Regressão quantílica com correção para a seletividade amostral: Estimativa dos retornos educacionais e diferenciais raciais na distribuição de salários das mulheres no Brasil

We estimate the returns to education for women and the racial wage differential among women over the wage distribution in Brazil by using quantile regression with semiparametric correction for sample selection. Our estimates show that the returns to education are high and that they are not constant along the wage distribution. Both returns to education and the racial wage differentials are higher at higher points of (the conditional) wage distribution. Black women seem to be facing a glass-ceiling in the higher wage segment of the distribution. In addition, quality of education seems also to play some role in the inequality observed among higher paid women. The paper also reveals that using a probit specification to the participation equation in order to correct selection issues yields very similar results to the semiparametric correction for both returns to education and racial discrimination.