Tanner Regan (George Washington University)
Andreas Diemer (Stockholm University (SOFI))
Cheng Keat Tang (Nanyang Tech. University)
Abstract: We study the extent of segregation in the social space of urban America. We measure segregation as the (lack of) actual personal connections between groups as opposed to conventional measures based on own neighbourhood composition. We distinguish social segregation from geographical definitions of segregation, and build and compare city-level indices of each. Conditional on residential segregation, cities with more institutions that foster social cohesion (churches and community associations) are less socially segregated. Looking at within-city variation across neighbourhoods, growing up more socially exposed to non-white neighbourhoods is related to various adulthood outcomes (jailed, income rank, married, and non-migrant) for black individuals. Social exposure to non-white neighbourhoods is always related to worsening adulthood outcomes in neighbourhoods that are majority non-white. Our results suggest that social connections, beyond residential location or other spatial relationships, are important for understanding the effective segregation of race in America.
JEL Codes: R23, J15
Key Words: Residential and Social Segregation, Networks, Social connectedness