ISSN 1178–2560 Decision Series Project no. 11.04/15933 Public version Draft Determination Note: This is a draft determination issued for the purpose of advancing the Commerce Commission’s decision on this matter. The conclusions reached in this draft determination are preliminary and take into account only the information provided to the Commission to date. NZME Limited and Fairfax New Zealand Limited The Commission: Dr Mark Berry Sue Begg Elisabeth Welson Graham Crombie Summary of application: NZME Limited and Fairfax NZ Limited and its parent company, Fairfax Media Limited, seek approval to merge the New Zealand operations of NZME Limited and Fairfax NZ Limited. Draft Determination: The Commerce Commission’s preliminary view is that, on the basis of the information provided to date, it is not satisfied that the merger will not have, or would not be likely to have, the effect of substantially lessening competition in a market. The Commission is also not satisfied that the merger will result, or will be likely to result in such a benefit to the public that it should be permitted. Therefore, the Commerce Commission’s preliminary view is that it would be likely to decline to grant authorisation for the merger pursuant to section 67(3)(c) of the Commerce Act 1986. Date of Draft Determination: 8 November 2016 2670529 2 Confidential material in this report has been removed. Its location in the document is denoted by [ ]. 3 CONTENTS SECTION 1: EXECUTIVE SUMMARY...............................................................................................8 TIMEFRAMES FOR SUBMISSIONS AND A CONFERENCE ..................................................................... 11 STRUCTURE OF THIS DRAFT DETERMINATION ................................................................................... 11 SECTION 2: THE COMMISSION’S APPROACH .............................................................................. 12 THE PROPOSED MERGER .................................................................................................................... 12 THE COMMISSION’S PRELIMINARY VIEW ........................................................................................... 12 THE COMMISSION‘S INVESTIGATION ................................................................................................. 12 LEGAL FRAMEWORK ........................................................................................................................... 13 Analysing the competition effects of a merger .......................................................................................... 14 Assessing the public benefit of a merger.................................................................................................... 15 SECTION 3: KEY PARTIES AND COMPETITION ISSUES .................................................................. 19 THE MERGING PARTIES ....................................................................................................................... 19 NZME .......................................................................................................................................................... 19 Fairfax ........................................................................................................................................................ 20 OTHER INDUSTRY PARTICIPANTS ........................................................................................................ 21 Television New Zealand (TVNZ) .................................................................................................................. 21 MediaWorks Holdings Limited (MediaWorks) ........................................................................................... 21 Radio New Zealand (RNZ)........................................................................................................................... 22 Allied Press Limited (Allied Press) ............................................................................................................... 22 KPEX Limited (KPEX) ................................................................................................................................... 22 INDUSTRY BACKGROUND ................................................................................................................... 22 Print is declining ......................................................................................................................................... 22 The move to online consumption of news and information ....................................................................... 23 A digital-first strategy ................................................................................................................................ 23 Inevitable consolidation ............................................................................................................................. 24 THE APPLICANTS’ RATIONALE FOR THE PROPOSED MERGER ............................................................ 24 WITH THE MERGER ............................................................................................................................. 25 WITHOUT THE MERGER ...................................................................................................................... 26 THE RELEVANT MARKETS .................................................................................................................... 27 How we define markets .............................................................................................................................. 27 Characteristics of two-sided markets ......................................................................................................... 28 Markets that could be affected by the proposed merger........................................................................... 28 SECTION 4: ADVERTISING MARKETS .......................................................................................... 30 MARKET DEFINITION IN ADVERTISING MARKETS ............................................................................... 30 Applicants’ view ......................................................................................................................................... 30 Previous Commission decisions – advertising markets............................................................................... 30 Advertisers’ views ....................................................................................................................................... 31 The Commission’s view............................................................................................................................... 33 DIGITAL ADVERTISING MARKETS ........................................................................................................ 34 Market definition ....................................................................................................................................... 34 Competition analysis – digital advertising ................................................................................................. 35 KPEX ........................................................................................................................................................... 40 Competition analysis – premium digital advertising .................................................................................. 41 PRINT ADVERTISING ............................................................................................................................ 46 Market definition ....................................................................................................................................... 46 Competition analysis – Sunday newspapers .............................................................................................. 49 Do the Fairfax Sunday newspapers currently provide a competitive constraint on the Herald on Sunday? ...................................................................................................................................................... 50 Competition analysis – metropolitan daily newspapers in the Waikato and Hawke’s Bay regions ........... 53 4 COMMUNITY NEWSPAPER ADVERTISING........................................................................................... 55 How the proposed merger could substantially lessen competition ............................................................ 55 Market definition ....................................................................................................................................... 58 Competition analysis .................................................................................................................................. 60 Investigation approach .............................................................................................................................. 60 Entry and expansion by competing community newspapers ..................................................................... 72 Conclusion .................................................................................................................................................. 76 Competition between The New Zealand Herald and Fairfax’s community network .................................. 77 SECTION 5: READER MARKETS ................................................................................................... 78 MARKET DEFINITION ........................................................................................................................... 78 Demand for location-specific content ........................................................................................................ 79 Media platforms and content format......................................................................................................... 80 The role of print .......................................................................................................................................... 82 Production versus distribution of online New Zealand news...................................................................... 84 The Commission’s approach to reader markets ......................................................................................... 84 COMPETITION ANALYSIS ..................................................................................................................... 85 The online New Zealand news market ....................................................................................................... 86 Existing competition ................................................................................................................................... 86 Entry or expansion by online New Zealand news producers ...................................................................... 98 Introduction of a paywall ......................................................................................................................... 101 Reduction in quality.................................................................................................................................. 105 Two-sided market..................................................................................................................................... 111 Sunday and community newspaper markets ........................................................................................... 114 Impact on price......................................................................................................................................... 114 Impact on quality ..................................................................................................................................... 114 Conclusion ................................................................................................................................................ 116 SECTION 6: VERTICAL AND CONGLOMERATE EFFECTS .............................................................. 117 VERTICAL EFFECTS............................................................................................................................. 117 Independent publishers ............................................................................................................................ 117 How the proposed merger could substantially lessen competition .......................................................... 118 Applicants’ views ...................................................................................................................................... 118 The Commission’s view............................................................................................................................. 118 CONGLOMERATE EFFECTS ................................................................................................................ 119 How the proposed merger could substantially lessen competition .......................................................... 119 Applicants’ views ...................................................................................................................................... 119 The Commission’s view............................................................................................................................. 120 OTHER EFFECTS ................................................................................................................................. 121 Withdrawal from News Works NZ............................................................................................................ 121 SECTION 7: PUBLIC BENEFITS AND DETRIMENTS ...................................................................... 124 THE COMMISSION’S ANALYTICAL METHOD ..................................................................................... 124 APPROACH TO QUANTIFYING BENEFITS AND DETRIMENTS............................................................. 125 QUANTIFIABLE BENEFITS: PRODUCTION EFFICIENCIES .................................................................... 125 Benefits – status quo ................................................................................................................................ 125 Sales ......................................................................................................................................................... 126 Marketing ................................................................................................................................................. 127 Editorial .................................................................................................................................................... 129 Printing, procurement and distribution .................................................................................................... 132 Management............................................................................................................................................ 133 Information technology ............................................................................................................................ 134 Premises ................................................................................................................................................... 135 Paper rationalisations .............................................................................................................................. 136 Finance costs ............................................................................................................................................ 137 Merger transactions costs ........................................................................................................................ 137 5 Total quantified benefits .......................................................................................................................... 137 QUANTIFIED DETRIMENTS ................................................................................................................ 138 Approach to quantifying detriments ........................................................................................................ 139 Detriments – status quo ........................................................................................................................... 139 Detriments – digital only .......................................................................................................................... 151 WEALTH TRANSFERS TO AND FROM NON-NEW ZEALANDERS FROM PRICE INCREASES................. 153 Approach to quantifying wealth transfers ............................................................................................... 154 Residency status of relevant groups ......................................................................................................... 155 Components of transfers .......................................................................................................................... 156 Wealth transfers – status quo .................................................................................................................. 157 Wealth transfers – digital only ................................................................................................................. 158 CONCLUSION ON THE QUANTIFIED BENEFITS AND DETRIMENTS.................................................... 159 UNQUANTIFIED BENEFITS ................................................................................................................. 163 Increased quality in reader markets ......................................................................................................... 163 UNQUANTIFIED DETRIMENTS ........................................................................................................... 163 Decreased quality in reader markets ....................................................................................................... 163 Detriments from varying forms of paywalls ............................................................................................. 165 Unquantified detriments: media plurality effects .................................................................................... 166 Applicants’ views ...................................................................................................................................... 169 Others’ views ............................................................................................................................................ 170 Assessing potential plurality effects of the proposed merger .................................................................. 174 Existing level of plurality .......................................................................................................................... 175 Change to plurality as a result of the proposed merger ........................................................................... 189 Conclusion ................................................................................................................................................ 190 BALANCING OF BENEFITS AND DETRIMENTS ................................................................................... 190 QUALITATIVE ASSESSMENT .............................................................................................................. 191 SECTION 8: DRAFT DETERMINATION........................................................................................ 194 THE COMMISSION MUST BE SATISFIED ............................................................................................ 194 DRAFT DETERMINATION ................................................................................................................... 195 6 Glossary Term Description App A type of software that allows you to perform specific tasks. Applications for desktop or laptop computers are sometimes called desktop applications, while those for mobile devices are called mobile apps. Banner or display advertising Advertisements that are embedded into a webpage and which typically occupy the top and sides of the webpage, or within articles and pages. Cpi or cpm ‘Cost per impression’ (cpi) or ‘cost per thousand impressions’ (cpm), which refers to the cost of advertising where advertisers pay each time an ad is displayed. Homepage takeovers Advertisements displayed for a day on a publisher homepage. Often used for branding campaigns. Mobile interstitials Full-screen advertisements that appear before a webpage is first loaded on a mobile phone browser or application. These advertisements must be dismissed by the reader before the webpage or application can be used. Native advertisements Content that is created for the paid promotion of a brand but which does not have the usual format of an advertisement. These are sometimes referred to as advertorials. NZME News Service NZME’s news sharing service with NZME’s daily, Sunday and community newspapers and online websites and independent publishers. News Works The trading name of the New Zealand Press Association Limited’s (NZPA) commercial entity. News Works advocates and promotes the brands of Newspaper Publishers’ Association members (amongst other functions). NPA Newspaper Publishers’ Association. The NPA’s members are Fairfax, Allied Press and various independently owned newspapers: The Gisborne Herald, Ashburton Guardian, The Greymouth Star, Westport News, The Whakatane Beacon and Wairoa Star. Paywall A system that prevents internet users from accessing a webpage’s content without a paid subscription. “Hard” paywalls allow minimal to no access to content without a subscription, while “soft” paywalls allow more flexibility in what users can view without subscribing. Programmatic The use of software to purchase digital advertising, rather than buying 7 Term Description advertising ads from salespeople. Reach The total number of different people or households exposed, at least once, to a medium during a given period. Search advertising A method of placing online advertisements on web pages that show results from search engine queries. Search engine A software system that is designed to search for information on the internet. Syndicated content Content that is not self-generated by the publisher, user generated or generated by freelance writers, but is collected from a range of other news, information and entertainment organisations and is made available in the publisher’s own print or online publication. 8 Section 1: Executive Summary 1. Fairfax and NZME (the Applicants) sought authorisation (or clearance) to merge their respective New Zealand operations. The merger would bring New Zealand’s two largest newspaper networks and most significant online news sites under common ownership. 2. Our preliminary view is to decline authorisation. We consider that the proposed merger would be likely to substantially lessen competition by increasing prices and/or decreasing the quality of content in the following markets: 2.1 premium digital advertising; 2.2 the provision of national online news and information services, through the stuff.co.nz and nzherald.co.nz websites; 2.3 Sunday newspapers in the North Island; and 2.4 community newspapers in: 2.4.1 Whangarei; 2.4.2 Hamilton; 2.4.3 Rotorua; 2.4.4 Taupo; 2.4.5 Napier; 2.4.6 Hastings; 2.4.7 Stratford; 2.4.8 Palmerston North; 2.4.9 Horowhenua; and 2.4.10 Kapiti. 3. We consider that NZME and Fairfax are each other’s closest competitors in these markets and that existing competitors would be unlikely to have sufficient coverage and reach to constrain the merged entity. Further, we consider that entry or expansion into these markets would not be likely to be of sufficient scale to constrain the merged entity. 4. We are therefore not satisfied that clearance should be given. Alternatively, the Applicants submitted that the Commission should authorise the merger, as it would result in significant public benefits to New Zealand. 9 5. We are not satisfied that the merger will result in such a benefit to the public that it should be allowed. We have closely examined the likely benefits and detriments that would arise from the merger. 6. NZME and Fairfax submitted that the merger is a response to the changing media landscape. Print readership and revenue is in decline while the consumption of online news has increased rapidly due to the growth of social media, New Zealand’s high internet penetration and consumer uptake of smart phones and other mobile devices. However, the revenue gained through this increase in online news readership is not currently able to replace the revenue being lost through declining print readership. 7. The Applicants also submitted that the proposed merger would enable them to adapt their businesses to the changing media environment and allow them to continue to invest in journalism and content creation. Their view is the Commission should authorise the proposed merger as it would result in better advertising products at competitive prices, improve the quality of news and information content and create a strong print-invested entity. They also consider there will be no material change to constraint on newspaper subscription prices. 8. We accept the Applicants’ submission that there would be significant cost savings for the parties from the proposed merger. We estimated the difference between the likely benefits and detriments to be at least $14.1 million over five years. While we cannot place a specific value on the cost savings, we accept that on a purely quantitative basis there is likely to be a financial benefit from the merger. However, in our view, the merger would likely result in detriments that cannot be quantified. These detriments arise from the reduction in content quality and plurality (diversity of views) that we consider would be likely to result from this merger. 9. If the merger was to proceed, NZME and Fairfax would have direct control of nearly 90% of daily circulation of all daily newspapers in New Zealand – a recent study suggests that this would be second highest concentration of print media ownership in the world, behind only China.1 10. The merged entity would also have indirect control over much of the national content provided to the remaining independently-owned print media that rely on this content for their regional audiences. Additionally, the Applicants would control New Zealand’s two largest news websites – nzherald.co.nz and stuff.co.nz – which together have a population reach more than four times larger than the next biggest New Zealand news website, and control one of New Zealand’s two largest commercial radio companies. This would be an unprecedented level of media concentration in a well-established liberal democracy. 11. It is unlikely that a media organisation would enter or expand into the print market to compete with the merged entity. It would also not be easy for another party to enter or expand to achieve the same online scale as the Applicants. Incumbency and 1 Noam (2016) Who Owns the World’s media?: Media Concentration and Ownership around the World. Oxford: OUP – the most recent international comparative study of media ownership. 10 the possession of an established news brand is a significant advantage in the media industry. 12. Our view is that competition post-merger would not be sufficiently robust to constrain a multi-media organisation, potentially with a single editorial voice, that would be the largest producer of national, regional and local news by some margin in New Zealand. 13. The detriments resulting from this concentration of media ownership are, in our view, likely to be significant and not in the public interest. In particular, the potential loss of plurality is a factor that weighs heavily in our decision. 14. An international comparison study undertaken this year by the Bertelsmann Foundation assessed plurality across 41 OECD and EU countries and ranked New Zealand last equal with Romania and Turkey. New Zealand already has limited diversity of views across its news media and the proposed merger would substantially reduce this. 15. Diversity of ownership drives competition between media organisations which helps ensure that a wider range of issues and topics are explored and that matters of public interest do not go unreported. Different media owners may also have different editorial priorities or preferences, whether or not driven by commercial or political objectives. 16. As it stands, Fairfax and NZME impose some journalistic constraint on each other on a national level. Both have a ‘digital-first’ strategy, whereby content goes up online first and print offerings are secondary. Competition between the Applicants to generate online content drives increased variety and volume across their own media platforms and the wider domestic news market as other media organisations pick stories up and seek new angles. 17. In our view, the check that NZME and Fairfax provide on each other would be lost under the proposed merger. The merged entity would have discretion as to what it does and does not cover, and the competitive tension that currently exists between NZME and Fairfax, which drives content creation, would be removed. The competitive tension between the Applicants that promotes accurate reporting would similarly be gone. In addition, anyone criticised in an article under the merged entity would have lost access to a direct competitor to tell their side of the story. 18. The media plays a substantial role in shaping public discourse and influencing the political agenda. Different perspectives and voices would ultimately be lost under the proposed merger, which in our view would impact on the quantity and quality of national, regional and local news coverage available in New Zealand. In an industry where the costs of achieving scale are substantial, we consider the loss of plurality that arises from the proposed merger is likely to be significant and potentially irreplaceable. 19. We also do not consider that the two-sided nature of the market would prevent these detriments from occurring. The merged entity would have a large reach that 11 would be appealing to advertisers. Consequently, we do not consider that advertisers would be likely to constrain the merged entity from reducing quality. 20. While we cannot quantify the detriments outlined above in monetary terms, we consider that they are likely to be so fundamental to a well-functioning New Zealand society that they outweigh the organisational efficiencies achieved by the merger. We therefore propose to decline authorisation for the merger. Timeframes for submissions and a conference 21. Having completed our investigation and analysis, we now seek written submissions on the Draft Determination on or before Tuesday 22 November 2016. Crosssubmissions must be received on or before Wednesday 30 November 2016. 22. Section 69B of the Act provides that the Commission may determine to hold a conference prior to making a final determination. We have currently scheduled a conference in respect of this matter for three days in Wellington: Tuesday 6 December to Thursday 8 December 2016. Further information and venue details will be posted on our website shortly. Structure of this draft determination 23. This report is divided into six sections: 23.1 Section 1 provides the executive summary; 23.2 Section 2 outlines the transaction structure, the Commission’s authorisation process and the legal framework we have followed; 23.3 Section 3 outlines the key parties to this merger and the other players in the industry, the Applicants’ rationale for the merger, the current state of the media industry, the Commission’s approach to market definition and the Commission’s assessment of the with and without the merger scenarios; 23.4 Section 4 defines the relevant markets on the advertising side, and considers whether the proposed merger would be likely to result in a substantial lessening of competition in those advertising markets; 23.5 Section 5 defines the relevant markets on the reader side and considers whether the proposed merger would be likely to result in a substantial lessening of competition in those reader markets; 23.6 Section 6 considers the potential vertical and competition effects arising out of the proposed merger; 23.7 Section 7 assesses and balances the benefits and detriments that are expected to result from the proposed merger; and 23.8 Section 8 sets out the Commission’s Draft Determination. 12 Section 2: the Commission’s Approach The proposed merger 24. On 27 May 2016 the Commission registered an application from Wilson & Horton Limited (trading as NZME) and Fairfax New Zealand Limited (Fairfax) and their ultimate parent companies, APN News & Media Limited and Fairfax Media Limited, seeking approval under section 67 of the Commerce Act 1986 (the Act) (the application). The Applicants seek clearance, or in the alternative, authorisation to merge the New Zealand operations of NZME and Fairfax. 25. On 29 June 2016 Wilson & Horton Limited changed its name to NZME Limited following its de-merger from APN News & Media Limited. NZME listed on the NZX on 27 June 2016. 26. NZME and Fairfax seek approval from the Commission to merge their New Zealand operations. 26.1 NZME proposes to acquire all of the shares in Fairfax. 26.2 In exchange, NZME will pay NZ$55 million in cash and issue shares equal to a 41% shareholding in NZME to Fairfax Corporation Pty Ltd (a wholly owned Australian subsidiary of Fairfax Media Limited).2 Fairfax Corporation Pty Ltd will nominate two directors to the Board of NZME on completion of the merger. The Commission’s preliminary view 27. For the reasons explained in this Draft Determination, the Commission’s preliminary view is that it would be likely to decline to grant clearance or authorisation for the proposed merger. The Commission‘s investigation 28. 2 In preparing this Draft Determination, the Commission has obtained information from a wide range of sources. In the course of this process, the Commission has, amongst other actions: 28.1 reviewed the information and analysis in the application, including the economic report submitted by the Applicants’ economic experts NERA, the indicative synergies analysis provided by the Applicants’ accounting experts PwC and the subsequent report submitted by the Applicants from Professor Randal C. Picker; 28.2 posted a public version of the application on the Commission website; Pre and post completion adjustments could result in a greater or lower amount of cash or shares being paid to Fairfax. https://www.nzx.com/companies/NZM/announcements/288557 13 28.3 sought further information and clarification from NZME and Fairfax on a range of subjects; 28.4 interviewed NZME, Fairfax and other interested parties; 28.5 considered submissions and data from NZME and Fairfax, academics, journalists and other interested parties, including economic evidence from Covec on behalf of Television New Zealand (TVNZ); 28.6 engaged BDO Wellington to assist in reviewing the Applicants’ claimed benefits; and 28.7 engaged Dr David Levy3 of the Reuters School of Journalism at the University of Oxford and Mr Robin Foster4 of Communications Chambers to assist in advising on reductions in quality of news content or reductions in media plurality that may arise from the proposed merger. Legal framework 29. Any person who proposes to acquire assets of a business or shares and considers that the acquisition may breach section 47 of the Act can make an application for an authorisation under section 67 of the Act. 30. Section 67(3)(a) of the Act requires the Commission to give clearance for a proposed acquisition if it is satisfied that the proposed acquisition will not have, or would not be likely to have, the effect of substantially lessening competition in a market. If the Commission is not so satisfied, clearance must be declined, although it may still grant an authorisation under section 67(3)(b) of the Act if the Commission is satisfied that “the acquisition will result, or will be likely to result, in such a benefit to the public that it should be permitted.” 31. If the Commission is not satisfied that the acquisition will result, or will be likely to result, in such a benefit to the public that it should be permitted, or the Commission is in doubt5 as to whether it is likely that the acquisition will create a public benefit, it must decline an authorisation under section 67(3)(c). 32. The burden of proof lies with the Applicants to satisfy the Commission on the balance of probabilities that the merger is not likely to substantially lessen competition and if it is likely to do so, that the likely public benefit is such that the Commission should authorise it.6 33. The Commission must decline to give authorisation where it is satisfied that the merger is likely to substantially lessen competition (section 67(3)(a)) and is unlikely to result in such a benefit to the public that it should be permitted (section 67(3)(b)). The Commission must also decline to authorise the merger if it is left in doubt as to 3 4 5 6 http://reutersinstitute.politics.ox.ac.uk/people/dr-david-levy-director http://www.commcham.com/our-members/ Commerce Commission v Woolworths Limited (2008) 12 TCLR 194 (CA) at [98] and [107]. Commerce Commission v Southern Cross Medical Care Society (2001) 10 TCLR 269 (CA) at [7]. 14 7 these matters. The existence of ‘doubt’ corresponds to a failure to exclude a real chance.8 34. Section 67(3) of the Act requires the Commission to issue a decision within 60 working days after the date of registration of the notice, or such other longer period agreed to by the Commission and the Applicant. The Applicants have agreed to an extension of time until 15 March 2017. 35. As outlined above, NZME and Fairfax have sought clearance or, in the alternative, authorisation for their proposed merger. Therefore, the Act requires us to: 35.1 first, determine whether to clear the merger, ie, to decide whether we are satisfied that the merger will not be likely to substantially lessen competition in any market; and 35.2 second, if we do not grant clearance, determine whether to authorise the merger, ie, to decide whether we can be satisfied that the merger will be likely to result in such a benefit to the public that it should be permitted.9 Analysing the competition effects of a merger 36. Our approach to analysing the competition effects of the proposed merger is based on the principles set out in our Mergers and Acquisitions Guidelines,10 and summarised below. We call this part of our analysis our competition assessment. 37. As required by the Act, we assess acquisitions using the substantial lessening of competition (SLC) test. 38. We determine whether a merger is likely to substantially lessen competition in a market by comparing the likely state of competition if the merger proceeds (the scenario with the merger, often referred to as the factual), with the likely state of competition if the merger does not proceed (the scenario without the merger, often referred to as the counterfactual).11 39. We make a pragmatic and commercial assessment of what is likely to occur in the future with and without the merger, based on the information we obtain through our investigation. We take into account factors including market growth and technological changes. 40. A likely counterfactual is one that has a real chance of occurring. This is something more than a possibility, but it does not need to be more likely than not.12 Because a counterfactual need not have a greater than 50% chance of occurring to be ‘likely’, there can be more than one likely counterfactual. 7 8 9 10 11 12 Brambles New Zealand Ltd v Commerce Commission (2003) TCLR 868 (HC) at [55]. Woolworths (CA) above n 5 at [98]. Godfrey Hirst NZ Ltd v Commerce Commission (2011) 9 NZBLC 103,396 (HC) (Godfrey Hirst 1) at [49]. Commerce Commission, Mergers and Acquisitions Guidelines, July 2013. Woolworths (CA) above n 5 at [63]. Ibid at [111]. 15 41. The present state of competition in a market (often referred to as the status quo) may not be the counterfactual that we use for our competition assessment. Where we regard it as likely that the future without the merger looks different from the status quo (because of technological developments or some other reason), we may identify a likely counterfactual that is different from the status quo. 42. A lessening of competition is generally the same as an increase in market power. Market power is the ability to raise price above the price that would exist in a competitive market (the ‘competitive price’), or reduce non-price factors such as quality or service below competitive levels. When a lessening of competition is substantial 43. Only a lessening of competition that is substantial is prohibited. A lessening of competition will be substantial if it is real, of substance, or more than nominal.13 Some courts have used the word ‘material’ to describe a lessening of competition that is substantial.14 44. Consequently, there is no bright line that separates a lessening of competition that is substantial from one that is not. What is substantial is a matter of expert judgement and depends on the facts of each case. Ultimately, we assess whether competition will be substantially lessened by asking whether consumers in the relevant market(s) are likely to be adversely affected in a material way. When a SLC is likely 45. A SLC is ‘likely’ if there is a real and substantial risk, or a real chance, that it will occur. This requires that a SLC is more than a possibility, but does not mean that the effect needs to be more likely than not to occur. 46. Our assessment of the competition effects of the proposed merger is set out in sections 4, 5 and 6. Assessing the public benefit of a merger 47. As described above, where the Commission declines to clear a merger, an applicant can ask the Commission to consider whether to grant an authorisation. The Act requires the Commission to authorise a merger where it is satisfied that the merger is likely to result in such a benefit to the public that it should be permitted. We refer to this as the ‘public benefit’ test. The relevant benefits and detriments are those that arise from the merger.15 48. The Act does not specify how the Commission should ascertain whether such a public benefit is likely to result from the merger. However, our approach to the public benefit test is described in our Authorisation Guidelines.16 13 14 15 16 Woolworths & Ors v Commerce Commission (2008) 8 NZBLC 103,128 (HC) at [137]. Ibid at [129]. Godfrey Hirst 1 above n 9 at [119]. Commerce Commission, Authorisation Guidelines, July 2013. 16 Quantified and unquantified benefits and detriments 49. We are required to have regard to all benefits and detriments – quantified and unquantifiable – and assess whether we are satisfied, on the evidence before us, that the merger results in such a benefit to the public that an authorisation should be granted. 50. In making this assessment, we have regard to the quality of the evidence and make judgements as to the weight to be given to items of evidence. 51. We must quantify benefits and detriments to the extent that it is practicable to do so,17 but this does not mean that we must disregard or give less weight to benefits and detriments that cannot be quantified. Detriments and benefits that cannot be expressed in monetary terms still need to be brought into the balancing exercise.18 The balancing exercise is not a purely arithmetical exercise. 52. The Commission directs itself – as it was put in Ravensdown – to the “task in hand”, and this task differs in each case depending on the markets and facts involved.19 53. Where we cannot quantify a benefit or detriment, we must exercise a qualitative judgement as to the nature and significance of those unquantified benefits or detriments relative to the quantified benefits and detriments.20 What amounts to a public benefit 54. Section 3A of the Act is the only section giving a specific indication of what constitutes a “benefit”. It refers to “efficiencies” that are likely to arise from the merger. However, efficiencies are not the only public benefits that can be counted. 55. A public benefit is:21 … anything of value to the community generally, any contribution to the aims pursued by the society including as one of its principal elements (in the context of trade practices legislation) 22 the achievement of the economic goals of efficiency and progress. 17 18 19 20 21 Telecom Corporation of New Zealand Ltd v Commerce Commission [1992] 3 NZLR 429 (CA) (AMPS-A CA) at 447 and Air New Zealand and Qantas Airways Limited v Commerce Commission (2004) 11 TCLR 347 (Air NZ No 6) at [319]. Ravensdown Corporation Ltd v Commerce Commission High Court, Wellington. Air NZ No 6 above n 17 at [415]. Ravensdown Corporation Ltd v Commerce Commission High Court, Wellington AP 168-96, 16 December 1996 at 48. We also note the comments of the Australian Competition Tribunal in Re Qantas Airways Limited [2004] ACompT 9 at [208]-[209]. The Court remarked that: “All other things being equal, detailed quantification is the best option. However, quantification at all costs is not required by the Act, and has never been sought by the Tribunal. There are diminishing returns to the quantification exercise. Benefits should be quantified only to the extent that the exercise enlightens the Tribunal more than the alternative of qualitative explanation. Where benefits cannot be quantified in monetary terms, they can still be claimed in qualitative terms. The authorisation test is, after all, a balancing exercise that requires judgment over a wide range of tangible and intangible factors. The final result will depend on the relative weight assigned to each of these factors”. Godfrey Hirst 1 above n 9 at [102]. Authorisation Guidelines, above n 16 at [35]. 17 23 56. As the High Court stated in AMPS-A, efficiency considerations (including detriments) “do not exhaust society’s interest in the business conduct the subject of the Commerce Act”. 57. Accordingly, we regard a public benefit as any gain to the public of New Zealand that would result from the proposed merger regardless of the market in which that benefit occurs or to whom in New Zealand it benefits.24 58. The Applicants claim quantified benefits in the form of cost-savings that would result from the merger. They also claim unquantified benefits in terms of an enhanced advertising alternative for advertisers, increase in journalistic quality, and enhanced content plurality. 59. Our assessment of the Applicants’ benefit claims is set out in the public benefits section of this Determination. Detriments that are relevant to our assessment 60. As set out in our Authorisation Guidelines, in assessing detriments we consider anticompetitive detriments that arise in the market(s) where we find a lessening of competition (whether substantial or otherwise).25 61. The detriments that arise from a lessening of competition include allocative efficiency detriments (welfare losses from increased prices/reduced quality), productive efficiency losses (higher costs over time), and dynamic efficiency losses (reduced incentive to innovate). In some circumstances wealth transfers from New Zealanders to non-New Zealanders may also give rise to a detriment to New Zealand.26 62. In this case, the assessable detriments also include: 22 23 24 25 26 62.1 reductions in the quality of the news content provided; and 62.2 reductions in media plurality. See Air NZ No 6 above n 17 at [319] and Telecom Corporation of New Zealand Ltd v Commerce Commission (1991) 4 TCLR 473 (HC) (AMPS-A HC) at 527-530 quoting Queensland Co-operative Milling Association Ltd (1976) ATPR 40-012 at 12,242 and In Re Rural Traders Co-operative (WA) Ltd (1979) ATPR 40-110 at 18,123. AMPS-A HC above n 22 at 528. An example of this is Air NZ No 6 above n 17. In that case, the High Court considered the Commission’s assessment of increased tourism as a benefit. That tourism benefit arose from increased numbers of New Zealand tourists as a result of them being deterred from travelling overseas by increased prices. While the Commission recognised that ‘benefit’, it netted off the detriment from fewer inbound visitors. The actual benefit figure ranged from a negative (ie, a detriment) to a positive benefit. There was no suggestion that the negative figure of the detriment from reduced inbound tourists had to be ignored. The High Court did not disagree with the Commission’s approach. Authorisation Guidelines, above n 16 at [38]. Authorisation Guidelines, above n 16 at [53]-[55]. The Commission’s approach was endorsed by the High Court in Godfrey Hirst NZ Ltd v Commerce Commission [2016] 3 NZLR 645 (Godfrey Hirst 2). This decision was appealed to the Court of Appeal, and was heard in September. The Court of Appeal has yet to release its judgment. 18 63. We treat quality detriments as falling within the identified reader and advertising markets. 64. Plurality effects are treated as extending beyond these markets, and as having an impact on New Zealand society more generally. For example, submitters and expert media advisers identified media plurality as an important contributor to democracy and Government accountability, and as an issue of public importance irrespective of the type of media coverage (reporting or opinion). 65. We recognise that a plurality detriment would not necessarily be the same as the types of inefficiencies we usually consider. However, in our view, media plurality is clearly relevant to the public interest of New Zealand, and is therefore an important consideration in our assessment of this merger. We consider plurality detriments are the sort of detriments the High Court was referring to in AMPS-A, when it noted that the concept of benefit to the public allows for netting out of any detriments to the public from the proposed acquisition.27 66. We therefore reject the Applicants’ submission that a reduction in plurality cannot be accommodated within the authorisation analysis envisaged under the Act.28 We also note that such a view is incompatible with the Applicants’ claim that the merger would increase content plurality to the benefit of New Zealand. Relevance of plurality as an unquantifiable detriment 67. The Applicants also argue that to the extent plurality concerns are relevant, those concerns should “be robustly tested against the facts, and if they cannot be properly evidenced then they should be given little weight as unquantified, subjective ‘detriments’ that will be outweighed by robust quantified benefits”.29 68. We agree with the Applicants that the impact on plurality should be based on the available evidence. We also accept that plurality and its value are, by their nature, unquantifiable. However, as described above, that does not mean that impacts on plurality should be disregarded, or given less weight. This is not consistent with the High Court’s approach in Air NZ (No 6). 69. We therefore assess the impact of the proposed merger on plurality and compare that impact to the other benefits and detriments arising from the merger. 27 28 29 AMPS-A HC above n 22 at 528. NZME/Fairfax Cross submission on TVNZ and MediaWorks SoPI submissions (10 October 2016) at [50(e)]. NZME/Fairfax Response to Submissions on SoPI (29 July 2016) at [6]. 19 Section 3: Key Parties and Competition Issues The merging parties NZME 70. 71. 30 31 32 33 NZME is a New Zealand media and entertainment business that offers radio broadcasting (Newstalk ZB, ZM, Radio Hauraki, etc), print publications (The New Zealand Herald, Herald on Sunday, Weekend Herald, etc), digital publications (such as www.nzherald.co.nz) and e-commerce services (such as www.grabone.co.nz).30 NZME also has an ownership interest in the:31 70.1 [ 70.2 The Beacon Printing & Publishing Company Ltd (21%); and 70.3 The Wairoa Star Ltd (40.41%). ]; A summary of NZME’s publications and brands is outlined below.32 71.1 six daily papers in the North Island; 71.2 two paid weekly papers; 71.3 11 online versions of newspaper websites; 71.4 two life and style websites; 71.5 10 radio station websites; 71.6 16 other websites;33 71.7 six magazines; 71.8 nine radio stations; 71.9 23 community publications: 71.9.1 two in Northland; 71.9.2 two in Coromandel; 71.9.3 three in regional Waikato; 71.9.4 six in Bay of Plenty/Rotorua; Application at [2.8]. Application at [2.9]. A full list can be found at Appendix One of the application. For example, WatchMe (www.watchme.co.nz) (NZME’s Video On Demand platform) and GrabOne (www.grabone.co.nz). 20 71.9.5 one in Taranaki; 71.9.6 four in Hawke’s Bay; 71.9.7 one in Manawatu; 71.9.8 one in Wanganui; and 71.9.9 three in Wellington. Fairfax 72. Fairfax is the New Zealand subsidiary of Fairfax Media Limited, which is a media company operating in New Zealand and Australia. 73. In New Zealand Fairfax has a variety of print publications, websites, tablet and smartphone apps. Its mastheads include The Dominion Post, The Press and The Sunday Star-Times. Fairfax’s primary online offering is stuff.co.nz.34 74. Amongst other companies, Fairfax also owns:35 74.1 50% of Times Newspapers Limited; the publisher of the Howick and Pakuranga Times and Botany and Ormiston Times; 74.2 44.99% of Neighbourly Limited (www.neighbourly.co.nz). Neighbourly is a private neighbourhood website that allows neighbours to talk and share online. 75. 34 35 36 A summary of Fairfax’s publications and brands is outlined below.36 75.1 nine daily papers – four in the North Island and five in the South Island; 75.2 three paid weekly papers; 75.3 seven websites; 75.4 10 magazines; 75.5 62 community publications: 75.5.1 three in Northland; 75.5.2 13 in Auckland (including Waiheke); 75.5.3 nine in Waikato; 75.5.4 one in Rotorua; Application at [2.3]-[2.4]. Application at [2.5]. A full list can be found at Appendix Two of the application. 21 75.5.5 two in Taranaki; 75.5.6 two in Hawke’s Bay; 75.5.7 four in Manawatu; 75.5.8 seven in Wellington; 75.5.9 two in Nelson; 75.5.10 seven in Marlborough; 75.5.11 six in Canterbury; 75.5.12 three in Otago; and 75.5.13 three in Southland. Other industry participants 76. The Applicants list a number of parties whom they consider to be competitors in appendix eight of the application. The appendix lists weekly and regional print publications, magazines and online and digital publications. We set out a summary of some of the main competitors below. Television New Zealand (TVNZ) 77. TVNZ is New Zealand’s free-to-view video content provider. TVNZ is owned by the Crown but it is not a public broadcaster. TVNZ operates as a self-funded, commercial entity by virtue of the Television New Zealand Act 2003. Its predominant source of revenue is from advertising.37 78. TVNZ has two main broadcasting channels, TVNZ 1 and TVNZ 2, as well its Duke channel and its TVNZ OnDemand and 1 News Now online channels.38 MediaWorks Holdings Limited (MediaWorks) 79. MediaWorks TV, Radio and Digital (MediaWorks) is New Zealand’s largest crossplatform media company.39 80. MediaWorks TV operates two free-to-air channels, TV3 and Edge TV and is a joint venture owner of free-to-air channel Bravo. 81. MediaWorks Radio operates nine commercial radio stations and its entertainment and news brands include Newshub, The Rock, More FM, The Edge, Mai FM, George FM and RadioLIVE. 37 38 39 Submission from TVNZ to the Commerce Commission (14 September 2016) at [1.1]. Submission from TVNZ to the Commerce Commission (14 September 2016) at [1.1]. Submission from MediaWorks to the Commerce Commission (30 September 2016) at [1.2]. 22 Radio New Zealand (RNZ) 82. RNZ is a Crown entity established under the Radio New Zealand Act 1995. RNZ provides listeners with independent radio programmes in accordance with the RNZ Charter. 83. RNZ broadcasts over three nationwide networks: Radio New Zealand National, Radio New Zealand Concert and the AM network which relays Parliamentary proceedings.40 Allied Press Limited (Allied Press) 84. Allied Press is a privately-owned Dunedin-based media company. It has interests in daily and community newspapers, a regional television station, internet websites and commercial printing operations.41 85. Allied Press publishes Otago Daily Times. Allied Press also holds a majority interest in Greymouth Evening Star and has a range of community and farming newspapers in the South Island.42 KPEX Limited (KPEX) 86. NZME and Fairfax are joint venture owners, alongside TVNZ and MediaWorks, in KPEX Limited.43 KPEX operates as an ad exchange to sell remnant digital advertising inventory across qualifying publishers' online and mobile websites. Each of the joint venture parties holds a 25% share in KPEX. Industry background 87. We set out below some of the key changes in technology and consumer demand that are taking place in the print and online markets for news and information content. 88. This background provides the context for our subsequent assessment of the competitive effects and the benefits and detriments associated with the proposed merger. Print is declining 89. 40 41 42 43 44 The media industry has been subject to significant change over the last five years, due to changes in technology and consumer demand. Print revenue and readership has been in decline, as has advertising revenue. In its Global entertainment and media outlook survey 2016-2020, PwC suggested that newspaper circulation is on track to fall by almost 50% in the decade to 2020.44 http://www.radionz.co.nz/about Submission from Allied Press to the Commerce Commission (1 July 2016) at [3]. Ibid at [4]-[5]. Short for Kiwi Premium Advertising Exchange. PwC Global entertainment and media outlook 2016-2020. 23 90. The Applicants submitted that the initial disruption to newspaper publishers’ revenue began a decade ago with the decline in classified revenue,45 and has now spread to all aspects of print media revenue – both circulation/subscription revenue and display advertising revenue.46 The Applicants’ advertising revenues are still earned predominantly from print. The move to online consumption of news and information 91. Traditional media models in the form of print newspapers are facing increasing challenges from other modes of delivery. In contrast to the print media industry, digital media has been growing, with many consumers switching their news consumption from print to digital media, or supplementing their print news consumption with digital news. 92. The Applicants submit that more New Zealand readers are choosing to read their news online on desktops, tablets and smartphones, and, in particular, via social media platforms.47 Consequently, advertisers are also choosing different media platforms to target potential customers. 93. The online platforms through which readers can access content created (or syndicated) by the Applicants are:48 93.1 publishers’ websites; 93.2 social media platforms, eg, Facebook (including Instagram and Whatsapp), Twitter, YouTube; 93.3 search engines, eg, Google and Yahoo; and 93.4 apps on smartphones and tablets. A digital-first strategy 94. 45 46 47 48 49 50 To cater to the change in reader demand, the Applicants have reconfigured their businesses. The Applicants now produce content for their online channel first, a ‘digital-first’ strategy, with an increasing focus on video and audio in addition to written content.49 This means that the Applicants prioritise the creation of digital editorial content instead of traditional daily newspaper formats. This content is then uploaded to the Applicants’ respective websites and social media accounts continuously throughout the day. Select content is then converted to a print format for inclusion in the NZME and Fairfax print publications.50 Application at [6.5]. Application at [6.8]. Application at [6.8]. Application at [9.6]-[9.11]. Application at [6.46]. Application at [9.3]. 24 Inevitable consolidation 95. The Applicants submitted that the above global trends are leading to consolidation and rationalisation of publishing businesses51 and that “… at some point in the future it is likely that some newspapers will no longer be printed every day and there may come a time where certain publications are only available online”.52 The Applicants’ rationale for the proposed merger 96. The Applicants view the proposed merger as a response to this changing media landscape.53 The Applicants pointed to declining print readership and revenue, as well as competition for online news (and information) becoming highly competitive with the growth of social media and New Zealand’s high internet penetration.54 55 97. Against this backdrop, the Applicants submitted that the proposed merger would allow them to continue to invest in journalism and content, but at the same time adapt their businesses to provide advertising customers with a “more targeted, lower cost, data rich, higher return on investment advertising offering”.56 98. In summary, the Applicants submitted that the Commission should clear the proposed merger because:57 99. 51 52 53 54 55 56 57 58 98.1 the merged entity would be able to provide advertisers with a better product at a competitive price point; 98.2 there will be no material change to constraints on newspaper subscription prices as a result of the merger; 98.3 the merger will improve the quality of news and information content produced; and 98.4 the merger will create a strong print-invested entity. In the alternative, the Applicants submitted that the Commission should authorise the merger. They submitted that the proposed merger would result in substantial public benefits to New Zealand, in the form of $[ ] annual cost synergies as well as providing an enhanced advertising alternative for advertisers, an increase in journalistic quality, and enhanced content plurality.58 Application at [6.43]. Application at [6.17]. Application at [1.2]. Application at [1.2]. The Applicants define news and information in their application to include local and regional news, life and style, sport, and entertainment content. The Commission’s analysis refers to “news”. Application at [1.6]. Application [1.12]. Application at 20.20. 25 100. The Commission has tested these benefits as part of its investigation. The Commission has also considered likely detriments that could arise as the result of the proposed merger. With the merger 101. As noted above, NZME proposes to acquire the business and assets of Fairfax. The proposed merger would result in overlap between NZME and Fairfax:59 101.1 in the provision of national online news and information services, through their respective websites, stuff.co.nz and nzherald.co.nz; 101.2 in Hawke’s Bay and Waikato for their respective daily newspapers: 101.2.1 Hawke’s Bay Today (NZME) and The Dominion Post (Fairfax) in the Hawke’s Bay region; and 101.2.2 The New Zealand Herald (NZME) and Waikato Times (Fairfax) in the Waikato region; 101.3 Sunday newspapers (paid weeklies) in the North Island; and 101.4 community newspapers in: 101.4.1 Whangarei; 101.4.2 Hamilton; 101.4.3 Rotorua; 101.4.4 Taupo; 101.4.5 Napier; 101.4.6 Hastings; 101.4.7 Stratford; 101.4.8 Palmerston North; 101.4.9 Horowhenua; and 101.4.10 Kapiti. 102. 59 The Applicants submitted that the merged entity would continue to develop and invest in new and innovative digital opportunities to maintain and grow its share of The Commission notes that on 25 October 2016 APN News and Media Limited announced that it had acquired the remaining 50% interest in Adshel Street Furniture Pty Limited. The Commission notes that NZME has the option of purchasing from APN the New Zealand business and assets of Adshel, an outdoor advertising and street furniture company. The Commission has not considered this possible purchase as part of its assessment of the with and without the merger scenarios in this draft determination. 26 digital revenue, which is essential given that some newspapers will no longer be printed every day.60 The Applicants also submitted that rationalisation of their respective print assets would enable the entities to extend the lifespan of their respective print publications.61 Without the merger 103. The Applicants presented their competition analysis and net benefit analysis on the basis that, without the merger, in the short term, NZME and Fairfax would continue to operate as separate entities.62 104. However, the Applicants also submitted that it is likely that each entity’s “[ ]”.63 105. In support of this statement, the Applicants supplied internal documents which note that [ ]64 [ ].65 66 106. We accept that the media industry has been subject to significant change in recent years due to changes in technology and consumer demand. However, it is not clear to us when and to what extent the Applicants will rationalise their print publications without the merger. 107. On the evidence we have, we cannot predict with any degree of precision what level of rationalisation would occur without the merger. We accept that the status quo will not necessarily prevail, but we cannot exclude the real chance that a scenario close to the status quo will prevail in the short term. Therefore, we have taken a pragmatic approach by conducting our competition analysis on the basis of the status quo scenario as we consider it to be the most conservative from a competition point of view. 108. We do not consider that rationalisation beyond the status quo would be likely to change our conclusions on the application of the substantial lessening of competition test to this merger, at least in the short term. 60 61 62 63 64 65 66 Application at [1.10]. The Applicants did not say which publications this may refer to. NZME/Fairfax cross-submission on TVNZ’s submission (14 September 2016) and MediaWorks’ submission (30 September 2016) dated 10 October 2016 at [54]. Application at [13.1]. The Applicants did not provide a specific timeframe for the short term. Application at [13.2]. NZME response to information request Annex 40 - NZME Strategy Day 22 March 2016. Fairfax response to information request Annex 48 - The Kiwi Way Koru: Redefining. Reimaging. Renewal April 2016. NZME response to information request Annex 39 - FY15 Overview & FY16 Topline Budget – Pre reading pack 13 October 2015. 27 109. However, we recognise that this approach will include some cost-savings that may occur without the merger. In recognition that the status quo may not necessarily prevail, and to assist us in determining the benefits and detriments of the proposed merger, we have considered what the detriments would be if the Applicants immediately ceased publishing all print products (except community newspapers)67 and shifted to ‘digital only’ businesses. As with the status quo, we do not consider this scenario to be likely, but we have used this scenario to assist with our quantitative analysis. The relevant markets How we define markets 110. Market definition is a tool that helps identify and assess the close competitive constraints the merged entity would face. Determining the relevant market requires us to judge whether, for example, two products are sufficiently close substitutes as a matter of fact and commercial common sense to fall within the same market. 111. We define markets in the way that best isolates the key competition issues that arise from a proposed acquisition.68 In many cases this may not require us to precisely define the boundaries of a market. What matters is that we consider all relevant competitive constraints, and the extent of those constraints. For that reason, we also consider products which fall outside the market but which still impose some degree of competitive constraint on the merged entity. 112. The standard means to define the market is to use the “hypothetical monopolist test” (HMT).69 The HMT asks whether a hypothetical sole supplier of a group of products could profitably raise prices by a small, yet significant, non-transitory (SSNIP) amount. If it could impose the SSNIP, the HMT is satisfied and a market is defined. If it could not, then the market is widened to include the next best substitute and the process repeated. The process continues until a group of products that satisfies the HMT is found. 113. Whether a SSNIP could be profitably imposed depends on the degree of demandside and supply-side substitution that would occur. Demand-side substitution is where customers switch to other products outside the candidate market in response to a price increase. Supply-side substitution is where rival firms offering products outside the candidate market could easily, profitably and quickly switch their production processes to supply those products in the candidate market, in response 67 We have assumed that the merged entity would continue to produce community newspapers in the ‘digital only’ scenario. While community newspaper advertising exhibits flat to declining revenues, information from the Applicants indicates that [ ], see application, Appendix 13 at [5] and [13-14]. We also note that many of NZME’s community newspapers are reported as [ ], see Annex 38, NZME F16 Regional Strategy Summary (February 2016). 68 Commerce Commission, Mergers and Acquisitions Guidelines, above n 10 at [3.10] - [3.12]. Ibid at [3.17] - [3.22]. 69 28 to a price increase. What matters is whether demand and supply-side substitution together is sufficient to defeat the SSNIP. 114. Where relevant, we also examine the ability of suppliers to discriminate between customers because their competitive alternatives vary. Characteristics of two-sided markets 115. 116. Fairfax and NZME operate a number of multi-sided media platforms in the form of hardcopy publications (newspapers, community papers, magazines and other publications), websites, and radio stations. Each of these platforms typically provides services to consumers in two markets: 115.1 an audience for advertisers; and 115.2 content for readers. On each side of these platforms there may be more than one specific separate market. For instance, there may be different customer markets amongst advertisers depending on their specific requirements, target audience, and ability to use other platforms as effective substitutes (eg Facebook, Google). Markets that could be affected by the proposed merger 117. Because the Applicants operate in a two-sided market, their success depends on their ability to attract both readers and advertisers to their publications (ie, all media the Applicants produce). In other words, revenue the Applicants earn from advertisers influences the quantity and quality of the content, which in turn impacts readership of a publication. 118. We outline specific market definitions in the relevant advertising and reader market sections below. However, in summary, our competition analysis focuses on those markets where the Applicants overlap in the provision of products or services, in particular advertising inventory in print and digital news and online and newspaper reader markets. The Applicants provide other products and services, such as NZME’s radio stations, Fairfax’s Neighbourly website and magazines and event management services. These markets are not further considered in this draft determination given the limited overlap in these areas. Advertising markets 119. Contingent on the level of competitive constraint provided by other providers of advertising inventory, the merged entity could use any market power that it had to raise advertising prices for its newspapers and online news websites. 120. As part of our analysis, we also specifically considered whether the merged entity could use any market power that it had to raise the price of certain forms of digital advertising, such as homepage takeovers, mobile interstitials and native advertising. 29 Reader markets 121. The proposed merger could give rise to a substantial lessening of competition in reader markets through increased prices and/or decreased quality of content. The merged entity could use any market power gained as a result of the merger to increase the subscription or cover prices of its newspapers, or choose to introduce a paid subscription model to its online products. 122. We also considered whether the merged entity could decrease the quality of its news content. We measured this by considering the possible impact of the proposed merger on the volume, variety, accuracy and objectivity of the news the Applicants produce. Foreclose of rival publishers 123. The proposed merger could give the merged entity the incentive and ability to foreclose rival independent newspaper publishers from NZME’s national news service either by worsening their terms of access for NZME’s News Service or by NZME withdrawing from the service altogether. As it is unlikely that the independents could obtain their national news from any other source, the quality of their publications could be eroded such that they become uncompetitive and are forced to exit downstream newspaper markets or are acquired by the merged entity. 124. The proposed merger may also afford the merged entity with the incentive and ability to foreclose rival independent newspaper publishers from downstream newspaper markets by ceasing its participation in News Works.70 Instead, the merged entity could sell its own national advertising campaigns and not book advertisements in the independents’ publications, affecting the independents’ viability through the loss of valuable advertising revenue. Bundling 125. 70 Post-merger, the merged entity may be able to offer a print, radio and/or online advertising bundle that could potentially foreclose other advertising sellers (such as radio stations, independent newspapers, and websites). Such a disadvantage could render rival advertising sellers less competitively effective due to reduced advertising revenue. News Works is the trading name of the New Zealand Press Association Limited’s (NZPA) commercial entity. News Works books national advertising campaigns in all metropolitan, national, regional and Sunday newspapers. 30 Section 4: Advertising Markets 126. Post-merger, NZME and Fairfax would cease being independent competitors for the supply of advertising inventory. If the merging parties are each other’s closest competitors in these markets, then the merger could provide the merged entity with increased market power. 127. Depending on the level of competitive constraint provided by other suppliers of advertising inventory, the merged entity could use any enhanced market power to raise the price of its advertising products. Market definition in advertising markets 128. To determine whether a substantial lessening of competition would be likely to arise in advertising markets as a result of the proposed merger, as a first step the Commission assessed: 128.1 whether it should analyse different media platforms as part of a wider advertising market; or 128.2 whether there are separate relevant markets for the different advertising platforms on which advertising is delivered, that is, different relevant markets for print advertising and digital advertising.71 Applicants’ view 129. The Applicants submitted that current market dynamics support a broad view of the advertising market, namely, a New Zealand market for the provision of print and online advertising services to advertisers. The Applicants suggested that this market is also likely to include other advertising sources, such as TV, radio and billboards.72 Previous Commission decisions – advertising markets 130. The Commission has previously considered advertising markets in a number of decisions. Consistent with the Applicants’ view, they submitted that the approach taken in Times Media73 is no longer appropriate. 131. In Times Media in 2005 the Commission cleared Fairfax’s acquisition of three overlapping community newspapers in the Rodney district. The Commission defined the relevant market to be print advertising services in the Rodney region. 132. In coming to this view, the Commission considered the extent to which other forms of print advertising, such as advertising flyers, were substitutable for advertising in community newspapers. The Commission was of the view that community newspapers and advertising flyers both had advantages and disadvantages for 71 72 73 And also relevant markets for advertising delivered by radio, television and outdoor. Application at [14]. Fairfax New Zealand Limited and Times Media Group Limited (Commerce Commission Decision 561, 14 October 2005). 31 advertisers but that they were sufficiently close substitutes for enough advertisers to be considered part of the same market. 133. The Commission also assessed whether or not other forms of media (namely the internet, radio, TV and billboards) were viewed as substitutes to print advertising. The Commission found that the different forms of media were compliments to print advertising (and each other) rather than substitutes. 134. The Commission considered that radio advertising was not in the same product market as print advertising as advertisers tended to utilise the two forms of media for different purposes. Radio was used to promote brand and simple messages and print was generally used to communicate details or display various prices of different stock. 135. For internet advertising, the Commission found that the internet was a constraint on classified advertising in newspapers, but that it only affected display advertising in a minor way. The Commission concluded that it was appropriate to define print advertising and internet advertising as complements, rather than substitutes. 136. Similarly, in older decisions involving newspapers, the Commission defined a wide print advertising market. In Power Plant Productions,74 the Commission defined the advertising market as print media advertising in the Hawke’s Bay region. In Marlborough Express, the Commission defined the market as the market for the provision of advertising services in the print medium in the Marlborough region.75 137. In Power Plant Productions and Marlborough Express, the Commission also held that there are no significant barriers to entry and expansion for the publishing of a community newspaper. 138. More recently, the Commission has taken a different approach with respect to mergers involving print advertising. In Bauer rather than defining a specific market the Commission considered the effect that the merger would have on the sale of magazine advertising to advertisers and focused on the constraint imposed by other magazines, newspaper-inserted magazines and the countervailing power of advertisers.76 Advertisers’ views 139. 74 75 76 Generally, the goal of advertising is to engage with consumers and persuade them to change their behaviour in some way. Advertisers view the different media platforms as achieving this in different ways: by engaging with specific groups or segments of the population, by describing the price and attributes of a product or service, or by imparting a sense of trustworthiness in the product or service through the environment in which it is advertised. Wilson and Horton Limited and Power Plant Productions Limited (Commerce Commission Decision 445, 7 December 2001). Independent Newspapers Limited and The Marlborough Express Newspaper Company Limited (Commerce Commission Decision 329, 2 October 1998). Bauer Media Group (NZ) LP and APN Specialist Publications NZ Limited [2014] NZCC 1. 32 140. The majority of the advertisers spoken to by the Commission split their advertising spend across a range of media, with most spending on a mix of digital, print, television and radio advertising. For each campaign, they consider which platforms best suit the message and the goal of the advertising campaign and seek bids or proposals from media firms within these platforms. 141. Normally, each of these platforms is purchased separately, and it is not uncommon for a campaign to have a number of different media vendors. Advertising suppliers generally price their products on a ‘cost-per-thousand’ basis. This means the cost for having the advertisement viewed or seen by a thousand people. 142. Industry participants informed the Commission that different types of media are used by advertisers to target different audiences or convey different types of messages. The different types of media also have different advantages and disadvantages in terms of reach, conversion and data utilisation. 143. Consistent with this, the majority of advertisers spoken to by the Commission use a mix of media types for their advertising to ensure that they are able to capture the appropriate audience for the individual campaign. Different advertisers will also have a preference for particular media due to past experience, or the particular industry in which they operate. 144. Advertisers, particularly those that are sophisticated purchasers or that use an advertising agency, will monitor the performance of the advertising product and it is common for the type of media used in a campaign to change as the campaign progresses in response to individual media’s performance or as part of the lifecycle of an advertising campaign. 145. One large advertising agency considered that for digital, its key benefit is the ability to target specific customer types through the use of data and to track the performance of the campaign on a day-to-day basis.77 146. Another advertising agency considered that mainstream media such as TV and print is more generic and broader than digital, which provides a more versatile and dynamic solution to advertisers. They considered that the benefits of print advertising were that it provides a combination of permanence and size and allows the advertiser to talk about the product and the price.78 147. [ ] stated that print advertising was more effective at targeting older customers.79 [ ] stated that print advertising is viewed by consumers as a more trusted source of information and the message gains some legitimacy from being in a print environment. For this reason, [ ] uses print advertising if it needs to convey a compliance message (for example informing customers of price increases or changes to terms of service). 77 78 79 Commerce Commission interview with PHD (28 June 2016). Commerce Commission interview with Draft FCB (27 June 2016). Commerce Commission interview with [ ] (22 August 2016). 33 148. In general, advertisers spoken to by the Commission had reduced their spend on newspaper advertising in recent years due to the decline in newspaper readership, as well as an increase in spend on digital platforms. The Commission’s view 149. Notwithstanding the conclusions reached in its previous decisions, the Commission examined the extent to which the various forms of advertising are seen as substitutes by advertisers to assess whether or not it is appropriate to view the different forms of advertising as being in separate product markets. 150. Advertisers appear to adjust their advertising spend from one platform to another in response to consumer engagement and the extent to which the platform is performing, in terms of changing consumer behaviour through increased sales or customer interaction. This may indicate that advertisers view the different platforms as part of a wider advertising market. 151. However, advertisers different methods of engaging with the consumer, advertisers’ views of the relative strengths of the different platforms and the extent to which many different forms of advertising media are selected as part of a wider advertising strategy, indicate that individual platforms could constitute separate product markets. 152. The Commission therefore considers that suppliers of advertising inventory on the same platform are likely to provide a stronger price and quality constraining effect on each other than those on a different platform. That is, print publications that are similar in geographic and demographic appeal are more likely to be closer competitors than a non-print platform even if that platform may have similar geographic and demographic appeal. Similarly, digital publications are likely to be a closer constraint on each other. 153. Consequently, for the purposes of considering the proposed merger, we have proceeded on the basis that there are separate relevant markets for print advertising and digital advertising.80 154. We have also proceeded on the basis that within each different platform market, there are separate relevant markets due to geographic or other considerations. We consider these relevant markets to be: 154.1 for digital advertising, including a separate market for the supply of premium digital advertising (encompassing homepage takeovers, mobile interstitials and native advertising); and 154.2 for print advertising, including separate markets for the supply of print advertising in: 154.2.1 80 Sunday newspapers; And also relevant markets for advertising delivered by radio, television and outdoor. 34 154.2.2 metropolitan daily newspapers in the Waikato and Hawke’s Bay regions; and 154.2.3 community newspapers in the relevant areas identified at paragraph 101.4 above. 155. These markets are discussed in more detail in the sections on digital, print and community advertising below. 156. We have conducted our competition analysis on the basis of these separate relevant markets. In doing so, as explained below, we recognise that within the advertising industry there will be a range of different forms of advertising that will be substitutes for different advertisers at different times and, in considering whether competition will be lessened in the relevant markets, we will have regard to this. Digital advertising markets 157. Digital or online advertising is a broad category that encompasses all advertising delivered over the internet, via a consumer’s computer, mobile phone or tablet. The advertising can be audio, video, image or text-based. Market definition 158. Industry participants outlined that within the broader category of online advertising there are a number of different types of advertising that provide different outcomes to the advertiser. The key categories identified by advertisers and publishers are search, social media, mobile, video, display, programmatic, mobile interstitial, homepage takeovers, and native advertising.81 159. Within the digital advertising market, there is a range of products that are utilised by advertisers for specific reasons and to achieve specific outcomes. For example, an advertiser may choose a social media campaign to engage with a younger audience, a homepage takeover to publicise a specific event, or programmatic advertising to maximise the number of eyeballs with the minimum amount of spend. 160. [ ] categorised the different types of digital advertising into two major types. The first type included advertising that is managed and purchased through its advertising agency. This type includes programmatic advertising, standard display advertising, social media and search. These advertising types tend to have a standard format, a standard measurement system and also a standard price.82 161. The second type involves a more in-depth relationship with the publisher and is built around specific campaigns or parts of the business. The types of products [ ] places in this category are native advertising and homepage takeovers. 81 82 Commerce Commission interviews with FCB (27 June 2016), OMD (27 June 2016), and PHD (28 June 2016). Commerce Commission interview with [ ] (05 September 2016). 35 162. The extent to which different media companies and publishers are active within these particular categories varies. For example, within the search category, the choice would typically be confined to Google, while in social media the choice would be between Facebook, Instagram and Twitter.83 163. We consider that certain types of digital advertising (such as homepage takeovers and mobile interstitials) can be distinguished from other forms of digital advertising in a number of ways. They have a high impact as they occupy a large amount of screen space and (in the case of mobile interstitials) must be closed by the user before the intended webpage can be viewed. 164. These types of digital advertising are also different in that they are limited in supply. For display, video, search and social media advertising, there is (theoretically) a limitless supply as a different advertising impression can be delivered each and every time a page is loaded. 165. By contrast, homepage takeovers and mobile interstitial advertisements are typically only available to one advertiser per day and are therefore valued by advertisers as they enable the guaranteed placement of the advertisement on a particular day. They are also typically sold on a per day basis (as opposed to other forms of digital advertising which is sold on a cost-per-thousand or cost per impression basis). 166. Advertisers considered these types of advertising valuable as they tie the scarcity value with high reach and impact.84 167. Also, there is value in being able to utilise pre-booked advertising such as this to target key events such as movie releases and changes to interest rates, or to target specific shopping dates (as is used by [ ]). 168. For the purposes of considering the competition effects of the proposed merger, we have proceeded on the basis that there are separate markets for digital advertising (encompassing video, search, social media, display and programmatic advertising) and for premium digital advertising (encompassing homepage takeovers, mobile interstitials, and native advertising). 169. In carrying out our competition analysis, we recognise that different types of digital advertising will be used by advertisers at different times. We have regard to this when considering whether competition will be lessened in the relevant markets. Competition analysis – digital advertising 170. We considered whether the proposed merger is likely to result in a substantial lessening of competition in the market for digital advertising. 171. NZME and Fairfax are both large suppliers of a wide range of digital advertising inventory. Post-merger, they would no longer be independent suppliers. If there is currently close competition between NZME and Fairfax for the supply of digital 83 84 Commerce Commission interview with Starcom (28 June 2016). Commerce Commission interview with PHD (28 June 2016). 36 advertising, then the merger could provide the merged entity with increased market power. 172. Depending on the level of constraint provided by other suppliers of digital advertising inventory, the merged entity could use any enhanced market power to increase the price of its advertising products. 173. Any price increases could be applied across all advertisers, or could potentially be targeted at those advertisers that face fewer or less attractive alternatives for digital advertising. For example, to those advertisers requiring premium digital advertising products which only the merging parties provide. 174. Based on feedback from advertisers, we separately considered the extent of competition in the narrower premium digital advertising market. Specifically, we considered whether the merged entity would be likely to have an incentive to exercise any market power it may obtain to increase the prices of premium digital advertising. 175. We have also considered how the proposed merger would impact on the KPEX programmatic exchange and whether, as a result of the merging parties obtaining a higher shareholding in KPEX, they would be able to operate the exchange to the detriment of the other owners (being TVNZ and MediaWorks). Applicants’ views 176. The Applicants asserted that the merged entity would not be able to raise prices above competitive levels in a converged New Zealand market. This is because of strong competition from, for example, Google and Facebook. The Applicants also submitted that New Zealand-focused advertising can be provided through a plethora of advertising options, including print, online, TV, radio, cinema, and outdoor.85 177. Specifically, in the online advertising market, the Applicants submitted that the market for digital advertising is highly competitive, with the merged entity representing less than 15% of all digital advertising inventory sales available to advertisers in New Zealand.86 Others’ views 178. The Commission interviewed a number of interested parties, including the largest advertising agencies in New Zealand, and a cross-section of the Applicants’ digital advertising customers. 179. Most of NZME’s and Fairfax’s customers in the digital advertising market tend to be large and sophisticated buyers that use rich data analytics to move advertising spend around in response to success or price. 85 86 Application at [14.11] - [14.12]. Ibid at [14.16]. 37 The extent to which NZME and Fairfax compete 180. The views expressed by advertising agencies and customers about the extent to which the Applicants compete in the digital advertising market varied depending on advertisers’ requirements. For example, if an advertiser is looking to run a social media campaign or simply optimise search results, NZME and Fairfax are not likely to be used or even approached. However, the merging parties were generally viewed as close competitors for the type of advertising and products that they offered. 181. When choosing whether or not to advertise on stuff.co.nz or nzherald.co.nz, OMD indicated that it assesses how well advertisements on these sites are converting to sales and it moves spend from the less well-performing site to the other.87 182. Starcom indicated that it does not pit NZME and Fairfax against each other regularly but when this does happen, it tends to occur in a particular ring-fenced area, such as rural-focused advertising. Starcom indicated that there would be an opportunity for the merged entity to increase the price of display advertising and, if that price rise was accompanied by an increase in the quality of the advertising product, then it would be welcome. However, Starcom advised us that in the event that the price rise was not accompanied by an increase in quality, it could use the size of its spend (circa [ ]) to explore other options.88 183. FCB was of the opinion that each of the merging parties provides a valuable advertising environment because of the strength of their news products. This agency said this credibility can rub off on the advertiser which is not always the case with lesser known publishers.89 184. [ ] also considered MediaWorks and TVNZ to be substitutes for display advertising and some video content but considered that NZME and Fairfax are the stronger parties in terms of news content.90 185. [ ] said that pricing of digital advertising can be mix of quality (in terms of effectiveness) but also security. Some premium digital products are in high demand and need to be booked in advance to guarantee that they are available when [ ] wants them. [ ] so as to 91 coincide with advertising on other platforms. 186. There are occasions when [ ] views the advertising products supplied by the merging parties as a must have, because of the reach that NZME and Fairfax can provide. Moreover, [ ] considered that the biggest differentiation between the merging parties and other suppliers of digital advertising is the type of content that they offer. [ 87 88 89 90 91 Commerce Commission interview with OMD (27 June 2016). Commerce Commission interview with Starcom (28 June 2016). Commerce Commission interview with FCB (27 June 2016). Commerce Commission interview with [ ] (31 August 2016). Commerce Commission interview with [ ] (26 August 2016). 38 ].92 187. Some concerns were also expressed by advertisers about the merged entity’s desire to continue to innovate. [ ] expressed some concern over whether the merged entity would continue to innovate in the provision of digital advertising. [ ] noted that no one really knows what new innovations are coming and that there are potential new products or ways of thinking that might not happen with the merger.93 188. [ ] considered that the competition between NZME and Fairfax has stimulated both parties to be innovative and invest in new methods of advertising. An example provided by [ ] is the real estate carousel on nzherald.co.nz which displays revolving display advertising of selected properties. [ ] stated that shortly after this was introduced on nzherald.co.nz, Fairfax followed suit with a similar product.94 Competition from other advertising suppliers 189. On the whole, the majority of advertising customers that the Commission spoke to considered that competitors such as Facebook, Google, TVNZ and MediaWorks, would continue to provide sufficient price or quality constraint on the merged entity for the majority of the digital advertising products that they purchased. 190. Advertising agencies considered that with the merger, they would still have a number of compelling alternatives for digital advertising. While the merging parties are considered “tier 1” at a “homepage mass reach”, advertising agencies consider that TVNZ, MediaWorks and other New Zealand-based providers are still effective.95 191. Some advertising agencies considered that while MediaWorks and TVNZ have considerably lower visitor numbers, there are customers for whom those providers are a better fit in terms of the environment they provide.96 192. Large advertising agencies did not consider that they would be vulnerable to price increases post-merger due to their size and ability to track and assess the performance of different publishers. If the price of advertising on stuff.co.nz or nzherald.co.nz were to increase post-merger, they would discuss what efficiencies the agency would gain as a result of the merger. If the price increase did not come with those efficiencies, they would look elsewhere (to Google, Facebook, MediaWorks and TVNZ).97 92 93 94 95 96 97 Ibid. Commerce Commission interview with [ ] (05 September 2016). Commerce Commission interview with [ ] (21 September 2016). Commerce Commission interview with Starcom (28 June 2016). Commerce Commission interview with PHD (28 June 2016). Commerce Commission interview with OMD (27 June 2016). 39 193. Advertising agencies considered that they will nearly always use Google as all clients require paid search advertising and in some circumstances that will be the only digital product used. FCB considered that while there is a concern that prices could increase with the merger, the reality is that they (the merging parties and the advertising agencies) need each other and there are enough alternatives for FCB to find an audience if the merged entity were to push prices up.98 194. Some corporate advertisers spoken to by the Commission had a slightly different view on the extent to which other providers of digital advertising would act as a constraint on the merged entity. [ ] considered that in the digital area there are only a small number of key New Zealand sites, being stuff.co.nz, nzherald.co.nz, trademe.co.nz, metservice.co.nz, tvnz.co.nz, TV3 and seek.co.nz.99 195. Nevertheless, advertisers generally viewed Facebook and Google as playing a significant role in the supply of digital advertising as they are “huge” digital advertising platforms, with one large advertiser noting that in recent years it has been able to reach a number of people through Facebook, Google and TVNZ onDemand.100 196. Some advertisers noted that Google and Facebook offer different things to the merging parties with Google dominant in search and able to be more direct in its tailored display options. Facebook was seen as offering a product that is tailored to a highly segmented audience with its advertising displayed in a very different social context.101 197. MediaWorks said that it views its Newshub brand and the ability to tie advertising to the range of MediaWorks’ products [ ]. For example, a home improvement show broadcast on TV3 with associated advertising for a building supply merchant or appliance retailer.102 198. TVNZ considered that the key to advertising is the ability to combine audience reach with a rich data set. [ ] to appeal to a large set of advertisers. TVNZ was concerned that with the merger, smaller competitors would be disadvantaged as an advertiser could obtain this large reach from a single entity, [ ].103 The Commission’s view 199. 98 99 100 101 102 103 The Commission considers that Fairfax and NZME are close competitors for the supply of the majority of digital advertising products they sell. This ranges from the Commerce Commission interview with FCB (27 June 2016). Commerce Commission interview with [ ] (31 August 2016) Commerce Commission interview with [ ](26 August 2016). Commerce Commission interview with [ ] (5 September 2016); Commerce Commission interview with [ ] (22 August 2016). Commerce Commission interview with MediaWorks (29 June 2016). Commerce Commission interview with TVNZ (27 June 2016). 40 supply of programmatic display advertising through to homepage takeovers and native advertising. 200. Nevertheless, the main competitive constraints cited by the Applicants – Google and Facebook – offer significant reach and scale advantages, particularly when combined with the rich data sets that each possesses regarding users and their interests, purchasing behaviour and networks. 201. Google, in particular, is able to leverage its Google AdWords and Ad Exchange services to be a significant competitor in the supply of a broad range of digital advertising services. Similarly, Facebook’s social element and ability to offer targeted, context-rich advertising is reflected in its large (and growing) use by advertisers. 202. The downside of this aggregation for advertisers is that they have less control over the placement of their advertising. In general, the presence of content aggregators104 means that a reader may not go direct to the publisher’s website to source content, but instead view content through an aggregator’s site or be directed to a range of other websites with similar content by the intermediary. This means that a publisher’s advertising brand may be diminished if a user has less need to go to stuff.co.nz or nzherald.co.nz to source their news. 203. We consider that other existing providers of digital advertising provide a constraint on the merging parties. The majority of this constraint stems from large aggregators or intermediaries such as Google and Facebook but also (to a lesser extent) from other news and information providers such as TVNZ and MediaWorks. While these other providers have considerably fewer visitor numbers (and therefore reach) than the merging parties or the aggregators and intermediaries, they provide a strong, brand-safe environment. 204. This constraint is diminished when the advertiser is seeking to advertise within a targeted news environment, as to achieve comparable reach an advertiser must combine purchases across a number of suppliers which adds cost and complexity for the purchaser. Nonetheless, the Commission considers that currently the market for the majority of digital advertising is competitive (and likely to remain so with the merger) with a number of alternative options for the majority of digital advertising. 205. Accordingly, our preliminary view is that the proposed merger is unlikely to give rise to a substantial lessening of competition in the market for digital advertising. KPEX 206. 104 105 NZME and Fairfax are joint venture owners, with TVNZ and MediaWorks, in KPEX Limited.105 KPEX operates as an ad exchange to sell remnant digital advertising inventory across publishers' online and mobile websites. Each of the joint venture parties holds a 25% share in KPEX. A content aggregator compiles content from a variety of sources and presents that content to readers in the form of a selection of website links. An example of this is Google News. Short for Kiwi Premium Advertising Exchange. 41 207. We have considered the extent to which the proposed merger would result in the merging parties having a greater control over KPEX than TVNZ and MediaWorks, and whether this would enable them to operate the exchange in such a way as to disadvantage TVNZ and MediaWorks. 208. Currently, all four shareholders in KPEX have an equal shareholding with each of them providing representation on the KPEX board and contributing to the running of the company. The day-to-day operations of the company are carried out independently from the shareholders. 209. An initial concern was that through the merged entity acquiring a 50% shareholding in KPEX (as opposed to the equal 25% shareholdings of the four shareholders at present), it could be afforded a degree of influence that it could in some way favour the placement of its own advertising inventory over that of other KPEX shareholders. 210. However, KPEX relies on sophisticated algorithms to match advertising inventory with the needs of the advertiser and it does this without reference to an individual publisher’s requirements. In addition, [ ]. [ ]. 211. To this extent, [ ] to prevent the merged entity from gaining a substantial degree of influence over KPEX. Competition analysis – premium digital advertising 212. Some of the large advertising customers expressed concern that they would be vulnerable to a price increase on certain forms of premium digital advertising postmerger. We discuss the premium digital advertising market below. 213. This type of advertising tends to be priced differently to other forms of digital advertising. In general, digital advertising is sold on a price-per-impression or priceper-thousand views basis. Mobile interstitials and homepage takeovers are generally priced on a per-day basis which means that their base price is generally higher. This per-day price does not take into account the actual or expected views that an advertisement may receive and, as a consequence, advertising charged on a per-day basis may be more or less expensive depending on the level of traffic to the website on that day. 214. As described above, the three different types of advertising that the Commission considers fall within this category are native advertising, homepage takeovers and mobile interstitial advertising. Advertising customers expressed concern that the merged entity could raise the prices of these particular advertising products postmerger. 42 Native advertising 215. Native advertising uses not only the advertising services of the supplier but also its expertise in content creation. Native advertising has the effect of camouflaging an advertisement within a news/information setting and is designed to add a measure of authority and trustworthiness to the advertisement. 216. Native advertising is currently primarily available in New Zealand through the merging parties and magazines. 217. [ ], both of which use native content across a range of platforms, expressed concern to the Commission that the proposed merger would reduce their options for this form of advertising. 218. [ ] is a significant purchaser of native content and expressed concerns that the proposed merger may lead to a price increase for this type of advertising. [ ] stated that post-merger, if it was looking to promote a product or service that is tailored to an environment around mass-mainstream news, there would be no alternative to the merged entity. [ ] stated: and I think to be very precise; I think the key issue here is the incredible similarity between these two businesses. You know, and the thing is that they are identical in their print and online news offerings. If you look around the edges of that you know sort of online digital, so digital video, there are quite a number of alternatives you can look at there or you know. So around other things sort of perhaps in the same sphere there are alternatives but really on 106 that core news, there isn’t really.” 219. [ ] does not see Facebook and Google as alternatives to the core news environment and this type of bespoke, advertorial content that the merging parties provide.107 220. MediaWorks and TVNZ are both active in the provision of native content and have the necessary journalistic expertise to create relevant content. However, advertisers do not consider that MediaWorks and TVNZ have sufficient reach to be effective substitutes for either NZME or Fairfax, let alone the merged entity. Homepage takeovers 221. A homepage takeover is a combination of different types of display advertising which is displayed on the main or homepage of a website (ie, it is displayed around the border of the text). This type of advertisement cannot be obtained by programmatic advertising. 222. Homepage takeovers in New Zealand are currently primarily sold by the Applicants and by MediaWorks and TVNZ. 106 107 Commerce Commission interview with [ Ibid. ] (05 September 2016). 43 223. In [ ] view, the only area of concern with the proposed merger is the supply of homepage takeovers. In this space NZME and Fairfax are the premium sites, offering a limited supply of inventory.108 224. [ ] noted that there are certain types of media that it needs to book in advance and this tends to be for products such as homepage takeovers where there is higher demand and a limited supply. [ ] is able to pencil in the dates it wishes to use the homepage takeover, but if another customer requests those dates, [ ] would have to confirm them or they would be sold to the other customer.109 225. [ ] would assess whether it was more efficient to invest the money in other digital advertising types.110 226. [ ] on the other hand, did not have a concern. It gave a recent example which illustrates the comparable effectiveness of the homepage takeover compared to other forms of digital advertising. [ ] The goal of the campaign was to get consumers to visit a website and view promotional videos.111 227. In response to the non-Facebook advertising, the website received [ ] views within the first 24 hours. In the same time period, the Facebook advertising resulted in [ ] views. [ ].112 Mobile interstitials 228. Mobile interstitial advertising is an advertisement that is displayed to the user prior to the loading of the desired webpage on a mobile device. These are typically displayed before the homepage of the website is loaded and must be closed by the user before the homepage is loaded. 229. Mobile interstitials are primarily sold by the merging parties, with few alternative providers. 230. [ ] considered that the key advertising product [ ] mobile interstitials which display a [ ] advertisement before nzherald.co.nz or stuff.co.nz website loads on a mobile device. This is an area that has grown very quickly and 108 109 110 111 112 Commerce Commission interview with [ ] (28 June 2016). Commerce Commission interview with [ ] (31 August 2016). Ibid. Commerce Commission interview with [ ] (13 September 2016). Ibid. 44 ]113 [ 231. [ ] has remained constant as ].114 [ mobile traffic has increased from [ ]% to [ ]% in the last three years. 232. [ ] nzherald.co.nz and stuff.co.nz mobile sites as it considers that they have different audiences. 233. [ ] said that a price increase of [ ]% would lead it to stop purchasing mobile interstitials.115 The Commission view 234. The Commission’s view is that the Applicants enjoy a strong position in the provision of premium digital advertising. Demand for this type of advertising is high from large corporate advertisers, and the merging parties are able to combine their large reach and general expertise in online advertising with large and well-resourced editorial expertise. 235. The Commission has assessed the extent to which other providers of premium digital advertising provide a competitive constraint on NZME and Fairfax, and, if not, whether they would be able to expand to provide such a constraint post-merger. 236. As outlined above, unlike other forms of digital advertising, there are few providers of this type of advertising in New Zealand that customers cited as providing a compelling digital advertising alternative to the merging parties. Aggregators such as Facebook and Google also do not act as a constraint in this market. 237. From a content perspective, TVNZ and MediaWorks provide the most comparable environment to the merging parties in that they also offer news and information content. Both of these parties offer some similar products to the merging parties in terms of homepage takeovers and some mobile content. 238. However, [ ] did not consider MediaWorks and TVNZ to be feasible alternatives for mobile interstitials as they do not appeal to customers at the key times that it is looking to target. [ ] said that TVNZ could be an alternative it they had a stronger news presence.116 113 114 115 116 Commerce Commission interview with [ [ Commerce Commission interview with [ Ibid. ] (22 August 2016). ]. ] (22 August 2016). 45 239. As discussed further in the section on reader markets, the Commission does not consider that TVNZ and MediaWorks are likely to expand their news and content offering greatly in the near future. As a consequence, their ability to expand their advertising offerings will be limited by their lack of reach compared to the merged entity. 240. Trade Me, which is a significant provider of display advertising and one of the most viewed websites in the country, [ ]. Currently, Trade Me offers some takeovers on its website but these are limited in number [ ] and they are embedded in certain parts of the Trade Me website such as property and motoring. Trade Me does not offer a ‘homepage takeover’, offering only [ ] 117 takeovers on certain areas of its website. 241. Metservice currently offers both homepage takeovers and mobile interstitials and considers that it competes with the merging parties for these. Metservice noted that it is different from the merging parties in that its homepage is [ ] bookmark the specific areas of the Metservice site (for example marine, rural, snow, etc) and bypass the homepage.118 242. Metservice considered that the stuff.co.nz and nzherald.co.nz websites are different in that users enter and then navigate around the site.119 243. The Commission considers that NZME and Fairfax are each other’s closest competitor in the market for premium digital advertising and that this competitive constraint will be lost with the merger. Countervailing power 244. The Commission has considered the extent to which the purchasers of premium digital advertising would be able to exercise countervailing power over the merged entity to resist any potential price rise over these types of advertising. 245. In this instance, when faced with a price increase in respect of premium digital advertising, an advertiser who purchases both digital and premium digital advertising from the merged entity may be able to threaten to shift its digital advertising (where the merged entity would face competition) to another competitor, such as MediaWorks or Google. In order for such an exercise of countervailing power to be effective, the advertiser must: 117 118 119 245.1 have credible alternative options for digital advertising such that the advertising would continue to be effective; and 245.2 represent a sufficient amount of the merging parties digital advertising revenue such that the removal of that revenue would be greater than the Commerce Commission interview with Trade Me (7 October 2016). Commerce Commission interview with Metservice (3 October 2016). Ibid. 46 increased revenue that the merging parties would gain from increasing the price of premium digital advertising to that advertiser. 246. In 2015, premium digital advertising accounted for $[ ] of NZME’s revenue, with other digital advertising bringing in $[ ]. For the same period, Fairfax received $[ ] from premium digital advertising, and $[ ] from digital advertising.120 247. Although the Commission considers that these advertisers generally have alternatives to the merged entity for digital advertising, we consider that these options may not be sufficiently substitutable in terms of type or reach. Therefore, switching to these alternatives could come at a cost to the advertiser because they would be unable to use their preferred advertising option for that campaign. 248. To this extent, we are not presently satisfied that these advertisers would have sufficient countervailing power such that they would act to constrain a price increase in premium digital advertising. Conclusion 249. The Commission’s preliminary view is that it cannot be satisfied that the proposed merger would not be likely to substantially lessen competition in the market for premium digital advertising. Print advertising 250. Print advertising is a broad category which encompasses a number of different types of publications. Within print the merging parties produce a number of different publications, often with different geographical footprints, subscription pricing, and content focus. The key categories of these relevant to this merger are: 250.1 Sunday newspapers – weekly publications which are published on a Sunday and carry a subscription or cover price, encompassing the Sunday Star-Times and the Sunday News (Fairfax), and the Herald on Sunday (NZME); 250.2 metropolitan daily newspapers – daily newspapers published in the main metropolitan centres and regions and which carry a cover or subscription price, in particular, The Dominion Post and Waikato Times (Fairfax) (Waikato region) and The New Zealand Herald and the Hawke’s Bay Today (NZME) (Hawke’s Bay region), for both the weekday and Saturday editions; and 250.3 community newspapers – typically weekly, free publications which are distributed within relatively small geographical boundaries, such as the titles listed at paragraph 101.4. Market definition 251. 120 Advertisers spoken to by the Commission viewed these different types of publication as delivering a different type of audience and advertising performance. For example, NZME and Fairfax data. 47 community newspapers are seen as providing low-cost (to both the advertiser and the reader), geographically-targeted advertising. Whereas a metropolitan daily, such as The Dominion Post, targets a more affluent audience that is spread across a wider region.121 252. The Commission considers that within the range of different print categories, certain types of publications (such as those that are distributed or sold in the same geographic area or which have a similar cover or subscription price) are likely to be viewed by advertising customers as closer substitutes. 253. This is because these publications generally attract a different audience, have different prices, and often target different geographical areas. In addition, a publication in any of these categories is likely to have as its closest rival a similar publication. 254. For the purposes of considering the competition effects of the proposed merger, we have proceeded on the basis that there are relevant markets for Sunday newspaper advertising, metropolitan daily newspaper advertising, and community newspaper advertising. 255. In reaching this view, we considered whether Saturday newspapers are in the same market for advertisers as Sunday newspapers. However, we have not defined the market on a ‘weekend’ newspaper basis because, as outlined below, feedback from both the Applicants and advertisers is that the Saturday and Sunday newspapers attract different types of advertisers. In any event, because the Applicants own all the metropolitan daily newspapers in the North Island, approaching the market in this way would not necessarily alter our competition analysis. 256. We have also proceeded on the basis that within each of these relevant markets there are likely to be separate geographic markets defined by the distribution or circulation boundaries of the individual publications. 257. In carrying out our competition analysis, we recognise that different types of print advertising will be substitutes for different advertisers at different times. Also, depending on the advertiser and the goals of the advertising campaign, different print publications in different geographic areas will be substitutes. We have had regard to this when considering whether competition would be lessened in the relevant markets. Geographic markets 258. 121 Within the three categories of print publications detailed above, the distribution of the individual papers will differ by the masthead’s location. For example, The Dominion Post is published in Wellington, and is primarily distributed in the lower North Island, whereas the The New Zealand Herald is published in Auckland and distributed mainly in the upper-North Island. Commerce Commission interview with Strategy Design and Advertising (02 September 2016). 48 259. Similarly, a community newspaper is typically based around a defined town or region and not available (at least in print form) outside of that region. 260. For Sunday papers, Herald on Sunday mirrors The New Zealand Herald in that it is an Auckland-based newspaper with a predominantly upper-North Island distribution. By contrast, the Sunday Star-Times and the Sunday News are distributed throughout the country. Table 1 below outlines the circulation of the different Sunday papers. Table 1: Circulation and distribution of Sunday Star-Times, Herald on Sunday and Sunday News CY2015122 Total123 Circulation Northern Region Newspaper (copies per year) Central Region (copies per year) Herald on Sunday [ ] [ ]% Sunday Star-Times [ ] [ ]% ] [ ]% Sunday News Total [ [ ] 100% [ [ [ [ Southern Region (copies per year) ] [ ]% ] [ ]% ] [ ]% ] 100% [ [ [ 0 0% [ ] ] [ ]% [ ] ] [ ]% [ ] ] 100% [ ] Source: NZME and Fairfax data 261. Table 1 shows that in the Southern region the Herald on Sunday is not present, with the Sunday Star-Times and the Sunday News being the only Sunday papers available. In the Northern region the Herald on Sunday has a much larger circulation with the balance shifting as you move down the North Island. This is consistent with the views of the Applicants and advertisers and indicates that the Herald on Sunday is the stronger paper in the Northern region. 262. The Commission considers that the relevant geographic scope within which to assess the impact of the acquisition on advertisers in Sunday newspapers is therefore in the North Island, as this is the area where the distribution of Fairfax’s and NZME’s Sunday newspaper overlap. 122 123 The Northern Region is defined as being the area north of Taupo, the Central Region as the central/lowerNorth Island, and the Southern Region as the South Island. Note, this total excludes the Herald on Sunday, Sunday Star-Times and Sunday News amounts of [ ], recorded as ‘other region’. 49 Competition analysis – Sunday newspapers 263. Post-merger, the merged entity would be the only provider of Sunday newspaper advertising in New Zealand. If Fairfax and NZME would have otherwise provided a competitive constraint on each other in the supply of this advertising for Sunday newspapers in the North Island, then the merger could provide the merged entity with enhanced market power. Herald on Sunday – NZME 264. The Herald on Sunday is published once a week on a Sunday and has an average circulation of [ ] per issue.124 It is a compact format newspaper and includes separate Living and Herald Homes lift-outs. Approximately [ ]% of sales of the Herald on Sunday are made in the Auckland and Northland regions. 265. In its advertiser overview, NZME describes the Herald on Sunday as: …a provocative, compact newspaper that provides uncompromising, agenda-setting news. It has a strong reputation built up in just 10 years, and one we are committed to building on further still. It provides readers with an informative, entertaining, lively Sunday companion that enriches their lives, and in turn provides advertisers with a rich environment to deliver 125 their message to a highly engaged audience. Sunday Star-Times – Fairfax 266. The Sunday Star-Times is also a weekly newspaper published on a Sunday and has an average weekly circulation of [ ]. Of this, approximately [ ]% is circulated in the Northern region (as defined above). The Sunday Star-Times is a broadsheet format newspaper and includes separate Sunday, Business and Escape lift-outs. 267. The Sunday Star-Times has three separate print runs (in the Northern, Central and Southern Regions). The Commission notes that [ ]. 268. Fairfax describes the Sunday Star-Times as: The Sunday Star-Times connects with New Zealanders every Sunday, offering the opportunity to sit back and relax, catch up with the latest news, be inspired and switch off from the 126 stressful week just gone. Sunday News – Fairfax 269. 124 125 126 Sunday News is a compact format newspaper with a national weekly average Sunday circulation of approximately [ ]. Fairfax describes the Sunday News as: NZME and Fairfax data. http://advertising.nzme.co.nz/media/361711/hosonepager2016-2.pdf http://advertise.fairfaxmedia.co.nz/s/Sunday-Star-Times-Product-Sheet-rlwg.pdf 50 Capturing a different audience to other Sunday papers, the Sunday News attracts a staunchly loyal, sports-loving audience. Sunday News features weekly columns, competitions, 127 horoscopes and puzzles, and is topped off with the ultimate sports lift-out. Type of advertising in Sunday newspapers 270. Sunday newspapers are viewed by advertisers as providing a more laid back, aspirational environment and are used by travel and motoring advertisers, for example, to engage a more leisurely audience. Sunday newspapers are not seen as valuable by retail advertisers since they are published too late in the week to engage with weekend shoppers.128 In contrast, travel companies will utilise Sunday papers as the new travel deals are released early in the working week.129 271. Sunday newspapers are viewed as presenting a more relaxed or contemplative environment130 and customers utilise them for specific purposes, such as campaign launches and events. For example, [ ].131 Do the Fairfax Sunday newspapers currently provide a competitive constraint on the Herald on Sunday? 272. To determine whether or not the proposed merger would increase market power and allow the Applicants to increase prices in the market for print advertising in Sunday newspapers in the North Island, we have considered the extent to which the Sunday Star-Times, the Sunday News and the Herald on Sunday are currently providing a competitive constraint on each other. Applicants’ views 273. The Applicants submitted that the primary competitive constraint on the Sunday newspapers comes from other forms of advertising rather than rivalry between the individual Sunday newspapers and that the ability for the merged entity to exert any market power with the merger is limited by the intrinsic link between the advertising and reader sides of the market. 274. The Applicants submitted that the Herald on Sunday and the Sunday Star-Times target different geographic regions and are therefore complementary choices for advertisers, rather than direct competitors. The Herald on Sunday has a strong upper-North Island focus with only [ ]%132 of its sales being made outside of the Auckland and Northland regions. In contrast, the Applicants submitted that only [ ]%133 of sales of the Sunday Star-Times and the Sunday News are made in the Auckland and Northland regions. 127 128 129 130 131 132 133 http://advertise.fairfaxmedia.co.nz/sunday-news Commerce Commission interview with [ ] (22 August 2016). Commerce Commission interview with Strategy Design and Advertising (02 September 2016). http://advertise.fairfaxmedia.co.nz/s/Sunday-Star-Times-Product-Sheet-rlwg.pdf Commerce Commission interview with [ ] (26 August 2016). Application at [14.41(a)]. Ibid at [14.41(b)]. 51 Advertising customers’ views 275. The Commission spoke to six of NZME’s top 10 Herald on Sunday advertising customers. The Commission also spoke to four of the top 10 advertising customers of the Sunday Star-Times. 276. In summary, these advertisers considered that Sunday newspapers were a key platform for delivering certain types of lifestyle-orientated advertising that was not easily replicated by other advertising platforms. 277. For example, [ ] considered that Sunday newspapers still have a strong readership and are a premium format that it is willing to pay for because of the impact that [ ] considers this advertising will have. [ ] does not advertise in the Sunday News as they do not consider that the content type is right for [ ].134 278. Although [ ] indicated that the readership of the Sunday Star-Times and Herald on Sunday is different as they have different geographic footprints, it considered that there is a risk that prices may increase post-merger and that this would be a concern to [ ]. In particular, in respect of specific campaigns or events where print is the key advertising medium that [ ] wants to use (such as its [ ]. 279. [ ] stated that it has a choice between the Sunday Star-Times and the Herald on Sunday and that in all its negotiations its plays NZME and Fairfax off against each other in a bid to negotiate prices down.135 280. [ ] considered that the choice of which Sunday newspaper to advertise in depends on the audience it is looking to target. It considered that the Sunday StarTimes is more focused on the lower-North Island and the South Island, while the Herald on Sunday is focused on the upper-North Island.136 281. PHD said that the decision of which newspaper to advertise in depended on the media strategy for a particular client and that if the strategy was about reaching people in a print environment, then it would choose both papers. PHD considered that it could reach more Auckland readers if it used both the Herald on Sunday and the Sunday Star-Times.137 The Commission’s view 282. 134 135 136 137 The Commission has not received any feedback from advertisers that the Sunday News acts as a strong competitive constraint to the Herald on Sunday. Advertisers noted that the content and environment of the Sunday News is different from the other two Sunday papers. The Commission considers that the Sunday News acts as a minor constraint on the Herald on Sunday. As can be seen from Table 1, the Sunday News is a long way behind the Herald on Sunday and the Sunday Star-Times in Commerce Commission interview with [ ] (26 August 2016). Commerce Commission interview with [ ] (31 August 2016). Commerce Commission interview with [ ] (05 September 2016). Commerce Commission interview with PHD (28 June 2016). 52 respect of circulation and a number of advertisers considered the Sunday News content to be unsuitable for their advertisements. 283. In terms of the Sunday Star-Times and the Herald on Sunday, the Commission considers that the proposed merger would remove each of the merging party’s closest competitor for the supply of Sunday newspaper advertising in the North Island. 284. Fairfax and NZME’s marketing material for the different Sunday newspapers indicates that the target audience of the Sunday Star-Times and the Herald on Sunday is the same.138 As a consequence, they tend to attract advertising from similar types of advertisers. Further, internal documents received from the Applicants indicated that they consider the newspapers compete closely.139 285. There are no other publishers of Sunday newspapers in New Zealand that would provide competitive constraint on the merged entity. 286. We also consider that Sunday papers serve a particular advertising purpose that may not be easily replicated by other forms of print advertising, or other advertising platforms. For example, travel advertisers noted that the Sunday papers are viewed as an important platform for advertising their products. Traditionally, new travel offers are released to the market early in the week, and advertising in the Sunday newspapers allows advertisers to prime the audience for these offers.140 287. Evidence submitted by the Applicants confirmed that while readership and circulation of Sunday newspapers are in decline, they are not (with the exception of the Sunday News) declining at the same rate as other newspapers.141 Conclusion 288. The Commission’s preliminary view is that we cannot be satisfied that the merger will not have, or would not be likely to have, the effect of substantially lessening competition in the North Island market for the supply of print advertising services in Sunday newspapers. 289. We also consider that due to the lack of price discrimination between Fairfax’s Sunday newspapers currently, the likely price increases resulting from this substantial lessening of competition would be likely to be felt nationwide. 138 139 140 141 See http://advertise.farifaxmedia.co.nz/s/Sunday-Star-Times-Product-Sheet-rlwg.pdf and http://advertising.nzme.co.nz/media/361711/hosonepager2016-2.pdf. Fairfax Annex 28, Fairfax Media Brand Health. Newspaper Summary Report. Wave 3. Sunday Star Times, Sunday News; NZME Annex 55, Readership Update; Q1-Q4 2015, contained in an letter from Russell McVeagh (on behalf of Fairfax) to the Commerce Commission (12 July 2016). Commerce Commission interview with Strategy Advertising (02 September 2016). http://www.scoop.co.nz/stories/BU1603/S00125/roy-morgan-readership-results-for-new-zealandnewspapers.htm 53 Competition analysis – metropolitan daily newspapers in the Waikato and Hawke’s Bay regions 290. As outlined in the market definition section above, the majority of metropolitan daily newspapers in New Zealand have discrete areas of geographic distribution, usually based around major cities or regions (for example, the NZ Herald in Auckland, the Otago Daily Times in Otago, and the Hawke’s Bay Today in Hawke’s Bay). 291. As a consequence, they tend to have a more specific regional focus in terms of their content and are also attractive to advertisers who are looking to target readers in particular geographic areas. 292. For the majority of the metropolitan dailies produced by the merging parties, the distances between their base cities are such that there is no material overlap. However, for two regions (being Hawke’s Bay and Waikato) the distance to the nearest rival metropolitan daily is less. As a consequence, there is a greater degree of circulation overlap between the Fairfax and NZME papers. 293. In Waikato, Fairfax publishes The Waikato Times, a daily newspaper based in Hamilton. NZME’s closest major metropolitan daily is The New Zealand Herald which is based in Auckland. In Hawke’s Bay, NZME publishes the Hawke’s Bay Today. Fairfax’s closest major metropolitan daily is The Dominion Post which is based in Wellington. 294. Post-merger, The New Zealand Herald, The Waikato Times, Hawke’s Bay Today and The Dominion Post would come under common ownership. The Commission has considered whether each of The New Zealand Herald and The Dominion Post currently act as a competitive constraint on The Waikato Times and the Hawke’s Bay Today, respectively, such that the removal of that constraint would be likely to substantially lessen competition in the markets for print advertising in metropolitan daily newspapers in each of the Waikato and Hawke’s Bay regions. Competition between the Fairfax and NZME metropolitan daily newspapers in the Waikato and Hawke’s Bay regions 295. To determine whether or not the proposed merger would have the effect of substantially lessening competition in the markets for print advertising in metropolitan daily newspapers in each of the Waikato and Hawke’s Bay regions, we have considered the extent to which The Dominion Post and Hawke’s Bay Today, and The New Zealand Herald and The Waikato Times, are currently providing a competitive constraint on each other. Advertisers’ views 296. None of the advertising customers spoken to by the Commission expressed any concern regarding the proposed merger in relation to The Waikato Times/The New Zealand Herald or The Dominion Post/Hawke’s Bay Today overlaps. These advertisers indicated that this was because purchasers of print advertising in metropolitan newspapers purchased advertising in the particular regions where readers live. 54 The Dominion Post and Hawke’s Bay Today 297. In its advertising material, Fairfax describes The Dominion Post as reaching Wellingtonians. In the same document, Fairfax presents a map of The Dominion Post’s footprint which stretches from Wellington through to north of Taupo.142 298. NZME’s marketing material for the Hawke’s Bay Today stated that it is distributed to the Hawke’s Bay area and surrounds and highlights that the paper is “unashamedly local in its coverage”.143 NZME informed the Commission that only around [ ] copies of the Hawke’s Bay Today are available outside of the Hawke’s Bay region, with none sold south of Dannevirke. 299. Data provided by the Applicants indicated that approximately [ ]% of total sales of The Dominion Post were made in the Hawke’s Bay region in 2015.144 300. Fairfax told the Commission that [ ].145 301. Fairfax also stated that it has [ outside of the Wellington Region.146 ] for The Dominion Post The Commission’s view 302. The Commission considers that while there is some overlap in the circulation of The Dominion Post in the Hawke’s Bay region, neither The Dominion Post nor the Hawke’s Bay Today are likely to be viewed as close competitive constraints on each other for the supply of print advertising in the Hawke’s Bay. 303. This is because a Hawke’s Bay based advertiser would be unlikely to advertise in The Dominion Post as they would likely miss out on advertising their message to the majority of their Hawke’s Bay customers. In addition, such an advertiser would be buying access to a large number of readers that are unlikely to make purchases in the Hawke’s Bay area. As newspaper advertising is sold on a cost-per-thousand basis, such an advertiser would be wasting the majority of their money. 304. Similarly, a Wellington-based advertiser would be unlikely to choose the Hawke’s Bay Today over The Dominion Post as no copies of the Hawke’s Bay Today are available in the Wellington Region. 142 143 144 145 146 http://advertise.fairfaxmedia.co.nz/s/GO-Local-Wellington-media-kit.pdf http://advertising.nzme.co.nz/media/361614/hbtodayonepager2016.pdf Fairfax data. Fairfax Response to Commerce Commission questions of 5 September 2016, attached to an email from Russell McVeagh (on behalf of Fairfax) to the Commerce Commission (14 September 2016). Ibid. 55 The New Zealand Herald and The Waikato Times 305. Similar to The Dominion Post and the Hawke’s Bay Today, The Waikato Times and The NZ Herald are predominantly based around their home cities of Auckland and Hamilton, respectively. However, the distance between these two cities is closer (125km) and as such the two papers’ distribution footprints are closer. 306. Approximately [ ]% of The New Zealand Herald’s sales are in the Waikato region. This is less than The Dominion Post achieves in Hawke’s Bay, despite the much greater distance between Wellington and Hawke’s Bay (294km). 307. For the same reasons outlined above, we consider that The Waikato Times and The New Zealand Herald do not provide a strong competitive constraint on each other for the supply of print advertising. Conclusion 308. The Commission’s preliminary view is that while there is physical overlap between the distribution of The New Zealand Herald and The Waikato Times in the Waikato region, and the distribution of The Dominion Post and the Hawke’s Bay Today in the Hawke’s Bay region, we are satisfied that the proposed merger would not be likely to substantially lessen competition in the market for print advertising in the Waikato and Hawke’s Bay regions. 309. While we do not consider that the proposed merger is likely to raise any concerns in respect of advertisers in these regions, the potential exists for the aggregation of metropolitan daily papers to raise concerns in reader markets. This is discussed further in the reader market at Section 5. Community newspaper advertising 310. The proposed merger would result in NZME acquiring Fairfax’s network of community newspapers across New Zealand, including 10 that overlap with an NZME community newspaper in the North Island. 311. Generally, community newspapers are published for free on a weekly or bi-weekly (twice per week) basis and almost exclusively cover news that is local to the community that they service. Given that community newspapers are distributed for free, their only source of revenue is derived from advertising.147 Therefore, this section deals with the impact that the proposed merger would have on community newspaper advertising in these areas of overlap. 312. Fairfax owns community newspapers throughout New Zealand, whereas NZME only owns community newspapers in the North Island. How the proposed merger could substantially lessen competition 313. 147 Post-merger, NZME would own and control the vast of majority of community newspapers in New Zealand, including in a number of areas where NZME and Fairfax With one exception: NZME’s The Northland Age which is a paid community newspaper. 56 community newspapers currently overlap. The areas of overlap submitted by the Applicants are set out in Table 2. 314. We have assessed whether the aggregation of NZME’s and Fairfax’s community newspapers in these overlapping areas would enable the merged entity to unilaterally increase the price of community newspaper advertising in each area. 315. In assessing the magnitude of any potential increase in advertising prices, we have looked at the competitive alternatives available to community newspaper advertisers. 57 Table 2: Areas of overlap Area Northland NZME Community paper Circulation The Northland Age 3,955 Frequency 148 Fairfax Community paper Circulation Frequency Northern News 14,702 Weekly The Bay Chronicle 12,719 Weekly Bi-weekly Whangarei Whangarei Report 26,675 Weekly Whangarei Leader 33,769 Weekly Hamilton Hamilton News 34,331 Weekly Hamilton Press 62,490 Weekly Rotorua Rotorua Weekender 24,394 Weekly Rotorua Review 23,221 Weekly Taupo Taupo and Turangi Weekender 16,801 Weekly Taupo Times 27,386 Bi-weekly Hauraki Coastal News 8,357 Weekly Hauraki Herald 31,790 Weekly Napier Napier Courier 24,913 Weekly Napier mail 28,529 Weekly Hastings Hastings Leader 18,883 Weekly Hastings Mail 25,812 Weekly Rural Hawke’s Bay CHB Mail 6,702 Weekly HB Country Scene - - Stratford Stratford Press 9,649 Weekly South Taranaki Star 15,421 Weekly Palmerston North Manawatu Guardian 37,062 Weekly The Tribune 37,001 Weekly Horowhenua Horowhenua Chronicle 15,745 Bi-weekly Horowhenua Mail 14,987 Weekly Kapiti Kapiti News 25,348 Weekly Kapiti Observer 28,904 Weekly Source: Application at Appendix 15. 148 Latest ‘ANC’ circulation figures from http://newspaper.abc.org.nz//. Retrieved on 4 November 2016. 58 Market definition 316. As outlined above, the Commission has previously looked at mergers involving community newspaper advertising. In assessing the relevant product market, the Commission considered the approach to market definition taken in Times Media. However, based on the information gathered in this investigation, we have taken an approach similar to Bauer. Rather than defining specific markets, we have explored the extent to which the merger would affect the sale of advertising in community newspapers. 317. In this case, we consider that there is no bright line that separates products that are within a market from those that are outside the market. Community newspapers face differing levels of advertising constraints, including from other community newspapers, other forms of print advertising (including advertising flyers and metropolitan and regional newspapers) and from other media platforms such as radio and digital. 318. The competition analysis below does not rely on exact delineation of the relevant product market to identify the competitive constraints acting on the merged entity with respect to community newspaper advertising. It is, therefore, unnecessary for us to reach a conclusion on the precise product scope of the market. Geographic areas of concern 319. Community newspapers are used by advertisers wanting to reach a local audience. Because of this, we consider that the proposed merger will only affect areas where NZME and Fairfax’s community newspapers cover a similar area. 320. In most of the areas of overlap set out in Table 2, we consider that NZME and Fairfax’s community newspapers are each other’s closest competitors. In each of these areas, we consider that NZME and Fairfax’s community newspapers provide a competitive constraint on each other and that constraint would be removed postmerger. 321. There are three areas that the Commission has assessed and in which we are satisfied that the proposed merger is unlikely to result in a substantial lessening of competition. 321.1 In Northland we were advised by local advertisers that NZME’s Northland Age covers a separate area to Fairfax’s Northern News and The Bay Chronicle. These advertisers said that the Northland Age is focused on the Kaitaia area whereas the Northern News and Bay Chronicle are focused on the Kerikeri and Bay of Islands areas, respectively. In addition, the local advertisers also noted that the Northland Age attracts different readers to the Northern News and The Bay Chronicle as it is a paid community newspaper. 321.2 In Hauraki, NZME’s Coastal News almost exclusively services Whangamata, whereas the Hauraki Herald covers the wider Hauraki district. Most advertisers in the Coastal News are based in Whangamata and explained that they do not consider the Hauraki Herald to be an option for advertising as it 59 would be a waste (as it overs a much larger area than Whangamata). Conversely, advertisers in the Hauraki Herald appear to be based in Thames and the wider Hauraki region and are unlikely to view the Coastal News as an alternative source of advertising. 321.3 The Applicants submitted that Fairfax’s HB Country Scene and NZME’s CHB Mail compete in rural Hawke’s Bay.149 Based on the Commission’s investigations, we have subsequently learned that the HB Country Scene ceased publication in 2013.150 322. In Taranaki, we consider that the proposed merger would only reduce the community newspaper advertising options in the Stratford area, rather than all of Taranaki. This is because NZME’s Stratford Press only covers the Stratford area, and therefore, advertisers in other parts of Taranaki (for example, New Plymouth) are unlikely to view the Stratford Press as a close alternative. 323. Similarly, in Manawatu, we consider that it is unlikely that the proposed merger would have a substantial impact on advertisers in the Feilding area. While there is some overlap between NZME’s Manawatu Guardian and Fairfax’s Feilding-Rangitikei Herald, the advertisers that we spoke with did not view the two newspapers as close substitutes. Advertisers that want to target a Feilding-based audience use the Feilding-Rangitikei Herald, whereas advertisers that want to target Palmerston North will use either the Manawatu Guardian or The Tribune. 324. To this extent, when assessing the extent to which the proposed merger would affect the sale of advertising in community newspapers, we have focused on each of the following areas: 324.1 Whangarei; 324.2 Hamilton; 324.3 Rotorua; 324.4 Taupo; 324.5 Napier; 324.6 Hastings; 324.7 Stratford; 324.8 Palmerston North; 324.9 Horowhenua; and 324.10 Kapiti. 149 150 Application at page 140. Email from Russell McVeagh (on behalf of Fairfax) to the Commerce Commission (20 October 2016). 60 Competition analysis 325. 326. To determine whether the proposed merger would affect the sale of community newspaper advertising in each of the areas of overlap, such that a substantial lessening of competition would be likely, we have considered whether the merged entity would be sufficiently constrained by: 325.1 other community newspapers within each area; and/or 325.2 other advertising options; and/or 325.3 the countervailing power of community newspaper advertisers. Having determined that the Commission is not satisfied that the proposed merger would not be likely to substantially lessen competition in any of the areas of overlap, we then considered whether potential entry by a new community newspaper in any of these areas would act as a constraint on the merged entity. Investigation approach 327. In conducting our competition analysis, the Commission spoke with a number of interested parties, including: 327.1 large national advertisers; 327.2 advertising agencies; and 327.3 local community newspaper advertisers. 328. Of these interested parties, we note that local advertisers are the biggest advertisers in community newspapers. For example, in 2015 only [ ]% of the total advertising revenue that NZME generates from its community newspapers in the areas of overlap is derived from national display advertising. Most of the remaining revenue is generated from local advertisers: [ ]% from local display advertising and [ ]% from classified advertising.151 329. To obtain the views of local advertisers, the Commission called approximately four to five local advertisers from NZME and Fairfax community newspapers across each of the areas of overlap. This resulted in us speaking with 45 local community newspaper advertisers from across the areas of overlap.152 330. For most parts of our analysis, we considered that the characteristics of the areas of overlap were sufficiently similar to assess the views of the local advertisers collectively. For example, our analysis of the extent to which advertising flyers compete with community newspaper advertising. 151 152 The remaining [ ]% is other forms of advertisements. NZME Annex 58 - Community Revenues (Excel), attached to an email from Russell McVeagh to the Commerce Commission (16 September 2016). The Commission also spoke with local advertisers in Northland, Hauraki and Feilding. 61 331. In other circumstances, where we considered that the location of an advertiser is important to the analysis, such as considering the relevant geographic dimension of the market or assessing competing community newspapers in a particular area, the Commission only considered the views of the local advertisers within each of the areas of overlap. Applicants’ views 332. The Applicants submitted that the merged entity would continue to face competition from existing community newspapers in a number of the areas of overlap. These competitors are set out in Table 3.153 Table 3: Existing competitors listed by the Applicants154 Area Competitor listed by Applicant Circulation Frequency Whangarei - - - Hamilton Western Community News 18,000 Twice per month Rotorua - - - Taupo - - - Napier & Hastings Bay Buzz 5,000 Every two months Opunake and Coastal News 8,800 Fortnightly Taranaki Farming Lifestyles 10,000 Monthly Patea and Waverly Press 2,700 Monthly Palmerston North Manawatu Farming Lifestyles 15,000 Monthly Horowhenua - - - Kapiti - - - Stratford 333. In addition to competition from other community newspapers, the Applicants submitted that in each area of overlap the merged entity would continue to face competition from a range of other advertising options, including digital advertising, TV, radio, and advertising flyers.155 334. Further, the Applicants submitted that the merged entity would also be constrained by the countervailing power of advertisers. In making this submission, the Applicants referred to Times Media, where the Commission held:156 The Commission considers that advertisers have a substantial degree of countervailing power. Advertisers are able to express this through a variety of ways, but underlying these different 153 154 155 156 Application at page 140-141 (Appendix 15). This table only includes those publications that the Applicants have listed as existing competitors (for example, we have not included neighbouring publications listed by the Applicants). Application at page 140 (Appendix 15). Application at page 58. 62 negotiation tactics is the knowledge that a community newspaper is dependent on advertisers for all its revenue. Others’ views View of advertising agencies 335. A number of the advertising agencies that we spoke with did not have a view as their clients rarely advertised in community newspapers. 336. Strategy Design & Advertising was one advertising agency that did have clients that advertised in community newspapers. [ ].157 337. Strategy Design & Advertising was of the view that [ ].158 For these local advertisers, Strategy Design & Advertising considered community newspaper advertising is cheaper and/or requires less technical experience compared to other forms of advertising, such as radio, Facebook, or using advertising flyers.159 338. [ ] is a small advertising agency with some smaller clients that advertise in community newspapers]. It was of the view that there are “not really any” alternatives that fill the ‘niche’ of community newspaper advertising. [ ] said that radio is completely different to community newspapers from a media perspective. In regards to Facebook, [ ] explained that significant time and costs are involved in making Facebook work and that it is a social media platform, rather than a sales tool.160 National advertisers 339. The national advertisers that we spoke with rarely advertise in community newspapers. Instead, they often opt for advertising flyers which are a cost-effective option compared to community newspapers when printed and distributed on a large scale. 340. PMP Limited, a large New Zealand-based printer and distributor of advertising flyers, confirmed that advertising flyers are cheaper than advertising in community newspapers for large scale advertisers. PMP Limited told the Commission that these national advertisers are printing up to a million advertising flyers at a time.161 157 158 159 160 161 Commerce Commission interview with Strategy Design and Advertising (2 September 2016). Ibid. Ibid. Commerce Commission interview with [ ] (1 September 2016). Commerce Commission interview with PMP Limited (15 September 2016). 63 341. Generally, national advertisers only use community newspapers on an ad hoc basis when they want to send an individualised message to specific local community. [ ] gave the example of using community newspaper advertising to promote the opening of a local store.162 [ ] did not consider radio to be an effective alternative to community newspapers, as it is not as targeted.163 342. We note that MediaWorks [ ]. It considers that “the propositions offered by radio and print advertising are quite different”. However, it did consider that in smaller markets where advertisers have fixed budgets, community newspapers are a competitor to radio advertising.164 Local advertisers 343. The local advertisers that we spoke with were of view that community newspapers are an effective way to advertise to people within their community. For many of these advertisers, their target audience is older and/or from a lower socio-economic group. The local advertisers explained that these audiences are more likely to read free community newspapers than paid newspapers and often don’t use the internet. 344. In response to a price increase of community newspaper advertising post-merger, none of the local advertisers said that they would switch to a competing independent community newspaper. Rather, faced with an increase in the price of community newspaper advertising post-merger, the majority of local advertisers said that they would continue to advertise in the merged entity’s community newspapers. 345. There was not a consensus among the advertisers spoken to by the Commission as to the extent of price increase needed before they would switch away from advertising in the Applicants’ community newspapers. However, the majority of those that said that they would continue to advertise in community newspapers, would do so at a price increase of at least 5-10%. 346. None of the 45 local advertisers we spoke with would switch to advertising flyers in response to a price increase of community newspapers.165 A number of local advertisers expressed the view that advertising flyers are an expensive option compared to community newspaper advertising. This view was supported by PMP Limited, which said that for local advertisers community newspapers are “quite a bit cheaper” than the cost of printing and distributing advertising flyers.166 167 162 163 164 165 166 Commerce Commission interview with [ Commerce Commission interviews with [ ] (9 September 2016). ] (9 September 2016) and [ ] (5 September 2016). NZME_Fairfax_Comcom questions (MW response) 30902016, attached to an email from MediaWorks to the Commerce Commission (30 September 2016). We note that in Northland (not an area of overlap) one local advertiser viewed advertising flyers as a close substitute to community newspaper advertising. Commerce Commission interview with PMP Limited (15 September 2016). 64 347. Around half of the local advertisers that we spoke with said that they either use or have used radio advertising. However, only one of these local advertisers said that it would switch to radio advertising in response to a price increase of community newspaper advertising. 348. The advertisers that continue to use both radio and community newspaper advertising were of the view that radio and community newspaper advertising serve different purposes. Those advertisers that have stopped advertising on radio advertising were generally of the view that it was expensive and ineffective. 349. While some local advertisers have an online presence, such as advertising on their own website, Facebook, Google and/or Trade Me, only five of the local advertisers (the majority of which were car dealers) indicated that they would consider switching to digital advertising in the event of a price increase. 350. Other car dealers had different views on the extent to which online advertising is a close substitute to community newspaper advertising. These advertisers explained that advertising on Trade Me means that they have to compete with cars that are being sold throughout New Zealand and it is often difficult to have a point of difference. Whereas community newspaper advertising allows these car dealers to target potential purchasers within their community. 351. The local retail and service advertisers explained that while Facebook is useful for connecting with pre-existing customers that are internet users, it is not as effective for targeting new customers or for getting their name ‘out there’. In addition, given that community newspaper audiences tend to be older, digital advertising was often not viewed as an effective way to reach these audiences. The Commission’s view 352. The Commission is of the view that community newspaper advertising is an effective way for local advertisers to reach local audiences. The Commission is not satisfied that local advertisers would be likely to have sufficient alternative advertising options to constrain the merged entity from increasing the price of advertising. 353. We consider that our view is consistent with Fairfax’s own internal document which states that community newspapers offer “[ ][ ][ ].”168 354. Further, we are of the view that NZME and Fairfax’s community newspapers provide a pricing constraint on each other and that this constraint would be removed post- 167 168 The Commission notes that Simon Ellis (Managing Director of PMP Limited) is the President of the New Zealand Community Newspapers Association. Fairfax presentation, Annex 26 - Communities Presentation. Trade Communications (November 2014), attached to an email from Russell McVeagh (on behalf of Fairfax) to the Commerce Commission (4 July 2016). 65 merger. We consider that this is consistent with NZME’s own internal document that states:169 [ [ 355. ][ ] ]. Consistent with NZME’s internal document, we note that [ ] in the 10 areas of 170 overlap, compared with the areas where it does not appear to face competition from an existing community newspaper,171 172 as set out in Table 4.173 Table 4: NZME yearly gross profit margins by competition faced, 2013-2015 10 Areas of overlap Areas of no competition 2013 2014 2015 2013 2014 2015 Average circulation 23,005 23,599 24,216 8,249 8,566 8,697 Advertising revenue $[ ] $[ ] $[ ] $[ ] $[ ] $[ ] Print costs $[ ] $[ ] $[ ] $[ ] $[ ] $[ ] ] $[ ] ] $[ ] Distribution costs $[ Gross profit $[ Gross profit margin [ ]% ] ] $[ $[ [ ]% [ ]% $[ $[ [ ]% ] ] $[ $[ [ ]% ] ] $[ $[ [ ]% Source: NZME Data 356. 169 170 171 172 173 The Commission understands that the main variable cost of producing a community newspaper is the cost of print and distribution. Advertisers are prepared to pay more for advertising in community newspapers that are circulated to a larger number of people. As such, we would expect advertising revenues and print and distribution NZME presentation, Annex 38 - NZME F16 Regional Strategy Summary (24 February 2016), attached to an email from Russell McVeagh (on behalf of NZME) to the Commerce Commission (19 August 2016). The Commission notes that the [ ]. Northland Age, Te Awamutu Courier, CHB Mail, Katikati Advertiser, Te Puke Times, Coastal News, Havelock North Village Press, Waihi Leader, Bush Telegraph. The Commission made this judgement with reference to NZME’s internal document – above n 169). [ ]. The Commission has been unable to undertake a similar assessment in respect of Fairfax’s community newspapers because Fairfax was unable to provide the distribution costs for each of its community newspapers. ] ] 66 costs to be roughly proportionate to the circulation of a community newspaper and, therefore, expect similar gross profit margins across different areas.174 357. The Commission notes that the characteristics of different markets could affect the profits earned by NZME. For example, a community newspaper that contained a higher proportion of real estate advertising (which is sold more cheaply than local display advertising) may have a lower profit margin compared with community newspaper that only sells local display advertising. However, in this case, real estate advertising only makes up around [ ]% of NZME’s total revenue in the 10 areas of overlap, but [ ]% in the areas of no competition.175 358. Further, we are not aware of any reason why constraints from other forms of advertising (such as digital, radio or advertising flyers) would be significantly different between the 10 areas of overlap and the areas where NZME does not face competition. 359. To this extent, it is our preliminary view that the difference in gross profit margins indicated in Table 4 is consistent with our view that Fairfax provides a constraint on NZME’s pricing of community newspaper advertising in the 10 areas of overlap. 360. As such, we are not satisfied that the merged entity would be constrained from increasing the price of community newspaper advertising because of the competition it would continue to face from other community newspapers, advertising flyers, radio and digital advertising and the countervailing power of advertisers. Constraint provided by existing community newspapers 361. In each of the areas of overlap, the Commission does not consider that the merged entity would face strong levels of competition from any existing independent community newspapers. 362. As set out in Table 5, we consider that the competing publications listed by the Applicants differ in circulation, geographic area, frequency and/or content from NZME and Fairfax’s community newspapers, such that they are unlikely to provide any significant constraint on the merged entity. 363. In addition, none of the local advertisers that we spoke with from each of the areas of overlap said that they would switch to advertising in an independent community newspaper post-merger.176 174 [ 175 Annex 58 - Community Revenues (Excel), attached to an email from Russell McVeagh (on behalf of NZME) to the Commerce Commission (16 September 2016). One advertiser said that it advertises in the Opunake and Coastal News when targeting audiences in the Opunake area, however it did not consider it to be a substitute for advertising in Stratford. 176 ] 67 Table 5: Existing competitors in each area of overlap Area Whangarei Existing competitors - Commission’s preliminary view on the constraint imposed by these competitors There are no competing community newspapers in Whangarei. The Commission does not consider that Western Community News is a close competitor to the Hamilton Press and Hamilton News, for the following reasons: Hamilton Western Community News • Between July 2015 and June 2016, the Western Community Centre earned $16,860 from the Western Community News.177 In contrast, in 2015, the Hamilton News and Hamilton Press earned around $[ ]m178 and $[ ]m179 in advertising revenue respectively. • Western Community News is focused on Hamilton West, and therefore, does not have the same coverage as NZME and Fairfax community newspapers. • It is a volunteer-run community newspaper with significantly less content than Hamilton News and Hamilton Press. 177 178 179 Rotorua - There are no competing community newspapers in Rotorua. Taupo - There are no competing community newspapers in Taupo. Napier Bay Buzz The Commission does not consider that the Bay Buzz is a close competitor to the Napier Courier and Napier Mail, for the following reasons: Western Community Centre 2016 Annual Report, retrieved from here http://www.westerncommunitycentre.org/annual-report-2016 on 4 November 2016. 3117377 Regionals L3Y P&Ls (confidential and subject to legal advice) – v1, attached to an email from Russell McVeagh (on behalf of NZME) to the Commerce Commission (4 July 2016). Annex 83 - Request28September_Communities, attached to an email from Russell McVeagh (on behalf of NZME) to the Commerce Commission (6 October 2016). 68 Area Existing competitors Commission’s preliminary view on the constraint imposed by these competitors • The Bay Buzz has a significantly smaller circulation area (5,000) compared with the Napier Courier (24,913) and Napier Mail (28,529). An advertiser wanting broad reach in the Napier area would be unable to achieve this by advertising the Bay Buzz. • The Bay Buzz is distributed less frequently (every two months) compared with the Napier Courier (weekly) and Napier Mail (weekly). For community newspaper advertisers that want to advertise on a frequent basis, the Bay Buzz would not be a close alternative. • The Bay Buzz is a paid magazine compared with the Napier Courier and Napier Mail which are free community newspapers. Many advertisers use community newspapers to target certain demographics that are more inclined to read free publications, it would be more difficult to target these audiences by advertising in the Bay Buzz. Hastings Bay Buzz The Commission does not consider that they Bay Buzz is a close competitor to the Hastings Leader or Hastings Mail, for the same reasons as in Napier. Opunake and Coastal News The Commission does not consider that the Opunake and Coastal News is a close competitor to the Stratford Press and South Taranaki Star as it is not distributed in the Stratford area.180 One advertiser said that it advertises in the Opunake and Coastal News when targeting audiences in the Opunake area, however it did not consider it to be a substitute for advertising in Stratford. Stratford Taranaki Farming Lifestyles The Commission does not consider that the Taranaki Farming Lifestyles is a close competitor to the Stratford Press and South Taranaki Star for the following reasons: • Taranaki Farming Lifestyles is only distributed to rural addresses, and therefore, is unlikely to be an effective option for advertisers wanting to target urban readers in Stratford. • Taranaki Farming Lifestyles covers farming news rather than Stratford-specific news, and therefore, is unlikely to be an effective option for advertisers wanting to advertise to the general 180 http://www.opunakecoastalnews.co.nz/ 69 Area Existing competitors Commission’s preliminary view on the constraint imposed by these competitors Stratford population. • Taranaki Farming Lifestyles is distributed less frequently (monthly) compared with the Stratford Press (weekly) and South Taranaki Star (weekly). For community newspaper advertisers that want to advertise on a frequent basis, the Taranaki Farming Lifestyles would be unlikely to be a close alternative. Patea and Waverly Press The Commission does not consider that the Patea and Waverly Press is a close competitor to the Stratford Press and South Taranaki Star as it is not distributed in the Stratford area. The Commission does not consider that the Manawatu Farming Lifestyles is a close competitor to the Stratford Press and South Taranaki Star for the following reasons: • Manawatu Farming Lifestyles is only distributed to rural addresses, and therefore, is unlikely to be an effective option for advertisers wanting to target urban readers in Palmerston North. Manawatu Manawatu Farming Lifestyles • Manawatu Farming Lifestyles covers farming news rather than Palmerston North-specific news, and therefore, is unlikely to be an effective option for advertisers wanting to advertise to the general Palmerston North population. • Manawatu Farming Lifestyles is distributed less frequently (monthly) compared with the The Tribune (weekly) and The Manawatu Guardian (weekly). For community newspaper advertisers that want to advertise on a frequent basis, the Manawatu Farming Lifestyles would not be a close alternative. Horowhenua - There are no competing community newspapers in Horowhenua. Kapiti - There are no competing community newspapers in Kapiti. 70 Constraint provided by advertising flyers 364. The Commission does not consider that, for local advertisers, advertising flyers would be likely to constrain the merged entity from increasing the price in any of the areas of overlap. 365. For local advertisers, advertising flyers are significantly more expensive than community newspaper advertising. With community newspapers, advertisers are able to share the print and distribution costs with other advertisers. Advertising flyers, on the other hand, require a single advertiser to bear all of the print and distribution costs. This view was shared by PMP Limited and is consistent with the fact that none of the local advertisers we spoke with would switch to advertising flyers in the event that the price of community newspaper advertising increased post-merger.181 366. For national advertisers, however, advertising flyers are a cost-effective alternative to advertising in community newspapers. This is because of the large scale by which these advertisers print and distribute their flyers (up to one million at a time). The large national advertisers we spoke with already preferred to use advertising flyers and this is reflected in the small proportion of community newspaper advertising that is derived from national advertisers. 367. We note that there is a point where an advertiser will have sufficient scale that advertising flyers become a cost-effective option compared to community newspaper advertising. This point likely lies somewhere between the scale of the national advertisers and the scale of local advertisers. 368. Based on the views of local advertisers we spoke with, we do not consider that a sufficient number of community newspaper advertisers would have the requisite scale such that advertising flyers would become a cost-effective option and, therefore, would be likely to constrain the merged entity from increasing the price of community newspaper advertising. Constraint provided by radio and digital advertising 369. Based on the views of the interested parties we spoke with, the Commission does not consider that, for local advertisers, radio and digital advertising alternatives would be likely to constrain the merged entity from increasing the price of community newspaper advertising in any of the areas of overlap. 370. While some local advertisers will view radio and digital advertising to be a close alternative to community newspaper advertising, we consider that there is unlikely to be a sufficient number of these advertisers to effectively constrain the merged entity. This is consistent with the views of a number of local advertisers we spoke with that would continue to advertise in community newspapers despite a 5-10% price increase. 181 Commerce Commission interview with PMP Limited (15 September 2016). 71 371. In respect of radio advertising, we consider that our view is consistent with other community newspaper publishers that we spoke with. NorthSouth Multi Media was of the view that radio needs the “support” of print advertising for details.182 Likewise, Wellington Suburban Newspapers and River City Press did not consider radio advertising to be an advertising competitor to their respective community newspapers. 183 184 372. In respect of digital advertising, we consider that our view is consistent with the NZME’s own internal document which states that “[ ]”.185 Countervailing power 373. Countervailing power assesses the extent to which a merged entity’s ability to profitably increase prices, may be constrained by the ability of certain customers to exert substantial influence on negotiations. This requires customers to have the ability to substantially influence the price the merged entity charges. 374. In assessing the extent to which community newspaper advertisers have sufficient countervailing power to constrain the merged entity, we reviewed the Commission’s findings in Times Media. In that decision, the Commission was informed that advertisers could effectively discipline an increase in price of community newspaper advertising by threatening to leave or actually leaving the publication. Based on the information we have gathered in this investigation, we have reached a different view. 375. In this investigation, we do not consider that community newspaper advertisers would have sufficient countervailing power to substantially influence the pricing decisions of the merged entity. While national advertisers or advertising agencies may have some degree of countervailing power, we do not consider that this would apply to local advertisers. 376. As discussed above, the majority of community newspaper advertisers are local advertisers. For these local advertisers, community newspapers are a valued form of advertising that is not easily substitutable with other forms of advertising. Therefore, although community newspapers are dependent on advertisers for revenue, we consider local advertisers are similarly dependent on the merged entity for advertising. 182 183 184 185 Commerce Commission interview with NorthSouth Multi Media (13 October 2016). Commerce Commission interviews with River City Press (16 September 2016) and Wellington Suburban Newspapers (28 September 2016). Sun Media, took a different view. It was of the view that it faced competition from radio and online advertising. That said, Sun Media also considered itself to be “a little bit different” to other community newspapers. In its view, because of its size The Weekend Sun attracts different advertisers to other community newspapers. Commerce Commission interview with Sun Media (16 September 2016). NZME presentation, Annex 38 - NZME F16 Regional Strategy Summary (24 February 2016), attached to an email from Russell McVeagh (on behalf of NZME) to the Commerce Commission (19 August 2016). 72 Real estate exception 377. The Commission notes that real estate advertisers may have more alternatives available to them than other local advertisers. In particular, the Property Press which is published throughout the North Island. Similarly, digital real estate options may be a closer substitute to community newspapers than it is for other local advertisers. 378. [ ].186 379. For these reasons, the Commission’s preliminary view is to exclude real estate advertisers from its analysis. However, in doing so we note that, in 2015, real estate advertising made up only [ ]% of advertising revenue generated from community newspapers across the areas of overlap.187 Entry and expansion by competing community newspapers 380. Having determined that the proposed merger would eliminate an existing competitive constraint within the overlap areas, the Commission has then considered whether potential entry by a new community newspaper would constrain the merged entity from acquiring market power in any of the markets of concern. 381. We assess whether entry by new competitors or expansion by existing competitors is likely to be sufficient in extent and occur in a timely fashion to constrain the merged entity and prevent a substantial lessening of competition. This is referred to as the ‘LET test’. Applicants’ views 382. The Applicants submitted that there are no “material barriers to entry” in starting a community newspaper, such that the threat of potential entry will constrain the merged entity.188 383. In making this submission, the Applicants referred to a number of existing independent community newspaper publishers that could readily expand into the markets of concern to constrain the merged entity. In particular, River City Press, Sun Media, Wellington Suburban Newspapers and NorthSouth Multi Media. 383.1 River City Press is an independent community newspaper based in Wanganui that was established in 1985. It competes with NZME’s community newspaper, the Wanganui Midweek. The Applicants submitted that River City Press could readily expand into Taranaki and Manawatu. 383.2 Sun Media is an independent news publisher based in Tauranga that was established in 2000. Sun Media publishes a weekly community newspaper, 186 187 188 Based on 2015 calendar year. NZME and Fairfax data. Application at page 73. 73 The Weekend Sun. Given the size of The Weekend Sun (80-90 pages with a circulation of 65000), Sun Media considers that it competes with both NZME’s daily and community newspaper in Tauranga region, The Bay of Plenty Times and Bay News, respectively. The Applicants submitted that Sun Media could readily expand into Rotorua, Hamilton, Taupo, Napier and Hastings. 383.3 Wellington Suburban Newspapers (WSN) is an independent news publisher that was established after acquiring four Wellington-based community newspapers from NZME in 2013.189 Owner, Les Whiteside, also owns a community newspaper in Marlborough called The Blenheim Sun. The Applicants submitted that WSN could readily expand into Kapiti and Manawatu. 383.4 NorthSouth Multi Media (NSMM) is a Northland-based independent publisher that owns The Kaipara Lifestyler (a weekly community publication in the Kaipara district with a circulation of around 10,000) and six monthly farming newspapers around New Zealand. The Applicants submitted that NSMM could readily expand into Whangarei. Others’ views 384. In assessing whether the LET test is likely to be satisfied in each of the markets of concern, the Commission spoke with Sun Media, WSN, River City Press and NSMM. 385. None of these independent publishers have plans to expand into any of the areas of overlap, nor were any of the view that they would start-up a new community newspaper or expand an existing community newspaper to replace competition lost by the merger. [ ]. 190 386. The independent publishers said that there would be considerable risk in starting-up a community newspaper in direct competition with the merged entity. In particular, the scale of the merged entity is such that a new entrant would not be able to compete with the merged entity on price. NSMM said that if it tried to start-up a competing community newspaper, Fairfax would respond aggressively by dropping prices.191 387. Each of the independent publishers has observed aggressive pricing of community newspaper advertising by NZME and/or Fairfax. 189 190 191 The Porirua News, The Independent Herald, The Cook Strait News and Wainuiomata News. Wellington Suburban News have subsequently closed the Porirua News. Commerce Commission interviews with Sun Media (16 September 2016) and WSN (28 September 2016). Commerce Commission interview with NSMM (13 October 2016). 74 387.1 Mr Whiteside explained that The Blenheim Sun [ ].192 387.2 Sun Media said [ ]. Sun Media said it could not compete with that price, as $[ ] is [ ].193 387.3 River City Press was of the view that the price that NZME [ ] and “[ ]”.194 387.4 NSMM said that it used to charge $[ ]. NSMM said that NZME and Fairfax “don’t play nice” with competition.195 388. The independent publishers said that they have been able to compete by having long-standing relationships with the local advertisers and being able to provide a better service. Each of the independent publishers’ community newspapers are wellestablished, the most recently established those published by Wellington Suburban Newspapers (formerly CityLife) which began publication 10 years ago. 389. In order to build these relationships with local advertisers, the independent publishers were of the view that a new entrant would need to be local to the community it services. WSN explained that a new publisher needs to be “getting involved in the community, going to events”.196 However, the independent publishers were of the view that it can be difficult to build these relationships and it can take a long time. 390. Sun Media provided an example of its unsuccessful attempt to enter the Hauraki region by launching a community newspaper called The Peninsula Sun. It stopped publication after nine weeks after being unable to attract readers and advertisers in the area.197 391. NSMM said that it recently took over a small community newspaper in rural Canterbury; however, it was misguided in its ability to run such a publication remotely. As such, it leased the newspaper to a local person.198 392. Les Whiteside said that The Blenheim Sun was able to get an advertising base in Marlborough because Countdown New Zealand supported the paper from ‘day one’. However, Mr Whiteside was of the view that entry in this fashion would be more 192 193 194 195 196 197 198 Commerce Commission interview with WSN (28 September 2016). Commerce Commission interview with Sun Media (16 September 2016). Commerce Commission interview with River City Press (16 September 2016). Commerce Commission interview with NSMM (13 October 2016). Email from WSN to the Commerce Commission (26 September 2016). Commerce Commission interview with Sun Media (16 September 2016). Commerce Commission interview with NSMM (13 October 2016). 75 difficult now as it is hard for independent community newspapers to attract national advertisers.199 393. Notwithstanding the above, the independent publishers said that if a new entrant was able to build an advertising base, it would take years to get a return on investment. 394. River City Press estimated that a new entrant would need to invest at least $100,000 to start up a new community newspaper and it would take three to five years to return a profit.200 WSN estimated that for the first three months of operation, a new entrant would need at least $50,000 to set up a new community newspaper. 201 NSMM was of the view that starting up a community newspaper in an area that does not face competition from another publisher would still cost a lot and take a long time to get established.202 The Commission’s view 395. We consider that the four independent publishers submitted by the Applicants would be the best positioned to enter or expand into each of the areas of overlap, post-merger. 396. We have not identified any local publishers in the areas of overlap that we consider would be better placed than WSN, Sun Media, River City Press and NSMM to start-up a community newspaper and effectively compete with the merged entity. This included assessing each of the competitors listed in Table 5. 199 200 201 202 203 204 205 396.1 Western Community News is a volunteer-run newspaper published by a local community centre.203 The community centre offers a range of services to “contribute to the wellbeing” of the Hamilton West community,204 including the publication of Western Community News. Given the community centre’s social objective to support its local community, we do not consider that Western Community News is more likely to expand than WSN, Sun Media, River City Press and NSMM. 396.2 Bay Buzz is a magazine rather than a community newspaper. Its focus is on investigative journalism rather local community news.205 If Bay Buzz were to expand to compete with the merged entity’s community newspapers, it would be a significant departure from its current course of business. Because WSN, Sun Media, River City Press and NSMM are already in the business of producing community news, we consider that they are more likely to expand than Bay Buzz. Commerce Commission interview with WSN (28 September 2016). Commerce Commission interview with River City Press (16 September 2016). Email from WSN to the Commerce Commission (26 September 2016). Commerce Commission interview with NSMM (13 October 2016). http://www.stuff.co.nz/national/politics/83295637/Western-Community-News-lashed-over-watermeters-stance http://www.westerncommunitycentre.org/what-we-offer A selection of articles from the Bay Buzz can be found here http://www.baybuzz.co.nz/articles/ 76 396.3 Neither Opunake and Coastal News or Patea and Waverly Press are distributed in Stratford. Therefore, both papers would need to expand into a new area to constrain the merged entity. This puts these two publishers in a similar position to WSN, Sun Media, River City Press and NSMM. Given the small size of Opunake and Coastal News and Patea and Waverly Press’s current operations, we do not consider that they would be more likely to expand than WSN, Sun Media, River City Press and NSMM. 396.4 Taranaki Farming Lifestyles and Manawatu Farming Lifestyles are published by NSMM. 397. The first element of the LET test is the likelihood of entry. Entry is likely if, given a price increase post-merger, the expected profitability of entry is positive.206 In considering the likelihood of entry we also assess the relative risk of entry and expansion compared to alternative investments. 398. In this case, even after a post-merger price increase, we consider that entry is unlikely to be profitable. This is because an independent community newspaper would face high costs of production compared with NZME and Fairfax but lower sales over which to spread those costs, and, therefore, a new entrant would be unlikely to be a cost-effective competitor. 399. This view was shared by all four of the independent publishers that we spoke with. Despite the respective scale that these independent publishers have, each expressed the view that they are currently unable to compete on price with NZME and/or Fairfax. We consider that this would remain the same post-merger. 400. In addition, the Commission notes that, in general, the sales of community newspaper advertising exhibit flat to declining revenues. Therefore, for new entry to occur a new publisher would have to win advertising customers from the merged entity. Given the aggressive pricing tactics observed by industry participants, we consider that this would be an added risk to a new entrant that would negatively affect the likelihood of entry. Conclusion 401. 206 For these reasons, the Commission’s preliminary view is that it is not satisfied that the proposed merger would not be likely to substantially lessen competition between NZME and Fairfax for the supply of community newspaper advertising in the 10 areas of overlap. Where expected profits are based on costs that include the opportunity cost of entry. 77 Competition between The New Zealand Herald and Fairfax’s community network 402. The proposed merger would result in NZME acquiring Fairfax’s network of community newspapers in Auckland (Auckland communities).207 Fairfax’s Auckland communities overlap with the NZME’s The New Zealand Herald. 403. Similar to other community newspapers, the Auckland communities are published for free on a weekly or bi-weekly basis and almost exclusively cover news that is local to the area that they service. The Applicants note that “by bundling up several [Auckland] community newspapers, an advertiser can get similar eyeball exposure to using the Herald”.208 404. The Commission assessed whether the aggregation of The New Zealand Herald and Fairfax’s Auckland communities would enable the merged entity to unilaterally increase the price of advertising in The New Zealand Herald and/or the Auckland communities. In particular, we have considered whether advertising across a bundle of the Auckland communities is substitutable for advertising in The New Zealand Herald. The Commission’s view 405. The Commission does not consider that The New Zealand Herald and the Auckland communities place a significant constraint on each other for advertising such that a substantial lessening of competition is likely. 406. None of The New Zealand Herald advertisers that we spoke with considered advertising in a group of the Auckland communities to be a substitute for advertising in The New Zealand Herald. These advertisers advised that readers of paid newspapers are more valuable, as they are more likely to be engaged (having purchased the newspaper) and are more likely to be from a higher-socio economic demographic. 407. Conversely, advertisers in the Auckland community newspapers seek to reach local readers in particular suburbs of Auckland. These local audiences are not as effectively reached through The New Zealand Herald as it covers the broader Auckland area, rather than focussing on particular suburbs. In addition, given that the Auckland communities are distributed for free, they target a different audience demographic than the paid The New Zealand Herald.209 408. For these reasons, the Commission is satisfied that the proposed merger is unlikely to result in a substantial lessening in competition in respect of the price of advertising in The New Zealand Herald and/or the Auckland communities. 207 208 209 Auckland City Harbour News, Central Leader, East & Bays Courier, Manukau Courier, North Harbour News, North Shore Times, Nor-West News, Papakura Courier, Western Leader, Rodney Times and Waiheke Marketplace. Application, at Appendix 13. Commerce Commission interview with The Cheesecake Shop (6 September 2016). 78 Section 5: Reader Markets 409. Post-merger, NZME and Fairfax would cease being independent competitors for the provision of news content across both their print and online products. If the merging Applicants are each other’s closest competitors in these markets, then the merger could provide the merged entity with increased market power. 410. Depending on the level of competitive constraint provided by other news providers, the merged entity could use any enhanced market power to either raise the price and/or reduce the quality for at least some of its print and online products. 411. The Commission has considered whether the proposed merger could result in a substantial lessening of competition in the following ways: 412. 411.1 a decrease in the quality of New Zealand online news content produced by the Applicants; 411.2 the introduction of a paid subscription model (paywall) for the stuff.co.nz and/or nzherald.co.nz sites; 411.3 an increase in subscription and/or retail prices for Sunday newspapers; and 411.4 a decrease in quality of Sunday newspapers and/or the community newspapers identified in paragraph 101.4 above. The Commission has also considered whether the merged entity’s incentive to exercise any market power by these methods may be tempered to some degree by the two-sided nature of the markets in which the Applicants operate. Because any increase in price or reduction in quality could result in fewer readers, or fewer readers of the type advertisers most value, such changes could reduce the value of the merged entity’s advertising inventory. Market definition 413. At a broad level, the Commission considers that there are reader markets for New Zealand news content. These markets can be further delineated by several characteristics, one of which is related to the demand from readers for locationspecific content at national, regional and local levels. 414. Another relevant characteristic is the degree to which this news is consumed across the various media platforms, ie, print, TV, radio and digital. The Commission has previously considered the traditional media platforms of print, TV and radio as distinct markets for the purposes of competition analyses; 210 however, the Applicants submitted that these distinctions are becoming increasingly blurred. 210 Fairfax NZ Ltd / Times Media Group Ltd, Decision 561, (14 October 2005), pg.13-14. 79 415. A further factor relevant to the Commission’s definition of the reader markets is the difference between the creation of content and the delivery of that content to consumers. 416. Each of these characteristics is explained in further detail below. Demand for location-specific content 417. There is an important location-specific aspect of the demand for news. In particular, there is a strong demand for national, regional and local content. Content that relates to areas outside of these locations is not substitutable for readers who want to read that specific content. 418. For instance, while there is a demand for international news, this is in addition to, and not substitutable for New Zealand news, including New Zealand perspectives on international news.211 Similarly, at regional or local levels, there is reader demand for news that is of local and regional interest for which news from another area is not substitutable.212 419. The importance of this demand for location-specific content is widely acknowledged throughout the industry. A Fairfax study of The Dominion Post and The Press reader preferences indicated that the primary topic of interest was national news, sought after by [ ]% and [ ]% of The Dominion Post and The Press readers, respectively.213 Further, local news was sought after by [ ]% and [ ]% of the readers of these publications, respectively. This study also stated:214 Newspaper readers ‘wants’ aren’t that different across the country… overall they are looking for detailed local news and involvement in their community… 420. In relation to meeting reader demand by creating content in-house or syndicating (purchasing) content from elsewhere, Murray Kirkness, New Zealand Herald digital editor, stated:215 On a regional sense it’s very difficult to buy content in. Even in a New Zealand-sense it’s very difficult to buy content in that is relevant to New Zealand audiences. … five years ago, no one would have thought that we would be competing with The Guardian. And we can obviously buy in similar content to The Guardian in terms of world news 211 212 213 214 215 Note that New Zealand news is not limited to news or events that occur domestically. For example “What does Brexit mean for Kiwis?” http://www.stuff.co.nz/business/81651887/the-great-oe-what-brexitmeans-for-kiwis. This article refers to an international event but provides a New Zealand-specific context that would not be produced by international media organisations. Fairfax presentation, Annex 26 - Communities Presentation. Trade Communications (November 2016), contained in a letter from Russell McVeagh (on behalf of Fairfax) to the Commerce Commission (12 July 2016). Fairfax presentation, Annex 68 - Project Meteor. Dom Post and The Press Reader Study (2 October 2015), attached to an email from Russell McVeagh (on behalf of Fairfax) to the Commerce Commission (14 September 2016). Ibid. Commerce Commission interview with NZME (15 June 2016). 80 etcetera… What we can’t do is buy in – the Bay of Plenty Times, can’t buy in Bay of Plenty Times specific [news] – you have got to pay to create that yourself. … on a slightly different scale, that’s the same even for The Herald say. If we looked at a pie chart of what content is created from our newsrooms as opposed to what we buy in, it would be something like 75% would be created by us. And then if you look at what our audience was, the digital analytics would tell you that it was about the same. The most valuable content to us is the stuff that is relevant to our audience, and by in large that is created by us. 421. Data indicates that around [ ] ([ ]%) of the approximately [ ] articles published on stuff.co.nz each month are produced by Fairfax, with the remaining [ ]% obtained from external sources, including overseas syndicated content.216 Similarly, [ ] articles published on nzherald.co.nz each month are produced by NZME.217 218 Further, data shows that of the 2.6 million people who access dedicated news and current event sites in New Zealand, 2.4 million access local New Zealand news.219 422. MediaWorks expressed the view that readers are primarily interested in local and hyper-local news content, with the key question that media firms typically need to consider when creating content being ‘how does this affect me [the reader]?’ 220 Given this characteristic of reader demand, MediaWorks considers that there is a distinct market for New Zealand news which is separate from the market for global news.221 423. These trends were supported by TVNZ which considered that “New Zealand news is a discrete service for which overseas news, blogs or general information services are not substitutes” and that there is a “distinct demand” for local content.222 This same sentiment was expressed in relation to local content produced by independent community newspaper publishers,223 which explained that readers of community newspapers desire news that is local to their community. Media platforms and content format 424. 216 217 218 219 220 221 222 223 Although traditional media platforms (print, TV, radio) are still used, a sizeable and increasing proportion of consumers are reading, viewing and listening to media content online and reducing their use of traditional media platforms. Based on average of May, June and July 2016. 14 September 2016 Fairfax Response to NZCC, attached to an email from Russell McVeagh (on behalf of Fairfax) to the Commerce Commission (14 September 2016). Based on average of May, June and July 2016. 16 September 2016 NZME Response to NZCC, attached to an email from Russell McVeagh (on behalf of NZME) to Commerce Commission (16 September 2016). However, NZME-created articles only account for around [ ]% of the [ ] articles published on nzherald.co.nz, as NZME [ ]. Nielsen Media Trends 2016: How New Zealanders consume newspapers, magazines, TV, radio & digital (2016) at page 12. Commerce Commission interview with MediaWorks (29 June 2016). Submission from MediaWorks to the Commerce Commission (30 September 2016) at [3.4]. Submission from TVNZ to the Commerce Commission (14 September 2016) at [5.4] and [5.6]. Commerce Commission interviews with Wellington Suburban Newspapers (28 September 2016) and River City Press (26 September 2016). 81 425. Fairfax research suggests that online news is now consumers’ primary news source.224 This study estimates that in a typical week the average New Zealander gets [ ]% of their news content from online sources, [ ]% from TV, [ ]% from traditional newspapers, and [ ]% from radio.225 426. The rise in online news consumption has occurred relatively recently. In 2015, 60% of the population read news online; this has increased from 25% in 2008.226 And, 2015 marked the first year that the number of people only reading online news exceeded the number of people only reading print news.227 Fairfax research indicates that this trend is continuing into 2016,228 with [ ]% of people surveyed stating that they have increased their consumption of online news, up from [ ]% in 2015.229 427. Conversely, while the daily reach of linear TV and live radio remains high (73% and 59% of New Zealanders respectively), the reach of both media platforms has decreased by 12% over the past two years.230 Similarly, the ownership or access to a TV and radio has also decreased over the same period.231 428. An advantage of online news compared to traditional media platforms is the ability for news websites to be updated regularly and not be confined to set bulletin or print times. This allows readers to access news at any time of the day and gives publishers the ability to engage with audiences in a dynamic way. MediaWorks explained that online news allows it to engage with its audience across the day, including the ability to track and understand what stories are popular and those which are not.232 429. As a consequence of these trends, New Zealand’s main media firms have changed the structure of their businesses and content publication processes to reflect the importance of digital media. Within the last two years NZME, Fairfax, TVNZ and MediaWorks have shifted to ‘digital-first’ operations to deliver content to consumers. This means that the each of the main media firms typically publish their stories online first, rather than waiting to break the story on TV, print or radio. 430. For NZME and Fairfax this means that they produce news for online consumption first and then select articles to be included in print in newspapers the next day. 224 225 226 227 228 229 230 231 232 Fairfax presentation, Annex 67 – Fairfax Media. Mini Brand Health Study. Wave 4 (June 2016), attached to an email from Russell McVeagh (on behalf of Fairfax) to the Commerce Commission (14 September 2016). Respondents were asked: In a typical week, what proportion of your news do you get from…? For each item listed please allocate a % figure to a total of 100%. The remaining [ ]% is recorded as ‘Other’. Nielsen Media Trends 2016: How New Zealanders consume newspapers, magazines, TV, radio & digital (2016) at page 11. Ibid. Fairfax presentation, Annex 67 - Fairfax Media. Mini Brand Health Study. Wave 4 (June 2016), attached to an email from Russell McVeagh (on behalf of Fairfax) to the Commerce Commission (14 September 2016). Respondents were asked: Compared to 6 months ago, are you reading each of the below MORE or LESS frequently (or the SAME)…?. [ ]% answered more and [ ]% answered less, resulting in a net share of [ ]% consuming more. These figures relate to all content, including not only news and information but also entertainment. Glasshouse Consulting, “Where are the audiences?” commissioned by NZ on Air (2016) at slide 27. Glasshouse Consulting, “Where are the audiences?” commissioned by NZ on Air (2016) at slide 64. Commerce Commission interview with MediaWorks (29 June 2016). 82 Similarly, TVNZ and MediaWorks publish stories online first and then use these stories for the basis of its TV and radio broadcasts. 431. However, TVNZ maintained that its online content is ancillary to its core broadcasting services,233 and considered that its 6pm TV news programme plays a different role to its online offering; summarising the day’s events and providing the “story behind the story”.234 432. We also observe that the main media firms have relative strengths in producing content in traditional news formats, such as video (TVNZ, MediaWorks), audio (MediaWorks, NZME, RNZ) and text (Fairfax, NZME), and these strengths are reflected in their online offerings. For example, TVNZ’s 1 News Now website is heavily focused on news in a video format. 433. From a reader-perspective, different content formats are likely to have differing strengths and weaknesses in their effectiveness of conveying content and for reaching readers. At various times and for various content, the different formats can be either substitutes, complements, or neither substitutes nor complements. 434. For instance, video footage of a spectacular news incident may be preferred by consumers to text-based or audio-based stories because of its ability to convey this incident. For other stories, or at certain times of the day such as the morning and evening commuting periods, audio content may be preferred as it does not require visual attention. Similarly, some stories may be most effectively conveyed through text, such as a large 600 word article.235 435. From a supply-side perspective, different content formats also have different advantages or disadvantages. For example, online news providers are able to collect more extensive data on reader engagement (for example, views and ‘likes’) and tailor both content and the promotion of content (for example, via site placement and/or notifications) to better suit consumers. 236 The role of print 436. 233 234 235 236 Despite the trends of increased digital consumption, leading to falling circulation amongst daily and weekend newspapers, we consider that print continues to play an important role in New Zealanders’ media consumption. Submission from TVNZ to the Commerce Commission (14 September 2016) at [6.17]. Commerce Commission interview with TVNZ interview (27 June 2016). Commerce Commission interview with E tū interview (2 August 2016). Submission from MediaWorks to the Commerce Commission (30 September 2016) at [3.3]. 83 Table 6: Yearly circulation of metropolitan newspapers, 2013-2015 2013 2014 2015 New Zealand Herald [ ] [ ] [ ] Waikato Times [ ] [ ] [ ] Dominion Post [ ] [ ] [ ] Christchurch Press [ ] [ ] [ ] Source: NZME and Fairfax data 437. There remains a sizeable group of readers for whom online and print platforms are not effective substitutes. Despite the majority of content contained within the Applicants’ newspapers being available free of charge on stuff.co.nz and nzherald.co.nz, the total circulation of daily newspapers is around 460,000 per day.237 This figure is equivalent to a quarter of all households in the country.238 438. Further, the large number of readers who are willing to pay for daily or weekly newspapers regardless of the provision of free news and information content both online and via TV and radio broadcasts indicates that, for a significant proportion of the population, content published on other platforms is not necessarily a close substitute for print products. 439. There are a number of potential reasons why some readers may prefer newspapers to either online, TV or radio platforms, including a preference to read the news in hardcopy at a time and place that is convenient for the reader, the age demographics of the reader and/or access to internet devices. 440. As indicated in a News Works report, 72% of New Zealanders aged 45 to 69 read newspapers daily.239 Further, in 2015, only 45% of adults over 55 years old had smartphones.240 441. Nevertheless, other readers continue to switch between print and online platforms. However, as outlined below, notwithstanding the number of readers that switch, we do not consider newspapers and online news to be in the same market for the purposes of our analysis. We consider that both newspapers and online news continue to have different characteristics that appeal to readers. 442. In any event, our analysis would not differ because the majority of daily newspapers in New Zealand are also owned by the Applicants. 237 238 239 240 See table 17. Statistics NZ estimates that there are 1.7 million households (Dwelling and Household Estimates, June 2016 quarter). Because there may be more than one reader per household/organisation, the effective reach of daily newspapers is likely to be higher than one and four individuals. Nielsen reports that in 2016, 53% if New Zealanders read a daily newspaper each month. Nielsen Media Trends 2016: How New Zealanders consume newspapers, magazines, TV, radio & digital (2016) at page 9. News Works presentation, ‘The growing importance of the over 45s’ (2016) at slide 10. Research New Zealand, Research New Zealand Special Report: use of smartphones (2015) at page 5. 84 Production versus distribution of online New Zealand news 443. We distinguish between the upstream market for the production of New Zealand news, and downstream markets for the distribution of this content. 444. The upstream production market is characterised as having a small number of large, New Zealand-based mainstream news producers – Fairfax, NZME, TVNZ, MediaWorks and RNZ – and a larger number of smaller producers such as NBR, BusinessDesk and Allied Press. These producers are also vertically integrated in downstream distribution markets via their own media platforms, ie, websites, print publications, TV broadcasts, and radio stations. 445. However, the downstream markets for the distribution of New Zealand news also include other firms that are not involved in the production of New Zealand news, such as third party aggregators (eg, Google, Yahoo, Buzzfeed) and social media networks (eg, Facebook, Twitter). These firms effectively operate as intermediaries between New Zealand news producers and readers. 446. We acknowledge the Applicants’ submission that they regard social media platforms, like Facebook, as a “significant competitor for advertising expenditure and a key competitor for consumer attention”.241 However, the Commission notes that Facebook (and other social media platforms) do not hire journalists nor do they produce New Zealand news content. For this reason, we consider that they are news aggregators rather than competitors in the production of New Zealand news content. 447. For these reasons, we consider that the production and distribution of New Zealand news are separate markets. We do not specifically consider the market for distribution of news further as part of our analysis, although we note that the loss of advertising revenues to third party distributors may reduce the incentive for new entrants in the production of New Zealand news. The Commission’s approach to reader markets 448. The Commission acknowledges that New Zealanders are increasingly sourcing their news from multiple formats and that digital platforms have allowed traditional print, video and audio news producers to provide their content online. 449. However, we consider that each format, including print, continues to have its relative strengths and consumer appeal. Further, we also consider that content provided in particular formats is likely to be more of a substitute than content published or broadcast via other media platforms. This is because the different platforms have different characteristics that consumers find appealing. 450. As a result, we consider that online content sites producing similar content are likely to have a stronger price and quality constraining effect on each other than the constraint provided by other traditional media platforms such as TV or radio, even if 241 Email from Russell McVeagh (on behalf of the Applicants) to the Commerce Commission (1 November 2016). 85 they produce similar content. We also consider that any substitutability is likely to be stronger one way than the other. Specifically, given news consumption trends, online news products are likely to provide a greater constraint on traditional media products than the other way around. 451. Therefore, for the purpose of analysing the competitive effects of the proposed merger, we consider online New Zealand news to be a separate market.242 Similarly, although print competes with other media, particularly digital, we consider that the closest substitute for a newspaper in any given area is likely to be other newspapers that focus on similar content in that area. 452. Based on the similar content and geographic areas covered by NZME and Fairfax’s Sunday newspapers and the 10 community newspapers listed in paragraph 101.4, the Commission has considered whether the proposed merger would be likely to result in a reduction of quality in these newspapers and, also, for the Sunday newspapers, an increase in the subscription/cover price paid by readers.243 453. As part of our analysis for both online and newspaper markets, we have considered the extent to which other media platforms also producing New Zealand news would provide a constraint on the merged entity. Competition analysis 454. Post-merger, NZME and Fairfax would cease being independent competitors in the reader markets for: 454.1 online New Zealand news; 454.2 Sunday newspapers; and 454.3 community newspapers in 10 areas of overlap. 455. The following two sections assess whether the loss of competition in these markets would be likely to result in a substantial lessening of competition. In particular, we assess whether merged entity could use any enhanced market power to either raise the price and/or reduce the quality of New Zealand news in these markets. 456. Furthermore, the Commission notes that given NZME and Fairfax’s digital-first strategies, a reduction in the quality of online New Zealand news resulting from the proposed merger may also affect the quality of New Zealand news that is published in NZME and Fairfax’s regional and metropolitan newspapers, despite the minimal overlap across these newspapers. 242 243 Given that the Applicants and their competitors for the production of New Zealand news are also vertically integrated into the distribution of New Zealand news, we have referred to the market for online New Zealand news more generally. For the avoidance of doubt, we do not consider that third party news aggregators and social media networks, such as Facebook and Google, are competitors in the market for online New Zealand news. We do not consider that it is likely that the merged entity would introduce cover prices for the community newspapers in the 10 areas of overlap. This is supported by the fact that currently almost all community newspapers are distributed for free, regardless of the extent of competition faced. 86 The online New Zealand news market 457. 458. In this section, we assess whether the proposed merger would be likely to result in a substantial lessening of competition in the market for online New Zealand news. In undertaking this assessment, we considered: 457.1 the existing competition in the market for online New Zealand news, in particular the extent to which the Applicants are close competitors and whether existing competitors would constrain the merged entity; 457.2 whether entry or expansion of other existing or new producers of online New Zealand news is likely; 457.3 whether the merged entity would be likely to have an incentive to exercise any market power it would obtain to introduce a paywall over at least some content and/or reduce quality; and 457.4 whether the two-sided nature of the online market would constrain any price increase and/or quality reduction. In undertaking this assessment, the Commission has obtained views from a number of interested parties including journalists, competitors, advertisers, other organisations and individuals, including industry experts. Existing competition 459. To determine whether the proposed merger would increase the Applicants’ market power in the provision of online New Zealand news, the Commission considered the extent to which the Applicants compete against each other and the position of the Applicants vis-à-vis other producers of online New Zealand news. Applicants’ views 460. The Applicants considered the relevant market to be print and online news and information services,244 and they did not submit specifically on competition in a market for online New Zealand news. The Applicants generally suggested that the New Zealand market for print and online news would remain “highly competitive” post-merger245 and that the merged entity would face direct competition from other existing rival producers of content.246 461. Nevertheless, the Applicants suggested that TVNZ, MediaWorks and RNZ are significant competitors in the online news market. The Applicants’ pointed to the extensive re-structuring and product development that all three of these companies have undertaken to suggest that they are becoming increasingly strong competitors online.247 244 245 246 247 Application at [14.4]. Ibid. Application at appendix eight; Fairfax/NZME Response to SOPI submissions (29 July 2016) at [16] to [25]. Application at [14.5(a)] and [14.5(b)]; Fairfax/NZME Response to SOPI submissions (29 July 2016) at [18]. 87 462. In addition, the Applicants listed a number of other New Zealand-based rivals with an online news presence. For example, Allied Press (www.odt.co.nz), Scoop.co.nz, Bauer’s online platforms for The New Zealand Listener, Metro and North & South, The National Business Review (www.nbr.co.nz) and Hive News (HiveNews.co.nz).248 Journalists’ views 463. The Commission spoke with E tū (a union which represents journalists in New Zealand) and a number of its members, including current and former employees of NZME and Fairfax.249 The view of the E tū journalists was that NZME and Fairfax each compete for readers’ attention by trying to: 463.1 be the first to cover ‘commodity’ news; 463.2 break new stories; and 463.3 provide alternative viewpoints on existing stories. 464. The E tū journalists considered that NZME and Fairfax are each other’s closest competitors for New Zealand news. These journalists explained that while radio, TV and other news websites compete with NZME and Fairfax to some extent, they are less of a constraint than what NZME and Fairfax are on each other. A key reason for this was the frequency by which stuff.co.nz and nzherald.co.nz are updated compared with TVNZ, MediaWorks and RNZ’s media platforms (whether online, TV or radio). 465. For example, one journalist said that TVNZ’s and MediaWorks’ 6pm news slots are “a different sort of competition” and their websites “aren’t as up-to-date and they’re not throwing stories up on their website every 5 seconds like we [Fairfax] are”. 250 Another journalist, referring to radionz.co.nz, said “I can never think of an instance when I’ve been told by my chief reporter, “oh there’s a story on Radio New Zealand you need to match”.”251 466. Further, the E tū journalists considered that there are limitations in what news can be covered on radio compared with text-based news. In regards to radio, one journalist said “we’re just not in the same market really in terms of a 600 word story compared to a … how many seconds news bulletin”.252 467. One area where the E tū journalists suggested that the Applicants’ rivals may be stronger was political coverage. One journalist said that the Parliamentary press gallery is one of the more competitive environments, noting TVNZ’s strong presence.253 248 249 250 251 252 253 Application at [14.5]. Commerce Commission interview with E tū (2 August 2016). Ibid. Ibid. Ibid. Ibid. 88 Competitors’ views 468. All three main online competitors considered that the merged entity either would be dominant or would have significant scale advantages. MediaWorks submitted that the merged entity would become the “dominant digital news organisation”, noting that it would account for 95% of online page impressions in the market.254 Likewise, TVNZ considered that the proposed merger would create a vast disparity between it and the merged entity in terms of audience reach.255 469. While TVNZ and MediaWorks said that their respective websites compete with NZME and Fairfax to break stories, neither considered that their TV news platforms compete in this way. 470. MediaWorks considered that its TV news platform is complementary to its online news offering. It advised that newshub.co.nz breaks stories during the day, and then these stories are repurposed for the 6pm TV news as a “different beast to what you might hear or see online during the day”.256 Similarly, TVNZ described its 6pm TV news as “the story behind the story”, whereas its online news is about immediacy. 257 471. Further, TVNZ said that its core offering is its broadcasting services, with its online content being an ancillary offering. 258 To this extent, TVNZ said “[ ].”259 472. RNZ indicated that it was putting a lot of “energy” into its website as well as distribution of its content on social media platforms.260 However, it considered that it was on a “different path” to the privately-owned media publishers. Because it is publically-funded, RNZ explained that it does not need to chase popular stories to attract eyeballs for advertisers. Rather, it can take a more “pure” news approach, covering stories that reflect its obligations as a publically-funded broadcaster.261 473. Competitors also considered that the Applicants’ strengths are their comprehensive national coverage that is derived from their newspapers around New Zealand. MediaWorks expressed the view that readers are primarily interested in local and ‘hyper-local’ content and has journalists located in Auckland, Wellington, Christchurch and Dunedin. 262 254 255 256 257 258 259 260 261 262 Based on total page impressions in July 2016 across newshub.co.nz, tvnz.co.nz/news, stuff.co.nz and nzherald.co.nz. MediaWorks submission at [4.1] (30 September 2016). We note that rnz.co.nz was not included in MediaWorks’ analysis. TVNZ Submission on SOPI 12 September 2016 at [6.15]. Commerce Commission interview with MediaWorks (13 October 2016). Commerce Commission interview with TVNZ (27 June 2016) Submission from TVNZ to the Commerce Commission (14 September 2016) at [6.17]. Ibid. We note that since 28 July 2016, RNZ has partnered with Fairfax to disseminate its content on stuff.co.nz. Commerce Commission interview with RNZ (12 July 2016). Commerce Commission interview with MediaWorks (29 June 2016). 89 474. TVNZ said that it could [ ]263 TVNZ employs reporters in Auckland, Wellington, Christchurch and Dunedin and only covers local stories if they become national news.264 We also note that news from TVNZ’s 1 News Now has a dedicated page on stuff.co.nz.265 475. Similarly, RNZ advised that it is only publishes stories that are of national interest, in accordance with its obligations as a publically-funded broadcaster.266 Therefore, it does not cover stories that are of purely regional or local interest. RNZ indicated that the vast majority of its staff are located in Wellington and Auckland.267 The Commission’s view 476. The Commission acknowledges that there is some validity to the Applicants’ claims that the existence of a range of rivals provides some constraint. However, the Commission considers that the Applicants are each other’s closest competitors for New Zealand online news and considers it unlikely that existing competitors have sufficient scale to act as a significant constraint post-merger. 477. In reaching these views, we considered: 477.1 the extent of the Applicants’ production of New Zealand news relative to their competitors; 477.2 use of the Applicants’ websites relative to other New Zealand news websites; and 477.3 the Applicants’ internal documents. The Applicants’ production of New Zealand news 478. The Commission is persuaded by the views of existing competitors as to their position in the market relative to the Applicants. 479. Measured by the number of stories published, Fairfax and NZME are the two largest New Zealand producers of online news by some margin. As outlined in Table 7, excluding externally sourced content (ie, syndicated stories), Fairfax and NZME produce more than three times the number of stories than TVNZ and MediaWorks.268 263 264 265 266 267 268 Commerce Commission interview with TVNZ (27 June 2016). Ibid. http://www.stuff.co.nz/national/1-news-now Commerce Commission interview with RNZ (12 July 2016). Ibid. MediaWorks was unable to provide the Commission with its article numbers split between syndicated and MediaWorks created content. Therefore, we have conservatively assessed all these as MediaWorks’ articles. 90 Table 7: Production of online New Zealand news Firm Total online articles Self-created published/month269 online articles/month270 Editorial and journalistic staff Fairfax [ ] [ ] [ ]271 NZME [ ] [ ] [ ]272 TVNZ [ ] [ ] [ ]273 MediaWorks [ ] [ ] [ ]274 Source: Fairfax, NZME, TVNZ and MediaWorks data. 480. NZME and Fairfax also hire relatively large numbers of editorial and journalist staff and spend considerable resources on the production of original content. The merged entity would control the largest newsroom and national network of reporters and journalists in the country, employing nearly twice the number of editorial staff than the three next biggest mainstream media organisations combined. 481. Based on current numbers, the merged entity would have [ ]+ editorial and journalistic staff compared with TVNZ which has [ ] journalists (described as a “crude make-up of producers and reporters”), MediaWorks which has [ ] (including “reporters, producers, editors, camera operators, directors, graphics experts and others”) and RNZ which has a newsroom of about [ ] staff.275 482. The merged entity would also have an extensive regional presence compared to TVNZ, MediaWorks and RNZ. Apart from a few independent media publishers such as the Otago Daily Times and the Ashburton Guardian, the evidence of the competing firms is that they do not have the same regional focus. This means that the merged entity is able to offer more comprehensive news coverage to readers. 483. As outlined above, the Applicants also referred to a number of smaller or niche news websites. The Commission does not consider that these websites would act as a competitive constraint on the merged entity. Websites such as the nbr.co.nz and scoop.co.nz produce a much smaller amount of original news than the Applicants. 269 270 271 272 273 274 275 Based on three month average during May, June, July 2016. Based on three month average during May, June, July 2016. 3117128 Staff Costs 23 March 2016 – v1.XLSX, attached to an email from Russell McVeagh (on behalf of Fairfax) to the Commerce Commission (4 July 2016). 3117363 Employee summary by function - v1.XLSX, attached to an email from Russell McVeagh (on behalf of NZME) to the Commerce Commission (4 July 2016). Described by TVNZ as “a crude make-up of producers and reporters”. Email from TVNZ to the Commerce Commission (5 October 2016). MediaWorks said that the Newshub newsroom is staffed by [ ] people, including reporters, producers, editors, camera operators, directors, graphics experts and others. Email from MediaWorks to the Commerce Commission (30 September 2016). Commerce Commission interview with Radio NZ (12 July 2016). 91 Furthermore, these smaller producers focus their content towards relatively niche groups of readers (for example NBR focuses on business news) and do not attract large, mainstream New Zealand audiences. 484. Further, the Applicants have pointed to limited examples of international news providers publishing New Zealand news, for example The Guardian publishes some New Zealand news from freelance journalists.276 Given the minimal amount of New Zealand news these providers produce, the Commission does not consider that these websites would act as a competitive constraint on the merged entity. Usage data 485. The Commission considers that online usage data supports the view that the Applicants are each other’s closest competitors in the delivery of New Zealand online news compared to TVNZ, MediaWorks and RNZ. 486. Over the previous year, the Applicants had an average combined reach on their websites of approximately 2.2 million unique online visitors per month (as outlined in Figure 1 below).277 This constitutes around 55% of New Zealand’s population aged 10 years and over.278 In contrast, the average monthly reach of the next biggest online New Zealand news brand, MediaWorks’ Newshub, is 12.6%. The equivalent figures for the TVNZ-news279 and RNZ websites are 10.7% and 8.8%, respectively. 276 277 278 279 Submission from Russell McVeagh (on behalf of the Applicants) (10 October) at page 18. In relation to Figures 1-3, an individual is counted as a viewer if they visited a website at least once in that month. Source: Nielsen CMI Fused Q2 15 - Q1 16 June 16 TV/Online –FFADNT. In relation to Figures 1-3, TVNZ refers to TVNZ-News which represents TVNZ’s own news classification or ‘custom rollup’ in the Nielsen CMI database. 92 NZ population 10 years+ (millions) Figure 1: Average monthly reach of New Zealand web brands, Q2 2015 – Q1 2016 (browsers only)280 2.2 1.9 1.6 0.5 0.4 0.3 0.2 0.2 0.2 Source: Nielsen CMI Fused Q2 15 - Q1 16 June 16 TV/Online -FFADNT 487. It is also useful to consider what proportion of nzherald.co.nz and stuff.co.nz readers access other New Zealand news websites, ie, multi-source. Figure 2 below indicates that of the readers that view stuff.co.nz at least once per month, two-thirds of these readers also visit nzherald.co.nz at least once a month.281 Less than one-quarter of these stuff.co.nz readers also visited the Newshub website at least once per month. Around one-fifth of stuff.co.nz readers also visited TVNZ-news website. 488. Similarly, Figure 3 below indicates that that 80% of readers who visit the nzherald.co.nz site at least once per month also visit stuff.co.nz. The next highest New Zealand news website visited by nzherald.co.nz readers is Newshub with 28% of readers multi-sourcing with this site. TVNZ-news and RNZ websites are next at 22% and 11%, respectively. 280 281 The merged entity total is calculated as the stuff.co.nz total plus 0.2*the nzherald.co.nz total to account for 80% of nzherald.co.nz viewers also visiting stuff.co.nz. In relation to Figures 2-3, the data does not account for frequency or volume of individuals’ consumption. Because of this, it likely provides a conservative estimate for highly consumed websites and likely overstates “true” multi-sourcing. 93 % stuff.co.nz audience 10 years+ Figure 2: Percentage of stuff.co.nz users that access other web brands at least once a month, Q2 2015 – Q1 2016 (browsers only) 66.7% 24.1% 20.1% 10.8% nzherald.co.nz Newshub website TVNZ - News Radio New Zealand 7.8% odt.co.nz 5.5% 4.2% NewstalkZB website Scoop Source: Nielsen CMI Fused Q2 15 - Q1 16 June 16 TV/Online -FFADNT Figure 3: Percentage of herald.co.nz users that access other web brands at least once a month, Q2 2015 – Q1 2016 (browsers only) % herald.co.nz audience 10 years+ 80.0% 27.5% 22.3% 10.9% Stuff.co.nz Newshub website TVNZ - News Radio New Zealand 8.5% odt.co.nz 5.4% 4.4% NewstalkZB website Scoop Source: Nielsen CMI Fused Q2 15 - Q1 16 June 16 TV/Online -FFADNT 489. Further, because monthly reach data only provides a limited picture of actual multisourcing, the Commission has also obtained ‘volume-based’ statistics regarding online New Zealand news publishers. These include measures such as total time spent on each site, average time spent per reader/viewer, session frequency, total page impressions (clicks) for each site, and average page impressions per reader/viewer. 94 490. Figure 4 below shows the total time spent on various New Zealand news websites accessed via web browsers on PCs and mobile devices (but not via apps). It shows that the total combined time spent on the Applicants’ news websites in the month of September 2016 was around 5.31 million hours. The next highest is the TVNZ website282 on which readers spent less than 0.53 million hours.283 Figure 4: Total time spent on New Zealand news websites, September 2016 (browsers only) total time spent - browser, September 2016 (million hours) Sun Allied FFX + Media Press RNZ MediaWorks TVNZ NZME NZME FFX 0 1 2 3 stuff.co.nz 5 2.26 0.01 merged entity 5.31 tvnz.co.nz newshub.co.nz tv3.co.nz 0.53 0.12 0.01 radionz.co.nz 0.08 odt.co.nz 0.05 sunlive.co.nz 0.04 Source: Nielson Online Ranking Report (market: New Zealand Domestic Traffic By Brand (website data), period: Monthly, 01/09/16 - 30/09/16) 491. 282 283 6 3.04 nzherald.co.nz newstalkzb.co.nz 4 Figure 5 shows the total page impressions (ie, number of pages viewed) obtained from online New Zealand news websites during September 2016. This indicates that the total combined number of page impressions obtained by the Applicants (285million) is over 15 times greater than those obtained by the next largest websites, TVNZ (18million). In relation Figures 4-7, TVNZ refers to the entire tvnz.co.nz website, not just news content. In relation to Figures 4-5, the Commission is not aware of any factor that would cause the month of September 2016 to significantly differ from other months of the year. 95 Figure 5: Total page impressions (page views) on New Zealand news websites, September 2016 (browsers only) page impressions - browser, September 2016 (millions) Allied Sun FFX + Press Media RNZ MediaWorks TVNZ NZME NZME FFX 0 50 100 150 stuff.co.nz 250 107 1 merged entity 285 tvnz.co.nz newshub.co.nz tv3.co.nz 300 177 nzherald.co.nz newstalkzb.co.nz 200 18 5 1 radionz.co.nz 5 sunlive.co.nz 4 odt.co.nz 3 Source: Nielson Online Ranking Report (market: New Zealand Domestic Traffic By Brand (website data), period: monthly, 01/09/16 - 30/09/16) 492. 284 Figure 6 shows the number of unique browsers (in millions) that each website attracts on a weekly basis. A unique browser is a proxy for the number of individual devices from which a website is accessed. Fairfax and NZME both attract the largest number of unique browsers, over 4 million and 3 million, respectively. TVNZ is the next highest with less than 1 million.284 In relation to Figures 6, the Commission is not aware of any factor that would cause the week of 19-25 September 2016 to significantly differ from other weeks of the year. 96 Figure 6: Total unique browsers (individual devices) on New Zealand news websites, 19-25 September 2016 (browsers only) unique browsers, week 19 - 25 September 2016 (million) Fourth Sun Allied Estate Media Press RNZ MediaWorks TVNZ NZME FFX 0 1 2 3 stuff.co.nz 3.2 0.1 tvnz.co.nz newshub.co.nz tv3.co.nz radionz.co.nz 5 4.1 nzherald.co.nz newstalkzb.co.nz 4 0.8 0.5 0.1 0.3 odt.co.nz 0.2 sunlive.co.nz 0.1 nbr.co.nz 0.1 Source: Nielson Online Ranking Report (market: New Zealand Domestic, Traffic By Brand (website data), period: weekly, 19/09/16 - 25/09/16 493. Figure 7 below shows the average number of hours each reader spends on the major online New Zealand news websites per month. The merged entity would capture around 3.5 hours per month from each viewer, on average. In contrast, the average viewer time of the next biggest New Zealand news website, Newshub, is approximately 42 minutes per minutes. 97 Figure 7: Average time per person (hours) spent on New Zealand news websites, August 2016 (browsers only)285 average time per person in August 2016 (hours) Allied Sun FFX + RNZ Press Media TVNZ MediaWorks NZME NZME FFX 0 1 2 stuff.co.nz 1.4 0.1 merged entity 3.8 Newshub TV 3 0.7 0.1 TVNZ 0.7 Sunlive 0.6 odt.co.nz Radio New Zealand 4 2.3 nzherald.co.nz NewstalkZB 3 0.5 0.2 Source: Nielson NOL Standard Metrics, frequency - monthly, country - New Zealand, audience - total, dataset - surfing, measurement period - August 2016 494. It is important to note that the data in Figures 1-7 do not include reach and consumption of content via the various media organisations’ mobile apps. However, as illustrated in Figures 8 and 9 below, consumption of New Zealand news via apps follows a similar trend to the consumption of New Zealand news via browsers.286 Figure 8: Total unique browsers (unique visitors) on New Zealand news apps, September 2016287 [ ] Source: Fairfax, NZME, MediaWorks and TVNZ data 285 286 287 In relation to Figure 7, the Commission is not aware of any factor that would cause the month of August 2016 to significantly differ from other months of the year. In relation to Figures 8 and 9, the Commission is not aware of any factor that would cause the month of September 2016 to significantly differ from other months of the year. Note: MediaWorks’ data was provided as a three-month average between July – September 2016 and includes both domestic and international users. Fairfax, NZME and TVNZ data relates only to domestic users in September 2016. 98 Figure 9: Total page views of New Zealand news apps, September 2016288 [ ] Source: Fairfax, NZME, MediaWorks and TVNZ data The Applicants’ internal documents 495. As well as usage statistics, the Applicants also provided a number of internal documents which support the proposition that they are each other’s closest competitors for online New Zealand news. 496. For instance, as described below, [ ]. Conclusion 497. Based on the data collected, the views of industry participants and the Applicants’ own internal documents the Commission’s considers that the Applicants are each other closest competitors in a market for New Zealand online news. 498. We consider that the Applicants are likely to place a substantial constraint on each other in the online New Zealand news market and this would likely be lost as a result of the merger. We also do not consider that existing competitors are sufficiently well placed to replace the competition that will be lost as a result of the merger. 499. In particular, the reach of these competitors is not as comprehensive as the merged entity. These competitors would have to grow significantly to have the same presence as the merged entity in the online New Zealand news market. Entry or expansion by online New Zealand news producers 500. To further evaluate whether the proposed merger would provide the merged entity with enhanced market power, the Commission has considered whether either potential expansion by existing rivals or new entry could constrain this market power. Applicants’ views 501. The Applicants suggested that barriers to expansion for online news are low and that TVNZ, MediaWorks and RNZ would continue to grow their online businesses.289 502. The Applicants submitted that if TVNZ and MediaWorks wish to grow their digital audience further, “they simply need to continue to produce attractive content that 288 289 Ibid. Submission from Russell McVeagh (on behalf of the Applicants) (29 July 2016). 99 engages New Zealanders.”290 In addition, the Applicants submitted that TVNZ, MediaWorks and RNZ have recently made “significant new investment” in their online news presence and are achieving “significant audience growth in New Zealand”.291 503. The Applicants also submitted that following entry into Australia of a number of international news organisations such as the BBC, The Guardian, the Huffington Post and Buzzfeed would inevitably expand into New Zealand.292 Competitors’ views 504. All of the main media firms identified by the Applicants suggested that there were significant barriers to entry in the online news market, given both the costs of running a national news organisation and the creation of content. 505. MediaWorks said that providing online news coverage on a national and local basis requires substantial scale and significant investment.293 It said that the Newshub newsroom employs more than 200 people across Auckland, Wellington, Christchurch and Dunedin at an annual cost of around $[ ] per annum.294 506. Furthermore, [ ].295 507. TVNZ was of the view that there are significant costs in setting up news gathering operation, from building content management systems to hiring journalists, producers and camera crews.296 TVNZ said that the business case to make the investment needed to develop the requisite level of online capability is dependent on the advertising revenue the platform is able to generate, which in turn is dependent on the audience reach.297 508. [ [ [ 290 291 292 293 294 295 296 297 298 ] ]. Rather, it ][ ][ ]”.298 Submission from Russell McVeagh (on behalf of the Applicants) (10 October 2016) at [23]. Ibid. at [24]. Application at [9.10]. Submission from MediaWorks to the Commerce Commission (30 September 2016) at [4.3]. Ibid. Commerce Commission interview with MediaWorks (29 June 2016). Submission from TVNZ to the Commerce Commission (14 September 2016) at [5.11]. Ibid. at [6.12]. Email from MediaWorks to the Commerce Commission (30 September 2016). 100 509. Similarly, TVNZ advised that it is [ [ 299 Further, TVNZ said [ ] and the ]. ] 300 510. RNZ held a slightly different view in relation to the costs required to set up an online news gathering operation. It advised us that its website cost $[ ] and indicated that “the digital world is dirt cheap compared to traditional broadcast”.301 However, RNZ agreed that there are significant costs involved with content creation and it indicated that it had [ ].302 The Commission’s view 511. The Commission does not consider that expansion by existing competitors or entry by new competitors on a sufficient scale would be likely to constrain the merged entity. 512. The Commission accepts that the creation of content is expensive. We note that in 2015, Fairfax’s own editorial costs were around $[ ].303 For TVNZ or MediaWorks to increase their coverage to the extent of the merged entity would require significant investments. These would also be sunk costs, increasing the risks associated with expansion. 513. In addition, creating additional content is not necessarily profitable. As outlined above and by the Applicants, media firms need to be able to attract readers to their website to monetise their advertising inventory. As New Zealand is a small market that is dominated by established media brands (as indicated by the usage data above), investment in brand development to attract readers and advertisers to compete against the merged entity is inherently risky. 514. [ Further, [ 299 300 301 302 303 304 305 ].304 ].305 Commerce Commission interview with TVNZ (27 June 2016). Ibid. Commerce Commission interview with RNZ (12 July 2016). Ibid. 3117115 Summary P&LCalendar2105 updated – vi, attached in an email from Russell McVeagh (on behalf of Fairfax) to the Commerce Commission (4 July 2016). Commerce Commission interviews with MediaWorks (29 June 2016) and TVNZ (27 June 2016). Commerce Commission interviews with MediaWorks (29 June 2016). 101 515. While the Applicants provide some Australian examples,306 we do not consider they point to any new or potential entry from any international news organisations providing New Zealand news other than Vice Media and freelance journalists.307 516. This is not to say that the Commission considers further expansion by existing firms or new entry to be unforeseeable.308 Rather, any new entry would be unlikely to be of sufficient extent, as per the LET test, to be an effective constraint on the merged entity. Conclusion 517. Given the current strength of the Applicants’ offerings and the lack of constraint from existing or potential competitors, the Commission considers that the merged entity would be likely to have increased market power in the online New Zealand news market as a result of the merger. We have considered below whether this increase in market power would be likely to result in a substantial lessening of competition in reader markets through: 517.1 increased prices via the introduction of a paywall; and/or 517.2 reducing the quality of news that is produced. Introduction of a paywall 518. The Commission considered whether the merged entity would be likely to increase the price of accessing online New Zealand news by introducing a paywall to stuff.co.nz and/or nzherald.co.nz. Applicants’ view 519. The Applicants asserted that consumers are unwilling to pay for online newsand information309 and that the merger would not alter the constraints the Applicants have faced to date in implementing a paywall. 520. The Applicants submitted that it is not competition between stuff.co.nz and nzherald.co.nz that has prevented either party from implementing a paywall. Instead, it is:310 520.1 306 307 308 309 310 the need for online news/information providers to attract attention to their content, via social media and other intermediated platforms, in a digital world where there is almost limitless competition for consumers attention (and therefore limitless competition for advertisers expenditure); Submission from Russell McVeagh (on behalf of the Applicants) (10 October 2016) at [37]. Ibid. [38]. For instance, the Applicants have referred to the potential entry of a new domestic news supplier Mark Jennings and Tim Murphy. Application from Fairfax/NZME (27 May 2016) at [16.1]. Authorisation Application from Fairfax/NZME (27 May 2016) at [9.20]; Submission from Russell McVeagh (on behalf of the Applicants) (29 July 2016) at [55(c]). Submission from Russell McVeagh (on behalf of the Applicants) (29 July 2016) at [54] – [59]. 102 520.2 consumers (un)willingness to pay for general news/information content in the context of significant competition; and 520.3 the impact on advertising revenues from the introduction of any paywall and, therefore, the inevitable and significant reduction in readers that would result. Commission’s view 521. The Commission considers that the evidence of the degree to which the Applicants currently constrain each other, and the lack of constraint that would be provided by existing and potential competitors post-merger, suggests that the merged entity would be likely to have the incentive to introduce some form of paywall postmerger. 522. [ ]. [ ] 523. [ 311 524. 312 525. 313 314 311 [ 312 [ 313 [ 314 [ ] ] ] ] 103 526. 315 527. 316 317 528. 318 529. 529.1 319 529.2 320 529.3 321 315 [ 316 [ 317 [ 318 [ ] ] ] ] 319 [ 320 [ ] ] 321 [ ] ] 104 [ ] 530. 322 [ 531. 323 532. ] Commission’s view – paywalls 533. [ 534. 535. 536. 324 325 537. 322 ] The Commission considers that if the merger were to proceed, other online news providers would not constrain the merged entity from implementing a paywall. In addition, if the merged entity chose to leave one of its websites (either stuff.co.nz or nzherald.co.nz) open for free access it would reduce the risk of losing advertising [ ] 323 [ 324 [ ] ] 325 [ ] 105 revenue by implementing a paywall on its other website. This is because its ‘free’ website would be able to capture a number of users who decide to switch away from the website with the paywall. 538. Furthermore, the merged entity could set a metered or soft paywall to minimise loss of unique viewers. For example, the merged entity could introduce a paywall for some content or introduce a paywall once a reader has exceeded a certain number of articles per month. A reduction in total hits/impressions may have little negative impact on advertising revenue as long as the merged entity could still retain its overall reach. 539. Our view is that introducing a metered or soft paywall over one of stuff.co.nz or nzherald.co.nz is the most likely approach the merged entity would take. Despite the Applicants submissions, we consider that because the merged entity would be the largest New Zealand online news provider and the website with the paywall would be able to provide unique content (as both websites currently do), readers would inevitably be prepared to pay to access at least some of that content. 540. However, in reaching this view, the Commission acknowledges that paywalls are not necessarily detrimental, and faced with a continuing decline in print revenues, media firms may seek to rely on paywalls for revenue. Nevertheless, we consider that a lack of competition in the New Zealand online news market could mean that a paywall may be more comprehensive or the price of a paywall may be higher than it would otherwise be in competitive market. 541. Therefore, the Commission’s preliminary view is that we are not satisfied that the proposed merger would not be likely to result in a substantial lessening of competition from an increase in prices for online New Zealand news content. Reduction in quality 542. The Commission considers that competition between the Applicants is a substantial driver of the quality of news produced in the online New Zealand news market. Quality of media content in reader markets 543. Assessing the quality of media content in reader markets presents several challenges. Unlike for some other products or services, news content cannot be objectively assessed based on factors like product reliability or timeliness of service. Opinions on the quality of news may vary widely across readers. These opinions are often related to the topics that are covered or are given prevalence, rather than how a given topic is covered. 544. With the exception of public service broadcasters, decisions around the presentation style and content covered by media firms are typically made in an attempt to most profitably meet the demand of media users (readers) so as to provide access to these readers for advertisers. 545. The Commission makes no judgement as to how individual media organisations attempt to profitably attract readers, but the Commission is interested in assessing 106 how the editorial and content choices made by the merged entity could be affected by the proposed merger to the detriment of readers across the online New Zealand news market. The factors that are relevant to assessing quality include: 545.1 Variety of content – could the merger lead to a narrower range of content? 545.2 Volume of content – could the merger lead to less content? 545.3 Accuracy of reporting – could the merger lead to less accurate content? 545.4 Objectivity of reporting – could the merger lead to increased bias in reporting? Applicants’ views 546. The Applicants submitted that the proposed merger would not lead to a reduction in quality. To the contrary, the Applicants suggested that the proposed merger would result in a reduction in duplication and free-up journalist resources to report on an increased number of stories. The Applicants suggested that the proposed merger would allow them “to invest in better quality journalistic content and greater breadth of coverage” to the benefit of competition.326 Moreover, the Applicants considered that investment in journalist training and enhanced investigative resources would lead to better quality journalism.327 547. The Applicants also submitted that they would not be commercially incentivised to reduce diversity or adopt a political bias as this would lead to readers switching to alternative news and information sources. 548. The Applicants also suggested that internal plurality exists within Fairfax and NZME today, and that this editorial independence would continue to exist post-merger. Both NZME and Fairfax currently have internal codes of ethics and state that these codes will continue to apply post-merger. The Applicants submitted that because of the competitive environment that the merged entity would operate in, high journalistic standards would be necessary to attract audience attention and, therefore, advertiser expenditure.328 549. In particular, the Applicants argued that the two-sided nature of the market would be sufficient protection to ensure that the merged entity would produce content that is appealing and accessible to readers to preserve advertising revenues.329 The Applicants stated that the merged entity would not risk doing anything that might reduce the “attractiveness, range, or reach of its product”.330 The Applicants filed a submission from Professor Randal C. Picker in support of this argument. We address 326 327 328 329 330 Application at [20.8]; Submission from Russell McVeagh (on behalf of the Applicants) (29 July 2016) at [34]. Submission from Russell McVeagh (on behalf of the Applicants) (29 July 2016) at [37]. Ibid. at [96]. Ibid. at [7], [25] and [29]. Ibid. at [33]. 107 this argument separately below in relation to both the issue of a reduction in quality and the introduction of a paywall as a result of the proposed merger. Journalists 550. Evidence provided by E tū journalists highlighted the extent to which competition between competing New Zealand news producers drives the pursuit of quality outcomes, particularly in terms of the variety of stories, volume of content, and plurality and diversity of voice. 551. E tū described journalists as “being generally competitive people, want to scoop the opposition, and they are embarrassed if the opposition scoops them”. This competition leads journalists to compete to find stories that would not otherwise have been gathered without the presence of competition.331 552. In addition, the E tū journalists explained that competitive pressures between the Applicants results in a greater number of journalists than there would be in a less competitive environment. This is because NZME and Fairfax are trying to win readers’ attention from each other by increasing the number of stories produced. Further, E tū considered that this competition between NZME and Fairfax results in stories being published sooner than would occur in the absence of competition. 553. E tū journalists gave examples of where the level of competition influences journalistic production: My comments are predicated on the undeniable fact that journalists produce their best work when in a competitive environment. Once the element of competition is removed, the tendency is for journalists to slip into “cruise” mode. That is a natural human reaction and there are countless examples in New Zealand journalism. In Wellington I worked at various times for The Evening Post and The Dominion and there is no doubt we worked better and were more driven when we had an opposing newspaper. In more recent years it was evident to me that reporters on The Dominion Post (a combination of the two newspapers) suffered because of the removal of that element of competition. The public therefore suffered too – it 332 was receiving only one version of significant stories, or sometimes no version at all. … I occasionally cover [ ] and one of the considerations my Chief of Board will want to know is were any other media there. And if not, then the tendency will be not to 333 cover the meeting immediately. And possibly hold the material for later use. And my concern would be with the merger that the pressure to actually be there would [be] less for obvious reasons … it may be that we cover it later, it may be we don’t cover it at 334 all. I mean this morning it’s David Bain, whether David Bain is getting compensation or not, I can guarantee that The Herald will have a live person there and say well we create a file as soon 331 332 333 334 Submission from E tū to the Commerce Commission (5 July 2016) at page 2. Ibid. at page 13. Commerce Commission interview with E tū (2 August 2016). Ibid. 108 as the decision is made. Now would we still do that if there was only one website? Well 335 probably not. We might file it 20 minutes later and be more relaxed about it. 554. The E tū journalists acknowledged that the competitive nature of journalists can also extend to internal competition within organisations, for example journalists within one organisation may compete with each other to be the first to publish a story. However, E tū journalists did not consider that internal competition would resolve the resourcing and timing implications that would result from the merger.336 555. The E tū journalists also addressed the Applicants’ argument that the merger would free-up journalists to cover a wider range of stories. These journalists did not consider that this was likely post-merger, as competition between NZME and Fairfax is what drives resourcing decisions and there would be no incentive on the merged entity to pay for an extra story if there is no other competitor that will beat them to it.337 556. In relation to the Applicants’ claimed reduction in duplication of stories that would result from the merger, E tū considered that reporters do not duplicate content. The journalists gave the following examples:338 Another example I’ve highlighted … which is about competitive pressures is the Louise Nicholas affair which was a story that was covered or broken by The Dominion and it was sort of essentially The Dominion’s story for a long time. But The Herald quickly sought to match that and this is just based on my discussion with one of The Herald investigative reporters, [ ], who is involved in covering for The Herald. (He’d recently come to The Herald when that story broke) but he managed to dig out a major sub-angle of it which was to do with Dewar and a fraud he’d committed by altering documents. So that was a new angle that may or may not have been found by The Dominion. … NZME recently published a series of articles on the Canterbury earthquakes and whether there was a link to an increased suicide rate in the region. And 90 minutes after the series began to be published about midday, the Christchurch Press posted a story discounting this link. 557. The E tū journalists were also concerned that as advertisers become more powerful they will be able to dictate what news is covered and what is not. Conversely, E tū also noted that one “commercial player” was worried that if the merged entity wrote a damaging article about it, then it would be unable to go to a rival to tell its side of the story.339 Competitors’ views 558. 335 336 337 338 339 The views of the E tū journalists were supported by TVNZ, MediaWorks and RNZ. Ibid. Ibid. Ibid. Ibid. Ibid. 109 559. TVNZ, MediaWorks and RNZ advised that one key source of news comes from monitoring competing New Zealand news websites. TVNZ advised that it “lives and breathes” other news websites for ideas for stories, looking “what angles do we have or don’t we have that they may have, so that happens every minute”.340 560. Similarly, MediaWorks said that it routinely monitors NZME, Fairfax, TVNZ and Radio New Zealand and will follow up with their own leads on stories that are of interest. MediaWorks explained that it is able to confidently monitor these websites as it knows that the stories will be credible and fact-checked. MediaWorks said that it monitors NZME and Fairfax the most often, given the large number of reporters they have.341 561. MediaWorks described this as the ‘eco-system’ of news, where the different media companies “feed and prompt each other into covering stories”. In its view, the merger would result in fewer stories being generated leading to a reduction in quality of news produced across the ‘entire eco-system’, as there would be fewer journalists “making calls, banging on doors, … checking sources.”342 562. MediaWorks said that the merged entity would face less competitive pressure to invest in journalists, resulting in a reduction in the quality of content that is published.343 MediaWorks expressed the view that the merged entity would only need to focus on producing cheap content and not need to invest in “expensive” journalists.344 Similar views were shared by RNZ. Also referring to the news ‘ecosystem’, RNZ said that there was a risk that the merger would result in fewer angles being covered as a result of a reduction in competition.345 563. MediaWorks also considered that the need for media diversity is “particularly true” in New Zealand, where relationships with commercial organisations can have a “significant impact on a media players’ advertising revenue.”346 Interested parties 564. Corporate advertisers not only interested in the media for advertising purposes, but also because the media generate publicity – positive and negative. For example, these entities often use the media for public relations purposes. 565. As a result, a number of large companies were concerned that the proposed merger would result in reduction in editorial diversity. Some were concerned that this would lead to the editorial decisions of the merged entity being increasingly influenced by commercial factors. 340 341 342 343 344 345 346 Commerce Commission interview with TVNZ (14 October 2016). Commerce Commission interview with MediaWorks (13 October 2016). Ibid. Commerce Commission interview with MediaWorks (29 June 2016). Ibid. Commerce Commission interview with RNZ (12 July 2016). Submission from MediaWorks to the Commerce Commission (30 September 2016) at [6.2]. 110 566. For example, [ ] said that competition between media firms creates “tension” between different media outlets competing to tell a story first. In terms of positive public relations, [ ] said the merger would reduce this competitive tension and potentially have a negative effect. Conversely, [ ] said that there would be “one less person digging around” in the event that a [ ] mistake led to potential negative public relations, so this would be an offsetting positive effect of the merger from [ ] perspective.347 567. [ ] explained that its “biggest concern” with the merger was that any negative public relations would only be covered by one editorial voice. It said that the merger would reduce the ability for companies to tell their side of the story. [ ] explained that “at the moment you could then go to the other people and give your side of the story, whereas if you take that option out, where else do you have to go and who’s held accountable?”348 568. Likewise, [ ] was concerned that the proposed merger would remove any “counter-balance” between NZME and Fairfax. [ ] gave an example of the media’s reaction to it [ ]. [ ] said that Fairfax “held on to the story for weeks” whereas other media publishers tended to show a different side to the story.349 569. Similar to some large advertisers, Trade Me considers that competition between NZME and Fairfax leads to its press releases being more widely covered by the media than they otherwise would be without competition.350 For example, Trade Me has on occasion offered exclusive rights to reports it produces to either Fairfax or NZME before they are made public. On these occasions NZME and Fairfax effectively compete with each other to offer Trade Me the greatest coverage of Trade Me’s information and the associated press release. Trade Me considered that TV and radio news bulletins were too limited to cover the content detail that is within its press releases.351 The Commission’s view 570. The Commission considers that competition between competing producers of New Zealand news can improve the quality of news produced. As such, we are not satisfied that the proposed merger would not be likely to result in a reduction in the quality of online New Zealand news. The evidence provided by the E tū journalists and competing media firms support this view. 571. Further, and irrespective of any leading role the Applicants play in the content of news, concerns were raised by a number of interested parties about media diversity, ie, that the proposed merger would result in a loss of voice and ability to “play” the Applicants off each other. 347 348 349 350 351 Commerce Commission interview with [ ] (13 September 2016). Commerce Commission interview with [ ] (23 September 2016). Commerce Commission interview with [ ] (28 June 2016). These releases relate to data held by Trade Me which may be of general news interest and/or specific interest to the business community. For example, data relating to the employment and property markets. Commerce Commission interview with Trade Me (7 October 2016). 111 572. Competition between the parties also promotes accuracy in their reporting. NZME and Fairfax monitor each other’s output and can react quickly to correct inaccurate information the other has reported, which is then corrected in the original article. This competitive oversight is beneficial for readers in improving the overall accuracy of reporting. 573. Competitors highlighted the expense associated with producing New Zealand news. Given the magnitude of these costs, the Commission’s view is that the merged entity could respond to the reduction in competition by reducing costs in a manner that could negatively affect the quality of New Zealand news produced. 574. [ ] The Commission considers that the elimination of this type of duplication carries with it a risk of quality reduction since different perspectives or angles on the same topic can be valuable. These reductions would be particularly detrimental in specialist areas such as investigative journalism, business and cultural news and political reporting where a reduction in journalists could have a significant impact on quality measures such as accuracy and objectivity. 575. The Commission considers that even further reductions of editorial staff (that are not strictly duplications) may be possible as a result of reduced competition in the online New Zealand news market. However, at this stage, the Commission has not assessed the likelihood of any further reductions. Two-sided market Applicants’ view 576. The Applicants submitted that the merged entity would have no incentive to reduce the quality of news produced and/or introduce a paywall. This is because both a reduction in the quality of news and a paywall would risk losing audiences, and, therefore, risk losing advertising revenue. 577. As an example, the Applicants said that “the adoption of a political bias would simply result in consumers that disagree with that bias switching to any of the alternative news / information sources available, eroding audience levels and ultimately advertising revenue.”352 578. Rather, the Applicants submitted that the merged entity would be “incentivised to increase the number of unique views, regardless of the number or extent of competitors that they face, as they need to attract consumer attention to drive advertising revenue … which therefore incentivises them to produce quality content that appeals to as many consumers as possible.”353 579. The Applicants further submitted a report from Professor Randal C. Picker of the University of Chicago on the consequences for quality and diversity of one-sided 352 353 Submission from Russell McVeagh (on behalf of the Applicants) (29 July 2016) at [46]. Authorisation Application from Fairfax/NZME (27 May 2016) at [20.6]. 112 payments in two-sided markets, and free-riding issues associated with the production of news stories with public-good characteristics.354 580. Professor Picker indicates that even if the market were to be defined as narrowly as to include only NZME and Fairfax, the proposed merger could have a positive impact on the quality of news produced.355 580.1 First, post-merger, if an advertiser-funded model were to remain, then the merger may increase the quality of news as the merged entity would not need to compete to provide content that appeals to the majority of reader preferences, but rather could provide a more diverse range of content that attracts a broader range of readers. Conversely, if the merged entity adopted a paywall, then the value that readers attach to content will be taken into account in producing that content. 580.2 Second, where there is limited legal protection for intellectual property, competition may reduce the quality of content that is produced. This is based on the idea that in competitive markets, news media firms have less incentive to invest in quality content of a durable nature,356 for example investigative journalism, because free-riding from competing publishers prevents the original firm from recovering the cost of producing the work in the first place. The Commission’s view 581. The Commission acknowledges that the two-sided nature of the merged entity’s online platform may provide a countervailing incentive to maintain content quality above a certain standard and that it may provide a downward pressure on the price of a paywall. 582. However, such countervailing incentive and, more particularly, the potential cost of any lost advertising revenue that may result from reduced readers, may not by sufficiently large to counter the benefits of the cost-savings associated with reducing quality or the revenues associated with increased price to readers. 583. Advertisers are also primarily concerned with reach, ie the number of unique viewers on a website. The number of articles on the Applicants’ websites could reduce; however, the total page views could remain high. 584. The Commission considers that the merged entity would have such a strong position in the market that there is a real chance that it could undertake quality-reducing cost 354 355 356 Randal C. Picker report, Commentary on News Media Quality Issues in Fairfax/NZME Proposed Acquisition (15 October 2016), attached to a submission from Russell McVeagh (on behalf of the Applicants) (October 2016). Professor Picker acknowledges that news media firms cannot cover every possible story and have to make decisions about how they allocate their journalists, and that these choices are influenced by the competition they face. The potential for free-riding depends on the durability of the content. News items that have a short-life, such as clickbait, may lose value before it can be appropriated. 113 reductions or introduce a paywall without putting a significant amount of advertising at risk. 585. The Commission is not persuaded by the arguments put forward by Professor Picker. First, Professor Picker uses content diversity as his measure of quality. However, as outlined above, it is appropriate to take a multi-dimensional approach to assessing quality (as noted above, it also includes volume, accuracy and objectivity). The merged entity could, for example, maintain diversity at a superficial level, but also cut investment into the accuracy of the content, degrading value to readers. 586. Second, we are also not persuaded by the argument that high quality content may be under-provided in a competitive market because of free rider problems. It is not necessarily the longevity of content that creates quality. Quality of less durable news items is also relevant. 587. A further reason as to why the two-sided nature of the market may not be sufficient to prevent a reduction in quality is that it may not be immediately obvious to readers that some elements of quality are reducing. Because readers are unaware of stories that go unreported, a reduction in the number of journalists uncovering stories, along with a reduction in the incentive of existing journalists to attempt to investigate and report stories from new angles, means that readers would be unaware of the content that would not be generated as a result of the merger. 588. Moreover, if quality is degraded, as outlined above, readers would have limited ability to obtain comprehensive national, regional and local news from elsewhere. Conclusion 589. The Commission accepts that the interrelationship between the proposed merger and reduction in quality is complex. 590. However, given the extent to which the Applicants currently compete and their prominence in the online New Zealand news market, we are not satisfied that the proposed merger would not be likely to result in a substantial lessening competition from a reduction in the quality of online New Zealand news content. 591. Further, we consider that the reduction in quality of online New Zealand news would likely have flow-on effects into other markets. We consider that NZME and Fairfax play a particular role in setting the agenda for online New Zealand news provided by other publishers, as evidenced by TVNZ, MediaWorks and RNZ monitoring and following up on stories published by the Applicants. Therefore, a reduction in the quality of news produced by the merged entity could also have the effect of reducing the quality of TVNZ, MediaWorks and RNZ’s source of news.357 592. In addition, given the digital-first strategies of the main media companies in New Zealand, a reduction in quality of online New Zealand news may have flow-on effects 357 This may have further flow on effects for content that is produced by referring to the New Zealand news publishers’ content, for example blogs. 114 to the news that is then published across other platforms, such as newspapers, TV and radio. Sunday and community newspaper markets 593. For the reasons set out in the advertising section above, we consider that the Applicants’ are each other’s closest competitors of Sunday newspapers in the North Island and community newspapers in 10 areas. We consider that the proposed merger would likely provide the merged entity with increased market power in these markets. 594. Consequently, in this section, we consider whether the proposed merger could result in a substantial lessening of competition through: 594.1 an increase in the price and/or a reduction in quality of Sunday papers in the North Island; and 594.2 a reduction in the quality of the 10 overlapping community newspapers. Impact on price 595. We consider that the merged entity would be likely to increase subscription prices and retail prices for Sunday newspapers in the North Island as a result of the loss of the competitive constraint currently imposed by the merging parties in this market. 596. We acknowledge that Fairfax currently appears to be operating a one-price model and it has not appeared to price discriminate in the South Island where it has considerably lower numbers of subscribers.358 597. However, we consider that, given declining reader numbers and advertising revenues, the merged entity would be likely to increase prices for Sunday newspaper post-merger in the North and South Islands. 598. This is consistent with the trends of other daily newspapers in New Zealand that face limited or no competition, where we have observed increasing cover and subscription prices. Impact on quality 599. The Applicants submitted that they would be constrained from reducing the quality of Sunday and community newspapers for the same reasons outlined in relation to online New Zealand news. The Commission’s view 600. 358 As noted above, the Commission considers that competition between the Applicants is a substantial driver of quality and that with the merger, there is the risk of a decrease in the quality of the online content produced by the merged entity. Around [ ]% of Fairfax’s Sunday Star Times subscription revenue is derived from subscribers in the South Island. SundaysBySuburb 6 Oct, attached to an email from Russell McVeagh (on behalf of Fairfax) to the Commerce Commission (6 October 2016). 115 601. Given that print publications largely reproduce the same content that is published online, any reduction in quality of online content would also negatively affect the quality of print content. 602. In addition, Sunday newspapers sometimes contain more in-depth, long-form journalism which is not necessarily suitable for inclusion in a daily metropolitan newspaper. With the merger, the competition between the newspapers for this type of journalism would be lost. 603. While other publications such as the Listener, Metro and North & South produce some long-form investigative journalism, we do not consider that they would sufficiently constrain the merged entity from reducing the quality of Sunday newspapers. 604. Two of these magazines (Metro, North & South) are monthly publications so are not published as frequently, and none of these magazines provide the same degree of up-to-date news as Sunday newspapers. For example, these magazines do not include new stories from the latest news cycle. Rather these magazines are designed to be read at any time over the issue period and do not have such short-lived, time specific content as Sunday newspapers. 605. In the 10 regions outlined in paragraphs 101.4 above, the proposed merger will result in a merger to monopoly of the competing community newspapers. While the Commission has not carried out a detailed analysis of the specific content of the different community newspapers, much of the content is very local in focus and is not generally replicated in other print publications, unless it is of national interest. 606. The Commission considers competition is also a substantial driver of quality at a community newspaper-level. This view was shared by independent community newspapers that we spoke with. 607. Les Whiteside, owner of WSN and the Blenheim Sun, said that community newspapers compete for readers attention. He considered that in Kapiti, where NZME and Fairfax community newspapers are competing “head on”, readers will “certainly be worse off” as a result of the merger.359 608. Further, Sun Media said that competition drives editorial diversity at a local level. It explained how competition between Sun Media and NZME in the Bay of Plenty region leads to value:360 Yes and we usually do [break stories first], we’re pretty good at getting first. [NZME] are little bit better at getting the story behind the story because that’s all that we leave them… they kind of sweep up the crumbs and every now and again they’ll get a gem out of that. Well that’s kind of competition when you look at it as a whole, it works, it keeps everybody on their toes. If that was taken away it will be quite sad really from an editorial ethic point of view. 359 360 Commerce Commission interview with Wellington Suburban Newspapers (28 September 2016). Commerce Commission interview with Sun Media (16 September 2016). 116 Conclusion 609. In summary, the Commission’s preliminary view is that we are not satisfied that the proposed merger would not be likely to substantially lessen competition in the reader markets for Sunday newspapers in the North Island and the 10 community newspapers in the areas of overlap. 610. As a result, we consider that it is likely that the subscription and/or cover prices of Sunday newspapers nationally would increase and the quality of both Sunday newspapers and the community newspapers in the 10 affected areas would be likely to reduce. 117 Section 6: vertical and conglomerate effects 611. Our analysis of the advertising and reader markets has considered the potential unilateral effects that could result from the proposed merger. There are also potential vertical and conglomerate effects that arise. We consider these below. Vertical effects 612. A vertical merger between an upstream input provider and a downstream producer of a final product could allow a merged firm to cease supplying the input to its rivals altogether or constructively refuse to supply by making the terms of access uncompetitive (hereafter, foreclosure refers to both refusal and constructive refusal to supply the product). Such foreclosure could have the effect of reducing competition in the downstream market. For a foreclosure strategy to be successful, the vertically integrated firm must have both the incentive and the ability to foreclose its rivals. 613. Presently, NZME is involved in a news sharing service, NZME News Service, with independent newspaper publishers.361 While the independents generate their own local and regional news, they are dependent on the NZME News Service for national content that NZME has produced for its own publications. As part of the news sharing service, the independents also provide local and regional news to NZME. 614. Each of the independents publishes daily newspapers that contain a bundle of news which is local, regional and national in nature. The independents rely on national news content to maintain their circulation. This is because many of their readers require the bundle of news. Independent publishers 615. The independents that participate in NZME News Service are Otago Daily Times (Allied Press), Ashburton Guardian, Gisborne Herald, Greymouth Star, Westport News and Wairarapa Times-Age. Concerns of independent publishers 616. Concern has been expressed by some of the independents362 that the merged entity may cease providing independent publishers with national news content and that a lack of national news content would degrade the quality of the independents’ newspapers to the extent that they would either exit the downstream regional newspaper market or otherwise be rendered less competitively effective. 617. Because of the different formats in which their news is gathered, the independents do not consider national content from TVNZ, MediaWorks, RNZ and the like to be substitutable for the content provided by the NZME News Service. 361 362 Fairfax sources and produces its own national and regional content which it has done since it withdrew from the New Zealand Press Association news syndicating service in 2011. Commerce Commission interview with [ ] (8 July 2016). Commerce Commission interview with [ ] (7 July, 2016). 118 618. One independent363 also submitted that in the event that NZME News Service was to cease and the independent exited, it doubted whether the merged entity would attempt to cover news from its area and, that if this was the case, then regional New Zealand and New Zealand as a whole would “be the losers”. How the proposed merger could substantially lessen competition 619. One theory is that the merged entity could cease providing national news content to the independents to shift the readers of the independent newspapers to the merged entity’s news websites in search of national news. It would do this to increase the number of viewers on its news websites,364 potentially enabling it to increase the price of its digital advertising inventory. The merged entity would pursue this strategy only if it were more profitable that the provision of its news service. 620. Alternatively, the merged entity could withdraw the provision of national news content from the independents to weaken their publications so it could either enter more profitably the independents’ markets with its neighbouring publications or acquire the independents at a reduced price. Applicants’ views 621. NZME and Fairfax submitted365 that there will be no change to their incentives to offer news syndication post-merger. This is because: 621.1 the news content provided by the independents provides NZME and Fairfax with additional content to show to consumers (ie, it grows audience); and 621.2 it is an additional revenue stream for content that has already been produced so the additional revenue has a zero incremental cost. The Commission’s view 622. Our preliminary view is that the merged entity is unlikely to withhold national news content from the independents. 623. As the Applicants submitted, the news service provides an additional revenue stream for which there is little cost.366 The merged entity may not be able recover the loss of this revenue through an increase in readers of its news websites and subsequent increase of its digital advertising inventory. 624. We acknowledge that the merged entity could, for example, seek to weaken the Otago Daily Times by no longer providing it with national news content and 363 364 365 366 Submission from Westport News Company, 24 June 2016. As the independents are the only publishers of daily newspapers in their respective regions, the merged entity would not have competing newspapers with which to capture the readers and so would not have an incentive to refuse to supply news content for purposes of enhancing its market power in regional daily newspaper markets. NZME and Fairfax, Response to Submissions, 29 July 2016. We note that NZME is presently negotiating contracts with the independents for the ongoing provision of news content. 119 increasing the number of its journalists in Dunedin with a view to establishing its own newspaper.367 625. Nevertheless, even if the merged entity managed to weaken and then acquire an independent in this way, the acquisition would not change the market structure, given that the independents are presently the only publishers of daily newspapers in their respective regions. As such, no substantial lessening of competition could arise from a transaction. 626. Accordingly, the Commission’s preliminary view is that the proposed merger is unlikely to give rise to vertical effects that would result in a substantial lessening of competition. Conglomerate effects 627. A conglomerate merger involves the acquisition of a complementary good or business, or the acquisition of a firm that trades in a separate but related market. Conglomerate effects often do not lead to competition problems by themselves and are often associated with efficiencies, but may also in some cases lead to competition concerns if the bundling of products prevented rivals obtaining sufficient scale to be viable. Alternatively, bundling might take the form of predatory pricing. How the proposed merger could substantially lessen competition 628. The merger would result in the overlap of the Applicants’ Sunday papers, as well as a number of their North Island community newspapers.368 There would be no overlap in metropolitan or regional newspapers. In addition, NZME’s radio network would be added to Fairfax’s print assets. 629. The concern of newspaper competitors is that the combination of NZME’s radio/ print/digital assets and Fairfax’s print portfolio (particularly in the South Island) and digital assets could enable the merged entity to offer bundles that other newspapers, radio networks or websites are unable to match either because they cannot offer an equivalent bundle or because a component of the bundle is priced below cost. To this extent, their ability to compete with the merged entity could be lessened to such an extent that they may exit the market. Applicants’ views 630. The Applicants submitted:369 630.1 367 368 369 The starting point is that if purchasers prefer such bundles, any increase in the availability of such bundles, or any reduction in price, is pro-competitive. Moreover, NZME already is able to offer an advertising bundle spanning print, online, and radio advertising in many parts of the North Island, and We understand that Fairfax currently has four or five journalists based in Dunedin. NZME does not currently own any newspaper assets in the South Island, with the majority of newspapers in the South Island being owned by Fairfax and Allied Press. NZME and Fairfax, Response to Submissions, 29 July 2016. 120 there has been no evidence of those bundles materially affecting competition in any of those areas. 630.2 Finally, there is nothing to prevent these print competitors from partnering with other radio providers (eg MediaWorks), to offer their own competing bundle to advertisers. For example, Allied Press and MediaWorks could partner together in Otago to offer a cross-media print, digital, radio and TV advertising bundle. The Commission’s view 631. The Commission’s preliminary view is that the proposed merger is unlikely to result in a substantial lessening of competition through the merged entity’s ability to bundle different types of advertising inventory. 632. Currently, in the North Island, NZME is able to (and does) bundle radio advertising with print advertising inventory to some customers. In addition, it offers metropolitan/regional and/or community print advertising bundles. We have found no evidence that NZME is pricing those bundles below cost. 633. It appears that NZME’s radio and print bundles are not presently impeding the ability of other newspaper publishers to compete with NZME. In the Bay of Plenty region, Sun Media has been able to compete successfully with NZME despite it having a regional newspaper, the Bay of Plenty Times, three community newspapers, the Bay News, Katikati Advertiser and Te Puke Times as well as radio stations. 634. Sun Media’s flagship newspaper, The Weekend Sun is a free weekend newspaper distributed in Tauranga with a circulation of 66,600. 635. Regarding its performance, Sun Media told us: The Sun is really, really strong we’re putting out 60-odd pages every week. This week we’ve put out our largest paper ever, about 80 pages. So we’re kind of flying in the face of the trend of newspapers supposedly being in decline. We consider ourselves to be one of the more successful community newspapers in the country and in particular with Sun Live it’s taken us even out of that to the category into a 370 serious news player. 636. 370 371 Not all print advertisers want a bundle of different types of advertising. As described in the print advertising sections above, advertisers choose the type of newspaper in which to advertise for a number of different reasons, including reader demographics, reach, the type of messaging the advertiser is trying to achieve, and timing – for instance, many retail chains would not advertise in Sundays because by Sunday “the action is over”.371 Commerce Commission interview with Sun Media (16 September 2016). Commerce Commission interview with Strategy Design and Advertising( 7 July 2016). 121 637. Similarly, radio is not suitable for a significant number of advertisers. Radio is better suited to advertisers wishing to emphasise brand or presence rather than say describing the attributes of a product. As outlined above, a similar rationale applies to an advertiser’s reasons for advertising online. 638. To this extent, it is likely to be a small subset of all advertisers for whom there is value in bundled print/radio and/or online advertising. Even if this were not the case, competitors to the merged entity would have counter strategies available to them through their ability to enter into arrangements with MediaWorks’ radio network372 and/or other providers of digital advertising. Other effects Withdrawal from News Works NZ 639. A number of newspaper publishers expressed concern that the combined entity would withdraw from the National Advertising Bureau’s media booking service, News Works NZ, and undertake the placement of national advertising campaigns itself. The publishers submitted that if the merged entity were to withdraw from that service then the independents would lose valuable revenue that they derive from advertising placed in their publications by News Works NZ as part of national advertising campaigns. This, they said, could cause them to founder or even exit the regional newspaper markets. Applicants’ views 640. The Applicants submitted that: 640.1 national advertisers require national advertising coverage and so postmerger would still require advertisements placed in the independent publications; 640.2 the Applicants have no incentive to make the process of booking print advertising more difficult for the customers that prefer to purchase nationwide print campaigns; and 640.3 each of the Applicants currently participate in News Works NZ as it advocates for the use of print newspaper advertising to drive sales of print advertising inventory, and that NZME and Fairfax would want that advocacy to continue post-merger.373 News Works NZ 641. 372 373 News Works NZ is the commercial arm of the Newspaper Publishers Association. Its roles are to coordinate the placement of national advertising campaigns in NZME faces vigorous competition throughout the country from MediaWork’s national radio network which has a market share of 53% compared with the 38% share of NZME’s radio network http://www.gfk.com/en-nz/insights/report/radio-audience-measurement accessed 20 October 2016. NZME and Fairfax, Response to Submissions, 29 July 2016. 122 newspapers throughout the country, advocate for and market newspaper advertising, and to undertake and disseminate advertising research. 642. News Works NZ provides a one-stop shop for advertisers and advertising agencies. Advertisers and agencies are also able to book advertising space directly with publishers. Typically News Works NZ places advertising in metropolitan, regional and Sunday paid newspapers but not community newspapers. Independent publishers’ views 643. The Ashburton Guardian told us that the volume of national advertising is rapidly declining. It advised that six years ago national advertising made up [ ] of the Ashburton Guardian’s revenue and that two years ago that had dropped to [ ]%. Around [ ] of the national advertising placed in the Ashburton Guardian is placed by News Works NZ with the other [ ] being placed directly by advertisers. The advertising placed by News Works NZ equates to approximately [ ] of the Ashburton Guardian’s total annual revenue.374 644. Allied Press also noted the rapid decline of national advertising placed in the Otago Daily Times. Grant McKenzie, CEO Allied Press, considered that if News Works NZ were to cease placing national advertising, national advertisers may choose to place their advertising in only the merged entity’s publications and not advertise at all in the Otago Daily Times which covers just 5% of the country’s population.375 645. [ ] obtains around [ ] of its revenue from advertising placed by News Works NZ. It told us that without the News Works NZ advertising, the [ ]would have to go directly to national advertisers such as [ ] and try to convince them that they need to advertise in [ ]. [ ] considers that if it lost the revenue it gets through advertising placed by News Works NZ it would be unprofitable.376 The Commission’s view 646. The Application attached a report from PwC which estimated the benefits arising from the proposed merger.377 [ ] 647. [ ] 374 375 376 377 Commerce Commission interview with Ashburton Guardian (20 September 2016). Commerce Commission interview with Allied Press (16 September 2016). Commerce Commission interview with [ ], 22 September 2016. See Section 7. 123 648. We consider that for the following reasons, the merged entity is likely to continue to participate in News Works NZ. 649. In our interviews with advertising customers who run national advertising campaigns we were advised that a national campaign typically requires placement of advertising across all regions in order to reach all customers. News Works NZ CEO, Brian Hill, confirmed this to be the case saying that the merged entity “could not altogether cut a particular region” such as those where the independents are because advertisers need to reach their customers in those regions.378 650. While advertisers can go directly to publishers to place advertisements, News Works NZ reduces transaction costs for national advertisers by being a one-stop shop. To this extent, the merged entity would be disadvantaging national advertisers if it were to leave News Works NZ and book advertising only in its own publications. We consider that this would be a risky strategy at a time when print advertising is declining markedly. 651. News Works NZ also advised us that in response to the decline in print advertising, it undertakes research and advocates for print advertising. For example, research undertaken by News Works NZ recently shows that advertising in regional papers is still a very effective way of reaching people in those regions. News Works has undertaken advertising to advertisers to this effect. 652. Given the decline in print advertising, we consider that the merged entity is likely to want to continue to benefit from the newspaper research and advocacy undertaken by News Works NZ. However, even if it did withdraw, the remaining members of News Works NZ would continue to be able to offer advertisers the benefits of onestop shop in their newspapers. The value of advertising in those publications does not depend on the participation of the merged entity in News Works NZ. 653. For these reasons, we consider that the merged entity is likely to continue participating in News Works NZ and, even if it did not, no competition issues are likely to arise in respect of national advertising. 378 Commerce Commission interview with News Works NZ (10 October 2016). 124 Section 7: public benefits and detriments 654. As we have identified a substantial lessening of competition in advertising and reader markets in terms of section 67(3)(a) of the Act, we must now consider whether we can be satisfied that the proposed merger will result, or be likely to result, in such a benefit to the public that it should be authorised in terms of section 67(3)(b) of the Act. 655. The Commission is tasked with assessing whether the public benefits, which arise primarily from rationalisation of the merged entity’s business operation are sufficient to outweigh the competitive detriments arising from the loss of competition in New Zealand’s media industry. The Commission’s analytical method 656. As described in the statutory framework section, the Commission is required to determine whether the merger will result, or be likely to result, in such a benefit to the public that it should be permitted. This public benefit test requires us to balance the benefits and detriments that may arise from the merger. In doing so, the Commission must quantify benefits and detriments in so far as is possible. However, benefits and detriments which cannot be quantified are still relevant to the Commission’s overall assessment. 657. In this case, the benefits claimed by the Applicants include cost-savings that can be quantified, an enhanced advertising alternative for advertisers, an increase in journalistic quality, and enhanced content plurality. The Applicants consider that there will be no material detriments arising from the merger.379 658. The Commission considers that the detriments fall into three categories. 658.1 Allocative efficiency detriments caused by price increases as well as productive and dynamic efficiencies that can be quantified. 658.2 Allocative efficiency detriments from a reduction in quality that we cannot, practically, quantify and which we make a qualitative assessment of without any quantification. 658.3 Detriments to New Zealand arising from any reduction in plurality. As with reductions in quality, these cannot be reliably estimated in any way and we have therefore addressed the magnitude of this detriment without any quantification. 659. We explain our approach to estimating each benefit and detriment in the sections which follow. 660. What is common is that we are not necessarily required to reach a point estimate for any particular benefit or detriment. Rather, we will generally estimate a likely range 379 Application, at [21.1]. 125 because further precision may be unwarranted. In Godfrey Hirst, the High Court indicated that: …unless the Commission has good reasons for excluding other values within the (likely) range that it has determined, it is the range rather than any point within the range that should form 380 the basis for the balancing exercise. Approach to quantifying benefits and detriments 661. As discussed previously, we recognise that there is a wide range of likely outcomes that could occur without the merger which vary according to the rate of decline in print. We acknowledge that the status quo, as it exists today, while a possible outcome absent the merger is not likely to continue given the trends evident in the market. We expect that reductions in print revenues and some rationalisation of newsprint publishing will continue to occur to some extent. 662. Consequently, cost-savings attributed to the merger as a result of newsprint rationalisation may be achieved to a greater or lesser extent with or without the merger. Therefore, these we cannot necessarily consider these values as a mergerspecific benefit. 663. However, in terms of estimating the benefits and detriments, the status quo figures provide a pragmatic basis for attempting to quantify the extent of benefits and detriments. This allows us to develop a sense of the scale of the quantifiable benefits and detriments, but not with the precision we would expect. 664. Further, as outlined at above, in recognition that the status quo may not necessarily prevail, to assist us in applying our judgement, we have also considered what the detriments would be if the Applicants immediately ceased publishing all print products and shifted to ‘digital only’ businesses. As with the status quo, we do not consider this scenario to be likely. 665. We consider that a digital-only New Zealand news environment is unlikely to emerge in the short term, if at all. However, since the likely without-the merger scenario will be somewhere between the status quo and digital only scenarios, this approach provides us with another reference point for the likely benefits and detriments. 666. We consider that the detriments calculated with respect to community newspaper advertising would remain unaffected under the ‘digital only’ scenario. As stated previously,381 we consider the Applicants would continue to publish community newspapers even under a ‘digital only’ business. Quantifiable benefits: production efficiencies Benefits – status quo 667. 380 381 The primary benefits of the proposed merger arise from the proposed reductions in duplication in a number of the Applicants’ business operations. Above n 9. See above at n 71. 126 668. The claimed productivity benefits from the merger have been grouped under the following operational categories to reflect those parts of the business where cost savings would arise. The production efficiency values were estimated by PwC based on information provided by the Applicants. Because the Applicants have not undertaken any specific post-merger implementation planning, there is uncertainty regarding the precise values for each of these categories, hence the Applicants provided estimate ranges. 669. Additionally, the Applicants provided updated estimates in relation to several of the cost saving categories outlined below. The total overall benefit figures estimated in the updated analysis are less, but differ by less than 10% from the original figures. The Commission has not yet had sufficient time to consider these new figures, and so has not included them in this Draft Determination.382 Sales Agency and direct sales to national customers 670. Both Applicants sell advertising inventory to the same advertising agencies and large corporate customers. These customers are relatively few in number.[ ] 671. [ ] 672. The Commission agrees with these figures as reasonable estimates and considers that the proposed merger would result in an agency and national customer sales cost saving benefit of between $[ ] million to $[ ] million over the course of one year, and one-off merger-related costs of between $[ ] and $[ ]. Sales operations 673. Both Applicants incur advertising costs in relation to their sales operations. [ ] 382 [ ] 127 674. The Commission agrees with these figures as reasonable estimates and considers that the proposed merger would result in a sales operations cost saving benefit of between $[ ] to $[ ] over the course of one year, and one-off merger-related costs of between $[ ] to $[ ]. Direct sales 675. Both Applicants sell advertising inventory to a large number of smaller customers. For example, each individual community paper typically has hundreds of customers each year. In many instances the direct sales teams of the Applicants target the same customers. 676. The duplication that would be [ ] 677. The Commission agrees with these figures as reasonable estimates, at least in the short term, and considers that the proposed merger would result in a direct sales cost saving benefit of $[ ] million over the course of one year, and one-off mergerrelated costs of $[ ]. 678. However, the Commission also notes that sales costs is an area where the Applicants may have significant scope to reduce costs in the without-merger-scenario. Marketing Subscriber management 679. Both Applicants undertake subscriber management activities to maximise subscription revenues. [ ] 680. The Commission agrees with these estimates and considers that the proposed merger would result in a subscriber management cost saving benefit of between 128 $[ ] and $[ $[100,000]. ] over the course of one year, and one-off merger-related costs of Research and insights 681. 682. Both Applicants have in-house teams which analyse AC Nielsen and other data for internal and external purposes. [ ] [ ] 683. The Commission agrees with these figures as reasonable estimates and considers that the proposed merger would result in a research and insights cost saving benefit of between $[ ] to $[ ] million over the course of one year, and one-off mergerrelated costs of $[ ]. Trade marketing 684. Both Applicants undertake trade marketing targeted at the same customer base which [ ] 685. The Commission agrees with these figures as reasonable estimates and considers that the proposed merger would result in a trade marketing cost saving benefit of between $[ ] and $[ ] over the course of one year, and one-off merger-related costs of $[ ]. Consumer marketing – staff 686. Both Applicants employ staff to undertake consumer marketing targeted at the same customer base which [ ] 687. The Commission agrees with these figures as reasonable estimates and considers that the proposed merger would result in a consumer marketing cost saving benefit of between $[ ] million to $[ ] million over the course of one year, and one-off merger-related costs of $[ ]. 129 Consumer marketing – external spend 688. Both Applicants undertake external consumer marketing spending which [ ] 689. The Commission agrees with these figures as reasonable estimates and considers that the proposed merger would result in consumer marketing cost saving benefit of $[ ] million over the course of one year. Editorial 690. The Commission has accepted the estimates provided by PwC for the purposes of quantifying the benefits arising from [ ]. However, based on the limited scope of the information provided in support of PwC’s estimates, the Commission is considering whether the methodology used to calculate these benefits is appropriate.383 [ 691. 691.1 691.2 691.3 691.4 691.5 692. ] PwC estimated that the merger would result in [ ]. Table 8: [ [ [ 383 ] ] [ ] ] [ ] The Commission notes the editorial numbers have been obtained from two different data sets and compiled using two different methods. PwC has estimated the likely editorial staff reductions with reference to the applicants’ editorial matrices. In our assessment of the reader markets, we have assessed journalist numbers using the applicants’ “all staff” lists. Advice from BDO suggests that the use of editorial matrices may not include some or all editorial employees within certain areas, such as those in Auckland, rural and digital areas. 130 [ ] [ [ [ ]-[ ] ] [ ]-[ ] ] [ ]-[ ] [ ] [ ]-[ ] [ ] [ ]-[ ] 693. As well as [ ] 694. Based on PwC’s estimates, the Commission considers that the proposed merger would result in a [ ] cost saving benefit of between $[ ] million and $[ ] million over the course of one year, and one-off merger-related costs of between $[ ] million and $[ ] million. [ 695. ] [ ] 696. Based on PwC’s estimates, the Commission considers that the proposed merger would result in a [ ]cost saving benefit of $[ ] million over the course of one year, and one-off merger-related costs of $[ ]. [ 697. ] [ ] 698. Based on PwC’s estimates, the Commission considers that the proposed merger would result in a [ ]cost saving benefit of between $[ ] to $[ ] over the course of one year, and one-off merger-related costs of between [ ] to $[ ]. 131 [ 699. ] [ ] 700. Based on PwC’s estimates, the Commission considers that the proposed merger would result in a [ ] cost saving benefit of between $[ ] to $[ ] over the course of one year, and one-off merger-related costs of $[ ]. Syndication and contributor costs 701. Both Applicants purchase syndicated content from content sources, including on a subscription basis and from freelance contributors. [ ]PwC estimated that this would result in a [ ]% reduction in syndication costs paid to content providers and a [ ]% reduction in contributor costs from an alignment of content. 702. Based on PwC’s estimates, the Commission considers that the proposed merger would result in a cost saving benefit of $[ ] million over the course of one year. [ 703. ] 704. PwC has estimated this benefit to be between $[ ] million to $[ ] million, with oneoff merger-related costs of between $[ ] million to $[ ] million. 705. However, the Commission considers that in the absence of the merger there is a real chance that [ ]. 706. Consequently, we consider that the proposed merger would result in a lower [ ] cost saving benefit of between $[ ] million to $[ ] million over the course of one year, and one-off merger-related costs of between $[ ] million to $[ ] million. [ ] 707. [ 132 ] 708. Based on PwC’s estimates, the Commission considers that the proposed merger would result in a staff cost saving benefit of $[ ] over the course of one year, and one-off merger-related costs of $[ ]. Printing, procurement and distribution Newsprint costs 709. Both Applicants currently purchase large volumes of newsprint under separate contracts. PwC modelleted that post-merger the Applicants may be able to obtain procurement savings from aligning newsprint costs. PwC has estimated that this would result in a cost saving of between [ ] to $[ ] million. 710. The Commission agrees with these figures as reasonable estimates and, correcting the upper bound figure to account for total tonnage figures (as opposed to the “good copy” quantities only), considers that the proposed merger would lead to cost-saving benefits of between [ ] to $[ ] million over the course of one year. Other procurement savings 711. Both Applicants purchase a large amount of inputs, including communications services, electricity, IT services, consumables, print materials costs, and repairs and maintenance services. The Applicants claimed that post-merger they may be able to obtain savings through [ ]. PwC estimated that this could result in savings of between [ ] to [ ]%. 712. The Commission agrees with these figures as reasonable estimates and considers that the proposed merger would result in potential procurements savings of between [ ] to $[ ] million over the course of one year. Consolidation of print distribution in Auckland 713. Both NZME and Fairfax currently independently distribute several publications in the upper North Island. [ ] 714. The Commission agrees with these figures as reasonable estimates and considers that the proposed merger would result in a [ ] cost saving benefit of $[ ] over the course of one year, and one-off merger-related costs of $[ ]. 133 Management Executive team 715. Both Applicants employ a number of people in senior executive roles. [ ] 716. PwC estimated that this would result in a cost saving of $[ ] million over the course of one year, with a one-off merger-related cost of $[ ]. The Commission considers that this figure overestimates the likely actual benefit to the extent that senior executives employed by the merged entity would have greater responsibilities than currently and would [ ] 717. We have not sought to estimate the exact extent to which [ ] to account for higher responsibilities due to inherent uncertainty in gauging the exact nature of the senior executives’ roles post-merger. We continue to use estimates provided by PwC, noting that they are likely to overestimate costs savings associated with [ ]. General management 718. Both Applicants employ a number of people in general management roles. [ ] 719. The Commission agrees with these figures as reasonable estimates and considers that the proposed merger would result in a [ ] cost saving benefit of $[ ] million over the course of one year, and one-off merger-related costs of $[ ] million. Human resources 720. Both Applicants incur staffing costs in relation to human resource functions. [ ] 721. The Commission agrees with these figures as reasonable estimates [ ], but notes that, in our view, there are likely to be additional oneoff merger related human resources costs relating to the integration of NZME and Fairfax. We consider that these additional one-off costs would relate to an additional six consultants required for between four to six months each at a rate of $20,000 per annum. 134 722. Consequently, we consider that the proposed merger would result in a [ ] cost saving benefit of between $[ ] million to $[ ] million over the course of one year, and one-off merger-related costs of $[ ] to $[ ]. Financial processing 723. Both Applicants incur costs in relation to financial processing functions although each has structured their financial processing functions differently. Fairfax has outsourced its operations offshore to TCS in India. In contrast, NZME has established its own financial processing centre in Whangarei (GFS) where it not only undertakes its own financial processing functions but also those of its former parent company APN. [ ] 724. [ ] 725. The Commission agrees with these figures as reasonable estimates and considers that the proposed merger would result in a [ ]cost saving benefit of $[ ] million over the course of one year, and one-off merger-related costs of $[ ].384 Group finance 726. Both Applicants undertake group finance functions. [ ] 727. The Commission agrees with these figures as reasonable estimates and considers that the proposed merger would result in a [ ] cost saving benefit of $[ ] million over the course of one year, and one-off merger-related costs of $[ ]. Information technology IT support services 728. Both Applicants employ staff to undertake considerable IT network support activities, including those relating to telephony, storage, printing and helpdesk functions, but excluding NZME’s radio engineers. [ ] 384 [ ] 135 729. The Commission agrees with these figures as reasonable estimates and considers that the proposed merger would result in a [ ] cost saving benefit of between $[ ] million to $[ ] million over the course of one year, and one-off merger-related costs of between $[ ] to $[ ]. IT: Platforms and apps 730. Both Applicants currently run separate IT platforms and apps. The Applicants claimed that the proposed merger would provide scope for the consolidation and harmonisation in this area. PwC estimated that this would result in a one-off benefit of $[ ] million, although they note that synergies would be determined by the extent to which the separate entities would build shared functionality, integrate NZME’s radio business and invest in merging IT roadmaps. 731. The Commission considers that the Applicants have not provided sufficient evidence to justify this claimed benefit. Any such benefit is dependent on the structure and organisational set up of the merged entity and its likely IT systems. To date, little detail has been provided on these matters. Consequently, we have not included this claimed benefit in our analysis. Premises 732. Both Applicants require premises in various locations throughout the country. [ ] 733. PwC estimated cost savings based on a calculation of current lease costs per person employed. PwC’s resulting estimate of the benefit of avoided lease costs is between $[ ] million to $[ ] million. The Applicants suggested that [ ] buildings are of high quality and would be sublet relatively easily. 734. The Commission considers that this range of avoided costs is too high for two reasons. First, we consider that lease costs per FTE is a more appropriate method of calculation than lease cost per person employed (headcount). This adjustment reduces both the upper and lower bounds of PwC’s estimated range by $[ ]. 735. Second, we consider that lease costs are not necessarily as scalable as suggested by the Applicants. [ ] To date, we have not obtained detailed information regarding the ability of the merged entity to cease Fairfax’s current lease arrangements. 736. Similarly, the Commission has not been presented with a specific merger integration plan. This means that there is considerable uncertainty as to the extent to which the merged entity would seek to cancel or sublet specific leases. Consequently, the 136 Commission has used a range of likely avoided lease costs of between [ ] to $[ ] million. Paper rationalisations Community papers 737. There are 12 areas where NZME and Fairfax both publish overlapping community papers. [ ] 738. PwC estimated [ 385 739. ] The Commission agrees with these figures as reasonable estimates, [ ] Consequently, we consider rationalisation of community papers would result cost saving benefits of between $[ ] million to $[ ] million over the course of one year, and one-off merger-related costs of $[ ]. Sunday papers 740. NZME currently publish the Herald on Sunday and Fairfax publishes the Sunday Star Times and the Sunday News. [ ] 741. PwC estimated [ 386 742. 385 [ 386 [ ] ] The Commission agrees with these figures as reasonable estimates and considers that the proposed merger would result in a cost saving benefit of between [ ] to $[ ] million over the course of one year, and one-off merger-related costs of $[ ]. ] 137 Finance costs 743. The Applicants currently incur finance costs in relation to lending. Post-merger these costs would increase as a result of: 743.1 a higher interest rate of [ ]% being charged on the $[ ] million of debt currently held by NZME as opposed to the current rate of [ ]%; and 743.2 increased borrowing costs (at [ ]%) as a result of the $[ ] million of lending required to fund the cash payment to be paid to Fairfax as part of the transaction. 744. The structure of the merger transaction had not been confirmed at the time of the PwC’s analysis, so the Applicants have not provided an estimate of this additional cost. The Commission considers that these additional finance costs would range from $[ ] million to $[ ] million over the course of a year. Merger transactions costs 745. The Applicants would incur costs involved with undertaking the merger transaction and integrating NZME and Fairfax. These include costs associated with merger integration planning and implementation, including hiring external consultants. These costs would not include advisory costs that have already been incurred and would likely be incurred during the merger authorisation process regardless of the Commission’s ultimate decision. 746. Based on estimates provided by the Applicants,387 the Commission considers that these costs would be likely to be in the range of $[ ] million to $[ ] million. Total quantified benefits 747. 387 The Commission estimates that under the status quo the total quantified benefits would likely be between $36.8 million and $55.7 million over the course of a year, whereas one-off merger-related costs would likely be between $13.6 million and $17.0 million. These costs are broken down by the major cost categories in Table 9 below. Interview: Dale Bridle and Ramesh Vedachalam, Fairfax, and Brendan Jones, PwC, 3 October 2016. 138 Table 9: Total quantified benefits Benefit (cost) category Annual – low ($m) Annual – high ($m) One-off – low ($m) One-off – high ($m) Sales [ ] [ ] [ ] [ ] Marketing [ ] [ ] [ ] [ ] Editorial [ ] [ ] [ ] [ ] Printing, procurement and distribution [ ] [ ] [ ] [ ] Management [ ] [ ] [ ] [ ] IT [ ] [ ] [ ] [ ] Premises - [ ] - Community paper rationalisation [ ] [ ] [ ] [ ] Sunday paper rationalisation - [ ] [ ] [ ] Finance costs [ (Merger transaction costs) - - [ Total $36.8m $55.7m ($13.6m) ] [ ] - - ] [ ] ($17.0m) Quantified detriments 748. The Applicants submitted that there would be no material detriments arising from the proposed merger. On the advertiser side, the Applicants submitted that they compete in a converged advertising market with global and local competitors. On the reader side, the Applicants submitted that cover and subscription prices are constrained by the potential effects on advertising revenues of losing circulation or readership. Further, quality is constrained by the numerous sources of news coverage.388 749. We assessed the level of detriment that could arise from the loss of competitive constraint that the Applicants impose on each other in the markets for: 388 749.1 Sunday newspaper advertising in the North Island; 749.2 premium digital advertising; 749.3 Sunday newspapers in the North Island; Application, at [21.1] – [21.3]. 139 749.4 online New Zealand news; and 749.5 community newspaper advertising in the Whangarei, Rotorua, Hastings, Napier, Manuwatu, Horowhenua, Kapiti, Taranaki, Taupo and Hamilton areas.389 750. We consider that the detriments arising from a loss of competition from the proposed merger in the above markets would result in both price and quality detriments. 751. In this section, we estimate only those detriments related to price effects, as well as productive and dynamic efficiency detriments. We have not attempted to quantify the likely quality detriments associated with a loss of competition as a result of the proposed merger, but consider them qualitatively. 752. In undertaking the assessment of detriments identified above, the Commission has used the following categories – loss of productive efficiency, loss of dynamic efficiency and loss of allocative efficiency. Approach to quantifying detriments 753. As discussed above, we estimate what the detriments would be under the status quo and if the parties were to immediately move to a ‘digital only’ businesses. 754. As above, we do not consider that either of these scenarios to be likely. However, estimating detriments under the status quo and the ‘digital only’ scenarios allows us to develop a sense of the scale of quantifiable detriments. 755. We proceed on the basis of quantifying detriments under the status quo, acknowledging that likely detriments would be lower. We also estimate detriments under the ‘digital only’ scenario, acknowledging that likely detriments would lie above the range estimated. Detriments – status quo Loss of productive efficiencies 756. One outcome that can be associated with a loss of competition is that a firm gaining market power has less incentive to minimise costs and to avoid waste. Organisational slack may creep into operations, and costs may increase, because a satisfactory level of profit is assured even when the firm is less than fully efficient. 757. On the other hand, a firm seeking to maximise its profits will have an incentive to minimise its costs, irrespective of the level of competition in the market. For this 389 As noted under Section 4, we consider that real estate advertisers have more alternatives available to them than other local advertisers. Examples include the Property Press, which is published throughout the North Island, and other digital options which we consider are closer substitutes to community newspapers for real estate advertisers than for other local advertisers. We exclude real estate advertising in community newspapers from the detriments calculated in this section as a result. 140 reason, the Commission does not assume that reductions in competition will necessarily lead to productive inefficiency.390 Setting the range for productive inefficiency 758. While we consider that competition is an important driver of productive efficiency, the weight given to competition as a factor driving productive efficiency is difficult to quantify. 759. As noted already,391 the media industry has been subject to significant change in recent years due to changes in technology and consumer demand, with print revenue and readership in decline. Traditional media models in the form of print newspapers are facing increasing challenges from other modes of delivery. 760. In contrast, digital media has been growing, with many consumers switching their news consumption from print to digital media, or supplementing their print news consumption with digital news. As discussed previously, this has led to the Applicants adopting a ‘digital-first’ strategy, with an increasing focus on video and audio in addition to written content.392 761. We consider that the changing trends in technology and consumer demand are likely to maintain pressures for internal efficiency, especially with respect to print advertising and production. 762. Nevertheless, we cannot fully discount the detrimental impact that a loss of competition may have on incentives to minimise costs and to avoid waste. Without day-to-day pressures from a close competitor and a lack of a competitive benchmark against which a firms’ management can be measured, it is possible that a firms’ management may become less productively efficient. 763. We have considered productivity growth across sectors when setting the range for likely productive efficiency detriments. These productivity growth measures may provide an indication of the range of productive efficiency losses that may arise from the proposed merger. Productivity growth across all industries from 1997 to 2012 was 0.8%, with productivity growth in the Information Media and Telecommunications industry higher at 2.4%, though not statistically significantly so.393 The above range of productivity growth measures indicates that there may be some productive efficiency loss from a loss of competition as a result of the proposed merger. 764. Overall, we consider that any loss of productive efficiency associated with the proposed merger would be minimal. Given the productivity performance observed across all industries, we propose to use a range of between zero and 1% of premerger variable costs to estimate productive efficiency losses as a result of the 390 391 392 393 Authorisation Guidelines, at [68]-[71]. See [91] – [93] above. Application, at [9.3]. Own calculations based on data from Statistics New Zealand. We note that the productivity growth measures mentioned are likely to also incorporate an element of dynamic efficiency. 141 proposed merger in the markets we consider there would likely be a substantial lessening of competition. Quantification of loss of productive efficiency 765. Based on the range which the Commission considers is appropriate, ie, that the likely productive efficiency losses would be between zero and 1% of pre-merger variable costs, we estimate that productive efficiency losses are likely to have a value of between zero to $[ ] million on an annual basis. Loss of dynamic efficiency 766. Dynamic efficiency typically refers to improvements made by firms over the long term concerning product quality, product variety, and cost efficiency through innovations in processes, equipment or managerial practices. A loss of a competitor might cause a firm to invest fewer resources in such improvements. 767. As is the case with the loss of productive efficiency, it is difficult to measure with any precision the cost to society of a lessening in dynamic efficiency attributed to a substantial lessening of competition in a market. Even if firms possess market power, they still have an incentive to innovate and achieve dynamic efficiency as doing so would lead to increased demand for their products and maximises profits. Consequently, a qualitative element is always a significant part of this assessment. Setting the range for dynamic inefficiency 768. We consider that there are a number of reasons why dynamic efficiency losses are likely to be limited with respect to community and Sunday newspaper advertising and the supply of Sunday newspapers. 769. We consider that newspaper print advertising and supply are likely to be characterised as ‘mature industries’. The Commission considers it unlikely that innovations in relation to newspaper print and supply would entail radically new products or processes. Rather, innovation is likely to be related to improvements of existing processes and incremental in nature. 770. As a result, the Commission’s view is that the dynamic efficiency losses associated with the merger with respect to community and Sunday newspaper advertising and the supply of Sunday newspapers is likely to be minor and in the range of zero and 0.5% of sales. 771. However the Commission considers that a higher range of 0.5% to 1% should be used to assess the likely dynamic efficiency detriment associated with the merger with respect to premium digital advertising. 772. As stated under Section 4, we consider that the proposed merger would result in the merging of the closest competitors for this service. 773. As outlined at above, evidence we gathered from [ ] suggested that competition between the two parties has resulted in innovations for premium digital advertising. [ ] also expressed concern that the merger might affect the merging parties’ 142 desire to continue to innovate in the provision of digital advertising.394 Quantification of loss of dynamic efficiency 774. Using the range of zero to 0.5% of total sales revenue for community and Sunday newspaper advertising, and the supply of Sunday newspapers, and a range of 0.5% to 1% for premium digital advertising, the estimated potential detriment is likely to have a value of between $[ ]million to $[ ]million on an annual basis. Loss of allocative efficiency 775. In general, when the price of a product increases because of a loss of competition (for example, as a result of a merger), demand for that product will fall as some consumers switch to alternative products. These alternatives may meet consumers’ requirements in a less satisfactory way and/or are more costly to produce than the products they replace. Alternatively, consumers may simply make fewer purchases, losing the benefit that they would otherwise obtain from a product. In effect, the net result is that the country’s resources are allocated less efficiently. 776. The size of the allocative efficiency loss depends to a large extent on the degree of price increases post-merger.395 The higher the price increase, the larger the loss of allocative efficiency. Another consideration is the ability of a firm with market power to price discriminate. Estimating allocative efficiency loss 777. To estimate the potential loss of allocative efficiency, we determine the likely price increases and determine the likely impact this would have on the quantity of products purchased in the various markets post-merger. 778. In modelling provided by NERA (on behalf of the Applicants), price increases postmerger are estimated through merger simulation modelling on the advertising side.396 Results of NERA’s modelling suggest price increases between 5% and 10%.397 NERA’s amended modelling uses a maximum price increase of 5%. 779. We consider each of the markets of concern below in turn and outline our approach to determining the likely allocative losses associated with the proposed merger. 394 395 396 397 See [189] above. It also depends on the elasticity of demand, that is, the sensitivity of buyers to changes in price. NERA does not apply merger simulation modelling to estimate price increases post-merger on the reader side due to the view that there is a relatively strong cross-platform externality between the reader and advertising sides, ie, the demand for advertising is strongly affected by the number of readers. In contrast, NERA’s survey of research finds evidence of weak or non-existent relationships between advertising and demand by readers, and, accordingly, applies merger simulation models to the advertising side of the market. See Application, Appendix 9, at 27. Application, Appendix 9, at 38. 143 Community newspaper advertising 780. As outlined under Section 4, we do not consider that other forms of advertising would be sufficient to constrain the merged entity in relation to community newspaper advertising.398 781. As noted above, there was no consensus among local community newspaper advertisers as to the extent of the price increase that would prompt them to switch to another form of advertising. However, the majority of those who said that they would continue to advertise in community newspapers, would do so at a price increase of at least 5 to 10%.399 In light of this, we propose to use a range of 5% to 10% as the likely price increase from the proposed merger. Sunday newspaper advertising 782. Our preliminary view is that the proposed merger would be likely to have the effect of substantially reducing competition for Sunday newspaper advertising in the North Island. 783. As outlined in Section 4, views received by advertisers suggested that the Herald on Sunday and Sunday Star-Times constrain each other, particularly when advertisers are seeking to attract a national audience.400 784. We therefore consider that it is likely the merged entity would raise advertising prices post-merger. However, we have not received views on the maximum price increases at which advertisers would absorb before switching to alternative providers. 785. We have considered the two-sided nature of the Sunday newspaper advertising market, and in particular the lack of definitive evidence that reduced advertising reduces reader demand.401 This suggests that there might not be a strong price constraining effect from a risk of reduced readership. However, in the absence of further evidence, we propose to use a range of 5% to 10% as the likely price increase from the proposed merger. Premium digital advertising 786. We consider the Applicants to be each other’s closest competitors for premium digital advertising. With the merger, the constraint provided by the Applicants on each other would be lost.402 787. We have relied on evidence from advertisers to estimate the likely price increase post-merger. As outlined under Section 4, various advertisers expressed views on the likely effects of the proposed merger on premium digital advertising. 398 399 400 401 402 See 352-355 above. See [346] above. See [276] to [281] above. See Application, Appendix 13, at 27 and footnote 54. See [234] to [244] above. 144 788. We have reached the view that it is likely that the merged entity could increase prices for premium digital advertising post-merger by different amounts for particular types of premium digital advertising. In particular, [ ]. 403 789. However, this does not necessarily mean that the merged entity would be able to raise prices to such an extent. Other advertisers indicated that they would re-assess the effectiveness of this type of advertising following a price increase.404 Consequently, we use a range for the likely price increase of between 5% and 10% to estimate the allocative efficiency loss associated with the merger for premium digital advertising. Sunday newspapers 790. As stated under Section 5, we also consider it likely that the Applicants would raise subscription and cover prices for Sunday newspapers. 791. We have calculated the likely maximum price increase by comparing the difference in cover and subscription prices between Sunday newspapers and the Applicants’ regional and metropolitan newspapers that are currently subject to little or no competition. 792. Research into declining readership, yet increasing prices, suggests that there is a rebalancing of prices between readers and advertising. The falling demand from advertisers for print advertising has reduced newspaper publishers’ ability to subsidise readers at the expense of advertisers.405 This is consistent with recent increases in cover and subscription prices the Applicants have implemented.406 793. The trends for regional and metropolitan newspapers of the Applicants suggest a similar rebalancing of how those papers are funded. The cover and subscription prices for Fairfax’s The Dominion Post and The Christchurch Press have risen between [ ]% and [ ]% since 2013. Similarly, cover and subscription prices for NZME’s print publications (The New Zealand Herald and regionals) have risen on average between [ ]% and [ ]%.407 This is despite a decline in circulation for all the above newspapers.408 Advertising yields have also decreased for these newspapers.409 403 404 405 406 407 408 409 See [234] above. See [228] above. See Adithya Pattabhiramaiah, S. Sriram and Shrihari Sridhar “Rising prices under declining preferences: The case of the U.S. print newspaper industry” (2013) Working Paper. Application, at [9.21]. Note that we excluded the increase in cover price for Saturday versions of these publications from this reported range. In particular, there has been an average decline of [ ]% for The Dominion Post and The Christchurch Press since 2013 and a [ ]% decline for The New Zealand Herald and NZME’s regional papers. Source: NZME and Fairfax data. Advertising yields (measures as total advertising revenue over column centimetres of advertising) have decreased by [ ]% and [ ]% for The Dominion Post and The Christchurch Press respectively since 2013. 145 794. Similar trends are observed for the Applicants’ Sunday newspapers. The cover and subscription prices for Fairfax’s Sunday Star-Times have, respectively, risen by [ ]% and [ ]% since 2013. Similarly, cover and subscription prices for NZME’s The Herald on Sunday have, respectively, risen between [ ]% and [ ]% since 2013. As with the regional and metropolitan newspapers of the Applicants, a decline in circulation is observed.410 A similar reduction in advertising yields is also observed.411 795. The average rates of increase in cover and subscription prices are larger for newspapers that face little to no competition when compared to Sunday newspapers where the Applicants’ mastheads compete in the North Island. 796. We consider that any change in costs are likely to be similar across the papers and we have no reason to consider that there exists another reason why the different sets of newspapers would face systematically different rates of price increases. 797. To determine the amount by which the merged entity would increase cover and subscription prices for Sunday newspapers post-merger, we have used the difference in the average rates of price increases for the print publications where the Applicants currently faces little to no competition and its Sunday newspapers since 2013. 798. Applying this approach, we find that the difference in average cover price increases between print publications that currently face little to no competition and the Sunday newspapers ranges from [ ]% and [ ]%.412 For subscription prices, we find that the difference ranges from [ ]% and [ ]%.413 799. The observed difference suggests that prices could be greater than 5% for both cover and subscription rates. However, as outlined above, we note that price increases for regional newspaper have differed. The recent price increases for Sunday newspapers outlined above may also indicate that the market may not support greater price increases. In light of these factors, we consider that the likely maximum price increase for both cover and subscription rates for Sunday newspapers is likely to be 5%. 410 411 412 413 The advertising yield for The New Zealand Herald has decreased by [ ]% in the same period. Source: NZME and Fairfax data. The Sunday Star Times has had a [ ]% decline in circulation since 2013 while The Herald on Sunday has had a [ ]% decline in circulation. Note the decline in circulation for The Herald on Sunday excludes the circulation of sponsored or free copies. Source: NZME and Fairfax data The advertising yields have declined by [ ]% and [ ]% respectively for The Sunday Star Times and The Herald on Sunday. Source: NZME and Fairfax data. The average cover price increase for Fairfax’s Monday to Friday The Dominion Post and The Christchurch Press since 2013 is [ ]% whereas the increase in price for The Sunday Star Times is [ ]%. The average cover price increase for NZME’s Monday to Friday The New Zealand Herald and regional newspapers is [ ]% whereas the increase in price for The Herald on Sunday is [ ]%. Source: NZME and Fairfax data The average subscription price increase for Fairfax’s The Dominion Post and The Christchurch Press since 2013 is [ ]% whereas the increase in price for the Sunday Star Times is [ ]%. The average subscription price increase for NZME’s New Zealand Herald and regional newspapers is [ ]% whereas the increase in price for the Herald on Sunday is [ ]%. Source: NZME and Fairfax data 146 800. At present, the Applicants’ respective Sunday newspapers overlap only in the North Island.414 However, the cover and subscription prices across the North Island and the South Island are the same. While the proposed merger would not change competitive conditions in the South Island for the supply of Sunday newspapers, we cannot discount the likelihood that price increases in the North Island would be applied equally in the South Island. As a consequence, the upper bound of the estimated allocative efficiency loss incorporates the likely maximum price increases to both the North and South Islands. National online news 801. We have been unable to calculate the allocative inefficiency loss associated with the introduction of a paywall on content on an Applicant’s website in the usual way, as we consider that the price for online news without the merger would be zero.415 802. [ 416 417 ]. 803. 804. 414 We view a metered paywall as a means of price discriminating across low and heavy users. We consider that three broad groups of readers are potentially affected by a paywall: 803.1 heavy users who would likely subscribe; 803.2 heavy users who would not subscribe; and 803.3 moderate to light users who would not be (or only rarely be) affected by the paywall. We do not consider this last group any further. The first group of heavy users who would be likely to subscribe demonstrate a willingness to pay that is equivalent or greater than the price, so there is no deadweight loss associated with this group. That is, given that the price is a fixed fee for unlimited consumption, no one in this group would have consumed more news See [261] to [262] above. Zero prices mean that calculations for elasticities and changes in quantities are mathematically undefined and indeterminate respectively. The usual approach to calculating quantity reductions and associated allocative efficiency detriments and wealth transfers therefore cannot be used. 415 416 [ 417 [ ] ] 147 had the price been lower. The subscription fee does, however, result in a transfer from these readers to the merged entity. 805. 806. The second group consists of readers who may have subscribed had the paywall subscription price been lower. We do not know, and have no information to help inform, the shape of their demand. We consequently adopt two extremes: 805.1 One, no readers value viewing a website more than [ ]. That is, no matter how low the paywall subscription price, they would be unwilling to pay for more than [ ]. 805.2 Alternatively, we assume that if the paywall subscription price had been marginally lower, then more readers would have viewed a website more than [ ]. We assume that this relationship is approximated by a linear demand curve.418 [ ]. 807. 418 419 420 We use the number of “unique browsers”419 that visited nzherald.co.nz in the year to September 2016420 and undertook a simple modelling exercise to estimate the number of readers that would choose to subscribe to a [ ] priced [ ]. We note that this is unlikely to be the true underlying demand curve. In particular, it is likely that demand is largely horizontal at price zero for a certain portion of consumers who choose to switch to alternative providers as a result of a paywall. This is consistent with Hsiang Iris Chyi “Willingness to Pay for Online News: An Empirical Study on the Viability of the Subscription Model” (2005) 18(2) JME 131-142, at 133, who suggests that demand for news is typically completely horizontal when free, with any increase in price resulting in a complete reduction of quantity consumed. We also consider that there is a subset of readers who have a positive willingness to pay for viewing online news, but whose willingness to pay does not meet the price of a digital subscription. These may be readers who have a willingness to pay for specialised or local content news or if the reader cannot access the same content on a free website (see the survey of reasons for willingness to pay for online news in Manual Goyanes “An empirical study of factors that influence the willingness to pay for online news” (2014) 8(6) JP 742-757, at 743). The underlying demand curve may consequently be more accurately described as relatively “L-shaped”. The estimated dead weight loss under the assumed linear demand may consequently overstate the true allocative efficiency loss. By “unique browser” we mean the number of unique devices accessing nzherald.co.nz. It is likely that some readers access nzherald.co.nz via multiple devices. The “unique browser” figure may consequently overstate the true number of unique consumers accessing nzherald.co.nz over a year. We have the total number of page impressions on nzherald.co.nz year to date September 2016. We also have the total number of page impressions and unique browsers that visited stuff.co.nz in the same period. Figures for stuff.co.nz suggest that unique browsers generate on average [ ] page impressions a year. We use this figure to calculate an estimated number of unique browsers that visited nzherald.co.nz for the year ended September 2016. Our calculations suggest that [ ] unique browsers visited nzherald.co.nz year to date September 2016. Source NZME and Fairfax data. 148 808. [ 421 ]. 809. [ 809.1 ]: [ ]. 809.2 [ ]. 809.3 [ ]. 809.4 [ ]. 810. 421 422 The resulting figure provides us with the number of unique readers that would be subject to a paywall.422 To estimate the number of unique readers that would choose to subscribe, we assume a [ ]% conversion rate.423 This calculation suggests that [ ] This resulting figure suggests that 6.4% of all readers would be subject to a paywall. [ ] 423 That is, we assume that [ ]% of all unique readers would choose to subscribe to a [ 149 [ [ ] unique readers would choose to subscribe after implementation of a ] paywall. 811. The difference between the number of unique readers that choose to subscribe and the number of unique readers that would be subject to a paywall provides an estimate of the group of readers who may have subscribed had the price of the paywall been lower. 812. As outlined above, we adopted two extremes when we calculated the allocative efficiency loss related to this group of readers: 812.1 One, that no matter how low the paywall, they would have no willingness to pay. Allocative efficiency loss would be zero as a result; 812.2 Two, we assume that more readers would have subscribed had the subscription price been lower. We assume that this is approximated by a linear demand curve.424 Allocative efficiency loss using this method amounts to $[ ] million.425 813. The revenue generated by the paywall from readers that choose to subscribe represents the wealth transfer as a result of the paywall. Absent the paywall, readers that subscribe would have viewed more than [ ] for a price of zero. The payment made, in the form of the subscription, therefore constitutes a transfer from consumers to the merged entity. 814. As discussed above, we cannot state with certainty what the profit maximising price or meter amount would be post-merger, nor the corresponding effect this has on reader demand. It may well be the case that allocative efficiency losses are higher with a higher subscription price. 815. [ ] to estimate the allocative efficiency loss under this particular form of a paywall. We discuss paywalls with a higher subscription price, and its ramifications on detriments as a result of the merger qualitatively below. ]. 424 425 The implicit slope of this demand curve is -0.0012. That is, a 1% reduction in the annual price of the paywall results in a 0.12% increase in the number of subscribers. This implies that a 1% reduction in price would lead to an additional [ ] unique consumers subscribing. The relatively large allocative efficiency loss is a result of the relatively larger number of readers, ([ ]) who would be subject to a paywall that choose not to subscribe when compared to readers modelled as subscribing ([ ]). As noted above, the use of a linear demand curve may also overestimate the true allocative efficiency loss should there be a subset of readers who do not have a positive willingness to pay regardless of the subscription price. 150 Elasticities 816. To estimate the allocative efficiency loss that would arise from these potential price increases, we must also assess the likely elasticities of demand for the various markets of concern. 817. In modelling provided by NERA on behalf of the Applicants, allocative efficiency losses are calculated using a range of elasticities. In particular, elasticities of: 817.1 -0.2 to -1.02 for community newspaper advertising; 817.2 -0.2 to -1.04 for Sunday newspaper advertising; 817.3 -0.2 to -1.68 for Sunday newspaper supply; and 817.4 -0.2 to -1.05 for premium digital advertising. 818. NERA’s estimates for community and Sunday newspaper advertising and Sunday newspaper supply are based on recent studies that capture the impact of online substitution, as well as critical loss analysis. The studies used suggest elasticities of 0.7 and -1.2426 for newspaper advertising and -1.75 for newspaper readers.427 819. We have undertaken critical elasticity analysis to help inform the range of elasticities used.428 On the basis of results from our critical elasticity analysis, the following range of elasticities are used: 819.1 -0.2 to -0.92 for community newspaper advertising; 819.2 -0.2 to -1.01 for Sunday newspaper advertising; 819.3 -0.2 to -0.98 for premium digital advertising; 819.4 -0.2 to -1.22 for retail Sunday newspaper supply; and 819.5 -0.2 to -1.68 for subscription Sunday newspaper supply. 426 427 428 Lapo Fillstrucchi, Tobias J. Klein and Thomas O. Michielsen “Assessing unilateral merger effects in a twosided market: an application to the Dutch daily newspaper market” (2012) 8.2 JCLE 297-329 [Assessing merger effects in two sided markets] and Ying Fang “Ownership consolidation and product characteristics: A study of the US daily newspaper market” (2013) 103.5 AER 1598-1628. Assessing merger effects in two sided markets, above n 440. We use the breakeven critical elasticity approach and a linear demand. See Gregory Werden “Beyond Critical Loss: Tailoring Applications of the Hypothetical Monopolist Paradigm” US DOJ Antitrust Division Economic Analysis Group Discussion Paper No. 02-9. We note, as do the Applicants, that applying the standard formula for critical loss analysis in a one-side market may not be the correct approach, as it does not account for the cross-platform externalities between newspapers and online websites, advertisers and readers. However, this approach is likely to be conservative, as the full impact of a price increase on demand is likely to be underestimated, meaning that the critical elasticity is over-estimated. Detriments are higher at higher elasticities and so this approach is conservative. 151 820. The overall range of total potential allocative efficiency losses on an annual basis is between $0.7 million to $20.9 million depending on the size of the price increase and the elasticity of demand. 821. The estimated allocative efficiency losses are broken into the following components in Table 10. Table 10: Estimated allocative efficiency losses by category Estimated price increase Annual detriment 5% to 10% price increase for community newspaper advertising [ ] 5% to 10% price increase for Sunday newspaper advertising [ ] 5% to 10% price increase for premium digital advertising [ ] 5% price increase for Sunday paper cover prices [ ] 5% price increase in Sunday paper subscription prices [ ] Allocative efficiency losses from paywall Total [ ] $0.7 million to $20.9 million Detriments – digital only 822. As outlined above, to assist us in applying our judgement, we have also estimated what the detriments would be if the merged entity immediately ceased publishing all its print products (except community newspapers) and shifted to a ‘digital only’ businesses. As noted previously,429 we consider that the likely detriments from the proposed merger would be greater than those calculated under the ‘digital only’ scenario. Detriments 823. 429 To meet unavoidable costs previously met by certain print publications, under the ‘digital only’ scenario prices may be raised. The result of this is that some price increases are likely to occur regardless of whether the merger proceeds and would not be considered detriments attributable to the merger as a result. We need to discount such detriments. See [752] above. 152 824. 825. 430 431 432 It is not certain what a ‘digital only’ scenario would look like. In order to estimate the likely detriments in this scenario, we have made the following assumptions: 824.1 The detriments calculated with respect to community newspaper advertising would remain unaffected. As stated previously, we consider the Applicants would continue to publish community newspapers. While it may be the case that increased prices for community newspaper advertising meet some of the costs previously met by other publications, we account for this in detriments calculated for New Zealand online news and premium digital advertising below. 824.2 We discount completely the detriments arising from the introduction of a paywall because we consider that a paywall may be implemented to meet costs previously met by print publications. The Applicants’ current newsroom costs suggest that the unavoidable costs that would have to be met by remaining ‘digital only’ operations would be in excess of the revenue generated by a paywall.430 824.3 However, as discussed previously,431 it is our preliminary view that the removal of competition from the proposed merger could lead to a digital subscription price that is higher than it would otherwise be in a competitive market. In such a case, there is likely to be detriments arising from the introduction of a paywall even under a ‘digital only’ scenario. While we are unable to quantify the detriments arising from a paywall with a higher digital subscription price in a ‘digital only’ scenario, we discuss it qualitatively below. 824.4 We discount completely the detriments arising from likely price increases for premium digital advertising. The Applicants’ unavoidable costs, such as newsroom costs, suggest that these are in excess of the revenue (ie, the wealth transfer) from increased prices for premium digital advertising.432 The net result of our simplifying assumptions is that detriments on an annual basis are: In particular, the total editorial costs for Fairfax for the year 2015 totals $[ ] million dollars. This is in excess of the estimated gross revenue the Applicants would derive from a paywall, ie, $[ ] million. We note that true newsroom costs in a ‘digital only’ scenario would likely be smaller, as news services/editorial costs would be smaller following the shutting down of print publications. In any case, we assume the revenue from a paywall is necessary to cover newsroom costs previously met by print publications. We assume that a paywall may be implemented to meet costs previously met by print publications. See [539] above. Editorial costs for Fairfax of $[ ] million outweigh both the estimated gross revenues the Applicants would derive from a paywall, ie, $[ ] million, and increased prices for premium digital advertising, ie, $[ ] million. We assume that the maximum likely price increases for premium digital advertising is equal to the appropriate cost of producing premium digital advertising. 153 ] in productivity efficiency detriments;433 825.1 zero to $[ 825.2 $[ 825.3 $0.2 million to $1.5 million in allocative efficiency detriments.435 ] to to $[ ] million in dynamic efficiency detriments;434 and Wealth transfers to and from non-New Zealanders from price increases 826. If the proposed merger allowed the merged entity to exercise market power by raising prices it would, in addition to the associated allocative efficiency loss, result in a transfer of wealth from consumers to the merged entity. Where one group gains at the expense of another in this manner, the transfer of wealth is not generally included in the Commission’s analysis of benefits and detriments. This is because there is typically no net impact to New Zealand from such a transfer.436 827. However, an exception to the approach of ignoring such transfers is if one or more groups are non-New Zealanders. The Authorisation Guidelines state:437 Wealth transfers may become relevant where the transfer is between New Zealanders and non-New Zealanders. This is because the public benefit test focuses on benefits to New Zealanders. As a result, transfers of wealth from non-New Zealanders to New Zealanders may be a public benefit. Similarly, transfers of wealth in the opposite direction may be a public detriment. However, in addition to considering the direct effects of wealth transfers, we also consider any effects on non-New Zealanders that may ultimately feedback to impact New Zealanders. For example, if a transaction would lead to a New Zealand firm charging higher prices to tourists, that would result in a transfer of wealth from those tourists to the New Zealand firm resulting in a public benefit. However, equally, those higher prices could lead to fewer tourists coming to New Zealand, which in the longer term could negatively affect New Zealanders. 828. As made clear by the Air New Zealand/Qantas438 decision, not only does the nationality of those who receive the wealth transfer as a result of their exercise of market power matter but so might the nationality of those who incur the price increase (in the case of Air New Zealand/Qantas, the foreign tourists). 829. To determine the nature (benefit or detriment) of wealth transfers, it is first necessary to determine the residency status of those who would receive the transfers, ie, the shareholders of the merged entity, as well as those who may incur 433 434 435 436 437 438 Productive efficiency losses are calculated using a range of zero to 1% of the pre-merger variable costs for community newspaper advertising and premium digital advertising. Dynamic efficiency losses are calculated using a range of zero to 0.5% of total sales revenue for community newspaper advertising and a range of 0.5% to 1% of total sales revenue for premium digital advertising. Allocative efficiency losses are calculated using a range of price increases of between 5% to 10% for community newspaper advertising. This is distinct from allocative efficiency impacts described above, which relate to the lost consumer and producer surpluses arising from lower quantities of output bought and sold in the market. Authorisation Guidelines, at [54] and [55]. Air New Zealand and Qantas Airways Limited v Commerce Commission (2004) 11 TCLR 347. 154 the burden of any price increase. The implications of this is that the nationality of both the owners of the merged entity and those purchasing advertising or news may be considered when determining how the wealth transfer is to be treated in the benefits and detriments analysis. 830. This is not to say that all returns that non-New Zealanders receive from investments in New Zealand should necessarily be considered detrimental transfers of wealth to foreigners. New Zealand benefits to a large degree from the inflow of foreign capital. 831. In addition to determining the residency status of those who receive wealth transfers and those who incur the burden of any price increase, we must determine the magnitude of the wealth transfer. Determining the magnitude of wealth transfers in turn entails an assessment of whether transfers comprise of supra-competitive returns. 832. Supra-competitive returns are those returns which are over and above those needed to incentivise efficient investment. They may arise from the exploitation of market power. These can be distinguished from competitive returns to capital that are necessary to incentivise efficient investment. Approach to quantifying wealth transfers 833. As with our estimation of the likely detriments arising from the merger, we proceed on the basis of quantifying detrimental wealth transfers under the status quo, acknowledging that the likely detrimental wealth transfers arising from the merger would be lower. 834. Similarly, we also proceed on the basis of quantifying detrimental wealth transfers under the ‘digital only’ scenario, acknowledging that the likely detrimental wealth transfers arising from the merger would be greater. 835. Our approach of quantifying wealth transfers under the status quo and the ‘digital only’ scenario has implications for the wealth transfers calculated in this section: 835.1 We consider that in the status quo, current prices would be more than sufficient to incentivise investment into the various markets affected by the merger. Any increase in price would suggest that shareholders would be deriving economic rents. In such a scenario, we consider that any incremental transfer of wealth from New Zealanders (eg, New Zealand advertisers or news readers) to non-New Zealand shareholders of the merged entity that arises from an increase in price would constitute a detriment of the transaction. 835.2 In the ‘digital only’ scenario, we consider that increases in price may be required to meet unavoidable costs previously met by revenues from print publications. That is, some or all of the price increase and resulting wealth transfer would be a competitive return as it would be necessary to incentivise investment in the remaining markets of operation of the merged entity. We detail how we distinguish competitive wealth transfers below. 155 Residency status of relevant groups 836. In determining benefits and detriments to the public of New Zealand, New Zealanders are considered to be those who are domiciled in New Zealand.439 837. As stated previously, post-merger NZME would acquire all of the shares in Fairfax. In exchange, NZME will pay $55 million in cash and issue shares equal to a 41% shareholding in NZME to Fairfax Corporation Pty Ltd. Fairfax Corporation Pty Limited is an Australian registered and owned company that presently holds 100% of the shares in Fairfax.440 838. As at 12 October 2016, NZME’s New Zealand-based shareholding amounted to [ ]%.441 For the purposes of calculating wealth transfers, we treat the percentage of NZME’s New Zealand-based shareholding as shares owned by New Zealand domiciled shareholders. 839. On this basis, post-merger, we understand that 41% of the shareholding of NZME would be held by Fairfax Corporation Pty Ltd. To achieve this, we assume that NZME would issue additional shares equivalent to 69.5% of the current shareholding in NZME to Fairfax Corporation Pty Ltd. Assuming no other material change in the current NZME share register, the net result of this would be that [ ]% of the merged entity would be New Zealand-owned post-merger. 840. In summary, for the purposes of calculating wealth transfers, we consider that: 841. 439 440 441 840.1 [ ]% of the merged entity would be owned by New Zealand-domiciled shareholders; and 840.2 [ ]% of the merged entity would be owned by overseas-domiciled shareholders. Figure 9 below provides a simplified view of the post-merger structure. Authorisation Guidelines, at footnote 44. Pre and post completion adjustments could result in a greater or lower amount of cash or shares being paid to Fairfax. https://www.nzx.com/companies/NZM/announcements/288557. Email from [ ] to [ ] (25 October 2016). 156 Figure 10: Simplified post-merger structure 842. We have also considered the residency of consumers who would purchase advertising and news from the merged entity. The Commission recognises that some purchasers of advertising and news may be non-residents. However, it is difficult to come to a firm conclusion on the likely residency of such consumers due the varying identity and ownership structures of such consumers. As a conservative measure, we assume that all consumers of advertising and news are New Zealanders, as under this assumption all transfers to foreign shareholders of the merged entity would be treated as detrimental.442 Components of transfers 843. The total net wealth transfer received by the merged entity can be divided into two components. The first component is the transfer from New Zealand consumers of advertising and news to foreign shareholders of the merged entity. This component accounts for [ ]% of all wealth transfers arising as a result of the proposed merger. We treat these transfers as detriments. This is because the negative impact arising from transfers of wealth from New Zealand consumers of advertising and news is not offset by the gains to foreign shareholders of the merged entity since they accrue to non-New Zealanders. 844. The second component that arises from the proposed merger is the transfer from New Zealand consumers of advertising and news to New Zealand shareholders of the 442 We also note that the economic incidence of increased prices to advertising consumers may be borne by ultimate consumers of the goods or services being advertised. We have not assessed the likely rate of any such pass-through. 157 merged entity. This component accounts for [ ]% of all wealth transfers arising as a result of the merger. We treat this transfer as neutral. This is because the negative impact arising from transfers of wealth from New Zealand consumers of news and advertising are offset by gains to the New Zealand shareholders of the merged entity. Wealth transfers – status quo Magnitude of transfers 845. We consider that in the status quo, current prices are sufficient to maintain incentives for efficient investment in the markets of concern. Any likely post-merger increases in prices, particularly when coupled with a reduction in costs, would suggest that shareholders would be deriving supra-competitive returns. 846. In light of the above, we consider that total net wealth transfer from New Zealand consumers to foreign shareholders of the merged entity should be considered as a detriment. Total wealth transfers 847. The potential size of the total transfer to the merged entity has been estimated for price increases and elasticities detailed under Table 11 below. 848. We consider that the total net transfer from New Zealand consumers to foreign shareholders of the merged entity should be considered as a detriment. Consequently, we have reduced transfers by [ ]%, to account for the purchases of advertising, print and news that are made by New Zealand consumers to New Zealand shareholders of the merged entity. 849. Finally, the transfer estimates are scaled down by an assumed effective marginal tax rate on foreign shareholders of 28%.443 This accounts for the fact that it is only the after-tax portion of these amounts that may be transferred to non-New Zealanders. 443 This is based on the corporate tax rate. 158 Table 11: Estimated price increases, elasticities and total wealth transfers Estimated price increase Elasticities 5% to 10% price increase for community newspaper advertising -0.2 to -0.92 [ ] 5% to 10% price increase for Sunday newspapers advertising -0.2 to -1.01 [ ] 5% to 10% price increase for premium digital advertising -0.2 to -0.98 [ ] 5% price increase for Sunday Paper cover prices -0.2 to -1.22 [ ] 5% price increase in Sunday Paper subscription prices -0.2 to -1.68 [ ] Wealth transfer from paywall 850. Estimated transfer [ ] In summary, we calculate that the net detrimental wealth transfer to non-New Zealanders under the status quo scenario ranges from $6.2 million to $7.8 million per year. As noted above, we consider that the likely detrimental net wealth transfer would be below this range because we consider the ongoing continuation of the status quo unlikely. Wealth transfers – digital only Total wealth transfers 851. Under the ‘digital only’ scenario, to meet unavoidable costs previously met by print publications, prices would likely need to be raised. The price increase and resulting wealth transfer would be considered competitive, as it would be necessary to incentivise investment in the remaining markets of operation of the merged entity. 852. We approach our consideration of competitive wealth transfers in the same way we approach the quantification of detriments in the ‘digital only’ scenario. In particular, we make the following assumptions: 852.1 All wealth transfers with respect to community newspaper advertising are a detriment to the extent that they go to non-New Zealanders. As stated previously, we consider the Applicants would continue to publish community newspapers. 159 853. 852.2 There are no wealth transfers associated with Sunday newspaper advertising and the supply of Sunday newspapers since these papers are assumed to no longer exist in this scenario. 852.3 As noted in the discussion of allocative inefficiencies, no such inefficiency arises from the introduction of a paywall since the price increase would be necessary to meet costs previously met by print revenues. This, in keeping with the above discussion of the treatment of transfers when price is below cost, indicates the wealth transfer from the paywall would be a competitive return as it would be necessary to incentivise investment in continuing operations. As a result we do not consider the wealth transfer from a paywall to be a detriment as a result. 852.4 However, as discussed previously,444 it is our preliminary view that the removal of competition from the proposed merger could lead to a digital subscription at a price higher than it would otherwise be in a competitive market. We discuss the implications of this further below. 852.5 As noted in the discussion of allocative inefficiencies under the ‘digital only’ scenario, no such inefficiencies arise either from likely price increases for premium digital advertising. This indicates that the wealth transfer from likely price increases for premium digital advertising would be a competitive return as it is necessary to incentivise investment in continuing operations. We do not consider the wealth transfer from likely price increases in premium digital advertising a detriment as a result. The net result of our assumptions is that that the post-tax wealth transfers accruing to non-New Zealanders on an annual basis are between $0.5 and $1.0 million under the ‘digital only’ scenario. Conclusion on the quantified benefits and detriments 854. This application involves a balancing of the public benefits and detriments which will, or will be likely to result, from the proposed merger. 855. In this section we bring together the quantified benefits and detriments. In the following section, we describe the unquantified benefits and detriments. We then balance all benefits and detriments – quantified and unquantified – to determine whether we are satisfied that the proposed merger will result, or be likely to result, in such a benefit to the public that it should be permitted. 856. As we have explained, we have estimated those quantifiable benefits and detriments on the basis of the status quo. We have also considered the likely detriments if the Applicants were to immediately cease all print production (with the exception of community newspapers). We have done this to help us inform our assessment of the magnitude of the benefits and detriments. 444 See [539] above. 160 857. We have treated all productivity efficiency gains that would be obtained by the merged entity as public benefits. This is despite the fact that some proportion of these gains flow directly to foreign shareholders in the first instance. This approach, consistent with the case law,445 recognises that enabling foreign shareholders to undertake such cost minimisation can provide significant feedback effects to New Zealand. 858. In keeping with the Commission’s view that the likely scenario without the merger would require some rationalisation of newspaper publishing to some extent, we consider that productivity gains obtained by the merged entity would be of the type that assists the merged entity with stemming the decline of the profitability of its print business. Consequently, we do not consider that productivity gains accruing to the foreign shareholders of the merged entity are likely to constitute supracompetitive returns, at the very least over the short term.446 859. Tables 12 and 13 below summarise the Commission’s quantitative assessment of the detriments and benefits arising from the merger under the status quo. We have provided a five year forecast to illustrate the net impact under the status quo for illustrative purposes only, noting again the artificiality of this scenario. Table 12: Summary of detriments – Status quo Category $0.7 million to $20.9 million $2.8 million to $87.0 million Productive efficiency [ [ Dynamic efficiency [ Total of quantified detriments 446 5-year NPV Allocative efficiency Net wealth transfers 445 Evaluation ] ] $6.2 million to $7.8 million [ ] ] $25.8 million to $32.7 million $29.2 million to $122.4 million Telecom Corporation of New Zealand Ltd v Commerce Commission (1991) 4 TCLR 473 (HC) and Telecom Corporation of New Zealand Ltd v Commerce Commission (1992) 4 TCLR 648 (CA) (AMPS-A). This approach was most recently upheld in Godfrey Hirst NZ Ltd v Commerce Commission [2016] NZHC 1262. We note that the treatment of productivity gains accruing to foreign shareholders is currently an issue on appeal at the Court of Appeal. We also note that another reason to treat productivity efficiency gains flowing to foreign shareholders of the merged entity as public benefits is that to do otherwise would effectively discriminate against such shareholders in comparison to domestic shareholders. Placing foreign owned businesses at a relative disadvantage in merger authorisations would create a disincentive for foreigners to undertake investment in New Zealand more generally. Such a disincentive could be detrimental given the wider benefits that arise from inbound foreign investment. These benefits include a higher stock of available capital and lower cost of capital for the New Zealand economy. 161 Table 13: Summary of benefits – Status quo Category Evaluation Sales Mix of one-off costs and ongoing benefits [ ] Marketing Mix of one-off costs and ongoing benefits [ ] Editorial Mix of one-off costs and ongoing benefits [ ] Printing, procurement and distribution Mix of one-off costs and ongoing benefits Management Mix of one-off costs and ongoing benefits [ ] Information technology Mix of one-off costs and ongoing benefits [ ] Premises Ongoing benefits [ ] Community rationalisation Mix of one-off costs and ongoing benefits [ ] Sunday rationalisation Mix of one-off costs and ongoing benefits [ ] Merger transaction costs One-off cost [ ] Finance costs Ongoing costs [ ] Total quantified benefits 860. 5-year NPV [ ] $136.5 million to $218.7 million The net impact of the quantifiable benefits and detriments under the status quo are summarised below: Table 14: Estimated net quantifiable impact under ‘status quo’ scenario Time frame High detriment/low benefits Low detriment/high benefits 5 years $14.1 million $189.5 million 162 861. Tables 15 summarises the Commission’s quantitative assessment of the detriments arising from the acquisition under the ‘digital only’ scenario.447 Table 15: Summary of detriments – Digital only Category Evaluation Allocative efficiency $0.2 million to $1.5 million $0.7 million to $6.3 million Productive efficiency [ [ Dynamic efficiency [ Net wealth transfers Total of quantified detriments ] ] $0.5 million to $1.0 million 5-year NPV [ ] ] $2.3 million to $4.1 million $3.5 million to $12.0 million 862. Under the status quo we estimate that the proposed merger would result in quantified net benefits of $14.1 million to $189.5 million over five years. 863. The figures above demonstrate that under both high detriments/low benefits and low detriments/high benefits in the status quo, the net impact is positive. This range is large, reflecting the uncertainty of the estimates. 864. As previously discussed, we consider the range of benefits and detriments calculated under the status quo are unlikely. We consider that the likely benefits and detriments arising from the merger are likely to fall below this range. 865. The true net impact depends on the magnitude of the benefits and detriments that would arise regardless of the merger. However, as previously outlined, we cannot predict this with any degree of precision, given the uncertainty surrounding the level of print-related rationalisation that would occur absent the merger. 866. In a ‘digital only’ world, quantifiable detriments are much lower. This is driven primarily by the reduction in allocative efficiency losses and wealth transfers as a result of less merger-specific price increases. Applying these detriments to the claimed cost-savings would also produce a materially positive net quantified benefit in the region of $16.9 to $41.3 million within the first year. 867. However, we consider merger-specific savings would also likely be substantially lower. While there are likely to be some benefits, it is likely that in the ‘digital only’ scenario, cost-savings arising from printing, procurement, distribution and Sunday newspaper rationalisation would occur regardless. We would also expect further 447 See [823.1] – [823.4] and [851.1] – [851.5] for the simplifying assumptions underpinning this assessment. 163 cost-savings to be realised regardless of the merger. These could include, for example, further costs savings related to the downsizing of management, marketing, and editorial functions, as well as reduced premises costs. 868. We have not attempted to speculate as to what the precise cost-savings in the ‘digital only’ scenario would be. However, we consider that they would be substantially below the cost-savings calculated in the status quo scenario. 869. While by no means definitive, we take the results suggested from our quantitative analysis under the status quo and ‘digital only’ scenarios as indicating that the net impacts are likely to be positive, with estimates under the status quo scenario suggesting that net benefits could exceed $15.7 million over five years. 870. We now turn to consider the unquantified benefits and detriments. Unquantified benefits Increased quality in reader markets 871. The Applicants submitted that the proposed merger would not result in a reduction in quality in the reader markets. Rather, the Applicants suggested that the proposed merger would result in a reduction in duplication and free-up journalistic resources to report on an increased number of stories. The Applicants considered that the proposed merger would allow them to “to invest in better quality journalistic content and greater breadth of coverage” to the benefit of competition.448 872. We recognise that there may be benefits associated with reductions in duplication of content and the freeing-up of journalistic resource. Studies confirm that such quality benefits may arise as a result of mergers.449 873. However, as discussed under Section 5, we consider that the proposed merger would merge the closest competitors for online and print news content. Further, the evidence we obtained suggests that the constraint the Applicants place on each other is a substantial driver of quality for online and print news content. Unquantified detriments Decreased quality in reader markets 874. 448 449 As outlined in Section 5 above, the Commission has reached the preliminary view that a substantial lessening of competition would be likely to arise in the reader market through decreased content quality. Further, as we discuss below in our Application, at [20.8] and Fairfax/NZME Response to SOPI Submissions, at [34]. For example, Przemysaw Jeziorski “Effects of mergers in two-sided markets: The US radio industry” (2014) 6(4) American Economic Journal: Microeconomics, 35-73, and Andrew Sweeting “The effects of mergers on product position: evidence from the music radio industry” (2010) 41(2) The RAND Journal of Economics, 372-397, find that mergers as a result of the removal merger restrictions under the US Telecommunications Act 1996 lead to more product variety. Similarly, Lisa George “What’s fit to print: The effect of ownership concentration on product variety in daily newspaper markets” (2007) 19(3) Information Economics and Policy, 285-303, finds that increased concentration may reduce the duplication of genres. 164 consideration of the potential the loss in plurality from the proposed merger, we consider that it is the competition between the Applicants that motivates the volume, variety and objectivity of news that the Applicants produce for online and print news content in reader markets. We do not consider that the threat of expansion or the two-sided nature of the reader market is likely to constrain these quality-reducing cost reductions. 875. As a consequence, we consider the removal of the competition between the Applicants post-merger would have a detrimental impact on the quality of online and print news content produced by the Applicants. 876. Further, given the influence the Applicants have in setting the news agenda in New Zealand, we consider that the likely reduction in quality as a result of proposed merger would have a wider detrimental impact on the quality of online and print news content. 877. Conceptually, we consider that reduced quality of online and print news content in reader markets can be seen as reducing the value of news content to consumers. We consider that this may be reflected in a reduced willingness to pay for news content, which in turn could be seen as a reduction in demand for online and print news content. While we do not have the means to practically, or with certainty, predict the likely reduction in demand due to the reduction in quality of news content associated with the proposed merger, the extent of the reduction in demand would depend not only on the change in quality, but also on the extent to which users value that quality and the extent to which change is visible to them. 878. Further we consider that the merger would reduce readers’ options for online New Zealand news. Therefore, readers may not be able to switch away from the merged entity’s publications. 879. We note that any reduction in willingness to pay for news content as a result of reduced quality is likely to lead to welfare losses. Readers have a quality adjusted willingness to pay that is higher than or just equal to what they effectively pay for the current quality of news content. With a reduction in quality and a corresponding reduction in willingness to pay, the surplus (ie, the difference between the higher willingness to pay and the effective price for the quality of news currently consumed) previously gained by consumers is lost. 880. We consider that the likely reduction in quality as a result of the proposed merger is likely to lead to a significant impact on total welfare, as suggested by recent research. In a recent paper, Ying Fan estimates a structural model for the United States newspaper market using data from 1997 to 2005. When analysing a proposed merger between two newspapers, she shows that the merger would reduce quality compared to the outcome without the merger, and that ignoring quality change can underestimate the welfare effects of a merger.450 450 Ying Fan, “Ownership consolidation and product characteristics: A study of the US Daily Newspaper Market” (2013) 103(5) American Economic Review, 1598-1628. 165 881. In contrast, we note, as above, that there are studies suggesting that mergers may lead to reduced duplication and increased variety. However, where there is evidence of likely detrimental impacts on quality, the recent research discussed above indicates that there is potential for detrimental welfare effects to be relevant. 882. Evidence on hand suggests that readers of online and print media currently, and post-merger, value the quality of online and print news content. While perhaps selfevident, [ ].451 883. [ ].452 884. Overall, we consider that the proposed merger is likely to result in a reduction in quality of news content in the reader markets. This, coupled with evidence that consumers perceive and value quality of news content, leads us to consider that the proposed merger is likely to lead to significant welfare losses. Detriments from varying forms of paywalls 885. In this section, we discuss detriments that may arise from a paywall with a higher subscription price. As detailed above, we have been unable to quantify allocative efficiency losses and detrimental wealth transfers from a paywall that differs [ ]. As a consequence, we discuss detriments that may arise from a paywall with a higher subscription price than there otherwise would be in a competitive market in a qualitative manner. 886. [ ]. Post-merger, it could therefore be the case that the profit maximising price for a digital subscription could be higher. We consider that a paywall with an increased digital subscription price is likely to lead to additional allocative efficiency loss and detrimental wealth transfers.453 451 [ 452 453 ] We note, as discussed earlier, that some detrimental impacts on quality, such as a reduction in the volume, accuracy and volume of news content, may not be immediately obvious to readers post-merger. We discuss the implications of this under our discussion of plurality detriments below. The wealth transfer from an increased digital subscription price will necessarily be greater than that modelled if it is considered to be profit-maximising post-merger. That is, the merged entity would only choose to implement a paywall with an increased subscription price if the increased profits from a higher 166 887. With a higher digital subscription price, we consider that some unique readers would likely choose to no longer subscribe, as their willingness to pay for articles above the metered amount is exceeded by the digital subscription. The result of this is that there would be additional allocative efficiency loss, as there would be a larger number of readers who would have subscribed had the subscription price been lower when compared to the situation under a lower subscription price. 888. We note that in our analysis above, we assumed two extremes. In particular, at one extreme, we assumed that the price increase would not result in any allocative inefficiency because those consumers who hit the free article limit would be unwilling to pay a higher price no matter how low the paywall. However, a higher price would likely cause some portion of those readers who would choose to subscribe at the lower price, to stop subscribing. There would be an allocative inefficiency associated with those customers. 889. We also considered that even under a ‘digital only’ scenario, a paywall with an increased digital subscription price may lead to allocative efficiency losses and detrimental wealth transfers.454 These allocative efficiency losses and detrimental wealth transfers may arise not simply because of the implementation of a paywall, but because of our view that it is likely to be implemented with a higher digital subscription price than would otherwise exist in a competitive market. 890. We do not speculate as to the precise nature of the additional allocative efficiency loss and detrimental wealth transfer that may result as a result of a paywall with a higher digital subscription price. We simply note that possible detriments associated with a paywall with a higher digital subscription price are potentially higher than those considered under quantified detriments and wealth transfers. Unquantified detriments: media plurality effects 891. 454 If the merger proceeded, it would provide the merged entity with direct control of: 891.1 the largest newsroom and network of reporters and journalists in the country, employing almost twice the number of editorial staff that are employed by the next three largest mainstream media organisations combined; 891.2 nearly 90% of daily circulation of New Zealand’s daily newspapers, and indirect control over much of the national content provided to the remaining independently owned print media; subscription price outweigh the amount of lost profits associated with the number of unique consumers who opt no longer to subscribe due to the higher price. See [823.3] and [851.4] above. 167 891.3 the two largest New Zealand news websites, which together have a population reach more than four times larger than the next biggest New Zealand news website; and 891.4 one of New Zealand’s two largest commercial radio companies. 892. This level of control over New Zealand’s news media output across a range of media platforms by a single commercial enterprise would be highly unusual when compared with other modern liberal democracies.455 It raises questions as to the possible impact of the merger on media plurality. 893. Media plurality is generally used to mean two slightly different things:456 893.1 A wide range of information, opinions and perspectives are available; and 893.2 No individual media owner should be in a position to shape or control the news and political agenda, either by influencing media users through editorial content or by directly leveraging the political process. 894. Both these facets of plurality are relevant to this merger. It is the Commission’s view that plurality in this context means more than the availability of a wide range of views and opinions. While access to and consumption of a range of views and opinions is important, equally important is that there is a range of organisations generating New Zealand news content, as well as controlling the dissemination of that content more widely throughout society. Plurality can manifest in, for example, differences in editorial choice, availability of different editorial perspectives on key issues of the day, news and story coverage, and the extent and choice of investigative journalism. 895. In the Commission’s view, the impact of the proposed merger on media plurality is a relevant consideration in assessing whether the merger is likely to result in such a benefit to the public that it should be permitted. The Commission’s concern is in regard to New Zealand news content where that includes local, regional and national news, including New Zealand perspectives on international news. 896. This assessment includes evaluating not only direct impacts on users of news media content, but also any flow-on effects that increased concentration may have on the wider public. The flow-on effects of plurality are considered to include increased government accountability, ensuring politicians, corporates and lobby groups find it 455 456 For example, in Ireland, a similar-sized, English-speaking country, the country’s largest newspaper company accounts for 52% market share of daily newspapers (Dr Julienne Molineaux et al., submission to the Commerce Commission (July 2016) referencing Noam (2016) Who Owns the World’s media?: Media Concentration and Ownership around the World. Oxford: OUP. Ofcom “Measuring Media Plurality” Ofcom’s advice to the Secretary of State for Culture, Olympics, Media and Sport, 19 June 2012, [3.8] – [3.10]. 168 more difficult to influence news media, and the better functioning of an effective democratic process.457 Evaluating such flow-on impacts is not straightforward. (a) Media plurality in changing markets 897. Assessing media plurality impacts is made more difficult by the degree of disruption currently occurring in the media sector. The ongoing shift in focus from traditional media platforms towards increased publication and consumption of news and information content online is continuing apace. This pattern applies to a greater or lesser degree across all content formats, ie, text, video, and audio. 898. The ongoing evolution of digital news media means that the structure and competitive dynamics in this sector are continuing to change and develop. However, there is no indication that either NZME or Fairfax would not exist without the merger. (b) Entry and expansion into online markets 899. The Applicants suggested that there are low barriers to entry to publishing content online and so new entry and/or expansion by existing organisations, including existing organisations that are not necessarily well-established, is relatively straightforward. This could mean that the ongoing shift to online consumption of news content could facilitate increased plurality. This may be true in relation to any public service broadcasters unconcerned with advertising revenue, like RNZ, as they are strongly incentivised to expand their audience through existing social media and third party aggregators. 900. However, as set out in Section 5, this view was not shared by the other main media firms in New Zealand. These firms suggested that there were significant barriers to entry in the online news market, given both the costs of running a national news organisation and of the creation of content. 901. Further, incumbency and the possession of an established news brand may be a major advantage in attracting and retaining audiences online. To the extent that users may favour using a small number of apps on mobile devices, particularly for news content, organisations whose apps have already achieved a high rate of market penetration may be in an advantageous position. Additionally, the rise of social media and third party aggregators as a major distribution channel, and the difficulty this presents in deriving advertising revenue from online content (as outlined by the Applicants), may make profitable new entry and expansion by commercially-focused organisations more difficult, not less. 902. These factors may mean that a continued shift by users from consuming news via traditional media platforms to online may further entrench or even enhance the strong position currently held by the Applicants’ brands. Furthermore, the relatively small size of the New Zealand market means that new entry from major international 457 Ofcom “Measuring Media Plurality” Ofcom’s advice to the Secretary of State for Culture, Olympics, Media and Sport, 19 June 2012, [3.14]. 169 news brands is less likely than elsewhere. This would exacerbate any plurality concerns. 903. The Commission recognises these competing forces. However, consistent with the Commission’s view that larger scale entry and expansion into online reader markets is unlikely post-merger, in the Commission’s view, it would be speculative to predict that the highly concentrated pattern of consumption of domestic news online will become less concentrated in the foreseeable future. 904. Consequently, we are of the view that there is a real chance that this merger would present a significant risk to media plurality. Given the importance of the news media, it is our view that any adverse effects from this have the potential to be substantial. Applicants’ views 905. The Applicants’ view was that issues of media plurality are not relevant to our analysis and fall outside of the scope of the merger authorisation process.458 It is their view that any media plurality or ‘fourth estate’ concerns are a public policy issue for the Government to address, which it currently does via organisations such as RNZ, TVNZ, Maori TV, NZ on Air, Parliament TV, Te Mangai Paho (including Maori radio), etc. 906. The Applicants noted that any reduction in journalists would be a small proportion of the claimed cost-saving efficiencies from the merger.459 Accordingly, such a cost reduction would be a pro-competitive productivity enhancing public benefit that would enable them to invest more in increased variety and quality. 907. The Applicants stated that there are strong commercial reasons why the merged entity would be incentivised to maintain not only the quality and quantity of content, but also the diversity of views within its publications.460 Because of the two-sided nature of the merged entity’s media products, the Applicants suggested that the merged entity would have a strong incentive to maintain and publish a range of views and opinions so as to attract a wide range of consumers. 908. The Applicants pointed to the fact that, notwithstanding the recent move by both NZME and Fairfax to integrated, digital-first newsrooms, there is currently editorial independence throughout their various organisations to the extent that there are different editors across the Applicants’ respective regional and community newspapers.461 Similarly, NZME reporters operate an Editorial Code of Ethics while Fairfax reporters are subject to its Media New Zealand Journalism Charter.462 909. The Applicants also pointed to the availability of an increasingly diverse range of options used by New Zealanders to access news and information and that there is 458 459 460 461 462 Fairfax/NZME, Response to submissions, 29 July 2016, pages 10, 28 – 33; Fairfax/NZME Cross submission 10 October 2016, pages [21] to [31]. Fairfax/NZME, Response to submissions, 29 July 2016, at [11 (c)]. Fairfax/NZME, Response to submissions, 29 July 2016, at pages [28] to [30]. Fairfax/NZME, Response to submissions, 29 July 2016, at [90] to [94]. Fairfax/NZME, Response to submissions, 29 July 2016, at [95] to [101]. 170 significant plurality of views available to New Zealanders.463 They further point to competition from TVNZ, Newshub, RNZ, and a range of other New Zealand print and digital news/information competitors (such as Allied Press, NBR, Bauer's The Listener, etc) that would continue to place significant competitive constraint on the merged entity in the provision of New Zealand news.464 Others’ views 910. The Commission received a number of submissions from interested parties regarding the impacts on media plurality of this merger. Dr Julienne Molineaux et al. 911. Dr Julienne Molineaux and five other academics (collectively referred to as the Academics) co-authored a submission opposing the proposed merger.465 All of the authors are academics in New Zealand universities who have conducted research on the New Zealand media industry and the profession and performance of journalism.466 912. The Academics submitted that the proposed merger would reduce competition resulting in detrimental consequences on product quality for news consumers and the New Zealand public at large. In their view, the proposed merger would reduce the plurality and diversity of news coverage to the detriment of New Zealanders. 913. In the Academics’ view, the public benefit of journalism is that it gives expression to different perspectives and voices, referred to as the “marketplace of ideas”. 914. The Academics considered that the public benefits derived from journalism are more likely to accrue in “diversified and open markets”, and are considerably less likely in more concentrated markets. They note that plurality is, in part, a function of the number of journalists employed to produce stories. In the Academics’ view, any reduction the number of journalists would reduce the quantity and quality of stories produced. 915. The Academics disagreed with the Applicants’ submission that the duplication of content between different media agencies is ‘wastage’. In the Academics’ view, democracy functions best with many voices and perspectives, and they considered that seemingly “‘objective’” reportage can be marked by all kinds of “subtle, and sometimes not so subtle, differences in their representation of an event or issue”.467 916. Post-merger, the Academics considered that the concentration of voices would risk reducing the plurality and diversity of New Zealand news coverage, to the detriment 463 464 465 466 467 Fairfax/NZME, Response to submissions, 29 July 2016, at [89]. Submission from Fairfax/NZME to the Commerce Commission (29 July 29) at [18]. Dr Julienne Molineaux et al., submission to the Commerce Commission (July 2016) Dr Julienne Molineaux (Auckland University of Technology), Associate Professor Donald Matheson (University of Canterbury), Dr. Merja Myllylahti (Auckland University of Technology), Dr. Sean Phelan (Massey University Wellington), Dr. Peter Thompson (Victoria University Wellington) and Associate Professor Geoff Lealand (University of Waikato). Submission from Dr Julienne Molineaux to the Commerce Commission (July 2016) at [page 16]. 171 of New Zealanders. Such levels of concentration increase the risk that media power will be abused, and reduces the likelihood of journalism holding the powerful to account. 917. The Academics submitted that the New Zealand media industry is already concentrated with weak levels of plurality on a global scale. The Academics note that if the merger were to proceed, the merged entity would have an 89.3% share of newspapers in New Zealand.468 As set out in Table 16, the Academics noted that this would put New Zealand second in the world behind China (state-owned) in terms of concentration of newspaper ownership. Table 16: Daily newspaper ownership – Market share of top company, 2016469 Country Owner Percentage China Government of China 100% New Zealand Merged entity 89.3% Egypt Government of Egypt Australia News Corporation and associated Murdoch companies 57.5% Chile El Mercurio SAP 54.9% Ireland Independent News and Media Turkey Dogan Group 46.4% Switzerland Tamedia 44.3% Portugal Cofina 42.4% Sweden Bonnier 41.3% Russia Komsomolskaya Pravda 39.6% South Africa Naspers Group UK Murdoch Group 32.5% Brazil Globo Group 28.4% 72% 52% 35% Source: Dr Julienne Molineaux et al., submission to the Commerce Commission (July 2016) New Zealand Political Studies Association 918. The New Zealand Political Studies Association (NZPSA) represents over 200 members. Its aims include fostering the study and research of political phenomena, and encouraging the dissemination of such information. NZPSA state that its members have published extensively on the media and on media competition in New Zealand.470 919. NZPSA explained that political scientists consider the media important because it “informs people about events, facilitates political learning, ensures surveillance of 468 469 470 Based on circulation. For overseas market shares Academics sourced Noam (2016). Who Owns the World’s media?: Media Concentration and Ownership around the World. Oxford: OUP. New Zealand Political Studies Association submission to the Commerce Commission (30 June 2016). 172 politics and economics, promotes public discussion and civic engagement, and helps create a sense of community.”471 920. NZPSA submitted that the proposed merger would “reduce competition and be damaging to the democratic process and the public interest”. It is of the view that a “well-functioning information market” should be considered a public good. As such, the “requisite threshold” should apply to mergers that threaten to reduce the breadth and depth of this public good. 921. NZPSA was of the view that the proposed merger would have negative political consequences and erode democracy as there would be fewer points of view represented, less discussion of politics and less surveillance of politics and business. Coalition for Better Broadcasting 922. The Coalition for Better Broadcasting (CBB) is an independent charitable trust which states that its primary interest is to “inform civic debate and policy formation in regard to public service broadcasting, although media convergence extends the scope of [its] concern to other platforms and content providers, particularly those which support public interest journalism.” 923. CBB considered that the role of news media to inform the public and hold powerful institutions to account is a “fundamental tenet of contemporary democratic society”.472 It considers that this is particularly important in New Zealand, where the concentrated and deregulated nature of our media sector has led to a “stronger dependence on the dominant companies not only to maintain competition and consumer choice but to act as a check on each other’s editorial integrity and journalistic quality.” 924. To allow a functioning democratic debate, CBB considered that a plurality of editorial views is “obviously important”. In its view, the proposed merger would result in the merged entity having an “unacceptable concentration of editorial power” which would threaten the effectiveness of the media to inform the public and hold powerful institutions to account. 925. While CBB acknowledged that media convergence has led to commercial challenges for the news media, in its view the merger would require the public to “shoulder the risk of detrimental market and democratic outcomes in order to allow two already major news media companies to consolidate their market position”. Additional views 926. 471 472 In addition to the views of the Academics, NZPSA and CBB, a number of other interested parties have expressed concern regarding a reduction in media plurality/diversity resulting from the proposed merger. These interested parties include: NZPSA submission – Kemp, 2010, pp. 385-397 Coalition for Better Broadcasters, submission to the Commerce Commission (July 2016). 173 926.1 journalists; 926.2 independent publishers; 926.3 advertising agencies; 926.4 advertisers; and 926.5 members of the public. The Commission’s view 927. As noted at above, in the Commission’s view, the effects of the proposed merger on media plurality are a relevant to our assessment of whether the merger is likely to result in such a benefit to the public that is should be permitted. 928. The Commission considers that the merging parties would control a significant portion of New Zealand news production across media, including print, online and radio. The Commission considers that a reduction in media plurality could have a significant adverse impact on society. Our view is informed by the experts we have engaged, the submissions we have received, interested parties with which we have spoken, the academic literature we have reviewed, and the regulatory approach applied to this issue in much, if not all, of the rest of the OECD. 929. Journalists employed by major media organisations play a key role in uncovering and reporting on information that may otherwise remain unknown by the wider population. Different media owners may have different editorial priorities or preferences, whether based on personal, commercial or political objectives. Therefore, more diversified media ownership may help ensure that a wider range of issues and topics are explored and addressed, and that specific topics or issues of public interest do not go unreported. This is particularly important given the media’s ability to shape public discourse and influence the political agenda. 930. The Commission considers that plurality considerations are particularly important in New Zealand given that, in contrast with a number of other jurisdictions, domestic news media organisations with a specific public service broadcasting objective play a lesser role than commercially-focused organisations. Unlike other jurisdictions, media ownership in New Zealand is also unregulated. 931. In particular, TVNZ, since 2001, has not had a public service objective.473 In 2011, the New Zealand Government abolished the TVNZ Charter in favour an annual Statement of Performance Expectations (SPE), which is more commercial in its focus.474 The SPE states that TVNZ must provide high quality content that: 473 Up until 2011, TVNZ had, among other things, an obligation to “provide independent, comprehensive, impartial, and in-depth coverage and analysis of news and current affairs in New Zealand and throughout the world and of the activities of public and private institutions” (The TVNZ Charter, http://images.tvnz.co.nz/tvnz/pdf/tvnz_charter_01.pdf accessed 1 November 2016). 474 http://images.tvnz.co.nz/tvnz_images/about_tvnz/Board%20Approved%20Statement%20of%20Performanc e%20Expectations%2031st%20May.pdf accessed 1 November 2016. 174 931.1 is relevant to, and enjoyed and valued by, New Zealand audiences; and 931.2 encompasses both New Zealand and international content and reflects Māori perspectives. 932. Although the SPE sets out the scope of TVNZ’s functions and intended operations, it does not refer to the provision of news or the requirement for independence and impartiality. Although, we note that TVNZ is still subject to broadcasting standards as set by the Broadcasting Standards Authority which include “balance”: “broadcasters should make reasonable efforts, or give reasonable opportunities, to present significant points of view either in the same programme or in other programmes within the period of current interest.”475 933. By contrast, Radio New Zealand still has a public service charter, the Radio New Zealand Charter (the RNZ Charter).476 One of the purposes of the RNZ Charter is to provide reliable, independent, and freely accessible news and information. In addition, RNZ is obligated to provide comprehensive, independent, accurate, impartial, and balanced regional, national, and international news and current affairs. However, we note that RNZ’s covers less news than the Applicants and its reach (radio only) is only around 13% of the population .477 934. The Commission does not accept the submission that the absence of specific media ownership regulations means that concerns about plurality do not apply to New Zealand. On the contrary, we are required to consider the public benefit of this proposed merger in a setting where media ownership is unregulated and there is limited media content regulation. 935. Existing policy settings regarding media plurality have been established in a media environment that could change substantially if the merger were to proceed. While it is possible that these policy settings could be altered post-merger, it is not possible for the Commission to predict whether and when this likely to occur, and how any alteration might impact a consummated merger. 936. Consequently, the Commission has assumed existing policy settings as the basis from which to assess both the with and without the merger scenarios. These settings include not only the absence of any specific plurality-related regulation, but also the current funding and operations of the various publicly owned or funded organisations, including RNZ, TVNZ, Maori TV, NZ on Air, Parliament TV, etc. Assessing potential plurality effects of the proposed merger 937. 475 476 477 To assess how the merger could affect media plurality, the Commission first evaluates the current state of media plurality. The Commission then considers how this level of plurality would be affected by the merger. Broadcasting Standards Authority, Broadcasting Standards in New Zealand Codebook (April 2016) page [39]. http://www.radionz.co.nz/about/charter accessed 1 November 2016. Glasshouse Consulting presentation to NZ on Air, Where are the audiences (2016) [slide 12]. 175 938. Assessing the potential plurality effects of the merger requires the Commission to consider not only changes to the status quo, but also how the merger could affect plurality within the New Zealand news media sector over the short term. This requires assessing both the likely changes to the media over this period as well as the likely editorial and operational decisions of the merged entity. Existing level of plurality 939. All else being equal, plurality is likely to be greater the more competing media organisations there are. Therefore, the starting point for any plurality analysis is to consider the number of media organisations under separate ownership. This is referred to external plurality, ie, plurality between organisations. 940. Evaluating the existing degree of media plurality requires assessing not just the number of different organisations that produce news and information content, but also assessing the consumption of this content by New Zealanders. Additionally, the Commission has sought to assess whether the impact of the content consumed varies across different organisations. 941. Along with external plurality, the Commission has also sought to assess the degree of internal plurality that is present within the sector, particularly with respect to the Applicants. Internal plurality refers to the range of perspectives and voices that can be presented within one media organisation. External plurality 942. There are few assessments of the state of media plurality in New Zealand. The JMAD New Zealand Media Ownership Report 2015 outlines that the overall ownership structure of the New Zealand media industry has been relatively stable since 2010; however the ownership of individual media firms has changed significantly within this time. 943. An international comparison of media plurality across OECD and EU countries has been undertaken as part of the Sustainable Governance Indicators project by the Bertelsmann Foundation.478 In a comparison across 41 countries New Zealand ranks last equal with Romania and Turkey on media pluralism, with a score of 4 out of 10. 944. In addition to these assessments, the Commission has sought to evaluate the existing degree of external plurality based on the framework established by Ofcom.479 This framework considers three categories of measures. These measures relate to 478 479 See: http://www.sginetwork.org/2016/Democracy/Quality_of_Democracy/Access_to_Information/Media_Pluralism. The Bertelsmann Foundation is a non-profit organisation based in Germany with programs designed to strengthen society (https://www.bertelsmann-stiftung.de/en/about-us/what-motivates-us/ourmission/). The Sustainable Governance Indicators project is a platform built on a cross-national survey of governance that identifies reform needs in 41 EU and OECD countries (http://www.sginetwork.org/2016/). Ofcom “Measuring media Plurality” Ofcom’s advice to the Secretary of State for Culture, Olympics, Media and Sport, 19 June 2012; Ofcom “Report on public interest test on the proposed acquisition of British Sky Broadcasting Group plc by News Corporation”, 31 December 2010. 176 production (availability) of content from different media organisations and the patterns of consumption of that content. Production/availability measures 945. As outlined in the discussion of the online reader market, producers of domestic news content can be broadly characterised as large, New Zealand-based mainstream news producers – Fairfax, NZME, TVNZ, MediaWorks and RNZ – and a larger number of smaller producers such as NBR, BusinessDesk and Allied Press. These producers are also vertically integrated in downstream distribution markets via their own media platforms, ie, websites, print publications, TV broadcasts, and radio stations. 946. The Commission has obtained data from the main media organisations regarding the number of editorial and journalistic staff employed and the total number of articles produced (see Table 7). Each of Fairfax and NZME produce more than three times the number of stories than either of TVNZ or MediaWorks, and collectively employ [ ] editorial and journalistic staff as compared to 510 similar staff members employed by TVNZ, MediaWorks and RNZ combined.480 947. Also relevant to an assessment of external plurality is the existence of funding for public interest content. NZ on Air is the government media funding agency that funds a range of domestic content to add diversity and choice for New Zealand audiences on radio, TV, and online.481 It has recently announced it intends to change its funding approach to remain flexible in light of what it considers to be “a fastchanging, converging media environment.”482 948. The Commission acknowledges that NZ on Air funding is likely to assist in widening the range of domestic news and information content that is available to New Zealanders. This is because, as outlined in its most recent proposed funding strategy, some of its funding will be directed to factual content, including documentaries, specialist journalism, and informational content.483 This includes funding RNZ. 949. However, the Commission does not consider that NZ on Air can be relied on to ensure that any reduction in media plurality arising from the merger would be offset. This is because of the relatively ad hoc and specialised nature of the content that is funded, and there is no indication that this would change or that its funding would 480 481 482 483 RNZ has a newsroom of about 100 staff (Radio NZ interview with the Commission, 12 July 2016). TVNZ has staff of about 170, described by TVNZ as “a crude make-up of producers and reporters” (TVNZ information response to the Commission 5 October 2016). MediaWorks said that the Newshub newsroom is staffed by 240 people, including reporters, producers, editors, camera operators, directors, graphics experts and others (MediaWorks response to Commerce Commission information request, received 30 September 2016). See: http://www.nzonair.govt.nz/about-nz-on-air/how-we-invest/ See: http://www.nzonair.govt.nz/news/articles/nz-on-air-responds-to-changing-audiences-with-change/ See: http://www.nzonair.govt.nz/publications/nz-on-air-funding-strategy-draft-for-feedback/#anchor-ata-glance---what's-different?; http://www.nzonair.govt.nz/news/articles/widely-diverse-topics-coveredin-new-and-returning-documentaries/ 177 increase.484 For instance, funding may be provided in relation to content regarding a specific issue that may be presented in, say, a one-off documentary. Consumption and reach measures 950. In order to help assess the extent of plurality in New Zealand news services, the Commission has first relied on data that relates to the consumption and reach of the various organisations that produce New Zealand news content across the online, TV, radio and print platforms to better understand their relative levels of usage. 951. In terms of consumption, the number of New Zealanders using various news sources is contained in Table 17. Newspaper consumption is indicated on the basis of the daily circulation figures of daily newspapers. Online consumption is based on monthly page views, and TV and radio consumption are, respectively, on the basis of number of monthly viewers and number of weekly listeners. 484 There is no apparent increase in funding in the pipeline for NZ on Air. NZ on Air has discretion over how it distributes its funding. 178 Table 17: Consumption of news by media type485 daily 486 circulation (1 July 2015 - 30 June 2016) Online NZME 188,707 Fairfax Fairfax 204,676 38,245 Print - daily newspapers Allied Press Independent TV viewers monthly (Q2 2015 - Q1 2016) Radio 177,150,489 TVNZ 1,457,000 RNZ 529,000 NZME 108,001,295 MediaWorks 679,000 NZME 482,000 TVNZ 18,142,624 Sky Network 387,000 MediaWorks 220,000 page views monthly (Sep 487 2016) MediaWorks 5,807,025 Ashburton Guardian 4,345 RNZ 4,842,514 Wairarapa Times-Age 5,766 SunMedia 3,616,210 Westport 489 News 1,600 Allied Press 3,022,551 Bauer 485 486 487 488 489 488 listeners weekly (2015 2016) 153,489 Sky Network N/A Scoop Media N/A The Spinoff N/A NBR N/A Print: The New Zealand Audit Bureau of Circulations, Daily Press circulation, Current Audit, ‘ANC to 30/06/2016’.http://newspaper.abc.org.nz/audit.html?mode=embargo&npa_admin=1&org=npa&show=li stpubs&letter Online: Nielson Online Ranking Report (market: New Zealand Domestic Traffic By Brand (website data), period: monthly, 01/09/16 - 30/09/16). TV: Free to Air TV and PayTV measures in Nielsen CMI Fused Q2 15 - Q1 16 June 16 TV/Online – FFADNT The CMI TV measure is described by Nielsen as “reach of programming in a month” with a population base 10+ years (email to the Commerce Commission from Nielsen (12 October 2016). Radio: Source for RNZ is http://www.radionz.co.nz/about/audience-research. Radio NZ reports weekly cumulative audience for RNZ National as 13.2% of the 10+ population, nationwide, for the period January 31 to July 2 2016 (equivalently, a cumulative audience of 529,000). Source for NZME and MediaWorks is the same as the TV source. An individual is counted as a listener in this data if they listened to the radio station at least once that week. Because this is daily circulation data, certain print sources of New Zealand news are excluded (NBR, Whakatane Beacon, and magazines including Bauer titles (The Listener, Metro, North & South). We exclude Sunday papers and the Weekend Herald. Titles included are: Allied Press: Greymouth Star, Otago Daily Times; Fairfax: Manawatu Standard, Taranaki Daily News, The Dominion Post, The Marlborough Express, The Nelson Mail, The Press, The Southland Times, The Timaru Herald, Waikato Times; NZME: Bay of Plenty Times, Hawke's Bay Today, New Zealand Herald, Rotorua Daily Post, The Northern Advocate, Wanganui Chronicle; Independent: Ashburton Guardian, Wairarapa Times-Age, Westport News. Online data comprises: Fairfax: Stuff.co.nz; NZME: newstalkzb.co.nz, nzherald.co.nz; TVNZ: tvnz.co.nz; MediaWorks: newshub.co.nz, tv3.co.nz; RNZ: radionz.co.nz; SunMedia: sunlive.co.nz; Allied Press: odt.co.nz, Bauer: listener.co.nz, metromag.co.nz. The Commerce Commission seeks page view data for the following online sources of New Zealand news: Sky Network, Scoop Media, The Spinoff, NBR. TV data comprises One News 6pm weekly coverage, Newshub 6pm weekly coverage, and Sky News NZ. E-mail from Westport News to the Commerce Commission (27 July 2016). 179 952. While it is clear that the merging parties dominate print and online, the range of the consumption metrics across the different types of media makes it difficult to ascertain the relative size and importance of the various news sources across platforms. The Commission, therefore, also turns to reach measures – the total number of different people exposed, at least once, to a particular platform during a given period – as a means of comparison. 953. At an aggregate level, a survey carried out by Fairfax indicated that online news now dominates as the primary news source (see Figure 10).490 Online’s share of news media consumption in a typical week in 2016 was [ ]%, versus the next highest source, TV at [ ]%. “Traditional” newspapers are at [ ]% and radio is at [ ]%.491 [ ] 954. In terms of disaggregated reach492 across the different news organisations’ products, stuff.co.nz leads, reaching 47% of the population on a monthly basis. This site is followed by nzherald.co.nz with 39% and by TVNZ’s One News 6pm with 37% on a monthly.493 The collective reach of the each of these organisations is greater were reach to be measured across all of their products in aggregate.494 For example, the reach of just one of NZME’s 31 papers, The New Zealand Herald, is 30%, and the Saturday version of that paper is 8%. Similarly, of Fairfax’s 75 papers, The Dominion Post has the greatest reach at 13% followed by The Press at 11%. TVNZ similarly has reach through its other TV news products and its website, which has a reach of 20%.495 955. An additional consideration is that website and TV reach is measured on a monthly basis, while that of newspapers is weekly. This tends to overstate the reach of websites and TV relative to newspapers, since it is more likely that a person will access a product over the course of a month than over the course of a week. However, since website and TV are measured on the same basis, the comparison of reach across those two media types should be like-for-like. This is similarly the case for comparisons across newspapers. However, comparing third parties’ newspaper reach to that for the Applicants websites tends to understate the relative reach of 490 Fairfax Annex 67, [ 491 What is contained in “traditional’ newspapers is not noted in the source material. Reach is measured for New Zealanders over the age of 15 for all print product and over the age of 10 for all other media (E-mail from Nielsen to the Commerce Commission (10 October 2016)). This data does not include reach of content via the various media organisation apps; however, as indicated in Section 5, the pattern of consumption of New Zealand news via apps is similar to the consumption of New Zealand news via browsers. Individual reach figures cannot be aggregated since it would entail double-counting of those people who reach an organisation’s brands in multiple ways. We only have reach information for TVNZ and MediaWorks’ main TV news products, One News 6pm and Newshub 6 pm, respectively. 492 493 494 495 ]. [ ]. 180 those third party products. For this reason, the Commission also separately examines reach for online products and newspapers in isolation below. 956. The Commission also separately considers the reach of radio. There are two separate concerns in regard to radio. The first issue is essentially the same as in the case of newspapers. That is, reach is understated relative to websites and TV because reach is measured on the basis of listenership over the last seven days. The second issue is that the data only capture the reach of RNZ’s website rather than the radio station itself. This is, too, is considered further below. 957. While the reach figures are particular to products rather than an organisation and print and radio reach risk being relatively understated, it is nonetheless clear that the merging parties’ products dominate, not only in terms of the relative popularity of their websites,496 but also because of the number of their products. Fairfax has a total of 78 news products and NZME has 35.497 For example, the monthly reach of Fairfax’s 63 community papers is 60%.498 958. No other organisation, with the exception of TVNZ, has a product with similar reach, and no other organisation has as a comparable number of products. TVNZ, which has a product with the third most reach, only has nine news products.499 MediaWorks has eight news products, and its main TV news product, Newshub 6 pm, has a reach of 17%, while its website has a reach of 13%.500 Allied Press, the next largest print publication, has three products. The largest of these, The Otago Daily Times, has a reach of 6%, followed by the odt.co.nz at 4%, and The Greymouth Star has a reach of 1%. 496 497 498 499 500 As noted, recognising that the reach of newspapers and radio stations are likely understated relative to websites and TV. For Fairfax this consists of nine daily newspapers, two paid weekly newspapers, two websites, two magazines and 63 community newspapers. For NZME this consists of six daily newspapers, two paid weekly papers, three websites, one radio station, and 23 community papers. We do not have similar information available for NZME’s community papers. As noted we only have the reach figure for TVNZ’s One News 6pm. The other news products are Breakfast, One News Midday, Q+A, Seven Sharp, Tagata Pasifica, Te Karere, and Tonight. The other news products are Newshub Late, Prime, Story, The Nation, and Paul Henry. 181 Figure 12: Total reach by media product, 2016501 % population 10 years+ (newspaper 15 years+) All Fairfax Monthly Communities… 60% The Dominion Post 13% The Press 11% Sunday Star Times AIR 8% Waikato Times newspaper (metro, regional, Sunday) Fairfax 5% Dominion Post Sat AIR 4% Sunday News AIR 3% The Southland Times Manawatu Standard Taranaki Daily News 2% The Timaru Herald 2% The Nelson Mail The Marlborough Express web brand 3% 2% 2% 1% 47% Stull.co.nz The NZ Herald… 30% NZ Herald Sat AIR 11% Herald On Sunday AIR Newspaper (metro, regional, Sunday) web brand radio 3% Bay Of Plenty Times The Northern Advocate NZME 8% Hawkes Bay Today 3% 2% Wanganui Chronicle 1% Rotorua Daily Post 1% 39% nzherald.co.nz NewstalkZB website Newstalk ZB 5% 12% One News 6pm weekly coverage… *Te Karere TVNZ free to air TV N/A *One News Midday N/A *Tonight N/A *Q + A N/A *Breakfast *Seven Sharp web brand web brand MediaWorks free to air TV 37% N/A *Tagata Pasifika N/A N/A TVNZ 20% Newshub website… 13% TV 3 Newshub 6pm weekly coverage 7% 17% *Story N/A *The Nation N/A *Paul Henry N/A *Newshub Late N/A *Prime News N/A radio radio Radio Live RNZ National… RNZ web brand Sky Network pay TV web brand Radio New Zealand Sky News NZ… Sky TV NZ Bauer magazine Otago Daily Times… Allied Press newspaper (metro, regional, Sunday) Scoop Media web brand web brand Odt.co.nz Scoop… 6% 13% 9% 10% 6% North & South… 6% NZ Listener 5% Metro Independent Spinoff Sun Media NBR 959. 501 502 newspaper (metro, regional, Sunday) web brand web brand newspaper (metro, regional, Sunday) Greymouth Star 3% 6% 1% 4% 4% The Gisborne Herald… 1% Ashburton Guardian 1% Wairarapa Times-Age 1% Westport News N/A The Spinoff… 2% SunLive… 2% The National Business Review AIR… 1% The Commission has also obtained data that indicates the degree of multi-sourcing by users.502 Multi-sourcing demonstrates the extent to which New Zealanders Nielsen CMI Fused Q2 15 - Q1 16 June 16 TV/Online – FFADNT. The Commission is still seeking reach statistics for these TVNZ and MediaWorks free to air TV news products and for Westport News. Newspaper figures are against a readership of people aged 15+, all other media is 10+. Newspapers – weekly reach; magazines - reach is an issue period; radio – stations listened to last 7 days, online – website visitation in a month; TV reach of news programming in a month The RNZ radio reach data is from Glasshouse Consulting presentation to NZ on Air, Where are the audiences (2016). Report by AC Nielsen, [1 November 2016] commissioned by the Commerce Commission. 182 benefit from the diversity of voice currently available to them in that it indicates what other sources of news they turn to given that they, for example, already obtain news from one of the merging parties. This data is in Table 18 below.503 Table 18: Multi-sourcing by major news organisation, 2016504 Percentage of Fairfax users who use other news products FAIRFAX NEWS Percentage of NZME news users who use other news products 86.4% NZME NEWS 74.1% TVNZ NEWS 83.2% 83.6% 82% 82.6% MEDIAWORKS NEWS Percentage of TVNZ news users who use other news products Percentage of MediaWorks news users who use other news products 79% 78.8% 68.1% 68.1% 90.6% 89.6% 960. These figures indicate that Fairfax and NZME users regularly turn to each other’s products, as well as the products of TVNZ and MediaWorks.505 In particular, 74.1% of Fairfax readers also use NZME products, and 86.4% of NZME users also use Fairfax products. While TVNZ and MediaWorks would remain highly used alternatives, the benefits of the dual usage of Fairfax and NZME would be lost. 961. As in the case of the reach data, the data are likely to overstate the degree of multisourcing to the extent that some New Zealanders use one or two primary sources for much of their news but occasionally view content from rival news brands. For example, a Fairfax user that uses Fairfax products on a daily or weekly basis and only use NZME products once a month, would still be captured as a multi-user.506 962. These figures are based on varying time periods over which reach is measured and consequently risks overstating the reach of those media that are measured monthly (websites and TV) versus those that are measured weekly (print and radio) and the risk of overstating the degree of multi-sourcing, the Commission also separately examines the reach of the various news organisations within online and within newspapers. We also separately consider the potential understatement of the importance of radio, RNZ in particular. 963. Figure 12 indicates that the average combined monthly reach of the merging parties’ web brands is 2.2 million New Zealanders.507 The reach of Newshub’s website at 0.5 million New Zealanders. TVNZ-News’ and RNZ’s websites have, respectively, 503 504 505 506 507 The table cells indicate the percentage of media users indicated in the column heading that also use the products of the organisation in the row heading. For example, the percentage of Fairfax users that also use TVNZ products is 83.2%. Nielsen CMI Q3 2015 - Q2 2016, August TV/Online Fused. Base: All People 10+. Newspapers – weekly coverage, Online – monthly fused Online Ratings, Radio – stations listened to last 7 days, TV – monthly fused TV Ratings RNZ is not included in these figures because Nielsen does not collect information regarding its radio ratings. Similar to the reach data, the multi-sourcing data does not include consumption of content via the various media organisations’ apps. An individual is counted as a viewer in this data if they visited the website at least once that month. 183 similarly low relative reach figures of 0.4 million and 0.3 million New Zealanders over the age of ten. NZ population 10 years+ (millions) Figure 13: Average monthly reach of New Zealand web brands, Q2 2015 – Q1 2016 (browsers only)508 2.2 1.9 1.6 0.5 0.4 0.3 0.2 0.2 0.2 Source: Nielsen CMI Fused Q2 15 - Q1 16 June 16 TV/Online -FFADNT 964. 508 The merging parties similarly dominate in newspapers. Figure 13 contains the readership figures for all major New Zealand print news publishers. In addition to the merging parties, this includes Bauer (a magazine publisher), Allied Press, NBR, and three independents publishers. The collective readership of the merging parties’ publications is considerable larger than that of any other remaining print publishers individually or when that group is taken as a whole. The merged entity total is calculated as the stuff.co.nz total plus 0.2*the nzherald.co.nz total to account for 80% of nzherald.co.nz viewers also visiting stuff.co.nz. 184 Figure 14: Total weekly readership of New Zealand news print media, Q2 2015 – Q1 2016509 NZ population 15 years+ (millions) 2.38 All Fairfax Monthly Communities… 0.53 The Dominion Post 0.43 The Press 0.31 Sunday Star Times AIR Fairfax newspaper (metro, regional, Sunday) Waikato Times 0.19 Dominion Post Sat AIR 0.17 Sunday News AIR 0.12 The Southland Times 0.11 Manawatu Standard 0.09 Taranaki Daily News 0.08 The Timaru Herald 0.07 The Nelson Mail 0.07 The Marlborough Express 0.03 1.19 The NZ Herald… 0.45 NZ Herald Sat AIR 0.32 Herald On Sunday AIR NZME newspaper (metro, regional, Sunday) Hawkes Bay Today Bay Of Plenty Times The Northern Advocate 0.11 0.11 0.08 Wanganui Chronicle 0.06 Rotorua Daily Post 0.05 0.23 North & South… Bauer magazine Metro Allied Press NBR Independent newspaper (metro, regional, Sunday) Greymouth Star 509 510 511 0.24 0.02 0.04 The Gisborne Herald… 0.04 Ashburton Guardian 0.03 Wairarapa Times-Age Westport News 965. 0.13 Otago Daily Times… newspaper (metro, regional, Sunday) The National Business Review AIR… newspaper (metro, regional, Sunday) 0.20 NZ Listener 0.03 N/A In regard to radio, because of the way reach data is gathered, there is the potential that the relative importance of radio is understated and, more importantly, Figure 13 does not contain the reach of RNZ’s radio station. Another source of information, however, indicated that RNZ’s radio reach in 2016 was 13%.510 The monthly reach RNZ may well be higher; however, the Commission has no reason to consider that it would be significantly so. This is similarly the case for MediaWorks’ Radio Live product, which has a weekly reach of 6%. Overall, while the total reach of all radio is 59%, listening is down 12% since 2014, and it is less popular among 15-39 year olds.511 Nielsen CMI Fused Q2 15 - Q1 16 June 16 TV/Online –FFADNT. Magazine readership figures are for a base of people aged 10+. Readership is defined as "readers of any copy of a publication anywhere for two minutes or more, within the publication's issue period". (Source: Email from Nielsen to the Commerce Commission (10 October 2016)). The Commission does not have readership data on NZME’s community papers and Westport News. Glasshouse Consulting presentation to NZ on Air, Where are the audiences (2016) [slide 12]. Glasshouse Consulting presentation to NZ on Air, Where are the audiences (2016) [slides 15 and 27]. 185 966. In general, the Commission considers the merging parties to be the dominant providers of news services, particularly in print and online. The Commission recognises the importance of linear TV news in terms of reach but, even taking that into consideration, the reach figures available to the Commission suggests that the diversity of media voice would be significantly impacted post-merger. This is particularly the case since TV news typically does not have the depth and breadth of news coverage as text. This is considered further in the below discussion of impact measures. Impact measures 967. In addition to measures of availability and consumption of news content, the Commission considers that it is useful to consider the extent to which content from different sources may have a different impact. This is relevant because content from different sources and sourced from different platforms may have a greater or lesser effect on New Zealanders. 968. For instance, listening to a short radio news bulletin may have less impact than 30 minutes spent reading an entire newspaper. Also, some sources may be considered more credible than others. 969. The Commission is not aware of any research that seeks to measure and compare the impact of different New Zealand news media organisations. In the absence of any such research, the Commission has reviewed various research that to some degree considers the importance of different sources. 970. A study commissioned by News Works512 found that New Zealanders rate newspapers, a medium in which the merging parties dominate, as more reliable, credible, dependable and trustworthy. The medium that rates the next highest on these features is TV (see Figure 15). 512 New Works is a commercial entity of the Newspaper Publishers' Association. It advocates news media brands throughout New Zealand like Stuff, NZ Herald, The Dominion Post, The Press, and Otago Daily Times among others (http://www.newsworksnz.co.nz/about/ accessed 2 November 2016). 186 Figure 15: How New Zealanders describe different media, 2013513 Source: News Works 971. In regard to attributes related to content and timeliness, the same survey found that New Zealanders rate the internet most highly, followed by, in all categories but one, newspapers (see Figure 16). As discussed in Section 5, the merging parties are dominant in online sources of New Zealand news. Figure 16: New Zealanders' views of content and timeliness, 2013514 972. As discussed, the Commission does not consider that radio and TV have the depth and breadth of news coverage as text-based news. 973. The Commission has also considered the importance of the Applicants in setting the news agenda and breaking stories that are picked up by other mainstream media organisations. As discussed in Section 5, the Commission understands that it is common place for journalists at each of the main media organisations to monitor 513 514 New Works, Newspapers are but one expression of a modern news brand, 2013 [slide 16]. Ibid [slide 17]. 187 each other’s news output to ensure that they are not missing important news stories. 974. This, according a senior industry figure, is because stories published by these organisations will have been confirmed, fact-checked and cleared by these organisations’ legal teams.515 This senior industry figure outlined that the media organisations that provide the greatest number of stories that are picked up are the Applicants. This is because of the greater output in part stemming from the larger network of journalists the Applicants have in comparison to the three broadcasters, TVNZ, MediaWorks and RNZ. 975. In contrast, the Commission does not consider that the smaller and niche domestic media organisations play as strong a role in the production and distribution of New Zealand news content. These organisations have far fewer resources, have much less reach amongst the public and do not appear to be monitored to the same extent by the major organisations. 976. Many of these smaller players, especially blogs, often provide opinion on news content produced by one of the main news organisations rather than original news content. Although there may be occasional isolated instances where smaller players produce news content that has a large impact, this cannot be relied upon to replace the loss in plurality consequent to the merger. 977. In contrast, the merging parties have substantially greater newsgathering resources, well-known national and regional brands, and operate across print, online and radio. Internal plurality 978. Another potential source of plurality is internal plurality; that is, the diversity of perspective from within a given organisation.516 Internal plurality depends on how far an organisation and/or regulatory factors enable, support or promote a range of internal voices and opinions.517 979. The Applicants submitted that they have internal charters in place that commit journalists and editors to certain standards. For example, the Fairfax charter, among other things, requires editors and journalists to maintain independence from political or commercial influences, maintain high standards of accuracy and ethical behaviour, and stand up for the communities they serve.518 The NZME Code of Ethics contains a principal responsibility to "maintain the highest ethical standard in our journalism while balancing the right of the individual with the public's right to know".519 515 516 517 518 519 Commerce Commission interview with MediaWorks (13 October 2016) Broadcasting Authority of Ireland, Report on Ownership and Control of Media Businesses in Ireland 20122014. Ofcom, Measurement framework for media plurality (5 November 2015) at [3.17], and Robin Foster, Measuring media plurality: a checklist (15 September 2016) at page [3]. Submission from Fairfax/NZME to the Commerce Commission (29 July 2016) at [97]. Submission from Fairfax/NZME to the Commerce Commission (29 July 2016) at [98]. 188 980. The Applicants further submitted that “[t]he merged entity will have the same economic incentives as each of Fairfax and NZME do today to commit to these ethical standards of balance, accuracy, independence, integrity, and adherence to Press Council decisions.”520 981. The Commission notes that plurality does not rest on a code of ethics since such a code, no matter how laudable, cannot assure diversity of voice. The Commission also notes that neither Fairfax nor NZME’s codes of ethics speak to promoting a range of internal voices and perspectives, and even if the merging parties were to make such commitments, they would be voluntary and so unenforceable. Ultimately, the Applicants can and do control what is contained (and not contained) within their publications. Regulatory considerations 982. The Applicants are currently subject to oversight by the New Zealand Press Council and state that this would continue post-merger. The Press Council is a voluntary, self-regulatory industry organisation that adjudicates on complaints about its members, who are print publishers and who may also operate websites. It is funded by the industry with the bulk of its funding sourced from the Applicants.521 983. The Applicants have pointed to the independent structure and operation of the Press Council and highlighted that it has ruled against both Fairfax and NZME, and that these rulings have been accepted.522 984. The Commission has considered the role of the Press Council and what influence it has on the activities and editorial decisions of its members. 985. The Press Council’s main relevant principle is that “[p]ublications should be bound at all times by accuracy, fairness and balance, and should not deliberately mislead or misinform readers by commission or omission.” Other principles relate to such items as privacy, confidentiality and treatment of children and young people.523 986. The Press Council cannot financially sanction breaches of its principles. While the Press Council would ensure that any complaints would be addressed, by way of requiring the publishing of retractions and corrections, it plays no role in editorial decisions and so cannot ensure that a range of views and perspectives are published or that topics that are potentially in the public interest are covered. 987. Even if the Press Council did play a role in plurality, because of the voluntary nature of the Press Council, there is no guarantee that the merged entity would continue to submit to it post-merger. 520 521 522 523 Submission from Fairfax/NZME to the Commerce Commission (29 July 2016) at [100]. Commerce Commission interview with Press Council (1 August 2016). Submission from Fairfax/NZME to the Commerce Commission (29 July 2016) at n 79. http://www.presscouncil.org.nz/principles accessed 1 November 2016. 189 Change to plurality as a result of the proposed merger 988. Based on the various measures outlined above, it is the Commission’s view that this merger would lead to a substantial increase in concentration across the news media sector generally as well as within several news media markets in particular. 989. As indicated above, the merged entity would control the overwhelming majority of domestic news content consumed via print and online. While a large proportion of the population would get some of their news from TV and/or radio broadcasts from other media organisations, this does not resolve our plurality concerns. 990. First, radio and TV news typically would not have the range and depth of news coverage of text-based news. 991. Second, as noted above, other large scale TV and radio organisations routinely monitor the Applicants’ websites and publications for news stories. Content published by the Applicants forms an important source of news for these organisations. Should the merged entity cease to uncover as many stories, whether as a result of cutbacks to journalistic staff or because of editorial choice, this would affect some of the content that these TV and radio broadcasters would otherwise cover. 992. Third, there is no likelihood that the merged entity’s strong position in the market for domestic news and information content published online would be eroded, especially in either the short to medium term, given the high costs of entry and expansion as discussion in Section 5. 993. Fourth, while TV continues to be an important news source, there appears to be a trend of decreased viewing of traditional (linear) TV broadcasts, and online has become the dominant news source, particularly among younger members of the public. While the decline in TV is yet to match the decline in print, if it continues, this could place TV as a source of news content in jeopardy. 994. Fifth, whereas radio does not appear to be facing any of the decline that TV and especially print appear to be experiencing, the two major news content stations that are not owned by the merging parties (MediaWorks’ Radio Live and RNZ) have a much smaller reach than that of the merging parties. For instance, Radio Live’s daily reach is 3% and RNZ’s is 13%. 995. Sixth, while the Applicants said that the merged entity would continue to be regulated by the Press Council, the remit and functions of the Press Council are limited (even if the merged entity continued to submit to it). The Press Council cannot ensure that a range of views and perspectives are published, and it plays no role in editorial decisions pertaining to what is not published or reported. 996. Moreover, the Commission notes that Fairfax currently has content sharing arrangements that provide it with access to (video) content from TVNZ and other content from RNZ. In this regard, TVNZ and RNZ utilise Fairfax’s stuff.co.nz website as another distribution channel to reach New Zealanders along with other distribution 190 channels such as social media sites or third party aggregators. If these content sharing agreements continue post-merger, TVNZ and RNZ will be to some degree dependent on the merged entity. Conclusion 997. The Commission is of the view that there is a real chance that the proposed merger would reduce media plurality. The merged entity would operate across number of media, dominating both online and print New Zealand news services, potentially with a single editorial voice. Given the importance of news media to wider issues of plurality, it is our view that any adverse effects are potentially substantial. Balancing of benefits and detriments 998. This application involves a balancing of the public benefits and detriments which will, or will be likely to result, from the merger. The Commission is required to decide whether the merger is likely to produce such a benefit to the public that it should be permitted. 999. Table 19 summarises the Commission’s quantitative assessment of the benefits and detriments arising from the merger under the status quo over a five year timeframe. Table 19: Estimated net quantifiable impact under ‘status quo’ scenario Time frame High detriment/low benefits Low detriment/high benefits 5 years $14.1 million $189.5 million 1000. The Commission also considered the likely detriments in a ‘digital only’ world to gauge the sensitivity of its results. Applying these detriments to the claimed quantifiable benefits also produces a materially positive net quantified benefit within the first year, ranging from $16.9 to $41.3 million. 1001. While we would expect merger-specific savings under a ‘digital only’ world to be substantially lower, we consider that the estimates calculated under the status quo and the ‘digital only’ scenarios suggest that the quantified net impacts of the proposed merger are likely to be positive. The quantified net benefit ranges are large, reflecting the inherent uncertainty of the analysis. 1002. Therefore, the Commission accepts that on a purely quantitative basis it is likely there is a positive financial benefit driven by cost-savings achieved by the merged entity. 1003. However, as previously outlined, this quantitative assessment does not determine the outcome of the application.524 Quantitative metrics are not perfect measures and quantification is only one tool to be used in reaching an authorisation decision. 524 Godfrey Hirst above n 9 at [115] to [117]. 191 The Commission is required to exercise its judgement (also referred to as making a qualitative assessment) as to whether it is satisfied on the evidence before it that the acquisition will result, or will be likely to result, in such a benefit to the public that the merger should be permitted.525 The Commission is not permitted to move directly from a quantitative analysis to a finding of satisfaction without exercising its judgement on the application in the round.526 1004. We have undertaken that qualitative assessment below. Qualitative assessment 1005. The Commission considers that the quantified net benefits are not reflective of the likely benefits and detriments of the merger. We reach this view because the benefits associated with this merger arise largely from the organisational efficiencies that are obtained through the merger of two businesses, for example reductions in sales and marketing, staff numbers, the cost of information technology and plant and buildings. Against these benefits, the unquantified detriments that we consider are likely to result from the merger have a much wider impact on New Zealand society. 1006. Through evaluating these unquantified detriments using measures of media quality (volume, variety and objectivity) and plurality (production/availability and consumption), we have been able to determine that the likely magnitude of these detriments is significant. In summary, as a result of the proposed merger, we consider that there would be the following likely unquantified detriments: 1006.1 a reduction in New Zealand news content and coverage; 1006.2 a negative impact on the New Zealand news ‘eco-system’, including online and print news, where news publishers benchmark against each other and are ‘news-takers’ from each other; 1006.3 a reduction in consumption – even if minimal – of New Zealand news and an associated impact on total welfare; and 1006.4 a loss of plurality of news media, through: 1006.4.1 a reduction in views and perspectives; 1006.4.2 a reduced check on objectivity or impartiality; and 1006.4.3 one media owner having a substantial amount of influence over the news and political agenda. 525 526 Section 67(3)(b); and note Godfrey Hirst above n 9 at [115] and Ravensdown Corporation above n 19 at [61] where the Court noted that “…it was legitimate for the Commissioners to stand back and notice that Ravensdown would be largely free from competitive restraint in the South Island markets and bring that consideration to account as part of the balancing exercise”. Godfrey Hirst above n 9 at [115] – [117]. 192 1007. In reaching this view, we recognise that media markets are changing and that print revenue and readership has been in decline, especially in recent years. We acknowledge that this change is affecting how news is consumed and is also creating opportunities for new media entities to have a voice. Further, we acknowledge that existing and emerging news media and platforms have a role to play in presenting a diversity of views and in enhancing news plurality. 1008. We have also taken into account the Applicants’ view that the proposed merger would allow them to invest in the delivery of news and that a combined business would be able to create “the most reliable and high quality New Zealand content” that can be sustained through these changes.527 1009. However, we consider that the merged entity would become the dominant provider of news services in New Zealand, particularly in print and online. And, after considering the evidence put forward by the Applicants and making our own enquiries, we are not convinced that competition from other news media would be sufficient to overcome the disparity. Our view is that competition would not be sufficiently robust to constrain a multi-media organisation, potentially with a single editorial voice, that would be the largest producer of national, regional and local news by some margin in New Zealand. 1010. We also do not consider that the two-sided nature of the market would prevent these detriments from occurring. The merged entity would have a large reach that would be appealing to advertisers. Consequently, we do not consider that advertisers would be likely to constrain the merged entity from reducing quality. 1011. The Commission would be concerned about a merger of this nature irrespective of the size of operational benefits – it would be difficult for the organisational integration achieved by the merger to offset the fundamental changes this merger would bring to New Zealand’s media landscape. That is an essential factor for the Commission to take into account in assessing this merger and is relevant to the importance (or weight) we are attributing to the unquantified detriments. 1012. Even in the face of a changing media landscape, the Commission cannot lose sight of the fact that is being asked to authorise a merger that would provide a single organisation with control of nearly 90% of all print media, New Zealand’s two largest news websites,528 and one of New Zealand’s two largest commercial radio companies. This would be an unprecedented level of media concentration in a wellestablished liberal democracy. 1013. Plurality of the media is essential to the maintenance of a well-functioning democracy, and a healthy democracy is dependent on a divergence of views. Representations made to us have highlighted the importance of these features in informing and influencing opinions and therefore contributing to democracy. 527 528 Application [20.3]. As outlined above, the merged entity’s websites would have a population reach more than four times larger than the next biggest New Zealand news website. 193 1014. We consider that Fairfax and NZME exert meaningful editorial influence over New Zealand’s news agenda. It is important that members of society – government, corporate and private citizens – are not able to influence one provider without the opportunity for differing views to be expressed. In that respect, it is essential that a wide range of issues and topics are able to be explored and addressed from different angles by the media, and that, similarly, specific topics or issues of public interest do not go unreported. 1015. Further, we do not consider that the self-regulatory structures that currently provide oversight of New Zealand’s media are sufficient to guard against a reduction in quality. It is unavoidable that a commercial provider has a set of editorial priorities or preferences based on its particular business objectives. Moreover, concerns regarding a media organisation’s subjectivity and preferences are not likely to be captured by any such regulations. 1016. We consider that the level of media concentration brought about by the proposed merger would not be in the public interest. We have weighed the cost-savings arising from the merger against the increased levels of media concentration, the ability of the merged entity to influence opinions and lead the news agenda and the overall detriments to plurality. In an industry where there are substantial costs of entry to achieve the scale of a large news publisher, we consider that the loss of plurality that arises from the proposed merger is likely to be significant and potentially irreplaceable. 1017. Therefore, although the net financial benefit is a positive value, our preliminary view is that we are not satisfied that the benefits of the proposed merger are likely to exceed the detriments. In the round, and standing back to assess the overall public benefits of the proposed merger,529 we consider that ultimately the unquantified (and quantified) detriments are likely to outweigh the quantified (and unquantified) benefits of this merger. Conclusion 1018. In our qualitative judgement, our preliminary view is that we are not satisfied that the merger will result, or will be likely to result, in such a benefit to the public that it should be permitted. 529 See n [532] above and Ravensdown Corporation above n [19 ] at 48. 194 Section 8: draft determination The Commission must be satisfied 1019. As set out in the competition analysis section above, the Commission’s preliminary view is that it is not satisfied that the merger will not have, or would not be likely to have, the effect of substantially lessening competition by increasing prices and/or decreasing the quality of content in the following markets: 1019.1 premium digital advertising; 1019.2 the provision of national online news and information services, through the Applicants’ stuff.co.nz and nzherald.co.nz websites; 1019.3 Sunday newspapers in the North Island; and 1019.4 community newspapers in: 1019.4.1 Whangarei; 1019.4.2 Hamilton; 1019.4.3 Rotorua; 1019.4.4 Taupo; 1019.4.5 Napier; 1019.4.6 Hastings; 1019.4.7 Stratford; 1019.4.8 Palmerston North; 1019.4.9 Horowhenua; and 1019.4.10 Kapiti. 1020. Nevertheless, the Commission has been asked to authorise the application under section 67(3)(b) of the Act. The Commission will grant authorisation if it is satisfied the merger will result, or will be likely to result, in such a benefit to the public that it should be permitted. 1021. Having applied our specialist judgement to assess the application in the round in the industry context, the Commission’s preliminary view is that it is not satisfied on the evidence that the proposed merger is likely to result in such a benefit to the public that it should be authorised. 1022. Some of the benefits and detriments arising from the merger can be quantified in monetary terms, while some benefits and detriments – including the reductions in 195 quality and plurality that we consider are likely to arise from the proposed merger – cannot. As can be seen from Table [19] above, the quantifiable benefits outweigh the quantifiable detriments in the five year timeframe by between $14.1 million (using high detriment assumptions) to $189.5 million (using low detriment assumptions). 1023. However, while we cannot quantify the detriments from a reduction in quality and plurality in monetary terms, we consider that they are likely to be so fundamental to a well-functioning New Zealand society that they outweigh the quantified net benefits from the merger. We therefore propose to decline authorisation for the merger. Draft determination 1024. The Commission’s preliminary view is that it is not satisfied that the merger will not have, or would not be likely to have the effect of substantially lessening competition in the markets identified at paragraph 1017 above. 1025. The Commission’s preliminary view is that it is not satisfied that the merger will result, or will be likely to result, in such a benefit to the public that it should be permitted. 1026. Following the Commission’s consideration of submissions on the Draft Determination, if its preliminary views are confirmed, the Commission would decline to grant authorisation for the merger under section 67(3)(b) of the Commerce Act 1986.