Decomposing wage residuals: Unmeasured skill or statistical artifact?

Abstract

The decomposition of wage residuals into standard deviation and percentile ranks can be misleading because the two measures are not necessarily independent. With rising wage inequality, the mean percentile rank of low-wage groups will rise simply because more dispersed distributions have thicker tails. This interpretation is consistent with the observed stability of gender and racial wage gaps. In contrast, the unmeasured skill interpretation of wage residuals would predict widening wage gaps in the face of rising wage inequality, unless one posits an increase in the level of unobserved skill for women and blacks.