Economic returns to schooling in urban China: OLS and the instrumental variables approach

This paper examines economic returns to schooling in urban China using ordinary least square (OLS) and instrumental variable (IV) methodologies. First, we find that OLS estimates of the returns to education are lower in China than in other transition economies, whereas IV estimates are higher in China. Second, we find that OLS, a method for estimating the returns to education without control for endogeneity bias, may underestimate the true rates of return for men. In addition, if we do not control for endogeneity bias and the sample selection bias, we may further underestimate the true rates of return for women. Finally, we find that OLS estimates of the returns to education for men are slightly higher than for women. The IV estimates for women are higher than those for men, and this difference increases after correcting for selectivity biases.