Heterogeneity Analysis with Heterogeneous Treatments

TL;DR

This paper introduces a new decomposition framework and debiased machine learning estimators to analyze effect heterogeneity at the group level, accounting for treatment heterogeneity and revealing underlying causes of observed differences.

Contribution

It develops a novel decomposition method and estimators that disentangle effect heterogeneity from treatment heterogeneity, improving analysis accuracy in complex treatment settings.

Findings

Gender gap in training returns is mainly due to differential treatment targeting.

Women are disproportionately assigned to lower-return vocational tracks.

The proposed methods effectively handle multiple and continuous treatments with limited overlap.

Abstract

Analysis of effect heterogeneity at the group level is standard practice in empirical treatment evaluation. However, treatments analyzed are often aggregates of multiple underlying treatments which are themselves heterogeneous, e.g. different modules of training programs. In these settings, conventional approaches such as comparing (adjusted) differences-in-means across groups can produce misleading conclusions when underlying treatment propensities differ systematically between groups. This paper develops a novel decomposition framework that disentangles contributions of effect heterogeneity and distinct components of treatment heterogeneity to observed group-level differences. We propose debiased machine learning estimators that adapt to many discrete and/or continuous treatments and limited overlap. We revisit a widely documented gender gap in training returns of an active labor market policy. The decomposition reveals that it is almost entirely driven by women being treated differently rather than different returns from identical treatments. In particular, women are disproportionately targeted towards vocational training tracks with lower unconditional returns.