We present evidence on the impacts of education on women's employment outcomes in developing countries using a robust econometric strategy to account for both the endogeneity of education and the selection bias in employment outcomes. Using micro-level data from Ethiopia, we estimate the causal effect of education on the probability that a female worker is wage-employed. The empirical strategy utilises the Ethiopian Government's Free Primary Education reform of the 1990s to obtain exogenous variation in levels education among women. Consistent estimation of the causal effect of education is achieved by combining this instrumental variable approach with the Heckman correction method for selection bias. We find that an additional year of schooling completed increases the probability that a female worker is wage-employed. Interestingly, neither marriage, nor motherhood instrumented by infertility has a significant effect on this probability.
Ouédraogo et al. (2025) studied this question.