At a lunchtime meeting on 23 March, Patrick Minford presented results of recent research on the effects of adopting a single currency in Europe. Minford is Edward Gonner Professor of Applied Economics at the University of Liverpool and a Research Fellow in CEPR's International Macroeconomics programme. His remarks were based on his Discussion Paper No. 656, `The Price of EMU Revisited', written jointly with Andrew Hughes Hallett and Anupam Rastogi. Financial support for this research was provided by the Commission of the European Communities under its SPES programme, while support for the meeting was provided by the UK Economic and Social Research Council. The views expressed by Professor Minford were his own, however, and not those of the above organizations nor of CEPR, which takes no institutional policy positions.

Minford began by considering the claims made by the European Commission's report, One Market, One Money, concerning the potential benefits of introducing a single European currency. This will clearly eliminate transactions costs, which the report estimates are currently 0.4% of Community GDP. It also claims that eliminating exchange rate uncertainty may increase Community GDP by up to 10%, but, according to Minford, offers no serious empirical support for this result. The report argues that the new, independent European Central Bank (ECB) will enhance price stability and policy credibility; but Minford noted that this new institution's anti-inflationary credibility remains to be established, while the Bundesbank, the Bank of Japan and the Federal Reserve System have all demonstrated that both these desirable objectives are attainable without EMU. Price stability is therefore essentially irrelevant to the debate over a single currency.

Minford then turned to the main potential cost of EMU: macroeconomic instability. In the traditional Keynesian literature on `optimal currency areas', with rigid wages and prices, flexible exchange rates and national interest rates enable a country to absorb a shock that disturbs its prices relative to those of its competitors. With repeated shocks, relative wages and prices may eventually adjust to each of them, but even short-run nominal and real rigidity in a monetary union may entail costs since wages and prices necessarily adjust more slowly than exchange and interest rates under a float. Labour mobility and fiscal transfers can reduce the need for such exchange and interest rate stabilizers, as in the US, which exhibits high labour mobility and large federal budget transfers, suggesting that it is an optimal currency area. The European Community, in contrast, has low labour mobility, negligible Community-level fiscal offsets to national shocks, and considerable short-run wage and price rigidity. The loss of the exchange rate and interest rate stabilizers under EMU may therefore have damaging effects on macroeconomic stability.

Minford noted that comparing the results of monetary policy coordination under EMU with uncoordinated or actual monetary behaviour under floating rates is the wrong comparison, since central banks can cooperate with floating rates. He estimated `macroeconomic variability' a weighted average of the variances of prices, output, real interest and exchange rates and money supply growth by simulating large numbers of shocks to the EC economy of the type incurred in the past under a single currency and coordinated floating. His results indicated that Community-level instability under EMU would be approximately double that under floating exchange rates. Simulations on the Liverpool world model indicated that such a floating rate regime outperforms EMU for all EC member countries, with a fixed money supply rule or a strategically responsive monetary policy. Even if the rest of the Community proceeds with EMU, the UK will do better to remain outside. Estimates on the more refined UK quarterly model reinforced this conclusion.

Minford noted the results of the European Commission's similar simulation exercise using the IMF's MULTIMOD world model, which usually exhibits greater wage and price rigidity than the Liverpool models and should therefore yield stronger results. Surprisingly, this exercise indicated that introducing a single currency had a stabilizing effect. This could derive from the inclusion of risk premium shocks under floating rates that were assumed away under EMU, or asymmetric assumptions about monetary policy under the two regimes, either of which would bias the resulting comparison. Minford reported Masson and Symansky's replication of the Commission's simulations using the MULTIMOD model but with consistent assumptions about shocks and policy coordination, which tended to support his own finding using the Liverpool model that instability was potentially minimized under floating.

Minford concluded that the capacity of a European floating rate regime to foster macroeconomic stability if properly exploited approximately doubles that of the proposed single currency. Europe's monetary authorities should not abandon a system that they have used badly in favour of one that severely restricts their ability to accommodate shocks. This need not rule out EMU as a long-term objective, but it certainly indicates that the timetable currently proposed is over-ambitious.