Objectives: Residential segregation is one of the fundamental features of health disparities in the United States. Yet little research has examined how living in segregated metropolitan areas is related to cognitive function and cognitive decline with age. We examined the association between segregation at the metropolitan statistical area (MSA) level and trajectories of age-related cognitive function. Method: Using data from Black and White older adults in the REasons for Geographic and Racial Differences in Stroke study (n = 18,913), we employed linear growth curve models to examine how living in racially segregated MSAs at baseline, measured by the degree of non-Hispanic Black (NHB) isolation and NHB dissimilarity, was associated with trajectories of age-related cognitive function and how the associations varied by race and education. Results: Living in MSAs with greater levels of isolation was associated with lower cognitive function (b =-0.093, p <. 05) but was not associated with rates of change in cognitive decline with age. No effects of living in isolated MSAs were found for those with at least a high school education, but older adults with less than a high school education had lower cognitive function in MSAs with greater isolation (b =-0.274, p <. 05). The degree of dissimilarity was not associated with cognitive function. The association between segregation and cognitive function did not vary by race. Discussion: Metropolitan segregation was associated with lower cognitive function among older adults, especially for those with lower education living in racially isolated MSAs. This suggests complex associations between individual socioeconomic status, place, and cognitive health.

We study a labor market characterized by search frictions in which firms choose between posting non-negotiable wage offers and bargaining wages with individual workers. We use the model to study the positive and normative implications of heterogeneous wage-setting strategies in labor markets, as well as the potential effect of policies that seek to regulate wage-setting. We analytically derive-and empirically validate-a testable prediction from the model regarding the cross-sectional prevalence of bargaining and renegotiation of wages among workers. We then estimate the model and use it to evaluate counterfactuals in which either wage-setting procedure is mandated. We find that eliminating bargaining reduces the overall gender gap in wages by 6%, the education gap by 3%, and residual wage dispersion by 12%, while leading to welfare losses for workers. Similar numbers are observed when bargaining is mandated, with ensuing welfare gains for workers. Either policy raises output by 1-3% by eliminating inefficient job mobility, but accounting for firm responses in vacancy creation can overturn these effects.