Research

Abstract: In this paper, we analyze the effects of a German job creation scheme (JCS) on the social integration and well-being of long-term unemployed individuals. Using linked survey and administrative data for participants and a group of matched non-participants, we find significant positive effects of being employed within this program. Participants with health impairments or an above-average duration of welfare dependence benefit more. The effects decline over the course of the program, which is explained by an increase in both the share of participants who leave the program, and of control individuals who find a job. The results suggest that JCSs should target those with the lowest employment prospects.

Abstract: This paper estimates how exposure to technological change affects the individual costs of job displacement. Using changes in the task structure of occupational labor markets to proxy for the gradual diffusion of new production technologies, I document substantial heterogeneity in post-displacement outcomes: Workers displaced from strongly changing occupations experience earnings losses roughly twice as large as those from stable ones. Larger losses stem mainly from longer unemployment and higher occupational mobility and cannot be explained by occupation tenure, age, or selection and sorting across firms. These findings are consistent with vintage human capital theory and technology-induced skills mismatch. They suggest that incumbent workers do not sufficiently invest in new skills to insure themselves against involuntary job loss even when technological change is gradual.

Abstract: The 2004 enlargement of the European Union created one of the largest institutional shifts in migration opportunities in recent years. While the mobility of workers from new member states has been widely studied, much less is known about how such reforms have affected the migration of highly skilled subpopulations. This paper examines how the staggered removal of transitional restrictions on labor market access in the EU shaped the international mobility of researchers from Central and Eastern Europe. Using newly compiled bibliometric data that tracks researcher affiliations across countries, we show that liberalization more than doubled the rate at which researchers moved from new to old member states. At the same time, the number of researchers moving in the opposite direction also increased, revealing a pattern of reciprocal “brain circulation” rather than a one-sided drain. Supplementary analysis using international patent data shows that this increase in circulation coincided with a marked rise in cross-border co-inventions, suggesting that enhanced mobility also fostered stronger transnational research collaboration. By exploiting the staggered timing of policy implementation, our study provides rare causal evidence of how institutional reforms redistribute highly skilled populations and strengthen international knowledge co-production and innovation.

Abstract: This paper examines how local labor market conditions shape the costs of job displacement. Using German administrative data and a matched difference-in-differences design, we measure the role of regional structural unemployment (SUE) at displacement. Displacement is substantially more costly in high-SUE regions, where re-employment and earnings remain persistently lower. Wage losses are up to 60 percent greater than in low-SUE regions, largely due to transitions into lower-paying firms and increased occupational and sectoral mobility. This reflects limited regional mobility and constrained job options, underscoring the importance of people- and place-based policies to mitigate scarring in structurally weak regions.

Abstract: The German Qualifications and Career Surveys (GQCS) have been widely used to study job tasks and to construct measures of occupational routine intensity. I argue, however, that this approach has important limitations: (i) task items in the GQCS are measured inconsistently across waves, making it difficult to track changes over time, and (ii) ad-hoc classifications of broadly defined tasks as 'routine' or 'non-routine' may be misleading, as they do not reliably reflect the actual routine intensity of jobs, which may vary within and between occupations and over time. As an alternative, I demonstrate that the GQCS contain a rich set of variables on job requirements and working conditions, including explicit measures of repetitiveness and programmability, which align more closely with the concept of routine intensity than ad-hoc task classifications. I provide code for data preparation and harmonization and illustrate that these variables predict occupational employment and wage growth in German social security data. 

Abstract: A large literature studies the impact of minimum wages on employment focusing on potential displacement and reallocation effects for workers. In this paper, we ask how firms respond to the adoption of a minimum wage, in particular whether it encourages firms to increase their capital intensity or to outsource some of their production steps. We study this question in the context of Germany, which adopted industry-level minimum wages between 1997 and 2014. Our analysis is based on rich balance sheet data on firms matched to administrative records for all employees.

Abstract: Routine-intensive occupations have been declining in many countries, but how does this affect individual workers’ careers if this decline is particularly severe in their local labor market? This paper uses administrative data from Germany and a matched difference-in-differences approach to show that the individual costs of job loss strongly depend on the task-bias of regional structural change. Workers displaced from routine manual occupations have substantially higher and more persistent employment and wage losses in regions where such occupations decline the most. Regional and occupational mobility partly serve as an adjustment mechanism, but come at high cost as these switches also involve losses in firm wage premia. Non-displaced workers, by contrast, remain largely unaffected by structural change.