The Color of Activity Space: Redlining and the Racial Segregation of Everyday Places
Racial segregation is conventionally located where people sleep, yet groups meet or miss one another in the places they pass through by day: the stores, clinics, churches, and restaurants that make up their activity spaces. We argue that this segregation in activity spaces is not merely a shadow of residential sorting but a lasting imprint of exclusionary institutions that organized American space by race, and trace it to a paradigmatic case: 1930s "redlining," in which federal appraisers graded neighborhoods in a color-coded hierarchy of mortgage risk steeped in racial assumptions. Using the inferred racial composition of visitors to four million establishments, we compare locations across historical redlining boundaries with undrawn counterfactual boundaries dividing neighborhoods with similar preexisting differences. Nearly a century later, lower-graded neighborhoods host activity spaces more racially segregated from their metros, and redlining explains about 6% of the standard deviation of place-level segregation at these borders. The effect works mainly through neighborhood racial composition: redlining left neighborhoods more Black and less white than their metros today, and visitors—even those from elsewhere—mirror that skew, suggesting racialized preferences over destination choices. A bureaucratic classification long abandoned still organizes whom Americans encounter each day.