Does urban sprawl cause obesity and unhealthy habits?
If public universities cap salaries for their Presidents, will they recruit less able leaders? The California State Universities may soon be running this experiment. The Chronicle also has an interesting article on my UCLA colleague Dick Jackson. Dr. Jackson is a leading scholar at UCLA's School of Public Health. Here is a quote from the Chronicle of Higher Education:
"His center had already been dealing with problems that he suspected had origins in the built environment—asthma caused by particulates from cars and trucks, water contamination from excessive runoff, lead poisoning from contaminated houses and soil, and obesity, heart conditions, and depression exacerbated by stressful living conditions, long commutes, lack of access to fresh food, and isolating, car-oriented communities. Treatments could come in the form of pills, inhalers, and insulin shots, but real solutions had bigger implications. "More and more, I came to the conclusion that this is about how we build the world that we live in," he recalls, speaking over the phone from San Francisco."
Can better urban design make us healthier?
This raises a fundamental selection versus treatment question. Put bluntly, do people with a proclivity to be sick self select to live in nasty neighborhoods featuring bad air quality, little access to good public transit and shopping opportunities? Or do, people who for random reasons choose to live in those areas subsequently become sick? The first is a selection effect while the second is a treatment effect.
Consider the example of smoking; if we observe that smokers tend to suffer from sickness --- again is this selection or treatment or both? Do the most impatient people in society smoke and such individuals tend to under invest in their health and subsequently are more likely to become sick or does smoking have an independent treatment effect on making you sicker? The answer is that both effects are likely to be playing out.
I find that public health researchers tend to ignore fundamental issues of self selection on unobserved attributes (one of Jim Heckman's early homeruns) and implicitly assume that households are randomly assigned across space so that any observed differences in outcomes are due to "treatment effects". In the absence of randomized trials (where residential locational choice is determined at random), this is a hard topic to work on.
Matthew Turner and co-authors have written an under-appreciated paper that was published in the Journal of Urban Economics. Here is the abstract of their paper titled "Fat City":
"We study the relationship between urban sprawl and obesity. Using data that tracks individuals over time, we find no evidence that urban sprawl causes obesity. We show that previous findings of a positive relationship most likely reflect a failure to properly control for the fact the individuals who are more likely to be obese choose to live in more sprawling neighborhoods. Our results indicate that current interest in changing the built environment to counter the rise in obesity is misguided."
Intuitively, Turner estimates a fixed effects regression using panel data where he tracks the same person over time for people who move from the center city to the suburbs or vice versa. If sprawl makes us fat, then the average person who moves from the center to the suburbs should be gaining more weight over time than the people who never leave the center city or never leave the suburbs. Turner rejects this hypothesis.
So, there is plenty of work to be done here but it remains an open question of how urban form affects our behavior. I've been especially interested in this question focused on our carbon footprint as a function of urban form.