• Graduate program
  • Research
  • Summer School
  • Events
    • Summer School
      • Applied Public Policy Evaluation
      • Deep Learning
      • Economics of Blockchain and Digital Currencies
      • Economics of Climate Change
      • Foundations of Machine Learning with Applications in Python
      • From Preference to Choice: The Economic Theory of Decision-Making
      • Gender in Society
      • Machine Learning for Business
      • Sustainable Finance
      • Tuition Fees and Payment
      • Business Data Science Summer School Program
    • Events Calendar
    • Events Archive
    • Tinbergen Institute Lectures
    • 16th Tinbergen Institute Annual Conference
    • Annual Tinbergen Institute Conference
  • News
  • Alumni
  • Magazine
Home | Events Archive | Revisiting the Determinants of Unemployment Duration: Variance Decomposition a la ABS in Spain
Seminar

Revisiting the Determinants of Unemployment Duration: Variance Decomposition a la ABS in Spain


  • Series
  • Speaker(s)
    Maia Güell (The University of Edinburgh, United Kingdom)
  • Field
    Empirical Microeconomics
  • Location
    Tinbergen Institute Amsterdam, Sydney room
    Amsterdam
  • Date and time

    May 10, 2022
    15:30 - 16:30

Abstract
Using the methodology developed in Alvarez, Borovickova, and Shimer (2014), wedecompose the variance of unemployment duration in Spain into three main compo-nents: labour market frictions, duration dependence and heterogeneity. Crucial to ouranalysis are the comparisons by different demographic groups over the business cycleto shed light in the mechanisms behind duration dependence. We offer a general ap-proach for interpreting administrative data that is closer to unemployment than non-employment and includes different types of unemployment spells, particularly shortspells without the right to claim benefits. Labour market frictions account for thebulk of the variance of unemployment duration, and become more prominent duringrecessions. Duration dependence constitutes 25% of the total variance, followed byheterogeneity, which represents 18%. Women and college graduates have larger sharesof duration dependence. Finally, the share of duration dependence does not vary overthe business cycle for non-college workers. But it is higher during expansions for col-lege workers, which is consistent with a statistical discrimination mechanism due to dynamic sample selection. (joint with Cristina Lafuente).

Link to paper.