After a brief period of decline in the second half of the 1980s, Spanish unemployment was on the rise again until recently. The unemployment rate reached a staggering 24.2% in 1994 and stands in 1995 at 22.5%, a figure which represents more than twice the average unemployment rate in the European Union and almost three times the corresponding rate in the OECD. What is really striking is that at the beginning of the 1970s, unemployment was around 2% and that during the last economic expansion in the 1980s, with GDP growing over 5% and employment creation taking place at 4% per year, it never fell below the 16% mark. Although conclusions from Spanish labour market studies differ somewhat, they tend to share a common theme. Namely, the evidence suggests that one must allow for the role of persistence and, thus, for the history of the shocks, in explaining such high structural unemployment. Increasingly, therefore, attention has switched away from investigation of the original sources of the rise in unemployment, towards the question of how the effects of shocks are propagated over time.

In Discussion Paper No. 1334, Research Fellow Juan Dolado and David López-Salido use long-run restrictions on a three-variable system containing output growth, real wage growth and the differenced unemployment rate, to isolate three `structural' shocks which drove business cycle fluctuations in Spain during 1970–94. These shocks are interpreted as aggregate demand, labour demand and labour supply disturbances in a framework where there is unit-root persistence in the unemployment rate. The basic finding is that disinflationary policies in an economy suffering from high persistence can become very costly in terms of unemployment, unless supply-side reforms, aimed at eliminating the sources of persistence, are implemented.