Hurting the needy? Regional inequities in sanction decisions against German welfare recipients

About this Session


Fri. 12.04. 10:20



Authors – Miriam Leue, Stefanie Rueß, Gerald Schneider.

Abstract :

The question “Who should get what, and why?” by van Oorschot (2000) is frequently discussed in the welfare state and public administration literature. However, the question of “where” remains largely unaddressed, even though several studies have shown how regional contextual factors such as the political ideology or the public mood within a municipality or province affect the treatment of marginalized groups by street-level bureaucrats. Well-studied cases of spatial political inequality are the widely different protection rates for asylum seekers within countries such as Germany, Switzerland, and the United States. In this article, we examine whether these spatial inequities also manifest in sanctions decisions for unemployment benefit recipients, thereby addressing the key question of “where”. Sanctions are an instrument that can be used to partially withdraw payment of unemployment benefits due to alleged violations of obligations.

We show with monthly data for the period from 2007 to 2022 that, despite a uniform legal framework, sanction rates against different groups of unemployment recipients have varied considerably across German job centers and over time. To understand this form of administrative inequality, we conceive of the responsible case workers as agents who consider regional political preferences and regional economic opportunities when deciding on sanctions against welfare recipients. This can result in similar groups of people being treated differently in various regions. Thus, we argue that case managers are more likely to impose sanctions when the political mood in the county is supportive of sanctions and when case managers attribute the causes of failure to circumstances within the beneficiary’s control such as good regional labor market conditions.

We examine our principal-agent hypotheses with a novel county-level dataset including administrative decisions and structural data, and employ a longitudinal model, using the prevalence of regional dialects as an instrument for regional sentiment. The preliminary analysis lends considerable support to our conjectures: (1) Beneficiaries who have violated their obligations and live in regions with good employment opportunities are more likely to be sanctioned. (2) Living in a region with high approval for parties supporting sanctions increases the risk of benefit cuts. Hence, this study reveals that regional patterns of administrative inequality emerge despite a uniform legal basis. This implies that the promise of a fair and equitable treatment of all recipients by street-level bureaucrats irrespective of external conditions cannot be realized.