Compact Development and BMI for Young Adults: Environmental Determinism or Self-Selection?

Problem, research strategy, and findings: The literature widely reports a statistical association between the built environment and obesity. What is less clear is the reason for the association. Is it environmental determinism—the effect of the built environment on individual behavior—with compact places inducing more physical activity and hence lower weight? Or is it self-selection, the tendency of healthy-weight individuals to select to live in compact places where they can be more physically active and possibly the tendency of overweight or obese individuals to opt for sprawling places? Both theories have been promoted in the literature. In this study we seek to address this issue using data from the National Longitudinal Survey of Youth. We study body mass indices (BMIs) of survey participants, all young adults, at two points in time and follow them longitudinally for 9 years as they move from place to place. We estimate models for the entire cohort and also for young adult movers and stayers separately. We find more evidence of self-selection than of environmental determinism. First, we find that compactness is not significantly associated with BMI in young adults for those staying in the same place for the entire period. Second, we find no significant association between changes in sprawl and the changes in BMI for the cohort of young adult movers. Third, our longitudinal analysis shows that young adults who are not overweight tend to move in the direction of greater neighborhood compactness, whereas overweight young adults tend to move in the direction of greater sprawl. Because young adults are at a unique stage in the life cycle, these findings cannot be generalized to other cohorts. Takeaway for practice: These findings do not suggest that place characteristics are unimportant; planners need to meet the growing demand for walkable, compact, and connected places, which are currently undersupplied. Still, they do not provide evidence that place characteristics influence the behaviors of young adults and hence their weight. Rather, they suggest that compact places appeal to more than half of the young adult population and that this latent demand should be met by providing dense, diverse, and well-designed residential options, even in the suburbs, through regional transportation plans and local planning and zoning. Education for primary and secondary school students, who are about to enter young adulthood, could also play a role.