Exchange Rate Management
Target Zones and the EMS

At a lunchtime meeting on 29 October, held to mark the launch of Exchange Rate Targets and Currency Bands , Marcus Miller discussed the results and progress of recent research into the theory of `target zones' and their implications for the development of the international financial system in particular the Exchange Rate Mechanism of the EMS. Miller is Professor of Economics at the University of Warwick and has been Co-Director of the International Macroeconomics programme of the Centre for Economic Policy Research. The meeting formed part of CEPR's research programme on Financial and Monetary Integration in Europe, which is supported by the Commission of the European Communities under its SPES programme and additionally by the Ford Foundation. Further financial support was provided by Cambridge University Press. The views expressed by Professor Miller were his own, however, not those of any of the above organizations nor of CEPR, which takes no institutional policy positions.
Miller first noted that research programmes in economics usually emerge from the intersection of a new analytical approach and a real economic problem. In the last few years such a programme has emerged in international monetary economics, which applies the analytical techniques of `stochastic process switching' used in modern finance theory to study the international arrangements under which national monetary authorities attempt to keep their exchange rates within currency bands or target zones. Such policies have assumed increased importance in the light of the unexpected success of the Exchange Rate Mechanism of the EMS.
Miller noted that Robert Flood and Peter Garber had first applied these techniques to study sterling's behaviour before the UK's return to the Gold Standard in 1925; but it was the world-wide move away from freely floating national currencies in the mid-1980s that provided the main stimulus to academic research in this area. In the US, there was a lively debate on the merits of stabilizing the dollar vis-à-vis the currencies of its major trading partners. The US authorities first sought to drive the dollar down with threats of coordinated intervention on foreign exchange markets and then tried to stabilize exchange rates with the 1987 Louvre Accord. Within Europe, the EMS proved much more durable than many had predicted; and in 1987 the ERM members moved further towards exchange rate stabilization by providing for increased reserve support and avoiding realignments within the ERM.
Paul Krugman developed his target zone model, presented in Chapter 2 of this volume, to explain how credible threats to intervene could help to stabilize exchange rates and to provide explicit, non-linear solutions for the relation between the exchange rate and the `fundamentals' when intervention is expected at the edge of the band. If this theory is correct, the exchange rate should stay closer to the centre of the band than would a freely floating rate; and this `honeymoon' effect can defer the need for intervention. Miller noted that other papers in the volume show how intramarginal intervention may be accommodated without changing the basic results, study the role of reserves in determining the sustainability of a fixed rate system, and test the theory's predictions for currency bands or regime switches.
Miller then discussed the major contributions to research in this field in the last year, some of which were presented at the CEPR workshop on `Exchange Rate Target Zones' on 27/28 October. Much of this research has sought to fit target zone models to the available data. Bertola and Svensson have found that the components left unexplained by simple, `first generation', non-linear models of the Krugman type are quite large. They therefore propose improving the empirical performance by including sources of exchange rate and interest rate fluctuations other than the fundamentals traditionally included. Their revised model, in which the likelihood and size of devaluations vary over time, may explain some of the conflicting findings in the empirical literature. Rose and Svensson have found that the behaviour of estimates of expected rates of depreciation within the band derived for daily FFr/DM exchange rate data for 1979-90 accords well with the Bertola-Svensson approach. Swedish daily data for the 1980s reveal that interventions occurred throughout the bands and almost every day. Miller concluded that for empirical applications Krugman's canonical model must be extended to allow both for intramarginal intervention and for the time-varying risk of devaluation.
Miller also argued that a central bank that is introducing a target zone may allow for a `learning process' by the public, which will initially doubt the credibility of the currency band. This `premium' should diminish over time, however, if the authorities defend it at the margin and there are no realignments. Any such realignment will revive fears of future realignments and hence reduce the current parity's credibility by setting back the public's process of `learning to trust' its central bank.
Miller argued that such adverse effects of realignments on perceived devaluation risk can be offset if a central bank can signal that future realignments are in fact being made more difficult. The Italian authorities sought to do this when they `devalued' the lira in 1990, when in fact they narrowed the band to the bottom 4.5% of its existing 12% range and argued strongly for increased monetary integration in Europe. Italy's subsequent experience of inflation suggests, however, that these signals did not work; although this may be a consequence of the fiscal deficits and not a reflection on the central bank. Any significant realignment by the UK after a year of ERM membership would certainly tend to undermine sterling's credibility; but with its very different fiscal position, the UK might be able to follow the Italian example without suffering such a credibility loss, particularly if this formed part of a general EMS `final realignment' before the move to a single currency.