Sources of wage inequality: decomposing the conditional Gini coefficient

Abstract

This paper introduces a new econometric method to identify factors influencing the disparities within the distribution of a positive random variable, focusing on US wages between 1986 and 2015. I relate the conditional Lorenz curve to the conditional quantile function to additively decompose the conditional Gini index. Moreover, this paper presents a technique to disentangle temporal changes in the distribution. The analysis shows that despite reduced impacts of race and gender on wages, persistent disparities require ongoing intervention, while higher education, especially college degrees, significantly reduces wage inequality.