Going beyond LATE: Bounding Average Treatment Effects of Job Corps Training

Abstract

We derive bounds on the population average treatment effect (ATE) and the average treatment effect on the treated (ATT) with an instrumental variable, and employ them to evaluate the effectiveness of the Job Corps (JC) training program using data from a randomized evaluation with non-compliance. We find positive effects of JC on earnings and employment, and negative effects on public benefits dependence, for eligible applicants (ATE) and participants (ATT). Some of our results also point to positive average effects on the labor market outcomes of “never takers” (individuals who never enroll in JC regardless of their treatment assignment).