Workers flee Sydney's unaffordable housing

The survey’s strength lies in its comparative methodology. It’s hard to dispute a consistent pattern emerging from so many markets with varying population sizes, histories, locations, industrial and economic orientations, interest rate regimes and taxation arrangements. In contrast to the dire Australian findings, 42 markets are assigned to the “affordable” category, including sizeable American and Canadian cities like Atlanta, Detroit, Dallas, Houston, Pittsburgh, Quebec and Ottawa. The survey’s message is clear, and consistent with Moran’s:

Research in the surveyed nations identifies the cause - the housing cost escalation is principally the result of supply factors. Where there are significant constraints on the supply of land for residential development, housing inflation has occurred. Where there are no such constraints, housing cost inflation has not occurred. Demand does not raise prices by itself. Demand can only raise prices where there is insufficient supply.

This conclusion undermines a popular Australian criticism of the structural or land supply explanation, namely that historically low interest rates in the early 2000s - which boosted the purchasing power of home buyers by 60 per cent, according to Rory Robertson - drove up demand and prices. The NSW Government and some media elements dismissed the survey on this ground. The Government also thought Sydney’s dismal ranking was misleading, since the “study included expensive harbourside areas”.

This objection would have more substance if the survey used mean rather than median prices - median values, unlike means, are barely affected by a few extreme observations. As the survey points out, the median multiple has been recommended by the World Bank and the United Nations as a tool for evaluating urban markets.

On the strength of his paper, Thinking About the Big Drop in Australian Housing Affordability, Robertson was trotted out by the Fairfax newspapers to trample on the structural or land supply perspective.

One Sydney Morning Herald article featured his assertion that releasing more land on the outskirts will do nothing to improve affordability. Desirable inner Sydney suburbs like Bondi, Bellevue Hill, Bronte, Mosman and Paddington, he said, would remain beyond the reach of most first home buyers.

This is a statement of the obvious, of course, and beside the point. Land releases improve affordability on fringe greenfields sites - where a large proportion, probably a majority, of low to middle income couples with children are willing to live.

Shortly afterwards, the Australian Financial Review quoted him deriding the Demographia survey because the “the 26 most-expensive cities were high demand locations with a “large, sexy and high profile”. In contrast, said Robertson, the affordable cities “were generally low-demand locations”, like Fort Wayne and other US cities which “were actually shrinking in size, such as Youngstown, Buffalo, Dayton, Rochester, Akron and Toledo”. He neglects to mention, though, that the affordable category also includes the booming cities of Atlanta, Dallas, Pittsburgh and Houston.

The NSW Government is not oblivious to the need for more residential lots on greenfields sites, and its comprehensive City of Cities plan earmarks two substantial north-western and south-western “growth centres” for residential development. But the plan envisages that over the next 25 years only 30 to 40 per cent of Sydney's new housing (160,000 lots) will be built in these centres, while the balance will be assigned to higher density developments in the city's established inner suburbs. This is unlikely to have much impact on the affordability problem, however.

In this respect at least, Robertson is correct. Yet the Government rejects expanded land supply as a solution to Sydney’s housing crisis, claiming that although 26,000 lots (including 17,000 rezoned by the Government last year) are currently available, there is simply no demand from developers and buyers in the weak property market. This reflects a narrow cyclical view of the market, however, and ignores the relationship between demand and prices. If lots were more freely available, their price would be lower and demand higher.

Housing affordability is on the agenda. But until planners and policymakers break free of their limited cyclical perspective, we aren’t likely to see more than stopgap proposals to increase the first home owners grant, cut taxes and stamp duty, increase the stock of public housing and rent relief - all of which are helpful but amount to little more than tinkering around the edges.

Freeing up the land supply disturbs powerful interests, who will resort to flawed economic analysis, and fears of social dysfunction and environmental degradation, including the prospect that suburban expansion will increase carbon emissions. All of these grounds are proved baseless on closer scrutiny.