Myth: Public transport makes housing unaffordable

Myth: Public transport makes housing unaffordable
Fact: Public transport needn’t spell the end of the quarter-acre block, nor is it responsible for the inflation in Australian house prices. Despite strong (though progressively weakened) planning measures in Melbourne to protect green wedges and restrain urban sprawl, the supply of housing now is greater than it’s ever been. Rapid population growth has likely been a factor in the most recent price hikes, but catering for this with car-dependent sprawl will only put more strain on household budgets in the longer term.

Few of us will have failed to notice the ‘affordability crisis’ that has struck the Australian real estate market in recent years. In Melbourne between 1996 and 2003 in particular, the median house price more than doubled, from $150,000 to $350,000. Growth in incomes has been less spectacular, and this has led to people having to borrow ever-greater multiples of their annual disposable income if they want to share in the ‘great Australian dream’ of home ownership.

There is a general consensus that the explosion in house prices is due to a number of factors, chief among them being a sustained period of low interest rates up to 2004 and again in 2009 (allowing more to be borrowed for the same monthly repayment), a boom in financial speculation – including from overseas – inflating asset prices, and various new government incentives to spend up big on real estate (such as the halving of capital gains tax on investment properties held for a year or more).

But even the multitude of credible explanations available hasn’t prevented property developers (who profit handsomely from real estate inflation), and the right-wing think tanks who act on their behalf, from offering self-serving ‘explanations’ of their own. So it’s become fashionable in some circles to blame high house prices on – of all things – sustainable transport-and-land-use planning.

The gist of this particular fairy tale is that anti-sprawl policies, such as Melbourne’s urban growth boundary and planning rules barring certain kinds of development in rural ‘green wedge’ areas, create an artificial scarcity in the supply of land for housing and this is what pushes prices up. Since public transport allegedly needs strong planning measures like this to support its use, it’s concluded that public transport is bad for ‘the battlers’ because it leads to unfairly inflated house prices.

Too many state and local politicians and their planning advisers place obstacles in the way of people seeking to buy new homes…. Many of this view have an emotional attachment to public transport.

—Alan Moran (Institute of Public Affairs), Herald Sun, 9 September 2006

It is important to remember that the scarcity that propelled land prices is artificial…. It is the product of restrictions invoked through planning regulation and zoning…. State and Territory Governments have been spurred along by an urban planning cheer squad which is obsessed with curbing the size of our cities and pushing a policy of urban consolidation. Between them they have excluded more low and middle income earners from home ownership than at any other time in Australia’s history.

—Bob Day (Institute of Public Affairs), address to Australian Christian Lobby, September 2006

No longer do we all troop to work to the urban centre using buses, trains and trams. For the most part, we work in dispersed locations and prefer to travel by car to work and for entertainment…. Imposing constraints on developments means an alarming inflation in the cost of land.

—Alan Moran, The Age, 23 January 2007

Much of this rhetoric originated in the USA, many parts of which lack the comprehensive urban planning that has been traditional in Australia, and where efforts to establish a planning system more akin to ours have been resisted by developers and the road lobby. But even in the Australian context the claims are nonsense. For a start, greater public transport use doesn’t actually require urban consolidation or higher population densities, as we explain on another page. Nor does our general preference for big houses in the suburbs mean that public transport can’t cope, as we also explain elsewhere. But regardless of all this, the evidence (which the developer lobby never refers to, preferring simply to belabour the obvious point that our house prices are high) shows there has been no scarcity of housing in Australia or in Melbourne to coincide with the price boom, and that there is no connection between urban consolidation policies (such as Melbourne 2030 or its weaker successors) and escalating house prices.

Consider the simple matter of timing. When the rapid escalation in Melbourne house prices began in 1997, Victoria was in the grip of the notorious, laissez-faire “I hate silly rules” regime of Kennett Government Planning Minister Rob Maclellan. This opened up a lot of land for development in ‘green wedge’ and other non-urban areas just before house prices went up. Melbourne 2030, with its green-wedge protection legislation and urban growth boundary, was not enacted until 2003, at which time the rapid growth in Melbourne house prices promptly stalled. Between 2003 and 2006 prices were essentially flat, and in real terms declined slightly.

Then in 2006 the government gave in to developer pressure, loosening the urban growth boundary and announcing new land releases to provide a rolling 15 years’ supply. Further loosening was flagged in 2008 and enacted in 2010. It’s not entirely fair to hold this responsible for the new price hikes that followed, since the same trend is found in other Australian capitals. Nonetheless, such evidence as exists suggests the exact opposite of what the developer lobby claims: laissez-faire development causes price inflation, while restrictive planning keeps prices under control!

Year Melbourne
house price
Change Development approvals in ‘green wedge’
(and other significant events)
1990 97.4
1991 96.9 -1%
1992 94.1 -3%
1993 95.9 +2%
1994 97.9 +2%
1995 97.3 -1% Laurimar Estate, Heritage Golf Estate
1996 101.2 +4% Sandhurst Golf Estate
1997 113.5 +12% Flora Park Estate
1998 124.6 +10%
1999 143.6 +15% Botanic Ridge, Kingston Waterways, Moonah Links
2000 157.5 +10% Castlebrook Estate / Pindara, Acacia Ridge
2001 189.9 +21% Keysborough Concept Estate
2002 213.6 +12% Eynesbury Station, Epping North
2003 240.2 +12% (Urban growth boundary established)
2004 236.6 102.4 -2%
2005 105.4 +3%
2006 114.8 +9% (Urban growth boundary relaxed)
2007 141.3 +23%
2008 137.0 -3% (Onset of global financial crisis)
2009 163.7 +20% (Further extension of growth boundary announced)
2010 176.6 +8%
2011 166.6 -6%
2012 166.6 0
2013 180.9 +9%

Sources: Australian Bureau of Statistics, Established House Price Index (December figures). Michael Buxton and Robin Goodman, Maintaining Melbourne’s Green Wedges. In 2005 there was a technical change in the ABS methodology: the new series has been used from this point, with all percentage changes based on figures from a single series.

The other major piece of contrary evidence comes from the Census data on households and dwelling sizes. It’s a reasonably well-known fact that through the last decades of the 20th century and well into the 2000s, there was a drop in average household size, with more people living alone and couples having fewer children. Despite this, the average house size (as measured by number of bedrooms) steadily rose over the same period. A 2009 ABS study found that new homes in Australia are on average the largest in the world, larger even than new homes in the USA. And within Australia, the biggest average house sizes are in Victoria, more specifically Melbourne.

As the table below shows, not only has the number of homes in Melbourne been growing faster than the number of households to live in them: because of the growing disconnect between house size and household size up to 2006, the total amount of available accommodation (bedrooms per person) grew even more rapidly.

Housing availability in Melbourne, 1991-2011
1991 1996 2001 2006 2011
Resident population 2,997,739 3,158,165 3,367,169 3,592,591 3,999,982
Resident households 1,115,531 1,199,271 1,298,999 1,391,929 1,533,138
Av. household size 2.687 2.633 2.592 2.537 2.609
Private dwellings 1,144,094 1,234,302 1,344,624 1,471,155 1,636,167
Dwellings per household 1.026 1.029 1.035 1.057 1.067
Av. bedrooms per dwelling 2.850 2.897 2.958 2.940 3.017
Number of bedrooms 3,260,668 3,575,773 3,977,398 4,325,196 4,936,316
Bedrooms per person 1.088 1.132 1.181 1.204 1.234

Source: Australian Bureau of Statistics. Census data for usual resident population, dwellings and number of bedrooms per dwelling. ABS derived figures for estimated resident households. All numbers relate to the Greater Melbourne Statistical Area (formerly the Melbourne Statistical Division).

Before the 2011 Census data appeared, the National Housing Supply Council had warned of an undersupply of housing throughout Australia due to rapid population growth after 2006. This warning was not borne out in the Census data, nor in subsequent Morgan Stanley research, which found that instead of an estimated undersupply of 228,000 houses in Australia there was in fact an oversupply of 341,000 houses. Confirming the overall picture was the news, reported in The Age in July 2012, that the number of unsold homes in Melbourne had reached a new record of 55,290 that June – with most of these being located in public-transport-poor outer suburbs.

It’s like a ghost town in the new developments. Nothing has moved for the past three, four, five, six months.

—John Halkidis (agent, Hocking Stuart), The Age, 8 July 2012

In the face of the evidence, Housing Supply Council chairman Owen Donald conceded that the housing situation was likely one of oversupply, in an article in the Australian Financial Review on 3 July 2012.

However one looks at the figures, they cannot be made to support the claim that a growing scarcity of housing in Melbourne caused the inflation in house prices between 1996 and 2006. What is evident instead is that housing provision more than kept up with population growth and with the shrinking size of households, and did so with houses larger than previous generations were accustomed to. Others who have looked at the Census data have come to similar conclusions, including for other parts of Australia.

The state of the land in Melbourne is substantially healthier and more positive than is the case in Sydney…. The Victorian Government has paid significant attention to the provision of sufficient land to satisfy demand in Melbourne’s growth areas.

—Urban Development Institute of Australia, State of the Land, 2006.

There is a genuine structural problem with housing affordability in Melbourne, but it is a distributional problem, not a land supply problem. Available evidence suggests on the one hand a concentration of demand for real estate in (mostly inner-city) locations with good local facilities and public transport access, and on the other a glut of unsold real estate in suburban locations that lack local facilities and where public transport is of a particularly poor standard. For example, in the first quarter of 2008, house prices in Ashburton (on the Alamein train line) increased by nearly 8%, while those in Hillside (an outer north western suburb with a handful of bus services) dropped by nearly 2%.

This is a troubling scenario of a city divided. We will wind up with two Melbournes, one that is prosperous and well served with infrastructure and employment opportunities, versus a second one on the fringes which is dislocated with property prices that are far less buoyant.

—Marcus Spiller (former president, Planning Institute of Australia), The Age, 8 July 2012

The big increase, however, has been in the price of land. This is not the result of policies that restrict the supply of new housing sites on the urban fringe. If it were, we would see prices rising most rapidly in the outer suburbs and a flattening of the price gradient from the fringe to the city centre. In fact, that gradient has risen dramatically, to the point where tiny blocks in the inner suburbs can be sold, vacant or with derelict buildings, for millions of dollars.

—John Quiggin, The Age, 15 June 2015

Recent US evidence also confirms this pattern. Following the US real estate crash that precipitated the Global Financial Crisis of 2008-09, house prices in car-dependent suburbs crashed a lot further than those in inner suburban ‘walkable neighbourhoods’ valued for their superior transport access. This blog post from the New Republic notes that this locational pattern of price falls has not been seen in any previous recession.

We are designing the world’s worst suburbs, the housing stock is terrible and they have very poor liveability. We are going down the same path as the US, where areas outside major cities have become suburban ghettos.

—Profssor Michael Buxton (RMIT University), The Age, 8 July 2012

It’s certainly also true that rapid post-2010 population growth has to be met with an increased supply of housing, whether that is inner-city apartments and townhouses or greenfield developments on the fringe. Population growth has likely been a factor in the most recent price increases, much more than in the late 1990s when Victoria’s population hardly grew at all. But no amount of tinkering with transport policy one way or the other will alter this pressure – the only question, if we are to have high population growth, is whether we accommodate it with well-serviced suburbs or car-dependent ones.

If anything, investment in public transport infrastructure and services should help make housing more affordable, by levelling out the locational benefits of particular areas and so tempering what commentator Julian Disney has called “bidding wars for well-located properties.”

Tackling the issue of affordability only through the lens of supply misses the point. Merely adding to supply, say, through the release of land on the fringes of our cities – ‘fixing the fringes’ – has been recognised as having little effect on affordability, because it does little to assuage the insatiable demand for well-located houses. It may even add to the financial woes of struggling, overstretched households in those outer suburbs where prices have fallen and where negative equity has become a problem for whole neighbourhoods.

—Fiona Allon, Renovation Nation: Our Obsession With Home (UNSW Press, 2008)

But if one wants a ‘supply-side’ explanation for the housing boom, there is an excellent orthodox candidate: the supply of money for home loans. At the 10% interest rate that prevailed in the mid-1990s, a family on the median household income of roughly $39,000 in Melbourne could afford to borrow about $130,000 on a typical bank formula (where one third of gross income is applied to loan repayments). But at the 5% rate applicable five years later, a family on the 2001 median income of roughly $45,000 could afford to borrow around $300,000. Throw in the proliferation of home loan providers and the relaxation of lending criteria and the climate was ripe for house price inflation to occur, even with plenty of houses to go around. In part this has been due to the proliferation of housing investors ‘crowding out’ owner-occupiers.

What sent house prices soaring was the invasion of the investors, driven by the tax breaks for negative gearing and capital gains. Landlords used to own 20% to 30% of the housing market. But by 2003 they were taking almost 50% of housing loans (excluding refinancing), and even now, they take close to 40%, outbidding the first-home buyers.

—Tim Colebatch, The Age, 18 March 2008

[In the past] credit was in such short supply, few were offered enough cash to buy a home. It wasn’t until our banks discovered cheap offshore credit in the mid-1990s and brought the cash onshore that we suddenly had “affordable” home loans. But the cheaper credit simply shifted the price of real estate higher. It was a windfall for the banks. For the real estate boom resulted in ever larger loans…. As a nation, it’s left us with a serious but not insurmountable foreign debt problem.

—Ian Verrender, The Age, 2 February 2012

The flood of negative gearers has pushed prices to levels that once wouldn’t have lacked support. In the decade before 1999 Melbourne prices rose by an average of 4.8 per cent per year. In the 15 years after they’ve climbed 7.6 per cent per year. They’ll keep climbing for as long as negative gearers (who don’t much care about prices) outbid owner-occupiers.

—Peter Martin, The Age, 9 June 2015

Another supply-side factor that shouldn’t be ignored is the practice of ‘land banking’, where large developers buy up vast tracts of land in new residential zones and then release it for construction at a rate calculated to maximise their commercial return. A December 2009 report by the Oliver Hume real estate group found that private developers were holding almost 70,000 vacant house blocks already approved for construction, with only 1,400 having been released to the market. So even if there were a shortage of land to build houses on, it would seem to be one created by developers themselves, not government policy!

Interest groups keep telling us the solution is to zone more fringe land for housing. But we already have zoned urban land in spades, owned by developers who are holding on to it so they don’t flood the market. Releasing more land will solve nothing.

—Tim Colebatch, The Age, 6 April 2010

In the 12 months to September 2010, the median price of land in Melbourne’s fringe suburbs inflated by 24 per cent. The context for this record-breaking increase was the Brumby Government releasing a new planning policy in December 2008 (Melbourne @ 5 Million) that foreshadowed a further massive extension of the urban growth boundary. After a year and a half of controversy, this extension was ultimately legislated in mid-2010. It was obvious to anyone with an interest in real estate that speculators had been bidding up the price of fringe land ever since it became apparent the government wanted it included in the urban area. But this didn’t stop the Master Builders Association coming up with its own self-serving explanation, complaining that “there is too little available for sale”!

The Australian housing market almost entirely avoided the turmoil associated with the Global Financial Crisis in the USA and Europe in particular. While this appears largely due to the greater prudence of Australian financial institutions, the anti-sprawl thrust of our planning culture (even if watered down) has played an important part. Not only has it helped restrain the urge toward oversupply in the boom years: it also ensured some basic level of access by (most) new home owners to shops, employment and public transport, even while their US counterparts found themselves struggling to afford the petrol to drive miles to buy food and other necessities.

In U.S. cities the proportion of household expenditure on transport increased from 10% in the 1960s to 19% in 2005…. Families with household incomes between $20,000 and $50,000 spend almost 30% on transport. In Atlanta, a city that features urban scatter within this income range, 32% of household incomes goes on transport. Households on the fringes of car-dependent cities are therefore much more vulnerable as the cost of transport escalates. The doubling and even tripling of gasoline prices this year… provided the spark that lit the bomb underneath the subprime mortgage crisis, hitting many with a double whammy of increased transport costs and ballooning mortgage payments.

—Professor Peter Newman, radio interview, December 2008

Look at the [US] cities with stable and recovering home markets. On [the west] coast, San Francisco, Portland, Seattle and San Diego come to mind. All of these cities have fairly strict development codes, trying to hem in their excess sprawl. Developers, many of them, hate these restrictions. They said the coastal cities would eventually price the middle class out, and start to empty.

It hasn’t happened. Just the opposite. The developers’ favorite role models, the laissez faire free-for-alls – Las Vegas, the Phoenix metro area, South Florida, [California’s central] valley – are the most troubled, the suburban slums.

—Timothy Egan, New York Times blog, 10 February 2010

A 2013 study of 37,000 US housing mortgages by the University of Arizona (reported by The Atlantic) confirms that those located near public transport were in fact substantially less likely to default. So rather than make it more difficult to provide affordable housing, transit-oriented planning does actually appear on the evidence to reduce financial stress on homeowners.

In summary, a return to the free-for-all urban planning approach of the Kennett era will not make housing more affordable, and might actually make it even less affordable. A bigger problem for our planners is the fact that many new home buyers in Melbourne lack access to public transport and are therefore particularly exposed
to higher petrol prices. Providing families with a public transport alternative costs less than building motorways and occupies less land, making it easier rather than more difficult to achieve urban growth.

Affordability may be a necessary condition of liveability, but it is not the only one. A suburb is not just a cluster of houses or residences, it is a community. To be successful, communities require infrastructure, and that means public transport to provide access to employment, shops, schools, medical facilities, open spaces and social amenities. Too many recent developments lack such things, leaving too many young people jobless, bored, alienated and angry, too many mothers in single-car families hopelessly isolated. A chorus of experts has warned that these places are outer-urban ghettos-in-the-making.

—Editorial, The Age, 9 July 2012

Return to index

Last modified: 25 October 2016