Hysteresis in Unemployment: Evidence from 48 U.S. States.

Studies using standard unit-root tests generally cannot reject the hypothesis of a unit root in unemployment rates. These findings have been interpreted as supporting the hysteresis hypothesis. The authors demonstrate that the failure to reject the null may be due to the low power of the tests. They analyze unemployment rates of forty-eight U.S. contiguous states using a more powerful panel-based test which exploits cross-section restrictions on the constituent series. The critical values are simulated tailored to the authors' sample size. They reject the null decisively, casting some doubt on the hysteresis hypothesis for the United States. Copyright 1997 by Oxford University Press.