2 European Developments
- Joseph Gold
- Published Date:
- March 1990
Fluctuation of Exchange Rates
For the purpose of this monograph, the fluctuation of exchange rates means all the movements in exchange rates that are possible under the exchange arrangements in force in the discretionary system. A characteristic of such a system is that it makes predictability of exchange rate behavior extremely difficult. This difficulty generates problems for parties engaged in international payments. They have to be aware that wide and erratic variations can take place in the exchange rates that affect their costs, revenues, and profit margins.
Fluctuating exchange rates have had an impact not only on parties that make international payments but also on both public international law and national law. The effects on national law have been felt in many of its specialized branches. Some of the legal effects considered in this monograph will be selected from international law and others from national law. Discussion will concentrate largely on legal developments that have occurred already. Numerous suggestions have been made for other changes in international law, usually to improve the international monetary system as a whole or the EMS, but these ideas are still largely in the political arena and are not the subject matter of this study. Some proposed changes in national law will be considered, but these are not obviously political.
Another prefatory comment that must be made is that many rules of law, particularly of national law, that apply to exchange rates antedate the decline and abrogation of the par value system but are still in force. The occasions for applying these rules, however, may be more frequent than in the past and, by demonstrating new facets of old problems, may produce refinements in the rules.
European Monetary System
In the relations among states, the most important legal consequence of the de facto fluctuation of exchange rates after August 15, 1971 has been the decision to validate the discretionary system of exchange rates by the Second Amendment of the IMF’s Articles. The second most important consequence has been the creation of the EMS. It may seem paradoxical to attribute to fluctuating exchange rates credit for the creation of the EMS and all that it implies for legal, economic, and political innovation. But it is irrefutable that the EMS has been created because of, and as a reaction against, the instability of exchange rates after the collapse and disappearance by law of the par value system. It is also irrefutable that the Second Amendment has provided room for the establishment of the EMS.
The EMS represents a departure in a sense from the spirit of the discretionary system, which is to return greater autonomy over exchange rates to individual governments than they retained under the par value system. It is sometimes said that a fundamental characteristic of the par value system was the surrender of a substantial measure of sovereignty over the external value of currencies to an international organization. The Second Amendment withdraws from the organization much of the authority that states had granted to the IMF by the original Articles. The participants in the EMS have exercised their freedom under the discretionary system to establish the EMS as a cooperative system based on a pooling of much national authority over exchange rates. If the par value system can be said to have involved a surrender by states of part of their sovereignty over exchange rates, so too does the EMS. The surrender, however, is to the collectivity consisting of the participants in the EMS and not to the IMF.
Much has been written about the EMS. Only some of its features that are of particular interest in connection with the fluctuation of exchange rates will be discussed here.
The Treaty of Rome, 1957, which established the European Economic Community, contained few provisions on exchange rates. It is safe to assume that the negotiators relied, as did the negotiators of so many other treaties, on the provisions of the IMF’s Articles and the IMF’s administration of them to preserve the stability of exchange rates, although not rigidly if there was economic justification for changes in rates. The member states of the European Community (EC) began to have doubts about the soundness, and perhaps even the permanence, of the par value system in the late 1960s, but the breakdown of the par value system made it essential to introduce a more structured system of exchange rates if the objectives of the EC were not to be frustrated. Instability in the exchange rates of European currencies that did not reflect the underlying economic conditions in those countries was attributable, at least in part, to the powerful effect of fluctuations in the exchange rate for the U.S. dollar.
After a meeting of the European Council on July 6 and 7, 1978, a statement was issued in which it was said that the Council had discussed a scheme for the creation of closer monetary cooperation “leading to a zone of monetary stability in Europe.”1 The scheme was outlined in an annex to the statement, but the scheme was greatly elaborated in a resolution of the Council of December 5, 1978 “on the establishment of the European Monetary System (EMS) and related matters.”2 Section A.3, which sets forth “The exchange rate and intervention mechanisms,” and Section A.5, entitled “Third countries and international organizations,” are reproduced in this chapter as Appendix D.
The core of the exchange rate mechanism of the EMS is that each participant establishes a central rate in terms of the composite of currencies that constitutes the ECU (European currency unit). The central rates create a grid of bilateral exchange relationships between each pair of participants in the EMS. Each participant must see to it that exchange rates in transactions between its currency and the currency of another participant are not above or below limits of 2.25 percent of the bilateral relationship between the currencies, but provision is made for some participants to maintain limits of 6 percent, subject to the condition that they must gradually reduce the limits to 2.25 percent as soon as economic conditions permit. Spain and the United Kingdom avail themselves of this option, and Italy did so at one time but decided on January 5, 1990 to apply margins of 2.25 percent.
The Agreement of March 13, 1979 among the central banks of the then eight members of the EC, signed at Basle, spells out the operating procedures for the EMS.3Section 1, which sets forth provisions on the exchange rate mechanism, will be found in Appendix E. The Basle Agreement was amended on June 10, 1985 and again on November 20, 1987, but changes were not made in Section 1.
There has been some discussion of the extent to which there was a legal foundation for the EMS in the Treaty of Rome. The problem can be considered to have been put to rest by Article 20 of the Single European Act signed in 1986, which has become effective. Article 20 provides that an Article 102A is to be included in the Treaty of Rome. The first paragraph of this new article provides that:
In order to ensure the convergence of economic and monetary policies which is necessary for the further development of the Community, Member States shall cooperate in accordance with the objectives of Article 104. In so doing, they shall take account of the experience acquired in cooperation within the framework of the European Monetary System (EMS) and in developing the ECU, and shall respect existing powers in this field.
In accordance with Article 104 of the Treaty of Rome,
Each Member State shall pursue the economic policy needed to ensure the equilibrium of its overall balance of payments and to maintain confidence in its currency, while taking care to ensure a high level of employment and a stable level of prices.
Differences Between EMS and Par Value Systems
It is clear that some aspects of the EMS have been inspired by the original par value system, but there are basic differences. Some of them will be described in the following subsections.
The common denominator of the par value system was gold, but for the purpose of the EMS it is the ECU. If the par value system of Schedule C of the Articles were to be brought into operation, the preferred common denominator would be the SDR, but there is no compulsion to choose the SDR. It is difficult to foresee what the common denominator would be if it were not the SDR, because the common denominator must not be gold or a currency.4 It is highly unlikely that the ECU would be chosen, because it is composed exclusively of European currencies.
Suppose that the par value system of Schedule C were called into operation. Participants in the EMS could have double and different obligations on the margins for exchange rates in exchange transactions between the currencies of partners in the EMS. The participants could observe the narrower margins of Schedule C or the EMS, as the case might be, in the hope that this practice would constitute compliance with the other margins. Compliance would not be assured by this strategy, because the common denominators of the two systems would be different. The margins under Schedule C would be 4½ percent above and below parities or such other margin or margins as the IMF might adopt.5 It is an open question whether the IMF could exercise its power to establish “such other margin or margins” by adopting for exchange transactions between participants in the EMS the margin imposed by the EMS and another margin for other exchange transactions. In view of the experience of the IMF and the discussions of multicurrency intervention in the work preceding the Second Amendment, the IMF probably has this power. If it were exercised in the way mentioned above, the problem of conflicting margins for participants in the EMS would not arise.
An alternative for participants in the EMS would be to refrain from establishing par values for their currencies under Schedule C. A member is not compelled to have a par value under the schedule.6 If a member does not establish a par value for its currency, the member is free to choose its exchange arrangement, but the obligations of the Articles would have to be observed. The principal obligations would be those included in Article IV, Section 1, but, in addition, the member would be obliged to consult with the IMF under the surveillance procedures and would be affected by specific principles adopted by the IMF for the guidance of members with respect to their exchange rate policies. The problems for participants raised here are not pressing, because there is no likelihood that the par value system of Schedule C will be called into being unless there is a radical change in the attitudes of the major industrialized countries.
For a decision of the IMF to call the par value system into operation, a majority of 85 percent of the total voting power of the membership is necessary.7 If it is supposed that the participants in the EMS decided that they would not establish par values for their currencies, they might be disinclined to vote in favor of a decision to activate the par value system of Schedule C. Voting by a member in favor of that decision, however, would not prevent the member from exercising its privilege of not establishing a par value for its currency.
It is implicit in the foregoing discussion that a participant in the EMS in its role as a member of the IMF is not, and cannot be, released from its obligations under the Articles. The right of members to join in cooperative arrangements like the EMS is recognized by the Articles,8 but exercise of the right in no way releases a participant from any of the obligations imposed on it by the Articles. This legal position is recognized by Paragraph 5.3 of the European Council’s Resolution of December 5, 1978:
The EMS is and will remain fully compatible with the relevant articles of the IMF Agreement.9
Paragraph 5.3 has a double function. Its first function is to assert that the participants are in no way acting contrary to their obligations under the Articles by collaborating in the EMS. It may be supposed that this manifesto was intended to counteract any suggestion that the EMS might be contrary to provisions of the Articles on exchange rates or to specific principles for the guidance of members with respect to their exchange rate policies that the IMF had adopted or might adopt. Another purpose the manifesto might have been intended to serve was to repudiate any suggestion that the creation of the ECU would be incompatible with the objective of making the SDR the principal reserve asset in the international monetary system.10
The second function of Paragraph 5.3 of the European Council’s Resolution of December 5, 1978 seems clearly to be to give an assurance to non-EMS countries and to the IMF that the EMS would remain fully compatible with the Articles whatever changes might be made in the EMS. No pronouncement by the IMF has declared that there was a problem of incompatibility of the EMS with the Articles.
In the par value system, the obligation of members to ensure observance of the margins for exchange rates was not accompanied by an explicit requirement that the obligation was to be performed by means of official intervention in the exchange market. A member’s obligation was to adopt “appropriate measures consistent with this Agreement”11 to see that transactions were conducted at exchange rates in accord with the margins. For example, instead of intervening, a member might perform this obligation by adopting and enforcing a law prescribing margins consistent with the Articles or by monopolizing exchange transactions and engaging in them at the proper exchange rates. As recalled in Chapter 1, if a member undertook to buy gold from and sell it to the monetary authorities of other members in return for its currency as the means of settling international transactions, the member did not have to intervene in the exchange market or take any other measure. In the operation of the par value system, the United States was the only member that gave the undertaking in respect of its currency, while most other members intervened, in most instances by buying and selling the national currency for U.S. dollars.
Schedule C of the Articles does not mention intervention, but times have changed and there is less support for such “appropriate measures” as the official monopolization of exchange transactions. As a consequence, the IMF is directed by Article IV, Section 4 to make its determination on whether to call the par value system of Schedule C into operation in the light of specified developments, which include “arrangements for intervention and the treatment of imbalances.” Nothing in Schedule C or Article IV resembles the special dispensation awarded by the original Articles to a member willing to buy and sell gold for its currency at prices consistent with the par value for the currency.
The EMS includes more specific obligations on intervention. Participants are required to intervene at the limits of the margins in whatever amounts prove to be necessary. If the exchange rate between the currencies of two participants reaches the limit of the margin either above or below the parity, both participants must intervene: the participant with the strong currency sells that currency in exchange for the weak currency, while the participant with the weak currency buys that currency in exchange for the strong currency. The obligation to intervene at the limit of a margin does not prevent intramarginal intervention. “In principle,” intervention is conducted with the currencies of participants, but “in principle” indicates that other currencies may be used. Intramarginal intervention is conducted in accordance with understandings reached among the participants and subject to a participant’s consent if its currency is to be used.
Intramarginal intervention has been attractive because it reduces fluctuations in the exchange rate of a currency and impedes a tendency of the currency to move to the limit of a margin. Speculation increases as an exchange rate moves toward a limit, and a member might have to intervene with larger amounts at the limit than would be necessary if it intervened within the margins. The greater proportion of intervention has been intramarginal and has been carried out with U.S. dollars. The EMS includes no provision resembling the one on gold transactions in the par value system that releases a participant in the EMS exchange rate system from the obligation to intervene because the participant is following a particular practice deemed to be equivalent to intervention.
Intramarginal intervention can take place before or after the divergence threshold, as discussed below, is crossed. Intramarginal interventions have decreased the importance of both compulsory intervention and the divergence indicator. Furthermore, the volume of intervention with U.S. dollars has reduced somewhat the role of the ECU in the operation of the EMS. For reasons such as these, some observers have questioned whether the ECU has succeeded in taking up its position at “the centre of the EMS” in accordance with the intention of the Council’s Resolution of December 5, 1978. It is equally, and perhaps even more, obvious that the SDR has not become “the principal reserve asset in the international monetary system” in accordance with Article VIII, Section 7 and Article XXII.
For the purpose of the present study, it is important to note that an understanding has been reached on a reduction in intramarginal intervention. One of the substitute measures to be observed is greater tolerance of the flexibility of exchange rates that is possible within the margins.
A feature of the EMS of considerable ingenuity and novelty is the “divergence indicator,” for which there was nothing comparable in the par value system and there is nothing comparable in Schedule C of the present Articles. The divergence indicator is the result of a conflict between participants that preferred an EMS based on bilateral parities and participants that argued in favor of a system based on central rates in relation to the ECU. Intervention points in the parity system would be at specified levels above and below a bilateral parity, while in the other system the intervention points would be at levels above and below the central rate for a currency in terms of the ECU.
The result of the dispute was a compromise in which, as has been seen, the compulsory intervention points are related to bilateral parities, while intramarginal intervention might be indicated on the basis of the deviation of the market exchange rate of a currency from its ECU-related value. The arguments for and against each solution were complex, but one of the objections to the parity solution was that if the limit of a margin was reached, it would not be clear which of the two participants whose currencies were involved in the parity relationship was responsible for this development. The solution of the ECU-related central rate, on which the divergence indicator is based, would succeed in showing which participant was allowing its currency to fall out of line with the currencies of other participants. If a currency reached the limit of a margin based on an ECU-related central rate, this development meant that the currency was diverging, whether by appreciation or depreciation, from an average of EC currencies, because the ECU is a weighted composite of EC currencies.
Each currency has an ECU-related central rate expressed as a given quantity of the currency per ECU. (For this reason, it is more correct to speak of a central rate per ECU in terms of each currency than a certain rate for each currency in terms of the ECU.) The central rates are the basis on which bilateral parities can be calculated between a unit of a currency and all other currencies of participants in the exchange rate mechanism of the EMS. The ECU-related central rate is a fixed value. The ECU also has a market rate in terms of each currency, which fluctuates. This rate for a currency is determined by the sum of the equivalents in that currency of the units (or fractions of units) of each of the currencies constituting the ECU. The rates used for making this calculation are the market rates between the currency in question and the other currencies. If, in the market of this currency, the other currencies in the ECU basket appreciate, the rate of the ECU in terms of the currency will appreciate. Similarly, if in the same market the other currencies depreciate, the rate of the ECU in terms of the currency in question will depreciate. Therefore, depending on the way exchange rates move, the market rate of the ECU in terms of a currency will stand at a premium or at a discount against the ECU-related central rate of the currency. There would be identity between the two rates only if the currency were to be at parity with all other currencies in the market.
As the degree of bilateral appreciation or depreciation between currencies in the market must not exceed 2.25 percent, it is possible to calculate the maximum spread of divergence or, in other words, the maximum percentage by which the market rate of the ECU in terms of a specific currency can be allowed to appreciate or depreciate against the ECU-related central rate of that currency. The maximum is reached when in its market the specific currency is at its margin of fluctuation of 2.25 percent in the same direction against all other currencies in the ECU.
The maximum divergence of the exchange rate of a currency from its ECU-related central rate, however, will not equal 2.25 percent even though the market rates for the other currencies have deviated by 2.25 percent in the same direction from their parities. The reason is that the component in the ECU of the currency in question is a fixed value, which by definition does not change. A currency cannot appreciate or depreciate against itself. The maximum divergence from ECU-related central rates, therefore, will not be the same for all currencies. The divergence will be closer to 2.25 percent the smaller the weight of a currency in the ECU basket. As a result, the lighter currency would reach its divergence threshold sooner than a heavier currency unless an adjustment was made that neutralized the weight of a currency in the ECU for the purpose of computing the threshold of the currency. Conversely, the heavier the weight of a currency in the ECU, the less will the value of the ECU in terms of the currency be influenced by fluctuations in the exchange rates of other currencies in the ECU.
Further adjustments must be made to prevent distortions that could result from the inclusion of the Greek drachma in the ECU even though Greece does not yet participate in the exchange rate and intervention arrangements of the EMS. Another adjustment is necessary if participants are availing themselves of wider margins of 6 percent above and below the parities for their currencies with other currencies in these arrangements. Only Spain and the United Kingdom do so now. The adjustments made are based on notional margins of 2.25 percent for the currencies of Greece, Spain, and the United Kingdom, and deductions are made from the appreciation or depreciation shown by the market rate of the ECU in terms of a currency that can be attributed to movements in the exchange rates for the three currencies outside their notional margins.12
The divergence threshold of a currency has been mentioned above. This concept is an essential element in the operation of the divergence indicator. It shows the extent to which a currency is nearing its maximum spread by relating the premium or the discount in the market rate of the ECU in terms of the currency to the maximum legal spread. When a currency reaches 75 percent of its maximum divergence from its ECU-related central rate according to the adjusted calculation described above, the currency is at its divergence threshold. Normally, a currency will reach its divergence threshold before reaching its bilateral limit against another currency under the parity system based on central rates, but there may be circumstances in which the development of exchange rates will depart from this sequence. Nevertheless, it was hoped that the warning signal given by a crossing of the divergence threshold would be sounded before bilateral margins were reached. It will be recalled that the theory of the divergence indicator is that it shows the extent to which the exchange rate of a currency is departing from the average of the exchange rates of currencies represented by the ECU.
No disciplinary consequences follow for a participant if its currency reaches its divergence threshold, but if the currency crosses the threshold, a presumption arises that the participant should take action to mitigate the tensions that can be, or are already being, provoked in the exchange rate system of the EMS. Before the threshold is reached, flexibility seems desirable, but beyond the threshold fluctuation to this extent is deemed to threaten the degree of stability that is the objective of the EMS as a zone of monetary stability. No more than a presumption to act arises in order to give a participant the opportunity to show that developments in the exchange rate for its currency are the consequence not of its policies but of an inflationary bias in the system resulting from the inflationary policies of other participants.
If a participant’s divergence threshold is crossed, the presumption is that the participant will choose among the ameliorative measures of diversified intervention, measures of domestic monetary policy, changes in central rates, and other measures of economic policy, or that the participant will give effect to some combination of these measures. If, in special circumstances, a participant does not take adequate measures, it must explain this inaction to the other participants in the various bodies in which their representatives meet.
Diversified intervention implies that a participant may intervene with a number of currencies, so as to distribute the effects among participants and not concentrate the effects on any one participant as the single involuntary creditor. There is no implication that the intervening participant should be constrained to choose those currencies for which the exchange rates are furthest away in the opposite direction from its own currency.
Intramarginal intervention is recognized as a measure that may be adopted if a participant’s currency crosses its divergence threshold, but the rule remains that the consent must be obtained of the participant that issues the currency to be used for such intervention. A participant approached for this purpose is not bound to consent, but the September 1987 amendment of the Basle Agreement suggests that usually consent will be forthcoming and that a refusal will be explained. The phrase “diversified intervention,” however, does not mean that intramarginal intervention in such circumstances can be conducted only with the currencies of participants. A non-EC currency, such as the U.S. dollar, could be used.13
The par value system of the original Articles (and of Schedule C) and the EMS are similar in that external values for currencies are fixed and margins are prescribed for exchange rates in exchange transactions. The EMS is an example, however, of a system referred to with approval by the Outline of Reform presented to the IMF’s Board of Governors by its Committee on Reform of the International Monetary System and Related Issues (the Committee of Twenty) on June 14, 1974. The Outline of Reform declared that a future exchange rate system would continue to be based on stable but adjustable par values, with provisions for floating exchange rates in particular situations, subject to authorization, surveillance, and review by the IMF.14 The Outline of Reform declared, however, that:
It is agreed that it would be desirable that the system of exchange margins and intervention should be more symmetrical than that which existed in practice under the Bretton Woods system.15
The Outline of Reform did not explain what was meant by greater symmetry, because the Committee had not reached agreement on this matter. The stage reached in discussions on topics that remained controversial was described in a number of Annexes to the Outline of Reform. Annex 316 set forth possible systems of exchange margins and intervention, one of which was called a system based on “multicurrency intervention.” This system was described as follows:
A number of countries (including those whose currencies were the most widely traded in world foreign exchange markets) would meet their obligations under paragraph 12 of the Outline by undertaking reciprocal responsibility to maintain the spot exchange rate for their currencies (the MCI currencies) within maximum margins of 4½ per cent above or below parity against each other by intervening in each other’s currencies at the margins. For this purpose they would publish mutually consistent limit prices at which they would be prepared freely either to buy, or to sell, or to both buy and sell, each other’s currencies.17
Such a system was considered more symmetrical than the par value system of the original Articles because all participants would undertake to intervene at the margin in accordance with the agreement on intervention, while under the original Articles the United States was able to refrain from intervention because of its policy on engaging in gold transactions with the monetary authorities of other members. Furthermore, there would be symmetry because the effective margins for all participants would be the same. Under the par value system, members intervened with U.S. dollars at rates consistent with the margins prescribed by the Articles, which meant that the margins for transactions between the currencies of two members that each intervened with dollars were the cumulation of the margins each member observed against the dollar. The result was that less flexibility for the fluctuation of exchange rates existed for the dollar than for other currencies.
The exchange rate and intervention arrangements of the EMS avoid both the asymmetrical aspects of the original par value system as described above and also a further problem of this character. The IMF recognized that under the par value system members were free to allow the exchange rates for their currencies in spot exchange transactions within their territories to fluctuate within the margins of 1 percent of parity above and below parity. This freedom seemed clear under the original Articles. Again, under what seemed to be the clear language of the Articles, members could exercise this freedom to the full extent in all exchange transactions.18 As noted above, if two members (Patria and Terra) intervened with the same intervention currency, the margins from parity for exchange rates in transactions between their two currencies would be the cumulation of each margin against the intervention currency. Suppose that Patria applied margins of 0.75 percent for exchange rates in transactions between Patria’s currency and Patria’s intervention currency. Did it follow that Terra had to observe margins of no more than 0.25 percent for exchange rates in transactions between Terra’s currency and its intervention currency? If Terra observed a margin wider than 0.25 percent, exchange rates in exchange transactions between Terra’s and Patria’s currencies would be outside the permitted margins. To have reached the conclusion that Terra was in violation of its obligation would have been inequitable, and yet the IMF never resolved this problem.
Instead, the IMF took a bolder approach to the problem. At all times in its history—which is to say under all versions of the Articles—the IMF has been able to approve multiple currency practices.19 In 1959, it seemed that a number of European and other members, which had been availing themselves of the dispensations of transitional arrangements,20 might be on the verge of making their currencies convertible for nonresidents and then fully convertible in accordance with the Articles.21 Members would be encouraged to take this step and allow greater freedom for their exchange markets if the members could permit greater fluctuation in the exchange rates for their currencies and therefore reduce the need for intervention in their markets, which meant that they could be somewhat more economical in the use of their monetary reserves.
A number of European countries agreed among themselves that each would impose a margin for exchange transactions between its own currency and its intervention currency that would make use of most of the permissible 1 percent. It would follow that from time to time exchange rates in transactions between two such currencies might move outside the margins of 1 percent from parity between the currencies. The IMF decided that it would approve as a multiple currency practice an arrangement in which a member permitted exchange rates for spot exchange transactions that exceeded 1 percent but were no more than 2 percent from parity. A condition of this approval was that the rates resulted from the maintenance of margins not in excess of 1 percent from parity for exchange transactions between the member’s currency and the convertible currency of another member, which by implication would usually be the first member’s intervention currency.22
The decision authorized greater fluctuation for exchange rates in order to make it easier for members to undertake the full convertibility of their currencies in accordance with the Articles. But there were some murmurs. Some of the European members that issued currencies of substantial international importance did not relish the analysis that their exchange arrangements constituted multiple currency practices.23 Such practices were more widespread among developing countries. There was also some suggestion that such a practice, by which all exchange rates were the result of market forces, was not a multiple currency practice in economic theory. In other words, the member’s exchange system was seen to be a unitary one by these critics even if the margins were not observed in all exchange transactions.
The decision did not resolve all potential problems: difficulties might have arisen under it because the IMF did not attempt to apportion the use of margins. For example, Terra and Patria both intervened with the same third currency, and observed margins of 0.75 percent, so that margins of 1½ percent could emerge in transactions between the currencies of Terra and Patria. If Regio intervened with the currency of Terra at margins of more than ½ of 1 percent, exchange rates in some transactions involving the currency of Regio could transgress the margins of 2 percent approved by the IMF in exercising its multiple currency jurisdiction. The episode suggests a less than full dedication to the observance of international economic obligations if they are seen to obstruct advance to a superior economic objective.
In this discussion of margins, it will be apparent that intramarginal intervention was not subject to regulation under the par value system. A member could choose freely its intervention currency and could decide whether, and at what exchange rate, to intervene, provided the member respected the legal margins. In contrast to this freedom, the EMS provides that intramarginal intervention by a participant with the currency of another participant can be conducted only with the consent of that other participant. It will be recalled that consent is necessary even though intramarginal intervention is foreseen as a measure that can be taken when a currency crosses its intervention threshold.
At least one deduction can be drawn from the foregoing discussion. The experience of the par value system, the IMF’s decision approving margins of 2 percent, Schedule C of the present Articles,24 and the example of the EMS show that in systems of fixed exchange rates reasonable margins for the fluctuation of exchange rates are desirable. The rigors of the original par value system inspired the negotiators of the Second Amendment to agree on margins as wide as 4.5 percent and even wider margins if 85 percent of the total voting power of members favored such a step. The participants in the EMS, however, have regarded margins of such width to be inconsistent with their concept of a zone of monetary stability.
Another deduction that can be drawn is that it may be desirable to provide for some discretion to establish margins considered desirable from time to time. The EMS contains no explicit provision on varying the margins of 2.25 percent, but it seems that they could be changed by the same legal process that permitted the establishment of the present margins or by whatever legal process might be in force.
Changes in par values could be made under the par value system, and changes in central rates can be made under the EMS. Differences between the two systems are evident in the criteria for changes and in the majorities required for endorsing proposed changes.
The criterion for changing the par value of a member’s currency was the existence of a fundamental disequilibrium.25 The Articles did not state that the fundamental disequilibrium had to be in the member’s balance of payments. A member might be pursuing policies that suppressed disequilibrium in the balance of payments. An obvious example of such a policy would be the imposition of sweeping restrictions on trade or payments. It is clear, however, that the criterion of fundamental disequilibrium was a severe one, and that this severity was intended to discourage the proposal of changes that were unnecessary if the member’s problem was temporary or cyclical or could be solved by different policies that were consistent with the letter and spirit of the Articles and were feasible. Exchange rates would not be stable if par values could be changed for insubstantial reasons.
The criterion of fundamental disequilibrium has been retained in Schedule C, but the severity of the original Articles has been mitigated in one respect. It has been made clear that there is no need to await the development of a fundamental disequilibrium to justify a change in par value. Under paragraph 6 of Schedule C a member may propose a change either to correct a present fundamental disequilibrium or to prevent the emergence of such a disequilibrium.
The provisions of the EMS say nothing about the criterion for changes in central rates. Nothing is said about the occasion for, or the size of, an adjustment. The original Articles did not say that a change had to be made if there was a fundamental disequilibrium. The criterion for the occasion of a change was stated negatively: a member was not to propose a new par value except to correct a fundamental disequilibrium, which meant unless a fundamental disequilibrium was in existence. The Articles dealt with the size of a proposed change as well as with the occasion for a change. A proposal to change a par value had to be sufficient to correct the fundamental disequilibrium, which was understood to mean that the change must be neither inadequate nor excessive. The amount, of course, could not be determined mechanically, and might well be subject to differences of opinion.
The negotiators of the EMS probably agreed that criteria were undesirable because they would have to be imprecise. Agreement on precise criteria was unlikely, and therefore little, if anything, was to be gained by vague formulations. If, nevertheless, precise criteria were technically feasible, they would be undesirable, because they were likely to become known, and then they would provoke speculation if the markets judged that a realignment of central rates in conformity with the criteria was in prospect. Similar reasoning was probably responsible for the absence of a definition of the concept of fundamental disequilibrium in the IMF’s Articles and for the IMF’s avoidance of a clarification during almost the whole life of the par value system.26
The hypothesis on which the exchange rate and intervention arrangements of the EMS are based is that if a desirable convergence in the policies of participants is attained, exchange rates are likely to be stable. The hypothesis resembles the assumption that underlies the provisions of the IMF’s present Articles on a stable system of exchange rates.
If, notwithstanding the hypothesis mentioned above, the convergence of policies is not achieved, or unexpected difficulties develop, the solution might have to be the adjustment of central rates to counteract the economic disturbance that has been caused. The idea of a zone of monetary stability might imply that economic disturbance should not be allowed to persist to the point at which more than modest changes would have to be made in central rates, so as to avoid agitation or bewilderment in the exchange markets because of large and convulsive changes.
With this approach, procedure becomes particularly important, because in providing protection procedure may be a substitute for meticulous economic criteria. For adjusting central rates, “mutual agreement by a common procedure” is prescribed. All participants in the exchange rate and intervention arrangements of the EMS and the Commission of the European Communities take part in the procedure. The quoted language is cumbersome and is probably intended to emphasize the requirement of unanimity. Still, the rule of unanimity is not made explicit. It would have been easy to express the rule, and to avoid the weightiness of referring to both mutual agreement and a common procedure. Perhaps the intention was that there must be a meeting, at which the participants can confront each other. Whatever the intention may have been, the procedure is extraordinary in the sense that exchange rates are determined by collective action. The Smithsonian Agreement reached in December 1971 after the par value system had broken down was considered a unique event because the agreement emerged from a meeting called to bring about a realignment of the exchange rates of a number of major currencies by collective consent. The establishment of the initial par values of currencies in December 1946 under the authority of the IMF has a certain similarity to the Smithsonian negotiation, but the approach to initial par values was to endorse existing exchange rates rather than to realign them.
If unanimity, on all occasions or in all but special situations, is the rule for changes in central rates of the EMS, the contrast with the par value system could hardly be stronger. For par values, a decision by the IMF to concur in or object to a proposed change required no more than the basic majority of voting power in the IMF’s practice. That a majority of the votes cast according to the IMF’s system of weighted voting power should suffice for a decision on a change is astonishing. The negotiators of the Articles were accepting international authority over a cherished aspect of sovereignty, and they were consenting that this authority could be exercised by a majority that was far short of a consensus or a widespread agreement among members. The modest majority meant that a substantial minority of voting power or even a majority of votes if not cast could not have the effect of vetoing a member’s proposal to change the par value of its currency, which gave the member an advantage in requesting the concurrence of the IMF. Viewed in this way, the basic majority can be considered a concession to sovereignty. The fact that no more than a majority of the votes cast was sufficient for a decision of the IMF to concur in a change of par value may help to explain why the original Articles placed so much emphasis on the avoidance of competitive devaluation.
A final difference between the par value system and the EMS for discussion here relates to the financing of intervention. Under the EMS, participants are required to intervene if the exchange rates for their currencies are at the limits of the prescribed margins, so as to ensure that the limits will not be transgressed. To enable participants to perform this obligation, participating central banks must open for each other “very short-term” credit facilities. Participants must make such resources automatically available in unlimited amounts for this form of intervention.
Participants with strong currencies were not eager to consent to intramarginal intervention with their currencies or to provide credit for this purpose because of the effect on the money supply of these participants and their fear of inflation. For the reasons already explained, participants with weak currencies were inclined to engage in intramarginal intervention. They considered the restraints of the Basle Agreement to be asymmetrical, because these participants would have to conduct intramarginal intervention with their monetary reserves and without the help of participants with strong currencies.
In September 1987 the Basle Agreement was amended. Participants had become increasingly disposed to make speculation more uncertain in circumstances in which capital was freer to move among participants because of reduced capital controls. In these conditions, exchange rates might become more unsettled. It would not be wholly advantageous for participants with strong currencies if all intervention took place at the limits of the margins, because these participants also would have to intervene and accept the economic consequences.
The right of a participant to refuse consent to intramarginal intervention with its currency has not been abrogated, but this form of intervention is now contemplated by the agreement of September 1987. Although consent is not mandatory, there is reason to think that participants have accepted a presumption that consent will be forthcoming. There is a similar presumption that if there is consent to the use of a currency in intramarginal intervention, the use will qualify for the very short-term financing facility for this purpose. Access to the facility, however, is subject to a prescribed maximum, and in some circumstances to certain conditions, such as the concurrent use by the intervening participant of its own reserves and reimbursement of the creditor in its own currency. A condition of concurrent use of a participant’s own reserves would resemble the provisions of the original Articles of the IMF under which in the year a member drew on the IMF’s resources the member was expected to make equal concurrent use of its own reserves, provided that those reserves were above a certain level. If the member did not observe this principle, an obligation to repurchase with reserves an appropriate amount of the IMF’s holdings of its currency would accrue that would put the member in the same situation as if it had observed the principle.
The IMF’s Articles did not refer to intervention and so did not mention the use of the IMF’s resources for this purpose. The Articles declared that a member’s use of the resources was justified if the member had a present need to draw on them.27 If resources were sought for intervention in support of a par value, the test of need was met. The original Articles did not provide that a member had an unchallengeable right to make use of the IMF’s resources, but the First Amendment, which became effective on July 28, 1969, authorized such use of the limited amount that was equivalent to a member’s “gold tranche.”28
The concept of the gold tranche was changed to the “reserve tranche” by the Second Amendment for the purpose of defining the limit of the unchallengeable right.29 The Second Amendment,30 like the First Amendment,31 forbids the IMF from recognizing further unchallengeable rights to use the IMF’s resources. This effect is achieved by the legal technique of providing that the IMF must have policies on the use of its general resources32 and that members’ uses of the resources must be consistent with the IMF’s policies, except that requests for reserve tranche transactions shall not be subject to challenge.33 The provisions of the EMS and the IMF’s Articles differ radically, therefore, on the extent to which unchallengeable use can be made of resources in support of exchange rates. The provisions differ also on such topics as the periods of use, the charges for use, and the means of settlement when use is terminated.
The amount of outstanding use that a member of the IMF and a participant in the EMS can make is yet another difference between the two monetary arrangements. It has been seen that the amount of its currency a participant must make available to another participant for mandatory intervention is unlimited, although a limit is imposed on the amount to be made available for intramarginal intervention. The Articles establish a limit on the outstanding use that a member can make of the IMF’s resources. The limit is equal to the member’s quota, but without prejudice to use of the reserve tranche and use under certain special policies.34 However, the IMF can, and frequently does, waive the limit.35
Another limitation on the use of the IMF’s resources must be noted. A member may not use the resources to meet a large or sustained outflow of capital,36 although a member may use its reserve tranche for this purpose.37 The limit has been largely ineffective, especially after the substantial reduction in capital controls occurred. Short-term capital flows in particular could have undesired effects on exchange rates. The necessary flexibility in permitting use of the IMF’s resources notwithstanding capital outflow has been found in the undefined adjectives “large” and “sustained” as the kind of outflows that cannot be financed with the help of the IMF’s resources. The apparent difference between the EMS and the Articles in this respect is not of great practical importance, and perhaps of no importance, except that the limitation is present in the Articles and could be invoked by the IMF. The EMS contains no provision that limits use of the very short-term credit facility by withholding resources because they are needed to deal with the consequences of capital outflow.
This discussion of the use of resources available under international auspices that can be employed in support of exchange rates must return to the subject of “need” to use the IMF’s resources, because it will be seen to have some relationship to the EMS. The Articles declare that a member seeking to use the IMF’s resources must have a need “because of its balance of payments or its reserve position or developments in its reserves.”38 These three phrases connote three separate justifications for use of the IMF’s resources, or, in other words, represent three different kinds of “need.” This interpretation is made clear by the Commentary included in the report of the IMF’s Executive Board to the Board of Governors on the proposed text of the Second Amendment.39 The Commentary states that:
Under the concept of need in Article V, Section 3(b) (ii), a member will be able to purchase the currencies of other members from the Fund if its balance of payments position or its reserve position is unfavorable, or if there is an unfavorable development in its reserves, e.g., because of an impending discharge of liabilities, even though it does not have a deficit in its balance of payments according to accepted definitions of the balance of payments.40
Participants in the European narrow margins arrangement (the “snake”), which preceded the EMS, used the IMF’s resources to settle obligations to other participants that arose as the result of intervention in accordance with the arrangement. The participants made these uses even though their reserves were high and the participants were not in deficit in the balance of payments.41 When the Second Amendment was drafted, the language in which the requirement of need is expressed was clarified, on the initiative of some European members, by referring expressly to “developments in…reserves.” This clause and the explanation in the Commentary were intended to place beyond doubt the propriety of using the IMF’s resources to settle obligations that accrued in accordance with the “snake.”
Obligations arising as the result of intervention under the EMS, now that the EMS has replaced the “snake,” would be equal justification for a participant’s use of the IMF’s resources, provided that the participant met the criteria of the IMF’s policy under which the use was requested. This fact is recognized in the Basle Agreement itself. The Agreement declares that insofar as settlement is only pardally effected by transferring holdings of the creditor central bank’s currency42 and ECUs from the debtor to the creditor central bank, the balance must be settled by transferring other reserve assets in accordance with the composition of the debtor’s reserves as they stood at the end of the month preceding the settlement. The reserves for this purpose are defined as assets denominated in SDRs or in currencies, and the debtor central bank may choose among the assets in each of the two categories to settle the amount to be settled with assets in the category.43 The reserve tranche of a member in the IMF is a reserve asset44 and it is denominated in SDRs.45
At this point, it is appropriate to evaluate briefly the success of the EMS so far. There is little doubt that the EMS has succeeded in establishing a zone of considerable monetary stability in Europe by means of cooperation in monetary matters and coordination in the management of exchange rates. Flexible administration of the EMS has been possible because of the discretions left to administrators by sparse and broadly drafted rules. The amendments of the Basle Agreement have contributed to, rather than restricted, the flexible management of exchange rates. The EMS can be regarded as an omen of the better regulation of exchange rates that is possible in the era of the discretionary system of exchange arrangements.
The next stage in the history of the EMS will arrive after the introduction of the single market in 1992 and the steps taken to bring about monetary union. A lesson that may be learned already is that greater freedom for capital flows need not be feared as disruptive because of the influence of exchange rate expectations. On the contrary, if the stability of exchange rates seems to be assured, as appears to be the condition achieved by the EMS, capital flows may be guided by less speculative motives. Capital flows will then be more beneficial to savers and to enterprises.46
Resolution of the European Council of 5 December 1978 on the Establishment of the European Monetary System (EMS) and related matters
A. The European Monetary System
3. The exchange rate and intervention mechanisms
3.1 Each currency will have an ECU-related central rate. These central rates will be used to establish a grid of bilateral exchange rates.
Around these exchange rates fluctuation margins of ±2.25 percent will be established. EEC countries with presently floating currencies may opt for wider margins up to ±6 percent at the outset of the EMS; these margins should be gradually reduced as soon as economic conditions permit.
A Member State which does not participate in the exchange rate mechanism at the outset may participate at a later date.
3.2 Adjustments of central rates will be subject to mutual agreement by a common procedure which will comprise all countries participating in the exchange rate mechanism and the Commission. There will be reciprocal consultation in the Community framework about important decisions concerning exchange rate policy between countries participating and any country not participating in the system.
3.3 In principle, interventions will be made in participating currencies.
3.4 Intervention in participating currencies is compulsory when the intervention points defined by the fluctuation margins are reached.
3.5 An ECU basket formula will be used as an indicator to detect divergences between Community currencies. A ‘threshold of divergence’ will be fixed at 75 percent of the maximum spread of divergence for each currency. It will be calculated in such a way as to eliminate the influence of weight on the probability of reaching the threshold.
3.6 When a currency crosses its ‘threshold of divergence’, this results in a presumption that the authorities concerned will correct this situation by adequate measures, namely:
- (a) diversified intervention;
- (b) measures of domestic monetary policy;
- (c) changes in central rates;
- (d) other measures of economic policy.
In case such measures, on account of special circumstances, are not taken, the reasons for this shall be given to the other authorities, especially in the ‘concertation between central banks’.
Consultations will, if necessary, then take place in the appropriate Community bodies, including the Council of Ministers.
After six months these provisions shall be reviewed in the light of experience. At that date the questions regarding imbalances accumulated by divergent creditor or debtor countries will be studied as well.
3.7 A Very Short-Term Facility of an unlimited amount will be established. Settlements will be made 45 days after the end of the month of intervention with the possibility of prolongation for another three months for amounts limited to the size of debtor quotas in the Short-Term Monetary Support.
3.8 To serve as a means of settlement, an initial supply of ECUs will be provided by the EMCF against the deposit of 20 percent of gold and 20 percent of dollar reserves currently held by central banks.
This operation will take the form of specified, revolving swap arrangements. By periodical review and by an appropriate procedure it will be ensured that each central bank will maintain a deposit of at least 20 percent of these reserves with the EMCF. A Member State not participating in the exchange rate mechanism may participate in this initial operation on the basis described above.
5. Third countries and international organizations
5.1 The durability of the EMS and its international implications require coordination of exchange rate policies vis-à-vis third countries and, as far as possible, a concertation with the monetary authorities of those countries.
5.2 European countries with particularly close economic and financial ties with the European Communities may participate in the exchange rate and intervention mechanisms.
Participation will be based upon agreements between central banks; these agreements will be communicated to the Council and the Commission of the European Communities.
5.3 The EMS is and will remain fully compatible with the relevant articles of the IMF Agreement.
Agreement between the central banks of the Member States of the European Economic Community laying down the operating procedures for the European Monetary System [Basle Agreement]
The central banks of the Member States of the European Economic Community,
Have agreed as follows:
1. Exchange rate mechanism
Article 1—Central rates in terms of the ECU
Each participating central bank shall notify the Secretariat of the Committee of Governors of the Central Banks of the Member States of the European Economic Community of a central rate in terms of the ECU for its currency. The Secretariat shall pass on this information to the other central banks and the European Communities.
Article 2—Intervention rules
2.1 Each participating central bank shall notify the Secretariat of the Committee of Governors of the rates for compulsory intervention expressed in its currency, and the Secretariat shall pass on this information to the other central banks. These rates shall be fixed in relation to the bilateral central rates derived from the central rates in terms of the ECU referred to in Article 1 of the present Agreement. The market shall be notified of them.
2.2 Interventions shall in principle be effected in currencies of the participating central banks. These interventions shall be unlimited at the compulsory intervention rates. Other interventions in the foreign exchange market shall be conducted in accordance with the relevant guidelines that were adopted by the Committee of Governors in its Report of 9 December 1975 or that may be adopted in the future, or shall be subject to concertation among all the participating central banks.
Article 3—Operation of the indicator of divergence
3.1 On either side of the central rate for its currency in terms of the ECU each participating central bank shall establish rates for its currency in terms of the ECU that will constitute ‘thresholds of divergence’. These thresholds of divergence shall be calculated in such a way as to neutralize the influence of the differences in weights on the probability of their being reached; they shall be set at 75 percent of the maximum divergence spread, this being measured by the percentage difference between the daily rate and the central rate of a currency against the ECU when that currency is standing at the opposite pole from all the other currencies at the compulsory intervention rates referred to in Article 2.1 of the present Agreement. The necessary steps shall be taken to take account of the effects of the adoption of different maximum margins of fluctuation for the participating currencies and of the possible non-participation of a currency in the exchange rate mechanism.
3.2 If a currency crosses a divergence threshold, this shall entail the consequences set out in paragraph 3.6 of the Resolution of the European Council of 5 December 1978.
Article 4—Method of calculating the values of the ECU in each currency
For the purposes of the operation of the indicator of divergence provided for under Article 3 of the present Agreement, the market value of the ECU in each currency shall be calculated by a uniform method as frequently as necessary and at least on the occasion of each daily concertation session among central banks.
Any central bank that is not participating in the exchange rate mechanism shall cooperate with the other central banks in the concertation and the other exchanges of information necessary for the proper functioning of the exchange rate mechanism.
“The credit facilities which the EC countries have granted each other are already so extensive and so easily accessible that they enable us, as noted in an earlier Report, largely to avoid recourse to the IMF with its more stringent conditionality and to postpone necessary policy adjustments within Europe. Ultimately this is bad both for Europe and for the IMF.
In this context it should also be noted that both the Netherlands and the other EC countries have undertaken to help the IMF special drawing rights (SDRs) evolve into the most important reserve asset and to reduce the role of reserve currencies. Promotion of the ECU as a new reserve asset will inevitably be detrimental to the cause of the SDR. No matter how fervently both goals are desired, in practice a choice will have to be made.
In summary: do we regard the ECU as the future common currency of Europe, as a substitute for the national currencies? Or do we see it as one of the world’s reserve currencies, contrary to our earlier intention to replace these reserve currencies by the SDR?”
“Each member that has a par value for its currency undertakes to apply appropriate measures consistent with this Agreement in order to ensure that the maximum and the minimum rates for spot exchange transactions taking place within its territories between its currency and the currencies of other members maintaining par values shall not differ from parity by more than four and one-half percent or by such other margin or margins as the Fund may adopt by an eighty-five percent majority of the total voting power.”
See also David Folkerts-Landau and Donald J. Mathieson, The European Monetary System in the Context of European Financial Markets, IMF Occasional Paper, No. 66 (Washington: International Monetary Fund, 1989); U.K. Treasury, An Evolutionary Approach to Economic and Monetary Union (November 1989); and ECU Newsletter, No. 29 (July 1989), Editorial, pp. 3–5.
“Upon completing the large internal market in 1992, the most visible factors separating national economies will be adjustable exchange rates. The economies of Europe will not be fully integrated until individuals and enterprises can take longterm economic decisions on matters such as investment, production and employment without having to worry about differing interest rates or changing exchange rates. Further progress towards economic and monetary union would considerably add to the benefits that will flow from the completion of the large internal market. The EMS in its ten short years, has already brought us a long way along this path, I am sure it can help us to go much further.”