Heterogeneity in Returns to Schooling in Cameroon: An Estimation Approach Considering Selection and Endogeneity Bias - Research Paper

The aim of this study is to show that the rate of the returns to education is not uniform, and that some people benefit most and others least from their education on the labour market. Using data from a survey on employment and the informal sector, the study provides evidence of the heterogeneity of the returns to education in Cameroon. By controlling for any selection bias attributable to endogenous choices of the employment sector in the school-to-work transition and the potential endogeneity of the education variable related to the individual unobserved heterogeneity, we used the ordinary least squares with robust standard errors and the quantile regression technique to estimate the Mincer earnings function. This estimation procedure based on the control function is suitable because of the robustness of the instruments used. Overall, the study found that the average rate of the returns of an additional year of education was 7.1%. The results of the quantile regression model showed that the returns to education differed according to the earnings quantile considered: they were highest for the highest-paid workers and lowest for the middle-income ones. In addition, the individual unobserved heterogeneity was observed to decrease the returns to education. The Wald test for the equality of coefficients significantly confirmed the heterogeneity of the returns to education by quantile. The study’s findings have many socioeconomic policy implications.