A quantile regression analysis of educational returns and income inequality

This paper starts from the methods of caculation and decompostion of Gini coefficient, in which we take urban pupulation as an increasing function of education, and analyzing the composition effect and compression effect of education on income inequality. We make a simulation of this model, using OLS and quantile regression method, and find that there is an inverse-U shape relationship between income inequality and education. We reckon that one cannot take education as a main cause of income disparity and income inequality, just because the educational returns of higher income group are larger than that of the lower income group. Our analysis focuses that, even if the income disparity between urban and rural keep expanding in the future, the improving of educational level will accelerate the transion of population which finaly reduce the urban-rural income inequality. In a word, the improving of educational level will finaly reduce the urban-rural income disparity and income inequality.