Publication

Male and female selection effects on gender wage gaps in three countries

Published at Labour Economics - https://doi.org/10.1016/j.labeco.2024.102506
Abstract

A vast literature on gender wage gaps has examined the importance of selection into employment. However, most analyses have focused only on female labour force participation and gaps at the median. The Great Recession questions this approach because of the major shift in male employment that it implied. This paper uses the methodology proposed by Arellano and Bonhomme (2017) to estimate a quantile selection model over the period 2007–2018. Using a tax and benefit microsimulation model, I compute an instrument capturing both male and female decisions to participate in the labour market: the potential out-of-work income. Since my instrument is crucially determined by the welfare state, I consider three countries with notably different benefit systems – the UK, France, and Finland. My results imply different selection patterns across countries and a sizeable male selection in France and the UK. Correction for selection bias lowers the gender wage gap and reveals a substantial glass ceiling with different magnitudes. Findings suggest that disparities between these countries are driven by occupational segregation and public spending on families.

Presentations Presentations: EALE 2023, LAGV 2023, Ninth ECINEQ Meeting of The Society for the Study of Economic Inequality, ECINEQ PhD Workshop participants at the London School of Economics, 14th Workshop on Labour Economics (IAAEU), the 4th QMUL Economics Workshop for PhD and Post-doctoral Students, the 2022 French Stata conference, PhD seminar at the Aix-Marseille School of Economics, Labour Chair seminar at the Paris School of Economics
Slide sel1 Slide sel2 Slide sel3 Slide sel4

Working Papers

Automating Inequality: Gender Bias in AI-mediated Labor Markets

With Germain Gauthier (Bocconi University), Debora Nozza (Bocconi University) & Paola Profeta (Bocconi University) - Link to the AEA Pre-analysis Plan, Previously "Generative AI & Labor Market Discrimination"

Accepted at PNAS (Proceedings of the National Academy of Sciences)

Working paper available upon request

Abstract

Generative AI is rapidly entering labor markets, with large language models (LLMs) now used to help write CVs and evaluate applicants. We study the risks this creates in a series of audit experiments in which applicant gender is randomly assigned while all other information is held constant. We find that LLMs generate gender-stereotypical CVs and exhibit gender bias in screening scores and wage recommendations. Screening scores track occupational gender segregation, penalizing women in male-dominated occupations and favoring them in female-dominated ones. Wage recommendations track occupational gender wage gaps when gender is explicit in the prompt, though this gradient weakens under name-based gender signaling and is sensitive to the LLMs being used. These disparities arise through two channels: evaluation bias, when identical CV content is evaluated differently by gender, and generation bias, when gendered content produced by the model affects later evaluation. Because the same systems shape both CV content and its subsequent assessment, these channels can reinforce one another. Taken together, our findings identify measurable risks of algorithmic amplification in AI-mediated job search.

Presentations TASKS VII Conference: The Economic Impacts of AI on Work and Labor Markets, 4th Workshop on Field Experiments in Economics and Business, 10th CEPR Text-As-Data Workshop, 2025 Workshop in Gender Economics at ENS de Lyon, 5th Discrimination and Diversity Workshop, LAGV 2025, Dondena AI and Society Initiative Seminar at Bocconi University, Advanced AI Methods Workshop, CREM Seminar at Rennes University, Applied Economics Lunch Seminar of the Paris School of Economics, Gender Inequality Workshop at the London School of Economics, Discrimination and the Inequalities in the Age of AI Workshop at Sciences Po

Supported by the Women and Science Chair Grant

What do women want in a job? Household constraints, gender-biased decisions and the reservation wage

This paper was awarded the Etta Chiuri Prize 2025 by the Italian Society of Public Economics, the Young Researcher Prize 2026 at the 5th Workshop in Gender and Economics (University of Luxembourg), and was runner-up for the Distinguished CESifo Affiliate Award in Labor Economics .

Draft available here: Dropbox link
Abstract

Recent explanations of the gender wage gap emphasize the role of gender differences in job search, yet the role of childcare constraints remains underexplored. This paper uses French administrative data to investigate how childcare constraints shape women’s reservation wage and job search strategies. First, I assess the types of occupations that men and women apply for and the implications for the reservation wage gap. Using text analysis, I create a novel dataset classifying occupations based on amenities and job content. Quantile decomposition methods allow me to document an unequal gap in reservation wage, intensifying along the distribution, partially explained by gender biased choices in the temporal flexibility associated with the desired job. Given that gender differences in targeted amenities may be shaped by childcare constraints, could a reduction in childcare costs change women's job search strategies? To address this question, I assess to which extent a 2018 reform, which increased childcare benefits for single-parent households by 30%, influenced the reservation wage and job-search strategies of women. Using a difference-in-differences strategy and spatial variation in childcare service availability, results indicate that the reduction in childcare costs led women to lower their reservation wages. However, I also find that more affordable access to flexible childcare increases the likelihood of targeting occupations requiring greater temporal flexibility and the desired maximum commute, thereby enabling them to secure more stable jobs and improve their reemployment outcomes. Lastly, I show that policies of childcare cost reduction are only truly effective when combined with childcare services availability.

Presentations
  • Bank of Italy Gender Economics Workshop, Workshop on Gender Gaps and Social Norms at Pompeu Fabra University, SIEP 2025 Conference, ECONtribute and C-SEB Design & Behavior Seminar at Cologne University, Areena JMC Symposium, EEA-ESEM 2024, EALE 2024, Junior Economist Meeting 2024, 2024 Junior Research Day at College de France, Afépop 2024, ADRES 2023, AFSE 2023, European Association of Labour Economists (EALE) Conference 2022, International Association for Applied Econometrics (IAAE) Conference 2022, LAGV 2022, JMA 2022, Food for Thought seminar at Bocconi University, Labour Chair Seminar at the Paris School of Economics, Firms and market seminar at CREST, Core Brown Bag Seminar at Louvain University, ADRES 2023 and PhD seminar at the Aix Marseille School of Economics
Paper
Slide 1 Slide 2 Slide 3 Slide 4 Slide 5 Slide 6 Slide 7 Slide 8

Gender gaps in the urban wage premium

With Cecilia García-Peñalosa (AMSE, CNRS, EHESS) & Christian Schluter (AMSE) - Link to the CEPR Working Paper - Link to the CESifo Working Paper - Appendix

Under Review

Abstract

We examine the economic geography of gender wage gaps to understand the role that location plays in gender earning differences. Using panelised administrative data for the universe of French workers, our findings indicate that women benefit relatively more from density than men, with an urban wage premium (return to urban density) 48% higher than for men. We identify a number of factors that explain this gap, with a large share being explained by the structure of the local labour market, notably, the extent of occupational segregation. Another important factor is commuting patterns, while childcare availability plays only a moderate role.

Presentations King’s Junior Research Day 2023, ADRES 2023, GRAPE 2023 Gender Gaps Conference, Bocconi Food for Thought seminar, PhD seminar at the Aix Marseille School of Economics, Paris School of Economics Labour Chair seminar; COSME workshop, 3rd Workshop on Public Policies, Urban Economics Association Conference 2024, IAAE Conference 2024, CEPR MESIE Conference
Communication:

Nada es Gratis

Slide 1 Slide 2 Slide 3



Other Writing - General Audience

Threat or opportunity? The impact of AI on women

With Paola Profeta (Bocconi University)

Abstract

The adoption of AI in various sectors has led to changes that present both opportunities and challenges for gender equality. Although AI appears to be less biased than human decision-makers, the literature also suggests it perpetuates stereotypes and inequalities between men and women. The unequal use of AI tools, combined with existing disparities in education and employment, may further disadvantage women in the labour market. In addition, AI's influence on employment, wage inequality and gender bias in healthcare has not been sufficiently studied, raising ethical questions about the fairness and transparency of these technologies. Addressing these challenges involves an increase in diversity in data-science teams, diverse and representative datasets, and promoting gender-inclusive training programs. Ultimately, the impact of AI on gender equality will depend on today's initiatives to neutralize the effects of AI.

Paper

Communication

La Repubblica

Selected Work in Progress

  • Generative AI's algorithmic bias on the labour market, with Max Thon (University of Zurich) & Paola Profeta (Bocconi University)
  • The Robot as a Referee: Gender Bias in AI-Assisted Peer Review, with Siri Isaksson (Burgundy School of Business)
  • Local Repercussions of AI Data Centres, with Mathieu Couttenier (ENS de Lyon)
  • Artificial Information: detecting and measuring news sites generated by LLMs during the 2027 presidential campaign, with Elisa Mougin (ENS de Lyon)