Volume 3, Issue 2

This paper gives an overview on the design and the evaluation of Active Labor Market Policy (AAMP) in Germany. While expenditure on AAMP is very high (43 billion DM in 2000), a comprehensive evaluation of its effects is missing. We discuss the methodological problems involved and the data requirements for a convincing evaluation. In the past, policy makers did not appear interested in a scientific evaluation of their sometimes erratic AAMP and no appropriate data were made available. This situation is changing slowly and awareness for the necessity of evaluation is increasing. However, there is the risk that some recent attempts by policy makers to foster evaluation may prove counterproductive.

The paper develops a scenario of an interaction between actual governmental active labour market programmes and econometric policy evaluations. It argues that this scenario could and indeed should be implemented. The paper shows also that realities in Switzerland and Germany are very different from this ideal situation. The good news is that both countries appear to move - albeit slowly - towards that vision. It is however apparent that currently Germany is considerably lagging behind.

Most evaluation studies for Germany's active labor market policy (ALMP) indicate that subsidized employment programs in the public sector (public works programs, „Arbeitsbeschaffungsmaûnahmen“) and publicly funded training programs have, on average, no or even negative effects on individual re-employment probabilities. This paper provides possible explanations for the ineffectiveness of these programs, where we focus on heterogeneous treatment effects, which are not accounted for in the German evaluation studies due to lack of data, and locking-in effects, in particular related to the relatively high level of income support for participants in these programs. Since there is very little direct evidence on these effects for Germany to date, we draw on results from evaluation studies for other European countries. We argue that the success of ALMP is to a large extent determined by design features like the targeting of particular groups and the incentives from the coordination with unemployment insurance as well as the incentives of program administrators and local governments.

This paper takes a fresh look at the often confused - and therefore confusing - debate on the role of wage policy in tackling Germany's unemployment problem. The key issues in this debate are the relative importance of wages and aggregate demand in the determination of employment, the appropriate behavior of wages in relation to productivity growth, and the significance of the purchasing power effect of wages. We argue that the most useful tool to address and resolve these issues is the basic textbook model of aggregate supply and aggregate demand. Most importantly, the model predicts that employment is governed by the ratio of aggregate nominal demand and the nominal wage level. The empirical picture strongly supports this assertion. In contrast, the relation between the rate of wage growth and the rate of productivity growth does not play the causal role in determining employment that is widely attributed to it.