This paper presents non-take-up rates of benefits from the German Income Support for Job Seekers scheme, called Unemployment Benefit II (Arbeitslosengeld II ). Eligibility to these benefits is simulated by applying a microsimulation model based on data from the Socio-economic Panel for the years 2005 to 2014. To ensure the quality of the results, feasible upper and lower bounds of nontake-up are shown for different simulation assumptions. By employing a binary choice framework, determinants of the decision (not) to take-up benefits are studied by means of a cross-sectional probit regression and a fixed effects linear probability model. The findings of this study indicate that rates of non-take-up are substantial and time-stable in the decade after the Hartz IV reform of 2005. On average, the share of households that do not claim benefits despite being eligible, amounts to 55.7 percent of all eligible households in that period. The issue of non-take-up has further important implications for the determination of the standard benefit rate. Since the legally defined calculation procedure does not account for non-take-up households in the reference group, the approximated consumption expenditure that is considered as necessary for a dignified life is calculated too low. The results of this study suggest that the legally defined monthly adult lump sum amount in the year 2014 would have been twelve euros higher if the issue of non-take-up was accounted for in the methodology. Based on the findings, the paper aims to give policy recommendations.
Keywords: Welfare participation, take-up, social policy, microsimulation