Can Urban Planning Reduce Obesity? The Role of Self-Selection in Explaining the Link between Weight and Urban Sprawl

I the United States, obesity has risen at an epidemic rate during the past 20 years (Flegal et al.). Recent studies by urban planning and public health researchers find that development patterns associated with urban sprawl are a significant determinant of obesity rates in the U.S. population (Ewing et al.; Frank, Andresen, and Schmid; Lopez). Several explanations are offered. First, poor connectivity in street networks and low-density and single-use development increase trip distances and reduce physical activity by making walking and bicycling impractical and unsafe. Second, low-density development reduces the viability of public transportation. As a result, people must commute to and from work by automobile, which can lead to traffic congestion and the diversion of time from activities such as exercising. Third, modern suburban developments do not adequately provide facilities such as parks that permit and encourage physical activity. An important assumption in the urban planning/public health studies is that urban form is exogenous to an individual’s weight.1 Thus, the researchers consider how an individual is affected by the pattern of land use where they reside, but do not account for the individual’s choice of residence. If weight affects an individual’s choice of the type of neighborhood to reside in, then the sprawl measures in earlier studies are endogenous regressors and the coefficient