How Taxes are Sending Housing Prices Through the Roof

This year looks set to repeat the national hand-wringing over housing affordability. We have been here before, most recently in 2016 and 2017. Then there was a torrent of angst about Australia’s house prices. This is not to deny that there is a real problem. However, most of the solutions proposed to deal with it are spurious or counterproductive.

In 2020 housing prices in Australia briefly and slightly dipped. As Covid lockdown restrictions were lifted, the housing market rebounded vigorously. There were a number of reasons for that: cheap money driven by low interest rates, high-debt to low-income loans, low-deposit borrowing, government home-buying incentive schemes, real estate and newspaper boosterism, and the re-routing of billions of dollars in savings from hospitality and travel into house purchases. What followed was a series of predictions in March 2021 that house prices would increase by 20 per cent over 2021 and 2022. Such predictions often fail. Nonetheless these ones at least reflected the escalating house prices of early 2021. Like night following day, angst over housing affordability followed quickly behind the reheated market. Higher housing prices in real terms mean that those entering the housing market find it increasingly difficult to afford deposits for homes. It also takes entrants into the housing market longer to pay off the asset they have bought. In addition, it means increasing rents in the major cities or else pressure on renters to move to poorly serviced outer-urban areas with weaker job markets or longer commutes to work.

In Australia a large gap exists between upward pressures on housing prices and the multiple mock solutions proposed to deal with these pressures. The recipes that are put forward are principally designed to get governments through the next election. They never address the major underlying structural problems of Australia’s housing market. Indeed, in many ways they can’t address those problems, when Australian governments at all levels are at least one of the major causes of so many of the problems in what has become, in the last thirty-five years, one of the most expensive housing markets in the world. Before 1985 it was inexpensive. Since then something has gone badly wrong.

Historically, Australia’s multiple of “median house price to median income” was in the twos and threes:[1] it took somewhere between two and four years of total household income to pay off a house. The Melbourne median “household price to household income” ratio in 1980 was three; by 1990 it was four; by 2000 it was five;[2] by 2004 it had catapulted to 6.9; by 2019 it was a precipitous 9.5.[3] Today Melbourne residents who are willing to relocate can find marginally cheaper housing outside the big metropolis in regional cities. Yet even there, the multiples are high:[4] Ballarat is 6.0, Geelong 7.4. Even further afield, Bendigo is 5.4.

Between 1987 and 2017, the cost of consumer goods and services dropped substantially in the world’s advanced economies. Against this propitious current, the prices of three big-ticket items rose sharply. Two of those were publicly-subsidised goods: education and health.[5] The third was housing. In Australia the cost of renting increased 57 per cent and mortgage-holding 43 per cent in real terms between 1994 and 2011, and 51 per cent and 40 per cent respectively between 1997 and 2017.[6] The number of bedrooms in a dwelling, an indicator of size, rose modestly from 3 to 3.2 between 1997 and 2017.The proportion of income spent by mortgagees and private renters on housing in 2011 and 2017 was almost the same as in 1994 and 1997 while real income per week had increased 50 per cent over the respective periods.[7] These trends meant, in effect, that increased housing costs absorbed the benefits of real income growth.

When the term “housing costs” is used, we naturally think of all the land and building costs that went into constructing a house. But in the case of Australia a remarkably high percentage of housing costs is not functional in nature but rather the result of taxation. Among OECD nations Australia relies heavily on property taxes.[8] In a detailed 2011 report, Taxation of the Housing Sector, the Centre for International Economics (CIE) concluded that housing was one of the most heavily taxed economic sectors in Australia.[9] It contributes 11 to 12 per cent of the total tax collected by all tiers of government. The Property Council of Australia estimates that, in 2013-14, 16 per cent of government taxes, fees and charges in Australia were related to property as a whole and that as much as 26 per cent of the cost of a finished house and 21 per cent of a finished apartment was due to these government imposts.[10] Today around half a million dwellings sell each year in Australia. About 40 per cent of these are new constructions. Heavily taxed and charged by government, those annual new builds add annually 2 per cent to the country’s total residential properties (around 10 million of them in 2018). In the financial year 2018-19 the mean price of a dwelling was $661,400.[11] The combined value of these new properties was around $132 billion. $67 billion in taxes and charges were levied on new and old housing in the same financial year.[12]

Taxes on dwellings include capital gains tax on investment properties, GST, stamp duty, land tax and council rates plus indirect taxes (such as insurance taxes, income taxes, fuel taxes, payroll taxes and import duties) on direct and intermediate inputs to dwelling construction as well as disguised taxes such as infrastructure charges, usage charges, permit and application fees, training and long service levies, document charges, zoning restrictions, planning delays and uncertainties, and exorbitant building code standards.[13] As well as being high, many of the taxes imposed on Australian dwellings are also expensive to administer.

The CIE calculated that the median tax impost on housing in Australia in 2011 represented 44 per cent of the purchase price of a new house in Sydney, 38 per cent in Melbourne and 36 per cent in Brisbane.[14] The tax component of new apartment costs was 35 per cent in Sydney, 33 per cent in Melbourne and 34 per cent in Brisbane. The tax share of infill residential development was 22 per cent in Sydney, 22 per cent in Melbourne and 29 per cent in Brisbane.[15] For a young couple on an average wage in Sydney, where the housing multiple in 2011 was over 12, the additional cost of a new home that was attributable to taxes amounted to a third of the couple’s after-tax income.[16] There is no sign that anything fundamental has changed since then.

In short, housing affordability in Australia is a problem to a significant extent created by government taxation. As housing costs rise, governments experience pressure from voters to “fix” the problem. But because taxation is the basis of government revenue, the remedies that governments offer invariably fall short. Either they don’t work or else they make the problem worse. Indeed governments have a vested interest in not fixing the problem. In 2011 new housing represented 1.2 per cent of the value added annually to the Australian economy but 2.8 per cent of government tax revenues.[17] Existing dwellings contributed 8.4 per cent of all government revenue while accounting for 7 per cent of the value added to the economy. Residential property taxes and charges represented over 40 per cent of state and local government taxation revenue.

Of Australia’s big industry sectors—defined as value-adding more than $10 billion annually—the CIE study calculated that the residential building sector was the second most heavily taxed, at 30.9 per cent of the value of output.[18] The sector at the time was the second-largest contributor of tax to Australian governments.[19] It furnished 12 per cent of all government revenues.[20] Only the wholesale and retail trade sector contributed a higher portion, 13 per cent. The next largest tax-contributing sector was transport, at 7.5 per cent.[21]

The US urbanist Wendell Cox has argued that the high-tax residential construction regime is a consequence of government regulation aimed at “urban containment”.[22] Government planners, he suggests, are obsessed with rules to limit or prohibit development on the metropolitan urban fringe. This translates into a reluctance by government to release land for development. This in turn creates an artificial scarcity of land that drives up land prices. This rule-driven price-hiking is effectively a tax on house owning. As population grows so does the demand for dwellings. Eventually pressure builds for land releases but the delays in these releases cause artificially high prices. This price escalation adds to both government revenue and housing costs. The eventual release of exurban land is inevitable but the delays in doing so cost time and money. The longer the application and approval processes, the higher the cost of the land released.

While that is a recognisable dynamic it is not clear that the preoccupation of government planners with “urban containment” is the primary or only reason for escalating house costs in Australia since the 1980s. It is true that the departure from affordable historical housing multiples closely follows the rise of bureaucratic environmentalism. Wide use of its signature word sustainability begins in the late 1970s and early 1980s. This usage rises steeply in the decades after that.[23] Bureaucratic environmentalism in practice means dragging out development approvals with time-consuming procedures. These processes expand government power and revenue. The effect of slowing down land-release approvals is to increase the gap between an escalating demand for houses and the supply of housing.[24] That said, though, the CIE report calculated that such processes represent only around 10 per cent of the total greenfield-housing cost in Melbourne. That leaves a further 28 per cent still attributable to government. Planning bureaucracy is only one part of the generic government appetite for tax revenue.[25]

If 30 to 40 per cent of the cost of a new detached house in Sydney and Melbourne is a function of direct and indirect government taxes and charges, how then do we account for the other 60 to 70 per cent of the cost? Are escalating real costs perhaps the result of the Australian taste for large dwellings? The average size of a new Australian home in 1984 was 162 square metres. In 2013 it was 214 square meters, a rise of 32 per cent.[26] But that does not explain the more than doubling of Australian housing multiples over the same period.[27] This is especially so as the logic of modern industrial societies is governed by the economic maxim that “less is more”. Rising productivity in principle over time should deliver more space for less cost. Australian residential construction has not done this. On the contrary, its productivity has failed to keep pace with and offset the consumer appetite for additional floor space let alone reduce real costs overall.

Industry productivity is principally a function of industry innovation. In Australia and globally there has been a long-term problem with construction productivity.[28] In 2017 the management consultancy firm McKinsey reported that since 1945 in the United States, productivity in agriculture, manufacturing and retail had grown by a factor of 1500 per cent.[29] Productivity growth on such a scale fuelled the country’s substantial long-term rise in standard of living. In contrast American construction productivity had barely moved since the end of the Second World War. According to McKinsey, the principal cause of this was the “specialised building trades” (such as carpenters, tilers, plasterers and plumbers). Their construction techniques, power tools withstanding, are essentially pre-industrial.[30] Innovation is not reserved to new building. A key component of Australian housing innovation is renovation. In calendar year 2019 the total value of residential construction was $101 billion of which 35 per cent was the value of renovations.[31]

America today continues to benefit from its traditional use of timber rather than masonry in house construction. This method is inexpensive and does not require costly high-skill labour. But compared with the era of the nineteenth-century balloon frame, innovation in US house construction in the past seventy years has been less than life-changing. The same is true internationally. Global labour-productivity growth in construction averaged 1 per cent a year between the mid-1990s and the mid-2010s. This contrasted over the same period with a 2.8 per cent per annum productivity growth in the world economy and 3.6 per cent in manufacturing. In Australia, between 1995 and 2005 the real gross value added per work hour in construction was minus 0.9 per cent.[32]

In Australia it is not just the self-employed specialised building trades (what the Australian Bureau of Statistics calls “construction services”) that is a low-productivity cohort.[33] The union-dominated “building construction” sector is even worse. The big Australian civil engineering enterprises working on infrastructure projects are pretty efficient. In 2018 they generated $279,791 gross value added per worker.[34] In contrast, enterprises engaged in domestic and non-domestic “building construction” along with the “construction services” supporting them (concreters, bricklayers, roofers, electricians, air-conditioning and fire-alarm installers) generated $102,684 and $108,642 in value added per worker respectively.[35] The larger part of the Australian construction industry, including most of its workforce, is not that far ahead of the chronically low-productivity education ($86,855) and health care ($76,655) sectors. The value added by builders is less than public administration ($120,155).

In construction, Australia is far from the worst performer in the world. Nevertheless its listless record of construction productivity fuels long-term rising house prices. The domestic construction sector globally is backward technologically and Australia is no exception. The construction industry has been slow to embrace information technologies and digital management tools to improve price estimating, project scheduling and procurement. There has been limited adoption of high-productivity auto-industrial technologies such as computer-controlled off-site manufacturing of housing as well as more recent developments such as 3D printing of buildings, robotic bricklaying and tiling, and the like.[36] There appears little appetite to pursue measures such as improved supply chain management, principles of lean construction, and automated defect management systems.

In addition, the hefty regulation of residential construction contributes to lethargic innovation. In the United States for example it is estimated that the construction industry is subject to seven times the number of laws that relate to agriculture and mining.[37] Heavily regulated sectors tend not to proceed by proactive widely dispersed self-organising experimentation with new technologies. Rather they behave much like government departments do. They wait impassively for slow-paced changes to industry rules and codes. High productivity and corresponding low prices are a function of lean regulation that is outcome-focused rather than fixated on bureaucratic procedures.[38] Major regulatory bottlenecks in construction occur in the delay-ridden permitting, approvals, zoning and public procurement processes.[39]

So why then do Australian governments not simply reduce the volume of regulation and taxation applied to housing? Because these two are the flip-side of revenue raising. The appetite for revenue—more than any other factor today—controls the behaviour of governments. This was not always the case. From the mid-1960s to the late 1970s government spending as a share of GDP in Australia was relatively moderate. It hovered around 26 per cent of GDP. Estimates of the optimal level of government spending range from 17 per cent to 30 per cent.[40] It is notable that the housing multiple rose as Australia trekked out of the optimal range of public expenditure. Spending ballooned to 36 per cent of GDP.[41] As government expenditure expanded as a portion of GDP, so did the level of tax. Housing is a soft target for ever-higher taxation and fee-generating regulation. Combined with poor construction industry innovation, the consequence of that, since the mid-1980s, is that Australia’s median “house price to household income” multiple has risen far beyond the historical norm.

In a sense Australians today are not paying for housing when they buy a dwelling. Rather they are funding the public sector. The high priority of investing in a house obliquely subsidises what in many cases are low-priority public-sector functions. This is a consequence of a historical development. Over the long term, as productivity in agriculture and manufacturing industries increased, employment shifted to services and the public sector. This post-industrial trend gathered pace after 1970. As it did, there was huge pressure on government to expand the public sector—a sector that is peculiarly resistant to efficiency gains.[42] What followed was the increasing appetite of governments for revenue to pay for a significantly enlarged and inherently low-productivity public sector. Notionally “limited government” administrations were able to resist the pressure from lobbies for more public spending for brief periods of time only. The price paid for this (literally) over the long run was the myriad tax and fee imposts on housing. Combined with a low-productivity residential construction sector, the effect was to dramatically drive up the housing multiple, especially in Australia’s metropolises.

In 2018 the median house price in Sydney was $950,000 and in Melbourne, $720,000, housing multiples of 10.4 and 7.9 respectively.[43] Australia in 2020 had three cities (Melbourne, Sydney and Adelaide) among the thirteen least affordable cities of ninety-two of the major housing markets in the world. Brisbane and Perth were not far behind. Only Hong Kong and Vancouver were worse than Sydney, and Auckland and Toronto worse than Melbourne. This is a terrible record. Possibly Australia’s greatest achievement has been its cities. They are highly liveable places and excellent crucibles for work and enterprise. However, as is often the case, this upside now has a downside. Other attractive and economically robust cities in the world have significantly lower housing multiples, including Atlanta (3.6) and Houston (4.0).[44]

If Australia’s major cities were to drop back to Singapore’s 4.7 multiple or Montreal’s 5.6, the reduction would make a striking difference to the standard of living of middle-income and low-income Australians. Australia’s historic norm was as low as one third of the current prevailing housing multiple, depending on city.[45] The old norm was comparable with contemporary Houston in Texas (4.0) and Minneapolis-St Paul in Minnesota (3.9). Both of these are major metropolises in high-performing state economies, one a “red state”, the other a “blue state”.[46] In Texas’s case the low multiple is a function of low state taxes and cheap land. Texas does not bank land. Building permits are swiftly approved.[47] Affordable housing is common in low-tax countries like Singapore and low-tax, lightly-regulated American states such as Florida, Texas, Georgia, Tennessee and Arizona, though as the case of Minnesota suggests, taxation and regulation are not the only determining factors.[48] Industry efficiency is also key.

One of the principal obstacles to more affordable housing in Australia is the emphatic wish of Australian governments not to reduce taxation or fee-generating regulation. In 2016, in the midst of one of the recurring panics in Australia about housing affordability, the then federal Treasurer, Scott Morrison, announced that he was discussing with the states “the potential to remove residential land use planning regulations that unnecessarily impede housing supply and are not in the broader public interest”.[49] The equivocal phrasing of this—with its hesitant terms and multiple qualifications—immediately pointed to its unlikely outcome. It was not that the diagnosis was wrong but that revenue-hungry state and local governments are unwilling to give up the bountiful fees and charges that regulation affords them. Rules are revenue, and revenue rules contemporary government thinking.

Government has a priority—revenue—that it pursues relentlessly. But in doing so it also curiously reflects public opinion. Opinion is ambivalent about house prices. When prices go up, opinion frets about the difficulties of young people purchasing their first home. At the same time opinion salivates at the thought that the value of the principal asset that most people own will rise. The ultra-low interest-rate policies of central bankers in the past two decades have underscored this view. Those policies have been cynically enacted to reduce the cost of government borrowing. The flow-on effect of super-low interest rates has been to reduce the cost, not of housing, but of housing loans. The ready availability of cheap loans smoothed the way for structurally-driven higher asset prices.

The other effect of abnormally low interest rates by historical standards has been to make bank savings an unattractive asset, pushing those savings into investments in other assets including houses to own or rent. The desire to see an increase in the nominal book value of houses to quell anxieties about low interest rates on savings has been a psychological driver of higher housing prices. Once individuals have purchased a home, their anxieties shift from the question of whether they can afford to buy a home to the hope and wish that the value of their property will rise over time so that they will eventually reap a substantial capital gain or at least feel their assets have grown, feelings that even term deposits in banks no longer satisfy. Via this social psychological route, established home-owners become a tacit lobby for high house prices, even if this is usually offset by the fear that if they sell their home for a substantially increased price, they may be forced to buy again for an even more forbidding price.

The mixed sentiment of the Australian buying public backed by central bank policy and cynical government convenience, compounded by boosterish media organisations that have a vested advertising interest in booming real estate markets, and allied with a lethargic low-performing building industry, has driven up the Australian ratio of house prices to household income to among the highest in the world. Paradoxically everyone benefits from this except society as a whole. The visible individual benefit of the high housing multiples to numerous interest groups and lobbies is matched by the shared detriment of a nominal rising standard of living that is consumed by inflated property values that can never be fully realised except perhaps at the end of life. Australians have become measurably wealthier in the past forty years. But often people say that it doesn’t feel like it. To the extent that this is true it is because the increase in the book-value of wealth has been offset by the increasing number of years of household income that are needed to pay off an average home.

Thus arises a Catch-22. Public opinion swings between demanding affordability and hoping for the most unaffordable prices possible. How is this circle squared? The answer is: by various sleights-of-hand. In no other area of Australian public policy do we see as many crank, pernicious, self-defeating and pointless policies devised as those meant to address the unaffordable prices of Australian homes. This problem seems to bring out both the most naive and the most deeply cynical forces in Australian politics.

Let’s consider the long list of panaceas for affordability that have been spruiked in the past few years. The worst of these—and it’s a competitive list—is the idea that young people should be allowed to draw down their superannuation retirement savings in order to pay for a house. The effect of this would be to run down one asset to buy another asset. It would have no effect on the cost of houses—on how efficiently they are built or on how much tax is levied on them. However, it would pump more money into the housing market, driving up prices still further. A similar paradoxical effect of government “help” is evident in the case of government “first home buyer” grant schemes.[50] It is widely understood that the only effect of these schemes, beyond that of aiding a government’s re-election chances, is to push up housing prices as more money enters the market to compete for a given supply of houses.[51] Then there is the idea of government “co-investing” with first home buyers struggling to afford a home. The same problem applies. More money in the housing market fuels higher prices rather than lower costs. Such schemes make housing less not more affordable.

It is astonishing the number of flaky schemes that get attention. There is pressure nationally to replace up-front taxes on the conveyancing of a property by scaling up annual state land taxes. Dubious claims are made about the resulting tax “savings”. Take the example of stamp duty of $31,000 levied upfront on a $600,000 median-priced house purchase in Victoria in 2015. On average Australian dwellings are sold once every eleven years. [52] The same house transacted 3.6 times in forty years represents in constant dollars a tax take of $111,600. The attraction of land tax to a purchaser is that it eliminates the big up-front levy on house conveyancing. The less-visible negative though is that while purchasers can borrow to pay the costs of stamp duty they need ready cash to pay their annual land tax. The professional services firm Deloitte calculated that to replace stamp duty in a “revenue-neutral” way Victoria would require an average land tax on owner-occupied properties of $2644 a year in 2015 prices or $105,760 over forty years.[53] Over a lifetime an expanded land tax would rearrange the tax burden on housing but would be unlikely to reduce it.

As well as the tax shell game, there is the unoccupied house flim-flam. Australia today is routinely subjected to moral panics instigated by the political class. A recent one, in 2016 and 2017, was the unoccupied house or unoccupied room panic. This was the political response to the last round of “affordability crisis” stories. The “cause” of unaffordable houses was a newly-coined social misdemeanour—owners sitting on unoccupied buildings without tenants or householders (with adult children who’d left home) occupying houses with empty bedrooms. Yes, empty bedrooms were driving the housing “crisis”. The “solution” to the “crisis” was to provide tax incentives for older Australians to downsize their properties and to legislate tax penalties for owners of unoccupied dwellings.[54] In 2017 Victoria introduced an annual tax on specified vacant residential properties.[55] In comparable countries over time a small and mostly constant percentage of dwellings are unoccupied.[56] Occupancy rates are indifferent to government interventions but sensitive to fluctuating rental yields in inner-city locales.[57] Any vacancy tax can be offset against higher resale prices and higher rental yields when market conditions are right.

False causalities afflict the housing debate. In recent decades Australia has had one of the highest rates of inward migration in the world. This has led to nativist claims that migration is a cause of high housing multiples.[58] But it is not. Over time there has been no correlation between metropolitan housing multiples and Australia’s net migration rate.[59] The universals of tax and technology are much more important to market dynamics than a fraction of buyers. Even in the peak migration decade of 2006 to 2015 the net migrant household component of current house sales was a less than dramatic fifth of the total.[60] The massive drop in immigration in 2020 has made no difference to escalating 2021 house prices.  

Two more ersatz “solutions” are popular. One is funding more public housing. Victoria, again leading the way, recently unveiled a public housing building program, spending $5.4 billion over four years on 12,000 social and affordable housing units at $450,000 a unit, the price of a median home in Melbourne in 2009. Not exactly a world-beater in the reduction of construction industry costs and a drop in the ocean of the 2.5 million homes in Victoria and the 175,000 of those that come on the market every year. Over the past two decades costs for public housing tenants have risen 52 per cent in real terms compared to 40 per cent for persons with a mortgage.[61] Mortgages are much more cost efficient. But even mortgage holders have had little effect on the twin pressures of high property taxation and low construction industry innovation in Australia. A reduction of the housing multiple in Melbourne to Montreal’s 5.6 would represent a massive improvement for low-income home buyers.

The last in this long cavalcade of non-solutions is relocation to regional Australia. The single most failed policy in the past fifty years in Australia must be regional decentralisation. I have no objection to urban devolution except in Australia it simply does not work. The nation is principally a land of large coastal cities. It always has been so and today more than ever. Some regional cities in Australia have significantly lower housing multiples than Australia’s capital cities.[62] Townsville is 3.9 and Darwin 3.8. But smaller cities and regions in commuting distance of the big metropolises have high multiples. The Sunshine Coast is 8.4 and the Gold Coast 8.0. Australia’s distinctive and vast coastal conurbations are much more powerful than any marginal regional dispersions. It’s the way we are.

Supposing all that, will housing affordability change? Given the desire of government to expand its revenue base and the lethargy of Australia’s building industrial complex, it seems unlikely. Housing in recent decades has done a lot for the productivity of bogus solutions. But collectively Australia continues to climb the mountain of high housing prices. Maybe in the not-too-distant future we will beat Hong Kong to the number one spot as the country in the world with the least affordable housing.

Peter Murphy is the author of Limited Government (2019) and The Political Economy of Prosperity (2020). A footnoted version of this article is available at Quadrant Online

Notes

[1]              If the constant quality real price of housing in 1880 was 100, in 1955 it was 100 and in 1980 it was 150. In 2010 it was over 400. Philip Soos, Table, Australian Constant Quality Real Housing Price Index 1880-2012, 2013 https://www.macrobusiness.com.au/2013/02/the-history-of-australian-property-values/ based on data from Australian Bureau of Statistics and N. Stapledon, “Trends and Cycles in Sydney and Melbourne House Prices from 1880 to 2011”, Australian Economic History Review 52(3), 2012, pp. 293–317.

[2]              R. Fox and R. Finlay, “Dwelling Prices and Household Income”, Reserve Bank of Australia Bulletin, December Quarter 2012, Graph 4.

[3]              W. Cox and H. Pavletich, Demographia International housing affordability survey, 2020, 2005.

[4]              W. Cox and H. Pavletich, Demographia International housing affordability survey, 2020.

[5]              Peter Murphy, Auto-industrialism: DIY Capitalism and the Rise of the Auto-Industrial Society, London, Sage, 2017, pp. 75-97.

[6]              Australian Bureau of Statistics, 4130.0 – Housing Occupancy and Costs, 2011-12, 2017-2018.

[7]              Australian Bureau of Statistics, 6523.0 Household Income and Wealth, Australia, 2013–14, Table 1.2.

[8]              The OECD defines taxes on property as “recurrent and non-recurrent taxes on the use, ownership or transfer of property. These include taxes on immovable property or net wealth, taxes on the change of ownership of property through inheritance or gift and taxes on financial and capital transactions.”  On average across the OECD in 2018 property taxes account for 1.8 percent of GDP. (OECD data, Tax on property, 2021. https://data.oecd.org/tax/tax-on-property.htm.) At 2.7 percent of GDP Australia sits well above average. Australia’s level of property taxation is not the highest in the OECD. That dubious honour falls to the United Kingdom at 4.1 percent of GDP, helping explain the UK’s long-running problem with housing affordability that stretches back at least to the 1950s. Property taxes in Sweden by contrast represent 0.9 percent of GDP. Generic property taxes are not exclusive to dwellings just as the cost of dwellings may be subject to all manners of fees, charges, duties and indirect taxes that are not property taxes. Nonetheless the level of property taxation is a rough indicator of where a nation’s tax burdens falls relative to other countries. In Australia, among Australian states almost as much tax revenue is collected annually from the generic category of property taxes as from taxes on good services and more than from the general category of payroll and income taxes. (Australian Bureau of Statistics, 55060DO001_201819 Taxation Revenue, Australia, 2018-19, Table 10 Taxation, Total All States State and Local Government). Also worthy of note, the revenue value of the generic category of property taxation in Australia rose from 1.9 percent of GDP in 1982 to 2.7 percent in 1988. (OECD data, Tax on property, 2021. https://data.oecd.org/tax/tax-on-property.htm) This rise paralleled the period in the mid-1980s when Australia’s housing multiple began to escalate above its historic level of between 2 and 3.

[9]              Centre for International Economics (CIE), Taxation of the Housing Sector, Canberra, CIE, 2011.

[10]             The Property Council of Australia, Fairer Taxes, March 2021.

[11]             Australian Bureau of Statistics, Residential Dwellings: Values, Mean Price and Number by State and Territories.

[12]             This assumes the CIE estimate of 12 percent of the total tax take of $559 billion in 2018-19 (ABS, Taxation Revenue, Australia, 2018-19 financial year).

[13]             “The Victorian planning authority has identified over 600 separate approval decisions for a new house in Melbourne. And that excludes the all-important strangulation of the first stage planning permission, the ‘release’ of land to allow it to be built upon.” A. Moran, ‘Australian housing’s regulatory price boost not about to end’, Catallaxy Files, 23 January, 2017.

[14]             CIE, 2011, Table 3.3, pp. 33, 36.

[15]             CIE, p. 36.

[16]             CIE, p. 53.

[17]             CIE, p. 59.

[18]             CIE, pp. 9, 61, 64.

[19]             CIE, p. 9.

[20]             CIE, pp. 7, 59, 64.

[21]             CIE, p. 65.

[22]             W. Cox, A Question of Values: Middle Income Housing Affordability and Urban Containment Policies, Winnipeg, Frontier Center for Public Policy, 2015.

[23]             Google N-Gram charts the steep escalation in the use of the word over time.

[24]             Cremins sums up the view of John Freebairn on how delayed supply affects house prices: “The supply of housing takes time to respond to an increase in demand. Often there are lags of several years to gain approval for new housing, to obtain finance, for the associated provision of infrastructure, and construction time. If there is an unanticipated jump in housing demand, with new housing supply restricted, most of the demand increase flows into higher prices. Only after some years to allow for supply to respond will the rate of price increase slow or fall.” N. Cremins, “House prices outpacing income growth”, Exchange Edition 1, Parkville, University of Melbourne, August 2015.

[25]             CIE, op. cit., Tables 3.12 and 4.1.

[26]             L. Wilson, “How big is a house? Average house size by country”, Renew Economy, 17 July 2013; E. Sorensen, “Why are our houses getting bigger?”, realestate.com.au, 10 December 2013.

[27]             R. Fox and R. Finlay, “Dwelling Prices and Household Income”, Reserve Bank of Australia Bulletin, December Quarter 2012, Graph 1.

[28]             The Housing Industry Association in 2016 in its submission to the Productivity Commission’s 5 year Productivity Review estimated that a 1 percent increase in productivity across the construction sector would be worth $2.36 billion of additional GDP a year. The HIA’s productivity solutions focus on workplace bargaining, independent contracting and streamlined regulation yet notably ignore industry technology, the great driver of modern growth economies.

[29]             McKinsey Global Institute, Reinventing Construction: A Route to Higher Productivity, February 2017.

[30]             McKinsey, p.4 and Exhibit E3.

[31]             HIA, Window into Housing, 2020. Residential construction in 2019 represented 5.3 percent of Australia’s GDP.

[32]             McKinsey, Exhibit 10.

[33]             Australian Bureau of Statistics, 1292.0 – Australian and New Zealand Standard Industrial Classification (ANZSIC), 2006 (Revision 1.0).

[34]             The Australian Bureau of Statistics “Heavy and civil engineering construction” and “Building construction” categories. D. Richardson, Productivity in the construction industry, Canberra ACT, The Australia Institute, August 2014, Table 3.

[35]             Australian Bureau of Statistics, 5204.0 Australian System of National Accounts, Table 5. Gross Value Added (GVA) by Industry; Australian Bureau of Statistics, 6291.0.55.003 Labour Force, Australia, Detailed, Quarterly, Table 06. Employed persons by Industry sub-division of main job (ANZSIC) and Sex.                   

[36]             Richardson, pp. 8-9.

[37]             Richardson, Exhibit 24.

[38]             Richardson, p. 45. It is not just the volume of regulations but rather “the confusing and arduous bureaucratic processes through which regulation is administered cause delays and compromise coordination among owners, construction firms, and regulators.”

[39]             Richardson, ibid.

[40]             This is exclusive of war-time periods when government spending naturally escalates. L. Di Matteo, Measuring Government in the Twenty-first Century, Toronto, The Fraser Institute, 2013; D. Chobanov and A. Mladenova, A. What is the Optimum Size of Government, Bulgaria, Institute for Market Economics, 2009; V. Tanzi, “The economic role of the state in the 21st century”, Cato Journal, 25: 3, 2005; R. Vedder and L. Gallaway, Government Size and Economic Growth, Washington DC, Joint Economic Committee, 1998; G. Scully, What is the Optimal Size of Government in the US, NCPA Policy Report No. 188, Dallas, National Center for Policy Analysis, 1994; E. Peden, “Productivity in the United States and its relationship to government activity: an analysis of 57 years, 1929–1986”, Public Choice 69,153-173, 1991.

[41]             J. Novak, Australia’s Big Government by the numbers, Melbourne, IPA, 2013, Figure 7.

[42]             W. D. Eggers and J. Jaffe, “Gov on the go: Boosting public sector productivity by going mobile”, Deloitte University Press, 19 February, 2013, Figures 1 (US) and 2 (UK); H. Simpson, “Spending cuts and public sector productivity”, CMPO Viewpoint, October 18, 2010; UK Office for National Statistics, “Public Service Productivity Estimates: Education 2013”, www.ons.gov.uk, Figure 1; Australian Government Productivity Commission, Annual Report on Government Services, www.pc.gov.au, 1995-2017.

[43]             Australian Bureau of Statistics, 6416.0 Residential Property Price Indexes: Eight Capital Cities, Tables 4 and 5; ABS, Household financial resources.

[44]             W. Cox, Demographia International housing affordability survey, 2021.

[45]             Asset-owners, having acquired an asset whose value is 30-40 percent taxation, presumably would not want to see a low-tax version of the same asset-class swamping the market. But as tax modification, technology innovation and asset creation are medium and long-term phenomena, market adaption invariably would outweigh short-term market shock.

[46]             Minnesota number 16 and Texas number 4, determined by a composite of weekly wages, house prices, GDP per capita, GDP growth, etc. (A. Kiersz, “Ranked: The economies of all 50 US states and DC from worst to best”, Business Insider, 3 August, 2015.)  Greater Houston has a population of 6.7 million, Minneapolis-St. Paul 3.5 million, Melbourne 4.3 million.

[47]             Texas has quite high annual property taxes. Like Australia’s land taxes, these are not part of the purchase price of a new house (in contrast to Australia’s stamp duties) but they do condition expectations of the re-sale price of a dwelling.

[48]             W. Cox and H. Pavletich, Demographia International housing affordability survey, 2017, Schedule 3; Cato Institute, Freedom in the 50 states index, www.freedominthe50states.org; United States Census Bureau, State Government Tax Collections 2015 and Annual Estimates of the Resident Population for the United States, Regions, States, and Puerto Rico: April 1, 2010 to July 1, 2016. Like all rules, there are exceptions to this rule. Many Florida cities are inexpensive but Miami is not. Take also the case of the high-tax, high-regulation state of Minnesota. It has an unusual history of organic multi-polar urbanism (A. Atkins, “St. Paul, Minneapolis, and Minnesota’s Urban Origins”, www.mnopedia.org). The bi-polar Minneapolis-St. Paul conurbation (a 3.9 housing multiple in 2020) reflects this. The twin cities in turn have spontaneously developed a cost-reducing hub-and-spoke pattern with multiple small satellite cities within commuting distance of the bi-polis, affording small-city house prices and big-city employment.

[49]             J. Massola, “Scott Morrison puts states on notice over house prices”, Campbelltown Macarthur Advertiser, 24 October 2016.

[50]             A parallel in the rental housing market is the federal government’s National Rental Affordability Scheme (NRAS), a program that provides financial incentives for large-scale housing investors to rent housing to low and middle income tenants at a 20 percent discount. Introduced in 2008, the scheme was begun to be phased out from the 2014-15 budget.

[51]             The long-term effects of government house subsidies in Australia bear this out. The home-ownership rate in Australia reached its highest level (72 percent) in 1961. That was three years before the first home-buyer subsidy scheme was introduced by the Menzies government in 1964.

[52]             CIE, p. 51. This is consistent with the evidence from R. Wilkins, D. Warren and M. Hahn, Families, Incomes and Jobs Volume 4, Parkville, Melbourne Institute of Applied Economic and Social Research, 2009, Table 28.4, indicating that mortgagees and outright house owners move three times in forty years.

[53]             0.65 percent on the unimproved value of land owned. Deloitte Access Economics, The revenue raising potential of a broad-based land tax, Barton ACT, Deloitte, March 2016, Table 3.3.

[54]             Then Treasurer Scott Morrison in 2016 suggested to the states a number of downsizing incentives including waiving land tax or stamp duty for retirees who downsized. With substantial political difficulty in 2016, the federal Treasurer managed to reduce the cut-off threshold for part-pensions back its historic norm. Less than a year later he was thinking about easing the pension means-test rules to encourage retired Australians to sell their family homes. The reasoning was that cash proceeds from the sale of a large family home were liable to reduce part-pension eligibility under the Treasurer’s more stringent pension rules. The tougher rules supposedly discouraged down-sizing house sales.

[55]             The annual tax was set at 1 percent of the capital improved value (CIV) of taxable land. That meant an annual tax of $5,000 on a vacant home with a CIV of $500,000. Vacancy was defined as not occupied for six months by the owner, permitted occupier, or tenant.

[56]             The Prosper surveys based on annual household water-usage data, concluded that 4.4 percent of Melbourne residential properties in 2012 were unoccupied, 4.3 percent in 2013 and 4.8 percent in 2014.* Between 1991 and 2011, Melbourne’s unoccupied properties as a fraction of all residential properties grew a miniscule 3 percent.** Melbourne’s 4 percent range was not unusually high. Zero percent un-occupancy or anything approaching it is a completely impractical prospect. The comparable 2013 figure for Auckland was almost identical to Melbourne’s rate.*** In Paris in 2012 it was 3 percent.**** In the United States, it was 7 percent.# The US vacancy fraction was 5 percent at the 1990 Census and 5 percent again at the 2000 Census at the peak of an economic boom. In 2015 as the US slowly emerged from a property bust and a prolonged period of stagnation it was 8 percent.##

*C. Cashmore, Speculative Vacancies 8 The Empty Properties Ignored By Statistics, Melbourne, Prosper Australia, 2015; C. Cashmore, Speculative Vacancies 7 Empty Investment Homes Ignored, Melbourne, Prosper Australia, 2014; P. Soos and P. Egan, Speculative Vacancies in Melbourne, Melbourne, Prosper Australia, 2013; **”Dwelling type”, profile.id.com.au; *** Auckland Council, Auckland Dwellings and Households: Initial results from the 2013 Census, May 2014. The 2013 New Zealand census figure for unoccupied residences was 6.5 percent. It is necessary to deduct the 25 percent of the owners who of census day are temporarily away. That leaves 4.8 percent of dwellings unoccupied for lengthy periods. ****Approximately 40,000 dwellings out of 1,336,000. S. Davies, “Paris hopes vacant homes can solve housing crisis”, www.france24.com, 19 December, 2012; #US Bureau of Census, Occupancy and Vacant housing units, 2011-2015, American Community Survey 5-Year Estimate. Number of dwellings vacant: 2015: 10,739,000 dwellings out of the total of 133,351,840; 2013: 10,326,971 dwellings out of the total of 132,057,804; 2009: 8,836,836 dwellings out of the total of 127,699,712. Vacant dwellings for this purpose exclude temporarily unoccupied dwellings for sale or sold or for rent or rented. ##US Bureau of Census, Occupancy and Vacant housing units, 2011-2015 American Community Survey 5-Year Estimate; US Census 1990 and 2000. 1990: 5,409,392 dwellings out of the total of 102,263,678; 2000:  5,925,000 dwellings out of the total of 115,904,641; 2009: 8,836,836 out of 127,699,712; 2015: 10,739,000 out of 133,351,840.

[57]             In big metropolises the highest volumes of unoccupied dwellings are concentrated in a narrow band of suburbs. Melbourne’s business districts and student districts, for example, figure prominently (Cashmore, Speculative Vacancies). The first are new apartment building hot-spots; the second are low-rent high-turnover student housing locales. The ultimate factor propelling house vacancy is simple economics. Low occupancy is the obverse of low rental yields. New South Wales data indicates that urban areas that have low rental yields at around 2 percent are 2.5 times more likely to have un-occupied dwellings compared to areas where rental yields are 6 percent or higher.* That’s not surprising. The rental yield (an investor’s annual return from rental income relative to property price) dropped from 4 percent in Melbourne in early 2009 to 2.7 in early 2017.** The correlate of this was a short-term rise in un-occupancy rates in low-yield suburbs.

*L. McKenny and I. Ting, “Thousands of empty homes adding to Sydney’s housing crisis, experts say”, The Sydney Morning Herald, 28 March, 2016, based on data from Laurence Troy and Bill Randolph, UNSW City Futures Research Centre; **M. Janda, “Superannuation for housing deposits would facilitate intergenerational theft”, ABC News, 16 March 2017.

[58]             To punish international investors in 2016 the then New South Wales state Treasurer, Gladys Berejiklian, introduced a foreign investor surcharge on stamp duty and land tax levied on residential real estate.

[59]             Sydney median housing multiples compared to Australia’s net immigration rate per 1000 population trended from 4.1/6.807 in 1985 to 7.2/6.420 (1990), 5.5/4.091 (1995), 9.1/4.928 (2000), 8.8/8.187 (2005), 9.1/10.299 (2010), 9.8/7.713 (2015) and 11.1/5.911 (2020). Demographia International Housing Affordability Survey, 2005, 2010, 2015, 2020; Fox and Finlay, “Dwelling Prices and Household Income”, Reserve Bank of Australia Bulletin, December Quarter 2012, Graph 4; Macrotrend net migration rate data based on United Nations, World Population Prospects.

[60]             In the years 2006-2015, the average net migration to Australia was 216,510 persons per year. Of that number approximately 35 percent were families and 65 percent were skilled migrants. Across the 2005-2014 decade, 81 percent* of the approximately 140,000 net migrant households a year for the period or 1,400,000 households in total would have exited the rental housing market (15 percent* do so in their first year in Australia) to enter the home ownership market of 5,000,000 unit sales for the decade**—in short over a peak net migration decade migrant households made up slightly more than one in five house purchasers. * Siew-Ean Khoo, Peter McDonald, Jeromey Temple and Barbara Edgar, Scoping Study of Migration and Housing Needs, Treasury, 2012. **Estimate based on Paul Bloxham, Daisy McGregor and Ewan Rankin, “Housing Turnover and First-home Buyers”, Bulletin, Reserve Bank of Australia, June Quarter 2010.

[61]             Australian Bureau of Statistics, Housing Occupancy and Costs, 2017-18 financial year.

[62]             W. Cox and H. Pavletich, Demographia International housing affordability survey, 2020.