Working Papers:

This paper presents a new efficiency argument for an accommodating taxation policy on high incomes. Job seekers, applying to different segments of a frictional labor market, do not internalize the consequences of mismatch on the entry decision of firms. Workers are not selective enough, resulting in a lower average job productivity and suboptimal job creation. The output-maximizing policy is anti-redistributive to improve the quality of the jobs prospected. As an income tax affects the sharing of the match surplus, a tax on production (or profits) is required to redress the slope of the wage curve. Neither a minimum wage nor unemployment benefits can fully decentralize optimal search behaviors.


  • Regional Equilibrium Unemployment with Agglomeration Effects, with Pierre Deschamps (Sciences Po Paris), draft upon request

Patterns of unemployment vary considerably across regions. Using an original model of the regional labor market with search and mobility frictions, we study the role of agglomeration effects on the dynamics of local unemployment. The impact of place-based subsidies and unemployment benefits crucially depends on the sign and strength of the agglomeration forces. With agglomeration productivity gains, negative regional employment shocks are amplified because profit opportunities deteriorate, inducing higher mobility out of the region. The model is able to reproduce the strong persistence of the shock on the unemployment rate and the region’s size.


  • Estimating On-the-job Search as an Intensive Margin, draft upon request

Job-to-job transitions are a major source of wage growth over a lifetime. This paper empirically assesses how job mobility responds to individual search efforts through a structural approach. I build a flexible model of job transitions in which forward-looking workers search optimally depending on the expected returns and household characteristics. The model provides a unique decomposition of job mobility disentangling job opportunities and private gains from searching. It relaxes the classical stationary-environment assumption in the literature to allow for a rich accounting of observed heterogeneity. The optimal search effort function is jointly estimated with the distribution of job offers from panel data. I find that job transitions are more frequent at the top of the job ladder, running up against the theoretical insights of the random search framework. 


On-going Projects:

  • Geographic and Labor mobility over the Life-cycle
  • Social Welfare for Independent Workers