Research
Research
with Alena Bičáková, Kamila Cygan-Rehm, and Klára Kalíšková,(supported by the DFG grant no. 531021587)
The Covid-19 pandemic was a severe shock to employers and employees in the form of substantial restrictions on face-to-face meetings, leading to an unexpected increase in work-from-home (WFH) arrangements. This may have affected the way individuals work and the skills they need to successfully adapt to the new working conditions. In this paper, we use the sharp increase in WFH in Germany in March 2020 as a natural experiment to examine its long-term effects on the labor market trajectories and human capital of affected workers. To this end, we combine survey data on the occupational WFH potential of a given job with individual data on labor market biographies from German social security records. Using event studies and a difference-in-difference design, we compare the career trajectories of individuals who held jobs with high and low WFH potential immediately before the pandemic. We pay particular attention to potential gender differences, as the pandemic disproportionately affected parents, especially women.
with Ewa Gałecka-Burdziak, Kamila Cygan-Rehm, Klára Kalíšková, Tomáš Lichard, and Robert Pater (supported by the DFG grant no. 531021587)
The Russian invasion of Ukraine resulted in an unprecedented influx of Ukrainian refugees into several Central and Eastern European countries. This created the need to absorb and integrate a substantial number of potential workers into local labor markets. In this paper, we study the impact of Ukrainian refugees on incumbent workers and firms in several heavily affected countries, such as the Czech Republic, Germany, Poland, and Slovakia. Specifically, we apply a shift-share instrumental variables design to country-specific datasets that combine administrative data on migrant stocks and flows with registry records on labor market outcomes of incumbent workers and performance measures of their firms. Cross-country comparisons allow us to test the role of specific institutions in shaping responses to the shock.
with Jörg Drechsler and Milan Quentel
In response to growing demand for detailed geographic data, institutions are increasingly geocoding microdata. However, access is limited due to confidentiality concerns. While the literature proposes a range of confidentiality-protection strategies, clear guidance on the selection and evaluation of these strategies is currently still lacking. This study aims to address this gap by conducting an empirical evaluation of confidentiality-protection strategies and assessing their efficacy in mitigating risk and preserving data utility. We apply spatial aggregation, geographic masking including the introduction of random noise and location swapping, and synthesis of the geographic information to a subset of geocoded German social security data. We evaluate the utility considering spatial pattern preservation and the geographic distribution of non-geographic information such as regional wages. The risk of disclosure is quantified by simulating a record linkage attack.
Maike Steffen, Konstantin Körner, Jörg Drechsler. An overview of data protection strategies for individual-level geocoded data. Statistics Surveys 19 (2025). DOI: 10.1214/25-SS151