The structure of wages in Chile 1960 - 1996: an application of quantile regression

This paper applies newly developed techniques in the estimation of earnings functions to compute private rates of return for education in Chile. Although studies concerning the rate of return for education in Chile are available, the quantile regression method used here permits a conditional breakdown of the wage structure that is of crucial importance in an economy which has experienced important changes in political regimes that have had different impacts on different groups of the population. The results show that: (i) mean and median wage estimation equations are stable; (ii) the differences in the rates of return to education by quantiles are systematic; (iii) wage inequality is more volatile than what we usually believe; and (iv) the labor market is more heterogeneous than originally thought.