Testing for unobserved heterogeneous treatment effects

TL;DR

This paper develops nonparametric tests for unobserved heterogeneous treatment effects using instrumental variables, allowing for endogenous treatments, and demonstrates their application with empirical examples.

Contribution

It extends existing testing methods by introducing nonparametric Kolmogorov--Smirnov--type tests that handle endogenous treatments under standard IV assumptions.

Findings

No unobserved heterogeneity in job training effects on earnings.

Significant unobserved heterogeneity in fertility's impact on family income.

Abstract

Unobserved heterogeneous treatment effects have been emphasized in recent policy evaluation literature. In this paper, we extend Lu and White (2014)'s testing method for unobserved heterogeneous treatment effects by developing nonparametric tests under the standard exogenous instrumental variable assumption and allowing for endogenous treatment. Specifically, we propose Kolmogorov--Smirnov--type statistics that are consistent and simple to implement. To illustrate, we apply the proposed test method with two empirical applications: treatment effects of job training program on earnings as well as the impact of fertility on family income. The null hypotheses, i.e., lack of unobserved heterogeneous treatment effects, cannot be rejected at a 10% significance level in the former case, but should be rejected at all usual significance levels in the latter.