Academic journal article Economic Perspectives

A Simple Model of Gross Worker Flows across Labor Market States

Academic journal article Economic Perspectives

A Simple Model of Gross Worker Flows across Labor Market States

Article excerpt

Introduction and summary

While standard macroeconomic models of labor markets typically assume two labor market states (employment/nonemployment or employment/unemployment), a recent literature has extended those models to incorporate three labor market states: employment, unemployment, and nonparticipation (or out of the labor force). See, for example, Tripier (2004), Veracierto (2008), Christiano, Trabandt, and Walentin (2010), Haefke and Reiter (2011), and Shimer (2013). Most of these papers have focused on modeling the number of workers in each of these labor market states, but not the gross flows of workers across them. A notable exception is Krusell et al. (2012), which introduces search frictions (for the process through which workers meet job opportunities) into a real business cycle model with borrowing constraints in the household sector. Their model is rich enough to explicitly determine the gross flows of workers across the three labor market states, potentially providing a deeper understanding of what drives unemployment and other labor market shifts.

Interestingly, Krusell et al. (2012) find that under certain specifications of aggregate shocks, their model is able to broadly reproduce the cyclical behavior of gross worker flows across labor market states in the U.S. economy. While this is an important result, the economic mechanisms behind it are somewhat obscured by the real business cycle structure and the borrowing constraints. The purpose of this article is to strip the model in Krusell et al. (2012) down to its bare bones--hat is, to develop (and analyze) a very simple version of it. The key difference between my model and theirs is that instead of embodying the search frictions in a real business cycle with borrowing constraints, I assume that technology and workers' preferences (for consumption and leisure) are linear. These assumptions allow for an analytical characterization of the model that makes the determination of gross worker flows transparent. Moreover, the simple structure of the model allows me to perfectly identify its shocks using U.S. data.

A key ingredient of any model useful for analyzing unemployment behavior is the presence of search frictions. While search frictions typically create bilateral bargaining situations between workers and employers, I (like Krusell et al., 2012) introduce them in such a way that wages are determined in perfectly competitive labor markets. (1) Moreover, the simple structure that is assumed allows for a single wage rate in the whole economy. Both features are obtained by assuming that all firms in the economy produce in a single geographical location where workers can move from one firm to another in a frictionless way.

Not all workers are present in the production location, though: Some of them are situated in a separate geographical location in which they are able to enjoy leisure. Search frictions are introduced by assuming that it is difficult to move from the leisure location to the production location. In particular, agents are assumed to be able to move from the leisure location to the production location with a fixed probability (interpreted as a job-finding rate). Also, workers who are present in the production location are forced to move to the leisure location with a fixed probability (interpreted as a job-separation rate). Aside from being subject to job-separation shocks that force individuals to move to the leisure location, workers always have the possibility of moving from the production location to the leisure location whenever they wish. Because workers are subject to idiosyncratic labor productivity shocks, they face nontrivial labor supply decisions. Also, because the shocks are idiosyncratic, workers end up moving in and out of employment in an unsynchronized way, generating gross flows across the different labor market states.

It turns out that calibrating the steady state (or long-run equilibrium) of the simple model's economy to average monthly U. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.