INSTITUTIONAL INVESTMENT IN THE HOUSING MARKET February 2019 Prepared by the Economics Division, Department of Finance finance.gov.ie i Contents Page Contents Tables, figures and boxes Executive Summary i ii iii 1. Introduction 1 2. Private Rented Sector 2 3. 3.1 3.2 Market Activity of Non-Households Apartments International experience 5 9 10 4. 4.1 4.2 Taxation of Institutional Investment REITS IREFS 11 11 11 5. 5.1 5.2 Private Rented Sector – Ownership Structure Dublin – overall ownership Dublin – apartment ownership 11 13 15 6. 6.1 6.2 6.3 6.4 6.5 6.6 6.7 6.8 6.9 6.10 Institutional Investment – impacts and related issues Supply Professionalisation of the rental market Competitiveness National Planning Framework Financial stability Local impact Pricing power Affordable housing Socio-economic polarisation Tenure neutrality 16 17 18 19 19 19 20 20 20 20 20 7. Conclusion 22 Tables, figures, and boxes Page Tables Table 1 Table 2 Table 3 Table 4 Table 5 Table 6 Net Unit Sales by ‘Financial & Insurance’ and ‘Real Estate’ firms, 2010-2017 Non-Household Purchase Shares, 2017 Non-Household Sale Shares, 2017 2017 Apartment Transactions, 2017 Individual and Company Landlords and Tenancies, September 2018 Apartment Stock and Company Tenancies, by Electoral Division 5 6 8 10 12 16 Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Figure 9 Figure 10 Owner-occupier rate, 1961-2016 Annual Property Price Inflation 2006-2018, by Property Type Headship Rate:1991-2016 Non-Household Purchases by Units, 2017 Non-Household Sales by Units, 2017 Net (sales less purchases) of all (new and existing) Dwellings, 2017 Share of Rental Stock Ownership, by number of tenancies owned RTB Tenancies – Company Landlords, by Electoral Division New Home Completions 1975-2017 (pre-2011 ESB connections) Residential Investment Activity, by investment in new or existing stock 2 3 4 6 7 9 13 14 17 18 Boxes Box 1 Headship Rates iii 4 Executive Summary This report examines the role of institutional investors and large-scale landlords in the Irish residential property market. It finds that the combined purchasing activity of property funds, real estate firms and real estate investment trusts is relatively small. Such firms purchased a net 1 per cent of transacted units in 2017. In fact, in every year since 2010 the ‘non-household’ sector as a whole sold more properties to the household sector than it purchased from it. The data show that even after excluding construction companies, non-households are net sellers in the residential market and have been for every year that the data are available. Ownership of rental properties by large scale landlords – those that own more than 100 rental units — is similarly minimal in the context of the wider market. Such firms hold 4.6 per cent of all tenancies nationally (excluding Approved Housing Bodies). The purchasing and ownership activity of such firms can, however, be far more pronounced at a local level. Using data from the Residential Tenancies Board, the report shows that, in some Electoral Divisions, the proportion of the rental stock in professional ownership is likely far greater than the national average. The perception of institutional investors purchasing large amounts of residential stock probably stems from their activity in the ‘new apartment’ sub-sector. Over 40 per cent of all new apartments are purchased by ‘Financial & Insurance’ and ‘Real Estate’ companies, while over 50 per cent of new apartments are sold intra-non-household and so are kept within the nonhousehold sector. This level of activity has caused some discomfort at a local level around the inability of first time buyers to purchase homes in new apartment blocks. Although many of these units would in the past have been purchased by buy-to-let investors and not owner-occupiers, some would-be purchasers have undoubtedly been displaced. However, the report outlines how institutional investment in the residential sector offers significant advantages. Such firms are investing in high quality, high density and well-located units for the rental sector. Following a similar evolution of the ‘Build-to-Rent’ business model that occurred in the UK, there is emerging evidence that such investors are the driving force behind a significant increase in the supply of new apartments in Dublin. Although in the main a positive development, the activities of real estate investment trusts, funds and private equity firms poses broader long-term questions for policymakers in relation to tenure, affordability and the socio-economic make-up of urban areas. This analysis examines these issues and outlines how, should it desire, the State has fiscal, regulatory and other policy levers available to address these issues. Institutional Investment in the Housing Market February 20191 Section 1 – Background and Introduction Large-scale professional landlords have not, historically, been a feature of the Irish housing market. Instead, the private rented sector has been dominated by relatively small-scale ‘buy-to-let’ (BTL) investors. Since the economic crisis, however, a noticeable feature of the housing market has been an up-scaling – albeit from a very low base – of the role played by companies such as pension funds, specialist private rental firms and Real Estate Investment Trusts (REITs) in the domestic residential property market. Although their presence is growing, these firms still represent a substantial minority of landlords – the total stock of properties owned by institutional investors is relatively low – and Ireland’s rental market continues to be dominated by BTL ownership. Indeed, the ‘non-household’ sector has been selling more properties to the household sector over the last number of years than it has been purchasing from it.2 3 Having said that, available evidence points to a significant role of institutional investors in certain segments of the residential property market such as the market for apartments and, in particular, the market for new apartments. This has led to concerns that such firms are ‘crowding out’ potential owner-occupiers of apartments. It also raises the question of whether some firms have developed sufficient scale to have price (rent) setting power in certain locations. At the outset, it is important to stress that institutional investment in apartments confers several advantages. Ireland has a shortage of apartments relative to other European countries, as well as an absolute shortage when assessed against the number of one-and two-person households. Furthermore, ‘compact living’ is a specific policy objective of the National Planning Framework (NPF). In contrast to the existing development model, higher-density housing located close to key employment clusters generates improved environmental, societal and economic outcomes. The purpose of this paper is to bring greater clarity to some of the issues associated with higher levels of institutional investment in the residential property market. Section 2 describes the structural decline in the owner-occupier rate over the past two decades and, by way of background, outlines recent trends in prices. Section 3 uses newly published Central Statistics Office (CSO) data to detail the market activity of the non-household sector, with a particular focus on the market for apartments. In section 4, a brief description of the tax treatment of some of the funds investing in Irish property is provided. Section 5 uses data from the Residential Tenancies Board (RTB) to describe the ownership structure of the Irish rental sector. Using RTB registrations as a proxy for ownership, the data show that large scale landlords play a marginal role in the national rental sector. However, some areas have levels of professional ownership significantly above the national average. Section 6 summarises all the pertinent issues around the increase in institutional investment. This section highlights the emerging data showing the significant increase in capital investment by ‘Build to Rent’ (BTR) investors. The level of additional units this investment may bring to the national housing stock 1 This report was produced by the Economic Division of the Department of Finance, and does not necessarily reflect the views of the Minister of Finance or the Irish Government. 2 The CSO release on non-household activity in the residential property market is available at: https://www.cso.ie/en/releasesandpublications/ep/p-rppi/residentialpropertypriceindexmay2018/non-householdsector/ 3 The non-household sector includes the activities of NAMA, which likely distorts the overall level of activity. 1 is potentially substantial. Finally, as a new structurally different housing market emerges, this section also posits some strategic questions for policymakers. Section 2 – Private Rented Sector In the early 1900s the vast majority of the Irish population lived in rented accommodation. However, for much of the 20th century the State encouraged home ownership, resulting in a shift in tenure profile. By the 1980s, Irish private rental accommodation was a ‘forgotten tenure’, with the majority of homes in the private rented sector being of poor quality and intended for temporary accommodation.4 In recent years there has been a shift back towards renting. As figure 1 illustrates, home ownership in Ireland has fallen from a peak of 79.3 per cent in 1991 to 67.6 per cent in 2016 – the lowest rate since 1971. There was a 2.1 per cent drop in home ownership between 2011 and 2016. Interestingly, the Irish owner-occupier rate is now below the EU average of 69.2 per cent. Figure 1: Owner-occupier rate, 1961-2016 85 80 75 70 65 60 55 50 2015 2010 2005 2000 1995 1990 1985 1980 1975 1970 1965 1960 State EU Average Source: CSO, Eurostat. In 2016, 29 per cent of the housing stock was used for rental accommodation – over 497,000 households — including 165,943 units owned by a local authority or voluntary body. Data from the 2016 Census show that renting is the majority tenure for those under 35 years of age. Furthermore, non-Irish nationals are proportionally more likely to be renting: in 2016, almost 40 per cent of households in the rental sector were non-Irish nationals. An important structural change over the past half century has been the decline in average household size, which has been on a downward trajectory since the 1960s. Of the 600,000 households formed since 1996, it is estimated that 400,000 comprise either one or two persons.5 Average household 4 5 Private Rented Accommodation; The Forgotten Sector of Irish Housing, O’Brien & Dillon, 1981 Lyons, R. Geary Institute for Public Policy lecture, 2017 size in Ireland is currently 2.75, well ahead of the European average of 2.3. There is no obvious reason why Irish rates will not continue to converge with the European norm, with the decline providing an additional source of housing demand (see Box 1). Other factors driving demand at present include strong population and economic growth, rising wages, as well as ‘pent-up’ demand from those unable to purchase during and since the recent economic crisis. As the increase in supply has not kept pace with the additional demand, residential property price inflation has remained elevated in recent years (Figure 2). Figure 2: Property Price Inflation 2006-2018, annual per cent change 35% 30% National - houses National - apartments 25% Year-on-Year Change 20% 15% 10% 5% 0% -5% -10% -15% -20% -25% -30% -35% 2018 2017 2016 2015 2014 2013 2012 2011 2010 2009 2008 3 2007 2006 Source: CSO Box 1: Headship Rate Several factors are expected to generate additional demand for housing in the short- and medium-term. One important factor is the so-called ‘headship rate’, defined as the proportion of the population who are heads of households. The headship rate is the reciprocal of household size, i.e. as average household size declines the headship rate rises. The headship rate for the total population has risen from 29 per cent in 1991 to 36 per cent in 2016 (figure 3). Figure 3: Headship Rate: 1991-2016 0.38 0.37 0.36 0.35 0.34 0.33 0.32 0.31 0.30 0.29 0.28 2016 2011 2006 2002 1996 1991 Source: CSO The headship rate is likely to continue its long-term trend upwards, primarily due to: • Demographics — a rise in the number of young adults • Continued economic expansion — enabling earlier household formation among young adults The impact of the economic crisis can be seen in the decrease in the headship rate between 2011 and 2016. As unemployment rose, incomes fell and immigration increased, relatively fewer households were formed. For example, Census 2016 showed that the number of adults, aged 18 years and over, who were working and still living at home rose by 19 per cent between 2011 and 2016, increasing from 180,703 to 215,088. Declining affordability of housing – in particular rental accommodation — can have a negative impact on household formation. Byrne et al. (2018) find that falling rents are linked with an increased rate of household formation*. The impact of affordability on household formation takes the form of students staying at home through third level education, employed persons continuing to reside in their childhood home or a rise in the number of adults sharing rental accommodation. Despite the recent decline in affordability, economic factors are forecast to reverse the recent trend in headship decline and resume the long-term increase evident before the crisis. As the headship rate rises, ceteris paribus, the demand for housing increases. The NPF assumes a headship rate of 40 per cent by 2040 (based on projected household size of 2.5). This represents a demand for an additional 210,000 housing units by 2040 against a scenario in which the headship rate was constant. Such an increase would involve demand for an additional 10,000 housing units per annum. * Byrne, D., Duffy, D., and Fitzgerald, J. (2018), “Household Formation and Tenure Choice: Did the Great Irish Housing Bust Alter Consumer Behaviour?” Economic and Social Review 49(3), pp. 287-317. Available at: https://www.esr.ie/article/view/979/196 Section 3 – Market activity of non-households In July 2018, the CSO published statistics on the activity of ‘non-households’ in the residential property market. Non-households include both public and private sector organisations and also the activities of NAMA. NAMA has been active in the sale of residential property in recent years through loan and asset sales, both through its debtors and via receivers acting on its behalf. NAMA has also facilitated the sale of residential units via its residential delivery programme, through which the Agency funds the development of residential housing. Such transactions are categorised under ‘Financial & Insurance’, ‘Real Estate’ or ‘Construction’, depending on the classification of the legal entity that sold the property. The data show that even after excluding construction, non-households are net sellers into the residential market and have been for every year that data are available. Table 1 shows the net activity of ‘Financial and Insurance’ and ‘Real Estate’ companies since 2010. The ‘Financial and Insurance’ sector has been a net seller of units every year. The ‘Real Estate’ sector was a net purchaser of units in 2014 and 2017. Overall, the ‘Real Estate’ sector is a net purchaser of units over the eight year period, but by less than 0.01 per cent, or 380 units. Table 1: Net Unit Sales by ‘Financial & Insurance’ and ‘Real Estate’ firms* Financial & Insurance Real Estate Total Market Transactions Net Activity (% terms) Financial & Insurance Real Estate 2010 2011 2012 2013 2014 2015 2016 2017 427 394 18,602 530 143 15,870 933 89 22,164 1,241 109 26,563 2,186 -997 40,847 2,957 239 45,448 4,288 149 46,658 4,168 -506 51,155 2.3% 2.1% 3.3% 0.9% 4.2% 0.4% 4.7% 0.4% 5.4% -2.4% 6.5% 0.5% 9.2% 0.2% 8.1% -1% * Sales minus purchases, negative number implies net purchase Source: CSO In 2017, non-households (ex-construction) accounted for 14.1 per cent of all purchases and 21.1 per cent of new dwelling purchases. However, they accounted for 18.8 per cent of all dwelling sales and 35.7 per cent of new dwelling sales. The ‘Financial & Insurance’ and ‘Real Estate’ sectors combined were also net sellers in 2017. Assessed individually, the ‘Real Estate’ sector was a net purchaser, purchasing a net 1 per cent of all stock. Excluding the ‘Financial & Insurance’ sector, which should exclude the majority of NAMA activity, shows that the remaining sectors were net purchasers of ‘all’ property in 2017, purchasing a net 1,802 units. Although the remaining sectors (ex-construction) were still net sellers of ‘new’ units, selling a net 268 units. Figure 4 illustrates the share of unit purchases by each non-household sector in 2017. Table 2 describes this in percentage terms and outlines the breakdown between new and existing properties. 5 Figure 4: Non-Household Purchases by Units, 2017 Households 1060 406 1209 43171 1691 7928 956 693 1913 Non-households Construction Financial & Insurance Real Estate Human Health & Social Work Extra-Territorial Public/Education Other Source: CSO Table 2: Non-Household Purchase Shares, 2017, per cent Construction Financial & Insurance Real Estate Human Health & Social Work Extra-Territorial Public/Education Other Total Share of nonhousehold purchases Share of new dwelling purchases Share of all purchases 8.7 24.1 21.3 13.4 5.1 15.2 12.1 100.0 0.9 4.2 7.2 5.3 0.5 2.3 1.6 22.0 1.4 3.7 3.3 2.1 0.8 2.4 1.9 15.5 Source: CSO  ‘Financial and Insurance’ and ‘Real Estate’ companies purchased 11.4 per cent of new dwellings and 7 per cent of dwellings overall.  The ‘Real Estate’ sector purchased 7.2 per cent of new dwellings and 3.3 per cent of all dwellings.  The public sector — comprising ‘Public and Education’ and ‘Human Health and Social Work’ — is a greater purchaser of homes than the ‘Real Estate’ sector.  The public sector purchased 7.6 per cent of new dwellings and 4.5 per cent of all dwellings. Within that: o ‘Public and Education’ bodies purchased 2.3 per cent of new dwellings and 2.4 per cent of all dwellings. o The ‘Human Health and Social Work’ sector, which includes Approved Housing Bodies (AHBs), purchased 5.3 per cent of all new dwellings and 2.1 per cent of all dwellings. Figure 5 illustrates the share of unit sales by each non-household sector in 2017. Table 3 describes this in percentage terms and outlines the breakdown between new and existing properties. Figure 5: Non-Household Sales by Units, 2017 Household 6081 1185 35838 15261 2003 5660 Non-households Construction Financial & Insurance Real Estate Human Health & Social Work Extra-Territorial Public/Education Other Source: CSO 7 54 146 132 Table 3: Non-Household Sale Shares, 2017, per cent Construction Financial & Insurance Real Estate Human Health & Social Work Extra-Territorial Public/Education Other Total Share of nonhousehold sales Share of all new dwelling sales Share of all sales 37.1 39.9 7.8 0.4 1.0 0.9 13.1 100.0 50.2 15.9 10.5 0.4 0.6 0.5 7.8 85.9 11.1 11.9 2.3 0.1 0.3 0.3 3.9 29.9 Source: CSO  Excluding construction, non-households accounted for 35.7 per cent of new dwelling sales and 18.8 per cent of all dwelling sales.  The ‘Financial & Insurance’ and ‘Real Estate’ sectors sold 26.4 per cent of new dwellings and 14.2 per cent of all dwellings last year.  In 2017, firms in ‘Financial & Insurance’ and ‘Real Estate’ purchased 3,604 dwellings and sold 7,266. They were net sellers of 3,662 units.  The ‘Real Estate’ sector sold 2.3 per cent of all sales and purchased 3.3 per cent of all purchases. The sector was a net purchaser of 1 per cent of transacted stock, or approximately 500 units. Figure 6 shows the net position of each sector. Figure 6: Net (sales less purchases) of all (new and existing) Dwellings, 2017 Sales Purchases Other Financial & Insurance = net sellers into market; Real Estate = net purchaser (1 per cent of transactions) Public/Education (O/P) Extra-Territorial (U) Human Health & Social Work (Q) Real Estate (L) Financial & Insurance (K) Construction (F) Household -40,000 -30,000 -20,000 -10,000 0 10,000 20,000 30,000 40,000 50,000 Source: CSO The above figures illustrate that nationally and in aggregate terms (houses and apartments) the impact of the ‘Real Estate’ industry is marginal. However, such companies have a disproportionate impact in the market for apartments, particularly new apartments. 3.1: Apartments There were 11,101 apartment sales (market) in 2017, 1,873 new dwellings and 9,228 existing units. Non-households were net sellers of apartments, even when excluding construction. However, 52.4 per cent of new apartments were sold intra-non-household i.e. sold and purchased within the nonhousehold sector. Table 4 below shows that although ‘Financial & Insurance’ and ‘Real Estate’ firms were also net sellers in 2017, it is significant that over 40 per cent of all new apartments were purchased by these firms. ‘Real Estate’ companies are net buyers of apartments, purchasing 9.5 per cent of all purchases while selling 3.95 per cent of all sales. In 2017, ‘Real Estate’ companies bought 25 per cent of all new apartments. 9 Table 4: 2017 Apartment Transactions, 2017, per cent (unless otherwise stated) Apartment transactions (number) Non-household purchases Non-household sales Financial & Insurance & Real Estate purchases Financial & Insurance & Real Estate sales Real Estate purchases Real Estate sales New Existing All 1,873 56.3 90.6 40.5 33.1 25.1 22.9 9,228 28.9 33.8 16.4 21.0 6.3 1.1 11,101 33.6 43.3 20.4 23.1 9.5 3.9 Source: CSO 3.2: International experience Comparable data on the activity of the non-household sector in other countries are unavailable. The work carried out by the CSO in this regard is at the forefront of research in this area. In the absence of robust official data, the Department must rely on some basic observations on the profile of institutional investors in other markets. Previous work by the Department of Finance examined the BTR sector in the UK, Germany and the US. BTR is a type of private rental investment, where typically, a whole block of apartments is built or bought by a single entity with the intention of leasing all units over the long-term. The paper found that the US and German BTR markets were very mature with significant involvement by large scale institutions, although the type of institution differed between countries. In the US, investors in residential stock are generally publicly listed companies, private equity investors or REITs. In Germany, the investor profile is more mixed with large private firms as well as incorporated public entities operating in the market. There are three channels by which BTR investors typically invest in residential property: purchase existing buildings, forward fund a developer to build stock, or purchase a new building via a forward purchase agreement. The UK experience shows that initial BTR investment tends to focus on purchasing existing stock. Once familiarity with the market is achieved, firms then begin to forward fund and forward purchase new stock. The market in the UK is very similar to the Irish market in terms of the growth of BTR. As of Q4 2018 there were almost 29,500 BTR units across the UK, a 29.4 per cent increase on Q4 2017. There were also a total of 110,100 BTR units either in planning or under construction. It is estimated that such developments now represent one in five building starts in London.67 6 Information on BTR in the UK is available at: https://www.barclayscorporate.com/content/dam/corppublic/corporate/Documents/Industry-expertise/BTR-Report.pdf 7 https://www.bpf.org.uk/what-we-do/bpf-build-rent-map-uk Section 4 – Taxation of institutional investment There are a number of different corporate entities investing in residential property. The most high profile are REITS but other corporate structures such as pension funds, joint ventures, Irish Real Estate Funds (IREFs) and publically listed BTR firms are also active. The tax status of REITs and IREFs, in particular, has attracted attention. 4.1 REITs REITs are collective investment vehicles designed to hold rental investment properties in a taxneutral manner. They are focused on long-term holding of income-producing property as opposed to short-term speculative gains. The function of the REIT framework is not to provide an overall tax exemption, but rather to facilitate collective investment in rental property. REITS provide the same after-tax returns to investors as direct investment in rental property and remove a double layer of taxation at corporate and shareholder level which would otherwise apply. REITs provide a number of benefits for investors when compared with direct property investment, including diversification of risk, increased liquidity, lower entry costs, and regular income streams. They are designed to provide investors with access to investment in real estate without buying property directly. Where a REIT complies with a number of requirements, such as the distribution of 85 per cent of its rental profits annually for taxation at the level of the investor, it is exempt from corporation tax on qualifying income and gains from rental property. 4.2: IREFs IREFs are investment undertakings (excluding UCITS) where at least 25 per cent of the value of that undertaking is made up of Irish real estate assets or where it is reasonable to consider that the main purpose, or one of the main purposes, of the investment undertaking is to acquire IREF assets, or to carry on IREF business. The legislation was introduced in the 2016 Finance Act to address concerns raised regarding the use of collective investment vehicles to invest in Irish property. Certain investors had been using the structures to minimise their exposure to Irish tax on Irish property transactions. IREFs must deduct a 20 per cent withholding tax on certain property distributions to non-resident investors. The withholding tax does not apply to certain categories of investors such as pension funds, life assurance companies and other collective investment undertakings. Section 5 – Private Rented Sector — ownership structure The vast majority of residential properties in the ownership of non-households are rental properties. The RTB is the primary source for data on rental market prices, number of tenancies and the associated number of tenants and landlords. The RTB has provided a breakdown of tenancies by type of landlord and number of tenancies. The data exclude the activities of the AHBs and are correct as of 13th September 2018. These data are presented in Table 5 below. 11 Table 5: Individual and Company Landlords and Tenancies, September 2018 Number of Tenancies Total Landlords 1 2 3 4 5 6 7 8 9 10-20 20 - 50 50 - 100 100 - 200 200 - 300 300 - 400 400 - 500 500+ Total 121,481 27,823 9,722 4,627 2,573 1,646 1,085 661 561 1,924 618 113 45 8 4 2 5 172,898 Tenancies Individual Landlords Company Landlords % Landlords % Tenancies 113,534 52,023 27,257 17,276 12,006 9,229 7,121 4,907 4,687 23,583 17,505 7,329 5,513 1,845 1,272 767 4,901 310,754 117,863 26,938 9,287 4,370 2,394 1,494 986 602 495 1,626 410 55 17 1 0 1 0 166,539 3,618 885 435 257 179 152 99 59 66 298 208 58 28 7 4 1 5 6,359 70.26% 16.10% 5.62% 2.67% 1.49% 0.95% 0.63% 0.38% 0.32% 1.11% 0.36% 0.07% 0.03% 0.01% 0.00% 0.00% 0.00% 100% 36.53% 16.74% 8.77% 5.56% 3.86% 2.97% 2.29% 1.58% 1.51% 7.59% 5.63% 2.36% 1.77% 0.59% 0.41% 0.25% 1.58% 100% Source: RTB An individual tenancy is not fully analogous to an individual property. There may be instances where there are multiple tenancies in one property. Similarly, some tenancies may have more than one individual landlord associated with them. Furthermore, the RTB database is a register of tenancies, not of ownership. However, the vast majority of tenancies are individual properties and are associated with a single landlord. As such, Table 4 can be used as an indicator of ownership of rental property. Individual landlords tend to register between 1 and 5 rental properties, with the vast majority (70 per cent) registering a single rental property. Over half of all companies (3,618) also register just a single rental property.8 Only 4.6 per cent of tenancies are held by landlords (both individual and company) with more than 100 units. There are just 5 companies that have registered more than 500 tenancies, accounting for a total of 1.6 per cent of the rental stock overall. The RTB cannot currently provide data on the concentration of registrations over time. In an attempt to investigate if such a trend exists, the Department of Finance analysed Local Property Tax (LPT) returns for the number of ‘surplus properties’ owned by individual tax entities. It shows that the number of properties owned by tax entities with over 200 units more than doubled between April 2014 and December 2016, from 2.2 per cent to 5.2 per cent. Differences in methodology and overall quantum aside, it is likely that any RTB data would display a similar trend. Figure 7 illustrates the share of total ownership of rental property owners, by number of rental properties registered. 8 Certification as either a company or individual landlord is self-reported. As such there may be inaccuracies caused by misclassification. Figure 7: Share of Tenancies by Number of Tenancies Held 5513 7329 1845 1272 767 4901 17505 23583 248040 1-10 10-20 20 - 50 50 - 100 100 - 200 200 - 300 300 - 400 400 - 500 500+ Source: Residential Tenancies Board Ireland has a total housing stock of just over 2 million units. Assuming that the vast majority of RTB registrations are equivalent to ownership of a single property, the above data indicate that landlords (both companies and individuals) with 20 units or more own approximately 2 per cent of the total housing stock, and 12.6 per cent of the total rental stock (excluding local authorities and AHBs). Large scale landlords (both companies and individuals with 200 units or more) own just 0.04 per cent of the total housing stock, and 2.8 per cent of the total rental stock (excluding local authorities and AHBs). The RTB registration data indicate that ownership by large scale investors (including institutional owners) in the rental market is still limited – both in their total number and in the number of rental properties they own. However, certain localities have a higher than average proportion of ownership by professional landlords. 5.1: Dublin – overall ownership Due to the small number of firms active in this area and the relatively low level of transactions, the CSO cannot collect data on the purchasing activities of non-households on a sub-county level. However, recent collaboration between Dublin City Council, the All Island Research Observatory in Maynooth University and the RTB as part of the Dublin Housing Observatory, has enabled a better understanding of the role of ‘company landlords’ at a more granular level.9 The data distinguish between individual landlords, company landlords and AHBs. The Dublin Housing Observatory map below illustrates the number of tenancies registered with a company landlord at an Electoral Division (ED) level in Dublin in 2017. These data exclude AHBs. 9 As above, identification as a company landlord is self-reported. 13 Figure 8: RTB Tenancies – Company Landlords, by Electoral Division Source: Dublin Housing Observatory10 The highest proportion of company landlords in Dublin is found in the ED of Ushers A, where 605 out of a total 958 tenancies (63 per cent) were registered with a company landlord in 2017. A number of other EDs indicate a high level of company ownership. In Sandyford, where 1,212 tenancies were registered with the RTB, 393 were done so with company landlords, or 32 per cent of tenancies in the area. Some 778 of the 1,878 tenancies (41 per cent) in Tallaght-Springfield are registered with companies. Stillorgan-Leopardstown had 273 out of 556 tenancies (49 per cent), while Chapelizod had 427 out of 880 tenancies (49 per cent) registered with company landlords. The data also suggest that individual landlords continue to have a high degree of ownership of the rental stock in many central areas of the city. In Merchant’s Quay, 705 out of 858 tenancies (81 per cent) were registered with individual landlords in 2017; in the South Docks individuals held 1,255 out of 1,535 units (82 per cent) in 2017; and in the electoral district of North City 749 of the 968 tenancies (77 per cent) were registered by individuals in 2017. It may be the case that individual landlords are more likely to own older stock, more prevalent in the city centre than in the suburbs. The RTB data indicate that company ownership of rental stock is relatively small on a national level. However, as the Dublin Housing Observatory data indicates, private firms can have a significantly higher level of ownership at a local level. Similarly, their ownership of apartments, particularly in specific localities, can be disproportionate to their overall share of the market. 10 https://airomaps.geohive.ie/dho/ 5.2: Dublin — apartment ownership The bulk of institutional investment in Ireland thus far has been in apartments, particularly new apartments. Just 12 per cent of the dwellings in Ireland are apartments, approximately one quarter of the EU average. If Ireland was in line with the EU average, there would be approximately 800,000 apartments in the country, 600,000 more than at present. Although Dublin has a higher proportion of the stock comprising of apartments (26 per cent), it is approximately one third of the average of many of its peer cities including Copenhagen, Brussels, Zurich and Amsterdam. 11 Census 2016 provides data on the stock of apartments in each ED. Comparing the number of tenancies registered by ‘company landlords’ with the apartment stock in each ED in Dublin provides a guide to the potential level of professional ownership in that area. The analysis makes the very strong assumption that company landlords exclusively invest in apartments, which is not likely to be the case. Furthermore, large scale landlords such as REITs and investment funds are only a sub-set of the broader ‘company landlord’ category. However, the analysis provides a guide to the potential market share of companies if their investment strategies focused purely on apartments, which is the case with most BTR investment. Examining a sub-sample of 65 EDs in Dublin (out of a total of 322) in which more than 40 per cent of the total apartment stock are rental units, gives a picture of the local ownership profile of company landlords. Using the above assumption — that company landlords exclusively invest in apartments — approximately 13 per cent of apartment units are held in the ownership of company landlords within these E.Ds. Given that this analysis uses company tenancies as a proxy for apartment ownership and is limited to those EDs where apartment rental is high, it suggests that professional owners of the apartment stock are still a substantial minority in Dublin. However, some localities have professional ownership rates far in excess of the county average. Table 6 provides information on 13 EDs in which the potential company ownership rate is above 20 per cent. 11 Lyons, R. Geary Institute for Public Policy lecture, 2017 15 Table 6: Apartment Stock and Company Tenancies, by selected Electoral Division Electoral Division Stock of Apartments (2016) Company Tenancies (2018) Company Tenancies as a proportion of the Apartment Stock Chapelizod Tallaght Springfield Ushers A North Dock A Pembroke West B Tallaght Glenview Pembroke West A Mansion House B Pembroke East D Grange B North Dock B Dundrum - Sandyford 826 1,923 1,691 125 907 545 1,218 338 1,227 949 2,111 1,770 427 774 605 39 262 153 332 85 287 220 485 393 52% 40% 36% 31% 29% 28% 27% 25% 23% 23% 23% 22% Dundrum - Balally 1,714 338 20% Source: RTB, Dublin Housing Observatory, Census 2016 The data implies that company landlords could potentially hold 36 per cent of the total apartment stock in Ushers A; 40 per cent in Tallaght-Springfield and 52 per cent in Chapelizod. Ownership shares in the rental market beyond a certain level in any particular locality may bestow rent-setting power to an individual or group of firms. It is beyond the scope of this paper to attempt to establish if such power exists in any particular locality. Any such analysis would be challenging given the difficulty in establishing a causal relationship between above average rents and the proportion of non-households or company ownership in a given area. Factors such as the quality of the stock and location would have to be controlled for.12 Section 6 – Institutional investment — impact and related issues Due to the pace at which the BTR sector has grown and the high profile nature of some recent transactions, concerns have been raised that such investment is ‘crowding out’ owner-occupier activity. In the absence of a more balanced view there is a risk that such high profile transactions could lead to imprecisely targeted regulatory interventions. At a local level there have been instances of investors purchasing large blocks of stock, removing some of those units from the owner-occupier segment of the market. However, at a national level, or in terms of the broader Dublin market, the advantages of institutional investment are manifold, broad-based and beginning to yield positive outcomes for the Irish housing system. 12 Given that Rent Pressure Zones (designated areas where rent can only be increased up to 4% per annum) restrictions do not apply to new stock, it is likely that localities with a high level of institutionally-owned new stock would display above average rents. Any analysis would need to compare new individually-owned stock with new institutionally-owned stock, as well as control for the other factors outlined above. 6.1: Supply Medium-term annual demand for housing in Ireland is estimated at between 30-35,000 units. Given current average household size, this is the equivalent of a need to house the population of Galway City every year. From 2011 to 2017 an average of 7,651 units were built annually (see Figure 9), resulting in a considerable level of ‘pent-up’ demand. The shortage created by this mismatch has resulted in high inflation in both the owner-occupier and rental sectors. The private rental sector in particular has witnessed a sustained period of high inflation. Rents are now at their highest since records began. In Q3 2018, average rents were 50 per cent higher than they were at the post-crisis trough in Q1 2012.13 In normally functioning markets this price signal would result in a supply response, but until recently capacity constraints in the construction and financial sectors hindered such a response. Market failure emerged as the price mechanism failed to induce sufficient supply resulting in an undersupply of housing. Recent analyses suggest that this response is now emerging, in part due to significant levels of funding from institutional investors. Figure 9: New Home Completions 1975-2017 (pre-2011 ESB connections) 100,000 90,000 80,000 70,000 60,000 50,000 40,000 30,000 20,000 10,000 0 Source: CSO Expanding the BTR sector is a specific objective of Rebuilding Ireland: An Action Plan for Housing and Homelessness.14 Policy appears to be effective in this regard as the sector plans and delivers an increasing amount of units. CBRE estimate that 30 per cent of commercial real estate investment in 2018— €1.13 billion— was invested in BTR developments.15 This large scale and unprecedented level 13 14 15 RTB Rent Index, Q3 2018 Rebuilding Ireland: An Action Plan for Housing and Homelessness, Pillar Four CBRE Research: Real Estate Market Outlook 2019 17 of investment is likely driving a significant increase in the number of apartments under development. In a recent analysis, AIB estimate that there are just under 13,000 apartments in the development pipeline in Dublin, representing 68 per cent of planned units.16 In the first 9 months of 2018, 57 per cent of units that were granted planning permission in Dublin were apartments, a total of 4,191. While it is not certain that all these apartments will ultimately be developed, these figures represent a significant level of potential medium to long-term supply. For context, in the three years between 2015 and 2017 just over 4,000 apartments were completed nationally. Throughout Ireland, the number of apartments as a proportion of all new home completions increased from 12 per cent to 16 per cent between 2016 and 2017. Figure 10 shows institutional investment in residential property from 2013 to the first half of 2018. Figure 10: Residential Investment Activity, by investment in new or existing stock and year 600,000,000 500,000,000 400,000,000 300,000,000 200,000,000 100,000,000 0 2013 2014 2015 Standing Stock 2016 2017 2018 H1 Forward Commit Source: CBRE Research Of particular note is the change in activity from the purchase of existing stock to investment in new stock. This trend mirrors the development of the BTR sector in the UK and represents a significant development in the evolution of the industry in Ireland. Institutional investors are now deploying large amounts of capital for the construction of new apartments. Given the growth in the volume of BTR investment, a continuation of this trend should result in a significant increase in the national apartment stock. 6.2: Professionalisation of rental market The rental market remains primarily comprised of small-scale landlords. Institutional investment provides an opportunity to professionalise the sector, to realise economies of scale, and to improve regulatory and taxation compliance standards. It would be erroneous to assert that, in the absence of institutional investment, the housing such firms currently purchase would necessarily be bought by first time buyers, or the household sector. Much of the stock would likely be purchased by BTL investors with access to equity, either through their household or business’ wealth, and not first time 16 ‘The Private Rented Sector in Ireland’, AIB Real Estate Finance buyers with limited access to such equity. Institutional investment is — to at least a significant degree — likely replacing BTL investors, not first time buyers. There are numerous positives to professionalisation of the sector. 6.3: Competitiveness The availability and affordability of suitable accommodation for one-and two-person households may act as a competitive disadvantage in attracting Foreign Direct Investment. In particular, more transient workforces such as those present in many ICT companies in Dublin and other cities require smaller, centrally located accommodation. Market participants have indicated to the Department of Finance that the lack of such accommodation is acting as a restraint on further job creation in these sectors. The type of housing that institutional investors have thus far delivered — higher density apartment stock — is more closely aligned to the needs of such workforces, as well as to the broader changes in household size outlined previously. 6.4: National Planning Framework A key objective of the NPF is the creation of liveable, diverse and integrated communities. In Dublin, the key employment clusters remain in the city centre resulting in a heavy reliance on commuting. Ireland’s boom-time development pattern was one of significant suburban development and there is emerging evidence that this is now being repeated. In contrast, the NPF specifically targets more compact growth, creating liveable cities in which workers live close to their place of employment. Given the location, density and characteristics of institutionally owned stock, such investment can play a positive role in helping the State meet its NPF objectives in this regard. 6.5: Financial stability The Department of Finance recently undertook an analysis of the potential rise in the loan to deposit ratio (LTD) of the Irish banking system under a hypothetical scenario in which 50,000 residential units were both demanded and supplied annually, by 2024.17 Using COSMO — the ESRI’s structural econometric model of the Irish economy — the analysis indicated that in order to provide the required credit to fund the supply (construction) and demand (mortgages) for housing, the LTD ratio would need to rise to 1.6. The current aggregate LTD ratio of the banking system is 0.8. Although the LTD ratio is not a sufficient measure of the overall stability of the banking system, a rise of this magnitude may be problematic. Institutional investment in the residential sector has the potential to reduce reliance on bank funding for development, which is important in building broader capital markets as set out in the EU’s Capital Markets Union agenda. By reducing the volume of a historically volatile asset class from the aggregate balance sheet of the banking system, it also helps to de-couple the financial health of the banking system from the vagaries of the residential investment cycle. In doing so, equity capital also releases bank capital for alternative forms of lending. Notwithstanding the positive impacts outlined above, there are legitimate concerns raised by the growing influence of institutional investors, as well as broader socio-economic issues policymakers need to consider. Some of these are set out below. 17 ‘The Funding Gap’, Department of Finance, Forthcoming 19 6.6: Local impact Given the limited number of new apartment completions overall and the relatively narrow geographic focus of institutional investor’s to-date, there have been instances in which such investors have purchased the vast majority of apartments that come to market in a particular area. Although many of these apartments would likely have been purchased by BTL investors and not owneroccupiers, such purchases may have the effect of reducing the choice available to some first time buyers purchasing in their own locality. 6.7: Pricing power Although as yet unproven, there is a risk that at sufficient scale an institutional investor or group of investors could, over time, develop monopolistic or oligopolistic pricing power. The highly dispersed nature of the Irish rental ownership structure to-date may have obviated such a development in the past. Theoretically, such price-setting power could be attained at a local level given certain market conditions and sufficient scale. 6.8: Affordable housing A previous analysis by the Department of Finance found that BTR investment tended to supply apartments at the premium end of the market, unsuitable for people on average incomes. Given the location of much BTR investment to-date, the relative expense of building apartments as opposed to suburban houses and the ancillary services often offered in BTR complexes, it is likely that this trend will continue. Institutional investment is unlikely to have a direct impact on increased affordability. Impacts will more likely be of an indirect nature. An increase in general supply should place downward pressure on rents by reducing competition between prospective tenants at lower price points. 6.9: Socio-economic polarisation There is a risk that, should BTR investment continue at current growth rates, market forces would over the long-term create socio-economic polarisation in some urban areas. Under such a scenario average income earners would be priced-out of purchasing or renting from the private market, while local authorities would continue to provide housing for people on low incomes. The State has responded to this with initiatives such as the cost rental model, the proposed Affordable Purchase Scheme and the Land Development Agency. 6.10: Tenure neutrality Government policy is one of tenure neutrality. At a national level, such a policy may lead to a natural ratio of owner-occupied housing to rental. At a local level, this may lead to areas almost entirely owner-occupied, or entirely rented. The State is not powerless in affecting tenure outcomes at a local level. Direct State interventions such as the proposed Affordable Purchasing Scheme or changes to the planning regime have the potential to introduce an element of tenure planning — should the State wish to do so. More broadly, given the ongoing structural change in the housing market, typified by the increasing presence of institutional investors, Irish policymakers need to consider the long-term implications of a shift in tenure profile. As currently constituted, welfare provision for older cohorts implicitly presupposes home ownership at retirement age. The fiscal implications of a change to this assumption are considerable. Section 7 – Conclusion The data outlined in this paper show that although institutional investors play an increasingly important role in the private rented sector, their purchasing activity represents a small proportion of the housing market overall. On a national level they also remain a significant minority of landlords. The growth of institutional investment is the result of a structural change in the market. The change has come from a combination of post-crisis capacity constraints in the financial and construction sectors; long-term societal changes such as increasing urbanisation and changing tenure profile; and a desire to avoid previous mistakes by improving spatial and urban planning. This changed environment has resulted in significant market dislocation and an imbalance between the demand and supply of suitable housing, particularly in the rental sector. Institutional investment has the potential to significantly increase the supply of high quality, high density and well-located units. However, such investment can only be one aspect of a multi-pronged response to addressing current issues in the market. Expansion of the sector must continue to be accompanied by a set of policies that can facilitate its positive impacts while addressing the broader issues it raises. 21 Government Buildings, T: 353 1 604 5626 Upper Merrion Street, F: 353 1 676 1951 Dublin 2, www.finance.gov.ie D02 R583 Ireland.