Forecasting Behavioural and Distributional Effects of the Bofinger-Walwei Model using Microsimulation

Summary Since Germany’s social assistance reform (“Hartz-IV-Reform”) in 2005 there has been a strong increase in the number of working poor and long-term unemployed. This development is often attributed to the remaining disincentives of the reformed social assistance to take up a low-paid full time job. Therefore, several proposals have been worked out to reduce these disincentives. In this paper we analyse an in-work benefit programme considered by the German government, which follows the proposal of Bofinger et al. (2006). We employ a microsimulation model for estimating labour supply as well as distributional and fiscal effects of this reform proposal.We provide “morning after effects”, i.e. fiscal effects without considering behavioural adjustments, and long run effects, which take into account the labour supply response following the introduction of the reform.We predict the labour supply responses by estimating a discrete choice model for different household types and find a moderate increase in labour supply (103,000 full-time equivalents) as well as overall low negative participation effects. The distributional analysis reveals an overall increase in poverty rates caused by lower earnings disregards as well as substantial deadweight losses, since a large part of the in-work benefit accrues to households who do not belong to the working poor in the status quo.