Taxation and Subsidies for Housing and Land: Market Impacts and Economic Efficiency Implications

Transcription

1 Taxation and Subsidies for Housing and Land: Market Impacts and Economic Efficiency Implications Peter Abelson Macquarie University Abstract This paper quantifies the major subsidies and taxes separately for the homeowner and rental sectors, analyses the effects of these fiscal measures on house prices and housing consumption in the owner-occupier and investor/renter sectors, and quantifies the efficiency effects. The paper estimates that the private housing sector receives a net subsidy of $7.0 billion per annum. Almost the whole subsidy accrues to homeowners, who receive about a 10 per cent subsidy on imputed gross rentals. There is minimal net subsidy to the rental sector. On plausible demand and supply assumptions, a 10 per cent subsidy to homeowners raises housing consumption by about 3.4 per cent and raises house prices also by an estimated 3.4 per cent. The housing subsidy produces an estimated deadweight loss from expenditure on renovations of about $100 million per annum. However, contrary to previous work, the paper finds that the housing subsidy produces welfare gains from expenditure on new housing in the order of $150 million a year. This arises because the subsidy offsets the over-regulated supply of new housing. However, transaction taxes on housing have a separate deadweight cost on asset exchange of $375 million per annum. Also, the unequal treatment of homeowners and renters has a small annual deadweight cost. JEL classification H20, H21, R31 Acknowledgements I thank Roselyne Joyeux for her assistance in the preparation of this paper. Any errors are my responsibility. Note: Paper to be presented to the 34 th Australian Conference of Economists in Melbourne in September Any comments on the paper are welcome.

2 Taxation and Subsidies for Housing and Land: Market Impacts and Economic Efficiency Implications 1 Introduction Residential property attracts concurrently numerous taxes and substantial subsidies. Governments typically view property as an irresistible sitting target for taxation. In Australia, taxes on housing include land taxes, local council rates, stamp duties on property transfers, numerous developer charges, the goods and services tax (GST) on sales of new dwellings and renovations of existing dwellings, and capital gains taxes on rental property and second homes. On the other hand governments recognise that housing is a fundamental component of household welfare. Many households spend a quarter or more of their income on housing. Housing is also a major vehicle for household saving. Thus, in Australia there is no income taxation of imputed rent, no expenditure taxes on rents, and no capital gains tax for owneroccupied housing. There are grants to assist first home buyers and rent subsidies for low income households in private and public housing. In this paper I analyse the market impacts and efficiency implications of these fiscal measures. Section 2 describes and quantifies the main taxes and subsidies that affect housing. Section 3 analyses the impacts of these fiscal provisions on house prices and housing consumption in both the owner-occupier and investor/renter sectors. Section 4 analyses the efficiency effects. 2 Housing Taxes and Subsidies in Australia Table 1 shows the main taxes and subsidies relating to housing in Australia. The taxes include local government rates and developer charges that may be regarded as user payments, which is an issue discussed further below. The subsidies are divided into tax concessions and public expenditure subsidies. As will be discussed, some tax concessions are economy-wide concessions, but they have a particularly strong impact on the housing sector. Before quantifying these measures, two general points should be noted. First, the notion and size of a tax concession depends critically on the definition of an appropriate tax base or benchmark. For example, to quantify capital gains tax concessions, most commentators compare the revenue in the concessionary case (nil for owner-occupiers) with the potential revenue from a tax on all nominal capital gains. Arguably, a preferred benchmark would be revenue from a tax on real capital gains. Similar issues arise with estimates of the magnitudes of other tax concessions. Second, the effects of tax concessions depend on various factors such as marginal income tax rates and inflation rates. The higher are marginal tax rates, the greater are the effects of the concessions. Also the effects of fiscal measures like capital gains tax concessions and negative gearing depend on the rate of inflation. 2

3 Table 1 Tax or subsidy Tax concessions Non-taxation of imputed rents Non-taxation of rental services Concessional capital gains taxes Tax treatment of losses on rental property (negative gearing) Public expenditure subsidies Major Australian taxes and subsidies affecting housing Main features Imputed rental income is not assessed as part of taxable income. There is no consumption tax on imputed or actual housing rents. (a) Home owner-occupiers pay no capital gains tax. (b) Property investors pay tax on half the nominal capital gains. Investors can deduct 100 per cent of nominal losses from rental property against other taxable income. First home owner grants Grants of $7000 to first home owner purchasers since 1 July Assistance to private renters Assistance to public housing tenants Main taxes on land and housing State land taxes Local government land taxes Stamp duties on transfers of land and housing GST on home maintenance and renovations, land sales and new buildings Infrastructure developer charges Australian Government provides rental subsidies to low income persons, who comprise about one quarter of all private renters. State governments provide public housing to 5 per cent of all Australian households - in most cases at below market rents. States taxes on value of land used for rental properties and second homes. Some land taxes on premium value owner-occupied property. Local governments levy land taxes (rates) on most residential properties. Most state governments levy stamp duty on value of property when it is transferred. 10% GST applies to (a) maintenance and renovation expenditure for existing housing and (b) sales of land and new buildings. Most state and local governments levy infrastructure charges on developers. Tax provisions that assist housing Four tax provisions substantially assist housing. They are the exclusion of imputed rents from the income tax base, the exclusion of all housing rents from any expenditure tax base, the exemption of owner-occupied housing from the capital gains tax (CGT), and the asymmetric income tax treatment of losses and gains by investors in rental housing. The exclusion of imputed rents from the income tax base is part of the general exclusion of imputed rents from consumer durables, such as motor vehicles, from the income tax base. However, housing is overwhelmingly the most important exclusion. Moreover, the rental income of investor landlords is taxed. Both Wang et al. (2004) and the Productivity Commission (2004) estimate that the non-taxation of imputed rents is worth a net $8 billion per annum to homeowners. 1 Wang et al. provide more detail. For 2001, they estimated that, with household equity averaging 81 per cent and a marginal tax rate on average income of 31.5 per cent, the non-taxation of net imputed rent (gross imputed rent less housing expenses) provided an after-tax benefit of $13 billion to owner-occupier households. Partially offsetting this is the non-deductibility of mortgage interest payments, which costs homeowners an estimated $5.0 billion after tax. The net result of $8.0 billion is sensitive to the marginal tax rate and gearing rate. Wang et al (2004) estimate that in 2000, when the marginal tax rate on 1 This Wang et al estimate applies to 2001; the Productivity Commission estimate to

4 average income was 35.5 per cent, the value to homeowners of the imputed rent exemption after tax was $10 billion. Of course, if net imputed rents were taxed, the benefit of equity financed housing would fall and households would likely reduce their equity in housing. Thus the value of the tax concession exceeds the revenue that government would collect by abolishing the concession. Secondly, existing housing services are not taxed as part of the consumption tax base. There is no GST or other tax on imputed or actual rents. Given an estimated gross annual rental of $91 billion on Australian housing stock (see Table 2 below), a 10 per cent GST would raise tax revenues of $9.1 billion per annum (assuming no change in the quantity, price or rental of the housing stock). 2 However, GST is charged on new houses and on inputs to house renovations and maintenance. If GST were charged on rents, presumably these taxes, valued below at $6.0 billion per annum, would be deductible. The net cost of excluding existing housing services from the tax base is in the order of $3.1 billion per annum. Turning to taxation of capital gains, home owners are exempt from CGT whereas, since September 1999, most investors in other assets pay CGT on half the nominal capital gains when the gains are realised. The Productivity Commission (2004) estimated that the value of this concession to homeowners was $10 billion in This result is sensitive to annual house price inflation, the marginal tax rate, and the benchmark. Wang et al (2004, Table A2) estimate that, between 1990 and 2001, the value of the CGT concession to homeowners compared with a CGT on 50 per cent of nominal gains (the present CGT) varied from $1 billion to $13 billion per annum and averaged $8.8 billion per annum in 2001 prices. Whether investors in rental housing receive a tax concession from the present CGT depends on the benchmark. Investors receive a concession (along with owners of most other forms of capital) if all nominal capital gains should be taxed. But taxation of nominal changes in value is hard to justify. There is a widespread view that the 50 per cent CGT on realised nominal capital gains is concessionary compared with the 100 per cent tax on realised real capital gains that it replaced. However, this is true only when the inflation rate (π) is less than the real rate of increase in house prices (r). If π > r, which it usually is (Abelson and Chung, 2004), the new CGT regime increases capital gains taxes. In practice, the current CGT is generally not a tax concession compared with a CGT on real gains. 3 The issue is further complicated by the treatment of borrowing costs. Australian income tax provisions allow investors generally to deduct losses from one activity, inclusive of borrowing costs, against income from another activity in the same period. When the losses arise from borrowing costs, the practice is known as negative gearing. In the rental property 2 Yates (2000) estimated that the subsidy from the non-taxation of rental services from all housing was about one-third higher than the subsidy from non-taxation of imputed rents. Neither Wang et al (2004) nor the Productivity Commission (2004) discuss this tax expenditure. 3 Suppose that the real percentage increase in house prices is r, the inflation rate isπ, and the capital gains tax rate is t. The nominal gain (n) equals (1+r) (1+π) -1. Thus n can be approximated by r+π. If the CGT is on real gains, the tax payable is P h rt, where P h is the price of the house at the start of the period and t is the tax rate. If the CGT is on half the nominal gains, the tax payable is approximately P h (r+π) 0.5t. If r =π, then tr = 0.5t ( r+ π). If r >π, the new CGT reduces tax payable. If r <π, the new CGT increases tax payable. Wang et al (2004, Table A.2) show that between 1990 and 2001 tax would have been lower with the CGT discount method introduced in 1999 than they were with the previous CGT based on taxes of real gains. 4

5 market it is common practice to deduct losses against other taxable income in the same period. Of the 17 per cent of taxpayers who report rental income, nearly half report losses on rental investment and in many years reported losses from rental property exceed reported incomes (Productivity Commission, 2004). Because negative gearing is an economy-wide practice, the Australian Treasury does not consider negative gearing to be a tax expenditure and the Productivity Commission (ibid) did not estimate an associated tax concession. However the combination of allowing investors to deduct 100 per cent of the cost of borrowing (whether making a loss of not) while taxing only 50 per cent of the capital gains is a substantial tax concession. Suppose that an investor borrows $100,000 at 7 per cent, the real rate of return on investment is 4 per cent, annual inflation is 3 per cent which is translated into a nominal increase in capital value, and the investor s marginal tax rate is 40 per cent. The investor records an operating loss of $3000 and reduces his tax payment by $1200. On the other hand, if he realises the nominal capital gain of $3000, he pays tax on only half that gain. The net gain is $600 or 0.6 per cent of the capital value. If he defers the capital gain, the present value of the gain is greater. The asymmetrical treatment of nominal costs and gains provides investors with an annual tax concession that equals (borrowing for rental investment the inflation rate the marginal tax rate) less the present value of the tax on realised nominal capital gains. Allowing for an estimated capital value of rental stock of $597 billion (see Table 2) and average investor gearing of 40 per cent, borrowing by rental investors is currently about $239 billion. Working with the same parameters as before, the tax concession is worth $1.43 billion (0.006 $239 billion). The present value of the gain will be larger in so far as the gain is deferred and the investor s discount rate exceeds the inflation rate. These benefits to investors arise from the general asymmetric treatment of nominal gains and costs and not just because investors can deduct losses against other taxable income. 4 Although the concession is an economy-wide one, households access the concession predominantly via investment in rental housing. A similar asymmetry occurs with depreciation allowances. Investors can deduct construction costs on new buildings at a rate of 2.5 per cent a year. When deductions for capital works are claimed for income tax purposes, the CGT cost base is reduced. However, investors can claim 100 per cent deduction from taxable income for depreciation and pay CGT on only half of this amount and later. Again there is an effective tax concession although a much smaller one than with negative gearing. But again similar concessions exist across the economy. Other housing subsidies Governments support home purchase in several ways, including the first home owner grant (FHOG) scheme, direct lending, deposit assistance, interest rate assistance measures, mortgage relief, and home purchase advisory services (Australian Institute of Health and Welfare, 2005a). Of these schemes, the FHOG has been by far the most substantial form of assistance to home purchase since its inception in July By January 2004, the Australian government had provided $4.3 billion to over 550,000 first home owners (Productivity Commission, 2004), which equates to a subsidy of about $1.2 billion per annum. 4 The estimated value of this tax concession is based on the benchmark assumption that only real income should be taxed and only real costs should be tax deductible. This is quite different from arbitrary policy proposals related to negative gearing that all costs should be tax deductible so long as the investor makes a profit and that no costs contributing to a loss should be tax deductible. 5

6 The Australian Government provides rental assistance directly to about one-quarter of all renters via means tested cash benefits. Wang et al. (2004) report that the Commonwealth provided over $1.8 billion in rent assistance to 943,000 income units in 698,000 households in for an average annual benefit of $2470 per household. 5 Public housing is a third major vehicle of housing assistance. About 5 per cent of households in Australia live in public housing with most receiving a rental subsidy. Drawing on a sample from the Australian Housing Survey (ABS, 2000), Wang et al. (2004) estimate that at end- June 2002, 342,500 households lived in public housing and 302,500 households received a subsidy. Comparing public housing rents with market rents, Wang et al. estimate the subsidy to be worth $1.25 billion per annum or about $4150 per household per annum. However, they note that some data on market rents are based on tenant estimates, which may not be reliable. Taxes on land and housing All States and Territories in Australia, except the Northern Territory, levy taxes on the value of land used for rental investments. The tax base is the value of improved land, which includes improvements produced by adding capital to land. Thus the land tax is also a tax on capital, albeit a much smaller tax on capital than is a full property tax. Land tax rates are usually progressive with marginal rates on high value property at around 5 per cent. Land tax is generally not imposed on owner-occupied residences though there are exceptions for second homes and for high priced properties. In 2004, land tax yielded state and territory jurisdictions about $3 billion per annum of which about 60 per cent was tax on residential land. On the other hand, the Productivity Commission (2004) estimated that the value of the home owner exemption from land taxes on their principal homes was $7 billion per annum. Evidently the land taxes can be viewed from two different perspectives. From the perspective of the tax system as a whole, given that other forms of household wealth are not taxed, land taxes are an impost on wealth held in the form of land. Alternatively, the exemption of home owners from land taxes can be regarded as a tax subsidy. The former more general position in adopted in this paper. In addition, most local governments in Australia raise rates via a tax on (improved) land values. These taxes are usually proportional to land value, but some jurisdictions levy a minimum amount per property regardless of land value. Nineteenth century British economists, such as Cannan and Marshall, tended to view local government rates as a benefit tax because the payments reflected services mainly to property owners. A more modern view, emanating from the United States and associated with the Tiebout hypothesis, is that the value of local services and taxes are capitalised into property values. This implies that households choose and pay for the local service/tax package that they want. It follows that even though local rates are a compulsory payment, they may be viewed as a user fee rather than as a tax. However, these perspectives have little current relevance to most Australian households. Most local services today are provided to people rather than to property. Moreover, services and local taxes do not vary a great deal across local governments within the cities. There is little 5 The Australian Government provides separately about another $80 million a year in various forms of private rental assistance, such as interest free loans for bonds, help with temporary problems and relocation expenses (Australian Institute of Health and Welfare, 2005b). 6

7 connection between local government rates paid and services received and no marginal relationship. In short, most Australian households would regard local rates as a tax to fund services rather than as a payment for particular services. In 2004, local government in Australia raised about $8.0 billion in land taxes. For the reasons just given, in this analysis these payments are treated as a tax on land rather than a payment for services. However investors can deduct land taxes, whether levied by the states or local jurisdictions, from their taxable rental income. Thus the after-tax cost of land taxes (LT) to investors is LT (1-mtr), where mtr is the marginal investor tax rate which we assume is 40 per cent in this analysis. All States and Territories traditionally charged stamp duty on transfers of land and housing. The tax base is the purchase price of the land or home. In 2003, stamp duty rates varied by jurisdiction from one per cent of sale value for low-priced properties to over 5 per cent of sale value for high-price properties. 6 Most jurisdictions provide stamp duty relief for first home owners. The NSW Government also levies a 2.25% vendor tax on sale of housing. Total revenue from stamp duty on all property related transactions rose from $4 billion in the mid s to over $10 billion in the early 2000 s. Of this, about $9 billion was collected on conveyancing of residential properties (Productivity Commission, 2004). 7 These stamp duties are not tax deductible against income. Another fiscal measure affecting housing is the GST. 8 This tax is imposed on vacant land sold by registered enterprises, new dwellings erected on vacant land, renovation expenditures, and inputs to home maintenance. Gross capital formation for private dwellings is currently about $50 billion per annum, inclusive of about $30 billion on new dwellings and $20 billion on renovations. This expenditure attracts GST of about $5.0 billion per annum. In addition, land development and repairs and maintenance each attract GST of about $0.5 billion per annum (Productivity Commission, 2004). Thus total GST on housing is about $6.0 billion per annum. However, as argued below, the GST on land and new dwellings is borne almost entirely by landowners rather than by the consumers (or the building industry). Most state and local governments in Australia levy developers of new housing for public provision of related economic and social infrastructure. Economic infrastructure includes water, sewerage, drainage, roads, public transport and other utilities. Social infrastructure includes open spaces and parks, libraries, community centres and sports facilities. The charges are mostly for services external to the development site and may serve several subdivisions, for example trunk water mains or major roads. Within sub-divisions, the developer organises and pays directly for most facilities. Development charges vary greatly with jurisdiction (Housing Industry Association, 2003). Charges are usually much higher for houses on the urban fringe than for units in established areas. Allowing an average state and local government charge of $ per new house, and new houses in a year, developer charges total some $2.0 billion per annum. Although the charges are levied as a capital charge on developers rather than as a recurrent charge on 6 The Housing Industry Association (2003, p.63) provides details of stamp duty rates by jurisdiction. 7 These figures do not include stamp duties on financial instruments (mortgages) associated with property purchase. 8 The 10% GST applies to about 55% of purchases in Australia. Thus GST is part of the general tax system rather than a specific tax on housing. 7

8 homeowners, they are generally designed to reflect costs and intended as payments for services provided. Housing subsidies and taxes: a summary To place the value of housing subsidies and taxes in perspective, Table 2 provides estimated capital and annual values of the housing stock in2004, by owner-occupied and investor/rental sectors. The estimates of housing stock are based on 2001 Census figures (Kennedy and Robertson, 2003) plus an average net increase of 100,000 dwellings per annum to 2004 that allows for demolitions. The same report shows that 70 per cent of private housing stock is in home ownership. Wang et al (2004) report that there are nearly 390,000 publicly rented dwellings. 9 Our estimates of average capital values are based on average house and unit prices from Abelson and Chung (2004) along with 2001 Census data showing that owner-occupiers occupy 81 per cent of separate houses and 30 per cent of units (with the balance rented). Our rental values are based gross rentals at 4 per cent of capital values (Reserve Bank, 2003). Table 2 Estimated capital and annual values of housing in 2004 Unit Owneroccupied housing a Private rental housing All private housing b Public rental housing No. of dwellings No mn Average (mean) value $ 300, , , ,000 Capital value $bn Rental value 4% $bn Investment in housing p.a. $bn Na Na 50 c Na (a) (b) (c) These include dwellings that are not occupied or rented, eg. second homes. Excluding caravans, cabins and houseboats. A trend figure that includes expenditure on new houses and alterations and additions. It does not include land development and home maintenance costs that total about another $10 billion. The estimated capital value of the private housing stock in mid-2004 was $2272 billion, including $1674 billion in owner-occupied housing and $598 billion in private rental housing. Annual investment in housing (including land development, building, renovation and maintenance) is about $60 billion per annum. Allowing a 4 per cent gross rental rate, the gross rental value of all private housing is $91 billion per annum, including $67 billion in imputed rent for owner-occupied housing and $24 billion in private rental housing. Table 3 summarises the estimated after tax value of housing subsidies and taxes per annum. As emphasised, the estimates depend on definitions of a benchmark tax regime. For example, the size of the negative gearing concession is estimated relative to a symmetric tax treatment of real gains and losses. Table 3 includes estimates of GST foregone by the lack of a consumption tax on imputed or actual rents, but this is offset to a large extent by GST paid on capital improvements and maintenance. Local government rates are included because they are viewed here as a tax rather than a user charge. Developer infrastructure charges are not included because they are primarily a user charge. 9 As Wang et al (2004) note, administration figures for public housing are higher than their survey figures. 8

9 Subsidies and taxes are estimated separately for homeowners and to investors/renters. Where attribution is not clear, the subsidy or tax is attributed approximately proportional to the value of the homeowner or rental stock. But I have used some judgement. For example I estimate that homeowners/landowners would bear a slightly greater proportion of GST on new homes than on renovations. For all private housing, estimated subsidies total $30.4 billion per annum and taxes total $23.4 billion per annum, making a net subsidy of $7.0 billion per annum. This net subsidy is equivalent to 7.7 per cent of estimated gross annual rental value of housing and just under about 0.9 per cent of GDP. The subsidy accrues almost entirely to the homeowner sector, which inclusive of landowners receives an estimated net subsidy of $6.9 billion per annum. If, as argued below, original landowners pay the $3.0 billion GST on new homes, homeowners receive a subsidy of $9.9 billion per annum. Allowing for second homes, this is equivalent to about $2000 per homeowner per annum. Investors and renters of private rental properties jointly receive an estimated net subsidy of only $0.1 billion per annum. This rises to $0.6 billion if GST on new homes is borne by the landowner, which equates to an estimated net subsidy of $250 per rental property per annum. 10 In addition, public housing tenants receive a subsidy of $1.25 billion, which translates to just over $4000 per subsidised household. Table 3 Estimated average value (after tax) of subsidies and taxes per annum Subsidy / tax Owner-occupied housing Private rental housing All private housing a Subsidies $bn % annual housing value $bn % annual housing value $bn % annual housing value Imputed rent tax concession Na Na No GST on imputed / actual rents Capital gains tax concession Na Na Asymmetric tax treatment of losses and gains Na Na First home owner grant Na Na Private rent subsidies Na Na Total subsidies Taxes Land taxes (state governments) Land taxes (local government) Na Na 1.2 b b Stamp duties GST on land / new houses GST on renovations and house maintenance c Total taxes Subsidies - Taxes (a) Excluding caravans, cabins and houseboats. (b) Estimated land tax payable by investors is $2.0 billion per annum. After Australian Government tax deduction, net tax payable is $1.2 million per annum. (c) A small part of this may be allowed as a tax deduction. 10 Because investor supply of rental housing is competitive, most of this subsidy accrues to renters. 9

10 3 Fiscal Measures and the Housing Market We now examine the impacts of these fiscal measures in the owner-occupied and rental sectors on house prices and housing consumption in the two sectors and across the market. As usual, outcomes depend on price elasticities of demand and supply for the taxed or subsidised commodity. In lieu of authoritative estimates of Australian house price elasticities, I start by assuming unitary elasticities and then review implications of other assumptions. The assumption of unitary supply price elasticity is probably more contentious than the assumption of unitary demand price elasticity, for which there is some evidence ( ). Much tax incidence literature assumes that the supply price elasticity of commodities is high because capital is mobile and capital markets are competitive. This could apply to housing renovations. It may also apply to the supply of new houses in marginal urban areas where land values for housing are close to values for other uses. However, most new houses in Australia are constructed in or close to major urban areas where land values far exceed land opportunity costs. Here, planning regulations determine the supply of land for new housing and the number of new houses constructed. Planners are influenced only weakly by house prices. The Berger-Thompson and Ellis (2004) study of the Australian housing market suggests that the price elasticity of new housing supply is low. Effects of fiscal measures affecting owner occupied housing Equilibrium in the market for owner-occupied houses requires that owners earn the same marginal rate of return on their housing investment as on other possible investments. This equilibrium relationship can be expressed as: R h / P h = ρ = i + m + t h - π h (1) where R h is the value of the rental services per period in owner-occupied homes, P h the price of existing houses, ρ the nominal after-tax rate of return on non-housing investment, i the weighted nominal rate of interest after tax, m house maintenance cost as a percentage of house price, t h the tax on housing as a percentage of house cost, and π h is nominal rate of house price inflation (which is not taxed for owner-occupiers). If housing is subsidised, t h is negative. In this equation, R h is determined by household demand in conjunction with the housing stock. All the right-hand side variables are exogenous. Note that i = (w.mr) + (1-w)ρ, where w denotes the proportion of debt in the house and mr the mortgage rate. Note also that tax is expressed here as a percentage of house price rather than of gross rental. 11 Thus the price of housing has to adjust to bring the owner-occupier (investor) market into equilibrium with other assets. This implies: P h = R h / (i + m + t h - π h ) (2) Holding housing constant, taxes on owner-occupiers directly reduce house prices. Conversely, subsidies increase prices. 11 Equation (1) can be expressed in terms of consumption instead of investment: as R h = (i + m + t h - π h ) P h. In this case, the real value of consumption in any period must equal the user cost of the asset. 10

11 For our purposes, it is more convenient to express Equation (2) in terms of changes in net rent: P h = (R h - T h ) / (i + m - π h ) (3) where T h is the amount of tax paid. If a subsidy or tax is permanent, and the owner-occupied housing stock is constant, house prices rise/fall pro rata with the proportion of subsidies/taxes to imputed gross rental values. A 10 per cent subsidy would raise house prices by 10 per cent. We now introduce interaction with rental housing and total housing supply. These effects will likely occur over one or two years rather than instantly. The relationship between the owneroccupied and rental market is complicated because houses are heterogeneous and owneroccupiers tend to own houses whereas most renters occupy apartments. However, to illustrate the impacts of subsidies for homeowners, houses are assumed to be equivalent units. Also, for ease of illustration, we take the current price/quantity relationship to be the initial equilibrium and estimate the impact of changes from this equilibrium. Thus Figure 1 shows an initial equilibrium in homeowner and rental markets with a house price of $300,000. As in Table 3, there are 5.58 million owner-occupied dwellings and 2.39 million private renter dwellings. 12 A homeowner subsidy of 10 per cent of gross rent raises the demand curve of homeowners from D o to D os and owner-occupier dwelling prices would rise by 10 per cent to $330,000. The renter demand curve for housing is assumed to be independent and unchanged. It slopes down because lower house prices imply lower rents and increased housing consumption. At a market price of $330,000 and with a demand price elasticity of -1.0, rental household consumption would fall by 10 per cent to 2.15 million housing units. However, this would result in 0.24 million unoccupied housing units. With unitary demand elasticity also in the owner-occupied market and no increase in overall supply, the new equilibrium market price is $321,000 per housing unit. Owner-occupier dwellings would increase from 5.58 million to 5.74 million and renter dwellings fall from 2.39 million to 2.23 million. Figure 1 Effects of 10 per cent subsidy to home owners on home owner and rental markets with fixed total supply of housing 12 These numbers include unoccupied dwellings. 11

12 Figure 2 introduces a housing supply response. It shows the total market demand curve for owner-occupiers and renters (D o + D R ) along with the market supply curve. The higher demand curve (D os + D R ) allows for a 10 per cent subsidy to homeowners. With unitary demand and supply elasticities, the new equilibrium price is $310,320. The housing stock increases to 8.24 million dwellings, including 5.93 owner-occupied dwellings and 2.31 investor owned dwellings. In this new equilibrium, both housing consumption and house prices rise by 3.4 per cent. However house ownership has risen by 6.3 per cent and rental dwellings have fallen by 3.3 per cent. The above estimates are based on solving for the following four equations where the β coefficient is the demand or supply price elasticity respectively. Initial owner housing demand: lnh od1 = 7.42 βlnp (4) Subsidised owner housing demand a : lnh 0d2 = 7.52 βlnp (5) Rental housing demand: lnh rd = 6.58 βlnp (6) Housing supply: lnh s = βlnp (7) (a) Allowing for a 10 per cent homeowner subsidy. Figure 2 Effects of subsidy to home owners with unit elasticity supply response Table 4 shows estimated market outcomes with various homeowner subsidies and demand and supply elasticities. For a given homeowner subsidy and supply price elasticity, the greater is the price elasticity of demand, the greater is the increase in owner-occupied dwellings and the higher the rise in prices. Holding demand price elasticity constant, a high supply elasticity increases housing consumption but reduces the rise in house prices. It should also be noted that, in percentage terms, estimated changes in housing consumption and house prices are not very sensitive to adoption of a different initial equilibrium price. 12

13 Table 4 Market outcomes from fiscal measures for home owners Market assumptions Owner subsidy / tax (%) a Demand elasticity Supply elasticity Market outcomes New house price ($) % change in house price Total housing (mn) Owner housing (mn) Rental housing (mn) % change: all housing % change owner housing % change rental housing (a) A + denotes a subsidy; is a tax. Effects of fiscal measures affecting investor / rental housing The effects of fiscal measures in the rental market depend on whether the subsidy or tax is targeted on investors or renters. Investors face a similar asset choice to homeowners. In equilibrium the return from housing must equal the return on other assets. Because they operate with different tax rules, the values of the variables may be different for investors, but the principle is the same. Figure 3 depicts a shift in the investor supply curve from S to S t as a result of a new 10 per cent tax on rental incomes. Given that homeowners own 70 per cent of dwellings and an assumed price elasticity of demand in homeowner and rental markets of -1, the equilibrium price for houses would fall by 3 per cent to $291,000. To recover the 10 per cent increase in costs, investors raise rents by 7 per cent and pay 3 per cent less for the properties. The rise in rentals reduces rental property consumption by 7.2 per cent, so rental properties fall from 2.39 million to 2.22 million. On the other hand, owner-occupied dwellings increase by 3.1 per cent from 5.58 million to 5.75 million. Note that this assumes no contraction of housing supply. Figure 3 Effects of 10 per cent tax on rental income with fixed housing supply 13

14 On the other hand, a 10 per cent reduction in taxes on investors in rental properties would lead initially to a 3 per cent increase in property prices, which would in turn induce some increase in supply. With unitary demand and supply price elasticities, the estimated final equilibrium price would be $304,460. Rental property consumption would rise by 8.4 per cent and owner-occupied housing would fall by 1.5 per cent. Table 5 Market outcomes from fiscal measures for investors Market assumptions Owner subsidy / tax (%) a Demand elasticity Supply elasticity Market outcomes New house price ($) % change in house price Total housing (mn) Owner housing (mn) Rental housing (mn) % change: all housing % change owner housing % change rental housing (a) A + denotes a subsidy; is a tax. Turning to the impact of monetary subsidies for renter households, the key point here is that the average rent subsidy of $2500 per annum per household ($50 per week) is less than the rent that most households pay. In these circumstances the rent subsidy is effectively an untied income cash benefit. Suppose, as shown in Figure 4, that a low income household has a disposable annual income of $16,000 from which it allocates one quarter to housing ($4000) and the rest to other goods. Even if a grant of say $2400 were tied to housing, the household could consume the same amount of housing as it would without the grant and consume an extra $2400 of other goods. With a unitary income elasticity of demand for housing, the household would increase its housing consumption by $600 per annum (or by just $12 per week). This scenario is depicted in Figure 4, where a utility maximising household moves from E 1 to E 2 as a result of the outward shift in the budget line and expenditure on housing rises by 15 per cent (in proportion with the increase in disposable income) from $4000 per annum to $4600 per annum. If all rent subsidies are in effect untied general grants and households spend a quarter of the subsidy on housing, a subsidy of $1.8 billion to private renters increases expenditure on private rental housing by $0.45 billion, which is only 1.9 per cent of estimated gross private rentals of $24 billion. Assuming unitary demand and supply elasticities, the estimated equilibrium house price across the market would rise by only 0.28 per cent from $300,000 to $300,850 and there would be a similar rise in gross rents. Renter dwellings would rise from 2.39 million to 2.43 million and there would be a very small fall in owner-occupier dwellings from 5.58 million to 5.56 million dwellings. 14

15 Expenditure on general goods ($ 000) E 2 E 1 I 2 I Expenditure on housing ($ 000) Figure 4 Housing expenditure with a private rental subsidy of $2400 per annum Effects of fiscal measures on new houses The Housing Industry Association (HIA, 2004) contends that new home buyers are inappropriately facing massive bills for upfront contributions to social and community infrastructure as well as GST and stamp duties and that the taxation of housing is impacting severely on housing affordability. The implication is that all such taxes on new houses are passed forward to the consumer. This would occur only if the demand for new housing were perfectly inelastic and supply perfectly elastic. Market conditions for new housing are almost the exact opposite of this. The demand for new houses is almost perfectly price elastic because existing houses are a very close substitute for new houses. Moreover, in any period sales of existing houses usually far outnumber sales of new houses. On the other hand, the supply of new houses is determined principally by planning regulation and is highly price inelastic. This is illustrated in Figure 5. The imposition of a tax on new housing would shift the supply curve only marginally, if at all, to the left. Accordingly, there would be would be minimal, if any, increase in the price of new houses. In these market conditions, nearly all taxes are passed backwards to housing suppliers, or more specifically to original land owners, rather than forwards to consumers. The HIA (2003) reports that infrastructure charges, GST and stamp duties range from $50,000 to over $100,000 per new house in But it also reports that the price of a typical greenfield building site ranged from $25,000 in Perth to $60,000 in Melbourne, $67,500 in Brisbane, and $110,000 in Sydney. These non-urban land values indicate that market forces have little effect on the supply of new houses. If housing supply responded to house prices, greenfield prices for housing would be a fraction of these values. 15

16 $ S t S P 1 P 2 D Q 2 Q 1 New houses Figure 5 The effect of a tax on new houses In a regulated market for new houses, taxes and subsidies for development affect raw land values, not house prices. Taxes reduce the price that developers are willing to pay for the land; subsidies increase it. Taxes and subsidies affect new house prices only when building land prices are close to opportunity cost and fiscal measures affect the supply of housing. 4 Fiscal Measures and Economic Efficiency There has long been concern that housing subsidies cause excess resources to be allocated to housing and too few resources to be allocated to allegedly more productive activities. In a major review of the topic, Henderson and Bourassa (1992) argued that that there were significant incentives to over invest in owner-occupied housing, but did not quantify the alleged distortion. The Reserve Bank (2003. p.5) expressed concern that tax incentives were causing excess investment in rental housing: When we observe the results (of current tax arrangements) resources and finance are being disproportionately channelled into this area. On the other hand, in reviewing various quantitative estimates of the welfare effects of potential excess investment in housing in relation to GDP, Yates (2000, p.515) concluded that there is little evidence to suggest that any efficiency costs (from housing tax expenditures) are large. A second efficiency concern is the use of the housing stock. Transaction taxes such as stamp duties affect not only amount of housing but also how it is used. In addition, unequal fiscal treatment between owned and rented property may distort the tenure choice decision. I examine below the possible deadweight costs of housing subsidies and taxes, inclusive of some important second best considerations. However, the paper does not attempt to estimate the value of possible positive externalities of house ownership or occupation. 16

17 Efficiency costs and gains from fiscal measures affecting investment in housing To estimate the efficiency implications of investment in housing, the paper adopts the classic public finance / cost-benefit approach and examines what consumers are willing to pay for housing compared with the real opportunity costs of the resources used to provide housing. Figure 6 exemplifies this approach for a competitive housing market, where the supply curve is also a marginal cost schedule. Subsidies to housing shift the demand curve for new housing from D to D s. Housing supplied increases from Q 1 to Q 2 and the price for housing rises from P 1 to P s. Housing producers gain the benefit of area P s BCP 1. However, the net price to house purchasers is only P c. Housing consumers gain the surplus area P 1 CFP c. Government pays the difference between P s and P c, which amounts to a total payment of area P s BFP C. The difference between government (taxpayer) subsidies and the sum of producer and consumer surpluses is the area BCF, which is the deadweight loss. $ S P S P 1 P C A C E F B D S deadweight loss D Q 1 Q 2 New housing per period Figure 6 Effects of a housing subsidy on housing quantity and price with competitive housing and land markets As shown in the public finance literature 13, this deadweight loss (DWL) is estimated by DWL = 0.5t 2 Q h P h / ((1/η d ) +(1/η s )) (8) Where t = the marginal tax (or subsidy) rate Q h = the number of new dwellings constructed with no subsidy P h = the initial price of housing η d = the demand price elasticity for housing η s = the supply price elasticity for housing If Q h equals 150,000 new dwellings per annum, P h is $300,000, the subsidy is 10 per cent, and there are unitary demand and supply elasticities, the deadweight loss from construction of new housing would be $225 million per annum. 13 For derivation of Equation (8), see for example Rosen (2002, p.292) 17

18 DWL NH =[0.5 (0.10) 2 $300, ,000 houses] / 2 = $225 million per annum The deadweight loss would rise to $450 million if the supply of new houses were perfectly price elastic. On the other hand, the loss tends to zero as the supply price elasticity for new houses tends to zero. However, we three other significant issues need to be considered. The first is expenditure on housing renovations. Second, the above estimates assume that the average initial house price of $300,000 represents the real opportunity cost of resources used in housing. Third, the estimates make no allowance for the deadweight cost of raising tax revenue. As will be seen, the second issue is fundamental and dealing with it radically affects the outcome. Annual capital expenditure on housing is about $55 billion per annum, which includes about $20 billion on renovations, $30 billion on construction of new housing and $5 billion on land development. To estimate the deadweight cost of renovations in the absence of convenient housing units, we substitute expenditure on housing renovation for Q h P h in Equation (8). The supply price elasticity of renovations is likely to be high and is assumed here to be infinite. Assuming again a subsidy of 10 per cent and a demand price elasticity of 1, the deadweight cost associated with expenditure on renovations is: DWL R = 0.5 (0.10) 2 $20 billion = $100 million per annum Returning to construction of new houses, what is the real opportunity cost? Obviously costs vary greatly. However, the widespread difference between the land price for housing and the opportunity cost of land indicates that the real opportunity cost of housing supply is generally below the price of housing. The following costs per housing are fairly typical. Private land development expenditure applies mainly to about 100,000 new houses on the urban fringe and is about $50,000 per house. However, some housing requires little land development, and the overall average cost per housing unit is assumed here to be $40,000. Public infrastructure costs are also higher on the urban fringe than for developments in established areas. These costs may average out at about $20,000 per housing unit. Building cost is assumed to be $150,000 per unit. Allowing $10,000 for the opportunity cost of raw land, the total land and building package is $220,000. Add $40,000 for management and sales, and the total opportunity cost of a house would average around $260,000 compared with the price of $300,000. The social surplus would be $40,000 per dwelling. In a study of a development of over 10,000 houses in south-west Sydney, I estimated that after allowing for all public and private costs there was a social surplus of about $20,000 per new dwelling unit. This was in an area of moderately priced houses and estimated costs included expensive public transport and negative externalities from private motor vehicles. Evidently, the supply curve in Figure 6 is not a marginal cost curve. Figure 7 shows the housing market with a marginal cost curve below the market price reflecting the regulation of new housing. In this case, a housing subsidy increases investment in housing from Q 1 to Q 2 and there is an efficiency gain equal to area ABCD. This is the difference between the price that consumers are willing to pay for housing and the real cost of providing the housing. 18