Unemployment-Rate Dynamics and Persistent Unemployment Under Rational Expectations

Abstract

This paper develops a model of unemployment rate dynamics that provides an explanation of persistent cyclical unemployment that does not involve persistent expectational errors or other nonoptimizing behavior. Our results are based on the interaction of search dynamics and inventory adjustments. An important element in these dynamics appears to be heterogeneity in the labor force which can be characterized as consisting of a relatively small group of high turnover individuals who comprise the bulk of normal unemployment and a larger group of low turnover individuals who dominate movements in cyclical unemployment. Our empirical results provide support for this theory as we demonstrate that the appropriately measured probability of becoming employed during a recovery falls relative to normal because of the unusually high proportion of low turnover individuals who have lost "permanent" jobs. As a result, recovery is much slower than is indicated by normal relationships although each individual is searching optimally.Macroeconomists typically build models of the determination of real output with unemployment-rate movements explained, if at all, by an appended Okun's Law relationship. En this paper we show that this approach is misleading because it misses important sources of persistence in cyclical unemployment --and hence -real output. In particular, we show that unemployment-rate dynamics imply "humped-shaped" cyclical unemployment characterized by persistence even though unemployed workers have faulty information on wage rates for only a comparatively brief period of time. This result is based on the interaction of two stock or state variablesinventories and cyclically unemployed workers --which result in a brief *The authors acknowledge helpful comments from

Unemployment durations and the pattern of duration dependence over the business cycle of British males

“…Turon (2003), using aggregate data, concludes that in Britain inflow heterogeneity is counter-cyclical. The analysis in this paper also identifies countercyclical inflow heterogeneity and this is not in support the claim of Darby et al (1985). 2 The empirical evidence for Britain concerning the variation over the business cycle in the hazard of leaving unemployment points to an increase in the probability of leaving unemployment with labour market tightness, i.e.…”

“…Darby et al (1985) argue that in a situation of a loose labour market, i.e. few vacancies relative to the number of unemployed, the newly unemployment have relatively lower probabilities of leaving unemployment (lower employability).…”

“…This is referred to as pro-cyclical inflow heterogeneity. Darby et al (1985) claim that the main reason for longer unemployment durations during an economic recession is pro-cyclical inflow heterogeneity. Turon (2003), using aggregate data, concludes that in Britain inflow heterogeneity is counter-cyclical.…”

Unemployed and Long-Term Unemployed in Croatia: Evidence from Labour Force Survey

“…Darby, Haltiwanger, and Plant (1985) argue that the job fi nding probability declines during recessions, because workers who become unemployed during recessions are different than workers who become unemployed during expansions. According to Shimer (2007), there are substantial fl uctuations in unemployed workers' job fi nding probability at business cycle frequencies, while the probability that a worker exits employment is comparatively acyclic.…”

Reassessing the Unemployment Fluctuations from a Different Perspective

“…In this framework, employment exit rates are the main responsible factor behind the unemployment rates. Darby, Haltiwanger and Plant (1986) point out that the number and the types of entries to unemployment pool is the most important factor in unemployment incidence [see also Darby, Haltiwanger and Plant (1985)] . According to their ‚heterogeneity hypothesis‛ the exit from employment and the entries to unemployment, are the main essential elements in the explanation of unemployment rates.…”