Heterogeneous Coefficients, Control Variables, and Identification of Multiple Treatment Effects

TL;DR

This paper develops identification conditions for multiple treatment effects in models with heterogeneous coefficients, using control variables and generalized propensity scores, extending classical binary treatment results.

Contribution

It introduces a novel identification framework for multiple treatments with heterogeneous effects, generalizing Rosenbaum and Rubin's binary treatment results.

Findings

Identification requires boundedness of generalized propensity scores.

Conditions hold for distributional and quantile treatment effects.

Results apply to treatment effects on the treated.

Abstract

Multidimensional heterogeneity and endogeneity are important features of models with multiple treatments. We consider a heterogeneous coefficients model where the outcome is a linear combination of dummy treatment variables, with each variable representing a different kind of treatment. We use control variables to give necessary and sufficient conditions for identification of average treatment effects. With mutually exclusive treatments we find that, provided the heterogeneous coefficients are mean independent from treatments given the controls, a simple identification condition is that the generalized propensity scores (Imbens, 2000) be bounded away from zero and that their sum be bounded away from one, with probability one. Our analysis extends to distributional and quantile treatment effects, as well as corresponding treatment effects on the treated. These results generalize the classical identification result of Rosenbaum and Rubin (1983) for binary treatments.