The Network Origin of Slow Labor Reallocation

Leonard Bocquet (1, 2)

1 PSE - Paris School of Economics
2 PJSE - Paris Jourdan Sciences Economiques

What explains slow worker reallocation after trade or technological shocks? In this paper, I explore the idea that imperfect skill transferability constrains the reallocation of workers between occupations. I model these frictions as forming a network (the “occupational network”), whose nodes are occupations and whose edges represent the possible occupational transitions. I show that the structure of the occupational network matters for the speed of worker reallocation and I make two contributions. First, I find that this network is very sparse, hinting at large frictions to occupational mobility, and that there exists central bottleneck occupations which can block the reallocation process. Second, I extend the search and matching model of the labor market with an occupational network. I find that asymmetric shocks can generate transition dynamics which are two orders of magnitude slower than in the standard model. Moreover, I show that the intensity of network bottlenecks is a key determinant of slow worker reallocation dynamics after asymmetric shocks. In other words, central occupations have a granular effect on worker reallocation speed.

Domaines: Economies et finances

Leonard Bocquet. The Network Origin of Slow Labor Reallocation. 2022. ⟨halshs-03703862⟩