Sex Segregation, Labor Process Organization, and Gender Earnings Inequality

This article revisits Tam's finding that occupational sex composition does not influence wages. This problem is approached in two quite different ways. First, a potential conceptual and methodological weakness in all research that focuses on national occupational, rather than local job and organizational, processes is pointed out. Second, the implications of organizationally relevant social closure and gendered labor process theories for our understanding of wage determination models is developed. The gendered devaluation and specialized human capital theories, which are stressed by Tam and his critics, do not represent the entire story. We find that the sex composition effect on wages exists, but it is indirect and relatively weak, operating largely through lower access of typically female jobs to extensive training. There is no strong evidence for the existence of a more generic gendered labor process in these cross‐sectional data. The evidence for social closure processes in this article is limited to the gendered nature of access to on‐the‐job training.