Discrimination over the Life Course: A Synthetic Cohort Analysis of Earnings Differences between Black and White Males, 1940-1990

Abstract

Using data from the 1940-1980 United States Microdata Extract File and the Current Population Survey's 1990 Annual Demographic File, this study examines three competing explanations of the disparity in black and white male earnings over the life course. The “legacy of discrimination” explanation suggests that current racial disparities in earnings reflect nothing more than past discrimination against older blacks, and that the earnings of younger black and white males should be similar over the life course. The “cumulative effects of discrimination” explanation suggests that the black-white earnings gap increases over the life course, and that this divergence in earnings exists for younger cohorts of males as well as older cohorts. The “vintage hypothesis” argues that the net black-white earnings gap reflects differences in self-investments in human capital and that the racial earnings gap should be virtually constant over time and over the life course for all cohorts (vintages) of black and white male workers. This study presents a synthetic cohort analysis of the effects of aging on the disparity in earnings for black and white males from 1940-1990. It shows that aging has a curvilinear effect on the black-white earnings gap. Younger black males in each year and cohort analyzed were closer to their white male counterparts than middle-aged blacks. However, there was a convergence in the earnings of elderly black and white males. While not completely consistent with any of the formulations, the findings most closely conform to the predictions of the cumulative effect of discrimination explanation.