Instrumental Variables with Treatment-Induced Selection: Exact Bias Results

TL;DR

This paper derives exact formulas for instrumental variables (IV) selection bias under various models and procedures, clarifying when IV remains preferable to OLS despite treatment-induced selection bias.

Contribution

It provides the first exact analytic expressions for IV selection bias across multiple models and selection procedures, expanding understanding beyond prior simulation studies.

Findings

IV selection bias varies with conditioning method

Bias due to covariate adjustment is a special case of sample truncation bias

In some models, IV and OLS estimators bound the true effect under selection

Abstract

Instrumental variables (IV) estimation suffers selection bias when the analysis conditions on the treatment. Judea Pearl's early graphical definition of instrumental variables explicitly prohibited conditioning on the treatment. Nonetheless, the practice remains common. In this paper, we derive exact analytic expressions for IV selection bias across a range of data-generating models, and for various selection-inducing procedures. We present four sets of results for linear models. First, IV selection bias depends on the conditioning procedure (covariate adjustment vs. sample truncation). Second, IV selection bias due to covariate adjustment is the limiting case of IV selection bias due to sample truncation. Third, in certain models, the IV and OLS estimators under selection bound the true causal effect in large samples. Fourth, we characterize situations where IV remains preferred to OLS despite selection on the treatment. These results broaden the notion of IV selection bias beyond sample truncation, replace prior simulation findings with exact analytic formulas, and enable formal sensitivity analyses.