 2017, noting that the approach being taken is designed to be repeatable as circumstances change (if the costs outweigh the benefits in the initial assessment) (pers. comm., 12 September 2016). This is a positive step. The issues and trade-offs in joining the Hague Agreement for Australia identified by the Commission (and outlined in boxes 11.3 and 11.4) will need to be assessed as part of that cost-benefit analysis. The Draft Designs Law Treaty and other harmonisation measures The draft DLT aims to enable easier access to the design protection system through the simplification of registration formalities, and provides a ‘ceiling’ on what countries may demand in the registration of design rights (ACIP 2014). In contrast to the Hague Agreement, which is a global registration treaty, the draft DLT is an administrative treaty that standardises the procedural aspects of designs registration among Contracting Parties. In October 2015, the WIPO General Assembly noted that the basic proposal for the DLT should be finalised in early 2016, and agreed to convene a diplomatic conference for the adoption of a DLT at the end of the first half of 2017. But a consensus on the basic proposal failed to emerge and commentators have questioned whether WIPO will be able to find a solution in order to convene a conference in 2017 (Ashton and Vekris 2016). If Australia acceded to the draft DLT in its present form it would need to institute a grace period of 6 or 12 months (WIPO 2016g, 2016h). A grace period provides a way for designers to disclose designs before filing a design registration, or protect against inadvertent disclosure. Most other jurisdictions have a grace period, which range from 6 to 12 months in duration. The rationale for aligning grace periods is to address the concern that the existence of different grace periods, and the absence of grace periods in some jurisdictions (including Australia), may cause an applicant to lose the possibility of obtaining protection for the industrial design abroad (WIPO 2016h). 354 INTELLECTUAL PROPERTY ARRANGEMENTS Grace periods have their disadvantages. Grace periods can prejudice third parties who act on the basis of a disclosure or a search of the designs register and are unaware that a valid IP right may subsequently be filed (ACIP 2015a). To the extent Australia elects to adopt a similar grace period for reasons independent of any harmonisation benefits (see below), this requirement will not be a binding constraint. ACIP recommended that Australia remain actively involved in the draft DLT process on the basis that Australian stakeholders are in favour of harmonisation of filing requirements internationally. It also suggested IP Australia identify ways to work with its international counterparts on ways to harmonise practices that the Hague Agreement and draft DLT do not address. These include addressing differences between the requirements of local IP Offices for claims, written descriptions, splitting of design applications and whether multiples are allowed, and requirements for drawings (ACIP 2015a). The Australian Government agreed to ACIP’s recommendations noting that ‘IP Australia will continue its active involvement in promoting harmonisation of international filing requirements through development of the Designs Law Treaty’ and ‘continue its efforts to encourage international harmonisation and streamlining of registered design systems in different jurisdictions’ (2016b, p. 1). The Commission supports efforts to encourage international harmonisation and streamlining of registered designs systems in different jurisdictions, where it is in Australia’s interests. The issue of how to weigh the benefits of increasing harmonisation though international agreements and treaties against the costs is common to all other areas of IP. Chapter 17 examines measures Australia should take to ensure consideration of such trade-offs occurs in an accountable manner and to reduce the risk of unintended consequences. Other measures to reduce the costs of acquiring design protection Excessive costs for acquiring design protection (such as filing fees) can discourage socially beneficial innovations. Conversely, very low costs can encourage activity with a low social value or strategic use of IP rights. During ACIP’s review of the designs system, and this inquiry, some participants suggested that the cost of acquiring design rights was too high and proposed a range of measures to reduce such costs (including reduced fees for applications that include multiple designs and introducing an unregistered designs system). Design fees are already partially subsidised by other IP rights holders (discussed below). The Commission does not consider that there are grounds for further cross-subsidisation of design fees. Hence in looking to lower costs, the issue is whether there are possible administrative efficiencies that could be better realised, or net benefits from a restructuring of design filing fees. REGISTERED DESIGNS 355 Reduced fees for multiple design applications Under the Designs Act, applicants are able to include multiple designs in a single application. This option was introduced to reduce the cost of filing applications and benefit industries which produce a large number of designs for the same product. The ALRC (1995) recommended that multiple design applications should be allowed and that each additional design included in a multiple application should be charged a reduced fee. In practice, filing a multiple design application does not result in cost savings because the fee for each additional design in a multiple design application is the same as the fee for an application for a single design ($250 using online services). A separate renewal fee applies to each design in an application. ACIP (2015a) concluded that the current fee structure may discourage use of multiple design applications, and may mean that the ALRC’s intention of making the system more accessible to industries with high design turnover is not being achieved. This view is consistent with the views of several participants to this inquiry, who identified the cost of registering multiple designs in Australia as an issue: The current fees for design registration I think would be one of the first ports to address. … I think the ACIP acknowledged that the current fees for multiple design registrations is turning people away from offering to do it. Currently each registration is an independent fee per product. If you design register a product which has a variation, say an armchair versus a lounge, you must register them. … If you have multiple products, there’s no incentive to get a cost reduction to do that. (Fereday, trans., pp. 390–391) ACIP (2015a) recommended improving the process for multiple design applications by reducing the fees for each additional design added to the application consistent with the ALRC’s original proposal. ACIP recognised that the recommendation would have a significant impact on IP Australia, which would need to make the system changes necessary to accept reduced fees and accommodate alternate numbering conventions. In May 2016, the Australian Government noted ACIP’s recommendation but indicated that it considered it appropriate to consider this recommendation further in the context of IP Australia’s fee review to be completed in 2016. IP Australia has since advised that the 2016 fee review did not include a consideration of reducing the fees for each additional design added to the application because the Government’s response to the ACIP review of the designs system was not approved and released until late in the fee review process (pers. comm., 30 August 2016). The Australian Design Alliance subsequently noted: IP Australia’s current fee review 2016 has offered no reduction in fees. This creates a barrier particularly for startups and SMEs. In addition to those companies launching a range of products. If the financial barrier to entering the design registration process was reduced, then greater revenue could be recovered at the examination stage … the 5-year mark when examination and filing is requested and designers have a better understanding of their designs financial performance. (sub. DR619, p. 4) 356 INTELLECTUAL PROPERTY ARRANGEMENTS Other jurisdictions have also sought to make filing applications with multiple designs more cost-effective. In the United Kingdom (which permits multiple designs in a single application), the Government recently agreed to reduce filing fees and renewal fees, with proposed changes expected to come into effect in October 2016. Under the new fee structure the cost of filing multiple designs will come down significantly. For example, an application containing one design will cost £50 and an application containing up to 10 designs will cost £70. Under the old fee structure, a single design application cost £60 and an application containing up to 10 designs cost £420 (UK IPO 2016). The Commission considers that fees for design rights should (at least) cover costs — this is consistent with the Australian Government’s Cost Recovery Guidelines. However, the guidelines also indicate that cost recovery fees, such as registration charges, should ‘reflect efficient unit costs of a specific good or service’ (Department of Finance 2014, p. 3). To the extent there are economies of scale in processing applications with multiple designs, then reducing fees for applications with multiple designs may be consistent with cost-reflective pricing. However, IP Australia has advised that it does not expect that there would be significant efficiency gains in processing an application with multiple designs because: … Even if separate searches are not required to examine an application with multiple designs, at the formalities stage an examiner must consider each design on its own merits (even if it is various embodiments of the same design). If there is an issue with one of the designs in an application, all of the other designs in the application are also held up while this issue is resolved. Extra administration and handling is therefore required to maintain the pending designs while the resolution process occurs (and separately register each of the designs of the application should this be required). (IP Australia, pers. comm., 30 August 2016) IP Australia further noted: Importantly, the cost of designs examination is subsidised by revenue from trade mark examination, as agreed with the Department of Finance and included within the recently approved Cost Recovery Implementation Statement. IP Australia obtained agreement to cost recover trade marks and designs in combination, rather than separately, in order to keep designs costs lower while not imposing a significant impost on trade marks customers. This is because the volume of design applications is relatively small when compared to the volume of trade mark applications. As such there are economies of scale for designs customers, as overhead costs are spread across all of the total business transactions, not just the designs transactions. If designs was instead operated as a completely separate business unit, the overhead costs per transaction would be higher. A new fee reduction may result in a net loss to IP Australia, which operates under a cost recovery model. Any changes in fee structures would also need to be considered carefully in the domestic and international context, including any impact that fee structure changes would have on applicant behaviour. (IP Australia, pers. comm., 30 August 2016) REGISTERED DESIGNS 357 Given the lack of economies of scale in processing applications with multiple designs and that design fees are already partly subsidised by trade mark applicants, there are not strong grounds for reduced fees for multiple design applications. The fact that the majority of design applicants are overseas companies or individuals adds further weight to the argument. Allowing the registration of design variants for a single fee Some participants suggested that the ability to register variants of a design or product ranges, and only pay a single application fee, would help reduce the cost of protecting designs. Some suggested that it is possible to register design variants in the same application for a single fee in Canada. 5 Bicycle accessories company, Knog, noted ‘we decided we’d register everything in Canada to begin with because we can register a range or variations on a theme’ (trans., p. 539). In Australia, the design examiners’ practice guidelines note that, generally, ‘variations’ and ‘alternate embodiments’ are too different to be considered one design (IP Australia 2016b). Each separate design attracts a separate fee. There are already provisions under the Designs Act to help protect product ranges. Applicants can apply to register a ‘common design’ in relation to more than one product and the application is treated as a single design and only one fee is payable (Crump 2014; IP Australia 2016c). 6 Examples of products which often have the same design include bedside tables, chests of drawers, dressing tables and wardrobes (shape of surfaces, handles, edges and legs) (IP Australia 2016b). Further, the ability to register variants of a design will not result in cost savings for designers who are seeking to register a range of product designs that are substantially different. For example, in Canada, to be accepted as variants, designs must be applied to the same article or set and not differ substantially from one another (CIPO 2015). If an application does not meet the requirements for variants, it is subject to a divisional application for the additional design. Around half of design applicants who claim variants in their design in Canada are subject to divisional applications because they do not meet the requirements under the Industrial Design Act 1985. The potentially limited benefits from allowing the registration of design variants for a single fee would need to be considered in light of the effects on the complexity of the designs application and examination process and resultant administrative costs. For example, the Commission understands that, in Canada, assessing variants requires 5 Variants are in essence already multiple designs in one application. The only difference is that these designs are deemed not substantially different and only one examination fee is charged. 6 For common designs, the design must be shown as applied to more than one product. A chair and a chair with arms are not two different products. The arms of a chair are simply visual features of one product, being a chair (IP Australia 2016a). 358 INTELLECTUAL PROPERTY ARRANGEMENTS examiners to spend significant time and effort assessing whether designs qualify as variants (for example, the drawings submitted are different but substantially similar designs applied to the same finished article) and the cost of this activity is not reflected in the price of the single examination fee. As noted, the costs of administering the designs system in Australia is already cross-subsidised by fees for other IP rights. The benefits would also need to be weighed against the potential to unintentionally expand the scope of rights afforded. For example, one commentator has noted common design protection is arguably broader than registering each of the individual products separately. According to the Examiner’s Manual the features of ‘the design’ are those features that are in common between the representations of the various products and the features of the products are those features that differ between the representations of the various products. The product to which the design is applied is of secondary importance to the design. Whilst it has not been interpreted by the Courts yet, this implies that the features of the products that are not part of the common ‘design’ are given less weight. (Crump 2014, p. 6) The Commission has not been able to identify a net benefit case for allowing the registration of design variants for a single fee. Unregistered design rights In 1995, the ALRC considered the option of supplementing or replacing the registered design system with an unregistered design right system, noting: There are particularly powerful arguments in favour of … giving designers the right to prevent others from copying their designs for a certain period (say, five years) without requiring them to register their design or satisfy an innovation threshold. (1995, s. 3.63) The ALRC (1995) suggested that an unregistered design right (UDR) would: • directly address the major problem faced by designers — copying • give designers protection without them incurring the costs and delay involved in a registration system (especially for products with a short commercial life) • be less legalistic than a registered system • impose a lower barrier on competition because another design will only infringe the anti-copying right where it has been copied, not where it has been independently created but happens to be the same or substantially the same product • be more likely to match developments in industrial design activity (such as increasing emphasis on rapid, interdisciplinary and interactive design). The ALRC’s argument was largely in principle, rather than based on empirical analysis. The ALRC (1995) noted that there was relatively little support for the introduction of a UDR, that enforcement would be uncertain due to the need to prove copying, and that introduction of a generalised ‘anti-copying’ right raised issues well beyond designs law. The ALRC therefore recommended a broader review of the idea. REGISTERED DESIGNS 359 ACIP (2015a) re-examined the option of UDRs both as a complement to registered rights (as is the case in the United Kingdom and the European Union) and as a replacement for registered design rights (although much of the focus was on the former). ACIP (2013b, 2014) noted that means of protecting unregistered designs are likely to be of particular interest to fast-moving design industries (such as the textile, clothing and footwear industries) as they avoid the cost of obtaining design protection. Several participants to this inquiry reiterated their support for a UDR (anti-copying right) system to complement Australia’s registered designs system (Australian Small Business and Family Enterprise Ombudsman, sub. DR403; Gilbert and Tobin, sub. DR565; Law Council of Australia, sub. DR490). It appears to be generally accepted that the costs of registering and certifying designs makes the registered designs regime unsuitable for industries with a high volume and turnover of designs and that designers’ understanding of the designs registration system is generally low. UDR protection would assist designers falling into these two categories. (Law Council of Australia, sub. DR490, p. 24) We believe that the IP system could be further improved to better address the needs of small business by: … Establishing a system for the protection of unregistered small business designs, ideally arising automatically (like copyright). (Australian Small Business and Family Enterprise Ombudsman, sub. DR403, p. 1, 3) Notwithstanding that a number of stakeholders have expressed support for a UDR regime in combination with registered rights, ACIP noted that overall there was significant opposition to the idea of a UDR. This opposition was due to concerns that a UDR would: • undermine incentives to register designs, which would be a cost to the public which may benefit from the existence of an informative register • impose costs on business, in the form of increased difficulty in establishing their freedom to operate (ACIP 2015a). ACIP (2014) noted even if a right is time‐limited the uncertainty created would not be so limited, since it may not be straightforward for an intending copier to determine when a product was first marketed. ACIP (2014) also noted that designers relying on UDRs can have difficulty enforcing their rights (due to a higher evidentiary burden) and that the introduction of a new legal system would increase the complexity of legal advice on rights. ACIP (2015a) ultimately recommended that Australia should not introduce protection for unregistered designs and the Australian Government accepted this recommendation. The Singapore Ministry of Law and IPO (2016) recently recommended not introducing unregistered design protection in Singapore for similar reasons. Some participants questioned the degree to which an anti-copying UDR would create uncertainty because people will generally be aware that they are copying another’s product (Law Council of Australia, sub. DR490; Gilbert and Tobin, sub. DR565). The Law Council of Australia (sub. DR490) proposed a detailed inquiry to determine whether Australia should introduce a UDR somewhat similar to such a right existing in the United Kingdom. 360 INTELLECTUAL PROPERTY ARRANGEMENTS But the case for adopting the United Kingdom model is not strong. The UK IPO Assessment of the Need for Reform of the Design Intellectual Property Framework noted the following in respect to UDRs: We have no evidence … for whether the balance is tilted in favour of design right owners or new market entrants or vice versa, and therefore currently have no insight into whether UK unregistered design right is good or bad for the UK economy. (2011, p. 10) The Commission considers that ACIP’s recommendation not to introduce protection for unregistered designs remains valid. As is the case with copyright, unregistered rights that automatically apply would be a blunt instrument for encouraging socially beneficial innovation, and adopting an unregistered design system risks importing the multitude of drawbacks embedded in the (unregistered) copyright system to the designs system (chapter 4). Protect designs under copyright law Some participants to this inquiry argued that Australia should provide full copyright protection to designs for a term on a par with other artistic works, noting that the United Kingdom now provides copyright protection for industrially produced artistic works for the life of the designer plus 70 years. As two participants noted: Copyright is something that allows protection without the requirement for registration and without having to pay fees. It’s been used by authors who write books, musicians who write songs, and all kinds of other – architects that design buildings, and I don’t understand why designers don’t fall under the same rules in Australia. So that’s something that I consider as an alternative. As far as I understand it’s actually the extension of copyright to design in the UK that has extended and created those new conditions in that market. (Archer, trans., p. 284) Design protection should automatically come under copyright law and not be a paid application process separate to copyright. (Terri Winter, sub. DR198, p. 1) The UK reforms to copyright in question involved the repeal of s. 52 of the Copyright, Designs and Patents Act 1988, which contained an exception which limited copyright protection to 25 years for certain artistic works when they have been industrially manufactured (box 11.5). Twenty five years is also the maximum term of protection for registered designs in the United Kingdom. REGISTERED DESIGNS 361 Box 11.5 Repeal of section 52 of the United Kingdom Copyright, Designs and Patents Act 1988 Section 52 of the UK Copyright, Designs and Patents Act 1988 (CDPA) contained an exception which limited copyright protection for certain artistic works to 25 years when they had been industrially manufactured (more than 50 copies). In May 2012, the UK IPO announced the UK Government’s plans to repeal s. 52 of the CDPA. Following consultation on transitional arrangements, the repeal of s. 52 came into force on 28 July 2016. From this date, all types of artistic works were granted copyright protection for the life of the creator plus 70 years, this included industrially manufactured works which had previously been limited to 25 years under s. 52 of the CDPA. It is against the law to make or import new copies of artistic works unless they were contracted before 28 October 2015, permission is granted by the rights holder, or an exception applies. From 28 January 2017, it will be against the law to deal with any replicas or unauthorised copies made in reliance on s. 52 of the CDPA. By this date, all of these items must be depleted (sold or destroyed), have permission from the rights holder to continue their trading or rely on an exception. For a design to be protected by copyright in the United Kingdom as an ‘artistic work’ (as opposed to a literary, dramatic or musical work), it must fall into a category in s. 4 of the CDPA: • a graphic work, photograph, sculpture or collage, irrespective of artistic quality • a work of architecture being a building or a model for a building • a work of artistic craftsmanship. The UK IPO notes that works most likely to be affected by the repeal of s. 52 of the CDPA are ‘works of artistic craftsmanship’. There is no statutory definition of a work of artistic craftsmanship. Consequently, it is up to the UK courts to decide what would be classified as a work of artistic craftsmanship (protected by copyright law). The UK IPO has noted the limited UK case law suggests it is not enough for a work (such as a piece of furniture) to look attractive to qualify as a work of artistic craftsmanship and that the phrase ‘artistic craftsmanship’ designates that a work must be both of artistic quality (determined in light of evidence) and a work of craftsmanship (which presupposes special training, skill and knowledge for production). Following the change in law, artistic works that had been industrially made but whose 25 year term of copyright had expired under s. 52, will resume the remainder of the complete lifetime plus 70 year term at the point at which the old 25 year term had expired. Take this example: person “W” created an artistic work in 1980. In the same year, “W” manufactured 51 copies and sold them. “W” died in 2010. Under section 52, this work would only be protected by copyright until 2005. Following the change in law this work would be protected until 2080. (IPO 2016c, p. 10) One commentator has suggested that replica furniture and limited-run designer products could be particularly affected by the repeal of s. 52 (Lomas 2016). Sources: Lomas (2016); UK IPO (2016c). 362 INTELLECTUAL PROPERTY ARRANGEMENTS Many commentators have been critical of the lack of evidence used to support the extension of copyright to industrially produced artistic works in the United Kingdom. For example, Bently was highly critical of how the impact assessments for the repeal of section 52 greatly underestimated the cost to consumers and follow on users of IP: In approaching the question of cumulation of design protection with copyright, neither the ECJ [European Court of Justice] nor the UK IPO appears interested in matters of policy. The Court is focussed on harmonization as an end in itself and the IPO on compliance with its understanding of the Flos judgment. Reading between the lines, the UK Government has been threatened by lobbyists and the reform of section 52 is driven by fear of liability. The Impact Assessment says that there is no need for consultation, and regards it as ‘proportionate’ to advocate a policy based on snippets of ‘evidence’ gleaned from previous submissions to it from lobbyists in the furniture industry and their compatriots. The IPO admits that the reform of section 52 will harm consumer welfare as ‘classic designs’ – those that are more than 25 years old will be re-monopolised. Replicas, currently available at 15% the price of the ‘original’ will no longer be available. But no opportunity has been given for consumers to be consulted – though it is not obvious who would represent those interests. Moreover, the Impact Assessment significantly underestimates the other costs that will arise, partly because of its focus on furniture and three-dimensional design (when section 52’s existing role also importantly extends to the exploitation of two-dimensional designs), and because it fails to acknowledge the section’s immunising effects on certain secondary uses (which formerly were permissible because subsection 2 added that ‘anything may be done in relation to articles so made, without infringing copyright in the work’). In short, the repeal of section 52 will create a whole host of new situations where permissions are required (and rightholders are difficult to locate). While these costs are evident, the public benefits from the extension are non-existent or speculative: in terms of incentives, there is no reason to think that design investment decisions are ever based on returns from exploitation more than 25 years into the future; and clearly, there is no benefit from lengthening the copyright protection afforded in relation to designs that already exist. The statements of the Minister that the change would ‘encourage innovation and investment in design’ are supported by the flimsiest of argument in the Impact Assessment. (2012, pp. 42–43) More recently, other commentators have noted concerns about the unintended consequences the new copyright laws may have on the ability of individuals who use 3D printing and other technologies to manufacture designs in the United Kingdom. Furniture is normally protected by something known as a design patent and not by copyright, and this has enormous ramifications for 3D printing: when something is under patent, you’re absolutely and one hundred percent free to make copies of it for your own use with your own tools and materials. When something is under copyright, you are not. Therefore, this move is a direct assault on the 3D printing revolution. (Falkvinge 2016) During the course of this inquiry, it seemed that at least some participants were under the impression that the UK copyright protection for industrially applied artistic works will have a broader application than might actually be the case. Ford (2016) notes the definition of artistic work in the United Kingdom is relatively narrow and many industrially produced REGISTERED DESIGNS 363 items (for example, parts for cars or items of furniture) may effectively be excluded from protection. The UK IPO has similarly noted: Some copyright experts say that the more constrained the designer is by functional considerations, the less likely the work is to be a work of artistic craftsmanship. A work designed to be mass-produced (rather than designed as a one-off or limited run and then copied multiple times) can be a work of artistic craftsmanship as a matter of principle, although designing for mass production may cast doubt on whether it is truly one of artistic craftsmanship. (IPO 2016c, p. 9) In Australia, the Copyright Act contains no definition of works of artistic craftsmanship and a precise definition of the term has not been settled by case law. However, the work must be of artistic quality and involve craftsmanship (Arts Law Centre of Australia 2016a). As noted above, there is no evidence demonstrating an additional five years of protection is required for designers to obtain a sufficient return on their investments (let alone the additional term of period that would apply if copyright was available to industrial designs). As acknowledged by ACIP (2014, p. 30), the ‘low innovation threshold required to qualify for copyright protection, the lengthy duration of such protection and the absence of any public register to record such protection have, by successive governments, been considered unsuitable for mass-produced items’. Moreover, the Commission has identified a range of flaws with the design of Australia’s copyright arrangements, which cannot be remedied by the Australian Government alone, due to international obligations (chapters 4 and 5). Extending copyright protection to designs would compound these problems. Measures to address poor public understanding of designs law ACIP’s review of the designs system, and participants to this inquiry, identified a number of options to address poor understanding or confusion in designs law. These include: • introducing a grace period for filings • enhancing awareness about designs law and practices. Introducing a grace period for filing designs One of the main arguments for introducing a grace period for filings is to protect designers who through ignorance or inadvertence publish their design before seeking legal protection. A 2014 survey of people engaged with the design registration system provides partial evidence that some loss of protection occurs in Australia through inadvertent disclosure (figure 11.7). ACIP suggested a grace period might be particularly important for SMEs who have less interaction with the IP system and may be less aware of the requirement not to disclose before filing (ACIP 2015a). Gilbert and Tobin (sub. DR565) noted that their experience corresponds with the ACIP survey results. 364 INTELLECTUAL PROPERTY ARRANGEMENTS Inadvertent disclosure of designs Figure 11.7 ACIP review survey results, in response to: ‘We often inadvertently disclose our designs to the public before we file for protection’a 60 56.2 50 per cent 40 30 20 14.1 9.1 10 0 9.4 4 1 2 2.7 3 4 5 Rating (1= disagree, 7 = agree) 6 4.4 7 a Sample = 326. Respondents were people engaged with the registration system. Source: ACIP (2014). A limited grace period would also enable designers to engage in some degree of market testing prior to deciding whether or not to incur the expenses associated with registering a design (Gilbert and Tobin, sub. DR565), particularly where they have multiple design prototypes. One participant noted: Conventionally what we do is we’ll have a few new designs which we will show at trade shows in Europe and America to get feedback before we invest any further in them. (Trubridge, trans., p. 408) The Commission heard from several designers that advocated a grace period for this reason (Goodrum, p. 393; Trubridge, sub. DR573; Fereday, trans., p. 390; Skeehan, trans., p. 388). For example, Tom Fereday noted ‘what a grace period allows for is a period to test to actually work [out] if you can actually afford to develop this product’ (trans., p. 390). Adam Goodrum noted a ‘grace period in the time that you’ve got to register your design would be very, very advantageous’ (trans., p. 393). As noted, grace periods have the drawback of potentially prejudicing third parties who act based on a disclosure and/or a search of the designs register and are unaware that someone else may subsequently file a valid IP right. A ‘prior user’ rule could help address this issue by allowing a third party to continue to use a design where that use commenced prior to the filing date of a registered design (ACIP 2015a). REGISTERED DESIGNS 365 ACIP recommended introducing a grace period of six months before the filing date, together with a prior user defence. ACIP suggested that limiting the grace period to six months (which is less than the 12 months grace period under patents law and for designs in some other jurisdictions) was appropriate to reflect the fact that the focus is on avoiding problems of inadvertent disclosure and minimising the period of uncertainty for third parties. In May 2016, the Australian Government agreed to introduce a grace period together with a prior user defence. The length of the grace period and requirement for an applicant to declare any disclosures at the time of filing will be determined at a later stage following further stakeholder consultation and development of international norms (Australian Government 2016b). Australia’s possible accession to the Designs Law Treaty (DLT), which currently requires a grace period of 6 or 12 months, is a relevant consideration for future implementation of a grace period. For example, offering a grace period of longer than 12 months may result in unnecessary confusion should Australia subsequently accede to the DLT and have to reduce the grace period. A six month grace period seems adequate if the primary purpose of the grace period is to address concerns about inadvertent disclosure, as opposed to allowing market testing prior to filing. To the extent that Australia adopts a grace period, there may be benefits in trialling a grace system independently as part of an overall assessment before committing to the DLT. This trialling would reduce the risk of locking in any unintended adverse effects. Gilbert and Tobin (sub. DR565) agreed that implementing the grace period before any finalisation of the DLT (and accession) would be a useful measure. The Law Council of Australia (sub. DR490) did not support the introduction of a grace period when there are no treaty obligations to do so, and further considered that if or when a grace period is introduced that there must be a prior user defence. The Australian Design Alliance (sub. DR619) supported ACIP’s recommendation and welcomed further consultation as long as it does not result in extensive delays. Publicly–funded education initiatives should be carefully targeted Some participants have suggested that measures to enhance awareness about designs law and practices could address concerns such as low uptake of design rights. The Design Institute of Australia advocated for the Australian Government (in consultation with the industry and IP Australia) to create an education program about the Designs Act and its provisions and to roll out the education program to Australian designers: This education program would seek to provide the legal knowledge and practice protocols to Australian designers about the design registration rights available to them. Designers, as well as all stakeholders in the design process, having greater and practical knowledge and awareness of the revised designs system will be more likely to use and adopt the designs registration system with better expectations of what the system will deliver for them. (sub. 131, p. 5) 366 INTELLECTUAL PROPERTY ARRANGEMENTS The Government would need to target carefully any publicly funded education initiatives that aim to raise awareness of designs law given the range of resources and services already available from public agencies, such as IP Australia, interest groups and legal advisers. For example, IP Australia’s website provides relatively comprehensive information on the design rights system. Interest groups, such as the Arts Law Centre of Australia, and law firms supplement this information with more tailored advice. In the first instance, governments should put the onus on businesses to make themselves aware of the different forms of IP protection available, as part of their due diligence and risk management activities and professional training. Design industry bodies also have a role to play in advancing understanding of the designs system among designers. If evidence emerges that businesses face significant impediments to obtaining relevant and reliable information on designs law, there might be a case for further government action beyond IP Australia’s current information efforts. However, participants have not identified such impediments. Terri Winter noted failure for designers to register was in part because they did not devote sufficient attention to administrative matters: To be honest, a large amount of the designers I’ve spoken to are not design registered and that is twofold from what my discussions are; one, the cost of registration; two, some designers aren’t the best bookkeepers in the world and that’s not part of this discussion because there’s nothing we can do about that. (trans., p. 380) Does the designs law need to be more adaptable? In the course of ACIP’s review of the designs system, and this inquiry, participants identified a number of options to make Australia’s designs system more adaptable to changing technologies and industry practices. One of these options relates to special provisions to protect virtual designs. Special provisions to protect virtual designs A virtual or non-physical design affects the appearance of a product through software displayed on the product’s screen, such as graphical user interfaces and screen icons. In Australia, designers’ ability to protect virtual designs through the design right system is uncertain. This uncertainty is partly because the courts have not yet clarified whether the definition of a product as something ‘manufactured or hand made’ applies to non-physical designs. Further, the IP Australia Designs Office has a practice that requires that the design of a product be visible when the product is in its ‘resting state’ rather than ‘in use’. The practice is untested in the courts (ACIP 2015a). Despite the legal uncertainty, there are examples of registered designs that relate to graphical user interfaces and other non‐physical designs in Australia, although very few of these registered designs are certified (ACIP 2015a). From 2004 to 2014, for example, IP Australia received and registered 107 design applications relating to graphical user REGISTERED DESIGNS 367 interfaces and icons (Locarno Classification 14, subclass 04 (14–04)), which is less than 0.2 per cent of all registrations. Most of these registrations were from multinational companies, with Microsoft alone accounting for over 60 (more than half of all) registrations. None of these registrations were from Australian residents. ACIP (2014, 2015a) found that some protection for non-physical designs is desirable, given the increasing focus of design efforts on software elements of products. It argued that drawing a distinction between two designs, one of which has design features which are permanent features of a product and the other of which is visible only when the product is ‘on’, is an artificial distinction and is not consistent with the definition of ‘product’ in the Designs Act. It also noted that several other jurisdictions, including China, European Union member states and the United States, allow design protection for graphical user interfaces. ACIP considered various options for providing protection to non-physical designs, ranging from revision of IP Australia’s practice regarding the resting state of devices, to amending the definition of a product to clarify that it includes a non-physical product. ACIP subsequently recommended that the Government reconsider the treatment of virtual or non-physical designs, for example by allowing consideration of the product in its active state, and not just its resting state, when considering validity. However, it noted that amending the definition of a product to clarify that it includes a non-physical product would represent a more fundamental change to Australia’s designs system (ACIP 2015a). Participants to this inquiry had materially disparate views on the need for special provisions to protect virtual designs. Apple argued that: … without strong intellectual property protection for design, it is difficult for companies to justify further investment to materially advance the state of GUI [graphical user interface] innovation for both existing and yet-undiscovered future platforms. (sub. DR554, p. 1) In contrast, the Law Council of Australia (sub. DR490) noted that it does not support the introduction of special provisions to protect virtual designs, having previously observed that: … The Designs Office is currently dealing with such designs under the 2003 Act. Further, to the extent they are not protectable under the 2003 Act, they may qualify for and retain copyright protection. Additionally, many items of this nature may be registrable as trade marks, but they would need to be used as trade marks in order to remain validly registered. (sub. 64, att. 3, p. 8) The additional effect on innovation from Australia providing virtual design protection might be limited if the main users of such protection are multinational businesses. For example, it seems likely that innovations embodied in popular smartphones would continue regardless of whether Australia protects virtual designs. In a study of design rights in the United States, which has granted several thousand design patents on virtual designs, Du Mont and Janis (2013, p. 137) found a small number of 368 INTELLECTUAL PROPERTY ARRANGEMENTS multinational firms (such as Microsoft, Apple and Samsung) ‘dominate the virtual design patent landscape’, while small companies are either not relying on IP protection for generated images or they are more heavily relying on copyright or trade marks. Responding to ACIP’s recommendation, the Australian Government agreed to re-consider the treatment of virtual or non-physical designs, although it did not commit to specific changes: Through its international engagement activities IP Australia will continue to assess whether specific considerations for the treatment of virtual or non-physical designs would benefit Australian interests. (Australian Government 2016b, p. 6). The Commission supports the Australian Government’s decision to further build the evidence base on whether specific considerations for the treatment of virtual or non-physical designs would be beneficial and, if so, how this should be done. In this regard, it is notable that the WIPO Standing Committee on the Law of Trademarks, Industrial Designs and Geographical Indications (SCT) agreed to a proposal by Japan and the United States to collate information on laws and practices regarding the protection of new technological designs (including graphical user interface and icon designs) through a survey of WIPO members. The SCT Secretariat is to prepare a document containing the replies to the questionnaire and present the document at the next session of the SCT in October 2016 (WIPO 2016f). Conclusion Participants expressed concerns about Australia’s design rights system, including the low uptake of design rights due to the cost of registration and enforcement, and a poor understanding of design law, which can lead to designers inadvertently losing their rights or failing to secure protection in the first place. The Australian Government has committed to making changes that would go some way to addressing these issues. Following a review by ACIP, the Government has agreed, among other things, to the introduction of a grace period for filing registered design applications. This will help ensure designers do not inadvertently lose eligibility for design protection and allow them to undertake some market testing prior to incurring the cost of filing. The Commission supports the introduction of a six month grace period. Such an initiative would benefit from a trial. The Commission is also recommending some general measures to improve dispute resolution processes (chapter 19). These reforms would go some way to addressing concerns among designers about enforcement costs and access to dispute resolution options. Many participants see joining the Hague Agreement as offering the potential for lowering the costs of registration. Under the Agreement, Australian designers would be able to seek protection in multiple countries through a single international application. But the benefits REGISTERED DESIGNS 369 to Australian firms, and in particular SMEs, are likely to be much smaller than some anticipate. Filing for protection under the Agreement is not necessarily cheaper than directly filing for protection, particularly where firms seek protection in a limited number of countries. More importantly, joining the Hague Agreement would involve extending the maximum term of protection for registered designs from 10 to 15 years. The Australian Government has already agreed to further investigate the costs and benefits before making a decision to sign on to the Hague Agreement. Consistent with the approach taken by the Commission in this inquiry, such a process should ensure the gains from ‘harmonisation’ outweigh the costs of extending term, and that the interests of Australian consumers are adequately considered. FINDING 11.1 The Australian Government has committed to implement many of the recommendations made by the Advisory Council on Intellectual Property in its recent review of Australia’s designs system. These measures will help address participant concerns about the cost of acquiring registered design rights, and the lack of understanding of design law. Recommendation 19.2 provides for a low-cost avenue for IP enforcement currently sought by designers. 370 INTELLECTUAL PROPERTY ARRANGEMENTS 12 Trade marks and geographical indications Key points • Trade marks help consumers find goods and services by making it easier to differentiate between the products and services of different firms. They can also encourage businesses to build and maintain a positive reputation. Geographical indications (GIs) play a similar role, helping consumers to better identify the provenance of goods and services. But both trade marks and GIs can be used to limit competition and reduce the welfare of the community. • An effective trade mark system needs to strike a balance so marks are not granted excessively and do not mislead consumers. A balanced system focusses on the quality of marks and GIs granted, as well as their number. • It is easier to register a mark than it is to identify and remove marks no longer being used. A better balance can be achieved by making it easier to identify marks not being used, and to reduce the period between registration and removal for non-use. • Evidence suggests trade marks are confusing consumers, especially in relation to geographic terms. This confusion can be redressed by empowering the Trade Marks Office to more rigorously test claims made about such marks. • There is also confusion around rights afforded by a business name relative to a trade mark. Linking business registrations with IP Australia’s trade mark database would reduce the risk of unintentional infringement of trade marks, especially by small business. • Parallel imports of legitimate trade marked goods are often prevented by law that has become unclear and unworkable. This is to the detriment of competition and Australian consumers. The Trade Marks Act 1995 needs to be amended to explicitly allow for parallel imports. • GIs for wine and spirits can be improved by remedying a legislative oversight that prevents changes to boundaries of Australian winemaking regions. • International systems of GIs inhibit Australian trade in some foodstuffs, and producers may face more barriers from international agreements currently ‘in the pipeline’. Governments should carefully consider the costs and benefits from clauses on GIs when negotiating and finalising trade agreements. This chapter describes Australia’s systems of trade marks and geographical indications (GIs), which are intellectual property (IP) rights largely designed to signal information between firms and consumers. TRADE MARKS AND GEOGRAPHICAL INDICATIONS 371 12.1 What are trade marks and GIs? Trade marks A registered trade mark is a right that is granted for a letter, number, word, phrase, sound, smell, shape, logo, picture and/or aspect of packaging. It is legally enforceable and gives the holder exclusive rights to commercially use, licence or sell it for the goods and services that it is registered under (IP Australia 2015o). A trade mark application is made to the Trade Mark Office, which is part of IP Australia. Applications are evaluated in two parts. The first is where the trade mark examiner assesses the application against the criteria in the Trade Marks Act 1995 (Cth) (‘Trade Marks Act’) and makes a decision whether registration should be allowed to proceed. If successful, an application moves to the second stage, where other parties can make oppositions to its registration based on the application and additional criteria. If a mark passes both stages, then it may be registered. (Appendix F provides greater detail on the trade mark application process, as well as details on use and enforcement.) An examination of a trade mark at application is made with a ‘presumption of registrability’, which was legislated in 1995 in response to concerns it was too difficult to register a mark. While the Trade Marks Act does not spell out how far the presumption of registrability extends, courts have interpreted that an application is to be accepted even if the registrar is in doubt about any of the criteria of an application (Burrell and Handler 2012). Following the ‘Raising the Bar’ amendments to IP laws in 2012, the presumption of registrability now extends to all criteria that are assessed as part of a trade mark application (IP Australia 2013b). The number of applications and registrations of marks in Australia has grown rapidly since the introduction of the presumption of registrability (figure 12.1). Australia has experienced among the strongest growth in the number of registered trade marks over the last decade compared to other countries (figure 12.2). Yet Australia’s net imports of trade mark-intensive commodities stand at about $30 billion, around triple the value in the late 1990s (appendix C). Marks can also be protected through common law and other statutory mechanisms. The tort of passing off allows a firm to seek a remedy if it can show that another party is using a mark that it owns, and that it has suffered damages as a result. The Competition and Consumer Act 2010 (Cth) also provides a means to prevent others using the mark of another where it could constitute misleading or deceptive conduct. One of the main benefits of a registered trade mark is that infringement actions are more straightforward, as it is not necessary to show confusion or damages; merely that the mark has been used without authorisation. 372 INTELLECTUAL PROPERTY ARRANGEMENTS Figure 12.1 (i) Trade mark applications and registrations in Australia by origin (ii) Registrations (1960–2014)a Applications (1960–2015) 50000 50000 40000 40000 30000 30000 20000 20000 10000 10000 0 1960 1970 1980 1990 2000 2010 Application year Australian origin 0 1960 1970 1980 1990 2000 2010 Application year Non-Australian origin a A registration is proxied by the status of a mark being ‘registered’ or ‘removed’, the latter taken as an indication that it once had been registered and since lapsed. Applications that are still pending are not counted as registrations. Data for 2015 is not shown, as it mostly comprises of pending applications. Source: Commission estimates based on IPGOD (2016 edition). Figure 12.2 (i) Registered trade marks in selected countries, 2005-2014 Growth in number of trade marks (ii) Growth in number of trade marks per capita United States United States Australia Australia United Kingdom United Kingdom Canada Canada New Zealand New Zealand Japan Japan 0 1 2 3 4 5 Average annual growth rate (per cent) 0 1 2 3 4 5 Average annual growth rate (per cent) Sources: Commission estimates based on WIPO (2015f); World Bank (2015). TRADE MARKS AND GEOGRAPHICAL INDICATIONS 373 Geographical indications IP Australia describes a Geographical Indication (GI) as: A Geographical Indication … identifies goods as originating in a specific territory, region or locality where a particular quality, reputation or other characteristics is essentially attributable to its geographical origin. (2013a) Australia provides different systems of GI protection for different goods. For grape-derived wine and spirits, there is a system of GIs administered by the Australian Grape and Wine Authority (AGWA), while for all other goods and services, a certification trade mark system is used. 1 Relatively few certification trade marks exist to provide GIs for non–wine and spirit products (Moir 2015), but geographic terms are sometimes used in standard trade marks. For example, there are around 70 registered trade marks that include the element ‘Hunter Valley’; around 80 that include ‘Byron Bay’, and 16 that include ‘King Island’. While this may not bestow an exclusive right to the term, it does demonstrate the way that the mark system can be used to provide a geographical ‘flavour’ to branding, without having to go through the more formalised (and rigorous) certification trade mark system. The GI regions for wine and spirits are determined by the GI Committee at Wine Australia. 2 They are required to take account of criteria that pertain to the origin of grapes (detailed in appendix F). This is different from the European system of GIs, which often encompass processes and heritage as well as origin (EC 2015b). For example, ‘Barossa’ wine is an Australian GI, but does not specify the type of grape used. In contrast, the Champagne GI defines a region in France, as well as the varieties of grape and the process to use in making the wine. The economic rationale for trade marks and GIs The economic rationale for trade marks — which is also applicable to Australian GIs — is summarised by Carter: The principal benefit of trademark protection is that it lowers consumer search costs. If goods were not marked, potential purchasers, unable to rely on any brand name (at common law, the trade name) or distinctive appearance of the packaging … to identify the producer, would need a means of testing the products directly. Moreover, the more valuable the mark, the greater the 1 A certification mark is a special type of mark to declare that a good or service has a particular quality — such as content, method of manufacture and/or geographic origin. A certification mark can be used by anyone, provided that they are certified as meeting the criteria required by the mark. In addition to assessment by the Trade Marks Office, an application for a certification mark is also assessed by the Australian Competition and Consumer Commission to ensure the claims it makes are not misleading or deceptive, and that the way it is licensed for use is not anticompetitive. 2 The Committee comprises members appointed by the AGWA, including a member nominated by each of the Wine and Grape Growers Association, and the Winemakers Federation of Australia (AGWA, sub. DR527). 374 INTELLECTUAL PROPERTY ARRANGEMENTS incentive for the producer to maintain the level of quality that creates the value and lowers the cost of search. Without legal protection, it would be difficult for the user of a mark to appropriate the full value that the mark represents. A trademark is not a public good, but it nevertheless exhibits nonexcludability, for in the absence of legal protection, if a firm of good reputation tried to mark its goods with a symbol to let consumers know them — that is … other firms could imitate the symbol and trade on the first firm’s reputation. This possibility would reduce the incentive for a successful firm to mark its goods and would thereby raise consumer search costs. 3 (Carter 1990, pp. 762–763) Although trade marks may reduce consumer search costs and encourage a firm to maintain a positive reputation, they can also impose costs on society. By providing an excludable right to use particular terms, shapes and colours, trade marks reduce the available set of signals that competitors can use (Barnes 2006). As put by Greenhalgh: … trade marks can be the basis for aggressive brand-building, resulting in market dominance by incumbent firms. In this case, the obstruction of introduction by new entrants of new qualities and varieties of products could reduce market competition. (Greenhalgh et al. 2011, p. 52) A balance is reached when the costs imposed by the trade mark are less than the benefits in reduced search costs. From an economic point of view, the mark should be granted if this is the case (Feinberg 1986). As put by Maskus: A balanced view recognizes that trademarks have positive impacts that offset the market power they might generate. Because trademarks indicate the inherent quality or other distinguishing features of identified products, the consumer’s costs of searching for preferred quality characteristics are lowered. This gives firms an incentive to maintain or improve quality over time in order not to erode the value of their marks. Thus, trademark protection may be expected both to raise the average quality of products on the market and to generate further product differentiation. (Maskus 2000, p. 48) In practice, the costs and benefits arising from trade marks will vary between consumers and producers. Choosing the laws and regulations relating to trade marks is a balancing act that attempts to lead to the greatest consumer benefit for the least amount of costs. Trade marks and GIs are by no means the only way to reduce consumer search costs. Alternatives include: • the common law protections of reputation, including the tort of passing off • consumer protection laws designed to prevent firms from making false claims • advertising and third–party reviews of products that provide information to consumers in making decisions without requiring an IP right. For example, the rise of the internet has made it easier for ‘digital word of mouth’ to spread faster and more accurately for a range of different products (Smith and Zentner 2016) 3 Search costs refer to the costs, including the cost of time, which consumers expend in looking for a product that satisfies their wants. TRADE MARKS AND GEOGRAPHICAL INDICATIONS 375 • government information schemes — such as food star labelling or mandating product origin that provide information to consumers, which in turn may mean that trade marks (and their potential impact on competition) are not warranted (Papandreou 1956, p. 510). Innovative activity could sometimes be supported by trade marks While trade marks are intended to reduce search costs, they can provide a way to market or signal an innovation that might lead to greater returns, and a greater incentive to undertake innovation in the first place. There is no empirical evidence that demonstrates a causal link between the use of trade marks and more innovation, but some studies have found a weak association between the two. For example: • a review of European countries found a positive correlation between innovation and trade mark activity (Mendonca, Pereira and Godhino 2004), as did a more recent analysis of UK firms in 2011 (Greenhalgh et al. 2011) • an Australian study found mixed evidence of an association between copyright and/or trade mark use, and firms completing a greater number of types of innovation, or more novel forms of innovation (Soames, Brunker and Talgaswatta 2011). More recent data indicate that innovating firms of all sizes are around three times more likely to use copyright and/or trade marks than non–innovating firms in Australia (figure 12.3) • an OECD study of the correlation between patent applications and particular classes of trade marks registered by firms in OECD countries concluded that R&D investors use patents and trade marks as complementary means of IP protection for some innovations (Dernis et al. 2015). This suggests that the economic rationale for trade marks could extend beyond reducing search costs in some cases, and may be an important signal to differentiate new and valuable innovations arriving on the market. But while some marks may be such a signal, most are not, and so such a role should not be an overriding consideration in assessing the effectiveness and efficiency of the trade mark system. What is needed to get the balance right? Trade marks and GIs can be used for anticompetitive rather than information purposes. An effective trade mark system is one that maximises the flow of information and innovation, and minimises the scope for anticompetitive conduct. This means that a balanced system is one that focusses on quality of trade marks and GIs granted rather than the quantity of such rights, and also takes into account the alternatives to these forms of IP rights. 376 INTELLECTUAL PROPERTY ARRANGEMENTS Figure 12.3 Share of firms in Australia that ‘use copyright or trade marks to protect their IP’ 2014-15 60 Not innovation active Innovation active Per cent of firms 50 40 30 20 10 0 0–4 persons 5–19 persons 20–199 persons 200 or more persons All firms Sources: Commission estimates based on ABS (Selected Characteristics of Australian Business, 2014-15, Cat. no. 8167.0). 12.2 Improving effectiveness by reducing the scope of cluttering What is trade mark cluttering and why is it a problem? Trade mark cluttering is where a register of marks contains a large number of unused or ‘overly broad’ marks (von Graevenitz et al. 2011). Cluttering undermines the effectiveness of the trade mark system as it makes it harder for new firms to establish a brand and it imposes extra costs and difficulties in finding a viable trade mark (Greenhalgh and Webster 2015). Ultimately, cluttering can undermine competition between incumbent firms and new entrants in a market, which is to the detriment of the community. Cluttering can arise for a number of reasons, including: • firms securing trade mark rights, but then not using those rights over part or all of what they sought • firms going out of business, but their marks remaining on the register of trade marks for the period left until they expire • changes to trade mark application processes that make it easier to apply for a series of marks rather than one at a time (von Graevenitz et al. 2011). TRADE MARKS AND GEOGRAPHICAL INDICATIONS 377 Australian trade mark law includes some measures to address these concerns. For example, the Trade Mark Act includes provisions to prevent the registration of common or generic terms, marks that are too similar to others, as well as mechanisms to remove marks that are not used by their owners. The effectiveness of these provisions affects the extent of cluttering that may occur (figure 12.4). A ‘lax’ system that makes it hard for new entrants (ii) An ‘effective’ system with more scope for new entrants Range of elements that the mark applies to (i) Trade mark cluttering – colourfully illustrateda Range of classes the mark applies to Range of elements that the mark applies to Figure 12.4 Range of classes the mark applies to a The concept of trade mark cluttering can be thought of in two dimensions: the way that marks apply to an overly broad number of goods and services (classes) and the number of elements to which they make a claim. This is demonstrated by the left panel of figure 12.4: different trade marks (represented by different colours) claim rights to ranges of elements and classes. A lax system encourages firms to seek rights as broadly as possible. As the number of ‘wide’ marks increases, the space left for other competitors diminishes. A trade mark system that has higher hurdles to register multiple terms and multiple classes could be thought of as the right panel of the figure. There are more marks which are more limited in scope. This provides space for future competitors to seek rights to differentiate themselves from incumbents. Because cluttering can potentially block firms from being able to distinguish themselves from other firms it erodes the effectiveness of the trade mark system by making trade mark searches, applications and registrations more difficult. Trade mark effectiveness, then, can be preserved by providing incentives for firms to register ‘as little as they need’ rather than ‘everything they can get’. What evidence is there of trade mark cluttering? Understanding the costs and difficulties in securing a trade mark is necessary to determine the extent of trade mark cluttering. Little data is collected specifically on this issue at the firm level in Australia, but other measures can help indicate if cluttering is, or is becoming, a problem. For example, an exploratory study of trade mark cluttering in the UK examined measures including feedback from trade mark users about difficulties they had faced; data on the number of classes in which trade marks are registered; and comparisons between 378 INTELLECTUAL PROPERTY ARRANGEMENTS jurisdictions with different stringencies of trade mark registration (von Graevenitz et al. 2011). One symptom of cluttering is rapid growth in the number of applications and registrations of trade marks, which Australia has experienced in recent decades. The growth in the number of registered trade marks in Australia has coincided with two events — the new Trade Marks Act in 1995, which introduced the presumption of registrability to applications, and the ratification of the Madrid Protocol by Australia in 2001 to harmonise and streamline international applications. These changes appear to have significantly increased the likelihood of successfully registering a trade mark (figure 12.5). Likelihood of trade mark registration has increaseda 100 Registrations by Australian firms Registrations by foreign firms 90 80 70 60 Presumption of registrability introduced 2014 2013 2012 2011 2010 2009 2008 2007 2006 2005 2004 2003 2002 2001 2000 1999 1998 1997 1996 1995 1994 1993 1992 40 1991 50 1990 Likelihood of successful registration (per cent) Figure 12.5 Madrid Protocol is ratified by Australia a A registration is proxied by the status of a mark being ‘registered’ or ‘removed’, the latter taken as an indications that it once was registered, but had since lapsed. Applications that are still pending are not counted as registrations. Dotted lines represent averages over the period. Source: Commission estimates based on IPGOD (2016 edition). Allowing for a greater number of trade mark applications to be granted would make sense if consumers are increasingly confused about the differentiation between brands, or that there are no alternatives to providing that information. Neither of these explanations seems applicable in the current circumstances. This is not to suggest that the pre-1995 level of registration success was appropriate — but rather the change in likelihood of success is difficult to rationalise after the event from an economic perspective. The key issue is whether the additional applications have contributed to cluttering of the trade mark system. IP Australia’s view is that it did not: IP Australia accepts that the presumption of registrability has had an impact on the proportion of trade mark applications that are accepted and notes that this is not an unexpected outcome. A deliberate decision was made to introduce a presumption of registrability with the 1995 Act TRADE MARKS AND GEOGRAPHICAL INDICATIONS 379 because the previous legislation was seen as being too strict and arguably prevented registration of marks that should have been registrable. IP Australia considers that the increased likelihood of a trade mark application being registered is not in itself evidence that the register is cluttered. (IP Australia, sub. DR612, p. 15) The Commission is less certain. Another indication of cluttering is trade mark applications being overturned at the oppositions stage on the grounds that they are too similar to existing marks or brands. An examination of published opposition hearings held by the Australian Trade Marks Office between 2010 and 2015 indicates that around two-thirds of oppositions are successful. Of these, the grounds for a successful opposition was usually related to the provisions of a mark being too similar to an existing mark (s. 44) or that another similar mark already has achieved a reputation in Australia (s. 60) (figure 12.6). Figure 12.6 Summary of published oppositions hearing outcomes, 1999 to 2015a Opposition failed Opposition successful under a different section Opposition successful under sections 44 or 60 The likelihood of successful oppositions occuring under sections 44 or 60 of the Trade Marks Act has increased through time. 1999-2003 (n = 378) 2004-2009 (n = 392) Between 2010 and 2015, around twothirds of successful oppositions applied to these sections. 2010-2015 (n = 391) 0% 20% 40% 60% 80% Share of opposition hearings outcomes 100% a Based on the ATMO oppositions hearings decisions under the Trade Marks Act 1995 (Cth) published by AustLII, which are a sample of hearings decisions. Section 44 of the Act requires a mark application to be not ‘substantially identical’ or ‘deceptively similar’ to an existing trade mark (or one seeking registration with an earlier application date). Section 60 of the Act prevents registration where a mark applied for already has a reputation in Australia. Source: Commission estimates based on AustLII (2016). This evidence suggests that oppositions are increasingly more successful than not, and that most successful oppositions are based on providing the Trade Mark Office with more information to highlight marks (registered or otherwise) that were either not considered during the application process, or were considered and judged not to be sufficiently problematic to warrant rejection at that stage. The Commission considers that both situations indicate greater difficulties in checking the stock of existing marks, and suggests that the trade mark register could be becoming more cluttered. 380 INTELLECTUAL PROPERTY ARRANGEMENTS Finally, there is some anecdotal evidence in Australia that new firms are finding it difficult to establish their branding, which also suggests a growing problem of clutter. As one designer put it: Designers should be able to use whatever colour they like, but trademarking is a hard issue … We work with most of Australia’s top brands and everything we do has to go through an IP lawyer these days to ensure it’s unique and can’t be challenged . . . it’s getting harder and harder to create a unique identity. … When colours, smells and shapes are owned by companies it becomes harder and harder for us to stay on top of the game and create something unique … (Hulsbosch in Redrup and Durkin 2014) How best to prevent cluttering? Cluttering can be mitigated by narrowing the scope of elements that can be ‘locked away’ under trade mark rights, reducing the number of classes in which marks are registered, and removing marks that may fall out of use or were never used in the first place. Policy levers to achieve these goals include: • fees to discourage overly broad applications and registrations • applying ‘disclaimers’ — conditions on a trade mark that mean certain parts of it are not protected • requiring more evidence of use of a mark, or providing better information to detect and remove marks that may not be used. Higher fees Trade mark fees — specifically the cost charged per class — are a tool that can encourage a narrower choice of classes at application and registration, which can reduce cluttering. For example, the World Intellectual Property Organization (WIPO) found that: Fees influence applicants’ decisions, not only on whether to apply for a trademark, but also on the number of classes in which they seek protection. For instance, in some offices, the initial application fee already covers goods or services belonging to more than one class, whereas in other offices the initial fee only covers goods or services belonging to a single class, and the fee for each additional class costs extra. As a result, offices in the former category see, on average, 0.63 more classes specified in each application than offices in the latter category’. (2013, p. 97) An example of a fee-based response to cluttering would be to charge more per class in a trade mark application as a means to discourage overly broad registrations, such as where firms seek protection in more classes than it needs to protect its reputation. Participants in the inquiry raised concerns about the consequences of such an escalating fee structure. One concern raised was that it would penalise firms that legitimately require trade marks in multiple classes (for example, IP Australia sub. DR612; FICPI, TRADE MARKS AND GEOGRAPHICAL INDICATIONS 381 sub. DR581). However, the current system of fees, which charge a flat rate per class to which a mark applies, could also be said to suffer from these same criticisms. Firms that need marks across a range of classes are implicitly penalised as they need to pay for more classes under the current fee and classification structure, but still face a level playing field in competition with each other, and would still do so under an escalating fee structure. And any particular concerns for small firms that may be adversely affected could easily be mitigated by not ‘escalating’ the fee until after a third class of goods or services is claimed, given that small firms rarely seek protection among multiple classes (figure 12.7). Another concern raised about escalating fees was that they would be inequitable, as they would have a greater effect in discouraging smaller firms from seeking trade marks relative to larger firms (for example, IPTA, sub. DR562). The Commission considers a more valid criticism is not that small firms are too responsive to changes in trade mark fees, but that larger firms are not responsive at all. Analysis undertaken by the Commission suggests that trade marking activity by large and overseas firms appear largely unresponsive to changes in trade mark fees, while the behaviour of small and medium enterprises are relatively more sensitive to fees (appendix G). Possible reasons for this include that the reputation and branding of large firms is worth so much that trade mark fees are entirely insignificant by comparison; and that larger firms are more likely to register a mark through a legal practitioner, the costs of which are greater than the fee itself. This suggests that fees are not readily amenable for use as a lever to counter trade mark cluttering. They may only be effective if targeted at large firms, and even then, only if set at a significantly higher level than presently. 382 INTELLECTUAL PROPERTY ARRANGEMENTS Figure 12.7 Trade mark applications and registrations have become broader The average number of classes per trade mark applicationa is rising… Classes per trade mark application 2.4 2.2 ... and over time classes per trade mark application have grown more quickly for large and foreign firms than the average. Legislative change allowed for more classes per trade mark application... 2.0 1.8 1.6 1.4 All trade mark applications Applications by large firms 1.2 Applications by foreign firms 1.0 1995 1997 1999 2001 2003 2005 2007 2009 2011 2013 …and the share of classes comprised of single-class trade marks has fallen Percent of registered classes 50 1996 2014 Of all classes registered in 1996, around half were registered under marks with a single class. By 2014, this proportion had fallen to a third. 40 30 20 10 0 1 2 3 4 5 Number of classes in a registered mark 6+ a Data on large firms is only available for applicants based in Australia. Source: Commission estimates based on IPGOD (2016 edition). TRADE MARKS AND GEOGRAPHICAL INDICATIONS 383 Disclaimers Another means to address cluttering is to employ ‘mandatory disclaimers’. These allow a trade mark examiner to require some elements of a trade mark application be ‘disclaimed’ — which means that they do not receive the right afforded by the mark — in order for registration of the entire mark to proceed. The virtue of mandatory disclaimers, is not that they reduce the number of registrations, but rather they curtail the scope of legal rights that those marks provide. 4 Trade mark examiners lost the ability to make mandatory disclaimers as part of the introduction of the 1995 Trade Mark Act, and the restoration of such powers have since been canvassed in subsequent inquiries: • the Ergas review found little quantitative evidence to suggest that mandatory disclaimers should be reintroduced, but recommended that voluntary disclaimers 5 be encouraged instead (IPCRC 2000, pp. 194–195) • the Advisory Council on Intellectual Property (ACIP) recommended that mandatory disclaimers be reintroduced as part of its 2004 review into trade mark enforcement, but this was not accepted by government (ACIP 2013a). Mandatory disclaimers have several positive features. They: • can provide, through time, a series of worked examples to demonstrate what is considered to be non-distinctive as a trade mark. This helps to provide certainty for future applicants • provide a tool for trade mark examiners to reduce the scope of trade marks in the interests of keeping the system uncluttered, and to preserve terms that all parties may wish to use in their branding • are not compulsory tools — trade mark examiners can use them when necessary. It is up to the Trade Marks Office to determine how best to use such a tool in the environment that it faces. Some participants in the inquiry advocated the restoration of mandatory disclaimers on the grounds that they improve the trade mark system by providing clarity around the scope of a mark, rather than as a means to specifically reduce cluttering (including Ai Group, sub. DR582; Alexander et. al., sub. DR505; FICPI, sub. DR581; IPTA, sub. DR562). The benefits of disclaimers were summed up by Telstra: 4 For example, in the case of ‘Diet Coke’, which was registered under the Trade Mark Act 1955 (Cth), the endorsement is included: ‘Registration of this trade mark shall give no right to the exclusive use of the word DIET’. (Trade Mark 385318) 5 Voluntary disclaimers allow an applicant or mark holder to disclaim any exclusive right to use, or authorise the use of, a specified part of the trade mark. While difficult to isolate in the trade mark register, IP Australia indicate that the number of voluntary disclaimers entered since the commencement of the 1995 Trade Marks Act is, at most, approximately 205 out of the million trade mark applications filed. In contrast, around 24 000 mandatory disclaimers were entered for the 117 000 marks applied for under the 1955 Act, between the years 1989 and 1995. (IP Australia, pers. comm., 8 July 2016) 384 INTELLECTUAL PROPERTY ARRANGEMENTS In the absence of a disclaimer, the non-distinctive elements of a mark may not be obvious, particularly when they form part of a composite mark, or when used in the context of a particular industry, or when applied to specific goods or services. While a disclaimer may not identify what rights flow from a registration, it would identify what rights do not flow. Currently it can be extremely cumbersome, complex and expensive to search the Trade Marks Register, identify potentially conflicting marks, and advise on adoption risks when it’s not clear on the face of the Register the scope of rights under a registration which includes non-distinctive elements. We agree that the reintroduction of mandatory disclaimers would assist with the ‘cluttering’ issue identified by the PC. Further, they would provide far greater certainty as to the validity and scope of trade mark rights, which in turn would reduce the potential for conflict between trade mark owners, and ultimately enforcement actions. (sub. DR316, p. 11) However, mandatory disclaimers also have costs. ACIP noted that the reintroduction of disclaimers would add more complexity to the registration system, and could lead to some parties being discouraged from pursuing trade mark protection if explanations of disclaimer requirements were not set out clearly (2004). Complexity arising from mandatory disclaimers for users of trade marks could be mitigated by providing more information as to what disclaimers can and cannot do, and by improving the search capabilities of the Australian Trade Mark On-line Search System (ATMOSS) database to make searching for disclaimers possible. The costs associated with administering disclaimers are more difficult to determine and to address. As noted by IP Australia: It is difficult to estimate the costs or administrative burden that might arise if mandatory disclaimers were introduced into the current Act. IP Australia can describe its experience with the mandatory disclaimer requirements of the previous Act, but this would not readily translate to the current Act because the overall context in which trade mark applications are examined and registered changed with the introduction of the new Act. Furthermore, any assessment of the impact of reintroducing mandatory disclaimers would be speculative in the absence of detailed information about the nature of the disclaimer requirements and any other associated amendments that might be made to the trade mark system in order to accommodate mandatory disclaimers. … It should also be noted that IP Australia operates on a cost-recovery basis and, other than extra fees charged per additional class, the fees for examining and registering trade marks are the same for all applicants regardless of the complexity or time spent on each application. Accordingly, the extra costs involved in administering mandatory disclaimers would be passed onto all trade mark applicants. (IP Australia, pers. comm., 8 July 2016) The bottom line is that mandatory disclaimers have clear benefits, but potentially high administrative costs that would affect all users of the system through higher fees. The policy question is whether the benefits from mandatory disclaimers to make ‘better-defined’ marks with a more narrow legal reach are greater than the costs imposed by higher fees across the board. Given that the latter are more likely to disproportionately affect small businesses, and concerns from participants about the impacts that higher fees may cause, the case to restore mandatory disclaimers is weak. Accordingly, the Commission does not support their reintroduction. TRADE MARKS AND GEOGRAPHICAL INDICATIONS 385 Use requirements Another option to limit cluttering is to change the information provided about the use of marks at application, and to make it easier and quicker to remove marks that are not in use. Unused marks represent clutter on the register, as they do not provide information to consumers, and may block entrants seeking to register similar terms. The problem caused by unused marks can arise from the rules that govern applications as well as the changing nature of firms using the trade mark system. In Australia, a firm may register a trade mark so long as it has intent to use that mark. A ‘grace period’ applies such that a mark may not be removed for non-use for five years following initial registration. Removal for non-use is not automatic — it requires another party to seek removal of a mark through the Trade Mark Office, with a hearing if the action is contested. 6 The exception to the non-use rules apply to defensive marks, which the Commission considers to be of limited value given they are rarely used, but that their removal would do little to mitigate cluttering (box 12.1). In Australia, changes in the nature of users of the trade mark system are playing a role in creating clutter from unused marks. Analysis of the Intellectual Property Government Open Data (IPGOD) dataset shows that the proportion of trade mark applications made by small Australian firms has increased from around 47 per cent in 1997 to 58 per cent in 2013. While firms are meant to report their ‘winding up’ or change in ownership of IP assets to IP Australia, in practice this does not always occur. The marks of ‘dead’ firms can remain on the register for years after a firm’s demise, serving no purpose except to potentially block others that may seek to register similar terms. IP Australia has attempted to quantify the number of marks that are still on its register that may be held by dead firms, by examining whether the ABNs or GST numbers linked to IPGOD data have been de-registered. This approach suggests that around 8000 active, registered marks — around 1.3% of registered marks held by Australian businesses — belong to firms that are no longer trading (IP Australia, sub. DR612). This is likely to be an underestimate, as it is likely there are dead firms that are yet to be removed from the business register, which may hold ‘live’ marks. As more small firms seek trade mark protection, the current grace period following registration has become inappropriately long. It makes little sense to afford a five-year term of protection given that around half of new firms fail within three years of start-up (ABS 2016), and ‘nascent’ firms that may secure trade mark rights have only a 33 per cent likelihood of having commenced operations after 3 years (DIISRTE 2012). The Commission considers that the grace period should be reduced to three years (the minimum period required under TRIPS). 6 Analysis of IPGOD by the Commission suggests non-use action is not frequently contested. Between 2002 and 2014, around 68 per cent of actions to remove a mark were not defended by the mark owner. 386 INTELLECTUAL PROPERTY ARRANGEMENTS Box 12.1 The use of defensive trade marks Defensive trade marks allow for owners of ‘well known’ marks to extend the scope of their trade mark to other goods and services that they have no intent to use, but where use by others could lead to consumer confusion (IP Australia 2016n, pt. 34). The sanctioned non-use of trade marks in particular classes of goods and services that defensive marks allow raises concerns as to whether they may promote cluttering. All else being equal, it seems reasonable at first glance to abolish such marks, especially if they are not well used. Indeed, when the Working Party to review trade mark legislation examined the issue in the early 1990s, they recommended: … defensive registration should be retained, and a review of the volume of such registrations should be conducted in, say, 10 years’ time, with a view to repeal of these provisions if there has been little use of them. (Working Party to Review the Trade Marks Legislation 1992, p. 56) Participants in the inquiry objected to the abolition of defensive marks. FICPI stated that defensive trade marks did not cause problems or lead to cluttering (sub. DR581). Others argued that defensive trade marks represent an effective means to mitigate consumer confusion, and impose less administrative burden on firms with well-known marks (Behan, trans., pp. 76–85; Queensland Law Society, sub. DR567; IPTA, sub. DR562). The Commission accepts this, but notes that owners of well-known marks are likely to have the strongest case to prevent registration of similar marks, and to be successful in common law actions. Put another way, defensive marks are affording an additional layer of protection to firms that need them the least. IP Australia argued that defensive marks form part of Australia’s international obligations to provide protections to well-known marks, under the Paris Convention and the Agreement on Trade Related Aspects of Intellectual Property Rights, and that they are the most efficient and transparent means of doing so (sub. DR612). It noted that the defensive mark regime provides an open and transparent mechanism for protecting well-known marks that benefits both the owners of these marks and third parties (IP Australia, pers. comm. 8 July 2016). While defensive trade marks may form part of Australia’s international obligations, it is not the case that they are necessary. One participant queried the use of statistics about defensive trademarks, suggesting that the number of classes where protection had been sought was a more reliable indicator than the number of applications (Queensland Law Society, sub. DR567). Analysis of the IPGOD dataset by the Commission indicates that there have been around 10 applications per year for defensive marks since the introduction of the 1995 Trade Mark Act, and that around 750 classes presently have defensive mark protection (compared to around 1.2 million classes protected by standard trade marks). Defensive marks are not extensively used, nor are they growing at the same rate as other trade marks applications – either in terms of the number of applications or classes applied for. While defensive marks are little used, the Commission considers the costs and benefits of dispensing with them are likely to be small. Another issue that contributes to the non-use of marks is the weak requirement at application and registration of an ‘intent to use’ the mark, rather than demonstrating actual use. It has been shown in other jurisdictions that intent to use frequently does not translate into eventual use of a mark, and also encourages broader claims in mark applications. For example: • more than half of trade marks filed with an ‘intent to use’ do not proceed to registration in the United States, where a demonstration of use is needed (WIPO 2013) TRADE MARKS AND GEOGRAPHICAL INDICATIONS 387 • a comparison of marks registered in both the US and the EU (where there is an intent to use system) shows that marks in the EU claim 50 per cent more goods or services than the same mark registered in the US (von Graevenitz, Ashmead and Greenhalgh 2015). An intent to use system therefore needs an effective means to remove marks for non-use. Some changes have been made recently to the trade mark system to this end. For example, applicants seeking the removal of a mark need not be an ‘aggrieved party’ or supply evidence that they have made enquiries about the use of the mark in question. Processes to remove marks where there is no opposition have also been streamlined (Burrell and Handler 2016). However, these changes have been more directed at applicants seeking removal of marks, rather than measures to ensure that registered mark holders are meeting the requirements of use. Nor do they assist those seeking to remove a mark in identifying which marks on the register are likely to be in use. The Commission considers further measures to strengthen non-use provisions are warranted. One approach would be to abandon the present ‘intent to use’ system and instead require evidence of use at registration and renewal. The Australian Working Party into trade mark legislation considered this issue in the early 1990s, and resolved that an ‘intent to use’ criterion was more appropriate given the nature of Australia’s international obligations (Working Party to Review the Trade Marks Legislation 1992, p. 53). ACIP examined the issue of demonstrated use at renewal, but found that the costs it would impose on trade mark owners would exceed the broader benefits (ACIP 2004, p. 25). Given the growth in the number of marks since the ACIP review, it is likely that the administrative costs of such a move would be even greater today. But there are less-costly alternatives that can achieve similar outcomes. IP Australia should collect information at the trade mark application, registration and renewal stages as to whether an applicant is using the mark or is intending to use the mark, with the resulting information to be held on the register and accessible by the public through ATMOSS. This would provide other parties seeking to use a mark that is held by another to have some information on the status of the mark and highlight those that may be more likely to be subject to removal for non-use. It would also be a consistent extension of current practice, which allows for searches of marks that are presently subject to non-use actions. IP Australia should also determine whether those marks registered under an ‘intent to use’ basis are indeed being used after the initial grace period. The simplest approach would be to contact the trade mark holder and seek a statutory declaration that the mark is indeed being used. Those that indicate that it is not in use should be invited to remove their mark from the register. This correspondence should be noted on ATMOSS as well, providing further transparency as to which marks may be subject to removal through a non-use action. 388 INTELLECTUAL PROPERTY ARRANGEMENTS 12.3 Improving effectiveness by reducing confusion among users of trade marks The protection afforded by trade marks is not well understood by consumers or business. Firms, especially small business, struggle to understand what protections a trade mark affords and frequently conflate the protections of a mark with that of business registration. Consumers, too, can be confused as to the role of standard trade marks, in that they are a tool for differentiating brands, rather than an explicit claim of quality. Consumer protection in trade marks A criterion for registration of a trade mark is that it must not be misleading or confusing (s. 43 of the Trade Marks Act). The standard for misleading or confusing is that the connotation of the mark must be obvious and direct, the danger of the public being misled must be immediate, and the confusion must stem from the mark itself (IP Australia 2016n, pt. 29). Such a test is applied from the point of view of an ‘ordinary person’. The presumption of registrability extends to the consideration of whether a mark is misleading or confusing. The examiner’s manual states that if the registrar has ‘doubt’ that a misleading or confusing connotation exists, then the grounds for rejection should not be raised (IP Australia 2016n, pt. 29). In applying this criterion, the Trade Mark Office is faced with firms that have a desire to push their marks beyond what is distinguishing, and more into the descriptive. For example, there are around 700 registered marks using the term ‘healthy’, around 200 using the term ‘sustainable’, and around 100 using terms similar to ‘good for you’. Standard trade marks are not indicators of quality beyond that of a brand’s reputation, and so the Trade Mark Office must assess marks pertaining to ‘quality’ very closely. As put by one participant: Studies indicate that misleading or deceptive ‘healthy’ trade marks can influence consumers’ perceptions of healthiness, as well as their choice and consumption of food. This means that misleading or deceptive ‘healthy’ trade marks are problematic if food manufacturers and producers misuse them to make unsubstantiated, exaggerated or misleading claims about the health qualities or status of their food products. While the relationship between health claims and sales is not linear or straightforward, the market for food products with health benefits or properties is growing, and consumers are inclined to pay more for, and purchase and consume more often, foods carrying ‘healthy’ trade marks. This provides an incentive for food manufacturers to use unsubstantiated, exaggerated or misleading ‘healthy’ trade marks. (Sanderson, sub. DR208, p. 3) Similar concerns were raised in the 2011 Review of Food Labelling Law and Policy. That review found that the standard of trade mark examination meant that marks could indeed be misleading, and recommended that mark applications be also scrutinised by relevant health and safety agencies and rejected if they violated the Food Standards Code (Blewett et al. 2011). This recommendation was considered by state and Australian governments, TRADE MARKS AND GEOGRAPHICAL INDICATIONS 389 but was not adopted following consultation with IP Australia (COAG & ANZFMRC 2014). Potential for the descriptive (and the dishonest) to persist in food marks remains. Another problem that confronts the Trade Mark Office is that their examination of a mark is at a point in time, and so does not consider whether a mark that is clear today may become confusing in the future. This issue is pronounced in circumstances where a firm seeks to register a mark in advance of operations, but subsequently changes its products and processes in a way that its mark could now be confusing. Rectification is left to the Australian Consumer Law’s misleading and deceptive conduct provisions. But these provisions are narrower than the misleading and confusing criteria at trade mark application. Geographical terms in trade marks are an especially contentious area when it comes to misleading and confusing connotations. Consumers are increasingly demanding locally-sourced food and beverages, which provides an incentive for firms to brand their goods to reflect such desires. This leads to a tension in the trade mark system and how it intersects with consumer protection laws more broadly: … many trademarks incorporating place names escape what might otherwise appear to be a justifiable application of the section 43 standard. Trademarks law here intersects with consumer law (in effect section 18 of The Consumer Law) in a somewhat tense symbiosis. Although consumer law aims at accuracy, and in particular, in relation to credence claims which consumers cannot readily verify, the trademarks law allows the registration of trademarks incorporating place names quite readily, on the basis of fairly sparse evidence of acquired distinctiveness, and by a rather subtle reading and application of section 43. This interpretation allows that the place name might create a certain impression as to origin, which other elements of the mark as actually registered, then dispel. It is to be questioned whether the consumer mind in reality operates in such a subtle fashion, and is not simply pushed in one direction by the most recognisable and striking part of the mark – which is commonly the (well-known) place name that is included. To some degree the two areas of law (consumer law and trademarks law) seem to pull in opposite directions. (van Caenegem 2015, p. 8) There have been some recent cases of confusion relating to geographical references where the Australian Competition and Consumer Commission (ACCC) has taken action. Prominent examples include products with a trade mark (and other branding) that gave the impression that they came from the Barossa, when they had been produced elsewhere (ACCC 2014b). Another example included beer with a trade mark that referenced Byron Bay, but was brewed in Melbourne (ACCC 2014a). In the case of the latter mark, an endorsement was applied following ACCC action: It is a condition of registration that in instances where the trade mark is used on beer that is not brewed by the applicant or that is not brewed in Byron Bay the product or packaging to which the trade mark is attached will include further information specifying the place of manufacture and/or the company that brewed the beer. (Trade Mark 1590666) 390 INTELLECTUAL PROPERTY ARRANGEMENTS The endorsement subsequently made clear that the exclusivity afforded by the trade mark did not extend to products that were produced outside of Byron Bay. Such endorsements are required already for products that are subject to the Wine Australia system of GIs. Geographical terms in trade marks have previously attracted more scrutiny at examination. Until around 15 years ago, section 43 of the Trade Marks Act was regularly used to raise oppositions to marks that contained references to geographical locations. However, this practice has since ceased on the understanding that labelling laws are perceived to be sufficient in ensuring the accuracy of marks (IP Australia, pers. comm., 23 February 2016). The cases pursued by the ACCC recently suggest labelling laws have not always been sufficient in this regard. The Commission considers that challenging geographic references in standard trade marks is a good practice, and that the inclusion of endorsements that require the goods and services defined under the mark to be produced in the region nominated makes for a more effective and transparent trade mark system. Such an approach would also provide clarity when a change in ownership of brand occurs. The Commission also considers there is merit in clarifying the Trade Marks Act such that the presumption of registrability does not apply to the assessment of the misleading and confusing criteria at the application stage. Apart from the most blatant cases, it appears difficult to reject a mark on the grounds that it is misleading and confusing. By removing the presumption, it does not give the Trade Marks Office the ability to reject applications at whim, but provides examiners scope to seek necessary detail from applicants to better ensure that registered marks are not misleading and confusing. Producers are confused too The effectiveness of trade marks is diminished when producers themselves become confused about what rights the trade mark system affords them. Such confusion is greatest where firms conflate the registration of a business name with that of a trade mark. This leads to circumstances where a firm uses its business name in marketing, but in doing so infringes on an existing mark. As put by the Australian Small Business Commissioner: In relation to IP disputes involving a small business, trade mark infringement is the most likely situation, particularly where a small business unintentionally infringes an existing IP right and the owner enforces that right. (sub. 101, p. 10) Such concerns have been examined previously, including by ACIP in 2006, which found that confusion could be costly for firms: A result of such misconceptions is that business owners conduct inadequate checks for common law and registered trade marks before embarking on a business activity. Sometimes this results in the owner only becoming aware that their name infringes someone else’s prior common law or registered trade mark once they have made a major investment in their business. This can have significant consequences, particularly for SMEs. The business or TRADE MARKS AND GEOGRAPHICAL INDICATIONS 391 company may be required to forgo use of the name and so lose the reputation associated with the name. They may incur their own legal and re-branding costs, and may even be liable to pay compensation to and the legal costs of the party whose rights they have infringed. (2006, p. 27) The extent of the problem is difficult to pinpoint, but most estimates point to a considerable overlap of registered marks and business names. The ACIP review panel asked state and territory registrars for the number of cases where they had been made aware of a conflict between a trade mark, business name and/or domain name. The consensus was that there were at most around ten cases per year (ACIP 2006, p. 47). However, surveys by ACIP found a greater frequency of potentially conflicting marks and registered business names. • ACIP (2006, p. 47) checked a sample of 511 business names registered in Victoria in a 6 month period against the trade mark register, and found 11.5 per cent of those names were confusingly similar with a registered or pending trade mark owned by another party in the same field of goods and services. • IP Australia and the Queensland Office of Fair Trading examined a sample of 267 business names in the state with the trade mark register. In this sample, 13.1 per cent of business names were found to be confusingly similar to a registered or pending trade mark owned by an unrelated party in the same field of goods and services (ACIP 2006, p. 47). Extrapolating this to the general population, ACIP (2006, p. 48) concluded that ‘ … the number of business names potentially infringing registered trade marks is likely to be in the tens of thousands’. To confirm the incidence, ACIP and IP Australia commissioned market research into the level of awareness and understanding among business name owners of the rights and obligations associated with business names and marks. The main conclusions from their survey results indicated: • a general over-estimation of the ‘rights’ value of business names; • a limited understanding of the nature of trade marks; and • mistaken assumptions about the business name registration process. (ACIP 2006, p. 49) In response to these concerns, those seeking to register a business name are advised by most business portals, at different levels of government, to check if the proposed name might infringe on a trade mark. IP Australia provides a simplified ‘TM Check’ service, which the Australian Securities and Investments Commission (ASIC) (2015) links to as part of its information on how to register a business name. But ‘TM Check’ is not without its limitations: … TM Check is a simplified search for trade marks that have been applied for or registered with IP Australia. It is intended to provide the user with an indication only - being simplified it does carry some risks. In particular, please be aware that a search using TM Check will return a maximum of 10 results, even if there are more trade marks that match your search criteria. … ATMOSS search results are likely to provide a fuller picture than TM Check, although searching ATMOSS can be complex. (IP Australia 2015l, emphasis in original) 392 INTELLECTUAL PROPERTY ARRANGEMENTS The Commission considers that the best approach to prevent accidental infringements like those raised in the ACIP review and by the Small Business Commissioner is to augment the present search capability of the ASIC website with that of IP Australia’s ATMOSS database, and to make such searches automatic when a business name is sought to be registered. Such functionality would replace the existing TM Check function. The criteria around when a name is judged to infringe would need to be determined. The ACCC — with its experience on what constitutes misleading and deceptive conduct — should participate in determining such criteria. 7 Data should also be collected on the searches conducted that trigger the criteria, so that IP Australia can see how frequently firms may be accidentally seeking to infringe on an existing mark. 12.4 Improving efficiency by reforming parallel imports Recent legal decisions (box 12.2) have made it more difficult to import legitimately marked goods — an outcome that can be anticompetitive and welfare eroding. Parallel imports, which in this context apply to the importation of legitimately trade marked goods 8, were notionally allowed under s. 123 of the Trade Marks Act. But now confusion reigns as to when imports are allowed, and this makes for a less efficient trade mark system. As IP Australia put it: IP Australia notes that the existing law around the parallel importation of trade marked goods has led to uncertainty and confusion. It is also arguable that section 123 of the Trade Marks Act is not effectively implementing the policy intention of allowing for the parallel importation of legitimate goods. … IP Australia notes that the complexity and uncertainty existing in this area makes it more difficult for potential parallel importers to operate with confidence that their activities will not lead to infringement of a trade mark. (sub. 23, p. 12) Legal professionals have also expressed concerns around how difficult the law has become to interpret. As put by Rothnie, and the Law Council of Australia: Trade Marks Act s123 provides an express defence against infringement for the use of a registered trade mark in relation to goods to which the trade mark has been applied by or with the consent of the trade mark owner. In a number of recent decisions, the Federal Court has struggled with the deceptively simple terms of the provision to the extent that, arguably, neither trade mark owners nor prospective parallel importers can confidently predict when the defence may be relied on. So, almost 20 years after its enactment, it is still not clear who is the ‘trade 7 One participant in the inquiry suggested that the ACCC would not be best placed to assist in this criteria, given that misleading and deceptive conduct under the Australian Consumer Law is a different standard to the misleading and confusing provisions of the Trade Mark Act (Alexander et al, sub. DR505). The Commission considers that consultation with the ACCC is necessary, as misleading and deceptive conduct is likely to be a subset of misleading and confusing, and it would be the role of the ACCC to pursue any misleading and deceptive conduct that could arise following registration of a trade mark. 8 This is a separate matter to the parallel import restrictions on books as part of the Copyright Act 1968 (Cth), discussed in chapter 5. TRADE MARKS AND GEOGRAPHICAL INDICATIONS 393 mark owner’ for this purpose nor the nature or scope of the ‘consent’ that is required. (Rothnie 2014, p. 39) … in light of several significant decisions by the courts, it has become very difficult to advise clients on what is, or is not, a legitimate parallel import. … In addition, section 123 operates as a defence so the onus lies on the importer or retailer to prove all the requirements of the defence have been satisfied. This is typically very difficult to satisfy. The members of the [Intellectual Property Committee] are aware that some Australian retailers and importers are therefore avoiding the risks associated with parallel imports for fear of engaging in criminal conduct and being labelled a counterfeiter. (Law Council of Australia 2014, pp. 1–3) The onus is currently on the parallel importer to prove that the product was manufactured under licence from the trade mark owner. This is often impractical for someone who has merely purchased products in a foreign market to prove. The current drafting of section 123 is also the source of uncertainty where an overseas manufacturer or trade mark owner registers its trade marks in Australia in the name of a related entity or local distributor. In many cases, this will preclude the clear application of section 123, and is becoming more frequently used as a means to circumvent the statutory intention of the section and control parallel imports. (Law Council of Australia sub. 64a, pp. 4–5) Costco provided an example of how such restrictions can affect importers and distributors in practice: A handbag brand manufactures its global supply of a particular design of handbag in a single factory in China and distributes its handbags throughout the world including in the US and Australia. The trade mark registrations in Australia are owned by one company within the larger brand company conglomerate while the trade marks in the US are owned by another. The importation of the particular handbag design from the US to Australia will be an infringement of the Australian trade mark. Without knowing the information about the single manufacturing source for the handbags (which is likely to be too costly and perhaps impossible to determine), the defence to infringement under section 123 cannot be relied upon – it is up to the infringer to assert and prove the defence – even though the mark was arguably applied by or with the consent of the Australian trade mark owner (since all handbags are manufactured at a single source for distribution to all countries). (sub. 31, p. 12) The issues around parallel imports have been examined previously in a number of reviews: • The Ergas review in 2000 recommended that the Trade Marks Act should be amended to ensure that its provisions were not used to circumvent the intent to allow parallel importation of legitimately trade marked goods (IPCRC 2000, pp. 190–191). While this recommendation was accepted by the government at the time, implementation did not occur (Law Council of Australia 2014, p. 3) • ACIP sought comments on the Ergas review recommendation about parallel imports as part of its 2004 review of trade mark enforcement. It found ‘mixed support’, noting that there was a lack of evidence to suggest that trade mark owners were commonly adopting practices to block parallel imports (ACIP 2004, p. 24) • The House of Representatives Standing Committee on Infrastructure and Communications (2013, p. 106) inquiry into IT pricing in 2013 recommended ‘that the 394 INTELLECTUAL PROPERTY ARRANGEMENTS parallel importation defence in the Trade Marks Act be reviewed and broadened to ensure it is effective in allowing the importation of genuine goods’ Box 12.2 Case studies of parallel imports of trade marked goods Sporte Leisure case This case involved trade marked clothing products imported into Australia by the retailer Pauls Warehouse. The trade marked clothing was imported from India. The Indian manufacturer had obtained a licence from the Australian trade mark licensor to use the trade mark, but had agreed to not supply the goods outside of India. The Federal Court held that even though the clothing products had been manufactured overseas with the consent of the Australian licensor, the unauthorised importation and sale of those goods in Australia may infringe the registered trade mark. Lonsdale case Similar to the Sporte Leisure case, this case also involved the importation of trade marked clothing into Australia. Here, a United Kingdom company, Lonsdale Sports Limited granted a German company, Punch, a licence to promote, distribute and sell goods bearing the Lonsdale trade mark within a defined territory in Europe. Pursuant to this licence, Punch sold Lonsdale branded clothing to a subsequent company in Europe. Ultimately, the Lonsdale branded clothing reached Paul’s Retail who offered and sold the trade marked clothing in Australia. Lonsdale Australia, the Australian trade mark owner commenced action for infringement. The Full Federal Court considered the application of section 123 of the Trade Marks Act. The court found that there was no consent by Lonsdale Australia as the use was outside the scope of the original licence, between Lonsdale Sports Limited and Punch, which was to sell the trade marked products within the specified territory. Scandinavian Tobacco Group and Trojan Trading Company case In this case, the Trojan Trading Company imported tobacco products into Australia that bore trade marks of the Scandinavian Tobacco Group, removed their packaging, and repackaged them in a manner that conformed to local tobacco packaging laws, including a reapplication of the original trade marks. Trojan was successfully able to defend a claim of trade mark infringement by relying on the defence for parallel imports in the Trade Marks Act 1995 (Cth). This ruling was affirmed by the full Federal Court in June 2016. Sources: ACCC (2014c, p. 61); Downie (2016); Rothnie (2016); Scandinavian Tobacco Group Eersel BC v Trojan Trading Company Pty Ltd [2016] FCAFC 91. The Commission has received submissions from participants that argue for caution in changing parallel import arrangements. These objections largely fall into two categories: that parallel imports devalue a trade mark in Australia, as marked goods tailored for overseas tastes are found to be unattractive to domestic tastes, or otherwise mislead consumers (INTA, sub. 20, p. 4). The second is that parallel imports are often of poorer quality, and may be dangerous (FCAI, sub. 88, attachment 1, p. 4). Were parallel import restrictions to be lifted, some participants argued in favour of clearly distinctive packaging to denote them as parallel imports, or clear disclaimers to be required in the case of warranties that may be void from using ‘unapproved’ imports (INTA, sub. DR516). TRADE MARKS AND GEOGRAPHICAL INDICATIONS 395 These arguments expect the Trade Marks Act to do too much. There is other information regarding parallel imports manufactured to different tastes — such as country of origin labelling requirements (DIIS 2016) and differences in price — to provide consumers with information on provenance and quality. Similarly, the argument that the Trade Marks Act should play a role in preventing dangerous goods being imported is without merit. The Commission notes that there are laws designed to prevent the importation and use of dangerous goods (‘parallel’ or otherwise), and there are regulatory agencies at the state and federal level designed to police such matters. Accordingly, the Commission sees no role for the Trade Marks Act to screen dangerous goods. Restrictions on parallel imports could be in the public interest if it could be demonstrated that they are overcoming the ‘free– riding’ problems inherent in creative ideas. But such a notion is irrelevant to trade marks — laws against counterfeit goods respond to free riding, whereas restrictions on parallel imports only serve to help rights holders price discriminate. As put by the ACCC: Legislative restrictions on parallel imports are not justified by the traditional ‘free rider’ concerns relating to IP which relate to preventing unauthorised reproduction. Instead, parallel importation restrictions extend IP rights into the process of distribution. They may also lead to inefficient outcomes by providing rewards to creators that are not proportional to the value or risk of their creation and create a public detriment. Parallel import restrictions grant an exclusive right to import to IP owners. By preventing international arbitrage these import monopolies may be used to support international price discrimination by firms with market power. The ACCC considers that restrictions on parallel imports prevent consumers gaining access to an alternative source of goods which can promote competition and potentially provide consumers with lower cost products and improve the international competitiveness of user industries. (sub. 35, pp. 13–14) The Commission agrees. There are clear benefits to allowing the parallel importation of trade marked goods where the importer has secured a license from the mark holder. Reforming the provision that hinders parallel imports could take different forms, including: • one that clarifies how ‘use of a mark’ and ‘consent of ownership’ applies to trade marks, and adjusts such terms to allow for parallel importation • one that spells out when the rights afforded to trade marks are exhausted. The Commission considers that the latter approach is preferable, and has greater scope to prevent future legal uncertainty around parallel imports. Such an approach is also consistent with the current practice in New Zealand and with the suggestion of the Law Council of Australia on how to rectify the problem: A simpler test is whether the goods are genuine in that they have originated from the trade mark owner or its licensee. This would be consistent with the principle that a trade mark is a badge of origin, not of geographic control. (sub. 64a, pp. 4–5) The Commission notes that the use of the provision in New Zealand does not appear to have engendered the issues suggested in the submissions objecting to such a change. And in any case, the resulting loss of consumer welfare from price discrimination that the status quo allows is potentially a worse outcome for the community than any confusion that may 396 INTELLECTUAL PROPERTY ARRANGEMENTS arise from relaxing parallel imports restrictions. Amending the Australian Trade Marks Act to reflect a similar provision would resolve confusion with the workings of the trade mark system, more closely reflect the original intent of the Trade Marks Act, and improve efficiency by fostering greater competition. 9 RECOMMENDATION 12.1 The Australian Government should amend the Trade Marks Act 1995 (Cth) to: • reduce the grace period from 5 years to 3 years before new registrations can be challenged for non-use • remove the presumption of registrability in assessing whether a mark could be misleading or confusing at application • ensure that parallel imports of marked goods do not infringe an Australian registered trade mark when the marked good has been brought to market elsewhere by the owner of the mark or its licensee. Section 97A of the Trade Marks Act 2002 (New Zealand) could serve as a model clause in this regard. IP Australia should: • require those seeking trade mark protection to state whether they are using the mark or ‘intending to use’ the mark at application, registration and renewal, and record this on the Australian Trade Mark On-line Search System (ATMOSS). It should also seek confirmation from trade mark holders that register with an ‘intent to use’ that their mark is actually in use following the grace period, with this information also recorded on the ATMOSS • require the Trade Marks Office to return to its previous practice of routinely challenging trade mark applications that contain contemporary geographical references (under s. 43 of the Trade Marks Act) • in conjunction with the Australian Securities and Investment Commission, link the ATMOSS database with the business registration portal, including to ensure a warning if a business registration may infringe an existing trade mark. 12.5 Challenges for trade marks in the digital age The way in which trade marks are used in the digital age poses fresh questions over what the right affords and when infringement may occur. As Telstra noted in its submission: Greater clarity is needed with respect to online infringement of Australian trade marks, to ensure the ongoing effective and efficient operation of our trade mark system. It is currently not 9 One participant to the inquiry suggested that reform to s. 123 alone may not be sufficient to allow for parallel import of goods that have already been brought to market elsewhere, as such imports may still be considered as ‘use’ for the purposes of infringement (Alexander et al., sub. DR505). In amending the Trade Mark Act to allow for parallel imports, the Government should consider whether amendments are needed to section 120 to ensure that such imports are not blocked on infringement grounds. TRADE MARKS AND GEOGRAPHICAL INDICATIONS 397 clear whether certain online uses of trade marks amount to infringement under the Trade Marks Act 1995. For example: • does use of a trade mark in a competitor’s metadata amount to trade mark infringement? • does the use of an Australian trade mark on an overseas website that is able to be accessed in Australia infringe an Australian trade mark? (sub. 76, p. 18) Metadata, put simply, is data that ‘labels’ or ‘describes’ other sets of data. Many websites contain metadata that state the format and content types on a web page. This information is usually not visible to consumers. Search engines, however, scour through all data of web pages, including metadata, and take note of the number of times that particular terms are referenced. By including or repeating particular terms in metadata, webpages can be engineered in a way to bolster their search ranking and drive more people to view a particular site. The intersection with trade marks and trade mark law occurs when a firm uses the trade mark of a competitor in its metadata, so that consumers looking for the competitor are (unwittingly) driven to the firm. Using a competitor’s mark on packaging is certainly infringing, but it is less clear cut when a competitor’s mark is used in metadata. The strategic use by some firms of a competitor’s trademark in online metadata has become problematic. It has introduced legal uncertainty as to whether it constitutes a form of trade mark infringement. In Australia, there have been some important legal cases relating to use of metadata and trade mark infringement. In Complete Technology Integrations Pty Ltd v Green Energy Management Solutions Pty Ltd 10, it was found that the act of embedding a competitor’s registered trade mark as a meta tag did not constitute trade mark use, and so was not an infringement (Kittikhoun 2015). But in Accor Australia & New Zealand Hospitality Pty Ltd v Liv Pty Ltd 11, it was found that the presence of a trade marked term in a meta tag did constitute use and infringement, despite a lack of evidence that consumers had ever viewed the term in accessing the website (Mancini 2016). More recently, in Veda Advantage Limited v Malouf Group Enterprises Pty Limited 12, it was found purchasing an advertisement that appeared when a competitor’s mark was searched for was not an infringement. To date, there is no guidance from an appellate court in relation to the use of a competitor’s trade mark in metadata (Fixler 2016).13 The scope of such infringement becomes relevant to the second question posed by Telstra — could an Internet service provider be infringing a mark by ‘importing’ it from an overseas website and providing it to a customer? Could an Internet Service Provider or a 10 Complete Technology Integrations Pty Ltd v Green Energy Management Solutions Pty Ltd [2011] FCA 1319. 11 Accor Australia & New Zealand Hospitality Pty Ltd v Liv Pty Ltd [2015] FCA 554. 12 Veda Advantage Limited v Malouf Group Enterprises Pty Limited [2016] FCA 255. 13 Notwithstanding Lift Shop Pty Ltd v Easy Living Home Elevators Pty Ltd [2014] FCAFC 75, where a firm used a trade marked term as part of search engine optimisation. The court found the use of the term was ‘descriptive’ rather than ‘distinctive’, and so did not constitute use (or infringement) of the trade mark. 398 INTELLECTUAL PROPERTY ARRANGEMENTS search engine be an unwitting parallel importer or infringer? All these questions cut to what constitutes use and consent of a mark in the modern age, and the cases above suggest that there is legal uncertainty in both. From an economic perspective, the use of metadata to drive search results has the scope to cause consumer confusion and reduce the effectiveness of a mark to distinguish between firms. But what is less clear is to what extent such use of metadata to confuse consumers is currently taking place. The Commission considers that a ‘watch and see’ approach by consumer protection agencies at the state and Commonwealth level is warranted at present. 12.6 Geographical indications Wine and spirit GIs The GI system that governs the provenance of Australian wine and spirits has prescriptive rules around how they are assigned and applied. AGWA raised concerns about some elements of the wine and spirit GI system in its submission to the inquiry: • that consumer confusion could arise from regulations about GI labelling • the law as implemented made it difficult to omit a GI or amend its boundaries (AGWA, sub. 72). The former issue relates to how the GIs are presented on a wine label. If multiple GIs are used, they must be displayed in descending order of content, and 95 per cent of blend must come from the GIs listed (AGWA, sub. 72). AGWA describes how confusion can then arise: This causes the potential for consumers to be misled where a small amount of the blend is from a sub-region (which is listed), and the remainder of the blend is from the wider region within which that sub-region sits. For example, ‘Barossa Valley’, ‘South Australia’ and ‘Australia’ all appear on the Register. Accordingly, if 35% of the grapes used to make a blend are derived from the Barossa Valley, 33% is from ‘South Australia’ and the remainder of the blend is from the greater ‘Australia’ GI, a producer may label ‘Barossa Valley, South Australia, Australia’, suggesting that the wine is entirely from the Barossa Valley, when in fact only 35% of the wine is from that region. (sub. 72, p. 4) The Commission agrees that such a rule can lead to confusion, but the extent and cost of such confusion is unclear. Exploiting the GI system in such a way may constitute misleading and deceptive conduct, and there is no impediment to winemaking bodies or firms to take any concerns along these lines to the ACCC. Greater evidence of the scope of such a problem is needed before legislative change is warranted. The other issue raised by AGWA — the lack of an effective mechanism to redraw or omit GI boundaries — appears to be more of an oversight in drafting of the enabling legislation than deliberate choice. In practice, it is difficult to amend or omit a GI as there are many TRADE MARKS AND GEOGRAPHICAL INDICATIONS 399 parties that effectively have ‘veto power’ to any changes once a GI is finalised. For example, AGWA noted that a proposed change could be blocked by objections from the Winemakers’ Federation of Australia, the Wine Grape Growers Association, state representative bodies, those that own or lease land of more than five hectares in the GI in question, or any producer of grapes or wine in the GI in question (AGWA, sub. 72). AGWA proposed that this should be remedied by allowing the GI Committee to omit or amend a GI (using the same process to determine a GI) on the grounds of: • environmental or production changes warrant the expansion or contraction of a GI; • a region becomes known colloquially by a different name to that by which it was determined (and/or there is regional support for its name to be altered); or • the GI ceases to meet the criteria set out in regulation 24 or 25 of the Australian Grape and Wine Authority Regulations 1981 (AGWA Regulations), in particular, it fails to have the requisite degree of homogeny. (AGWA, sub. DR527, pp. 1–2) The existing process of determining a GI has led to costly litigation and disputes to date. An example of this was the determination of the Coonawarra region (box 12.3). Given the ‘mega-litigation’ that has occurred previously, some caution is warranted when it comes to amending or omitting existing GIs. With this in mind, the Commission considers that the GI Committee is the appropriate organisation to omit and amend existing GIs. This approach has the support of the main winemaking bodies in Australia, the Winemakers Federation and the Wine and Grape Growers Association (AGWA, sub. DR527; WFA, pers. comm., 14 July 2016; WGGA, pers. comm., 15 July 2016). The Commission understands that changes to GIs are also likely to be made sparingly and with approval by industry, rather than unilaterally or without consultation. 400 INTELLECTUAL PROPERTY ARRANGEMENTS Box 12.3 Where is the Coonawarra, exactly? The boundaries for the Coonawarra geographical indications (GIs) were finalised in 2000 by the GI Committee (the dotted line, right). Winemakers outside the region objected to the determination, and took action in the Administrative Appeals Tribunal. As put by one of the attorneys at the time: By the end of [2000], there were over 70 parties locked in dispute in front of the Federal Administrative Appeals Tribunal. The parties included 45 vignerons endeavouring to be included in what was to become the Coonawarra wine region as originally mapped out by the Australian Geographical Indications Committee, and 25 vignerons within the region as already mapped by the GIC, who were fighting to keep the other 45 out. The novelty and complexity of the litigation as well as the enormous amount of ongoing coordination between the two dozen lawyers representing the 45 ‘outsiders’ had resulted in significant legal expenditure. (Stern 2016, pp. 249–250). Stern (2016) went on to describe that the decision on whether to proceed with litigation was less to do with the merits of the case, and more to do with the additional revenue that holding a Coonawarra GI would bring. After lengthy hearings that included expert witnesses in the fields of geography, soil science, hydrology, viticulture and mapping, the tribunal concluded that there was no ‘absolutely correct boundary’ for the region, but on the basis of the evidence presented, provided for an expanded area of the GI (the unbroken line). However, winemakers outside this new region brought a case in the Federal Court to argue that the Administrative Appeals Tribunal had erred and that they deserved inclusion in the region as well. They were ultimately successful, and the GI region was expanded to include an even greater area (the shaded regions). The Coonawarra case has been described as a case of ‘mega-litigation’ and demonstrates the tensions that can arise in determining a valuable GI. Sources: Rimmer (2009a, 2009b). To provide greater certainty to those firms presently subject to GIs, a period of time should be required before any omissions/amendments decided by the GI Committee are implemented; an approach that AGWA describes as ‘a viable option’ (AGWA, sub. DR527, p. 2). The question then becomes one of how long a ‘grace period’ — to allow for implementation and appeal — should be offered after an amendment or omission has been made. In practice, such a grace period will be dependent on the circumstances of the change to a GI. Expanding a successful GI is likely to create more ‘winners’ than ‘losers’ and thus only a short period would be necessary. Conversely, the omission of a GI TRADE MARKS AND GEOGRAPHICAL INDICATIONS 401 could cause costs to those affected, and a longer period would be appropriate. A case-by-case assessment is appropriate in this regard, and should form part of the GI Committee consideration and decision-making when it comes to amending or omitting GIs, which the Commission considers should be no less than a year. RECOMMENDATION 12.2 The Australian Government should amend the Australian Grape and Wine Authority Act 2013 (Cth) and associated regulations to allow the Geographical Indications (GIs) Committee to amend or omit existing GIs in a manner similar to existing arrangements for the determination of a GI (including preserving the avenues of appeal to the Administrative Appeals Tribunal). Any omissions or amendments to GIs determined in such a manner should only take effect after a ‘grace period’ determined by the GI Committee on a case-by-case basis. Trade barriers stemming from GIs During the negotiation of TRIPS, there was disagreement between parties as to how much protection should be afforded to GIs. ‘Old world’ countries — principally those of the European Union — sought to include ‘strong form’ protection for GIs, which would have encompassed criteria around heritage, traditional methods and location. In contrast ‘new world’ countries pressed for GIs that pertained only to location (Jay and Taylor 2013). The result was a compromise, in that TRIPS represents stronger form GIs for wines and spirits, but weaker form GIs for other goods, as well as a number of exemptions to the applications of GIs: TRIPS has good grandfathering safeguards, such that even strong form GIs do not have to be adopted. TRIPS safeguards generic names and existing trademarks, protecting the community and trademark owners from expropriation. Article 24 provides that those with existing trademarks, or having used a name continuously for at least 10 years, may continue to use those names, with no provisos, even if they are for wines or spirits. (Moir 2015, p. 17) While TRIPS sets a minimum standard for GI protection, there is nothing preventing countries entering into bilateral agreements that strengthen GIs further. For example, the European Union has pursued trade agreements with other jurisdictions that seek to extend GI protection for particular goods. Of the 1400 registered GIs in the European Union, 173 were included in the Comprehensive Economic and Trade Agreement between the European Union and Canada in 2015, and around 60 were included in the EU–Korea trade agreement in 2010 (Borderlex 2015; O’Connor 2014). Such agreements impact on Australian firms, with dairy goods being particularly affected (box 12.4). As put by IP Australia: In some cases, protection of GIs can act as a barrier to trade in relation to a particular product. Terms may be GIs in one territory but generic terms in another. An example is feta, which is a protected GI in the EU, but a generic name for a type of cheese in Australia and other 402 INTELLECTUAL PROPERTY ARRANGEMENTS jurisdictions. Australia cannot export cheese under the name of ‘feta’ to the EU and, as the EU concludes more trade agreements that recognise feta as a protected term, the number of markets where Australians can sell product labelled as feta continues to narrow. The protection of generic terms as GIs leads to increased costs for consumers due to lack of competition and can damage the interests of legitimate producers and exporters. (sub. 23, p. 14) Box 12.4 Not having a Gouda time: Australian dairy and GIs Australia is a major exporter of dairy products, especially cheeses, to both the EU and other countries to which the EU has bilateral trade agreements (ABARES 2016). Both the system of GIs in the EU, and the terms included in these trade agreements have affected, or are likely to affect, Australian exports. Dairy Australia has launched numerous objections to the registration of GIs in the EU, which could be considered as descriptive or generic terms rather than appellations of origin. These include terms such as ‘gouda’, ‘gruyere’, ‘havarti’ and ‘danbo’. In some cases, Dairy Australia has objected on the grounds that the registration of such terms is at odds with other agreements, such as the World Trade Organization’s Agreement on Technical Barriers to Trade, where the nomenclature used under EU GIs are generic descriptors for trade. Generally speaking, Dairy Australia’s efforts have been largely unsuccessful in preventing registration of these GIs in the EU. The bilateral agreements between the EU and other countries have also affected Australian cheese exports. For example, the EU–Korea agreement prevents Australian dairy manufacturers from exporting ‘feta’ to Korea, instead having to label it as ‘white cheese’ (Harris 2014). Restraints on the use of terms such as feta, brie, gorgonzola, provolone, asiago, parmesan, munster and havarti have been a result of trade agreements between the EU and states including Singapore, Columbia, Peru, and Korea. In at least one case, however, there is some evidence to suggest that the resulting restrictions are being applied in different ways to different countries. In response to the EU-Korea agreement, the US government sought clarification on the terms of the agreement, and secured an assurance that some generic terms listed in the agreement — including particular varieties of cheese exported by the US — were not considered by Korea to be protected (Kim 2011). The Australian Government later received clarification on the use of particular terms in that agreement as well (Woo 2014), but not to the extent that allowed the export of Feta and some other cheeses to Korea. The European Union argues that GIs are a powerful tool for rural development and to increase economic activity; and that it would welcome the extension of GI protection in Australia to agricultural products (sub. DR495). Many Australian producers, however, have expressed alarm that terms currently viewed as generic in Australia could be ‘locked away’ as the result of current trade negotiations between Australia and the EU (AFGC 2016; Dairy Australia, sub. 38; NFF 2016). Given that these negotiations are not well advanced, it remains to be seen what sort of GI protections may be proposed, if any. Recent trade negotiations between the EU and Canada suggest that the former may be softening its position when it comes to GIs. As part of that agreement, a number of EU GI terms were specifically nominated that could be used by Canadian producers, provided that TRADE MARKS AND GEOGRAPHICAL INDICATIONS 403 they used terms including ‘like’ or ‘style’ — for example, ‘gorgonzola-style’ cheese for Canadian gorgonzola, whereas the EU product would have exclusive rights to be called ‘gorgonzola’. Disputes over GIs in the ongoing trade agreement negotiations between the EU and the United States have also led some to question whether a strong commitment to GIs is in the EU’s interests — as put recently by the German minister for food and agriculture: If we want to take the chance to make the most of free trading with the huge American market, we can’t protect every sausage and cheese as a specialty anymore. (Schmidt in Deutsche Welle 2015) Despite some evidence to suggest that the EU position on GI protection may be weakening, other countries have made efforts to bolster protections against GI provisions in trade agreements. While TRIPS provides some protections against terms that are considered to be generic from being locked away, it is at the discretion of a country as to whether to ‘sign their cheeses away’. The proposed Trans-Pacific Partnership (TPP) agreement adds further criteria to the administration of GIs — as put by the Department of Foreign Affairs and Trade (DFAT), on the TPP chapter dealing with IP: TPP Parties will provide robust and transparent application systems with opposition and cancellation procedures and grounds, including where a term would be likely to cause confusion with an already protected term, and where a term is generic. There are also safeguards for terms in translation and multi-component terms. The Chapter includes a clarification regarding GIs for wines, namely that a term need not be protected by a country where it is the same as the name of a grape variety that is considered to be generic in that country. With respect to new terms proposed for protection as GIs under international agreements, TPP Parties are required to provide additional transparency mechanisms and procedures for opposition or opportunities for comment. These provisions will enable Australian agricultural exporters to better protect access for their products in TPP markets. (2015b, p. 3) Other countries that have been affected adversely by EU GIs have also reacted positively to these proposed provisions in the TPP: U.S. dairy and wine producers have expressed support for new due-process and transparency provisions governing the recognition of GIs, particularly GIs that may conflict with trademarks or common food names in TPP markets. The U.S. Dairy Export Council and the National Milk Producers Federation, for example, state that prior FTAs left a vacuum in this area and that TPP’s new requirements provide an “equitable international model” for resolving disputes between GIs and trademarks. They favorably contrast this model with the “horse-trading protection” the European Union has sought for common names (such as asiago, feta, fontina, and gorgonzola) in trade agreement negotiations with Canada, Japan, Malaysia, Mexico, Peru, Vietnam, Singapore, and others. They state that the new provisions will “significantly strengthen” the ability of the United States to combat barriers and help to preserve market access opportunities for U.S. companies. (United States International Trade Commission 2016, p. 469) How GI protections between the EU, Australia and other countries develop will have a bearing on Australian exports, income and welfare. As with any trade agreement, the costs 404 INTELLECTUAL PROPERTY ARRANGEMENTS and benefits of its provisions on the community should be closely evaluated in advance (chapters 17 and 18). The Commission understands that IP Australia is currently undertaking research into the impacts of stronger-form GIs in Australia. This research aims to supplement the negotiations between the European Union and Australia, and should provide timely information on the costs and benefits of implementing stricter GI protections to make better evidence–based policy. A better understanding of the potential impacts of stronger form GIs and meaningful consultation with stakeholders are two essential precursors to inform the Australian Government’s negotiating position on GIs. This further reinforces the overarching need for the Australian Government to advance its approach to assessing the net benefits of potential negotiating positions and final form of bilateral and plurilateral trade agreements as outlined in chapter 17. TRADE MARKS AND GEOGRAPHICAL INDICATIONS 405 13 Plant Breeder’s Rights Key points • Plant Breeder’s Rights (PBR) are a sui generis regime of intellectual property (IP) rights used to protect innovations in plant breeding. PBR: – provide rights-holders with exclusive control over the commercial use of a registered plant (for up to 20 to 25 years, depending on the type of plant) – include a ‘breeder’s exception’ — reflecting the incremental and long-term nature of conventional plant breeding — which allows new plant varieties to be used immediately as inputs to further breeding programs. • PBR were introduced to encourage greater investment and private sector involvement in plant breeding, and promote faster rates of genetic gain. – The potential to profit under PBR from the development of successful new varieties has provided a strong incentive for private sector innovation. However, the evidence on rates of genetic gain is more mixed. • While plant breeders are broadly supportive of PBR, they have expressed some concerns with the scope of rights and with enforcement and compliance. – A particular concern is that PBR are not adapting to changes in technology, opening the door on unauthorised copying of protected plant varieties. • The Government is in the process of closing a loophole that allows breeders to copy and sell existing PBR protected varieties (so long as they do not register them). This will go some way to addressing plant breeder concerns. • Greater use of genetic marker technologies could also improve the quality of PBR protection by improving tests of difference between related plant varieties and reducing uncertainty about what is and is not an essentially derived variety (EDV). – IP Australia should continue to work with breeders to ensure that EDV rules are transparent, consistent and well understood, and adequately balance the interests of initial and follow-on plant breeders. • In conjunction with the introduction of PBR, the development of end point royalty (EPR) systems has been central to the success of commercial plant breeding in Australia. – In much of the agriculture sector, grower compliance with royalty systems is comparatively high, and revenue from EPRs now fully funds commercial plant breeding operations. – But there is scope for greater use and efficiency of EPR systems, particularly in the horticulture and nursery sectors. • In sectors where end point royalties are not well suited — such as perennial pasture and forage crops — industry groups should lead education and awareness campaigns to improve grower compliance with licensing conditions and reduce unauthorised use of protected varieties. PLANT BREEDER’S RIGHTS 407 13.1 Plant Breeder’s Rights: A primer Plant Breeder’s Rights (PBR) are a sui generis (dedicated) regime of intellectual property (IP) rights introduced to protect innovations in plant breeding. To be eligible for protection a plant variety must satisfy a number of criteria, including being ‘new, distinctive, uniform and stable’ (ACIP 2010a). PBR grant successful applicants a range of time-limited rights, including the right to exclude others from producing or selling protected plant varieties. In the case of eligible trees and grapevines, the period of protection is 25 years from the date of granting. For all other eligible plant species, the period of protection is 20 years (IP Australia 2014c). Why protect new breeds of plants? Developing new plant varieties is generally expensive, often takes many years and does not come with a guarantee of success. As noted by Kingston: It is characteristic of innovation in plant breeding to be predominantly incremental, proceeding by progressive enhancement of a particular variety through the introduction of desired traits from other sources. This requires investment at high risk, because even the process of trying such introductions may take several years, and in the end it may not result in an improvement that is commercially successful. (2007, pp. 295–296) When desirable new varieties do eventuate, absent some form of legal protection, commercially oriented breeders would be constrained in their ability to earn an adequate return on their investment. This is because ‘ … seeds have the peculiarly self-destructive (for breeders) characteristic that they quickly and automatically create their own competitors’ (Maskus 2012, p. 284). In effect, once a new crop or plant variety is sold to farmers or plant buyers for the first time, they generally have the means (in the form of saved-seed or cuttings) to grow subsequent crops or plants. The retained propagating material may be for their own use, or for trading or sharing with other farmers and growers. This potential for market failure partly explains why agricultural plant breeding has traditionally been undertaken within the public sector, with new plant varieties effectively made freely available to growers as they are developed (Sanderson and Adams 2008). The emergence of IP rights over plant varieties Global influences In the history of IP rights, PBR are a comparatively new development. Plant breeding has been practiced for thousands of years and scientific or evidence-based plant breeding has 408 INTELLECTUAL PROPERTY ARRANGEMENTS been practiced since the late middle-ages (Murphy 2007). 1 However, it was only during the 20th century that legislators around the globe began to implement systems of formal property rights over newly developed plant varieties (Sanderson 2011). The International Convention for the Protection of New Varieties of Plants 1961 (UPOV Convention) was a key development in this regard. The UPOV Convention outlined a sui generis regime of IP protection specifically adapted to the vagaries of plant breeding. As conventional plant breeding is largely incremental and uses procedures that are known and obvious, it was thought that plant breeding would not meet the ‘inventive step’ requirement for patentability. As a signatory to the World Trade Organization’s Agreement on Trade-related Aspects of Intellectual Property Rights (TRIPS), Australia is also required to protect new plant varieties, either by patents or by an effective sui generis system or by any combination thereof (Stewart et al. 2015). Domestic factors were also important The timing of these global developments and international agreements 2 also coincided with a period of economic policy reform in Australia. The latter favoured, among other things, a smaller role for government, greater competition in markets, and wider application of the user-pays principle. The introduction of plant variety rights, it was envisaged, would allow plant breeding to move out of the public sector (where it had traditionally been and where financial support was in decline) and into the private sector, where it would depend for its success on the willingness of growers to pay (Coles 2007; Kingwell 2003; Lindner 2004). Beyond stimulating a shift to greater private sector participation in plant breeding, it was thought that the introduction of plant variety protection (PVP) would increase plant breeding effort in Australia and lead to the development of new varieties that offered greater physiological adaptability and improved disease resistance (Ockwell 1982). An important added benefit for Australian farmers and horticulturists was expected to be an improvement in access to new varieties from overseas. Australia — like the vast majority of countries — implemented PVP based on the UPOV Convention model. Although the Australian Government initially passed an act supporting a system of plant variety rights in 1987, this legislation was substantially modified and extended by the Plant Breeder’s Rights Act 1994 (Cth) (PBR Act) — the basis of the scheme that persists to this day (Alexandra, Lee and Vanclay 2002). 1 Murphy (2007) characterises scientific plant breeding as being based upon at least a partial understanding of the traits that regulate the agronomic performance of crops, coupled with some knowledge of how to manipulate them. 2 Australia acceded to the UPOV convention in 1989, while TRIPS was negotiated at the end of the Uruguay Round of the General Agreement on Tariffs and Trade in 1994. PLANT BREEDER’S RIGHTS 409 Key differences to patents Although notionally similar to patents (table 13.1), the protection offered by PBR is considerably less extensive. A key feature of the UPOV regime is the ‘breeder’s exception’ which allows propagating material from a protected plant variety to be used for the purpose of breeding other varieties. Table 13.1 Plant breeder’s rights and patents A comparison of requirements PBR: Patents: To be a registrable plant variety, a variety must: • be newa To be patentable an invention must be: • have a breeder • be distinct, uniform and stableb • not have been exploited or have been exploited only recently. • a ‘manner of manufacture’ • novel • inventive • useful, meaning that it fulfils its promise • not secretly used beforehand by or on behalf of the patentee. Exemptions: • using a protected plant variety privately and for non-commercial purposes, or for experimentation, or to breed other varieties • farmers and growers can, in certain circumstances, retain seed to grow further crops. However, common law contracts between the PBR owner and other parties can be used to limit such uses. Exemptions: • limited, but includes acts for experimental purposes. Duration: • tree and vine crops – 25 years • all other crops – 20 years. Duration: • 20 years (extensions for pharmaceutical products are possible – see chapter 10). a ‘Newness’ implies the variety cannot have been commercialised for longer than a prescribed period for the species. b The UPOV Convention states that, ‘the variety shall be deemed to be distinct if it is clearly distinguishable from any other variety whose existence is a matter of common knowledge at the time of the filing of the application’. A variety is ‘uniform’ if, subject to the variation that may be expected from the particular features of its propagation, it is uniform in its relevant characteristics on propagation (PBR Act s. 43(3)). A variety is ‘stable’ if its relevant characteristics remain unchanged after repeated propagation (PBR Act s. 43(4)). Source: ACIP (2010a). The breeder’s exception allows for the inherently cumulative nature of conventional plant breeding, whereby existing varieties are used as the starting point for breeding new and improved varieties. According to UPOV: The existence of the breeder’s exemption optimizes variety improvement by ensuring that germplasm sources remain accessible to all the community of breeders … [and] helps to ensure that the genetic basis for plant improvement is broadened and is actively conserved, thereby ensuring an overall approach to plant breeding which is sustainable and productive in the long term. (2016) 410 INTELLECTUAL PROPERTY ARRANGEMENTS 13.2 PBR in practice PBR have encouraged the development of a more commercially-oriented plant breeding sector in Australia. According to the Australian Seed Federation, they have contributed to: … a highly competitive business culture to plant breeding where breeding organisations compete with each other for market share by developing and commercialising attractive varieties that improve grower returns. (sub. 42, p. 2) In the grains sector (worth $11 billion in 2015-16 and representing around 40 per cent of the value of agricultural crops in total), breeding and testing of new varieties has changed dramatically since the creation of the PBR legislation in 1994. State and university-based breeding programs have largely been replaced by fewer and larger privately owned breeding companies or public-private partnerships (Agtrans Research 2012b; ASF, sub. 42). In the early 1990s, there were nine wheat-breeding programs in Australia that were either university based or based within State Government agriculture departments. Today there are four major wheat-breeding companies in Australia and one smaller specialist company (Alston, Gray and Bolek 2012; GRDC 2011). In pasture crop breeding, the introduction of PBR has facilitated the entry of private seed companies to the market and led to improved access to overseas-bred material for use in Australian breeding programs (RIRDC 2014; PGG Wrightson Seeds, sub. 82). Outside agriculture, Jarakad (sub. 78) note that PBR have encouraged investment in ornamental grasses and other plants, including for export and licensing to overseas growers. Public breeding programs remain important While PBR have been successful in encouraging more commercial breeding, some plant breeding has stayed within the public sector (such as in State Government agriculture departments, universities, and the CSIRO). Public sector breeding often occurs within commercially oriented business units in alliance or partnership with private sector businesses. For example: • government and university based plant breeding remains important for oats, triticale and durum wheat (Agtrans Research 2008), and for horticulture crops, including apples, stone fruits, nuts, and grape rootstock • public/private partnerships in crop improvement are important for forage crops such as lucerne and clover (RIRDC 2014), while breeding programs for pulses are being undertaken as joint ventures involving State Government departments of agriculture and Australian universities (Agtrans Research 2012a) • cotton breeding in Australia is almost exclusively performed by the CSIRO in collaboration with industry bodies and the private life-sciences company, Monsanto. This alliance supplies genetically modified seed to Australian cotton growers while CSIRO also exports non-GM (genetically modified) cotton varieties. PLANT BREEDER’S RIGHTS 411 In general, public breeding is focused on crop and plant varieties that have smaller potential markets and for which the private sector is less inclined to invest. However, as the PBR system has matured (and as compliance with royalty collection regimes has increased), private sector plant breeders have shown greater interest in expanding the range of crops they offer. Governments and grower organisations also remain involved in pre-breeding and basic research — or the development of ‘essential plant breeding infrastructure’, to use the terminology of Lindner (2004). This is in part because this type of research has a larger public good component and is typically less attractive to private sector businesses. PBR are now routinely sought by plant breeders Plant breeders (in both the public and private sectors) have generally embraced the use of PBR to protect their IP. The registration of new plant varieties under the PBR regime is now commonplace among plant breeders. New PBR applications in Australia are currently around 350 per year and are roughly evenly split between domestic and overseas-based breeders (figure 13.1). Nursery and ornamental plants dominate applications by number, accounting for around one half of annual applications. However, the nursery-ornamental sector represents a small share of the total sales value of plant varieties covered by PBR. 3 In contrast, agricultural plant varieties account for a small share of PBR applications, but represent a much larger share of the value of production and exports derived from PBR protected varieties. 3 In 2013–14, the value of nursery and garden sector production was estimated to be $660 million, while the value of agricultural crop production was $30 billion (Plant Health Australia, (2016); ABARES (2015b)). 412 INTELLECTUAL PROPERTY ARRANGEMENTS Figure 13.1 Applications for Plant Breeder’s Rights By applicant residence status 500 Resident: private sector Resident: public sector 400 Non-resident 300 200 100 0 1988 1992 1996 2000 2004 2008 2012 Source: Commission estimates using IPGOD (2016 edition). Though not all plant breeders rely exclusively on PBR Not all crops or plant varieties need protection in order to stimulate private sector interest in plant breeding. Some plant varieties — notably hybrid varieties of sunflowers, corn, soybeans and canola — cannot usually be grown from saved-seed, and hence have natural protection against unauthorised use. And hybrid seeds cannot be reproduced on-farm because the process used to derive them requires two different parent lines, which are generally kept secret and closely guarded by seed companies. Growers must purchase new seed each year in order to regrow these crops. This in-built protection stimulates a commercial breeding culture for these crops, independent of the existence of PVP systems. 4 Some plant breeders also rely on other forms of IP protection. For example, GM varieties of cotton and canola embody patent-protected genes that improve resistance to herbicides or improve resistance to insect and disease threats. In these cases, the patent system provides protection against unauthorised use. 4 Murphy (2007) reports commercial interest in the idea of producing hybrid maize in the USA as far back as 1825, although it took another century to realise the concept as a viable commercial operation. Today, hybrid corn is the norm in the United States, and the majority of Australia’s canola crop is currently based on hybrid varieties. PLANT BREEDER’S RIGHTS 413 In the ornamental sector, not all plant breeders seek PBR protection over their varieties, often because the markets for some of these plants are small (and certainly much smaller than markets for new grain crop varieties) and breeders believe that it is not worth the cost (Prescott and Christie 2015). In some cases, ornamental breeders use trademarks to protect their IP in preference to PBR, or rely on first mover advantage in what is often a fashion-oriented market. Plant breeding — bigger but not necessarily better? While PBR have stimulated greater investment and private sector participation in plant breeding, it is less clear that they have delivered more and ‘better’ plant varieties compared with the counterfactual. 5 Stakeholder views are mixed on the question of whether PBR have driven improvements in the quality of new varieties. Thomson (2014, p. 22) finds a negative outcome on agricultural crop breeding in the post-PBR era, and notes that ‘farmers are increasingly choosing to cultivate older varieties rather than quickly adopting newly released varieties’. This is attributed to a lack of agronomic performance in recent plant varieties. On the other hand, a major commercial wheat breeder, Australian Grain Technologies (sub. 15), has argued that recent increases in investment in plant breeding capacity are beginning to translate into greater rates of genetic gain in new crop varieties. Similarly, PGG Wrightson Seeds (sub. 82) point to stronger yield gains in forage and pasture crop varieties following the introduction of PBR and the entry of private sector plant breeders. International evidence, while limited, supports the notion that PBR have led to improvements in the quality of new plant varieties. For example, a UPOV report found the introduction of UPOV-based plant variety rights regimes (Australia was not included in the analysis) had a positive influence on the quality and quantity of new plant varieties registered. The report states: Put simply, farmers, growers and breeders have access to the best varieties produced by breeders throughout UPOV member territories. … It is also important to note that membership of UPOV provides important technical assistance and maximizes opportunities for cooperation, which enables PVP to be extended to the widest range of plant genera and species in an efficient way thereby enabling the benefits to be maximized. (2005, pp. 16, 19) On the other hand, Maskus found: … little systematic econometric evidence about the roles played by [intellectual property rights] in promoting innovation and diffusion in new plant varieties and biotechnological crops. (2012, p. 285) 5 The counterfactual being what would have happened in the absence of the introduction of PBR and assuming agricultural plant breeding had largely remained in the public sector with funding from general revenue and/or levies on growers. 414 INTELLECTUAL PROPERTY ARRANGEMENTS Maskus concluded, nevertheless, that the benefits of adopting PVP systems probably did not hinder technical progress in plant breeding. Trends in crop yields and farm productivity Another way to assess the impact of PBR on genetic gain in plant breeding is to consider longer term trends in observed crop yields and farm productivity. While many other factors influence crop yields and farm productivity from year to year, particularly seasonal conditions, longer term trends in crop yields are heavily influenced by genetic gain. Higher rates of genetic gain in new crop varieties can raise on-farm productivity by increasing crop yields, by reducing aggregate inputs (say, because new varieties have higher pest or disease resistance, reducing the need to apply pesticides and herbicides), or by some combination of both. On average, crop yields and crop farm productivity have increased over the past five decades (figure 13.2). In considering the role that PBR may have played in this, it is important to note that, although the PBR Act came into existence in 1994, it is unlikely that this policy change could have had any demonstrable effect on average crop yields until more recently. Figure 13.2 Crop sector productivity growth and crop yieldsa 2.5 250 200 150 Index Tonnes per hectare 2.0 1.5 100 1.0 Wheat yield 50 Coarse grains yield 0.5 Total Factor Productivity (index, right hand axis) 1970 1975 1980 1985 1990 1995 2000 2005 2010 0 a ‘Total factor productivity’ is in index form, and measures changes over time in the ratio of output to inputs, where both output and inputs are measured in quantity (volume) terms. The index is estimated for broadacre agricultural crop producers only, such as growers of wheat, coarse grains, oilseeds and pulses. Crop yields are averages across Australia. All series are five-year moving averages. Sources: ABARES (2015a, 2015b). PLANT BREEDER’S RIGHTS 415 As discussed in more detail below, the time taken to breed new agricultural plant varieties typically ranges from 8 to 22 years, depending on the plant. Hence, the impact of the introduction of PBR on average crop yields and farm productivity would be unlikely to be manifest until well into the 2010s. This more truncated period has been characterised by highly variable crop yields with little trend growth. However, total farm productivity and average crop yields reached record levels in 2011-12, and this may be an indication that underlying improvements in plant varieties in the PBR era (combined with a return to more favourable growing conditions) are now paying dividends to growers. More time will be needed to definitively answer this question. 13.3 Are reforms to PBR required? As noted earlier, the Australian system of PBR is popular with breeders, and has played an important role in the development of commercial and private-sector plant breeding. More broadly, UPOV-based PVP regimes, such as Australia’s, have proven popular around the globe, and the number of counties signing on to the UPOV convention has grown steadily over time. However, PVP regimes have their limitations, and some critics have gone so far as to suggest that rapid technological change could eventually make them redundant. A major concern is with the scope of rights, and whether PVP regimes and their ‘breeder’s exception’ can provide sufficient protection against copying, particularly when plant breeding technologies and practices are changing rapidly. These concerns are often associated with calls for substantive changes to the rules used to determine whether a newly developed plant variety is sufficiently different to the ‘initial variety’ from which it was derived. Another concern with PVP regimes relates to the enforcement of rights, particularly in regard to the payment of royalties or licences fees by plant growers. For crops and plants that can be regrown from saved seeds or cuttings, commercial plant breeding may not be viable if compliance is low and enforcement is costly or difficult. The following sections examine these two issues in turn. Concerns with the scope of rights The plant breeder’s exception is a key feature of UPOV-based regimes, and allows protected varieties to be used immediately as a source of variation in other breeding programs. The breeder’s exception is thought to bolster long-run genetic gain by allowing plant breeders to freely share in the available germplasm, and reflects the time-consuming and incremental nature of conventional plant breeding (box 13.1). 416 INTELLECTUAL PROPERTY ARRANGEMENTS Box 13.1 Plant breeding characteristics Breeding time and market life The time taken to produce marketable new plant cultivars using conventional (crossing and selection) plant breeding strategies varies according to a range of factors, including crop type, the desired traits to be introduced, the crossing type and the propagation system. The ‘market life’ of an individual plant variety also varies, depending on factors such as crop type, agronomic performance and consumer tastes. Some agricultural plant varieties can be very popular with growers initially, but quickly succumb to disease or develop other negative traits. Other varieties lose favour relatively quickly because better varieties are developed. For fashion-oriented nursery plants, the effective market life can also be quite short, mirroring shifts in tastes and preferences. On the other hand, some horticultural tree crop varieties — such as granny smith apples and imperial mandarins — can stay in the market for many decades. Indicative breeding and market life times for major crop and plant types in Australia are as follows: Breeding time Market life Cereals, coarse grains, pulses, oilseeds, sugarcane 8 to 14 years 9 years Rice Pasture and forage crops Ornamental and nursery plants Tree and vine crops 7 to 10 years Up to 15 years 3 to 9 years 12 to 22 years 15 years + (not available) 3 to 8 years (not available - decades in some cases) Breeding costs For major agricultural crops, the cost of bringing a new variety to market can be in the millions of dollars. For example, Australian Grain Technologies (sub. 15) report an average cost of breeding a new cereal or oilseed variety of around $3 million. The current annual investment in wheat breeding programs in Australia is estimated to be around $45 million, while expenditure on oilseeds breeding averaged $3 million per year between 2002 and 2009. The annual rice breeding program in Australia costs around $3 million per year (in 2016 dollar terms), with a new variety released every two years, on average – an implied cost of $6 million per variety. Ornamental plant breeding in Australia is considerably smaller in scale and aggregate cost. A national survey found five breeding programs had budgets in excess of $100 000 per annum, four spent between $50 000 and $100 000 per annum, and four more between $10 000 and $50 000. Sources: ACIP (2010a); Government of Western Australia (2016); SRA (2015); PGG Wrightson Seeds (sub. 82); Ricegrowers’ Association of Australia (2016); Nursery & Garden Industry Australia (2009); pers. comm. Ben Ovenden, NSW DPI, March 2016. However not all plant breeding is time-consuming, and it is possible to develop new varieties that are physically ‘distinct’ but identical in agronomic terms to existing protected varieties. This raises the prospect or possibility of copying or free-riding on the efforts of breeders engaged in longer-term conventional (crossing and selection) breeding programs. Lawson (2014) argues that early (pre-1991) versions of the UPOV Convention did not provide much protection to initial breeders, as they allowed very minor differences PLANT BREEDER’S RIGHTS 417 between protected varieties. The rise of new molecular technologies in the 1970s and 1980s also heightened concerns among breeders that the scope of breeder’s rights was too narrow. In particular, breeders were concerned that molecular technologies could be used to rapidly alter existing elite plant varieties in ways that were sufficient to achieve ‘distinctness’, and thereby earn the ‘new’ varieties PBR protection in their own right. In the words of UPOV: … the development of genetic engineering [at the time] offered new tools with the possibility to transfer a single gene to an existing variety and to get new varieties very close to the initial one in a very short time. (2013) This was leading to a decline in the ‘quality’ of protection (UPOV 2013). The importance of essentially derived varieties In response to these concerns the 1991 UPOV Convention introduced the concept of essentially derived varieties (EDVs). This expanded the scope of a breeder’s right to any new variety that was selected on the basis of a very minor difference (Lawson 2014). The intention of the EDV concept was to reduce the incentive for free-riding on the efforts of initial breeders, without limiting the scope for genuine follow-on innovation via the breeder’s exception. Sanderson notes that one of the specific reasons given by UPOV to justify introducing the EDV concept was that it would: … prevent the exploitation of mutations of protected varieties and varieties that had undergone a minor or trivial change in relation to the initial variety, for example by using biotechnology, without the first plant variety rights holder being able to share in the profits. (sub. DR208, p. 4) However, the UPOV Convention does not specify the precise rules that member countries must use to assess essential derivation. Nor does it provide guidance on how, and under what conditions, an initial breeder and a follow on breeder might ‘share in the profits’ in cases where essential derivation was, in fact, found to exist. Ultimately, the determination of essential derivation is left to the administrative discretion of the regulating authority in each member country (Godden 1998). In Australia, the PBR Act allows any new variety to avoid being declared an EDV as long as it exhibits an ‘important’ feature not exhibited in the initial variety. ‘Important’ features are not defined in the PBR Act, but are interpreted by IP Australia as ‘significant changes that affect performance, value or place in the market’ (ACIP (2010a), Waterhouse (2013)). The intention of the ‘important’ features test is to ensure that: … those derived varieties that do have an important difference from the original variety do not have PBR on the original variety extended to them, thus encouraging improvements to existing varieties. (ACIP 2010, p. 63) 418 INTELLECTUAL PROPERTY ARRANGEMENTS The inclusion of the ‘important feature’ rule in determining EDV status appears to be unique to Australia and gives rise to some important differences: • a mutation or genetically modified plant would generally not be declared an EDV in Australia, whereas under the UPOV guidelines and in PVP regimes in other jurisdictions, mutations and genetically modified plants are by definition considered to be EDVs (Kock, Porzig and Willnegger 2006) • so long as the test is deemed to be met, the follow-on breeder is granted full rights over the ‘new’ variety. These points of difference mean that, in some cases, the balance of rights is tipped in favour of follow-on breeders over initial breeders, particularly in an era where new technology can be used to rapidly introduce traits of interest into existing varieties. In essence, the time it takes for a competitor’s derived variety to enter the market is a key determinant of the effective protection available to the breeder of the initial variety. In their 2010 review of PBR, the Advisory Council on Intellectual Property (ACIP) acknowledged concerns that the ‘important feature’ test could discriminate against the initial breeder, noting stakeholder concerns that: … the use of the term ‘important’ allows the breeder of a variety where one important change has been made to receive a disproportionate share of the return on investment in the breeding of the variety. (2010a, p. 66) Similarly, in relation to the random discovery of ‘sports’ 6 in horticultural plants, Department of Agriculture and Food Western Australia (DAFWA) observed: DAFWA recognises the importance of rewarding the breeder of a discovered sport. However, DAFWA does not believe that it is appropriate if that sport demonstrates an important difference that that entitles the breeder of the derived variety to ignore the importance of the initial variety and of the work involved in producing that variety. Without the initial variety there would be no derived variety. (ACIP 2010a, p. 68) However, setting the threshold of difference for an EDV too low could lead to an ‘endless cascade of PBR over all derived varieties’, which could also discourage investment in plant breeding (ACIP 2010a, p. 66). This led ACIP to conclude that: … a [PBR] system in which the owner of the derived variety is in some way required to share its benefits with the owner of PBR on the initial variety would make the scope of protection provided by PBR too broad and out of proportion with the level of innovation achieved. Such a system would also be difficult to implement. (2010a, p. 70) 6 ‘Sports’, ‘breaks’ or ‘chimeras’ can arrive in plants by freaks of nature. A mutation occurs causing a random change in the plants’ chromosomes brought about by insect damage, the weather or other factors. New varieties can be developed reasonably quickly from such mutations. In contrast, new varieties that are arrived at via conventional plant breeding approaches involving crossing and selection tend to take considerably longer to produce. PLANT BREEDER’S RIGHTS 419 Evidence raised in submissions to this inquiry Participants in this inquiry have again raised concerns about the scope of PBR protection, arguing that the existing rules on distinctiveness are interpreted too leniently by IP Australia, and effectively allow unauthorised copying of existing cultivars. Concerns have also been raised that IP Australia’s interpretation of rules on essential derivation favours follow-on breeders over initial breeders, and that this compromises the ability or willingness of breeders to engage longer-term breeding programs necessary to address complex plant problems, such as improving drought or frost tolerance, increasing yields, or improving resistance against pests and diseases. In regard to copying, PGG Wrightson Seeds (sub. 82) note that for a number of important pasture crop species, some seed companies can make simple in-paddock crosses between competitor cultivars and market the subsequent seeds under their own name. This undermines their ability to earn a return on their investment. In a post-draft submission they went on to argue that the PBR system is, ‘primarily concerned with morphological traits, most of which are economically inconsequential for the financial wellbeing of farmers’, and that determinations of essential derivation in forage and pasture crops too readily favour cosmetic differences over important economic differences (sub. DR547, p. 3). In regard to balancing the interests of initial versus follow-on breeders, Prescott criticises IP Australia for being too ‘narrow’ in their view of distinctiveness and derivation, arguing that: … all mutations and GMOs [genetically modified organisms] by definition are EDVs as they are predominantly derived from the initial variety, particularly when they retain a very high genetic conformity to the initial variety. The current provision that a mutation or a GMO cannot be declared a EDV whenever it contains an important characteristic which differentiates it from the initial variety, does not take into consideration that the commercial viability of a plant variety depends upon a series of important characteristics, and that a change in one regardless as to how significant it is, does not preclude the “breeder” of the mutation or GMO from exploiting the initial breeder’s genetic inheritance of all the other important characteristics. (sub. DR470, p. 1) In a similar vein, AGT have argued that patent-protected genome editing technology could be used to insert novel traits into PBR protected plants, such that: [The patent owner] would then have rapidly achieved an improved version of the original PBR owner’s variety and have no obligation to seek a license from that original owner or to return the new variety back into the breeding community via PBR. … The ‘free ride’ of a patent owner to use the PBR IP of [a] plant breeder puts at risk the level of investment a plant breeder is likely or willing to make. It is unreasonable that a plant breeder should invest 10 to 12 years developing a new variety obtaining PBR and then having the variety ‘sniped’ by a patent owner thereby preventing the original breeder [from] obtaining a fair return on their investment. This is a MAJOR concern for ongoing investment in plant breeding in Australia and the improvements in performance required by Australian farmers. (sub. 15, pp. 3–4) 420 INTELLECTUAL PROPERTY ARRANGEMENTS On the other hand, owners of patented GM technologies are concerned that any moves to reduce the level of protection that patent holders currently receive would allow plant breeders to appropriate their investments. For example, CropLife Australia argued: If existing patent protection were to be weakened or removed, then there would be nothing stopping a competitor from cross-breeding [a] GM trait into a different variety and claiming plant breeder rights. This process would take one growing season and would completely undermine the original technology provider’s investment. With such a significant ‘free rider’ effect, no company would invest in developing the technology in the first place. (sub. 25, p. 8) Surveys undertaken on behalf of CropLife suggest that the mean cost associated with the discovery, development and authorisation of a new biotechnology derived crop trait was in the order of $136 million and the mean time taken was just over 13 years (Phillips McDougall 2011). Are changes to PBR arrangements needed? Participants in this inquiry have raised a number of suggestions for reforming the PBR regime in order to address perceived problems with the scope of rights. The solutions focus on two areas — improving tests of difference using developments in molecular marker technologies; and changing words in the PBR Act to make clearer what is and is not essential derivation. In respect of better using technology, PGG Wrightson Seeds advocated greater use of DNA-based testing by the regulator, noting that: Given the relatively recent emergence and rapid progression of molecular and DNA based technologies capable of determining the relatedness of plant populations, the ability to utilise DNA technology to inform [PBR] decisions and aid enforcement must be considered a high priority. (sub. 82, p. 10) The Institute of Patent and Trade Mark Attorneys of Australia also argue for greater use of technology in tests of new varieties, noting that: The current Australian statutory regime for PBR does not take advantage of the available technology. In particular, the availability of affordable DNA tests would make determinations of distinctiveness of an allegedly ‘new plant variety’ simpler and more determinative and would be a straightforward way to confirm the asserted breeding of the new plant variety. (sub. 73, p. 16) PGG Wrightson Seeds (sub. DR547) argued that a relatively recent genotyping technology — Genotype-by-sequencing (GBS) — has been demonstrated to be an effective tool for both determining which variety a sample of seed is from, and for determining the relatedness of varieties of perennial ryegrass. They believe GBS will have broader application in other forage crop species, and could assist IP Australia to better enforce PBR in this sector. PLANT BREEDER’S RIGHTS 421 In the nursery/ornamental sector, Prescott and Christie (2015) suggest the use of economic or agronomic differences in tests of distinctiveness, rather than relying predominantly on physical or morphological differences. PGG Wrightson Seeds (sub. 82, p. 14) also argue for a PBR system that provides greater protection over economic characteristics ‘instead of differentiating products based on easily measured and altered morphological traits of negligible economic consequence’. To deal with the possible threat from patent-related ‘sniping’ (whereby follow-on breeders might build on recently developed PBR-protected varieties using techniques or processes that allow them to patent their resulting plants, thereby avoiding the breeder’s exception themselves) AGT suggest: The PBR Act state that ‘The use of a PBR variety with a patented technology will require a license from the PBR owner unless the breeder (creator) of the new variety, that incorporates the patented technology, makes the new variety and patented technology freely available to all plant breeders under the breeders rights exemption of the Act[’]. (sub. 15, p. 6) In their 2010 review of PBR, ACIP recommended that the test of ‘important features’ in the PBR Act used to differentiate EDVs should be replaced by a test for ‘essential characteristics’. They considered that the revised wording would reduce uncertainty and make it easier for the regulator to assess essential derivation. The then Australian Government agreed that the current test was not well understood, but rejected the recommendation, believing that the solution to the problem was through education and awareness campaigns (Stewart et al. 2015). Despite the problems and concerns raised by participants, it is worth noting that, in many ways, Australia’s PBR regime has proven to be resilient and flexible in the face of change. In a review of the use of PBR, Sanderson and Adams (2008) argue that amendments and extensions to the Act have helped the system contend with advances in science and controversies over biopiracy and enforcement, as well as deal with specific legal disputes. As a result, they argue that the PBR system is ‘fluid and dynamic and is able to respond to various controversies and challenges’ (Sanderson and Adams 2008, p. 994). Further, there have only been three decisions by the regulator regarding alleged EDVs since the inception of PBR in 1994. Two cases related to non-agricultural crops (both turf varieties), while the third case involved a claim brought by an apple breeder against himself. All cases were rejected by the registrar of PBR. (pers. comm. IP Australia, January 2016). The strategic use of patents to circumvent the breeders exception in PBR appears to be a potential rather than actual problem. CSIRO (sub. DR575) have noted that the use of a new technology to add a trait to an existing cultivar will still take many years (to test for the effectiveness of the new trait, and to bulk up commercial quantities of an improved line), during which time the owner of the original variety can generate a return on their investment. 422 INTELLECTUAL PROPERTY ARRANGEMENTS However, the lack of EDV claims or legal action may not fully reflect the level of copying and free-riding in Australian plant breeding. As the Institute of Patent and Trade Mark Attorneys of Australia submitted: … the pecuniary relief available for infringement is minimal compared to the breeding and application costs and to the costs of pursuing an infringement action. IPTA understands that all of these costs are in fact a disincentive to pursuing PBR protection. (sub. 73, p. 15) Addressing a significant loophole is an important first step Notwithstanding the mixed evidence on the extent of the problem, avoidance of policy mistakes requires remaining alert to potential improvements. Submissions to this inquiry support the view that, if left unchecked, problems could compromise the effectiveness and adaptability of the system. One reform to PBRs that warrants immediate action is amending the PBR Act to enable EDV declarations to be made in respect of any plant variety, not just varieties that are nominated for PBR protection. This was a recommendation made by ACIP following their review of PBR in 2010, which was accepted by the then Government (Australian Government 2010). The reform would remove a loophole in the PBR Act that allows downstream breeders to copy or make minor or cosmetic changes to existing PBR-protected varieties and then freely market the resulting seeds or plants, simply by not seeking to register the copied varieties. (This reform is of particular interest to breeders of outcrossing species, such as pasture and forage crops, which are relatively easy to copy (PGG Wrightson Seeds, sub. DR547).) As noted by ACIP, a declaration process cannot commence unless the derived variety is subject to a PBR application or grant (2010). The Commission considers that this recommendation, which was supported by participants in this inquiry 7 and the ACIP review, should be implemented as soon as possible. The Commission understands that the Australian Government is currently in the process of developing legislation to amend the PBR Act along these lines. RECOMMENDATION 13.1 The Australian Government should proceed to implement the Advisory Council on Intellectual Property’s 2010 recommendation to amend the Plant Breeder’s Rights Act 1994 (Cth) to enable essentially derived variety (EDV) declarations to be made in respect of any variety. 7 Including: Law Council of Australia (sub. DR490); PGG Wrightson Seeds (sub. DR547); International Association for the Protection of Intellectual Property (sub. DR551); Swinburne University of Technology (sub. DR557); CropLife Australia (sub. DR561); CSIRO (sub. DR575); and FICPI Australia (sub. DR581). PLANT BREEDER’S RIGHTS 423 Technology is already being used to improve tests of difference As highlighted above, another mechanism for improving the operation of PBR is the use of new molecular technologies for characterising and measuring plant differences. IP Australia has confirmed that DNA tests, such as those using molecular markers, are already available for use in the PBR system, and noted that in some cases DNA tests are making the examination of new plant varieties more efficient (pers. comm. IP Australia, 23 February 2016). UPOV is also continuing to investigate and evaluate the use of genetic marker technologies to improve tests of difference between plants. While molecular marker technologies and associated rules for establishing threshold genetic differences between related plant varieties are likely to play an increasing role in future tests of derivation, they are unlikely to resolve all questions of difference. Janis and Smith (2007) report that the technical literature expresses enthusiasm about the potential for molecular data to test differences, but also contains reservations about implementation complexities. According to Lawson (2014) ‘it is certain … that EDV is not merely a technical question that can be resolved with a technical answer, such as a statistical index or a DNA sequence’. Improved transparency and communication will also help While these changes will go some way to help address some concerns with the operation of PBR, they will not be a complete solution. The EDV concept is complex, and uncertainty and differences of opinion will likely remain unresolved. According to the Chairman of the Technical Committee of UPOV, the rules on EDVs have been ‘deeply discussed’ since the concept was first introduced in 1991, and are still ‘a source of discussion’ (Guiard 2013). Plant breeders and IP Australia should continue to work collaboratively to ensure that the implementation of rules regarding distinctiveness and essential derivation are consistent, transparent and well understood. Regular review and communication on the impact of new technologies — both new breeding technologies and new technologies for testing plant differences — will enhance the adaptability of the system, and reduce uncertainty regarding what is and is not copying or essential derivation. As noted by PGG Wrightson Seeds: Given the recent progress in molecular techniques, we suggest IP Australia should seek to regularly review how contemporary techniques can contribute to the objectives of PBR. (sub. DR547, p. 6) A new consultative committee on PBR — comprising plant breeders and other industry experts — is currently being formed (pers. comm. IP Australia, 2 February 2016). This body is ideally placed to facilitate communication and consultation on new technologies and their impact on the scope of rights. 424 INTELLECTUAL PROPERTY ARRANGEMENTS Concerns with compliance and enforcement A second area of concern with Australia’s PBR system relates to issues of enforcement and compliance. A sustainable plant breeding sector depends on both the willingness and ability of growers to pay license fees or royalties for protected varieties, as well as efficient systems for collecting revenue. Hence, the mechanisms by which plant breeders are compensated for the varieties they develop are also important (Alston, Gray and Bolek 2012). Inquiry participants have suggested a number of specific reforms targeted at improving grower compliance with licensing and royalty arrangements, and improving the efficiency with which revenue is collected. Seed royalties (one-off payments for seeds or plants) can be an effective way to appropriate a return on plant breeding, particularly for those crops or plant varieties where growers need to purchase new supplies of seed or new plants each time they wish to grow them. As noted earlier, hybrid crops cannot generally be regrown from saved-seed and in these cases a seed royalty is the standard model used by breeders to generate revenue. However, for self-reproducing cereal crops such as wheat and barley, farmers need only purchase seed once as they can then rely on saved-seed to produce future crops. 8 This means there is greater scope for revenue leakage under a seed royalty model, undermining the ability of commercial breeders to earn a return on their investment. The use of comparatively large up front seed royalties for perennial or self-pollinating crops can also be financially challenging for growers who may feel that they bear a disproportionate share of the production and marketing risk under such an arrangement. This problem may be particularly acute in Australia given its climatic variability and the frequent chance of large-scale crop failures. And in the case of horticultural tree and vine crops, the decision to adopt a new plant variety generally represents a costly long-term commitment on the part of growers. It can take many years before the crops mature and begin to yield, and there is uncertainty regarding longer-term production and market (price) outcomes. A high one-off seed or plant royalty adds to the cost of investing in the new variety and may discourage some growers from making an otherwise advantageous change. To avoid these problems, breeders are increasingly relying on end point royalty (EPR) systems — a system of payments based on crop production rather than seed or plant sales. Payments can be based on the area planted, the volume, quantity or weight of grain, or crop production sold (GRDC 2008). Payments continue for as long as the variety is grown, or for as long as the variety remains under the protection of a PBR. Contracts are generally required to enforce payment conditions, and payment systems often rely on centralised collection agencies to increase efficiency (Sanderson 2007). 8 Growing crops from saved-seed is deeply embedded in Australian farming culture. According to the Australian Seed Federation (sub. 42) grain growers in Australia use retained seed to plant in excess of 90 per cent of the crop each year. PLANT BREEDER’S RIGHTS 425 EPRs have become an integral feature of the plant breeding landscape in Australia (Lawson 2013b). According to Variety Central: Both growers and the extended grain industry are recognising the value of an EPR system to the Australian grains market as a risk sharing mechanism and also as the most effective way to support high quality breeding programs. (2016a) In 2014, there were 202 agricultural crop varieties in the Australian market that required payment of an EPR, including in cereals, pulses and brassicas (McGrath 2014). Approximately 70 per cent of the Australian wheat harvest in 2009-10 was made up of EPR bearing varieties (ASF, sub. 42), and according to Alston, Gray and Bolek (2012) EPRs have become ‘the primary source of funding for wheat breeding in activities in Australia … and now provide sufficient revenue to support all downstream commercial wheat-breeding activities.’ In 2015, EPRs collected on wheat and barley crops generated revenues of around $50 million. A significant proportion of this revenue flows back to plant breeders. It also appears that Australia is well advanced in using EPR regimes to fund plant breeding compared with other agricultural exporters. Apart from encouraging domestic plant breeders, the possibility of earning reliable EPR revenue streams on successful new plant varieties has encouraged overseas plant breeders to invest in Australia. For example, in 2014 Bayer CropScience opened a $14 million wheat and oilseeds breeding centre at Longerenong College, near Horsham, Victoria. Royalty rates were increasing, but appear to have slowed While EPRs have become an integral feature of the PBR system, concerns have been raised about a number of issues, including increases in royalty rates over time and the possibility that too large a share of royalty payments is being appropriated by commercialisation agents or other intermediaries in the supply chain (Coles 2007). Royalty and license fees for PBR protected varieties did increase relatively quickly between 2000 and 2012 (figure 13.3). However, the increase likely reflects the maturing of a new system of funding for plant breeding. In the early days of the system, PBR protected varieties had to compete with freely available varieties or with varieties that were covered by PBR but did not attract license fees or royalties. This limited the extent to which commercially oriented breeders could charge royalties (Alston, Gray and Bolek 2012). As successful new varieties were developed however (and as growers became more willing to support plant breeding by paying royalties) royalty rates tended to increase. More recent data suggest that EPR rates for newly released varieties have stabilised. For example, EPR rates for new wheat varieties in 2015 ranged from $3 per tonne to just over $4 per tonne, similar to rates observed over the last four to five years. 426 INTELLECTUAL PROPERTY ARRANGEMENTS Figure 13.3 End point royalties for wheat by year of variety release 5.0 4.0 3.0 $/t 2.0 1.0 0.0 1996 1998 2000 2002 2004 2006 2008 2010 2012 2014 2016 Sources: Arnold (2015); Gray and Bolek (2012); Variety Central (2016b). Competition should constrain the capacity of breeders to generate excessive profits from PBR protected varieties and ensure that the benefits arising from the development of superior new plant varieties are shared between breeders and growers. That being said, the long lead times involved in plant breeding could lead to individual breeders holding market power from time to time, particularly if they produce new varieties that are well above average in terms of desirable characteristics (yield, disease resistance, ease of growing etc.). Even then, the returns available from growing the next best available varieties set limits on the maximum EPR rates that individual breeders can charge. Grower compliance is key to the success of commercial plant breeding Despite the effectiveness and popularity of EPR systems as mechanisms to fund plant breeding, there have been ongoing concerns regarding grower compliance and the efficiency with which royalties are collected. Parts of the agriculture industry have successfully sought to improve EPR compliance and collection efficiency over time. Anecdotal and other evidence suggests that cereal and other grain crop breeders are now recovering around 80 per cent of possible EPR revenue, a good outcome given that the cost of collecting the remaining 20 per cent is likely to be reasonably high. However, misrepresentation of varieties and refusal to pay royalties remain a concern in select sections of agriculture. PGG Wrightson Seeds (sub. 82) argue that for cross-pollinating species like forage crops, free riding by other seed companies and over-the-fence trading by growers deprives the breeder of a return on their investment. In PLANT BREEDER’S RIGHTS 427 the nursery and ornamental sector, breeders and right holders remain concerned about the difficulty of obtaining evidence necessary to bring infringement actions (Prescott and Christie 2015). A number of recommendations have previously been made to improve compliance and increase the ease with which royalties can be collected more generally. These include: • a procedure similar to the Information Notice Scheme contained in the UK’s PBR legislation, allowing a PBR owner who suspects an infringement to serve on the suspected infringer an information notice seeking confirmation of the source of the harvested material and products made from harvested material. It was thought that an information notice scheme would enhance the ability of PBR owners to exercise their rights, in a way that ‘balances the legitimate interests of PBR owners and growers’ (ACIP 2010a) • a ‘purchase’ right (PGG Wrightson Seeds, sub. 82; AGT, sub. 15) that would allow plant breeders or their agents to recover end point royalties from grain ‘purchasers’, such as traders and accumulators, rather than from grain growers. This potentially makes the collection of royalties more efficient, as grain purchasers number in the tens or hundreds, in contrast to grain growers, who can number in the tens of thousands (Lawson 2013b) • ongoing education and awareness campaigns. At present, IP Australia is monitoring developments in relation to the UK’s Information Notice Scheme and is waiting for the development of a new consultative body (to replace the Plant Breeder’s Rights Advisory Committee) for further guidance. The Australian Government previously rejected the need for a purchase right, believing that contract-based approaches could address breeder’s concerns, and that positive responses to other recommendations (including industry-supported education campaigns) would strengthen the system and should be allowed time to demonstrate their impact (Stewart et al. 2015). Since then, IP Australia has continued to monitor developments and considers that the reasons given by the Australian Government to reject the need for a purchase right still stand (pers. comm., IP Australia, 5 April 2016). The Commission agrees with this assessment. Indeed, ongoing education and awareness campaigns have already proven effective as a solution to issues of compliance and enforcement. For example, in relation to the larger broadacre crops, plant breeders and other industry stakeholders have continued to educate growers and others in the supply chain about the role of PBR and royalty arrangements in the modern plant breeding system. As a result, grower recognition and acceptance of royalty systems has improved. In other agricultural sectors, reinforcing the need for all participants — breeders and growers — to agree to and abide by the rules will improve the effectiveness and efficiency of PBR. In the dairy, sheep and beef sectors, industry associations and plant breeders 428 INTELLECTUAL PROPERTY ARRANGEMENTS should continue to promote awareness among farmers of the benefits of a competitive and profitable pasture and forage crop breeding sector funded by royalties or license fees. Initiatives to improve the availability of accurate and independent information regarding the performance of new plant varieties is likely to improve grower compliance over time, and reduce the potential for copycat breeding and copying. In the nursery and ornamental sector there is room to improve grower understanding of the PBR system, particularly with respect to the payment of royalty and license fees. For example, Prescott and Christie (2015) believe the development of an EPR type system for the cut flower industry is stymied by (among other things) a significant level of misunderstanding in the sector regarding the nature and function of IP in general, and PBR in particular. Efforts that involve industry (who directly benefit from higher levels of compliance) in improving outcomes is the preferred approach. With this in mind ACIP recommended that the Australian Government should: Encourage PBR owners to make clear to growers the conditions of sale of propagating material and their obligations in relation to future generations of it. This includes making clear that growers require the authorisation of the PBR owner to sell crops grown from farm-saved seed. (2010a, p. 8) The evidence suggests that there is scope to increase the use and efficiency of EPR systems, particularly in horticulture and nursery sectors. The Commission also agrees that compliance with royalty and licensing agreements is best achieved through closer cooperation and consultation, with industry groups best placed to promote awareness and understanding of the rules regarding PBR. This is particularly important in relation to livestock producers growing pasture and fodder crops. Technology change and market forces could make PBR redundant In the medium to longer term it is possible that the PBR system may decline in importance relative to other forms of protection. As noted earlier in this chapter, some consider that PBR do not offer sufficient protection, and alternative forms of protection can be used by breeders — particularly for GM and hybrid varieties — to guard against copying and unauthorised use of new plant varieties. If future trends in plant breeding favour the development of patentable and hybrid varieties, PBR will potentially become less attractive to breeders as a primary means of protection. Recent innovations in Australian pasture breeding are already resulting in high-quality GM and hybrid varieties (Dairy Futures CRC 2016). It is not clear whether the relevant researchers targeted the development of hybrid and GM varieties specifically to overcome perceived weaknesses in the capacity of PBR to protect conventionally bred pasture varieties, or whether this was a serendipitous outcome. Kock et al. (2006) note that breeders in Europe are increasingly turning to hybrid technologies to overcome the limited scope of protection and difficulties in enforcing PVP rights there. Whatever the motivation, PLANT BREEDER’S RIGHTS 429 if pasture crop breeding in Australia comes to be dominated by hybrid and/or patentable GM varieties, growers may bypass the PBR system. Other developments in plant breeding, such as the CRISPR gene editing technology (box 13.2), could hasten a more general shift to other forms of protection. Technologies like CRISPR allow plant breeders to cheaply and accurately rearrange or delete specific genes, with the resulting new plant varieties likely to be patentable. If CRISPR-derived plant varieties also avoid being declared as genetically modified organisms (and hence avoid costly regulatory regimes), this will potentially make the technology even more attractive to breeders. Box 13.2 CRISPR-Cas9 gene editing CRISPR-Cas9 (CRISPR for short) is a comparatively new gene-editing technology that has the potential to speed up the process by which ‘genes of interest’ can be modified or deleted from an existing plant variety (Hsu, Lander and Zhang 2014). In essence, CRISPR makes it ‘easy, cheap, and fast to move genes around — any genes, in any living thing, from bacteria to people’ (Maxmen 2015). CRISPR is thought to have great potential in agriculture, particularly in relation to improving crop varieties. For example: • Chinese scientists have used the method to produce a wheat strain that is resistant to powdery mildew — a fungal disease • agricultural biotech giant DuPont is set to field-test CRISPR-edited drought-resistant corn and high-yield wheat, which the company says could be on the market in five years • the US Department of Agriculture recently gave permission for a CRISPR-edited mushroom to be cultivated and sold without GMO regulation, because developers removed, rather than introduced, new genetic material. It is unclear whether CRISPR will complement or replace conventional longer-term plant breeding programs. CRISPR could become more widespread and popular in Australian plant breeding if CRISPR-edited plants avoid the costly and time-consuming regulatory requirements of ‘conventional’ GMOs. This may be possible, as CRISPR can be used in ways that do not introduce non-plant genetic material to an existing plant (as with other gene editing technologies), but simply to rearrange or delete existing plant genetic material. Sources: Maxmen (2015); Richter (2016). On the other hand, predictions of the demise of PVP-type regimes have been made many times in the past. 9 Similarly, new technologies often do not live up to expectations. If plant breeding based on crossing and selection remains the most cost-effective way to improve and enhance existing plant varieties, PBR will remain an important form of IP protection for some time. 9 See, for example, Samuel, Kingma and Crellin (1983); Cornish (1989); Fowler (1994); Godden (1998); Reichman (2000); Kock, Porzigq and Willnegger (2006); Janis and Smith (2007); and Kingston (2007). 430 INTELLECTUAL PROPERTY ARRANGEMENTS 14 Circuit layout rights Key points • Australia introduced circuit layout rights in 1989 as part of multilateral trade negotiations. − Dedicated rights to protect intellectual property (IP) in circuit layout designs were first introduced in the United States in the 1980s in response to concerns about unauthorised copying. − At the time, the United States also pushed hard for other countries to recognise circuit layout rights. Australia responded by introducing the Circuit Layout Act 1989 (Cth). • There is little evidence that unauthorised copying of integrated circuits remains problematic in Australia or internationally. A number of factors have seen unauthorised copying become increasingly unworkable and unprofitable. − Customers are demanding unique and customised integrated circuits, which have a shorter commercial life. − The technology required to manufacture integrated circuits has become more complex, and research, production and setup costs have increased. • In addition to the ‘natural’ protections afforded by these changes in market circumstances, circuit layouts can also be protected by other IP rights — for example, patents can protect semiconductors within the integrated circuit. • While there is little evidence that the circuit layout rights system is used much at all, and the Commission questions the need for a sui generis right, Australia’s obligations under TRIPS requires some form of protection for integrated circuits. − Retaining circuit layout rights in their present form is the preferred option, as the alternatives may create more problems than solutions. • Australia’s experience with circuit layout rights provides a cautionary tale. The community would be better served by IP rights that are motivated by the underlying economics of protecting the embodiment of ideas, rather than a desire to be party to multilateral agreements. 14.1 What are circuit layouts and how are they protected? Put simply, a circuit layout is the blueprint for an integrated circuit (figure 14.1). It is defined in the Circuit Layouts Act 1989 (Cth) (CL Act) as: a representation, fixed in any material form, of the three-dimensional location of the active and passive elements and interconnections making up an integrated circuit. (s. 5) CIRCUIT LAYOUT RIGHTS 431 Figure 14.1 How semiconductors, circuit layout and integrated circuits differ A semiconductor is a material that conducts electricity under some conditions, but not others, thus allowing for the control of electrical currents. A circuit layout is the blueprint design illustrating the locations and interconnections between various components, including semiconductors. These layout designs are protected by circuit layout rights. The physical form of the circuit layout, including the interconnections and its components, is known as an integrated circuit. Integrated circuits are the backbone of a variety of electronic devices, ranging from mobile phones to fridges. Circuit layouts are protected by dedicated (sui generis) rights, once circuit layouts are created in a material form. Protection is automatic, and operates in a similar way to copyright in that no registration is required. Circuit Layout Rights (CLRs) provide holders with a time-limited exclusive right to commercially exploit the design (effectively protecting the integrated circuit). Protection lasts 10 years from the creation of an eligible layout. 1 However, if the layout is commercially exploited during that time, it is protected for 10 years from the date of commercial exploitation (or for a maximum of 20 years). There is a number of exemptions to the rights afforded under CLRs, including allowing for copying for research or teaching purposes and reverse engineering of protected layouts. Section 23 of the CL Act allows for third parties to ‘deconstruct’ a protected integrated circuit and to recreate an original layout based on resulting evaluation and analysis (Bowen 1988b). The resulting layout can then be commercially exploited without infringing on the rights of the original maker. 1 An eligible layout is defined in s. 5 of the CL Act as one where a layout was first commercially exploited in Australia or other eligible countries (which are defined in regulation as members of the World Trade Organization), or where the maker of the layout was an individual or firm in Australia or in an eligible country. 432 INTELLECTUAL PROPERTY ARRANGEMENTS The genesis of CLRs The early 1980s saw growing concerns about unauthorised copying of integrated circuits, particularly around the designs of circuit layouts, leading to widespread calls in the United States for IP protection (Radomsky 2000). Having successfully legislated for a sui generis right (in the form of the Semiconductor Chip Protection Act 1984), the United States turned its attention to the international landscape, advocating for multilateral change. While the proposed Washington Treaty on Intellectual Property in Respect of Integrated Circuits failed to get traction in its own right, it was subsequently referenced in the Agreement on Trade Related Aspects of Intellectual Property Rights (TRIPS). Australia was quick to conform, introducing the CL Act in 1989, despite the Washington Treaty not being in force. The reasoning provided in the explanatory memorandum is indicative of the level of consideration given at the time to the introduction of a new IP right: This Bill provides a new copyright-style of intellectual property in original layouts for integrated circuits. The Bill is consistent with the main elements of a draft Treaty on the topic developed by the World Intellectual Property Organisation, and with the laws of our major trading partners. (Bowen 1988a, p. 1) The CL Act means Australia conforms with the later introduced 1994 TRIPS. 14.2 The role of CLRs today The market for CLRs has changed Even if a rationale for introducing CLRs had been established in the 1980s, circumstances have since changed. The technology required to manufacture integrated circuits has become significantly more complex, and production and setup costs have increased. At the same time, customers have increasingly sought unique and customised integrated circuits, which have a shorter commercial life (Rauch 1993). Together, these factors have made unauthorised copying of integrated circuits less profitable and so less appealing. Smaller and more complex integrated circuits require specific processes and sophisticated manufacturing equipment. The equipment used to manufacture integrated circuits today, generally comprises of multi-chambers, robotic parts, ultra-high vacuum manufacturing apparatuses and clean rooms (Radomsky 2000). This specialised equipment is extremely expensive — in some cases, costing billions of dollars. In 2012, the costs of developing and fabricating state of the art integrated circuits was in excess of US$10 billion (McKinsey & Company 2013). CIRCUIT LAYOUT RIGHTS 433 The integrated circuit industry has adapted to the significant growth in production costs by adopting a ‘foundry model’, which separates the semiconductor fabrication plant operation (foundry) from the circuit layout design operation (designers). This setup allows for foundries to capitalise on their expensive equipment by manufacturing for multiple designers. Not only has the increasing costs of production made unauthorised copying less attractive, the foundry model itself provides a barrier to entry and therefore an added protection. To make an unauthorised copy of an integrated circuit, parties must either gain access to an existing manufacturing foundry or invest in their own. The reduced life cycle of integrated circuits has also discouraged unauthorised copying. In the 1970s, the typical life of an integrated circuit was a few years (Radomsky 2000). Since then, the life cycle of an integrated circuit has declined to less than one year, reflecting the shift in demand from a business to a consumer market (Boston Consulting Group 2012). By way of example, some manufacturers, such as Sony, are aiming for a six month life cycle for their flagship phone (PhoneArena 2014). The reduction in time has significantly curtailed the ability to copy and commercialise integrated circuits — by the time a copy is brought to the market, new and more advanced circuits can be on the market. The shortening life cycle of integrated circuits is also reflected in the declining asset life of computers, in which circuit layouts play an important role. The ABS uses a measure of asset life for all capital assets as part of the national accounts, noting that the average lifespan of computer equipment is assumed to have gradually declined from eight years in the 1960s. The Bureau currently uses an average lifespan of 4.9 years (ABS 2015b, pp. 371–372). Finally, for business users of integrated circuits, there has been a shift away from selling standardised integrated circuits. Integrated circuit companies now offer customised programmable integrated circuits tailored towards client’s needs (known as ‘field-programmable gate arrays’), along with after sales technical support and other services (Radomsky 2000). The move towards custom-designed circuit layouts, which by their nature have limited alternative applications, have further dampened incentives to copy. 434 INTELLECTUAL PROPERTY ARRANGEMENTS Other forms of protection tend to be used instead The narrow and uncertain protection of CLRs discourages their use. Similar to copyright, only a direct copy constitutes an infringement, and even then, copies based on reverse engineering are permitted. The complexity of integrated circuits makes it hard to distinguish between a direct copy and one that has been reverse engineered (Radomsky 2000), and thus harder for parties to enforce their rights. Some designers have taken to using technological means to protect their designs from unauthorised copying. For example, circuit layouts increasingly include mechanisms such as: • ‘encryption and authentication’, where software built into the integrated circuit requires a key or code to enable use • ‘obfuscation’, which is the process of adding complexity to the design of a circuit layout for the purpose of making it difficult to copy • ‘watermarking and fingerprinting’, which adds features to an integrated circuit for the purpose of checking if it is legitimate. While protection for the entire design or layout of a circuit is not afforded through other IP rights (IP Australia 2016k), in some cases, components of integrated circuits can be protected (figure 14.2). For example, protection exists for the: • components of an integrated circuit through patents • drawings of, and firmware embedded in, integrated circuits, through copyright • branding and legitimacy of the integrated circuits being sold, through trade mark and design rights (Kiat et al. 2010). CIRCUIT LAYOUT RIGHTS 435 Figure 14.2 Integrated circuits: market and forms of protection Sources: Adapted from Bowen (1988a); Linear77 (2012); Wilkof and Basheer (2012). There is little evidence that designers rely on CLRs Examining changes in business practices and reliance on other IP rights provides one perspective on the value of CLRs. Another perspective can be gleaned by examining the extent to which they are used and enforced. However, the automatic granting of CLRs makes it difficult to determine how many eligible integrated circuits are protected in Australia — the absence of a registration 436 INTELLECTUAL PROPERTY ARRANGEMENTS scheme means little is known about the extent of their use. Determining the importance of the integrated circuit manufacturing industry is also difficult, but the limited available evidence suggests that it is likely to be small as: • the value of Australian exports of integrated circuits is small, and has declined both in real terms and as a share of merchandise exports (figure 14.3) • participants in the inquiry noted ‘the absence of a circuit layouts fabrication industry of any note in Australia’ (FICPI, sub. DR581, p. 27), and suggested that there was only one fabrication plant in the country (Jones, sub. DR412) • ABS data indicate that the ANZSIC class in which integrated circuit manufacturing is defined employed 5083 people in June 2014, with value added of $774 million (ABS 2014). However, this class also includes other goods unrelated to integrated circuit manufacturing, and so the actual level of employment and output is likely to be smaller still. Figure 14.3 Trade in integrated circuit products Value of trade (2015 $b)a As a share of merchandise trade (per cent) 3.0 5.0 4.5 4.0 3.5 2.0 3.0 2.5 2.0 1.0 1.5 1.0 0.5 0.0 1989 0.05 0 1994 1999 2004 2009 Export 2014 0.0 1989 1994 1999 2004 2009 2014 Import a Three digit SITC code 772 (Electrical apparatus for switching or protecting electrical circuits or for making connections to or in electrical circuits; electrical resistors, other than heating resistors; printed circuits; boards, panels, consoles, desks, cabinets and other bases, equipped with two or more apparatus for switching, protecting or for making connections to or in electrical circuits, for electric control or the distribution of electricity) and 776 (Thermionic, cold cathode or photo-cathode valves and tubes diodes, transistors and similar semiconductor devices; photosensitive semiconductor devices; light-emitting diodes; mounted piezoelectric crystals; electronic integrated circuits and micro assemblies; parts thereof) are included to capture the integrated circuit industry. Such a definition, however, is broader than just the integrated circuit industry, and so may include other manufactures not strictly applicable to CLRs. Price deflators to determine real values of merchandise trade were SITC codes 76 & 77 for exports, and SITC code 77 for imports. Sources: Commission estimates based on ABS (International Trade Price Indexes, Australia, Dec 2015, Cat. no. 6457.0; unpublished trade data at the three-digit level). CIRCUIT LAYOUT RIGHTS 437 The lack of legal cases involving unauthorised copying of integrated circuits also suggests the scope of infringement is not a major issue in Australia. The Commission examined cases brought before the Federal Court and found only 10 cases where the CL Act was cited. Of these cases, most involve other IP rights, and only made passing reference to a circuit layout. Only a handful of cases that have come before the courts could be classed as genuine disputes arising from the application of CLRs, with the most prominent being Nintendo Co Ltd v Centronics Systems Pty Ltd [1994] HCA 27 (box 14.1). Box 14.1 Nintendo Co Ltd v Centronics Systems Pty Ltd Between the passing and commencement of the Circuit Layout Act 1989, Centronics imported video game machines known as Spica Entertainment Units, which included an unauthorised chip containing Nintendo’s circuitry layout. Nintendo claimed Centronics Systems infringed on their original circuit layout. Centronics Systems made several claims of defence, including that: • they imported gaming units prior to the commencement of the CL Act, and so their act does not constitute an infringement • it was a case of secondary infringement, where the gaming units happened to include an unauthorised copy of Nintendo’s circuit • they were unaware the Spica Entertainment Units contained an unauthorised copy of Nintendo’s circuit layout, and so were innocent violators. The High Court of Australia ruled Centronics’ defences as invalid, and found that they had infringed on Nintendo’s circuitry rights, as they possessed the required constructive knowledge of Nintendo’s property rights. The High Court’s decision is significant in several respects. It found: • that providing IP rights for particular expressions of ideas (for example, circuit layouts) could be legislated • that there was now an effective way to contest the use of IP rights for circuit layouts, and that further sui generis rights might provide protection for other forms of computer technology • that rights could be crafted in a way that reflects the needs of both consumers and producers, and that judicial interpretation could take account of both. Source: Clark (1994). The Australian experience of limited reliance on CLRs is not unique. Even among the world’s largest producers of circuit layouts — such as the US, Korea, Japan, Taiwan and China — the data suggest CLRs are little used. In these countries, there is a requirement to register the designs of integrated circuits in order for protection to be valid. The data reveal that the number of registrations in most of these jurisdictions has been declining. For example, in the US, the number of registrations fell by over 90 per cent from around 1000 to less than 100 per year between 1997 and 2013 (figure 14.4). In other jurisdictions, the pattern is more mixed: Taiwan has had a relatively 438 INTELLECTUAL PROPERTY ARRANGEMENTS steady number of registrations, while China has experienced growth. Today, there are very few registrations in most developed jurisdictions — as Karnell noted: It was never used much, as the legislation is based on an outdated business and technology model. It is still in use though, but most developed countries with a chip industry see three or four registrations per year only. Litigation is in these circumstances very rare. (Karnell in Derclaye and Leistner 2011, p. 5) Figure 14.4 2000 Number of CLR registrations in United States, Korea, Japan, Taiwan and China US Korea Japan Taiwan China 1800 1600 1400 1200 1000 800 600 400 200 0 1985 1990 1995 2000 2005 2010 Sources: KIPO (2015); SIPO (2016); SOFTIC (2014); TIPO (2015); USCO (2016). 14.3 Where to from here — policy lessons rather than policy change The case for sui generis protection for circuit layouts in Australia is questionable. The large upfront costs required to exploit a copied circuit and the presence of technological solutions to prevent others from reproducing circuits suggest that the ideal IP protection could well be ‘none at all’. Dispensing with redundant rights is not straight forward, as Australia is a party to TRIPS. Under TRIPS, Australia is obliged to provide protection for the layout of integrated circuits. TRIPS does not require that protection be afforded through a dedicated right, and so, Australia could (in theory) repeal the CL Act 1989. But Australia could only do so by providing comparable protection through another avenue or unilaterally departing from this TRIPS obligation. In the absence of the CL Act, different aspects of integrated circuit CIRCUIT LAYOUT RIGHTS 439 design would be protected under the Copyright Act 1968 (Cth) or the Designs Act 2003 (Cth), but it is unclear if these would be sufficient, as presently drafted, to meet the TRIPS obligation (Law Council of Australia, sub. DR490). It is also not clear that copyright and design rights are appropriate for the protection of circuit layouts, given their scope and limited exceptions. CLRs provide for a shorter term of protection (relative to copyright) and specific exceptions for reverse engineering, teaching purposes and personal use (table 14.1). Accordingly, abolishing the CL Act — and allowing greater use of copyright and design rights — is likely to do more harm than good. As noted by FICPI: Although as identified in the PC Draft Report there are other ways to prevent copying of circuit layouts, legally, there do not appear to be better mechanisms to protect circuit layouts other than by way of the Circuit Layout Act. (sub. DR581, p. 27) Table 14.1 Are CLRs still the best fit? Term of protection Scope of protection Exceptions for education and teaching Reverse engineering Targeted at specific products CLRs Design Rights ~10-20 years Narrow 5-10 years Form rather than function       Patents Copyright 8-20 years Life + 70 years Very broad All creative works, no functionality a  c  c  b  c  c  a,b,c Notwithstanding compulsory licensing provisions, fair dealing exceptions, and allowances for derivative works respectively. In light of the restrictions placed by TRIPS and the relatively benign redundancy of CLRs, attempting to change the mechanism for protecting integrated circuits may create more costs than benefits. FINDING 14.1 Dedicated intellectual property protection for circuit layouts is not ideal and seldom used, but given Australia’s international commitment to protect circuit layouts and no superior alternatives, the best policy option is to maintain the status quo. 440 INTELLECTUAL PROPERTY ARRANGEMENTS Lessons for the future Australia’s experience with CLRs provides a cautionary tale. As already highlighted, the justification for CLRs proved not to be well founded: • The legislation was motivated more by a desire to be compliant with the expected ratification of the Washington Treaty — which did not eventuate — rather than evidence of a significant problem. • The implementation of the right was expected to help stimulate the nascent integrated circuit manufacturing industry in Australia, which did not occur. As put by the then Attorney-General in the second reading speech of the bill: The Australian computer chip industry is recognised by this Government as an important, innovative and growing one. The Government is committed to fostering its development. The Circuit Layouts Bill is a significant and timely demonstration of this commitment … (Bowen 1988b) Two lessons emerge from this experience. First, accession to international agreements should be a means to an end, not an end in and of itself. Second, greater consideration needs to be given to the role of IP rights, and whether such rights are always the best way of encouraging and rewarding innovative activity. Finally, Australia’s experience with CLRs underscores the importance of transparent and robust assessment to inform the consideration of any new, dedicated, forms of IP rights. The Commission has recommended a number of changes to governance arrangements to improve IP policy decision-making, including in respect of IP protections agreed in international treaties. These recommendations are discussed in chapter 17. CIRCUIT LAYOUT RIGHTS 441 15 Intellectual property rights and competition law Key points • Section 51(3) of the Competition and Consumer Act 2010 (CCA) exempts licensing or assignment of intellectual property (IP) from certain competition provisions of the CCA. • The rationale for the exemption has largely fallen away. IP rights and competition are no longer thought to be in ‘fundamental conflict’. IP rights do not, in and of themselves, have significant competition implications. – Rather, competition implications arise in those cases where there are few substitutes or where the aggregation of IP rights may create market power. • The Commission considers that commercial transactions involving IP rights, including the assignment and licensing of such rights, should be subject to the CCA in the same manner as transactions involving other property and assets. • At present, the immediate costs and benefits of removing the exemption under s. 51(3) are finely balanced. – However, looking ahead, the benefits could rise as the level of licensing and cross licensing increases, especially in pharmaceutical and communications markets. • A range of measures could mitigate initial and ongoing regulatory costs, further tipping the balance in favour of removing the exemption. – Guidance by the ACCC on the application of competition law to IP could address any potential compliance costs and uncertainty. This proposed approach is similar to current approaches in Europe and the United States. – Reforming per se provisions in the CCA along the lines suggested by the Competition Policy Review would address legitimate concerns that socially valuable activities are not impeded. • Accordingly, the Commission considers that the Government should repeal s. 51(3) of the CCA at the same time as giving effect to the Competition Policy Review recommendations on per se prohibitions in Part IV of the CCA. • No case currently exists for extending the National Access Regime to IP. – The imperatives for compulsory access and the effects on incentives to invest in IP differ across the spectrum of IP rights. The existing access provisions for individual IP rights allow for these differences. – Replacing the existing compulsory access arrangements with part IIIA would likely impose higher barriers to access and impede innovation. INTELLECTUAL PROPERTY RIGHTS AND COMPETITION LAW 443 As discussed in chapter 2, the economic characteristics of knowledge are such that it is likely there will be an under-provision of ideas and innovation within an economy without some form of intervention by government. Intellectual property (IP) rights, and the excludability that they allow, are designed to encourage creative activity by providing a right to exclude others from using ideas without permission and/or payment. A well-functioning IP system must take account of the long term effects on competition that exclusivity can create. Competition is important because it improves choice and prices for consumers, and encourages the efficient allocation of resources and innovation, both drivers of economic growth. The excludability allowed by IP rights can, in some cases, have adverse implications for competition: … IP rights can be used in a way that deters competition and limits consumer choice. For example, this could manifest in owners of IP rights extracting excessive royalties from IP licences or placing anticompetitive restrictions on knowledge dissemination. This would have adverse knock-on effects for innovation. (Harper et al. 2015, p. 101) Ideally, the IP system would embody a balance between rights holders and users. When the balance is tilted too far in favour of rights holders there are dangers that the IP system can hinder competitive outcomes: IP rights can help to break down barriers to entry but, when applied inappropriately, can also reduce exposure to competition and erect long-lasting barriers to entry that fail to serve Australia’s interests over the longer term. (Harper et al. 2015, p. 41) A number of policy levers affect the balance of the IP system. These include the breadth and strength of IP rights that are afforded (as discussed in earlier chapters). The balance of the IP system can also be influenced by the way in which competition policy and law is applied to IP. This chapter examines two exemptions IP arrangements currently enjoy from competition law. • Licensing or assignment of IP property is currently exempt from most laws dealing with anticompetitive business practices in part IV of the CCA (s. 15.1). • IP is exempt from the National Access Regime in part IIIA of the CCA (s. 15.2). Competition policy issues, which relate to specific types of IP arrangement, are dealt with in other chapters. ‘Pay-for-delay’, where a pharmaceutical company might pay another to delay market entry as part of a patent infringement settlement, to the detriment of government and consumers, is dealt with in chapter 10. Territorial restrictions on imports of IP-protected goods legitimately produced overseas, such as parallel importation restrictions on books and trade marked goods, are considered in chapters 5 and 12. Geo-blocking, where access to websites and digital goods and services is restricted to a consumers’ ‘home market’, is examined in chapter 5. 444 INTELLECTUAL PROPERTY ARRANGEMENTS 15.1 The exemption for licensing or assignment of IP Part IV of the CCA prohibits companies from engaging in certain types of conduct that reduce competition (box 15.1). Section 51(3) of the CCA provides an exemption from part IV for conditions in licences and assignments of patents, registered designs, copyright, or eligible circuit layout rights. The exemption is limited to conditions of licences and assignments insofar as they ‘relate to’ IP rights. The exemption does not apply to ss. 46 (misuse of market power), 46A (misuse of market power in a Trans-Tasman market) or 48 (resale price maintenance) of the CCA. The exemption also does not apply to rights granted under the Plant Breeders Rights Act 1994 (Cth). Box 15.1 Part IV of the Competition and Consumer Act 2010 Part IV of the Competition and Consumer Act 2010 (CCA) prohibits businesses from engaging in certain restrictive trade practices that would reduce competition. Of most relevance to agreements dealing with IP: • section 45 prohibits a corporation from making a contract including a provision that has the purpose or likely effect of substantially lessening competition • section 46 prohibits a corporation that has a substantial degree of market power from taking advantage of that power to eliminate or substantially damage a competitor, prevent the entry of a person into a market, or deter or prevent a person from engaging in competitive conduct in a market • section 47 prohibits a corporation from supplying goods or services whilst imposing restrictions on the ability of the person supplied to acquire goods or services from a competitor or restrictions on resupply of the goods or services with a purpose or likely effect of substantially lessening competition. The consequences of breaching a provision of part IV of the CCA are significant. Section 76 provides for a court imposing a maximum penalty of $10 million for each act or omission that contravenes a provision of part IV. What was the genesis of the exemption? The exemption for IP from aspects of competition law was imported from similar UK law when the predecessor to the CCA, the Trade Practices Act (Cth) was enacted in 1965. The prevailing view at the time the exemption was enacted was that IP rights and competition policy were in fundamental conflict (IPCRC 2000, p. 206; NCC 1999, p. 149). However, it is now the generally accepted view that IP rights do not, of themselves, create economic monopolies (NCC 1999, p. 149). Rather, the effect on competition depends on the nature and extent of IP rights granted and the extent to which close substitutes are, or are likely to be, available (Landes and Posner 2003, p. 374). INTELLECTUAL PROPERTY RIGHTS AND COMPETITION LAW 445 For example, competition concerns may arise where a standard of manufacture for an entire industry incorporates patented technology, or in respect of the patenting of pharmaceutical products where it is difficult and costly to produce a substitute product without breaching existing patent rights. On the other hand, competition concerns are unlikely to arise in respect of the copyright afforded to a cookery book for which there are many alternatives. Even where an individual IP right may have several substitutes and not pose competition problems, the aggregation of IP rights may create market power. This may occur where a single company acquires the licences to manufacture and distribute a range of competing products in Australia or where rights are collectively administered. It is for this reason that the operation of copyright collecting societies is subject to ongoing scrutiny, including by the Australian Competition and Consumer Commission (ACCC) (chapter 6). Similarly, patent pools, which aggregate patent rights held by an individual or organisation for the purpose of licensing patents can have anticompetitive effects (ACCC, sub. 35, p. 9). The nature of the competition problems that arise in those cases where rights holders enjoy market power is varied. They can include excessive prices, price discrimination, raising barriers to entry in immediate and downstream markets, and restrictions on access, thereby impeding competition and follow on innovation. The exemption has been subject to scrutiny and review More finessed views about the intersection between IP rights and competition policy now prevail. Reflecting this, the exemption under s. 51(3) has been subject to seven reviews that have recommended its removal or significant narrowing over the past 15 years. In 1999, in order to comply with the Competition Principles Agreement 1995 (CPA), the National Competition Council (NCC) reviewed ss. 51(2) and 51(3) of the then Trade Practices Act 1974. Under the CPA, jurisdictions were required to review and reform all existing legislation that restricted competition. Clause 5(1) provided a ‘guiding principle’ that legislation should not restrict competition unless it can be demonstrated that: (a) the benefits of the restriction to the community as a whole outweigh the cost; and (b) the objectives of the legislation can only be achieved by restricting competition. The NCC considered the costs and benefits of the s. 51(3) exemption and recommended that: … the exemption in section 51(3) be retained, but amended to remove protection of price and quantity restrictions and horizontal restrictions. (1999, p. 243) Section 51(3) has been the subject of a number of further reviews since the NCC reported in 1999 culminating in recommendations for repeal or amendment (table 15.1). These recommendations have not been adopted by the Australian Government. 446 INTELLECTUAL PROPERTY ARRANGEMENTS Table 15.1 Recommendations on section 51(3) from past reviews Year Report Title Summary of Recommendations Government Response 1999 National Competition Council Section 51(3) should be retained but amended to remove price and quantity restrictions and horizontal agreements No Government response Review of ss. 51(2) and 51(3) of the Trade Practices Act 1974 Section 51(3) should be extended to cover rights granted under the Plant Breeder’s Rights Act 1994 (Cth) The ACCC should release guidelines dealing with when IP licenses and assignments might breach part IV of the CCA and the types of conduct that might be authorised, despite a possible breach 2000 Intellectual Property and Competition Review Committee Review of intellectual property legislation under the Competition Principles Agreement Repeal s. 51(3) and replace it with a provision that would ensure that conditions in a contract do not give rise to a contravention of part IV of the TPA (now CCA) ‘so long as those conditions do not result, or are not likely to result in a substantial lessening of competition’ Accepted (June 2000), but legislation never introduced 2004 Australian Law Reform Commission: Genes and Ingenuity – Gene Patenting and Human Health The Commonwealth should amend s. 51(3) of the TPA to clarify the relationship between part IV of the TPA and IP rights Recommendation noted (November 2011) 2013 Australian Law Reform Commission: Copyright and the Digital Economy Repeal of s. 51(3) should be considered as an integral aspect of equipping copyright law for the digital economy No formal government response 2013 House Standing Committee on Infrastructure and Communications: Inquiry into IT Pricing The Committee recommends repeal of s. 51(3) of the CCA No formal government response 2015 Competition Policy Review Panel: Competition Policy Review Section 51(3) of the CCA should be repealed The Government notes this recommendation and will have regard to the findings of the Productivity Commission’s inquiry into Australia’s intellectual property arrangements (November 2015) The ACCC should develop guidelines to clarify the relationship between part IV of the TPA and IP rights. The guidelines should address when the licensing or assignment of IP rights might be exempted under s. 51(3) or might breach part IV and when conduct that would otherwise breach part IV might be authorised by the ACCC INTELLECTUAL PROPERTY RIGHTS AND COMPETITION LAW 447 The initial costs and benefits of the exemption are finely balanced The absence of a fundamental conflict between IP rights and competition policy means that both the costs and benefits of retaining or dispensing with the exemption under s. 51(3) are likely to be modest. Why have parties argued to get rid of the exemption? The number of arrangements that are affected by removal of the exemption is likely to be small as ‘ … the vast majority of arrangements where IP rights are licensed or assigned to other entities are likely to be pro-competitive’ (ACCC, sub. 35, p. 14) and therefore not likely subject to enforcement action by the ACCC under the competition provisions of the CCA. In economic terms, a pre-condition for an anti-competitive effect is that one of the parties to the arrangement has sufficient market power to influence prices in a market. Most businesses are not in that position. Reflecting this, Many of the arguments in favour of dispensing with the exemption under s. 51(3) rely on identifying instances where anticompetitive conduct might occur rather than instances where such conduct has occurred. For example, the ACCC argued: … some licensing arrangements can unduly damage efficiency and welfare. Where these arrangements are exempt because of … section 51(3) this conduct cannot be addressed. (sub. 35, p. 14) Similarly, the NCC Review focused on identifying examples of the types of arrangements that could have competition implications rather than practical examples. These included arrangements where competitors face few or no substitutes for their products and: • exclusively cross-licence their IP rights • include quantity and/or price restrictions in their licence agreements, which enables them to restrict output of particular products and fix prices • include territorial restrictions in exclusive cross-licences, which enables them to allocate territories to each other and reduce competition within those territories. (NCC 1999, p. 202) Two further examples of arrangements that could have competition implications are ‘grant-back’ and ‘hold-up’ (box 15.2). Repeal of s. 51(3) would allow part IV of the CCA to deal more effectively with these types of arrangements. 448 INTELLECTUAL PROPERTY ARRANGEMENTS Box 15.2 Examples of arrangements with potential competition implications Grant-back Grant-back obligations are clauses in licence agreements stating that licensees are required to license the IP in any improvements made to the technology back to the licensor. The European Commission’s Competition Directorate considers some grant-back obligations can reduce licensee’s incentives to innovate and are potentially anticompetitive (DG Competition 2014). Grant-back was the main issue in the US Pilkington case, whereby Pilkington licensed the ‘float glass process’ to numerous manufacturers of glass, under the condition that they kept the process secret and licensed any improvements back to Pilkington. This allowed Pilkington to maintain a dominant market position in glass manufacture from the 1950s to 1994 (long after the expiry of the original patents), when the US Department of Justice took action under the Sherman Anti-Trust Act (US) to prevent Pilkington from enforcing the relevant licence conditions. Hold-up ‘Hold-up’ refers to patent holders imposing higher license fees on users after they have made up front investments that depend on the patented invention. The risk of this problem is commonly raised as a concern for patents that form part of a technology standard (such as the standards for Wi-Fi and 3G). These patents are commonly referred to as ‘standard essential patents’. A follow on innovator may be exposed to the risk of economic hold-up where it makes an investment that relies on access to — and thus locks it into using — a patented technology (Bessen 2004; Lemley and Shapiro 2007). In some cases, the threat of hold-up can discourage innovation and investment, as an innovator or investor will anticipate hold-up by the patent holder and will require a higher rate of return from the up front investment costs, such that less investment and therefore innovation will proceed. Territorial licensing conditions Territorial restraints are often used in copyright, where films, television programs, books and music are distributed in certain geographies at different prices (chapter 5). Where substitutes exist in the relevant market, such practices are unlikely to have anticompetitive outcomes. However, as pointed out by the US Federal Trade Commission, licensing arrangements can be used to anti-competitive effect where rights holders divide the market geographically between manufacturers to ensure that they do not compete with each other in the supply of a product. Sources: US v Pilkington plc and Pilkington Holdings Inc CV 94-345-TUC-WDB; (US Department of Justice and the Federal Trade Commission 1995). Some participants in the Commission’s hearings suggested inadequate ACCC oversight of anti-competitive behaviour in the pharmaceutical drugs and biotech sectors, which may be, in part attributable to the s. 51(3) exemption: … it seems to be very problematic at the moment that various IP monopolies are not necessarily subject to proper competition oversight by the ACCC and we really do need to change — and that’s particularly critically important in relation to pharmaceutical drugs and the INTELLECTUAL PROPERTY RIGHTS AND COMPETITION LAW 449 biotech sector, and which there has been a lot of concern about anti-competitive behaviour by one another, price gouging. (Rimmer, trans., p. 47) The Competition Policy Review also noted that the absence of an exemption in other jurisdictions has not been problematic: Most comparable jurisdictions have no equivalent to subsection 51(3). None of the US, Canada or Europe provide an exemption from competition laws for conditions of IP transactions. In those jurisdictions, IP assignments and licences and their conditions are assessed under competition laws in the same manner as all other commercial transactions. The courts in those jurisdictions distinguish between competitively benign and harmful IP transactions, taking account of all relevant circumstances of the transaction and the conditions imposed. There is no evidence that this has diminished the value of IP rights in those countries. (Harper et al. 2015, p. 109) The absence of an exemption in the US has allowed a body of case law to develop around the type of arrangements that breach antitrust law (box 15.3). Box 15.3 Typology: treatment of licensing and assignment of IP under US case law Exclusive licences Exclusive licences may breach antitrust law if the licensor and licensee are actual or potential competitors and the licence creates or enhances the exercise of market power.a Price restrictions Downstream price restrictions on products by licensors are generally unlawful. However, if a patent holder manufactures and sells a product and the price restriction affects the price at which the patent holder can sell its own goods, the restriction may be allowable.b Retail price maintenance can also be allowable if there is a procompetitive effect.c Field of use and territorial restrictions Field of use and territorial restrictions do not raise antitrust concerns.d Tying arrangements only breach antitrust laws if the defendant has market power in the tying product.e Whether tying arrangements where two or more patented products are combined into a new product breach antitrust law is considered by balancing the benefits of the combination against the anticompetitive effects.f a American Bar Association, Antitrust Issues in International Intellectual Property Licensing Transactions Handbook 34 (2013); b United States v General Electric Corporation 272 U.S. 476; c Leegin Creative Leather Products v PSKS Inc 551 U.S. 877 (2007); d General Talking Pictures v Western Electric Company 304 U.S. 175, 180-182; e Illinois Toolworks v Independent Ink 547 U.S. 28 (2006); f Medtronic Minimed v Smiths Medical MD Inc 371 F Supp 2d 578 (D.Del 2005). 450 INTELLECTUAL PROPERTY ARRANGEMENTS Why have parties argued to retain the exemption? While there is little evidence to suggest that the costs of retaining the exemption are significant, at least at this time, there is also little to suggest that the exemption gives rise to much in the way of benefits. The benefits of the exemption are likely to be minor and relate to the exemption removing the need for those assigning or licensing IP to consider part IV of the CCA. The NCC (1999, p. 150) concluded that the exemption provides ‘greater certainty [and so] can help reduce the costs associated with compliance with trade practices law and encourage more licensing activity.’ licensing or assignment of IP rights can provide benefits by enhancing economic efficiency and encouraging more innovation. Submissions to previous reviews focused heavily on the implications of the exemption for business certainty and compliance costs. For example, The Law Council of Australia submission to the NCC review of ss. 51(2) and 51(3) stated: The legal inquiry which is required to be undertaken to determine whether the entry into an agreement could substantially lessen competition is an arduous and expensive one, and the increased cost may well deter exploitation of the rights by way of licensing or assignment, especially for small to medium sized businesses. (1998, p. 1) Similarly, submissions to this inquiry argued that repealing s. 51(3) would result in increased risk and uncertainty for rights holders, would increase transaction costs, and may act as a disincentive to some forms of licensing or assignment. 1 However, the benefits from legal certainty provided by the exemption in s. 51(3) may be more illusory than real. As noted by the ACCC: … the extent of the exception contained in section 51(3) is highly uncertain, given limited jurisprudence, but potentially very narrow. As a result, rights holders face significant uncertainty if they rely on section 51(3) to protect them from competition law claims being brought against them. (sub. 143, p. 3) Indeed, fears about the consequences of removing the exemption appear to be grounded in misunderstanding. As Eagles and Longdin noted: While most of these prohibitions have long been tamed into rules of reason, they can, as currently in Australia still induce panic (a panic largely borne of confusion) in right holders and licensees who confront them for the first time. Panic rises to even higher levels when intellectual property owners are threatened with the sudden removal of the statutory shield which up until now has protected most (but significantly not all) of their dealings with licensees from the attention of regulators and competitors. (2003, p. 28) 1 The Australian Copyright Council (sub. 36, p. 11), Hachette Australia (sub. 41, p. 35), the Coalition of Major Professional & Participation Sports Inc. (sub. DR312, p. 20), Harper Collins (sub. DR418, p. 6), and Swinburne University of Technology (sub. DR557, p. 5) INTELLECTUAL PROPERTY RIGHTS AND COMPETITION LAW 451 This misunderstanding is likely a result of rights holders and licensees having a limited understanding of the competition provisions of the CCA. The exemption has, thus far, allowed them to ignore this aspect of the law when licensing or assigning IP. Proposed ACCC guidance will address this misunderstanding. The balance of costs and benefits is subject to change The initial costs and benefits associated with retaining or dispensing with the exemption under s. 51(3) are likely to be minor. However, the costs of the exemption may increase over time, as certain growth industries engage in cross-licensing that is currently exempt from competition law: In fields in which there are multiple and competing IP rights, such as the pharmaceutical or communications industries, cross-licensing arrangements can be entered into to resolve disputes but which impose anticompetitive restrictions on each licensee. Subsection 51(3) can operate to exempt those arrangements from competition law. (Harper et al. 2015, p. 109) Moreover, a more nuanced approach — which gives the ACCC the power to address genuinely anticompetitive conduct while at the same time minimising uncertainty for rights holders and licensees where practices are socially valuable — would provide more meaningful benefits. As the Federal Trade Commission in the United States has observed, the objective is to achieve a balance: Agencies must apply antitrust principles to identify illegal collusive or exclusionary conduct while at the same time supporting the incentives to innovate created by intellectual property rights. Condemning efficient activity involving intellectual property rights could undermine that incentive to innovate, and thus slow the engine that drives much economic growth in the United States. However, failure to challenge illegal collusive or exclusionary conduct, involving intellectual property as well as other forms of property, can have substantial negative consequences for consumers. (2007, p. 2) This approach is consistent with the advice put forth by the Business Council of Australia, which stated: The proponents of repeal argue that intellectual property rights can be used anticompetitively and should be fully subjected to the Competition and Consumer Act. However, intellectual property rights are also different to other rights in that their protection provides an important incentive to innovate. In the event that it is made more difficult to obtain a patent, it may result in less inventors filing for patents, and subsequently lead to less innovation. Changes to the law that reduce the incentive to innovate should be avoided, and all potential costs to the economy need to be tested. In this context, the Productivity Commission may wish to consider whether there are ways of improving the effective operation of licensing, rather than simply removing the exemption. (sub. 59, p. 8) 452 INTELLECTUAL PROPERTY ARRANGEMENTS Guidelines would help minimise uncertainty One way of addressing genuinely anticompetitive conduct while also minimising uncertainty for rights holders and licensees where practices are socially valuable, would be to repeal s. 51(3), at the same time as the ACCC issuing guidelines and addressing concerns about the per se prohibitions under the CCA. Proposals for the use of guidelines to minimise uncertainty following repeal of s. 51(3) have been a feature of reviews of s. 51(3) since at least 1999. Guidance should set out clearly the factors that would guide the regulator in determining whether there may be anticompetitive effects in breach of the law. This approach would be similar to that adopted in a range of other jurisdictions. For example, the United States and Canada both provide an exemption through regulatory guidance (box 15.4). The ACCC and a number of academics supported the proposed approach of repealing s. 51(3) and issuing guidance on application of the section to IP: … guidelines issued by the ACCC could provide rights holders with guidance on whether conduct is likely to breach the CCA, and where authorisation under Part VII of the CCA might be available. The ACCC may grant authorisation, broadly speaking, if it is satisfied that the likely public benefits of the conduct outweigh the likely public detriment. Guidelines addressing the application of competition laws to IP have previously been issued in Australia … The ACCC notes also that guidelines addressing the potential treatment of various types of conduct involving IP have been issued in the United States and Canada. (ACCC, sub. DR603, p. 6) We strongly endorse this recommendation [to repeal s. 51(3)] … we support the implementation of guidelines by the ACCC in order to provide much-needed clarity to IP holders as to when their conduct is likely to contravene the Competition and Consumer Act 2010 (Cth). (Alexander et al., sub. DR505, p. 34) The Copyright Council (sub. DR543, p. 12) also indicated its support for ACCC guidance if s. 51(3) were repealed. INTELLECTUAL PROPERTY RIGHTS AND COMPETITION LAW 453 Box 15.4 Relief from competition law for IP in overseas jurisdictions Other jurisdictions provide relief from competition law to reduce uncertainty and compliance costs. This is often implemented in regulatory guidance, rather than by a statutory exemption. The United States Department of Justice and the Federal Trade Commission issued the Antitrust Guidelines for the Licensing of Intellectual Property (Guidelines) in 1995, which provides licensors of IP with guidance on when action will be taken in respect of anticompetitive effects. The guidelines note that licensing of IP rights is generally pro-competitive and innovation enhancing. The Guidelines state that identifying whether a relationship is vertical or horizontal is an aid to determining whether there may be anticompetitive effects arising from a licensing arrangement. Horizontal arrangements are more likely to give rise to anti-trust concerns. The Guidelines put in place an anti-trust ‘safety zone.’ The safety zone is ‘designed to provide owners of IP with a degree of certainty in those situations in which anticompetitive effects are so unlikely that the arrangements may be presumed not to be anticompetitive without inquiry into particular industry circumstances.’ As noted in the Guidelines, the safety zone applies if: The restraint is not facially anticompetitive; and The licensor and its licensees collectively account for no more than twenty percent of each relevant market significantly affected by the restraint; or There are four or more independently controlled technologies in addition to the technologies controlled by the parties to the licensing arrangement that may be substitutable for the licensed technology at a comparable cost to the user; or Four or more independently controlled entities in addition to the parties to the licensing arrangement possess the required specialized assets or characteristics and the incentive to engage in research and development that is a close substitute of the research and development activities of the parties to the licensing agreement. The European Union provides an exemption from competition law for certain licensing arrangements for IP. The exemption takes the form of a Commission Regulation known colloquially as the ‘Technology Transfer Block Exemption’ (TTBE). Article 3(1) of the TTBE provides a presumption that technology transfer agreements between competitors with a market share of less than 20 per cent ‘ … generally lead to an improvement in production or distribution and allow consumers a fair share of the resulting benefits.’ Article 3(2) provides the same presumption for non-competitors with a market share of 30 per cent. Article 4 provides that the exemption is not available for certain ‘hardcore restrictions’, including restrictions on the ability of a party to determine prices when selling products to third parties and limitations on output. In Canada, the Intellectual Property Enforcement Guidelines (2016) (Enforcement Guidelines) provide that ‘The Bureau generally does not challenge the conduct of a firm that possesses less than a 35 per cent market share.’ Paragraph 5.2.3 of the Enforcement Guidelines state that ‘the Bureau focuses on whether the conduct will result in horizontal anticompetitive effects, in other words, consequences for firms producing substitutes or firms potentially producing substitutes.’ Sources: US Department of Justice and the Federal Trade Commission (1995); Commission Regulation (EU) No 316/2014 (‘Technology Transfer Block Exemption’); CBC (2016). 454 INTELLECTUAL PROPERTY ARRANGEMENTS Reforming per se prohibitions would also help Reforming the per se prohibitions in the CCA would also help ensure that socially valuable IP licensing or assignment activities would not be foregone following removal of the s. 51(3) exemption. The per se provisions prohibit certain conduct without requiring a ‘substantial lessening of competition’. Other provisions of part IV, by contrast, require that conduct has that effect before a prohibition applies. Per se prohibitions in part IV of the CCA of relevance to IP include: • cartel conduct that amounts to price fixing (Division 1 of part IV) • exclusionary conduct (s. 45(2)) • third line forcing (ss. 47(6) and 47(7)). The per se prohibitions have been a brake on the repeal of s. 51(3). The review of IP legislation under the Competition Principles Agreement (2000) pointed to the possible negative effects of the per se prohibitions if the s. 51(3) exemption were repealed: The Committee does not believe that simple repeal of the section would be desirable. Any assessment of repealing the section must take account of the effect repeal would have on licensing and assignment decisions. More specifically, it seems reasonable to suppose that the per se prohibitions embodied in the Act, and the potentially burdensome requirements for administrative review, would catch many license conditions that are usually socially beneficial. For example, tying and exclusive dealing arrangements in patent licenses. Over the longer term, this could both reduce innovation and distort competition between those (typically smaller and more specialised) firms that depended on licenses and assignments and those that did not. (2000, p. 212) The Competition Policy Review (Harper et al. 2015) considered each of the per se prohibitions and recommended either a competition test (with respect to price fixing and third line forcing) or repeal (with respect to exclusionary conduct). While not the primary motivation of the Competition Policy Review Panel, Giving effect to these recommendations would remove the remaining impediment to the repeal of s. 51(3). The relevant recommendation from the Competition Policy Review is set out in box 15.5. INTELLECTUAL PROPERTY RIGHTS AND COMPETITION LAW 455 Box 15.5 Competition Policy Review recommendations on the cartel provisions of the Competition and Consumer Act 2010 Recommendation 27 – Cartel conduct prohibition The prohibitions against cartel conduct in part IV, Division 1 of the CCA should be simplified and the following specific changes made: • The provisions should apply to cartel conduct involving persons who compete to supply goods or services to, or acquire goods or services from, persons resident in or carrying on business within Australia. • The provisions should be confined to conduct involving firms that are actual or likely competitors, where ‘likely’ means on the balance of probabilities. • A broad exemption should be included for joint ventures, whether for the production, supply, acquisition or marketing of goods or services, recognising that such conduct will be prohibited by s. 45 of the CCA if it has the purpose, effect or likely effect of substantially lessening competition. • An exemption should be included for trading restrictions that are imposed by one firm on another in connection with the supply or acquisition of goods or services (including intellectual property licensing), recognising that such conduct will be prohibited by s. 45 of the CCA (or s. 47 if retained) if it has the purpose, effect or likely effect of substantially lessening competition. Source: Harper et al. (2015, pp. 363, 597). Few participants commented on the intersection between the per se provisions and the removal of s. 51(3), however, Microsoft noted: Applying per se prohibitions to IP licensing is likely to result in overbroad prosecution of procompetitive, socially and economically beneficial conduct. We therefore agree with the Draft Report’s recommendation that if section 51(3) is repealed than reforms of the per se prohibition in the Competition and Consumer Act, along the lines of the recommendations in the Harper Competition Policy Review should be instituted at the same time. It would be helpful to include this point in Recommendation 14.1. (sub. DR420, p. 8) The recommendations of the Competition Policy Review (Harper et al. 2015), in respect of the per se prohibitions, address the barrier that the per se prohibitions may have previously posed to repeal of s. 51(3). These recommendations have been accepted by the Government, but legislation is yet to be introduced into the Parliament. 456 INTELLECTUAL PROPERTY ARRANGEMENTS RECOMMENDATION 15.1 The Australian Government should repeal s. 51(3) of the Competition and Consumer Act 2010 (Cth) (Competition and Consumer Act) at the same time as giving effect to recommendations of the (Harper) Competition Policy Review on the per se prohibitions. The Australian Competition and Consumer Commission should issue guidance on the application of part IV of the Competition and Consumer Act to intellectual property. 15.2 Compulsory access to intellectual property In addition to the exemption that applies under s. 51(3), IP is also exempt from part IIIA of the CCA — the ‘National Access Regime’. The Regime provides a mechanism via which the Government can potentially set the price for access to services provided by infrastructure that meets certain criteria (including ‘national significance’) in order to promote competition. The Regime is intended to address an economic problem that can arise with certain infrastructure. The Hilmer Committee described the problem associated with so-called essential facilities: Some facilities that exhibit these [natural monopoly] characteristics occupy strategic positions in an industry, and are thus ‘essential facilities’ in the sense that access to the facility is required if a business is to be able to compete effectively in upstream or downstream markets … Where the owner of the ‘essential facility’ is vertically-integrated with potentially competitive activities in upstream or downstream markets … the potential to charge monopoly prices may be combined with an incentive to inhibit competitors’ access to the facility. (1993, p. 240). Examples of infrastructure services that have been ‘declared’ under the Regime (which provides access seekers with a right to seek arbitration in the event of an access dispute) include railway track, airport and sewage transmission services. The ACCC has suggested that there could be a case in the future for extending the Regime to allow it to arbitrate access disputes that involve IP. The ACCC states: … if access to particular IP becomes more restricted in the future due to the pace of technological advancement, there may be a need to consider the effectiveness of existing access mechanisms. … In the event that existing frameworks prove not to be effective in ensuring efficient access in the future, some legislative change to access regimes may require further consideration. (sub. 35, pp. 16–17) INTELLECTUAL PROPERTY RIGHTS AND COMPETITION LAW 457 It is not clear what forms of IP would meet the criteria for declaration under the Regime. In addition to needing to be nationally significant, it must be uneconomic for anyone to develop another facility to provide the service, and access would need to promote a material increase in competition in at least one market other than the market for the service. IP specific provisions While part IIIA of the CCA does not apply to IP, other arrangements can provide analogous outcomes. The Commission recently completed a review of the National Access Regime in which it found it to be working relatively well. In commenting on the exemption for IP from part IIIA, the Commission noted that: Apart from goods, other exclusions from Part IIIA include the use of intellectual property. … The [rationale] for these exclusions [is that] … access to intellectual property is already covered by dedicated licensing arrangements … (PC 2001, p. 152) Indeed, provisions of the Patents Act, Plant Breeder’s Rights Act and the Copyright Act all provide for compulsory access to IP in given circumstances. The conditions under which compulsory access can be granted differ across IP rights, as do the parties responsible for settling disputes. The Patents Act provides for compulsory licensing of patents if the holder cannot give ‘a satisfactory reason for failing to exploit the patent’ and the ‘reasonable requirements of the public have not been met.’ Similarly, the Copyright Act provides access to sound recordings in some circumstances. The Plant Breeder’s Rights Act provides broad powers for the Secretary of the Department of Industry, Innovation and Science to provide compulsory access to plant varieties (table 15.2). 458 INTELLECTUAL PROPERTY ARRANGEMENTS Table 15.2 Comparison of compulsory licensing regimes for IP Patent What are the Applicant has sought authorisation conditions for from patentee for a reasonable time on reasonable terms and conditions access? Reasonable requirements of the public have not been satisfied Patentee has not given a satisfactory reason for failing to exploit the patent OR The patentee has breached part IV of the CCA Who decides terms of access? Federal Court As agreed between the parties or ‘just and reasonable having regard to the economic value of the licence and the desirability of discouraging contraventions of part IV of the CCA.’ How often is it Seldom (3 applications; none used? successful) What are the Moderate ($100 000 - $1 million) transaction costs? What is the price? Copyright Plant Breeder’s Rights Rights of access for government and educational institutions for broadcasts, literary, artistic and dramatic works Grantee is failing to provide reasonable public access to a plant variety and it affects the interests of the applicant Negotiated between collecting societies and users, with Copyright Tribunal to arbitrate disputes Negotiated between collecting societies and users, with Copyright Tribunal to arbitrate disputes Often Secretary of the Department of Industry, Innovation and Science Low Low Reasonable prices, as determined by Secretary of the Department of Industry, Innovation and Science Never Little use has been made of these provisions. So far, no one has sought access for plant breeds, and there are only a few instances of parties advancing matters to the Copyright Tribunal (chapter 6). In respect of patents, there have only been three applications for a compulsory license in Australia since they became available and none of the applications resulted in an order for a compulsory licence. While the provisions under the Patents Act, Plant Breeder’s Rights Act and the Copyright Act are seldom used, this is partially explained because it is usually in the interests of a rights holder to license an invention or work, and so there is a limited range of circumstances where the provision would need to be invoked. As the Commission (2013a) explained in its review of Compulsory Licensing of Patents, provisions which allow for compulsory access are an important safeguard that provides incentive for rights holders to engage in voluntary negotiations with potential licensees. It is not clear that relying on access provisions under part IIIA would provide the same incentives. Part IIIA is complex, the transaction costs of an application are high and the barrier to having a particular facility ‘declared’ under part IIIA is extremely high, including, for example, a test of ‘national significance.’ As noted in table 15.2, compulsory access for plant breeder’s rights and copyright, to the extent to which it is available, involves low transaction costs. Even the relatively restricted access regime for patents is likely more permissive than part IIIA. INTELLECTUAL PROPERTY RIGHTS AND COMPETITION LAW 459 Accordingly, the Commission does not consider that part IIIA of the CCA should be extended to IP. Instead, the existing, narrowly targeted, provisions of each of the relevant acts should continue to govern when compulsory access should be available. Is section 46 an alternative access regime? Several authors have pointed to the potential for s. 46 of the CCA to also operate as an alternative access regime (Corones 2005; Nielsen and Nicol 2008). Section 46 prohibits a corporation that has a substantial degree of power in a market from taking advantage of that power for the purpose of: • eliminating or substantially damaging a competitor • preventing the entry of a person into that or any other market or • deterring or preventing a person from engaging in competitive conduct in that or any other market. Nielsen and Nicol point out that a breach of s. 46 of the CCA can be the basis for making an order for a compulsory licence (Nielsen and Nicol 2008, p. 332). In its inquiry into Compulsory Licensing of Patents, the Commission found that although not widely used, s. 46 had been used in some instances to provide access to copyright-protected information. The Australian Government has announced that it will strengthen s. 46. The proposed new law would allow the ACCC to intervene if a corporation with substantial market power engages in conduct that has the purpose, effect or likely effect of substantially lessening competition. The availability of s. 46 to address instances of anticompetitive denial of access to IP, in addition to the specific compulsory licensing regimes, further strengthens the case that Part IIIA should not be extended to IP. 460 INTELLECTUAL PROPERTY ARRANGEMENTS 16 IP and public institutions Key points • Australian governments provide significant research funding. Some publicly-funded research can be commercialised and, where this is the case, IP arrangements can play a facilitating role. • Commercialisation incentives generated by IP arrangements must be balanced against their potential to limit knowledge dissemination. • IP policy can alter the balance of activities undertaken by Australian research institutions. IP policy should be neutral in its influence. • Publication is an important mechanism for knowledge diffusion from publicly-funded research. But access to published material is often limited by copyright. − All governments should adopt open access policies (free access once a decision to publish is taken) for the results of publicly-funded research. The Australian Government should encourage international agencies it co-funds to adopt a similar policy. − Importantly, such an open access policy does not in itself compel publication. • Current policy settings allow public-funded institutions to assert ownership rights over publicly-funded IP. International experience suggests that this is more effective than the alternatives of government ownership or ‘professor privilege’. • OECD indicators suggest research institution collaboration with industry is relatively low in Australia. This concern has prompted government and industry-led initiatives to improve the commercialisation of publicly-funded research. It is too early to assess the efficacy of these recent initiatives. • Other interventions, such as US-style ‘march in’ arrangements are not warranted at this point. They would impose costs and are not needed in the presence of Australia’s compulsory licensing arrangements. Governments play a major role in shaping the innovation system through the design and governance of institutions, supporting the education and training of scientists and experts, and funding high-value research that would not otherwise be undertaken (OECD 2015c; PC 2007a). Governments also play a direct role through their own public sector research agencies and by funding research and development (R&D) (and sometimes commercialisation) in universities and businesses. Table 16.1 sets out the many policy levers, including (intellectual property) IP policy, available to government to encourage innovation. IP AND PUBLIC INSTITUTIONS 461 Table 16.1 Policy levers to encourage innovation Main features Research direction Ownership of results Main advantages Main disadvantages Generic IP owner (firm or institution) Decision on R&D decentralised Static misallocation of resources Requires private ex ante financing of R&D Depends on Mobilises contract competitive forces for the provision of public goods Difficult to write perfect contracts Privately funded and executed IP rights Market exclusivity Applied Publicly funded and privately executed Procurement Government purchases of well-defined innovative goods — for example, military equipment Generic Applied Research subsidies and direct government funding Public support for targeted Generic research Applied Usually firm Mobilises competitive forces for public benefit Governments are imperfectly informed about potential success of R&D projects Prizes Prizes for targeted solutions to specific problems Usually public Mobilises competitive forces for public benefit Subsequent competitive provision of technology Difficult to write perfect contracts Requires private ex ante financing of R&D R&D tax credits and related fiscal incentives Reduced taxation of profits Generic linked to investment in Applied R&D Firm Decisions on R&D decentralised Does not address firms’ appropriability problem Requires private ex ante financing of R&D Generic Applied Publicly funded and executed Public Public goods such as research defence and health organisations Does not undertake commercialisation of knowledge Basic Generic Public Institution Advance fundamental scientific knowledge Uncertain impact Academic research Basic Generic Public Institution Advance fundamental scientific knowledge Uncertain impact Aimed at increasing basic scientific knowledge Does not undertake commercialisation of knowledge Source: WIPO (2011, p. 85). Given the wide-ranging scope of this inquiry, and the recent review of the R&D tax incentive (Australian Government 2016d), this chapter limits its focus to the intersection 462 INTELLECTUAL PROPERTY ARRANGEMENTS between IP rights and publicly-funded innovation. This interplay has important implications for how resources are allocated across the many forms of research: The system has to encompass all varieties of research: short term and long term; high risk and low risk; curiosity driven, investigator led research and experimental development; research in different fields and across different sectors. Achieving the right balance is not easy … (CSIRO 2006, p. 37) The chapter is structured in four parts. Section 16.1 examines the general role IP rights play in the presence of public funding. The remainder of the chapter explores how policy settings for the creation and use of IP can influence the extent to which the benefits of publicly-funded research reach the community. The main policy choices are whether IP rights should be deployed (section 16.2) and which party should hold, if any, the IP rights (section 16.3). The chapter concludes by assessing the policy case for purported safeguards, such as ‘use it or lose it’ provisions (section 16.4). The consideration of a ‘use it or lose it’ provision is in response to a recommendation in the review of Research Policy and Funding Arrangements (Australian Government 2015c) and a subsequent request from the Treasurer for the Commission to examine the issue as part of this inquiry. 16.1 IP rights and public funding of research Public funding of research is a means of correcting the under-provision of certain types of research that offer benefits that cannot be sufficiently captured by private investors even in the presence of IP rights. Understanding the basis for public funding is important in determining the desirable interplay between IP policy and the broader innovation system. As noted in recent work by the OECD, the motivations for public investment in innovation are more complex than commonly understood (OECD 2015c). The orthodox arguments for publicly-funded (and sometimes provided) innovation centre on the characteristics of knowledge and (to a lesser extent), the role government plays sometimes as a producer of goods and services: • In many areas of new knowledge, it is inherently difficult to enforce IP rights (such as mathematical concepts and the science of climate change) — so-called non-excludable knowledge. This is often intrinsic to basic research. 1 This characteristic of knowledge means that there is sometimes little scope for private agents to appropriate sufficient returns to motivate the production of such knowledge. • In some instances, the benefits of research cannot be appropriated by a business because there is no market for the ‘products’ to which that knowledge relates. For 1 Just under 50 per cent of public funding for research by higher education institutions is for basic research, around 45 per cent for applied research and less than 10 per cent for experimental development (ABS 2012). IP AND PUBLIC INSTITUTIONS 463 instance, environmental externalities are not priced in markets. The absence of a return means the incentives for commercial innovation in this area are muted. • In some cases, it is more efficient for government to act as an implicit agent for multiple users of new knowledge, and to innovate on their behalf, reflecting the large number of users, and the trade-offs between the near zero cost of diffusing knowledge and the significant transaction costs in determining payments for the use of new ideas. • Governments may also sometimes be the only or dominant producer of certain services and have a comparative advantage in undertaking (or commissioning) innovative activities in these areas (for example, in defence and security). The key questions for this inquiry relate to: • Does the IP system frustrate the achievement of the goal for public-funded research? • Would changes to the IP system accentuate the benefits of such public funding? These questions are not (materially) relevant to some of the factors listed above. For example, it is not clear that the IP system frustrates the generation or use of ideas where there is no market for the ‘products’ to which knowledge relates, or where governments are dominant producers. In respect of the other factors, a common theme is that the public ‘pays twice’ for research, once to fund research and again through higher prices for goods or services protected by the IP derived from that research (Siepmann 2004, p. 236). As outlined later, this is an overly simplistic representation of the tension between IP policy and the policy intent of government funding, but it nevertheless serves as a useful starting point for consideration of many of the issues. The policy relevant issues emerging are fourfold: • A major mechanism for diffusion of ideas is through publication. Research is paid for by governments, by philanthropic and not-for-profit bodies, and sometimes by businesses. Yet access to this information can be limited through copyright protection, which allows parties to charge for access. • Even though much of the knowledge generated by public-funded entities (including universities) is of a basic nature, this is not always the case, and parties undertaking basic research sometimes identify follow-on commercial opportunities. How far should publicly-funded agencies go in securing IP protection, when public funding, career advancement and prestige already motivate the creation of the ideas? • Where it is deemed appropriate to patent an ‘invention’ made by a publicly-funded entity, the question arises as to which party should own the property rights — the researcher, the institution, or the funder (or some agent representing their interests). There is no parallel complexity for private businesses in this area. 464 INTELLECTUAL PROPERTY ARRANGEMENTS • There is a broader, related question about whether the behaviour and culture of public research agencies may be inimical to the realisation of commercial opportunities associated with their knowledge generation. While there are undoubtedly other complementarities and tensions between statutory IP arrangements and public funding of research, these four issues are the basis for the sections that follow. 16.2 Should IP rights be afforded for the results of publicly-funded research? Copyright for publicly-funded research Publication is an important mechanism for disseminating the results of publicly-funded research. Around one-third of innovation-active business identified these as a source of ideas (Australian Government 2015c, p. 48). Moreover, given the often-iterative nature of basic research — each academic’s research builds on another’s — the greatest public benefit arises from the publication of research results. This allows follow-on research that may eventually lead to discoveries that offer significant public benefits. Copyright has always played an idiosyncratic role in academic publishing in that creators have very strong motivations to produce the relevant knowledge, as do the institutions in which they sit. The researchers receive funding for their work, so they do not need an economic return to stimulate their innovation. This is reinforced by the desire by academic institutions to gather prestige to attract government funding, students and philanthropic donations; and by the desire by academics to secure tenure and status through publication in leading journals. ‘Publish or perish’ encapsulates the incentives nicely. As the University of Sydney stated: For certain individuals it is not the intellectual property system that encourages creativity, innovation and authorship. An example is the case of university academics in relation to peer-reviewed journal articles and the dissemination of findings of university or Australian Government funded research. In our experience, an academic’s research and creative output is largely driven by an incentive to establish and enhance their reputation and career, and encourage open and informed debate in their discipline. The receipt of remuneration or royalties from scholarly works is not a primary incentive to publish. (sub. 104, p. 8) Equally, the role of (and costs borne by) publishers in marketing copyright works in fiction, music and video content to consumers are not as obvious in the academic journal arena, where bibliographic searching tools or web-based search engines are the dominant method for identifying relevant material. Nevertheless, there has been a compelling longstanding argument for copyright protection of academic knowledge in that there are costs associated with sifting through paper IP AND PUBLIC INSTITUTIONS 465 submissions by academics, and in coordinating the publication process (finding and chivvying referees for example) that must be funded through a revenue stream. Were there no protection, a competing publisher could simply copy the work of another, and would only have to meet the typesetting and printing costs. Accordingly, copyright protection was seen as a vehicle for dissemination of research work. Whether it did this efficiently has been questioned. Even in a world in which research results were disseminated through paper, there were alternative models that may have been more efficient, such as ‘author pays’. 2 The digital age has disrupted this model — or more precisely, has the potential to do so. • Physical copies are no longer necessary as papers can be provided online, which makes free access feasible. • The online environment creates alternative ways in which papers can be reviewed and ranked for quality (for example, not just by reviewers, but by readers). It has typically been the case that referees are unpaid, so the key task in producing an online journal is the establishment of a reputation and the coordination of blind refereeing and coherent editing standards. The latter has now been aided by freeware such as LaTex, which is widely used in academic publications. Universities and some publishers now provide free or low cost access to online peer-reviewed research papers. For example, the Directory of Open Access Journals covers around 11 500 journals and 2.3 million papers (though many prestigious journals published by for-profit publishers are not included). To provide an illustration, in the discipline of economics, the key journals from the American Economic Association (the American Economic Review and the Journal of Economic Perspectives) are available free online. Nevertheless, many private publishers still continue with a ‘pay-for-article’ online model (for example, as in Elseviers’ European Economic Review). While new technologies have facilitated free open access to copyright material, many research publications continue to be sold in both paper and online versions. As Australasian Open Access Strategy Group observed (and as demonstrated in figure 16.1), universities continue to spend considerable funds on journal subscriptions: The fact that universities are able to provide access to journals may be seamless, but it is at great cost. In fact the vast majority of research journals require a subscription. In 2014, Australian universities paid AUD 221 million (data from the Council of Australian University Librarians, CAUL) for access to electronic journals. (Australasian Open Access Strategy Group, sub. DR431, p. 3) 2 Notably, some US universities rejected the author pays model, exemplified by Yale’s withdrawal from BioMed Central in 2007 after it found the costs prohibitive (Gawrylewski 2007). However, Yale re-joined in 2015 (Yale University 2015). 466 INTELLECTUAL PROPERTY ARRANGEMENTS Figure 16.1 Expenditure on e-resources by Australian university libraries 250 Millions ($A) 200 150 100 50 0 2004 2005 2006 2007 2008 2009 2010 … 2014 Year Source: Council of Australian University Libraries, accessed at www.caul.edu.au. Recognising that further incentives are not needed to encourage dissemination of publicly-funded research through publication, but that charging practices nonetheless continue, jurisdictions are increasingly requiring research results to be available for free under ‘open access’ arrangements within a given period after publication. For example, in the United States, since 2013, federal agencies with over $100 million in annual research and development expenditure must develop a plan for making peer-reviewed manuscripts or final published papers publicly available within 12 months of publication (Holdren 2013, p. 2). Open access arrangements have also been adopted by Research Councils in the United Kingdom, and in Europe under the ‘Horizon 2020’ program (EC 2013). While immediate open access publishing is preferable, these open access arrangements sometimes allow embargo periods (often of 6 or 12 months, depending on the field of research), during which publishers can charge a price for exclusive access. The trend towards open access arrangements is transforming the way in which research results are communicated. As Universities Australia points out: In just a few short years, open access publishing has dramatically changed the scholarly communications landscape. More than 50 funding agencies around the world require open access to peer-reviewed articles arising from the research they fund. According to Dr Peter Suber, Director of the Office for Scholarly Communication at Harvard University, this number is not only growing, but the growth is accelerating. Dr Suber says: Funders are charities or philanthropies, and that explains why they grasp the logic of open access. If a research project is worth funding, then its results are worth sharing. Funders IP AND PUBLIC INSTITUTIONS 467 have no reason to hold research back, in order to generate a revenue stream or meter it out to paying customers. On the contrary, they have every reason to make it available to everyone who could make use of it. (sub. 71, att. 1, p. 28) A similar trend has also emerged in Australia. In 2013, the National Health and Medical Research Council (NHMRC) and the Australian Research Council (ARC) introduced limited open access policies, requiring any publication arising from research to which they contribute funding to be freely accessible to the public within 12 months of publication. 3 Many other government providers of information now voluntarily make their material available online at zero cost because their previous cost recovery model was principally for publication costs of physical volumes. The Australian Bureau of Statistics and the Productivity Commission itself are examples where this has occurred. Even so, there is not yet a comprehensive policy covering all publicly-funded research. For example, CSIRO provides open access to its publications, but has author pays arrangements. 4 Reflecting this, the Australian Government has indicated that: The Government will develop a policy to ensure that more publicly-funded research findings are shared openly and available to be used commercially or in other ways that will bring the greatest benefit to Australians. (Australian Government 2015d, p. 16) The Commission supports measures to release the results of publicly-funded research under open access arrangements. This imperative goes beyond the Australian Government, and applies to publicly-funded research by state and territory governments, and universities (whose research funding extends beyond the ARC and NHMRC). It should also include international agencies to which Australia is a contributory funder, but which still charge for their publications (such as the OECD). Some participants in the inquiry misinterpreted the Commission’s proposal as limited to publications in peer-reviewed academic journals. It is intended that the proposal apply to all publicly-funded research results in a publication. This includes, but is not limited to, journal articles, books, book chapters, websites, pamphlets and so-called ‘grey literature’ (see chapter 4 for a discussion of ‘grey literature’). Inquiry participants also argued that there was a need to cater for situations where commercial imperatives required restricted access until other IP rights are sought: … the University of Queensland is supportive of publicly funded research being open access after publication so long as there are provisions to allow for situations which also involve industry partners who for commercialisation reasons need to restrict access until appropriate registered rights are sought. (University of Queensland, sub. DR200, p. 1) However, open access arrangements do not compel publication, rather, they provide free access once a decision to publish has been taken (an important point of difference). Responsibility for compliance and implementation rests with funding recipients who must 3 National Health and Medical Research Council, 2014, NHMRC’s Policy on the Dissemination of Research Findings and Australian Research Council, 2015, ARC Open Access Policy. 4 http://www.publish.csiro.au/nid/247.htm. 468 INTELLECTUAL PROPERTY ARRANGEMENTS justify why they have not complied with the open access policies. To reinforce open access policies, the recent review of Research Policy and Funding Arrangements (Australian Government 2015c) recommended that reporting arrangements take into account the relative share of research output made available through publication or open source repositories. Publishers who participated in this inquiry argued that flexibility should be allowed in the embargo period for open access (Wiley Australia, sub. DR494; Association of American Publishers, sub. DR338; Oxford University Press Australia and New Zealand, sub. DR448). The Association of American Publishers provided evidence of the half-lives of published journal articles to support an argument that a 12 month embargo is too short: In the field with the shortest half-life – health sciences – it took more than 24 months for the majority of articles to experience half their total downloads. In fields such as mathematics, physics and the humanities, median half-lives exceeded 48 months. Overall, articles in the majority of journals received more than 50% of their usage at least 36 months after publication. Only 3% of journals in all fields experienced half-lives aligned with the arbitrary 12-month embargo period proposed in the Draft Report. (sub. DR338, p. 8) An alternative interpretation of the evidence presented on the half-lives of publications is that it demonstrates the extent of public benefit to be derived from a move to open access. As noted above, a 12 month period is emerging as both the Australian and international norm. The 12 month period is not intended to provide protection for the half-life, or any particular proportion of the life of a publication. Instead, it is intended to provide sufficient exclusivity to allow publications a revenue stream to fund the now much reduced costs of the administrative tasks associated with a publication (as distinct from the already funded cost of the research itself). One consideration is how open access would apply to research that is partly privately-funded. Where private parties make a substantial funding contribution, any open access requirements would need to be agreed between the parties to any funding arrangements. Implementation of an open access policy would also need to consider what transition period is required to enable funders of research, researchers, publishers and institutions to adapt their practices and procedures to the new arrangements. The University of Melbourne suggested a three year transition period would be appropriate (sub. DR560, p. 2). The Commission considers that a three year transition period represents more than a reasonable period. Open access to other types of publicly-funded research output (or input) Several submissions suggested that open access should extend to the results of publicly-funded research which may take forms other than publications: IP AND PUBLIC INSTITUTIONS 469 [the open access recommendation] should be re-scoped or extended beyond publications to include publications (the ‘traditional’ form of research outputs) plus research data, techniques, algorithms and software (the non-traditional outputs). (Australian National Data Service, sub. DR314, p. 4) Such a policy should apply to all works of any form developed with public funds for any purpose (other than those which, for reasons beyond the scope of IP policy, must not be published at all in order to protect the privacy of citizens or the secrecy of sensitive military operations). (Open Source Industry Australia Ltd, sub. DR486, p. 31). The policy case for requiring public release of research outputs other than publications is not as clear. Publications can be distinguished from other research outputs because they are, by their very nature, public. As noted above, open access arrangements do not compel publication. By contrast, research techniques, algorithms and software may never become public in the absence of a specific requirement to make them available. This allows significant private and institutional investments to be made in these resources, which are often inputs to the research process. As discussed below, there needs to be more nuanced and critical assessment of the notion that the public pays twice. In certain circumstances, IP is needed to encourage investments. Universal requirements that they be freely available might also significantly reduce incentives for private sector collaboration and critical commercialisation support. The broader issue of access to publicly-funded data is being canvassed as part of a concurrent Commission inquiry on data availability and use. RECOMMENDATION 16.1 The Australian, and State and Territory governments should implement an open access policy for publicly-funded research. The policy should provide free and open access arrangements for all publications funded by governments, directly or through university funding, within 12 months of publication. The policy should minimise exemptions. The Australian Government should seek to establish the same policy for international agencies to which it is a contributory funder, but which still charge for their publications, such as the Organisation for Economic Cooperation and Development. While the Commission’s proposed open access initiative would be a major step forward, there may be future options for a more liberal regime that permits instantaneous access to a broader range of manuscripts through different models of publishing. For example, this may be through: • 470 author pays (or effectively ‘research institution pays’) with budget supplementation for this purpose. It is unlikely that such a model could initially have much reach since INTELLECTUAL PROPERTY ARRANGEMENTS many prestigious international journals do not operate on an author pays basis. That said, there are grounds for investigating closely the feasibility of encouraging and facilitating this model for those publications where that is not the case • encouragement of different ways in which the goals of academic publishing may be realised (Casella and Calvi 2009; Ware and Mabe 2009). Some of these objectives are readily met (such as those that create awareness, since there are already free high quality search engines, such as Google Scholar, for academic material). There are also already initiatives in this area. For instance, Philica is a multidisciplinary online ‘journal’ that achieves the certification objective by using anonymous professional referees, with the weight given to any given referee’s judgment determined by the quality certification of that referee’s works. Government may not need to play any role in this area other than through recognition of the legitimacy of new publishing models in its funding arrangements. But nor should it necessarily be passive. Patent protection for publicly-funded research A separate consideration is the circumstances in which patent protection should be granted to publicly-funded research. The National Principles of Intellectual Property Management for Publicly Funded Research (the National Principles) currently allow for patenting of publicly-funded research. There are three main drawbacks to current arrangements. First, IP holders can respond to uncertainty in a risk-averse way. IP protection may be sought ‘in case’ an invention has value. Further, once protection is obtained, IP owners may be reluctant to assign or license in return if the amounts offered are small even if these offers are commercially realistic. Second, the owner of a publicly-funded patent may not take into account costs and benefits that accrue to those external to themselves or their institutions. This can lead to IP being used to generate a return through scarcity of the relevant IP in circumstances where broader public dissemination would generate larger public benefits. For example, a particular invention patented by a publicly-funded institution might be capable of improving the effectiveness of all pain relieving drugs on the market. An exclusive licence might generate a large return for the institution, whereas broader licensing might generate larger public benefits. Third, cultural aspects of Australian business and universities can be a barrier to negotiating access to university-held IP, as noted by an association of technology universities and an association of scientists and engineers: … [the] cultural element is really important. IP is a really complex issue and sometimes the negotiations will take longer because there’s more complexity to that particular agreement and IP AND PUBLIC INSTITUTIONS 471 negotiation. But it’s not just the IP arrangements that are preventing collaboration. (Australian Technology Network of Universities, trans., p. 471) … universities can be difficult to deal with on IP issues. Now, again, that varies widely within the sector. There are some universities and tech transfer offices that are better than others, but often the complaint is that the starting point is lawyers at 20 paces and everything will be negotiated to the nth degree before we start any work … It’s more around how IP sharing and IP ownership is negotiated between parties. (Australian Academy of Technology and Engineering, trans., p. 498) It has also long been recognised that IP rights can stifle innovation. Several submissions to the Commission’s 2007 research report on Public Support for Science and Innovation expressed concerns about the adverse impact of IP rights on dissemination of knowledge and further research. Rooney and Mandeville noted that ‘overly strong IP can block the knowledge flow and thereby new knowledge creation as well’ (2006, p. 9). As the 2007 Commission research report noted, universities should, in some circumstances, ‘give their research away’: … for example, if the knowledge or technology is generally applicable to a wide range of firms and the costs of further development and replication of the resulting innovation are low. In this case, seeking to protect the IP and sell or license it delays its transfer and diffusion, potentially imposing costs on firms and the community. (PC 2007b, p. 290) Recognising that creation of IP over publicly-funded research results can sometimes be against the public interest, the ALRC, in 2004, recommended that the ARC and NHMRC should have the power to place conditions on funding to require that results are licensed widely or placed in the public domain: The ALRC also considers that in exceptional circumstances the ARC and NHMRC should be prepared to place conditions on grant funding to direct how any resulting technologies are exploited where it is considered that greater public benefit would result from the resulting research being placed in the public domain either with no patent being sought or, where a patent is sought, from being widely licensed. Provision for such conditions to be placed on the grant of public research funding should be incorporated into the National Principles. (ALRC 2004, p. 290) This recommendation was not given effect. There needs, however, to be more nuanced and critical assessment of the purported wisdom that the public pays twice when universities license patent inventions based on research that has already been publicly-funded. In certain circumstances, IP rights are needed to encourage commercialisation of the results of publicly-funded research. This is particularly the case given that public funding rarely extends to the commercialisation phase of a research project — usually on the basis that private support should be available if an idea is able to be commercialised. For instance, as noted by CSIRO: Even in cases where it is sought to obtain a public benefit, e.g. community or environmental benefits, by making a technology or practice freely available, it may be that enabling 472 INTELLECTUAL PROPERTY ARRANGEMENTS technologies or services are required to achieve the intended benefit. In the absence of a market for such enabling technologies or services, the primary technology may not be able to be deployed – and will thus fail to achieve the desired impact. The existence of IP rights can provide a mechanism here also to encourage investment in, and hence availability of, the required enabling technologies or services. (sub. 126, p. 2) Similarly, the University of Melbourne submission states: Effective patent protection is critical to supporting commercialisation and attracting further funding into development of most commercial products and services that arise from university inventions. In the absence of patent protection the commercial development would not occur. It should be said that effective IP rights encourage investment in innovation rather than encouraging innovation itself. In the absence of a patent system, private investment in innovation would not flourish even though public funding may still be available through government and philanthropic grants. (sub. 100, p. 5) There are also other considerations. For example, new inventions have value for users overseas, which should be exploited. It may sometimes be in the public interest to obtain rents through patent licensing to overseas users so long as the gains exceed those of open access in Australia. There is, therefore, no clear imperative that universities and public research agencies provide open access to their ‘inventions’ (either by simply giving the ideas away or through patenting and free licensing). Institutions should have flexibility to decide how to make inventions available to the public. One method would be public access without IP protection; another may be to seek protection, but broadly license IP rights at a minimal cost. The former would involve lower transaction costs but involves no control over what entity accesses the invention. The latter would involve higher transaction costs but offer a degree of control over access. A degree of control may be desirable, for example, to direct benefits of research funded by the Australian public towards the providers of that funding rather than making research results publicly available worldwide. The optimal actions should depend on the context, and in that vein, the decisions are best delegated to research agencies or their nominated agents. Australian universities have been active in this area. Seven Australian universities use the Easy Access IP model, which offers a simplified one page contract for IP. The research entities may elect to provide early-stage or high-risk IP free of charge to companies to evaluate the IP and to use it quickly. The businesses must pay for patenting costs and acknowledge the source of the ideas. If a business does not exploit the IP, it reverts back to the university in three years. Regardless, the universities can use the IP for research and teaching purposes as they wish (Australian Government 2015c, p. 59). IP AND PUBLIC INSTITUTIONS 473 16.3 Who should own IP arising from publicly-funded research? Given that there are some circumstances in which IP rights should be sought for publicly-funded research, a subsidiary policy question is who should own the IP. While institutional ownership is becoming the global norm, there are two additional options: • the individual inventors, often referred to as ‘professor privilege’ • the government agency providing the research funding As set out in more detail in chapter 3, the best option is that which provides the greatest incentive to engage in additional socially valuable innovation without creating excessive barriers to innovation in the form of ‘hold up’ rights that prevent inventors from building on earlier successful innovation. Ownership of IP by inventors — ‘professor privilege’ Until the mid to late 1980s, in most European countries, publicly-funded IP was owned by individual researchers under so-called ‘professor privilege’. Such arrangements continue to operate in a small number of countries, notably Italy and Sweden. A version of professor privilege also operated in Australia — ownership of IP was largely decided by the policies of individual institutions, with copyright often residing with individual researchers, but patents with institutions. However, following the publication of The National Principles of Intellectual Property Management for Publicly Funded Research (the Principles) in 2001, the practice of vesting IP rights with individual researchers all but ceased. Support for professor privilege still exists. The submission by the National Tertiary Education Union advocated adoption of professor privilege as an exception to the Copyright Act: The Act should be amended so that the ownership of all IP generated by university academic staff and researchers remains with creators and not with their employer as is the case in most other circumstances. (sub. 24, p. 7) Professor privilege has some benefits. It has the obvious appeal of providing individuals with the strongest financial incentive to commercialise their inventions. Although some studies have shown that commercialisation by an entrepreneur is more likely to be successful than commercialisation by an inventor, these studies also tend to show that continued involvement of an inventor has a positive impact on performance (Braunerhjelm and Svensson 2009; Damsgaard and Thursby 2013). Survey evidence suggests that three-quarters of all licensed inventions require further input from the inventor in order to be commercialised (Thursby, Jensen and Thursby 2001). 474 INTELLECTUAL PROPERTY ARRANGEMENTS But professor privilege also has several drawbacks. First, parties wishing to utilise publicly-funded IP incur transactions costs in sourcing the most effective property to meet their needs. These costs would likely be higher under professor privilege, as there would be a large number of inventors to approach, and each inventor may have idiosyncratic demands that would make contracting costly. Second, and as noted by the Australian Law Reform Commission, there is the potential for a lack of financial capacity and skills to act as a potential barrier to commercialisation if property rights are vested in inventors: … placing the onus to exploit intellectual property on researchers may be problematic where they lack the financial capacity to take their research results through to the commercialisation stage. Researchers also may not possess the business and legal expertise required for successful commercial negotiations. (ALRC 2004, para. 11.51) Indeed in cases where individual researchers overvalue their IP, particularly early stage IP, in the belief it will generate significant revenue, it could be that third parties are effectively priced out of the market or denied access. Professor Dianne Nicol and Dr Jane Nielsen highlight the costs involved when parties refuse to deal with downstream developers: Blocking can occur when the owner of a patent over foundational technology refuses to deal with a developer of downstream technology … The risk is that the timely delivery of new products and processes could be significantly hindered in these new areas of technology, which has both economic and social consequences. (sub. 61, p. 5) Third, professor privilege could result in some academics capturing economic value that has been created by their institution rather than through their individual intellectual endeavour. The often collaborative and incremental nature of research means that publicly-funded research results are rarely attributable solely to individual effort. Finally, professor privilege runs counter to a longstanding principle in Australian employment law under which an employer owns the IP created in the course of employment. Providing an exception to this principle for certain types of employees would open the question of why other types of employees do not have the same rights. Government ownership of publicly-funded IP An alternative to professor privilege is government ownership. Under this approach — which was the norm in the United States up until 1980 — the government agency that funds research retains ownership of any relevant IP. Superficially, this model appears attractive. The government funds multiple agencies and so it can avoid the complexities that individual researchers or institutions face when determining the ownership of jointly-produced IP. The government could hire high-quality experts in IP assessment, and be a central and clear point for negotiations with commercial and other IP-using entities. The government could have a clear mandate for the public interest in its decisions on patenting, which may not be as easy to achieve for IP AND PUBLIC INSTITUTIONS 475 funding-constrained research agencies or academics looking to secure higher returns from their work. But, theory aside, these types of arrangement do not work. Experience in the United States prior to 1980 revealed multiple problems with the government ownership approach. These problems have been well documented (Siepmann 2004). According to the then Commonwealth Department of Education Science and Training: • There were a large number of government funding agencies, each with a different policy on ownership • The government funding agency had the power to manage the commercialisation process but was often not in the best position to do so • Universities and other research institutions were reluctant to invest in commercialisation because there was no guarantee of obtaining exclusive rights. (Department of Education Science and Training 2003, p. 68) The deficiencies with government ownership arrangements in the United States prompted the introduction of the Bayh-Dole Act, which shifted ownership to institutions. The Bayh-Dole Act is generally considered to have been a significant success, with The Economist commenting: Possibly the most inspired piece of legislation to be enacted in America over the past half century was the Bayh-Dole Act of 1980 … .this unlocked all the innovations that had been made in laboratories throughout the United States with the help of taxpayers money. More than anything, this single policy measure helped to reverse America’s precipitous slide into industrial irrelevance. (The Economist 2002) The status quo: institutional ownership The move towards an institutional ownership model in the United States has subsequently been emulated in almost every OECD country, including Australia. In Australia, the National Principles of Intellectual Property Management for Publicly Funded Research have been adopted by the NHMRC and the ARC. Clause 2(c) of the Principles deal with the ownership of IP resulting from NHMRC, ARC or ‘other government research funding schemes’: Ownership and the associated rights of all IP generated as a result of Australian Government competitively funded research will initially be vested in the research institutions receiving and administering the grants as a way of recognising the inventive contribution made by the research institutions. IP generated as a result of collaborative endeavours between research institutions will vest as agreed between those institutions. The ARC and the NHMRC do not wish to hold a stake in direct ownership of IP nor do they intend to benefit directly from commercial outcomes of the research funded through their financial support. (Coordinating Committee on Innovation 2013, sec. 2(c)) 476 INTELLECTUAL PROPERTY ARRANGEMENTS There are several arguments in favour of institutional ownership of IP rather than by individual research teams: • Institutions are well placed to negotiate with their employees to address any principal-agent problems if ongoing assistance is required for commercialisation efforts. For example, profit sharing arrangements are common, with academics often retaining between 30 and 50 per cent of the income from commercialisation of their research (PC 2007b, p. 283). • Institutional ownership provides an incentive for institutions to commercialise the results of publicly-funded research because the institution retains any revenue from commercialisation efforts. • Institutional ownership allows for specialisation of the commercialisation role through the establishment of ‘technology transfer offices’ (TTOs). In Australia, each of the major research universities, and some of the smaller universities, has a TTO that specialises in the creation, management and enforcement of IP rights. Specialist staff with skills in business, law and marketing allow these institutions to effectively perform the role of commercialising the results of academic research. Some Australian TTOs have been very successful. For example, the University of Queensland’s UniQuest claims a 30-year history, over 700 patent applications, more than 400 contracts per year, 70 start-up companies and more than $11 billion in product sales (UniQuest 2014). • Institutions can more effectively manage the risks associated with the commercialisation of research. As TTOs generally manage a large portfolio of IP, losses on unsuccessful commercialisation efforts can be offset by gains on other commercialisation efforts (Collier 2007, p. 59). The current policy settings put a significant emphasis on TTOs making the right decisions about IP protection. The fact that the institutions to which TTOs belong bear a significant proportion of the costs (application and administrative fees) and receive the rewards (revenue from licensing or assignment) from exploitation of IP ensures that they take these costs and benefits into account. However, institutional ownership of IP does not always align with the public interest. Like individual researchers, TTOs may engage in risk-averse behaviour by routinely seeking IP protection just in case an invention has value, and being reluctant to assign or license IP without considerable payment in return. Institutional ownership of IP also means that any costs and benefits that are external to the relevant institution may not be taken into account. Concerns of this nature have been raised in the past. For example, the 2007 Commission Research Report on Public Support for Science and Innovation noted: In this study, two broad concerns were raised about the management of IP by universities, namely that there is a tendency for universities to: • overestimate the commercial value of their research; and, consequently IP AND PUBLIC INSTITUTIONS 477 • seek an unrealistic financial return from the sale or licensing of IP or equity stake in commercialisation projects. To the extent these concerns are valid, the transfer and diffusion of knowledge and technology to the business sector could be significantly impeded. (2007, p. 284) It is not necessarily the case that these problems reflect flaws in the National Principles. Rather, IP-owning entities may have unequal expertise in understanding the commercial value of their IP and thereby reaching collaborative arrangements for exploiting IP. On balance, while institutional ownership has some shortcomings, it remains preferable to government ownership or professor privilege to assign IP ownership for publicly-funded research. 16.4 Improving access to publicly-funded research Most analysis of Australia’s science system finds that Australia’s publicly-funded research organisations have generally high productivity as measured by their academic outputs. Overall, the sector is highly productive, internationally connected, and recognised globally for high quality research. For example, in 2013 we contributed to 3.9 per cent of the world’s research output (in terms of publications) from 0.3 per cent of the world’s population, ranking 9th in the OECD. Our research sector is also building on this strength and has improved its share of the top 1 per cent of publications from 3.8 per cent in 2004 to 6.7 per cent in 2013 (measured by relative citation impact). (Australian Government 2015b) Notwithstanding this high productivity, there is evidence that Australia performs poorly compared to other countries when translating publicly-funded research into collaboration with business: • Australia ranks 29th and 30th out of 30 OECD countries on the proportion of large businesses and small-to-medium sized enterprises (SMEs) collaborating with higher education and public research institutions on innovation, respectively (Australian Government 2015b, p. 3). • Australia ranks 23rd out of 32 OECD countries on the percentage of total research publications that are co-authored by industry and the research sector (Australian Government 2015b, p. 4). The most recent report into Australia’s research sector concluded: The overall quality of the Australian research sector is high by OECD standards but Australia’s performance is poor when it comes to translating publicly-funded research into collaboration with business. We rank last out of 26 OECD countries on the proportion of businesses collaborating with higher education and public research institutions on innovation. (Australian Government 2015c, p. 1) Limited collaboration and commercialisation by publicly-funded research organisations may reflect a strong focus on basic research, which is more consistent with their role in the innovation system. Further, reflecting the focus of Australian universities on 478 INTELLECTUAL PROPERTY ARRANGEMENTS non-commercial basic research, particularly in the health field, some have pointed to evidence that makes the lack of collaboration unsurprising: The disconnection between the research agendas of industry and the university sector is acute. In 2010, businesses spent 52 per cent of their R&D outlay on engineering and 28 per cent on ICT. Correspondingly, universities spent 9 per cent on engineering and only 4 per cent on ICT. On the other hand, while universities spent 38 per cent of their research expenditure on medical and health sciences and biological sciences, the comparable figure for business is 6 per cent. (Australian Technology Network of Universities and Ai Group 2015, p. 8) Nevertheless, there are signs of a genuine problem. If it were merely that Australian universities do not engage in patentable activities, the issues associated with poor collaboration in respect of IP would be moot. However, Australia ranks 12th out of 38 OECD countries on the number of patents filed by public research institutions (OECD 2012). So the issue appears to be one of ‘patenting without impact’. Certainly, participants in this inquiry and other recent reviews have identified IP arrangements as either actual or potential barriers to greater private sector collaboration with universities and publicly-funded research organisations (box 16.1). Box 16.1 Concerns about IP and collaboration Concerns that IP is an impediment to collaboration between business and public institutions are not limited to Australia, nor are they a recent phenomenon (they were, for example, noted in the 2007 Commission Report Public Support for Science and Innovation). Stakeholders also raised a number of concerns about universities and publicly-funded research organisations as part of the Review of Research Policy and Funding Arrangements (Watt, 2015), including that: • IP negotiations can be long and costly • some universities overvalue their IP, particularly early stage IP • considerable IP is held by universities that could be released and commercialised • there is a lack of clarity on IP ownership for students on industry placements and researchers on secondment • some universities are inept at managing commercialisation of IP and few university staff have commercial experience. However, as that review noted, concerns were often based on anecdotal evidence, some of which was dated. Other stakeholders (particularly universities) argue that there is not a problem with collaboration. They note: • parties tend to rely on traditional means of disseminating results of academic research such as through publication, conferences and industry events • new digital marketplaces (such as Source IP and WIPO Green) are increasingly being used by universities to bring patent information to the attention of business • commercial opportunities are not as pronounced in Australia, causing Australian universities to explore international, rather than domestic opportunities. Sources: Australian Government (2015c); University of Melbourne (sub. 100); University of Sydney (sub. 104). IP AND PUBLIC INSTITUTIONS 479 Is more needed to improve commercialisation of publicly-funded research? The Australian Government has recently made changes to address some of the concerns about the level of collaboration involving publicly-funded research, including the introduction of: • Easy Access IP (as described above) • Source IP, launched by IP Australia in November 2015, which aims to provide information to businesses about public sector IP. An online database allows businesses to search for patents filed by Australian public-research organisations • the IP Toolkit, which was developed to simplify and improve discussions around IP in research collaborations. It helps to establish the terms for managing and using IP in collaborative ventures between business and researchers. Most recently, the Review of Research Policy and Funding Arrangements (Watt, 2015) recommended that reporting arrangements (for universities) take into account the relative share of research exploited through IP arrangements. WIPO has emphasised the importance of university IP policy to foster commercialisation of IP: … the evidence stresses the importance of a well-defined university IP policy. Universities with internal rules regulating the participation of researchers in the transfer of technology perform better than universities without such rules. (WIPO 2011, p. 163) There have been some promising developments in university IP policies over the course of this inquiry. The Australian Technology Network of Universities (ATN), which is made up of Queensland University of Technology, University of Technology Sydney, RMIT, University of South Australia and Curtin University, introduced ‘ATN National IP Principles’: Having recognised this problem, the ATN has been leading the charge to improve university/industry collaboration and we’re now making it easier for industry to work with our researchers and staff by reducing the barriers to commercial research. I’m delighted to share the ATN National IP Principles with you today. (trans., p. 469) The ATN National principles are set out in box 16.2. Such policies can assist to address broad historic and cultural issues that could otherwise inhibit commercialisation. As noted above, some parties believe these are significant issues in Australia. Their adoption by many Australian universities demonstrates the potential for organic change within the sector, without the need for Government intervention. 480 INTELLECTUAL PROPERTY ARRANGEMENTS Box 16.2 The ATN National IP Principles The ATN’s approach to managing intellectual property (IP) is based on the following principles: 1. We actively encourage students and staff to undertake research that is relevant to challenges faced by society and in partnership with industry, government and community groups. 2. As guided by our industry partners, we encourage them to own and take the lead in commercialisation of IP generated from industry funded research when they are best placed to do so. 3. Where access to university owned or jointly owned IP is necessary or beneficial for commercialisation we support access to the IP based on fair and equitable terms, in a timely manner. 4. Our interactions with industry will be governed by a transparent, flexible and user-friendly system that supports and encourages engagement using a range of IP models. 5. Each university will make public our IP Policies and Standard Commercial Agreement templates, to provide a simple and transparent framework. 6. We actively encourage and promote an entrepreneurial culture for our staff and students. This includes a system of support to facilitate the creation of new ventures where our staff and students are appropriately involved. 7. All partnerships and resultant commercial agreements will be developed and negotiated in a prompt manner and in keeping with these core principles. Is there a case for a ‘use it or lose it’ provision? Some countries include ‘use it or lose it’ provisions as part of research funding agreements. For example, in the United States a funder (usually a government body) can grant a licence over a patented invention if the funding recipient has not taken ‘effective steps to achieve practical application of the subject invention’. However, this provision has never been used since its introduction in 1980. The Review of Research Policy and Funding Arrangements went on to consider the potential for such a provision in Australia, raising a proposal that: … would require universities to make the IP arising from publicly-funded projects openly accessible to potential end users (e.g., through Easy Access IP arrangements) within a specific timeframe of the project’s completion, unless the university has taken steps to commercialise the IP. (Australian Government 2015c, p. 60) The Review concluded that ‘universities are moving in that direction anyway and that such a policy would be difficult to implement’ (Australian Government 2015c, p. 60). However, the Government asked the Commission to consider the feasibility of such arrangements in this inquiry. IP AND PUBLIC INSTITUTIONS 481 The Commission agrees that the case for a ‘use it or lose it’ provision is not strong. Australia’s current National Principles of Intellectual Property Management for Publicly Funded Research (National Principles) already require research institutions to ‘make every reasonable effort to gain benefit for Australia from IP’. For example, this may involve protection or making the IP publicly available in a timely manner. (Of course, imprecation aside, this may be hard to enforce.) A ‘use it or lose it’ scheme may also entail costs not justified by the magnitude of the problem. For example, some have argued the US scheme has costly reporting and compliance arrangements (GAO 1999, p. 15). Submissions to the Review of Research Policy and Funding Arrangements (Watt 2015) considered the issue of costs a problem with a use it or lose it approach: Universities would need more resources and budget for the patent filing process. (UTS 2015, p. 14) The Bayh-Dole Act is often cited as a reason for university intransigence over IP ownership and use and a significant obstacle to innovation in the university/industry collaboration space. (UNSW 2015, p. 5). The deployment of such measures requires a government to have very effective information systems allowing tracking of significant IP commercialisation opportunities. (Deakin University 2015, p. 12) Participants in this inquiry did not provide any further material arguments in favour of the introduction of a use it or lose it approach in Australia. On the contrary, the ATN supported the view that it would involve additional compliance costs with little benefit: I note that in the United States, where they do have the use it or lose it provisions, they haven’t been used since 1980. So I think that the use it or lose it provisions are probably adding an additional layer of complexity to address a problem that’s overstated somewhat. (trans., p. 472) Recent commentary in the United States reveals frustration with use it or lose it arrangements (referred to as march in rights). It is argued that they have a ‘chilling effect’ on collaboration between the private sector and publicly-funded institutions, with no observed benefit from actual release of technology (Bloch 2016, p. 260). This frustration has led to some suggesting compulsory licensing be adopted in the United States as an alternative (Thomas 2006, p. 349). Interestingly, others have rejected suggestions for compulsory licensing as too extreme, and instead proffer other alternatives to march in (Bloch 2016, p. 259). Comparisons of compulsory licensing and the US ‘march in’ rights in the Bayh-Dole Act 1980 (US) (as outlined in table 16.2) reveal the main difference is the mechanism of obtaining access. Compulsory licensing under the Patent Act 1990 (Cth) requires an application to a Court, whereas the funder retains a right to ‘march in’ under the BayhDole Act 1980 (US). 482 INTELLECTUAL PROPERTY ARRANGEMENTS Table 16.2 A comparison of compulsory licensing and ‘march in’ Compulsory licensinga Who can seek access? What are the criteria for access? March inb Anyone with standing in an Australian Court Anyone can petition a funding agency to march in; only a funding agency can exercise the right The applicant has tried for a Patentee has not taken, or is not reasonable period, but without expected to take within a success, to obtain from the reasonable time, effective steps to patentee an authorisation to work achieve practical application of the the invention on reasonable subject invention terms and conditions AND OR Action is necessary to alleviate the health or safety needs which are The reasonable requirements of not reasonably satisfied the public with respect to the patented invention have not been satisfied OR OR The patentee has contravened part IV of the Competition and Consumer Act 2010 (which deals with anticompetitive conduct) Which body decides? Federal Court What appeal rights exist? Full Federal Court, High Court How often have the rights been exercised? 3 applications, none successful (over a 26 year period) Action is necessary to meet requirements for public use specified by Federal regulations Funding agency — on own initiative or after ‘petition’ from entity seeking access United States Court of Federal Claims 5 march in petitions; march in rights never exercised (over a 36 year period) a See Chapter 12 of the Patents Act 1990 (Cth). b See US Code, Title 35, Part II, Chapter 18, s.203 – March in rights. In order for ‘march in’ to occur, an applicant must convince the funding agency to take action. 5 In practice, this has proven to be an insurmountable hurdle — experience in the United States reveals that of the five petitions received by the National Institute of Health over the past 36 years none have resulted in that agency exercising march in rights. Further, the Bayh-Dole Act 1980 (US) includes provision for the patent holder to appeal to a court if march in rights are exercised. Given that appeal rights are likely to be exercised if a patent is valuable, compulsory licensing and march in are similar, in that both result in a court-based determination of access arrangements. 5 Similar access barriers existed in Australia from 1952 to 1990 — with the Patents Act 1952 requiring an applicant to convince the Commissioner of Patents to seek a compulsory licence. These provisions were changed to allow direct access to a Court. IP AND PUBLIC INSTITUTIONS 483 Another difference between march in and compulsory licensing is that march in becomes unavailable if the patent holder takes, or is expected to take, effective steps to apply the invention. Compulsory licensing continues to be available where the reasonable requirements of the public have not been met, even if the patent holder has taken steps to apply the invention. Any policy change that reduces the control that recipients of public funding have over their inventions would need to grapple with the issue of impacts on research neutrality. At an extreme, if researchers in certain fields found that they were frequently losing the right to exploit their inventions, this could lead managers of public institutions to redirect research efforts towards areas that are less exposed to march in or compulsory licensing. The status quo, in which loss of control is very rare, avoids impacts distorting decision making about research priorities. The main benefit of both march in rights and compulsory licensing is the incentive created for holders of IP, and those seeking access, to voluntarily negotiate an arrangement that is mutually beneficial. It is not clear that march in rights would be a more effective mechanism to create this incentive than the existing compulsory licensing provisions. Existing mechanisms should be given an opportunity to work The poor ranking of Australia in OECD indicators on collaboration between research institutions and industry has raised awareness of the issue and prompted a range of policy responses, both governmental and self-regulatory initiatives by public funded institutions. Existing measures such as Easy Access and Source IP, and self-regulatory measures like the ATN National IP Principles are still in their infancy and should be given time to work before being assessed on their efficacy and whether any further policy is needed. In the Commission’s view, the adoption of an additional ‘use it or lose it’ provision for patents owned by publicly-funded organisations (including in the form that applies in the United States) is not warranted at the present time and risks research neutrality. FINDING 16.1 The adoption of an additional ‘use it or lose it’ provision for patents owned by publicly-funded organisations is not warranted. 484 INTELLECTUAL PROPERTY ARRANGEMENTS 17 Intellectual property’s institutional arrangements Key points • An integrated and coherent approach to developing intellectual property (IP) policy is needed if Australia’s IP system is to strike the right balance between the interests of IP originators, follow on innovators and consumers. • A range of public institutions have a role in Australia’s IP system. The three main institutions are IP Australia, the Department of Industry, Innovation and Science (DIIS), and the Department of Communications and the Arts. • IP Australia is responsible for administering Australia’s patents, trade mark, designs, and plant breeder’s rights systems and shares responsibility for policy development and advice for these rights with the DIIS. • The Department of Communications and the Arts is responsible for policy development and advice on copyright and circuit layout rights. • Inquiry participants raised a number of concerns with institutional and governance settings, including: – IP policy-making responsibility is fragmented and in some cases inappropriately resourced – IP policy development often lacks transparency, meaningful consultation and supporting evidence, especially IP negotiations in international trade agreements – potential conflicts between IP Australia’s dual roles of administrator and policy adviser – a lack of independent and integrated policy advice. • Better governance arrangements are needed for IP policy development and implementation. The Australian Government should promote a coherent and integrated approach to IP policy by: – establishing and maintaining greater IP policy expertise in the DIIS – ensuring the allocation of functions to IP Australia has regard to potential conflicts arising from IP Australia’s role as IP rights administrator and its involvement in policy development and advice – establishing a standing IP Policy Group and formal working arrangements to ensure agencies work together within the policy framework outlined in this report. The Group would comprise those departments with responsibility for industrial and creative IP rights, the Treasury, and others as needed, including IP Australia. • IP policy development would benefit from expert independent input and external scrutiny. However, expert advice does not necessarily have to come from a standing body. • The Australian Government, in consultation with industry and the broader community, should develop best practice guidance for developing IP provisions in international treaties. INTELLECTUAL PROPERTY’S INSTITUTIONAL ARRANGEMENTS 485 Good governance arrangements are critical. The OECD (2014, p. 3) notes that ‘Good regulatory outcomes depend on more than well-designed rules and regulations’. It is also important that institutional and governance settings support high-quality decision making and provide ‘confidence that regulatory decisions are made on an objective, impartial and consistent basis, without conflict of interest, bias or improper influence’ (OECD Regulatory Policy Committee 2012, p. 4). This chapter examines the institutional and governance settings for developing and administering intellectual property (IP) policies in Australia. It first describes existing roles and functions assigned to different domestic institutions (section 17.1). It then considers participant concerns about current institutional and governance arrangements (section 17.2) and whether changes to the current arrangements are warranted (sections 17.3 and 17.4). 17.1 The institutional landscape The domestic context IP Australia and the Department of Communications and the Arts play key roles The three main public institutions responsible for Australia’s IP system are IP Australia, the Department of Industry, Innovation and Science (DIIS) and the Department of Communications and the Arts. IP Australia is responsible for administering Australia’s patents, trade mark, designs, and plant breeder’s rights systems. IP Australia operates as a listed entity 1 within the Industry, Innovation and Science portfolio and is primarily funded by the fees it charges for administering registered rights (IP Australia, sub. 23). In addition to granting exclusive rights under the statutes it administers, IP Australia: • works with the DIIS to advise the Australian Government on IP policy (IP Australia is the main source of IP research within the Australian Government) • provides IP information and education services to business and the broader community • regulates the IP attorney profession • contributes to bilateral and multilateral negotiations and development cooperation programs to promote a more harmonised global IP system (IP Australia, sub. 23) (figure 17.1). 1 IP Australia is a non-corporate entity, which operates independently of DIIS on financial matters, and with delegated authority on other matters. The Deputy Director General (IP Rights Division) of IP Australia holds the offices of Commissioner of Patents, Registrar of Trade Marks, Registrar of Designs and Registrar of Plant Breeder’s Rights (IP Australia, sub. 23; IP Australia nd). 486 INTELLECTUAL PROPERTY ARRANGEMENTS The Department of Communications and the Arts is responsible for administering Australia’s copyright and circuit layout rights systems, and advising the Australian Government on related issues. (The Department of Communications and the Arts took over these responsibilities from the Attorney-General’s Department in September 2015 — this change was the first time since Federation that responsibility for copyright was not in the Attorney-General’s portfolio) (Australian Copyright Council, sub. 36). As copyright and circuit layout rights are free and arise automatically with no formalities, the Department of Communications and the Arts does not have administrative functions relating to granting IP rights and its role is largely one of policy development and advice. 2 Many other institutions have interests and responsibilities that relate to IP policy and administration … Other public institutions with an interest and responsibilities that relate to Australia’s IP policy or administration include: • the Department of Foreign Affairs and Trade (DFAT), which has an interest in IP policy settings that affect international trade and investment • the Department of Agriculture and Water Resources, which administers legislation relevant to geographical indications 3 • the Department of Health, which has a policy interest in data protection for pharmaceuticals and the registration of pharmaceuticals for sale in Australia • the Department of Immigration and Border Protection, which provides technical advice on enforcement during policy development • the Treasury, whose policy interests include small business matters • the Australian Competition and Consumer Commission (ACCC), which assesses rules for certification trade marks to ensure that they do not constrain competition, and may request to be a party to a matter before the Copyright Tribunal of Australia. In some cases, private entities also have a role in administering Australia’s IP rights. For example, copyright collecting societies collect royalties for uses of copyright material on behalf of authors and copyright owners (chapter 5). 2 The copyright section in the Department of Communications and the Arts has some administrative functions, including maintaining the Statement of IP Principles for Australian Government Agencies and the Australian Government Intellectual Property Manual, and negotiating and administering lead agreements with the Copyright Agency Limited and Screenrights Australia under the government statutory licence on behalf of the Commonwealth (Department of Communications and the Arts, pers. comm., 15 July 2016). 3 Wine Australia is a statutory body within the Agriculture portfolio, which administers grape-based wine and spirit geographical indications (chapter 12). INTELLECTUAL PROPERTY’S INSTITUTIONAL ARRANGEMENTS 487 Figure 17.1 IP rights policy and administration: institutional arrangements Minister for Agriculture and Water Resources Minister for Industry, Innovation and Science Minister for Communications Minister for the Arts Dept. of Industry, Innovation and Science Dept. of Communications and the Arts Dept. of Agriculture and Water Resources IP Australia Copyright Copyright Act 1968 Circuit layouts Circuit Layouts Act 1989 Designs Patents Patents Act 1990 Designs Act 2003 Trade marks Trade Marks Act 1995 Policy advice and input Registered IP right IP rights allocation and administration Unregistered IP right Wine Australia Plant breeder’s rights Geographical Indications Plant Breeder’s Rights Act 1994 Australian Grape and Wine Authority Act 2013 … external advice and review … Some public institutions have an external advisory or review role. Parliamentary committees and institutions such as the Australian Law Reform Commission and the Productivity Commission undertake ad hoc reviews of IP matters, at the request of Government. Until its abolition in April 2015, the Advisory Council on Intellectual Property (ACIP) was an independent, expert body appointed by the Australian Government, which advised the Federal Minister for Industry and Science — and relevant Parliamentary Secretary — on IP matters and the strategic administration of IP Australia. ACIP undertook a number of IP reviews including into the Designs and Innovation Patent systems. The Plant Breeder’s Rights (PBRs) Advisory Committee, which advises the Minister for Industry on issues that may arise under the Plant Breeder’s Rights Act 1994 (Cth) and on the desirability of making regulations that enhance the PBR scheme, is scheduled to be abolished (Cormann 2014). Australia’s courts hear appeals and undertake judicial reviews of decisions to grant IP rights made by the statutory office holders in IP Australia. The Federal Court of Australia 488 INTELLECTUAL PROPERTY ARRANGEMENTS has jurisdiction over decisions made by the Commissioner of Patents and the Registrars of Trade Marks, Designs, and Plant Breeder’s Rights. The Administrative Appeals Tribunal may review some decisions of each of the officers. The Federal Circuit Court may also hear appeals of decisions of the Registrars of Trade Marks and Designs (DIS 2015). The recently established IP stakeholder forum is a principal forum for consultation, discussion and information exchange on IP matters that relate to IP stakeholders. The IP stakeholder forum includes representatives from a cross section of professional and business associations and an independent attorney. Its functions include validating IP Australia’s self-assessments under the Australian Government’s regulator performance framework. The Forum Chair is the IP Australia Director General, who is responsible for setting the strategic direction of the forum. In June 2016, the forum membership was expanded to include a representative from the DIIS (IP Australia, pers. comm., 12 September 2016). … or enforcement of IP rights The courts also deal with IP disputes. The Federal Court of Australia is the primary court for disputes about IP infringement. The Federal Circuit Court and state and territory Supreme Courts are also able to hear cases. The Australian Copyright Tribunal, a division of the Federal Court, hears disputes over terms and conditions (including royalty rates) under voluntary and statutory copyright licences (setting copyright fees) (chapter 19). The Australian Border Force (Department of Immigration and Border Protection) has responsibility for seizing commercial quantities of allegedly infringing or counterfeit copyright or trade mark protected imports under Australia’s Notice of Objection scheme. The international context International treaties have significantly influenced Australia’s domestic IP settings and have been a major driver of stronger IP protections (appendix B). DFAT leads whole-of-government engagement in Australia’s international trade and investment negotiations and institutions, with the support of agencies responsible for IP policy where relevant. DFAT noted that: Australia’s engagement on international intellectual property issues is typically led by either IP Australia (which has policy responsibility for industrial property issues such as trade marks, patents and designs) or the Department of Communications and the Arts (which has policy responsibility for copyright). (sub. 65, p. 3) DFAT (2016c) has indicated that, consistent with other treaty negotiating processes, the Australian Government consults extensively with stakeholders (including peak industry bodies, individual companies, academics, unions and consumers groups) before deciding whether to enter into negotiations on trade agreements. It has also noted that the INTELLECTUAL PROPERTY’S INSTITUTIONAL ARRANGEMENTS 489 Government consults with stakeholders in the public and private sectors when developing negotiating positions (box 17.1) and bases the decision on whether a treaty is in the national interest on information obtained during consultations (DFAT 2016d). Current requirements for assessing the impacts of prospective trade agreements include the preparation of Regulation Impact Statements (RISs), which are assessed by the Office of Best Practice Regulation (OBPR), a National Interest Analysis (NIA) (prepared by DFAT) and a review by the Joint Standing Committee on Treaties (JSCOT) to consider whether the agreement is in the national interest prior to ratification (PC 2015b). Box 17.1 Trade-related IP negotiations and the public consultation process The Department of Foreign Affairs and Trade (DFAT) provided the following comments regarding trade-related intellectual property (IP) negotiations: In cases where DFAT has responsibility for leading trade-related intellectual property negotiations, it engages in extensive stakeholder consultations and negotiates on the basis of parameters approved by the Government. Other government departments and agencies, including IP Australia and the Department of Communications and the Arts, play an integral role. Negotiating positions are informed by domestic policy and best practice approaches, with advice from specialists within Australian agencies that have responsibility for intellectual property policy settings. Consultations with public and private stakeholders are essential to determining Australia’s defensive and offensive interests and positions in a particular negotiation. Consultations, along with information posted on the Department’s website, also update stakeholders on the progress of negotiations. For example, the Department held over 1000 stakeholder briefings and consultations between May 2011 and mid-2015 in connection with negotiation of the Trans-Pacific Partnership Agreement. Treaty text is put before the public and Parliament for scrutiny before any action is taken that would legally bind Australia. The Joint Standing Committee on Treaties reviews the treaty text and consults widely before making a recommendation as to whether Australia should enter into the treaty. A publically available National Interest Analysis is also prepared, which identifies overall the obligations, costs and benefits of the treaty for Australia and any regulatory change required. Source: DFAT (sub. 65, p. 11). 17.2 Participant concerns about current arrangements Participants to this inquiry, and others, have identified four specific issues regarding aspects of current institutional and governance settings: • IP policy-making responsibility is fragmented and, in the case of the DIIS in particular, inappropriately resourced • IP policy development often lacks transparency, meaningful consultation and supporting evidence, especially IP negotiations in international trade agreements • potential conflicts between IP Australia’s dual roles of administrator and policy adviser • a lack of independent and integrated policy advice. 490 INTELLECTUAL PROPERTY ARRANGEMENTS IP policy-making responsibility is fragmented Businesses are increasingly deploying multiple IP protections over the same good as a source of competitive advantage. Technological advances coupled with growth in trade (and global value chains) in goods and services across jurisdictions have seen IP rights become more interrelated. The interrelated nature of IP rights means an integrated and coherent approach to developing IP laws and policies is needed to strike the right balance between the interests of innovators and the wider public. Despite this, the 2015 Competition Policy Review noted the absence of an overarching policy framework for IP policy development in Australia: We remain concerned that there is no overarching IP policy framework or objective guiding changes to IP protection. (Harper et al. 2015, p. 104) Commentators overseas have argued the absence of an integrated and coherent approach to developing IP results in unintended consequences. Beckerman-Rodau has argued that in the United States the absence of an integrated approach to IP law and policy development has resulted in overlapping IP rights and overprotection: [The] expansion of covered subject matter under each specific area of intellectual property law has occurred with little regard to its effect on the other areas of intellectual property law. The unintended result has been the ability to protect certain subject matter simultaneously under patent, copyright, and/or trademark law. Such overlapping protection undermines the careful balance individually developed under each body of intellectual property law. (2010, p. 88) In Australia, participants have argued that the division of responsibilities for IP rights makes it difficult to take an integrated and coherent approach to policy development. Professor Andrew Christie argued that the problem of fragmented policy-making responsibility is particularly acute in respect of copyright. The separation of responsibility for copyright from responsibility for patents, trade marks, designs and plant breeder’s rights inevitably has the effect that policy for copyright is developed largely ignorant of the policy, principles and practices of the other IP regimes – meaning an holistic view of IP policy is missing. (sub. 29, p. 3) The Australian Federation of Intellectual Property Attorneys (FICPI Australia) similarly argued the separate development of copyright law policy from other IP policy has resulted in inconsistencies with other IP rights. For example, copyright in architectural plans for houses continues for the life of the author plus 70 years. In commercial terms, this is a monopoly in perpetuity. If houses were the subject of design protection, the monopoly right in the plans would extend for only 10 years (and then only on registration). (sub. DR581, p. 49) The Law Council of Australia (sub. DR490 p. 36) observed that ‘the transfer of copyright from the Attorney General’s Department to the Department of Communications and the Arts was made without any clear rationale being expressed’ and ‘it is not clear what was considered in making the change’. However, the result is that ‘areas of copyright relevant INTELLECTUAL PROPERTY’S INSTITUTIONAL ARRANGEMENTS 491 to innovation and technical matters, such as circuit layouts, databases and unregistered designs, remain outside the purview of IP Australia’. Differences in how different institutions view IP issues (including how rights should be afforded, administered, and enforced) can contribute to a lack of policy coherence. For example, PricewaterhouseCoopers (2014) noted that locating responsibility for copyright in the Attorney General’s Department (rather than in IP Australia) meant that copyright was regarded as a legal issue that was the domain of lawyers rather than an issue of innovation and commercialisation. The Australian Copyright Council (sub. 36, p. 5) noted that it is ‘yet to be seen what impact, if any, [copyright moving to the Department of Communications and the Arts] will have on the administration of copyright in Australia’. Some inquiry participants have highlighted the need to bring an economywide perspective to bear on IP policy, drawing parallels with experiences in competition policy (Moir, sub. 130; Open Source Industry Australia, sub. DR486). In 2008, the panel for the Review of the National Innovation System argued that: … it is imperative that IP policy make the transition that competition policy made over a decade ago now, from a specialist policy area dominated by lawyers, to an important front of micro-economic reform. (Cutler 2008, p. 85) IP policy development often lacks transparency, meaningful consultation and supporting evidence Transparent and evidence-based policy helps ensure the public can hold Ministers and government agencies to account for policy decisions and use of public resources. Transparency is also an important means of securing legitimacy and public confidence in actions taken by government. A number of participants in this inquiry expressed concern that IP policy development often lacks transparency, meaningful consultation and a robust evidence base. Concerns about evidence-based IP policy mostly relate to trade agreements Participants’ concerns about lack of transparency, meaningful consultation and evidence-based policy mostly related to Australia’s decisions to enter into international trade agreements that incorporate IP provisions. The Business Council of Australia (BCA) noted that some of its members expressed concerns over a lack of transparency around the process for developing Australia’s negotiating position, while others argued that consultations with government did not result in a change in the government’s position. One inquiry participant commented that it was not clear how or why DFAT identified particular stakeholders for consultation for Trans-Pacific Partnership (TPP) negotiations. Others questioned the value of the stakeholder consultation (box 17.2). 492 INTELLECTUAL PROPERTY ARRANGEMENTS Commentators have raised similar concerns about Australia’s treaty-making processes previously. A Senate Committee inquiry also identified significant shortcomings in these processes. For example, it noted that although DFAT consults widely, the effectiveness and usefulness of private briefings with stakeholders was called into question by many and ‘in consulting with stakeholders, quantity was a poor substitute for quality’(SFADTRC 2015, p. 73). The Productivity Commission’s Trade and Assistance Review also identified issues with the rigour of impact assessments. … current processes fail to adequately assess the impacts of prospective agreements. They do not systematically quantify the costs and benefits of agreement provisions, fail to consider the opportunity costs of pursuing preferential arrangements compared to unilateral reform, ignore the extent to which agreements actually liberalise existing markets and are silent on the need for post-agreement evaluations of actual impacts. (PC 2015b, p. 82) Charlesworth et al (2006) have observed the time frames for reviewing and consulting on treaties has, in some cases, left limited scope for public scrutiny and input. With respect to the Australia-United States Free Trade Agreement (AUSFTA) they noted: The timing of JSCOT’s scrutiny suggests that the government never saw the JSCOT process as anything other than a formality … JSCOT’s first official briefing on the AUSFTA was held on 2 April 2004 … JSCOT concluded its public hearings six weeks later, on 14 May 2014. Four days later, Australia committed itself in principle to the AUSFTA by signing the agreement in Washington DC. Australia’s signature of the AUSFTA did not bind Australia to the agreement, but it was a powerful signal that the government fully supported the agreement and intended to become party to it. JSCOT issued its conclusions and report on the AUSFTA on 23 June 2004. That same day, the legislation implementing the AUSFTA was introduced to the House of Representatives. The remarkably short time between the tabling of JSCOT’s report and the tabling of the AUSFTA legislation gave members of parliament insufficient time to digest and debate the many recommendations made by JSCOT in relation to the AUSFTA. (pp. 129, 133-134) As noted in appendix B (box B.3), Australia made substantial changes to its copyright regime as a result of AUSFTA. 4 4 The Parliament of Australia’s (2016) 20 year history of the Joint Standing Committee on Treaties summarises the inquiry timelines for subsequent trade agreements (such as with Japan, Korea and China). INTELLECTUAL PROPERTY’S INSTITUTIONAL ARRANGEMENTS 493 Box 17.2 Views on lack of transparency and consultation A number of participants to this inquiry offered their views on the consultation processes and evidence base used to inform decisions on intellectual property (IP) arrangements in international agreements. Concerns included lack of transparency around the process for developing Australia’s negotiating position and lack of meaningful consultation. Some [Business Council of Australia] members have expressed concern that there was a lack of transparency around the process for developing Australia’s position. Others argued that consultations with government did not result in a change in the government’s negotiation position. (BCA sub. 59, p. 7) In connection with the recent TPP Agreement, IPTA [The Institute of Patent and Trade Mark Attorneys of Australia] contacted the Department of Foreign Affairs and Trade (DFAT) and indicated an interest in being involved in working with DFAT to help analyse the consequences of any IP rights provisions in the agreement. However, IPTA was not consulted. IPTA understands that there were consultations in relation to provisions in TPP with other groups and individuals, but it is far from clear who, how and why these particular individuals and groups were identified for consultation. IPTA believes there should be more transparency associated with the negotiation of IP provisions in international agreements and, if such transparency is not permitted, that negotiators should at least consult with bodies representing the patent and trade mark attorney profession in Australia, of which IPTA is the peak body. (IPTA, sub. 73, p. 18) CHOICE became involved in [the TPP] process due to significant consumer concerns arising from leaked sections of the draft text of the agreement. The possibility of medicine price rises, criminal punishments for minor, private copyright infringement, and threats to the government’s ability to pass laws like the tobacco plain packaging legislation were all raised as concerns – and all have intellectual property at their core … CHOICE’s experience in the TPP negotiations demonstrates that current levels of transparency are inadequate. (CHOICE, sub. 26, p. 13) Throughout the negotiations, the TPP was subject to extensive criticism from health, development and consumer organisations, both internationally and within Australia. Much of this criticism focused on the proposed content of the TPP, particularly provisions proposed by the United States for the intellectual property and investment chapters. Criticism also focused on the lack of transparency in the negotiations and the imbalance in terms of input from large corporations and industry associations in contrast with the public. (Gleeson, sub. 128, p. 1) CHOICE attended several of these meetings [consultations on the TPP], and they were of extremely limited use. Australian stakeholders were given no access to the negotiating documents during this consultation period, and consequently our ability to engage constructively in consultation was severely constrained. Departmental staff were unable to provide CHOICE with any negotiating documents, position papers, issues papers, or the wording of any sections of the agreement. They were not able to provide descriptions of the content of the agreement, or directly answer questions on this. Despite this, CHOICE was asked to raise concerns about the specific wording of particular sections. General questions could not be answered. (CHOICE, sub. 26, p. 14) The principal factor constraining responsible and transparent evaluation of IP rights extensions has been the negotiation of bilateral and regional treaties with IP provisions in secret. This practice should be abandoned altogether. (Open Source Industry Australia, sub. 21, p. 11) The details of domestic copyright law are increasingly being influenced by bi- and plurilateral agreements such as the Australian US Free Trade Agreement (AUSFTA) and the Trans-Pacific Partnership (TPP). There is little transparency in the negotiation of these agreements, and where consultation does occur it is frequently biased towards private industry, without including the public sector (eg schools and libraries) or civil society in discussions. (Australian Digital Alliance, sub. 108, p. 30) Foxtel considers that there has been very little transparency with respect to Australia’s entry into such [AUSFTA and TPP] agreements. (Foxtel, sub. 115, p. 9) (continued next page) 494 INTELLECTUAL PROPERTY ARRANGEMENTS Box 17.2 (continued) Telstra remains concerned by the lack of transparency in the negotiation of IP Chapters in international and trade agreements. In particular, the failure of some negotiating parties to facilitate open public discussion about the specific issues being negotiated, or to release the draft text for public review and comment. Without knowing or understanding the issues, or more importantly understanding how those issues are dealt with in specific text, it’s very difficult for stakeholders to assess the potential commercial impacts of an IP Chapter, or to make meaningful submissions. (Telstra sub. DR316, p. 14) I was involved in endless DFAT consultations about the TPP and various other agreements and it’s a very Kafkaesque experience, very Kafkaesque. You’re invited into a room … and they say, “What are your thoughts?” and you say, “Well, what have you been negotiating?” and they say, “We can’t tell you”. You say, “Well, we came and saw you six months ago. These are our concerns. These are the things that we think matter and that hasn’t changed, so what are we talking about here?” It’s a very, very strange process. (Gruen, trans., p. 723) The evidence base for domestic IP policy has also been low on occasion In terms of domestic IP policy, Open Source Industry Australia (sub. 21, p. 11) pointed to the process undertaken by ACIP and IP Australia for the review of the innovation patent system as a good example of an ‘evidence-based, reasonably transparent reform process’. However, it also noted that ‘in general a stronger evidence base is highly desirable (noting the review of the innovation patent system as a prominent exception, where the economic modelling was of a high standard)’. Charles Lawson (sub. 7, p. 1) argued that previous reviews of the Patents Act 1990 (Cth) have failed to comply with the Competition Principles Agreement by demonstrating ‘that the benefits of the restriction on competition outweigh the costs and that the objectives of patent privileges can only be achieved by restricting competition’ (chapter 15). There are signs of improvement in evidence-based policy, but challenges remain Participants acknowledged that the quality of evidence used to support IP policymaking in Australia has generally improved in recent years and that IP Australia has strengthened its capacity to provide evidence-based policy advice. In 2014, the Australian Public Service Commission observed that IP Australia increasingly brought evidence and analysis to bear to inform external decision making and had developed greater internal policy capability, including through the establishment of its Office of the Chief Economist and efforts to improve data and reporting capability (APSC 2014). For example, analysis undertaken by IP Australia’s Office of the Chief Economist provided a key source of evidence on the effects of innovation patents, after an ACIP review (from 2011 to 2014) was unable to find sufficient empirical evidence to assess their effectiveness. Nonetheless data and research gaps remain (box 17.3). Improving the evidence base for copyright policy has proven particularly challenging. Assembling data on copyright is difficult because copyright arises automatically and so is INTELLECTUAL PROPERTY’S INSTITUTIONAL ARRANGEMENTS 495 not well documented, nor traceable to its owners (Hargreaves 2011). Richard Hooper (previously of the Copyright Hub Foundation London) noted that there could also be a lack of incentive to improve the evidence base. There is evidence around the world that some people make money from poor data and therefore are not in a rush to improve it. Also poor data probably keeps more people in jobs and that may also reduce the incentive to create and keep high quality data. (sub. 6, p. 4) Further, the resources available to undertake ongoing copyright policy work have arguably been more susceptible to shifting government priorities than those for other IP rights. Unlike IP Australia, which is a cost recovery agency dedicated to IP, the Attorney General’s office (which had carriage for copyright policy until recently) is a government department that needs to adjust resources allocated to different policy areas in response to changing priorities. Christie was particularly critical of the administration of copyright policy in recent years, noting ‘ … copyright is a basket case in terms of its policy administration. I don’t think I should mince my words about that’ (trans., p. 445). Box 17.3 Data and research gaps Previous chapters have identified areas of intellectual property (IP) policy where additional data and research would help to address policy questions. The following are examples of areas for attention: • Copyright: The drivers and impacts of infringements, and innovative policy ideas such as mandatory copyright registration beyond life plus 50 years, to support a multilateral reform agenda (chapters 4, 5 and 18). • Patents: Data to better calibrate patent fees and facilitate collaborative research with other countries to identify what the ‘optimal’ level of inventive step is from a global perspective (chapter 7). Collecting and publishing information on the grants of software patents in Australia (chapter 9). Reforms to legislative requirements for data collection regarding research and development expenditure on drugs that receive pharmaceutical patent extensions of term (chapter 10). • Trade marks and Geographical Indications: Research on the scale of anticompetitive behaviour that the trade mark system can induce, and the resulting costs to society. Further research on how trade marking behaviour changes in response to changes to trade mark fees in Australia. Quantifying the effects of protecting geographical indications on Australian industries (chapter 12). • Other IP rights (Designs, Circuit Layout Rights): Research on the economic effects of design rights (chapter 11). Collecting data from designers and manufacturers of circuit layouts to understand how frequently users rely on their protection (chapter 14). 496 INTELLECTUAL PROPERTY ARRANGEMENTS Potential conflicts between IP Australia’s dual roles of administrator and policy adviser IP Australia’s dual roles of IP rights administrator and IP policy adviser has on occasion raised concerns about the potential conflicts in these functions (APSC 2014). For example, the Department of Health argued that: The independence of policy development from the administration and enforcement of regulation is an important principle in ensuring that regulators can maintain the confidence and trust of the community. As the granting of IP rights is a regulatory activity, this principle is important to ensuring that Australia’s IP arrangements strike an appropriate balance for IP holders, IP users and the public interest. (sub. 84, p. 3) The BCA (sub. 59) argued that, to ensure regulation is administered in the most efficient manner possible, there should be a clear separation of roles between policymakers and regulators, and the government’s expectations of a regulator should be transparent and clearly within the scope of the regulator’s powers. In practice, rights administrators and regulators will always have at least some input into policy development, even where there is functional separation of policy and regulatory functions (OECD 2014). The real issue is the extent to which they should be involved, the channels through which policy advice should be provided, and the transparency and documentation of those arrangements. Should the administrator or regulator have primary responsibility for developing policy and of the regulatory instruments meant to achieve the government’s objectives? Or should it contribute to the public policy process through its parent department (or some other agency), which is responsible for providing policy advice to the Minister? (CAV 2008). IP Australia has assumed a lead role in some areas of IP policy The role of IP Australia in IP policy has evolved over time. To illustrate, the Industry portfolio budget in 2006-07 noted ‘IP Australia’s role in supporting quality research and providing policy advice is growing. This specialist advice enables Australia to keep on the forefront of IP issues and be influential in international activities’ (DITR 2007, p. 91). The 2014 capability review for IP Australia noted that ‘greater focus on policy engagement and effectiveness would be welcomed by a number of other government agencies, including the Department of Industry’ (APSC 2014, p. 21). Around 25 to 30 people currently work on IP policy at IP Australia (IP Australia, pers. comm., 4 April 2016). The resources dedicated to policy development outside of IP Australia are quite limited: • DIIS has an IP policy unit that consists of four people. The DIIS IP policy unit is the coordination point on IP issues in the department, interacting with other policy and program areas in the DIIS that intersect with IP issues (DIIS, pers. comm., 12 April 2016). The Commission understands that the IP policy unit was recently formed. INTELLECTUAL PROPERTY’S INSTITUTIONAL ARRANGEMENTS 497 • The copyright section of the Department of Communications and the Arts comprises about 7 full time equivalent staff (Department of Communications and the Arts, pers. comm., 15 July 2016). IP Australia acknowledges that it has a ‘significant role in policy advice to government’ compared to some other patent and trade mark offices (Kelly 2015, p. 2). IP Australia’s policy advice functions include: providing advice on IP matters to the Minister; supporting research into the current and future use of IP rights; and developing legislation to implement IP changes and improvements as required (IP Australia 2015d). In some areas, IP Australia has assumed a lead role in progressing IP policy. For example, IP Australia coordinated the Australian Government’s response to ACIP’s review of Australia’s designs system and has undertaken further stakeholder consultation on ACIP’s recommendation to abolish the innovation patent system (IP Australia, sub. 23). IP Australia also consulted on a proposal to introduce an objects clause to the Patents Act (IP Australia 2013c). One commentator has noted ‘IP Australia has, in recent times, taken a leading role in driving IP policy, and was the primary force behind the Raising the Bar IP reforms that were passed in 2012’ (Summerfield 2016). As noted, IP Australia leads Australia’s international engagement on IP issues relating to industrial property (DFAT, sub. 65). The Commission understands that IP Australia and the DIIS have recently developed an internal working document outlining their respective responsibilities for IP. 5 However, this information is not in the public domain. IP Australia has advised that: IP Australia works with DIIS to advise the Government on IP matters, IP Australia generally takes the lead on technical policy issues, and DIIS takes the lead on broader policy issues that deal with other aspects of innovation policy or relate to the role of IP in Australia’s innovation system. DIIS has responsibilities across industry and innovation policy, including national policy issues relating to the digital economy, as well as extensive capabilities in economic analysis, which complement IP Australia’s more technical IP expertise. (sub. DR612, p. 21) The main argument for having an administrative or regulatory agency involved in policy development is that the expertise developed at one stage of the policy process can be used to inform other stages, thereby making regulation more effective and responsive (CAV 2008). IP Australia (sub. 23, p. 16) noted that the technical nature of the subject matter, and the complex international context of IP policy work, causes policy makers to work closely with subject matter experts, to ensure IP policy is ‘technologically-informed’. The OECD (2014, p. 38) has argued regulators’ experience means they ‘should have a specific and explicit advisory role on government policy [or] input in developing government policy’. Other regulatory and administrative agencies in Australia generally do not take the lead on policy advice or development. For example, agencies in the Treasury portfolio, such as the Australian Securities and Investments Commission (ASIC), Australian Prudential 5 This document is periodically updated. It was first created in 2014. 498 INTELLECTUAL PROPERTY ARRANGEMENTS Regulation Authority (APRA), ACCC and the Australian Taxation Office (ATO), have statements of expectations with the Minister which clarify that they will contribute to policy development by advising Treasury on the operational implications of Government policy initiatives (Australian Government 2016e). Under these arrangements, the agencies provide valuable technical and expert input to policy development by the Treasury. Blurring policy and regulator/rights administrator functions can have undesirable consequences Australian and international literature on the governance of regulatory and rights administrator bodies 6 — across areas such as tax, financial services, utilities, competition, health and safety, the environment and consumer protection — have highlighted the importance of regulators having a clear role without conflicts and the risks associated with regulators having input into policy development and review (ANAO 2014; CAV 2008; IOSCO 2010; NZPC 2014; OECD Regulatory Policy Committee 2012; PC 2013b; VCEC 2005, 2009). Common arguments against combining regulatory functions and policy advice include: • the increased risk of regulatory ‘creep’, whereby a regulator aligns policy preferences with its institutional interest to maintain or expand its role • the potential for a regulator to be drawn into the political process, possibly compromising its capacity to make impartial decisions • the greater likelihood of a narrower policy perspective being applied by a regulator compared to its portfolio department (this may manifest in inadequate cost-benefit assessments of alternatives through lack of awareness of other government objectives and actions) • the risk of reduced accountability as there is an incentive for a regulator to less rigorously specify objectives against which its subsequent regulatory performance can be assessed • regulated stakeholders may be unwilling to substantially engage in policy debates due to concerns that to do so may affect the regulator’s attitude towards them or even influence enforcement decisions • the potential distortion of risk assessment in policy responses — a regulator may be more risk averse and advocate regulation simply because it does not want to be criticised for missing a problem after deciding not to regulate a risk that later materialises (CAV 2008; PC 2013b; Pratt and Berg 2014; VCEC 2009) • the increased risk of a regulator being captured by the regulated who will perceive the regulator as able to heavily influence policy development and therefore devote commensurate resources to exerting influence. A similar point was made by Open 6 While participants had different views on the extent IP Australia fits the traditional archetype of a regulator, as opposed to an administrative body, the literature nonetheless provides useful insights. INTELLECTUAL PROPERTY’S INSTITUTIONAL ARRANGEMENTS 499 Source Industry Australia, which argued that ‘fragmented IP regulation increases the risk of regulatory capture, where an agency largely benefits a small number of specialists and very large multinational companies to the detriment of the Australian public and smaller Australian businesses’ (sub. DR486, p. 17). Views on the extent of these risks vary Stakeholders’ views about whether these risks are real, or largely theoretical, vary. As part of a broad based Capability Review of IP Australia in 2014, the Australian Public Service Commission found that IP Australia’s stakeholders — relevant ministers, private sector companies, state organisations, peak bodies, interest groups, clients and central agencies — viewed the agency as impartial and as operating without agenda or bias (APSC 2014). However, the Capability Review did not examine governance arrangements in detail. Other commentators have argued that the large businesses that dominate patent applications have a strong influence on the decisions of patent offices around the world (Drahos 2010; SCARC 2010). Christie (trans., p. 445) noted that IP Australia are ‘not necessarily wanting to accrete all power to themselves … but you do need to understand this stuff and basically that’s where the only people are in Canberra who do understand it live’. He argued that this heavy reliance on IP Australia for policy advice has meant IP policy has lacked a broader perspective: At the moment with the greatest respect I think [expertise about IP] only resides in one place, and that is IP Australia, and there are challenges there because of the divide, or lack of divide between administration and policy making. IP Australia is very focused on how you do examination and it’s not that focused on - and yet it needs to be because the policy expertise on this issue doesn’t reside elsewhere … whether the things they’re being asked to examine are the right things. (trans., p. 445) IP Australia and the DIIS have taken steps to address concerns about potential conflicts between its policy and regulatory (or rights administrator) functions by establishing a separate Deputy Director General responsible for policy within IP Australia in 2015 (box 17.4). While IP Australia’s status as a (non-corporate) listed entity with statutory office holders means it operates with some independence from government, it is essentially part of the DIIS. In contrast, other non-corporate entities, such as the ACCC, APRA, ATO and ASIC are independent statutory agencies with enabling legislation, outside of direct Ministerial control (unless there is a specific power of direction in favour of the relevant Minister in respect of certain matters) (APSC 2016). 500 INTELLECTUAL PROPERTY ARRANGEMENTS The Government reviewed IP Australia’s governance structure in 2005 7 as part of a broader review of statutory authorities and considered no changes to its governance arrangements were required. Reasons for this decision include the existing governance arrangements gave the Minister clear and firm control over the agency and its executive. The Government confirmed its preferred classification of Commonwealth entities as part of preparation for the Public Governance, Performance and Accountability Act (PGPA) Act 2013 (IP Australia, pers. comm., 12 September 2016). Box 17.4 IP Australia’s organisational structure The Industry, Innovation and Science portfolio comprises several agencies, including IP Australia, and the Department of Industry, Innovation and Science. The Minister for Industry, Innovation and Science is responsible for the Department and IP Australia. IP Australia has a Director General and two Deputy Directors General. • The Director General of IP Australia holds the statutory position of Designated Manager for the regulation of the IP attorney profession and sits on the Professional Standards Board for Patent and Trade Marks Attorneys. The Director General of IP Australia also sits on the DIIS Executive Board. • The Deputy Director General (IP Rights Division) holds the offices of Commissioner of Patents, Registrar of Trade Marks, Registrar of Designs and Registrar of Plant Breeder’s Rights. The Deputy Director General (IP Rights Division) exercises functions and powers under the legislation that IP Australia administers. • In 2015, the Government introduced the role of Deputy Director General (Policy and Corporate Division) with responsibility for IP policy and cooperation activities. This new role was to create a structural separation of IP Australia’s policy and service delivery functions. The Director General of IP Australia is the accountable authority under the Public Governance, Performance and Accountability Act 2013, and is directly responsible to the Minister for Industry, Innovation and Science for the proper use and management of the entity’s resources. The Director General has agency head powers under the Public Service Act 1999, delegated from the Secretary of the Department. IP Australia, which has existed since February 1998, and its predecessors, have traditionally been affiliated with the Industry Portfolio. The Department and IP Australia report annually on performance in accordance with the Public Service Act 1999 and guidelines approved by the Joint Committee of Public Accounts and Audit. Source: IP Australia (nd), Lawson (2008a). IP offices (IPO) in other countries are typically units within government departments (as opposed to independent government agencies). The United Kingdom IPO is an executive agency of the Department of Business, Innovation and Skills (box 17.5). The United States Patent and Trademark Office is an agency within the Department of Commerce (Department of Commerce 2016; US PTO 2016). 7 IP Australia is currently a ‘listed entity’ under the PGPA Act. Under the terminology of the previous legislation, the Financial Management and Accountability Act, IP Australia was a ‘prescribed agency’. INTELLECTUAL PROPERTY’S INSTITUTIONAL ARRANGEMENTS 501 Box 17.5 Governance of the UK Intellectual Property Office The Intellectual Property Office (IPO) is the official government body responsible for IP rights (including Patents, Designs, Trade Marks and Copyright) in the United Kingdom. The IPO operates on a cost-recovery basis and is responsible for implementing the national framework governing IP rights and for promoting the UK’s interests in the development of the international IP rights system. The UK IPO’s specific responsibilities include: • granting patents, registering trademarks and designs, and maintaining granted and registered rights • formulating advice on, and implementing UK IP Policy • providing UK input into the development of international legislation (including treaties) • providing advice on IP issues to businesses and raising awareness of IP issues. The IPO is an executive agency of the Department of Business, Innovation and Skills. Executive agencies are distinct both from non-ministerial government departments and non-departmental public bodies, each of which have legal separation from ministerial control. The Secretary of State for the Department determines the policy framework within which the IPO operates, agreeing strategic objectives, setting financial and performance targets and approving the corporate plan. The Chief Executive of the UK IPO, members of the UK IPO and the Steering Board meet with the responsible minister for IP at least once per year. The Chief Executive has the right to direct access to the Minister. The UK IPO is sponsored within the Department by the Director General responsible for IP. The sponsor is the key link between the IPO and the Department ensuring that both have a clear understanding of each other’s objectives and methods of working. These functions are mainly fulfilled through the departmental sponsor’s memberships of the Steering Board. The Chief Executive and Comptroller General of the IPO (who is also the registrar of trade marks and designs) is appointed by the Secretary of State generally for a fixed term. They are responsible for the day-to-day operations of the IPO and accountable to the Secretary of State/responsible minister and the permanent secretary of state as the Principal Accounting Officer. As comptroller general, they are responsible for the administration of statutes and advise the secretary of state on all aspects of the national IP system, related EU legislation and other relevant policy issues. Sources: UK Intellectual Property Office (2016b), UK Cabinet Office (2015). Lack of independent and integrated policy advice Some suggest there is insufficient independent expert input into IP policy development … Impartial and credible expert advice can be valuable in policy development where there is a need to draw on additional policy expertise (such as economic, legal or scientific expertise) and/or to manage any real or perceived risk of bias or undue influence on the policy development process. With regard to the latter, Hargreaves has noted: In the case of IP policy and specifically copyright policy … there is no doubt that the persuasive powers of celebrities and important UK creative companies have distorted policy 502 INTELLECTUAL PROPERTY ARRANGEMENTS outcomes. Further distortion arises from the fact (not unique to this sector) that there is a striking asymmetry of interest between rights holders, for whom IP issues are of paramount importance, and consumers for whom they have been of passing interest only until the emergence of the internet as a focus for competing technological, economic, business and cultural concerns. (2011, p. 93) The broad purpose of independent expert advisory bodies is to provide a well-informed and impartial view on policy or regulatory issues. Christie (sub. 29, p. 2) argued that Australia’s IP policy-making suffers ‘an absence of independent expert input’. He argued that the abolition of the Copyright Law Review Committee (CLRC) in 2005 has led to poor policy-making in copyright. It is not coincidental that the decade following the abolition of the CLRC is one in which poor policy-making (or, more accurately, the absence of good policy-making) in copyright has occurred. At least part of the reason Australia has a poor record of copyright reform in the past decade is due to the absence of independent expert advice. (sub. 29, p. 3) Christie also noted that, following the abolition of ACIP and the Plant Breeder’s Rights Advisory Committee, there will be no standing independent expert body providing input on IP policy and limited scope for IP Australia to consider policy issues at the ‘macro level’. While it is true that IP Australia regularly convenes “working groups” to provide input into patents, trade marks and designs, these working groups are almost exclusively comprised of practitioners (lawyers and attorneys) and are confined largely to consideration of micro-level issues of practice. The ability for the working groups to provide a broader perspective (including from economics, industry and civil society) at the macro-level on policy-oriented issues is very limited. (sub. 29, p. 3) The Commonwealth Department of Health reflected that it is unclear how the independence of policy advice will be maintained with future reviews of IP matters being coordinated by IP Australia (sub. 84, p. 4). … and oversight of IP Australia’s regulatory decisions and operation Governments delegate powers to regulators through legislation on the assumption that regulators will use them to achieve the longer-term goals that justified their establishment. Because even the most detailed legislation is unlikely to be a sufficiently complete ‘contract’ to ensure a regulator performs exactly as the government desires, there is a possibility of regulators departing from governmental objectives or priorities. This risk increases where legislation is vague, general, ambiguous, and/or internally inconsistent (CAV 2008). Some of IP Australia’s administrative functions — including searches and examinations that inform the decision whether to register a right — involve a degree of discretion in the application of the rules. The way that the rules are applied can have direct bearing on the INTELLECTUAL PROPERTY’S INSTITUTIONAL ARRANGEMENTS 503 ‘quality’ of the right (IPCRC 2000). For example, in the past stakeholders raised concerns about IP Australia’s application of the ‘manner of manufacture’ test, which ultimately resulted in a High Court ruling and updates to IP Australia’s Patent Examination Guidelines (as discussed below, this occurred in an emerging and contentious area of law). In the Pharmaceutical Patents Review, GMiA and Alphapharm suggested that quality issues in the patent examination process may have been contributing to the grant of low quality patents (Harris, Nicol and Gruen 2013). The reviews of gene patents and pharmaceutical patents recommended external oversight of IP Australia’s regulatory decisions in addition to the appeals processes provided by the courts, to ensure the quality of patents (Harris, Nicol and Gruen 2013; SCARC 2010). Some stakeholders have argued that external oversight arrangements are needed to mitigate the risk of interest groups having an undue influence on the patent system (SCARC 2010). Palombi (trans., p. 211) suggested oversight is necessary partly because IP Australia has a financial incentive to grant patents. Christie argued there is also a lack of external oversight of IP Australia’s broader operations. He noted that, while ACIP was notionally responsible for strategic oversight of IP Australia, it was never able to discharge those responsibilities for structural reasons. When I was on ACIP … I was always thinking … we are supposed to be oversighting IP Australia … and we weren’t. Now, why weren’t we? Well, because we were completely resourced by IP Australia. The Secretariat came from IP Australia. All the funding for ACIP came from IP Australia. All the directions from the Minister to ACIP first came from IP Australia to the Minister, down to ACIP. The response went from ACIP to the Minister. We didn’t even have [oversight of IP Australia] when we notionally had it, and we certainly don’t have it now. And it troubled me, and I think that we should have it. (trans., p. 447) 17.3 Reform options — domestic context The absence of an overarching policy objective, policy framework and reform champion, coupled with some strong vested interests, have collectively contributed to poor policy outcomes. Reform to Australia’s institutional settings need to focus on establishing: • an overarching objective and policy framework for the IP system • an alignment between responsibility for IP policy and the objectives of the IP system • clear and transparent separation of responsibility for administering the IP system from lead responsibility for IP policy development and advice • greater transparency and accountability in international negotiations. It is necessary to clarify what it is that Australia wants its IP system to achieve, who should be responsible for achieving that outcome and how should they go about achieving it. 504 INTELLECTUAL PROPERTY ARRANGEMENTS This section considers options for addressing the first three points. Section 17.4 considers options for addressing concerns about transparency, consultation, and evidence-based policy making in the international context. An overarching objective and policy framework for the IP system To promote a more coherent, economywide perspective, there would be value in specifying the overarching objectives of the IP system to inform the broader community and guide agencies and departments involved in IP policy development and administration. A common framework for formulating IP policy would also assist; the four principles employed by the Commission throughout this report provide a ready starting point (chapter 2). The DIIS (sub. DR615, p. 9) acknowledged that ‘clarity about whole-of-government objectives for [the] IP system would be of benefit to the department and other agencies with an interest in IP issues’. The Government’s response to this Report provides an opportunity to articulate the overarching objective and policy framework for the IP system. This would require the relevant agencies to have specific regard to the overarching objective and principles for the IP system when: • evaluating proposals to change the scope or duration of IP protection • determining the appropriate charging framework for IP rights administration processes (complementing whole-of-government charging policies under the PGPA Act). The Government’s response to this report could also clarify the roles of agencies involved in IP policy and outline specific actions to build the evidence base for IP policy (discussed below). The governance structures for non-corporate Commonwealth entities under the PGPA Act would allow the government to readily apply an IP policy framework. For example, a non-corporate Commonwealth entity listed under its enabling legislation or the PGPA Rule (such as IP Australia) is subject to the policies of the Australian Government under section 21 of the PGPA Act (Department of Finance 2015). Alignment of responsibility for IP policy with the objectives of the IP system Part of establishing a robust governance framework for IP is aligning responsibility for IP policy development with the objectives for the IP system. The Commission examined two options: • consolidating responsibility for IP policy advice into one agency with a single Ministerial ‘policy champion’ • introducing measures to increase coordination of policy development across agencies. INTELLECTUAL PROPERTY’S INSTITUTIONAL ARRANGEMENTS 505 Consolidating responsibility for IP policy advice into one agency Many participants to this inquiry, and others, have advocated locating primary responsibility for all areas of IP in one agency and/or designating a single Minister to have responsibility for all IP regimes (FICPI Australia, sub. DR581; Law Council of Australia, sub. DR490; National Association of Cinema Operators Australasia, trans., p. 74; Christie, sub. 29; PwC 2014; Terri Winter sub. DR198; University of Melbourne, sub. 100). For example, the Law Council of Australia submitted that the Government should place IP under one administrative head and noted that this model has been implemented successfully in the United Kingdom. Past reviews of the UK IP system, and consultation undertaken for this inquiry, suggest there are several potential benefits from consolidating responsibility for copyright and other major industrial IP rights into one agency, with one Minister. These benefits include: • a more holistic approach to issues that cut across the boundaries of all IP rights, such as policy, research, education, enforcement, business support, and awareness raising (for example, the United Kingdom has an enforcement strategy covering all IP) • staff can enhance their skills by working across copyright and other areas of IP • more cost-effective interactions with industry, the general public and other government agencies because there is a single contact point for IP issues (for example, businesses may have queries relating to both copyright and industrial IP rights) • a stronger presence at international fora by being the official national authority on IP. The Law Council of Australia also highlighted the operational benefits available from combining responsibility for copyright and industrial IP rights. Vesting separate or overlapping responsibilities in multiple government departments in relation to IP rights is inefficient and may result in multiple departments becoming involved and thus diluting responsibility or hampering efforts to address issues. By way of example, the [IP Committee] has had experience, when consulting with IP Australia in relation to trade marks issues, that copyright implications have arisen. Unless representatives from another department (recently changed) are included in consultation meetings, these implications may be overlooked. If representatives are included they may not be properly briefed on the trade mark issues which may result in double-handling of the issue and potential derailment of the consultation. (sub. DR490, p. 36) Other participants questioned whether creating or merging agencies responsible for IP policy would be worthwhile, noting that: • any benefits from changes to the machinery of government (the allocation of government functions between departments and ministers) would not be enduring as government could easily decide to reverse such arrangements in the future • having copyright in the Communications and Arts portfolio has advantages that would be lost if IP policy responsibility were consolidated in another agency (for example, 506 INTELLECTUAL PROPERTY ARRANGEMENTS there is arguably a healthy tension in copyright policy debates that arises from the competing interests of stakeholders in the communication and arts sectors) • departments that are industry or science-based (such as the DIIS) would likely prove ill-placed to deal with the cultural issues necessarily embedded in the copyright policy considerations (AIPP, sub. DR387) • copyright policy would receive a lower priority and/or copyright holders would have less influence in policy debates. There is very little evidence with which to test the proposition that consolidating responsibility for copyright and industrial IP rights leads to better outcomes. While the UK performs relatively well on various international IP indexes, such indexes have questionable normative value because they often reflect heavily the interests and perspectives of right holders as opposed to the community as a whole. Moreover, there is no clear link between international IP rankings and whether the same entity administers copyright and industrial property rights. For example, the United States and Germany also rank highly on several IP indexes and have a separate copyright office (Cornell University, INSEAD and WIPO 2015; EIU 2009; Levy-Carciente 2015; USCC 2015; WIPO 2016b). Among the participants who supported consolidating responsibility for IP policy advice into one agency, there was debate about which department is best placed to assume that role. Several participants suggested locating responsibility for all IP regimes in the Industry, Innovation and Science portfolio (which includes the DIIS and IP Australia) (Christie sub. 29; Law Council of Australia, sub. DR490; PwC 2014). (As discussed below, several participants suggested this option should coincide with a clearer separation of IP policy and administration, with the DIIS taking a more prominent role in providing IP policy advice and IP Australia focusing more on administration.) One argument for locating responsibility for IP regimes in the Industry, Innovation and Science portfolio is that IP policy fits within the DIIS’s broader responsibilities for innovation policy. For example, the DIIS has several programs that aim to facilitate the development and uptake of new ideas and technology and translate them into commercial activity. Retaining IP policy in the Industry, Innovation and Science portfolio could therefore help ensure a holistic view with regard to various policy ‘levers’ for stimulating innovation. IPTA noted: In view of the very close relationship between IP rights, innovation and industry, it makes most sense for responsibility for the IP rights systems, particularly those dealing with patents designs and trade marks to remain with the Department of Industry Innovation and Science. It is noted that the UK IPO is part of the roughly equivalent Department for Business, Innovation and Skills. (sub. DR562, p. 4) IP Australia (sub. DR612, p. 21) similarly noted that ‘the strong links between IP policy and innovation policy are enhanced by retaining responsibility for these policy areas within the same portfolio, and there may be risks in separating them’. INTELLECTUAL PROPERTY’S INSTITUTIONAL ARRANGEMENTS 507 Christie (trans., pp. 446–47) observed that locating responsibility for IP policy in the DIIS could also help minimise disruption and transitional costs if the government chose to more clearly separate IP policy and administration, noting: ‘it should be easier to transfer the expertise … and presumably transfer the revenue required to support the expertise if you’re all in the one portfolio’. A counterargument to locating responsibility for IP policy in the DIIS is that the DIIS’s focus on the competitiveness of Australian industries is not necessarily aligned with an economywide perspective on IP policy and overall competiveness (table 17.1). Table 17.1 Looking at IP policy through the right lens Objectives of IP compared to Agency portfolio outcomes Outcome IP related deliverables Objective of IP Maximise wellbeing of Australians by providing appropriate incentives for innovation, investment and the production of creative works while ensuring further innovation, competition, investment and access to goods and services are not unreasonably impeded. IP Australia Increased innovation, investment and trade in Australia, and by Australians overseas, through the administration of the registrable IP rights system, promoting public awareness and industry engagement, and advising government. Rights administration performance criteria include: processing of IP rights meets specified timeliness and quality standards. Advice to Government performance indicators include: satisfaction of stakeholders with quality and timeliness of advice. Department of Industry, Innovation and Science Enabling growth and productivity for globally competitive industries through supporting science and commercialisation, growing business investment and improving business capability and streamlining regulation. Programs include those that promote the growth of internationally competitive industries by facilitating nationwide action on deregulation, collaboration, commercialisation and international engagement. Performance criteria include increased investment by businesses in intangible assets (IP products). Department of Promote an innovative and competitive Communications communications sector, through policy and the Arts development, advice and program delivery, so all Australians can realise the full potential of digital technologies and communications services. Strategy includes reforming the copyright regulatory framework to promote creativity, commercial activity and access, particularly in relation to the digital environment. Treasury Performance indicators include: advice is timely, of high quality, and is based on an objective and thorough understanding of issues and a whole-of-government perspective. Informed decisions on the development and implementation of policies to improve the wellbeing of the Australian people, including by achieving strong, sustainable economic growth, through the provision of advice to government. Sources: Portfolio budget statements 2016-17. 508 INTELLECTUAL PROPERTY ARRANGEMENTS Reflecting such concerns, others participants highlighted the possibility of locating responsibility for IP policy in the Treasury portfolio to provide an economywide perspective (Moir, sub. 130, Gruen, trans., pp. 725–26). In 2008, the Review panel of the National Innovation System recommended that IP policy ‘make the same transition as competition policy did in the 1980s and 90s’, which was to move into the Treasury portfolio (Cutler 2008, p. 19) (box 17.6). The transitional costs of consolidating responsibility for IP policy in Treasury would likely be orders of magnitude higher than consolidating all IP policy in the DIIS as it would require relocating policy expertise for all industrial IP rights, as well as transferring copyright expertise. While such institutional changes are not unprecedented (as demonstrated by the history of competition policy), the key question is whether the benefits of assigning responsibility for IP policy to an agency with a strong economywide perspective are sufficient to justify the downside risks and these transitional costs. Box 17.6 Why did competition policy move to the Treasury portfolio? A key impetus for Australia’s competition policy reforms in the 1990s was the concern that the limited purview of federal competition policy arrangements would severely constrain the scope for further economic reform and there was a need to expand the scope of competition law to areas of the economy that had been immune to the Trade Practices Act (Kain, Kurruppu and Billing 2003). Subsequent competition reforms resulted in (among other things): • the scope of the Trade Practices Act being extended so that it applied to all forms of business activity in Australia • at the institutional level, the Trade Practices Commission and the Prices Surveillance Authority being merged to form the Australian Competition and Consumer Commission (ACCC). Competition policy reforms initiated in the 1990s had pervasive effects across many sectors of the economy and involved considerable cooperation across multiple levels and branches of government and other stakeholders. Former chairman of the ACCC, Professor Allan Fels has noted that the shift in responsibility for competition law from the Attorney General’s department to Treasury in the early 1990s was a positive step in advancing competition policy: Traditionally the ACCC and its predecessors were part of the Attorney-General’s portfolio … Treasury tends to understand the broad economy wide need for competition law more than most. Further, in Australia, most business regulation is now within the Treasury although most came after the ACCC. Treasury is more able than most to look at economy wide issues. It is not beholden to any one sector and is sensitive to the global ramifications as well. Treasury is the natural home (Fels 2001, p. 18). While arguably not fully analogous with IP policy, the move of competition policy to the Treasury is illustrative of the enduring importance of getting the policy institutional settings right. INTELLECTUAL PROPERTY’S INSTITUTIONAL ARRANGEMENTS 509 Other approaches to promoting integrated IP policy development Some participants suggested that the Government could achieve a more integrated approach to IP policy through greater coordination across agencies, rather than consolidating policy responsibility (DIIS, sub. DR615; IP Australia, sub. DR612; Open Source Industry Australia, sub. DR486). IP Australia (sub. DR612) suggested that a standing interdepartmental committee to further systematise collaboration between relevant agencies, which would be led by the key policy agency, could help achieve a more integrated approach. Governments in Australia and overseas have previously used standing interdepartmental committees to progress reforms that require cross-government cooperation and coordination. To help implement the national innovation and science agenda, for example, the Australian Government announced a new subcommittee of the Cabinet called Innovation Science chaired by the Prime Minister and an interdepartmental committee in recognition that the issue covers nine different portfolios (DPMC 2015). In 2003, the Japanese Government established the Intellectual Property Strategy Headquarters within Cabinet, to plan, develop and implement Japan’s IP strategy and coordinate policies among concerned ministries. Members included ministers and experts from the private sector (Arai 2005; JPO 2012a). Establishing more formal arrangements for collaboration and cooperation across the different agencies responsible for IP policy, such as through a memorandum of understanding, might also promote greater integration of IP policy. Such arrangements might include: • identifying areas of common interest (such as overlaps or interactions between IP laws) that would benefit from a coordinated and consistent response (such as Indigenous IP) • affirming agencies’ commitment to the overarching objective and principles of the IP system (discussed above) • specifying how the parties will consult with each other and share information and resources. The direct cost of developing formal arrangements for collaboration and cooperation, such as memoranda of understanding, is likely be a relatively low. Further, the DIIS recently advised that it is looking to increase integration of IP policy through a standing interdepartmental committee (IDC) that would discuss IP policy issues at senior levels. This approach offers a number of advantages including promotion of integrated policy development with a strategic focus, and mitigation of risks involved with ad-hoc and reactive consultation. The IDC would be guided by terms of reference and one of its first tasks could be the development of an overarching IP policy framework, as suggested by the Commission. It could consider the interaction between IP and broad issues related to competition, trade, industry, and digital economy policies to aid Ministerial consideration and government decision making. … 510 INTELLECTUAL PROPERTY ARRANGEMENTS In supporting a retention of the current arrangements, the department is investigating options for increased collaboration with Treasury. This will enable competition and consumer issues to be properly considered during IP policy development, while maintaining a focus on providing incentives for innovation and creativity. (sub. DR615, p. 9) A public commitment by the Australian Government to concrete actions to encourage more effective working arrangements between agencies involved in IP would provide a signal of its desire to address participant concerns about fragmented IP policy. It would also be consistent with section 17 of the PGPA Act that requires the accountable authority of a Commonwealth entity to encourage officials of the entity to co-operate with others to achieve common objectives, where practicable. Where to from here? The evidence both for and against policy consolidation is limited and judgment is required. On balance, the Commission considers that government should not consolidate responsibility for IP policy into a single entity at this time. • Participant concerns about copyright policy in Australia relate to outcomes when copyright was the purview of the Attorney General’s Department. While it is too early to tell whether the transfer of copyright to the Department of Communications and the Arts will result in better policy outcomes for copyright, there are some reasons to at least allow the Department time to demonstrate its performance in addressing existing concerns. For example, there does appear to be a healthy tension in copyright policy debate that arises from the competing interests of stakeholders in the communication and arts sectors that was not present when copyright was in the Attorney General’s Department. • Measures to increase policy coordination and appropriate resourcing could provide many of the same benefits, but with relatively low costs and disruption to current administrative arrangements. The Commission considers measures to increase policy coordination and contribute to integrated policy should include: • establishing a standing (interdepartmental) IP Policy Group and formal working arrangements to ensure agencies work together within the policy framework outlined in this report. The Group would comprise those departments with responsibility for industrial and creative IP rights (DIIS and the Department of Communications and the Arts), the Treasury, and others as needed, including IP Australia and other agencies mentioned in section 17.1 (such as the Department of Health and the ACCC) • the Minister for DIIS would assume lead Ministerial responsibility for overall IP policy development (except for standalone copyright matters, which would remain with the Minister for Communications). INTELLECTUAL PROPERTY’S INSTITUTIONAL ARRANGEMENTS 511 The functions of the IP policy group could include, inter alia: • informing Australia’s negotiating positions on IP in international agreements and advising on best practice approaches to including IP provisions in international agreements (discussed below) • developing a multilateral reform agenda (to pursue with like-minded countries) for areas of substantive IP law, such as standards of protection, with the aim of achieving more balanced IP policy settings (chapter 18) • maintaining a watching brief on IP policy issues and ensuring research and analysis is undertaken by the relevant Departments and agencies (including IP Australia) (such as monitoring the use of software patents, reviewing the ongoing necessity of pharmaceutical patent extensions of term and changes to data collected to assess that scheme, and the assessment of Australia’s decision to join the Hague Agreement) (chapters 9, 10 and 11). • bringing together a range of agencies to examine the issues of Indigenous cultural intellectual property more holistically (chapter 2). Establishing a clearer separation of IP rights policy and administration Regardless of where policy responsibility for IP ultimately resides, there is the question of whether responsibility for IP policy should be more clearly separated from responsibility for IP rights administration. In making any separation clear, it is important to note that the division will not be absolute but a matter of degree. IP Australia (like other regulators and rights administrators) has technical and working expertise that is an integral input to policy development and design. In respect to policy development, the OECD recommends: The responsibility for setting or advising on government policy, particularly relating to the nature and scope of the regulator’s powers and functions, should not principally sit only with the regulator even though the regulator has the most up to date knowledge of the issues in the regulated sector. The principal responsibility for assisting the executive to develop government policy should sit with the responsible executive agency and the regulator should have a formal advisory role in this task. In all cases such policy should be advanced in close dialogue with affected regulatory and other agencies, and there should be specified mechanisms for regulators to contribute to the policy-making process. (OECD 2014, p. 30) Clearer separation of IP policy from administration could guard against the risks outlined in section 17.2. It would also clarify the respective roles of the policy Department (for example, the DIIS) and the IP rights administrator (IP Australia). The Department would have a clear responsibility to maintain policy-making capacity in IP, which is important for ensuring a broader government perspective is brought to bear on IP issues including the role IP plays within the broader innovation system. 512 INTELLECTUAL PROPERTY ARRANGEMENTS One way to establish a clearer articulation and separation of IP policy and rights administration would be for the relevant Minister to issue a formal statement clarifying the respective roles of IP Australia and the Department in IP policy development. This statement could include obligations for the Department and IP Australia to develop transparent protocols for developing policy advice. Publishing this information would increase transparency by making IP Australia’s remit and the resourcing of policy capacity in the Department clearer to stakeholders and the broader community. Such measures would seek to complement rather than duplicate agencies’ existing performance reporting and strategic planning. The Australian Government has traditionally used devices such as ‘statements of expectation’ to clarify the roles of Commonwealth statutory agencies, such as the ATO, ASIC and ACCC, recognising that statutory agencies enjoy greater independence from Government than regulators or administrator agencies in government ministries (Uhrig 2003). However, the OECD (2014) and others have highlighted that statements of expectations (or similar documents) can also be of value in clarifying the role of ministerial regulatory units (box 17.7). For example, the Victorian Government uses statements of expectation and intent for both departmental and statutory agencies (DEDJTR 2015a; Victorian DTF 2014). Another way of establishing a clearer separation of policy and regulatory and administrative functions is through structural separation. This would involve establishing IP Australia as a statutory agency with enabling legislation outlining its objectives, powers and functions and accountability requirements. There is scope for greater transparency Many participants suggested that the respective roles of the DIIS and IP Australia in IP policy should be clearer. However, they had different views on how to achieve a clearer separation of IP policy and administration. Some participants suggested making existing roles and responsibilities more transparent would be a useful first step. Open Source Industry Australia (sub. DR486, p. 18) argued that ‘IP Australia and DIIS should make an in-principle public statement outlining their respective responsibilities for IP’. IP Australia (sub. DR612) noted that having a transparent protocol about responsibilities with DIIS could assist in addressing concerns about the dual role of IP Australia as an administrator and provider of policy advice. Other participants said that the Department should take a more active role in IP policy. FICPI Australia (sub. DR581) argued that IP Australia should focus more on administrative functions, leaving policy to a specific department dedicated to dealing with IP policy issues. The Law Council of Australia (sub. DR490) proposed that a policy department (responsible for all IP matters) have dedicated and co-ordinated units that are responsible for policy formulation and implementation within each area of IP, including INTELLECTUAL PROPERTY’S INSTITUTIONAL ARRANGEMENTS 513 legislation and international engagement. It suggested the regulatory body for IP rights should continue to have a role in policy making but should be counterbalanced by a body providing independent advice. Box 17.7 The relationship between policy departments and administrators and regulators: the example of taxation Statements of Expectation The OECD (2014, p. 81) notes statements of expectation are ‘a good mechanism for ministers and regulators [both independent and ministerial regulatory units] to achieve clear expectations’. Statements of Expectations (SOE) are a formal and public statement made by a responsible Minister to a regulator outlining relevant government policies, regulatory objectives and government’s expectations of how the regulator should conduct its operations. The regulator formally responds by outlining its intent to meet these expectations in its corporate plan or similar document, such as a Statement of Intent. SOEs enable Ministers to provide greater clarity about the role and responsibilities of regulatory or administrator agencies and their relationship with the Government. For example, the SOE for the Australian Taxation Office (ATO) sets out the Minister’s expectations regarding the relationship with the responsible Minister, the relationship with Treasury, regulatory cooperation and transparency and accountability. The OECD (2016b) notes statements of expectations need to bring clarity to the respective roles of the regulator and the department and serve as an incentive to strengthen and improve internal processes. It cautions they could have little use if the statements become a ‘shopping list’ of vague and broad expectations and could end up being counterproductive if they are perceived as heavy handed, suggesting outcomes on technical areas which are clearly within the scope of the regulator’s functions. Some agencies have more detailed protocols detailing the working arrangements between agencies. ATO-Treasury protocol Since 2002, Treasury has had responsibility for advising relevant government ministers on both tax policy and the design of tax laws, while the Office of Parliamentary Counsel (OPC) has responsibility for drafting legislation. The ATO has continuing responsibility for administering the tax law, and brings its experience with that administration to bear to assist Treasury with the development of tax policy and the design of law. The ATO-Treasury protocol outlines arrangements for four areas: The integrated tax design process; Quality assurance of new law; Revenue costing; and The law and its administration. Within the framework set out in the Protocol: • Treasury has accountability for providing advice to government on policy and law design • the ATO’s administration of laws includes forming views about the interpretation of those laws, recognising that the courts are the final arbiter on matters of statutory interpretation • the nature of the ATO’s input into the policy development process may include commenting on drafting instructions, draft law and explanatory memoranda, and may involve direct discussions with OPC where appropriate. The ATO’s input into tax law design extends to the passage of the law through the Parliament. Sources: Australian Government (2016e); ATO (2012); OECD (2014). 514 INTELLECTUAL PROPERTY ARRANGEMENTS No participant advocated establishing IP Australia as a statutory agency. The Law Council of Australia (sub. DR490) suggested the regulatory body could continue to sit within the DIIS or be outside subject to ministerial direction, similar to the ATO and ASIC. However, it did not express a preference for either model. Open Source Industry Australia opposed formally separating the regulator and policy department, noting the risk of blurring policy and administration does not disappear if the two are formally separated. Although separation might help, other processes are necessary as well. Those other processes could just as well be implemented in one agency as in two. (sub. DR486, p. 19) Where to from here? The Commission considers that the respective roles of DIIS and IP Australia in IP policy development should be distinct and transparent. The allocation of policy functions to IP Australia should have (and be seen to have) regard to real or perceived conflicts arising from IP Australia’s role as IP rights administrator. There is also a good case that the DIIS should play a more active role in shaping IP policy and dedicate resources to IP policy accordingly. However, the Commission is mindful that strict demarcations between policy and administration functions could unnecessarily limit the DIIS’s ability to draw on IP Australia’s expertise. The management of perceived conflicts in responsibilities should not come at the expense of well-informed IP policy. To help clarify the respective roles of IP Australia and the DIIS, particularly with respect to IP policy development, the Minister responsible for IP should outline the functions and responsibilities for IP Australia through a public statement of expectations or similar document. The statement could cover issues such as the Government’s overall objectives of the IP system (mentioned above) and how IP Australia should contribute to IP policy development. IP Australia should publicly respond to the statement. The DIIS and IP Australia should also develop and publish protocols on working arrangements. The protocol should, among other things, outline the DIIS’s approach to identifying and managing potential conflicts when allocating policy-related functions (including ad hoc responsibilities) to IP Australia. The statement of expectations and working arrangements would be akin to those outlined in box 17.7, but would be tailored to the specific circumstances and governance issues arising in IP. Reflecting the DIIS’s responsibility for leading IP policy on industrial IP rights, the Government should ensure steps are taken to establish and maintain ongoing and robust IP policy expertise in the DIIS. IP Australia would retain its research capability and inform policy deliberation based on their experience in administering IP laws. The case for establishing IP Australia as a statutory agency The main purported benefit of the independent regulator model is protecting regulatory activities from direct political influence and influence of interest groups. This can in turn INTELLECTUAL PROPERTY’S INSTITUTIONAL ARRANGEMENTS 515 contribute to objectivity, consistency and predictability in the administration of regulation and build public trust. The independent regulator model can also accommodate the separation of policy and regulation and support transparency by specifying the regulator’s objectives and mandate in legislation. However, some consider that the independent regulator model does not address, but rather contributes, to such concerns. For example, in 2011 the UK Government opposed providing the UK IPO greater statutory independence. The Government’s preference would be to retain IPO as a body combining practical experience of the IP system through its rights granting and advisory functions with policy responsibility for IP. Separation of the two would create further risks of lobbying leading to disadvantageous outcomes, and it is right for Ministers to carry ultimate responsibility for IP policy as part of the Government’s wider innovation and growth policies. (2011, p. 13) Independent regulators are also likely to be more costly because a separate organisation is established with new accountability arrangements (although in the case of IP Australia there is already a degree of physical and administrative separation from the DIIS). Establishing an independent regulator can also have the effects of reducing flexibility to allocate tasks and resources across the regulator and Department, and to respond to changing government and community priorities (Department of Finance 2015; Uhrig 2003). Legislation may become dated and can be difficult to change. For these reasons, the Uhrig review of statutory authorities in 2003 noted: Consideration should be given to whether functions can be accommodated successfully within a departmental structure or an executive agency, reducing the need for the creation of a separate authority and the associated costs and demands placed on the public sector. (2003, p. 58) OECD (2014) guidelines suggest that the decision to establish an independent regulatory agency will depend on the level of risk (or perceived risk) to the regulator’s integrity. In particular, they argue that independent regulatory agencies should be considered in situations where: • there is a need for the regulator to be seen as independent, to maintain public confidence in the objectivity and impartiality of decisions • both government and non-government entities are regulated under the same framework and competitive neutrality is therefore required or • the decisions of the regulator can have a significant impact on particular interests and there is a need to protect its impartiality. IP Australia already enjoys a degree of independence though its statutory office holders and designation as a listed entity that operates on a cost recovery basis. The Department of Finance notes: A common misconception is that independence can only be achieved by creating a separate Commonwealth entity. This is not the case. Enabling legislation sets out the roles and 516 INTELLECTUAL PROPERTY ARRANGEMENTS responsibilities of an activity, function or body and cannot be altered without parliamentary amendment. A statutory office holder can operate within a Commonwealth entity, and even be supported by staff of the entity, while exercising their statutory role independently as described in the enabling legislation. One benefit of this approach may be that the activity can be carried out at a lower cost within the entity (which will already have support functions like a chief financial officer and a human resources section). (2015) The Commission was unable to identify an example of a country that has successfully implemented the independent regulator model for IP. The Commission considers that a clearer separation of policy and rights administration can be achieved without having to establish IP Australia as a statutory agency. RECOMMENDATION 17.1 The Australian Government should promote a coherent and integrated approach to IP policy by: • establishing and maintaining greater IP policy expertise in the Department of Industry, Innovation and Science • ensuring the allocation of functions to IP Australia has regard to conflicts arising from IP Australia’s role as IP rights administrator and involvement in policy development and advice • establishing a standing (interdepartmental) IP Policy Group and formal working arrangements to ensure agencies work together within the policy framework outlined in this report. The Group would comprise those departments with responsibility for industrial and creative IP rights, the Treasury, and others as needed, including IP Australia. Facilitating independent expert input on IP policy development Some participants questioned the scope for independent expert input following the recent abolition of ACIP and called for the establishment of an independent advisory body. Other than standalone IP reviews undertaken by organisations such as the Australian Law Reform Commission and the Productivity Commission, the Government could obtain independent expert advice by establishing a specialised advisory group with expertise in IP policy. There are different forms the advisory group could take. These include a standing body (such as ACIP) or a panel of experts, which Government could convene and draw upon as needed (such as the Financial Advice Market Review Expert Advisory Panel in the UK) (FCA 2015; PC 2013b). Christie (sub. 29, p. 2) proposed that a body similar in operation, responsibility and membership to the recently-abolished ACIP should be established, to provide ‘informed, frank and fearless advice to government on all key issues of IP policy’. Open Source INTELLECTUAL PROPERTY’S INSTITUTIONAL ARRANGEMENTS 517 Industry Australia (sub. DR486) similarly called for a ‘transparent, arms-length process for policy review and commentary, similar to the former Advisory Council on Intellectual Property’. The Institute of Patent and Trade Mark Attorneys of Australia (sub. 73, p. 18) said ‘it may be preferable to establish panels of experts containing particular expertise relevant to the policy area being investigated’. In principle, the Commission believes that there is a good case for facilitating independent expert input on IP policy development and exposing policy ideas to external scrutiny. However, it does not consider that such input needs to come from a standing body. A standing body of independent experts has advantages, such as the ability to take a longer-term and broader perspective on IP issues and to maintain a watching brief on emerging issues. However, it requires an ongoing commitment of public resources. Given the number of major IP rights reviews that have been undertaken in recent years (the recommendations of which, if accepted, will take time to implement), future workflow might be insufficient to justify the resources to maintain a standing body of independent experts. Without a clear work program, there is a risk that a standing independent advisory body would be diverted to lower level or largely academic issues. A broad panel of experts could achieve many of the benefits of a standing advisory body, at a lower cost. For example, an expert panel could be drawn upon to examine broader policy issues during the policy development stage. The expert panel model also has the advantage that the Government could draw on the panel to establish the requisite skill set from project to project, which could contribute to higher quality and/or more timely advice. (The IP policy group proposed above could help guide and oversee the research agenda and monitor emerging issues for investigation, including those outlined in box 17.3). Emerging contemporary practice has seen a trend towards selectively drawing upon broader panels of experts, on a tailored and as-needed basis. For example, the capability review of ASIC recommended that ASIC recalibrate its advisory panel setup such as through a larger pool of experts that can be called upon to advise on various issues as needed based on issue-specific needs and expertise gaps, coupled with regular performance assessment and enhanced internal responsibility to act on recommendations (ASIC Capability Review Panel 2015). Additional external scrutiny of IP Australia decisions In the past there have been some concerns about the quality of patents granted by IP Australia (section 17.2). These concerns predominantly relate to pharmaceutical and gene patents and have prompted calls for additional external oversight of IP Australia’s regulatory decisions: • 518 The Pharmaceutical Patents Review Panel called for the establishment of an external body, the Patent Oversight Committee — potentially as part of ACIP — to audit the INTELLECTUAL PROPERTY ARRANGEMENTS patent grant processes to help ensure new standards under the Intellectual Property Amendments (Raising the Bar) Act 2012 (Cth) are achieved and to monitor evergreening in respect of patents for follow-on pharmaceuticals (Harris, Nicol and Gruen 2013). • Professor Drahos (as part of the Gene Patents inquiry) called for a Patent Audit Committee to undertake independent audits of the quality of patents granted in a chosen area (Professor Peter Drahos, Committee Hansard, 20 August 2009, p. 15). The Senate Community Affairs References Committee endorsed the proposal and further suggested a Patent Audit Committee could assess the operation and performance of the patent system, particularly in relation to areas of complex or emerging technology (SCARC 2010). In responding to the recommendation of the Senate Community Affairs References Committee to establish a Patent Audit Committee in November 2011, the Government noted that ACIP already had the powers to undertake quality reviews where directed by the Minister and to co-opt temporary members with expertise in the relevant subject area of a review. In particular, ACIP could be tasked with providing advice to the Minister on matters such as: whether the patent system appropriately balances economic considerations with the needs of the community (including benefits to the community); emerging technologies and access issues; and compulsory licensing (Australian Government 2011). As noted above, a former ACIP member has questioned whether ACIP was able to fulfil its oversight function in practice. The Government also noted that there are a range of measures to ensure the quality of individual patents. … any such reviews would be in addition to existing avenues to assure the quality of individual patents in Australia including substantive patent examination, re-examination, pre-grant opposition procedures, third party notification under section 27 of the Patents Act 1990, the administrative and judicial review system, and IP Australia’s internal quality audits and transparency in the prosecution of patent applications through the eDossier facility (which provides on-line, free of charge, public access to relevant documents and correspondence on the patent application prosecution file). The Intellectual Property Research Institute of Australia (IPRIA) also has an active and varied research program looking at various topical patent issues, including issues of quality. (Australian Government 2011, p. 14) Appendix F outlines current measures to ensure the quality of individual patents. In May 2013, the then Parliamentary Secretary for Climate Change, Industry and Innovation announced that the Government would appoint a Patent Audit Committee to ‘advise on patent policy settings and undertake audits of patent approvals for certain technology groups’. However, the committee never came into being. In recent years, IP Australia has taken steps to improve its internal quality auditing processes (including trialling a pilot quality benchmarking and review process with other Vancouver Group countries, chapter 18). While the Commission recognises concerns INTELLECTUAL PROPERTY’S INSTITUTIONAL ARRANGEMENTS 519 relating to pharmaceutical patents, as discussed in chapter 10, the Commission considers the relatively low inventive step has contributed to this outcome and recommends addressing the issue directly. The recent High Court decision on gene patents has clarified issues in that area. The Commission, therefore, does not consider that further changes are warranted at this time. 17.4 Options for reform — international context International treaties have strongly influenced Australia’s domestic IP settings (section 17.1) and many inquiry participants have expressed concerns about Australia’s approach to negotiating IP provisions in international agreements (section 17.2). Participants’ concerns about the treaty-making process include a lack of independent impact assessment and a lack of meaningful stakeholder consultation and public access to negotiation documents. Independent assessments Independent impact assessments can help promote public trust in the treaty-making process. The Commission and others have previously made a number of suggestions for improving treaty-making processes, including the use of independent assessments of agreements (before and after negotiations) (box 17.8). For example, Charlesworth et al. (2006) proposed that Joint Standing Committee on Treaties (JSCOT) should make greater use of independent experts to critique and highlight issues in NIAs. The Harper Review made similar recommendations with particular reference to the process of negotiating IP provisions in trade agreements. In particular, the Panel recommended that: • a separate independent review should assess Australian Government processes for establishing negotiating mandates to incorporate IP provisions in international trade agreements • trade negotiations should be informed by an independent and transparent analysis of the costs and benefits to Australia of any proposed IP provisions. Throughout this inquiry, a number of stakeholders affirmed their support for the Harper recommendations (BCA, sub. 59; Australian Digital Alliance, sub. 108; CHOICE, sub. 26). To date, successive Australian Governments have resisted proposals for independent assessments of agreements, indicating that current arrangements are appropriate. For example, the Australian Government response to the Harper review stated that: The Government does not support a separate independent review of the Australian Government processes for establishing negotiating mandates to incorporate intellectual property (IP) provisions in international trade agreements. The Government already has robust arrangements in place to ensure appropriate levels of transparency of our negotiating mandate while 520 INTELLECTUAL PROPERTY ARRANGEMENTS protecting Australia’s negotiating position. These include public and stakeholder consultation; feasibility studies and cost benefit analyses; and whole of government agreement to negotiating positions. Once a free trade agreement (FTA) is signed, regulation impact statements and national interest analyses are published and the agreement is scrutinised by the Parliament through the Joint Standing Committee on Treaties, prior to ratification. The Government does not support an independent cost benefit analysis being undertaken and published before negotiations are concluded. Such an analysis would reflect incomplete or inaccurate outcomes, signal Australia’s position to our negotiating partners and potentially compromise our capacity to achieve Australia’s national interest. It would also duplicate the processes outlined above. (Australian Government 2015a, p. 8) The Australian Government did not accept any of the 10 recommendations made by the Senate Report on reforming Australia’s treaty-making process, including that cost-benefit analysis of trade agreements be undertaken by an independent body and tabled in parliament prior to the commencement of negotiations or as soon as is practicable afterwards (SFADTRC 2015). The Government noted ‘Under the existing treaty-making system it has been common practice under successive governments to conduct a feasibility study prior to the commencement of trade negotiations and for this study to be made public’(Australian Government 2016a, p. 4). The Commission has previously expressed concerns about the quality of assessments under current arrangements and suggested areas for improvement. Stakeholder consultation and public access to negotiation documents Transparency is an important means of securing legitimacy and public confidence in actions taken by government including the negotiation of treaties. As Smith notes: … it lets the light in and provides for scrutiny and insight into activities which would otherwise be hidden from view or only revealed by the actors in a manner which they themselves controlled. (Smith 2016) The Senate inquiry into Australia’s treaty-making process noted that a lack of access to information about confidential negotiations, and the impact of a lack of information on the quality of stakeholder consultation, was of concern to the majority of submitters. Participants to this inquiry expressed similar concerns (section 17.2). The Senate inquiry recommended several measures to improve transparency drawing on international approaches. These recommendations included: • tabling the final draft text of the agreement in parliament prior to authorisation for signature (or tabling a document outlining why it is in the national interest for Australia to enter negotiations) • publishing additional supporting information on treaties under negotiation, such as plain English explanatory documents and draft treaty text • providing key stakeholders access to draft treaty text under conditions of confidentiality during negotiations INTELLECTUAL PROPERTY’S INSTITUTIONAL ARRANGEMENTS 521 providing parliamentarians and stakeholders access to treaty text on a confidential basis during negotiations (SFADTRC 2015). • Box 17.8 The Commission’s and other’s suggestions for improving treaty-making processes Productivity Commission Bilateral and Regional Trade Agreement report The Commission’s view was that a more transparent and strategic process to Australia’s trade policy is required to ensure an appropriate focus on policies that are most in Australia’s interests. The Commission’s proposed approach to improve the processes for establishing trade agreements included: • Pre-negotiation modelling should include realistic scenarios and be overseen by an independent body. Alternative liberalisation options should be considered. • A full and public assessment of a proposed agreement should be made after negotiations have concluded — covering all the negotiated provisions. • An overarching trade policy strategy should be developed and published to better coordinate and track the progress of trade policy initiatives, and to ensure that efforts are devoted to areas of greatest likely return. Productivity Commission’s Chair statement to the Joint Select Committee on Trade and Investment Growth More recently, the Commission’s Chair observed that the process of entering into trade agreements could entail a two-stage process, involving comprehensive analysis independent of trade negotiators, but closely informed by them. Above all, this would be transparent and would enable interested parties to take a well-informed look at what are complex agreements to promote understanding of the opportunities and limitations to what might be achieved. The knowledge that such an assessment would occur before signing would bring added discipline to negotiations. The pre-negotiation analysis would be published after Cabinet triggers a negotiation, and the final text analysis published as an input to the pre-ratification Parliamentary review. The analysis to be undertaken before negotiations commence for a particular agreement would include all areas that could be covered by the agreement. The base case would be the status quo informed by assessments of the contemporary trade and investment relations between the relevant nation(s) and Australia, against which different scenarios would be assessed. Senate Report on reforming Australia’s treaty-making process In 2014, the Senate referred an inquiry into the Commonwealth’s treaty-making process to the Foreign Affairs, Defence and Trade References Committee. The Committee made a number of recommendations relating to improving treaty-making processes and examined new methods of consultation and negotiation adopted overseas. It recommended that the Joint Standing Committee on Treaties engage earlier and more comprehensively in the oversight of trade agreements under negotiation; that parliamentarians and stakeholders be given access to treaty text on a confidential basis during negotiations; that trade agreements be subject to an independent cost-benefit analysis prepared at the commencement of negotiations; and that a model agreement be developed as a template for all future agreements that deal with complex issues such as investor-state dispute settlement and intellectual property arrangements. Sources: Harris (2015), PC (2010), SFADTRC (2015). 522 INTELLECTUAL PROPERTY ARRANGEMENTS The Australian Government did not accept any of the Senate Report recommendations on consultation and transparency and noted that: Australia’s existing treaty making system is working well and is sufficiently flexible to accommodate the different approaches needed for the wide variety of treaties to which Australia becomes a party. The existing system allows for extensive consultations and enables briefing of stakeholders where appropriate. (Australian Government 2016a, p. 1) The Commission recognises that the Australian Government (and DFAT in particular) may wish to maintain some flexibility in how it approaches consultation and public disclosure of information for each treaty negotiation. It may therefore be wary of committing to overly prescriptive rules about how it conducts negotiations. In some cases, for example, the Government may wish to withhold information from public view in the short term to ensure that internal discussions on policy proposals can be full and frank, without fear of prejudicial exposure. There may also be instances when the other party seeks to place limits on the disclosure of information. Further, the level and extent of transparency and consultation needs to be proportionate. Some agreements will have larger potential impacts and attract more public interest than others. The drawback of a ‘case-by-case’ approach to each negotiation, however, is that there is limited incentive for government agencies to make information available to other Australian stakeholders and a risk that they will err on the side of not disclosing information. In the European Union (EU), public concern about the Transatlantic Trade and Investment Partnership (TTIP) negotiations between the EU and US has led to government agencies adopting a more proactive approach to transparency in treaty negotiations. In 2014, the EU Ombudsman undertook an inquiry into the arrangements for the public disclosure of information concerning the TTIP negotiation and recommended a range of practical measures, encouraging the European Commission to publish documents proactively and to make information about meetings available. The Ombudsman also opened an inquiry into the fact that the European Council had not disclosed the TTIP negotiating directives. Smith (2016) observes ‘positive responses to both these initiatives and pressure from the European Parliament have significantly increased the amount of information about the TTIP negotiations officially put into the public domain by the EU.’ The proactive approach to transparency in the EU is consistent with the view that negotiators need to consciously consider whether the potential harm of disclosing information outweighs the public interest of transparency. For example, Smith notes: … the Ombudsman made clear in her TTIP Decision, an international partner cannot have an unfettered veto over the disclosure of documents in the possession of the EU institutions … The public interest both in disclosure and in the protection of international relations have to be taken into account. (Smith 2016) INTELLECTUAL PROPERTY’S INSTITUTIONAL ARRANGEMENTS 523 Guidance for developing IP provisions in international treaties To address concerns about a lack of transparency, and facilitate a more comprehensive consideration of domestic IP interests, the Government should develop best practice guidance for including IP provisions in international treaties — an option supported by many participants in this inquiry. Guidance could take different forms. For example, the Senate inquiry into Australia’s treaty-making process recommended that the Government develop template draft text for IP (and other ‘controversial issues’) that could be used for future negotiations. Draft text could be designed to embody Australia’s strategic interests in IP and avoid inconsistency across agreements which target the same policy outcome with respect to IP. Several participants were wary of such an approach. Alexander et al (sub. DR505) argued that draft text would likely encourage repetition of unnecessary text in multiple agreements. Gleeson (trans., p. 462) suggested that prescriptive rules could constrain Australia’s flexibility to respond to changing circumstances in areas such as pharmaceuticals. The extent of these risks will depend on the level of prescription in drafting advice. For example, guidance need not require agencies to draw on template text for IP issues where they would otherwise have been omitted from the agreement. Other commentators suggested that such guidance could be a statement of principles that informs Australia’s negotiating stance. Alexander et al noted: … public discussion at a level somewhere between the attempt to pre-draft detailed treaty text, and the current standard of rhetorical platitudes used to describe IP commitments in treaties in Australia’s National Interest Analyses, is more likely to be successful. Perhaps … the establishment (via a suitable process of consultation) of a statement of principles to inform Australia’s negotiating stance would be more useful. (sub. DR 505, pp. 37–38) Participants suggested a range of principles (including those developed by the Max Planck Institute for Intellectual Property and Competition Law) could guide the development of IP provisions, including those relating to: • avoiding the inclusion of IP provisions in bilateral and regional trade agreements and leaving negotiations on IP to multilateral fora (Open Source Australia, sub. DR486) • protecting flexibility to achieve socioeconomic goals, such as by reserving the right to draft exceptions and limitations (Open Source Australia sub. DR486; MPI 2013) • identifying no go areas that are likely to be seldom or never in Australia’s interests, such as retrospective extensions of IP rights (Lateral Economics, sub. DR187; Open Source Australia, sub. DR486) • identifying best practice procedures relating to IP negotiations such as: – explicitly considering the long-term consequences for the public interest and the domestic IP system in cases where IP demands are accepted in exchange for obtaining trade preferences or other benefits (in impact assessments) 524 INTELLECTUAL PROPERTY ARRANGEMENTS – conducting negotiations, as far as their nature makes it possible, in an open and transparent manner and ensuring that right-holders and industry groups do not enjoy preferential treatment over other stakeholders • ensuring consistency with domestic policies (such as competition policy principles) (Dairy Australia, sub. DR247). Best practice principles for IP negotiations, should be developed in consultation with industry and the broader community, with a view to: • refining principles to ensure they can be applied in practice (for example, overly prescriptive principles regarding consultation could render compliance excessively costly or infeasible; in some cases, it may be more effective to put the onus on the Department to publicly indicate how it will give effect to a principle) • determining whether and how to integrate the principles into the processes for assessing treaty proposals (for example, the OBPR might assess compliance with the principles during the RIS process) • identifying issues that are relevant to Australia’s treaty-making processes generally (for example, disclosure of draft texts for bilateral and regional trade agreements) and determining whether those issues are better addressed though broader changes to Australia’s treaty-making process. The process of developing best practice principles for IP negotiations would help clarify Australia’s strategic interests in IP internationally as well as the specific types of information and consultation valued by business and the broader community. It would also provide an opportunity for the government and business and community groups to identify ways of facilitating independent scrutiny of key documents while managing risks to Australia’s negotiating position. While unilateral action by the Australian Government can help achieve more balanced IP policy settings, some reforms are best pursued through multilateral forums in collaboration with like-minded countries (chapter 18). FINDING 17.1 Australia’s approach to negotiating IP provisions in international treaties could be improved through greater use of independent impact assessment and more meaningful stakeholder consultation. INTELLECTUAL PROPERTY’S INSTITUTIONAL ARRANGEMENTS 525 RECOMMENDATION 17.2 The Australian Government should charge the interdepartmental IP Policy Group (recommendation 17.1) and the Department of Foreign Affairs and Trade with the task of developing guidance for IP provisions in international treaties. This guidance should incorporate the following principles: • avoiding the inclusion of IP provisions in bilateral and regional trade agreements and leaving negotiations on IP standards to multilateral fora • protecting flexibility to achieve policy goals, such as by reserving the right to draft exceptions and limitations • explicitly considering the long-term consequences for the public interest and the domestic IP system in cases where IP demands of other countries are accepted in exchange for obtaining other benefits • identifying no go areas that are likely to be seldom or never in Australia’s interests, such as retrospective extensions of IP rights • conducting negotiations, as far as their nature makes it possible, in an open and transparent manner and ensuring that rights holders and industry groups do not enjoy preferential treatment over other stakeholders. 526 INTELLECTUAL PROPERTY ARRANGEMENTS 18 International cooperation in IP Key points • International cooperation can reduce the transaction costs of seeking and licensing intellectual property (IP) in multiple jurisdictions, improve the quality of examination of applications for IP rights and facilitate trade of IP-intensive goods and services. • Slow progress in multilateral forums has led to increasing reliance on IP provisions in preferential trade agreements (PTAs). But this has given rise to some problems. – Some PTAs embody stronger standards of protection than multilateral agreements, tipping the balance in favour of rights holders and elevating the costs of IP protection. This imposes greater costs on Australia as a net importer of IP, and impedes further creation and innovation. – IP provisions in PTAs substantially constrain domestic policy flexibility and increase the complexity and costs of negotiating IP arrangements in international agreements. • Changes to standards of IP protection should be pursued in multilateral forums — where outcomes are less likely to be driven by the interests of a few — and pursued separately to efforts to align administrative processes. – In line with good practice, international IP agreements should be periodically and independently reviewed. • There is scope for the Australian Government to achieve more balanced IP arrangements within the confines of existing international obligations. – However, pronounced changes in IP rules in areas such as patentability criteria, copyright term and data exclusivity for pharmaceuticals will require proactive engagement with Australia’s international partners. – In the first instance, Australia should seek a review of the TRIPS Agreement by the WTO secretariat under Article 71.1 of the Agreement. • Further collaboration and cooperation through the World Intellectual Property Organisation, World Trade Organization and international IP offices can reduce duplication of search activity, lower transaction costs, and facilitate the licensing and transfer of IP rights across borders. – However, efforts to align administrative processes should not compromise the quality of rights granted. Collaborating IP offices must have sufficiently robust search and examination procedures. Intellectual property (IP) law applies solely within designated territories, whereas IP is embodied in goods and services traded across borders, and so has international reach. Reflecting this, the rules and procedures that apply in foreign IP systems have an important bearing on domestic innovation and policy settings. INTERNATIONAL COOPERATION IN IP 527 In keeping with the global reach of IP, there has been a long history of international cooperation. Cooperation typically occurs with a view to: • streamlining IP administrative processes, including for sourcing, granting and exchanging IP rights • aligning standards of protection, such as the nature of activity afforded protection, minimum terms of protection, and enforcement mechanisms • improving the quality of examination standards for applications for IP rights, for instance, by IP offices sharing information relating to applications lodged in multiple countries, and expertise to build examination capabilities. Cooperative efforts have traditionally progressed through multilateral treaties and international forums such as the World Intellectual Property Organisation (WIPO) and World Trade Organization (WTO). Recently, efforts have advanced through preferential trade agreements (PTAs), resulting in standards of protection that typically exceed those set out in the Agreement on Trade-Related Aspects of Intellectual Property Rights (TRIPS). This chapter considers how to achieve potential gains from international cooperation, while minimising the costs associated with excessive IP protection and a loss of sovereignty. The chapter explores the motivation for different jurisdictions to cooperate on IP arrangements (section 18.1). The following two sections outline various approaches to international cooperation and their outcomes (sections 18.2 and 18.3). The final section discusses opportunities to improve Australia’s approach to international cooperation (section 18.4). Improvements to develop the evidence-base for policy choices and best practice guidance for transparency and consultation processes when negotiating IP provisions in international agreements are explored in chapter 17. Enforcing IP rights internationally is discussed in chapter 19. 18.1 What motivates international cooperation? Globalisation is making international cooperation in IP more important. Technological innovation, reduced trade barriers and economic liberalisation have facilitated geographical fragmentation of production processes globally (PC 2015b). As such, world trade and production are increasingly structured around global value chains, with trade growth since the 1990s dominated by intra-regional trade in intermediate inputs. Global value chains increasingly involve a wider and more diverse range of international suppliers (WIPO 2015h). IP rights are an important foundation for international transactions involving IP transfer, which may occur through licensing and contractual arrangements, foreign direct investment or direct trade in IP-intensive goods and services. Greater participation by firms in global value chains has increased the demand for IP protection in multiple jurisdictions, as reflected in the increase in cross-border filings (WIPO 2015g). 528 INTELLECTUAL PROPERTY ARRANGEMENTS The international reach of IP is evident in applications and grants … Among the clearest indicators of the international reach of IP are applications for, and grants of, IP rights in overseas jurisdictions. Many Australians seek IP rights in other countries. In 2014, Australians filed over four times as many patent applications abroad as they filed domestically. Major destinations for applications were the United States (38 per cent), the European Patent Office and New Zealand (each with 9 per cent) (IP Australia 2016a). Around 65 per cent of applications were filed under the Patent Cooperation Treaty (PCT), which is designed to make it easier to apply for patent rights in multiple jurisdictions (see below). The remaining applications were made direct to destination patent offices (figure 18.1). Figure 18.1 Patent applications by filing route Applications per year (thousands) (i) 10 9 Australian applications filed abroad Direct PCT 2005 2006 8 7 6 5 4 3 2 1 0 2004 Applications per year (thousands) (ii) 3000 Direct PCT 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2012 2013 2014 World applications filed abroad 2500 2000 1500 1000 500 0 2004 2007 2008 2009 2010 2011 Sources: IP Australia (2016a); WIPO statistics database (2016e). INTERNATIONAL COOPERATION IN IP 529 The propensity to seek protection overseas is also evident across other IP rights. For example, Australian trade mark applications filed abroad increased by around 20 per cent between 2005 and 2014. China is the top destination for Australian trade mark filings, overtaking New Zealand in 2011 (IP Australia 2016a). In the same vein, many overseas residents seek IP protection in Australia. The majority of patents granted in Australia are to non-residents (around 93 per cent in 2015) (figure 18.2). In 2014, the United States was the largest filer of patent applications in Australia, followed by Australia and Japan (table 18.1) (IP Australia 2015b). The United States was also the most prominent overseas filer for trademarks, designs and plant breeder’s rights. Figure 18.2 Patents granted in Australia to residents and non-residents Patents granted per year ( thousands) Resident Non-resident 25 20 15 10 5 0 2011 2012 2013 2014 2015 Source: IP Australia (2016a). Australia’s increase in demand for IP protection abroad has been part of a global trend. Patent filings worldwide almost tripled between 1985 and 2014. The combined share of the five largest IP offices (the ‘IP5’) 1 increased from 70 per cent of all filings in 2000 to 82 per cent in 2014 (WIPO 2015g). In terms of international rankings of the number of IP applications filed, Australia ranked 21st for patents, 15th for trademarks and 19th for designs (table 18.2). 1 The IP5 consists of the five largest IP offices: the US Patent and Trademark Office, European Patent Office, Korean IP Office, Japan Patent Office and the State IP Office of the People’s Republic of China. These IP offices together handle around 80 per cent of the world’s patent applications, and 95 per cent of all work carried out under the PCT (fiveIPoffices 2016). 530 INTELLECTUAL PROPERTY ARRANGEMENTS Table 18.1 Origin of IP rights filed in Australia, 2014 IP right Rank Country of origin Patents 1 United States 2 3 1 2 Australia United States 3 Trade marks Designs Plant breeder’s rights No. of filings % growth 2005-2014 % growth 2010-2014 11 544 9 10 Australia 1 968 -19 -17 Japan 1 679 -1 15 41 614 6 981 13 23 8 11 United Kingdom 1 977 55 38 1 Australia 2 616 -7 -8 2 3 United States United Kingdom 1 801 266 64 40 54 -5 1 2 Australia United States 136 76 -20 21 -23 90 3 Netherlands 41 41 37 Source: Commission estimates based on IPGOD data (2016 edition). Table 18.2 Global IP filing activity, 2014a Ranking of total (resident and abroad) IP filings by origin, selected countries Origin Patents Trademarks Registered designs China USA Germany 1 2 5 1 2 4 1 5 2 Japan Republic of Korea France 3 4 6 5 10 3 7 3 9 Italy 10 11 4 United Kingdom Switzerland India 7 8 14 8 12 9 11 8 13 Australia 21 15 19 a Rankings are based on the total number of applications filed by origin. Patent data refer to numbers of equivalent patent applications. Trade marks data refer to numbers of equivalent trademark applications based on class counts — the number of classes specified in applications. Design data refer to number of equivalent industrial design applications based on design counts — the number of designs contained in applications. The table lists selected countries among those with the highest rankings for which at least two types of IP filing data are available. Source: WIPO (2015g). … and in receipts for licence and royalty fees … There has also been strong growth in global licensing of IP. Global trade in patent and creative licences alone was estimated to be more than $A930 billion a year in 2008 or over five per cent of all world trade (Hargreaves 2011). Australia is no exception to this trend, INTERNATIONAL COOPERATION IN IP 531 with total payments and receipts from IP licence and royalty fees increasing over time. Nearly all of this growth has been driven by higher payments made to rights holders in other countries (figure 18.3). Most patent license agreements with other jurisdictions reported to IP Australia involve parties from Europe and the United States (figure 18.4). Figure 18.3 Australia’s international IP licensing payments and receiptsa 5 US$ billion 4 Payments Receipts 3 2 2015 2014 2013 2012 2011 2010 2009 2008 2007 2006 2005 2004 2003 2002 2001 2000 1999 1998 1997 1996 1995 1994 1993 1992 1991 0 1990 1 a Charges for the use of intellectual property (balance of payments) at current US prices. Data for 2002 and 2003 not reported in the World Bank database. Source: World Bank (2016). Figure 18.4 Patent license agreements with other jurisdictionsa 220 165 110 55 0 Europe USA All other Japan China India a Transactions reported to IP Australia where a non-owner of the patent has been granted authoritative permission to use, but not own, the IP rights. Source: IPGOD table 124 (2016 edition). 532 INTELLECTUAL PROPERTY ARRANGEMENTS … as well as trade data Another indicator of the global reach of IP is trade in IP-intensive goods and services. According to this measure, and consistent with figure 18.3, almost all of the growth has been driven by imports of IP-intensive goods and services. IP-intensive imports have more than doubled over the past 15 years (figure 18.5). Figure 18.5 Australia’s trade in IP-intensive goods and servicesa 80 Exports $ billion 60 Imports Imports more than doubled over this period and the IP trade gap has increased 40 20 0 a Total gross value of trade in IP-intensive goods and services includes patent-intensive goods, copyright-intensive goods and services (consisting of merchandise goods and copyright services as measured by the value of royalties, licence fees and other charges) and trade mark-intensive goods. The value of trade with a high IP content is estimated using a quantitative approach based on recent studies. Estimates are provided separately for patent, copyright and trademark intensive goods in appendix C. Source: Appendix C. Protecting IP in multiple jurisdictions is costly Obtaining and protecting IP in other jurisdictions entails transaction costs. The nature and extent of these costs depends on how IP rights are transferred — whether through trade, licensing the technology to an overseas business, or foreign direct investment (FDI). The main transaction costs include: • application-related costs that are incurred when applying for IP rights through overseas IP offices. Costs can arise from different application processes, fees and deadlines, translation procedures, and rules for local legal representation (Maskus 2012) • costs associated with finding a licensing partner and negotiating an agreement in cases where IP rights are licensed to a local distributor. Licensing is a common strategy for rights holders transferring their technology to countries with strong IP rights systems, INTERNATIONAL COOPERATION IN IP 533 such as Australia (chapter 3). Difficulties finding licensing partners is one of the main barriers to international technology licensing (Kani and Motohashi 2012; Pluvia Zuniga and Guellec 2009) • costs associated with enforcing rights in foreign jurisdictions. These costs can include ex ante risks associated with the unfamiliar and/or unpredictable operation and jurisdiction of foreign legal systems. Studies have found legal diversity and uncertainty, including in contract law, imposes financial risk in cross-border transactions that increase costs (Meyer 2006; Wagner 2007; WIPO 2010). These findings suggest that efforts to create platforms for identifying licensing partners and improving legal predictability are likely to be helpful in reducing the costs of protecting IP in multiple jurisdictions. 18.2 Approaches to international cooperation International IP cooperation can include efforts to streamline administrative processes, increase the quality of examination, and align standards of IP protection. The approach to all three can vary according to: • the regulatory cooperation mechanisms used (for example, legally binding treaties versus voluntary standards) (table 18.3) • the countries and parties involved (for example, agreements can be bilateral, plurilateral or multilateral, and cooperation can occur at the government or industry level). Regulatory cooperation mechanisms are not mutually exclusive. Particular mechanisms can be used in conjunction with others to achieve regulatory or policy objectives. For example, regulatory cooperation between Australia and New Zealand involves the use of mutual recognition, harmonised standards, information sharing and memoranda of understanding — though not all prove equally successful (box 18.1). Cooperation on administrative processes and examination quality Government led initiatives The Australian Government is party to a number of administrative agreements that seek to simplify and streamline the process of sourcing, granting and exchanging IP rights internationally. These agreements include a mix of multilateral, plurilateral and bilateral initiatives. 534 INTELLECTUAL PROPERTY ARRANGEMENTS Table 18.3 Comparison of regulatory cooperation mechanisms Mechanism Examples Advantages Disadvantages Regulatory integration and legally-binding treaties (entailing alignment or harmonisation of standards) European Union institutions and directives Multilateral (WIPO Patent Cooperation Treaty) Bilateral agreements Same rules apply to all parties promoting greater compliance Loss of domestic policy flexibility to adapt to changing circumstances Lengthy negotiation processes High costs of enforcement Tendency to force higher standards of protection beyond optimal level Full harmonisation (global property rights) is considered unachievablea Mutual Recognition Agreements Trans-Tasman Mutual Recognition Agreement European Union approach to technical harmonisation and standardisation Preserves state sovereignty in rule making Minimises adjustment costs and duplicative regulatory effort Requires broadly similar regimes and extensive trust between parties Time consuming and costly to negotiate Could reduce the quality of rights if other countries have weak examination standards Non-legally binding and ‘soft law’ instruments WIPO joint recommendations PCT and European Patent Office guidelines for patent examination Lower cost tools that can be easily adapted May provide guidance on interpretation of treaties (for example WIPO joint recommendations) Signals political commitment by parties for future law-making Difficulty achieving compliance and enforcement Countries may selectively adopt parts of internationally agreed standards a Previous attempts headed by the United States, European Union and Japan to negotiate a substantive patent law treaty through WIPO have failed on this front. Sources: OECD (2013); WIPO (2003). The Patent Cooperation Treaty (PCT) is a multilateral arrangement administered by WIPO. Under the PCT, parties can file a single international patent application, rather than separate national or regional patent applications, and so avoid some of the transaction costs of filing in multiple jurisdictions. Applications are sent to one of a number of patent offices designated as ‘International Search Authorities’ to determine if the invention claimed is novel and inventive. However, granting of patents remains under the control of national or regional patent offices. In the past decade around 22 per cent of total world patent applications filed abroad were made via the PCT system, with the rest filed directly with destination patent offices (figure 18.1). INTERNATIONAL COOPERATION IN IP 535 Box 18.1 Trans-Tasman institutional and regulatory cooperation on IP Australia and New Zealand co-operate on IP issues through formal and informal arrangements and treaties, information sharing and memorandums of understanding. The Commission (2015a) considered removal of the exclusion of IP from the Trans-Tasman Mutual Recognition Agreement and concluded that it should be retained. Since there are material differences between Australian and New Zealand patent law, mutual recognition would compromise each country’s ability to meet its international treaty obligations. These differences include differing terms for patent protection and exclusions from patentability. For example, while Australia allows patenting of methods of medical treatment, New Zealand does not. Efforts to pursue regulatory cooperation on IP under the Australia New Zealand Single Economic Market (SEM) have involved proposals to align registration procedures and examination practices rather than IP laws themselves. This includes: • a single joint registration regime for patent attorneys in both countries with the same qualification requirements and professional standards • a single application and examination process for patents filed in both countries • a single trademark and plant variety regime. Australia has passed legislation to give effect to SEM initiatives. However, following the introduction of a bill to the New Zealand Parliament, a parliamentary committee recommended that New Zealand not continue with the single application and examination processes for patents, on the basis it would be unlikely to provide significant benefits to New Zealand. The main reasons it recommended not to proceed were: • lack of benefits in terms of saved time and costs to business and patent attorneys, as differences in patent law and practice between New Zealand and Australia mean that applicants would still be required to consider separate objections to both applications • increased administrative and IT costs to New Zealand firms due to examination and filing fees • patent examiners in Australia and New Zealand already benefit from work-sharing practices and initiatives such as the Global Patent Prosecution Highway and electronic filing procedures under the Patent Cooperation Treaty. IP Australia is now awaiting a decision on the recommendation from the NZ Government. Sources: IP Australia (2016g); NZ Parliament (2016); PC (2015a). Since 2014, IP Australia has participated in the Global Patent Prosecution Highway — a plurilateral initiative that allows patent applicants who have a patent accepted in one country to request expedited examination in other countries. Around 20 countries participate in the initiative, including the United States, Japan and Korea. IP Australia is also involved in a number of collaborative projects with the UK and Canadian IP offices under the umbrella of the Vancouver Group. • 536 In partnership with WIPO, the Group developed a system called Centralized Access to Search and Examination (WIPO CASE). The system provides a single portal through which examiners can access search and examination reports of other patent offices that have examined the same invention. It includes participating offices from more than 20 countries and links to the IP5’s ‘One Portal Dossier’ system. Since June 2015 any patent office may join the system. INTELLECTUAL PROPERTY ARRANGEMENTS • The Group also engages in examination quality benchmarking and review initiatives. There has been a program of examiner exchanges to discuss and assess patent examination and quality review processes. Following up on this work, the participating patent offices exchange application cases for audit twice a year. While its participation has been limited in recent years, Australia is also part of the Asia-Pacific Economic Cooperation (APEC) Intellectual Property Experts Group. The Group aims to promote the protection of IP rights in the Asia-Pacific region through initiatives that include capacity building and international cooperation on patent acquisition procedures (APEC 2016). Australia is also involved in bilateral initiatives that seek to streamline the process of obtaining IP rights. The Australian Government recently signed a memorandum of understanding with the European Patent Office (EPO) covering, among other things, reciprocal access to patent information and establishment of a patent prosecution highway pilot program. Under this program, applicants whose claims have been examined by IP Australia or the EPO may ask for accelerated processing of their corresponding application at the other office. Industry-led initiatives Industry-led cooperation spans the spectrum of IP rights. In the area of patents, ‘patent pools’ have emerged. These embody agreements between two or more patent owners to license one or more of their patents to each other or a third party. Patent pools tend to be created for complex technologies where complementary patents must be combined to produce a new product or innovation, and where there are common technological standards and essential patents are easy to identify. In the copyright space, copyright collectives facilitate cooperation between copyright collecting societies by providing a licensing ‘clearinghouse’ that administers copyright on behalf of multiple rights holders (chapter 5). Australia’s Copyright Agency is a collecting society with authority to license copyright works and collect royalties. It is party to a number of international reciprocal agreements with other copyright management organisations (Copyright Agency 2016). Cooperation on IP standards As noted above, another avenue for international cooperation is to harmonise standards of IP protection. Australia is party to a number of international agreements that set common rules. • Early treaties, such as the Paris Convention for the Protection of Industrial Property (1883) and the Berne Convention for the Protection of Literary and Artistic Works (1886) sought to harmonise IP standards on a multilateral basis. • TRIPS establishes a minimum set of obligations on WTO members for the protection and enforcement of IP. Parties (including Australia) can provide higher levels of protection, but cannot provide less without risking a dispute with another WTO member. INTERNATIONAL COOPERATION IN IP 537 A stocktake of Australia’s multilateral cooperation on IP standards and international commitments is outlined in appendix B. While IP protections have traditionally been the province of multilateral treaties, in recent years IP provisions have been included in PTAs.2 The number of agreements (including both regional and bilateral trade agreements) that contain IP provisions accelerated after the WTO and TRIPS came into force (Valdés and McCann 2014). The number and types of IP provisions vary widely, producing a complex web of agreements. Australia has been a part of this broad trend — almost all the PTAs that Australia has concluded over the past decade have included IP provisions (figure 18.6). 3 Figure 18.6 Evolution of international IP agreements 1998: Trademark Law Treaty 1980: Patent Cooperation Treaty for international patent filing ratified by Australia 1980 1985 2007: Australia ratifies the WIPO Copyright and Performances and Phonograms Treaties Most recent 2006: adoption of Singapore UPOV Treaty on Law of Trademarks 1995: WTO established 2001: Adoption of Madrid Protocol TRIPS (Trademarks) Agreement 1990 1995 1983: ANZCERTA replaced the NZAustralia FTA Multilateral Regional Bilateral 2000 2005 2005: AustraliaUS FTA ThailandAustralia FTA 2003: SingaporeAustralia FTA 2010: AustraliaEurope Wine Agreement 2015: (GIs) Trans-Pacific Partnership negotiations conclude 2010 2015 2012: MalaysiaAustralia FTA 2009: AustraliaChile FTA 2015: Japan-Australia Economic Partnership Agreement China-Australia FTA 2014: Korea-Australia FTA Sources: DFAT (2016b); WIPO (2015e). 2 Treaties are legally binding agreements between two or more countries. Treaties may be bilateral agreements between Australia and one other country, or multilateral agreements between three or more countries (commonly termed conventions) (DFAT 2016d). PTAs include bilateral and regional agreements. 3 Australia has 9 PTAs that include IP-specific chapters currently in force. These are agreements signed with Singapore, Thailand, US, Chile, the Association of South East Asian Nations (with New Zealand), Malaysia, Korea, Japan and China. These countries account for 67 per cent of Australia's total trade (DFAT 2016a). The ANZCERTA, which dates back to 1983, does not include an IP-specific chapter. 538 INTELLECTUAL PROPERTY ARRANGEMENTS Obligations in PTAs tend to go beyond those in TRIPS (often referred to as ‘TRIPS-plus’ provisions). For example, the Australia-United States Free Trade Agreement (AUSFTA) includes an IP chapter that significantly strengthens the protection given to IP rights holders (table 18.4). 4 While AUSFTA is the only PTA that has required substantive changes to Australia’s IP laws, many of the provisions in that agreement have been included in subsequent PTAs with countries such as Chile and Korea and in the TPP, with some resulting in overlapping and complex rules. Table 18.4 TRIPS-plus provisions in the Australian-US Free Trade Agreement Area TRIPS-Plus requirements Copyright term Increase in the duration of copyright protection to the life of author plus 50 years to life plus 70 years (Article 17.4.4) Copyright infringement Criminal penalties and procedures in relation to copyright and trade mark infringement, including in relation to Internet Service Providers (Articles 17.11.26-28; Article 17.11.29 and Side Letter 1) Requirement to provide procedures for dispute settlement for cybersquattinga and to provide public access to an online database on domain-name registrants (Article 17.3) Restrictions on the manufacture and provision of devices or services used to circumvent effective technological measures, and restrictions on the use of such devices or services (Article 17.4.7) Cybersquatting and online databases Anti-circumvention of effective technological measures Protection of encrypted Tighter protections through extension of the previous protection program-carrying satellite signals regime to include foreign and other transmissions not covered by the Broadcasting Services Act 1992 (Cth) and to criminalise end users of unauthorised decryptions (Article 17.7) Second use patents Patent protection for new uses or methods of using a known product (Article 17.9.1) Data protection Obligation to provide five years protection of undisclosed test data for new pharmaceutical products and 10 years for a new agricultural product, including new uses of the same product (Article 17.10.1)b Marketing of generic versions Requirement to provide measures in the marketing approval process to prevent a generic medicine from entering the market before patent expiry (Article 17.10) Objection to Geographical Indications (GIs) Changes to Australia’s system for protecting GIs for wine, with the introduction of procedures allowing third parties to object to protection of a GI, and grounds for refusing a GI application where it is likely to cause confusion with a mark (Article 17.2.12 (b)) a Also known as ‘domain squatting’ refers to the use of an Internet domain name with bad faith intent to profit from the goodwill of a trademark. b The five year data protection period was not a requirement of TRIPS but was consistent with Australia’s existing data protection system when AUSFTA was signed. Sources: Australia United States Free Trade Agreement 2004; DFAT (2004); Richardson (2004). 4 Changes to Australian law resulting from the AUSFTA are outlined in more detail in appendix B. INTERNATIONAL COOPERATION IN IP 539 18.3 The outcomes of Australia’s IP cooperation Streamlining administrative processes has helped reduce transaction costs Some positive outcomes have emerged from Australia’s involvement in initiatives to improve administrative processes and the quality of examination. The WIPO CASE system has enabled patent offices to share search results and exchange examination documentation related to patent applications to facilitate work sharing programs. According to WIPO data, over 13 million applications are accessible through WIPO CASE and as of March 2016, over 45 000 documents have been accessed by participating offices (IP Australia, sub. DR612). As noted, Australia is a participant in the Global Patent Prosecution Highway (PPH). As of June 2015, Australian patent applications had been used as the basis for 974 prosecution highway requests made overseas, the majority of which were received by the United States Patent and Trademark Office (USPTO). IP Australia submitted that the PPH improves administrative efficiency through reduced duplication, while achieving greater consistency of outcomes and access to accelerated pathways (IP Australia, pers. comm., 10 June 2016). In comparison with other IP offices, prosecution highway requests for applications examined by IP Australia have a higher grant rate, lower average pendency of claims, and fewer administrative actions (figure 18.7). 5 The USPTO estimated that the cost savings to applicants from the PPH in 2014, based on the reduced number of administrative actions, were between US$1711 and $10 865 (depending on the complexity of the application) (Trevisian 2014). The USPTO also reported benefits in terms of improved timeliness of patents issued. While cooperative efforts under the PPH have resulted in cost savings and more timely outcomes for patent applicants, it is important to ensure that such efforts do not compromise the integrity of IP administrative processes and the quality of IP rights granted (section 18.4). 5 ‘Average pendency’ is the average period from filing an examination request (for PPH applications, this is the PPH request) to acceptance of the application by IP Australia. Applications that are not accepted that is, are withdrawn or lapse) are not included in the calculation of ‘average pendency’ (IP Australia, pers. comm., 30 August 2016). 540 INTELLECTUAL PROPERTY ARRANGEMENTS Figure 18.7 Global Patent Prosecution Highway (PPH) requestsa,b Selected countries,c July-Dec 2015 (i) Grant rate (per cent) 100 80 60 40 20 0 Australia Canada Columbia PPH requests (ii) Japan Russia USA Taiwan All applications including PPH and non-PPH Average pendancy of PPH requests Taiwan USA Russia Japan Columbia Canada Australia 0 1 2 3 4 5 6 7 Months PPH requests All applications including PPH and non-PPH a Grant Rate = (number of granted patents) / (number of granted patents + number of patents withdrawn or abandoned after first office action). This definition reflects that examination in Australia generally results in one of three outcomes: acceptance, withdrawal or abandonment, but not refusal. b Average pendency is months from time the applicant makes the PPH request to the final decision to approve the application. c Offices with more than 50 PPH requests and provided specific statistical data on PPH. Source: JPO (2016). There is also growing use of international platforms aimed at facilitating the licensing and transfer of IP rights. Easy Access IP is an international collective of universities and research institutions that allows companies free licensing of select technologies, subject to certain conditions. While still a relatively new initiative, there are currently seven INTERNATIONAL COOPERATION IN IP 541 Australian universities registered under the Easy Access IP initiative with international partners in Canada, China, Denmark and Germany (Easy Access IP 2016). Source IP is another initiative with an international dimension — developed by IP Australia to help identify potential collaboration opportunities between businesses seeking to work with public sector partners. IP Australia is pursuing opportunities to promote further use of Source IP to facilitate licensing in international markets. In March 2016, an agreement was made with the UK IP Exchange — a global marketplace for IP trade and licensing. Opportunities with other digital marketplaces are also being explored (IP Australia, sub. DR612). The University of Sydney (sub. DR566) pointed to other international initiatives to facilitate the licensing and transfer of IP, including WIPO Green (which promotes green technology innovation and transfer) and the Philippines IP Depot. Aligning standards of protection has imposed costs In almost all cases, efforts to harmonise standards of protection have entailed a strengthening of Australia’s IP rights. For example, TRIPS extended the term of protection for patents from 16 to 20 years, while AUSFTA extended the term of copyright from life of the author plus 50 years to life plus 70 years. More subtle changes in AUSFTA included weakening the licensing provisions for patents and extending the scope of performers’ rights. As highlighted in this report, these outcomes are at odds with a well-functioning IP system. They tip the balance too far in favour of rights holders, imposing undue costs on users. This imbalance imposes greater costs on Australia as a net importer of IP, and impedes further creation and innovation in Australia. A notable example is the retrospective extension of copyright term under AUSFTA — as discussed in chapter 4, this extension provided no additional incentives for creators, but imposed additional costs on users. Another cost of including IP standards in international agreements is it significantly constrains Australia’s flexibility to reform its IP arrangements. A number of participants commented on this issue: New international rules have also closed off various sources of flexibility Australia would otherwise have had to reform domestic IP law, and as a result have created real barriers to reform of Australian IP law in ways that would make domestic law more effective, efficient, and adaptable. (Weatherall, sub. 99, p. 11) ‘IP rights’ provisions in treaties … are written as old-fashioned heavy-handed regulation. A modern approach to regulation specifies desired outcomes that should be achieved and leaves it to the person or institution implementing the regulation to determine how best to achieve this … the AUSFTA and the proposed Trans Pacific Partnership Agreement are extremely detailed, leaving no room for each signatory nation to frame the approach that will best achieve agreed goals, given their institutions, culture and laws. (Moir, sub. 137, p. 15) 542 INTELLECTUAL PROPERTY ARRANGEMENTS The unnecessary focus on ‘harmonisation’ of laws in international treaties on IP has clearly served to erode Australia’s sovereignty. (OSIA, sub. DR486, p. 4) Even where obligations in international agreements do not necessitate legislative amendment, they may constrain domestic policy flexibility while offering little in the way of benefits. Most recently, 12 nations signed on to the Trans-Pacific Partnership (TPP) Agreement, which further anchors TRIPS-plus provisions. Submissions to this and other inquiries have raised concerns that such provisions limit Australia’s flexibility to adapt to changing circumstances. The Business Council of Australia argued: While the Australian government was successful in negotiating an outcome that allowed it not to change its domestic settings for biologic data exclusivity, these commitments now bind Australia’s domestic system and make our IP regime less adaptive to change. (sub. 59, p. 6) While some TPP provisions appear to be in Australia’s interests (such as commitments towards greater transparency of IP systems), in some areas the TPP goes beyond other agreements such as TRIPS. And the TPP further locks in past bilateral commitments, complicating any renegotiating efforts that could be taken to strike a better balance. As such, it is unclear whether Australia is a net beneficiary on the IP provisions taken collectively. There is little evidence to suggest that recent increases in the strength of IP rights have increased incentives for innovation and, as explained in chapter 3, evidence suggests efforts have done little to promote technology transfer. Just as problematic is that harmonisation efforts have focused on the strength of IP regimes and appear to have come at the expense of targeting those issues that matter most for reducing transaction costs. As noted above, transaction costs are influenced by information asymmetries with respect to identifying potential licensing partners and enforcement rules and practices used in other jurisdictions. On this matter, the Law Council of Australia said: It is notable that recent trade agreements to which Australia has been a party, including in particular the TPP, appear to have focused on setting baselines for substantive law with little attention to harmonisation at the level that would allow either cross-border trade or streamlining of multi-jurisdictional applications for rights. There is evidence that detailed substantive commitments cause problems for law reform. (sub. DR490, p. 37) And in the Commission’s public hearings, Gruen argued: I think it’s pushing the envelope a bit to say that the terms in AUSFTA or the TPP have got anything to do with lowering transactions costs … they’re about rent creating. (trans., p. 723) Indeed, by focusing almost exclusively on the rights of IP rights holders — and in so doing ignoring the rights of users — recent international agreements have missed important opportunities to meaningfully lower transaction costs that arise when private parties seek to commercialise IP in global markets. INTERNATIONAL COOPERATION IN IP 543 18.4 Where to from here? The Commission has considered how Australia can improve its approach to international collaboration. Achieving better outcomes on standards of protection Efforts to align standards of protection should be separated from other initiatives The preceding discussion highlighted the significantly different outcomes from efforts to streamline administrative processes, compared with efforts to align standards of protection. All parties involved can benefit from streamlining administrative processes (and improving the quality of examination), while aligning standards of protection can benefit some countries at the expense of others. Despite this, efforts to streamline administrative processes and align standards of protection are often bundled together in the same set of negotiations and agreements. For example, the Hague Agreement on design rights seeks to streamline international applications for design protection, as well as align the duration of protection for design rights to 15 years (currently 10 years in Australia) (chapter 11). Efforts to align standards of protection should be separate from efforts to streamline IP administrative processes. This would improve the effectiveness of Australia’s international cooperation in IP by making it easier to target efforts to align administrative processes with jurisdictions which Australia conducts most technology trade, or jurisdictions where technology trade is expected to increase. It would also increase the flexibility in the mechanisms used, including by allowing for less formal mechanisms, such as mutual recognition agreements and voluntary standards. Where we do consider IP standards, multilateral forums should be used A further benefit of separating cooperation on administration and standards of protection is that the latter is best pursued transparently in multilateral forums, where outcomes are less likely to be driven by the interests of a select few. For example, IPTA commented that some discussions between key country groupings have not represented the interests of smaller users of the IP system: Unfortunately, discussions in these groups often take place without input from those representing the interests of smaller users of the system, such as SMEs, and the proposals and policies developed by these groups are sometimes not ones that would suit Australian innovators. It will be important for Australia to monitor carefully what is going on within these country groups and do whatever can be done to minimise negative impacts on Australia. Australia’s interest would seem to be best served by bringing negotiations back into WIPO and the WTO. (sub. 73, pp. 18–19) 544 INTELLECTUAL PROPERTY ARRANGEMENTS Moreover, as Weatherall, Alexander and Handler (sub. 99) highlighted, multilateral approaches avoid the web of commitments that can occur with bilateral approaches to aligning IP standards. As they point out: • International IP obligations are subject to a most favoured nation rule (TRIPS, article 4), and so IP rules concluded in a bilateral agreement must be accorded to nationals of all WTO member countries. • Where plurilateral agreements operate alongside bilateral agreements, IP obligations undertaken in bilateral agreements may bind in addition to rules in plurilateral agreements. These factors in combination can make it difficult to determine which rules apply and to whom. This not only adds to the complexity and costs faced by governments in negotiating subsequent IP agreements, but inevitably adds to the compliance costs faced by Australian firms seeking IP protection abroad and overseas firms seeking to do IP embedded trade with or investment in Australia. However, there are practical difficulties in reaching multilateral agreements. IP Australia noted: Australia supports the multilateral system because multilateral norm-setting outcomes potentially have greater impact and significance for Australian businesses seeking to trade in a wide range of foreign markets. However, the negotiation of multilateral treaties is complex and can be slower than bilateral or plurilateral treaties. (sub. 23, p. 24) Shelston IP commented: … the difficulty in obtaining multilateral agreements via WIPO or WTO is the very reason Australia has sought bilateral agreements or multilateral agreements outside WIPO and WTO … most of the multilateral agreements have been fairly noncontroversial and relate to the front end or clerical processes for obtaining patent protection. Whilst some of the costs associated with these processes may be reduced, it is not clear how further multilateral agreements would benefit Australian businesses or in particular the IP community in Australia. (sub. DR483, p. 23) In practice, countries tend to pursue a combination of approaches. In this context, the Department of Foreign Affairs and Trade (sub. 65) views bilateral and regional negotiations as a means to reinforce multilateral efforts. Indeed, as the example of the PPH illustrates, collaboration that begins as a bilateral initiative can expand to involve a broader range of countries. INTERNATIONAL COOPERATION IN IP 545 Australia is not formally part of major cooperative forums, such as the IP5 and the Tergernsee Group,6 but IP Australia (sub. DR612) commented that it takes a keen interest in these groups and is active in Group B+, 7 which ultimately feeds into Australia’s participation in multilateral forums. As noted above IP Australia has joined with the IP offices of Canada and the UK to form the Vancouver Group of similar mid-sized offices. While these are important initiatives, there is scope for Australia to more actively promote the role of WIPO (and the WTO) in pursuing further efforts for global cooperation on substantive elements of IP law, most notably standards of protection. Stronger multilateral approaches would avoid outcomes being driven by the interests of a few key nations and addresses concerns about domination of international groupings such as the IP5 in multilateral negotiations. As many participants made clear, harmonisation should not be pursued for its own sake, and need not always involve agreeing to prescriptive treaties (box 18.2). In line with good practice, international IP agreements should be periodically and independently reviewed. Box 18.2 Participant views on harmonisation of IP laws Australia should continue to aim to harmonise its IP law with its major trading partners, but … efforts to harmonise should, where possible, be at a policy level … or by less formal agreement such as the administrative and non-binding agreements that exist between patent offices … rather than by committing to prescriptive treaties. In this context it is submitted it is exactly the wrong approach to adopt laws which are simply the minimum to which we are bound by international treaty. (Law Council of Australia, sub. DR490, p. 38) … FICPI … supports the view that total IP harmonisation at the expense of losing domestic policy flexibility is not in the interests of Australian citizens and industry. However, FICPI Australia is supportive of harmonisation to a reasonable extent that dispenses with duplication of formality matters that add to the expense of Australian and other applicants seeking IP rights in other countries. Maintaining a degree of discretion to formulate IP policy and systems to suit its own nationals, while maintaining equal treatment for all users of the system is a desirable outcome … in practical terms, moving towards multilateral agreements on major reform is likely to be the most expeditious way in achieving policy outcomes of advantage to Australia. With sufficient weight behind multilateral reforms, there is a greater likelihood that such reforms will ultimately be taken up by the international community through WIPO. (FICPI Australia, sub. DR581, pp. 21, 50) There may be an asymmetry of influence when Australia negotiates with a much larger country. That can be tempered when we are part of a community of countries with similar interests. Harmonisation of laws should always be secondary to drafting good laws, and indeed eliminating bad ones. (OSIA, sub. DR486, p. 23) 6 The Tegernsee Experts Group consists of representatives from the USPTO, the JPO, the EPO and the patent offices of the UK, Denmark, Germany and France. The Group was formed in 2011 for the purpose of furthering the dialogue on substantive patent law harmonization and has considered particular issues such as the grace period, duration of publication of applications, and prior user rights. 7 Group B+ was established to promote and facilitate progress on issues under consideration by WIPO particularly efforts aimed at substantive patent law harmonisation. The Group consists of all members of WIPO’s Group B, EU members states, the European Commission, the EPO and South Korea. WIPO and Singapore attend the Group’s meetings as observers (EPO 2016b). 546 INTELLECTUAL PROPERTY ARRANGEMENTS RECOMMENDATION 18.1 The Australian Government should: • pursue international collaborative efforts to streamline IP administrative and licensing processes separately from efforts to align standards of IP protection. In so doing, it should consider a range of cooperative mechanisms, such as mutual recognition • use multilateral forums when seeking to align standards of protection. Improving the quality of examination Hasten slowly and carefully — the PPH initiative International initiatives to share patent examination results and mutual recognition agreements with respect to IP application assessment processes — most notably the PPH — have the potential to reduce transaction costs and improve the quality of IP rights granted. Given their importance they should continue to be pursued and improved upon. However, a nuanced approach is required. Research suggests there are substantial differences in the quality of examination across patent offices (van Pottelsberghe 2010; de Rassenfosse, Jaffe and Webster 2016). Since PPH agreements seek to harmonise decision making, they may propagate a poor quality decision into the whole patent family, further weakening patent rights. Indeed, the Hargreaves Review (2011) revealed a lack of trust in the PCT system, and found that patent offices have been reluctant to commit to relying on international standards for granting of patents — based on the belief that the quality of examinations do not meet their own standards. To avoid compromising patent quality, patent offices should ensure their search and examination procedures converge to a rigorous level before entering into mutual recognition arrangements. IP Australia should take a lead role in quality review Concerns about the quality of examination are being addressed by the PCT Quality Sub-group (a group of WIPO members, including Australia). The group is looking at ways to increase confidence in patent search reports. IP Australia is also actively involved in initiatives that focus on ensuring International Searching Authorities comply and continue to improve practices consistent with PCT guidelines. As IP Australia (sub. DR612) noted, improving the quality of products delivered under the PCT can help to foster more confidence and trust in the work of other IP offices. INTERNATIONAL COOPERATION IN IP 547 IP Australia should take a lead role in PCT quality review processes by ensuring that PCT guidelines for International Searching Authorities are not only enforced, but regularly reviewed and updated for best practice. Advocating for changes to standards for IP protection As noted in this chapter and the report more broadly, IP rights have swung too far in favour of rights holders, imposing undue costs on users. While unilateral action by the Australian Government can help achieve more balanced IP policy settings, some reforms are best pursued through multilateral forums in collaboration with like-minded countries. Since poorly set standards can impose significant costs on the Australian community, Australia should be more proactive in seeking to influence the tenor and scope of the rules that apply to IP protection internationally. The Commission has identified a reform agenda that would be mutually beneficial for Australia to progress with other countries in the ‘long-game’ of achieving more balanced IP policy settings (table 18.5). This should not be seen as an exercise in horse-trading or cajoling. Many of the issues that the Commission’s reform agenda seeks to address are equally problematic in other countries. 548 INTELLECTUAL PROPERTY ARRANGEMENTS Table 18.5 Potential areas for international collaboration on IP reform Potential reform area Policy flexibility under existing international obligations Benefits of change Raising the inventive step beyond the level applied in other countries (chapter 7) Some policy flexibility — key concepts such as obviousness are not defined in international agreements. While there are reforms for raising the threshold for inventive step that Australia can and should pursue unilaterally, these would leave the inventive step below the ideal level. Australia should explore opportunities with likeminded countries to raise the threshold further. Allow manufacture for export (MFE) for generic drugs during the patent extension period (chapter 10) Limited policy flexibility — attempts to allow MFE could be subject to challenge under AUSFTA which limits exports from a third party for the purpose of marketing approval requirements. Publication of clinical trial data (chapter 10) Striking a better balance in copyright arrangements (chapter 4) Allowing MFE during the patent extension period would avoid some costs of delayed entry of generics into overseas markets, primarily lost sales. In the short term, most of the potential benefits could be secured by a more targeted extensions of term system. In the longer term, Australia should pursue the steps needed to explicitly allow MFE in the patent extension period. Limited policy flexibility — some Allowing researchers access to IP agreements include restrictive clinical trial data could provide provisions to maintain substantial public health benefits. confidentiality of clinical trial data. International approaches to the Unilateral action by Australia to publication of trial data would publish data could lead to a loss avoid sequencing problems while of protection in other markets and ensuring that researchers (in delay some drugs coming to the Australia and overseas) could Australian market. access clinical data, encouraging more follow-on innovation. Limited policy flexibility — any substantive changes to term would require amending international agreements such as the Berne Convention, TRIPS and AUSFTA, and would be met with significant resistance. There are considerable benefits to consumers in terms of reduced costs and improved access from shorter copyright term, which would not materially reduce the revenue of rights holders or the incentive to create. Initially this could entail introduction of formalities (such as registration of copyright works) for term in excess of life plus 50 years. Review provisions in the TRIPS Agreement provide a mechanism to initiate an international dialogue on IP reform. Article 71.1 obliges the parties to review the agreement every two years. Despite this, there has never been a review of TRIPS under Article 71.1. A review is overdue, and would be a helpful first step in achieving more balanced IP policy arrangements. Australia should advocate and propose that the WTO Council undertake a review of the TRIPS Agreement. Efforts to rebalance IP arrangements are best guided by an economywide perspective (chapter 2). The agencies and departments involved also need to build their capabilities and INTERNATIONAL COOPERATION IN IP 549 dedicate adequate resources from within their existing budget envelopes to the task. As outlined in chapter 17, greater use of independent and public reviews, and more effective consultation, would improve the information used in, and transparency of, Australia’s international cooperation on IP, ultimately leading to better policy outcomes. RECOMMENDATION 18.2 The Australian Government should play a more active role in international forums on intellectual property policy — areas to pursue include: • calling for a review of the TRIPS Agreement (under Article 71.1) by the WTO • exploring opportunities to further raise the threshold for inventive step for patents • pursuing the steps needed to explicitly allow the manufacture for export of pharmaceuticals in their patent extension period • working towards a system of eventual publication of clinical trial data for pharmaceuticals in exchange for statutory data protection • identifying and progressing reforms that would strike a better balance in respect of copyright scope and term. 550 INTELLECTUAL PROPERTY ARRANGEMENTS 19 Compliance and enforcement of IP rights Key points • International comparisons suggest Australia has robust intellectual property (IP) enforcement arrangements. But more can be done to improve the ability of rights holders to utilise Australia’s enforcement infrastructure. • The cost of enforcement is integral to the overall efficiency of Australia’s IP arrangements. • Little data exists on IP infringement in Australia, and even less on the economic impact of infringement for IP rights holders or those seeking to innovate and create. • Online copyright infringement remains problematic for rights holders. Evidence suggests most people infringe copyright online in the absence of timely and competitively–priced access to content. This requires action by rights holders, publishers and other online intermediaries. • ‘Safe harbour’ arrangements are an important part of balancing the interests of users, rights holders and online intermediaries. Australia’s current arrangements unnecessarily restrict safe harbour protection to Internet service providers. – The Australian Government should proceed with its proposal to expand the safe harbour scheme to encompass the full range of online service providers, as occurs in other countries. • While Australia’s enforcement system works relatively well, reform is needed to improve access, especially for small- and medium-sized enterprises. – Recent initiatives of the Federal Court to improve the efficiency of IP litigation should improve enforcement outcomes for high value IP disputes. – Introducing (and resourcing) a specialised IP list within the Federal Circuit Court (akin to the United Kingdom’s Intellectual Property Enterprise Court) would provide a timely and low cost option for resolving IP disputes. The ability of rights holders to enforce their intellectual property (IP), and users to defend their use of IP, is a key element in how well Australia’s IP arrangements work. The ability to enforce the rights granted under law materially impacts on the value of IP. This chapter reviews the evidence on IP infringement and enforcement in Australia (section 19.1), and assesses how well we fare in limiting infringing activities (section 19.2). The chapter then considers two issues raised by participants that warrant particular attention: online copyright infringement (section 19.3) and the time and cost of enforcing IP rights, particularly for small- and medium-sized enterprises (SMEs) (section 19.4). COMPLIANCE AND ENFORCEMENT OF IP RIGHTS 551 19.1 The extent and impact of IP infringement IP infringement varies significantly, both in its nature and impact. Infringement may be intentional, such as importing counterfeit trade marked goods, while others may be inadvertent or done with the belief an exception applies, such as a consumer using a copyright-protected song in a video clip. In other cases, an infringement may be in response to the unavailability of a work in Australia. Many inquiry participants, particularly Australian rights holders, commented on the extent of IP infringement in Australia and, in some cases, overseas. While most submissions discussed copyright infringement, and to a lesser extent patent and trade mark infringement, some submissions and public hearing participants did raise issues relating to registered designs and plant varieties. Many of these comments were at a high level, reflecting the difficulty in establishing evidence on IP infringement rates in Australia and its impact on incentives to innovate and create. Useful data on the rates of IP infringement in Australia is lacking. The limited research that is available frequently relies on surveys, although court and customs seizures data also provide insights. The prevalence of copyright infringement Many participants raised concerns about copyright infringement in Australia, including Dreamtime Public Relations (sub. 2), Foxtel (sub. 115), News Corporation (sub. 119), Sony Music Entertainment Australia (sub. 124), FreeTV Australia (sub. 129), the Phonographic Performance Company of Australia (sub. 123) and the Arts Law Centre of Australia (sub. 117). Most of these participants focused on ‘low value/high volume’ copyright infringement, typified by unauthorised downloading of music, television programs and movies. Illustrative of the views of many rights holders and intermediaries, the Australian Screen Association stated: The threats to the creative industries are greater than ever before. Australia, sadly, has a reputation for being one of the per capita leaders in internet infringement of films and TV shows. (sub. 43, p. 4) Research undertaken by TNS Social Research for the Department of Communications and the Arts (2015) found around one quarter of Australian Internet users aged 12 or above (approximately 5.2 million people) infringed at least one item of online content over the first 3 months of 2015. Responses further suggested around 7 per cent consumed infringing content exclusively, with movies the most likely content to be infringed, followed by TV programs and then video games. By comparison, the number of Internet users aged 12 or above who had consumed at least one item of online content illegally in the UK was 17 per cent in 2013 (IPO 2016a). 552 INTELLECTUAL PROPERTY ARRANGEMENTS While reductions are modest, rates of copyright infringement appear to be falling. CHOICE (sub. 26) detailed the results of its own survey, which found 17 per cent of Australians were regularly infringing copyright-protected television programs in 2015, compared to 23 per cent when CHOICE undertook the survey in 2014 (CHOICE 2015). Patent infringement Survey research provides the most robust source of information on the extent of patent infringement in Australia. A 2010 study of Australian patent applicants found approximately 28 per cent of respondents believed some level of copying of their invention had occurred (Weatherall and Webster 2010). SMEs and individuals are more likely to be aware of their patent being infringed than large companies and public research bodies. Sale by another company, catalogues, trade fairs and information from customers and suppliers were the main ways patent holders became aware of potential infringements. However, claims of patent infringement are often not clear cut. They rarely involve straightforward reproduction of a patented invention and frequently the alleged infringer counter-claims revocation of the patent. As such, it is often unclear whether an infringement has occurred. Data from court cases provides another perspective on the extent of patent infringement. A 2007 study of patent litigation found 399 patent cases were filed in the Federal Court of Australia between 1995 and 2005 (Rotstein and Weatherall 2007). Although data identifying the nature of each individual case was not always available, 47 per cent of those cases related to an infringement, with 10 per cent of cases involving one party opposing the grant of a patent, and only 3 per cent being an appeal from a previous case. The 2010 study by Weatherall and Webster estimated the litigation filing rate to be about 0.5 per cent (filings per stock of patents in force in any year). Trade mark infringement Trade mark infringements range from circumstances where a counterfeit mark has been applied (around one third of actions undertaken in trade mark proceedings), to those where a trader’s mark is inadvertently close to that of a registered mark. The way these cases play out is quite different, with counterfeit cases being fastest to resolve (Bosland, Weatherall and Jensen 2006). Little evidence exists on the extent of trademark infringement. A 2006 study found Australian courts made 81 trade mark enforcement decisions from January 1997 to December 2003 (Bosland, Weatherall and Jensen 2006). Another study of trade mark cases between 1990 and 2010 found 353 court actions (Huang, Weatherall and Webster 2011). COMPLIANCE AND ENFORCEMENT OF IP RIGHTS 553 As noted in chapter 12, court actions for trade mark infringement generally also involve a claim for the tort of passing off. In addition to the civil actions for IP infringement brought by rights holders, the Department of Immigration and Border Protection (DIBP) and the Australian Border Force (ABF) have processes to prevent the importation of goods that allegedly infringe trade marks or copyright. Rights holders must register a ‘Notice of Objection’ with the Department before the ABF can seize goods. The DIBP reports on the number of seizures undertaken by the ABF in recent years for copyright- and trade mark-protected goods, and the estimated value of those goods had they been genuine (figure 19.1). Most seizures relate to trade mark infringing imports. 4000 Border seizures of infringing goodsa Total seizures Total Value of seizures (RHS) 45 3500 40 Annual seizures 3000 35 2500 30 2000 25 20 1500 15 1000 10 500 0 50 5 2012-13 2013-14 2014-15 Value of annual seizures ($million) Figure 19.1 0 Source: Australian Customs and Border Protection Service (2013; 2014; 2015). Infringement by overseas parties As Australian companies seek opportunities in emerging markets, they are confronted by different legal systems and enforcement regimes for IP rights. Many participants raised concerns about infringing activities in other countries, particularly patent and trade mark infringement in China. The China-Australia Chamber of Commerce summed up the difficulties: Despite innovation in China flourishing over the past five years, intellectual property rights (IPR) laws are largely ill-enforced. IPR violations and counterfeiting remain a major problem and concern for Australian exporters, Australian businesses currently operating in China and technology transfer companies alike. (sub. 50, p. 2) 554 INTELLECTUAL PROPERTY ARRANGEMENTS The China-Australia Chamber of Commerce pointed to specific concerns with the Chinese trade mark regime : The current ‘first-to-come first-to-file’ system of trademark registration is open to abuse and manipulation. The first company or person to register a trademark owns the rights to that trademark in China. Unlike Australia and other western countries, misleading and deceptive conduct surrounding intellectual property is rarely recognised in China. It is not uncommon for professional trademark pirates to bring an action towards a legitimate company attempting to enter the Chinese market. Consequently, risks surrounding brand protection is also a major concern. Currently there is no workable solution, and trademark piracy is a growing industry in China. (sub. 50, p. 4) Examples of the concerns raised are easy to come by, especially in the field of winemaking: • Australian winemakers have to ‘smash their bottles’ after tastings and exhibitions to prevent counterfeiters from salvaging and re-using the bottles and labels (Puddy 2016). • some surveys have suggested that 90 per cent of ‘high-profile’ wine brands in China could be fake (Jordan 2014). • Penfolds wine had to purchase its mark from a trade mark ‘squatter’, and suffers from imitators including ‘Benfolds’ wine (Grigg 2014; Port 2010). Participants in this inquiry have also highlighted the practice of copying goods protected by trade marks and other IP rights being reimported into Australia (for example, Cameron Canvas, sub. 1; Trayon Campers, sub. 4). Such problems are likely to become more pronounced following a recent legal ruling in China that has been interpreted to allow for trade mark infringement where goods are produced solely for export (Murdoch 2016). These concerns are not unique to Australia. The OECD found that those countries whose companies had their IP rights most infringed between 2011 and 2013 were the United States, Italy, France and Switzerland and that most counterfeit and pirated goods originate in middle-income or emerging market economies. The UK IPO commissioned research on UK-based businesses operating in China and found around a quarter of firms had experienced a bad faith trade mark application in China. A majority of such firms indicated that their brand reputation was damaged in China, with around 40 per cent indicating that their reputation had also been damaged in other jurisdictions as a result (Future thinking 2015). Trade in counterfeit goods is expected to grow, rather than decline (OECD & EUIPO 2016). Understanding how infringement impacts the community Determining the impact of IP infringement on incentives to create and innovate is not straightforward. Like all investors, creators and innovators invest in IP based on an expectation of earning a return from their activities. For a range of reasons, many investments in IP fail to make their expected return, similar to many investments in the economy more broadly; infringement being only one possible factor. COMPLIANCE AND ENFORCEMENT OF IP RIGHTS 555 From an economic perspective, IP infringement has short– and long–term consequences. In the short term, infringement reduces profits to rights holders, but increases the welfare of those who infringe (see Fink (2008) for a more detailed discussion of the welfare impacts of IP infringement). While rights holders likely expect some infringement (factoring this into their decision whether or not to invest), expectations around sales and profits for new works will be lower as the level of infringement rises. Consequently, as incentives to invest fall, so too does the supply of new inventions and works. Governments can also lose through foregone taxes and incurring enforcement expenses. As Foxtel noted: A range of studies have shown the significant impact of piracy on the Australian economy. … Forgone consumer spending impacts content creators and distributors and ‘ripple effects’ are felt across the economy. Taxes are also forgone, impacting the Government’s ability to invest in services for Australians. (sub. 115, p. 7) Foxtel (sub. 115) also noted a study of the impact of online copyright infringement on the film sector in Australia, costing the Australian economy an estimated $1.37 billion in gross output (lost sales) and 6100 full time equivalent jobs (Ipsos MediaCT and Oxford Economics 2011). However, determining the extent to which infringement displaces sales and impacts the economy is practically impossible. As the Senate Standing Committee on Legal and Constitutional Affairs reported: The statistics that are available have been prepared by various industry groups, and each set differs in the method of its preparation. AGD and the Department of Communications, Information Technology and the Arts (DoCITA) submitted that the methodology is also in many cases untested, in the sense that it has never been fully explained. ADA submitted that for this reason industry statistics should be subjected to a “high level of critical scrutiny”. (2000, p. 7) In addition to lost sales, negative effects of infringement may include lost brand value, increased development costs (to make products more difficult to copy), and reduced incentive to innovate, though the effects vary widely among sectors and companies (AIC 2008; GAO 2010). 19.2 Addressing infringement — how do we fare? It is important to recognise that there will always be some level of IP infringement. In patents for example, there will continue to be disputes about the boundaries of IP rights — where both parties claim that the other is infringing their rights. Further, measures to reduce infringement are not costless. In deciding how best to respond, it is necessary to weigh the costs of taking action against the avoided harm to the community. 556 INTELLECTUAL PROPERTY ARRANGEMENTS As already outlined, disputes arising under Australia’s IP arrangements vary widely, and different types of infringement pose policy challenges that can warrant different solutions. Despite the focus on enforcement, a broader range of options exist for addressing IP infringement. Two often overlooked options include improving market access and improved community understanding and acceptance of the law. Market access Poor market access to IP-protected goods and services plays a role in parties’ decisions about whether or not to infringe, particularly in the realm of copyright. Commenting on the connection between copyright pricing, access and infringement, Electronic Frontiers Australia (EFA) noted: EFA believes that there is clear evidence that the vast bulk of Australian consumers are very willing to consume content legally, where it is available at a fair price, and in a convenient and timely manner. EFA therefore believes that the balance of Australia’s copyright regime should be adjusted significantly to ensure that the rights of consumers and other content users to access content according to the principles of fairness, convenience and timeliness are greatly enhanced. (sub. 114, p. 3) Research undertaken by both CHOICE and TNS Social Research (for the Department of Communications and the Arts) underscore the relationship between timely and cost effective market access and infringement. In CHOICE’s 2015 survey, 38 per cent of respondents cited expense of legitimate content as the reason they infringed copyright; 32 per cent cited the time taken to release new content and 23 per cent the ongoing availability of material (CHOICE 2015). TNS Social Research (2015) found similar results in its study. Several participants highlighted the commercial steps they had put in place to release copyright material to consumers. Free TV Australia (sub. 129) noted the introduction of ‘catch-up’ online streaming services released by all of Australia’s major TV broadcasters, as well as strategies such as ‘fast-tracking’ overseas content for broadcast in Australia, and releasing content online prior to broadcast. Foxtel (sub. 115) detailed its ‘on-demand’ television program and film streaming service. Initiatives such as catch-up streaming services and ‘fast-tracking’ overseas content for broadcast are positive measures, but there remains scope to improve community awareness of mechanisms for accessing legitimate content. While online services such as YouTube, Foxtel’s Presto and Apple’s iTunes are relatively well known, only around one third of consumers are aware of services such as Pandora and SBS on Demand (TNS 2015). Community understanding and acceptance of the law Not all IP infringement is deliberate. Many Australians are keen to ‘do the right thing’, but lack an understanding of what activities are permissible. In respect of copyright, survey COMPLIANCE AND ENFORCEMENT OF IP RIGHTS 557 results show that 43 per cent of users are not confident that they know what is legal and what is not when downloading, streaming and sharing content through the internet (TNS 2015). Similarly in trade marks, firms conflate the registration of a business name with that of a trade mark, and inadvertently infringe. As put by the Australian Small Business Commissioner: In relation to IP disputes involving a small business, trade mark infringement is the most likely situation, particularly where a small business unintentionally infringes an existing IP right and the owner enforces that right. (sub. 101, p. 10) The Advisory Council on Intellectual Property (ACIP) (2006, p. 48) concluded that ‘ … the number of business names potentially infringing registered trade marks is likely to be in the tens of thousands’. Participants also pointed to the legal uncertainty around parallel import provisions for trade marks, making it impractical for importers to determine whether a good was likely to infringe. Another factor that can contribute to infringement is where the law is significantly out of step with community norms. For copyright, in particular, legal boundaries sometimes fail to reflect how people access content today. Australia’s pre-2006 copyright arrangements provide one of the best examples of the incongruity between community norms and the law. Format shifting — converting and transferring digital media from one format to another, such as a song from a CD converted and transferred to an mp3 player — was common practice yet technically a copyright infringement. Time shifting, or recording broadcast television to be watched later, was another salient example of normal practice, which the community expected to be able to do, yet infringed copyright. Some inquiry participants pointed to similar links between enforcement and the broader policy settings for patents. For example, Ausbiotech stated: A patent system that sets a relatively low bar for the grant of patent, resulting in enforcement litigation in which many or even most granted patents are found by the court to be invalid and therefore unenforceable, is less likely to be taken seriously by markets and potential infringers. It not only leads to uncertainty because many granted patents are known to exist, but are suspected to be invalid and unenforceable, it also wastes valuable resources of the government and companies in an often long and expensive process. This is also true for the Australian Innovation Patent system, whereby Innovation Patents are granted without undergoing examination and only require a very low threshold of inventiveness. (sub. 37, p. 3-4) The accessibility of enforcement options International comparisons suggest Australia has robust IP enforcement laws and institutions. For example, the US Chamber of Commerce’s Global Intellectual Property Centre assesses the legislative, regulatory and administrative strength of 30 countries’ IP systems. In assessing each country’s enforcement arrangements, the Centre measures the prevalence of IP rights infringement, criminal and civil procedures available to rights 558 INTELLECTUAL PROPERTY ARRANGEMENTS holders and the authority of customs officials to detain infringing goods at the border. Its 2016 assessment scored Australia’s enforcement system 4.66 out of 6, or 10th overall (figure 19.2). Figure 19.2 2016 Global IP Index of enforcement systems 6.0 5.0 5.7 5.5 5.5 5.5 5.4 5.3 4.9 4.8 Score (out of 6) 4.0 4.7 4.7 4.4 4.2 3.8 3.4 3.3 3.2 3.0 2.0 1.0 0.0 Source: Global Intellectual Property Center (2016, p. 41). Notwithstanding the normative basis of such rankings, evidence suggests Australia has all the necessary features of an effective enforcement system — strong laws and institutions. Yet research, evidence and information from participants suggest more can be done to improve the ability for rights holders to utilise Australia’s enforcement infrastructure. Many inquiry participants argued the enforcement mechanisms for Australia’s IP arrangements were inadequate, including the Australian Publishers Association (sub. DR435), Andrew Christie (sub. DR580), AusBiotech (sub. DR419) and the Department of Communication and the Arts (sub. DR154). Illustrative of many of the comments made, Professor Andrew Christie stated: Australia fails to deliver on the promise, implicit in the provision of a sophisticated system for the grant of IP rights, that there will be an accessible system for enforcement of those rights in the event they have been transgressed. (sub. 29, p. 2) The Commission heard about the difficulty some individuals have in enforcing IP rights, notably professional photographers (Australian Institute of Professional Photographers, trans., p 752). Many others raised the particular difficulties SMEs face when pursuing IP infringement, a situation reflected in the academic research (box 19.1). For example, IP Australia stated: COMPLIANCE AND ENFORCEMENT OF IP RIGHTS 559 … there remains a long-term issue with IP enforcement and whether IP right holders, particularly SMEs, are able to effectively make use of IP enforcement mechanisms. Anecdotal evidence indicates that smaller IP stakeholders in Australia find it too hard to enforce their IP rights due to the expense and uncertainty of IP litigation. This is particularly challenging for SMEs operating globally. (sub. 23, p. 19) Box 19.1 Smaller firms and IP enforcement Academic research supports the view that small- and medium-enterprises (SMEs) are infrequent users of Australia’s intellectual property (IP) enforcement system, particularly when compared to the experiences of larger firms. A 2005 study of Australian SMEs found the cost of enforcing rights was the number one factor inhibiting SME use of IP protection (IPRIA 2005). The study highlighted the cash flow issues faced by many SMEs, as well as the lack of management time and expertise, as being the primary reasons why SMEs failed to pursue enforcement against alleged infringements. In particular, SME’s considered the benefits of pursuing litigation as an enforcement option as highly uncertain. The study noted, unlike for many other commercial risks, Australia does not have a well-developed IP insurance market, which would otherwise assist SMEs to manage the risks of both their IP being infringed, and inadvertently infringing the IP of other firms (IPRIA 2005). Comparable overseas research on SME enforcement reveals similar experiences. In survey research conducted in the United Kingdom (UK) in 2010, approximately 25 per cent of SMEs had been involved in an IP dispute in the previous 5 years, with patent rights holders the most likely to be involved (Greenhalgh, Pitkethly and Rogers 2010). In the creative industries, around 20 per cent of SME copyright holders had been involved in a dispute. The UK study found SMEs tended to send letters or negotiate with the other party to resolve disputes, with only 13 per cent of those SMEs resolving their cases in court (Greenhalgh, Pitkethly and Rogers 2010). Around 80 per cent of respondents said they were happy with the result of their efforts. Again, the time and cost of litigation, as well as the management time and cost diverted in the pursuit of litigation, were cited as the primary reasons for avoiding court. And similar to Australia, very few SMEs held IP insurance to cover the cost of either defending an IP right or pursuing an alleged infringer. The cost of insurance was the primary barrier and many respondents were unaware of its existence. More recently, an evaluation of the UK’s Intellectual Property Enterprise Court (IPEC) showed increased use of the courts by SMEs following reforms to that court. While the number of cases varied from year to year, there was an increase of cases filed in the IPEC from 110 cases in 2010 to 272 cases in 2013, of which about 75 per cent were copyright or trade mark cases. Patent cases increased from 8 in 2010 to 17 in 2013, with a high of 27 in 2011. (Helmers, Lefouili and McDonagh 2015) For individuals, the Commission has previously reported that IP, together with matters of privacy, court processes and complaints against independent bodies, accounted for only 2 per cent of the unmet legal need in Australia involving civil disputes over individuals rights (PC 2014). 560 INTELLECTUAL PROPERTY ARRANGEMENTS Others thought Australia’s enforcement mechanisms were not as bad as others had claimed. For example, AusBiotech stated: Based on our understanding and experience, Australia sits somewhere in the middle of the spectrum between countries where IP enforcement is quicker and cheaper but still seen as broadly fair (an example being Germany) and countries where the court systems are dysfunctional to the point where enforcement of any legal rights are problematic. Leaving aside general differences among legal systems, Australia is probably towards the reliable of the spectrum in that IP rights are not generally regarded by the courts as inherently suspect or illegitimate … (sub. 37, p. 3) IP enforcement cases are often lengthy, complex and expensive, although this is not unique to IP cases. As the Commission noted in its 2014 inquiry into Access to Justice Arrangements, the time, cost and uncertainty of legal proceedings can act as a barrier to many businesses and individuals seeking to resolve disputes (PC 2014). That said, a number of features of IP cases may cause additional complexity and expense. Many IP enforcement cases (particularly for patents) require the rights holder to establish the validity and extent of their right before a case of infringement can be determined. Cases can involve competing firms and competing patents, each alleging the other infringes their patent, and seeking to have the other’s right declared by the court to be invalid. Patent and trade mark cases often rely on expert witnesses, technical information and surveys to establish the nature of and scope of rights. These add expense and delay to cases, especially where multiple experts are required or used. In its 2014 inquiry, the Commission (2014) noted evidence from the Australian Law Reform Commission that discovery processes in Federal Court proceedings typically represented up to 20 per cent of total litigation costs, but that IP cases typically had disproportionate discovery efforts. Several participants commented on the high costs of patent enforcement in particular: The costs of running a patent dispute is in the vicinity of $500,000 to a million dollars and it’s just not economical for a start-up business who is in their first two to three years of operation to go and take someone to Federal Court. (Alder, trans., p. 241) In essence, what we’re told is we have a Rolls Royce system called the Federal Court. You go there. The starting price will be $200,000 minimum. … Take it from there. $400,000, and then you might have the costs of the other side, et cetera, et cetera. (Christie, trans., p. 438) Many parties are deterred from taking enforcement action The time and cost of dispute resolution, along with uncertainty about the extent of infringement dissuade many from pursuing the enforcement of their IP rights. While around a third of patent holders experience some degree of patent infringement, only around 20 per cent of SMEs had sent a letter to the alleged infringer, the lowest cost step in enforcing an IP right (although this was higher than the 11 per cent of large companies that COMPLIANCE AND ENFORCEMENT OF IP RIGHTS 561 took the same action). In a follow-up telephone survey, the primary reasons for not sending a letter to an alleged infringer were because the cost was too high, the infringer was based overseas, or the infringement would be too difficult to prove (Weatherall and Webster 2010). Of those survey participants who had sent a letter, 37 per cent said the alleged infringer either stopped their use, or agreed to licence the invention from the patent holder, with around 61 per cent saying the alleged infringer only temporarily stopped copying, ignored the letter or questioned the validity of the patent. The low rate of sending cease and desist letters is surprising, and may suggest Australian businesses consider some level of patent infringement not worth pursuing. Many rights holders have difficulty quantifying their losses and the impacts of infringement (AIC 2008). In relation to online copyright infringement, each infringement is likely to be individually insignificant and represent only a small financial cost to a rights holder. But proving an infringement has occurred and claiming damages against thousands or millions of infringers is also unlikely to be feasible for many rights holders. Rights holders may also be deterred by public relations problems that have accompanied large-scale litigation campaigns in other jurisdictions (Suzor and Fitzgerald 2011). 19.3 A case to do more on copyright infringement? Online copyright infringement remains problematic and prominent, involving rights holders, intermediaries such as Internet service providers (ISPs), consumers and infringers. Few areas of IP policy attract as much media and policy attention. Unlike patents and trade marks, where alleged infringers can often be easily identified, copyright holders often do not have direct knowledge of who is infringing their copyright-protected works. This helps to explain why few civil cases have been brought against individuals that infringe and why, notwithstanding claims about the rate of online infringement, very few (if any) civil cases have been brought against individual infringers. Technology allows rights holders to identify the Internet protocol addresses and the ISP an infringer is using when their works are being illegally shared on peer-to-peer networks. However, accessing customer details to pursue an individual infringer requires either the consent of the ISP or a court order. This situation is further complicated by Australia’s safe harbour arrangements (discussed below). The Australian Communications Consumer Action Network commented on the recent approach to involving ISPs in deterring copyright infringement in Australia: Recent steps in copyright enforcement have attempted to make Internet Service Providers (ISP) play a central role in enforcement of copyright. The Copyright Notice Scheme Industry Code, for example, places significant obligations on ISPs in assisting the identification and enforcement against customers believed to be infringing copyright. At the time of writing, 562 INTELLECTUAL PROPERTY ARRANGEMENTS funding for this Code process is yet to be established; however there is an expectation from rights holders that ISPs should make a significant contribution. When coupled with the recently adopted Copyright Amendment (Online Infringement) Act 2015, there is a growing expectation that ISPs should have an unfunded role in enforcing copyright. (sub. 68, p. 7) However, the proposed Copyright Notice Scheme Industry Code is yet to be agreed, and ISPs and content providers reportedly informed the Australian Government in April 2016 they cannot agree on how the costs of the scheme should be shared between ISPs and rights holders (CNET 2016). The Australian Government’s Department of Communications and the Arts highlighted this ongoing tension between ISPs and copyright rights holders: Concerns have been raised by both copyright owners and users about the enforcement of copyright in Australia. Civil enforcement takes time and is a costly exercise. This operates as a disincentive for individual creators and smaller copyright owners to take action. Larger copyright owners are also unwilling to enforce their own rights in court where the alleged infringers are individual end users, preferring instead to seek the cooperation of internet service providers (ISPs) in deterring online copyright infringement. (sub. DR154, p. 4) Australia’s safe harbour arrangements Australia’s safe harbour scheme indemnifies ISPs from being held liable for alleged copyright infringement occurring over their networks, if they comply with certain requirements (box 19.2). Without the safe harbour scheme, ISPs could be found liable for the copyright infringement of their users (known as ‘authorisation liability’). The Copyright Act 1968 (Cth) (Copyright Act) establishes that copyright is infringed if someone ‘authorises’ another to copy or make available a copyright-protected work. Whether or not an ISP would be found to have infringed copyright is determined based on the facts of each case. In assessing the facts, the courts give regard to the ISPs power or ability to prevent the infringing act, the nature of the relationship between the parties, and whether reasonable steps were taken to prevent the infringement, including complying with any relevant industry codes. COMPLIANCE AND ENFORCEMENT OF IP RIGHTS 563 Box 19.2 How does Australia’s safe harbour scheme work? The Copyright Act 1968 (Cth) defines four categories of activities eligible for safe harbour protection: • Category A – acting as a conduit for Internet activities by providing facilities for transmitting, routing or providing connections for copyright material (for example telecommunications infrastructure providers and ISPs) • Category B – caching through an automatic process (for example, virtual private networks, proxy services or ISPs) • Category C – storing copyright material on their systems or networks (for example websites hosts or cloud storage providers) • Category D – referring users to an online location (for example, search engine linking). Australia’s safe harbour scheme was introduced in 2006 and applies to carriage service providers, defined in the Telecommunications Act 1997 (Cth) as ‘ … a person that supplies a listed carriage service to the public using a network unit owned by one or more carriers, or a network unit that has a nominated carrier declaration.’ Due to the adoption of the definition in the Telecommunications Act 1997 (Cth), in practice, only telecommunications infrastructure providers and ISPs in Australia enjoy the benefit of protection under the safe harbour system (category A), rather than all of the categories (as cited above) intended at the time the scheme was introduced. Safe harbour protection for carriage service providers is not automatic, and to benefit from protection carriers must satisfy various conditions under the Copyright Act, depending on the category of activity for which protection is sought. These conditions can include adopting and implementing policies, such as terminating the accounts of repeat infringers, if warranted by the scale of infringement. When Australia introduced the safe harbour scheme, adopting the definition used in the Telecommunications Act 1997 (Cth) was considered appropriate. At the time, the Internet was more ‘static’ with few streaming services — options for individuals to self-publish were more limited and cloud-hosting and streaming were not widespread. As noted by participants in this inquiry, today many businesses operating over the Internet undertake activities covered in the four categories listed in box 19.2, including cloud storage providers and search engines, but because they do not meet the definition of a carriage service provider, do not qualify for protection under the scheme. Google Australia highlighted the costs that come with Australia’s more narrow system: Australia is required by the Australia United States Free Trade Agreement (AUSFTA) to include all online service providers in the scheme. This is the position in the US and other countries where safe harbours have been introduced such as Singapore, South Korea, the UK and other EU countries. However, Australia has only given safe harbour protection to commercial ISPs (namely carriage service providers). Excluding other online service providers from the safe harbour scheme makes Australia a much more high-risk legal environment for hosting content when compared to countries that have safe harbour schemes with broad application. It also creates an uneven playing field for local 564 INTELLECTUAL PROPERTY ARRANGEMENTS innovations, placing them at serious commercial disadvantage when compared to commercial ISPs and global competitors. (sub. 102, p. 4) In 2011, the Attorney-General’s Department sought feedback on a proposal to expand the definition of online businesses protected by Australia’s safe harbour regime. In December 2015, the Department of Communications (having been afforded responsibility for copyright policy) sought further feedback on an exposure draft of amendments to the Copyright Act to broaden the safe harbour protections. APRA AMCOS argued the current separation under the scheme, between those providing telecommunications facilities and the broader category of online service providers, was the appropriate and balanced outcome the scheme intended: … extending the Safe Harbour Scheme to content providers will categorically disturb any notion of balance inherent in the Act. First it will subvert the policy behind the Scheme, which distinguishes between content providers and those who merely facilitate the communication of content. Secondly and more tangibly, it will reduce the incentives for such entities to enter into commercial agreements with copyright owners. APRA AMCOS does not understand Australia’s existing Safe Harbour Scheme to be inconsistent with Australia’s obligations under its international treaty obligations … the fact that the scope of the Australian Scheme may be more restrictive than is possible while remaining consistent with Australia’s obligations under the AUSFTA and TPP is not in itself a reason for reform. (sub. DR404, p 14) Several participants argued in this inquiry that Australia should not extend the protections offered by the safe harbour scheme to additional online service providers until the law around authorisation liability is strengthened. For example, Free TV Australia stated: Free TV’s view is that authorisation liability under the Copyright Act should operate in a manner that ensures that ISPs and other service providers are required to take reasonable steps to ensure that their online services are not being used for the purposes of copyright infringement. This is consistent with the purpose of ss 36 and 101 of the Copyright Act. Free TV therefore supports any necessary amendments to achieve technologically neutral authorisation liability provisions which operate effectively in the online environment, consistently with the purpose of those provisions and with Australia’s international obligations Free TV notes that in the absence of such amendments to the authorisation liability provisions it does not support any expansion of the safe harbour scheme to cover other service providers. This would simply lead to the safe harbour scheme being used as a mechanism to circumvent the authorisation provisions. (sub. 129, p. 10) News Corporation argued safe harbour and authorisation liability are ‘inextricably linked’ (sub. 119, p. 7), and changes to extend the safe harbour regime must be ‘balanced’ by changes to authorisation liability. Several participants argued the High Court’s decision in Roadshow Films Pty Ltd v iiNet Ltd meant authorisation liability in Australia no longer functioned as intended (box 19.3). COMPLIANCE AND ENFORCEMENT OF IP RIGHTS 565 Box 19.3 Roadshow Films v iiNet In this case, several copyright rights holders (predominantly film studios and distributors) alleged that Internet service provider iiNet had ‘authorised’ the copyright infringement of its customers by not passing on copyright infringement notices sent by rights holders, and by failing to suspend or terminate the accounts of alleged infringers. However, in a judgment upheld by the High Court, the Full Federal Court found iiNet was not liable for the alleged copyright infringement of its customers. The Court found that rights holders had not provided sufficient detail about how they had proven infringement occurred, and that given this lack of detail, it was unreasonable for iiNet to suspend or terminate customer accounts. Source: Roadshow Films Pty Limited v iiNet Limited [2011] FCAFC 23. However, concern with authorisation liability is not universal, as Telstra Corporation submitted: The Issues Paper notes that rights holders have increasingly demanded that ISPs take responsibility for copyright infringement over their networks. Part of the genesis of this demand is an argument that in the digital era, efficient and effective enforcement of copyright requires the extended liability of ISPs, or the extension of authorisation liability. We disagree with this proposition. We do not support any change to the law of authorisation liability, including any change to reverse the High Court’s decision in the iiNet case. (sub. 76, p. 13) A number of other participants also supported extending Australia’s safe harbour regime to protect other online service providers, as envisioned in the AUSFTA, such as Telstra Corporation (sub. 76), Google Australia (sub. 102), the Digital Industry Group Incorporated (sub. 111), the Australian Digital Alliance (sub. DR578), the Copyright Advisory Group to the COAG Education Council (sub. 429), and legal academics (sub. DR505). In particular, during inquiry consultations the Commission heard that many online service providers already operated under foreign safe harbour schemes which were broader than Australia’s, such as in the United States, and that as part of their global operations they already provided mechanisms for rights holders to have infringing content removed from their services. Subsequent to the decision in the Roadshow Films case, ISPs and copyright rights holders began negotiating a ‘three strikes’ copyright notice scheme. Under a draft of the scheme, ISPs would match the IP addresses of alleged infringers with their customer details, and pass on copyright infringement notices. An initial ‘education’ notice would alert the customer to the alleged infringement and provide information to the customer on legal alternatives for accessing copyright protected material. A second ‘warning’ would alert customers that they had previously received an ‘education’ notice and that a further infringement has been detected. A final notice would warn customers that a rights holder may initiate court proceedings to enforce their IP, including seeking customer details from ISPs. The draft scheme also included a mechanism for customers to challenge an infringement notice. 566 INTELLECTUAL PROPERTY ARRANGEMENTS However, in February 2016, the draft code was abandoned due to the expected costs of the scheme, and disagreements over who would be liable for such costs. Illustrative of the debate, Telstra Corporation submitted: We do not, however, believe that ISPs should be responsible for the costs associated with that enforcement. An IP rights holder must bear the cost of enforcing its own property rights. A general principle of IP enforcement across copyright, patents, trade marks, designs, etc. is that the IP owner bears the costs of enforcing their rights. The costs of any enforcement process may be recoverable from an infringer, following a successful enforcement action. A rights holder is the sole beneficiary of an enforcement action by way of a reduction in infringement and an uplift in its royalty revenues. (sub. 76, p. 21) In the Commission’s view, extending the coverage of Australia’s safe harbour regime, along the lines proposed in the Australian Government’s exposure draft amendments, will improve the system’s adaptability as new services are developed. Such an expansion is consistent with Australia’s international obligations and is an important balance to the expanded protections for rights holders Australia has accepted as part of its international agreements. As such this is a legislative amendment that should be made without delay. RECOMMENDATION 19.1 The Australian Government should expand the safe harbour scheme to cover not just carriage service providers, but all providers of online services. Blocking access to websites facilitating infringement In 2015, amendments to the Copyright Act were passed to allow rights holders to seek an order from the Federal Court requiring ISPs to block access to an overseas website if the website is primarily for facilitating copyright infringement. Before granting an order to block a site, the court must give regard to a number of factors, including: • the flagrancy of the infringement, or the flagrancy of the facilitation of the infringement • whether the site makes available or contains directories, indexes or categories of the means to infringe, or facilitates an infringement of, copyright • whether the owner or operator of the online location demonstrates a disregard for copyright generally • whether access to the online location has been disabled by orders from any court of another country or territory on the ground of or related to copyright infringement • whether disabling access to the online location is a proportionate response in the circumstances • whether it is in the public interest to disable access to the online location. COMPLIANCE AND ENFORCEMENT OF IP RIGHTS 567 Introduction of the ‘website take down’ scheme was controversial. At the time, concerns were raised with the lack of specificity in the scheme, and in particular, the lack of a definition as to what constituted the ‘primary purpose’ of a website, whether any infringement was ‘flagrant’ or not, and whether or not the website showed ‘disregard for copyright generally’. Concern was also expressed that rights holders might inadvertently (or inappropriately) target other websites providing socially useful functions, or capture virtual private network services. Others argued taking down websites would achieve little, because blocked websites would simply change their address and be quickly discoverable again. The Australian Publishers Association pointed to the UK as an example where such legislation has been successful: In the UK, the main redress to online copyright infringement has been through the High Court. The UK PA successfully brought a Blocking Order under Section 97A of the UK’s Copyright, Designs and Patents Act 1988 in May 2015. No damages in the case were awarded. The legislation only allows for an injunction against the ISPs to prevent access to the sites. Though the process is expensive and complicated both factors are reducing as more cases are brought. Indications show that the actions can reduce use of the infringing sites by up to 75%. (sub. 48, p. 22) While the Australian Communications Consumer Action Network highlighted research suggesting overseas approaches had only had limited success: Dutch research has attempted to measure the effectiveness of blocking access to torrenting website The Pirate Bay … The research did not find a strong indication of a long-lasting effect of website blocking in preventing piracy. This research confirmed the findings in other studies which found that legal action against file sharing often has an immediate effect, but this typically fades out after a period of six months as new sources for pirated content emerge. (sub. 68, p. 7) Research by CHOICE suggests that blocking access to websites is unlikely to significantly curtail infringement behaviour in Australia. CHOICE asked infringers what they would likely do if a website they use was blocked, with 54 per cent indicating they would find another website with infringing material, and 49 per cent suggesting they would use other tools to unblock or circumvent the restriction (CHOICE 2015). Timely and competitively-priced access is the key to reducing infringement Copyright rights holders have been engaged in an almost 20 year-long battle to eliminate online infringement internationally. However, pursuing court-based enforcement against websites, ISPs and individuals has not been successful, and possibly less successful in Australia than elsewhere. 568 INTELLECTUAL PROPERTY ARRANGEMENTS What has appeared to limit infringement is timely and competitively–priced access to content. Various survey studies consistently demonstrate that where copyright-protected content is available to consumers, the vast majority prefer paid, legal consumption. • In CHOICE’s (2015) study, around one third of respondents stated that they download or stream pirated movies or TV shows much less often since subscribing to streaming services. • Research by the Communications Alliance in 2014 found 43 per cent of respondents believed a market-based response was the best way to address online copyright infringement, as opposed to 19 per cent who believed stronger penalties were the answer (Communications Alliance 2014). • TNS Social Research (2015) found, in asking respondents the best way to reduce online copyright infringement: – 39 per cent said reducing the price of legal content – 38 per cent suggested improving availability – 36 per cent suggested releasing content in Australia at the same time it is released overseas. While there will always be some who continue to infringe, even in the presence of accessible legitimate options, they constitute a small group. For example, TNS found only 5 per cent of respondents indicated nothing would stop them accessing infringing material, and in the CHOICE (2015) study, only 11 per cent of respondents indicated that they infringed copyright because they did not want to pay. As discussed in chapter 4, new technologies hold the potential to provide more consumer licensing options and to address the poor data about rights holders and the types of licences available. For example, in the UK (and with the support of Australian copyright collecting societies), the Copyright Hub is seeking to provide a ‘digital copyright exchange’ service, allowing consumers to ‘right click’ on copyright protected content (including images, music, audio-visual and text content) and see what licences are available, at or close to zero transaction cost (sub. 6, p. 5). In the absence of these industry-led initiatives, other technological improvements (such as further advances in audio visual compression) will potentially accelerate infringing activity. In the Commission’s view, the case for further policy change or Government action on copyright infringement is weak. Rights holders, their publishers and other content providers are best placed to bring content to Australian consumers in a timely and competitively priced way. This approach is the most efficient and effective way to reduce online copyright infringement. COMPLIANCE AND ENFORCEMENT OF IP RIGHTS 569 FINDING 19.1 Timely and competitively-priced access to copyright-protected works is the most efficient and effective way to reduce online copyright infringement. 19.4 Improving dispute resolution Several options exist to lower the length, cost and complexity of IP dispute resolution in Australia. These include: • facilitating alternative dispute resolution (ADR) • lower cost court procedures. Alternate dispute resolution and IP advisory services ADR is an established process for resolving low-level disputes that do not warrant the time or cost of a court judgment, making it a potentially attractive option, especially for SMEs, to enforce their rights. ADR offers a number of advantages, including cost and time savings and confidentiality of outcomes, provided both sides are willing to constructively engage in the process. As IP Australia (sub. 23) noted in its submission, the Federal Court and Federal Circuit Court already have the ability to order parties to attend ADR processes, and some IP offices overseas provide tribunal or advice services to litigants. In 2010, the Advisory Council on Intellectual Property (ACIP) reviewed both post-grant patent enforcement strategies and enforcement of plant breeder’s rights, and recommended that an IP dispute resolution centre be established within IP Australia (ACIP 2010a, 2010b) ACIP noted that IP rights holders often did not know about the enforcement options available to them and struggled to source low-cost advice about the likelihood of successfully bringing an enforcement action. According to ACIP, an IP dispute resolution centre would facilitate ADR and provide a low-cost, transparent and easily accessible source of information and potential advice about enforcement. ACIP argued that two key elements of the dispute resolution centre would benefit rights holders, and SMEs in particular: • a register of experts to provide advice to rights holders considering pursuing a potential IP infringement dispute. IP experts could help parties to narrow and clarify the issues in dispute, and provide an external perspective on the potential outcome of a case, the extent of the rights involved, and the respective strengths and weaknesses of each party’s case • establishment of a tribunal to provide non-binding determinations and undertake arbitration according to commercial arbitration rules (ACIP 2010). 570 INTELLECTUAL PROPERTY ARRANGEMENTS In responding to the ACIP recommendations to establish an IP tribunal, the Australian Government (2013) questioned the likelihood that parties would use such a service if determinations were non-binding, and raised concerns about Australia’s constitutional prohibition on vesting judicial power in a non-judicial body. Specifically, the Government stated: On balance, the Government considers that the costs of a Patent Tribunal to the parties in a dispute, in particular the potential uncertainty created by such a body, outweighs the potential benefits at this time. (Australian Government 2013) During the course of this inquiry, IP Australia (sub. 23) and Professor Christie (trans., pp. 443–444) suggested that IP Australia could play a greater role in providing advisory services to assist business, as IP offices in other countries had done. The UK Intellectual Property Office, for example, provides fee-based services for mediation and non-binding opinions on infringement and validity of patents. The opinion service is designed to assist potential litigants to negotiate a settlement or to decide whether to proceed with full legal proceedings. The Commission notes that a number of ADR services already exist. Both the Australian Small Business and Family Enterprise Ombudsman (sub. DR403) and IP Australia provide information about dispute resolution and links to private providers on their websites. The Australian Small Business and Family Enterprise Ombudsman also has an online tool to help small business find the most appropriate low cost dispute resolution service. The World Intellectual Property Organization (WIPO) Arbitration and Mediation Center provides ADR services for both domestic and cross-border IP disputes. The Commission (2014) has also recently made recommendations to facilitate greater ADR in legal disputes. In respect of validity opinions, re-examination of a granted patent is already available by IP Australia for a fee of $800. IP Australia has recently consulted the public on extending re-examination processes to other registered rights (IP Australia 2015i). The Commission has not received evidence on the demand for additional services such as infringement opinions. As far as the Commission is aware, there is no impediment to IP Australia offering such services. Lower cost court procedures As the Commission has previously reported, courts provide a mechanism for creating, interpreting and applying the law. In doing so, they provide a public service that goes beyond the interests of the parties in a dispute. However, concerns remain that court-based dispute resolution is excessively resource intensive (PC 2014). As discussed above, participants in this review expressed concern over the expense of IP litigation and the negative affect that had on enforcing IP rights. COMPLIANCE AND ENFORCEMENT OF IP RIGHTS 571 Improved Federal Court procedures The majority of IP enforcement cases are brought in the Federal Court of Australia. This may be due to the processes of the Federal Court in determining commercial matters, an awareness that there is a greater degree of IP expertise among Federal Court judges, or because practitioners are generally familiar with Federal Court rules and procedures. Given cost is one of the biggest barriers to accessing court-based enforcement across all facets of the law, the Federal Court has broad discretion in taking steps to limit the costs faced by parties in a case. Rule 40.51(1) of the Federal Court Rules allows parties in a case to request an order from the Court on the maximum amount of costs each party can recover from the other. This allows parties to limit their liability in a case early in proceedings. However, it is not clear how often such orders are requested or granted by the Court. It may be that neither party in a court case, both of whom believe they have a reasonable prospect of success, would seek to limit the amount of costs they could claim from the other party. The Federal Court is currently implementing measures aimed at better managing the court’s caseload, while reducing the time and cost for parties seeking resolution to cases. Under the National Court Framework, the Federal Court is making three main changes to the way it works, including: • the creation of a national operations registrar, to better allocate the matters heard by each judge, aiming for a balanced workload across the Court • simplified practice notes for the specialised areas of the Court, providing clearer guidance to litigants, the legal profession and judges about the conduct and operation of cases • a managed docket system to support judges based on the character of matters, their workload and specialised areas of expertise (Federal Court of Australia 2015b). The Court recognises a degree of specialisation, with subject matter areas of the court organised into eight National Practice Areas, including an IP practice area. In October 2015, the Federal Court released further changes it intends to make to the conduct of IP cases. The proposed changes are aimed at ‘identifying the genuine issues in dispute between the parties at the earliest possible stage’, with a view to parties agreeing processes, expert evidence and all other matters that encourage speedy resolution of a dispute (Federal Court of Australia 2015a). These changes will encourage judges to take a much more active hand in determining the issues actually in dispute between the parties — a key element to shortening the length of trials and lowering their cost. 572 INTELLECTUAL PROPERTY ARRANGEMENTS Greater use of the Federal Circuit Court While the majority of IP enforcement cases are brought in the Federal Court of Australia, the Federal Circuit Court (FCC) also has jurisdiction to hear civil disputes under Australia’s copyright, trade mark, registered designs and plant breeder’s rights schemes. The FCC cannot hear patent disputes, unless the dispute is transferred from the Federal Court under section 32AB of the Federal Court of Australia Act 1976 (Cth). As noted on its website, the FCC (2014) was established: … to provide a simple and accessible alternative to litigation in the Federal Court of Australia and the Family Court of Australia and to relieve the workload of those courts. The FCC Act directs the Court to operate informally and to use streamlined procedures. This complements the Parliament’s initiatives to encourage people to engage in a range of dispute resolution processes. Judges in both the Federal Court and the FCC may order parties to attend mediation (a form of ADR) in the early stages of a case. Mediation provides a mechanism whereby an independent person assists parties to negotiate an agreement to resolve their dispute. The FCC refers about a third of IP cases to mediation. Hence greater use of the FCC may itself facilitate ADR. The FCC already includes many features similar to those of the UK’s lower cost IP list, the Intellectual Property Enterprise Court (IPEC). For example, the FCC may grant injunctive relief and damages of up to $750,000. The court may also specify at the outset the maximum costs that may be recovered. However, this is left to judicial discretion, rather than a mandatory cap on recoverable costs. Discovery is not allowable by way of right. Rather, it is only allowed where a judge declares that it is appropriate in the interests of the administration of justice. The FCC has established a number of specialist panels. The Commercial panel includes judges that specialise in copyright. While the FCC has a small claims procedure for other matters, there is currently no small claims procedure for IP disputes. In practice, few IP cases are brought in the FCC. 1 Participants suggested several reasons: the lack of judges with IP expertise and a resulting increased likelihood of decisions being reversed on appeal; that in practice, the FCC was no less formal or expensive than the Federal Court; and awarded costs were considerably less than the Federal Court scale (Australian Film/TV Bodies, sub. DR497; Australian Institute of Professional Photography, sub. DR387; FICPI Australia, sub. DR581, The Institute of Patent and Trade Marks Attorneys of Australia (IPTA), sub. DR562, Law Council of Australia, sub. DR490). In commenting on the role of the FCC, the Australian Screen Association stated: 1 52 IP cases were filed in 2014-15 (53 in 2013-14) of which 31 per cent were referred to mediation. The court delivered 7 written judgements on IP matters in 2014-15 (FCC 2015). COMPLIANCE AND ENFORCEMENT OF IP RIGHTS 573 Australia already has a lower cost Court (the Federal Circuit Court) that has equivalent functions in copyright and trade mark matters to the UK Intellectual Property Enterprise Court. More could be made of this Court to take on a range of civil and criminal enforcement matters with more streamlined processes and more realistic cost recovery for IP owners choosing that pathway. (sub. 43, p. 20) The Commission heard that in major cases, rights holders prefer to initiate court-based proceedings in the highest court possible. Initiating proceedings in the court with higher standing may in some cases reduce the total cost of litigation, particularly if a decision is likely to be appealed. There is potential for greater use of the FCC (including ADR processes within that court) to reduce the expense, and improve access to justice, particularly for SMEs, in relation to IP disputes. To address the key concerns that the Commission heard, particularly from legal advisors, this would at a minimum require development of greater IP expertise within the FCC (itself further facilitated through) a dedicated list and a more proactive approach to case management to narrow the issues in dispute. Looking abroad Many governments have established separate courts to deal with IP disputes. These take numerous forms, including specialised (trial or appeal) courts and specialist divisions or lists within a general court. The proclaimed advantages of specialised courts include development of judicial expertise, consistency and predictability of outcomes, and enhanced efficiency through specialised knowledge and rules and procedures suited to IP cases (Ma, Arievich and Karlhuber 2016). Some studies report a positive correlation between specialised IP courts and the efficient and effective resolution of IP cases (Zuallcobley et al. 2012). Australia has taken some steps in this direction, with the use of specialist IP panels in the Federal Court. Many participants highlighted the UK’s IPEC, effectively a dedicated IP list, as a model Australia should adopt. Submissions noted the role cost- and damages-capping plays in reducing the cost and uncertainty for rights holders pursing enforcement actions, especially among individuals and SMEs. The Commission has examined the UK IPEC in detail, including meeting with His Honor Judge Hacon (Presiding Judge of the IPEC) to get a working understanding of the court. (box 19.4). Experience thus far with the IPEC suggests that cases were typically being heard with costs below £40 000. Moreover, the active case management limits and clarifies claims early in the process and speeds up the litigation process. This, together with a two day cap on trials, also reduces expenses (Birss 2015; Helmers, Lefouili and McDonagh 2015a). 574 INTELLECTUAL PROPERTY ARRANGEMENTS Box 19.4 The United Kingdom’s Intellectual Property Enterprise Court The UK has experimented for 25 years with ways to reduce the cost and time to resolve disputes over IP rights, particularly for small and medium enterprises. The Patents County Court (PCC) was established in 1990 as an alternative to High Court litigation. However, several shortcomings prevented the PCC from achieving its aims, including the court’s inability to: • control the issues parties filed in a case, or to keep cases moving • place limits on the value of cases brought before it • modify procedural rules so the court operated differently to other courts. In practice, these limitations resulted in the PCC operating under the same rules and at the same cost as litigation in the High Court. In 2013 the PCC was reformed as a specialist list in the Chancery Division of the High Court, and renamed the Intellectual Property Enterprise Court (IPEC). The court has jurisdiction over a wide range of IP rights disputes including: claims and counterclaims of copyright, patent, trade marks and registered designs infringement; validity of patents, trade marks and registered designs; amendment of patents; declarations of non-infringement; misuse of trade secrets; and unjustified threats of infringement. Arguably the most significant reform to the Court is the introduction of caps on recoverable costs, and damages. The 2010 (UK) Jackson Review of Civil Litigation Costs identified the two largest risks to SMEs taking court-based action were: • the risk of paying a defendant’s costs if their case was unsuccessful • the potential damages payable. Changes to the court rules following the review means parties taking a dispute to the IPEC under its multi-track procedure are no longer liable for the other party’s costs in excess of £50 000. The court is also limited in the amount of damages it can award in a case, with a cap of £500 000. Another key aspect of the IPEC is active case management – the judge takes a much more ‘hands-on’ role at the case management stage to limit discovery, expert evidence and even the arguments to be made at trial. IPEC endeavours to hear trials in under two days and many cases are heard in a single day. A small claims track was also introduced into the IPEC with a cap on damages of £10 000. The small claims track is intended to provide a forum for resolving straightforward IP claims with low financial value and without the need for legal representation. The general principle that an unsuccessful party will pay the legal costs of the successful party does not apply in the small claims track. The small claims track hears similar cases to the multi-track except cases concerning patents, registered designs or plant variety rights. Although final injunctions and damages can be awarded, interim injunctions are not available under the small claims track. The limit on damages available in the IPEC provides a clear distinction between IPEC and other parts of the High Court. Parties in either the IPEC or another court can apply to transfer proceedings if they think the other court is more appropriate. Sources: Birss (2015), Helmers, Lefouili and McDonagh (2015), HM Courts & Tribunals Service (2014a, 2014b). COMPLIANCE AND ENFORCEMENT OF IP RIGHTS 575 The German system is another that is considered appealing to users due to the fast resolution of cases and relatively low cost of litigation. This is in part due to the parsimonious use of expert opinions and the absence of discovery procedures. The IPEC attempts to balance the efficiency of European civil law systems with the common law traditions of the UK, which are shared by Australia. According to Birss J, this is achieved by better targeting of discovery and expert witnesses to the key issues in dispute through more active case management: The two things which mark out the common law approach to a trial are documentary disclosure (discovery) and the cross-examination of witnesses. I believe they are both important components in a fair system of justice, but they can be awfully expensive. You could make cases much quicker and much cheaper by abolishing them both. The real trick in the IPEC is that you get to keep them and still have a system which is cheaper and quicker. (2015, p. 11) An evaluation of the reforms to the IPEC found an increase in the quantity of cases filed in the UK by SMEs, with increases in the number of claimants and the cases per claimant. Quantitative analysis and interviews with legal professionals (including judges, solicitors, barristers, patent and trade mark attorneys) suggested that the introduction of IPEC had greatly improved SME access to the court system, with an established upfront maximum liability for costs being the key attraction. It was also suggested that out-of-court settlements likely increased (Helmers, Lefouili and McDonagh 2015a, 2015b). As Birss J recounted: Now another thing the increasing case load demonstrates is this. Some people expressed serious doubts that there was actually a demand for any of this at all. They doubted that the court would attract any work because they doubted there was any work to attract. Litigants were apparently happy with their disputes being resolved just the way they were. The significant increase in cases shows that those doubts were wrong. There was … a constituency of parties who were not accessing the courts and who now are. (2015, p. 4) The small claims track of IPEC is even more streamlined, with case management done on the papers and trials completed in hours. It has dealt mainly with copyright and trade mark matters, becoming popular with professional photographers as a forum to enforce their rights (Birss 2015). This limited evidence appears to suggest the benefits to rights holders and users delivered by the IPEC derive more from its procedural rules than its status as a specialised list within the UK High Court. That is, it is not the specialisation per se that has improved access to dispute resolution for SMEs, but the ability of the court to minimise the financial exposure of rights holders pursuing a claim. Nevertheless, the specialist nature of the IPEC has been seen by some as an important contributor to its success (Birss 2015, IPTA, sub. DR562; Law Council of Australia, sub. DR490). 576 INTELLECTUAL PROPERTY ARRANGEMENTS Where to from here? While Australia has all of the necessary features of an efficient and effective IP enforcement system — a system of IP laws and courts — current practices mean that some IP rights holders still face barriers when seeking to resolve their disputes. The demand for a lower cost forum for resolving IP disputes is difficult to determine. There is evidence pointing to the cost of enforcement as being the biggest factor inhibiting SMEs’ use of IP protection (box 19.1). But other evidence suggests that many rights holders are not taking advantage of low cost options already available to them, such as cease and desist letters (section 19.1). Some researchers have suggested positive welfare effects from effective enforcement regimes, but that high costs are a barrier to justice. It is important to realize that the parties most likely to suffer from economic uncertainty are SMEs (small and medium-sized enterprises) and YICs (young innovative companies) since they typically lack the financial resources to fight extended litigation battles. The preferred litigation forum for these types of enterprises is a low-cost system that generates resolution quickly. (Harhoff 2009, p. 51) The character of IP litigation is also changing. Due to rising numbers of applications for rights and the presence of thickets, determining freedom to operate has become difficult or expensive for many businesses. This can result in inadvertent infringement. A lower cost system should reduce expenses for these businesses, though it may also increase the risk of a legal case being raised in the first place. Survey evidence from the UK suggests that the IPEC has also made it more practical to defend a case, rather than concede due to cost of litigation (Helmers, Lefouili and McDonagh 2015). The Commission (2014) has previously considered the costs and benefits of specialised courts within the Australian court system. Specialisation through a specialist court or specialised court lists has the potential to enhance court efficiency. Judges repeatedly hearing the same kinds of cases will potentially make quicker and more accurate decisions, and be more consistent over time. However, these benefits must be balanced against the potential rigidities such specialisation introduces. Judges benefit from being exposed to a wide range of legal cases and issues, particularly when cases (such as those involving IP) often cut across multiple areas of law (such as taxation, company and contract law), and while specialisation may lower the time and cost of original hearings, judges involved in hearing appeals are less likely to be specialists in the original subject matter. Critics of specialist IP courts have argued that specialisation can narrow a judges’ view so that developments in other areas of law are not integrated into IP cases, or that such courts become biased towards rights holders. This criticism has been most pronounced in relation to the United States Court of Appeals for the Federal Circuit which was established in 1982 with jurisdiction to hear appeals on patent cases. Any such bias is less likely to arise in an inferior court subject to more frequent review. Further, concerns that the IPEC would encourage inappropriate cases do not appear to have eventuated (Birss 2015). COMPLIANCE AND ENFORCEMENT OF IP RIGHTS 577 If elements of the UK IPEC were adopted in Australia, the question turns to the appropriate court to introduce such reforms. The Federal Court was noted by participants as a venue that commands respect with long standing expertise on IP matters, but it is a court with a ‘high cost DNA’. While the FCC appears to have relatively less experience with IP cases and may lack immediate breadth of expertise, it has a ‘low cost DNA’. The Federal Court is likely to remain the preferred venue for large companies, particularly for high value disputes. The reforms proposed by the Federal Court to improve efficiency of IP litigation are welcome. But they are unlikely to result in the level of savings to litigants that the UK IPEC has achieved through the imposition of mandatory caps on costs and damages, two-day trials and strict case management. In the Commission’s view, more should be made of the FCC. Established to hear cases more quickly and at lower cost, including through the use of ADR and mediation, evidence suggests the court is being underutilised when it comes to IP disputes. The FCC’s objectives are closely aligned with the operation of the IPEC — to operate informally and to use streamlined procedures. The Commission believes reforms to exploit the opportunities this court offers will improve access to justice for lower value IP matters and especially be of benefit to Australian SMEs. The Commission supports adopting reforms similar to those implemented in the UK through the IP Enterprise Court. This would be best done by introducing a specialist IP list in the FCC, together with a small claims procedure suitable for self-represented litigants. The Commission anticipates that these reforms will result in some additional demand for the Court’s services. The Court should be adequately resourced to ensure that an increase in its workload does not result in longer resolution times. Following implementation of these reforms, they should be reviewed after five years to ensure the arrangements are providing the expected advantages and are not having unintended effects. This review should also consider the effectiveness of the reforms proposed by the Federal Court for management of IP cases within the National Court Framework. 578 INTELLECTUAL PROPERTY ARRANGEMENTS RECOMMENDATION 19.2 The Australian Government should introduce a specialist IP list in the Federal Circuit Court, encompassing features similar to those of the United Kingdom Intellectual Property Enterprise Court, including limiting trials to two days, caps on costs and damages, and a small claims procedure. The jurisdiction of the Federal Circuit Court should be expanded so it can hear all IP matters. This would complement current reforms by the Federal Court for management of IP cases within the National Court Framework, which are likely to benefit parties involved in high value IP disputes. The Federal Circuit Court should be adequately resourced to ensure that any increase in its workload arising from these reforms does not result in longer resolution times. The Australian Government should assess the costs and benefits of these reforms five years after implementation, also taking into account the progress of the Federal Court’s proposed reforms to IP case management. Implementation issues — making it work While the government will need to consult further with the court and with users on the implementation of these reforms, the Commission believes there are several critical elements to include. Some of these may be implemented by adjusting procedures within the existing framework of the FCC. Others such as a mandatory cap on costs and the extension of jurisdiction to patents may require legislative change. Appointment or development of specialist IP judges Appointing judges with expertise in IP, or the opportunity and ability to develop such expertise, was seen as critical by some participants. For example, the Law Council of Australia said: The greatest current barrier to adoption of the FCC jurisdiction to date has been that cases are not managed and decided by a judge with IP expertise. (sub. DR490, p. 39) While the IPEC has a single judge based in London to hear cases, in the context of Australia’s federal system a small panel of experts may be more appropriate. Greater use of technology, such as tele- or video-conferencing, would also streamline the litigation process and overcome geographic issues (PC 2014). Extending the jurisdiction of the Federal Circuit Court to include patents Much of the concern regarding the costs of enforcing IP rights relates to patents. Extending jurisdiction to patents will provide an accessible venue for smaller-value patent claims. COMPLIANCE AND ENFORCEMENT OF IP RIGHTS 579 ACIP (2003b) has previously recommended that the Federal Magistrates Court (the predecessor to the FCC) hear patent cases. The Government preferred to wait and give further consideration in light of the Court’s experience with trade marks and designs, as patent cases were typically longer (Australian Government 2016c). The introduction of specialist IP judges, along with the observed experience in the UK, would help address previous concerns that patent disputes are ‘too complex’ for the streamlined procedures in the FCC. Procedural rules to reduce case costs The success of the UK IPEC has been largely attributed to the proactive and disciplined approach to managing cases taken by Judge Hacon and his predecessor, Judge Birss. Similar judicial leadership will be needed in Australia to curb the propensity of IP litigants to adopt gold-plated approaches. Justice Barker highlighted this difficulty in a recent Federal Court decision on innovation patents: … these parties are never satisfied unless they are continually turning stones. Certainly it is the case that no stone in the proceeding has been left unturned by them. Even after closing submissions were made, further submissions came in. 2 Strict case management by the FCC and early identification of issues in dispute will be critical to the success of these reforms. This has been assisted in the UK by limiting trials to one or two days. The FCC should adopt similar limits. Capping recoverable costs and damages The level of fixed costs should be set after careful consultation with court users and IP practitioners, but a maximum cap on recoverable damages and costs should be mandated. Consistent with IPEC, recovery of costs above the cap should be permissible only in exceptional circumstances, such as unreasonable behaviour. A cap on damages (together with a cap on costs) allows parties to know their total financial exposure to a dispute. This provides certainty to SMEs in deciding whether to pursue action. Consideration should be given as to whether the current limit on damages in the FCC is appropriate for IP cases. The Australian Film/TV Bodies (sub. DR497) argued against a cap on damages, given the actual size of damages is itself uncertain, mostly due to the potential awarding of ‘additional damages’ in IP cases. However, the FCC already deals with additional damages for copyright and trade mark matters within its existing limit on damages. 3 The Federal Court would continue to be the appropriate forum where the rights holder is seeking damages beyond the limit. 2 Australian Mud Company Pty Ltd v Coretell Pty Ltd (No 7) [2016] FCA 991 3 For example, in Vertical Leisure Limited & Anor v Skyrunner Pty Ltd & Anor [2014] FCCA 2033, an award of $300,000 was made by way of additional damages (FCC 2015). 580 INTELLECTUAL PROPERTY ARRANGEMENTS A small claims track A separate small claims procedure, such as that provided in the IPEC, will provide an informal forum for low value cases. The small claims track would be most appropriate for those rights over which the FCC currently has jurisdiction, namely copyright, trade marks, registered designs, and plant breeder’s rights. Given the greater complexity of patent cases, they are less likely to be suitable for small claims procedures. Like the IPEC small claims track, interim injunctions should not be available. 19.5 Better enforcement of rights abroad Rights holders’ experience with enforcement overseas is a different story. The Department of Foreign Affairs and Trade commented on the difficulty Australian businesses have enforcing IP rights overseas: But, despite the wide membership of global intellectual property agreements, inadequate intellectual property protection and enforcement remains a problem for Australian exporters and investors. Further, and more generally, deeper cooperation and coordination between nations in the area of intellectual property can deliver increased efficiencies and greater certainty in trade and investment relationships. (sub. 65, p. 9) AusBiotech suggested the best many Australian businesses could hope for overseas was to be treated no differently to domestic rights holders: It is inevitable that there will be variation in the ease with which IP rights of Australian firms can be enforced internationally across regions and/or countries. All Australia can reasonably ask for is that, in the context of an overseas jurisdiction, Australian firms are not discriminated against and are treated equally with local firms (and that the laws of the overseas jurisdiction comply with international treaty obligations). Australia should continue to actively support international efforts encouraging transparency and strong enforcement of valid IP rights internationally. (sub. 37, p. 3) Lengthy trade mark opposition processes are among the difficulties that have been faced by firms doing business in China. Until recently, opposition periods of ‘three to four years’ and attempts to cancel the mark of squatters of ‘five to seven years’ have been the norm (IP Australia 2006, p. 3). Recent changes to Chinese trade mark law have reduced the period over time over which opposition and cancellation of marks occurs (Lam and Chatterton 2013). However, it is still common for legitimate companies to experience trade mark actions being brought against them when trying to enter the Chinese market (China-Australia Chamber of Commerce, sub. 50). Forewarned is forearmed The Commission is not in a position to make recommendations that other jurisdictions change the way they use and enforce IP rights. But the Commission does note that these COMPLIANCE AND ENFORCEMENT OF IP RIGHTS 581 issues can be taken up bilaterally with the relevant countries and in international fora where parties come together to co-operate on cross border aspects of IP enforcement. Australian businesses would also benefit from more guidance on how to navigate foreign enforcement systems. Shelston IP (sub. DR483) called for a regional enforcement process whereby parties could seek non-binding opinions on the infringement or validity of patents. It suggested a multi-jurisdictional body providing non-binding opinions could encourage greater filings and exports into those jurisdictions where it provides an alternative to costly patent enforcement through the courts. The WIPO has recently extended its network of External Offices, establishing an office in Beijing in mid-2014. These offices provide dispute resolution services such as arbitration, mediation and expert determinations. The China-Australia Chamber of Commerce (sub. 50) raised a range of China-specific proposals, including IP best practice checklists, updates on recently decided IP cases, and providing links to relevant Chinese government department websites and contacts. These proposals point to a need to improve information and raise awareness of foreign IP and enforcement systems among Australian business. The Government has recently announced the appointment by IP Australia of an IP Counsellor in China, to give Australian businesses greater confidence to commercialise their IP. The Counsellor will support Australian businesses in China by: • providing guidance on local Chinese IP arrangements and related matters such as registration of rights, and enforcement • raising awareness of IP through business outreach • liaising with the Chinese Government and stakeholders about the Chinese IP environment. Industry-led initiatives to raise the visibility and understanding of legal rules in other jurisdictions would also help to better inform business and improve the efficiency of IP transactions. 582 INTELLECTUAL PROPERTY ARRANGEMENTS A Conduct of the inquiry The Commission received the terms of reference for this inquiry on 18 August 2015. It subsequently released an issues paper on 7 October 2015 inviting public submissions and highlighting particular matters on which it sought information. In total, 620 public submissions were received and placed on the inquiry website. A list of public submissions is contained in table A.1. In addition, the public were able to make brief online submissions, which are listed on the inquiry website. During the course of the inquiry, the Commission held informal consultations and roundtable discussions with governments, regulatory bodies, peak industry groups in the government sector, as well as a number of private and government organisations. Tables A.2 and A.3 list these participants. The Commission would like to thank all those who contributed to this inquiry. CONDUCT OF THE INQUIRY 583 Table A.1 Public Submissions Participants Submission no. Adam Goodrum Ai Group DR598 DR582 AIPPI Australia DR551 Alder IP Alexandra Lavau 140 DR255 Alice Godwin DR305 Alison Clifford Alison Croggon DR292 DR381 Alison Lyssa Alison Stegert DR493 DR474 Allayne Webster DR254 Allen & Unwin DR473 Allison Tait Alphapharm Pty Ltd Amanda Curtin DR294 93, DR584 DR438 Amanda Francey Amanda Graham Amanda Holohan Amanda Stewart DR230 DR540 DR307 DR353 Amgen Australia Pty Ltd Amie Johnstone Andrea Smith DR337 DR372 DR183 Andrew Christie Andrew Dalgleish Andy Griffiths Angela Daly DR580 DR201 DR456 DR392 Angela Sunde Angelo Loukakis DR210 DR513 Ann Villiers DR214 Anna Funder Anne Morgan DR302 DR399 Annie White DR226 Anthony Pisani DR195 Apolonia Niemirowski Apple Inc DR600 DR554 APRA AMCOS 113, DR404 Ariel Katz Aristocrat Leisure Limited DR605 139,DR572 Arnold Bloch Leibler Arts Law Centre of Australia 46, DR349 117, DR536 Ashley Kalagian Blunt Association of Heads of Independent Schools of Australia DR264 136 (continued next page) 584 INTELLECTUAL PROPERTY ARRANGEMENTS Table A.1 (continued) Participants Submission no. Association of American Publishers (AAP) Association of Australian Musicians, Australian Independent Musicians Association and Australian Guild of Screen Composers Association of Liquor Licensees Melbourne Inc Aura Parker DR338 DR512 62, DR544 DR289 AusBiotech Ltd 37, DR419 Australasian Music Publishers Association Limited (AMPAL) Australasian Open Access Strategy Group DR535 DR431 Australia Children’s Television Foundation DR548 Australia Council for the Arts 105, DR553 Australian Association for the Teaching of English (AATE) Australian Blindness Forum 110, DR401 DR390 Australian Booksellers Association Australian Broadcasting Corporation (ABC) Australian Centre for Intellectual Property in Agriculture (ACIPA) Australian Chamber of Commerce and Industry DR466 107, DR604 7 70, DR569 Australian Communications Consumer Action Network (ACCAN) Australian Competition and Consumer Commission (ACCC) Australian Copyright Council 68 35,143,144, DR603 36, DR543 Australian Curriculum, Assessment and Reporting Authority Australian Design Alliance Australian Digital Alliance Australian Directors Guild 94 DR619 108,141, DR578 DR185 Australian Directors Guild (ADG) and Australian Screen Directors Authorship Collecting Society Australian Fair Trade and Investment Network (AFTINET) 10 DR471 Australian Fiction Writers DR205 Australian Film/TV Bodies DR497 Australian Grain Technologies Pty Ltd 15 Australian Grape and Wine Authority 72, DR527 Australian Industry Group Australian Information Industry Association (AIIA) 60 89, DR379 Australian Institute of Aboriginal and Torres Strait Islander Studies (AIATSIS) Australian Institute of Professional Photography Australian Libraries Copyright Committee Australian Literacy Educators' Association Australian Literary Agents' Association (ALAA) DR583 DR387 125, DR602 DR449 74, DR427 Australian National Data Service Australian Nuclear Science and Technology Organisation (ANSTO) DR314 17 Australian Policy Online (APO) Australian Property Institute NSW Division DR444 95 Australian Publishers Association Australian Recording Industry Association (ARIA) 48, DR435, DR614 122, DR499 (continued next page) CONDUCT OF THE INQUIRY 585 Table A.1 (continued) Participants Submission no. Australian Screen Association Australian Seed Federation 43 42 Australian Society of Authors Australian Subscription Television and Radio Association (ASTRA) 121, DR343 85, DR530 Australian Technology Network of Universities (ATN) 91, DR484 Australian War Memorial DR563 Australian Writers Guild and Australian Writers Guild Authorship Collecting Society Australian Writers’ Guild 53 DR549 Authentic Design Alliance DR 588 Avid Reader Barbara Hermon & John Hermon DR380 DR590 Benjamin Stevenson Bernadette Foley Bernd Winter DR179 DR259 DR506 Beth Spencer DR518 Bettina Deda Bill Concannon Bindu Narula DR330 DR568 DR323 Black Inc Blanche d'Alpuget Books Kinokuniya Boyer Educational Resources 9 DR318 DR485 58 Brolly Books Brook Martin Brunswick Bound DR279 DR176 DR462 Building Designers Association of Victoria Inc Buro North 5 DR253 Business Council of Australia Business SA Calum Drummond Cambridge University Press Australia 59, DR587 DR389 DR296 22, DR421 Cameron Canvas Pty Ltd 1 Cameron Raynes Canberra Society of Editors DR258 DR319 Candice Lemon-Scott DR209 Caroline Magerl DR271 Cary J Lenehan Cass Moriarty DR609 DR430 Cassandra Brooke DR313 Cate Whittle Cath Ferla DR231 DR369 Cecilia Clark DR280 (continued next page) 586 INTELLECTUAL PROPERTY ARRANGEMENTS Table A.1 (continued) Participants Submission no. Cengage Learning Australia Pty Ltd Centre for Law and Genetics, University of Tasmania DR189 61 Cheryll Johns * in confidence content omitted China-Australia Chamber of Commerce DR371 50 Choice 26, DR269 Chris Bell DR507 Chris Dent Chris Prescott 30, DR286 DR470 Chris Snow 127, DR324 Chris Tugwell Christina Booth DR251 DR252 Christine Balint * in confidence content omitted Christine Bongers Christopher Heathcote DR180 DR410 DR197 Christopher Jones DR412 Christopher Wilkinson Claire Boston Coalition of Major Professional and Participation Sports (COMPPS) DR359 DR181 DR312 Colin Thompson Commercial Radio Australia Communications Alliance Ltd Complementary Medicines Australia DR299 103, DR303 DR374 DR534 Copyright Advisory Group to the COAG Education Council Copyright Agency Copyright Agency Viscopy with APRA AMCOS, Foxtel, News Corp Australia, PPCA, Screenrights DR429 47, DR510 134 Copyright Hub Foundation London Copyright Licensing Limited New Zealand Cosmic Enterprises 6 DR239 109 Costco Wholesale Australia Pty Ltd 31 Council of Australian University Librarians (CAUL) Creative Commons Australia Cristina Cappelluto 120, DR426 DR504 DR514 CropLife Australia CSIRO Cult Design 25, DR561 126,DR575 DR596 Currency Press Curtis Brown (Australia) Pty Ltd DR178 DR285 D S Craig Dairy Australia DR365 38, DR247 Danielle Clode DR215 (continued next page) CONDUCT OF THE INQUIRY 587 Table A.1 (continued) Participants Submission no. Danielle Freeland Darren Goossens DR228 DR165 David Greagg David Harrison DR245 DR589 David Merrylees and Katherine Downey DR386 David Trubridge Ltd David Webber Davies Collison Cave DR573 40, DR447 DR417 Deborah Abela DR156 Deborah Gleeson Deborah Kalin DR616 DR297 Dee White Delegation of the European Union to Australia Delia Falconer DR170 DR495 DR361 Democratic Pirates Australia DR162 Denise Kirby Department of Communications and the Arts Department of Foreign Affairs and Trade DR262 DR154 65 Department of Health Department of Immigration and Border Protection Department of Industry, Innovation and Science Design Institute of Australia 84 52 DR615 131, DR594 Digital Industry Group Incorporated (DIGI) Digital Rights Watch Dimitrios Eliades 111, DR528 DR414 DR579 Dimity Powell Donald Richardson DR211 138 Donna Best DR327 Dr Deborah Gleeson 128 Dr Kathryn Fox Dr Matthew Rimmer DR166 145,146,147,148 Dreamtime Public Relations Pty Ltd 2 Dymocks Eddy Krajcar DR613 DR541 Edwina Preston DR335 Eleanor Curtain Publishing Pty Ltd 69 Electronic Frontiers Australia Elise McCune 114 DR227 Elizabeth Fensham DR277 Elizabeth Hatton Ellen Tyrrell DR356 DR433 Emma Viskic DR326 (continued next page) 588 INTELLECTUAL PROPERTY ARRANGEMENTS Table A.1 (continued) Participans Submission no. Envato Pty Ltd Elsevier 80 DR585 Eve Lamb Federal Chamber of Automotive Industries DR212 88 FICPI Australia DR581 Fiona Hazard DR442 Fiona McCallum Fiona Wood DR288 DR370 Firefly Education DR436 Foxtel Fraser Old 115, DR550 DR161 Fraunhofer-Gesellschaft Free TV Australia Freedom Publishers Union DR452 129, DR570 DR163 Garth Nix DR405 Generic and Biosimilar Medicines Association (GBMA) Geoff Hook Geoffrey Atherden 67, DR396 DR577 DR339 Getty Images Gilbert and Tobin Lawyers Glynis Traill-Nash Good Design Australia DR491 96, DR565 DR440 DR599 Google Australia Pty Ltd Gordon Tait Graham Seal Greg Holfeld Griffin Press Griffith University Hachette Australia Haese Mathematics Pty Ltd Harlequin Enterprises (Australia) Pty Ltd HarperCollins Publishers Australia Hazel V J Moir Helen O'Neill 102, DR523 DR597 DR218 DR191 49 106 41, DR393 DR306 45 56, DR418 130, DR295 DR334 Helena McAuley DR333 ICMP (International Confederation of Music Publishers) DR344 IFPI Independent Scholars Association of Australia Inc DR503 DR358 Inga Simpson DR409 Insight Publications Intel Corporation DR282 66 Interactive Games and Entertainment Association Ltd 77, DR437 (continued next page) CONDUCT OF THE INQUIRY 589 Table A.1 (continued) Participants International Association of Scientific Technical and Medical Publishers (STM) International Confederation of Music Publishers (ICMP) International Confederation of Societies of Authors and Composers International Federation of Pharmaceutical Manufacturers and Associations International Generic and Biosimilar Medicines Association Submission no. DR366 32 DR268 87, DR391 DR526 International Publishers Association 57, DR375 International Trademark Association IP Australia 20, DR516 23, DR612 Isabella Alexander, Catherine Bond, Robert Burrell, Michael Handler, Graham Greenleaf, Dianne Nicol, Jane Nielsen, Kimberlee Weatherall iSignthis DR505 DR443 J Albert and Son Pty Ltd DR193 Jackie French Jaclyn Moriarty Jacqueline Harvey DR382 DR517 DR315 Jaime Lumsden Kelly James Guest James Knight Jane Sullivan DR532 DR394 DR320 DR261 Jane Turner Goldsmith Janeen Brian Jarakad Pty Ltd DR276 DR219 78 Jay Sanderson Jennifer Scoullar Jenny Barry Jeremy Fisher DR208 DR186 DR332 18 Jessica White DR331 Jill B Bruce Jo Jette DR237 DR363 Jo McClelland Phillips DR340 Joanna Burns Jodie Wells-Slowgrove John Chapman DR244 DR217 DR158 Jolyon Sykes Josephine Moon J.R Poulter DR476 DR192 DR601 Judith Daley Julie Haydon DR478 DR250 Julie Van Mil * in confidence content omitted June Perkins DR455 DR284 Kane Waterworth Karen Collum Karen Foxlee 135 DR384 DR586 (continued next page) 590 INTELLECTUAL PROPERTY ARRANGEMENTS Table A.1 (continued) Participants Submission no. Katherine Scholes Kathryn Apel DR350 DR509 Kay Crabbe * in confidence content omitted Kaye Baillie DR360 DR164 Kim Rackham DR234 Kim Wilkins DR311 Kimberlee Weatherall, Isabella Alexander and Michael Handler Kinokuniya Book Stores of Australia Pty Ltd 99 DR508 Kirsty Murray DR373 Kristin Weidenbach Krys Saclier DR300 DR317 Kurt Johnson Lateral Economics Laurine Croasdale DR378 DR187 DR283 Law Council of Australia 64, DR490 Law Institute of Victoria Leanne Albers Leanne Bridges DR558 DR278 DR475 Leeza Baric Lia Weston Lincoln Law Linda Jaivin DR576 DR263 DR309 DR458 Linda Wells Linux Australia Loretta Re DR220 DR488 DR243 LoveOzYA Lynn Savery DR182 DR620 MacLean's Booksellers DR355 Macmillan Publishers Australia Pty Ltd * in confidence content omitted DR552 Macmillan Science And Education Australia Malla Nunn 16 DR267 Mandy Foot DR249 Maree Kimberley Margaret Hamilton DR406 DR167 Margareta Osborn DR342 Margi Prideaux DR434 Margot Hilton Margret Best DR190 DR368 Marie Alafaci DR229 Mark A Summerfield Mark Isaacs DR388 DR411 Mark Smith DR395 (continued next page) CONDUCT OF THE INQUIRY 591 Table A.1 (continued) Participants Submission no. Mark Swanton Markus Zusak DR287 DR232 Martin Goetz Mary Abel DR160 DR290 Mary Bayley DR464 Mary-Lou Stephens DR272 Matthew Rimmer McGraw-Hill Education (Australia) DR611 14 McKay Innovation DR520 Media, Entertainment and Arts Alliance Medicines Australia DR559 44, DR529 Melanie Hill Melanie Joosten Melbourne University Publishing DR408 DR257 DR240 Melbournestyle Books DR216 Melina Marchetta Melissa Sargent Mem Fox DR402 DR310 DR413 Meredith Appleyard Meryl Harris Michael Adams Michael Barr DR241 DR329 DR222 DR521 Michael Bauer Michael Dwyer Michael Robotham DR291 DR496 DR207 Michelle Diener Michelle Michau-Crawford DR155 DR450 Michelle Worthington DR481 Microsoft DR420 Miriam Sved * in confidence content omitted Monica McInerney DR531 DR377 Music Australia Music Rights Australia Pty Ltd Must do Brisbane.com Pty Ltd Nadia Wheatley Name withheld * in confidence content omitted Natasha Lester National and State Libraries Australasia National Archives of Australia National Association for the Visual Arts (NAVA) DR606 51, DR407 134 DR428 33 DR364 55 DR525 142, DR467 (continued next page) 592 INTELLECTUAL PROPERTY ARRANGEMENTS Table A.1 (continued) Participants Submission no. National Copyright Unit, COAG Education Council National Film and Sound Archive of Australia (NFSA) 97 DR480 National Tertiary Education Union (NTEU) News Corp Australia 24 119, DR546 New South Wales Department of Justice DR610 Nick Earls DR266 Nicky Luckie Nicola Tierne DR354 DR235 Nikki Gemmell DR242 Nikos Koutras Nomadic Solutions Pty Ltd 132 28 Nova Weetman NSW Department of Justice NSW Writers' Centre DR472 39 DR460 Office of the Australian Small Business and Family Enterprise Ombudsman DR403 Office of the Australian Small Business Commissioner Open Source Industry Australia Ltd Oxford University Press Australia and New Zealand 101 21, DR486 8, DR448 Pamela Freeman Pamela Rushby Pantera Press Pat Lowe DR152 DR199 DR424 DR168 Paul Harpur Paul Sharrad Paula Weston DR174 DR172 DR347 P B Mitchell Pearson Australia DR618 63, DR398 Penguin Random House Australia 75, DR498 Peter Carnavas DR246 Peter Donoughue Peter Jaszi, Michael Carroll, Sean Flynn, Michael Palmedo, Kimberlee Weatherall and Ariel Katz Peter King DR149 DR275 Peter Murphy Peter Taylor Peter Treloar DR177 DR376 DR479 Pfizer Australia PGG Wrightson Seeds (Australia) Ltd Phillip L'Estrelle Phonographic Performance Company of Australia (PPCA) Pirate Party Australia Polly Seidler Printing Industries Association of Australia 11 83, DR298 82, DR547 DR157 123, DR511 DR522 DR555 90 (continued next page) CONDUCT OF THE INQUIRY 593 Table A.1 (continued) Participants Submission no. Prof Andrew Christie Prof Kathy Bowrey 29 86 Prof Robert Burrell and Associate Prof Michael Handler * in confidence content omitted Professional Historians Association NSW & ACT Qualcomm Incorporated Queensland Law Society 92 3, DR469 DR345 116, DR567 R.I.C. Publications Pty Ltd Rachael Blair 12 DR188 Rachel Walsh DR221 Raelyene Kutzer Readings Pty Ltd DR328 DR441 Rebecca Borona Rebecca Giblin and Kimberlee Weatherall Rebecca Regan-Coe DR385 DR524 DR542 Reinvent Australia DR538 RIC Publications Richard Bourke Richard Griffiths DR336 DR159 DR492 Richard Newsome Riverbend Books Roanna Gonsalves Robert Banks DR463 DR461 DR519 DR487 Robert Hind Robert J Bouvet Robert Macklin DR445 DR273, DR608 DR539 Robert Watkins Robyn Cadwallader DR400 DR304 Rod Black DR322 Romance Writers of Australia Inc DR416 Rosanne Hawke Rosetta Books DR465 DR537 Ruth Clare DR260 Ruth Skilbeck Sally Heinrich DR457 DR489 Sam Sochacka DR308 Samantha Forge 79 Sandi Wallace Sandra O'Grady DR153 DR236 Sandy Fussell DR357 Sarah Neilson Sarah Sandford-Bell DR545 DR422 Special Broadcasting Service (SBS) 112 (continued next page) 594 INTELLECTUAL PROPERTY ARRANGEMENTS Table A.1 (continued) Participants Scholastic Australia Pty Limited Screen Producers Australia Screenrights Scribe Publications Pty Ltd Submission no. DR213 DR556 118, DR454 DR574 Sean Cotcher DR500 Secret Girls Business Partnership DR515 Shane W Smith Shannon Stein DR184 DR451 Shelly Unwin DR362 Shelston IP Sherry Landow DR483 DR482 Sherryl Clark Sheryl Gwyther Society of Children's Book Writers and Illustrators (SCBWI) DR223 DR204 DR367 Society of University Lawyers 98 Sony Music Entertainment Australia Pty Ltd Sonya Bates Sophie Hamley 124 DR423 DR351 Sophie Masson Sophy Williams Space Furniture Special Broadcasting Service (SBS) DR150 DR248 DR595 DR502 Spinifex Press Stefano Boscutti Stephanie Alexander 19, DR397 DR341 DR607 Stephanie Smith Steven Hocking DR224 DR233 Sue Whiting DR281 Sulari Goonetilleke (Gentill) DR203 Susanna Elliott-Newth Susanna Rogers DR169 DR196 Suzanne Leal DR439 Swinburne University Tania McCartney DR557 DR265 Telstra Corporation Limited 76, DR316 Terri Winter DR198 Text Publishing Company The Author People DR346 DR206 The Chamber of Arts and Culture of Western Australia DR533 The Federation Press Pty Ltd The Heffernan Group DR173 DR194 (continued next page) CONDUCT OF THE INQUIRY 595 Table A.1 (continued) Participants The Institute of Patent and Trade Mark Attorneys of Australia The Software Alliance (BSA) Submission no. 73, DR562 DR446 The Walter and Eliza Hall Institute of Medical Research Thomas Bailey-Smith DR571 DR348 Tim Sinclair DR301 Tim Winton DR325 Tom Fereday Tom Skeehan DR591 DR592 Tomek Archer DR593 Toni Jordan Tony Healy DR256 DR202 Tracey Jackson Trayon Campers Trish Anderson DR383 4 DR270 Tyrion Perkins DR293 Universities Australia University of Melbourne University of Queensland 71, DR453 100, DR560 DR200 University of South Australia University of Sydney University of Technology Sydney University of Wollongong 13, DR432 104, DR566 DR564 54 UNSW Press Ltd US Chamber of Commerce Viacom International Media Networks Australia (VIMN) 27 DR415 DR468 Victorian Curriculum and Assessment Authority Village Roadshow Limited DR459 DR352 Virginia Rigoni DR617 Vivienne Kelly DR321 WeCreate Wendy Blaxland DR238 DR501 Wendy Orr DR151 Wild Pure Heart Wiley Australia DR477 DR494 WiseTech Global DR274 Working Title Press 81 Writers Victoria Xavier Brouwer DR175 DR171 Yvonne Low DR225 596 INTELLECTUAL PROPERTY ARRANGEMENTS Table A.2 Stakeholder consultations Participants ACT Attorney-General’s Department Australian Competition and Consumer Commission Australian Small Business Commissioner Department of Communication and Arts Department of Education and Training Department of Foreign Affairs and Trade Department of Health Department of Industry, Innovation and Science Department of Immigration and Border Protection Department of Treasury Dr Hazel Moir Institute of Patent and Trade Mark Attorneys IP Australia Office of the Chief Scientist Professor Ian Hargreaves CBE Telstra Corporation Limited The Australian Property Institute Therapeutic Goods Administration Winemakers Federation of Australia New South wales Adam Lieberman APRA AMCOS Aristocrat Technologies Australian Centre for Intellectual Property in Agriculture Australian Law Reform Commission Choice Copyright Agency Ltd Copyright Council Costco Australia Department of Education Facebook Australia/New Zealand Google Australia Justice Annabelle Bennett Justice Nye Perram Law Council of Australia Nestlé Australia Ltd Policy Australia (continued next page) CONDUCT OF THE INQUIRY 597 Table A.2 (continued) Participants South Australia Kym Anderson Paula Zito Wine Australia Queensland Queensland Law Society William Van Caenegem Victoria Australian Competition and Consumer Commission Ambercite Association of Australian Medical Research Institutes Bill Ferris AC Deborah Gleeson Department of State Development, Business and Innovation (Victorian Government) Dr Alan Finkel AO Intellectual Property Research Institute of Australia Justice Middleton Michael O’Bryan Tony Kitchener (AKGK Pty Ltd) Walter & Eliza Hall Institute of Medical Research International Competition Bureau Canada (Teleconference) Copyright Hub UK European Patent Office (Germany) Fraunhofer Institute (Germany) Justice Arnold (UK) Justice Carr (UK) Kenneth Crews Max Planck Institute (Germany) Ministry of Business, Innovation & Employment (NZ) New Square Chambers (UK) Professor Mark Schankerman (UK) PRS for Music Ltd (UK) South Square Chambers (UK) The Copyright Hub Foundation (UK) The Organisation for Economic Cooperation and Development (France) UK Intellectual Property Enterprise Court (UK) UK Intellectual Property Office (UK) United States Patent and Trademark Office (Teleconference) US Federal Trade Commission (Teleconference) World Intellectual Property Organization (Switzerland) World Trade Organization (Switzerland) 598 INTELLECTUAL PROPERTY ARRANGEMENTS Table A.3 Roundtables Individual or organisation Melbourne – 21 October 2015 Allens Lawyers CSIRO Dr Nicholas Gruen, Lateral Economics Professor Beth Webster, Swinburne University Professor Jason Potts, RMIT University Research Australia Canberra – 21 October 2015 Australian Academy of Technological Sciences and Engineering Australian Competition and Consumer Commission Dr Hazel Moir, Australian National University Department of Communications and the Arts Department of Industry, Innovation and Science IP Australia Universities Australia Intellectual Property Research Institute of Australia (IPRIA) hosted roundtable Melbourne – 25 November 2015 Andrew Mitchell, University of Melbourne/Global Economic Law Network Arlen Duke, University of Melbourne Graeme Austin, Victoria University of Wellington Joshua Gans, University of Toronto Kwanghui Lim, Melbourne Business School/IPRIA Megan Richardson, University of Melbourne/IPRIA/Centre for Media and Communications Law Owen Malone, IPRIA Paul Jensen, University of Melbourne/IPRIA/Melbourne Institute of Applied Economic and Social Research Peter Yu Texas, A&M University School of Law Philip Williams, Frontier Economics Sam Ricketson, University of Melbourne Tania Voon, University of Melbourne Melbourne – 17 June 2016 Alphapharm AusBiotech Consumers Health Forum of Australia Department of Health Generic and Biosimilar Medicines Association Institute of Patent and Trade Mark Attorneys Lateral Economics Law Council of Australia Medicines Australia (continued next page) CONDUCT OF THE INQUIRY 599 Table A.3 (continued) Individual or organisation Pfizer Australia Professor Andrew Christie Professor Dianne Nicol Public Health Association of Australia Sydney – 15 June 2016 APRA AMCOS Australian Copyright Council Australian Digital Alliance Australian Home Entertainment Distributors Association Australian Publishers Association CHOICE Copyright Agency Limited Google Monash University National Copyright Unit, COAG Education Council Screenrights Telstra University of New South Wales University of Sydney 600 INTELLECTUAL PROPERTY ARRANGEMENTS Table A.4 Public hearings Individual or organisation Page numbers Brisbane – 20 June 2016 Dr Nicolas Suzor 2-13 Hatchette Australia Dr Matthew Rimmer 13-36 36-50 Ms Sheryl Gwyther, Mr Michael Bauer and Ms Angela Sunde 50-66 National Association of Cinema Operators Australasia 66-75 Queensland Law Society Ms Candice Lemon–Scott and Mr Bill Concannon 76-85 86-97 Ms Christine Bongers, Ms Jacqui Carling-Rodgers, Ms Melanie Hill and Ms Andrea Smith Queensland Law Society 98-125 126-127 Avid Reader Bookshop 127-128 Sydney – 21 June 2016 Arts Law Centre Of Australia Ausbiotech 130-141 142-153 Australian Writers Guild IPTA Mr James Kellow, Ms Nikki Gemmel and Mr Michael Robotham Ms Kimberlee Weatherall 153-160 161-175 176-195 195-207 Alpapharm Australian Screen Association Copyright Agency 208-218 218-228 228-239 Mr Anthony Alder Screenrights Dr Ruth Skilbeck Ms Carol O’Donnell 239-248 249-259 259-263 263-265 Authentic Design Alliance and Design Institute of Australia 266-277 Space Furniture and CULT Mr Tomek Archer 277-283 284-285 Canberra – 22 June 2016 AHEDA Australian Digital Alliance APRA & AMCOS 287-302 302-311 312-327 Medicines Australia, IFPMA & JPMA Genetic & Biosimilar Medicines Association Australian Publishers Association 327-345 346-355 356-372 Top3 by Design Independent Designers 373-384 385-396 Archer Office Mr David Trubridge 396-405 406-411 Mr Chris Snow 412-415 (continued next page) CONDUCT OF THE INQUIRY 601 Table A.4 (continued) Individual or organisation Melbourne – 23 June 2016 Australian Policy Online Page numbers Music Australia 418-426 427-436 Professor Andrew Christie Open Source Industry Australia 436-451 451-459 Dr Deborah Gleeson 459-468 Australian Technology Network of Universities 468-475 Cambridge University Press Australia 475-483 Dr Rebecca Giblin Australian Academy of Technology & Engineering 484-494 494-505 Melbourne Style Books 505-515 Cengage Learning Australia P/L Mr David Webber Australian Design Alliance 515-525 525-534 534-545 Ms Jo Jette Spurling Mr Robert Bouvet 545-552 552-554 Melbourne – 24 June 2016 Mr Mark Summerfield Australian Booksellers Association Electronic Frontiers Australia Mr Peter Donoughue 557-571 572-590 591-602 602-609 Ms Dee White Mr Peter Gleeson Association Of Liquor Licences Melbourne 609-616 617-624 624-632 Penguin Random House Australia Law Council Of Australia 633-646 647-659 Qualcomm Incorporated 659-672 IPTA 673-686 Text Publishing Company of Australia Ms Wendy Orr 687-697 698-704 Scribe Publications 705-712 Happy Finish Design Mr Nick Gruen 712-719 719-727 Mr David Day 727-730 Sydney – 27 June 2016 Australian Society of Authors Australian Institute of Professional 732-746 732-754 Aristocrat Leisure 755-762 (continued next page) 602 INTELLECTUAL PROPERTY ARRANGEMENTS Table A.4 (continued) Individual or organisation Page numbers Mr Peter Treloar Association of Australian Musicians 762-769 769-781 Australian Copyright Council Mr Fraser Old 781-792 792-799 Ms Chenoa Fawn 800-805 COAG Education Council 806-818 Pearson Australia Kawat Enterprises 818-826 827-829 Universities Australia 829-839 Free TV Australia Australian Directors Guild 839-848 848-859 Ms Delia Falconer National Association for the Visual Arts DDI Australia 860-870 870-876 877-885 Australian Publishers Association Australian Copyright Council Australian Literary Agents’ Association 885 885-886 886-887 CONDUCT OF THE INQUIRY 603 B International Commitments Many aspects of Australia’s intellectual property (IP) arrangements are embodied in international agreements, which set out minimum IP protections. This appendix sets out the main multilateral organisations dealing with IP (section B.1), the treaties administered by multilateral bodies (sections B.2 and B.3), Australia’s obligations under bilateral and regional trade and investment agreements (section B.4), and the flexibility Australia has to alter its IP policy settings (section B.5). B.1 What are the main multilateral agreements? Australia is a member of two important international organisations that set the framework for IP in the multilateral system. • The World Trade Organization (WTO) Agreement on Trade-Related Aspects of Intellectual Property Rights (TRIPS) entered into force on 1 January 1995. TRIPS is one of the set of agreements that make up the WTO system of trade rules, established at the conclusion of the Uruguay Round of multilateral trade negotiations. Australia is bound by the obligations in TRIPS by virtue of its membership of the WTO. • The Convention establishing the World Intellectual Property Organization 1967 (WIPO Convention). Australia became a member of WIPO in 1972. WIPO, a specialised agency of the United Nations with 188 member states, is the main multilateral body in terms of IP norm-setting. WIPO administers 25 treaties in addition to the WIPO Convention. These treaties are developed by consensus among the WIPO member states. Australia has ratified 16 of the WIPO IP treaties. The WTO system also allows for preferential trade agreements (PTAs) between members. Australia has entered into a number of PTAs containing IP chapters. They include bilateral agreements with Chile, China, Japan, Republic of Korea, Malaysia, Singapore, Thailand and the United States and regional agreements with New Zealand-ASEAN and the recently concluded Trans–Pacific Partnership (TPP). The network of Australia’s international IP agreements is shown in figure B.1. INTERNATIONAL COMMITMENTS 605 Figure B.1 Selected multilateral and bilateral agreements a Association of Southeast Asian Nations – New Zealand (ASEAN-NZ). Source: Department of Foreign Affairs and Trade (2016b). The TRIPS Agreement TRIPS was a game-changer in efforts to establish a global framework for IP. It establishes a minimum set of obligations on WTO members for the protection and enforcement of IP. Countries can provide higher levels of protection, but they cannot provide less without risking a dispute with another WTO member. TRIPS builds on earlier international IP law. The earliest significant multilateral treaties are the Paris Convention for the Protection of Industrial Property 1883 and the Berne Convention for the Protection of Literary and Artistic Works 1886. The Paris and Berne Conventions arose in recognition that IP rights protection within nations has little effect in preventing imitation or copying abroad. 606 INTELLECTUAL PROPERTY ARRANGEMENTS Two non-discrimination rules underpin TRIPS — national treatment and most favoured nation (MFN). National treatment, long a feature of IP treaties including the Paris and Berne Conventions, prohibits a country treating foreign nationals less favourably than its own nationals in the implementation of IP laws. MFN requires that any advantages given to the nationals of any other country must also be given to the nationals of all other WTO members. Subject to some exceptions, MFN prevents Australia applying a principle of reciprocity. So by adopting higher standards, particularly through bilateral agreements, Australia must also provide these higher standards to nationals of third countries, possibly without receiving reciprocal benefits from those countries. TRIPS embodies a significant number of new or higher standards for IP protection relative to the earlier Paris and Berne Conventions. It sets out standards of protection for copyright, trade marks, geographical indications, industrial designs, patents, integrated circuit layouts, and undisclosed information (box B.1). Box B.1 What does TRIPS do? According to the Department of Foreign Affairs and Trade (DFAT), TRIPS is intended to maximise the contribution of IP systems to economic growth through trade and investment by: • establishing minimum standards for IP rights protection in the national systems of WTO members • prescribing agreed elements of an effective mechanism for administration and enforcement of IP rights • creating a transparency mechanism — each WTO member is required to provide details of their national IP laws and systems, and to answer questions about their IP systems • creating a predictable, rules-based system for the settlement of disputes about trade-related IP issues between WTO members • allowing for mechanisms that ensure that national IP systems support widely accepted public policy objectives, such as stamping out unfair competition, facilitating transfer of technology, and promoting environmental protection. In setting minimum standards, TRIPS resulted in a number of changes to Australian law including extending the term of patent protection from 16 to 20 years. Source: DFAT (2015a). Controversially, TRIPS reframed IP as trade law. TRIPS is subject to the WTO dispute settlement regime, which provides a framework for determining whether a country’s measures comply with the agreement (Taubman 2008). While TRIPS establishes standards for private rights and obligations, dispute settlement is reserved to member states of the WTO. Private rights holders cannot invoke a dispute under the WTO rules. The dispute mechanism gives TRIPS teeth. A country can impose trade sanctions on another country for violation of TRIPS, provided authorisation is obtained from the WTO. INTERNATIONAL COMMITMENTS 607 The WTO dispute mechanism replaced the more unpredictable bilateral approach to dispute settlement of IP differences that preceded TRIPS. Australia made significant changes to patent and copyright rules to implement TRIPS. Patent term was increased from 16 to 20 years (pharmaceuticals were already protected for up to 20 years) and conditions were placed on the grant of compulsory licences for patents (such as a requirement to compensate the patent owner). The extension of patent term was a significant change and previous analysis by Commission staff found it resulted in large net costs to Australia (Gruen, Bruce and Prior 1996). Copyright laws were amended to grant rental rights for computer software and sound recordings, and to extend the scope of performers’ rights and raise the duration of their protection from 20 to 50 years (Revesz 1999). TRIPS also allows Members to provide for exceptions and limitations to the exclusive rights covered under the agreement. Along with a number of other multilateral, regional and bilateral agreements, TRIPS uses language known as the ‘three-step test’ to define the freedom of countries to provide for exceptions and limitations to copyright law (box B.2). Box B.2 The three step test for exceptions and limitations The three step test was first enacted in the 1967 revision of the Berne Convention in relation to reproduction of authors works. Variations have since been included in a number of multilateral, regional and bilateral trade agreements and in a number of WIPO copyright treaties. The three step test in Article 13 of TRIPS applies to exceptions and limitations to any of the exclusive rights associated with copyright. Members shall confine limitations and exceptions to exclusive rights • to certain special cases • which do not conflict with a normal exploitation of the work • and do not unreasonably prejudice the legitimate interests of the right holder. TRIPS also applies modified versions of the test to trade marks (article 17), designs (article 26.2) and patents (article 30). Source: Electronic Frontier Foundation (2016). World Intellectual Property Organization Membership of WIPO treaties is not mandatory. Unlike TRIPS and the WTO, membership of WIPO does not require membership of any of the IP treaties. Australia can decide which treaties it wishes to join. 1 Australia may also denounce any of the WIPO treaties it has joined. While the mechanism for denouncing a treaty is set out in each agreement, it 1 Although Australia has agreed in bilateral agreements such as the Australia-United States Free Trade Agreement (AUSFTA) to ratify certain WIPO treaties. 608 INTELLECTUAL PROPERTY ARRANGEMENTS typically requires written notification to the Director General of WIPO, with denunciation taking effect one year later. Instances of countries denouncing WIPO treaties are rare. Some treaties, such as the Patent Cooperation Treaty (PCT) and the Madrid Protocol, require membership to take full advantage of the benefits they provide. For others, it is not necessary to join to obtain the benefits. In these cases, the primary benefits of membership are having a seat at the table to help shape future reforms and influence other countries to join. B.2 Australia’s membership of World Intellectual Property Organization treaties WIPO treaties fall into four groups (table B.1): treaties that provide basic standards of IP protection (IP Protection); systems for filing one international application (Global Protection Systems); administrative agreements that simplify and streamline the process of obtaining rights internationally (Administrative); and treaties that create international classification systems (Classification). The first group of treaties provide for national treatment and a right of priority, 2 extend the rights protected under copyright and related rights, and set an international standard of duration for copyright protection of life of the author plus 50 years. The WIPO Internet treaties (as the WIPO Copyright Treaty and WIPO Performances and Phonograms Treaty are collectively known) both incorporate the three-step test for determining exceptions and limitations from the Berne Convention. The WIPO Internet treaties provide terms of copyright protection of at least 50 years and, like Berne, protection cannot be subject to formalities such as a registration system. The other groups of WIPO treaties are largely administrative in nature. They can lead to international harmonisation through cooperation between countries and setting maximum standards of formality. 2 A right of priority allows an applicant to file in a first country then, within a specified time (12 months for patents, 6 months for designs and trade marks) file in a second country. The priority date is used to assess when an invention or design was new and who was first to file an application for the right. INTERNATIONAL COMMITMENTS 609 Table B.1 Australia’s membership of WIPO treaties Ratified by Australia Total members 10 October 1925 14 April 1928 176 172 22 June 1974 78 26 October 1990 30 September 1992 37 92 26 July 2007 94 26 July 2007 10 December 2015 94 20 31 March 1980 7 July 1987 150 80 11 July 2001 97 21 January 1998 16 March 2009 16 March 2009 53 45 37 8 April 1961 12 November 1975 84 62 IP Protection Paris Convention (industrial property) Berne Convention (literary and artistic works) Phonograms Convention Brussels Convention (Satellite Signals) Rome Convention (Performers, Producers of Phonograms and Broadcasting Organisations) WIPO Copyright Treaty (WCT) WIPO Performances and Phonograms Treaty (WPPT) Marrakesh Visually Impaired Persons Treaty Global Protection System Patent Cooperation Treaty Budapest Treaty (Deposit of Microorganisms for Patents) Madrid Protocol (Trade Marks) Administrative Trademark Law Treaty Singapore Treaty on the Law of Trademarks Patent Law Treaty Classification Nice Agreement (Trade Marks) International Patent Classification Source: WIPO (2016k). B.3 Australia’s membership of UPOV treaties Australia is also a member of the International Union for the Protection of New Varieties of Plants (UPOV). UPOV is a separate body to WIPO. Australia acceded to UPOV 91 on 20 December 1999, following accession to the earlier UPOV 78 in 1989. UPOV 91 provides rights to breeders for new varieties of plants that are new, distinct, uniform and stable. It sets out compulsory exceptions for non-commercial use, for experimental purposes, and for breeding other varieties. It also provides an optional exception allowing farmers to save seed to be replanted on their own holdings. 610 INTELLECTUAL PROPERTY ARRANGEMENTS B.4 Bilateral and Regional Trade Agreements Preferential trade agreements In recent years, the Australian Government has entered a range of PTAs. These agreements seek to reduce trade barriers between partner countries on a preferential basis. Most also contain provisions affecting broader areas of policy including IP. Australia has implemented nine PTAs with IP chapters, all since 2003. It has also concluded negotiations on the TPP which includes an extensive IP chapter. The TPP is a regional trade agreement between 12 countries — Australia, Brunei, Canada, Chile, Japan, Malaysia, Mexico, Peru, New Zealand, Singapore, the United States and Vietnam — with the potential for further countries to join. Negotiations are progressing on the Regional Comprehensive Economic Partnership, an ASEAN-centred proposal including the ten ASEAN member states and those countries that have existing trade agreements with ASEAN — Australia, China, India, Japan, Republic of Korea and New Zealand. The Australian Government has announced plans for an agreement with the European Union (EU). The EU has previously used PTAs to expand IP protection for the benefit of certain European industries. While most Australian PTAs have IP chapters, the nature of commitments varies from cooperation based agreements to prescriptive standards of protection. The Australia-United States Free Trade Agreement (AUSFTA) includes a detailed IP chapter that strengthened the protection given to holders of IP rights (box B.3). AUSFTA is the only PTA that has required changes to Australia’s IP laws (DFAT, sub. 65). While the lock-in effects of AUSFTA are well-known, many of the provisions have been repeated in subsequent PTAs, resulting in overlapping and complex rules. For example, Australia first agreed to copyright duration of 70 years beyond the life of the author in AUSFTA and has replicated that commitment in subsequent trade agreements with Chile, Singapore, Korea and the recently concluded TPP. INTERNATIONAL COMMITMENTS 611 Box B.3 Changes to Australian law resulting from the AUSFTA Australia made substantial changes to its copyright regime as a result of the AUSFTA. The main changes to Australia’s copyright regime included: • extending the term of protection for most copyright material by 20 years • new economic and moral rights for performers in sound recordings • a scheme for limitation of remedies available against Carriage Service Providers for copyright infringement • wider criminal provisions for copyright infringement • broader protection for electronic rights management • protection against a wider range of unauthorised reproductions. Changes to the patent system were minimal, with amendments to introduce new grounds to refuse a patent being granted based on the invention not being useful, or being secretly used. These grounds had previously only been available for revoking a patent once granted, not for preventing a patent being granted. There were also a number of new international obligations that did not require changes to Australian patent law. AUSFTA went beyond TRIPS in areas such as patents for plants and animals, limits on compulsory licensing, restrictions on international exhaustion of patent rights, and adjustment of the term of pharmaceutical patents. AUSFTA obligated Australia to maintain five years of protection for test data submitted to the Therapeutic Goods Agency for marketing approval of pharmaceuticals and introduced ten years of protection for data submitted for the approval of agricultural chemicals. Australia introduced a notification scheme in its regulation of pharmaceuticals whereby patent owners are notified when a generic manufacturer obtains marketing approval during the patent term. Amendments were made to Australia’s system for protecting geographical indications for wine, with the introduction of procedures allowing third parties to object to protection of a geographical indication. Sources: IP Australia (2016e); Taubman (2008). Australia-Europe Wine Agreement The Australia – European Community Agreement on Trade in Wine commenced on 1 September 2010 and provides rules for the trade in wine between Australia and the EU. The agreement replaces an earlier agreement that entered into force in 1994. Both agreements require Australia to protect agreed European geographical indications (GIs) for wines. The 2010 agreement requires Australia to protect certain ‘sensitive’ EU GIs, including Chablis, Champagne, Port, Sherry and Burgundy. Australia now protects more than 2500 European GIs on the Register of Protected Geographical Indications and Other Terms, administered by Wine Australia (DAFF 2010). 612 INTELLECTUAL PROPERTY ARRANGEMENTS B.5 International flexibility While international agreements set out minimum standards of protection that Australia has agreed to provide, there is some flexibility in how those obligations are implemented (box B.4). Box B.4 Flexibilities under TRIPS TRIPS sets out minimum standards of protection that each WTO member must provide to the intellectual property of other members. However, TRIPS also incorporates flexibilities. These flexibilities generally fall within the following categories: • Flexibilities as to the method of implementing TRIPS obligations. This allows WTO members to exploit creative solutions when implementing provisions that TRIPS enunciates but does not define. • Flexibilities as to substantive standards of protection. These can operate upwards or downwards. That is, they allow measures that limit the rights conferred (exceptions and limitations), or measures that raise the level of protection beyond the minimum standards (more extensive protection). • Flexibilities as to mechanisms of enforcement. WTO Members can, for example, maintain their own judicial system or adopt IP specific courts. Flexibilities as to the standards of protection can also be implemented through enforcement measures, for example through limitations on remedies or by addressing abuse of enforcement procedure. • Flexibilities as to areas not covered by TRIPS. Where protection is afforded to areas that fall outside of TRIPS, countries do not need to conform to the principles in TRIPS, such as national treatment or most favoured nation. • Control of anticompetitive practices. Countries can adopt measures to prevent or control practices in the use or enforcement of IP rights that are abusive and anticompetitive. TRIPS also recognises that some licensing practices or conditions of IP rights which restrain competition can have adverse effects on trade and may impede the transfer and dissemination of technology. Source: WIPO (2016a). TRIPS enunciates a number of concepts but leaves them undefined. Examples are patent concepts such as ‘invention’ and ‘inventive step’. The TPP goes further in requiring that each Party shall consider whether the claimed invention would have been obvious to a person skilled in the art. This provides a minimum standard for inventive step, rather than a maximum (DFAT 2015b). As one submission noted, terms such as ‘obvious’ and ‘person skilled in the art’ are undefined (Gleeson, sub. 128). Australia therefore has flexibility as to how it sets the appropriate level of inventive step. There is also scope to exclude certain subject matter such as business methods and software. While some proponents of software patenting have argued that TRIPS requires patent protection of such innovations, many countries have adopted exceptions without INTERNATIONAL COMMITMENTS 613 challenge (Bakels and Hugenholtz 2002, p. 15). In fact, Article 10 of TRIPS recognises that computer programs should be protected under copyright. While TRIPS and the Berne Convention place restrictions on formalities (such as registration systems) for copyright, this is not the case for other rights. Formalities can be employed as a policy lever to achieve certain outcomes, for example, renewal fees can be imposed periodically to maintain a patent. Economists have found that a fee structure where renewal fees increase over the life of a patent is a desirable feature of the patent renewal process (chapter 7). TRIPS does not address the issue of exhaustion of IP rights. This allows countries to provide for parallel importation of goods. Australia is unfettered in allowing for parallel importation of copyright or trade marked goods and services. The three step test from the Berne Convention (and variations thereof) has become the international standard for assessing permissibility of copyright limitations and exceptions. The bounds of the three-step test have not been precisely defined. Terms such as ‘certain,’ ‘normal’ and ‘unreasonable’ are all open to interpretation. Following a detailed analysis, the Australian Law Reform Commission (2014a) considered that fair use and fair dealing exceptions would be consistent with Australia’s international obligations (chapter 6). WTO members can provide more extensive protection than required by TRIPS, as long as the additional protection does not contravene provisions in TRIPS. These are often referred to as ‘TRIPS-plus’ measures. This can provide additional flexibility to address specific aspects of national economic interest. The innovation patent system is an example of a TRIPS-plus measure. There are no binding requirements to maintain an innovation patent system. 614 INTELLECTUAL PROPERTY ARRANGEMENTS C The IP-intensity of Australia’s trade Key points • Australia is a significant net importer of intellectual property-intensive goods and services. • Estimates of patent-intensive goods show that real imports nearly tripled over the period 1998-99 to 2014-15. Growth in exports was more subdued. The real value of the trade gap (the difference between imports and exports of patent-intensive goods) increased by about 5 times (from about $4 billion to $20 billion) between 1998-99 and 2014-15. • Copyright-intensive goods are becoming relatively smaller in magnitude than copyright-intensive services, for both exports and imports. This likely reflects growth in music and television subscription services, offset by declines in consumer purchases of merchandise goods, such as compact discs. – In 2014-15, the value of Australian trade (exports and imports) in copyright-intensive services was nearly $10 billion, whereas trade in copyright-intensive goods was only about $2.5 billion. – For merchandise goods, copyright-intensive exports declined slightly in real terms, and imports increased slightly between 1998-99 and 2014-15. – Copyright-intensive service imports more than tripled in real terms between 1998-99 and 2014-15. In contrast, over the same period, copyright-intensive service exports increased by about 60 per cent. – The trade gap between imports and exports of copyright goods and services increased from about $1 billion to $5 billion between 1998-99 and 2014-15. • Exports of trade mark-intensive goods were estimated to have declined between 1998-99 and 2014-15. Over the same period, imports of trade mark-intensive goods increased by almost 150 per cent. The trade gap nearly tripled over the period, to about $30 billion. C.1 Introduction The terms of reference for this inquiry ask the Commission to consider the relationship between intellectual property (IP) and trade. As noted throughout this report, while IP rights are territorial, the IP embodied in goods and services flows across borders. This appendix provides estimates of the value of Australian trade in goods and services deemed to have a high IP content. Estimates are provided separately for patent-, copyright- and trade mark-intensive goods (sections C.2, C.3 and C.4 respectively). The results show that Australia is a significant net importer of IP. THE IP-INTENSITY OF AUSTRALIA’S TRADE 615 The results also show an increasing IP ‘trade gap’ (the difference between imports and exports of IP-intensive goods) has emerged since the early 2000s. While the Commission’s estimates employ a different methodology, its findings are consistent with results published by the Organisation for Economic Co-operation and Development (OECD) (box C.1). Box C.1 Australia’s technology balance of payments The OECD presents data on the ‘technology balance of payments’ (TBP) for OECD countries. The TBP is one way to measure Australia’s receipts and payments for some of the types of intellectual property considered in this inquiry. The TBP consists of money paid and received for various uses, including for patents, licences, trade marks and designs. Specifically, there are four main categories of technology receipts and payments: • transfer of techniques (through patents and licences, disclosure of know-how) • transfer (sale, licensing, franchising) of designs, trademarks and patterns • services with a technical content, including technical and engineering studies, as well as technical assistance • industrial R&D (OECD 2009b, p. 118). Australia’s TBP is presented in the figure below. Payments exceeded receipts by a small amount up until 2006. Since that time, the intellectual property trade gap (difference between payments and receipts) has been growing. Australia’s technology balance of payments 12 10 Receipts Payments $ billion 8 6 4 2 0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 Source: Commission estimates using data from OECD Main Science and Technology Indicators database. 616 INTELLECTUAL PROPERTY ARRANGEMENTS C.2 Estimates of traded goods with high patent usage To examine the value of trade in patent-intensive goods, it is first necessary to determine what constitutes a ‘patent-intensive’ good. Some taxonomies exist (see for example Maskus (1993)) but are dated and arguably subjective. 1 For this reason, the Commission has constructed its own measure of patent-intensive goods. Patent-intensive goods were identified in two steps. First, the total number of patent applications in Australia for each type of traded good was aggregated over the period 2005–14. This gives an estimate of the absolute patent intensity for each good. Second, for a given good, the total number of patents attached to it was divided by the number of persons employed in the industry that produced that good (yielding a ‘relative patent intensity’). 2 Those goods with relative patent intensities above the average were categorised as patent-intensive. The above approach borrows heavily from that used in two recent studies — for the United States (US Department of Commerce 2012) and for Europe (OHIM and EPO 2013). Those studies identified industries (rather than goods) that most intensively use patents and trade marks. More detail on the methodology is provided in section C.5. The product groups identified as being patent-intensive were mapped to trade data for 2014-15. The trade data were taken from the ABS, and are at a disaggregated level (Standard International Trade Classification (SITC) five digit). The results show that Australia is overwhelmingly a net importer of patent-intensive goods. The estimated value of patent-intensive imports was $29.7 billion and the value of patent-intensive exports was $7.1 billion in 2014-15. The IP trade gap — the difference between imports of goods with a high patent intensity and exports of the same type of goods — has been increasing over time (figure C.1). In the past 15 years, imports of patent-intensive goods have nearly tripled in real terms whereas exports have remained relatively unchanged. 3 In real terms, the IP trade gap has increased around five-fold since 1998-99. 1 These older studies typically focus only on merchandise goods, and not services, and so ignore an increasingly important area of trade (section C.3). 2 Alternative measures to determine relative intensity (such as value added) were not used because the data were not sufficiently disaggregated for the analysis (section C.5). 3 In nominal terms, the trade gap increased by a lesser amount. THE IP-INTENSITY OF AUSTRALIA’S TRADE 617 Figure C.1 Australia’s trade in patent–intensive goods 2014-15 dollars 35 30 $ billion 25 Imports Exports 20 15 Imports almost tripled over the period, increasing the IP trade gap 10 5 0 Sources: Commission estimates based on ABS (International Trade in Goods and Services, Australia, Cat. no. 5368.0, unpublished data at the five digit level; Balance of Payments and International Investment Position, Australia, Cat. no. 5302.0). C.3 Estimates of goods and services with high copyright usage The World Intellectual Property Organization (WIPO) has released a guide to help categorise copyright industries (WIPO 2015a). There are four categories of copyright industries, depending on the degree to which their activity depends on copyright — core copyright industries, interdependent copyright industries, partial copyright industries and non-dedicated support industries. Core copyright industries are those which are ‘wholly engaged in the creation, production and manufacture, performance, broadcasting, communication and exhibition, or distribution and sale of works and other protected subject matter’ (WIPO 2015a, p. 51). The core copyright industries, as defined by WIPO (2015a), are: • press and literature • music, theatrical productions, operas • motion picture and video • radio and television • photography 618 INTELLECTUAL PROPERTY ARRANGEMENTS • software, databases and computer games • visual and graphic arts • advertising services • copyright collective management societies. In a 2012 study to identify copyright- (and patent- and trade mark-) intensive industries in the United States (US Department of Commerce 2012), only the core copyright industries were used. Industries only engaged in the distribution of copyright works were excluded from the study. Trade in copyright-intensive goods and services The Commission has also chosen the core copyright industries identified by WIPO to determine, and separately estimate trade in, copyright-intensive goods and services. To estimate trade in copyright-intensive goods, merchandise trade data for three product types — printed matter, sound recordings and other goods (mainly video games) — were identified as being produced by the core copyright industries outlined above. Some goods, such as photographic supplies, were not included. That is because products such as photographic equipment, television sets, and cameras (used to produce material which may be subject to copyright) are not part of the subcategories of core copyright industries. The goods are similar to those that Revesz (1999) selectively identified as copyright-intensive goods. The results show Australia is a net importer of copyright-intensive goods. The traded value of copyright-intensive goods has decreased over time for exports and increased slightly for imports (figure C.2). 4 In addition to copyright-intensive goods, there are also some services provided that are likely to be subject to copyright. The ABS has balance of payments information on copyright-intensive services. These being licence fees to reproduce and/or redistribute computer software services; music and other charges; royalties for audiovisual and related services; 5 and computer and information services. Again, these are similar to the services chosen by Revesz (1999). Computer and information services includes subscription services, software publishing services and computer design and development, but also support services, which may rely less on copyright. Thus, the figures should be considered as upper bound estimates. The results show Australia is a net importer of these services. 4 Although the value of imports and exports of non-core copyright industries was not considered in this report, it has been by others. For example a report by PricewaterhouseCoopers (PwC 2015b) found that interdependent industry imports declined, but partial copyright industry imports increased slightly between 2002 and 2014. Imports declined overall. Exports fell across all types of copyright-related industries. 5 This includes cable (pay TV) and other subscription broadcasting services. THE IP-INTENSITY OF AUSTRALIA’S TRADE 619 Figure C.2 Australia’s trade in copyright–intensive goods 2014-15 dollars 3.5 3.0 Imports Exports $ billion 2.5 2.0 1.5 1.0 0.5 0.0 Sources: Commission estimates based on ABS (International Trade in Goods and Services, Australia, Cat. no. 5368.0, unpublished data at the five digit level; Balance of Payments and International Investment Position, Australia, Cat. no. 5302.0). Compared with merchandise goods, which show a relatively flat trend over time (figure C.2), imports of copyright-intensive services have increased sharply since the early 2000s while exports have remained relatively constant (figure C.3). This is likely driven by consumers moving towards using newer copyright services instead of buying goods. An example is the music industry — relatively new services that provide music (for example, Spotify) are likely to have displaced sales of physical products such as compact discs. When considering goods and services together, royalties and licence fees contributed a much larger share than merchandise goods to the total value of copyright goods and services in 2014-15 (table C.1). Further, the difference between imports and exports of copyright services has increased over time, reflecting the large growth in imports of copyright-intensive services (royalties, licence fees and other charges). In real terms, growth in copyright-intensive imports (both goods and services) outpaced exports between 1998-99 and 2014-15 (figure C.4). The increasing trade gap between copyright-intensive imports and exports largely reflected growth in copyright-intensive service imports. Copyright goods and services exports have remained fairly stable, as have imports of copyright goods. In contrast, imports of copyright-intensive services increased significantly over the past 15 years (figure C.5). 620 INTELLECTUAL PROPERTY ARRANGEMENTS Figure C.3 Australia’s trade in copyright–intensive servicesa 2014-15 dollars 7 6 $ billion 5 4 Imports Exports IP trade gap has emerged, largely due to growth in imports 3 2 1 0 a The spike in exports for 2000-01 reflects a one-off increase in ‘other film, television and multimedia royalties’. This was due to the broadcasting rights for the Sydney Olympic Games. Source: Commission estimates based on ABS (Balance of Payments and International Investment Position, Australia, Cat. no. 5302.0). THE IP-INTENSITY OF AUSTRALIA’S TRADE 621 Table C.1 Trade in copyright–intensive goods and services 2014-15 dollars Description of copyright-intensive good or service 1998-99 2014-15 Change Exports $ million $ million Per cent Computer and information services Audiovisual royalties 599 1 013 278 617 2 306 222 3 128 -20 Sub-total services and charges 1 890 3 145 66 Licences to reproduce and/or distribute computer services (software); charges for music and other IP services Printed matter 459 256 -44 Sound recordings 243 126 -48 35 71 104 737 453 -39 2 628 3 598 37 818 456 659 2 542 2 559 1 359 211 462 106 1 933 6 460 234 946 819 1 056 812 12 -1 Other Sub-total merchandise goods Total copyright exports Imports Licences to reproduce and/or distribute computer services; charges for music and other IP services Computer and information services Audiovisual royalties Sub-total services and charges Printed matter Sound recordings Other 133 363 174 Sub-total merchandise goods 1 898 2 231 18 Total copyright imports 3 831 8 691 127 Sources: Commission estimates based on ABS (Balance of Payments and International Investment Position, Australia, cat. no. 5302.0, ABS International Trade in Goods and Services, Australia, Cat. no. 5368.0, unpublished data at the five digit level. 622 INTELLECTUAL PROPERTY ARRANGEMENTS Figure C.4 Australia’s trade in copyright–intensive goods and services 2014-15 dollars 10 9 Imports 8 Exports $ billion 7 IP trade gap has increased, with strong growth in imports 6 5 4 3 2 1 0 Sources: Commission estimates based on ABS (Balance of Payments and International Investment Position, Australia, Cat. no. 5302.0; International Trade in Goods and Services, Australia, Cat. no. 5368.0, unpublished data at the five digit level). Figure C.5 Composition of trade in copyright goods and services 2014-15 dollars (i) Exports (ii) Imports 10 10 Goods 9 Services 8 8 7 7 6 6 5 4 $ billion $ billion 9 Goods Services 5 4 3 3 2 2 1 1 0 0 Sources: Commission estimates based on ABS (Balance of Payments and International Investment Position, Australia, Cat. no. 5302.0; International Trade in Goods and Services, Australia, Cat. no. 5368.0, unpublished data at the five digit level). THE IP-INTENSITY OF AUSTRALIA’S TRADE 623 C.4 Estimates of traded goods with high trade mark usage Goods deemed to be intensive users of trade marks were also identified using the same methodology for patents. That is, the relative intensity (number of trade mark applications divided by employment, for each good) was estimated. The concordance between product classification (by Nice descriptor6) and SITC goods was only available at the two digit SITC level. 7 This meant that the total number of product groups (by SITC) were small, but tended to be large in value. Because the product groups were so broad in scope, some judgment was used to omit goods within the broad categorisations that appeared to be primary products or raw materials. However, to offset these omissions, two categories (rubber tyres, and office and other machines) were introduced as they also have a relatively high trade mark intensity, only slightly below average (section C.5). Because the data were aggregated at a higher level, some goods within a category had already been identified as being copyright-intensive or patent-intensive. These goods were omitted (section C.5). Table C.2 describes the broad categories of trade mark-intensive goods identified, and their value in trade. The goods identified are somewhat consistent with those obtained by Dernis et al. (2015). For example, they found that some of the highest trade mark intensity ratios (the number of trade marks filed divided by sales) were in industries including other manufactures; textiles and apparel; wood and paper; and rubber, plastics and minerals. 8 Similar to patent-intensive goods and copyright-intensive services, there has been a divergence between imports and exports of trade mark-intensive goods. Over the past 15 years imports of trade mark-intensive merchandise increased around 2.5 times, while exports decreased slightly (figure C.6). The ABS collects balance of payments data on franchise and trade mark licensing fees. These payments are relatively small compared with trade in goods, but also show that Australia is a net importer. Fee payments to Australia (service exports) were only $70 million in 2014-15, whereas payments to non-residents (imports) were about $1.5 billion. 6 The Nice Classification was established by the Nice Agreement (1957), and is an international classification of goods and serviced applied for the registration of trade marks. 7 Lybbert et al. (2014) outline how the concordance was done, with accompanying concordance files available at: http://www.wipo.int/econ_stat/en/economics/publications.html. 8 Note, however, that other goods deemed to be copyright- or patent-intensive above also featured prominently on the list (for example, pharmaceuticals and publishing and broadcasting), as did service industries. 624 INTELLECTUAL PROPERTY ARRANGEMENTS Table C.2 Trade in trade mark-intensive goods 2014-15 Exports ($m) Imports ($m) Tobacco Perfumes, soap 5 571 219 1 832 Insecticides, starches, explosives etc 652 598 66 2 425 Paper, paper products Woven fabrics, textiles, floor coverings 841 250 2 303 2 956 Glass, glassware, pottery Iron/steel products 143 403 1 029 987 Tools, for use in hand or machines; cutlery, household equipment 156 1 436 Office machines, automatic data processing machines Clothing 644 261 8 601 7 422 Ammunition, plastic articles, toys, office supplies, artworks, jewellery, musical instruments 1 170 6 261 Total 5 163 36 069 Rubber manufactures (tyres) Source: Commission estimates based on ABS (International Trade in Goods and Services, Australia, Cat. no. 5368.0, unpublished data at the five digit level). Figure C.6 Australia’s trade in trade mark–intensive goods 2014-15 dollars 40 35 Imports Exports $ billion 30 25 20 IP trade gap has increased, due to growth in imports 15 10 5 0 Sources: Commission estimates based on ABS (Balance of Payments and International Investment Position, Australia, Cat. no. 5302.0; International Trade in Goods and Services, Australia, Cat. no. 5368.0, unpublished data at the five digit level). THE IP-INTENSITY OF AUSTRALIA’S TRADE 625 C.5 Method to estimate relative intensities for patents and trade marks This section contains a more detailed explanation of how relative intensities were determined for patent- and trade mark-intensive goods. Estimating relative intensities for patents The results in table C.1 were generated using various data sources and concordance tables. Patent classifications Information on the number of patents filed was extracted from IP Australia’s database — Intellectual Property Government Open Data (IPGOD) (2015 edition). IPGOD has information on patents filed in Australia, which were used to estimate how many patents are used for various types of goods. Each patent is classified according to an International Patent Classification (IPC). The IPC divides technology into eight sections, with approximately 70 000 subdivisions. Results were aggregated to a higher level for analysis. Around 600 IPC categories were used. Data on the number of patents (including those not necessarily granted) for the 10 year period covering 2005–2014 were used in the analysis. Mapping patent classifications to traded goods A paper prepared for WIPO (Lybbert and Zolas 2012) outlines a concordance between IPC and traded goods (classified by SITC). This concordance generally maps each IPC category to one or multiple SITC goods. The mapping is done using probabilistic matching procedures. Put simply, this means that key words are extracted from the IPC category name and then matched probabilistically to industry or trade classifications, by using probability weights. Thus the process relies on actual data, rather than manually selecting a concordance between each IPC and industry or good. The concordance between IPC and SITC was used to assign each patent in the database to one or multiple goods (by SITC), by weighting the results. By mapping each patent to one traded good (or more partially traded goods), the absolute number of patents by SITC was obtained. This gives an estimate of the absolute patent intensity for each type of traded good (at the 5 digit SITC level). Estimating the relative patent intensity of each good Absolute intensity is not a good indicator of whether a good is highly patent-intensive, because it does not account for the amount of inputs used to produce the good. For example, each industry producing a good will vary in size. A good assigned with a high 626 INTELLECTUAL PROPERTY ARRANGEMENTS number of patents might also use a lot of inputs relative to others. Ideally, measures of patent-intensity would account for these inputs. One way to do this is to adjust for the size of the industry, for example, by using employment, sales, value add or capital inputs. For the analysis here, a ‘relative intensity’ measure was defined as the number of patents per good divided by employment in the industry that produces the good.9 The choice to use employment was largely driven by data availability — other indicators did not have disaggregated data to the same Australian and New Zealand Standard Industrial Classification (ANZSIC) industry level. While data for employment by industry are available, data on employment by SITC category are not. Hence another concordance table was used to link the traded good (by SITC) to industry of employment (using ANZSIC codes). Two issues with this approach were addressed. • First, the concordance between SITC and ANZSIC was not at the same level — the industry classification was at a higher (more aggregated) level. In some cases, multiple SITC goods that were patent-intensive had a concordance that matched to the same industry. In these instances, each SITC good was assumed to be produced by the same industry. As such, employment (obtained from industry data) for those goods was also assumed to be the same. • Second, the concordance between a SITC good and ANZSIC was not always one-to-one. There were multiple industries (and therefore employment levels) that some SITC goods matched with. The method used to generate a one-to-one concordance was to select the first industry listed in the concordance table that a good matched with as the industry from which to impute employment. Although this method is not ideal, it affected relatively few goods. In the final step, for each traded good, the number of patents was divided by employment to obtain relative patent intensity. To be consistent with the traded goods data, the average of the most recent ten years of employment data was used (2006–2015). A traded good was defined as being ‘patent-intensive’ if its relative patent intensity was above the average for all traded goods (and only for those goods which had at least one patent assigned to them). An illustration of the process, including the data sources and concordances, is outlined in figure C.7. In this example, the good with IPC classification ‘A’ has a concordance to three different goods by SITC class. They have probability weights (p1, p2, and p3) associated with each SITC good that sum to one. Finally, the number of patents for each SITC good is then matched with employment in the industry that produces it. The relative intensity is defined as the number of patents for a good (by SITC) divided by employment (by ANZSIC). 9 Employment data were sourced from the ABS (Labour Force, Australia, Detailed Quarterly Nov 2015, Cat. no. 6291.0.55.003). THE IP-INTENSITY OF AUSTRALIA’S TRADE 627 Figure C.7 Illustrative example of the concordance used to estimate patent intensity Estimating relative intensities for trade marks The conceptual process used to estimate patent-intensive goods, as described above, was also applied to trade marks. However, data limitations necessitated some adjustments. • Like patents, the SITC goods did not have a one-to-one concordance with employment. This required judgements to be made in order to select the ANZSIC industry of employment. • The concordance between trade mark product (defined by Nice classification) and SITC goods was only available at the two digit SITC level. This meant that the goods deemed to be trade mark-intensive were very highly aggregated. • There was some overlap between trade mark-intensive goods and those already deemed to be patent-intensive or copyright-intensive (from above). The second and third issues in particular meant that the estimates for trade mark-intensive goods included a broader group of products than was ideal. Keeping these issues in mind, the results for trade mark-intensive goods showed that imports were about $51.2 billion and the value of exports was about $9 billion in 2014-15. Thus the IP trade gap was over $40 billion, with imports more than five times larger than exports. These figures are likely to be overestimates of the magnitude of trade in mark-intensive goods. Using two digit rather than five digit data means that some two digit goods that comprise a broad five digit category are likely to be goods other than only those deemed to be ‘trade 628 INTELLECTUAL PROPERTY ARRANGEMENTS mark-intensive’. 10 Further, if trade mark-intensive goods that have already been identified as patent-intensive or copyright-intensive goods are included this will lead to double counting for overall IP-intensive goods in trade. In order to account for this some adjustments were made. • First, the composition of all two digit SITC goods was examined. Where it comprised a five digit good that was already deemed to be either copyright-intensive or patent-intensive (by the process described above) it was omitted, to avoid double counting. Further, the (two digit) good ‘plastics in primary forms’ was omitted entirely because many goods at the five digit level that comprise this two digit good (one-third) were deemed to be patent-intensive. • Where a product was deemed to be trade mark-intensive (at the five digit level), and was comprised of components that appeared to be more primary or intermediate goods than final products (at the two digit level), they were excluded. Goods that more closely resembled a manufactured product were retained. For example, the broad two digit level good ‘rubber product manufacturing’ consists of various goods at the five digit level, including rubber in primary forms, unhardened rubber tubes, and tyres (for example, used on motor vehicles and bicycles). Only the latter good was retained. Dernis et al. (2015) found that industries such as ‘basic metals’ and ‘rubber, plastics and minerals’ featured high patent propensities, but low trade mark intensities. The authors stated that this ‘might mirror the very nature of these industries, which generally rely on sophisticated technologies to produce intermediate goods, and hence do not reach out to final consumers to the extent that others [do]’ (p. 58). • To offset the omission of primary products (including one entire good at the two digit level) an additional two goods were included. These were rubber manufactures not elsewhere specified (tyres) and office machines and automatic data processing machines. These items were chosen because they had a relative intensity just below the average. Although this approach relied on some judgment to exclude and include various goods after applying the overarching framework, the trends in the underlying results were robust to the method used. The main difference was a change in the magnitude of trade. Compared with retaining all broad two digit category goods and their components, the changes had the effect of reducing trade mark-intensive exports from $9 billion to $5.1 billion. Trade mark-intensive imports were reduced from $51.2 billion to $36 billion. Expressed differently, the IP trade gap was reduced from over $40 billion to about $30 billion. In the past 10 to 15 years there was an increase in trade mark-intensive imports while exports remained flat, regardless of the method used. 10 Although conversely it is possible that a trade mark-intensive good at the five digit classification level will be omitted if the other goods that combine with it to comprise a two digit good have a low intensity. THE IP-INTENSITY OF AUSTRALIA’S TRADE 629 D Evidence on patents: social value, additionality and thickets This appendix draws together some of the empirical evidence on patents. This evidence supports the Commission’s assessment that many patented inventions do not create material social benefits (section D.1) and that, regardless of their social value, many others do not need patent protection to encourage their development or commercialisation (section D.2). The appendix concludes by examining the extent of patent thickets in Australia (section D.3). D.1 Evidence on social value Survey evidence Moir (2013) assessed the knowledge contributed by 72 granted business method patents filed in Australia between 2003 and 2006. She used documented exchanges between applicants and patent examiners to identify the features of various patents that were integral to protection being granted. The study concluded that many of the features integral to a patent being granted were trivial in nature, and that it was difficult to discern any new knowledge in any of the 72 patented inventions. While the conclusions necessarily rely on the judgment of the author, some of the examples provided are compelling. For example: • an absence of features in an online employment register compared to earlier systems enabled the applicant to successfully establish an inventive step (Australian patent application number 2005234625) • generating a report on a server, rather than on site was the key feature that made a process for combining date-stamped photos and written reports sufficiently inventive for a patent (Australian patent application number 2003246060). It was further argued that some inventions received patent protection despite: • being a combination of well-known features. An example was an invention to teach children about finance that included features like credit advances, sickness insurance and buying shares. The examiner was unable to find documented evidence that said combining these features in the environment of pocket money is obvious • appearing to simply involve the use of a well-known process in a different context. An example was a rewards system for tickets, which was argued to already be a well-known process used in consumer loyalty schemes. EVIDENCE ON PATENTS: SOCIAL VALUE, ADDITIONALITY AND THICKETS 631 A similar US study examined 50 software patents to assess the degree to which they were obvious and incremental (Campbell-Kelly and Valduriez 2005). This was assessed using the number of previous patents and other information sources cited in the patent documents. The study focused on patents that had the most forward citations of all patents in the wider sample, as these were considered to be a proxy for best practice in the granting of patents. While the authors concluded that only two patents could be described as obvious, all were argued to be incremental in nature. Proxy measures of private and social value A further approach to better understand the social value of patents is to estimate the value of patented inventions using proxy measures. The OECD (2015a) proposed a set of measures that proxy for the private and social value of patented inventions (box D.1). The OECD argues that the revenue generated from a patented invention should be commensurate with the invention’s technological contribution to society, and therefore that the private and social value of patented inventions are closely related (OECD 2009a). While there is empirical support for a strong link between the private and social value of patents, there is also evidence that the link may be weaker in industries where innovation builds on previous innovations in an iterative and cumulative fashion (Hall, Jaffe and Trajtenberg 2005; Lanjouw and Schankerman 1999; Nagaoka 2005). Private and social value will differ to the extent that a patented innovation causes spillovers. Nonetheless, the measures proposed by the OECD can be a useful input into assessments of the distribution of patent value across inventions when considered alongside other measures. The Commission drew on patent application administration data to construct indicators of a patent’s social value (figure D.1). While no single measure provides definitive evidence on the value of patented inventions, as a collective they provide an indication of the distribution of value across patented inventions. Results suggest that the bulk of patents are of low social value. The results are consistent with empirical studies that find a highly skewed distribution of patent values, with value disproportionately concentrated within a small proportion of patents (Dahlin and Behrens 2005; Hall, Jaffe and Trajtenberg 2005; Schankerman and Pakes 1986). The evidence from these studies and the Commission’s analysis suggest that while the patent system may play an important role in promoting some socially valuable inventions, many patented inventions are of little value. 632 INTELLECTUAL PROPERTY ARRANGEMENTS Box D.1 Patent value measures • Forward citations. The number of times a patent is cited is considered to reflect the importance of the patent for subsequent technology and the social value of inventions (Hall, Jaffe and Trajtenberg 2005; Harhoff, Scherer and Vopel 2003; Trajtenberg 1990). Forward citations are considered to indicate the existence of downstream research efforts (OECD 2009a). There is evidence that forward citations, especially where they are weighted by their own number of citations, are associated with technological improvement (Benson and Magee 2015; Jaffe and Rassenfosse 2016; Moser, Ohmstedt and Rhode 2014; Trajtenberg 1990). The more forward citations, the higher the social value. • Citations to non-patent literature. Patents that cite non-patent literature (such as scientific papers) may contain more complex and fundamental knowledge and be higher value than patents that do not cite such literature (Branstetter 2005; Cassiman, Veugelers and Zuniga 2008). Harhoff, Scherer and Vopel (2003) provide empirical evidence that citations made to non-patent literature is predictive of private value as stated by patent holders and based on profit flow data. The more non-patent literature citations, the higher the social value. • Generality index. Patent generality indexes are based on the number and distribution of forward citations and the technology classes of the patents these citations come from. The measure has been used to identify general purpose technologies (Hall and Trajtenberg 2004). The higher the value of the generality index the higher the social value. • Radicalness index. Measured as a time invariant count of the number of WIPO technology classes in which the patents are cited by the given patent, but in which the patent itself is not classified. Sabrine (2015) concludes that the ‘radicalness’ of a patent has a positive effect on firm performance. The higher the value of the radicalness index the higher the social value. • Patent scope. The broader the scope of a patent, as proxied by the number of distinct technology classes it cites, the higher its potential technological and market value (OECD 2015a). The greater the patent scope the higher the social value. • Patent family size. The set of patents filed in several countries that are related to each other by one or several common priority filings is generally known as a patent family. Large international patent families have been found to be correlated with patent value, possibly because only valuable patents will be filed in multiple countries (Harhoff, Scherer and Vopel 2003; Lanjouw and Schankerman 2004). Patent family size is proxied by the number of patent offices at which the invention has been protected. The greater the patent family size the higher the social value. • Years a patent is in force. Some studies suggest that patents with higher private value are renewed for longer periods (Bessen 2006; Pakes 1986). As every renewal is costly and in many cases renewal fees increase over the life of a patent, patent renewal decisions reveal the willingness of the patent holder to pay for patent protection (Baron and Delcamp 2011). The higher the number of years a patent is in force the higher the social value. • Patent value index. A composite indicator based on forward citations, generality index, radicalness index, citations to non-patent literature and patent family size. The higher the value of the patent value index the higher the social value. EVIDENCE ON PATENTS: SOCIAL VALUE, ADDITIONALITY AND THICKETS 633 Figure D.1 Proxy measures for the social value of patented inventionsa,b,c a Data on forward citations, generality index, radicalness index, non-patent literature citations and family size were obtained for USPTO patents from the OECD’s patent quality indicators database. These patents were matched to patents granted in Australia by Patent Cooperation Treaty (PCT) number (for patents filed through the PCT) and using an OECD-prepared equivalence table that matches USPTO patents to IP Australia patents based on their priority documents. Where there were multiple USPTO matches for a single Australian patent (due to continuation or divisional applications) the match with the most forward citations was chosen. Recently granted patents were more likely to be matched — 56 per cent of patents granted in 2015 could be matched to a USPTO patent compared to 48 per cent in 2005. Data on patent scope and years patent in force were sourced from Intellectual Property Government Open Data (IPGOD). b Forward citations and the generality index are based on patents granted in Australia between 2005–2010. This is to provide a window for patents to receive forward citations. Years patent in force is based on patents granted in Australia between 1995–2005, to provide time for renewal. Remaining measures are based on patents granted in Australia between 2005– 2015. c The generality index is only calculated for patents that have at least one forward citation. The family size measure is positively biased: patents granted in Australia that have a USPTO match will necessarily have a patent size of at least two. To construct the patent quality index forward citations, non-patent literature and family size measures were normalised according to the maximum value of the measure in the same cohort (filing year and technology field). These indexes were corrected for extreme values. Patents with no forward citations were assumed to have a generality index of zero for the purposes of the quality index. All components were given equal weight. The quality index is only calculated for patents granted in Australia between 2005– 2010 (because it includes forward citation measures) and where the matched USPTO patent was granted. Source: Commission estimates based on IPGOD (2016 edition) and OECD Patent Quality Indicators database. 634 INTELLECTUAL PROPERTY ARRANGEMENTS D.2 Evidence on additionality Surveys of firms Survey evidence shows that patents are seldom the most important means for appropriating returns to innovations. In a series of surveys of manufacturing firms, lead-time and superior sales and service are generally nominated as the most important appropriation mechanisms for product innovations (table D.1). A survey of managers of large Australian firms between 2001–2006 found patents were considered the least effective appropriation mechanism on average for both product and process innovations (Jensen and Webster 2009). Other survey evidence suggests that patents are an even less important appropriation tool in the service sector (Blind et al. 2003). Table D.1 Appropriating the returns to product innovationsa Relative importance by means Survey (year) Country Ranking of mean importance 3 4 Yale (1982) Harabi (1988) US sales/service lead time Switzerland sales/service lead time patents secrecy secrecy patents Dutch CIS (1992) Netherlands lead time retaining employees secrecy patents Carnegie-Mellon (1993) US lead time secrecy/ sales/service complementary manufacturing Japan C-M Japan lead time patents SESSI/INSEE EFA (1993) France lead time patents StatCan Innovation (1999) Canada confidentiality trademarks agreement patents secrecy CIS 3 2000 (2000) EU12 lead time secrecy trademarks complexity Melbourne Institute (2001-2006) Gonzalez-Alvarez and Nieto-Antolin (2007) Australia know-how brand name lead-time secrecy Spain lead time complexity secrecy patents 1 2 patents sales/service secrecy complementary manufacturing secrecy complexity a There are differences in the wording of questions across surveys. For example in some surveys the question is phrased as what share of innovations are protected by the various appropriation mechanisms. Sources: Cohen et al. (2000); Hall (2009); Jensen and Webster (2009); Levin et al. (1987). The importance of patents in appropriating returns to innovations varies across industries. Surveys have found that patents are more important in the pharmaceutical industry, followed by specialised machinery and instruments and other chemicals (table D.2). These findings support earlier survey research that suggests innovation in the pharmaceutical and chemical industries is more reliant on patent protection than innovation in other industries (Mansfield 1986; Mansfield, Schwartz and Wagner 1981). The industries where patents are EVIDENCE ON PATENTS: SOCIAL VALUE, ADDITIONALITY AND THICKETS 635 found to be important are typically characterised by large sunk costs. While in the services sector patents seem less important overall, there is evidence that the business service, telecommunications and media service industries are more reliant on patents than others (Baldwin et al. 1998; Blind et al. 2003; Hipp and Herstatt 2006). Table D.2 Appropriating the returns to product innovations Relative importance by industry Survey (year) Country Industry preferences for patents in descending order Yale (1982) Harabi (1988) US Switzerland pharmaceuticals, plastics, chemicals, steel, oil research labs, machinery, chemicals, watches, paper Dutch CIS (1992) Netherlands pharmaceuticals, chemicals, instruments, rubber and plastics, oil Carnegie-Mellon (1993) US SESSI/INSEE EFA (1993) StatCan Innovation (1999) CIS 3 2000 (2000)a France Canada EU12 pharmaceuticals, medical instruments, special machinery, computers, chemicals pharmaceuticals, instruments, transport, chemicals, machinery, paper machinery, electronics, pharmaceuticals, communications, instruments, chemicals, motor vehicles transport, instruments, chemicals a Pharmaceuticals and chemicals are combined. Source: Hall (2009). The evidence on the importance of patents across industries is consistent with evidence at the technology level. • Firms that use mostly explicit or codified technologies, which characterises many innovations in the machinery and pharmaceutical industries, are more likely to use patents (Gonzalez-Alvarez and Nieto-Antolin 2007). These technologies can be more easily reverse engineered, providing the innovator with a shorter lead time in the absence of patent protection. • Firms that supply products that rely on only a handful of patents (called ‘discrete products’, such as some pharmaceuticals) generally report that they patent to exclude competitors and prevent litigation, whereas firms that supply products that rely on a large number of patents (called ‘complex products’, such as information technology) are more likely to patent for cross-licensing and negotiation purposes (Cohen et al. 2002). • Patents are a less important appropriation tool for process innovations than for product innovations (Hall 2009). Firms that mostly develop process innovations are more likely to use trade secrets (Arundel 2001; Byma and Leiponen 2007; Combe and Prister 2000). 636 INTELLECTUAL PROPERTY ARRANGEMENTS Economic models In addition to using surveys, economic researchers have used theoretical and empirical models of innovation to shed light on the relationship between patenting and innovation. This literature provides mixed evidence on the relationship between patents and innovation (box D.2). Box D.2 Economic literature on patents and R&D Estimating the causal relationship between patenting and innovation is challenging. • Measures of innovation rely on proxy measures such as research and development (R&D) expenditure. As an input into innovation, this is an imperfect measure. • Systematic differences between firms that patent and firms that do not patent bias estimates. A potential solution is to find a variable that randomly affects the probability of a firm receiving a patent, and use this source of variation to conduct a more robust comparison of R&D expenditure across firms with and without patents. There are few variables that randomly affect the probability of receiving a patent (one example might be the ‘leniency’ of a randomly assigned patent examiner). • Most firms conduct R&D before applying for a patent. It is the prospect of receiving patent rights that influences behaviour. This makes it harder to test the effect of the patent system on innovation. So even if there is a variable that randomly affects the probability of receiving a patent, the outcome associated with this variable has to happen before innovation occurs. Despite these challenges, a body of literature examines the impact of patent protection on R&D using economic models. • Evidence at the industry level. Using a structural model that combines responses from a US survey with R&D data, Arora, Ceccagnoli and Cohen (2008) estimates the ‘patent premium’ (increase in an invention’s value due to a patent) and the effect this has on R&D across sectors. Results suggest the highest premiums are in health-related industries like pharmaceuticals and biotechnology. Arora and Athreye (2012) obtains similar results using UK data. Some studies conclude that patents are likely to provide little if any boost to software innovation (Bessen and Hunt 2007; Lemley 2013). • Evidence at the technology level. Theoretical evidence suggests patents promote innovation in discrete technologies, but under certain conditions can decrease welfare where innovation is of a cumulative or complementary nature (Bessen 2004; Scotchmer 1991). Moser (2005) analysed the records for 15 000 European innovations displayed at two international fairs during the 19th century. Results imply that patents encourage innovations for technologies that cannot be protected by trade secrets. Palangkaraya, Webster and Jensen (2011) compared European and Japanese patent examination decisions between 1990 and 2004. They find that false positives are more likely in sectors where the speed of technological change is fast. The above evidence suggests that in some industries patents are less important for promoting R&D. However, evidence from the literature shows that even in some of these industries there are some firms that appear to rely on patents. Arora, Ceccagnoli and Cohen (2008), for example, estimates that the patent premium for electronics (40 per cent) is sufficient to stimulate R&D. Using data from a survey of Australian inventors who submitted a patent application between 1986 and 2005, Jensen, Thomson and Yong (2009) found evidence that the effect on R&D from granting a patent is similar across technology areas. EVIDENCE ON PATENTS: SOCIAL VALUE, ADDITIONALITY AND THICKETS 637 While some studies conclude that patents do little to promote innovation in general, others find a positive and significant effect. Evidence from models that use industry-level data supports the findings from the survey literature, with patents found to be most important for promoting innovation in sectors like pharmaceuticals and biotechnology (Arora and Athreye 2012; Arora, Ceccagnoli and Cohen 2008). There is, however, evidence that patents are also important in other sectors, such as computers and electronics (Arora, Ceccagnoli and Cohen 2008; Bessen 2006; Jensen, Thomson and Yong 2009). D.3 Evidence on patent thickets A patent thicket is essentially a cluster of patents in a given technology space. Shapiro described them as: … a dense web of overlapping intellectual property rights that a company must hack its way through in order to actually commercialise new technology. (2004, p. 120) Thickets can impede innovation by making it difficult for firms to enter and compete in a given market, including by increasing the costs incurred in negotiating access to multiple patented inventions. These costs are greater when the ownership of patents for a given technology is more dispersed. Overcoming patent thickets can be especially difficult for small-to-medium enterprises (SMEs) and potential market entrants. A number of empirical studies find that patent thickets inhibit market entry, especially for SMEs (Cockburn and MacGarvie 2009, 2011; Hall, Helmers and Graevenitz 2015; IPO 2013). Identifying patent thickets Some researchers and patent offices identify patent thickets using a measure called a ‘triple’ (Graevenitz, Wagner and Harhoff 2011, 2013; Hall, Helmers and Graevenitz 2015; IPO 2013). A triple consists of three firms that each hold patents that cite patents held by the other two firms. The relative value of the patents held by any two firms in the triple depends on the actions of the third, which makes bargaining more difficult. This difficulty is compounded where there are other closely related patents in the same technology space. Indeed, the UK IPO (2013) notes that a triple is the most basic type of a patent thicket — the boundaries of a thicket can extend from a triple to encompass an even larger number of firms. While thickets can also be measured using the raw number of patents in a technology field, this approach does not account for the bargaining difficulties that arise from overlapping patent rights. Researchers have found evidence that patent thickets, as measured by triples, can inhibit market entry. Hall, Helmers and Graevenitz (2015), for example, found evidence that increases in the size of thickets can decrease the likelihood of market entry by up to 20 per cent (chapter 7). The Commission has used administrative data for patents lodged after 2003 to identify triples. It has estimated triples at both the World Intellectual Property Organization 638 INTELLECTUAL PROPERTY ARRANGEMENTS (WIPO) technology field level, and at the WIPO sector level. The broad steps taken in identifying triples are: 1. within each WIPO technology/sector classification, identify pairs of firms where both firms hold patents that cite the other’s patents (a ‘double’) 2. identify groups of three firms — again within the same WIPO technology/sector — where each firm has a double with each of the other two firms (forming a triple). Use of patent administrative data likely underestimates the number of triples. This is because identifying triples is made more difficult by the likelihood that a patent holder has different name variations in administrative data. Variations may arise due to spelling error, punctuation variation, different names by geographical area, and extensive networks of subsidiaries. As a result, the set of firms that can be matched by name in the steps above is limited. The ability to identify triples is also limited by patent citation data only being available for patents filed after 2006. The Commission’s analysis The Commission identified 154 triples at the technology field level, and 366 triples at the sector level (table D.3). The estimated number of triples at the technology level is lower because all patents in the triple must belong to the same field of technology — a more stringent restriction than belonging to the same sector. Using either measure as a proxy to gauge the extent of patent thickets requires caution. • The number of triples at the sector level is likely to be an overestimate of patent thickets because some triples will include unrelated patents. • Conversely, the number of triples at the technology level is likely to be an underestimate of patent thickets because some products are underpinned by multiple technologies. For example, large pharmaceutical firms hold patents in a number of different technology fields — including pharmaceuticals, organic fine chemistry and biotechnology — each of which are considered separately, rather than collectively, in attempting to measure the extent of patent thickets. The technology fields with the most identified triples are digital communication, basic materials chemistry and organic fine chemistry. For the electrical engineering and chemistry sectors, there are many more triples identified at the sector level. Four examples of thickets that have formed around triples — including two that are comprised of an innovation patent — are illustrated in figure D.2. EVIDENCE ON PATENTS: SOCIAL VALUE, ADDITIONALITY AND THICKETS 639 Table D.3 Estimated thickets by sector and technology fielda,b Sector Technology field Chemistry Thickets identified at the sector level Thickets identified at the technology level 222 Basic materials chemistry Organic fine chemistry 29 26 Biotechnology 12 Food chemistry 1 Pharmaceuticals Macromolecular chemistry, polymers 1 1 Electrical engineering 114 Digital communication Electrical machinery, apparatus, energy Audio-visual technology Instruments 52 5 1 14 Medical technology Optics Other fields 12 2 11 Civil engineering 7 Furniture, games Other consumer goods 1 1 Mechanical engineering 5 Thermal processes and apparatus 1 Engines, pumps, turbines 1 Transport 1 Total 366 154 a Thickets are identified at the WIPO technology and sector levels using the ‘triples’ measure. The difficulties involved in identifying triples mean that the Commission has likely underestimated the number of triples and thus thickets. As noted in the text, the estimated number of triples at the technology level is lower because all patents in the triple must belong to the same technology, a more stringent restriction. b The estimated number of thickets differs from the estimate in the draft report because an extra year of IPGOD data became available after the draft was released. Sources: Commission estimates using IPGOD database (2016 edition) and unpublished IP Australia citations data. 640 INTELLECTUAL PROPERTY ARRANGEMENTS Figure D.2 Schema of four Australian patent thicketsa a The firms on either side of an interconnecting line cite each other’s patents — that is, each firm pair represents a bilateral patent relationship. The thicket is initially identified by the interrelationships between firms that are part of ‘triples’ — three firms that each hold patents that cite patents held by the other two firms. The dark green circles denote firms that are either part of a triple relationship or a broader relationship that involves more than three firms. The light green circles denote bilateral patent relationships. The red interconnecting line indicates that the bilateral patent relationship includes at least one innovation patent. For these examples, the Commission checked the documentation for each patent in the thicket and judged that the patents were related to one another. Some firms were omitted from some of the above thickets to aid with presentation. Sources: Commission estimates using IPGOD database (2016 edition) and unpublished IP Australia citations data. EVIDENCE ON PATENTS: SOCIAL VALUE, ADDITIONALITY AND THICKETS 641 E Copyright evidence The Commission’s recommendations to reform Australia’s copyright arrangements were strongly contested by inquiry participants. Many Australian authors, publishers and booksellers opposed the repeal of parallel importation restrictions (PIRs) on books and the proposed introduction of a fair use exception. In arguing against the Commission’s recommendations, some participants have undertaken or commissioned empirical analyses of the proposed costs and benefits of reform. For example: • the Australian Publishers Association (APA) (sub. DR435) and the Australian Booksellers Association (ABA) (sub. DR466) compared the price of books sold in Australia and overseas, concluding that removal of PIRs in Australia would provide few benefits to Australian consumers • many other participants made other claims about the potential consequences of repealing PIRs, drawing parallels to other reform episodes, such as removal of PIRs in New Zealand, and removal of the restrictions on importing sound recordings in Australia • the Copyright Agency, APRA AMCOS, Foxtel, News Corporation, PPCA and Screenrights Australia (subs. 133 & DR510) commissioned a report by PricewaterhouseCoopers (PwC) on fair use (PwC 2016). Many of these analyses have significant limitations. This appendix considers the available analysis and presents the Commission’s own updated analysis of the price effects of PIRs (section E.1). Section E.2 examines the usefulness of comparing changes in the New Zealand book industry to inform analysis of changes that may occur in Australia. Section E.3 considers studies that evaluate the impacts of fair use and section E.4 examines the potential impacts of some of the Commission’s proposed copyright reforms. E.1 Analysis of book prices Book price comparisons submitted by inquiry participants Several inquiry participants provided their own price comparisons of book titles across different countries. They argued that book prices were not higher in Australia, and thus the removal of PIRs would not offer benefits to Australian consumers. COPYRIGHT EVIDENCE 643 • The ABA (sub. DR466) compared the 2 June 2016 price of about 75 books sold in Australia and in the US and/or the UK. They included new release, cookery, international new release and new release wall display titles stocked by Readings Books and Music Carlton, as well as some bestsellers from the Nielsen BookScan database. • The APA (sub. DR435) compared the price of 200 books published in Australia, the US, the UK, Hong Kong and New Zealand using prices obtained between 1 November 2015 and 31 March 2016. They included titles from adult literary and general fiction, biography, politics, young adult and illustrated children’s genres. Both studies converted overseas recommended retail prices (RRPs) to an equivalent Australian dollar price, and made other adjustments as necessary, to enable comparisons. For instance, the ABA removed GST from Australian RRPs (US and UK RRPs do not include sales tax) and converted foreign RRPs to Australian dollars using exchange rates at 2 June 2016. The APA converted foreign RRPs to Australian dollars using exchange rates averaged over 12 years. From their analyses, both the APA and ABA conclude books are not, on average, more expensive in Australia, and in many cases are cheaper. The ABA found about 54 per cent of books sold in both the UK and Australia were cheaper in Australia, and for books sold in both the US and Australia, prices were lower in Australia for 83 per cent of titles in the sample. Likewise, the APA concludes that book prices in Australia are comparable — sometimes cheaper — than the price of overseas editions. The APA analysis sought to provide evidence that Australian editions were relatively price comparable irrespective of their time since first publication or retail price. For instance, they argued there was no relationship between the affordability of Australian books and their publication date (sub. DR435, att. 1, figure 1), and that the price of Australian books remains comparable with overseas editions (on average) regardless of whether the book retails at A$15 or $A40 (sub. DR435, att. 2, figure 2). However, each of the studies provided to the Commission suffer from methodological problems that hamper their value for policy analysis. • 644 All of the studies provided by participants were based on very limited samples of titles that are not representative of the broader retail book market. While in 2015, more than 550 000 different titles were sold in Australia, the ABA drew its conclusions from only 75 titles, while the APA did the same with less than 200 titles. INTELLECTUAL PROPERTY ARRANGEMENTS • Nearly half of the matches used by the ABA compare the price of paperbacks in Australia with the price of hardcovers in the US or UK. As a result, Australian books may appear price comparable but only because consumers are actually purchasing a different product. For instance, the data shows Rich Cohen’s ‘Sun and the Moon and the Rolling Stones’ had a lower paperback price in Australia than the hardcover in the US and UK (sub. DR466), when in fact the Australian hardback version is more expensive than in the UK or US (Readings 2016). • Neither the APA nor ABA studies compared the Average Sale Prices (ASPs) of books in different markets, instead solely comparing RRPs. In some markets only RRP data is available, but where available, ASPs better reflect the actual prices paid by consumers. • Both studies focused on the average market price of Australian books relative to the average overseas price. However, in assessing the impact of PIRs, comparing the prices of books that could be sourced more cheaply overseas more accurately demonstrates the welfare effects of the restrictions. If removing PIRs means that some books can be sourced more cheaply (either because the threat of imports reduces domestic wholesale pricing, or from overseas directly), this will benefit book retailers and ultimately consumers. Updating the Commission’s previous analysis The Commission has sought to undertake a comprehensive comparison of book prices in Australia and overseas, consistent with the approach used in the Commission’s 2009 study Copyright Restrictions on the Parallel Importation of Books (PC 2009). This updated analysis finds substantially similar results to the Commission’s previous work (PC 2009). Namely, that at any given time, a significant proportion of book titles available in Australia could be purchased at a lower cost in the UK or the US. As the Commission noted in its 2009 study, the results of these comparisons are an indication of gaps between the observed retail prices of books in the different countries. Data In updating its previous analysis for this inquiry, the Commission has used title, volume and pricing data on the top 5000 titles sold in Australia, the UK and the US between 31 May 2015 and 28 May 2016, drawn from the Nielsen Company’s BookScan database. In Australia, the top 5000 books sold represent over 60 per cent of all books sold on a volume basis, and nearly 55 per cent of all books sold on a value basis. In the UK, the top 5000 books sold represent over 50 per cent of the market by volume, and over 40 per cent by value, while in the US the top 5000 books sold represent over 40 per cent of the market by volume (value data unavailable). COPYRIGHT EVIDENCE 645 For each book, the BookScan database includes information on: • the title, author, publisher, imprint (the publisher’s trading name) and the International Standard Book Number (ISBN) • sales — the total volume (for all countries) and value (for Australia and the UK only) of books sold in the period covered by the BookScan data • prices — RRP and ASP (for Australia and the UK only) • the format • the publication date and country. Matching books between markets To make price comparisons, books from Australia are matched with books from the US and the UK. When identical book editions are sold in multiple markets, the matching process is simple and done using the ISBN — a unique identifier assigned to each edition of a book. In the Commission’s dataset, matching by ISBN results in matches for 909 and 109 Australian titles with the UK and US respectively. An alternative approach is required to find appropriate matches for books where identical editions are not sold in multiple markets. Different editions of the same book can have minor variations in the wording of titles and authors’ names (for instance, the word ‘colour’ in Australian titles is frequently spelled ‘color’ in US titles). Multiple editions of a book can also have different covers, formats (hardcover or paperback), paper quality and/or dimensions. In some cases, these differences are sufficiently significant such that comparing prices for these books would be misleading. For books without an ISBN match, the Commission matched books by title and author using an ‘approximate string matching’ algorithm to account for variations in the wording of titles and author names. The resulting set of matches were further restricted so that matches with different publishing formats (for example, hard cover as distinct from softcover), or books that were published more than two years apart or had missing RRPs (this applies for the US matches only) were excluded. Where more than one match was found, the lowest cost edition was used. Matches were manually checked to ensure that clearly incomparable editions of books (such as a children’s book with a t-shirt included when sold in one country but not another) were not included. The remaining variation between matched books is restricted to differences in content (such as cultural reediting) and small differences in dimensions. The resultant sample (which includes both ISBN and non-ISBN matches) includes 483 matches between Australian and US titles, and 1126 matches between Australian and UK titles. This sample is larger than the available sample if only ISBN matching is used (however, results for the ISBN-only sample are presented for sensitivity analysis), and 646 INTELLECTUAL PROPERTY ARRANGEMENTS significantly larger than the samples used in all book comparisons provided by inquiry participants. Comparing prices PIRs prevent Australian booksellers from importing books from overseas, either from foreign wholesalers or publishers. Therefore, the best assessment of the impact of removing PIRs would compare the wholesale book prices currently paid by Australian retailers to the wholesale prices they would pay if they could purchase books from overseas. The benefit to Australian consumers from removing PIRs would then depend on: • the number of books that Australian retailers can source more cheaply overseas • the wholesale price differential Australian retailers face between Australian and foreign publishers • the extent of retail competition resulting in wholesale book savings being passed through to consumers (given the number of book retailers in Australia and other sources for consumers to purchase books, assuming most savings are passed on to consumers is reasonable). Unfortunately, data on wholesale prices faced by Australian retailers are unavailable. The price data available in the Nielsen BookScan database relates to: • RRPs are set by publishers and frequently printed on the back of books. Publishers sell books to booksellers, often on a ‘sell or return’ basis, at a discount to the RRP, with discounts often determined by volume and other factors. RRPs may proxy wholesale book prices if publishers calculate wholesale prices at a consistent discount to RRPs across countries. • ASPs are the average price that a book is sold for by retailers. ASPs will represent the better proxy for wholesale prices if the profit margins on books are similar across countries. 1 The Commission estimated price differentials using both RRP and ASP. • Australian RRPs were compared to UK RRPs and US RRPs for matched books. • Australian ASPs were compared to UK ASPs for matched books (ASP was not available for US books). The Commission made some adjustments to overseas prices (RRPs and ASPs) to make them comparable to Australian prices. In particular, overseas prices were converted into AUD and adjusted for differences in applicable taxes. 1 This would also mean that the level of service provided in stores is similar across countries as there is more scope for discounting if less service is provided to the book buyer. COPYRIGHT EVIDENCE 647 An estimate in the cost of freight from the comparison country to Australia would further improve the comparability of prices between countries. However, several issues arise in determining an appropriate freight adjustment as freight costs depend on: where books are printed (which may be outside the UK/US); the number of books imported at any one time; and whether they are shipped by air or sea. For this reason, the Commission has not included estimates of freight costs in its average comparisons. Exchange rate fluctuations The Commission converted foreign book prices to AUD using average exchange rates for the period covered by the BookScan data: 31 May 2015 to 28 May 2016. Over this period the Australian dollar was, on average, worth about 73 US cents (US $0.73), and about 49 UK pence (£0.49). These exchange rates are about 18 per cent lower (USD) and 9 per cent lower (GBP) than the average exchange rate over the 10 years from July 2006 to June 2016 (figure E.1). However, exchange rates frequently change, and this can have a noticeable impact on relative book prices. 2 Therefore, results using the highest and lowest exchange rate over the sample period were also calculated. Sales taxes Consistent with the approach of the ABA (sub. DR466), sales taxes were removed from all prices. Australia’s 10 per cent GST was excluded from Australian prices. As there is no equivalent tax levied in the UK, and because US states apply sales taxes on top of RRP (PC 2009), prices in these countries were not adjusted. 2 Under current policy setting (with PIRs) exchange rates influence the price of local book prices to the extent to which consumers can (or are willing to) import books from online retailers. If PIRs were removed, exchange rate fluctuations would have a larger impact on local book prices as some books would be sourced by Australian retailers from overseas. The size of this effect would depend on: (a) the proportion of books sourced overseas by booksellers (and this could change in response to large changes in exchange rates); (b) the share of final sales prices that are attributable to wholesale book prices; and (c) how quickly changes in costs were passed to consumers. 648 INTELLECTUAL PROPERTY ARRANGEMENTS Figure E.1 Australian dollar exchange rates July 2006 to June 2016 1.2 US dollar 1.0 0.8 0.6 UK pound 0.4 Analysis period 0.2 0.0 Jul-06 Jul-07 Jul-08 Jul-09 Jul-10 Jul-11 Jul-12 Jul-13 Jul-14 Jul-15 Source: RBA (2016). Results For the Commission’s sample of matched books, the average Australian ASP exceeds the average UK ASP by about 20 per cent, or $2.73, while the average RRPs are similar in Australia to the UK and US (the average Australian RRP is two per cent higher than the average UK RRP and one per cent higher than the average US RRP) (table E.1). The difference between these results are due to the stronger discounting that occurs in the UK compared to Australia. For the matched book titles, the average ASP is 30 per cent lower than the RRP in the UK, compared with 18 per cent lower for Australia (a slightly larger difference than for the whole sample of books in each country). However, these average price differentials obscure significant differences in price at the individual book level and that a substantial number of books are cheaper overseas (table E.1, figure E.2). For example, over three-quarters of books are more expensive in Australia than in the UK using ASP. For these books, the average price differential is $4.20 (30 per cent of the average ASP in the UK). Similarly, using RRP, approximately 50 per cent of book titles are cheaper in the UK than they are in Australia, and around 45 per cent of book titles are cheaper in the US. The average price differential for these more expensive books is $3.41 and $4.85 for the UK and US, respectively. COPYRIGHT EVIDENCE 649 Table E.1 Comparisons of matched book pricesa 31 May 2015 to 28 May 2016 Number Average price of books difference (%) Average Share of books that price are more expensive in Australia (%) difference (AUD) Average price difference for more expensive books (AUD) Australia – UK RRP ASP Australia – US RRP 1 113 1 126 2.4 19.6 0.48 2.73 51.1 77.7 3.41 4.20 483 1.2 0.24 45.5 4.85 a UK and US prices are converted to AUD using exchange rates of US dollar (USD) 0.7312 and UK pound (GBP) 0.4892. The averages presented in this table are simple averages (that is they are not weighted). The average price difference (%) is calculated using the average Australian price and the average foreign price. Sources: Commission estimates based on the Nielsen BookScan databases. Exchange rate data from RBA (2016). There is more dispersion in price differences between Australia and the US than there is between Australia and the UK. The middle 50 per cent of the book price differences between Australia and the UK was in the range of -6 to 9 per cent of the average UK RRP, while the middle 50 per cent of book price differences varied between -18 and 15 per cent of the average US RRP (figure E.2). 650 INTELLECTUAL PROPERTY ARRANGEMENTS Figure E.2 Range of price gaps for like-with-like comparisonsa 31 May 2015 to 28 May 2016 Median 50 Mean 40 Per cent 30 20 10 0 -10 -20 -30 AUS-UK (ASP) AUS-UK (RRP) AUS-US (RRP) a The edges of the boxes represent the range of price gaps between the 25th and 75th percentiles of the sample. The ‘whiskers’ extending out from the boxes show the values for the 10th and 90th percentiles. Source: Commission estimates based on the Nielsen BookScan database, using exchange rate data from RBA (2016). How have price differences in each market changed over time? This analysis updates the Commission’s analysis in 2009. In that study, for two yearly time periods (2007–2008 and 2008–2009), the Commission estimated price differences using comparisons adjusting for GST. Notably the AUD-USD and AUD-GBP exchange rates used for 2008-09 are similar to those used in the updated analysis (USD 0.74 compared to USD 0.73 and GBP 0.46 compared to GBP 0.49). Between 2008-09 and 2015-16, the average difference between Australian and UK and US RRPs has decreased by approximately 10 percentage points (figure E.3). In contrast, ASP differences between Australia and the UK decreased by 5 percentage points over that time period. Changes in the composition of the Australian retail market, including increased competition from online book retailers (including Ebooks) may have affected Australian retailers more strongly than foreign retailers. COPYRIGHT EVIDENCE 651 Figure E.3 The extent to which Australian prices exceed UK/US pricea Per cent 40 35 2007-08 2008-09 2015-16 30 25 20 15 10 5 0 US - RRP UK - RRP UK - ASP a 2015-16 year is for end of May 2015 to end of May 2016. Sources: Commission estimates based on the Nielsen BookScan databases; PC (2009); RBA (2016). Sensitivity analysis Sensitivity analysis was undertaken to analyse how results change when key assumptions are varied. These assumptions include: • the type of matching used (ISBN versus the full sample) • the exchange rate used (the maximum and minimum over the year rather than the average) • the use of simple versus weighted averages. Price comparisons based on only ISBN-matched books produced very similar results to the full sample for the UK (this is unsurprising given that 80 per cent of UK matches were ISBN-matches) (table E.2). The results for the US matches indicate that US RRPs are on average higher than Australian RRPs. However, this result should be interpreted with caution because it is based on a small sample (109 titles) that may not be representative. 652 INTELLECTUAL PROPERTY ARRANGEMENTS Table E.2 ISBN matches compared to full sample ISBN matches only Number of books Average price difference (%) Full sample Share of books Number more expensive of books in Australia (%) Price difference (%) Share of books more expensive in Australia (%) Australia – UK RRP 897 -1.4 47.8 1 113 2.4 51.1 ASP 909 15.9 77.1 1 126 19.6 77.7 109 -6.8 38.5 483 1.2 45.5 Australia – US RRP Sources: Commission estimates based on the Nielsen BookScan databases; exchange rate data from RBA (2016). Compared with the full sample, ISBN-only matches show less price dispersion. 3 The middle 50 per cent of price differences between Australia and the US is between -14 and 5 per cent of the average US RRP (figure E.4). Figure E.4 Range of price gaps for ISBN matchesa 31 May 2015 to 28 May 2016 Median 50 Mean 40 Per cent 30 20 10 0 -10 -20 -30 AUS-UK (ASP) AUS-UK (RRP) AUS-US (RRP) a The edges of the boxes represent the range of price gaps between the 25th and 75th percentiles of the sample. The ‘whiskers’ extending out from the boxes show the values for the 10th and 90th percentiles. Source: Commission estimates based on the Nielsen BookScan database, using exchange rate data from RBA (2016). 3 Price dispersion for the full sample is presented in figure E.2. COPYRIGHT EVIDENCE 653 Altering the exchange rate assumptions has a noticeable impact on the price comparison results (table E.3). Using the maximum (minimum) exchange rate for the period, results in the average price difference being about 5-10 percentage points higher (lower). Table E.3 Varying exchange rate assumptions Books Highest exchange rate Average exchange rate Lowest exchange rate number % % % 1 113 1 126 14.1 33.3 2.4 19.6 -5.1 10.8 483 8.1 1.2 -5.0 Australia – UKa RRP ASP Australia – USb RRP a The high, mid and low AUD-GBP exchanges rates are 0.5450, 0.4892 and 0.4532 respectively. b The high, mid and low AUD-USD exchanges rates are 0.7812, 0.7312 and 0.6867 respectively. Sources: Commission estimates based on the Nielsen BookScan databases; exchange rate data from RBA (2016). The average price differentials in the base set of results are calculated using simple averages (not weighted averages). Weighting by volume or by value (which gives a larger weighting to high-selling and more expensive books) reduces the differences in the average ASP in book titles between Australia and the UK by 5-8 percentage points, though UK book titles continue to be significantly cheaper on average. There is only a marginal effect on the differences in RRP for both US and the UK comparisons (table E.4). The smaller price difference that exists when using weighted (rather than unweighted) ASPs indicates that in Australia, larger discounts tend to be applied to books with the highest sales volume (most likely because book sellers face the most competition on these books). The fact that the ASP reflects competitive pressures and is a price that is actually paid by consumers suggests that it is a more appropriate measure to use for price comparisons than RRP. 654 INTELLECTUAL PROPERTY ARRANGEMENTS Table E.4 Australia – UK RRP ASP Australia – US RRP Average price differentials — Simple averages versus weighted averagesa Number of books Simple average Weighted, by volume Weighted, by value number % % % 1 113 2.4 -0.1 -0.5 1 126 19.6 11.9 14.5 483 1.2 -0.0 0.1 a Weighted average price differences are calculated as the difference between the weighted average of Australian and foreign prices. Two sets of weights are used — the volume and value of Australian sales. Sources: Commission estimates based on the Nielsen BookScan databases; exchange rate data from RBA (2016) Adding freight costs to estimate welfare impacts The price differentials presented in the previous sections represent a simple comparison of the prices paid for books by consumers in their home country. Freight costs need to be incorporated into this analysis to assess the welfare impacts of removing PIRs. 4 Freight costs are hard to calculate because they depend on a multitude of factors (such as the volume shipped, the frequency of shipping and when in a month the goods are shipped). The Commission has therefore considered three possible price points for freight ($0.50, $2.00 and $5.00 freight per book) to analyse how different freight costs affect the welfare impacts of removing PIRs. 5 These rates were applied to UK ASPs in the Commission’s matched sample of books to proxy the price that Australian retailers would be able to sell books for if PIRs were removed. 6 Australian consumers would have paid around $15 million less in 2015-16 for the 1100 books with a UK match if freight costs were $2.00 per book and around $25 million less if freight costs were only 50 cents per book (table E.5). Indeed, freight costs would need to be in the order of $10 per book for there to be no improvement in Australian welfare. These numbers are an underestimate because not all books that can be sourced 4 Some freight costs are already included in the BookScan price data. For example, some Australian books (typically children’s ‘board books’) are printed overseas and sold in Australia by Australian publishers. Similarly, books printed in Australia need to be shipped from the printer to the publisher and from the publisher to booksellers around Australia. 5 The Commission has estimated lower and upper bounds for book freight costs, based on both sea and air freight modes. 6 US matched books were not used to estimate welfare impacts because the ASP (the price paid by consumers) was not available. COPYRIGHT EVIDENCE 655 more cheaply from overseas are included in the Commission’s matched sample. These books not in the sample can either be: • sourced more cheaply from the UK, but no match was found in the Commission’s analysis because the book is not in the top 5000 books sold by volume in either Australia or the UK and is thus not in the BookScan database • sourced more cheaply from a country other than the UK (such as the US). Table E.5 Welfare impacts of removing PIRs Shipping costs per book Number of books cheaper if sourced from UK Change in Australian welfare number $m 0.50 812 24.7 2 5 611 276 15.3 5.9 $AUD Source: Commission estimates. Conclusions It is the Commission’s view that removal of Australia’s PIRs would put downward pressure on prices for a significant number of books. This would lead to a reduction in retail book prices (which would benefit consumers) and promote efficiency in the Australian publishing sector. Indeed, the analysis presented in this appendix suggests the number of books that could be sourced more cheaply overseas could be substantial, leading to an annual savings of around $25 million for Australian consumers. Further, if PIRs are removed and ultimately only a small selection of books can be sourced more cheaply overseas, there would be no additional costs for Australians — Australian retailers would continue to source their books from Australian publishers. E.2 The New Zealand book industry case study During earlier inquiries into reforming Australia’s PIR arrangements, the Australian publishing industry has argued New Zealand’s experience with removing PIRs on books in 1998 resulted in a reduction in New Zealand publishing, fewer books available to New Zealand consumers, and no real reduction in book prices. The claims made by publishers and authors during this inquiry are no different: How is it wise then to adopt a system that prevails nowhere else on the planet except New Zealand? They adopted it in 1998, and there it has been a resounding failure. The New Zealand book industry is in continuing, long-term decline in consequence, and book prices have not 656 INTELLECTUAL PROPERTY ARRANGEMENTS dropped to anywhere near the same extent as they have here in Australia, where books are now 18 per cent cheaper than they are across the Tasman (Flanagan 2016). If in 1998 you could have foreseen the very different experiences of the book industries in NZ and Australia, would you still abandon territorial copyright? Or is the Productivity Commission of the view that the decline in New Zealand publishing is to be welcomed, as this is the inevitable outcome of the efficient allocation of resources into higher value activities than writing? (Heyward 2016) The New Zealand industry is routinely cited as best practice. The facts supplied by AC Nielsen Bookscan prove the contrary. While since 2008, prices in that market have fallen by 14 per cent, the reality is the range of available books has reduced by 34.5 per cent and the volume of sales has declined by 17.9 per cent. In comparison, Australian book prices have come down by 25 per cent during the same period. New Zealand is a case study in how to kill off a highly competitive and commercial industry. (Adler 2016) These arguments were raised with, and considered by, the Commission in its 2009 analysis of PIRs (PC 2009, pp. C.4-10). At the time of that study, the New Zealand industry had operated without PIRs for around 10 years, and numerous reviews had concluded their removal had not negatively impacted the publishing sector as alleged (box E.2). Box E.1 Earlier reviews of the New Zealand book reforms ‘Opponents to the removal of PIRs in New Zealand predicted a flood of cheap imports, and warned of trade sanctions. Neither has materialised, and the effect of introducing parallel importing, to date, appears muted. However, although tangible impacts on volumes and prices have been small, the possibility of parallel importing has exerted pressure on the market place … Delivery speed has improved, with customer orders able to be filled in a matter of days instead of weeks or months. Access to a full range of titles is possible, with availability enhanced by electronic links with suppliers around the world.’ (NZIER 2000, p. 269) ‘… there was no evidence of any substantial detriment to the financial performance or investment activity … income earned by publishers has risen substantially from 1998 to the present while export growth in this sector since liberalisation has exceeded overall export growth.’ (NECG 2004, p. 6) ‘NECG’s assessments of various indicators suggest that there are no adverse impacts on investment in this industry from the liberalisation of parallel importing. Many publishers that were interviewed noted that insofar as there were any major impacts from parallel importing, they arose from the greater availability of remainders.’ (NECG 2004, p. 37) ‘There is agreement among industry that book sellers will generally stock books from New Zealand publishers if they have the titles in stock. Commercial parallel importation is predominantly of niche titles and the bulk purchase of discounted backlist titles (especially bestsellers from the United States).’ (LECG 2007, p. 17) ‘… the bulk purchase via parallel importation of backorder titles (and end of line remaindered books) has allowed book sellers to provide books to consumers at prices that they would not otherwise have been able to. Book sellers do not see this discount selling affecting their sales of new release titles and full price books, and believe that parallel importing has allowed them to service a new market for books (at lower price points).’ (LECG 2007, p. 19) ‘The book industry in New Zealand appears to be growing with no major detrimental impacts as a result of parallel importing … in the last ten years the number of children’s titles published has trebled with significant export growth in these products.’ (MED 2008) COPYRIGHT EVIDENCE 657 Since the Commission’s 2009 study, the New Zealand publishing industry has changed, but to ascribe those changes to the removal of PIRs is misleading and inaccurate. For example, in 2013, multinational publishers Pearson Education and Hachette closed their New Zealand publishing operations; the latter retaining a sales and marketing presence (sub. DR393, p. 7). Around the same time Harper Collins moved most of its distribution operations to Australia but retained a local publishing presence in New Zealand (sub. 56, p. 10). But these changes in New Zealand reflect a period of global change in publishing. Sales of electronic books have continued to increase, and consumers are increasingly comfortable with purchasing books online from the cheapest overseas seller. These factors have put pressure on publishers in all countries, and have no doubt forced some to cut costs by closing down operations in smaller countries. For example, Hachette blamed its decision to close its New Zealand arm on ‘the increased sourcing of books from overseas at the expense of the local trade, and the rapid growth of e-books’ (New Zealand Herald 2013). New Zealand is likely to have felt these global effects more strongly than Australia given the relative size of its publishing sector. While New Zealand’s removal of PIRs would be expected to impact the size of its publishing industry, attributing recent industry changes to reforms that occurred almost 20 years ago is implausible. In particular, attributing a decline in print book sales since 2008 is particularly problematic given the global drop in consumer demand caused by the Global Financial Crisis (GFC). Moreover, it is not clear the GFC had a permanent impact: Booksellers NZ chief executive Lincoln Gould says paper’s outlook is promising. “Booksellers are feeling as though they’re thriving, rather than just surviving.” Gould says print sales slumped around the time of the financial crisis in 2008 and 2009. “Things were pretty bleak through those years and so it was only last year that it began to pick up.” (Picken 2015) And analysis of book sales or prices in New Zealand in the years after 1998 are not possible, as data has only been collected since late 2007. However, a 2012 study of the impact of removing PIRs in New Zealand found: Overall, the 1998 changes in New Zealand do not appear to have had significant negative effects on domestic creative effort in the book publishing industry. The number of new NZ book titles that published annually has remained fairly steady between 2005 and 2008. Data on the number of authors shows that following the changes the share of authors in overall employment has increased in New Zealand. (Deloitte Access Economics 2012, p. 6) In the Commission’s view, recent changes in the structure of New Zealand’s publishing sector are not reflective of the changes in PIRs 20 years ago. As such, suggestions that those recent changes provide guidance on the potential impacts from removing PIRs in Australia do not withstand scrutiny. 658 INTELLECTUAL PROPERTY ARRANGEMENTS E.3 Studies evaluating the impacts of fair use Several Australian studies assess the likely effects of moving to a more permissive copyright regime. Some, such as EY (2016) and PwC (2016), assess the effect of replacing fair dealing with a fair use regime, while others such as Lateral Economics (2012) focus specifically on the impacts on Internet intermediaries (box E.1). A lack of comprehensive quantitative evidence makes assessments in this area difficult. The EY and Lateral economics studies instead provide qualitative insights into the likely effect of reform. On the other hand, the PwC report contains a number of methodological flaws that undermine its conclusions. EY report The Australian Government commissioned EY to evaluate the costs and benefits of implementing copyright reforms suggested by the Australian Law Reform Commission (ALRC 2014). EY evaluate two reforms in particular: a ‘first best’ option of replacing fair dealing arrangements with fair use; and a ‘second best’ option of broadening current fair dealing exceptions. They conclude that the fair use reforms are likely to produce net benefits relative to fair dealing (even if current fair dealing is reformed to be less ‘close-ended’). Evaluating the ALRC reforms The ALRC suggests more purposes should be exempt from copyright under fair dealing and that fairness should be established using four new fairness criteria. The new fair dealing purposes are for: • quotation • non-commercial private use • incidental or technical use • education • libraries and archives • access for people with a disability. EY directly evaluates the likely effects of implementing these purposes to determine whether — relative to the status quo — they have either a positive, limited positive, neutral, ambiguous, limited negative or negative net benefit (assuming that the ALRC’s reforms to statutory licensing and treatment of orphan works are also adopted). Each purpose is evaluated by considering the costs and benefits that will accrue to copyright owners and users following the reforms. These costs and benefits can be COPYRIGHT EVIDENCE 659 categorised as reduced incentives to produce new works, reduced administrative/transaction costs, increased incentives to create new uses of existing works, increased use of existing works and changes to social welfare more generally. EY considers whether these costs and benefits accrue in the short or long run, who they are likely to accrue to and their likely magnitude (although this is limited by a lack of quantitative data). The magnitude of each offsetting economic effect is estimated using a ‘bottom-up’ approach based on available information and clearly stated assumptions. For instance, EY estimate that academics currently use 7000 third party quotes per annum by combining an estimate of the proportion of UK research material that includes quotes (10 per cent) with an Australian Research Council estimate of the works produced by Australian universities (70 000 per annum). EY combines the estimates of the net benefit of each new purpose and determines that implementing fair dealing reforms would have a small positive effect on social welfare. Implementing each new purpose is expected to have at least a limited positive effect when assessed individually (except the new data and text mining purpose, which had an ambiguous effect). Implementing fair use would have its own effect over and above those expected under reformed fair dealing. While it is difficult to quantify the net benefits of this reform, EY surveys available evidence and concludes that: • legislative uncertainty would temporarily increase relative to the status quo, but would return to pre-reform levels once a body of case law is established • fair use would increase flexibility of the copyright system, which likely stimulates future innovation and reduces regulatory burden. EY’s approach is more robust than that taken by some other inquiry participants as they: • use a bottom-up approach which means that readers can evaluate the validity of their data and assumptions themselves; making it less likely significant errors remain undetected. • identify all of the most relevant costs and benefits that accrue over the short and long run relative to the status quo. • use measured qualitative evaluation to supplement the analysis of the ALRC reforms where there is an absence of quantitative evidence. Scope for further research While EY’s analysis is relatively more robust, some aspects could be improved. • 660 Results do not take into account that copyright works are often owned by overseas entities. This means that many of the costs of a more permissive copyright system that INTELLECTUAL PROPERTY ARRANGEMENTS are born by copyright owners would not flow on to effect Australian social welfare and the net benefit of reforms is likely understated. • The degree of transitional legislative uncertainty is probably understated because EY assumes that proposed fairness factors would codify existing case law. In reality it may be difficult to fully codify existing case law, which may mean it takes longer for the level of legislative uncertainty to revert to pre-reform levels. • Some parameter values are determined based on qualitative evidence. For instance, when calculating the cost of a diligent search under proposed orphan work reforms, EY use qualitative evidence that some archives already perform diligent searches to discount their initial estimate of costs by 10 per cent. While qualitative evidence suggests some discount is appropriate, it is unclear whether 10 per cent is correct. While these assumptions are largely unavoidable, future research should empirically verify these parameters or at least vary them to gauge their effect on results. These issues do not detract from EY’s findings. While specific benefits and costs may be slightly over- or under-stated, the main costs and benefits of copyright reform are accurately presented. PwC report APRA AMCOS, Copyright Agency Viscopy, Foxtel, News Corp Australia, PPCA and Screenrights commissioned PwC to assess the effect of replacing current fair dealing provisions in the Copyright Act with a fair use exception modelled on US law (sub. 133). PwC conclude that introducing fair use would reduce the supply of original works and increase compliance, transaction and enforcement costs; while not providing commensurate economic benefits. The PwC report contains a number of serious methodological flaws which not only call into question the accuracy of its conclusions but also undermine arguments that may rely on its evidence (such as some of those made by the Copyright Agency (sub. DR510)). The framework Introducing fair use would have offsetting economic effects. On one hand, fair use may reduce incentives to supply original works by restricting the rights of some content owners to prevent their work being copied, performed, published, communicated or adapted without consent. On the other hand, fair use promotes enhanced consumption of copyright works and increased production of transformed secondary works. In some cases introducing fair use would result in a transfer between Australian copyright owners and users that does not change total social welfare (for example a copyright user who currently pays licensing fees to use copyrighted works but who would no longer be required to under fair use). In other cases, fair use will increase the welfare of Australians when the owners COPYRIGHT EVIDENCE 661 of copyrighted works are not Australian. The net effect on social welfare depends on which of these effects dominate. PwC assess the impact on net social welfare using a cost-benefit analysis that overstates the costs and understates the benefits of fair use, and places undue emphasis on the education book industry (which is underpinned by an incorrect assumption about educational licensing under fair use). The remainder of this section discusses these weaknesses against the benchmark principles of effective cost benefit analysis which the Commission views as a more sound basis for assessment (box E.2). Box E.2 The principles of effective cost-benefit analysis The Australian Government recommends using cost-benefit analysis to evaluate the effect of regulatory change, particularly when it has many offsetting effects. Cost-benefit analysis is best implemented by: 1. defining the regulatory change 2. identifying the costs and benefits of the change; predicting how these are likely to change over the life of the proposal 3. attaching a monetary value to each cost and benefit (if possible), considering that they may accrue in the future 4. calculating the net benefit of each option as the benefit less cost, and conducting sensitivity analysis 5. suggesting regulatory change if the net benefit is positive. As a guide, the Australian Government recommends cost-benefit analysis clearly state the benefits and costs of all proposals ‘in a balanced and objective manner’ (DPMC 2016). Further, cost-benefit analysis should consider the effects on all stakeholders rather than just those that accrue to specific groups. This provides a transparent framework for determining the effect of regulatory change. Cost-benefit analysis is useful even when the effect of policy is difficult to quantify or cannot be valued. For instance, if a lack of data makes likely effects difficult to quantify, stakeholders should still fully outline costs, benefits and their determinants. Even a qualitative description is helpful. Emphasising costs and benefits that can be quantified at the expense of those that cannot should be avoided; as should drawing inappropriate generalisations from any specific examples. Sources: Australian Government (2014b); DPMC (2016). Canadian education book industry PwC argue the 2012 Copyright Modernization Act extension to fair dealing caused a significant reduction in the supply of original works by the Canadian education book industry, and that a similar outcome would be expected if Australia adopted fair use. Their argument relies on back-of-the-envelope estimates that state Canada’s education book 662 INTELLECTUAL PROPERTY ARRANGEMENTS industry’s contribution to GDP fell from Can$740 million in 2011 to Can$620 million in 2013 7 following the fair dealing extensions (PwC 2015a). But is this an accurate estimate of the effect of fair dealing on the Canadian education book industry? Several factors besides fair dealing extensions have been identified as causing revenues of the Canadian education book industry to decline over the past decade (figure E.5). Shifts in provincial curriculum, growth in the used book market, rental programs for textbooks, growing demand for supplementary education resources and significant market consolidation all affected the industry (PwC 2015a; Rollans and de la Cheneliere 2010). This is reiterated by inquiry participants, such as the National Copyright Unit, COAG Education Council (sub. 97, pp. 41–42), Australian Digital Alliance (sub. 141, p. 3) and Carroll et al (sub. 149, p. 13) who suggest that a range of factors are responsible for the declining fortunes of the Canadian publishing industry (chapter 5). Despite claims to the contrary (sub. DR510, p. 16), identifying the true effect of fair dealing needs to take these other factors into account. The profile of large Canadian education publishers’ primary school sales revenue suggests that if other factors were taken into account the effect of fair dealing extensions would likely be less than those reported by PwC (figure E.5). Importantly, the changes to fair dealing in Canada are not directly comparable to the proposed changes to fair use in Australia. While the Canadian education sector responded to fair dealing extensions by opting to no longer pay licensing fees to Canadian publishers — which according to the Copyright Agency (sub. DR510, p. 16) caused ‘irreversible’ damage to the industry — the proposed Australian fair use system requires that the potential economic harm to the value of copyright material be explicitly considered when determining fair use. As discussed in chapter 5, the Copyright Advisory Unit to the COAG Education Council has repeatedly stated the education statutory licence scheme will coexist with a broader fair use exception (sub. DR429). Finally, by focusing almost solely on costs, PwC does not present a balanced assessment of the potential effect of fair use on the Australian education book industry. For instance, PwC does not consider how educational institutions might spend any funds no longer paid in licensing fees (such as the fees currently paid for the use of freely available online materials). In any case, the estimated reduction in licensing fees is overstated and does not take into account that educational institutions will continue to pay license fees for all copying not considered fair use. 7 These figures are cited from a previous PwC report (PwC 2015, pp. 44 - 47) and seem to be calculated by applying a national input-output multiplier for the newspaper, periodical, book and directory publishing industry to estimates of spending on educational books sourced from Access Copyright, the Canadian Publisher’s Council and the Canadian Educational Resources Council. COPYRIGHT EVIDENCE 663 Figure E.5 Primary sales revenue of large Canadian education publishersa,b By consumer type 140 Index (2012 = 100) 120 100 80 60 40 Copyright Modernisation Act 20 0 2007 2008 2009 2010 K-12 2011 2012 2013 2014 Post-secondary a Index set so that 2012 revenue equals 100. b As explained in (PwC 2015a), this series includes the aggregate primary sales revenue of large publishers only. Data limitations mean revenue from the K-12 publishing industry excludes Quebec while revenue for the post-secondary publishing industry does not. Revenue from print, digital and blended formats are included. Source: PwC (2015a). Overstated costs Transitioning from current fair dealing arrangements to a fair use system would undoubtedly impose costs that would reduce the welfare of some. For instance, introducing fair use would disadvantage copyright owners who currently receive license fees but who would no longer receive these fees under fair use. PwC also overstate the effects on transaction, compliance and enforcement costs: • PwC suggest fair use would substantially increase the transaction costs of copyright licensing because collection agencies would close and copyright owners and users would need to negotiate independently. PwC use evidence from the UK to argue collection agencies reduce transaction costs by 1800 per cent (or $940 million per annum if this factor is applied to the costs of all Australian collection agencies). But PwC provide no evidence that collection agencies operating outside the education sector would close under fair use. • PwC argue that greater legal uncertainty would significantly increase the costs of copyright enforcement. However, as chapter 5 notes, claims of significant increases in legal uncertainty under fair use are overstated. These claims overlook inherent uncertainty in the current fair dealing system, and ignore that illustrative examples of 664 INTELLECTUAL PROPERTY ARRANGEMENTS fair use, well defined fairness factors and overseas case law would counteract the uncertainty of moving to fair use. Understated benefits PwC significantly underestimate the benefits of fair use, by ignoring several key benefits of moving to a fair use system. For instance, they do not take into account that a fair use system would be better placed to adapt to new uses — or methods of use — of copyright material. Unless foreseen by legislators, these future fair uses would be considered copyright infringement under the status quo. Moving to a fair use system provides the flexibility to support the transformative uses of future innovations, but PwC do not take this into account in their cost benefit analysis. Further, they do not take into account the broader benefits of fair use that have led countries like Israel to adopt fair use. For instance, Zemer (2011) conclude that, broadly speaking, the fair use doctrine enables the fair use of copyright works where it was previously too costly to clear rights, and helps promote the use of copyright works when there is a strong social or ethical rationale. And as noted in chapter 6, Israel’s adoption of fair use in 2007 was specifically to ‘advance culture and knowledge’ (Nair 2012, p. 30). PwC draw inappropriate conclusions from statistical analysis to claim the introduction of fair use would not yield economic benefits. PwC find a weak statistical association between a Consumers International measure of copyright flexibility and GDP per capita across 44 countries, and conclude that copyright flexibility is not a significant determinant of GDP per capita. They take this to mean that a more flexible copyright system (such as fair use) has little economic benefit. But the flexibility of copyright is not expected to have a significant impact on GDP per capita because works protected by copyright represent only a minor proportion of national output 8. Hence it is inappropriate to use a weak statistical relationship between copyright flexibility and GDP per capita to conclude fair use has limited economic benefit. Similarly, inferring from countries that implement fair use and have a smaller value share of industries that benefit from the exception relative to Australia is not evidence that fair use has little benefit. Such examples do not take into account factors that bear on the uptake of a fair use exception, such as wider patterns of industrial formation, geography and differences in the nature of innovation between countries relative to the status quo. 8 PwC’s estimate of the size of Australian industries relying on copyright exceptions industries — 12.9 per cent of GDP (2016, p. 37) — is likely an overestimate. PwC rely on a Lateral Economics estimate of the value of industries that rely on copyright exceptions, rather than the value of the activities covered by the exceptions (Lateral Economics 2012). For instance, Lateral Economics reported $17 billion of value add by the private education and training industry in 2009-10, but the value add of the specific activities within private education and training that rely on copyright exemption is much less. COPYRIGHT EVIDENCE 665 Lateral Economics report While EY and PwC evaluate the net benefit of moving to a more permissive copyright system generally, Lateral Economics focuses on evaluating the effect of introducing more flexible exceptions and extended safe harbour provisions for Internet intermediaries (Internet service providers, Internet hosts, search engines and portals, E-commerce intermediaries, Internet payment systems and user created content platforms). In a hypothetical scenario, a more permissive treatment of copying for the purposes of Internet intermediation is assumed to produce a one percentage point increase in real output of exception dependent industries, which translates to a $593 million increase in social welfare. Many of their results are driven heavily by unverified assumptions. For instance, they provide no evidence that a more permissive treatment of copying for the purposes of Internet caching, searching or hosting will necessarily cause a one percentage point increase in real output for exception dependent industries. As a result, quantified results should be seen as indicative of potential effects rather than modelled estimates. Notwithstanding these issues, Lateral Economics outlines some potential benefits flowing from adopting more flexible exceptions and extended safe harbour provisions for Internet intermediaries (although more could be done to illustrate how this affects Australian firms in particular). These include that: • unlike other third party uses, copying for the purpose of Internet search is likely to increase revenues to copyright owners because search usually makes content more discoverable (which attracts more users) • adopting a more permissive system would recognise that compliance with the status quo introduces prohibitive transaction costs for intermediaries (Lateral Economics estimates search engines would incur transaction costs of over $150bn per year if they sought permission to copy Australian content, but this is not a precise estimate) • intermediaries need to establish whether copying for purposes of caching, indexing or search falls under current prescriptive fair dealing purposes whenever technology changes. Moving to a more permissive exception would reduce this legal uncertainty • introducing a more permissive system would lower the investment hurdle for intermediaries looking to operate in Australia but who perceive copyright litigation risk. The resulting increased investment could provide long term benefits for Australia. These benefits highlight that current copyright legislation embodies a distinction between copy and use that no longer applies in the Internet age where to ‘use is to copy’. A reform case study: unlocking access to orphan works Many of the benefits (and indeed some of the costs) from adopting fair use are, by their very nature, difficult to quantify. One benefit that arises from fair use that better lends itself to quantification relates to unlocking the value of orphan works. 666 INTELLECTUAL PROPERTY ARRANGEMENTS There are an estimated 52 million items (written works of various types, still photos, audio and video items etc.) held in 1900 libraries and archives across Australia. Of these, it is likely that at least 40 per cent (21 million items) are ‘orphaned’. Diligently searching for the copyright owners to seek permission to use these works is time consuming and costly. Moreover, in cases where the copyright owner is not found, the legal consequences of use remain unclear. The extent of community interest in accessing such material is illustrated by the success of the Australian National Library’s Trove project. Trove, a searchable online database of digitised books, newspapers, pictures, music, maps and other material, has proven immensely popular with everyday Australians. Many Trove users volunteer hundreds if not thousands of hours of their free time editing and correcting the machine translated text, improving the experience for subsequent users. Trove is the fourth most popular Australian Government website. By the end of April 2016, Trove volunteers had corrected over 196 million lines of text — equivalent to 505 standard work years (Jessica Coates, Australian Digital Alliance, pers. comm., 2016). The National Library’s Australian Women’s Weekly project provides another example of broader community benefits that might otherwise go untapped in the absence of more flexible rules on copyright. In this project, digital copies of every edition of the Australian Women’s Weekly from its inception in 1933 to the final weekly edition in 1982 were up-loaded to Trove’s open and searchable database. This was done with the assistance of newspapers and microfilm owned by the State Library of NSW and Australian Consolidated Press (National Library of Australia 2016). Private use of the material is permitted and encouraged, and there were over 1.5 million views of the digital editions during the first 12 months of the trial. Importantly, any individual, organisation or business wishing to use the digitised material for purposes other than personal use (such as commercial use) must still seek a license to do so from the copyright owner — just as they would have been required to do if the material had never been made available online by the National Library. In essence, making this material more accessible to the general public for uses or purposes that might be considered ‘fair’ has not made it ‘free’ to all users, or for all purposes. It has, however, provided pleasure and enjoyment to huge numbers of Australians, without compromising the incentives of current creators to generate new works — the primary goal of copyright laws. Trove and the Women’s Weekly project are clear examples of the kinds of personal uses that a ‘fair use’ regime would likely permit. While both of these projects have occurred under the current copyright arrangements, it is likely many other socially beneficial projects have not. EY (2016) estimated the annual gains from better access to orphan works to be between $10 million and $20 million. This estimate was derived by using a ‘search cost’ approach — multiplying an estimate of the average cost of a diligent search for a rights holder (an assumed time in hours multiplied by an average hourly wage rate) by the estimated number COPYRIGHT EVIDENCE 667 of orphaned works that cultural institutions might seek to use each year for commercial purposes, such as for exhibitions and producing items for sale. A key assumption underlying the calculation is that the ‘use value’ of each cleared work is at least equal to the cost of searching for the missing rights holder. The study did not attempt to measure the flow on benefits to growth and new business creation that might arise from commercialising this type of content. Improved access to orphan works would also have distributional consequences. Currently, statutory licenses under Parts VA and VB of the Copyright Act allow the copying and communication of materials for education, whether or not they are orphaned, subject to the payment of reasonable remuneration to the declared collecting societies. Fees collected from uses of orphan works are held in trust, and, if the rights holder is not identified after four years, redistributed among the current members of collecting societies (even though they played no part in the creation of the orphaned works). In 2014, the amount in question was around $9.2 million (EY 2016). Under a fair use regime, it is possible this activity would no longer be remunerable. 668 INTELLECTUAL PROPERTY ARRANGEMENTS F Registered rights in Australia: a primer The three main registered IP rights in Australia are patents (chapters 7–10), designs (chapter 11) and trade marks (chapter 12). This appendix summarises these rights, including the application process used to register these rights. F.1 Patents What rights do patents confer and what qualifies for protection? A patent provides its owner the exclusive right to commercially exploit an invention. 1 These rights only apply in the country in which the patent is granted — protection in one country does not confer protection in another. Patents can be granted to a broad range of invention types, including devices, substances, methods and processes. Applications for patents are filed with IP Australia (figure F.1). Several types of patent applications can be submitted. • Provisional application — an application filed prior to filing a complete application. A provisional application establishes the earliest possible ‘priority date’, which proves who was first to file a patent application for an invention and is the date from which the application is assessed against the patent criteria (see below). A complete application must be filed within 12 months of a provisional application or the provisional application lapses, resulting in a loss of the priority date. • Complete application — can result in a standard patent being granted. Complete applications are published in the Australian Official Journal of Patents, and must include: – a full description of the invention that would enable a person skilled in the relevant field to replicate the invention (box F.1) – one or more ‘claims’ to the invention, which determine the scope of exclusive rights claimed by an applicant. • International application — an application for patent protection overseas. Applicants have two choices: 1 The Commission has in many places used the term ‘invention’ in keeping with the language of the Patents Act. This means that in some places invention is used in place of the broader concept innovation. REGISTERED RIGHTS IN AUSTRALIA: A PRIMER 669 – Paris Convention application — if protection is only sought in a few countries, it might be most cost effective to make separate patent applications in each country. Countries that are party to the Paris Convention allow applicants to use their original filing date as the priority date for each country if an application is made within 12 months. – Patent Cooperation Treaty (PCT) application — allows an applicant to file a patent application with IP Australia (or another patent office recognised by the World Intellectual Property Organization) and elect for protection in over 140 countries. PCT applicants must apply to the patent office in each country they wish to receive protection within two and a half years from the priority date. The PCT application may claim priority from a provisional application or a Paris Convention application. Firms and individuals can also apply for an innovation patent, which, in exchange for a lower cost and quicker patent application procedures, provides a shorter term (chapter 8). Figure F.1 Standard patent application process in Australia Source: IP Australia website. 670 INTELLECTUAL PROPERTY ARRANGEMENTS Box F.1 Disclosure rules Patent applicants must disclose sufficient information for a person skilled in the relevant field to replicate the invention. This information, which is publicly released 18 months after the priority date, is widely considered to be the quid pro quo of patent rights. While a number of participants highlighted the importance of disclosure, few pointed to specific issues with current disclosure rules. Reforms introduced as part of the ‘Raising the Bar’ initiative2 increased the obligations on patent holders to disclose how their inventions work. • Specifications must now disclose the invention in a manner clear and complete enough for it to be performed by a person skilled in the relevant art. Claims must also be fully supported by the description of the invention. • A provisional application now only provides a valid priority date for an invention later claimed in a complete application if the provisional specification discloses the invention clear and complete enough for the invention to be performed by a person skilled in the relevant art. • The ‘usefulness’ patent criterion now requires that a specific, substantial and credible use for the invention be disclosed in the application. There is evidence these reforms have been effective in raising disclosure standards. In a recent case, 3 IP Australia rejected an application on the grounds it did not satisfy provisions introduced as part of Raising the Bar. IP Australia found the claims were not fully supported by the description, and the invention was not disclosed in a way that enabled it to be performed by a person skilled in the relevant art. IP Australia grants patents to inventions that meet various criteria outlined in the Patents Act 1990 (Cth). To satisfy the criteria for a standard patent, inventions must: • be a ‘manner of manufacture’ — defined by the courts to be an artificially created state of affairs in a field of economic endeavour.4 Jurisprudence recognises some subject matters that fail to satisfy the test, including mere discoveries, ideas, scientific theories and laws of nature. The Patents Act excludes from patentability human beings and the biological processes for their generation • be novel — the invention must be novel in light of a single piece, or a combination of two or more pieces (that a skilled person could have been reasonably expected to have combined), of ‘prior art information’ (information about the current state of technology). Prior art information includes documents as well as information publicly available through doing an act (including a prior use) • involve an inventive step — the invention must not be obvious to a person skilled in the relevant art in light of common general knowledge. Common general knowledge can be considered separately or together with the prior art information 2 Intellectual Property Laws Amendment (Raising the Bar) Act 2012 (Cth). 3 CSR Building Products Limited v United States Gypsum Company [2015] APO 72. 4 National Research Development Corporation v Commissioner of Patents [1959] HCA 67. In D’Arcy v Myriad Genetics Inc & Anor [2015] HC 35, the High Court said that an artificially created state of affairs in the field of economic endeavour was not exhaustive of the concept of manner of manufacture [at 20]. REGISTERED RIGHTS IN AUSTRALIA: A PRIMER 671 • be useful — there must be a specific, substantial and credible use for the invention disclosed in the specification (description of the invention) • have not been secretly used — the invention cannot be commercially used before the priority date. A patent applicant must request examination within five years of when the application is first filed (or earlier if directed by IP Australia). In assessing an application against the patent criteria, patent examiners consider the prior art and common general knowledge that applied at the time of the priority date. In deciding whether the conditions for the patent criteria are met, the Patents Act requires the examiner to be satisfied ‘on the balance of probabilities’. Each year IP Australia receives between 25 000 and 30 000 applications for standard patents. The majority of applications are successful, with about 60 per cent of standard patent applications filed in any given year eventually granted. 5 An application may fail to result in a granted patent for a number of reasons, including because it is abandoned by the applicant, an opposition proceeding is upheld, or it fails to meet the patent criteria. With regards to opposition proceedings, there are several mechanisms for third persons to challenge a patent application or patent: • Intervention in the examination process. Third persons can provide information relevant to novelty or inventive step under section 27 of the Patents Act. This is usually considered during examination, but if filed too late it may be considered in a re-examination. • Opposition. Once the examination process is completed and an application is accepted, third persons have three months to oppose a patent being granted. There is then a process for each side to file evidence before a hearing is held to decide the opposition. • Re-examination. IP Australia can re-examine an accepted patent application or a granted patent. The process is instigated at IP Australia’s discretion, upon request by the patentee or any interested person, or by the direction of a court before which the validity of a patent is in dispute. The procedure is ex parte — it does not involve a third party who requests re-examination. • Revocation. Third persons can challenge the validity of a patent by applying to the Federal Court or a State Supreme Court to revoke the patent. Re-examination may lead to amendment of the application or patent, refusal to grant a patent, or revocation of a patent. 5 Based on applications filed between 2000 and 2008. Filings after 2008 are excluded from the sample due to the time it can take for a patent application to reach a conclusion. For applications that received a first report, 73 per cent are on average eventually accepted. 672 INTELLECTUAL PROPERTY ARRANGEMENTS IP Australia’s decisions are subject to legal challenge. In 2014-15, the courts decided four appeals of IP Australia’s decisions for patents. IP Australia also has its own internal review mechanisms for patents and other rights (box F.2). Box F.2 IP Australia’s internal review mechanisms IP Australia operates a Quality Review System (QRS) as part of its overarching quality management system. The QRS was introduced in 2011 and has been refined since. Under the QRS, examination work is assessed against Product Quality Standards by staff in the Quality Improvement section through a sampling inspection regime. Product Quality Standards are arranged in three tiers. • Tier 1: matters which could adversely affect the validity of the intellectual property right. • Tier 2: matters which would require a considerable amount of rework and/or inconvenience to the applicant or IP Australia. • Tier 3: other important procedures required by the Patents Manual. Task complete Sampling Arbitration process Quality review (standards) Quality review results and feedback Acceptable quality levels Supervisor validation Quality improvement Recognising good quality Key elements of the scheme are as follows. • Examiner completes an examination task (examination report, search report). • The examiner’s work is audited according to a sampling regime based on ISO 2859. • Work is assessed by the Quality Improvement section (QIS). • If the QIS identifies a nonconformity with the quality standards, it is sent to the examiner and their supervisor. The supervisor responds to the reviewer, either agreeing or disagreeing with the assessment. (continued) REGISTERED RIGHTS IN AUSTRALIA: A PRIMER 673 Box F.2 (continued) • If the supervisor agrees, corrective action is taken. • If the supervisor disagrees, the case is sent to an Assistant General Manager for arbitration. Decisions by arbiters are published internally for access by all Supervising Examiners of Patents (Directors) and a summary of the issue is communicated to all examiners. • If an examiner reaches certain threshold levels of non-conformance (depending on the tier), their work is subject to additional sampling for 3 months, escalating to work being fully supervised until they obtain a satisfactory assessment. • The quality section is responsible for feeding any improvements back into the system. For example, through changes to the Examiners manual or training. Similar arrangements apply to trade mark, designs and Plant Breeder’s Rights examination. IP Australia reports quarterly on its performance against the quality standards. IP Australia publishes the results and outcomes of the Quality Review System in Customer Service Charter Reports, which it posts on its website. Source: IP Australia website. Who holds patents in Australia? In 2015 there were over 130 000 active standard patents in Australia. The highest numbers of patents were in the fields of medical technology, biotechnology, pharmaceuticals and organic fine chemistry (figure F.2). Of the standard patents granted to Australians between 2010 and 2015, 67 per cent were filed by small– and medium–sized enterprises (SMEs), 16 per cent by large firms, and 17 per cent by individuals. The number of patents in Australia has grown rapidly, increasing by about 100 per cent between 1990 and 2015 (figure F.3). The system has also expanded to cover a wider range of sectors and technologies. The technologies with the greatest growth rates are information technology (IT) methods for management, semiconductors, medical technology and micro-structural and nano-technology. Some of these technologies have grown off a low base — for example, in 2000 there were only 30 patents in the IT methods for management technology field. On average, Australians file fewer patents on a per capita basis than most other developed countries (figure F.4). 674 INTELLECTUAL PROPERTY ARRANGEMENTS Figure F.2 Distribution and growth of patents in Australiaa (i) The largest concentration of patents in 2015 are in health–related technologiesb (ii) Patents have grown rapidly in some technologies over the past 15 yearsc a Technologies are based on WIPO International Patent Classification codes. b Technologies with largest stock of standard patents. The percentage of total patents that each technology field comprises is provided in brackets. c Percentage increase in standard patents granted by technology between 2000–2015. Source: Commission estimates using IPGOD (2016 edition). Figure F.3 The stock of patents has growna Stock of standard patents in force in Australia, 1990–2015 a The stock of patents have not been adjusted for growth in the economy. The net change in the stock of patents accounts for the granting and expiration of patents. Source: Commission estimates using IPGOD (2016 edition). REGISTERED RIGHTS IN AUSTRALIA: A PRIMER 675 Figure F.4 Australians file relatively few patents per capitaa a Patents filed domestically and overseas in force in 2014. Source: WIPO (2016). Australia’s patent system interacts with innovators and patent systems in other countries. The rules and procedures that apply in patent systems overseas thus have an important bearing on innovation and policy settings in Australia. • Most patents granted in Australia are held by non-residents. Between 2011 and 2015 non-residents received about 93 per cent of patents granted (IP Australia 2016a). The United States was the largest non-residential filer with 13 781 applications in 2015 (about 52 per cent of non-residential applications), followed by Japan (1733), Germany (1339) and the United Kingdom (1155) (Australians filed 2291 applications in 2015). Between 1998-99 and 2014-15, net imports (the real value of imports minus exports) of patent-intensive commodities increased by about 400 per cent (appendix C). • Most Australians that file patents do so overseas. In 2014 Australians filed 9253 applications overseas, compared to 1979 domestic applications (IP Australia 2016a). 6 Major destinations were the United States (38 per cent of Australian overseas applications), the European Patent Office (EPO) (9 per cent) and China (9 per cent). 6 For an invention with applications in multiple jurisdictions, each application is counted separately. For applications filed at the EPO and the African Regional Intellectual Property Organization, each application is counted as only one application. For applications filed at the Eurasian Patent Organization and African Intellectual Property Organization, each application is multiplied by the corresponding number of member states. 676 INTELLECTUAL PROPERTY ARRANGEMENTS F.2 Registered designs Registered design rights protect the appearance (rather than function) of products that have an industrial or commercial use. Applications for design rights are made to IP Australia (figure F.5). To be registered, an application for a registered design right only has to pass a formalities check. Registered designs must be examined and certified before they can be enforced. To pass substantive examination, a registered design must: • be ‘new’ — it must not be identical to any design previously disclosed anywhere in the world (including on the Internet), nor any design previously used in Australia (the prior art) • be ‘distinctive’ — it must not be substantially similar in overall impression to any design previously published anywhere in the world (including on the Internet), nor any design previously used in Australia • not be a product that is excluded from designs protection, which includes medals, emblems, flags or graphics that might be regarded as scandalous. The prior art includes all previously disclosed work, so registration is not valid if the design was disclosed in the marketplace prior to the application being filed. The Designs Act 2003 provides guidance to design examiners in determining whether a design is ‘substantially similar in overall appearance’. In particular: • more weight is to be placed on the similarities between designs than to differences • where a design application makes special reference to a particular visual feature of the design, particular regard must be had to that visual feature • the standard to be applied when making the comparison is that of a person who is familiar with the product to which the design relates, that is, the ‘informed user’ • in infringement proceedings, deciding whether a design is new and distinctive is one of fact determined by the judge from the perspective of an informed user. REGISTERED RIGHTS IN AUSTRALIA: A PRIMER 677 Figure F.5 Registered design application processa a If a notice of deficiency is issued following a request for registration or publication, the application will lapse if the objections are not overcome within two months. b Details published and the application date established. c Certificate issued and notice published. Source: Based on ACIP (2015a, p. 58). 678 INTELLECTUAL PROPERTY ARRANGEMENTS F.3 Trade marks and geographical indications Trade marks and the application process There are four types of trade marks (table F.1), and 45 ‘classes’ of goods or services to which they can be applied. These classes are known as the Nice Classification. Table F.1 The different types of trade marks Type of trade mark Description Standard Provides an exclusive right for the mark owner or licensee to use the mark in the nominated classes of goods and services. They are the most common type of trade mark. Used to distinguish goods or services provided by members of an association, and are frequently used by professional associations. Companies limited by shares or individuals cannot own collective trade marks. Used to declare that a good or service has a particular quality — such as content, method of manufacture and/or geographic origin. A certification mark can be used by anyone, provided that they are certified as meeting the criteria required by the mark. In addition to assessment by the Trade Marks Office, a certification mark is also assessed by the Australian Competition and Consumer Commission to ensure the claims it makes are not misleading or deceptive, and that the way it is licensed for use is not anticompetitive. The ‘Australian Made’ logo is an example of a certification trade mark. Prevent an existing well-known trade mark being used in another class. For example, a popular battery brand holds a trade mark for batteries, as well as a defensive trade mark to prevent its brand being used to market energy drinks. Collective Certification Defensive Source: IP Australia (2014a). Trade mark applications are processed by the Australian Trade Marks Office (ATMO), which is part of IP Australia (figure F.6). Applications are initially examined against application criteria, which if successful leads to the mark being accepted. An opposition period allows other parties to challenge the acceptance of the mark, based on the application criteria or on additional criteria specific to the oppositions process (including that the applicant is not the true owner of the mark, and that the mark applied for is too similar to another mark — registered or otherwise — that has already gained a reputation in the community). If oppositions are unsuccessful, then the mark is registered, and can only be removed for ‘non-use’ or a successful challenge that the original application was made in bad faith. In 2014-15 there were 21 appeals from the Registrar of Trade Marks’ decisions. A registered trade mark has a term of 10 years from the date of application, but there is no limit to the number of times it can be renewed. REGISTERED RIGHTS IN AUSTRALIA: A PRIMER 679 Figure F.6 The trade mark application process Criteria for acceptance: • does not include specific signs listed in trade mark regulations (s. 39 of the Trade Marks Act) • be represented graphically (s. 40). • be capable of ‘distinguishing’ the applicant’s goods or services (s. 41). • does not contain scandalous matter or provide for a mark where its use would be contrary to law (s. 42). • is not likely to deceive or cause confusion (s. 43). • is not ‘substantially identical’ or ‘deceptively similar’ to an existing trade mark (or one seeking registration with an earlier application date) (s. 44). Criteria for oppositions prior to registration: • any of the criteria for acceptance, and • applicant is not the owner of the mark (s. 58) • opponent has made an earlier use of the mark in question (s. 58A) • applicant does not intend to make use of the mark (s. 59) • mark applied for already has a reputation in Australia (s. 60) • mark contains a false geographical indication (s. 61) • application itself is defective (s. 62) • the application was made in bad faith (s. 62A). Criteria for oppositions following registration • that the mark has not been used for a period of three years (or five years in the case of new registrations) (s. 92) • that the application for a mark was not in good faith (s. 92). Sources: Adapted from IP Australia (2014a); Trade Marks Act 1995 (Cth). 680 INTELLECTUAL PROPERTY ARRANGEMENTS International registration of trade marks The Madrid Protocol treaty provides for international registration of a trade mark (IP Australia 2016f). The Protocol has 97 members that covers 113 countries, including Australia, which together account for about 80 per cent of world trade (WIPO 2015c). Under the Protocol each application is examined according to the trade mark laws in the jurisdictions for which it is applied. IP Australia facilitates applications made by Australians to register trade marks through the Protocol. To be valid, applications must: • also be applied for, or registered, in Australia • have the identical mark to that of the mark applied for or registered in Australia • pertain to goods and services classes that are covered by the Australian application or registration, and have the same applicant as the Australian application or registration (IP Australia 2016f). Use and enforcement of trade marks A successful registration of a trade mark — denoted by the use of the ‘®’ symbol — allows the owner or licensee to use the elements of the mark exclusively. Trade marks are also subject to a ‘use it or lose it’ provision. A mark must have been used in the last three years, or may be liable to be removed from the trade mark register upon application by a third party to the Trade Marks Office. A grace period of five years of non–use is provided to newly registered marks. While a trade mark does provide the owner or licensee of the mark exclusive rights to use it in the classes of goods and services nominated, it is also the responsibility of the owner to defend those rights (IP Australia 2015h). Legal action is the mechanism to resolve disputes about trade mark infringement, and disputes are often resolved by ‘cease and desist’ letters rather than taking the matter to court (Queensland Law Society, sub. 116). Geographical indications The criteria evaluated by the Geographical Indications Committee are set out in regulation 25 of the Australian Grape and Wine Authority Regulations 1981 (Cth): For the purposes of subsection 40T(2) of the Act, the Committee is to have regard to the following criteria: (a) whether the area falls within the definition of a subregion, a region, a zone or any other area; (b) the history of the founding and development of the area, ascertained from local government records, newspaper archives, books, maps or other relevant material; REGISTERED RIGHTS IN AUSTRALIA: A PRIMER 681 (c) the existence in relation to the area of natural features, including rivers, contour lines and other topographical features; (d) the existence in relation to the area of constructed features, including roads, railways, towns and buildings; (e) the boundary of the area suggested in the application to the Committee under section 40R; (f) ordinance survey map grid references in relation to the area; (g) local government boundary maps in relation to the area; (h) the existence in relation to the area of a word or expression to indicate that area, including: (i) any history relating to the word or expression; and (ii) whether, and to what extent, the word or expression is known to wine retailers beyond the boundaries of the area; and (iii) whether, and to what extent, the word or expression has been traditionally used in the area or elsewhere; and (iv) the appropriateness of the word or expression; (i) the degree of discreteness and homogeneity of the proposed geographical indication in respect of the following attributes: (i) the geological formation of the area; (ii) the degree to which the climate of the area is uniform, having regard to the temperature, atmospheric pressure, humidity, rainfall, number of hours of sunshine and any other weather conditions experienced in the area throughout the year; (iii) whether the date on which harvesting a particular variety of wine grapes is expected to begin in the area is the same as the date on which harvesting grapes of the same variety is expected to begin in neighbouring areas; (iv) whether part or all of the area is within a natural drainage basin; (v) the availability of water from an irrigation scheme; (vi) the elevation of the area; (vii) any plans for the development of the area proposed by Commonwealth, State or municipal authorities; (viii) any relevant traditional divisions within the area; (ix) the history of grape and wine production in the area. Once an application is received by the geographical indicators (GI) Committee, trade mark owners are given an opportunity to object (through the Registrar of Trade Marks) to the proposed GI on the basis that it is likely to cause confusion with a trade mark. Objections that the proposed GI is used in Australia as a common name of a type or style of wine are also considered. Once the GI Committee decides to accept an application, objections can be raised in the Administrative Appeals Tribunal within 28 days of the decision, with further appeals held in the Federal Court (Wine Australia 2015). 682 INTELLECTUAL PROPERTY ARRANGEMENTS G Additional material about trade marks and geographical indications This appendix provides additional detail regarding trade marks and geographical indications, and supplements chapter 12 of the inquiry report. Information around the assignment, use and enforcement of trade marks is provided in the registered intellectual property (IP) rights appendix (appendix F). How responsive are applications to changes in fees? As discussed in chapter 12, trade mark fees have been shown to influence the number of classes in trade marks applied for by firms in some jurisdictions (WIPO 2013), but there is little evidence about the relationship between fees and the scope of marks in Australia. One cross-country study that has examined how international fees affect trade mark activity is described in de Rassenfosse (2015). This study uses a panel data approach to measure the fee elasticity of demand for trade mark applications based on marks using the Madrid Protocol, and found a fee elasticity of demand between -0.31 and -0.42 — meaning that a 1 per cent increase in trade mark fees would be expected to reduce applications by between 0.31 and 0.42 per cent. Some key features of that study’s approach include: • the variation in the data is driven by panel fixed effects of different countries (which serve as indicators for different trade mark regimes and criteria) and exchange rate and purchasing power variation which affect the ‘real’ level of fees paid by a firm seeking registration of a trade mark. The data do not include domestic registrations or registrations outside of the Madrid Protocol process • explanatory variables for the number of trade mark fee applications include the (natural logs of) real trade mark fees, real GDP and applications made in the previous year. This approach is a ‘partial adjustment’ model — trade marks and economic output are assumed to be linked, but trade mark applications are slow to reflect changes in economic growth. The Commission has adapted this methodology by using the IPGOD database to differentiate between applications made by small Australian firms, large Australian firms, and overseas firms. • To get real trade mark fees, assumptions are made about the fees paid by different types of firms (small firms are more likely to use the online system to register their mark, which attracts a lower fee than larger firms that go through a lawyer). The trade mark ADDITIONAL MATERIAL ABOUT TRADE MARKS AND GEOGRAPHICAL INDICATIONS 683 fee payable is adjusted by the average number of classes in which a firm seeks an application. Registration fees are applied in proportion to the number of marks that have a successful application. • Trade mark fees have not changed frequently, and so variation in the real cost of fees is largely driven by inflation, changes in the number of classes applied for, and the likelihood of a successful registration. – Inflation is applied using two deflators: the GDP deflator and the Consumer Price Index (CPI). – The fees applicable for trade mark applications from overseas are scaled by the trade weighted index in order to adjust for differences in currency. The results of the model (table G.1) indicate that: – small firms are found to have a negative fee elasticity of demand (between -0.51 and -0.80) – fees do not appear to affect trade mark behaviour of large Australian and overseas firms. – Australian firms’ trade mark activity is affected by previous Australian trade mark activity. This relationship is not found for overseas firms. These results tend to confirm what the Commission was told by participants in the inquiry: that small firms are more responsive than larger firms, and that changes in trade mark fees would have more of an impact on the former than the latter. That said, these results should be treated with caution for a number of reasons: • the analysis is based on few observations • the observed relationship between real gross domestic product (GDP) and trade mark applications is unexpectedly weak or statistically insignificant. This may suggest that Australia has a relatively slower rate of adjustment compared to other jurisdictions. • elasticity estimates of this sort are an average across the observed data — some firms may be more or less responsive to fees in practice. 684 INTELLECTUAL PROPERTY ARRANGEMENTS Table G.1 Model resultsa Dependent variable: natural log of number of trade mark applications Model # Explanatory vars Natural log of fees, deflated by GDP deflator Natural log of fees, deflated by CPI deflator Natural log of real GDP Natural log of lagged trade mark applications Constant Stationary? b Misspecified? c R-squared Small Australian Businesses Small Australian Businesses Large Australian Businesses Large Australian Businesses Overseas Businesses Overseas Businesses 1 2 3 4 5 6 -0.51* 0.05 -0.80* -0.01 -0.20 -0.04 -0.15 -0.06 -0.24 -0.29* 0.91* 0.86 0.80*** 0.76*** 0.87*** 0.87*** 0.30 0.30 7.49 8.40 4.15 6.54 -5.89 -5.06 Yes Yes Yes Yes Yes Yes No No No No No No 0.96 0.96 0.82 0.82 0.89 0.89 a Where *, **, and *** denote statistical significance at the 10%, 5% and 1% levels of significance, respectively. Based on 18 annual observations (1995–2015). b Based on a Dickey-Fuller test with one lag. Model declared stationary if the test statistic of the augmented Dickey-Fuller test with one lag is rejected at the five per cent level of significance. c Model declared misspecified if the test statistic of the Ramsay RESET test is rejected at the five per cent level of significance. Sources: Commission estimates based on IPGOD (2016 edition); IP Australia (2014b, 2015f, 2015m, 2015n). For reference, a schedule of fees is summarised in table G.2. Since the analysis, IP Australia has announced changes to trade mark fees, the most important of which is to combine the application and registration fees into a single payment (IP Australia 2016d). While this change will slightly reduce the cost of a successful application and registration, it also increases the cost of an unsuccessful application by the higher fee. ADDITIONAL MATERIAL ABOUT TRADE MARKS AND GEOGRAPHICAL INDICATIONS 685 Table G.2 A summary of trade mark fees Fee type Fee Applicationa $120-370 per class Registration $300 per class Extensions of timeb $100 per month Filing notice of intentions to oppose a mark, apply for removal due to non-use, or providing notice of opposing removal $150-250 per trade mark Requesting or attending a trade mark hearingc $400-600 per trade mark Purchasing a commemorative certificate of trade mark registrationd $44–137.50 a The size of the fee depends on whether the trade mark application uses the online application process (‘eServices’), whether the application is for a series of trade marks, and whether the ‘pick list’ is used. A series of trade marks is made of two or more similar trade marks on the same application or registration. The pick list refers to an application where the classes for registration are picked from a default list that maps to the Nice Classification. Those without the pick list require more work as they must be assigned by the trade mark examiner into the appropriate class. b Extensions of time refer to extensions to respond to an examiner’s report, to pay an initial registration fee, or to pay a late renewal fee. c The cost depends on whether the hearing is about an opposition to a trade mark or for some other purpose. d Printed by a third party outside of IP Australia. Sources: IP Australia (2014b, 2015f, 2015m, 2015n). Geographical indications An example of terms protected in bilateral agreements Table G.3 lists the terms proposed to be protected in the Canada-EU free trade agreement. For Greek and Cypriot geographical indications (GIs), the transliteration of indications from Greek are also protected under the agreement. Particular goods are also allowed to be used by Canadian producers if they include words such as ‘style’ or ‘like’ to accompany them (e.g. ‘Feta style’ cheese could be produced and marketed in Canada under the terms of the proposed agreement). 686 INTELLECTUAL PROPERTY ARRANGEMENTS Table G.3 GI terms in the proposed Canada-EU Free Trade Agreement Indication Product Class Origin České pivo beer Žatecký Chmel hops Hopfen aus der Hallertau Nürnberger Bratwürsteb hops fresh, frozen and processed meats Czech Republic Czech Republic Germany Germany Nürnberger Rostbratwürste Schwarzwälder Schinken Aachener Printen Nürnberger Lebkuchen Lübecker Marzipan Bremer Klaben Hessicher Handkäse Hessicher Handkäs Terttnanger Hopfen Spreewälder Gurken fresh, frozen and processed meats Germany fresh, frozen and processed meats Germany confectionery and baked products confectionery and baked products confectionery and baked products confectionery and baked products cheeses cheeses hops fresh and processed vegetable products cheeses table and processed olives confectionery and baked products cheeses Germany Germany Germany Germany Germany Germany Germany Germany oils and animal fats oils and animal fats Greece Greece oils and animal fats Greece oils and animal fats spices cheeses cheeses cheeses cheeses cheeses fresh and processed vegetable products fresh and processed vegetable products Greece Greece Greece Greece Greece Greece Greece Greece table and processed olives confectionery and baked products oils and animal fats oils and animal fats oils and animal fats Greece Cyprus Spain Spain Spain Danablu Ελιά Καλαμάτας Μαστίχα Χίου Φέταa Ελαιόλαδο Καλαμάτας Ελαιόλαδο Κολυμβάρι Χανίων Κρήτης Ελαιόλαδο Σητείας Λασιθίου Κρήτης Ελαιόλαδο Λακωνία Κρόκος Κοζάνης Κεφαλογραβιέρα Γραβιέρα Κρήτης Γραβιέρα Νάξου Μανούρι Κασέρι Φασόλια Γίγαντες Ελέφαντες Καστοριάς Φασόλια Γίγαντες Ελέφαντες Πρεσπών Κονσερβολιά Αμφίσσης Λουκούμι Γεροσκήπου Baena Antequera Sierra Mágina Transliteration Elia Kalamatas Masticha Chiou Feta Kalamata olive oil Kolymvari Chanion Kritis Olive Oil Sitia Lasithiou Kritis Olive oil Olive Oil Lakonia Krokos Kozanis Kefalograviera Graviera Kritis Graviera Naxou Manouri Kasseri Fassolia Gigantes Elefantes Kastorias Fassolia Gigantes Elefantes Prespon Florinas Konservolia Amfissis Loukoumi Geroskipou Denmark Greece Greece Greece Greece (continued next page) ADDITIONAL MATERIAL ABOUT TRADE MARKS AND GEOGRAPHICAL INDICATIONS 687 Table G.3 (continued) Indication Product Class Origin Aceite del Baix EbreMontsía Oli del Baix EbreMontsía Aceite del Bajo Aragón Priego de Córdoba Sierra de Cádiz Sierra de Segura Sierra de Cazorla Siurana Aceite de Terra Alta Oli de Terra Alta Les Garrigues Estepa Guijuelo Jamón de Huelva Jamón de Teruel Salchichón de Vic Llonganissa de Vic MahónMenorca Queso Manchego Cítricos Valencianos Cîtrics Valancians Jijona Turrón de Alicante Azafrán de la Mancha Comté Reblochon Reblochon de Savoie Roquefort Camembert de Normandie Brie de Meaux Emmental de Savoie Pruneaux d'Agen Pruneaux d'Agen micuits Huîtres de MarennesOléron Canards à foie gras du SudOuest: Chalosse Canards à foie gras du SudOuest: Gascogne Canards à foie gras du SudOuest: Gers Canards à foie gras du SudOuest: Landes Canards à foie gras du SudOuest: Périgord Canards à foie gras du SudOuest: Quercy Jambon de Bayonnec oils and animal fats oils and animal fats oils and animal fats oils and animal fats oils and animal fats oils and animal fats oils and animal fats oils and animal fats oils and animal fats oils and animal fats oils and animal fats oils and animal fats fresh, frozen and processed meats fresh, frozen and processed meats fresh, frozen and processed meats fresh, frozen and processed meats fresh, frozen and processed meats cheeses cheeses fresh and processed fruits and nuts fresh and processed fruits and nuts confectionery and baked products confectionery and baked products spices cheeses cheeses cheeses cheeses cheeses cheeses cheeses fresh and processed fruits and nuts fresh and processed fruits and nuts fresh, frozen and processed fish products fresh, frozen and processed meats fresh, frozen and processed meats fresh, frozen and processed meats fresh, frozen and processed meats fresh, frozen and processed meats fresh, frozen and processed meats Spain Spain Spain Spain Spain Spain Spain Spain Spain Spain Spain Spain Spain Spain Spain Spain Spain Spain Spain Spain Spain Spain Spain Spain France France France France France France France France France France France France France France France France drycured meats oils and animal fats essential oils cheeses France France France France cheeses cheeses France France Huile d'olive de HauteProvence Huile essentielle de lavande de HauteProvence Morbier Beaufortc Epoisses (continued next page) 688 INTELLECTUAL PROPERTY ARRANGEMENTS Table G.3 (continued) Indication Product Class Origin Maroilles Marolles Munstera cheeses cheeses France France cheeses cheeses cheeses cheeses cheeses cheeses cheeses cheeses cheeses cheeses cheeses cheeses cheeses cheeses cheeses cheeses Spices fresh and processed vegetable products vinegar vinegar fresh, frozen and processed meats fresh, frozen and processed meats fresh, frozen and processed meats fresh, frozen and processed meats drycured meats drycured meats drycured meats drycured meats cheeses cheeses France France France France France France France France France France France France France France France France France France Italy Italy Italy Italy Italy Italy Italy Italy Italy Italy Italy Italy cheeses Italy cheeses Italy cheeses cheeses cheeses cheeses cheeses cheeses cheeses fresh and processed fruits and nuts fresh and processed fruits and nuts fresh and processed fruits and nuts fresh and processed vegetable products fresh and processed fruits and nuts Italy Italy Italy Italy Italy Italy Italy Italy Italy Italy Italy Italy Munster Géromé Abondance Abondance Bleu d'Auvergne Livarot Cantal Fourme de Cantal Cantalet Petit Cantal Tomme de Savoie Pont L'Evêque Neufchâtel Chabichou du Poitou Crottin de Chavignol SaintNectaire Piment d'Espelette Lentille verte du Puy Aceto balsamico Tradizionale di Modena Aceto balsamico di Modena Cotechino Modena Zampone Modena Bresaola della Valtellina Mortadella Bologna Prosciutto di Parma Prosciutto di S. Daniele Prosciutto Toscano Prosciutto di Modena Provolone Valpadana Taleggio Asiagoa Fontinaa Gorgonzolaa Grana Padano Mozzarella di Bufala Campana Parmigiano Reggiano Pecorino Romano Pecorino Sardo Pecorino Toscano Arancia Rossa di Sicilia Cappero di Pantelleria Kiwi Latina Lenticchia di Castelluccio di Norcia Mela Alto Adige (continued next page) ADDITIONAL MATERIAL ABOUT TRADE MARKS AND GEOGRAPHICAL INDICATIONS 689 Table G.3 (continued) Indication Product Class Origin Südtiroler Apfel Pesca e nettarina di Romagna Pomodoro di Pachino Radicchio Rosso di Treviso Ricciarelli di Siena Riso Nano Vialone Veronese Speck Alto Adige Südtiroler Markenspeck Südtiroler Speck Veneto Valpolicella Veneto Euganei e Berici Veneto del Grappa Culatello di Zibello Garda Lardo di Colonnata Szegedi téliszalámi Szegedi szalámi Tiroler Speck Steirischer Kren Steirisches Kürbiskernöl Queijo S. Jorge Azeite de Moura Azeites de TrásosMontes Azeite do Alentejo Interior Azeites da Beira Interior Azeites do Norte Alentejano Azeites do Ribatejo Pêra Rocha do Oeste Ameixa d'Elvas Ananás dos Açores / S. Miguel Chouriça de carne de Vinhais Linguiça de Vinhais Chouriço de Portalegre Presunto de Barrancos Queijo Serra da Estrela Queijos da Beira Baixa Queijo de Castelo Branco Queijo Amarelo da Beira Baixa Queijo Picante da Beira Baixa Salpicão de Vinhais Gouda Holland Edam Holland Kalix Löjrom Magiun de prune Topoloveni fresh and processed fruits and nuts fresh and processed fruits and nuts fresh and processed vegetable products fresh and processed vegetable products confectionery and baked products cereals fresh, frozen and processed meats fresh, frozen and processed meats fresh, frozen and processed meats oils and animal fats oils and animal fats oils and animal fats fresh, frozen and processed meats fresh, frozen and processed meats fresh, frozen and processed meats fresh, frozen and processed meats fresh, frozen and processed meats fresh, frozen and processed meats fresh and processed vegetable products oilseeds cheeses oils and animal fats oils and animal fats oils and animal fats oils and animal fats oils and animal fats oils and animal fats fresh and processed fruits and nuts fresh and processed fruits and nuts fresh and processed fruits and nuts fresh, frozen and processed meats fresh, frozen and processed meats fresh, frozen and processed meats fresh, frozen and processed meats cheeses cheeses cheeses cheeses cheeses fresh, frozen and processed meats cheeses cheeses fresh, frozen and processed fish products fresh and processed fruits and nuts Italy Italy Italy Italy Italy Italy Italy Italy Italy Italy Italy Italy Italy Italy Italy Hungary Hungary Austria Austria Austria Portugal Portugal Portugal Portugal Portugal Portugal Portugal Portugal Portugal Portugal Portugal Portugal Portugal Portugal Portugal Portugal Portugal Portugal Portugal Portugal Netherlands Netherlands Sweden Romania a Denotes terms that can be used when accompanied by expressions of ‘kind’, ‘type’, ‘style’ or ‘imitation’. b, c May still be used in a transitional period lasting 5 and 10 years, respectively. Source: European Commission (2016). 690 INTELLECTUAL PROPERTY ARRANGEMENTS H Economic impacts of recommendations The reforms to Australia’s intellectual property (IP) arrangements outlined in this report have been made on the grounds that, collectively, they lead to a more balanced set of IP arrangements that improve the wellbeing of the community as a whole. The economic benefits expected to flow from implementing some of the more significant reforms are summarised in table 1 of the Overview, as well as in chapters dealing with specific IP rights. Quantitative estimates of reform impacts are provided in a number of cases, while in others the assessments are largely qualitative. This appendix explains in more detail how empirical estimates of impacts were derived. The explanations and evidence provided are intended to support independent assessment of the conceptual and measurement issues involved. It is also important to note that many of the benefits expected to flow from the Commission’s proposed reforms — particularly those arising from changes to copyright laws — are private or non-market benefits. While such reforms enhance the welfare of the Australian community overall, non-market benefits do not appear in standard measures of economic output or activity. This does not, however, diminish their importance or value to the community. The discussion of empirical work to follow focusses on two areas: • general reforms to patents • reforms specific to pharmaceutical patents. Detailed descriptions of the methodologies and data used to estimate the impacts of two other important reforms — introducing a ‘fair use’ exception in copyright law, and removing parallel import restrictions on books — are contained in appendix E. H.1 The impacts of general reforms to patents The Commission outlined a package of reforms in chapters 7 and 8 to improve the patent system’s effectiveness and efficiency. This section explores some of the potential impacts of those reforms. Most impacts cannot be quantified, and the Commission has been ECONOMIC IMPACTS OF RECOMMENDATIONS 691 deliberately conservative in its assumptions — hence this section only gives a partial indication of the benefits from reform. Reductions in the number of low-value patents The Commission is recommending a package of reforms to reduce the number of low-value patents. The reduction in patents for the standard patent system is estimated as follows. • A lower bound estimate is derived by assuming the reform package eliminates the greater tendency for IP Australia to grant patents compared to the European Patent Office (EPO). This is calculated as the difference between the annual number of patents that IP Australia accepts but the EPO rejects (810) and the number that the EPO accepts but IP Australia rejects (434).1 This provides an estimate of 376 patents per year. • An upper bound estimate is derived by assuming the reform package closes the gap in the patent application acceptance rate (post Raising the Bar) between IP Australia and the EPO. This provides an estimate of 1144 patents per year. The upper bound is a more comprehensive estimate. It accounts for all applications, while the lower bound only accounts for applications made to both IP Australia and the EPO. Even the upper bound is likely to be conservative. This is because the differences in grant rates between the EPO and IP Australia (which the upper bound estimate is based on) is likely to be currently biased downwards. This bias arises because it is unlikely that there has been enough time since Raising the Bar reforms for more ‘line ball’ applications to have been resolved in both offices. Since IP Australia is still more likely to accept such applications than the EPO (chapter 7), as they are resolved over time the grant rate differential would increase. A higher grant rate differential would increase the upper bound estimate. For the purposes of estimating the flow-on impacts of a reduction in low-value patents, the Commission took the halfway point between the upper and lower bound estimates (760). This is equivalent to around 4.5 per cent of annual patents granted.2 Over time this would materially reduce the patent stock, as existing low-value patents expire. Given current expiry rates, after 10 years the patent stock would be reduced by around 6700 patents. 3 1 These figures were estimated as follows. In the twenty months prior to Raising the Bar reforms, 1398 applications were accepted by IP Australia and rejected by the EPO, and 675 applications were accepted by the EPO and rejected by IP Australia. On an annual basis these figures convert to 839 and 405 applications. Post Raising the Bar, IP Australia has granted patents when the EPO has not in around 65 per cent of cases where there has been a different outcome across the two offices. Applying this percentage to the annual number of applications where there is a difference in outcomes (839 + 405), provides the estimates 810 and 434. 2 This is based on patents granted between 2011 and 2013. There was an increase in granted patents in 2014 and 2015 due to a surge in applications just before Raising the Bar reforms came into effect. 3 This amount is estimated using average yearly renewal rates between 2005 and 2015. 692 INTELLECTUAL PROPERTY ARRANGEMENTS Abolishing the innovation patent system (IPS) would also reduce low-value patents. While some inventions would instead receive protection under the standard patent system, most would not. Around 1700 innovation patents are granted annually. In estimating the flow-on impacts of patent reforms below, the Commission has in many places not accounted for the reduction in innovation patents. This, combined with a number of other careful assumptions, means the estimates are highly conservative. Increased returns to genuine innovation Reducing low-value patents improves the environment for innovation through a number of channels: fewer patent thickets, improved credibility signal in patents, reduced pendency in the patent office, and reduced scope for strategic misuse. These outcomes in turn increase the private returns to innovation. Private returns to innovation do not always correlate well with the social value of innovation, but in some circumstances they can be a reasonable proxy (appendix D). • Patent thickets are estimated to reduce over the long term by around 7 per cent. The largest estimated reductions are in electrical machinery and energy systems (14 per cent) and engines, pumps and turbines (11 per cent). 4 In addition to improving the general environment for innovation, the reduction in thickets is estimated to increase market entry rates by around 1 per cent on average, with increases of around 1.4-1.8 per cent for the above technologies. 5 Greater market entry in turn boosts competition, putting downward pressure on prices and driving innovation. • An improved credibility signal in patents would lead to more favourable finance terms for genuine innovations. A number of studies suggest that improvements in the signal value of patents leads to quicker and larger initial public offerings and venture capital funding for start-ups and small and medium-sized enterprises (Cockburn and MacGarvie 2009; Comino and Graziano 2015; Useche 2014). The decrease in the stock of patents is estimated to increase the number of firms receiving external financing in the long run by around 2–6 per cent. 6 These figures provide only a small indication of the benefits from reform. Most benefits cannot be quantified — the full impact of fewer low-value patents would be significant. 4 The reduction in patent thickets was estimated by drawing on the Commission’s estimates of the number of patent thickets in Australia (appendix D). All patents belonging to triples were identified, then some of these patents were randomly removed to proxy the estimated reduction in grant rates (based on the Commission’s estimate of 760 fewer low-value patents granted each year). Each removed patent and all of its citations were then dropped from any triples they belonged to. After this adjustment, if any firm in the triple no longer cited another then the triple was discarded. 5 These estimates are based on research suggesting that a 1 per cent reduction in thickets increases the probability of market entry by 0.11–0.13 per cent (Hall, Helmers and Graevenitz 2015; IPO 2013). 6 These estimates are based on the Commission’s estimated reduction in low-value patents and an estimate that a 1 per cent decrease in the stock of patents increases the number of firms receiving initial funding from outside investors by 1.263 per cent (Cockburn and MacGarvie 2009). These are rough estimates, as the data used in Cockburn are limited to firms in software markets. ECONOMIC IMPACTS OF RECOMMENDATIONS 693 Administrative and compliance cost savings A reduction in low-value patents would discourage firms and individuals with low-value inventions from applying for patent protection. This would reduce total administrative and compliance costs incurred from patent applications. Estimates of administrative and compliance costs for the standard patent system are as follows. • IP Australia estimates that (IP Australia 2016i). it costs $4463 to examine a typical patent • The costs to patent holders from complying with IP Australia’s regulatory processes, including an estimate of attorney fees, are derived from a regulatory costing assessment of IP Australia’s activities (KPMG 2014). The costs to domestic applicants from complying with filing, examination and opposition procedures are estimated to be around $17 115 per application. 7 Combining these estimates with the estimated reduction in standard patents suggests the reforms would reduce annual administrative and compliance costs in the order of $4.3 million. 8 IP Australia (2015k) estimated that abolishing the innovation patent system would result in compliance cost savings of around $11.1 million per annum (to the extent that some innovation patent holders instead seek a standard patent, some of these costs would migrate to the standard patent system). Reduced payments to international patent holders Restructuring patent renewal fees would reduce the risk that poor quality patents remain entrenched, as well as limit the scope for strategic misuse. By encouraging patent holders to release their patents early, restructuring fees would also reduce licensing and other payments made by Australians to international patent holders. The reduction in payments to international patent holders is estimated by multiplying average payments per patent by the decrease in patents held by international patent holders. Average payments and the decrease in patents are estimated for each year of the 20 years of a patent. 7 This estimate accounts for attorney fees and applicants’ in-house costs associated with application, examination and opposition proceedings. 8 The reduction in domestic patents is estimated as 7 per cent of the total reduction (0.07 x 760). Thus, the $4.3 million figure is derived as: $4463 x 760 + $17 115 x (0.07 x 760). 694 INTELLECTUAL PROPERTY ARRANGEMENTS • Average total payments across the life of a patent are estimated to be $309 000. 9 This amount is allocated across the 20 years of a patent using estimated distributions of patent revenues (figure H.1). • The decrease in patents is estimated using patent fee elasticities, which vary across the life of a patent (chapter 7). The decrease in international patents is estimated to be 91 per cent of the total decrease in each year. 10 Depending on the assumed distribution of patent revenues, restructuring renewal fees is conservatively estimated to reduce annual payments to international patent holders somewhere between $45-80 million. More significant benefits would arise from the reduced risk of patents being resurrected and reinterpreted to cover technology not envisaged, and the improved environment for innovation from fewer low-value patents. Figure H.1 Estimated distributions of patent revenues Distribution 0.120 0.080 0.040 0.000 Harris et al. (2013) Abrams (2013) Giummo (2014) 1 5 10 Year of patent 15 20 a The estimated distribution from Harris, Nicol and Gruen is for the ‘type 2’ patent. Data sources: Abrams (2013); Giummo (2014); Harris, Nicol and Gruen (2013). 9 Total payments are proxied by the private value of patents. The Australian Inventor Survey reported an average private patented invention value of $6.3 million and a median value of $800 000 (Jensen, Thomson and Yong 2009). Since the Commission’s reforms are expected to mainly target low-value patents, the Commission used the median value (about $966 000 in 2015 dollars). To isolate the value of the patent from the value of the invention, an estimate of the patent premium (the value that a patent adds to the value of an invention) is applied. After converting to 2015 dollars, this gives a median patent value of about $309 000. 10 IP Australia (sub. DR612) noted that 91 per cent of renewals from the fourth year on are from international firms. ECONOMIC IMPACTS OF RECOMMENDATIONS 695 H.2 The impacts of reforming pharmaceutical patents As noted in chapter 10, the Pharmaceutical Patents Review (PPR) estimated that the total savings to Government if extensions of term (EoTs) were entirely abolished would be $244 million (in 2012-13) per annum. Taking account of more recent data, the Commission has estimated that the savings from abolishing the EoT scheme would be $260 million per annum. Rather than abolish the scheme, the Commission has recommended that EoTs be granted only in instances of unreasonable regulatory delay. The Commission has estimated that the saving to taxpayers from implementing recommendation 10.1 would be $258 million per annum. The Commission reached this estimate by examining the difference between the costs to the Pharmaceutical Benefits Scheme (PBS) from the current system and the recommended approach. Broadly, this involves two elements: • the composition of extensions — the recommendation alters which drugs would receive extensions, and how long those extensions would be • the effects of generic entry — under the PBS statutory price reductions upon generic entry (at or near patent expiry), combined with subsequent rounds of price disclosure reductions result in reduced payments. These elements, and how they are combined to calculate the saving, are considered in turn. Changes in the composition of extensions EoTs are currently awarded where the time between the filing date of the patent and listing on the Australian Register of Therapeutic Good is at least 5 years. If the Commission’s recommendation 10.1 were adopted, the criteria for eligibility would change. Instead of ‘starting the clock’ at the filing date of the patent, the Commission’s recommendation focuses solely on those drugs for whom there has been a ‘unreasonable regulatory delay’. That is, where the delay attributable to the Therapeutic Goods Administration’s (TGA) processes (and not applicants’ actions) exceeds 255 working days. The Commission was provided with data on individual drugs that were granted an extension of term between January 2009 and May 2016, and the length of those extensions (IP Australia, pers. comm. 8 August 2016). The Commission was also provided with data (for the same set of drugs) on the total time that each application was with the TGA, time allocated to applicants for responses to requests from the Secretary of the Department of Health, and time mutually agreed that should be disregarded when calculating the time taken for regulatory approval (‘stop clock’ times) (Therapeutic Goods Administration, pers. comm., 18 August 2016). 696 INTELLECTUAL PROPERTY ARRANGEMENTS This data allowed the Commission to calculate the length of patent term extensions that would be available to those drugs within the set of existing extensions that are subject to regulatory delay of more than 255 working days, after allowing for stop clock times. Figure H.2 shows the distribution of extensions of term under the existing and proposed approaches. The changed pattern of extensions of term is explained by the fact that regulatory delay by the TGA of more than 255 days is relatively rare. Lengthy delays in approvals are often explained by the applicant and the regulator mutually agreeing that the clock should stop or the applicant delaying responses to requests. These delays are currently rewarded by compensatory additions to extensions of term. Under the recommendation, delays in regulatory approval that are not attributable to the regulator are not added to the term. The distribution of effective patent lives (figure H.3) demonstrates the improved targeting of the Commission’s recommendation. The average effective patent life under the proposed approach would be 10 years (in line with the recommendation of some of the PPR panel), compared to 14 years under the existing approach, but almost a third of patented drugs that received an extension would still have an effective life of 12 to 15 years. Note that the Commission’s estimates have only examined changes in the composition of the existing set of extensions. Beyond this, there may be rare cases where some drugs that were not previously eligible for extensions under the old formula would receive extensions under the Commission’s recommended approach. 11 11 Specifically, those drugs where TGA approval was sought relatively shortly after patenting, but the TGA’s process takes more than one year, would receive extensions where previously they had not. ECONOMIC IMPACTS OF RECOMMENDATIONS 697 Figure H.2 Distribution of EoTs – existing and recommended approaches Percentage of patents that received an EoT Per cent 70 60 Existing approach 50 Recommended approach 40 30 20 10 0 0 0-1 1-2 2-3 3-4 4-5 5 Years Data source: Commission estimates based on data provided by IP Australia and the Department of Health. Figure H.3 Distribution of effective patent lives – existing and recommended approaches Percentage of patents that received an EoT 90 Existing approach Recommended approach 80 70 Per cent 60 50 40 30 20 10 0 0-3 3-6 6-9 9-12 Effective patent life (years) 12-15 Data source: Commission estimates based on data provided by IP Australia and Department of Health. 698 INTELLECTUAL PROPERTY ARRANGEMENTS Effects of generic entry Reducing the length of extensions of patent term would allow generic drugs to enter the market in competition with drugs that were initially protected by patents. Division 3A of Part VII of the National Health Act 1953 (Cth) provides for an immediate 16 per cent price reduction when a new brand of a drug enters the market. The price reduction applies to any new or existing brand of the item. In addition to the statutory 16 per cent price reduction, the entry of a competitor drug triggers price disclosure requirements under Division 3B of that Part. Price disclosure involves the PBS applying additional price reductions by reference to market prices for drugs. The PBS checks market prices every 6 months and the prices for individual drugs can be reduced every 6 months until they exit the market. In order to determine the impact of the Division 3A and 3B price reductions, the Commission drew on data on price reductions for over 300 individual drugs following price disclosure rounds between December 2009 and October 2016 (Department of Health, pers. comm. 1 August 2016). These were drugs where a competitor product had entered the market — effectively meaning the patent had expired. As such, the set of drugs used for price reductions (off patent drugs) represents a different set of drugs to those who had been granted an extension (on patent drugs). Calculating an average price reduction for drugs subject to Divisions 3A and 3B is complicated by the fact that some drugs experience only the initial price reduction, whereas others reduce in price for up to 5 years after a competitor enters the market. Whilst theoretically possible, few drugs reduce in price more than 5 years after the entry of a competitor. Price reductions were calculated on a year-by-year basis for individual drugs, matched by the number of years from competitor entry. This resulted in an average price reduction of around 18 per cent in the first year, 29 per cent in the second year and around 40 per cent from the third year onwards. Estimating the savings to taxpayers Reducing the length of EoTs for patented drugs causes the price reduction mechanisms set out above to start earlier in the life of a drug. This results in savings for the PBS. For each drug that received a term extension between 1 January 2011 and 31 December 2016, the Commission calculated how many years earlier patent protection would have expired if its recommendation were applied. 12 The profile of savings resulting from the price reduction analysis described above was applied on a drug-by-drug basis to the PBS 12 The Commission examined a five-year window back from the most recently available data. A period of five years was selected as it reflected a full cycle of price disclosure reductions. ECONOMIC IMPACTS OF RECOMMENDATIONS 699 expenditure on each drug in 2015-16 (roughly equating to the application of an 18 per cent saving in the first year, 29 per cent in the second year and 40 per cent thereafter, following generic entry). 13 The average per year saving for each drug was then summed to arrive at a total saving of $258 million per annum for the PBS. Using 2015-16 PBS expenditure data likely underestimates both the nominal and real saving that could be achieved, as it does not account for increases in PBS expenditure between now and when drugs would be subject to a reduction in extension of term (which may be much larger due to ageing of the population and increasing expectations of consumers around health outcomes). On the other hand, the estimate also does not seek to account for the time value of money by discounting savings that occur in the future. The analysis uses a past set of decisions about EoT to consider the effect if the recommended approach had been in force for that period. Implicitly, such an approach assumes that the future pattern of drugs subject to extension of term is the same as those granted extensions of term over the period 2011-2015 inclusive. 13 For example, one drug granted an extension of term during the period under analysis was granted an extension of 2.98 years. Under the proposed approach, it would it would receive an extension of 0.11 years, resulting in price reductions occurring 2.87 years earlier. The hypothetical average price reduction over those 2.87 years would be around 28 per cent. PBS expenditure on that drug in 2015-16 was $1.8 million. Saving 28 per cent of $1.8 million each year results in a saving of around $500 000 per annum. 700 INTELLECTUAL PROPERTY ARRANGEMENTS References ABARES (Australian Bureau of Agricultural and Resource Economics and Sciences) 2015a, Agricultural commodities: March quarter 2015, Canberra. —— 2015b, Agricultural commodity statistics 2015, December, Canberra. —— 2016, Agricultural commodities: March quarter 2016, Canberra. Abrams, D., Akcigit, U. and Popadak, J. 2013, Patent value and citations: creative destruction or defensive disruption?, PIER Working Paper 13-065, Penn Institute for Economic Research. ABS (Australian Bureau of Statistics) 2012, 8111.0 - Research and Experimental Development, Higher Education Organisations, Australia, 2010, http://www.abs.gov.au/AUSSTATS/abs@.nsf/Lookup/8111.0Main+Features12010 (accessed 15 April 2016). —— 2014, 8158.0 Innovation in Australian Business, 2012-13, http://www.abs.gov.au/AUSSTATS/abs@.nsf/allprimarymainfeatures/99990EDED8F7 67E7CA257FF60017FC61?opendocument (accessed 31 August 2016). —— 2015a, 5204.0 - Australian System of National Accounts, 2014-15, http://www.abs.gov.au/AUSSTATS/abs@.nsf/productsbyCatalogue/110953FFA28D4E 52CA2572110002FF03?OpenDocument (accessed 30 March 2016). —— 2015b, 5216.0 - Australian System of National Accounts: Concepts, Sources and Methods, 2014, http://www.abs.gov.au/AUSSTATS/abs@.nsf/allprimarymainfeatures/ 1A9681CB1164057CCA257F7D00177CBF?opendocument (accessed 29 March 2016). —— 2016, 8165.0 - Counts of Australian Businesses, including Entries and Exits, Jun 2011 to Jun 2015, http://www.abs.gov.au/ausstats/abs@.nsf/mf/8165.0 (accessed 12 July 2016). ACBPS (Australian Customs and Border Protection Service) 2013, Australian Customs and Border Protection Service: Annual Report 2012–13, Annual Report, Canberra. —— 2014, Australian Customs and Border Protection Service: Annual Report 2013–14, Annual Report, Canberra. —— 2015, Australian Customs and Border Protection Service: Annual Report 2014–15, Annual Report, Canberra. ACCC (Australian Competition and Consumer Commission) 1999, Potential Consumer Benefits of Repealing the Importation Provisions of the Copyright Act 1968 as they Apply to Books and Computer Software, Canberra. REFERENCES 701 —— 2001, Summary of the Commission’s March 1999 Report on The Potential Consumer Benefits of Repealing the Importation Provisions of the Copyright Act 1968 as they Apply to Books and Computer Software – Including Price Updates for Books, Computer Software and Sound Recording, Canberra. —— 2014a, ACCC acts on beer labelling, https://www.accc.gov.au/media-release/acccacts-on-beer-labelling (accessed 1 March 2016). —— 2014b, Maggie Beer Products acknowledges labelling likely to be misleading, https://www.accc.gov.au/media-release/maggie-beer-products-acknowledges-labellinglikely-to-be-misleading (accessed 1 March 2016). —— 2014c, Submission to the Competition Policy Review on Reinvigorating Australia’s Competition Policy, 25 June, Canberra. Access Copyright 2016, Access Copyright Annual Report 2015, Canada. ACIP (Advisory Council on Intellectual Property) 1995, Review of the Petty Patent System, Canberra. —— 2003a, Report on a review of the patenting of business systems, Canberra. —— 2003b, Should the jurisdiction of the Federal Magistrates Service be extended to include patent, trade mark and design matters?, November, Canberra. —— 2004, Review of trade mark enforcement, Final Report, Canberra. —— 2006, A review of the relationship between trade marks and business names, company names, and domain names, Final Report, Canberra. —— 2010a, A review of enforcement of Plant Breeder’s Rights, Final Report, January, Canberra. —— 2010b, Patentable Subject Matter, Final Report, Canberra. —— 2010c, Post-Grant Patent Enforcement Strategies, January, Canberra. —— 2013a, Government Response - review of the enforcement of trade marks, https://www.ipaustralia.gov.au/about-us/public-consultations/archive-of-ip-reviews/ipreviews/Government-Response-review-of-the-enforcement-of-trademarks/government-response (accessed 11 July 2016). —— 2013b, Review of the Designs System, Issues Paper, Canberra. —— 2014, Review of the Designs System, Options Paper, December, Canberra. —— 2015a, Review of the Designs System, Final Report, March, Canberra. —— 2015b, Review of the Innovation Patent System: Final Report, May, Canberra. Adler, L. 2016, ‘The evidence is in: copyright proposal won’t mean cheaper books’, Sydney, Sydney Morning Herald, 3 June, http://www.smh.com.au/comment/ourcopyright-regime-needs-to-be-strengthened-not-weakened-20160602-gp9x7l.html (accessed 20 September 2016). 702 INTELLECTUAL PROPERTY ARRANGEMENTS AFGC (Australian Food and Grocery Council) 2016, Submission to DFAT on the Australia-European Union Free Trade Agreement, March, Canberra. Agtrans Research 2008, An Economic Analysis of Investment in Breeding of Durum, Triticale and Oats (Other Cereals), April, GRDC Project Code: ATR00002, Grains Research and Development Corporation, Brisbane. —— 2012a, An Economic Analysis of GRDC Investment in National Lupin Breeding for Southern Australia, GRDC Project Code: ATR00010, April, Grains Research and Development Corporation, Brisbane. —— 2012b, An Economic Analysis of Investment in the National Variety Trials, April, GRDC Project Code: ATR00010, Grains Research and Development Corporation, Brisbane. Ai Group 2014, Submission to the Senate Standing Committee on Economics inquiry into Australia’s Innovation System, Canberra. AIC (Australian Institute of Criminology) 2008, Intellectual property crime and enforcement in Australia, Research and Public Policy Series No. 94, Canberra. Akerlof, G., Arrow, K., Bresnahan, T., Buchanan, J., Coase, R., Cohen, L., Friedman, M., Green, J., Hahn, R., Hazlett, T., Hemphill, C., Litan, R., Noll, R., Schmalensee, R., Shavell, S., Varian, H. and Zeckhauser, R. 2002, The Copyright Term Extension Act of 1998: An Economic Analysis, May, AEI-Brookings Joint Center for Regulatory Studies, Washington D.C. Albini, S. 1993, The Problem with Music, The Baffler, http://thebaffler.com/salvos/theproblem-with-music (accessed 23 October 2015). —— 2014, Steve Albini on the surprisingly sturdy state of the music industry – in full, The Guardian, http://www.theguardian.com/music/2014/nov/17/steve-albinis-keynoteaddress-at-face-the-music-in-full (accessed 23 October 2015). Alexandra, A., Lee, J. and Vanclay, F. 2002, Innovation, Exclusion and Commodification of Plant Types: A Social and Philosophical Investigation of Plant Variety Rights in Australia, Working Paper No. 2002/18, Centre for Applied Philosophy and Public Ethics (CAPPE), Canberra. Allegra, F. 2016, How to find Pokémon Go’s gyms and PokéStops, Polygon, http://www.polygon.com/2016/7/7/12118576/pokemon-go-pokestop-gym-locationsmap-guide (accessed 9 July 2016). ALRC (Australian Law Reform Commission) 1995, Designs, June, Report No. 74, Sydney. —— 2004, Genes and Ingenuity: Gene Patenting and Human Health, Report No. 99, Sydney. —— 2010, Designs, https://www.alrc.gov.au/inquiries/designs (accessed 23 February 2016). —— 2013, Copyright and the Digital Economy, Discussion Paper No. 79, May, Sydney. —— 2014, Copyright and the Digital Economy, February, Report No. 122, Sydney. REFERENCES 703 —— 2015, Connection to Country: Review of the Native Title Act 1993 (Cth), Report No. 126, Sydney. Alston, J., Gray, R. and Bolek, K. 2012, Farmer-Funded R&D: Institutional Innovations for Enhancing Agricultural Research Investments, Working Paper, March, University of California, Giannini Foundation, University of Saskatchewan. ANAO (Australian National Audit Office) 2014, Administering regulation achieving the right balance, Better Practice Guide, Canberra. APEC (Asia-Pacific Economic Cooperation) 2016, Intellectual Property Rights Experts Group, Intellectual Property Rights Experts Group, http://www.apec.org/Groups/Committee-on-Trade-and-Investment/IntellectualProperty-Rights-Experts-Group.aspx (accessed 15 April 2016). APRA AMCOS 2015, 2014-15 APRA AMCOS Year http://sustainability.apraamcos.com.au/ (accessed 13 April 2016). in Review, APSC (Australian Public Service Commission) 2014, Capability review: IP Australia, April, Canberra. —— 2016, APS agency listing - agencies covered by the Public Service Act 1999, http://www.apsc.gov.au/publications-and-media/current-publications/australian-publicservice-agencies (accessed 16 September 2016). Arai, H. 2005, ‘Intellectual Property Strategy in Japan’, International Journal of Intellectual Property - Law, Economy and Management, vol. 1, pp. 5–12. ARIA (Australian Recording Industry Association) 2014, Australian Wholesale Sales for the Years Ended 31 December, http://www.aria.com.au/pages/documents/ physdigsalesxunits.pdf (accessed 31 August 2016). —— 2015, ARIA Charts - End Of Year Charts - Top 100 Albums 2015, ARIA Top 100 Albums 2015, http://www.aria.com.au/aria-charts-end-of-year-charts-top-100-albums2015.htm (accessed 15 July 2016). Arnold, A. 2015, ‘A Game-Theoretic Approach to Modelling Crop Royalties’, Doctor of Philosophy thesis, The University of Adelaide, Adelaide. Arora, A. and Athreye, S. 2012, Patent incentives: returns to patenting and the inducement for research and development, Intellectual Property Office Research Paper No. 2012/20, UK Intellectual Property Office, Cardiff, UK. ——, Ceccagnoli, M. and Cohen, W. 2008, ‘R&D and the patent premium’, International Journal of Industrial Organization, vol. 26, pp. 1153–1179. Arts Law Centre of Australia 2003, Code of Conduct for Copyright Collecting Societies, http://www.artslaw.com.au/articles/entry/code-of-conduct-for-copyright-collectingsocieties/ (accessed 13 April 2016). —— Protecting your designs, http://www.artslaw.com.au/info-sheets/infosheet/protecting-your-designs/ (accessed 6 September 2016). 704 INTELLECTUAL PROPERTY ARRANGEMENTS —— Submission, Submission to the IP Australia consultation on Indigenous Knowledge and the intellectual property system, Canberra. Arundel, A. 2001, ‘The relative effectiveness of patents and secrecy for appropriation’, Research Policy, vol. 30, pp. 611–624. Ashton, T.P. and Vekris, E. 2016, Yet another bump in the road to achieving a simplified global filing procedure for designs, http://www.lexology.com/library/detail.aspx?g= 1d22cd4c-edc0-467a-acb9-604ed6e01c20 (accessed 31 August 2016). Ashurst Australia 2013, EU Commission issues first fines for ‘reverse payment’ settlements, Media release, 25 June 2013, Sydney. ASIC (Australian Securities & Investment Commission) 2015, How to protect your business name, http://asic.gov.au/for-business/registering-a-business-name/before-youstart/how-to-protect-your-business-name/ (accessed 7 March 2016). —— Fit for the future A capability review of the Australian Securities and Investments Commission, December, Canberra. Atal, V. and Bar, T. 2014, ‘Patent quality and a two-tiered patent system’, The Journal of Industrial Economics, vol. 62, no. 3, pp. 503–540. ATO (Australian Taxation Office) 2012, ATO-Treasury https://www.ato.gov.au/General/New-legislation/In-detail/ATO---Treasuryprotocols/ATO---Treasury-protocol/ (accessed 31 August 2016). protocol, —— 2015, Guide to depreciating assets 2015, Canberra. AusBiotech 2013, AusBiotech response to consultation on an objects clause and an exclusion from patentability, Canberra. —— 2015, AusBiotech submission to IP Australia’s consultation on the ACIP recommendation to abolish the Innovation Patent System, Melbourne. —— 2016, Bio-therapeutics - Fast pageid=174 (accessed 7 April 2016). Facts, http://www.ausbiotech.org/content.asp? AUSTLII (Australasian Legal Information Institute) 2016, Australian Trade Marks Office, http://www.austlii.edu.au/au/cases/cth/ATMO/ (accessed 17 August 2016). Austrade (Australian Trade Commission) 2015, Australian Intellectual Property laws, https://www.austrade.gov.au/International/Invest/Guide-to-investing/Running-abusiness/Understanding-Australian-business-regulation/Australian-IntellectualProperty-laws (accessed 12 November 2015). Australian Government 2010, Government Response: A Review of Plant Breeder’s Rights, http://www.acip.gov.au/reviews/government-response-enforcement-pbr/ (accessed 27 July 2015). —— 2011, Australian Government Response to Senate Community Affairs References Committee Gene Patents Report, November, Canberra. REFERENCES 705 —— 2013, Government Response - Review of post-grant Patent Enforcement Strategies, http://www.ipaustralia.gov.au/about-us/public-consultations/archive-of-ip-reviews/ipreviews/Government-Response-review-of-the-enforcement-of-trade-marks/ (accessed 21 January 2015). —— 2014a, Government statement on the Pharmaceutical Patent Review final report, https://www.ipaustralia.gov.au/sites/g/files/net856/f/government_statement_on_the_ph armaceutical_patent_review_final_report.pdf (accessed 10 February 2016). —— 2014b, The Australian Government guide to regulation. ——2015a, Australian Government Response to the Competition Policy Review, Canberra. ——2015b, Boosting the commercial returns from research, Canberra. —— 2015c, Review of Research Policy and Funding Arrangements, Canberra. —— 2015d, Vision for a science nation: responding to science, technology, engineering and mathematics: Australia’s future, Canberra. —— 2016a, Australian Government Response to the Senate Foreign Affairs, Defence and Trade Reference Committee Report: Blind Agreement: reforming Australia’s treatymaking process, February, Canberra. —— 2016b, Government response - ACIP Review of the Designs System, https://www.ipaustralia.gov.au/about-us/news-and-community/news/governmentresponse-acips-review-designs-system (accessed 28 June 2016). —— 2016c, Government Response - consideration of extending the jurisdiction of the Federal Magistrates Service to patent, trade marks and design matters, Text, https://www.ipaustralia.gov.au/about-us/public-consultations/archive-ip-reviews/ipreviews/consideration-of-extending-the-jurisdiction/government-response (accessed 1 September 2016). —— 2016d, R&D Tax Incentive Review, http://www.business.gov.au/grants-andassistance/innovation-rd/RD-TaxIncentive/Pages/randd-tax-incentive-review.aspx (accessed 15 April 2016). —— 2016e, Statements of Expectations, http://www.treasury.gov.au/PolicyTopics/PublicPolicyAndGovt/Statements-of-Expectations (accessed 31 August 2016). Australian Technology Network of Universities and AI Group 2015, Innovate and prosper - ensuring Australia’s future competitiveness through university-industry collaboration. Bakels, R. and Hugenholtz, P. 2002, The patentability of computer programmes: Discussion of the European-level legislation in the field of patents for software, Working Paper prepared for the European Parliament, Luxembourg. Baldwin, P. 2014, The Copyright Wars: Three Centuries of Trans-Atlantic Battle, Princeton University Press, Princeton. Baldwin, R., Gellatly, G., Johnson, J. and Peters, V. 1998, Innovation in dynamic service industries, Statistics Canada, Ministry of Industry, Ottawa, Canada. 706 INTELLECTUAL PROPERTY ARRANGEMENTS Bandcamp 2015, Pricing, https://bandcamp.com/pricing (accessed 20 October 2015). Banks, G. 2009, Evidence-Based Policy-Making: What Is It? How Do We Get It?, Productivity Commission Chairman’s Speech, Canberra. Barnes, D. 2006, ‘A new economics of trademarks’, Northwestern Journal of Technology and Intellectual Property, vol. 5, no. 1, pp. 22–67. Baron, J. and Delcamp, H. 2011, The private and social value of patents in discrete and cumulative innovation, Cerna Working Paper, 2010–07, Paris, France. Baudry, M. and Dumont, B. 2009, A Bayesian Real Option Approach to Patents and Optimal Renewal Fees, LEMNA Working Paper 2009/09, Nantes, France. Baum, J.. and Silverman, B.. 2004, ‘Picking winners or building them? Alliance, Intellectual, and Human Capital as selection criteria in venture financing and performance of biotechnology startups’, Journal of Business Venturing, vol. 19, no. 3, pp. 411–436. Baxter IP 2016, Innovation Patent Applications in Australia Patent Process & Costs, Baxter IP Patent & Trademark Attorneys, http://www.baxterip.com.au/innovationpatent-applications (accessed 18 July 2016). Beckerman-Rodau, A. 2010, ‘The Problem with Intellectual Property Rights: Subject Matter Expansion’, Yale Journal of Law & Technology, vol. 13, no. 1, pp. 36–89. Benson, C. and Magee, C. 2015, ‘Quantitative determination of technological improvement from patent data’, PLoS ONE, vol. 10, no. 4, pp. 1–23. Bently, L. 2012, The Return of Industrial Copyright?, University of Cambridge Faculty of Law Research Paper No. 19/2012, July, Cambridge, UK. ——, Blakeney, M., von Borrie, C., Chesser, R., Cloonan, M., Cook, N., Gibson, J., Hugenholtz, B., Kay, J., Kretschmer, M., MacQueen, H. and Waelde, C. 2008, Creativity stifled? A Joined Academic Statement on the Proposed Copyright Term Extension for Sound Recordings, 16 June, Bournemouth, UK. Bernaski, K. 2014, Saving Mickey Mouse: The Upcoming Fight FOR Copyright Term Extension in 2018, Seton Hall University, Newark. Bessen, J. 2004, ‘Holdup and Licensing of Cumulative Innovations with Private Information’, Economics Letters, vol. 82, pp. 321–326. —— 2006, The value of US patents by owner and patent characteristics, Working Paper, 06–46, Boston University, USA. —— and Hunt, R. 2007, ‘An Empirical Look at Software Patents’, Journal of Economics and Management Strategy, vol. 16, no. 1, pp. 157–189. —— and Maskin, E. 2009, ‘Sequential innovation, patents, and imitation’, RAND Journal of Economics, vol. 40, no. 4, pp. 611–635. —— and Meurer, M. 2008, Patent Failure How Judges, Bureaucrats, and Lawyers Put Innovators at Risk, Princeton University Press, United States. REFERENCES 707 Bilir, L. 2014, ‘Patent Laws, Product Life-Cycle Lengths, and Multinational Activity’, American Economic Review, vol. 104, no. 7, pp. 1979–2013. Birss, C. 2015, ‘To boldly reform IP dispute resolution: experience in the IP Enterprise Court’. Blewett, N., Goddard, N., Pettigrew, S., Reynolds, C. and Yeatman, H. 2011, Labelling Logic: Review of the Food Labelling Law and Policy, Canberra. Blind, K., Edler, J. and Friedewald, M. 2005, Software Patents: Economic Impacts and Policy Implications, Edward Elgar Publishing, Cheltenham, UK. ——, Edler, J., Schmoch, U., Andersen, B., Howells, J., Miles, I., Roberts, J., Hipp, C., Green, L., Herstatt, C. and Evangelista, R. 2003, Patents in the service industries, EC contract No ERBHPV2-CT-1999-06, European Commission, Brussels. Bloch, D. 2016, ‘Alternatives to March in Rights’, Vanderbuilt Journal of Entertainment and Technology Law, vol. 18, no. 2, pp. 247–265. Boldrin, M. and Levine, D. 2008, Against Intellectual Monopoly, Cambridge University Press, New York. Bollinger, T., Nelson, R., Self, K.M. and Turnbull, S.J. 1999, ‘Open-Source Methods: Peering Through the Clutter’, IEEE Software, vol. 16, no. 4, pp. 8–11. Borderlex 2015, Geographical indications in TTIP: a new European controversy, http://www.borderlex.eu/expert-ttip-worsen-position-eu-registered-gis-us/ (accessed 30 March 2016). Bosch, J. 2014, ‘Continuous Software Engineering: An Introduction’, in Bosch, J. (ed), Continuous Software Engineering, Springer International Publishing, Switzerland. Bosland, J., Weatherall, K. and Jensen, P. 2006, ‘Trade mark and counterfeit litigation in Australia’, Intellectual Property Quarterly, vol. 4, p. 347. Boston Consulting Group 2012, Shifting Sands: New Challenges in Semiconductors, https://www.bcgperspectives.com/content/articles/technology_digital_economy_shiftin g_sands_new_challenges_in_semiconductors/ (accessed 29 March 2016). Bovens, M. 2005, Public Accountability: A framework for the analysis and assessment of accountability arrangements in the public domain, Draft discussion paper for Connecting Excellence on European Governance Research Group, Utrecht, The Netherlands. Bowen, L. 1988a, Circuit Layouts Bill 1988, Explanatory Memorandum, Parliament of Australia, Canberra. —— 1988b, Second Reading Speech - Copyright Amendment Bill 1988, 3 November, House of Representatives, Proof Hansard, Parliament of Australia, Canberra, pp. 2397– 2398. Bowker 2012, Bowker Research Shows Australia is a Global Leader in e-Book Adoption, http://www.bowker.com/news/2012/290244851.html (accessed 21 October 2015). 708 INTELLECTUAL PROPERTY ARRANGEMENTS Boyle, J. 2015, ‘(When) Is Copyright Reform Possible? Lessons from the Hargreaves Review’, http://www.thepublicdomain.org/wp-content/uploads/2015/02/Is-CopyrightReform-Possible1.pdf (accessed 18 December 2015). Branstetter, L. 2005, ‘Exploring the link between academic science and industrial innovation’, Annals of Economics and Statistics, vol. 79/80, pp. 119–142. ——, Fisman, R. and Foley, F. 2006, ‘Do Stronger Intellectual Property Rights Increase International Technology Transfer?’, The Quarterly Journal of Economics, pp. 321–349. Braunerhjelm, P. and Svensson, R. 2009, ‘The inventor’s role: Was Schumpeter Right?’, Journal of Evolutionary Economics, vol. 20, no. 3, pp. 413–444. Breitwieser, A. and Foster, N. 2012, Intellectual property rights, innovation and technology transfer: a survey, 20 January, MPRA Paper No. 36094, University of Vienna, Vienna Institute for International Economic Studies. BSA The Software Alliance 2015, BSA The Software Alliance’s Comments on ACIP’s Recommendation on the Innovation Patent System, Singapore. Buccafusco, C. and Heald, P. 2012, Do Bad Things Happen When Works Enter the Public Domain?: Empirical Tests of Copyright Term Extension, Chicago-Kent College of Law Legal Studies Research Paper, No. 2012-04, Illinois Institute of Technology, Chicago. Burrell, R. and Handler, M. 2012, ‘Rethinking the presumption of registrability in trade mark law’, Monash University Law Review, vol. 38, no. 1. —— and —— 2016, Australian Trade Mark Law, 2nd edn, Oxford University Press, Melbourne. Byma, J. and Leiponen, A. 2007, Can’t block, must run: small firms and appropriability, Working Paper, 1–07, The Mario Einaudi Center for International Studies, New York, USA. van Caenegem, W. 2015, Quality local food products - some aspects of trademarks law and GIs, Faculty of Law Corporate Governance eJournal Paper 35, Bond University, http://epublications.bond.edu.au/cgej/35 (accessed 19 February 2016). Campbell-Kelly, M. and Valduriez, P. 2005, ‘A Technical Critique of Fifty Software Patents’, Marquette Intellectual Property Law Review, vol. 9, no. 2, pp. 249–305. Carr, K. 2011, Intellectual Property Laws Amendment (Raising the Bar) Bill 2011, Explanatory Memorandum, Parliament of Australia, Canberra. Carroll, M. 2011, ‘A framework for tailoring intellectual property rights’, in Kur, A. and Mizaras, V. (eds), The Structure of Intellectual Property Law: Can One Size Fit All?, Edward Elgar, Cheltenham, UK, pp. 15–39. Carter, S. 1990, ‘The trouble with trademark’, The Yale Law Journal, vol. 99, pp. 759–800. Casella, M. and Calvi, L. 2009, ‘New journal models and publishing perspectives in the evolving digital environment’, presented at World Library and Information Congress: 75th lfla General Conference Council, Milan, Italy. REFERENCES 709 Cassiman, B., Veugelers, R. and Zuniga, P. 2008, ‘In search of performance effects of (in)direct industry science links’, Industrial and Corporate Change, vol. 17, no. 4, pp. 611–646. Castro, D. 2016, Chile: Trends Favoring Protection Of Pharmaceutical Patents. Why Apply For A Patent In Chile?, http://www.mondaq.com/x/466600/Patent/Trends+ Favoring+Protection+Of+Pharmaceutical+Patents+Why+Apply+For+A+Patent+In+Ch ile (accessed 6 April 2016). Caulkins, J.P., Feichtinger, G., Grass, D., Hartl, R.F., Kort, P.M. and Seidl, A. 2013, ‘When to make proprietary software open source’, Journal of Economic Dynamic & Control, vol. 37, pp. 1182–1194. CAV (Consumer Affairs Victoria) 2008, Institutional arrangements for consumer protection agencies, April, Research Paper No. 13, Melbourne. CBC (Competition Bureau Canada) 2014, ‘Patent Litigation Settlement Agreements: A Canadian Perspective’, presented at the A paper prepared for the Global Antitrust Institute, Washington D.C., 23 September. —— 2016, Intellectual Property Enforcement Guidelines, Technical Guidance Document, http://www.competitionbureau.gc.ca/eic/site/cb-bc.nsf/eng/04031.html (accessed 18 July 2016). Chalmers, R. 2007, ‘Evergreen or Deciduous? Australian Trends in Relation to the “Evergreening” of Patents’, Melbourne University Law Review, vol. 30, pp. 29–61. Chandler, K. and Golder, T. 2012, ‘Comparison Tables: Australia’, in Wilkof, N. and Basheer, S. (eds), Overlapping intellectual property rights, Oxford University Press, UK. CHOICE 2015, Desperately Seeking Streaming - Research update: CHOICE Digital Consumers Paying for Content Behaviour & Attitudes, Sydney. Christie, A. 2011, ‘Maximising permissible exceptions to intellectual property rights’, in Kur, A. and Mizaras, V. (eds), The Structure of Intellectual Property Law: Can One Size Fit All?, Edward Elgar, Cheltenham, UK, pp. 121–135. ——, Dent, C. and Lim, A. 2013, An empirical comparison of the outcome of patent examination in the USPTO, the EPO and IP Australia, TILEC Conference 2013, Tilburg, The Netherlands. ——, Dent, C., McIntyre, P., Wilson, L. and Studdert, D. 2013, ‘Patents associated with high-cost drugs in Australia’, PLoS ONE, vol. 8, no. 4. —— and Dias, E. 2005, ‘The new right of communication in Australia’, The Sydney Law Review, vol. 27, no. 2, pp. 237–262. CIPO (Canadian Intellectual Property Office) 2015, Industrial Design Office Practices, CIPO, Gatineau, Canada. Clark, E. 1994, ‘Nintendo Wins the Circuit Layout Game: Nintendo Co Ltd v Centronics Systems Pty Ltd’, Journal of Law, Information and Science, vol. 5, no. 2. 710 INTELLECTUAL PROPERTY ARRANGEMENTS CLRC (Copyright Law Review Committee) 1987, Report on Performers’ Protection, Canberra. —— 1988, Report on Moral Rights, Canberra. —— 2002, Copyright and Contract, April, Canberra. CNET 2016, ISPs and rights holders quietly ditch three-strikes piracy scheme, http://www.cnet.com/au/news/isps-and-rights-holders-quietly-ditch-three-strikespiracy-scheme/ (accessed 8 September 2016). COAG & ANZFMRC (Council of Australian Governments and Australia and New Zealand Food Regulation Ministerial Council) 2014, Food labelling - progress report, http://www.foodlabellingreview.gov.au/internet/foodlabelling/publishing.nsf/Content/P rogress_report_December_2014 (accessed 2 August 2016). Cockburn, I. and MacGarvie, M. 2009, ‘Patents, Thickets and the Financing of Early-Stage Firms: Evidence from the Software Industry’, Journal of Economics and Management Strategy, vol. 18, no. 3, pp. 729–773. —— and —— 2011, ‘Entry and Patenting in the Software Industry’, Management Science, vol. 57, no. 5, pp. 915–933. Cohen, J.E. and Lemley, M. 2001, ‘Patent Scope and Innovation in the Software Industry’, California Law Review, vol. 89, pp. 1–57. Cohen, W., Goto, A., Nagata, A., Nelson, R. and Walsh, J. 2002, ‘R&D spillovers, patents and the incentives to innovate in Japan and the United States’, Research Policy, vol. 31, pp. 1349–1367. ——, Nelson, R. and Walsh, J. 2000, Protecting their intellectual assets: appropriability conditions and why US manufacturing firms patent (or not), NBER Working Paper No. 7552, Washington D.C. —— and Walsh, J. 1998, R&D Spillovers, Appropriability and R&D Intensity: A Survey Based Approach, Carnegie-Mellon University, Pittsburgh, USA. Coles, D. 2007, ‘Funding Plant Breeding in Australia – A Review of End Point Royalties and Future Funding Requirements’, http://www.seedquest.com/forum/c/ ColesDonald/07jul.htm (accessed 4 February 2016). Collier, A. 2007, ‘Australian Framework for the Commercialisation of University Scientific Research’, Prometheus, vol. 25, no. 1, pp. 51–68. Combe, E. and Prister, E. 2000, Patents against imitators: an empirical investigation on French data, Working Paper, Cahiers de la MSE, Paris, France. Comino, S. and Graziano, C. 2015, ‘How many patents does it take to signal innovation quality?’, International Journal of Industrial Organisation, vol. 43, pp. 66–79. Commonwealth Courts Portal 2011, Applications to File: Herman Miller, Inc & Anor v Tebi Pty Ltd ACN 001 478 404 t/as Matt Blatt. REFERENCES 711 Communications Alliance 2014, Online Copyright Infringement Research Report, http://www.commsalliance.com.au/__data/assets/pdf_file/0003/46857/OnlineCopyright-Infringement-Summary.pdf (accessed 28 March 2016). Concept Economics 2009, Economic Modelling on Estimated Effect of Copyright Term Extension on New Zealand Economy, New Zealand Ministry of Business, Innovation and Employment, New Zealand. Conti, A., Thursby, M.. and Rothaermel, F.. 2013, ‘Show me the right stuff: signals for high tech startups’, Journal of Economics and Management Strategy, vol. 20, no. 2, pp. 341–364. Coordinating Committee on Innovation 2013, National Principles of Intellectual Property Management for Publicly Funded Research. Copyright Advisory Office, Columbia University 2016, Fair Use Checklist, Fair Use Checklist, https://copyright.columbia.edu/basics/fair-use/fair-use-checklist.html (accessed 13 April 2016). Copyright Agency 2015, 2014-15 Annual Review, 2014-15 Copyright Agency Annual Review, http://www.annualreview2014-15.com/copyright.html (accessed 13 April 2016). —— 2016, International copyright, http://copyright.com.au/about-copyright/internationalcopyright/ (accessed 21 March 2016). Copyright Licensing Agency (UK) 2015, Accounts for the year ending March 2015, Annual Report, Copyright Licensing Agency (UK), UK, http://www.cla.co.uk/data/pdfs/accounts/cla_statutory_accounts_year_to_2015.pdf (accessed 5 September 2016). Cormann, M. 2014, Smaller Government - Towards a sustainable future, December, Australian Government, Canberra. Cornell University, INSEAD and WIPO 2015, The Global Innovation Index 2015: Effective Innovation Policies for Development, Fontainebleau, Ithaca, and Geneva. Cornelli, F. and Schankerman, M. 1999, ‘Patent renewals and R&D incentives’, RAND Journal of Economics, vol. 30, no. 2, pp. 197–213. Cornish, W.R. 1989, Intellectual property : patents, copyright, trade marks, and allied rights, 2nd edn, Sweet & Maxwell, London. Corones, S. 2005, Australian Business Law Review, vol. 33, no. 2, pp. 150–155. Cotters 2016, Innovation patents, http://www.cotters.com.au/innovation-patents/ (accessed 30 May 2016). Council of Ministers of Education, Canada nd, Fair Dealing Guidelines, Council of Ministers of Education, Canada, http://www.cmec.ca/docs/copyright/ TEACHERS_PRESENTATION_FDG_EN.pdf (accessed 13 April 2016). Crennan, S. 2007, ‘Obviousness - different paths through Scylla and Charybdis’, Intellectual Property Forum: journal of the Intellectual and Industrial Property Society of Australia and New Zealand, no. 71, pp. 12–20. 712 INTELLECTUAL PROPERTY ARRANGEMENTS CropLife 2015, Consultation on the Advisory Council on Intellectual Property’s (ACIP) recommendation on the innovation patent system, Canberra. Crump, D. 2014, Designing Applications: Tips for filing design applications in Australia, http://www.fpapatents.com/resource?id=352 (accessed 8 September 2016). CSIRO 2006, Public support for science and innovation: a submission to the productivity commission from CSIRO, July, Canberra. Cutler, T. 2008, Venturous australia, building strength in innovation: Review of the National Innovation System, Cutler and Company, Melbourne. DAFF (Department of Agriculture, Fisheries and Forestry) 2010, Australia - European Community Agreement on Trade in Wine, Canberra. Dahlin, K. and Behrens, D. 2005, ‘When is an invention really radical? Defining and measuring technological radicalness’, Research Policy, vol. 34, pp. 717–737. Damsgaard, E. and Thursby, M. 2013, ‘University entrepreneurship and professor privilege’, Industrial and Corporate Change, vol. 22, no. 1, pp. 183–218. Davies Collison Cave 2015, Design infringements and enforcement, http://www.davies.com.au/services/design-infringements-enforcement (accessed 21 August 2016). Davis, M. 1996, Indigenous Peoples and Intellectual Property Rights, Parliamentary Library Research Paper 20 1996-97, Parliament of Australia, Canberra, http://www.aph.gov.au/About_Parliament/Parliamentary_Departments/Parliamentary_ Library/pubs/rp/RP9697/97rp20 (accessed 2 August 2016). Dawes, M. 2010, ‘Setting the Orphans Free’, Australian Law Librarian, vol. 18, no. 4. Deakin University 2015, Deakin University Submission Review of Research Policy and Funding Arrangements for Higher Education, Melbourne. DEDJTR (Victorian Department of Economic Development, Jobs, Transport and Resources) 2015a, Ministerial Statements of Expectations. —— 2015b, Submission to IP Australia consultation on innovation patent system, Melbourne. Dee, P. 2004, The Australia-US Free Trade Agreement: An Assessment, Paper prepared for the Senate Select Committee on the Free Trade Agreement between Australia and the United States of America, Canberra. Deloitte 2014, 2015 life sciences outlook: Southeast Asia, http://www2.deloitte.com/content/ dam/Deloitte/global/Documents/Life-Sciences-Health-Care/gx-lshc-2015-life-sciencesreport-southeast-asia.pdf (accessed 16 September 2016). Deloitte Access Economics 2012, The Costs and Benefits of Preventing Parallel Imports into New Zealand, Canberra. Department of Commerce 2016, DOC Organizational Chart, https://www.commerce.gov/page/doc-organizational-chart (accessed 13 September 2016). REFERENCES 713 Department of Commerce Internet Policy Task Force 2016, White Paper on Remixes, First Sale, and Statutory Damages, White Paper, January, Washington D.C. Department of Education Science and Training 2003, Analysis of the Legal Framework for Patent Ownership in Publicly Funded Research Institutions, Canberra. Department of Finance 2014, Australian Government Cost Recovery Guidelines, Resource Management Guide No. 304, June, Canberra. —— 2015, Considering governance structures, http://www.finance.gov.au/resourcemanagement/governance/policy/flowchart2/ (accessed 29 July 2016). Department of Health 2013, Pharmaceutical Benefits Scheme Venlafaxine and Venlafaxine Derivative Therapeutic Group, http://www.pbs.gov.au/info/publication/factsheets/ venlafaxine-therapeutic-group (accessed 19 September 2016). —— 2015a, Biosimilar Medicines - Factsheet for Patients and Consumers, 21 August, Canberra. —— 2015b, Submission to the Senate Community Affairs References Committee Inquiry into the Availability of New, Innovative and Specialist Cancer Drugs in Australia, March, Canberra. —— 2016a, Pharmaceutical Benefits Scheme Expenditure and Prescriptions twelve months to 30 June 2015, http://www.pbs.gov.au/info/statistics/pbs-expenditureprescriptions-30-june-2015 (accessed 19 September 2016). —— 2016b, The Pharmaceutical Benefits Scheme: Home, http://www.pbs.gov.au/pbs/ home (accessed 20 September 2016). Derclaye, E. and Leistner, M. 2011, Intellectual Property Overlaps: A European Perspective, Oxford and Portland, Oregon, USA. Dernis, H., Dosso, M., Millot, V., Squicciarini, M., Vezzani, A. and Hervas, F. 2015, World Corporate Top R&D Investors: Innovation and IP bundles, JRC and OECD common report, Luxembourg. Deutsche Welle 2015, ‘We can’t protect every sausage,’ says German agriculture minister over TTIP deal, http://www.dw.com/en/we-cant-protect-every-sausage-says-germanagriculture-minister-over-ttip-deal/a-18169728 (accessed 30 March 2016). DFAT (Department of Foreign Affairs and Trade) 2004, Australia-United States Free Trade Agreement - Guide to the Agreement: 17. Intellectual Property Rights, http://dfat.gov.au/about-us/publications/trade-investment/australia-united-states-freetrade-agreement-guide-to-the-agreement/Pages/17-intellectual-property-rights.aspx (accessed 7 April 2016). —— 2015a, Intellectual Property, http://dfat.gov.au/trade/topics/intellectualproperty/pages/intellectual-property.aspx (accessed 1 December 2015). —— 2015b, Trans-Pacific Partnership Agreement, Chapter Summary: Intellectual Property, http://dfat.gov.au/trade/agreements/tpp/summaries/Documents/intellectualproperty.PDF (accessed 15 April 2016). 714 INTELLECTUAL PROPERTY ARRANGEMENTS —— 2016a, About Free Trade Agreements, http://dfat.gov.au/trade/agreements/Pages/about-ftas.aspx (accessed 7 April 2016). —— 2016b, Free Trade Agreements, Free Trade Agreements, http://dfat.gov.au/trade/agreements/Pages/trade-agreements.aspx (accessed 14 April 2016). —— 2016c, FTA negotiations and the public consultations process, http://dfat.gov.au/trade/agreements/Pages/public-consultations.aspx (accessed 31 August 2016). —— 2016d, Treaty making process, http://dfat.gov.au/international-relations/treaties/ treaty-making-process/Pages/treaty-making-process.aspx (accessed 18 April 2016). DIIS (Department of Industry, Innovation and Science) 2014, Australian Pharmaceuticals Industry Data Card, Canberra. —— 2016, Country of Origin Labelling, http://www.industry.gov.au/industry/ IndustrySectors/FoodManufacturingIndustry/Pages/Country-of-Origin-Labelling.aspx (accessed 9 April 2016). DIISRTE (Department of Industry, Innovation, Science, Research and Tertiary Education) 2012, Australian Small Business: Key Statistics and Analysis, Canberra. Dinosaur Designs 2013, Submission to IP Australia on the Review of the Designs System – Issues Paper, Melbourne. DIS (Department of Industry and Science) 2015, Department of Industry and Science: 2014-15 Annual report, Canberra. DITR (Department of Industry, Tourism and Resources Portfolio) 2007, Portfolio budget statements 2006-07, Canberra. Doctors without Borders 2014, R&D Cost Estimates: MSF Response to Tufts CSDD Study on Cost to Develop a New Drug, http://www.doctorswithoutborders.org/article/rd-costestimates-msf-response-tufts-csdd-study-cost-develop-new-drug (accessed 16 April 2016). Dourado, E. and Tabarrok, A. 2014, ‘Public choice perspectives on intellectual property’, Public Choice, vol. 163, no. 1–2, pp. 129–151. Downie, R. 2016, Full Federal Court confirms limitation of trade mark rights with respect to repackaging of goods by third parties, http://www.herbertsmithfreehills.com/ insights/legal-briefings/full-federal-court-confirms-limitation-of-trade-mark-rights (accessed 11 August 2016). DPMC (Department of Prime Minister and Cabinet) 2015, Launch of the National Innovation and Science Agenda. —— 2016, Guidance note: cost-benefit analysis, Canberra. Drahos, P. 2010, ‘Reclaiming the patent social contract’, The Global Governance of Knowledge, Cambridge University Press, pp. 285–317. REFERENCES 715 Du Mont, J. and Janis, M. 2013, ‘Virtual Designs’, Stanford Technical Law Review, vol. 17, no. 107, pp. 107–182. Dukes, G. 1996, ‘Drug regulation and the tradition of secrecy’, International Journal of Risk and Safety in Medicine, vol. 9, no. 3, pp. 143–149. Eagles, I. and Longdin, L. 2003, ‘Competition in information and computer technology markets: intellectual property licensing and section 51(3) of the Trade Practices Act 1974’, QUT Law and Justice Journal, vol. 3, no. 1, pp. 31–50. Easy Access IP 2016, About Easy Access IP, http://easyaccessip.com/?page_id=9 (accessed 31 August 2016). EC (European Commission) 2012, Innovative S&T indicators combining patent data and surveys: Empirical models and policy analyses, Project reference no. 217299, Luxembourg. —— 2013, Fact sheet: Open Access in Horizon 2020, https://ec.europa.eu/programmes/horizon2020/sites/horizon2020/files/FactSheet_Open _Access.pdf (accessed 16 April 2016). —— 2014, Directive on collective management of copyright and related rights and multiterritorial licensing of rights in musical works for online use in the internal market, 26 February, Brussels, Belgium. —— 2015a, 6th Report on the Monitoring of Patent Settlements, 2 December, Luxembourg. —— 2015b, Geographical indications and traditional specialities, http://ec.europa.eu/agriculture/quality/schemes/index_en.htm (accessed 12 January 2016). —— 2015c, Study on the economic impact of utility model legislation in selected Member States, Final Report, April, Luxembourg. —— 2016, Draft of Comprehensive Economic and Trade Agreement (CETA) between Canada, of the one part, and the European Union (and its member states), http://trade.ec.europa.eu/doclib/docs/2016/february/tradoc_154329.pdf (accessed 22 August 2016). EFF (Electronic Frontier Foundation) 2016, The Three-Step Test, https://www.eff.org/document/international-ip-infosheets-three-step-test (accessed 15 April 2016). EIU (Economist Intelligence Unit) 2009, A new ranking of the world’s most innovative countries, London, UK. Eli Lilly Australia 2014, Submission to Parliament of Australia on Australia’s Innovation System, Submission, 28 August, Sydney. EMA (European Medicines Agency) 2016, European Medicines Agency - Human regulatory Clinical data publication, http://www.ema.europa.eu/ema/ 716 INTELLECTUAL PROPERTY ARRANGEMENTS index.jsp?curl=pages/special_topics/general/general_content_000555.jsp&mid=WC0b 01ac05809f363e (accessed 19 July 2016). Engel, P. 2014, ‘Taylor Swift Explains Why She Left Spotify’, Australia, Business Insider Australia, 14 November, http://www.businessinsider.com.au/taylor-swift-explainswhy-she-left-spotify-2014-11 (accessed 12 April 2016). Entsch, W. 2000, Patents Amendment (Innovation Patents) Bill 2000, Second Reading Speech, 29 June, House of Representatives, Canberra. EPO (European Patent Office) 2016a, European Patent Convention, Sixteenth edition, Munich, Germany. —— 2016b, Group B+, https://www.epo.org/news-issues/issues/harmonisation/group-bplus.html (accessed 5 August 2016). Ergas, H. 2002, ‘Intellectual Property Rights and Competition’, presented at Federal Trade Commission/Department of Justice Hearings on Antitrust and Intellectual Property Law and Policy in the Knowledge-Based Economy, Washington D.C., 23 May. EY 2016, Cost benefit analysis of changes to the Copyright Act 1968, April. EUIPO (European Union Intellectual Property Office) 2016, Designs in the European Union, https://euipo.europa.eu/ohimportal/en/designs-in-the-european-union (accessed 9 April 2016). Fabrizi, S., Lippert, S., Norback, P. and Persson, L. 2013, ‘Venture capitalists and the patenting of innovations’, The Journal of Industrial Economics, vol. LXI, no. 3, pp. 623–659. Falkvinge, R. 2016, As 3D printers break through, EU expands copyright to furniture and extends term by a century, https://www.privateinternetaccess.com/blog/2016/08/3dprinters-break-eu-expands-copyright-furniture/ (accessed 21 August 2016). Farrell, J. and Shapiro, C. 2008, ‘How Strong Are Weak Patents?’, American Economic Review, vol. 98, no. 4, pp. 1347–1369. Farre-Mensa, J., Hegde, D. and Ljungqvist, A. 2016, The bright side of patents, USPTO Economic Working Paper, No. 2015-5. FCA (Financial Conduct Authority) 2015, Financial Advice Market Review (FAMR) Expert Advisory Panel – Terms of Reference, http://fca.org.uk/yourfca/documents/famr-expert-advisory-panel-terms-of-reference (accessed 6 April 2016). FCC (Federal Circuit Court of Australia) 2014, About the Federal Circuit Court, http://www.federalcircuitcourt.gov.au/wps/wcm/connect/fccweb/about/about-fcc (accessed 10 April 2016). —— 2015, 2014-15 Federal Circuit Court of Australia Annual Report, 2 September, Commonwealth of Australia, http://www.federalcircuitcourt.gov.au/wps/wcm/connect/ fccweb/reports-and-publications/annual-reports/2014+-+15/2014-15-annual-reportpart3 (accessed 2 August 2016). REFERENCES 717 Federal Court of Australia 2015a, Consultation Draft Practice Note, Intellectual Property National Practice Area, 16 October, Sydney. —— 2015b, Introducing the National Court Framework, http://www.fedcourt.gov.au/lawand-practice/national-court-framework (accessed 19 January 2016). Feinberg, R. 1986, ‘Trademarks, market power and information’, Review of Industrial Organization, vol. 2, no. 176–185. Fels, A. 2001, ‘Competition Policy: Governance Issues – What are the alternative structures? Australia’s Experience’, https://www.accc.gov.au/system/files/ Fels_Canada_20_6_01[1].pdf (accessed 4 April 2016). Fialkoff, M.L. 2014, ‘Pay-For-Delay Settlements in the Wake of Actavis’, Michigan Telecommunications and Technology Law Review, vol. 20, no. 2, pp. 523–546. Fink, C. 2008, Enforcing Intellectual Property Rights: An Economic perspective, July, ICTSD Programme on IPRs & Sustainable Development, International Centre for Trade and Sustainable Development, Switzerland. Fitzsimmons, C. 2015, ‘Knock it off: designers fight back against replica furniture industry’, Sydney, Financial Review, 16 January. fiveIPoffices 2016, About IP5 Cooperation, http://www.fiveipoffices.org/about.html (accessed 29 March 2016). Fixler, D. 2016, Can I use a competitor’s name or trade marks for Google AdWords?, http://www.lexology.com/library/detail.aspx?g=20a58caa-f6d4-4fc5-a7a3f9a9cb40a50e (accessed 15 April 2016). Flanagan, R. 2016, ‘Books are too important to be left to economists’, Sydney, Sydney Morning Herald, 2 June, http://www.smh.com.au/comment/books-are-too-importantto-be-left-to-economists-20160602-gp9ycs.html (accessed 20 September 2016). Flattery-O’Brien, J. and Zammit, M. 2016, Is the term of your US patent critical? If so, you may want to carefully consider how the US patent is prosecuted in order to maximize Patent Term Adjustment. Here’s how…, http://www.lexology.com/library/detail.aspx? g=058cee82-e75a-4ad3-a5b9-0fd043f4fe5d (accessed 16 September 2016). Ford, M. 2016, Changes to Copyright Protection of Industrial Designs, http://ahpatent.com/news/2016/06/20/changes-to-copyright-protection-of-industriallymanufactured-designs/ (accessed 7 July 2016). Fowler, C. 1994, Unnatural Selection: Technology, Politics and Plant Evolution, International Studies in Global Change, Gordon and Breach, Switzerland. FPA Patent Attorneys 2016, Innovation patents, http://www.fpapatents.com/innovationpatents (accessed 30 May 2016). FTC (Federal Trade Commission) 2009, Emerging Health Care Issues: Follow-On Biologic Drug Competition, A Federal Trade Commission Report, Washington D.C. 718 INTELLECTUAL PROPERTY ARRANGEMENTS —— 2016, Agreements Filed with the Federal Trade Commission under the Medicare Prescription Drug, Improvement, and Modernization Act of 2003, Overview, Washington D.C. Future thinking 2015, Bad Faith Trade Mark Applications in China, United Kingdom. Galasso, A. and Schankerman, M. 2015, ‘Patents and Cumulative Innovation: Causal Evidence from the Courts’, The Quarterly Journal of Economics, vol. 130, no. 1, pp. 317–369. Gans, J., King, S. and Lampe, R. 2004, Patent Renewal Fees and Self-Funding Patent Offices, University of Melbourne Legal Studies Research Paper, No. 64, February. GAO (US Government Accountability Office) 1999, Reporting requirements for federally sponsored inventions need revision, August, Washington D.C. —— 2010, Intellectual Property: Observations on Efforts to Quantify the Economic Effects of Counterfeit and Pirated Goods, Report to Congressional Committees, April. —— 2013, Intellectual Property: Assessing factors that affect patent infringement litigation could help improve patent quality, 22 August, GAO-13-465, Washington D.C. Gawrylewski, A. 2007, ‘Yale dumps BioMed Central’, The Scientist. Geist, M. 2015, Why Canada’s high drug prices are about to soar higher, thestar.com, http://www.thestar.com/business/2015/12/21/why-canadas-high-drug-prices-are-aboutto-soar-higher-geist.html (accessed 31 August 2016). Generic Medicines Industry Association and GMIA 2013, Consultation on an objects clause and an exclusion from patentability, Pymble. Giblin, R. 2015, Rethinking Duration: Disaggregating Copyright’s Rewards and Incentives via a System of Rolling Rights, 20 November. Ginarte, J. and Park, W. 1997, ‘Determinants of patent rights: A cross-national study’, Research Policy, no. 26, pp. 283–301. GIPC (Global Intellectual Property Center) 2016, Infinite Possibilities: U.S. Chamber International IP Index, February, Fourth Edition, Washington D.C. Giummo, J. 2014, ‘An examination of the intertemporal returns of patented inventions’, Research Policy, vol. 43, no. 8, pp. 1312–1319. Gleeson, D., Lopert, R. and Reid, P. 2013, ‘How the Trans Pacific Partnership Agreement could undermine PHARMAC and threaten access to affordable medicines and health equity in New Zealand’, Health Policy, vol. 112, no. 3, pp. 227–233. Global Academics 2013, Open letter from international intellectual property scholars and experts supporting Brazil’s proposed patent reform, http://infojustice.org/support-brazil (accessed 14 September 2016). Godden, D. 1998, ‘Growing Plants, Evolving Rights: Plant Variety Rights in Australia’, Agribusiness Review, vol. 6, no. 3. REFERENCES 719 Gonzalez, A. 2006, ‘The software patent debate’, Journal of Intellectual Property Law and Practice, vol. 1, no. 3, pp. 196–206. Gonzalez-Alvarez, N. and Nieto-Antolin, M. 2007, ‘Appropriability of innovation results: an empirical study in Spanish manufacturing firms’, Technovation, vol. 27, pp. 280–295. Google 2016, Google developers: Open source programs https://developers.google.com/open-source/ (accessed 14 April 2016). office, Government of Canada 2015, Trans-Pacific Partnership (TPP): Intellectual Property Chapter, http://www.international.gc.ca/trade-agreements-accords-commerciaux/agracc/tpp-ptp/understanding-comprendre/17-IntellectualProperty.aspx?lang=eng (accessed 7 April 2016). Government of Western Australia 2016, Successful apple breeding – patience, perseverance and luck, https://www.agric.wa.gov.au/pome-fruit/successful-applebreeding-%E2%80%93-patience-perseverance-and-luck (accessed 4 January 2016). Gowers, A. 2006, Gowers Review of Intellectual Property, HM Treasury, London, UK. Grabowski, H. 2009, ‘Data Exclusivity for Biologics: What is the Appropriate Period of Protection’, vol. American Enterprise Institute for Public Policy Research Health Policy Outlook, no. 10, pp. 1–6. von Graevenitz, G., Wagner, S. and Harhoff, D. 2011, ‘How to measure patent thickets - A novel approach’, Economics Letters, vol. 111, no. 1, pp. 6–9. ——, —— and —— 2013, ‘Incidence and growth of patent thickets: The impact of technological opportunities and complexity’, Journal of Industrial Economics, vol. 61, no. 3, pp. 521–563. ——, Ashmead, R. and Greenhalgh, C. 2015, Cluttering and Non-Use of Trade Marks in Europe, No. 2015/48, UK Intellectual Property Office, London, UK. ——, Greenhalgh, C., Helmers, C. and Schautschick, P. 2011, Trade Mark Cluttering: An Exploratory Report, UK Intellectual Property Office, London, UK. Graham, S.J.H. and Harhoff, D. 2014, ‘Separating Patent Wheat from Chaff: Would the U.S. Benefit from Adopting a Patent Post-Grant Review?’, Research Policy, vol. 43, pp. 1649–1659. Gray, R. and Bolek, K. 2012, ‘Grain Research Funding in Australia: Lessons from International Experience’, presented at Australian Agricultural and Resource Economics Society 56th Annual Conference, Fremantle, 7 February. GRDC (Grains Research & Development Corporation) 2008, PBR Agreement, Fact Sheet, March, http://www.grdc.com.au/Resources/Publications/2008/03/Plant-Breeders-Rights (accessed 5 February 2016). —— 2011, Wheat Breeding Information Sheet, http://www.grdc.com.au/uploads/ documents/grdc-fs-wheatbreeding.pdf (accessed 22 December 2015). 720 INTELLECTUAL PROPERTY ARRANGEMENTS Greenberg, G. 2013, ‘Small Firms, Big Patents? Estimating Patent Value Using Data on Israeli Start-Ups’ Financing Rounds’, European Management Review, vol. 10, no. 4, pp. 183–196. Greenhalgh, C., Pitkethly, R. and Rogers, M. 2010, ‘Intellectual property enforcement in smaller UK firms – findings from a survey in 2009-10’, presented at 5th EPIP Conference, Maastricht. ——, Rogers, M., Schautschick, P. and Sena, V. 2011, Trade Mark Incentives, July, Intellectual Property Office, London, UK. —— and Webster, E. 2015, ‘Have trademarks become deceptive?’, The WIPO Journal, vol. 6, no. 2, pp. 109–116. Grigg, A. 2014, Penfolds suffers from China squatters, http://www.afr.com/lifestyle/foodand-wine/wine-and-spirits/penfolds-suffers-from-china-squatters-20140713-j0sws (accessed 3 March 2016). Gross, A. 2014, Singapore’s pharmaceutical industry 2014 update, Pharmaphorum, http://pharmaphorum.com/views-and-analysis/singapores-pharmaceutical-industry2014-update/ (accessed 12 July 2016). Gruen, N., Bruce, I. and Prior, G. 1996, Extending Patent Life: Is it in Australia’s Economic Interests?, Industry Commission Staff Information Paper, Canberra. Guiard, J. 2013, The development of the provisions on essentially derived varieties, http://www.upov.int/meetings/en/doc_details.jsp?meeting_id=29782&doc_id=253272 (accessed 8 September 2016). Haeussler, C., Harhoff, D. and Mueller, E. 2009, To be financed or not - the role of patents for venture capital financing, ZEW Discussion Paper No. 09-003. Hakes, T. 2016, How Google’s Patent Stash is (Probably) Making Them a Ton of Money on Pokémon Go, IP Street Blog, http://www.ipstreet.com/blog/how-googles-patentstash-is-probably-making-them-a-ton-of-money-on-pok%C3%A9mon-go (accessed 7 September 2016). Halfacree, G. 2013, System Shock 2 gets reborn after 14 years, Bitgamer, http://www.bittech.net/news/gaming/2013/02/14/system-shock-2/1 (accessed 12 April 2016). Hall, B. 2009, The use and value of patent rights, Prepared for the UK IP Ministerial Forum on the Economic Value of Intellectual Property. —— 2014, ‘Does patent protection help or hinder technology transfer?’, Intellectual Property for Economic Development, Edward Elgar Publishing, pp. 11–32. ——, Helmers, C. and Graevenitz, G. 2015, Technology entry in the presence of patent thickets, NBER Working Paper No. 21455. ——, Helmers, C., Rogers, M. and Sena, V. 2014, ‘The choice between formal and informal intellectual property: a review’, Journal of Economic Literature, vol. 52, no. 2, pp. 375–423. REFERENCES 721 ——, Jaffe, A. and Trajtenberg, M. 2005, ‘Market value and patent citations’, RAND Journal of Economics, vol. 36, pp. 16–38. —— and Trajtenberg, M. 2004, Uncovering GPTS with patent data, NBER Working Paper, 10901, NBER, Cambridge, MA. Hargreaves, I. 2011, Digital Opportunity: A Review of Intellectual Property and Growth, May, London, UK. Harhoff, D. 2009, Economic Cost-Benefit Analysis of a Unified and Integrated European Patent Litigation System, 29 February, MARKT/2008/06/D, Institute for Innovation Research, Technology Management and Entrepreneurship. —— 2016, ‘Patent quality and examination in Europe’, American Economic Review: Papers and Proceedings 2016, vol. 106, no. 5, pp. 193–197. ——, Scherer, F. and Vopel, K. 2003, ‘Citations, family size, opposition and the value of patent rights’, Research Policy, vol. 32, no. 8, pp. 1343–1363. Harper, I., Anderson, P., McCluskey, S. and O’Bryan, M. 2015, Competition Policy Review: Final Report, March, Canberra. Harrap, C. and Irvine, S. 2015, New appeal decision on computer implemented inventions in Australia, http://www.freehillspatents.com/resource?id=407 (accessed 29 July 2016). Harris, P. 2015, ‘Opening Statement by the Productivity Commission Chairman’, Hearing of Joint Select Committee on Trade and Investment Growth’s inquiry into Business Experience in Utilising Australia’s Free Trade Agreements, Canberra, http://www.pc.gov.au/news-media/speeches/free-trade-agreements (accessed 1 February 2016). Harris, R. 2014, Dairy industry fears European attack on cheese names could hurt exports, http://www.weeklytimesnow.com.au/agribusiness/dairy/dairy-industry-fears-europeanattack-on-cheese-names-could-hurt-exports/newsstory/dfc6b39d18b5f8e8114e44a763e74bb1 (accessed 5 April 2016). Harris, T., Nicol, D. and Gruen, N. 2013, Pharmaceutical Patents Review Report 2013, Canberra. Harrison, M. 2010, Valuing the Future: the Social Discount Rate in Cost-Benefit Analysis, Visiting Researcher Paper, April, Productivity Commission, Canberra. Haruvy, E., Sethi, S. and Zhou, J. 2008, ‘Open Source Development with a Commercial Complementary Product or Service’, Production and Operations Management, vol. 17, no. 1, pp. 29–43. Heald, P. 2008, ‘Property rights and the efficient exploitation of copyrighted works: an empirical analysis of public domain and copyrighted fiction bestsellers’, Minnesota Law Review, vol. 92, pp. 1031–1063. —— 2014, The Demand for Out-of-Print Works and Their (Un)Availability in Alternative Markets, Illinois Public Law and Legal Theory Research Papers Series No. 14-31, 14 March, University of Illinois College of Law, United States. 722 INTELLECTUAL PROPERTY ARRANGEMENTS Heikkila, J. 2014, The relationship between first and second tier patent protection: The case of the Dutch short-term patent system abolition, Jyvaskyla University School of Business and Economics, Finland. Hein, E. 2009, The natural history of the Funky Drummer break, Weblog, The Ethan Hein Blog, http://www.ethanhein.com/wp/2009/the-natural-history-of-the-funky-drummerbreak/ (accessed 14 January 2016). Helland, E. and Seabury, S.A. 2015, ‘Are Settlements in Patent Litigation Collusive? Evidence from Paragraph IV Challenges’, RAND Justice, Infrastructure, and Environment Working Papers, vol. WR-1099. Heller, M. and Eisenberg, R. 1998, ‘Can patents deter innovation? The anticommons in biomedical research’, Science, vol. 280, pp. 698–701. Helmers, C., Lefouili, Y. and McDonagh, L. 2015, Evaluation of the Reforms of the Intellectual Property Enterprise Court 2010-2013, 22 June, London, UK. Herbert Smith Freehills 2015, CCC v Pfizer – some guidance to pharmaceutical companies on navigating competition risk, Press Release, 3 March. Heyward, M. 2016, What We said to the Productivity Commission on 24 June 2016, Blog, https://www.textpublishing.com.au/blog/what-we-said-to-the-productivity-commissionon-24-june-2016 (accessed 20 September 2016). Hilary Charlesworth, Madelaine Chiam, Hovell, D. and Williams, G. 2006, No country is an island: Australia and international law, University of New South Wales Press. Hilmer, P.F., Rayner, M., Taperell, G. and Smith, W. 1993, National Competition Policy Review Report, August. Hipp, C. and Herstatt, C. 2006, Patterns of innovation and protection activities within service companies: Results from a German study on service-intensive companies, Working Paper, 45, Technishe Universitat Hamburg-Harburg, Germany. HM Courts & Tribunals Service 2014a, Guide to the Intellectual Property Enterprise Court Small Claims Track, July, United Kingdom. HM Courts & Tribunals Service 2014b, The Intellectual Property Enterprise Court Guide, April, By authority of the Chancellor of the High Court, United Kingdom. Holdren, J. 2013, Memorandum for the Heads of Executive Departments and Agencies, 22 February, Executive Office of the President, Washington D.C. Hopewell, L. 2015, Netflix Has 10 Times More Aussie Subscribers Than Foxtel, Stan Or Quickflix, Gizmodo, http://www.gizmodo.com.au/2015/06/netflix-has-over-ten-timesmore-aussie-customers-than-foxtel-stan-or-quickflix/ (accessed 2 November 2015). Hospira 2014, Biosimilars Compared to Generic Drugs, http://www.hospira.co.uk/en/ about_hospira/biologics/biosimilars_generic_drugs (accessed 22 February 2016). Hottenrott, H., Hall, B. and Czarnitzki, D. 2016, ‘Patents as quality signals? The implications for financing constraints on R&D’, Economics of Innovation and New Technology, vol. 25, no. 3. REFERENCES 723 House of Representatives Standing Committee on Infrastructure and Communications 2013, At what cost? IT pricing and the Australia tax, July, Commonwealth of Australia, Canberra. Hsu, D. and Ziedonis, R. 2013, ‘Resources as dual sources of advantage: Implications for valuing entrepreneurial-firm patents’, Strategic Management Journal, vol. 34, pp. 761–81. Hsu, P.D., Lander, E.S. and Zhang, F. 2014, ‘Development and Applications of CRISPRCas9 for Genome Engineering’, Cell, vol. 157, no. 6, pp. 1262–1278. Huang, V., Weatherall, K. and Webster, E. 2011, The Use of Survey Evidence in Australian Trade Mark and Passing Off Cases, Working Paper no. 5/11, Intellectual Property Research Institute of Australia, Melbourne. Hunt, R. 2001, ‘You Can Patent That? Are Patents on Computer Programs and Business Methods Good for the New Economy?’, Federal Reserve Bank of Philadelphia Business Review, vol. Q1 2001, pp. 5–15. Hutchinson, J. 2011, Federal Court overturns Samsung Galaxy Tab injunction, http://www.itnews.com.au/news/federal-court-overturns-samsung-galaxy-tabinjunction-281727 (accessed 14 April 2016). IBM (International Business Machine) 2016, Open source development, http://www.ibm.com/developerworks/opensource/ (accessed 14 April 2016). IFPMA (International Federation of Pharmaceutical Manufacturers & Associations) 2007, Data Exclusivity: Encouraging Development of New Medicines, July, Switzerland. IIP (Institute of Intellectual Property) 2014, ‘Desirable fee system and description requirement for claims’, IIP Bulletin, vol. 23. Intellectual Property Office Of Singapore and Ministry Of Law 2016, Review of Singapore’s registered designs regime, https://www.mlaw.gov.sg/content/minlaw/ en/news/press-releases/registered-designs-regime--to-provide-greater-protection-andcla.html (accessed 9 September 2016). IOSCO (International Organization of Securities Commissions) 2010, Objectives and Principles of Securities Regulation, Madrid, Spain. IP Australia 2006, How to Protect Trade Marks in China, https://www.wineaustralia.com/en/Regulatory%20Services/~/media/0000Industry%20S ite/Documents/Production%20and%20Exporting/Market%20Access/IP%20Australia% 20Trade%20Mark%20Protection%20China.ashx (accessed 3 March 2016). —— 2009, Getting the balance right: Toward a stronger and more efficient IP rights system, Consultation Paper, Canberra. —— 2012a, Cost Recovery Impact Statement and Fee Changes: June 2012 - June 2016, Canberra. —— 2012b, Innovation patents - raising the step, Canberra. —— 2013a, Geographical Indications, http://www.ipaustralia.gov.au/get-the-right-ip/trademarks/tools-and-resources/geographica-indications/ (accessed 16 October 2015). 724 INTELLECTUAL PROPERTY ARRANGEMENTS —— 2013b, Intellectual Property Reform in Australia, January, Canberra. —— 2013c, Patentable Subject Matter: Consultation on an Objects Clause and an Exclusion from Patentability, Canberra. —— 2014a, A guide to applying for your trade mark, Canberra. —— 2014b, Commemorative certificates, http://www.ipaustralia.gov.au/manage-yourip/commemorative-certificates/ (accessed 6 January 2016). —— 2014c, Understanding PBR, Plant Breeder’s Rights, https://www.ipaustralia.gov.au/ plant-breeders-rights (accessed 4 February 2016). —— 2015a, About designs, https://www.ipaustralia.gov.au/designs/understandingdesigns/design-basics (accessed 1 September 2015). —— 2015b, Australian Intellectual Property Report 2015, Canberra. —— 2015c, Guide to Patented Pharmaceutical Invention (PPI) Compulsory Licences: Australia’s implementation of the TRIPS Protocol, 25 August, Canberra. —— 2015d, IP Australia Corporate Plan: Covering period 2015-16 to 2018-19, August, Canberra. —— 2015e, Keeping a lid on patent costs, http://www.ipaustralia.gov.au/about-us/newsmedia-and-events/publications/publications-listing/leeping-a-lid-on-patent-costs/ (accessed 9 March 2016). —— 2015f, Other types of trade marks: Series of trade marks, https://www.ipaustralia.gov.au/trade-marks/understanding-trade-marks/types-of-trademarks/other-types-trade-marks#Series of trade marks (accessed 6 January 2016). —— 2015g, Patents for business methods, https://www.ipaustralia.gov.au/patents/aboutpatents/what-can-be-patented/patents-business-methods (accessed 15 February 2016). —— 2015h, Protecting your trade mark, http://www.ipaustralia.gov.au/get-the-rightip/trade-marks/maintain-your-trade-mark/protecting-your-trade-mark/ (accessed 6 January 2015). —— 2015i, Public Consultation: Proposals to streamline IP processes and support small business, Canberra. —— 2015j, Source IP, http://www.ipaustralia.gov.au/understanding-intellectualproperty/commercialise-your-ip/source-ip-site/ (accessed 22 January 2016). —— 2015k, The economic impact of innovation patents, Economic Research Paper 05, IP Australia, Canberra. —— 2015l, TM Check, http://pericles.ipaustralia.gov.au/ols/tmcheck/ (accessed 7 March 2016). —— 2015m, Trade mark costs, https://www.ipaustralia.gov.au/trade-marks/understandingtrade-marks/trade-mark-costs (accessed 6 January 2016). REFERENCES 725 —— 2015n, Trade Mark Price Calculator, https://www.ipaustralia.gov.au/toolsresources/tm-price-calculator (accessed 6 January 2016). —— 2015o, Trade Marks, http://www.ipaustralia.gov.au/get-the-right-ip/trade-marks/ (accessed 10 September 2015). —— 2016a, Australian Intellectual Property Report 2016, Canberra. —— 2016b, Designs Examiners’ Manual, http://manuals.ipaustralia.gov.au/ designs/designs_exam_manual.htm (accessed 6 September 2016). —— 2016c, Designs registration: FAQ, https://www.ipaustralia.gov.au/toolsresources/faqs/design-faqs (accessed 6 September 2016). —— 2016d, Fee Changes – IP Australia Fee Review Update, https://www.ipaustralia.gov.au/about-us/news-and-community/official-notice/feechanges-ip-australia-fee-review-update (accessed 19 August 2016). —— 2016e, Free trade agreement, Text, https://www.ipaustralia.gov.au/about-us/globalengagement/free-trade-agreement (accessed 15 April 2016). —— 2016f, Getting an international trade mark, https://www.ipaustralia.gov.au/trademarks/managing-your-trade-mark/international-trade-marks/getting-an-internationaltrade-mark (accessed 14 September 2016). —— 2016g, Global Engagement: Australia-New Zealand Single economic market, IP Australia, https://www.ipaustralia.gov.au/about-us/global-engagement#Vancouver group (accessed 18 April 2016). —— 2016h, Indigenous knowledge consultation, https://www.ipaustralia.gov.au/aboutus/public-consultations/indigenous-knowledge-consultation (accessed 2 August 2016). —— 2016i, IP Australia Corporate Plan 2016-2020, Canberra. —— 2016j, Patent Examiners Manual, http://manuals.ipaustralia.gov.au/patents/ Patent_Examiners_Manual.htm (accessed 29 July 2016). —— 2016k, Patents for computer-related inventions, https://www.ipaustralia.gov.au/ patents/about-patents/what-can-be-patented/patents-computer-related-inventions (accessed 27 July 2016). —— 2016l, Regulation reform - Cutting red tape, https://www.ipaustralia.gov.au/aboutus/public-consultations/regulation-reform-cutting-red-tape-1 (accessed 15 March 2016). —— 2016m, Streamlining IP processes and supporting small business, https://www.ipaustralia.gov.au/about-us/public-consultations/consulting-proposalsstreamline-ip-processes-and-support-small (accessed 12 September 2016). —— 2016n, Trade Mark Examiners Manual, http://www.ipaustralia.gov.au/pdfs/ trademarkmanual/trade_marks_examiners_manual.htm (accessed 24 February 2016). —— 2016o, Trial aims to speed up patent examination, Media Release, 11 February, https://www.ipaustralia.gov.au/about-us/news-and-community/news/trial-aims-speedpatent-examination (accessed 16 April 2016). 726 INTELLECTUAL PROPERTY ARRANGEMENTS —— 2016p, Types of patents, https://www.ipaustralia.gov.au/patents/understandingpatents/types-patents (accessed 12 September 2016). —— nd, The agency plan, https://www.ipaustralia.gov.au/about-us/accountability-andreporting/freedom-information/information-publication-scheme (accessed 15 September 2016a). —— nd, Submission to the Department of Industry, Innovation and Science’s Functional and Efficiency Review, Unpublished. IPAC (Industrial Property Advisory Committee) 1984, Patents, Innovation and Competition in Australia, 29 August, Canberra. IPCRC (Intellectual Property and Competition Review Committee) 2000, Review of intellectual property legislation under the Competition Principles Agreement, Canberra. IPO (United Kingdom Intellectual Property Office) 2013, A study of patent thickets, Newport, UK. —— 2016a, Online copyright infringement tracker survey (6th wave), 20 June, Newport, UK. —— 2016b, Our governance, https://www.gov.uk/government/organisations/intellectualproperty-office/about/our-governance (accessed 15 April 2016). —— 2016c, Repeal of section 52 of the Copyright, Designs and Patents Act 1988: Guidance for affected individuals, organisations and businesses, United Kingdom. —— 2016d, UK accession to the Hague agreement, Newport, UK. IPRIA (Intellectual Property Research Institute of Australia) 2005, Factors Affecting the Use of Intellectual Property (IP) Protection by Small and Medium Enterprises (SMEs) in Australia - Volume 1, April, Melbourne. Ipsos MediaCT and Oxford Economics 2011, Economic consequences of movie piracy, January, London, UK. IPTA and FICPI (Institute of Patent and Trade Mark Attorneys of Australia and FICPI Australia) 2015, Submission to IP Australia on the Innovation Patent Review, Melbourne. IPTA (Institute of Patent and Trade Mark Attorneys of Australia) 2013, Submission to IP Australia consultation: Patentable subject matter, Melbourne. —— (The Institute of Patent and Trade Mark Attorneys of Australia) 2015, Submission to Public Consultation: Proposal to streamline IP processes and to support small business, 7 April. Jaffe, A. and Rassenfosse, G. 2016, Patent citation data in social science research: overview and best practices, Working Paper No. 21868, NBER, Cambridge, Massachusetts. —— and Trajtenberg, M. 1996, ‘Flow of knowledge from universities and federal laboratories: modelling the flow of patent citations over time and across institutional REFERENCES 727 and geographic boundaries’, Proceedings of the national academy of sciences, vol. 93, pp. 12671–12677. Jaiya, G. 2007, An Overview of National and International Intellectual Property Systems and the Role of WIPO, http://www.wipo.int/edocs/mdocs/sme/en/ wipo_unido_smes_msk_07/wipo_unido_smes_msk_07_www_73614.ppt (accessed 27 November 2015). Janis, M. and Smith, S. 2007, ‘Technological Change And The Design Of Plant Variety Protection Regimes’, Chicago-Kent Law Review, vol. 82, no. 3, pp. 1557–1615. Janke, T. 2012, New tracks: Indigenous knowledge and cultural expression and the Australian intellectual property system, May, Terri Janke and Company Pty Ltd, Rosebery. Jay, T. and Taylor, M. 2013, A case of champagne: a study of geographical indications, Faculty of Law Corporate Governance eJournal Paper 29, Bond University, http://epublications.bond.edu.au/cgej/29 (accessed 19 February 2016). Jefferson, T. 1813, Thomas Jefferson to Isaac McPherson, http://presspubs.uchicago.edu/founders/documents/a1_8_8s12.html (accessed 20 September 2016). Jensen, P., Thomson, R. and Yong, J. 2009, Estimating the patent premium: evidence from the Australian inventor survey, Melbourne Institute of Applied Economic and Social Research and Intellectual Property Research Institute of Australia, Melbourne. —— and Webster, E. 2004, ‘Achieving the Optimal Power of Patent Rights’, The Australian Economic Review, vol. 37, no. 4, pp. 419–426. —— and —— 2006, ‘Firm size and the use of intellectual property rights’, The Economic Record, vol. 82, no. 256, pp. 44–55. —— and —— 2009, ‘Knowledge management: does capture impede creation?’, Industrial and Corporate Change, vol. 18, no. 4, pp. 701–727. —— and —— 2014, ‘Patents, transaction costs and academic research project choice’, Economic Record, vol. 90, no. 289, pp. 179–196. Johnson, M., Mitra-Kahn, B., Bialowas, A., Man, B., Nicholson, P. and Bakhtiari, S. 2015, The economic impact of innovation patents, Economic Research Paper 05, IP Australia, Canberra. Johnson, S. 2015, The Creative Apocalypse that Wasn’t, The New York Times, http://www.nytimes.com/2015/08/23/magazine/the-creative-apocalypse-thatwasnt.html?_r=1 (accessed 23 October 2015). Joint Standing Committee on Treaties 2004, The Australia-United States Free Trade Agreement, 23 June, Report No. 61, Canberra. Jones, J. and Lucas, M. 2015, Pharmaceutical Patent Term Extension and Data Exclusivity in Singapore - Davies Collison Cave Australia, http://www.davies.com.au/ipnews/pharmaceutical-patent-term-extension-and-data-exclusivity-in-singapore (accessed 12 July 2016). 728 INTELLECTUAL PROPERTY ARRANGEMENTS Jordan, R. 2014, China bottle tech checks smash fake wine market, https://au.news.yahoo.com/thewest/wa/a/22829109/china-bottle-tech-checks-smashfake-wine-market/ (accessed 3 March 2016). JPO (Japan Patent Office) 2012a, Formulation and Implementation of National IP Strategy in Japan. —— 2012b, Utility Models in Japan, Tokyo. —— 2016, Patent Prosecution Highway Portal Site, https://www.jpo.go.jp/ppphportal/statistics.htm (accessed 15 July 2016). Kain, J., Kurruppu, I. and Billing, R. 2003, Australia’s National Competition Policy: Its Evolution and Operation. Kalanje, C. 2005, Role of Intellectual Property in Innovation and New Product Development, http://www.wipo.int/export/sites/www/sme/en/documents/pdf/ip_innovation_developm ent.pdf (accessed 23 June 2016). Kani, M. and Motohashi, K. 2012, ‘Understanding the Technology Market for Patents: New Insights from a Licensing Survey of Japanese Firms’, Research Policy, vol. 41, no. 1, pp. 226–235. Kaufman, G. 2013, Mac Miller settles $10 million Lord Finesse lawsuit, MTV News, http://www.mtv.com/news/1700290/mac-miller-lord-finesse-lawsuit/ (accessed 12 April 2016). Kelly, P. 2015, ‘IP Australia - what role in value-adding services’, 2015 LESANZ conference, Melbourne. Kiat, S.Y., Kim, T.N., Zhi, H.K. and Tricia, B.Y.D. 2010, Intellectual Property of Integrated Circuits, J.Ross Publishing, Plantation, Florida. Kim, J.H. 2011, Letter to the Honorable Ron Kirk, United States Trade Representative, 20 June, Ministry of Foreign Affairs and Trade, Seoul, Korea. Kim, Y.K., Lee, K. and Choo, K. 2010, ‘Intellectual property rights and innovation in economic development in Korea’, The economics of intellectual property in the republic of Korea, WIPO. ——, ——, Park, W. and Choo, K. 2012, ‘Appropriate intellectual property protection and economic growth in countries at different levels of development’, Research Policy, vol. 41, pp. 358–375. Kingston, W. 2007, ‘Repairing Incentives to Invest in Plant Breeding’, Intellectual Property Quarterly, vol. 10, no. 3, pp. 294–311. Kingwell, R. 2003, Institutional change and plant variety provision in Australia, Conference paper, 47th Annual Conference Of the Australian Agricultural and Resource Economics Society, Fremantle. REFERENCES 729 Kinnaird, J. 2015, Extending patent term in Southeast Asia, Media release, Marks & Clerk, http://www.marks-clerk.com/Home/Knowledge-News/Articles/Extending-PatentTerm-in-Southeast-Asia.aspx (accessed 16 April 2016). KIPO (Korean Intellectual Property Office) 2015, Registrations, http://www.kipo.go.kr/upload/en/download/Registrations.xls (accessed 19 April 2016). Kittikhoun, A. 2015, The art of concealment: use of a competitor’s trade marks in metatags held not to amount to copyright or trade mark infringement in Canada, IP Whiteboard, http://ipwhiteboard.com.au/the-art-of-concealment-use-of-a-competitorstrade-marks-in-metatags-held-not-to-amount-to-copyright-or-trade-mark-infringementin-canada/ (accessed 9 March 2016). Klein, J. 2002, ‘Band overcame an avalanche of hurdles in sampling songs’, Chicago, Chicago Tribune, 8 August, http://articles.chicagotribune.com/2002-0808/features/0208080123_1_samples-avalanches-clearing (accessed 12 April 2016). Kock, M.A., Porzigq, S. and Willnegger, E. 2006, ‘The Legal Protection of PlantBiotechnological Inventions and Plant Varieties in Light of the EC Biopatent Directive’, International Review of Intellectual Property and Competition Law, vol. 37, no. 2, pp. 135–156. Kou, Z., Rey, P. and Wang, T. 2013, ‘Non-obviousness and screening’, The Journal of Industrial Economics, vol. LXI, no. 3. KPMG 2014, Regulatory costing project, Final Report, 23 July, Sydney. Krill, P. 2008, Microsoft exec touts mixed source ventures, InfoWorld, http://www.infoworld.com/article/2653682/operating-systems/microsoft-exec-toutsmixed-source-ventures.html (accessed 21 July 2016). Lam, H. and Chatterton, E. 2013, China amends trademark law: practical implications for brand owners, DLA Piper, https://www.dlapiper.com/en/australia/insights/ publications/2013/12/intellectual-property-and-technology-news-united__/chinaamends-trademark-law-practical-implication__/ (accessed 2 September 2016). Lamine, A. and Goran, S. 2013, ‘Motivations for Open Source Project Entrance and Continued Participation’, Informatics Masters thesis, Lund University, Sweden. Landes, W. and Posner, R. 2002, Indefinitely Renewable Copyright, John M Olin Law & Economics Working Paper No 154, University of Chicago, Chicago. —— and —— 2003, The Economic Structure of Intellectual Property Law, Harvard University Press, Cambridge, Massachusetts. Lanjouw, J. 1993, Patent Protection: Of What Value and For How Long?, NBER Working Paper, 4475, Cambridge, MA. —— 1998, ‘Patent protection in the shadow of infringement simulation estimations of patent value’, Review of Economic Studies, vol. 65, no. 4, pp. 671–710. —— and Schankerman, M. 1999, The quality of ideas: measuring innovation with multiple indicators, NBER Working Paper, Cambridge MA. 730 INTELLECTUAL PROPERTY ARRANGEMENTS —— and —— 2004, ‘Patent quality and research productivity: measuring innovation with multiple indicators’, The Economic Journal, vol. 114, no. 495, pp. 441–465. Lateral Economics 2012, Exception industries: the economic contribution to Australia of industries relying on limitations and exceptions to copyright. Law Council of Australia 1998, Submission 33 to the National Competition Committee Review of Sections 51(2) and 51(3) of the Trade Practices Act 1974, Canberra. —— 2014, Submission to the Competition Policy Review on the Parallel Imports of Trade Marked Goods, 19 June, Canberra. Law Institute of Victoria 2015, Consultation on the Advisory Council on Intellectual Property’s (ACIP) recommendation on the innovation patent system, Melbourne. Lawrence, A., Houghton, J., Thomas, J. and Weldon, P. 2014, Where is the evidence: realising the value of grey literature for public policy and practice, Swinburne Institute for Social Research, Melbourne. Lawson, C. 2005, ‘Patent Privileges and the National Competition Policy - Patent Scope and Allocation?’, Australian Business Law Review, vol. 33, no. 1, pp. 7–29. —— 2008a, ‘Managerialist influences on granting patents in Australia’, Australian Journal of Administrative Law, vol. 15, no. 2, pp. 70–99. —— 2008b, ‘Quantum of Obviousness in Australian Patent Laws’, vol. 19, pp. 43–65. —— 2013a, ‘How are pharmaceutical patent term extensions justified? Australia’s evolving scheme’, Journal of Law and Medicine, vol. 21, pp. 379–398. —— 2013b, ‘The Evolution of a Workable Scheme for End Point Royalties for Plant Varieties’, The Intellectual Property Society of Australia and New Zealand. Inc., no. 94. —— 2014, ‘Plant Breeder’s Rights and Essentially Derived Varieties: Still Searching for Workable Solutions’, European Intellectual Property Review, vol. 32, no. 8, pp. 499– 517. LECG (Law and Economics Consulting Group) 2007, MED Parallel Importing Review: Impact upon Creative Industries, November, Wellington, New Zealand, http://www.mbie.govt.nz/info-services/business/intellectual-property/parallel-importingin-new-zealand/copyright-parallel-importing-of-films/documents-and-images/lecgreport-impact-parallel-importing-nov07.pdf (accessed 15 September 2016). Leibowitz, J. 2009, ‘Pay-for-Delay’ Settlements in the Pharmaceutical Industry: How Congress Can Stop Anticompetitive Conduct, Protect Consumers’ Wallets, and Help Pay for Health Care Reform (The $35 Billion Solution), 23 June, Federal Trade Commission. Lemley, M. 2013, ‘Software patents and the return of functional claiming’, Wisconsin Law Review, vol. 2013, no. 4, pp. 905–964. —— and Shapiro, C. 2007, ‘Patent holdup and royalty stacking’, Texas Law Review, vol. 85, pp. 1991–2049. REFERENCES 731 Lennon, L., Eade, L. and Smyth, A. 2013, ‘3D printing - Design revolution or intellectual property nightmare?’, Law Society Journal, pp. 60–63. Lerner, J. 2000, 150 Years of Patent Office Practice, Working Paper, January, 7477, National Bureau of Economic Research, http://www.nber.org/papers/w7477 (accessed 7 April 2016). Letzing, J. 2009, Microsoft defends right to brandish open-source patents, MarketWatch, http://www.marketwatch.com/story/microsoft-defends-nuanced-open-source-approach (accessed 27 July 2016). Levin, R., Klevorick, A., Nelson, R., Winter, S., Gilbert, R. and Griliches, Z. 1987, ‘Appropriating the returns from industrial research and development’, Brookings Papers on Economic Activity, vol. 1987, no. 3, pp. 783–831. Levy-Carciente, S. 2015, 2015 International Property Rights Index, Property Rights Alliance, Washington D.C. Lieberman, M. and Montgomery, D. 1988, ‘First-Mover Advantages’, Strategic Management Journal, vol. 9, pp. 41–58. Lim, K., O’Sullivan, D. and Falk, M. 2014, The market for design: insights from interviews with Australian firms, IP Australia Economic Research Paper 03, IP Australia, Canberra. Lindgren, K. 2015a, Report of Review of Copyright Collecting Societies’ Compliance with their Code of Conduct for the Year 1 July 2014 to 30 June 2015, 23 December, Sydney. —— 2015b, Supplementary Report of the Code Reviewer (The Hon K E Lindgren AM, QC, Formerly a Justice of the Federal Court of Australia) upon a Review of the Operation of the Code of Conduct of the Copyright Collecting Societies of Australia, 28 October, Sydney. Lindner, R. 2004, ‘Economic Issues for Plant Breeding - Public Funding and Private Ownership’, Australasian Agribusiness Review, vol. 1, no. 2. Linear77 2012, Major steps in the IC design flow, https://en.wikipedia.org/wiki/Integrated_circuit_design#/media/File:Integrated_circuit_ design.png (accessed 19 September 2016). Lomas, M. 2016, UK copyright law repealing section http://www.lexology.com/library/detail.aspx?g=feac51ab-5381-463a-b939e72926b273e8 (accessed 6 September 2016). 52, López, A. 2009, ‘Innovation and Appropriability, Empirical Evidence and Research Agenda’, in World Intellectual Property Organisation (ed), The Economics of Intellectual Property: Suggestions for Further Research in Developing Countries and Countries with Economies in Transition, Geneva, Switzerland, pp. 1–32. Love, B. 2013, ‘An Empirical Study of Patent Litigation: Could a Patent Term Reduction Decimate Trolls Without Harming Innovators?’, University of Pennsylvania Law Review, vol. 161, pp. 1309–1359. 732 INTELLECTUAL PROPERTY ARRANGEMENTS Lybbert, T. and Zolas, N. 2012, Getting patents an economic data to speak to each other: An ‘algorithmic links with probabilities’ approach for joint analyses of patenting and economic activity, Working Paper no. 5, October, World Intellectual Property Organization. ——, —— and Bhattacharyya, P. 2014, An ‘Algorithmic Links with Probabilities’ Concordance for Trademarks For Disaggregated Analysis of Trademark and Economic Data, Working paper no. 14, January, World Intellectual Property Organization, Geneva, Switzerland. Ma, H., Arievich, E. and Karlhuber, M. 2016, Adjudicating Intellectual Property Disputes: An ICC report on specialised IP Jurisdictions worldwide, International Chamber of Commerce. MacMillan, I.., Siegel, R. and Subbanarasimha, P.. 1985, ‘Criteria used by venture capitalists to evaluate new venture proposals’, Journal of Business Venturing, vol. 1, no. 1, pp. 119–128. Mancini, L. 2016, Can Website Meta Data Infringe Registered Trade Marks, http://www.cullens.com.au/trade-marks/can-website-meta-data-infringe-registeredtrade-marks/ (accessed 9 March 2016). Manheim, B.S., Granahan, P. and Dow, K.J. 2006, ‘“Follow-On Biologics”: Ensuring Continued Innovation In the Biotechnology Industry’, Health Affairs, vol. 25, No. 2, pp. 394–404. Mann, R. 2005, ‘Do Patents Facilitate Financing in the Software Industry?’, Texas Law Review, vol. 83, no. 4, pp. 961–1030. —— and Sager, T. 2007, ‘Patents, venture capital, and software startups’, Research Policy, vol. 36, no. 2, pp. 193–208. Mansfield, E. 1986, ‘Patents and innovation: an empirical study’, Management Science, vol. 32, no. 2, pp. 173–181. ——, Schwartz, M. and Wagner, S. 1981, ‘Imitation costs and patents: an empirical study’, The Economic Journal, vol. 91, pp. 907–918. Maskin, E. 2012, Patents on Software: A Nobel Laureate’s View, http://www.nytimes.com/2012/10/15/opinion/patents-on-software-a-nobel-laureatesview.html (accessed 16 February 2016). Maskus, K. 1993, ‘Trade-related intellectual property rights’, European Economy, vol. 52, pp. 157–184. —— 2000, Intellectual Property Rights in the Global Economy, Peter G. Peterson Institute for International Economics, Washington D.C. —— 2012, Private rights and public problems: The global economics of intellectual property in the 21st century, Peter G. Peterson Institute for International Economics, Washington D.C. REFERENCES 733 —— and McDaniel, C. 1999, ‘Impacts of the Japanese patent system on productivity growth’, Japan and the World Economy, vol. 11, pp. 557–574. Maxmen, A. 2015, The Genesis Engine, WIRED, http://www.wired.com/2015/07/crisprdna-editing-2/ (accessed 8 September 2016). Mazzone, J. 2012, Copyfraud and Other Abuses of Intellectual Property Law, Vimeo, https://vimeo.com/40175400 (accessed 12 April 2016). MBIE (New Zealand Ministry of Business, Innovation & Employment) 2016, Implementation of the Trans-Pacific Partnership Intellectual Property Chapter, Discussion Paper, Wellington, New Zealand. McEneiry, B. 2012, ‘An Empirical study of business method patent applications filed in Australia 2000-2009’, Intellectual Property Forum, no. 89, pp. 1–21. McGrath, D. 2014, ‘The Australian grain industry end point royalty (EPR) collection system – a collective industry approach’, http://acipa.edu.au/abstracts/201409-endpoint-royalties.html (accessed 4 March 2016). McHugh, S. 2011, Herman Miller v Matt Blatt, King & Wood Mallesons. McKeon, S., Alexander, E., Brodaty, H., Ferris, B., Frazer, I. and Little, M. 2013, Strategic Review of Health and Medical Research – Better Health Through Research, Final Report, February, Canberra. McKinsey & Company 2013, McKinsey on Semiconductor, No. 3, Autumn 2013, New York. McLeod, K. and DiCola, P. 2011, Creative License: The Law and Culture of Digital Sampling, Duke University Press, Durham, North Carolina. MED (Ministry of Economic Development) 2008, Parallel Importing, Cabinet Paper, Wellington, New Zealand. Medicines Australia 2013, Submission to the Pharmaceutical Patents Review, 23 January, Canberra. —— 2015, Submission to IP Australia review into Innovation Patent System, Canberra. Menand, L. 2014, ‘Crooner in Rights Spat - Are copyright laws too strict?’, The New Yorker, October 20, 2014, http://www.newyorker.com/magazine/2014/10/20/croonerrights-spat (accessed 12 April 2016). Mendis, D. and Secchi, D. 2015, Legal and Empirical Study of 3D Printing Online Platforms and an Analysis of User Behaviour, The Intellectual Property Office, Bournemouth University, UK. Mendonca, S., Pereira, T. and Godhino, M. 2004, Trademarks as an Indicator of Innovation and Industrial Change, Working Paper 2004/15, Laboratory of Economics and Management, Sant’Anna School of Advanced Studies, Pisa, Italy. Meyer, L. 2006, ‘Soft law for solid contracts? A comparative analysis of the value of the UNIDROIT principles of international commercial contracts and the principles of 734 INTELLECTUAL PROPERTY ARRANGEMENTS European contract law to the process of contract law harmonization’, Denver Journal of International Law and Policy, vol. 119, no. 123. Moir, H. 2013, ‘Fabricating invention: the patent malfunction of Australian patent law’, Agenda, vol. 20, no. 2, pp. 21–38. —— 2015, European Trade Treaties: Key Intellectual Property Demands, ANU Centre for European Studies Briefing Paper Vol. 6 No. 4, Australian National University, Canberra. Molloy, S. 2016, Territorial copyright for books in Australia: a real-world analysis informed by the theory of impure public goods, http://www.copyright.org.au/acc_prod/ AsiCommon/Controls/BSA/Downloader.aspx?iDocumentStorageKey=5f022aba-a4bb49d6-aa80-ea1fa34a9d80&iFileTypeCode=PDF&iFileName=Territorial%20copyright %20for%20books%20in%20Australia (accessed 30 May 2016). Morgan, S., Grootendorst, P., Lexchin, J., Cunningham, C. and Greyson, D. 2011, ‘The cost of drug development: A systematic review’, Health Policy, vol. 100, no. 1, pp. 4–17. Moritz, S. and Christie, F. 2006, ‘Second-tier patent systems: the Australian experience’, European Intellectual Property Review, no. 4, pp. 645–655. Moser, P. 2005, ‘How do patent laws influence innovation? Evidence from nineteenthcentury world fairs’, American Economic Review, vol. 95, no. 4, pp. 1215–1236. ——, Ohmstedt, J. and Rhode, P. 2014, Patent citations and the size of patented inventions: Evidence from hybrid corn, Cambridge MA. Moultrie, J. 2011, Design Economics, Chapter three: design right case studies, Intellectual Property Office, Cardiff, UK. MPI (Max Planck Institute for Intellectual Property and Competition Law) 2013, Principles for intellectual property provisions in bilateral and regional agreements, Germany. Murdoch, N. 2016, Chinese ruling threatens trademark chaos, http://www.lawyersweekly.com.au/opinion/17965-chinse-ruling-threatens-trade-markchaos (accessed 19 February 2016). Murphy, D. 2007, Plant Breeding and Biotechnology: Societal Context and the Future of Agriculture, Cambridge University Press. Nagaoka, S. 2005, Patent quality, cumulative innovation and market value: Evidence from Japanese firm level panel data, IIR Working Paper 05-06, Institute of Innovation Research, Hitotsubashi University. Nair, M. 2012, Canada and Israel: Cultivating Fairness of Use, 1 February, PIJIP Research Paper Series, American University Washington College of Law. National Film & Sound Archive 2010, Statement on Orphan Works, June, Canberra. National Library of Australia 2016, The Australian Women’s Weekly - Trove, http://trove.nla.gov.au/aww (accessed 8 September 2016). REFERENCES 735 NCAFP (National Congress of Australia’s First Peoples) 2013, Position Paper on a National Indigenous Cultural Authority, Sydney. NCC (National Competition Council) 1999, Review of sections 51(2) and 51(3) of the Trade Practices Act 1974 - Final Report, March, Melbourne. NECG (Network Economics Consulting Group) 2004, The Impact of Parallel Imports on New Zealand’s Creative Industries, October, New Zealand. New Zealand Herald 2013, ‘Hachette to close NZ publishing unit’, 26 July, http://www.nzherald.co.nz/business/news/article.cfm?c_id=3&objectid=10903768 (accessed 13 September 2016). Newman, J. 2011, The Lost History of System Shock, G4TV.com, http://www.g4tv.com/thefeed/blog/post/713030/the-lost-history-of-system-shock/ (accessed 12 April 2016). —— 2013, ‘Thirteen Years Later, System Shock 2 Lives Again’, Time, http://techland.time.com/2013/02/13/13-years-later-system-shock-2-lives-again/ (accessed 12 April 2016). NFF (National Farmers’ Federation) 2016, Submission to the Department of Foreign Affairs and Trade Inquiry into the Free Trade Agreement Negotiations between Australia and the EU, Submission to DFAT on the Australia-European Union Free Trade Agreement, February, Canberra. NGIA (Nursery & Garden Industry Australia) 2009, Ornamental Plant Breeding in Australia, April, 3, Nursery Papers Technical, Sydney. Nielsen, J. and Nicol, D. 2008, ‘Wither Patent Use Without Authorisation in Australia’, Federal Law Review, vol. 36, no. 3, p. 331. Nordhaus, W. 1969, Invention growth and welfare: a theoretical treatment of technological change, MIT Press, Cambridge MA. Nowak, D. and Doucas, M. 2015, Pharmaceuticals, Patents, Data protection and the TPP, Media release, Henry Hughes, http://www.henryhughes.co.nz/Site/News_Articles _Case_Notes/Pharmaceuticals-Patents-Data-protection-and-the-.aspx (accessed 16 April 2016). Nufarm 2015, Submission responding to IP Australia’s consultation paper on ACIP’s recommendation on the innovation patent system, Melbourne. NZ Parliament 2016, Patents (Trans-Tasman Patent Attorneys and Other Matters) Amendment Bill, http://www.parliament.nz/en-nz/pb/legislation/bills/00DBHOH_ BILL66518_1/patents-trans-tasman-patent-attorneys-and-other-matters (accessed 18 April 2016). NZIER (New Zealand Institute of Economic Research) 2000, The Economic Effects of Lifting the Ban in NZ: A Report to the Intellectual Property and Competition Review Committee, Wellington, New Zealand. 736 INTELLECTUAL PROPERTY ARRANGEMENTS NZMFAT (New Zealand Ministry of Foreign Affairs & Trade) 2015, Trans-Pacific Intellectual Property Fact Sheet, Wellington, New Zealand. NZPC (New Zealand Productivity Commission) 2014, Regulatory institutions and practices, Final Report, Wellington, New Zealand. Ockwell, A.P. 1982, Plant Variety Rights - A Review of Issues, Occasional Paper No. 65, Bureau of Agricultural Economics, Canberra. O’Connor, B. 2014, Geographical Indications in CETA, the Comprehensive Economic and Trade Agreement between Canada and the EU, http://www.origingi.com/images/stories/PDFs/English/14.11.24_GIs_in_the_CETA_English_copy.pdf (accessed 30 March 2016). OECD (Organisation for Economic Co-operation and Development) 1997, Competition Policy and Intellectual Property Rights, Policy Roundtable, Paris. —— 2009a, OECD Patent Statistics Manual, Paris. —— 2009b, OECD Science, Technology and Industry Scoreboard 2009, Paris. —— 2011a, How’s Life? Measuring well-being, Paris. —— 2011b, OECD Science, Technology and Industry Scoreboard 2011, Paris. —— 2012, OECD Patent Database, February, Paris. —— 2013, International Regulatory Co-operation: Addressing Global Challenges, Paris. —— 2014, The Governance of Regulators: OECD Best Practice Principles for Regulatory Policy, OECD Publishing, Paris. —— 2015a, Enquiries into Intellectual Property’s Economic Impact, OECD Publishing, Paris. —— 2015b, OECD Science, Technology and Industry Scoreboard 2015: Innovation for growth and society, OECD Publishing, Paris. —— 2015c, The Innovation Imperative: Contributing to Productivity, Growth and WellBeing, OECD Publishing, Paris. —— 2016a, OECD Patent Quality Database, http://www.oecd.org/sti/inno/intellectualproperty-statistics-and-analysis.htm (accessed 16 September 2016). —— 2016b, Being an independent regulator, 11 April, OECD, Paris. —— 2016c, Glossary of Statistical Terms, https://stats.oecd.org/glossary/index.htm (accessed 21 March 2016). OECD Regulatory Policy Committee 2012, Recommendation of the Council of the OECD on Regulatory Policy and Governance, http://www.oecd.org/gov/regulatorypolicy/recommendations-guidelines.htm (accessed 12 January 2016). OECD & EUIPO (Organisation for Economic Co-operation and Development and European Union Intellectual Property Office) 2016, Trade in Counterfeit and Pirated Goods: Mapping the Economic Impact, Paris. REFERENCES 737 OHE (Office of Health Economics) 2012, The R&D Cost of a New Medicine, December, London, UK. OHIM and EPO (Office for Harmonization in the Internal Market and European Patent Office) 2013, Intellectual property rights intensive industries: contribution to economic performance and employment in the European Union, September, Melbourne. Ollila, E. and Hemminki, E. 1996, ‘Secrecy in drug regulation: Licensing documentation on the Norplant® contraceptive’, International Journal of Risk and Safety in Medicine, vol. 9, no. 3, pp. 161–172. O’Shea, A. 1995, Protecting Intellectual Property in an Emerging Digital Environment, Cook, Canberra. OSIA (Open Source Industry Australia Ltd) 2013, Submission to the Australian Council on Intellectual Property Review of the Innovation patent system, October, Melbourne. Pakes, A. 1986, ‘Patents as options: some estimates of the value of holding European patent stocks’, Econometrica, vol. 54, pp. 755–784. Palangkaraya, A., Webster, E. and Jensen, P. 2011, ‘Misclassification between patent offices: evidence from a matched sample of patent applications’, The Review of Economics and Statistics, vol. 93, no. 3, pp. 1063–1075. Pallante, M.A. 2013, ‘The Next Great Copyright Act’, The Columbia Journal of Law and the Arts, vol. 36, no. 3, pp. 315–344. Papageorgiadis, N., Cross, A. and Alexiou, C. 2015, Index of patent systems strength 201415, http://www.sciencedirect.com/science/article/pii/S1090951613000990 (accessed 1 April 2016). Papandreou, A. 1956, ‘The economic effect of trademarks’, California Law Review, vol. 44, no. 3, pp. 503–510. Park, W. 2008, ‘International patent protection: 1960-2005’, Research Policy, no. 37, pp. 761–766. —— 2010a, International patent protection, http://fs2.american.edu/wgp/www/ (accessed 16 September 2016). —— 2010b, ‘On patenting costs’, WIPO Journal, vol. 2, no. 1. —— and Lippoldt, D. 2005, International Licensing and the Strengthening of Intellectual Property Rights in Developing Countries During the 1990s, OECD Economic Studies No. 40, 2005/1, OECD, pp. 7–48. Parliament of the Commonwealth of Australia 2016, A history of the Joint Standing Committee on Treaties: 20 years, Commonwealth of Australia. Patentec Patent Attorneys 2016, Innovation patent, http://www.patentec.com.au/ patents/patent-applications/innovation-patent/ (accessed 30 May 2016). PC (Productivity Commission) 1996, Stocktake of Progress in Microeconomic Reform, Research Report, Canberra. 738 INTELLECTUAL PROPERTY ARRANGEMENTS —— 2001, Review of the National Access Regime, Report no. 17, Canberra. —— 2003, Evaluation of the Pharmaceutical Industry Investment Program, Research report, Canberra. —— 2007a, Productivity Perspectives 2007, Conference Proceedings, Canberra. —— 2007b, Public Support for Science & Innovation, Research report, Canberra. —— 2009, Restrictions on the Parallel Importation of Books into Australia, Research report, Canberra. —— 2010, Bilateral and Regional Trade Agreements, Research report, Canberra. —— 2013a, Compulsory licensing of patents, Inquiry Report no. 61, Canberra. —— 2013b, Major Project Development Assessment Processes, Research report, Canberra. —— 2014, Access to Justice Arrangements, Inquiry Report no. 72, Canberra. —— 2015a, Mutual Recognition Schemes, Research Report, Canberra. —— 2015b, Trade and Assistance Review 2013-14, Annual Report Series, Canberra. Pearce, A., van Gool, K., Haywood, P. and Haas, M. 2012, ‘Delays in access to affordable medicines: putting policy into perspective’, Australian Health Review, vol. 36, no. 4, pp. 412–418. Pfizer Australia 2015, Consultation on the Advisory Council on Intellectual Property’s (ACIP) recommendation on the innovation patent system, Sydney. Pharmaceutical Research and Manufacturers of America (PhRMA) 2013, Pharmaceutical Research and Manufacturers of America (PhRMA) Special 301 Submission 2013, Washington D.C. Phillips McDougall 2011, Cost of Bringing a Biotech Crop to Market, CropLife International, https://croplife.org/plant-biotechnology/regulatory-2/cost-of-bringing-abiotech-crop-to-market/ (accessed 9 September 2016). PhoneArena 2014, Flagship life cycle: Would a 6-month time frame prove too short?, PhoneArena, http://www.phonearena.com/news/Flagship-life-cycle-Would-a-6-monthtime-frame-prove-too-short_id56988 (accessed 30 March 2016). Picken, D. 2015, ‘The book strikes back’, New Zealand, Bay of Plenty Times, 16 November, http://www.nzherald.co.nz/bay-of-plenty-times/news/article.cfm?c_id= 1503343&objectid=11546174 (accessed 20 September 2016). Plant Health Australia 2016, Production nurseries, http://www.planthealthaustralia.com.au/ industries/production-nurseries/ (accessed 13 September 2016). Pluvia Zuniga, M. and Guellec, D. 2009, Who Licenses out Patents and Why? Lessons from a business survey., 31 March, 2009/05, OECD Science, Technology and Industry Working Papers, OECD Publishing. REFERENCES 739 PMPRB (Patented Medicine Prices Review Board of Canada) 2015, Annual Report 2014, 16 November. —— 2016, Patentee’s Guide to Reporting - Updated July 2015, http://www.pmprbcepmb.gc.ca/view.asp?ccid=523 (accessed 15 February 2016). Pollock, R. 2007, Forever Minus a Day? Some Theory and Empirics of Optimal Copyright, 24 September, MPRA Paper No. 5024, Munich Personal Research Papers in Economics Archive, Munich. Port, J. 2010, Chinese fake it with counterfeits of Australian wines, http://www.executivestyle.com.au/chinese-fake-it-with-counterfeits-of-australianwines-13im7 (accessed 3 March 2016). Posner, R. 2012, Do patent and copyright law restrict competition and creativity excessively?, The Becker-Posner Blog, http://www.becker-posnerblog.com/2012/09/do-patent-and-copyright-law-restrict-competition-and-creativityexcessively-posner.html (accessed 23 August 2016). van Pottelsberghe, B. 2010, The quality factor in patent systems, Bruegel Working Paper, 2010/03. —— and Francois, D. 2009, ‘The cost factor in patent systems’, Journal of Industry, Competition and Trade, vol. 9, pp. 329–355. —— and van Zeebroeck, N. 2008, ‘A brief history of space and time: the scope-year index as a patent value indicator based on families and renewals’, Scientometrics, vol. 75, no. 2, pp. 319–338. PPCA (Phonographic Performance Company of Australia) 2014, Phonographic Performance Company of Australia Annual Report 2014, Annual Report, NSW, http://www.ppca.com.au/IgnitionSuite/uploads/docs/2014%20PPCA%20Annual%20Re port.pdf (accessed 13 April 2016). Pratt, R. and Berg, A. 2014, Governance of Securities Regulators: A Framework, Policy Research Working Paper 6800, The World Bank, Washington D.C. Prescott, C. and Christie, A. 2015, End Point Royalty Collection in the Nursery & Garden Industry, October, Plant Breeder’s Rights Advisory Committee, Melbourne. Priapanjta, P. 2000, ‘Trade Secret: How does this apply to drug registration data?’, presented at ASEAN Workshop on the TRIPS Agreement and its Impact on Pharmaceuticals, Jakarta, Department of Health and World Health Organization, 24 May. PRS for Music 2015, Performing Right Society 2015 Financial Review, UK, http://www.prsformusic.com/SiteCollectionDocuments/About%20MCPS-PRS/ financial-results/prs-for-music-financial-review-2015.pdf (accessed 5 September 2016). Prud’homme, D. 2014, Creating a ‘model’ utility model patent system: A comparative analysis of the utility model patent systems in Europe and China, Working Paper Series 2014, IP Key, Alicante, Spain. 740 INTELLECTUAL PROPERTY ARRANGEMENTS Puddy, R. 2016, Smashing solution to China’s imitation game, http://www.theaustralian.com.au/news/nation/smashing-solution-to-chinas-imitationgame/news-story/4c5da325b66b5dbfd0d50bb56288db78 (accessed 3 March 2016). PwC (PricewaterhouseCoopers) 2014, Maintaining Australia’s advantage: Institutions and Innovation, Policy Paper prepared for the Australian Council of Learned Academies, July, Sydney. —— 2015a, Economic impacts of the Canadian educational sector’s fair dealing guidelines. —— 2015b, The Economic Contribution of Australia’s Copyright Industries 2002-2014, April, Australia. —— 2016, Understanding the costs and benefits of introducing a ‘fair use’ exception. Radomsky, L. 2000, ‘Sixteen Years after the Passage of the U.S. Semiconductor Chip Protection Act: Is International Protection Working?’, Berkeley Technology Law Journal, vol. 15, no. 3, pp. 1049–1094. Rai, A.K. 2013, ‘Improving (Software) Patent Quality Through the Administrative Process’, Houston Law Review, vol. 51, pp. 203–543. de Rassenfosse, G. 2015, On the price elasticity of demand for trademarks, Working Paper, Ecole polytechnique fédérale de Lausanne, Lausanne, Switzerland. —— and Jaffe, A. 2015, Are Patent Fees Effective at Weeding out Low-quality Patents?, Motu Working Paper, January, 15–01. ——, —— and Webster, E. 2016, Low-quality patents in the eye of the beholder: Evidence from multiple examiners, NBER Working Paper No. 22244, Cambridge. —— and van Pottelsberghe, B. 2010, ‘On the price elasticity of demand for patents’, Oxford Bulletin of Economics and Statistics, vol. 74, no. 1, pp. 58–77. —— and —— 2013, ‘The role of fees in patent systems: theory and evidence’, Journal of Economic Surveys, vol. 27, no. 4, pp. 696–716. Rauch, C. 2015, Hague to Break It to You: International Design Applications Are Not a Silver Bullet for Multijurisdictional Protection, HodgsonRuss. Rauch, J. 1993, ‘The Realities of our Times: The Semiconductor Chip Protection Act of 1984 and the Evolution of the Semiconductor Industry’, Fordham Intellectual Property, Media and Entertainment Law Journal, vol. 3, no. 2, pp. 403–439. RBA (Reserve Bank of Australia) 2016, Historical Data, Historical http://www.rba.gov.au/statistics/historical-data.html (accessed 22 July 2016). Data, Readings 2016, The Sun & the Moon & the Rolling Stones by Rich Cohen, http://www.readings.com.au/products/19729307/the-sun-and-the-moon-and-the-rollingstones (accessed 2 August 2016). Redrup, Y. and Durkin, P. 2014, ‘Not easy being green: BP loses bid to trademark colour’, The Sydney Morning Herald, 3 June, http://www.smh.com.au/business/not-easy-being-greenbp-loses-bid-to-trademark-colour-20140702-zsu3i.html (accessed 10 September 2015). REFERENCES 741 Reeves, P. and Mendis, D. 2015, The Current Status and Impact of 3D Printing Within the Industrial Sector: An Analysis of Six Case Studies, Intellectual Property Office, Newport, UK. Reichman, J.H. 2000, ‘Of Green Tulips and Legal Kudzu’, Vanderbilt Law Review, vol. 53, no. 6, pp. 1743–1798. ResMed 2015, ResMed submission on Innovation Patent System, Melbourne. Revesz, J. 1999, Trade-Related Aspects of Intellectual Property Rights, Productivity Commission Staff Research Paper, Canberra. Ricegrowers’ Association of Australia INC 2016, Research & Development, http://www.rga.org.au/the-rice-industry/research-development.aspx (accessed 9 March 2016). Richardson, A. 2014, Should you buy replica furniture?, Daily Life, http://www.dailylife.com.au/life-and-love/dl-home/should-you-buy-replica-furniture20140908-3f2o0.html (accessed 6 July 2016). Richardson, D. 2004, Intellectual property rights and the Australia-US Free Trade Agreement, Research Paper No. 14 2003-04, http://www.aph.gov.au/binaries/ library/pubs/rp/2003-04/04rp14.pdf (accessed 7 September 2015). Richter, V. 2016, What is CRISPR and what does it mean for genetics?, Cosmos Magazine, https://cosmosmagazine.com/biology/what-crispr-and-what-does-it-meangenetics (accessed 8 September 2016). Ricketson, S. and Suthersanen, U. 2012, ‘The design/copyright overlap: Is there a resolution?’, in Wilkof, N. and Basheer, S. (eds), Overlapping intellectual property rights, Oxford University Press. Riker, D. 2014, Intellectual Property Rights and International Receipts of Royalties and Licensing Fees, Office of Economics Working Paper, August, 2014–08C, US International Trade Commission, Washington D.C. Rimmer, M. 2005, ‘The Grey Album: Copyright Law and Digital Sampling’, Media International Australia Incorporating Culture and Policy, no. 114. —— 2009a, Submission to the Joint Standing Committee on Treaties on the agreement between Australian and the European Community on Trade in Wine, Canberra. —— 2009b, ‘The grapes of wrath: the Coonawarra dispute, geographical indications and international trade’, in Kenyon, A., Richardson, M. and Ricketson, S. (eds), Landmarks in Australian intellectual property law, Cambridge University Press, Melbourne, pp. 209–232. RIRDC (Rural Industries Research & Development Corporation) 2014, Pasture Seeds Program Five Year Plan 2013-18, RIRDC Publication, June, 14/053. Rollans, G. and de la Cheneliere, M. 2010, Study of the Canadian K to 12 Education Book Publishing Sector, Department of Canadian Heritage. 742 INTELLECTUAL PROPERTY ARRANGEMENTS Romandini, R. 2016, ‘Flexibilities under TRIPS: An analysis of the proposal for reforming Brazilian patent law’, John Marshall Review of Intellectual Property, vol. 15, pp. 150– 212. Rothnie, W. 2014, ‘Trade marks and parallel imports’, Competition and Consumer Law Journal, vol. 22, pp. 39–62. —— 2016, Repacking into tobacco plain packaging is still parallel importing, http://ipwars.com/2016/06/29/repackaging-into-tobacco-plain-packaging-is-stillparallel-importing/ (accessed 11 August 2016). Rotstein, F. and Weatherall, K. 2007, ‘Filing and Settlement of Patent Disputes in the Federal Court: 1995-2005’, Intellectual Property Forum, no. 68. Sabrine, R. 2015, The market value of technological innovation: evidence from European patents, University of Paris Dauphine. Sakakibara, M. and Branstetter, L. 2001, ‘Do stronger patents induce more innovation? Evidence from the 1988 Japanese patent law reforms’, RAND Journal of Economics, vol. 32, no. 1, pp. 77–77. Sampat, B. and Williams, H. 2015, How do patents affect follow-on innovation? Evidence from the human genome, NBER Working Paper No. 21666, Cambridge, MA. Samuel, S.N., Kingma, O.T. and Crellin, I.R. 1983, Government intervention in rural research : some economic aspects of the Commonwealth’s role, Occasional Paper, 76, Australian Bureau of Agricultural Economics, Canberra. Sanderson, J. 2007, ‘Back to the Future: Possible Mechanisms for the Management of Plant Varieties in Australia’, UNSW Law Journal, vol. 30, no. 3, pp. 686–712. —— 2011, ‘Intellectual Property and Plants: Constitutive, Contingent and Complex’, Emerging Issues in Intellectual Property, Oxford University Press, pp. 164–182. —— and Adams, K. 2008, ‘Are plant breeder’s rights outdated? A descriptive and empirical assessment of plant breeder’s rights in Australia, 1987-2007’, Melbourne University Law Review, vol. 32, no. 3, pp. 980–1006. Sansom, L., Delaat, W. and Horvath, J. 2015, Review of Medicines and Medical Devices Regulation: Report on the regulatory framework for medicines and medical devices, 31 March, Canberra. Sauer, R. 2007, Why Develop Open Source Software? The Role of Non-Pecuniary Benefits, Monetary Rewards and Open Source Licence Type, IZA Discussion Paper No. 3197, Institute for the Study of Labor (IZA), Bonn, Germany. SCARC (Senate Community Affairs References Committee) 2010, Gene patents, November, Canberra. Scellato, G., Calderini, M., Caviggioli, F., Franzoni, C., Ughetto, E., Kica, E. and Rodriguez, V. 2011, Study on the quality of the patent system in Europe, Torino, Italy. Schankerman, M. and Pakes, A. 1986, ‘Estimates of the value of patent rights in European countries during the post-1950 period’, The Economic Journal, vol. 96, pp. 1052–1076. REFERENCES 743 —— and Schuett, F. 2016, Screening for Patent Quality: Examination, Fees, and the Courts, Centre for Economic Policy Research Discussion Paper, forthcoming. Scherer, F. 2011, R&D Costs and Productivity in Biopharmaceuticals, RPP-2011-10, Regulatory Policy Program, Mossavar-Rahmani Center for Business and Government, Harvard Kennedy School, Cambridge MA. Schmidt, C. 2011, ‘Comment: some economic considerations regarding optimal intellectual property protection’, in Kur, A. and Mizaras, V. (eds), The Structure of Intellectual Property Law: Can One Size Fit All?, Edward Elgar, Cheltenham, UK. Scotchmer, S. 1991, ‘Standing on the shoulders of giants: cumulative research and the patent law’, Journal of Economic Perspectives, vol. 5, no. 1, pp. 29–41. Screen Australia 2016, Box office in Australia for Australian feature films, Australian Box Office Share, http://www.screenaustralia.gov.au/fact-finders/cinema/industrytrends/box-office/australian-box-office (accessed 15 July 2016). Screenrights 2015, 2014-15 Screenrights Highlights Report, 2014-15 Screenrights Highlights Report, https://www.screenrights.org/sites/default/files/uploads/20142015_Screenrights_Highlights_Report.pdf (accessed 13 April 2016). SCRGSP (Steering Committee for the Review of Government Service Provision) 2016, Report on Government Services 2016, Productivity Commission, Canberra. Sewell, A. 2014, ‘Paul’s Boutique and Fear of a Black Planet: Digital Sampling and Musical Style in Hip Hop’, Journal of the Society for American Music, vol. 8, no. 1, pp. 28–48. SFADTRC (Senate Foreign Affairs, Defence and Trade Reference Committee) 2015, Blind agreement: reforming Australia’s treaty-making process, Canberra. Shapiro, C. 2004, ‘Navigating the Patent Thicket: Cross Licenses, Patent Pools, and Standard Setting’, Innovation Policy and the Economy, vol 4, The MIT Press, Cambridge, Massachusetts. Shavell, S. 2010, ‘Should Copyright of Academic Works be Abolished?’, Journal of Legal Analysis, vol. 2, Number 1, pp. 301–358. —— and van Ypersele, T. 2001, ‘Rewards versus intellectual property rights’, Journal of Law and Economics, vol. 44, pp. 521–547. Sherman, B. 2015, Computer Programs As Excluded Patentable Subject Matter, http://www.wipo.int/edocs/mdocs/scp/en/scp_15/scp_15_3-annex2.pdf (accessed 7 April 2016). Shin, Y.S. 2014, Patent Term Extension in Korea, January, American Intellectual Property Law Association - International Subcommittee. Siepmann, T. 2004, ‘The Global Exportation of the U.S. Bayh-Dole Act’, University of Dayton Law Review, vol. 30, pp. 209–244. SLCLC (Senate Legal and Constitutional Legislation Committee) 2001, Inquiry into the Provisions of the Copyright Amendment (Parallel Importation) Bill 2001, Canberra. 744 INTELLECTUAL PROPERTY ARRANGEMENTS Smith, A. 2013, Many Questions: System Shock 2 Comes To GOG, Rock Paper Shotgun, https://www.rockpapershotgun.com/2013/02/13/many-questions-system-shock-2comes-to-gog/ (accessed 12 April 2016). Smith, G. 2016, The Role of the European Ombudsman in bringing greater Transparency to EU Negotiations. Smith, M. and Zentner, A. 2016, ‘Internet Effects on Retail Markets’, in Basker, E. (ed), Handbook on the Economics of Retailing and Distribution, Edward Elgar, Cheltenham, UK, pp. 433–454. Soames, L., Brunker, D. and Talgaswatta, T. 2011, Competition, Innovation and Productivity in Australian Businesses, Joint research paper, Australian Bureau of Statistics & Productivity Commission, Canberra. SOFTIC (Software Information Centre) 2014, Registration of the semiconductor circuit layout rights (by year), http://www.softic.or.jp/ic/ic-layout/toukei/suii-fy.html (accessed 30 March 2016). Sony 2016, Open Source, http://developer.sonymobile.com/knowledge-base/open-source/ (accessed 14 April 2016). SoundCloud 2015, Go Pro on, SoundCloud, https://soundcloud.com/pro?ref=t204 (accessed 19 October 2015). SPTO (Spanish Patent and Trademark Office) 2015, Basic Issues about Patents and Utility Models, Madrid, Spain. SRA (Sugar Research Australia) 2015, Varieties and plant breeding update, December, http://www.sugarresearch.com.au/page/Growing_cane/Varieties/Plant_breeding_newsl etter/ (accessed 4 February 2016). Stamelos, I., Angelis, L., Oikonomou, A. and Bleris, G.L. 2002, ‘Code quality analysis in open source software development’, Information Systems Journal, vol. 12, no. 1, pp. 43–60. Standing Committee on Legal and Constitutional Affairs (House of Representatives Standing Committee on Legal and Constitutional Affairs) 2000, Cracking down on copycats: enforcement of copyright in Australia, November, Parliament of Australia. Stanford University 2015, Summaries of Fair Use http://fairuse.stanford.edu/overview/fair-use/cases/ (accessed 16 April 2016). Cases, State Intellectual Property Office of the P.R.C and SIPO 2016, Table 11 Registration of Applications for Layout Designs of Integrated Circuits, http://english.sipo.gov.cn/statistics/2016/2/201603/t20160318_1252634.html (accessed 30 March 2016). Stern, S. 2016, ‘A history of Australia’s wine Geographical Indications legislation’, in Gangjee, D. (ed), Research Handbook on Intellectual Property and Geographical Indications, Edward Elgar, Cheltenham, UK, pp. 245–291. REFERENCES 745 Stewart, A.J., Griffith, P., Bannister, J. and Liberman, A. 2015, Intellectual Property in Australia, 5th edn, Chatswood, NSW LexisNexis Butterworths, Sydney. Stiglitz, J. 2008, ‘Economic foundations of intellectual property rights’, Duke Law Journal, vol. 57, pp. 1693–1724. Stockton, R. 2015, The U.S. and Japan join the Hague System, Lexology. Stoianoff, N. and Roy, A. 2015, ‘Indigenous knowledge and culture in Australia - the case for sui generis legislation’, Monash University Law Review, vol. 41, no. 3, pp. 745–784. Stone, B. 2009, ‘Amazon Erases Orwell Books from Kindle’, New York, The New York Times, 17 July, http://www.nytimes.com/2009/07/18/technology/companies/ 18amazon.html?_r=0 (accessed 12 April 2016). Summerfield, M. 2015, Submission to IP Australia on ACIP’s recommendation on the innovation patent system, Canberra. —— 2016, Government Backs Australian FinTech Innovation – It’s Just a Pity It’s Not Patentable!, http://blog.patentology.com.au/2016/03/government-backs-australianfintech.html?utm_source=feedburner&utm_medium=twitter&utm_campaign=Feed%3 A+Patentology (accessed 16 March 2016). Suzor, N. 2006, ‘Transformative Use of Copyright Material’, LLM Thesis, Queensland University of Technology. —— and Dootson, P. 2013, Inquiry into IT Pricing: Submission to the House Standing Committee on Infrastructure and Communications, http://nic.suzor.net/wpcontent/uploads/2013/03/Suzor-Dootson-2013-IT-Pricing.pdf (accessed 4 November 2015). —— and Fitzgerald, B. 2011, ‘The legitimacy of graduated response schemes in copyright law’, University of New South Wales Law Journal, vol. 34, no. 1, pp. 1–40. Sweeny, K. 2013, The impact of Further PBS Reforms: Report to Medicines Australia, May, Centre for Strategic Economic Studies, Victoria University, Melbourne. Tabarrok, A. 2002, ‘Patent Theory v Patent Law’, Contributions to Economic Analysis & Policy, vol. 1, no. 1, Article 9. —— 2013, Innovation Policy and Australia, http://grattan.edu.au/wpcontent/uploads/2014/05/908_tabarrok_alliance_21.pdf (accessed 6 November 2015). Tansey, C. and Dixon, G. 2015, Differences in trans-Tasman patent law: extension of term for pharmaceuticals, Media release, Shelston IP, http://www.shelstonip.com/news/ differences-in-trans-tasman-patent-law-extension-of-term-for-pharmaceuticals/ (accessed 28 January 2016). Taubman, A. 2008, ‘Australia’s interests under TRIPS dispute settlement: Trade negotiations by other means, multilateral defence of domestic policy choice, or safeguarding market access?’, Melbourne Journal of International Law, vol. 9, pp. 217–270. Taylor Wessing 2013, Global Intellectual Property Index - 4th Report, UK. 746 INTELLECTUAL PROPERTY ARRANGEMENTS —— 2016, Global Intellectual Property Index - 5th Report, UK. Tehranian, J. 2007, ‘Infringement Nation: Copyright Reform and the Law/Norm Gap’, Utah Law Review, vol. 2007, no. 3, pp. 537–550. The Economist 2002, Innovation’s Golden Goose, The http://www.economist.com/node/1476653 (accessed 12 April 2016). Economist, The Law Library of Congress 2016, Patent Term Extensions and Adjustments, March, Washington D.C. The Wall Street Journal 2014, A Study Says It Costs $2.6 Billion to Develop a Drug, but Will This Become Gospel?, Media release, 21 November 2014. Thomas, J. 2009, Patent ‘Evergreening’: Issues in Innovation and Competition, 13 November. —— 2013, The Role of Patents and Regulatory Exclusivities in Pharmaceutical Innovation, 7 January. —— 2014, Follow-On Biologics: The Law and Intellectual Property Issues, Report No. R41483, 15 January, Congressional Research Service, Washington D.C. Thomas, K. 2006, ‘Protecting Academic and Non-Profit Research: Creating a Compulsory Licensing Provision in the Absence of an Experimental Use Exception’, Santa Clara High Technology Law Journal, vol. 23, pp. 347–370. Thomson, R. 2014, ‘The Yield of Plant Variety Protection’, American Journal of Agricultural Economics, vol. 97, no. 3, pp. 762–785. Throsby, D. and Zednik, A. 2010, Do you really expect to get paid? An economic study of professional artists in Australia, Australia Council for the Arts, Sydney. Thursby, J., Jensen, R. and Thursby, M. 2001, ‘Objectives, characteristics and outcomes of university licensing: A survey of major U.S. universities’, Journal of Technology Transfer, vol. 26, pp. 59–72. —— and Thursby, M.C. 2006, Here or There? A Survey of Factors in Multinational R&D Location: Report to the Government/University/Industry Research Roundtable, SSRN Scholarly Paper, 1 December, ID 1456460, Social Science Research Network, Rochester, NY, http://papers.ssrn.com/abstract=1456460 (accessed 25 August 2016). TIPO (Taiwan Intellectual Property Office) 2015, Annual Statistics, Taiwan. TNS 2015, Online Copyright Infringement Research, June, Canberra. Trajtenberg, M. 1990, ‘A penny for your quotes: patent citations and the value of innovation’, RAND Journal of Economics, vol. 21, no. 1, pp. 172–187. Treasury 2011, Treasury Strategic Framework 2012-13, Canberra. Tresidder, G. 2011, Designs: When is a replica not a replica?, http://www.hwlebsworth.com.au/latest-news-a-publications/publications/intellectualproperty-and-trade-marks/item/284-when-is-a-replica-not-a-replica?.html (accessed 8 September 2016). REFERENCES 747 Trevisian, P. 2014, ‘Experiences with the Patent Prosecution Highway (PPH) at the USPTO’, Tel Aviv, Israel. Uhrig, J. 2003, Review of the Corporate Governance of Statutory Authorities and Office Holders, Canberra. UK Cabinet Office 2015, Machinery of Government Guidance, October. UK CMA (Competition and Markets Authority) 2016, CMA fines pharma companies £45 million, Media Release, 12 February. UK Government 2016, Register a design, GOV.UK, https://www.gov.uk/register-adesign/check-if-you-can-register-your-design (accessed 9 April 2016). UK IPO 2011, IPO Assessment of the Need for Reform of the Design Intellectual Property Framework, UK IPO, Cardiff, UK. UK IPO 2016, Proposal for changes to Registered Design Fees Government Response. —— (Intellectual Property Office) 2012, Joining the Hague Agreement on Designs: Impact Assessment. UniQuest 2014, Our Track Record - Uniquest, http://www.uniquest.com.au/our-trackrecord (accessed 14 April 2016). United States International Trade Commission 2016, Trans-Pacific Partnership Agreement: Likely Impact on the U.S. Economy and on Specific Industry Sectors, Publication Number 4607; Investigation Number TPA-105-001, Washington D.C. UNSW 2015, Submission to the review of research policy and funding arrangements for higher education: issues paper, Sydney. UPOV (International Union for the Protection of New Varieties of Plants) 2005, UPOV Report on the Impact of Plant Variety Protection, International Union For The Protection Of New Varieties Of Plants, Geneva, Switzerland. —— 2013, Seminar on essentially derived varieties (EDVs) (UPOV/SEM/GE/13), http://www.upov.int/meetings/en/details.jsp?meeting_id=29782 (accessed 8 September 2016). —— 2016, About the UPOV System of Plant Variety Protection, http://www.upov.int/about/en/upov_system.html (accessed 8 September 2016). US Copyright Office 2015, Orphan Works and Mass Digitization, June, United States. —— 2016, Public Catalog, http://cocatalog.loc.gov/cgi-bin/Pwebrecon.cgi?DB= local&PAGE=First (accessed 30 March 2016). US Department of Commerce (United States Department of Commerce) 2012, Intellectual Property and the U.S. Economy: Industries in Focus, March, United States. US Department of Justice and the Federal Trade Commission 1995, Antitrust Guidelines For The Licensing Of Intellectual Property. 748 INTELLECTUAL PROPERTY ARRANGEMENTS US Department of Justice and the Federal Trade Commission 2007, Antitrust Enforcement and Intellectual Property Rights: Promoting Innovation and Competition, Washington D.C. US PTO (United States Patent and Trademark Office) http://www.uspto.gov/about-us (accessed 13 September 2016). 2016, About us, USCC (US Chamber of Commerce) 2015, GIPC International IP Index, Global Intellectual Property Center, Washington D.C. Useche, D. 2014, ‘Are patents signals for the IPO market? An EU-US comparison for the software industry’, Research Policy, vol. 43, no. 8, pp. 1299–1311. USPTO (United States Patent and Trademark Office) 2015, Patent term, uspto.gov, http://www.uspto.gov/web/offices/pac/mpep/s2701.html (accessed 9 April 2016). —— 2016, USPTO Fee Schedule, http://www.uspto.gov/learning-and-resources/fees-andpayment/uspto-fee-schedule (accessed 10 February 2016). UTS 2015, Submission to the review of research policy and funding arrangements for higher education issues paper, Sydney. Vaile, M. 2004, Letter to the Honourable Robert B. Zoellick, United States Trade Representative, 18 May, Canberra, Australia. Valdés, R. and McCann, M. 2014, Intellectual Property Provisions in Regional Trade Agreements: Revision and Update, Staff Working Paper ERSD-2014-14, World Trade Organization, Washington D.C. Variety Central 2016a, End Point Royalties, http://varietycentral.com.au/end-pointroyalties (accessed 3 March 2016). —— 2016b, Varieties and Rates, http://varietycentral.com.au/varieties-and-rates (accessed 4 April 2016). VCEC (Victorian Competition and Efficiency Commission) 2005, Housing Regulation in Victoria: Building Better Outcomes, October, Melbourne. —— 2009, A Sustainable Future for Victoria: Getting Environmental Regulation Right, Melbourne. Verve Economics 2013, The economic value of the Australian innovation patent, Canberra. Victorian DTF (Victorian Department of Treasury and Finance) 2014, Statement of Expectations for regulators, http://www.dtf.vic.gov.au/Publications/Victoria-Economypublications/Statement-of-Expectations-for-regulators (accessed 11 September 2016). Viscopy 2015, 2014-15 Annual Review, 2014-15 Viscopy Annual Review, http://www.annualreview2014-15.com/copyrightviscopy.html (accessed 13 April 2016). Wagner, G. 2007, ‘Transaction costs, choice of law and uniform contract law’, Modern law for global commerce: Proceedings of the congress of the United Nations Commission REFERENCES 749 on international trade law held on the occasion of the fortieth session of the Commission, vol 39. Wakasugi, R. and Ito, B. 2007, ‘The effect of stronger intellectual property rights on technology transfer: evidence from Japanese firm-level data’, The Journal of Technology Transfer, vol. 34, pp. 145–158. Waldfogel, J. 2012, ‘Music piracy and its effects on demand, supply and welfare’, Policy and the Economy, vol. 12, pp. 91–110. —— 2014, ‘Digitization, Copyright, and the Flow of New Music Products’, Handbook of the Economics of Art and Culture, vol 2, North Holland, Oxford, pp. 277–298. Wang, O. 2013, 20 Years Ago Biz Markie Got The Last Laugh, The Record: Music News From NPR, http://www.npr.org/sections/therecord/2013/05/01/180375856/20-yearsago-biz-markie-got-the-last-laugh (accessed 12 April 2016). Ware, M. and Mabe, M. 2009, The STM Report: An Overview of Scientific and Scholarly Journal Publishing, The Hague, The Netherlands. Waterhouse, D. 2013, ‘Experience on essentially derived varieties in Australia’, International Union for the Protection of New Varieties of Plants, http://www.upov.int/meetings/ en/details.jsp?meeting_id=29782 (accessed 3 March 2016). Weatherall, K. and Webster, E. 2010, ‘Patent Infringement in Australia: Results From A Survey’, Federal Law Review, vol. 38, no. 1, pp. 21–70. Werde, B. 2004, ‘Defiant Downloads Rise From Underground’, The New York Times, 25 February, http://www.nytimes.com/2004/02/25/arts/defiant-downloads-rise-fromunderground.html (accessed 16 February 2016). WHO (World Health Organization) nd, WHO World Health Organization Assesses the World’s Health Systems, WHO, http://www.who.int/whr/2000/media_centre/ press_release/en/ (accessed 16 September 2016). WhoSampled 2016, Samples of Funky Drummer by James Brown on WhoSampled, http://www.whosampled.com/James-Brown/Funky-Drummer/sampled/ (accessed 12 April 2016). Wilkof, N. and Basheer, S. (eds) 2012, Overlapping intellectual property rights, Oxford University Press, UK. Wine Australia 2015, Geographical Indications, http://www.wineaustralia.com/en /Production%20and%20Exporting/Register%20of%20Protected%20GIs%20and%20Ot her%20Terms/Geographical%20Indications.aspx (accessed 14 September 2016). WIPO (World Intellectual Property Organization) 2003, Standing Committee on the Law of Patents, Ninth Session Geneva, May 12 to 16, 2003: Status of the Practice Guidelines under the Draft Substantive Patent Law Treaty: Options for discussion, Geneva, Switzerland. —— 2010, Transfer of Technology, 25 June, Standing Committee on the Law of Patents, Fourteenth Session, Geneva, Switzerland. 750 INTELLECTUAL PROPERTY ARRANGEMENTS —— 2011, World Intellectual Property Report: The Changing Face of Innovation, 944E/2011, Geneva, Switzerland. —— 2012, The Hague Agreement Concerning the International Registration of Industrial Designs: Main Features and Advantages, 944(E), WIPO, Geneva, Switzerland. —— 2013, World Intellectual Property Report: Brands - Reputation and Image in the Global Marketplace, Geneva, Switzerland. —— 2015a, Guide on Surveying the Economic Contribution of the Copyright Industries, 893 E, Geneva, Switzerland. —— 2015b, Hague Yearly Review: International Registration of Industrial Designs, 930E/15, Economics & Statistics Series, WIPO, Geneva, Switzerland. —— 2015c, Members of the Madrid Union, http://www.wipo.int/madrid/en/members/ (accessed 6 January 2016). —— 2015d, Study on Inventive Step, SCP/22/3, Geneva, Switzerland. —— 2015e, WIPO-Administered Treaties, http://www.wipo.int/treaties/en/ (accessed 2 October 2015). —— 2015f, WIPO IP Statistics Data Center, IpsStatsResultvalue (accessed 25 February 2016). http://ipstats.wipo.int/ipstatv2/ —— 2015g, World Intellectual Property Indicators, 2015, Geneva, Switzerland. —— 2015h, World Intellectual Property Report: Breakthrough innovation and economic growth, 944E, Economics & Statistics Series, Geneva, Switzerland. —— 2016a, Advice on Flexibilities under the TRIPS Agreement, http://www.wipo.int/ipdevelopment/en/legislative_assistance/advice_trips.html (accessed 15 April 2016). —— 2016b, Directory of Intellectual Property Offices, http://www.wipo.int/directory/ en/urls.jsp (accessed 19 January 2016). —— 2016c, Hague Agreement Concerning the International Registration of Industrial Designs Geneva Act of the Hague Agreement Concerning the International Registration of Industrial Designs, Media Release, 13 January, The Hague Notification No. 126, Geneva, Switzerland. —— 2016d, IP Statistical Country Profiles, WIPO, en/statistics/country_profile/ (accessed 7 April 2016). —— 2016e, IP Statistics Data Center, keysearch.htm?keyId=222 (accessed 9 August 2016). http://www.wipo.int/ipstats/ http://ipstats.wipo.int/ipstatv2/ —— 2016f, Standing Committee on the Law of Trademarks, Industrial Designs and Geographical Indications (SCT),Thirty-Sixth Session, Geneva, Switzerland. —— 2016g, Standing Committee on the Law of Trademarks, Industrial Designs and Geographical Indications: Thirty-Fifth Session , April 25 to 27, 2016: Industrial Design Law And Practice – Draft Regulations, Geneva, Switzerland. REFERENCES 751 —— 2016h, Standing Committee on the Law of Trademarks, Industrial Designs and Geographical Indications: Thirty-Fourth Session Geneva, November 16 to 18, 2015, Geneva, Switzerland. —— 2016i, Traditional Knowledge Frequently Asked http://www.wipo.int/tk/en/resources/faqs.html (accessed 13 July 2016). Questions, —— 2016j, U.S. Extends Lead in International Patent and Trademark Filings, http://www.wipo.int/pressroom/en/articles/2016/article_0002.html (accessed 25 May 2016). —— 2016k, WIPO Administered Treaties, http://www.wipo.int/treaties/en/ (accessed 9 September 2016). —— nd, Protecting Innovations by Utility Models, http://www.wipo.int/sme/en/ ip_business/utility_models/utility_models.htm (accessed 14 April 2016a). —— nd, What is Intellectual Property?, http://www.wipo.int/about-ip/en/ (accessed 2 October 2015b). Woo, T. 2014, Letter in response to request for clarifications on the geographical indications provisions of the Korea-EU Free Trade Agreement, 24 March, Ministry of Trade, Industry & Energy, Seoul, Korea. Working Party to Review the Trade Marks Legislation 1992, Recommended Changes to the Australian Trade Marks Legislation, Final Report, Canberra. World Bank 2015, Population, total, http://data.worldbank.org/indicator/SP.POP.TOTL (accessed 9 April 2016). —— 2016, World Development Indicators, http://databank.worldbank.org/data/reports.aspx ?source=2&series=BM.GSR.ROYL.CD&country= (accessed 14 September 2016). Yale University 2015, New discount for Yale authors on article-processing charges, http://library.medicine.yale.edu/blog/new-discount-yale-authors-article-processingcharges (accessed 16 April 2016). Yen, A. 1991, ‘When Authors Won’t Sell: Parody, Fair Use, and Efficiency in Copyright Law’, University of Colorado Law Review, vol. 62, pp. 79–108. Zemer, L. 2011, ‘Copyright departures: the fall of the last imperial copyright dominion and the case of fair use’, DePaul Law Review, vol. 60, no. 4. Zuallcobley, R.W., Castañeda, H.J.A., Davis, A.J., Dean, O., Fysh, H.M., Harms, H.L., Kallinikou, D., Mimura, H.R., Ombija, H.N., Ono, S., Pacón, A.M. and Poonsombudldert, K. 2012, Study on Specialized Intellectual Property Courts, 25 January, International Intellectual Property Institute and the United States Patent and Trademark Office, Washington D.C. 752 INTELLECTUAL PROPERTY ARRANGEMENTS