By: Anton Cheremukhin
The original Mortensen-Pissarides model possesses two elements that are absent from the commonly used simplified version: the job destruction margin and training costs. I find that these two elements enable a model driven by a single aggregate shock to simultaneously explain most movements involving unemployment, vacancies, job destruction, job creation, the job finding rate and wages. The job destruction margin’s role in propagating aggregate shocks is to create an additional pool of unemployed at the onset of a recession. The role of training costs is to explain the simultaneous decline in vacancies and slow response of job creation.
Applications of the Mortensen-Pissarides matching model do not use endogenous job destruction because there is little evidence it varies over the cycle and it would involve countercyclical vacancies. This paper shows that these disadvantages can be fixed once training and hiring costs are included, and they provide interesting dynamics.