Four essays on the social structure of urban labor markets

The four chapters of this dissertation present tests of the hypothesis that social structure---the connections and ties between individuals---plays an important role in labor market outcomes. The first chapter is a reassessment of the literature of the effect of contacts-help from friends or relatives to find work---on wages. After the first chapter, the rest of the dissertation focuses on how social and residential segregation affect racial inequality. Because of the prevalent use of contacts to find work and high levels of racial social segregation, a significant amount of job information is transmitted through highly segregated social networks. The second chapter explores the actual degree of racial segregation among labor market contacts. In addition, using a comparison of cross-sectional data and panel data on multiple job searches by the same individual, I discuss the effect of contacts on the racial wage gap. The third chapter looks at the other side of the labor market, using a data set of 2,700 companies in four cities to test whether the use of employee referrals to fill job vacancies generates inter-firm racial segregation among black, white, and Hispanic workers. The final chapter extends the focus from social segregation to residential segregation. Using longitudinal neighborhood-level data from Detroit and Chicago, I present an empirical test of the spatial mismatch hypothesis, which argues that employment decentralization and persistent residential segregation combine to adversely affect the labor market outcomes of black workers.