Association for Public Policy Analysis & Management

Activation policies such as sanctions, workfare employment and counselling and monitoring schemes have been found to speed up labour market reintegration. At the same time, it has been suspected that this quicker reintegration is paid for with worse job quality, e.g. in terms of lower wages. If this concern holds true, activation policies would push unemployed worker into low-wage traps, hereby contributing to increased scarring effects of unemployment and rising inequalities within the workforce. In spite of a huge body of ALMP evaluations, empirical evidence on the effect of activation programmes on job quality is surprisingly rare. I contribute to this discussion by analysing the effects of a large-scale counselling and monitoring scheme from Germany on quantity and quality of labour market integration, measured by an indicator of being regularly employed and post-unemployment wages.

The programme under discussion is a rather typical counselling and monitoring scheme, which combines more intense job search assistance with the threat of sanctions in case of non-compliance. It has been carried out in Germany between July 2010 and June 2013, and has been one of the largest ALMP programmes from Germany at that time, with more than 138,000 participants scattered throughout the country. To identify the effect on the mentioned outcomes, I rely on administrative data on a treatment and control group. I follow the advice of recent Monte-Carlo-experiments and estimate the programme effect by regression-adjusted matching estimations. Moreover, I exploit the institutional setting and the high quality of the data to conduct placebo tests on endogenous selection and substitution effects.

The results point to a strongly positive effect on employment probability but no effect on wages. After six months, absolute integration rates are estimated to be about three percentage points higher in the treatment group, which translates into a relative effect of 35 percent. At the same time, there is clearly no effect on post-unemployment wages. These findings are robust to changes in the matching algorithm. Moreover, placebo tests refute concerns about endogenous selection or substitution effects. These findings contrast the results from previous research on sanctions, which confirmed a negative effect on job quality. This puzzle suggests that the existence of adverse effects on job quality depends on the type of activation programme. While it may indeed be there for very intense kinds of activation programme, it can be avoided if the right balance between pressuring and supportive components is obtained. Therefore, the major implication for future policy-making and research on activation policies is that more effort should be made in order to identify which kinds and components of activation programmes successfully speed up labour market integration without pushing unemployed worker into precarious employment and low-wage traps.