Decomposition of Differences in Distribution under Sample Selection and the Gender Wage Gap

TL;DR

This paper develops methods to decompose outcome distribution differences between groups considering self-selection, applying them to analyze the gender wage gap and identifying participation and self-selection as key factors.

Contribution

It introduces new decomposition techniques for outcome distributions accounting for self-selection, with a focus on both participants and the entire population.

Findings

Self-selection significantly influences the gender wage gap.

Changes in female participation have reduced the wage gap.

The proposed methods enable valid inference using quantile regression.

Abstract

I address the decomposition of the differences between the distribution of outcomes of two groups when individuals self-select themselves into participation. I differentiate between the decomposition for participants and the entire population, highlighting how the primitive components of the model affect each of the distributions of outcomes. Additionally, I introduce two ancillary decompositions that help uncover the sources of differences in the distribution of unobservables and participation between the two groups. The estimation is done using existing quantile regression methods, for which I show how to perform uniformly valid inference. I illustrate these methods by revisiting the gender wage gap, finding that changes in female participation and self-selection have been the main drivers for reducing the gap.