RDP 2011-03: Urban Structure and Housing Prices: Some Evidence from Australian Cities 5. Conclusion

In this paper we use a simple model that highlights some long-run influences on the supply and demand for urban housing. While the model abstracts from some aspects of reality, we show in Section 4 that it is consistent with many of the key patterns observed in Australia's large cities.

The model highlights the importance of transport infrastructure. While we do not model the financing of such infrastructure, the model points to the potentially significant effects of transport systems on households. In particular, in cities with better transport infrastructure it is more feasible to live further from the city, where land and housing prices are lower and dwelling sizes can be larger. It is noteworthy that reports on the development process for Australian cities have pointed to transport problems being a major issue for development at the city fringe (e.g. Applied Economics (2010)).

The analysis also highlights the effect of land usage policies on housing density and prices. We show that zoning limits on the amount of housing built close to the CBD imply that more of the population lives in middle and outer rings, increasing the overall footprint of the city and resulting in higher housing prices. These effects can be expected to become more pronounced as the population of cities increase. In particular, as population grows and land prices increase, the expected response would be for cities to become denser and for land to be used more intensively. However, any fixed set of zoning restrictions will slow this adjustment.[25] While the model predicts city-wide benefits – in terms of more affordable and better-located housing (and larger average dwelling sizes) – from less restrictive zoning, it is noteworthy that existing residents often oppose the easing of zoning restrictions in their own neighbourhoods. This highlights an important point: there is a trade-off between density and housing prices.[26]

The model shows that policies associated with the development process that act as a friction on the supply of new housing result in higher housing prices for consumers and a reduction in the supply of housing. In the model, which assumes a fixed number of people per household, this is reflected in smaller dwelling sizes, but one can also think of these policies as resulting in an increase in the average number of people living in dwellings of any given size.

While demand-side factors, especially changes in the cost and availability of finance, have clearly been very important in explaining developments in the Australian housing market over the past two decades, our results lend support to other work pointing to the role of structural factors in influencing housing market outcomes. There is a growing body of international evidence on the role of supply-side constraints in limiting construction and driving up prices: for example, see Gitelman and Otto (2010) for Australia; Grimes (2007) for New Zealand; Barker (2004) for the United Kingdom; Green, Malpezzi and Mayo (2005), Glaeser et al (2005a), Glaeser, Gyourko and Saks (2005b) and Saiz (2010) for the United States; and Andrews, Caldera Saánchez and Johansson (2011) for a large sample of OECD economies. The evidence for Australia presented in Section 4 is also consistent with various supply-side aspets boosting the cost of housing, although these findings are tentative given the paucity of data on land prices and zoning, and on the relationship between these variables.

Finally, we note that the effects on the cost of housing of any given set of development, land use and transport policies will tend to be more pronounced as the population grows. Arrangements which might have been considered broadly appropriate at some point in the past are likely to have more pronounced effects as the population expands. This suggests that any set of policies that were thought optimal at one time, in terms of the trade-offs, might no longer be so as the population grows.


In this regard, a recent report by the Productivity Commission (2011, p xviii) calls for ‘broad and simple land use controls to: reduce red tape, enhance competition, help free up urban land for a range of uses and give a greater role to the market in determining what these uses should be’. More broadly, the report suggests that planning systems in Australian cities suffer from ‘objectives overload’, which has been increasing over time. [25]

See NHSC (2010, p 123) for a discussion of some options for stimulating higher density and infill developments in our cities. [26]