Job duration and the cleansing and sullying effects of recessions

A central question in economics is how business cycles affect the allocation of resources. Focusing on the labor market, an unresolved issue is whether recessions lead to above or below average productive arrangements. Typical models of the labor market imply that recessions cleanse the labor market as low quality employer-employee matches are destroyed and only exceptionally high quality matches are created. These models, however, ignore the potential sullying effect of recessions through on-the-job search, i.e. the process by which workers transit between jobs without an intervening spell of unemployment. Theory suggests this process is stymied in recessions and leads to worse matches. ; Distinguishing between these two possible effects of recessions is important for monetary policy. If recessions are cleansing, then accommodative policy may be delaying the natural process of creative reallocation. Alternatively, if recessions are sullying, then accommodative policy may improve labor market outcomes by stimulating job creation and promoting worker reallocation. ; This paper quantifies these alternative views of recessions by using data from the National Longitudinal Survey of Youth (NLSY) from 1979 to 2006. Using the duration of an employer-employee relationship as a proxy for its quality, the analysis tests for the cleansing and sullying effects of recessions. The results provide no systematic evidence for the cleansing effect, but do suggest a role for the sullying effect. As predicted by theory, the quality of matches formed from workers transiting between jobs, without an intervening spell of unemployment, falls in recessions. This suggests an active role for monetary policy in the labor market during recessions for the aforementioned reasons.