At a lunchtime meeting on 18 March, George Alogoskoufis presented results of recent research on the impact of exchange rate regimes and electoral cycles on UK inflation. Alogoskoufis is Reader in Economics at Birkbeck College, London, and a Research Fellow in CEPR's International Macroeconomics programme. His remarks were based on his CEPR Discussion Paper No. 657, `Wage Inflation, Electoral Uncertainty and the Exchange Rate Regime: Theory and UK Evidence', written jointly with Ben Lockwood and Apostolis Philippopoulos. Financial support for the research presented in this paper was provided by grants from the Ford Foundation and Alfred P Sloan Foundations to CEPR's research programme in International Macroeconomics. Financial support for the meeting was provided by the UK Economic and Social Research Council. The views expressed by Dr Alogoskoufis were his own, however, and not those of the above organizations nor of CEPR, which takes no institutional policy positions.

Alogoskoufis noted that recent developments in the theory of inflation have focused on its interactions with institutional structures, public and private sector incentives and in particular governments' inability to precommit to price stability. Governments have incentives to create `surprise' inflation to reduce unemployment, but if the private sector knows this and sets wages accordingly, the cost of such inflation to the authorities will rise to eliminate such incentives. To make a credible commitment to price stability, a government must build a reputation for keeping its promises or create an institution that will constrain its actions after private sector wages have been set such as an independent central bank.

Alogoskoufis described the `partisan' model of inflation, in which `socialists' are more concerned than `conservatives' about unemployment and therefore have a greater incentive to inflate, but both governments achieve the same level of unemployment. Once this model is extended to allow for unemployment persistence, Alogoskoufis maintained that a `socialist' government should generate higher inflation throughout its term of office, and inflation should be higher for either party in the period following an election the `election year' than in the rest of its term of office.

He also described the `exchange rate regime' model as a possible institutional mechanism to commit to price stability. As in the case of the appointment of independent, inflation-averse central bankers, participation in a fixed exchange rate regime `ties the hands' of domestic policy-makers. Monetary policy is determined by an anti-inflationary foreign central bank, as the `leader' in the system. Alogoskoufis maintained that inflation should be lower and less persistent under a fixed rate regime than under a managed float when governments are more likely to attempt `surprise' inflations.
Alogoskoufis then reported the results of an empirical investigation of these models for the UK during 1952-90. The choice of exchange rate regime appears to have exerted a significant effect on expected price inflation and hence on nominal wage growth. Expected price inflation was independent of unemployment, the party in power and the timing of elections during the Bretton Woods era, when UK governments were effectively `buying' US anti-inflationary credibility through their commitment to the exchange rate. Under the post-Bretton Woods managed float, expected inflation has been both higher and more persistent, and it seems to have depended positively on the unemployment rate.

Alogoskoufis also reported that differences between the anti- inflationary behaviour of Labour and Conservative administrations since the collapse of the Bretton Woods system have been very small in non-election years, which may indicate that their levels of commitment to reducing unemployment were rather similar, while election years were associated with higher expected inflation for both parties, with Labour administrations slightly more inflationary than Conservative ones.

Extending the `partisan' model to consider the Thatcher administrations as a third, distinct regime, Alogoskoufis found that there were no significant differences in expected inflation between Labour and pre-Thatcher Conservative governments, in election or non-election years. Elections appear to have had no impact under the Thatcher governments, however: since 1979, election years have been no more inflationary than non-election years. This may indicate that nobody believed these governments could lose these elections.

Alogoskoufis concluded that his results provided strong support for the `exchange rate regime' and `dynamic partisan' models; the lasting anti-inflationary contribution of the Thatcher regime may have been to eliminate the electoral inflation cycle that had pervaded the UK economy since the collapse of Bretton Woods. It would therefore be dangerous and short-sighted to devalue and thereby give away all the anti-inflationary credibility gained from the last eighteen months' membership of the ERM. Such a policy would probably not work, and the government would need another eighteen months to recover its credibility. The need for a realignment is a problem not for the UK but for all the ERM member countries outside Germany: a revaluation of the Deutschmark would be the least inflationary solution for all.