Publications
Earnings Expectations of "First-in-Family" University Students and Their Role for Major Choice
(with Fabian Kosse, Markus Nagler, and Johannes Rincke)
Abstract: How do students’ earnings expectations differ by being the first in their family to attend university (FiF) and how do they affect field of study choice? We leverage unique survey and administrative data from a German university to document sizable gaps in expected earnings between FiF and non-FiF students. Our data can explain two-thirds of this gap, with the largest share attributable to field of study choice. We show that FiF students sort less into study fields based on their earnings expectations. Investigating potential explanations, we find that they expect lower own ability and worse non-wage amenities in high-earning fields.
Labour Economics 95: 102743, 2025
Published Version | IZA Discussion Paper No. 17720
Current Papers
Born for this? Sorting into Career Paths on Personality and Preferences
(with Fabian Kosse, Markus Nagler, and Johannes Rincke)
Abstract: We study how personality and preferences shape sorting into careers. Combining rich survey data with administrative records on university entrants in Germany, we document three key findings. First, students strongly sort into study fields along a broad set of personality and preference measures, even after conditioning on expected earnings, amenities, and ability. Second, sorting is productive: misalignment between individual and field profiles predicts dropout and weaker academic progress. Third, patterns are strikingly persistent: Around 70% of endline differences reflect initial selection, while dropout, non-response, and within-individual changes during university play minor roles. Endline differences closely match patterns across corresponding occupations.
Draft coming soon!
Gender Discrimination in Online Negotiations - Evidence from a Field Experiment
Abstract: This paper studies gender discrimination in negotiations using a field experiment. To do so, I post job advertisements regarding a home improvement service on an online platform for classified advertisements in Germany. I negotiate offers with all potential sellers of the service who react to the advertisement and randomly vary buyer gender. In the negotiations, two different treatment messages aim at examining statistical discrimination, one with and one without a justification of the negotiation. I find evidence for sellers discriminating against female buyers of the service. Female buyers are less likely than male buyers to receive a counteroffer unless they justify why they negotiate. There are no differences by buyer gender in final offers and price discounts offered by sellers. Analyses of chat data from the platform using large language model (LLM) classifications suggest that sellers engage in statistical discrimination.
Draft coming soon!
Work in Progress
Sorting into Collaboration - Individual Traits vs. Gender
(with Daniel Banko-Ferran and Pauline Madiès)