May 2012

IZA DP No. 6565: Endogenous Exit Bias in Training Programs for Unemployed Workers

revised version published as 'The impact of training duration on employment outcomes: Evidence from LATE estimates' in: Economics Letters, 2013, 120 (3), 487-490

This paper assesses the importance of reverse causality when evaluating the impact of training duration for unemployed workers. We use planned duration as an instrumental variable for actual duration. Our results suggest that the potential endogeneity of exits seems to be only relevant in the lower and upper part of the treatment duration distribution.