A considerable body of research has tried to explain the behaviour of UK unemployment in the 1980s, which rose sharply to levels unprecedented in the post-war period and remained at very high levels for a considerable period. Those arguing that the long-run equilibrium unemployment rate rose and actual unemployment followed the equilibrium rate have for the most part been unable to identify the economic fundamentals that caused equilibrium unemployment to rise. Others have argued that the UK economy experienced a series of temporary adverse shocks, which caused the initial rise in unemployment, which returned only slowly to its equilibrium rate. Both explanations emphasize the role of `persistence' and therefore cannot easily explain the periods sharp rise and fall in unemployment, in 1979-82 and 1987-9 respectively.

In Discussion Paper No. 540, Research Fellow Alan Manning departs from traditional explanations which assume that unemployment has a single equilibrium to consider the possibility that there exist numerous possible equilibrium rates and that the rise in UK unemployment can be viewed as a move from a low-unemployment to a high-unemployment equilibrium with no change in the economic fundamentals and no need for recourse to theories of persistence and sluggish adjustment.

For a simple macroeconomic model based on imperfect competition in product and labour markets, in which the economy may have multiple equilibria whenever production exhibits increasing returns to scale, Manning estimates a multiple equilibrium model and tests it against a traditional model for UK data for 1951-87. He finds some evidence of increasing returns to scale, but it is difficult to reject the hypothesis of a single equilibrium. This uncertainty may have important policy implications: for example, the response of unemployment to changes in exogenous variables may differ in sign depending on the nature of the economy's equilibrium.

Manning uses the estimated model to compute the two equilibrium unemployment rates and compares them to actual unemployment. He finds that actual unemployment in the 1980s was reasonably close to the high-equilibrium rate, which suggests that the UK economy did switch equilibria. For the earlier period, however, neither computed equilibrium rate is very close to the actual rate, although movements in the low-equilibrium and actual rates seem to be correlated. Manning concludes that the UK may have moved from a low- to a high-unemployment equilibrium in the Thatcher years, but the evidence is far from conclusive. Finally he notes that while traditional single equilibrium models suggest that the Thatcher government's policies of reducing union power will if successful reduce equilibrium unemployment, the multiple equilibria model suggests that such policies may even exacerbate the problem.

Multiple Equilibria in the British Labour Market: Some Empirical Evidence
Alan Manning

Discussion Paper No. 540, April 1991 (AM)