Melbourne, like other large Australian cities, is growing outwards more than upwards. Together with the shift from manufacturing to services, this leaves a large proportion of the population to choose between unaffordable housing in inner areas with high concentrations of employment, or long and expensive commutes from the outer suburbs. The result is erosion of living standards and rising environmental costs. State and local governments acknowledge the need for measures to increase housing supply in established suburbs and to improve the efficiency of transport systems.

There are currently limited tools available to assess the spatial economic impacts of urban policies or infrastructure investments in Australia. We have developed a spatial computable general equilibrium (SCGE) model of Melbourne featuring housing, employment and travel between over three hundred residential and employment zones for work and other purposes. Our ‘VU-Cities (Melbourne)’ model also accounts for positive externalities of density affecting both firm productivity and residential amenity.

We demonstrate application of VU-Cities by comparing population growth under three different policies: (i) a continued heavy reliance on peripheral greenfield developments; (ii) extensive rezoning for denser residential developments in Melbourne’s established ‘middle ring’ suburbs; and (iii) a uniformly applied transport tax. Both densification and transport tax policies increase residential densities, but the transport tax has much stronger impacts on employment densities. It also reduces aggregate travel demand substantially. The densification policy alleviates pressure on housing costs but has small productivity benefits, whereas both housing costs and productivity are substantially increased with the transport tax. The densification and transport tax policies thus have complementary benefits.