Working Papers
Colleges, Business Dynamism, and Local Growth [September 2026]
Selected for the 2023 EALE job market Tour
Abstract: Can new colleges transform a local economy? This article studies the largest higher-education expansion in French history, a policy that opened hundreds of teaching-oriented colleges across 145 cities in the 1990s. Exploiting the staggered timing of openings, I find a persistent 17% increase in the share of college graduates. The policy induced creative destruction: firm creation rose by 16% while incumbent firms contracted. Entry was concentrated in skill-intensive and in tradable industries, led by experienced entrepreneurs. On net, value added grew by 5% and productivity by 6%. These gains were unevenly distributed: skilled employment rose among younger workers, while older, less-educated workers experienced lasting losses. Twenty-five years later, cohorts of enrollment age when the colleges opened have more education, higher employment rates, and more-skilled jobs. Overall, local effects were larger in initially less-developed areas, contributing to a narrowing of spatial disparities.
Location Effects or Sorting? Evidence from Firm Relocation [July 2026] R&R Econometrica
With Pauline Carry and Benny Kleinman
Abstract: What role do firms play in geographic wage disparities? This paper exploits establishment mobility as a novel source of identification to separate firm sorting from location treatment effects, i.e., the local wage premium for a given worker at a given firm (due to, e.g., geography, infrastructure, amenities). Using data from France and the U.S., we first document key facts about establishment relocation: 4% of establishments move each year, retaining their activity and structure while adjusting their workforce and wages. Combining establishment and worker mobility in France, we estimate a model that decomposes wage variation due to workers, firms, and locations. Spatial wage differences are primarily driven by sorting: worker composition accounts for 30%, firm composition for 17%, and their co-location for 34%. Residual location effects explain only 2–4%. Establishment sorting also accounts for most of the urban wage premium and the higher wage growth in larger cities.
The Effects of a Shorter Workweek on Employment, Output, and Productivity [October 2026]
With Pauline Carry, Claire Montialoux, Alexandra Roulet and Nina Roussille
Abstract: Do shorter workweeks generate work-sharing and productivity gains, as intended by policymakers? We study France's pay-preserving reduction in the workweek from 39 to 35 hours, bundled with employer payroll tax cuts. We first exploit the staggered implementation by firm size to identify the impact of the reform package. We then use differential exposure to the hours cut and tax cut to separate their effects. We combine matched employer--employee, balance-sheet, and survey data to study worker, firm, and local labor market effects. The reform package left employment unchanged. This zero masks large offsetting effects: absent the tax cuts, the shorter workweek would have reduced employment by 3.8%. The effects are skill-biased, raising employment among managers and professionals while reducing it among other employees. Finally, shorter hours raised total factor productivity by 2%, limiting the output decline to 1% despite a 6% fall in total hours. Surveys point to work intensification and organizational changes as potential drivers of productivity gains.
Publications
The Effects of Mandatory Profit-Sharing on Workers and Firms: Evidence from France
With David Sraer and David Thesmar.
Quarterly Journal of Economics (2026)
Charles River Associates Award for Best Paper on Corporate Finance (WFA 2024)
Abstract: Since 1967, all French firms with more than 100 employees have been required to share a fraction of their excess profits with their employees. Through this scheme, firms with excess profits distribute, on average, 10.5% of their pre-tax income to workers. In 1990, the eligibility threshold was reduced to 50 employees. We exploit this regulatory change to identify the effects of mandated profit-sharing on firms and their employees. The cost of mandated profit-sharing for firms is evident in the significant bunching at the 100-employee threshold observed prior to the reform, which completely disappears post-reform. Using a difference-in-difference strategy, we find that, at the firm level, mandated profit-sharing (a) increases the labor share by 1.8 percentage points, (b) reduces the profit share by 1.4 percentage points, and (c) has no significant effect on investment and productivity. At the employee level, mandated profit-sharing increases low-skill workers' total compensation and leaves high-skill workers' total compensation unchanged. Overall, mandated profit-sharing redistributes excess profits to lower-skill workers in the firm without generating significant distortions or productivity effects.
Policy brief (in French), Financial Times, Stanford Social Innovation Review, VoxTalks, Vox Column, NBER working paper
Inequality and Earnings Dynamics in France: National Policies and Local Consequences
With Francis Kramarz and Thomas Delemotte
Quantitative Economics (2022) -- Part of the Global Repository of Income Dynamics Project
Abstract: This paper provides new stylized facts about labor earnings inequality and dynamics in France for the period 1991-2016. Using Linked Employer-Employee Data, we show that (i) Labor inequality in France is low compared to other developed countries and has been decreasing until the financial crisis of 2009 and increasing since then. (ii) Women experienced high earnings growth, in particular at the bottom of the distribution, in contrast to the stability observed for men. Both result from a decrease in labor costs at the minimum wage and an increase in the hourly minimum in the aftermath of the 35h workweek policy. (iii) Top earnings (top 5 and 1%) grew moderately while very top earnings (top 0.1 and 0.01%) experienced a much higher growth. (iv) Inequality between and within cohorts follow the same U-shaped pattern as global inequality: it decreased before 2009 and then increased until 2016. (v) Individual earnings mobility is stable between 1991 and 2016, and very low at the top of the distribution. (vi) The distribution of earnings growth is negatively skewed, leptokurtic, and varies with age. Then, studying earnings dispersion both within and between territories, we document strong differences across cities as well as between urban and rural areas, even after controlling for observable characteristics. We also observe a continuous decrease in earnings inequality between territories. However, a larger inflation in rural territories mitigates this convergence. Finally, we document a strong reduction in inequality within rural and remote territories, again driven by changes at the bottom of the wage distribution.
Selected Work in Progress
Safety at Work: Does Taxing Accidents Reduce Injuries?