Over the past twenty years, there has been an increase in the level of unemployment, and in the persistence of changes in the unemployment rate in most OECD countries. The two possible explanations for these medium-term changes in unemployment have involved hysteresis, on the one hand, and structural factors on the other. In Discussion Paper No. 1082, Marco Bianchi and Gylfi Zoega evaluate the empirical evidence for these two sets of explanations without assuming one specific model as the correct one for changes in equilibrium unemployment. Using data for 17 OECD countries over the period 1960–90, they find that there have been infrequent shifts in the mean rate of unemployment in OECD countries. Most countries share two such shifts, one occurring after 1973, the other after 1979. The size of these shifts can be explained by variables which determine real wage flexibility. They also calculate that the persistence of unemployment is much reduced by taking into account the fairly infrequent shifts in the mean rate. Now unemployment becomes a stationary process with low values for coefficients of lagged unemployment for all countries except France, Italy, Norway and Spain.

The authors suggest that this leaves two possible explanations. First, the global rise in unemployment was caused by changes in the price of energy, in the mid- and late-1970s, and possibly also in the price of capital in the early 1980s. Second, the adverse supply shocks caused a depression and high levels of unemployment through monetary channels which then had a persistent effect on unemployment through one of the unemployment hysteresis effects.

Unemployment Persistence: Does the Size of the Shock Matter?
Marco Bianchi and Gylfi Zoega

Discussion Paper No. 1082, December 1994 (HR)