In this paper, I examine the school attainment of foster children using data from the Cape Area Panel Study (CAPS), collected in the Metropole of Cape Town. Based on cross-sectional data, I show that foster children have a lower grade attainment level than children who live with both their parents, even when they reside in the same household. In particular, double orphans and children with two absent parents perform relatively poorly. However, these disparities may be driven by unobservable factors that differ between foster and biological children. ... Based on an individual fixed effects analysis in which children are examined between the ages of 7 and 17, there is no evidence that separating from parents has a significant effect on the probabilities of advancing from the previous grade. However, non-constant effects of fostering appear to be present.
Includes abstract.