The theory and practive of floating exchange rates and the role of official exchange-market intervention

BIS Economic Papers No 5
February 1982


After a short spell of relative calm that gave rise to hopes that the initial teething problems of the floating rate system had finally been overcome, the period since the second half of 1977 has again been one of pronounced exchange-market unrest. Some of the exchange rate movements that occurred during this period were undoubtedly justified by changes in basic underlying conditions and were therefore in the interests of longer-term equilibrium. At the same time, however, there can be little doubt that a substantial proportion of these fluctuations was unnecessarily large or even undesirable and entailed on occasion serious conflicts between external and domestic economic policy requirements. Moreover, this instability occurred despite a level of official exchange-market intervention that sometimes dwarfed the amount of intervention that had to be undertaken under the fixed rate system.

This paper seeks to shed some light on the causes of these problems and to evaluate the case for official intervention or other exchange rate policies by comparing how a floating rate system should ideally respond to certain kinds of "disturbances" with the kind of reaction patterns that are likely to ensue in a world of major economic uncertainties and long lags in adjustment. The approach used is a highly pragmatic one. No heroic effort is made to set up a general equilibrium model of exchange rate determination, and the assumption throughout this paper is that prior to the disturbance the exchange rates were at some kind of longer-term equilibrium level, or were moving along some sort of longer-term equilibrium path.

Section I deals with temporary changes in international interest rate differentials, Section II considers the role of international inflation differentials, Sections III and IV explore the effects of disequilibria caused by temporary and permanent changes in the underlying real sector of the economy and Section V discusses the scope for more permanent changes in real international interest rate differentials. Particular attention is paid throughout to the role of the forward markets and the question of the bias of the forward rate structure. Section VI seeks to illustrate some of the points made in Sections I to V by reviewing recent exchange rate experience. A concluding section attempts to sum up some of the findings.

One assumption that is basic to this paper is that unnecessary exchange rat instability, ie movements in real exchange rates that do not contribute to the adjustment process and are not in the interests of longer-term equilibrium, entails very real economic costs and should therefore be regarded as undesirable. For one thing, it will distort international competitive conditions, reduce the efficiency of resource allocation and add to general economic uncertainties. In a world in which the setting-up of new production facilities requires large fixed capital commitments, pronounced uncertainty about one of the most important parameters determining the commercial soundness of such investment, namely the country's real exchange rat level, is bound to substantially increase the uncertainty premium that is necessary to justify the risks incurred. Exchange rate instability and the related uncertainties will therefore have a negative impact on the level of investment and economic growth. This will tend to be a problem particularly in the case of smaller countries where import penetration is very high and where, in view of the narrowness of the domestic market, commercial viability will very often depend on the ability to sell a large part of output abroad. Moreover, for firms heavily engaged in international trade, exchange rate instability may mean that the possible losses or gains deriving from exchange rat movements will dwarf the revenue from the firm's ordinary business. There will therefore be a danger that foreign exchange management and speculation might absorb a large portion of the prime managerial talent and attention that would otherwise be devoted to innovation in the field of production processes, products and new markets.

Finally, in a world of much less than perfect competition, of widespread oligopolistic market structures, on both the input and the output levels, and pronounced price rigidities in a downward direction, exchange rate instability is bound to give a boost to worldwide inflation. While the domestic inflation rate will tend to react very promptly in those countries with depreciating exchange rates, prices in the appreciating countries will be slow to respond and the response will be rather muted one. Even within one and the same country an exchange rate depreciation that is subsequently reversed is bound to leave domestic prices at a higher level than would have prevailed had exchange rates remained unchanged in the first place. And similar ratchet effects may arise in the field of restrictive trade practices. There is a real danger that protectionist measures taken at a time when an unrealistically high exchange value of the national currency is threatening the survival of certain domestic industries may not be readily unwound when the real exchange rate returns to a more realistic level.

Despite the importance attached to avoiding unnecessary exchange rate instability, this essay is, however, not intended as a plea for a fixed rate system, which, of course, would not afford protection against undesirable real exchange rate movements either. It is the author's view that in the present world environment of pronounced economic and political uncertainties a return to a genuine fixed parity system is out of the question. The sole aim of this paper is therefore to explore to what extent the floating should be "pure" and to what extent there is a case of official intervention and/or for subordinating domestic policies to external constraints.

In fact, to anticipate somewhat the conclusions of this essay, the issue of floating versus fixed exchange rats would seem to have been very much overplayed. A genuine fixed rate system would undoubtedly be feasible in a stable and predictable political and economic environment, in which a pure floating rate system would function very satisfactorily too. Conversely, in a world such as the present one, in which wide inflation differentials and all manner of uncertainties and instabilities preclude a genuine fixed rate system, pure floating is also bound to produce unacceptable results. Thus, the authorities today find themselves in a kind of "no man's land" that may at times combine the disadvantages of both systems, where it is not possible to rely mechanically on a few simple rules but where survival requires the constant exercise of judgement and discretion, although this necessarily entails the risk of error. Of course, all this may not be very palatable to theoretical purists, but the situation here is no different from any other sphere of social relations. While the yearning for simple "black and white" answers to all the complex questions that arise in modern society helps to explain the appeal of ideologies, the simplistic answers and mono-causal explanations offered by these ideologies are usually spurious ones and divert attention from the real problems. There are no simple, ready-made solutions, and truth has to be conquered anew every day. Genuinely fixed parities and pure floating are undoubtedly both very useful as analytical concepts and tools, but neither of the two systems, if applied exclusively, can provide satisfactory solutions to all the problems of international economic relations that arise in the complex and troubled world in which we live today.