[Journal:] Zeitschrift für ArbeitsmarktForschung - Journal for Labour Market Research [ISSN:] 2510-5027 [Volume:] 39 [Issue:] 3/4 [Publisher:] Institut für Arbeitsmarkt- und Berufsforschung (IAB) [Place:] Nürnberg [Year:] 2006 [Pages:] 477-490
Institut für Arbeitsmarkt- und Berufsforschung (IAB), Nürnberg
The so-called Hartz reform introduced fundamental changes for publicly sponsored training in Germany. This concerns primarily the issuing of training vouchers as an exclusive allocation device, allowing potential participants free choice between training providers within a given training target range. In addition, the issue of training vouchers is made conditional on a job-seeker having sufficient re-employment prospects. We make a comparative examination of the impact of participation for the most important type of training program in quantitative terms before and after the reform by using administrative data from the Federal Employment Agency. Based on conditional propensity score matching, we investigate average treatment effects by means of non-parametrically estimated state probabilities and semi-parametrically estimated transition rates. Our results show that participation in training programs already had a positive impact on employment prospects before the reform. Results for the relatively short observation period since the reform point to a further increase in effectiveness. The driving force behind this seems to come from reducing the duration of training programs, which has decreased the lock-in effect without affecting the program effect.