Both Sydney and Melbourne express pathologies of modern urban planning. Sydney more than Melbourne: it has become a prime case of how to decay a city.
Sydney’s transport systems are a mess of gridlock and inadequacy. Its housing market has become an erratic, monstrously overpriced disaster. It has become a city of crime and social tension: expressed most dramatically in the February 2004 Redfern riot, December 2005 Cronulla riots, the Macquarie Fields “disturbances” of February 2005, a recurrent incidence of drive-by shootings, and an apparently inexorable growth of gang warfare. None of this was remotely inevitable. Melbourne functions better as a city, though it also provides object lessons in bad public policy.
The curse of beauty
Sydney is sited amid famous natural beauty. Sydney Harbour is a series of vistas of breathtaking splendour. This creates Sydney’s first problem, from which so much else flows: it is a city with a geography that divides. It divides in a physical sense: creating valleys and peninsulas that complicate and channel traffic. It also divides in a social sense: living with harbour or beachfront views, or even just on top of hills, is very different from living in the oppressive flatness of the Liverpool plain.
Harbour, beachfront and hilltop views are what economists call positional goods: desirable things it is not possible for everyone to have. It is precisely their “if I have it, you don’t” nature which helps make them so desirable. The holder of a positional good has a strong interest in preserving it. Sydney’s urban geography ensures that the wealthiest, most influential people have a vested interest in highly regulated land-use to protect their views. Nobel Laureate economist Paul Krugman famously divided the USA into “Flatland” and “The Zoned Zone”. In Flatland cities, housing construction responds to demand so housing prices are also relatively flat—the median price of a family home in Houston or Dallas is around three times median household income.[i] In the Zoned Zone, land use is highly controlled and housing prices surge into housing bubbles—many have since burst, but Californian prices are still up to seven times median household income. Krugman writes:
By my rough estimate, slightly under 30 percent of Americans live in the Zoned Zone, which comprises most of the Northeast Corridor, coastal Florida, much of the West Coast and a few other locations.[ii]
Median house prices in the Zoned Zone reached over eleven times median household income in Los Angeles and San Francisco.[iii] In Australia, median house prices range from nine to ten times median household income (Melbourne and Sydney) to four times median income (Mildura), with major metropolitan cities (those with more than one million residents) ranging upwards from six times median income in Perth (it is a triumph of regulation to create land “shortage” in Western Australia). Australia, the developed economy with the lowest population density, has on average the least affordable housing in the Anglosphere.[iv]
The discretion trap
The importance of such zoning in housing prices can be seen in comparing Britain with Germany. In Britain nothing can be built without official permission, and housing prices reach amazing heights, with periodic collapses. In Germany there is a constitutional right to build and house prices move about the rate of inflation (with some drop in the national average after the poor-quality East German stock was included).
A frequent public policy delusion is that discretionary control by officials leads to more even economic outcomes. Nonsense: discretionary control by officials, particularly over market entry, is typically more chaotic than an open market, since feedback mechanisms are narrower and more erratic.[v] Just capture a few key decision-makers—who inevitably operate on a narrower information and issue base than the wider society—and you are away. In the last half-century, Japan and Britain, the most regulation-rationed land markets among the major developed economies, have had far more chaotic housing price experiences than Germany, the least regulation-rationed national housing market in the developed world.
Back in the USA, the Zoned Zone cities are typically waterfront cities—on the coasts and around the Great Lakes: they typically have notable positional goods based on location. As coastal cities, they are also typically “gateway” cities, hence migrant cities. Incoming migrants have two notable features. They are housing market entrants; they are also non-citizens, who cannot vote. If your housing market entrants are disproportionately non-citizens, would that help or hinder the “positional good” push for strong land use controls? There is a startlingly strong correlation between housing affordability in the USA (measured by the ratio of median house prices to median household income) and the proportion of foreign-born residents. There is no correlation with population growth.[vi] Apparently it does not affect housing affordability if many people are trying to move to your city: it does affect housing affordability if they cannot vote.
As Texas—very much part of Flatland—shows (it has lots of migrants), non-voting migrants alone don’t affect housing affordability. It is the combination of real estate positional goods and non-voting migrants that make a city’s entry into the Zoned Zone likely. Moreover, Texas has many Hispanic migrants but already has many Hispanic voters. So its foreign-born migrants typically have networks they can “plug into” relatively easily, which weakens any “non-voter entrant” effect.
Once regulatory land rationing becomes a major factor in local jurisdiction—once you have entered the Zoned Zone—the policy dynamics change in invidious ways. If one restricts the supply of housing for land, that raises the value of land that is already approved for housing. Which benefits people who already own houses. They vote. Discretionary power by officials typically advantages the well-connected and disadvantages those that are not. Inner-city residents benefit most from the increased housing prices of land-rationing and are disproportionately well connected—both politically and in public debate, whose terms they have considerable power to frame. This creates a large vested interest in land-rationing policies, and in asserting and combating the evil of “urban sprawl”.
Regulation then creates a set of new “positional goods”. A classic case is taxi licences. In Melbourne, the cost of a taxi licence is about the same as the median house price.[vii] Rationing of market entry creates a positional good (the taxi licence) and a group (licence holders) with a strong interest in defending their value. The discretionary power of officials that “protects” us from “too many” taxis has similar effects to that of the discretionary power of officials that “protects” us from “too many” houses. (It seems that we don’t need protection from taxi drivers with inadequate English or knowledge of the city, or from rising house prices and rents.) They “protect” us so well from “too many” houses that at least 445,000 lower income households pay 30 per cent or more of their gross household income in rent; around 170,000 Australian households pay more than half their gross household income in rent.[viii]
Note that public housing is not part of the infrastructure of a city: housing is very much a private good, not (as infrastructure typically is) a network. Rental subsidies are generally a much better way of assisting the disadvantaged than public housing. Such subsidies can be delivered quickly, adjust readily to changes in people’s circumstances and reduce the danger of creating dysfunctional local social environments. Nowadays, the real point of public housing in Australia is to provide concentrations of Labor voters. (If public housing residents start voting Green, or if Coalition-voting aspirational Asian, or conservative African, immigrants start to dominate public housing, I predict that the ALP will adopt privatisation of public housing and provision of housing assistance via rental subsidies.) Much of modern welfare is about creating the exchange of tax-paid services for votes: public housing is just a particularly egregious example.
One consequence of strong official discretionary control over land use is the suppression of architectural and design diversity and innovation. Approval tends to be restricted to what planners are used to and comfortable with. With the growth of single-person households and inherent bushfire risks, Australia should be a place for experimentation in co-residency and fire resistant construction. Anyone attempting building innovation in Australia rapidly finds themselves enmeshed in bureaucratic red tape as officials “prove” how “needed” their control is. Bureaucratic narrow-mindedness makes innovation an often painful and expensive process (if a government department wants to construct some modernist monstrosity, however, that will be fine). Even a traditional form of construction, such as using bluestone in Gippsland, can prove “too hard” if it is outside the comfort zones of university-trained council planners[ix] armed with discretionary control over building approval.
Revolutions in transport technology since the second half of the nineteenth century have led to re-conceptions of urban property rights. Trains, trams and cars encouraged the growth of residential-only neighbourhoods. From the 1910s onwards, buses and trucks made industry, commerce and apartments more unpredictably mobile, which led to the creation of zoning to protect the amenity of the residential neighbourhood. (In the USA this was upheld in a key 1926 US Supreme Court decision.)[x] The growth of motorways and freeways in the 1950s and 1960s led to another wave of zoning. Restrictions on the use of property can increase the value of particular attributes of property—such as by increasing its scarcity value or by protecting other desirable characteristics. In Texan housing developments, binding covenants that are part of what you choose to purchase have created a competitive market in local rules.
There is a crucial distinction here. Zoning to protect the amenity of existing residential neighbourhoods is one thing, zoning to control the creation of new residential neighbourhoods quite another. If one takes the attribute suitable for housing and shares control of it between the landowner and officials, it is not hard to see that land-for-housing will be underprovided (that is, supply will not meet demand). Increasing housing land scarcity drives up the value of housing land, leading to more revenue for officials to use (rates, land taxes, stamp duty), more value for the houses of existing homeowners (who vote and like their house values to rise) and more political donations (from developers who need political access to get development approvals). By adopting land-rationing, one has both increased transaction costs (which will mean fewer such transactions) and gives decision-power to those who get little benefit from using individual land blocks for housing and lots of benefits from restricting land use.
Welcome to the real world of regulation. Welcome to the Zoned Zone.
Discretionary disaster (1): house prices
This approach also sets things up to create housing price bubbles (a form of asset bubble). Asset bubbles are based on belief in “one-way” bets: something government, as a risk-suppressing agency, can easily create from its actions. The expectation of rising value—due to land-rationing that impedes supply responding to demand—becomes part of what drives prices apparently ever upward, especially where the long-term experience of constant, if generally low, inflation creates a demand for inflation-proof or—even better—inflation-beating assets. Expectations of a “sure-fire” winning asset can easily overshoot, leading to a collapse of the bubble as people lose confidence that prices will continue to rise.
One of the difficulties in sorting out the deleterious effects of poor regulation is that its effects manifest as activity within markets, and there is typically a chorus of folk willing to blame markets for the problem: a case of blaming the effects for the cause. Comparing jurisdictions allows us to see effects more clearly. Texas has a higher population than Australia, higher population growth, more of its people in its five largest metropolises, higher per capita income, and yet housing in its cities is between a half and a third the price—relative to household income—of housing in Australia’s major cities. Whether officials have discretionary control over land use makes a powerful difference.
Postwar Britain enthusiastically adopted the model of discretionary control over land use. The experience of winning the war, memories of the Depression and the Labour landslide of 1945 combined to legitimise control by beneficent officials. Australian states successively adopted the British approach, with Sydney leading the way in restrictive controls. The “Harbour” effect, combined with Sydney’s increasing domination of postwar migration, was followed by Adelaide, with the “Adelaide Hills” effect.
There is a plausible argument that the effects of land-rationing on various metropolitan areas help explain differences in voting in the November 2007 general election. Cities with significant housing stress had larger swings to Labor: Sydney with a deflation of its bubble and a long history of expensive housing,[xi] Melbourne and Adelaide both with a long history of expensive housing, and Brisbane and south-east Queensland experiencing soaring costs. Perth, in the early stage of a housing boom, and Hobart, experiencing one for the first time, both had smaller swings against the Coalition. Conversely, in the August 2010 federal election, the largest swings against the first-term Labor government were in the areas with the most housing stress—Sydney and south-east Queensland.
If landownership is sufficiently concentrated, and there is some enforcement mechanism to stop members “breaking ranks”, then effective land cartels can operate, restricting supply. This was part of what happened in Ireland. In his recent book on the Irish housing boom and bust, Irish journalist Fintan O’Toole refers to:
certain landowners [who] had accumulated large landbanks at the outskirts of urban areas which they then released in dribs and drabs in order to manipulate the market and artificially to maintain high land prices.[xii]
In Australia we have a name for such people. We call them “state governments”. In the case of VicUrban, the Victorian government’s land commission, the Age has reported that:
Government developer VicUrban is sitting on a further stockpile of 25,000 housing lots listed for development across Melbourne, but selling just over 700 lots a year, or 3 per cent of its stock.[xiii]
The point of such government activity was originally to promote affordable housing: such institutions have made housing much less affordable because revenue and other imperatives have taken over:
The government developer said it released its land on a commercial basis and declined to say how much land was released to the housing market at one time.
‘‘VicUrban’s Act requires it to operate commercially. We release land in a commercial manner that responds to local area needs and market demand,’’ VicUrban chief executive Pru Sanderson said.
We do not need a government body to reproduce commercial behaviour: but such bodies are excellent mechanisms for manipulating land markets. In Ireland, government had no interest in changing the processes that were inflating house prices, since it benefited from increased property tax revenue and the wealth effect for voters of inflated house prices. (Or, so it appeared at the time: now, with its current fiscal problems, not so much.) Alas, the same applies to the control and manipulation of land markets by state and territory governments in Australia: in the ten years after 1995–96, property taxes doubled their share of state and territory revenue from 12 to 20 per cent to 25 to 45 per cent of taxation receipts. Driving up housing costs is very much in the revenue interest of state and territory governments. As economist Leith van Onselen has commented:
Australia’s governments have become Ponzi merchants: attempting to keep the Great Australian Housing Bubble going by pumping demand and restricting supply in order to preserve government finances.[xiv]
But remember, as you or yours struggle with mortgage costs and high rents: they are from the government and they are there to help you.
In Victoria, the new Baillieu government has announced it intends to greatly expand land release in order to drive down land prices and improve housing affordability.[xv] While the intent is to be commended, the problem is with the notion of “land release” itself. Abandoning such discretionary official control over quantity would fundamentally change expectations and drive land prices down much more permanently and reliably. In order to understand the effect of regulatory controls, one has to look at the entire regulatory package: zoning and land releases are obvious quantity controls, but a vast range of regulatory requirements can act as impediments to housing supply.[xvi]
A great deal of value can be “created” by regulatory decisions. The more value at stake in any particular regulatory decision or rule, the more regulations will be “gamed”. So anyone involved in the land market in a major way—developers, real estate agents, and so on—has a vested interest in getting the ear of officials. Such interests thus become a major source of revenue for political parties: buying access through political donations. This occurs well inside the borders of outright corruption.
It is widely accepted that price control, in the form of rent control, can have a very deleterious effect on housing markets. The Swedish economist Assar Lindbeck famously wrote, “Next to bombing, rent control seems to be the most efficient technique so far for destroying cities.”[xvii] Rent control blocks investment in rent-controlled housing by destroying its profitability: in its worst cases, landlords can abandon buildings to minimise their losses.[xviii] The notion that quantity controls are somehow benign or innocuous makes little sense. Just as price controls have quantity effects, so quantity controls have price effects: they block supply responses to demand, driving up prices. For example, in Ireland, local government areas with the highest vacancy rates were the most liberal in releasing land, while the areas with the lowest vacancy rates were the most restrictive. This led to the worst of both worlds, as housing estates were built where they were permitted, rather than where the demand was, while quantity controls in the more desirable areas still drove prices up. The post-bust result is empty estates of unwanted housing.[xix]
A well-developed ideology of planning, which believes in the ability of wise, well-informed planners to somehow juggle all the disparate pressures and come up, on a continuing and systemic basis, with the correct discretionary decisions, helps propagate belief in quantity controls. There is often a fundamental economic illiteracy involved, as a necessary buttress to this self-belief, with little sense of the inevitability of political pressures, the price effect of quantity controls, the role of expectations, the problems of knowledge in an economy or the evidence that controlled markets are more chaotic than open ones.[xx] Little wonder that the university-training of planning officers often arms them with far more doctrine than practical sense or analytical capacity.[xxi]
Bubble and bust
How does all this create booms and busts in housing prices? There are two reasons to buy an asset: for the income stream that it generates and because of expectations of capital gains. If the only value of an asset is its income stream (as is the case with bonds), then that will determine its value.[xxii] With other assets, income streams are subject to much greater variability. If it is expected that some asset will generate increased income, its value will rise. If there is a general expectation that an asset will show capital gain over time, its current value can become largely determined by its expected capital gain, rather than its income return. A capital gain based on a pattern of expectations of capital gains is very real: people can borrow against it, they can realise it by selling the asset. The capital gain is real—until it is not.
Asset bubbles occur where there is a widespread expectation of capital gains and downside risk is discounted. Discretionary control by officials that restricts land use generates continual experience of rising prices of land approved for housing, and thus expectations of capital gains, which then feed into price levels.
Leith van Onselen, in his Unconventional Economist blog, has drawn attention to the surge in housing prices from 1997 onwards:
Notice how land value/GSP [Gross State Product] ratios began rising in 1997—just as the non-bank lenders entered the mortgage market en masse and led the reduction of mortgage lending standards? The reduction of lending standards, supported by the banks’ heavy offshore borrowings and tax-based investment (e.g. negative gearing) was the demand trigger for the Great Australian Housing Bubble. But the supply-side restrictions—including zoning rules, urban growth boundaries, up-front infrastructure charges, and impact fees—ensured that the increased demand fed into rapidly rising prices, in turn encouraging mass speculation by investors and “panic buying” by first home buyers.[xxiii]
If it is easier to make bets, more (and larger) bets will be made.[xxiv] But what made housing look like such a good bet in the first place? The expectations of capital gains created by the regulatory quantity restrictions retarding supply responding to demand.[xxv] General expectations of capital gains are part of the information feeding into the market.[xxvi] Implicitly or explicitly guaranteeing an asset expands the value of that asset and the willingness of economic agents to discount downside risks. A 2008 survey conducted for the Australian Securities and Investment Commission found that one in ten Australians owns one or more investment properties.[xxvii] Since 1990, owner-occupied and investment property credit has expanded its share of total credit from 23 per cent to 58 per cent. (Business credit has dropped from 63 to 34 per cent.)[xxviii] Australians have been taking on large amounts of debt to invest in houses whose prices are largely a product of quantity controls: Australia has become a country highly leveraged on regulatory approval.
At some point, we can expect that the expectations will no longer sustain themselves, and a “tipping point” will be reached. The bubble then bursts as prices drop back towards prices reflecting expected income. But if we knew from available information what will generate that tipping point, that would then feed into expectations about capital gains. Expectations about expectations simply collapse into expectations. If we could reliably predict such tipping points, no one would buy an asset at a price to get “caught”, which would make the tipping point happen earlier in a regress which would stop such price surges happening in the first place. Bubbles occur because we cannot reliably predict such tipping points. (To start, there is the little difficulty of predicting information that does not yet exist.)
But what generates the housing bubble in the first place is the discounting of downside risk due to a pattern of rising prices caused by officials restricting supply so it cannot fully respond to demand. It is not monetary policy that creates the bubble.[xxix] In the USA, housing price bubbles occurred in some jurisdictions (such as California) and not others (such as Texas) under the same monetary policy regime.[xxx] But Australia has become a land where homeowners expect their houses to achieve capital gains, expect house prices to rise, and typically must borrow extensively against expectations of capital gains to enter into this “guaranteed” wealth effect. Hence interest rates (and what the Reserve Bank does) matter in a very direct and obvious way to people—as we can see in the role of interest rates in various federal election campaigns.
That Australian housing prices did not collapse in the lead up to, or during, the recent global financial crisis meant that neither the Australian economy, nor the financial system, nor government revenues, had to deal with the consequences of such a collapse. But this is no guarantee that house prices won’t collapse—at some point.[xxxi] The best policy can do is not create the conditions which foster creation of bubbles—in particular, not encourage systematic discounting of downside risk.
How land markets operate, how housing markets operate, how they are regulated, and how they interact with financial markets, can have profound effects on the life of a city.[xxxii] Evidence from the USA, for example, suggests that housing affordability has a major impact on domestic migration patterns.[xxxiii] Land use regulation can have anti-competitive effects well beyond housing provision, such as for shops, hotels and other commercial services.[xxxiv]
All those networks
Basic to how a city operates are the networks that service them. For infrastructure—roads, trains, trams, buses, water, phones, gas, electricity—is typically a network good, or, as with ports and airports, nodal points in networks. What shapes provision of such network assets matters profoundly for the life of a city. The creation of infrastructure is an act of imagination: of expectations about where people will live and what they will want and need, about what creating (or not creating) infrastructure will do to the dynamics of a city. It is not only about risks and expectations over time, it is also about whose wants and needs count and how much.
As network goods, such infrastructure has an inherent tendency towards monopoly, for it is much cheaper to add an extra user to an existing network than add another network or to add an extra user to a smaller network: an advantage that increases the larger a network is. Infrastructure also typically involves large, static investments that deliver services to lots of people, so have an inherent tendency to attract disproportionate political attention. (In other words, politically they are big, static targets: ask Telstra shareholders how much fun that is.) The typical regulatory response to these two features is to have infrastructure publicly provided or, if privately provided, highly regulated.
Frank Knight’s classic definition of the difference between risk and uncertainty says that (ordinary) risk is where one can put a value on expectations about it (that is, it is calculable) and uncertainty is where you cannot. Taking mathematics as the science of pattern and structure, we can say that ordinary risk is where the expected dangers are sufficiently structured that a pattern of expected risks can be derived from it (so that, even if specific values cannot be calculated, general rankings and ranges can be reasonably derived) and uncertainty is where there is insufficient confidence in knowledge of how the dangers are structured so as to frustrate calculation, even in general terms.
Uncertainty thus makes long-term economic activity (particularly investment) much more problematic. Hence business will often prefer policy clarity—even if the policy is hostile or otherwise problematic—to policy uncertainty, since the former gives some structure within which to calculate likely results from actions over time (particularly investment). This can be another reason to have publicly provided infrastructure: if a constant tendency to political interference generates ongoing policy uncertainty within an industry, internalising the costs of that within the political structure through public provision (and so generalising them across taxpayers as a whole, whose interests are thereby easier to discount) may be preferable to having specific private shareholders bear the brunt.[xxxv]
Infrastructure networks also generate benefits that the provider cannot normally capture. Providing rail lines, roads, piped water, sewerage and so forth raises the value of the surrounding land: an increase in value that a private infrastructure provider normally does not benefit from, but a tax-collecting government can. Infrastructure will tend to be under-provided by private operators, unless they are given some further benefits. In the USA, during the nineteenth century railway boom, railway entrepreneurs were given extensive land grants that they could sell off once their rail line had increased the land’s value.
That governments benefit from increased tax collection is another reason to have infrastructure publicly provided. (Indeed, it is the original reason why rulers first built roads, bridges, ports and so on.) There are, however, notorious and continuing problems with public provision: over-staffing and other inefficiencies, ineffective or inappropriate quality control, long-term tendencies to declining productivity, being conduits for political interests of dubious social benefit,[xxxvi] and so on. Governments also periodically suffer fiscal crises that retard their debt-financing abilities.[xxxvii] For these reasons, there has been a long-term tendency for infrastructure provision to shift back and forth from public to private depending on which set of difficulties—those from public or private provision—have become the more salient at any given time.[xxxviii]
While it is true that prevailing ideas can change, the underlying issues identified above often matter more. One sometimes sees the claim that, in Victoria, the Bolte Liberal government had a much higher rate of state debt than the Cain–Kirner Labor government and so it is only a “shift in the ideology about debt” which led to the fuss about the latter’s debt levels. This is nonsense. The Bolte government used debt to finance useful infrastructure that increased population and economic activity—and therefore revenue. The Cain–Kirner government incurred debt while financing policy disasters. Debt that does not lead to increased revenue is a very different matter from debt that does: there is nothing “ideological” about that except, perhaps, in the policy delusions that lead to such failures.
This is where official discretionary control over land has a particularly invidious effect. It is a lot easier and cheaper to drive up land prices (and thus tax revenues) from quantity controls on land use than by building infrastructure. Such quantity controls thus undermine one of the prime motivations for governments to build infrastructure—increased tax revenues—by providing a much easier alternative.
Debt is appropriate to finance infrastructure, since the benefits accrue over time and to future generations. (Provided, of course, that it does actually provide such benefits.) Requiring local infrastructure to be funded upfront by developers can, however, be a splendid device to increase scarcity value and thus land prices, with all the attendant tax revenue and other political benefits.
Discretionary disaster (2): infrastructure
Using regulation-created scarcity to raise the prices of land-for-housing has a series of invidious effects on infrastructure provision for a city beyond undermining the increased tax revenue incentive to provide infrastructure. It makes acquiring land for infrastructure much more expensive. It tempts governments to sell off land previously set aside for infrastructure, because of the greatly increased revenue available from selling it off. It encourages residents to engage in the politics of NIMBY (not-in-my-backyard) or BANANA (build-absolutely-nothing-anywhere-near-anyone) since they have so much value (and expectation of further capital gain) invested in their houses and so are extra fearful of anything that threatens that expected wealth effect: there is a difference between being serviced by a railway or freeway and being next to it. The notion that land-use control is a tool to improve infrastructure provision is nonsense on stilts: such controls greatly impede such provision.
Especially as such highly regulated land rationing encourages the notion that a city can be “moulded” to the whims of the planners, which are declared to be more important than the preferences and decisions of the residents (otherwise, folk could be left to buy and sell land for housing or non-housing uses as they wish). This encourages the notion that infrastructure is not about servicing people’s wants and needs, but a tool for the whims of planners.
Which then become more, not less, open to political manipulation. If infrastructure is no longer about servicing the general community (according to fairly clear criteria, where provision of general benefits to people has the moral authority), but about “doing good” (a much more nebulous and variable notion) then the “good” can just get redefined to suit the most politically engaged and effective interest groups: especially those with greater ability to frame the terms of policy debate.
Which is precisely what has happened, such as in the demonisation of “urban sprawl” (the “wickedness” of people wanting houses-with-gardens at prices they can afford). The failure of infrastructure provision in Sydney and Melbourne provides object lessons in “the good” turning out to be the convenience of the politically connected.
Part 2 of Michael Warby’s “Why Our Major Cities Are in Decay” is here…
[i] Except where otherwise indicated, median house prices and median household income ratios from 6th Annual Demographia International Housing Affordability Survey: 2011 (Data for 3rd Quarter 2010) online at www.demographia.com/dhi.pdf.
[ii] Paul Krugman, “No Bubble Trouble?”, New York Times, January 2, 2006 online at select.nytimes.com/2006/01/02/opinion/02krugman.html?_r=1.
[iii] 4th Annual Demographia International Housing Affordability Survey: 2008 (Data for 3rd Quarter 2007) online at www.demographia.com/dhi2008.pdf.
[iv] Australia’s national mean ratio of median house price to median household income for metropolitan areas of 1m residents or larger is 7.1, New Zealand 6.4, the UK 5.1, Ireland 4.8, Canada 4.6, the US 3.3. For all markets it is 6.1, New Zealand 5.3, the UK 5.2, Ireland 4.0, Canada 3.4, the US 3.0. 7th Annual Demographia International Housing Affordability Survey: 2011 (Data for 3rd Quarter 2010) online at www.demographia.com/dhi.pdf.
[v] Recent research has found that high levels of land use control are associated with more chaotic house prices, Edward L. Glaeser & Joseph Gyourko, Rethinking Federal Housing Policy: How to Make Housing Plentiful and Affordable, AEI, 2008, p.94 online at www.aei.org/book/971. Haifang Huang & Yao Tang, Residential Land Use Regulation and the US Housing Price Cycle Between 2000 and 2009, University of Alberta Working Paper No. 2010-11 online at www.bowdoin.edu/~ytang/Huang-Tang-2010-November.pdf found that US housing markets with more regulated land use had larger housing price booms and busts. The chaotic internal processes of command economies provide a more general example.
[vi] Research by author using data from the US Census Bureau’s American Community Survey and Demographia median house value and median household income data.
[vii] Taxi licences in Melbourne increased in value from $265,000 in January 1999 to approximately $478,333 in June 2008. Victorian Essential Services Commission, Final Report: Taxi Fare Review 2007-08, August 2008, online at www.esc.vic.gov.au/NR/rdonlyres/06733700-9E2F-4DDC-B912-C44EA7F410AA/0/FinalReportTaxiFareReview200708.pdf.
[viii] National Housing Supply Council, 2nd State of Supply Report, p.17 online at www.fahcsia.gov.au/sa/housing/pubs/housing/national_housing_supply/Pages/default.aspx.
[ix] Local government planning departments can be very comfortable forms of employment where no amount of failure is too much. For a scathing insider expose of UK local planning bodies see Matthew Walker, “The Great Inertia Sector: A whistleblower’s account of council work where staff pull six-month sickies”, Daily Mail, 26th June 2010 online at www.dailymail.co.uk/news/article-1289702/Public-sector-inertia-council-office-employees-month-sickies.html.
[x] Village of Euclid, Ohio v. Ambler Realty Co. online at caselaw.lp.findlaw.com/scripts/getcase.pl?navby=CASE&court=US&vol=272&page=365.
[xi] One study found that swings to Labor in NSW and the ACT were more strongly associated with housing stress than unemploynment or low incomes, though that low income cities vote National might have affected the result. Stuart Washington, “Fear of losing homes drove Labor win”, Sydney Morning Herald, December 8, 2007 online at www.smh.com.au/news/national/fear-of-losing-homes-drove-labor-win/2007/12/07/1196813021229.html.
[xii] Fintan O’Toole, “The landed class who blew the bubble”, Irish Times, 10 Oct 2009, online at www.irishtimes.com/newspaper/weekend/2009/1024/1224257361998.html. There was a complex interaction with planning approvals: cartels generally need some enforcement mechanism, given the incentive to “cheat”, which official discretions can provide.
[xiii] Marika Dobbin and Jason Dowling, “Huge land bank puts squeeze on buyers”, The Age, March 18, 2010 online at www.theage.com.au/victoria/huge-land-bank-puts-squeeze-on-buyers-20100317-qflq.html.
[xiv] “Why Not Copy Houston”?, January 11, 20111, online at www.unconventionaleconomist.com/2011/01/why-not-copy-houston.html.
[xv] Peter Rolfe, “Baillieu Government releases land for two new Melbourne suburbs”, Sunday Herald-Sun, January 23, 2011, online at www.heraldsun.com.au/news/victoria/boom-town-new-land-for-homes/story-e6frf7kx-1225992943557.
[xvi] For example, control of lot sizes—much of Massachusetts requires lot sizes to be an acre or more. Edward Glaesar, “If we build it, they will come”, Boston Globe, January 23, 2011 online at www.boston.com/realestate/news/articles/2011/01/23/if_we_build_it_they_will_come/.
[xvii] Assar Lindbeck, The political economy of the new Left, Harper and Row, 1977  p.39 quoted in Edward L. Glaeser & Joseph Gyourko, op cit, p.77.
[xviii] One of the worst examples was New York in the 1970s. New York City had extensive rent control, the US Federal Reserve was running a high inflation policy and New York State gave priority for public housing to those whose apartment had burnt down. Landlords stopped maintaining, then abandoned, buildings which began burning so that residents could get public housing. City blocks were devastated.
[xix] Rob Kitchin, Justin Gleeson, Karen Keaveney and Cian O’Callaghan, A Haunted Landscape: Housing and Ghost Estates in Post-Celtic Tiger Ireland NIRSA Working Paper 59 July 2010 online at irelandafternama.wordpress.com/2010/07/29/a-haunted-landscape/.
[xx] The Kitchen et al study is a particularly fine example of planning ideology . The authors, without any sense of self-consciousness, write of “laissez-faire approach to planning” while detailing a system of zoning controls which became dominated by political pressures; presume that a “proper” planning system can balance supply and demand; have no conception that quantity controls can have price effects or their role in creating expectations and discounting downside risk; note that those areas with highest vacancies were the most generous in providing zoning approvals (which meant that new housing estates were built where they were permitted rather than where there was likely to be sustaining demand) without also noting that the areas with highest demand (as measured by vacancy rates) also had the most restricted supply, encouraging upward pressure on prices. To the authors, anything that suggests markets have value is “neoliberal ideology” and the base problem, to which wise planning following the ideas they agree with is the solution. As the authors note, Ireland had strong population growth (16.8% increase from 1996 to 2006), shrinking household sizes and rising average incomes. The simple existence of quantity controls in that situation—given the advantages in restricting approval to less than the level of demand—would be enough to set off rising prices, consequent expectations of capital gain and discounting of downside risk. Which led to overshooting of supply—mismatched as previously noted—and a price collapse. The authors detail a complete failure of a planning system, including quantity controls, and label it a failure of market ideology.
[xxi] Hugh Pavletich reports that a retired academic recently wrote the following comments to him:
“Our universities are churning out social scientists not as informed citizens who might then look for a career, but as people who have the expectation that they can turn their degree in to a career. Many of them head in to public sector employment and policy jobs, because there is nowhere else for them to go. And any time they are asked to look at an issue, they generate policy recommendations that lead to regulations. Because their knowledge is bestowed on them by weak (or doctrinaire) teachers, they do not have the critical capacity necessary for the job. Nevertheless, they can become part of the club which exerts considerable power, and persists (often becoming stronger) through changes in government. This applies to local and central government.”
“The result is not just over regulation, but poor regulation. Policy is based on imitation, workshops and group think.”
“While my leanings are (or were) more liberal than libertarian, this has led me to the position of seeing governments as inept and potentially a deadweight as a result of an entrenched and pedestrian bureaucracy that has circumscribed the discretion of the executive, ably supported by a lazy media that in turn limits the range of issues we might debate. There are exceptions to all of this I guess.” Respecting The Median Multiple Housing Measure, January 30, 2011 online at www.realestatebuzz.com.au/respecting-the-median-multiple-housing-measure/.
[xxii] The value of bonds is, therefore, entirely determined by the reliability of their income stream. Since bonds are nothing more than a congealed specified income stream, you do not get “bubbles” in bond prices that have continued to conform to their promised obligations, though you can get “busts” if the promised income stream is cut off. (For example, because the company goes bust or the jurisdiction defaults.)
[xxiii] “The Great Australian Land Racket”, Friday, January 7, 2011 online at www.unconventionaleconomist.com/2011/01/great-australian-land-racket.html.
[xxiv] Thus both Texas and Georgia largely avoided the housing price boom-and-bust but Georgia suffered far more from the sub-prime crisis because of much more permissive home lending laws. Alyssa Katz, How Texas avoided the worst of the real estate meltdown, Posted Tuesday, March 30, 2010 – 3:01pm online at www.thebigmoney.com/articles/judgments/2010/03/30/lone-star-secret?page=full. Paul Krugman “Georgia on My Mind”, New York Times, April 11, 2010 online at www.nytimes.com/2010/04/12/opinion/12krugman.html?_r=1.
[xxv] The issue of the existence (or not), nature and appropriate policy maker response to bubbles generates considerable debate among economists, much of which is fairly arcane and turns on definitional issues. It also involves the badly named ‘efficient market hypothesis’ (EMH), the weak version of which just says that, in reasonably open markets, available information gets reflected in market prices. There may well be “friction” in the transfer of such information to prices, but once people in the market start acting on specific information, prices will reflect that. The corollary to this is that there is no systematically better way than reasonably open markets to price assets. But the notion that the efficient market hypothesis (at least in its weak form) precludes asset price bubbles is a misreading of what EMH implies. (It says nothing about the quality of the information, for example.) General expectations of capital gains are part of the information feeding into the market.
[xxvi] It is all very well to say that prices have diverged from the value of expected income from said asset but that says nothing about how long they will continue to do so, or how far they will diverge, or whether income will rise to conform to the asset value. We cannot reliably and systematically predict the future of prices because we cannot reliably and systematically predict future information, the information that does not yet exist. Expectations, however, are part of current information. While prices continue to reflect the expectations of capital gains, such capital gains will continue to be realised.
[xxvii] ASIC, Australian investors: at a glance, April 2008 online at www.asic.gov.au/asic/pdflib.nsf/LookupByFileName/rep_121_Australian_investors_at_a_glance.pdf/$file/rep_121_Australian_investors_at_a_glance.pdf.
[xxviii] RBA statistics, online at www.rba.gov.au/statistics/tables/index.html#money_credit.
[xxix] Monetary policy is a general phenomenon while bubbles occur in specific asset markets. It is always a pertinent question, why did a bubble occur in this asset market and not another? The US, for example, does not have a single housing market, it has hundreds.
[xxx] Federal Reserve Chairman Ben Bernanke usefully goes through the international evidence that monetary policy does not create housing bubbles in his January 3 2010 speech at the Annual Meeting of the American Economic Association, Atlanta, Georgia online at www.federalreserve.gov/newsevents/speech/bernanke20100103a.htm.
[xxxi] Can we know whether house prices will go up or down in the next x months? No, for we cannot predict future information. Can we know if and when the “tipping point” will occur when house prices collapse back to their long-term rental income value? No, for the same reason. Hence the difficulty for policy to “target” bubbles.
[xxxii] Since assets provide benefits over time (this is one of the defining characteristics of an asset), understanding risk and expectations becomes particularly important in thinking about asset provision and markets.
[xxxiii] Wendell Cost, Metropolitan Area Migration Mirrors Housing Affordability 03/28/2010 online at www.newgeography.com/content/001485-special-report-metropolitan-area-migration-mirrors-housing-affordability.
[xxxiv] One study found measurable anti-competitive effects for hotel location and size in Texas: far from the most restrictive jurisdiction. Junichi Suzuki, Land Use Regulation as a Barrier to Entry: Evidence from the Texas Lodging Industry, University of Toronto Dept. of Economics Working Paper 412, October 21, 2010 online at epec.economics.utoronto.ca/files/tecipa-400.pdf.
[xxxv] Murray J. Horn, The Political Economy of Public Administration: Institutional Choice in the Public Sector, Cambridge University Press, 1995, particularly his discussion of the patterns of which industries are significantly government owned Pp134ff. Statutory monopolies are far more common than “natural” monopolies for government-owned businesses as revenue is thereby more easily redistributed for non-commercial (i.e. political) reasons. The existence of “sunk” costs vulnerable to expropriation, connection to a range of interests seeking benefits via non-commercial objectives, centralized production (reducing the difficulty of, and costs from, outside monitoring and control of production) and producing a standard product (making monitory of cost and quality easier)—all common characteristics of infrastructure provision—seem to be more important explanatory features for government ownership in market economies than conventional “market failure” analysis.
[xxxvi] Horn’s discussion of the patterns and problems of government control is enlightening. In particular, the issue of endemic uncertainty due to commitment problems—that governments cannot be trusted to keep promised tax or legislative benefits over the longer term, especially not to narrow private interests such as shareholders: ibid, Pp135-145.
[xxxvii] Horn points out that public ownership is not good for the interests of taxpayers and creditors; if public debt becomes a more urgent issue, that can abruptly shift political benefit in favour of privatisation: ibid, Pp165-8.
[xxxviii] Especially as, while the empirical evidence on cost, efficiency, productivity, etc strongly favours private over public ownership, the evidence is much more equivocal regarding public ownership versus heavily regulated private ownership: ibid, p.152.