Work in progress

  • Convergence to collusion in algorithmic pricing

Artificial intelligence algorithms are increasingly used by firms to set prices. Previous research shows that they can exhibit collusive behavior, but how quickly they can do so has so far remained an open question. I show that a modern deep reinforcement learning model deployed to price goods in a repeated oligopolistic competition game with continuous prices converges to a collusive outcome in an amount of time that matches empirical observations, under reasonable assumptions on the length of a time step. This model reliably shows cooperative behavior supported by reward-punishment schemes that discourage deviations.

How related are different jobs in terms of skills? To what extent training programs allow to move across jobs that differ in skills, and to what extent can this reduce the mismatch unemployment - that is, the unemployment due to unbalances in labor demand vs. supply across occupations? The existing literature often answered the first question based on expert knowledge and existing job classification systems (O-NET, ROME classification in France etc.). Instead, we propose to build new measures of skill proximity across jobs based on job descriptions from vacancy data using state-of-the-art Natural Language Processing techniques. Making use of the skill distance measure produced and of comprehensive administrative data on unemployment spells, training use and employer-employee data, we describe the labor supply reallocations associated with the use of training programs by French job seekers. Comparing such occupational transitions in relationship with labor market tightness measures, we aim to assess the extent to which public funded training programs contribute to the reduction of mismatch unemployment.

  • Matching outcomes in the Italian job market for teachers

Italian teachers are centrally allocated to schools using the deferred acceptance algorithm, which makes it possible to formulate the matching between teachers and schools as the outcome of a discrete choice model with personalized choice sets. I assemble a novel dataset containing rankings and hiring decisions by combining data from multiple sources, and use it to estimate a structural model of heterogeneous teacher preferences for school attributes. My estimates show that the determinants of teacher assignment are primarily geographical, with a strong distaste for distance from the province of birth, as well as a general dislike for more rural and less populated localities. Descriptive evidence points to a shortage of competent teachers especially for technical fields, and estimates confirm that competent teachers are drawn outside the profession but show that wages and employment rates in non-teaching jobs are not relevant in making individuals choose to exit the teaching labor market.

R packages

  • MST-based k-means clustering (code, CRAN)
  • Principal curves of oriented points, as introduced in Delicado and Huerta (2003) (code, CRAN)