Year: 2010
Working paper number: 287
Author: Startz, Meredith
Unit: SSU

This paper uses a regression discontinuity design to measure the impact of exemptions from school fees on the desegregation of formerly white and coloured schools in post-apartheid South Africa. Schools which were designated for whites under apartheid often charge high fees, equal to a substantial portion of average black and coloured households' income. Consequently, policy makers have worried that fees act as a mechanism for de facto racial segregation in schools, and have created income-tested fee exemption policies to counteract this effect. However, I find no evidence that eligibility for a fee exemption increases the probability that a black student will attend a formerly coloured or white school, or that a coloured student will attend a formerly white school. I also find no statistically significant relationship between income and school choice for black students, and a small, significant relationship for coloured students. The results of this analysis show that income and fees, contrary to conventional wisdom, may not be key factors in the choice between racially defined school categories. This finding calls into question the relevance of the current fee exemption policy, and suggests that we need to develop a more nuanced understanding of the dynamics of race and inequality in South African schools.

Publication file: WP287.pdf