Abstract
Estimation of causal effects in controlled or natural “experiments with imperfect compliance” usually relies on an Instrumental Variable strategy, which can yield imprecise and uninformative inference when compliance rates are low. We tackle this problem by proposing a Test-and-Select estimator that exploits covariate information to restrict estimation to a subpopulation with non-zero compliance. We derive the asymptotic properties of our proposed estimator under standard and weak-IV-like asymptotics. We provide conditions under which it dominates the usual 2SLS estimator in terms of precision. Under an assumption on the degree of treatment effect heterogeneity, our estimator remains first-order unbiased with respect to the Local Average Treatment Effect estimand. We illustrate finite sample properties, robustness to treatment effect heterogeneity and precision gains using Monte Carlo simulations. Applying our methodology to the returns to schooling example, we document that a reduction in standard errors of 12% to 48% depending on specifications.
The seminar will be held in room 1249 (12th floor) at Eilert Sundts Hus. The address is Moltke Moes vei 31.