There is general agreement that the EMS has helped reduce exchange rate instability within the System. It is also claimed that this stabilization has facilitated convergence on German inflation rates, but convergence may have resulted from a general shift towards inflation control around 1979. In Discussion Paper No. 296, Research Fellow Michael Artis and D Nachane investigate whether the EMS has contributed independently to the convergence of inflation rates.

The basis for supposing some counter-inflationary effect of the EMS, the authors note, is its effect on expectations. Parity realignments in the System do not completely offset inflation differentials. If member countries allowed their inflation rates to diverge strongly, the result would be mounting and unsustainable losses of competitiveness relative to low- inflation Germany. The declared commitment to EMS membership may therefore convince agents to expect an inflation rate consistent with the maintenance of stable nominal exchange rates against Germany.
Artis and Nachane first examine whether a robust forecast of inflation in an EMS member country can be improved by adding information about the outlook for German inflation. They estimate forecasting equations for EMS countries based on lagged values of the inflation rate itself; this technique generally provides robust inflation forecasts. They then add to these equations terms reflecting comparable forecasts of German inflation. The estimates reveal that since 1979 these forecasts are closer to the actual path of inflation once they are supplemented by the information on German forecasts. Adding information on German inflation forecasts in the pre-EMS period has either a much smaller effect or none at all, and it is irrelevant to the forecasts of UK inflation in both periods. These results lend support to the hypothesis of German counter-inflationary leadership.
Cointegration tests have proved quite successful in establishing the presence of underlying `equilibrium' relationships between variables which are subject to high short-run volatility. Artis and Nachane use this technique to establish whether there is a long-run relationship between prices and wages in other EMS countries and those in Germany. The tests do suggest a long-run relationship between German and non-German inflation rates under the EMS. But they also reveal such a relationship with Italy in the pre-EMS period and with the United Kingdom since 1979, so this result does not support the hypothesis of an independent effect of the System in inducing inflation convergence.
The hypothesis of German leadership also depends on intra-EMS nominal exchange rates being sufficiently stable, and being expected to remain so. The authors conduct cointegration tests of cross rates between pairs of currencies, but these do not appear to uphold either nominal or real stability in EMS bilateral exchange rates. Although there are further aspects that could still be explored, Artis and Nachane conclude that convergence of inflation rates could as well be explained by a homogeneous group of countries reacting similarly to the external inflation shock associated with the 1979 oil price increase as by an independent effect of the EMS

Wages and Prices in Europe: A Test of the German Leadership Thesis
Michael J Artis and D Nachane

Discussion Paper No. 296, March 1989 (IM)