Decomposing Identification Gains and Evaluating Instrument Identification Power for Partially Identified Average Treatment Effects

TL;DR

This paper analyzes how instrumental variables contribute to identifying average treatment effects in partially identified models, providing a decomposition framework and practical insights for IV selection and relevance detection.

Contribution

It introduces a novel decomposition of IV contributions to ATE identification, combining graphical, simulation, and empirical methods to enhance understanding and practical application.

Findings

Decomposition of IV contributions clarifies their roles in ATE identification.

Simulation results suggest methods for detecting irrelevant instruments.

Empirical analysis illustrates the decomposition's practical utility.

Abstract

This paper examines the identification power of instrumental variables (IVs) for average treatment effect (ATE) in partially identified models. We decompose the ATE identification gains into components of contributions driven by IV relevancy, IV strength, direction and degree of treatment endogeneity, and matching via exogenous covariates. Our decomposition is demonstrated with graphical illustrations, simulation studies and an empirical example of childbearing and women's labour supply. Our analysis offers insights for understanding the complex role of IVs in ATE identification and for selecting IVs in practical policy designs. Simulations also suggest potential uses of our analysis for detecting irrelevant instruments.