My entry in Ted Baillieu's (proposed) competition. Flinders St Station - originally planned Swanston St elevation

A new US study has found that there is a significant association between public transport use and reductions in Body Mass Index over time.

This study has excited a lot of attention because it used a before-and-after “natural experiment” (see here, here, and here).

The researchers did an initial “before” telephone survey of residents living within one mile of a proposed new light rail line in Charlotte NC. They followed up with an “after” telephone survey 6-8 months after the new line opened.

There are some major methodological limitations with the study. Respondents self-reported their weight. The initial sample of 839 fell to 498 respondents in the follow-up phase. Only 26 respondents used the new line to commute on a daily basis.

Nevertheless, I have no difficulty with the proposition that those who choose to commute by transit are likely to be thinner than those who choose to drive to work. After all, transit requires the expenditure of more calories on walking and standing than driving does.

But in my view, the key issue is to what extent better health outcomes – and in this context specifically weight reduction – should shape transport policy. In order to look at that issue it is essential to understand what’s driving the “obesity epidemic” in Australia.

As I’ve pointed out before, the main culprit isn’t exercise but what goes in our mouths. You have to walk the dog for an hour and a half, or cycle for an hour, to burn off the calories in just one Big Mac.

This article by Deirdre Macken in last weekend’s AFR notes that Australia’s obesity problem began in the 1980s. She documents some interesting stats on our food supply:

  • The first McDonalds opened in Australia in 1971, and by 1980 there were 105 restaurants and 228 by 1990
  • By 1988, Australia had 1100 fast food outlets and the fast food market was growing at 20% p.a.
  • Sales of pre-prepared foods in supermarkets doubled from 1982 to 1992
  • It wasn’t just McDonalds that super-sized food. Muffins ballooned to three times their traditional size, family packs became common, bread was sliced thicker, the most popular size of chip packets doubled, and soft drink bottles went from 200 ml in the 1960s to 375 ml, and, later, 600 ml.

While what we eat is the key problem, activity levels have nevertheless fallen across-the-board – and not just in commuting. We certainly drive more, but we also spend more hours sitting in front of the TV and more hours in front of the computer.

Portion size - then and now

The vast majority of us operate our electronic gear, air-conditioning, heating and garage doors with remote controls. We spend less effort on preparing our food – pre-packaged food is heated in microwave ovens and much of the time we eat out anyway. Gardens are watered automatically and pools are filtered automatically.

Most jobs are now in the service economy and many are performed from a seat. Farmers harvest from air-conditioned cabs. Loads are lifted and moved with forklifts. Chippies use nail guns and powered screwdrivers. Gardeners use whippersnippers and powered leaf blowers.

The underlying causes of these changes include rising incomes, marketing, technological advances, feminisation of the workforce and ageing of the baby boomer cohort. My theory is that all those smokers who quit the habit and consequently gained weight are somewhere in the mix too.

It’s true that walking to the rail station will burn up more calories. So will using a push mower instead of a motor mower, taking the stairs at work rather than the lift, putting the garage door on manual, or throwing away every remote control in the house. And that’s before considering the most proximate cause – what goes in our mouths.

What should be clear from the foregoing is that the obesity epidemic isn’t something that should be sheeted home to transport or urban form. They’re implicated, but they’re only a part of the deeper problem. The obesity epidemic isn’t going to be solved by higher public transport use or denser suburbs.

In fact my suspicion is that the sorts of people who currently use transit are the sorts of people who, on average, are more likely to eat carefully and be active, anyway. I’ve argued much the same before in relation to the link between density and obesity.

If transit use were increased dramatically, (say) to a 30% share of all travel, it may be that the aggregate health benefits could start to disappear. New users from different demographics to those of current users might get no benefit from transit, perhaps because it would take time away from other healthy activities, or because they would continue bad eating habits, or because they would drive to the station.

By all means add health to the various impacts considered when developing transport policy and evaluating projects, but get its importance right first. It needs to be recognised that there are cheaper and more direct ways of tackling obesity. We also need much better information on the relationship between health and transport. I expect when looked at across the whole population, the health benefits are going to be pretty small compared to other criteria like time savings and environmental impact.

We should also be careful what we wish for. Where the health benefits of (say) walking to transit stops are included in the calculus, it will also be reasonable to count the social costs of time lost in commuting e.g. time away from family or from activities that build social capital. Public transport commutes in Melbourne, for example, are almost twice as long as car commutes, largely because of the time involved in walking to stops.