Abstract:
Persistently sizeable unemployment attracts interest to active labor market policy as an instrument to reduce unemployment. Moreover, sustainable economic growth requires an effective re-training system, a part of which is usually associated with state employment offices’ programs. Little is known, however, about the effects of active labor market programs (ALMPs) on the unemployed in Russia. The paper is the first attempt to shed some light on effectiveness of ALMP in Russia from micro perspective. The influence of ALMPs on the probability of re-employment is estimated using administrative individual-level data from employment service register on two Russian regions. Overall and group treatment effects of the programs are estimated using the nonexperimental exact matching approach. Two cases - assuming that the first program has the major effect (single program participation) and examining sequences of programs (multiple program participation) – are considered. A matching design allowing taking advantage of duration nature of administrative data to compensate for informational restrictions associated with the dataset is proposed. We find that the programs under consideration seem to prolong the unemployment spells in one of the regions, and help to leave unemployment quicker in the other, with the size of the effects differing 3-5 times. The sizable difference in treatment effects prompt for substantial institutional differences: there seems to be high discretion in interpretation of employment service role in the local labor market revealed in procedures of program assignment.