Along the business cycle, labor markets face a large amount of reallocation: firms create and destroy vacancies, work relationships are formed and resolved, and workers change jobs and careers. While some argue that recessions generate a cleansing effect—the most unproductive jobs are destroyed and reallocation speeds up, counteracting the effect of the downturn, others argue that recessions generate a sullying effect—job search becomes more disorganized and movements along the job ladder slow down, deepening the recession.
Understanding the frictions underlying labor market reallocation and their impact on the cyclicality of productivity and wages is key for informing the design of policies aimed at stabilizing economic fluctuations (e.g., monetary policy, unemployment protection) as well as promoting long-run growth (e.g., retraining programs, structural labor market reforms).
To study how business cycles affect the allocation of workers to jobs, we first develop a novel theoretical framework. The model incorporates two key features. First, workers differ along multiple skill dimensions (e.g., math, verbal, social) and sort into careers with heterogeneous requirements along those dimensions. Second, workers and firms have incomplete information and only learn about the workers’ skills currently used in production; thus, switching careers entails learning about untried skills. In equilibrium, workers reallocate up and down the job ladder of a given career path as well as across different career paths, based on beliefs. Using our theoretical framework together with worker-level data from the US, we find that, in recessions, underqualified workers are fired, specifically those at the bottom rungs of the job ladder, which reduces mismatch among ongoing work-relations (cleansing).
Among these workers, there is an endogenous increase in career switching, which leads to higher mismatch when rehired at the bottom of a new job ladder (sullying). Therefore, job transitions in and out of bottom job rungs, combined with career mobility of workers, are important to account for the empirical behavior of mismatch. While both the cleansing and sullying effects are present, overall the cleansing effect dominates, and mismatch is procyclical.
Our theoretical framework also provides a novel narrative narrative for the “scarring effect of unemployment”. In line with empirical evidence showing that earnings losses after job displacement are large and persistent, our model predicts that workers displaced from their careers suffer large and persistent earnings losses, even after they have been re-employed. This is because rebuilding a career in a new sector implies high levels of uncertainty, which slows down a worker’s climb through the job ladder.
- Professor
- Professor
- More information