Sebastian F. Daza , University of Wisconsin-Madison
Alberto Palloni, University of Wisconsin-Madison
Despite substantial research, the drivers of the widening gap in life expectancy between rich and poor in the U.S. —the so-called longevity gap— remain unknown. Recent research has suggested that contextual income mobility (e.g., county-level socioeconomic mobility) may play an essential role in explaining the longevity gap. However, previous studies have used mostly aggregate or cross-sectional data to examine the link between exposure to a given income mobility regime and health and mortality. Some of the critical issues of those studies are the ecology fallacy associated with aggregate analyses, but also the overlook of selection processes related to residential mobility over time. This paper aims to extend previous research by estimating the effect of average exposure during childhood and adolescence on health during young adulthood. For that, we use both the National Longitudinal Survey of Youth 1997 (NLSY) and the Panel Study of Income Dynamics (PSID) with geocode data to assess the link between county-level income mobility (Chetty's estimates) and health outcomes and behaviors such as health self-report, BMI, depression, and smoking. Also, we use cohorts that match better the ones used by Chetty's estimates of income mobility in the U.S. (1980-1982) and account for selection and time-varying confounders using marginal structural models (MSM). Thus, we provide a more precise assessment of the hypothesis that exposure to income mobility may determine health later in life and explains the longevity gap.
Presented in Session 94. Educational and SES Differences in Health and Wellbeing Over the Lifecourse