Fabian Böhme - Effects of the German tax-benefit system on labour force participation of unemployed welfare benefit recipients

  • Presenting author: Fabian Böhme (IFO)

  • Authors: Fabian Böhme, Kerstin Bruckmeier, Andreas Peichl

  • Session: C03B - Behavior [2] - Wednesday 14:00-15:30 - Marietta-Blau Hall

Germany has one of the lowest unemployment rates in the European Union. Despite significant macroeconomic risks, such as the Russia-Ukraine war, the German labour market is generally robust with a high number of people in employment and a large number of job vacancies. However, while the shortage of skilled workers is increasingly becoming a problem for economic development, long-term unemployment remains a concern. Long-term unemployment rose significantly for the first time in almost a decade after the Corona crisis during the year 2021 and has remained at higher pre-crisis levels since then. In 2023, almost 900 thousand people have been officially registered as unemployed for more than a year.

The long-term unemployed are not covered by unemployment insurance benefits, but receive

benefits under the basic income scheme for job-seekers (until 2022 ALG II) in Germany. High

long-term unemployment combined with high labour demand has brought the labour supply of benefit claimants into the focus of the political discussion. The monetary incentives provided by the ALG II benefit system are often seen as one obstacle contributing to only a slow decline in long-term unemployment.

In this context, our paper analyses the impact of monetary incentives inherent in the German

Tax-Benefit-System on welfare recipients’ labour supply. A large number of studies from the

optimal taxation literature examines the relationship between labour supply and taxation. Empirical studies focus on the link between work incentives set by taxes and benefits and individual labour market participation. To measure incentives, effective marginal tax rates and participation tax rates resulting from the tax-benefit-system are simulated for various hypotheticaljobs with different working intensity. Usually studies make use of survey data and microsimulation models, e.g. EUROMOD, to simulate participation tax rates (unemployed) individuals are faced with when taking up employment. We contribute to this literature by presenting for the first time results for Germany based on extensive, large administrative data.

To this we apply the newly developed open-source German tax-benefit-microsimulation model ‘GETTSIM’ to administrative data of welfare recipients. ‘GETTSIM’ is developed by several researchers from different Universities and Institutes.1 We use ‘GETTSIM’ to simulate counterfactual net incomes for unemployed welfare recipients based on the information provided in the administrative data. The administrative data stems from the Sample of Integrated Welfare Benefit Biographies (SIG) of the Institute for Employment Research (IAB) of the German Federal Employment Agency (BA). We use a randomly drawn 10% sample of recipients of ALG II and observe individuals between the years 2017 and 2019.

The data contains information on the benefit claimant himself as well as on his family members and the household income. This is due to the fact that ALG II is means-tested in the household context.

Our simulation results show high participation tax rates, which are above the tax burdens found in the previous literature for Germany. The reason for this finding is the relevance of high benefit withdrawal rates within ALG II and other means-tested benefits. In addition, we find heterogeneous results in the participation tax rates across different household types and years. The simulated participation tax rates are then used as an explanatory variable in a regression analysis that models the transition from unemployment to employment. We find mixed evidence for the effect of the participation tax rates, and hence monetary incentives, both in terms of statistical significance and magnitude of the estimated effect. The application of different estimation methods points to the importance of the consideration of individual unobserved heterogeneity and the possible endogeneity of the participation tax rate.