Counterfactual decomposition of changes in wage distributions via quantile regressions with endogenous covariates
Elena Martinez-Sanchis1, Juan Mora1, Ilker Kandemir
1University of Alicante, Spain, 2University College London, UK
Counterfactual decompositions allow the researcher to analyze the changes in wage distributions by discriminating between the effect of changes in the population characteristics and the effect of changes in returns to these characteristics. In this paper we derive counterfactual distributions by recovering the conditional distribution via a set of quantile regressions, and correcting for the endogeneity of schooling decisions using a control function approach. With this methodology we analyze the sources of the changes in wage distribution that took place in the United States between 1983 and 1993, using proximity to college for different parental background as instruments. Our proposal allows us to obtain the counterfactual changes in the wage distribution that would have prevailed if there had been a change in the distribution of the unobserved ability affecting both years of education and earnings conditional on schooling. Our results show that this change had a negative effect on wages at the low quantiles, which almost compensates the positive effect of the observed change in the schooling distribution. Although we allow for the distribution of the unobservables to depend on observable covariates, we find that the change in the conditional distribution of the residuals accounts for most of the increase in wage inequality during the eighties in the United States.
View full paper