Identifying Effects of Multivalued Treatments

TL;DR

This paper develops a method to identify treatment effects in complex models with multidimensional unobserved heterogeneity, using threshold-crossing rules and continuous instruments, expanding beyond traditional assumptions.

Contribution

It introduces a novel identification strategy for multivalued treatments that accommodates multidimensional unobserved heterogeneity and generalizes existing models.

Findings

Identification achieved under threshold-crossing and continuous instruments

Applicable to various classes of multivalued treatment models

Extends beyond ordered choice and monotonicity assumptions

Abstract

Multivalued treatment models have typically been studied under restrictive assumptions: ordered choice, and more recently unordered monotonicity. We show how treatment effects can be identified in a more general class of models that allows for multidimensional unobserved heterogeneity. Our results rely on two main assumptions: treatment assignment must be a measurable function of threshold-crossing rules, and enough continuous instruments must be available. We illustrate our approach for several classes of models.