Estimating the effect of racial classification on labour market outcomes: A case study from Apartheid South Africa

Abstract

What were the effects of being officially classified as White on labour market outcomes during apartheid in South Africa? South Africa's apartheid government implemented a comprehensive system of discrimination against "non-Whites" that covered every major facet of life. Discrimination in educational opportunities, healthcare, and neighbourhood quality were designed to create productivity differentials across race groups; and these effects would not be included in most estimates of labour market discrimination. We quantify the cumulative effect of all of these forms of discrimination by estimating the causal effect of being classified as White on education, employment and income. Our identification strategy is based on a policy change that privileged ancestry over appearance in the process of racial classification for those born after the 1951 Census. We use census data from 1980, 1991, and 1996, and restrict our sample to Whites and Coloureds. The data exhibits a discontinuity as well as a trend change in racial shares for cohorts born after 1951. Combined, these imply a 6 percentage point lower likelihood of being classified as White for people born 10 years after 1951. Our preferred estimates indicate that being classified as White instead of Coloured resulted in a more than threefold increase in income for men. This corresponds to approximately 65% of the difference in mean incomes between the two population groups. Our findings for women are inconclusive.