Macroeconometric evaluation of active labour market policies in Germany - a dynamic panel approach using regional data

Most evaluation studies of active labour market policies (ALMP) focus on the microeconometric evaluation approach and work with individual data. The main question is if the interesting outcome variable for an individual is affected by the participation in an ALMP programme. That being done, the direct gain can be compared with the associated costs and the success of the programme can be judged. However, microeconometric approaches estimate in nearly all cases the effect of treatment on the treated, ignoring impacts on the non-participants. Therefore this type of analysis should only be seen as a first step of a complete evaluation, because general equilibrium effects are neglected. These are very likely to occur regarding the immense amounts spend on ALMP in Germany. One possible solution to overcome the pitfalls of the microeconometric approach is to carry out a macroeconometric analysis. Instead of looking at the effect on individuals performance we would like to know if the ALMPs represent a net gain to the whole economy. This is likely to be the case only, if the total number of jobs is positively affected by labour market policies. Empirical work on the macroeconomic effects of ALMPs is rare and this might be due to several reasons. The first obstacle is the absence of an obvious theoretical framework within which to couch the analysis. Leaving the traditional way of 'cheating the Phillips curve' aside we use a standard labour market framework to consider the effects of ALMP on the matching, the job creation and the wage setting process. A second problem which has to be solved is the inherent simultaneity problem. Spending on ALMP should influence the labo market situation but might also be determined by it. One way to solve this, is an instrumental variable approach, where the issue of finding the right instruments is of major importance. A further problem is the availability of suitable data, whic allows to take regional heterogeneity into account. Especially in Germany this is problematic due to permanent adjustments in th regional delimitations of the labour office districts (Arbeitsamtsbezirke). In contrast to other evaluation studies we use adjusted data, which is corrected for these changes. This should allow us to obtain more accurate results. The aim of the study is to add a new perspective to the evaluation of ALMP in Germany. This is done by using regional data to obtain macroeconomic or net-effects of these measures. We hope to give advice for future policy implementation.