Turbulence and Unemployment in Matching Models

Ljungqvist and Sargent (2007) show that increases in turbulence, in the sense of worse skill transition probabilities for workers who suffer involuntary layoffs, generate higher unemployment in a welfare state. den Haan, Haefke and Ramey (2005) challenge this finding and argue that if turbulence also exposes voluntary quits to a tiny risk of skill loss, then higher turbulence leads to a reduction in unemployment. In this paper we explore the source of these disparate results within the two adopted matching models. We find that the latter authors' assumption of additional exposure of the high-skilled unemployed to skill losses following unsuccessful job market encounters, together with the parameterization of their model, implies small incentives for labor mobility in tranquil times. Hence, any small cost to mobility would cause voluntary separations to shut down, e.g., tiny government mandated layoff costs would have counterfactually large effects of suppressing unemployment. Once the additional exposure to turbulence is dismissed and the parameterization is adjusted to account for the unemployment dynamics in data, the positive relationship between turbulence and unemployment reemerges.