The new Australian Infrastructure Plan has some good recommendations, but it isn’t notable for the sophistication and depth of its supporting analysis.

One of the tropes repeated by Infrastructure Australia is the idea that residential development infrastructure costs on the urban fringe are much higher than those in the inner suburbs. It says:

“The research finds that the average cost (in 2007 dollars) of providing energy, telecommunications, water and transport infrastructure services to a unit of housing in existing inner urban areas is $26,500, while for outer urban greenfield locations the cost is $69,500. This represents a difference of $43,000 per unit.”

The research Infrastructure Australia relies on is a 2007 report, Assessing the costs of alternative development paths in Australian cities, written by three Curtin University academics, Roman Trubka, Peter Newman and Darren Bilsborough.

The authors compared the costs of providing infrastructure in the first 10 kilometres from the CBD of a notional city with the costs in the outermost 10 kilometers (they also looked at the economic costs of the two locations).

A number of other influential reports have also relied on their findings to support their conclusions and recommendations e.g. Transforming Australian Cities; Residential Intensification in Tramway Corridors; and Macro-Urban Form, Transport Energy Use and GHG Emissions: an Investigation for Melbourne.

I looked at the Curtin University report back in 2010 and 2011 and wasn’t impressed. I pointed out in this article, that the underlying data is out of date and the methodology isn’t transparent.

As I said then, Trubka et al didn’t calculate costs themselves. They sourced their estimates from a 2001 report, Future Perth, prepared by the WA Planning Commission to assess infrastructure costs in Perth.

Future Perth didn’t calculate its estimates from first principles either; rather, it surveyed 22 earlier studies, some dating from as far back as 1972 and some relating to costs in the United States and Canada.

Future Perth is a working paper and hasn’t been published — hence the rigour of its methodology and the quality of the 22 studies it drew from hasn’t been tested.

Unfortunately, this area of policy and research seems to be a black hole. The sort of reliable data on contemporary infrastructure costs that’s required to inform policy-making simply doesn’t exist. I don’t know if they ever were, but I think it’s unlikely the relativities estimated by Trubka et al are even close to the mark now.

The implicit assumption of urban consolidation is that there’s spare infrastructure capacity in inner areas. After all, the inner suburbs used to support larger populations 40 or more years ago, the argument goes, when average household size was larger than it is today.

However the frequent stories in the media about traffic congestion, competition for on-street parking, and over-crowded schools suggest that assumption doesn’t hold anymore.

While there are also city-specific factors, there are a number of likely reasons for that conclusion. As I’ve explained before:

  • Most areas within 10 kilometres of the CBD now have the same or higher population as they had three, four or five decades ago;
  • Households today consume more electricity, gas, water and road space than previous generations, e.g. air-conditioning systems, flat-screen TVs, second cars;
  • The sorts of households that locate in the inner suburbs today tend to be much more affluent than their predecessors. As the Australian Conservation Foundation points out, prosperous residents use more resources per capita than average income residents;
  • Average household size is smaller now. The same number of people is spread across more dwellings, meaning economies of scale in infrastructure provision are lower, e.g. it takes almost as much gas or electricity to centrally heat two people living in a dwelling as it does if four live in it; and
  • Some inner-suburban infrastructure has been converted to other uses, e.g. primary schools converted to apartments.

There’s also another critical factor that bears on residential development costs: it’s difficult and expensive to retrofit infrastructure. Inner-suburban areas are more intensively developed, values are higher, sites are smaller, access is harder, opposition from neighbouring land uses is more intense, and the cost of avoiding disruption of other activities is higher.

I favour policies to increase housing supply in inner suburbs and in established areas more generally because there’s evident demand and because it’s likely social costs are lower. It’s vital, though, to have reliable evidence. Right now we don’t know the relative costs of providing development infrastructure in outer and inner areas (and there are city-specific influences too).

Infrastructure Australia and other policymakers shouldn’t rely on shaky data. This is a basic and obvious gap. There’s a serious need for governments and universities to put resources into providing the necessary information to guide policy on a city-by-city basis.