University of California Transportation Center

Using micro data from the US Census, this paper tests the importance of the spatial isolation of minority and poverty households for youth employment in the largest US metropolitan areas. We first estimate a model relating youth employment probabilities to individual and family characteristics, race, and metropolitan location. We then investigate the determinants of the systematic differences in employment probabilities by race and metropolitan area. We find that a substantial fraction of differences in youth employment can be attributed to the isolation of minorities and poor households. Minority youth residing in cities in which minorities are more segregated or in which minorities have less contact with non-poor household have lower employment probabilities than otherwise identical youth living in similar but less segregated metropolitan areas. Simulations suggest that the magnitude of these spatial effects is not small. It may explain a substantial fraction of the existing differences in youth employment rates for white, black, and hispanic youth.