Heterogeneous treatment effects: instrumental variables without monotonicity?

Abstract

"Imbens and Angrist (1994) were the first to exploit a monotonicity condition in order to identify a local average treatment effect parameter using instrumental variables. More recently, suggested estimation of a variety of treatment effect parameters using a local version of their approach. We investigate the sensitivity of respective estimates to random departures from monotonicity. Approximations to respective bias terms are derived. In an empirical application the bias is calculated and bias corrected estimates are obtained. The accuracy of the approximation is investigated in a Monte Carlo study." [author's abstract]