% obese persons in Australia by age group: 1995 versus 2004–2005 (source: Preventative Health Taskforce)

A common view among politicians, the media, planners and health professionals is that urban sprawl is a key cause of the modern obesity epidemic. Higher population densities and more walkable neighbourhoods, many argue, are an essential strategy for fighting this scourge of the affluent lifestyle, e.g. see here and here.

The trouble is both propositions are dubious. There are good reasons to pursue higher densities in Australia’s major cities, but addressing obesity isn’t one of them. It’s unlikely tightening the belt on sprawl will have much, if any, impact on the average BMI of Australians.

It’s not really surprising this view of sprawl is so entrenched. After all, there’s plenty of evidence showing suburbanites are generally fatter than inner-city residents. But correlation is not causation, and households aren’t randomly distributed across cities.

An alternative explanation is suburbanites aren’t fatter because they live in the suburbs, but because people who are more likely to be fat self-select into the suburbs. Conversely, people who are likely to be thin self-select into denser neighbourhoods such as the inner-city. Changes in urban form will accordingly have little if any impact on obesity.

This fascinating study published in the leading Journal of Urban Economics, Fat city: questioning the relationship between urban sprawl and obesity, supports the alternative explanation. The research team, which included internationally prominent academics Henry Overman, Diego Puga and Mathew Turner, examined changes in the geographic location of 6000 respondents in the US between 1978 and 1994.

They had access to a remarkable data base, which, among other important variables, provided the precise street addresses, height and weight of respondents at intervals over a 16-year period. They were able to track changes in the location of individuals and construct measures of the neighbourhood density and walkability of each address.

As expected, the authors found residents of more sprawling neighbourhoods are indeed heavier on average than people who live in less sprawling neighbourhoods (although they found this applied to men but not women). However their results strongly suggest urban sprawl does not cause weight gain:

Rather, people who are more likely to be obese (e.g. because they do not like to walk) are also more likely to move to sprawling neighborhoods (e.g. because they can more easily move around by car). Of course the built environment may still place constraints on the type of exercise that people are able to take or the nature of the diet that they consume. The key point is that individuals who have a lower propensity to being obese will choose to avoid those kinds of neighborhoods. Overall, we find no evidence that neighbourhood characteristics have any causal effect on weight.

Observers of Australian cities who argue that density and walkability affect obesity usually compare the inner city with the suburbs e.g. see here. But the two populations are not the same. For example, compared to suburban populations, inner-city residents are more likely to be young, single, have no dependants, have a higher level of education and enjoy a higher income.

It’s easier to be thin when you’re young (see exhibit!), haven’t had a baby, don’t have kids badgering you for fast food, are well informed about nutrition, can afford good food and have the time to cook. These sorts of factors are more likely to explain why on average inner-city residents are thinner, not the fact they live at higher densities (I’ve expanded on this line of thinking before, here, here and here).