Cutting benefits may boost employment but can reduce job quality, increase inequality, and shift costs and risks to workers
Germany’s unemployment insurance system has long been at the centre of political debate. The so-called “Hartz reforms” of the early 2000s tightened eligibility, increased job-search obligations, and reduced benefits. Supporters argued that the reforms helped bring more people into work. Critics stressed the risks for vulnerable households and the rise in inequality.
This debate has returned with the German government’s decision to replace the “Bürgergeld” (citizens’ benefits) with a stricter system of basic income support. Once again, the discussion focuses mainly on employment and public spending. Will tougher rules push more people into work? Will they reduce welfare costs? These are important questions. But they are not the whole story.
In a forthcoming paper in Econometrica, we study the broader economic effects of unemployment benefit reforms in Germany. Our analysis uses a model in which workers search for jobs, move between jobs, bargain over wages, and save to protect themselves against unemployment risk. This framework makes it possible to look beyond immediate employment effects and ask a wider question: do benefit cuts make society better off?
The answer is nuanced. A 10% reduction in unemployment benefits does increase employment. Employment rises by 2.7% and output by 2.9%. Lower benefits make unemployed workers search harder and accept jobs more quickly. Firms respond by creating more jobs. Households also save more because public insurance has become weaker, which raises the capital stock.
These results explain why policymakers often find benefit cuts attractive. They can raise employment and output while reducing dependence on public transfers. But the same reform also has less visible costs.
One important cost is that workers become less selective. Some accept jobs that are not very productive or not well matched to their skills. As a result, average labour efficiency falls. Average wages also decline, and earnings inequality rises. More people work, but not necessarily in better jobs.
A second cost is the additional effort required from workers. When public insurance is weaker, people must search more intensively and bear more income risk. This effort is not free. It uses time, energy, and attention, and it can be especially burdensome for workers facing unstable employment prospects. Once these costs are included, the reform reduces welfare. In the model, a newborn worker would lose the equivalent of 1.32% of lifetime consumption.
The effects also differ across workers. Younger and poorer workers increase their job-search effort the most and experience larger improvements in job-finding rates. But older and wealthier households can suffer large welfare losses because they are exposed to falling wages, lower interest rates, and greater labour income risk. Aggregate employment numbers therefore hide important distributional effects.
The main lesson is not that unemployment benefits should never be reformed. Rather, policymakers should be clear about the trade-off. Cutting benefits can increase employment, but it can also reduce job quality, raise inequality, and make workers bear more risk and search costs.
A labour market reform should therefore not be judged only by how many more people enter work or how much the government saves. It should also be judged by the quality of the jobs created, the risks shifted onto households, and the welfare consequences for different groups of workers.
© Leo Kaas, Etienne Lalé, and Nawid Siassi
Leo Kaas is Professor at the European University Institute, Italy, and IZA@LISER Research Fellow
Etienne Lalé is Associate Professor at York University in Toronto, Canada, and IZA@LISER Research Fellow
Nawid Siassi is Associate Professor at Technical University Vienna, Austria
Please note:
We recognize that World of Labour articles may prompt discussion and possibly controversy. Opinion pieces, such as the one above, capture ideas and debates concisely, and anchor them with real-world examples. Opinions stated here do not necessarily reflect those of the LISER. The underlying paper is forthcoming as: Kaas, Leo, Etienne Lalé, and Nawid Siassi. 2026. "Job Ladder and Wealth Dynamics in General Equilibrium." Econometrica.
Related World of Labour content:
https://wol.iza.org/articles/unemployment-benefits-and-job-match-quality by Konstantinos Tatsiramos
https://wol.iza.org/articles/unemployment-benefits-and-unemployment by Robert Moffitt and Wonsik Ko
https://wol.iza.org/articles/impact-of-monitoring-and-sanctioning-on-unemployment-exit-and-job-finding-rates by Duncan McVicar
https://wol.iza.org/articles/labor-market-policies-unemployment-and-identity by Ronnie Schöb
Photo by courtneyk on iStock