Misspecification and the Propensity Score: The Possibility of Overadjustment ∗

The popularity of propensity score matching has given rise to a robust, sometimes informal, debate concerning the number of pre-treatment variables that should be included in the propensity score. The stan-dard practice when estimating a treatment effect is to include all avail-able pre-treatment variables, and we demonstrate that this approach is not always optimal when the goal is bias reduction. We characterize the conditions under which including an additional relevant variable in the propensity score increases the bias on the effect of interest across a variety of different implementations of the propensity score method-ology. Moreover, we find that balance tests and sensitivity analysis provide limited protection against overadjustment