It is amazing how quickly supposed facts become accepted wisdom. A feature article in The Age last week, Fringe Benefits, repeated the highly questionable claim that “for every 1,000 dwellings, the cost for infill development (in existing suburbs) is $309 million and the cost of fringe development is $653 million”.

These numbers are doing the rounds at the moment, appearing in a number of recent reports, including Transforming Australian Cities, prepared jointly by the Victorian Department of Transport and the Melbourne City Council; Residential Intensification in Tramway Corridors, prepared by SGS Consultants for the Victorian Department of Planning and Community Development; and in Macro-Urban Form, Transport Energy Use and GHG Emissions: an Investigation for Melbourne, published by the Victorian Department of Transport.

The claim is doubtful for a number of reasons. The reports cited above all sourced the numbers from another report, Assessing the Costs of Alternative Development Paths in Australian Cities, written by three academics (Roman Trubka, Peter Newman and Darren Bilsborough) and published by Curtin University.

Trubka et al in turn sourced estimations of infrastructure capital costs from a 2001 report, Future Perth. As the title suggest, this report does not relate to Melbourne. It is also unpublished and hence it is difficult to evaluate the soundness of its methodology and findings. However Trubka et al do tell us that it did not actually calculate development costs from first principles, but rather derived estimates by surveying 22 earlier studies. These date from as far back as 1972 and cover the USA and Canada as well as Australia.

Further, around half the $653 million fringe development cost quoted does not actually relate to infrastructure outlays but is made up of private costs (like vehicle operating expenditures) and ‘notional’ economic costs (like travel time). These are important issues, but they are not infrastructure outlays and the two should not be confused.

Finally, the figures quoted actually compare development costs on the fringe against those in the inner city, not against the existing suburbs as the article in The Age claims. Inner city infrastructure development costs are irrelevant to any discussion of sprawl because households on the fringe can generally neither afford nor want to settle in the inner city – the established parts of the outer and middle suburbs are the likely alternative for them if they cannot settle on the fringe.

In any event, vigorous population growth over the last 10-15 years has weakened the longstanding premise that there is boundless spare capacity in inner city infrastructure. The problem now is that increasing capacity in the inner city is very expensive given its high dwelling densities, high land values and extensive stock of heritage sensitive buildings.