Segregated jobs or ethnic niches?. The impact of racialized employment on earnings inequality

Using data from the Multi-City Study of Urban Inequality (MCSUI), this study examines the extent to which the racial or ethnic composition of jobs affects racial and ethnic-linked earnings inequalities among whites, blacks and Hispanics. Four types of jobs are distinguished according to the racial/ethnic composition of jobs in work establishments: predominantly white, multi-racial or mixed, predominantly black, and predominantly Hispanic. We found considerable differences among the four types of jobs. Jobs composed predominantly of white workers are characterized by the highest earnings, the highest status occupations, and the highest levels of education. In contrast, jobs predominantly composed of Hispanic workers are characterized by the lowest salaries, the lowest status occupations, and the lowest levels of education. The data analysis supports the hypothesis that job segregation is responsible for earnings disparities in the case of blacks versus whites, but only partial support for this hypothesis is found in the case of Hispanics versus whites. The analysis also provides support for the "devaluation hypothesis" which suggests that all workers experience pay penalties in jobs in which minority workers are predominant. Further analysis reveals that had most workers been rewarded like whites employed in predominantly white jobs, their earnings would have increased considerably. The only groups of workers who "benefit" from job segregation are Hispanic workers employed in predominately Hispanic jobs. In the absence of competition with others, Hispanics employed in predominantly Hispanic jobs earn more than they would earn in other jobs. The differential effects of the ethnic composition of jobs on economic outcomes of minority populations are evaluated and discussed in light of the roles played by sheltered and protected ethnic economies.