« Read This Outdoors | Main | $10 billion? That's nothing! »

September 15, 2005

Charming 3-bedroom, 2 bath with only 5 pounds of weight gain a year

Obesity has many causes, but current research indicates that sprawl may play a part. So far, researchers have concluded that people who live in sprawling, car-dependent neighborhoods are more likely to be obese, while people who live in walkable neighborhoods are apt to do more walking. But researchers still trying to tease out cause and effect: do walkable neighborhoods encourage people to walk? Or do people who like to walk move to walkable neighborhoods? I suspect it is both.

However, one recently study claims that sprawl doesn't, in fact, cause obesity. The authors, who base their findings on a complex theoretical model, don't dispute that sprawl and obesity are linked. But they claim that people who move to sprawling neighborhoods are simply making a more-or-less conscious choice to put on more weight:

"[R]esidents are willing to accept locations that result in weight gain because they face lower housing prices and can purchase more housing."

In other words, people buying a house don't mind putting on a few extra pounds in order to get the house they want.

But here's the catch: the link between sprawl and obesity just isn't that widely known. Researchers have just been started exploring these connections over the last few years; the literature is growing, but it's still in its infancy. So it's hard to imagine that most home-buyers, over the last several decades, were weighing the concrete effects of neighborhood design on their health. How could they, when the information just didn't exist?

While I haven't worked through the entire model line-by-line, I also have some quibbles with the methods. As with any theoretical model, it's based on a host of assumptions.  And one of them -- that calorie consumption increases as income increases -- clearly poses problems. As Clark blogged about a while ago, food is really cheap, and cheapest foods are the most calorie dense (think greasy fast food). So there's ample reason to believe that calorie consumption could actually increase as income goes down. 

Theory can be very useful; and there may well be some truth to the notion that people who don't like exercise don't mind living in places that discourage walking. But in this case, only studies that track actual people over time as they move among actual neighborhoods will yield reliable answers.

Posted by Jessica Branom-Zwick | Permalink

TrackBack

TrackBack URL for this entry:
http://www.typepad.com/services/trackback/6a00d834573a7069e200d8348c9c5e69e2

Listed below are links to weblogs that reference Charming 3-bedroom, 2 bath with only 5 pounds of weight gain a year:

Comments

Excellent synopsis, Jessica. Here's the ref for scholarly types:

Plantinga and Bernell 2005. A Spatial Economic Analysis of Urban Land Use And Obesity. Journal of Regional Science 45:3 pp. 473–492.

IMHO this is an important paper. This paper is valuable in at least three ways (only 3 to save bandwidth):

1. It offers an alternative explanation for obesity in our spatial arrangements. Not a complete one, but an explanation nonetheless and something to think about. This means the sprawl-obesity picture is not complete and we must search harder for linkages.

2. It offers - as is the case for most microecon papers – a partial explanatory variable for human behavior. It is not complete, but it is something to consider when we think about the sprawl-obesity issue.

In this case, the paper attempts to explain how people value time. We must remember to consider what tradeoffs people make when we think about spatial arrangements - e.g. as the paper explains, those that place more importance on appearance would be constrained by a house in the suburb, because they spend more time in their car and thus have less time to exercise. If this is true, sprawl has an indirect effect on obesity, as folk have choices to exercise but choose not to - in micro terms, their disutility is not so great as to cause a change in behavior or to rearrange their schedules to exercise.

This time thing is important as we think about spatial arrangements - if we have a TND in the 'burbs but the residents are in their cars for 2 hours, is the TND going to make them walk if all their time is taken?

3. It reminds us that there is a segment of society that wishes to be separate from others, if only a little bit. As the paper explains, folks sort themselves based on preferences, esp. public goods provided by local governments - schools, sewer, roads. If land rents are comparatively low, larger houses can be built. Which is what you see in many sprawling neighborhoods. Big houses on big lots, and when you open your bathroom window your neighbor isn't 6 feet from your nose. I hear that a lot out here. Maybe isolated people don't feel the need to exercise, or they don't want to exercise anyway so they move out to large houses on large lots.

Failing to consider these preferences makes it difficult to justify policy options, and this paper illuminates how some of these choices are made.


Also, it helps to consider the approach the authors took: the assumption is that our time is completely booked, and how we choose to spend our leisure time affects how we sort ourselves (Tiebout sorting in my 3. above).

It's important to consider that homeowners aren't *explicitly* weighing neighborhood design per se, but they are looking at separation from other homes - for privacy and quietude. Living in a quiet suburb means our chances of seeing people are lower than in a crowded, noisy city. Seeing fewer people means our time expenditure considerations are different: there's less of a social imperative for physical fitness (this is where we can argue it's possible our built environment enables sedentary living and isolationism).

Thus, thru this paper, we indirectly see the effects of the built environment. The microeconomic implication is that if there is extra time in the suburb, will agents (folks) choose exercise? This can be tested empirically. And the authors imply this test will further illuminate whether there is a sprawl=obesity link - a link that, in my mind, is tenuous.


Posted by: Dan Staley | Sep 15, 2005 4:41:05 PM