Treatment Evaluation at the Intensive and Extensive Margins

TL;DR

This paper develops a method to evaluate treatment effects in selective samples without instruments or parametric assumptions, using sharp bounds and machine learning estimators, especially accounting for units indifferent to treatment.

Contribution

It introduces a novel approach for treatment effect bounds under conditional monotonicity, accommodating indifference units and high-dimensional covariates with efficient estimators.

Findings

Identifies the prevalence of indifferent units in labor market policies.

Provides sharp bounds for treatment effects under minimal assumptions.

Demonstrates the effectiveness of machine learning estimators in this context.

Abstract

This paper provides a solution to the evaluation of treatment effects in selective samples when neither instruments nor parametric assumptions are available. We provide sharp bounds for average treatment effects under a conditional monotonicity assumption for all principal strata, i.e. units characterizing the complete intensive and extensive margins. Most importantly, we allow for a large share of units whose selection is indifferent to treatment, e.g. due to non-compliance. The existence of such a population is crucially tied to the regularity of sharp population bounds and thus conventional asymptotic inference for methods such as Lee bounds can be misleading. It can be solved using smoothed outer identification regions for inference. We provide semiparametrically efficient debiased machine learning estimators for both regular and smooth bounds that can accommodate high-dimensional covariates and flexible functional forms. Our study of active labor market policy reveals the empirical prevalence of the aforementioned indifference population and supports results from previous impact analysis under much weaker assumptions.