Exploiting Regional Treatment Intensity for the Evaluation of Labor Market Policies

Abstract

We estimate the effects of active labor market policies (ALMPs) on subsequent employment and earnings by nonparametric instrumental variable estimators. Very informative administrative Swiss data with detailed regional information are combined with exogenous regional variation in program participation probabilities to generate an instrument within well-defined local labor markets. We find that implementation of an ALMP increased individual employment probabilities by about 15% for unemployed that might be considered “marginal” participants.