Research
Research
Embedded Files
Discussion Papers
Discussion Papers
Returns to Higher Education and Dropout: A Double-Debiased Machine Learning Approach
Most recent version here.
Abstract: This paper estimates the short- and medium-term individual returns to Higher Education (HE) for a cohort of young people in the United Kingdom, with a particular focus on those entering HE but not completing a first degree. Using a double-machine learning multi-valued treatment approach to overcome endogeneity issues, I investigate heterogeneity in labor market outcomes for dropouts and graduates relative to those who have never attended HE, where outcomes are employment, wages, and occupational status. Results reveal marked socioeconomic and gender differences: in the early-career stage, female dropouts show no employment or wage differences relative to those with no HE experience, but attain higher occupational status, similar to that of graduates. Conversely, male dropouts experience an employment penalty and do not benefit from higher occupational status. In general, however, most of the negative effects accrue to those from less advantaged backgrounds. By the mid-career stage, wage and occupational status premiums exist independent of gender and socioeconomic background, though the magnitude varies, and dropouts are more likely to be employed. This suggests that any negative penalty associated with dropout attenuates over time. These findings challenge the view of dropout as a failed outcome, instead framing it as a distinct educational trajectory with heterogeneous labor market consequences.
With Guido Neidhöfer and Patrick Lehnert
Abstract: We construct a new longitudinal database of intergenerational mobility of education in European regions that enables previously impossible analyses of social and economic phenomena over time and space. Developing a panel of intergenerational mobility indices at a microregional level, we make two contributions. First, we extend existing knowledge on the development of intergenerational mobility in European regions over time. Second, we use the new database to provide novel evidence on the relationship between intergenerational mobility and innovation and find that higher mobility is associated with increased innovation.
Current R&R at Nature.
Sorting, Status, and Shadow Education: How Track Placement Shapes Parental Investment
With Thilo Klein
Abstract: Educational tracking - separating students into tracks or schools by ability - is commonplace, but access and preferences for top programs often depend on socioeconomic status (SES), reinforcing inequality. We study shadow education in the context of an early-tracking system, exploiting score cutoffs using a pseudo-regression discontinuity design to isolate the causal effect on parental investments. We find that assignment to the highest track disproportionately increases private tutoring among families in the lowest tercile of SES. This suggests tracking activates a behavioral response among disadvantaged households, which may amplify between-track achievement gaps.
Access, Achievements, and Aspirations: The Impacts of School Tracking on Student Outcomes
With Thilo Klein and Maximilian Bach
Abstract: Though the use of tracking policies to stratify students is commonplace, evidence concerning the effects of ability-based tracking on student performance is mixed. Using rich data from the Hungarian secondary school centralized assignment mechanism and a quasi-experimental framework, we find that attending the highest track noticeably improves standardized test scores and university aspirations two years post-match. Heterogeneity analyses find this effect is independent of socioeconomic status, prior achievement, and parents’ educational attainment, and we find only limited evidence of peer spillover effects in terms of academic ability. Given socioeconomic disparities in track placement, tracking may reinforce educational inequality.
Labor Market Integration of Asylum Seekers in Europe: Recent Trends and Barriers
With Martin Lange and Philipp Schmidt
Abstract: The labor market integration of asylum seekers remains a contested issue. Using the EU-Labor-Force-Survey, we characterize the state of asylum seekers’ labor market integration in Europe, and provide representative statistics on several dimensions of integration. We compare asylum seekers to natives and economic migrants and find that asylum seekers struggle to integrate across European states, exhibiting employment rates of 10 percentage points lower than that of natives, on average, as well as a notable gap in job quality. Analyzing self-reported barriers to employment, we document that asylum seekers’ lower employment rates and job quality are likely the result of institutional hurdles.
Work in Progress
Work in Progress
Bit by Bit: Evolving Wage Polarization, Employment Shifts, and Job Satisfaction in the Age of Automation
Abstract: This paper investigates the effects of exposure to digitalization and robotization in an inter-industry analysis of regional economies in Switzerland. Using individual panel data, I first demonstrate the evolving transition patterns of routine workers: while some remain in routine occupations, others downgrade into worse-paying jobs, and some exit into (early) retirement. Only a small fraction of the displaced routine workers exit into long-term non-employment. Second, I examine the distributional consequences for wages and job-satisfaction of exposure to technological change providing worker-level, demographic group-level, and local labor market-level evidence. Worker-level evidence is based on individual records in the Swiss Household Panel (SHP) and takes into account the panel structure of the data. A cross-sectional analysis uses the adjusted penetration of robots (APR) to examine regional differences in wage distributions over time, wherein the unit of interest is the local labor market. Finally, I investigate heterogeneities where the unit of analysis is demographic groups, controlling for relative task displacement as a weighted function of exposure to APR by group.
Digitalization, Automation, and ?: The Intergenerational Origins of Regional Resilience
Abstract: When individual opportunities are not constrained by an individual’s childhood circumstances, it permits a better match between talents and opportunities, suggesting a greater degree of individual productivity. Conversely, when talent is misallocated, individual productivity is limited to a level below an individual’s productivity potential. Existing research shows that greater mobility is associated with higher levels of regional innovation. In aggregate, therefore, an endogenous growth model suggests a more economically efficient allocation of labour to jobs, driven by regional mobility differences and the presence of knowledge spillovers, may, on the one hand, improve the resilience of a region, and thus its capacity to absorb an exogenous shock, and, on the other, may drive a shorter recovery period following exogenous economic disruptions. We compute weighted effective mobility indices and human capital stocks at the micro-regional level for the last three decades for Germany, and control for technical capital, and growth in technical capital, using a combination of ICT investments (EU-KLEMS) and the localised adjusted penetration of robots (APR) (using a shift share approach to account for regional differences in sectoral structure and size). We then use a triple differences framework to examine the response of local labour market areas to exogenous shocks, and the effects on output, employment, and innovation.
Geographies of Discontent: The Local Opportunity Structure, Populism and Democratic Resilience
Abstract: The rise of populist movements across European democracies has prompted extensive inquiry into the structural determinants of political discontent. While existing research has identified broad socioeconomic factors associated with populist voting, less attention has been paid to the role of local opportunity structures in shaping political attitudes and democratic engagement. The concept of opportunity structure, rooted in sociological theories of social mobility, encompasses the institutional and economic arrangements that determine individuals' life chances within specific geographic contexts. Nevertheless, areas characterised by limited intergenerational mobility may foster environments conducive to anti-establishment sentiment, resentment, and reduced confidence in democratic institutions. The goal of this project is to examine how exposure to local opportunity structures, operationalised through effective intergenerational educational mobility indices, influences individual political attitudes and voting behaviour. I employ multilevel modelling techniques to account for the nested structure of individuals within regions, allowing for the simultaneous examination of individual and contextual effects. This enables the identification of cross-level interactions between personal characteristics and regional opportunity structures, providing insights into how local contexts moderate the relationship between individual circumstances and political outcomes.
Page updated
Google Sites
Report abuse