Does short-term training activate means-tested unemployment benefit recipients in Germany?

This paper estimates for a sample of means-tested unemployment benefit recipients the effects of their participation in short-term training programmes in Germany. We apply propensity score matching and rely on a large sample of treated and controls from administrative data, which in contrast to data used in many comparable evaluation studies is rich in terms of information on household members. We regard a period after the beginning of the year 2005 just after a reform of the means-tested benefit system, which aimed at activating employable people in needy households. Short-term training programmes intensively target such persons. We study whether the programme has an impact on the 'regular employment' rate of the treated. Moreover, we also quantify whether it reduces their job-seeker rate and their rate of unemployment benefit II receipt. We estimated effects for within company and classroom training separately and find that mainly the former programme that establishes a contact to an employer has a considerable impact on the regular employment rate of the participants. The impacts on the other outcome variables are usually weaker. Our analysis considers effect heterogeneity. We generally distinguish between men and women in East and West Germany. But we also regard effect heterogeneity by age, migration background, qualification, unemployment rate, family status/children and time since last job. Both programmes tend to be less effective in particular for people aged younger than 25 years than for others. This may reflect that the programmes are also a tool to avoid that young adults are registered as unemployed for longer than three months.