One of the major benefits of the EMS, it is argued, is that it has enabled its members to engage in disinflation at a lower cost in terms of lost output and employment than other developed countries. In Discussion Paper No. 326, Research Fellow Paul De Grauwe reassesses the empirical evidence for the widespread belief that there is a disciplining feature in the Exchange Rate Mechanism of the System that facilitates deflation.
De Grauwe presents graphical comparisons of the inflation and unemployment experiences of the EMS countries with those of the other OECD countries as a whole and with non-EMS members in Europe. Although inflation within the System declined from 11% in 1980 to 2% in 1988, this is not exceptional compared with the rest of the OECD  during the System's existence. The inflation/unemployment trade-off in all OECD countries worsens after the oil shocks of 1973 and 1979, but the increase in unemployment in EMS members after 1979 is significantly higher and their growth rates of output and employment have declined by more than the average in the other OECD countries, including those in Western Europe. Favourable movements in non-EMS countries' unemployment and inflation/unemployment trade-offs after 1983 are not found for EMS members.
Other adverse developments, for example in the labour market, cannot provide a complete explanation for EMS members' relatively poor economic performance, according to De Grauwe. To do so they would have to explain why these factors became so important in the 1980s, since during the 1970s very little difference between EMS and non-EMS OECD countries could be observed. They would also have to explain why they appear to have had less impact on European countries outside the EMS, which are generally agreed to have very similar structures to those of EMS members.
Interpreting this evidence, De Grauwe notes that, in contrast to countries like the United Kingdom and the United States, who let their exchange rates float, the EMS countries could not use a `shock therapy' involving sharp monetary contraction and a steep real appreciation of the currency. A shock therapy quickly establishes the authorities' anti-inflationary reputations, and thereby leads to a quicker reduction of inflationary expectations. The EMS has forced its high-inflation members to follow gradual deflationary policies, he maintains, with the disadvantage that their authorities gain anti-inflationary reputations correspondingly slowly and deflationary policies must be applied for longer.
This does not mean, however, that the EMS may not have advantages it may also make it harder for countries to lose their reputation once they have acquired it. Nor does the evidence suggest that the System is more costly. While the gradual approach of EMS members in the 1980s has tended to postpone the unemployment cost of disinflation, De Grauwe argues, the shock therapy followed by other countries has a tendency to produce a more sudden time path for these losses. It is unclear which strategy has been the least costly over time.

The Cost of Disinflation and the European Monetary System
Paul De Grauwe

Discussion Paper No. 326, July 1989 (IM)