Book
Share
IMF History (1972-1978) Volume 3
Chapter

Report of the Executive Directors

Author(s):
International Monetary Fund
Published Date:
February 1996
Share
  • ShareShare
Show Summary Details

On August 18, 1972, the Managing Director as Chairman of the Executive Board sent to the Board of Governors the following report of the Executive Directors on reform of the international monetary system.

Reform of the International Monetary System

(August 18, 1972)

A Report by the Executive Directors to the Board of Governors

CONTENTS

INTRODUCTION

At the 1971 Annual Meeting of the Fund, the Board of Governors adopted Resolution No. 26-9 on the “International Monetary System,” the full text of which is reproduced in Annex I to this Report. Parts I and II of the Resolution dealt with immediate difficulties. In Part III the Governors addressed themselves to long-term reform and requested the Executive Directors:

  • (a) to make reports to the Board of Governors without delay on the measures that are necessary or desirable for the improvement or reform of the international monetary sysem; and

  • (b) for the purpose of (a), to study all aspects of the international monetary system, including the role of reserve currencies, gold, and special drawing rights, convertibility, the provisions of the Articles with respect to exchange rates, and the problems caused by destabilizing capital movements; and

  • (c) when reporting, to include, if possible, the texts of any amendments of the Articles of Agreement which they consider necessary to give effect to their recommendations.

In response to this request, the Executive Directors have been engaged since the last Annual Meeting on a study which is still under way. Their present report on that study is thus necessarily preliminary in character. It is difficult at this stage to foresee which of the possible courses for long-term development of the international monetary system would best accommodate the diverse positions and interests of member countries, while safeguarding the global character of the system and minimizing the risk of any retreat into protectionism and inward-looking blocs. Much progress has, however, been made in the identification and examination of major issues on which further study and discussion will be required from both policy and technical points of view. The Committee of the Board of Governors on Reform of the International Monetary System and Related Issues provides a new forum at a high policymaking level in which further progress can be made on major policy issues relating to international monetary reform.1

This Report reviews certain options for reform in the main problem areas of the system, setting out the arguments on both sides when there appear to be both advantages and disadvantages in any course of action and eschewing conclusions that would narrow prematurely the field of choice. The Report consists of six chapters. The first chapter reviews the need for reform and the main objectives and major characteristics of any reformed system. Chapter II discusses issues arising in achievement of an improved functioning of the exchange rate mechanism that would be necessary in any reformed international monetary system. Chapter III reviews the desirability of certain changes in arrangements relating to convertibility, the official settlement of imbalances, and exchange market intervention, examining possible approaches and the circumstances in which they might be adopted. Chapter IV reviews the possible roles of the principal reserve assets in any reformed system, in particular in the light of possible changes in arrangements relating to convertibility and official settlements, and discusses possible arrangements that would permit the substitution of SDRs for other reserve assets. Chapter V examines the problems arising for the international monetary system as a result of disruptive capital flows, and discusses possible means of checking such flows and of mitigating both their domestic and external impact. Chapter VI examines the position and interest of the developing countries in the working of the international monetary system and considers possibilities for contributing, within the framework of any reformed system, to an increased flow of resources for development purposes.

The elements of the international monetary system that are considered in these separate chapters are closely interrelated. Thus, for example, decisions relating to any new arrangements for the settlement of payments imbalances would depend heavily on the working and prospective future working of the adjustment process, in particular the exchange rate mechanism. Similarly, a system involving greater exchange rate flexibility for all currencies, including the U.S. dollar, might tend to diminish both the need for reserves and the role of reserve currencies. The future role of reserve currencies, together with other reserve assets, would also to some extent be determined by any settlement arrangements that were adopted.

An important series of questions will arise with respect to the appropriate phasing of the introduction and implementation of various interrelated aspects of a reformed system. A number of the policy options for reform would require amendments to the Fund’s Articles of Agreement and, in view also of the interrelationships among particular elements, it might be considered advantageous to introduce these amendments simultaneously, and to make all the legal changes that would be required in a single set of amendments to the Articles. An important question would be whether these amendments should make it easier under acceptable procedures to adapt the system to changing circumstances in the light of experience.

But though progress toward reform hinges to some extent on the timing of amendment of the Articles, some aspects of reform could, if this were considered useful, be put into practice as soon as they were agreed in the course of the reform discussions, and before any amendment that might be required had become effective. Beyond this, much could be achieved in a variety of areas in advance of any changes in legal provisions. The most substantial contribution should come from the progressive reduction of international payments imbalances in response to the 1971 realignment of exchange rates of major currencies and to other factors. For example, whether members make par value changes promptly when these are needed depends to a large extent on the attitudes of individual members and of the community of members as a whole. The development of more effective ways of dealing with destabilizing capital movements may be largely a matter for concerted national and international action, rather than for amendment of the Articles. Thought should be given to possible preparatory measures that might pave the way to future arrangements for the settlement of payments imbalances by all countries. It should also be noted that two of the possibilities discussed in Chapter VI for increasing the flow of financial resources to the developing countries could be implemented without amendment. Outside the specifically monetary field, progress would, in addition, be promoted by international understandings and agreements aiming at the reduction and gradual elimination of trade and other restrictions that impede not only economic well-being but also the adjustment of payments imbalances.

More generally, there will be a need for collaboration among members in dealing on a pragmatic basis with problems as they arise. Without the continuing cooperation of members to this end, the prospects for any reformed system and the credibility of specific reform proposals would be seriously prejudiced.

CHAPTER I The Need for Reform

1. Features of the Bretton Woods system

The international monetary system created at Bretton Woods was based on a strong affirmation of the desirability of cooperation among all members. The system as originally established and as it subsequently developed has been dominated by three major features:

(1) The par value system was designed to permit the adjustment of exchange rates—but only if necessary for the correction of a fundamental disequilibrium in a member’s economy—and to avoid the competitive depreciation that characterized the interwar period. Under this system exchange rates can be altered when exclusive reliance on domestic policies to correct an external imbalance would produce consequences (for example, in terms of unemployment or inflation) that are unacceptable to the country in question. The system provides that exchange rates are matters of international concern and that, therefore, par values can be adopted or changed, on the proposal of the member, only with the prior concurrence of the Fund.

(2) The stability of exchange rates in the market was sought by obligating each member (unless it were freely buying and selling gold for the settlement of international transactions) to maintain in its territory spot rates between its currency and the currencies of all other members within margins of 1 per cent on either side of parity. In 1959, after most European currencies had become de facto convertible, margins of up to 2 per cent were permitted when they resulted from margins of not more than 1 per cent vis-à-vis the member’s intervention currency. This arrangement presupposed market intervention as the main technique to ensure observance of the margins. It was understandable that the U.S. dollar, which in postwar conditions had acquired a special role, should become the principal medium for intervention and the currency in terms of which most margins for intervention were stated. At the same time, the United States was not required to intervene to ensure the observance of margins in respect of the dollar because it had opted to buy and sell gold freely for the settlement of international transactions. In these circumstances the United States assumed a basically passive role in the exchange market.

(3) The need for early elimination of any restriction and discrimination in payments and transfers for current international transactions is strongly emphasized under the system, and members are obliged to make their currencies convertible as soon as their balances of payments permit. By contrast, members are left with virtually complete autonomy to restrict capital transfers and the system favors such restriction in certain circumstances to safeguard the resources of the Fund. In practice, however, members that made their currencies convertible generally moved in the direction of liberalizing capital movements.

2. The functioning of the Bretton Woods system

The quarter century since establishment of the Bretton Woods system has been characterized by a massive expansion in world trade and payments, but it became increasingly apparent over time that the operation of the system was impeded in certain cases by the mutual incompatibility of members’ domestic and exchange rate policies, and also by the inability of the system to accommodate certain major changes in the international economic and financial environment. These weaknesses in the working of the adjustment process reflected in particular the following specific difficulties.

(1) In the past, especially in the 1960s, the problem of achieving timely and adequate changes in exchange rates became increasingly apparent in cases of both surplus and deficit countries. It arose in acute form in the case of the U.S. dollar. Nothing in the Fund’s Articles precluded a change in the par value of the U.S. dollar, and indeed the history of the negotiations preceding the realignment of December 1971 shows that the willingness of the United States in the end to undertake such a change was a positive factor in promoting the general readjustment of parities which was ultimately achieved. However, it was widely assumed, throughout the postwar period, that par value adjustment was as a practical matter unavailable to the United States. This presumed inability of the United States to adjust the par value of the dollar reflected various asymmetries deriving principally from the position of the United States as the largest economy in the system and from the associated major importance of the United States in world trade and capital transactions. Other special factors were, first, that the United States was the only major country which chose the option of freely buying and selling gold as a means of fulfilling its exchange stability obligations, and, second, the predominant role of the U.S. dollar as a reserve currency to which other currencies were pegged. Inflexibility in the par value of the dollar did not matter in the immediate postwar years when the economic position of the United States—which was providing resources to other countries for postwar reconstruction—was overwhelmingly strong. But it became increasingly troublesome in the 1960s as structural changes and developments in relative costs and prices led to a diminution in the competitive position of the United States. In effect, it left with other countries the main burden of initiative for altering exchange rates; yet such action as other countries considered it possible to take was not in itself sufficient to end the underlying imbalance, and the exchange rate relationship between the dollar and other currencies thus became increasingly inappropriate.

(2) The rise in the volume of international transactions and thus in the magnitude of the financing that might be needed for balance of payments disequilibria made it necessary for the stock of international reserves to rise over time. With the monetary price of gold remaining unchanged, the supply of newly mined gold and the use of Fund facilities met only part of the need for reserves, and the bulk of global reserve creation took the form of increases in holdings of dollars. But though accrual of reserves in this form met a global reserve need and was thus a beneficial influence on economic activity and expansion, it could continue only as long as the United States incurred balance of payments deficits and surplus countries were willing to accept and retain a large proportion of dollars in the financing of their surpluses. It became increasingly apparent, however, that the system was not able to withstand the strains of a method of reserve creation that involved persistent balance of payments disequilibria. Dissatisfaction with reliance on this process as the prime means of reserve creation and the desire to achieve a more rational and efficient mechanism for adapting the world’s supply of international reserves to long-term global needs led to the establishment in 1969 of the facility for special drawing rights. The first decision to allocate SDRs was predicated on the assumption that the earlier process of reserve creation had substantially come to an end. The magnitude of the reserve creation in the form of reserve currencies in 1970 and 1971 showed, however, that no mechanism existed to ensure that the external deficit of the United States and the accumulation of dollars would diminish. Indeed, both assumed unprecedented proportions in 1970 and 1971 under the influence of exchange rate misalignment and massive capital outflows. Moreover, global reserve creation was accelerated further by the buildup of official balances in currencies other than the dollar and, in some degree, by changes in official currency balances in the Euro-dollar market.

(3) The difficulties of the system have been exacerbated by factors that have tended to increase the extent of external imbalances requiring adjustment. For example, countries have been affected in varying degrees by the emergence of cost-push inflation, and experience in curbing inflation by means of incomes and prices policies has been mixed. Increasingly complex problems have arisen in dealing with the external effects of the differences in cyclical phasing as between countries. These difficulties have been compounded by the increased resort that many countries have had to monetary policy for the pursuit of domestic stabilization, as a result, in part, of the relative inflexibility of their fiscal policies. The consequential reduction in the ability and willingness of countries to attune their monetary policies to balance of payments objectives has made balance of payments adjustment more difficult.

(4) The substantial increase in imbalances attributable to capital flows was in addition more generally associated with an enhanced responsiveness of such flows to incentives of all kinds and diminished confidence in the maintenance of fixed par values as the need for adjustment came increasingly to be recognized but needed adjustments were delayed. The marked deterioration in the U.S. basic balance coupled with capital flows of this kind from time to time greatly intensified pressure on the U.S. dollar and on the U.S. gold reserves. Intervention in the private gold market was withdrawn in March 1968. The convertibility of officially held dollar balances into gold or other reserve assets was suspended by the United States on August 15, 1971.

These specific difficulties of the system as it has worked in practice need to be taken fully into account in the design of any viable future system. The international monetary arrangements based on decisions taken on December 18, 1971 constitute a set of arrangements for coping with the most immediate aspects of these difficulties which could otherwise have readily degenerated into situations of serious conflict. But though a more appropriate exchange rate structure of major currencies was achieved in the framework of these decisions, the major long-term problems of the system were left unresolved.

3. Aims for a new system

The need for reform has to be seen in the context of the broad aims of economic policy which, as stated in the Articles of the Fund [Article I], include

the expansion and balanced growth of international trade

and the contribution this makes

to the promotion and maintenance of high levels of employment and real income and to the development of the productive resources of all members….

In the last quarter century members have benefited from their cooperation in a wide range of monetary, trading, and other economic relationships. These cooperative efforts need to be carried further and redirected in various ways. Action may be required not only in the payments field, which is of prime concern to the Fund, but also in respect of the rules governing trade, investment, and aid in which other organizations also are involved. In considering appropriate action in each of these fields it will be necessary to be aware of the close interrelationships among them.

Particular attention should be given to the position and interests of the developing countries. Full account should be taken of various features of the present or any new international monetary system that might involve special problems and difficulties for the developing countries. In addition, consideration should be given to the question whether and how a direct contribution might be made in the framework of the reform to the needs of the developing countries for additional financial resources.

In general terms, the avoidance or elimination of the difficulties that impeded the system in the past and the achievement of the aims mentioned above will depend on the adoption of improved arrangements for the adjustment and settlement of payments imbalances. The essence of an efficient adjustment mechanism is that countries refrain from policies that are likely to lead to severe balance of payments disequilibria and take early action to correct such disequilibria as nevertheless occur. A clearer understanding is needed of the appropriate roles and uses of the various policy tools that are available for basic adjustment, as well as policies intended to deal with temporary disequilibria. Attempts should also be made to find clearer criteria for allocating responsibilities for initiating adjustment among countries in different balance of payments situations. In addition, settlement arrangements for all countries should be such as to promote the objectives of the adjustment process.

A major task will be to ensure that the balance of payments targets of countries, in particular industrial countries, will be compatible with one another as well as realistic both domestically and within the framework of the international economic situation. With few exceptions countries feel more comfortable with an overall balance of payments surplus than with an overall deficit. Insofar as this attitude reflects a general desire of countries for rising reserves it should be possible to meet this desire by an appropriate amount of global reserve creation in the form of SDRs without the need for reliance on some countries to run payments deficits to accommodate the reserve needs of others.

The structure of a country’s balance of payments reflects in a large measure the degree of development of its economy. Countries that are short of capital generally have a current account deficit financed by capital inflows and, in the case of low-income countries, by grants. Most of the highly developed countries normally have a surplus on current account to offset capital exports and the flow of aid to the developing countries. In addition to these basic forces relating to economic structure, countries may be induced by domestic economic, social, and political motives—including the desire to promote manufacturing industry, to further a particularly dynamic sector of the economy, to protect employment, etc.—to follow policies that stimulate exports and discourage imports and, more generally, to aim at a relatively large surplus on current account. Policies to this end tend to be facilitated for an individual member by an exchange rate for its currency which errs on the side of being undervalued in terms of other currencies.

Such preferences in this direction as some industrial countries may have had in the past were in many cases neutralized by domestic inflation and hesitation to resort to devaluation. There has been little in the history of the postwar period to bear out the fears of outright competitive depreciation which preoccupied the authors of the Fund’s Articles of Agreement. Indeed, countries have often postponed devaluation until massive capital outflows superimposed on a deteriorating current account have reduced their reserves to critical levels. However, when a country decided to devalue it often sought to devalue to the maximum extent consistent with the Articles in order to have some safety margin, and this was often considered desirable by other countries, e.g., to make the parity adjustment convincing to the market. The resulting pattern of overvalued and undervalued exchange rates among the industrial countries tended to lead to large balance of payments disequilibria. These were sometimes enlarged by capital flows, with consequent efforts to control or discourage such flows—efforts which in the circumstances were unlikely to be fully successful.

If appropriate par value adjustments by industrial countries were to be made more promptly in future, certain potential conflicts between the balance of payments objectives of these countries would emerge more clearly. Countries would, then, have to face much more directly than in the past the task of gearing their policies—including their exchange rate policies, trading arrangements, and other measures addressed to the balance of payments—to payments targets that are realistic and compatible with one another. These needs, and the severe problems involved, were manifest in the discussions leading up to the Smithsonian agreement.

While the means of improving the adjustment process should be actively pursued, it would be unrealistic to expect that all impediments to efficient adjustment could be overcome in the short run. One basic difficulty that has to be faced is a difference of view about the manner in which divergent tendencies in the current and capital accounts of the balance of payments should be reconciled. Some believe that it is neither practicable nor even desirable to impede at least persistent market tendencies in the flow of capital by an apparatus of controls or artificial incentives and that the current and capital accounts should be accommodated to each other through market forces. Others, while agreeing that many types of capital movements, and particularly long-term movements, are useful and must not be hampered unduly, believe that all countries should take measures to influence or control capital flows (especially speculative flows and perhaps also direct investments) and that, when overall disequilibria arise, the capital as well as the current account should be made to bear its share of the burden of adjustment. At least some of the difficulties arising from these policy divergencies are likely to persist. Moreover, a reformed system could not be expected to eliminate disruptive short-term flows altogether, although it should help to reduce them. Moreover, while there has been notable progress in the postwar period in the reduction of tariff barriers on a multilateral basis, nontariff barriers and regional preferential trading arrangements have increasingly emerged as serious impediments to global international trade. One view is that, as a consequence, some sectors of the domestic economies of industrial countries tend to be shielded from the impact of exchange rate changes by administrative restrictions or other measures which can at times impede needed exchange adjustment in two ways: by making initiation of parity changes more difficult or by increasing the amplitude of change required to achieve a given change in the external position.

In the light of these considerations it seems necessary to accept the conclusion that, if the monetary system of the future is to work effectively, it will need to be able to accommodate national policies which in many cases may not be optimal from the point of view of the adjustment process.

Finally, it must be recognized that however well any new arrangements are designed, their success cannot be ensured without the determined cooperation of members in a more effective functioning of the system as a whole.

CHAPTER II The Exchange Rate Mechanism

The exchange rate mechanism plays a major role in the adjustment process and the unsatisfactory functioning of this mechanism was an important factor in the weaknesses of this process, which have been widely analyzed in recent years. The role of exchange rates in the adjustment process was extensively reviewed by the Executive Directors in their 1970 Report on that subject.2 Their main conclusion was that the par value system was the most appropriate general exchange rate regime but that acceptance of this conclusion carried with it an obligation on all parties to make the system as effective as possible. For the Fund, this was seen as “the need to review and where necessary to adjust its criteria, procedures and operational practices to ensure that they are as well suited as possible” for the effective working of the system. For individual countries, it would involve a “willingness, in instances where exchange rate changes are appropriate to restore equilibrium, to make such changes at a time and in a manner most likely to enhance their effectiveness.”3 The Executive Directors suggested that, in certain cases, prompter and smaller adjustments in members’ par values would help to avoid the building up of large disequilibria and the eventual recourse to large adjustments.

But developments in the period since publication of the 1970 Report have introduced new elements into discussions of the exchange rate mechanism. The widespread recourse to floating rates after August 1971 showed that, even with an intensification of capital controls in many countries, the system as previously operated was subject to pressures beyond its power to handle, and underlined the urgency of making it less crisis-prone. Two inferences may be drawn from this experience. First, protracted uncertainties about future exchange rate relationships may unfavorably affect the general economic climate, in particular when they coincide with a weak cyclical position. Second, floating rates (like fixed rates) require some means of reconciling countries’ interests and avoiding conflicting policies, and it is virtually impossible to negotiate the necessary arrangements at short notice under the impact of sudden shocks to the system.

The experience of international consultation on par value changes in recent years has emphasized the fact that members are concerned not only with their own par value adjustments but also with the par value adjustments proposed by other members. The process leading up to the realignment of December 1971 underlined the considerable difficulty of achieving an adequate effective change in the value of the U.S. dollar against other currencies even when there was wide agreement that such adjustment was needed. For the system to be able to cope in future with changes in the relative value of the dollar as well as of other currencies without recurrence of the sort of disturbance which preceded the 1971 realignment, it is essential that the exchange rate regime, in conjunction with other elements in the system, should reconcile what might otherwise be conflicting policies on the part of major member countries. The difficulties that are likely to arise in this connection have been set out in Chapter I. One further conclusion from that discussion needs to be drawn here. While the point of departure for the Bretton Woods system was to prevent competitive depreciation (in the form of unjustified downward parity adjustments), it is equally important that countries do not fail to adjust the parities of their currencies when these have become undervalued.

Thus a fresh approach is required to the criteria and procedures for orderly change which will accord to the United States, as well as to other members, a due measure of initiative in the effective exercise of exchange rate flexibility. This in turn requires that needed parity adjustment of the dollar should not be vitiated by defensive adjustment of the par values of other currencies.

Against the background of these general reflections and those to which reference was made in Chapter I, a number of specific aspects of increased exchange rate flexibility need to be considered. Consideration of these aspects may help to shed light on the conditions necessary for an effective functioning in present circumstances of any par value system, whether slightly or substantially modified, and hence on the acceptability of such a system.

1. The basis for the determination of par values4

The basic principles of the par value system as it now operates are that any changes in par values should be made with the concurrence of the Fund and that they should be related to the correction of basic payments and economic imbalances, a concept that is expressed in the present Articles in terms of the correction of a “fundamental disequilibrium.” This rules out par value changes—though not fluctuations of market rates within margins—as one means of dealing with cyclical imbalances or with temporary capital flows5 and also provides assurance against the disruption involved in attempts at competitive depreciation. But whereas the principles of the system permit par value changes in certain circumstances and preclude them in others, the system does not provide assurance that adjustment will be made in cases where, with the passage of time and changing circumstances, the par value of a currency has become inappropriate. Experience suggests that the working of a par value system could be improved if its basic principles were made more symmetrical, i.e., if members were not only expected to abstain from par value changes when, in the language of the Articles, there was no “fundamental disequilibrium,” but if, also, they were expected to make such adjustments when they were in “fundamental disequilibrium.”

But though it would be a major improvement, symmetry in this sense would not be sufficient in itself to ensure satisfactory functioning of the par value system, as has been demonstrated in particular by recent experience of capital movements. Two additional conditions would need to be fulfilled to achieve this improvement. First, par value changes would have to be made before the evidence became overwhelming that adjustment was necessary. The resulting prompter and therefore smaller par value changes would tend to reduce the need for and burden of defense of par values that had become inappropriate and would also prevent the buildup of substantial disequilibria which would ultimately require large and disruptive adjustments. Second, in cases where existing par values were regarded as appropriate, the necessary measures would need to be taken, nationally and internationally, to prevent disruption of the system by temporary capital flows. Issues involved in the control and financing of such flows are discussed in Chapter V.

For par value changes to be made more promptly than in the past, understandings would be needed on the identification of situations requiring adjustment and on the magnitude of the change required. These understandings would relate: first, to the magnitude of the maladjustments with respect to which exchange rate correctives should be considered; second, to the procedure for assessing the need for exchange rate adjustments; and third, to the responsibility on both surplus and deficit countries for making par value changes.

To identify situations in which adjustment of the exchange rate of any major currency is required or to assess the magnitude of the par value change that is needed, it would be necessary to take a view about the development of rates for at least the other major currencies. Evaluation and assessment in respect of the structure of these rates could be made as a fairly continuous process and should not be undertaken only when a crisis had occurred or when there was prima facie evidence of a substantial distortion of rates. It must be recognized, however, that inappropriate disclosure of the results of such a review could seriously disturb exchange markets, and might stimulate such large capital flows as to render continued observance of one or several par values impracticable. This risk would be particularly serious in cases where a given member considered that a change in its par value advocated by others would be unjustified. The value of any review of this kind is therefore to a large extent predicated on the existence of adequate arrangements to ensure confidentiality and discretion. These risks would, however, tend to be reduced over time to the extent that the principal of continuous international review of the exchange rate structure came to be accepted, especially if par value changes were in general prompter and smaller than in the past.

A general review of the exchange rate structure of the main industrial countries was undertaken in advance of the December 1971 realignment, when there was general agreement that relatively large parity changes were needed and should be implemented without delay. The exercise pointed to the difficulties involved, even in those circumstances, in forecasting underlying balance of payments trends and in assessing the probable repercussions and interactions of possible exchange rate changes. Moreover, particular difficulties arose—and still remain—in ensuring that the balance of payments objectives of the various countries are both compatible with one another and realistic in terms of what can in practice be achieved, and in ensuring that appropriate account is taken of cyclical variations and of the delayed effects of past developments in exchange rates and in national cost levels.

It will take time to improve techniques of evaluation and assessment involved in reviews of this nature and they are likely to remain subject to a considerable margin of uncertainty. Since, however, the exchange rate of each currency is—to a larger or smaller extent—a matter of concern to all members, it is unavoidable that some judgment as to appropriate rates be made by the international community, and in accordance with the best techniques available.

2. Objective indicators for par value changes

In the face of the difficulties which have been found in practice, both in the discretionary assessment of the need for and in the implementation of par value changes, numerous possible schemes have been suggested which, to a greater or lesser extent, rely on one or more objective indicators to provide the criteria for appropriate parity change. Proposals in this category aim to promote prompter and smaller changes by introducing an element of automaticity in the process of determining parities, thereby establishing a change in par value as a normal instrument in the adjustment process, devoid of political trauma. The schemes which have been suggested rely in the main on objective indicators which are balance of payments oriented, such as movements in a member’s official reserves or in the spot rate for a member’s currency in the foreign exchange market, but some also link par value changes to movements in relative price indices.

Some schemes envisage more or less continuous automatic adjustment of par values in very small steps in response to objective indicators. The advocates of such schemes consider that, although the indicators used cannot be relied upon at all times to pinpoint situations that require exchange rate correction, reliance on such indicators for the determination of exchange rate changes would over time reduce balance of payments problems.

Other proposals for the use of objective indicators envisage somewhat larger and less frequent adjustment of parities and provide for a more or less substantial discretionary element. Under schemes of this nature a specified movement in an objective indicator (or parallel movement in a number of indicators) would trigger international consultation with a member. The movement in the indicator would constitute presumptive evidence of the need for a par value adjustment, in response to which the member would either make an appropriate adjustment or present a case to the effect that, in the circumstances, adjustment was unwarranted.

One category of proposals in this area would envisage the establishment for each country of internationally agreed upper and lower limits for reserves. Reserve movements within these limits would raise no question at the international level as to the need for exchange adjustment. If, however, the reserves of a country reached the upper or lower limit, this would be regarded as an indication, or as prima facie evidence, that a parity change was needed.

It may be noted that the Fund and some of its members have gained a certain amount of experience in the use of objective indicators as a guide for making exchange rate decisions. Developing countries whose rapid price inflation has necessitated relatively frequent devaluations have in some cases used net foreign asset objectives as well as other criteria (in particular, relative price trends) for this purpose, although these objectives and criteria have not worked satisfactorily in some cases. The extent to which the experience of these countries would be applicable to countries in different circumstances would require careful study.

Objective indicators based on exchange rate or reserve movements may reflect short-term features of the balance of payments rather than the development of an underlying situation or its causes. The evidence they provide would, therefore, not necessarily be an appropriate basis for par value adjustment if it remains an important objective that such adjustment be based on longer-term developments in and prospects for the balance of payments. This aspect may affect their acceptability even as presumptive guides for such adjustment. These comments do not to the same degree apply to indicators based on appropriate relative prices and, indeed, relative price movements play an important—but never an exclusive—role in the assessments discussed in the preceding section.

A further problem with reliance on indicators is that they may be susceptible to influence or control by members. There would also be a serious risk that public knowledge of actual or prospective movements in the specific indicators that had been chosen would provoke adverse market reactions.

The main judgment to be made with respect to the use of objective indicators is whether the added incentive that they give to prompt par value changes offsets the disadvantage of linking such changes, even presumptively, to indicators that, by their nature, can at best give only a broad indication, rather than a balanced assessment, of a country’s situation. Some would answer this question in the negative and would, moreover, object to the sacrifice of national and collective discretion that such indicators would be likely to entail; others would stress that the experience with balanced assessment so far has not been impressive and that, consequently, the search for effective and acceptable objective indicators should be pursued with vigor and imagination.

3. Responsibilities for making par value changes

Even where a judgment is made as to the extent to which parities (i.e., relative par values among currencies) are out of line, this does not in all cases immediately imply a judgment about which individual currency or currencies should undergo a par value adjustment. In situations where, as was the case in 1971, a substantial number of par values were out of line with each other, it is necessary to take a view about an appropriate division of responsibility for par value changes among the countries that are in deficit and those that are in surplus. Such a view would need to be based on some external criterion. One approach would be to require that all members whose relative par values were out of line would contribute to the realignment in such a way as to ensure an approximate balance between revaluations and devaluations. This approach would have the incidental effect that the value of the SDR in terms of currencies would be kept approximately constant. Under some alternative approach, in which there would be greater emphasis on devaluations than on revaluations, the value of the SDR would tend to appreciate in terms of currencies.6

Problems of general misalignment of parities such as the one that had built up by 1971 have, however, been exceptional. Most commonly, balance of payments surpluses or deficits have been sufficiently focused on particular members to indicate misalignment of their exchange rates vis-à-vis the generality of other currencies. In such cases it is clearly desirable that the member that is out of line should make the necessary par value change. Responsibility for making a par value change in such circumstances arises irrespective of whether the external imbalance takes the form of a surplus or a deficit.

4. The role of the Fund in respect of the initiation of par value changes

In view of the growing awareness of the importance of relative exchange rates, and the increasing interest each country has in the par values of other countries as well as in its own, the question arises whether the Fund’s legal powers with respect to changes in par values should remain confined to responding to proposals for changes in par values made by members.

Differing views are held on this question.

Some consider that surveillance of exchange rates would work more satisfactorily if it encompassed certain initiatives by the Fund with respect to par value changes that it considered to be required. To some extent there has always been informal activity in this direction, and it may be possible and desirable to extend this practice. A question for consideration is whether the Articles should be amended to provide more explicit authority for bringing the influence of the international community, exercised through the Fund, to bear on the initiation of changes in par values: an aspect that would require particular attention in that event would be whether such influence should be exercised through the Managing Director, through the Executive Board or, perhaps, in certain circumstances through an appropriately constituted Committee of the Board of Governors. The Fund’s activity in this direction could take the form of informal consultations with any member whose par value, in the light of the techniques for judging the parity structure referred to above, was becoming or had become out of line. Experience with such consultations would indicate whether, as a further step, it would be desirable for the Fund to make formal recommendations in respect of a member’s par value.

Serious misgivings have, however, been expressed by others as to any formal extension of the Fund’s powers of surveillance with respect to par values and par value changes, though these misgivings would not apply to informal consultations between the Fund and members. Some consider, moreover, that the influence and initiative of the Fund should continue to be exercised principally through analysis and persuasion. In their view, the initiative of the Fund has to be cautiously exercised and should normally be addressed to the need for balance of payments adjustment, as far as possible leaving the choice of instruments, including proposals for par value changes, to members themselves. It has also been noted that the considerations relating to confidentiality—as mentioned above—arise with similar force in the context of the exercise of Fund initiative.

If the formal responsibilities of the Fund with respect to par values and par value changes were extended, an important issue would arise as to whether and to what extent the Fund should be enabled in some way to enforce the rules and promote the objectives of the future system, and the means it should employ to this end. Questions as to the circumstances in which and the means through which such collective pressure might be exercised are beset with many difficulties. It would appear that the type of action envisaged in Article VII, Section 3 of the Fund Agreement, which authorizes other members to discriminate in exchange operations against a member whose currency has been declared scarce, does not in its present form provide a practicable means of applying pressure on surplus currencies. Another means of pressure already available to the Fund would be the preparation of a report—which could be confidential or published—on a member’s policies as related to adjustment under Article VII, Section 1, or Article XII, Section 8; these provisions have so far never been used.

In a future system, enforcement could conceivably take the form of pressure applied to members that failed to implement changes in par values which the Fund believed to be necessary, and perhaps also to members that in other respects failed to fulfill their responsibilities under the adjustment process. It is clear from the experience of the last decade that the capacity to apply such pressure, if and when appropriate, would need to be available in respect of both surplus and deficit, and large and smaller, countries. It would depend on circumstances, especially on the strength of the incentives to adjustment inherent in the situations themselves, whether in future such pressure would need to be applied more on surplus or on deficit countries.

One question is how definite the rules inscribed in the Articles would have to be, and how precisely the circumstances of their application would have to be delineated, in order to warrant the application of measures or procedures that might be regarded as penalizing a member that had defaulted in performing its obligations or whose balance of payments policies conflicted with criteria that had been internationally agreed in the framework of the reformed system. It would also be for consideration whether any such measures or procedures should involve progressively firmer pressure on a member that failed to respond. Other questions that arise are, first, whether members would be willing to take decisions in the Fund, under new provisions, to bring pressure to bear on a particular member; second, whether members would be willing to take the action enjoined, or permitted by, any decisions that were adopted; and, third, whether it would be desirable or possible to design procedures under which rules, once agreed, could to a large extent apply without the exercise of further discretion.

Notwithstanding the difficulties mentioned earlier, the importance to the system of having at its disposal some form of collective leverage is such that the subject deserves study and consideration.

5. Width of margins

In the framework of the realignment of December 18, 1971 the Executive Directors adopted a decision establishing a temporary regime under which a member may permit the exchange rates for its currency to move within margins of 2¼ per cent on either side of the established relationship of its currency to its intervention currency. The decision also provided that a member that takes appropriate measures to maintain the exchange rates for its currency within the margins of 2¼ per cent in terms of its intervention currency may also permit the resulting exchange rates for its currency in terms of other currencies to fluctuate within margins of not more than 4½ per cent of the relationship based on their relative par values or central rates, with further margins of 1 per cent each way in certain circumstances.

Since the adoption of this temporary regime wider margins have been extensively used. More recently, with the intention of promoting monetary and economic integration, members of the European Economic Community have agreed among themselves to reduce the margins for rate movements among their currencies from 4½ per cent to 2¼ per cent,7 though this is still in excess of the margins of approximately 1½ per cent that applied among these currencies before May 1971.

The temporary regime of wider margins was introduced to facilitate the cooperation of members in observing the purposes set out in Article I of the Fund Agreement during the period of adjustment after the realignment of exchange rates and pending review of the provisions of the system. It is now for consideration whether wider margins should become a permanent feature of the par value system by means of an amendment of the Articles of the Fund, and how wide such margins should be; whether alternatively, the Fund might be authorized to permit wider margins in certain circumstances either for currencies generally or in particular cases; and whether any such authority should be exercised on the basis of a decision taken by some specified majority and should include the right to withdraw any permissions given under it.

The advantages that could be expected from an appropriate widening of margins fall in three main categories. First, by increasing the scope for exchange rate movements, they would tend—where there is confidence in the maintenance of the parity—to promote equilibrating short-term capital flows and discourage disequilibrating flows, thus absorbing pressures that would otherwise be reflected in larger changes in official reserve levels or further recourse to capital controls. Second, by creating greater uncertainty as to the development of exchange rates in the near future, they would also reduce the sensitivity of short-term capital movements to divergences of conditions in national money markets, thus allowing somewhat greater freedom of action for national monetary policies. Third, the increased scope for exchange rate movements in response to market pressures would reduce the prospective profitability of speculation on possible changes in parities if these were normally small and could also be of some help in smoothing the transition from one par value to another.

Widening of margins could, however, involve a number of disadvantages, particularly for the developing countries. In the first place, the increased scope for movement in market rates would have certain disturbing effects on trade and current payments. With only a moderate widening of margins, these effects are in general unlikely to be substantial, but they may bear more heavily on the interests of particular groups of members such as developing countries. Moreover, there may at times be circumstances in which the inhibiting effect of wider margins on short-term capital flows is itself unwelcome—for example, if it prevents a reflux of funds following a previous speculative outflow. In the face of these disadvantages, some Executive Directors would prefer that the developed countries put greater emphasis on means other than wider margins for the purpose of limiting disruptive capital flows.

In their 1970 Report, the Executive Directors indicated that they had not then reached a common view as to whether and how far margins could advantageously be widened beyond the 1 per cent from parity vis-à-vis the intervention currency and 2 per cent vis-à-vis other currencies permitted under the decision of 1959. The basic arguments for and against a regime of wider margins are essentially the same now as they were in 1970. Experience since that time is susceptible to substantially different interpretations. For these and other reasons, Executive Directors have not reached a common view as to an appropriate regime of margins which might succeed the present temporary arrangements. In further work on this issue, attention will in particular need to be given to the precise width of margins that might be adopted, to the generality of their application, and to the possibility of variation, in appropriate circumstances, in the permitted width of margins.

6. The role of the Fund in relation to temporary deviations from par values

In their 1970 Report, the Executive Directors reviewed briefly the issues involved in the question whether the par value system might be equipped to meet circumstances in which exceptional pressures induce an individual country to suspend observance of its exchange rate obligations and to move temporarily to a floating rate. This question arises in respect of any of the kinds of floating rates that have occurred in the experience of the Fund in the past and that were introduced by members with the intention of re-establishing, sooner or later, an effective par value. Under the Articles, the Fund is not authorized to approve a unitary fluctuating rate. It is necessary to consider whether it would be desirable to amend the Articles of Agreement so that the Fund would have the authority to approve such departures from the normal par value regime; and, if so, under what circumstances and subject to what conditions.

Between May and December 1971, most major currencies floated for certain periods. These instances reflected both the fundamental disequilibria that prevailed and the pressure of overwhelming capital movements. If a future system were such as to incorporate both wider margins and prompter and smaller adjustments in par values, occasions involving pressure for deviations from par value obligations would be substantially reduced.

Nevertheless, situations may arise—largely but not exclusively as a result of capital flows—in which individual members find that they have little practical alternative but to deviate from their exchange rate obligations. If such situations occur, it might be better that they fall within rather than outside the ambit of the authority of the Fund; specifically, the granting of legal authority to the Fund to approve temporary deviations under an express power to approve, subject to conditions, might enable the Fund to deal more effectively with them. This authority could help to ensure that such departures were flanked by special safeguards as temporary substitutes for the normal safeguards for the international community which are an integral feature of the par value system when it is being observed. Such special safeguards, which an all cases would include an obligation of the member concerned to remain in close consultation with the Fund, would also provide assurances against the imposition or intensification of restrictions on trade and current payments. Such consultation might relate, inter alia, to the member’s intervention policies in the exchange market.

The practical problems involved in effective surveillance of floating rates and the application of appropriate safeguards must not be underestimated. Unless one could be confident that the Fund would be successful in this respect, and would deter recourse to floating in situations in which floating was considered inappropriate, the granting of legal authority for the Fund to approve departures from the basic par value regime might be questioned; in particular, it could make such departures appear a less serious breach than if unitary floating rates remained clearly outside the legal regime of the Fund.

Finally, exercise of jurisdiction in the Fund under the present Articles has involved an element of imbalance; whereas the Fund is unable to approve a unitary floating rate, it is empowered to approve multiple rates of exchange, and has approved their introduction even when they included two or more floating rates. This imbalance is unfortunate since in some cases a unitary floating rate and avoidance of the controls necessarily associated with a multiple rate system may be the preferred course.

CHAPTER III Convertibility and the Settlement of Imbalances

1. Convertibility in the past

Convertible currencies in the central sense in which this term is used in the Articles are the currencies of members that have undertaken to perform the obligations of Sections 2, 3, and 4 of Article VIII. These obligations involve the avoidance in general of restrictions on payments and transfers for current transactions, multiple currency practices, and discriminatory currency arrangements, and, in order to facilitate payments and transfers for current transactions, a readiness of the members concerned to convert balances of their own currencies held by other members into gold or the currencies of the members requesting conversion. The provisions relating to the conversion of official balances have in practice rarely been used. Their purpose—that of ensuring the multilateral settlement of payments imbalances at rates in the vicinity of par—has been achieved through the transferability of currency holdings and their conversion, particularly for private holders, through the foreign exchange markets rather than by the monetary authorities of the issuing countries: a holder of a convertible currency who desires conversion into some other currency normally sells his holdings in the exchange markets and does not present them for conversion to the issuing central bank. That bank stands ready to sell foreign exchange in the market to prevent the rate from falling below its lower margin, thus ensuring the convertibility of its currency at a rate not far from parity.

Market convertibility has always been a feature of the U.S. dollar. Any holder (official or private) of U.S. dollars can sell them to acquire other currencies. Confidence that conversion can be carried out in this way at a rate close to parity has been provided, however, by the action of central banks of countries other than the United States, who have intervened in the market, as necessary, to maintain the rates for their respective currencies within the prescribed limits. For its part, the United States declared its readiness under Article IV to convert into gold the dollar balances of official holders. Convertibility into gold or other reserve assets was suspended on August 15, 1971; market convertibility of the dollar, however, continues and, since December 18, 1971, again takes place within determined margins from parities (based on par values or, in certain cases, on central rates) as a result of the action of other countries.

Another special feature of the U.S. position should be noted. Countries that are not reserve centers cannot count on having any part of their deficits financed by accumulations of their currency in the hands of official holders. Even a secondary reserve center cannot rely on such financing since changes in official holdings of its currency tend to vary with the overall payments balances of members of its currency area and both they and it may be in deficit at the same time. But since the end of the war the United States financed most of its deficits by means of U.S. dollars which were to a large extent held without conversion by other countries that ran overall surpluses. This arrangement worked on the whole satisfactorily for both sides for a considerable period during which the amounts involved were relatively small and their mode of financing brought about a desirable redistribution of world reserves and a moderate increase in these reserves in the form of U.S. dollars. In the course of time, however, doubts arose as to the desirability of a system that relied to a large extent on this mechanism as a source of additional reserves and that tended to insulate the United States from the need to take measures to correct its payments position, and thus also facilitated the perpetuation of the surpluses of certain other countries. The decision to establish the facility for special drawing rights reflected a desire to introduce a greater degree of collective responsibility for the future provision of reserves and to promote a better working of the adjustment process. Nevertheless, decisive action to terminate existing practices with respect to reserve currencies was not taken. In the end this led to imbalances of a size that could not be sustained by either side, and to reserve creation in the form of dollars on an unprecedented scale, thus causing the suspension of convertibility of the U.S. dollar on August 15, 1971.

As has been indicated, the balance of payments deficits and surpluses of reserve centers other than the United States are substantially settled in reserve assets and foreign official holdings of their currencies cannot be counted upon to vary in such a way as to finance their deficits. This means that the question of suitable arrangements with respect to convertibility and the settlement of payments imbalances in future is one that arises principally with respect to payments imbalances between the United States and other countries. However, variations in official holdings of subsidiary reserve currencies may be substantial, and, like variations in holdings of U.S. dollars, they may involve unplanned changes in the stock of world reserves. Consequently, the question of appropriate convertibility and settlement arrangements arises also with respect to the secondary reserve centers.

2. Settlement arrangements in a reformed system—general considerations

The convertibility arrangements that prevailed with respect to the U.S. dollar prior to August 15, 1971 became increasingly unsatisfactory from the point of view both of the adjustment process and of their effects on the stock of reserves. It follows from this that, in any reformed international monetary system, the convertibility arrangements of reserve centers, or the operation of those arrangements, should differ in some major respects from the position that prevailed with respect to the U.S. dollar prior to August 15, 1971.

One view is that any new arrangements must also differ from those now prevailing which have inherent inflationary tendencies. On this view, the absence of provision for settlement of U.S. dollars acquired by other countries unduly shields the United States from the working of the adjustment process and would not be satisfactory in the framework of a reformed system. According to this view, any new system should be one in which the surpluses and deficits of the reserve centers, including the United States, like those of all other countries in the past, are settled in reserve assets or by means of credit operations, with fluctuations in the global amount of reserve currency holdings serving only as means of interim financing pending periodic settlements (apart from changes in reserve currency holdings to meet any needed increases in working balances). Although it would be essential that U.S. surpluses and deficits be reflected in the U.S. official reserves, satisfactory future operation of the adjustment process would not require that the United States offer to official holders of dollars facilities for the conversion of holdings accumulated in the past into gold or, for that matter, into SDRs or other reserves.

Another view is that, whatever the weaknesses of the present arrangements, there is a question whether an asset settlement system universally applied would function consistently with the generally accepted broad objectives of national and international well-being. Those holding this view do not necessarily preclude the introduction of an asset settlement system on a global basis, but point out that there is no precedent for a regulated asset settlement system. They emphasize the substantial improvement in the adjustment mechanism that would be required for the system to function to the relatively tight specifications of asset settlement, and are concerned at the possibility that such a system, if rigidly adhered to, might contain a deflationary bias. They question how far governments would be prepared to undertake the necessary obligations of asset settlement and would be able to deal with their difficulties in times of strain without departing from these obligations or from the obligations of the exchange rate regime, or from both. In particular they question whether and to what extent governments would be prepared to accept the additional flexibility in either or both of the two directions which, according to this view, would be necessary for effective operation of an asset settlement system, viz., much greater exchange rate flexibility or very much larger reserves or similar unconditional financing facilities. They also note that the willingness and ability of the United States to extend swap credits to other countries and to give them access to its capital market has been an important element of flexibility in recent years. Finally, they observe that, in addition to these policy questions, the technical problems associated with asset settlement schemes are such that the search for possible alternative approaches of a different nature should be continued.

3. Asset settlement—considerations relating to reserve centers

While the principle of asset settlement would preclude the financing of U.S. deficits by means of reserve liabilities, it would not preclude financing of deficits by negotiated credits, e.g., drawings on the Fund in the credit tranches. Correspondingly, the United States would receive reserve assets when it was in balance of payments surplus (except insofar as the surplus was applied to redemption of dollar balances or repayment of any loans); this would enable that country, as a general rule, to accumulate reserves when in surplus rather than merely experience a decline in its liabilities. The same principle of asset settlement of the imbalances could apply, with suitable variations, to other reserve centers as well.

Whatever one’s views as to the desirability of a system of asset settlement, it is obvious that certain conditions would, in any event, have to be fulfilled in the case of the United States to make transition to asset settlement possible. These conditions relate to:

(1) the balance of payments and reserve strength of the United States in the period that lies ahead;

(2) the adequacy and reliability of the adjustment mechanism;

(3) suitable arrangements regarding temporary imbalances, in particular those arising from short-term capital flows.

As regards (1), the U.S. balance of payments is likely to have been fundamentally strengthened as a result of the realignment and other measures. The assumption of the obligations of asset settlement by the United States would, however, require in addition an adequate improvement in the U.S. reserve position. Such improvement might come from asset settlement of future U.S. balance of payments surpluses and from the U.S. share in SDR allocations. If mutually satisfactory loan arrangements could be devised, long-term lending by other countries of their currencies or, possibly, of SDRs would be a further possible means of improving the U.S. reserve position.

In respect of (2), it would be necessary for the United States not only to overcome its present deficit but also to be in no less secure a position than other countries to defend its balance of payments and to earn needed reserves. This would require a more effective adjustment mechanism involving all members and include a readiness on the part of other members to accept U.S. surpluses as well as deficits and to accept changes in the effective exchange rate of the U.S. dollar in appropriate circumstances. The United States, in its turn, would have to be willing to propose timely changes in its exchange rate when these became necessary.

In relation to (3), with an efficient mechanism for exchange rate adjustment under which chronic payments imbalances would be avoided and with a proper functioning of the SDR mechanism, the reserves and credit facilities available to members should normally be sufficient to ensure settlement when they are in deficit. But from time to time capital movements might develop on a scale which members would not be able to finance from their reserves and normal credit facilities. Some possible methods of limiting and of financing such movements, and the problems involved, are discussed in Chapter V. As one example, if the U.S. dollar were to remain the principal intervention currency, movements between non-U.S. private and official holders of dollars might at times pose considerable problems for the international management of an asset settlement arrangement.

The many uncertainties as to the timing and magnitudes involved in the three factors mentioned make it difficult to assess the prospects for a satisfactory functioning of a full asset settlement system.

In this situation, one possibility would be to move toward full asset settlement in phases. This would offer an earlier contribution to the objectives discussed above than if the transition from the present situation to full asset settlement were to be delayed until it could be made in one jump—but the contribution would of course only be partial. Earlier phases might involve proportional asset settlement of any U.S. deficits, or settlement in dollars, or in assets, up to a specified absolute amount. Another possible approach toward full asset settlement could be found if the United States took on an increasing share in any needed support of the U.S. dollar in the exchange market. Partial asset settlement of U.S. deficits would not preclude full asset settlement of U.S. surpluses if this were desired as a means of accelerating progress toward full asset settlement of future U.S. deficits.

4. Asset settlement—considerations relating to other countries

Any asset settlement arrangement that was adopted would have to be geared not only to the achievement of the objectives discussed earlier, but also, as far as possible, to the direct interests of members other than the reserve centers. In particular, in a situation in which countries continue to hold substantial amounts of currencies in their reserves, it would be necessary to consider how far account could be taken of differences of preference in asset composition. These differences to some extent reflect institutional factors but they also reflect the concern that some countries would feel at any loss of income entailed in holding reserves in the form of SDRs instead of reserve currencies, the differing degrees of confidence felt in different assets, and the desire to maintain flexibility and spread risks.

Three possible technical approaches to asset settlement are described below, together with some appraisal in terms of their ability to (a) achieve full asset settlement and (b) leave members as far as possible free to determine the composition of their reserves. These approaches involve differences of method and extent in substituting SDRs for reserve currencies as discussed in Section 1 of Chapter IV.

One approach would be for the reserve center to undertake to convert an amount equivalent to any net increase in balances of its currency outstanding in official hands by transferring a corresponding amount of reserve assets (gold, SDRs, reserve positions in the Fund, or currencies of other members) to official holders in exchange for its currency. The countries which would have to accept the reserve assets, and the amounts they would have to accept, would be determined under the administration of the Fund by procedures possibly analogous to present methods of SDR designation. The objective that reserve centers should earn reserve assets when in surplus would be realized in part by channeling to these centers any SDRs that other countries wanted to use, e.g., by the SDR designation mechanism. But over and above this, further arrangements would be needed: these could readily be provided by measures analogous to those in the substitution facility discussed in Chapter IV. Such a facility could be empowered to create and sell SDRs to reserve centers in exchange for their own currencies in amounts equivalent to any net diminution in aggregate holdings of these currencies in the official reserves of other members. This approach would ensure full asset settlement of reserve center surpluses and deficits (except insofar as there were international agreement that there was a need for some increase in balances of a particular reserve currency), but would accommodate asset preferences of individual members only to the extent that this could be done through the designation mechanism (or a similar administrative procedure).

Under a second approach, each country would agree not to let its holdings of any reserve currency diverge from a previously specified level. Countries would be obliged to present to the reserve centers for conversion into reserve assets any foreign exchange holdings which accrued to them above these levels and, to the extent that they fell below these levels, to reconstitute their balances by selling primary reserve assets to the reserve centers. A variant on this approach would be to specify only an upper limit to the reserve currency holdings of each country, and to rely on a substitution facility similar to that described above to ensure asset financing of reserve center surpluses. Different asset preferences could be accommodated insofar as members were initially enabled to specify their own limits of reserve currency holdings. Moreover, flexibility could be introduced by admitting a degree of variation from specified holdings. But the greater such flexibility, the less would be the assurance that asset settlement would correspond precisely with payments disequilibria.

Under a third approach, which would rely to a greater extent on voluntary action, a substitution facility in the Fund would allow members to convert reserve currency balances into SDRs at any time or, perhaps, at specified times. Settlement of reserve center imbalances would then be achieved through the Fund: a reserve center, when in deficit, would buy from the Fund in exchange for reserves an amount of its currency equivalent to the increase in its liabilities outstanding to official holders, and, when in surplus, would be able to sell its currency to the Fund in the amount of any decline in its liabilities outstanding to official holders, along the lines of the arrangements earlier described. This approach would leave members with a high degree of freedom in the determination of their reserve composition and would achieve full asset settlement as long as the cumulative total of currencies sold to the Fund for substitution into SDRs permitted the Fund to sell the required amount of its currency to a reserve center in deficit. Only if currency holdings of the Fund under the facility became at any time inadequate to match a reserve center deficit would there be an increase in global reserves in currency form.

Under any of these approaches, there would have to be rules that would determine which of their various reserve assets (gold, SDRs, reserve positions in the Fund and, where appropriate, holdings of foreign currencies) would be used by countries in settlement of deficits. These rules could range from a complete freedom for the deficit country to choose the assets to be used, to strict rules for their determination.

A wide range of technical issues—in addition to those discussed in this and the previous section—would arise under any of these schemes. Considerable further exploration of technical aspects would thus be required before a choice could be made among these and any other possible approaches to asset settlement.

5. Alternative exchange market intervention systems

Since the gradual restoration of free exchange markets in the course of the 1950s, and, in particular, since the dismantling in 1958 of the intervention and settlement arrangements of the European Payments Union, the relative exchange rates of most countries, including all the main industrial countries, have been kept within agreed margins of variation almost exclusively through their intervention on exchange markets in U.S. dollars. The United States, for its part, has maintained a basically passive role on exchange markets.

This asymmetrical arrangement has grown up very largely because of the convenience to monetary authorities of being able to control the exchange value of their respective currencies through intervening in a single currency, which has an extremely broad and stable market, and which could conveniently serve as a reserve asset.

Nevertheless, the arrangement is not without its drawbacks from the standpoint of the international monetary system. Its main disadvantages are two:

(1) Under this arrangement the United States lacks, to a considerable extent, the degree of control over its exchange rate which is available to other countries and which the United States needs no less than they. This applies both to the size and utilization of the permitted margin of exchange rate variation in relation to parity as a means of dealing with destabilizing capital flows and to the possibility of small par value changes; it also applies to the possibility, in a situation of extreme pressure, of effective recourse to a floating rate. It is noteworthy that not only is the maximum margin of the U.S. dollar vis-à-vis each of the currencies that are pegged to it limited to half that of these currencies vis-à-vis one another, but also that the decision to what extent this margin will be used lies entirely with other countries.

(2) Intervention arrangements involving reserve currencies would tend to induce the abandonment of asset settlement when reserve centers were in difficulties whereas, in principle, at least, alternative intervention arrangements would automatically preclude any lapse into liability financing.

These factors have led some to consider proposals for the introduction of more symmetrical intervention arrangements whereby certain countries, including the United States, whose currencies were dealt with on a substantial scale in world markets, would undertake to maintain prescribed margins among their currencies by standing ready to buy or sell any of the other participating currencies at the margin.8 Settlements could be made frequently in primary reserve assets or in currencies other than those of the bilateral debtor or creditor. If any country, even the United States, failed to continue settlement under the terms of the arrangement established, a decision on the part of other countries not to deal in the currency of that country on the market would not itself affect their exchange rates with one another. The fact that cross rates could be maintained among currencies in respect of which the settlement arrangement continued to be observed would thus facilitate the adoption of such a decision.

Multicurrency intervention systems raise a number of technical and policy problems relating to such matters as consistent rules for intervention within the margins, settlement arrangements as between participating and nonparticipating countries, valuation of assets used in settlement, etc. Recent limited experience with such arrangements among present and prospective members of the European Economic Community has indicated some of the problems which can be encountered, but has not been sufficient as yet to provide a clear guide to solutions. Only after all the issues arising have been more fully explored will it be possible to form a judgment whether, to what extent and in what way, the practice of multicurrency intervention might be incorporated into the international monetary system.

CHAPTER IV The Roles of Various Assets in Reserves

The present chapter is concerned with the respective roles that would be played by individual reserve assets—notably reserve currencies, gold, and SDRs—in a reformed international monetary system. As a preliminary, the matter of global reserve creation in such a system needs to be set in perspective. If the conditions for assumption of asset settlement obligations by reserve centers were met and, in particular, if the working of the adjustment mechanism improved sufficiently to ensure consistent observance of any asset settlement arrangement that might be adopted, situations of undue reserve ease as a result of excessive accumulation of reserve currencies would thereafter be avoided. The extent to which further increases in reserve currency holdings would take place would then depend on the need for increased working balances and on the particular asset settlement arrangement chosen. In general, the tighter the obligations of any asset settlement system, the greater would be the degree of dependence of the system on collective methods of reserve creation. It would thus be a major function of a reformed system to determine and provide for needed increases in reserves through the SDR mechanism.

1. The role of reserve currencies

If, as a result of the adoption of an asset settlement system, the growth of global reserves were in the form of SDRs rather than reserve currencies, the relative role of currencies would as a result decline over time. Moreover, the absolute amount of reserve currency holdings would fall insofar as such holdings were replaced by SDRs (under arrangements to be discussed below) or used to finance the combined deficits of other countries vis-à-vis reserve centers. The phasing and extent of the diminution in the role of reserve currencies would also be affected by any special arrangements instituted during a period of transition. For example, if countries were to make reserve loans of their currencies to the United States, this might involve a temporary expansion of the use of these currencies in reserves.

The special arrangements with respect to reserve currencies could take various forms but would as a minimum consist of a facility that would enable a reserve center to acquire assets when in surplus. To this end this facility would have to be empowered to create and sell reserve assets to a reserve center in exchange for its own currency for that part of its surplus that had led to a diminution of holdings of its currency in the reserves of other members. The necessary qualities of the reserve asset that would be needed for this purpose would be effectively the same as those of the SDR and it would thus appear natural to use SDRs. However, one possibility for a transitional period—before any SDR-based substitution facility could be introduced—would be to create a new reserve asset under a facility that would be based on informal arrangements among members that chose to participate.

The substitution of newly created SDRs for reserve currency holdings—whether on a limited or on a wider scale—would require the establishment of a supplementary facility in the Fund. Without prejudice to the question of the way in which such an account would be established, the term “substitution account” will be used in what follows to indicate the sets of transactions in which the Fund might engage under such a facility.

In addition to the minimum need for a substitution facility to enable reserve centers to earn assets when in surplus, countries collectively might desire to reduce their balances of reserve currencies faster than would be required to finance any combined deficits that they might have vis-à-vis reserve centers. This would require greater recourse to a substitution facility. A desire for such reserve currency substitution could result, for example, from the spread of multicurrency intervention arrangements; these might involve a reduction in the use of existing reserve currencies in settlements between third countries, and, if reserve centers were to participate, would reduce or eliminate such use in settlements with the reserve centers themselves. This would reduce the demand for reserve currencies, and increase the demand for primary assets, in the reserves of other members. The extention of substitution and the concomitant reduction of reserve currencies in countries’ reserves would strengthen any asset settlement arrangement and ensure that reserves would increasingly be insulated from the effects of any changes in the par values of reserve centers.

The substitution of SDRs for official holdings of reserve currencies could take place in a number of ways. For example, reserve centers could be entitled to sell to the substitution account, for SDRs, an amount of their currency equivalent to any decline in the volume of their official liabilities, and on this basis could undertake to convert into primary reserve assets any balances of their currencies presented to them by official holders. It would also be possible to entitle holding countries to sell reserve currencies to the substitution account in exchange for SDRs newly created for this purpose.

The operations of this account could also be adapted to meet the opposite desire on the part of holding countries, namely to acquire reserve currencies from the substitution account, within the limits of its holdings, in exchange for SDRs. It might be convenient from a number of points of view for the account to be able to operate in both directions, i.e., buying as well as selling SDRs in exchange for currencies. It could, to a large extent, take the place of the designation mechanism as a means by which participants could acquire currencies against SDRs and it might also provide a means for effecting agreed increases in currency holdings. If it were, however, agreed that reserves in the form of SDRs should, over time, increasingly be substituted for reserves held in the form of currencies, two-way operation of the account would require safeguards—such as the requirement of need—in respect of the purchase of currencies from the account to ensure that possibilities for undesirable switching between SDRs and currencies were avoided.

A major portion of existing official holdings of reserve currencies could be replaced if the principal trading countries agreed on a concerted program of substitution of a substantial part of their currency holdings into SDRs. Reserve substitution beyond this stage might be more difficult to achieve, particularly in the case of smaller countries that had traditionally held their reserves in the form of currency and maintained their parities in terms of their reserve currency. Replacement of reserve currencies by SDRs would tend to be larger if the attractiveness of the SDR as a reserve asset were enhanced along the lines suggested in Section 3 below.

The creation in the Fund of a substitution facility along the lines described above would require provision for the maintenance of value of currencies acquired by the account against newly issued SDRs and would involve, inter alia, the specification of the servicing obligations of reserve centers. These obligations would include the payment of interest at a rate appropriately related both to the market interest rate on the type of currency obligations that had been replaced and the (possibly modified) rate on SDRs (see Section 3 below). One view is that suitable provision for amortization would be a further essential element in the arrangement, partly to avoid what might otherwise be regarded as concessionary treatment of reserve centers. However, from a purely economic standpoint, the servicing of a perpetual obligation—with no separate provision for amortization payments—need not be concessional unless the interest rate is concessional. Moreover, the obligation of making amortization payments might be such as to require reserve centers to earn external surpluses larger than other countries would want to accommodate. Any amortization arrangements might, in addition, be complicated by two-way operation of the substitution account, which could involve frequent purchases as well as sales of reserve currencies against SDRs.

The institution of any facility for the replacement of currency reserves along the lines described above would involve moving toward a wider concept of the SDR, which some countries might not favor as a matter of principle, and which might not be without its difficulties. One difficulty is that such a facility might be represented by some as unduly benefiting the reserve centers, whereas the contrary view might also be taken that it would unduly benefit other members. Other difficulties—which might differ substantially depending on the method of asset settlement selected—relate largely to the strain that a sizable expansion of SDRs might put on the arrangements for transfer and acceptance of SDRs. In this connection, it has to be borne in mind that any replacement of currency holdings by SDRs created for the purpose could require the assumption of additional acceptance obligations. The distribution of these additional obligations would cause problems if, as a result of the increased holding of SDRs in members’ reserves, the SDR became a less-favored asset.

A related, though different, question for consideration is the possibility of replacing some part of the reserves that are regarded by individual members as excessive by illiquid long-term claims on reserve centers, in particular the United States. It is, however, questionable whether there are many countries in a position to invest a substantial part of their present reserves by funding them into illiqud form. Moreover, even if a funding operation could be negotiated, it is unlikely that it would be sufficient to ensure that the United States would earn reserves to the full amount of any surplus.

2. The role of gold

Gold performs two main functions in the present system. First, it is the unit of account or standard in terms of which par values are expressed, the value of the SDR is defined, and the value of the Fund’s assets is maintained.9 In examining the role that the SDR could play in the future system, it will be necessary to consider whether the SDR, rather than gold, should be made the standard of par values. In the Articles, the value of a special drawing right is stated to be 0.888 671 gram of fine gold, with intentionally no provision for a change in the equivalence. As a result, par values are de facto as firmly linked to SDRs as they are to gold and any change in the value of gold in terms of currencies involves automatically a corresponding change in the value of the SDR in terms of currencies. Thus, a change in the Articles which made the SDR the standard of par values—and which consequently defined the value of gold in terms of SDRs rather than that of SDRs in terms of gold—would derive importance only from its presentational effects, unless it were accompanied by some loosening in the link between the two. The desirability of such a loosening may, however, be questioned since it might add a new dimension of uncertainty to the relative values of different reserve assets.

Second, and more important, gold is a major reserve asset, constituting nearly a third of international reserves at the end of 1971. Views differ on the future role of gold in this capacity. Some consider that, while gold might well continue as a major though declining part of reserves for some time to come, this would not preclude a movement of gold from official holdings into nonmonetary use, and some would welcome such a development. Those holding these views would rule out action that would increase the share of gold in reserves through an increase in the official price of gold in terms of currencies in general.10 They consider that recent developments of the private gold price have been importantly influenced by market uncertainty regarding the official price and some believe that a commodity for which private demand is relatively large and highly speculative has a disadvantage for use as a major reserve asset.

Others wish to keep a central role for gold in the international monetary system. They base their conviction on the tradition, universal use, limited production, and physical characteristics of gold. They also maintain that gold is the only international asset held by monetary authorities that is not a liability of another monetary institution and that, in addition, it is the only instrument of reserves that is subject to complete national control.

In recent months, the private market price of gold has far exceeded its monetary price. Such a wide discrepancy makes countries hesitant to use gold in official settlements and might also give countries an incentive—in circumstances where they needed to use gold to meet a payments deficit—to sell the gold on the private market rather than at the official price. It might, moreover, make countries unwilling to use their SDRs and reserve positions in the Fund and hesitant to use their credit tranches. Some see no inherent harm for the international monetary system in any tendency for gold to become the residual element in official reserves, or even for gold to be eliminated from reserves, as long as the system functions adequately in other respects.

Some partial offset to these tendencies could be brought about if countries participating in settlement arrangements were to use the various reserve assets, including gold, in an agreed manner, thus countering the natural tendency for debtors to settle in the least preferred assets. It has also been suggested that the Fund might institute an arrangement under which members would be entitled to swap gold with the Fund against SDRs. Such an arrangement—which would require amendment of the Articles—would enable a member that used SDRs acquired in this way to reacquire the gold when it earned back SDRs later.

The tendencies mentioned earlier are, however, likely to continue unless the gap between the official and the private price of gold is reduced. According to some, any difficulties arising from what they consider to be the present undervaluation of monetary gold would be resolved by increasing the official price of gold. While an increase in reserves would result as a by-product of such action, those holding this view consider that any inflationary dangers or distributional inequities created by the adjustment in the value of gold reserves could be overcome. They, as well as some others who do not share the preceeding views, are convinced that it is unrealistic to believe that an enduring reduction in the present price discrepancy could be effected by means of sales in the private market as suggested below.

Any approach involving an increase in the official price of gold is, however, strongly opposed by all those who argue that it would undo the progress made so far toward a rational monetary system, undermine the SDR facility, and be highly inequitable among members. They do not believe that the potentially serious inflationary and distributional consequences of such an increase could effectively be overcome. Moreover, they believe that any such price increase would tend to stimulate expectations of further changes. They consider that clear action in the context of the reform which precludes an increase in the official price would eliminate such expectations and narrow the existing differential between the official and the private market price. To the extent it was considered desirable to reduce the spread between the two prices, the action required would in their view consist of measures to bring down the price in the private market, which would call for the sale of monetary gold into that market, perhaps in substantial amounts. While this could be done by the major gold-holding countries individually or collectively for their own account, it might be desirable for such sales to be organized through the Fund, acting as an agent for all members wishing to participate. Alternatively, the Fund might on its own account sell either gold from the General Account or gold acquired in exchange for newly issued SDRs through a reserve substitution facility: these two last-mentioned arrangements would require an amendment to the Articles of Agreement.

A possibility (though perhaps remote) is that member countries would come to prefer to hold SDRs rather than gold as a monetary reserve. In these circumstances, the Fund might find practical difficulties in using gold in replenishment, and its liquidity could thus be impaired. This would be remedied if a substitution facility—as discussed in Section 1 of this chapter—were also enabled to issue SDRs in exchange for gold held in the General Account.

3. The role of SDRs

If an asset settlement arrangement were introduced for all members, the SDR would—as indicated—be established as the principal means of reserve creation. In addition, asset settlement would involve the substitution of SDRs for reserve currencies at a pace determined by the timing and scale of reserve center surpluses and any substitution arrangements that might be adopted. The required characteristics of an asset that served as the main reserve instrument are likely to be different from those of an asset that constituted only a marginal supplement to other reserve assets. For these and other reasons it would be necessary to reconsider certain features of the SDR system in the context of reform, in particular those bearing on the use and acceptance of the SDR and on its relative attractiveness as a reserve asset. These features are discussed in the remainder of this chapter. In addition it might be desirable to broaden the provisions in the Articles to include a wider group of official holders of SDRs. It should be noted further that the provisions on designation and reconstitution will in any event be reviewed in accordance with the Articles before the end of this year, and that certain other provisions may deserve reconsideration in the light of experience.

a. Use and acceptance of SDRs

Any arrangements adopted with respect to the asset settlement of payments imbalances are likely to require certain mandatory transfers of SDRs to which the operating provisions of the SDR system would have to be adapted. Under the first two approaches to asset settlement (as described in Section 4 of Chapter III) mandatory transfers would involve both reserve centers and holders of reserve currencies whereas, under the third approach, such transfers would take place between reserve centers and the Fund.

Other consequences would follow from certain of the arrangements discussed above and from any major increase in the proportion of reserves held in the form of SDRs:

(1) The acceptance obligation constitutes a basic safeguard of the usability of the SDR. Under the present Articles, this obligation is subject to a limit which is linked directly to a participant’s net cumulative allocation.11 If provisions were adopted under which SDRs could be created other than by allocation to participants, the method by which the magnitude of acceptance obligations is determined would need to be modified. This problem might be dealt with in any one of three ways:

  • (i) The ratio of acceptance obligations to net cumulative allocations could be increased to accommodate nonallocated SDRs.

  • (ii) The required total of acceptance obligations could be distributed among participants on the basis of some criterion other than their net cumulative allocations (for example by reference to current quotas).

  • (iii) The limit on acceptance obligations could be abolished. In the event that the SDR had become established as the major reserve asset and its general acceptance was assured by its use as the principal medium for official settlement, acceptance limits could become unnecessary and exclusive reliance could be placed on the designation mechanism to distribute SDRs used among participants in whatever manner might be decided.

(2) If a substitution facility were introduced under which members would be able to exchange their holdings of reserve currencies into SDRs, the main reason for limiting the scope of transactions among participants without designation and without requirement of need would be removed.12 It would then be possible to allow much greater freedom for such transactions between willing participants and to reduce correspondingly reliance on the designation mechanism. But, however far the system might develop toward voluntary transfers, it would always be necessary to provide for designation in cases where a participant was unwilling or unable to rely on these methods of use of SDRs, and for such guided transactions it would be logical to continue to apply a requirement of need.

(3) A substantial rise in the volume of SDRs, together with the possible decline in the volume of reserve currencies and (as indicated in Section 2—The role of gold) of gold in members’ reserves, might make it appropriate to extend the range of transactions and operations in which the General Account of the Fund can accept and use SDRs in lieu of currency and in lieu of gold. This in turn might be the occasion for a broader review of the provisions of the Articles relating to the operations and transactions of both the General Account and the Special Drawing Account.

(4) There would also be room for a broadening of the categories of transactions and operations in SDRs—for example, to accommodate grants, loans, or pledges of SDRs.

b. The attractiveness of the SDR as a reserve asset

The acceptability of SDRs will be determined not only by their usability but also by other aspects of their attractiveness as reserve assets, in particular the interest rate they carry together with their expected future value in terms of other assets. In the less than three years since SDRs were first allocated, many participants have used them without encountering technical difficulties. Nevertheless, a broadened and freer use, along lines indicated in the preceding subsection, would no doubt enhance the usability of the SDR as a reserve asset. It is not suggested at this stage that the holding of SDRs should move beyond the official circle: private holdings of and private transactions in SDRs are, however, seen as a possible development for the more distant future.

It should be noted that if the exchange rate mechanism in future were such as to lead to an approximate balance between revaluations and devaluations, the value of SDRs would move roughly in line with that of other currencies in general. Insofar as devaluations on balance exceeded revaluations, the value of the SDR would appreciate in terms of currencies, though if this result were desired, there would be other means of achieving it. One approach would be to make uniform proportionate changes in the par values of all currencies from time to time under the provisions for such changes in the Articles. This would make it possible, for example, to keep the value of the SDR in line with that of the strongest major currency. Unless some such approach to the value of the SDR were introduced, or unless agreed provisions were instituted to limit foreign currency holdings, it seems clear that to encourage holdings of SDRs on a scale substantially above that at present might require an increase in the rate of interest on SDRs to a rate close to that on average reserve currency holdings. Some have, however, observed that this problem would be reduced if countries became less influenced by interest rate and valuation aspects in their reserve management. The question of the flexibility of the interest rate on SDRs would also assume considerable importance.

Although the similarity of creditor positions in the General Account of the Fund and SDRs should not be overemphasized, large differences between interest rates payable on them could give rise to certain difficulties; thus, the remuneration paid on net creditor positions (and perhaps also the charges payable on debtor positions) might have to rise to some extent along with that on SDRs. Under the present provisions the interest charged on net use of SDRs exactly equals the interest credited on amounts held in excess of allocations. If it were desired to levy interest charges on net use of SDRs, or on part of net use, at a lower rate than that credited on excess holdings, the resulting net creation of SDRs could either be offset by assessment on all participants (e.g., in proportion to their net cumulative allocations), or, if the amounts were significant, taken into account in the decisions on future SDR allocations. The problems and consequences for the system involved in such an approach would require careful study.

CHAPTER V The Problem of Disequilibrating Capital Movements

The alarming growth, in recent years, in the frequency and magnitude of payments imbalances has largely though not exclusively reflected a growth in the scale of temporary and reversible capital flows, which have more and more played a disequilibrating rather than an equilibrating role in balances of payments. Most of the short-term variability in capital flows has occurred with respect to short-dated banking or commercial claims, partly “leads and lags” in payments for current transactions. Some of the instability, however, has occurred in respect of medium-term loans and portfolio assets. The problems raised by this growing instability in capital flows, while very severe, should not, however, be allowed to impede a recognition of the beneficial effects of the general expansion of international investment both on the growth of international trade and on the economic well-being of developed and developing countries.

1. Causes and consequences of the growth of disequilibrating capital flows

Three main factors underlie the recent growth and volatility of capital flows. First, the growing integration of the international economy since the establishment of external convertibility for major currencies in 1958 has led to a massive growth in the volume of mobile funds, and has enhanced the propensity of capital to move from country to country and from currency to currency in response to incentives of various kinds. The development of banking in foreign currencies (Euro-currency markets) has played an important part in fostering the mobility of short-term funds. Second, the inducements to interest arbitrage have increased as differences in the cyclical phasing of economic activity and the widespread use of monetary policies in demand management have involved the emergence, and sometimes persistence, of relatively wide interest rate disparities among financial centers. Third, the emergence of basic disequilibria in the 1960s and early 1970s, and the resulting exchange rate adjustments, especially since they were sometimes unduly delayed, have awakened an increased sensitivity to the possibility of further changes. The incentive to vary currency positions for precautionary and speculative motives has thus increased, and action in the expectation of exchange rate changes has come to be regarded as a more or less normal aspect of the efficient husbandry of liquid funds by banks and other business enterprises which operate internationally. The interest-sensitive, hedging, or speculative transactions to which these factors may lead can now be so large as almost to overwhelm the ability of national authorities to finance or curb them.

However, some aspects of capital mobility involve substantial benefits even in the case of short-term funds, and in practice it is often extremely difficult to distinguish desirable from less desirable types of capital flow. The Euro-currency and related Euro-bond markets provide efficient intermediation in channeling funds to investment projects, including some in developing countries, which might otherwise be frustrated by the inadequacy of domestic sources of financing; arbitrage-induced flows sometimes exert an equilibrating influence on payments imbalances; and even disequilibrating flows have served at times to prompt decisions on new policy measures, including par value changes, in circumstances in which the need for policy changes has been inadequately appreciated or such changes have been unduly delayed.

But large capital movements have aggravated very considerably the difficulties of domestic economic management, and the pressure which they put on par values has not always been salutary. Countries whose monetary policies have been geared primarily to the achievement of their domestic economic objectives have found that arbitrage-induced flows tend to undermine these policies and call for more extensive offsetting movements in domestic credit than can be achieved in all cases. Moreover, especially where such arbitrage movements magnify rather than compensate reserve changes attributable to basic payments imbalances, underlying doubts and uncertainties on exchange rates may be inflamed. The danger is then that the cumulation of speculation on top of arbitrage-induced flows may involve such pressure on a country’s reserves as to precipitate exchange adjustment which might not be justified by reference to underlying economic conditions. Even when there is a question as to the existence of a basic disequilibrium, it is clearly undesirable that the impact of speculative flows should be such as to deprive the authorities of all discretion as to the timing and magnitude of the required adjustment.

2. Means of offsetting and accommodating short-term flows

Countries differ in their ability to offset the domestic monetary effects of short-term flows where these run counter to the objectives of the authorities. Open market operations and changes in reserve requirements are the most powerful instruments, but variations in the amounts and terms of central bank rediscounting and other lending facilities may also be effective, and quantitative credit ceilings may be used. Such offsetting is, however, constrained by technical, institutional, and political factors. The difficulties may be particularly great from a technical standpoint in the case of payments surpluses and in the case of countries where short-term capital flows are large relative to the stock of domestic assets. Moreover, the fuller development and use of monetary techniques to offset short-term flows will diminish the natural self-correcting tendency of such flows and can thus prolong them. Nevertheless, offsetting measures have generally enabled countries to stay closer to their monetary and domestic demand targets than would otherwise have been possible. It is to be expected that countries will wish to continue to develop and use such techniques—which may also prove useful in dealing with purely domestic problems of stabilization.

Whatever its merits in protecting the domestic economy, monetary offsetting will do nothing to moderate the massive reserve movements that are associated with the payments imbalances provoked by disruptive capital flows. Even countries that possess reserves sufficient to finance swings in the basic balance of payments (as well as cyclical arbitrage flows of short-term capital) seldom have reserves in the amounts that might at times be required to meet sustained speculative capital outflows. Reserves sufficient to buffer all disruptive flows might be well in excess of normal reserve needs, and might thus in the aggregate tend, in periods relatively free from speculative disturbances, to exercise directly or indirectly an unduly inflationary influence. It is therefore necessary to review the adequacy and suitability of existing short-term credit arrangements whereby countries can limit the effects on their reserves.

Insofar as disequilibrating flows have taken place between the United States and other countries they have, in the main, been financed through the accumulation or reduction of U.S. dollar holdings by other countries through exchange market intervention. Some of these holdings have, on occasion, been provided with an exchange guarantee through swaps and other arrangements. In the case of disequilibria arising among other countries the financing, other than that provided by reserve use, has taken the form of activation of short-term swap arrangements and drawings from the Fund. These methods have in the past decade helped substantially to reduce the drain on the reserves of deficit countries, but most of them are not without drawbacks. In the first place they do not have a corresponding impact on the reserve influx, or the resulting problems, of the surplus countries to which the capital is being attracted. Moreover, since the creditor claims associated with such methods of financing usually take the form of reserve assets, their use has unplanned and possibly disturbing effects on the level of world reserves. Another aspect of these arrangements (with the exception of credit tranche drawings on the Fund) is that they necessarily involve the provision of resources, at least initially, without conditions as to the policies of borrowing countries. Finally, whatever the merits of these arrangements for meeting situations of genuinely temporary and reversible short-term capital flows, there is the danger that they may be used inappropriately to stave off the needed correction of a more basic disequilibrium.

The question arises whether such facilities should be supplemented and could advantageously be brought under more deliberate and coordinated multilateral surveillance or control. The main objectives would be (a) to ensure that adequate facilities for the financing of occasional exceptional flows would be promptly available in the case of need on appropriate terms as to interest and period of repayment, (b) to minimize the risk that such temporary finance would be used to accommodate a basic deficit, and (c) to assist members in the choice of measures to abate further flows. The safeguards that would be necessary under a multilateral facility should not, however, be such as to impair unduly the willingness of deficit countries to use the facility.

3. Measures to influence capital flows

There is a wide range of possible measures available to countries to influence capital flows, and each country that uses them is likely to adopt a combination of techniques depending on its special preferences, situation, and capabilities.

a. Wider margins

The present temporary wider margins potentially provide a means of countering disruptive short-term flows: the issues involved are discussed in Section 5 of Chapter II.

b. Harmonization of interest rate policies

To the extent that countries were able to make fiscal techniques more flexible and hence more serviceable as instruments of demand management, they would be better placed to harmonize their monetary policies more closely than in the past and to reduce incentives to arbitrage arising in particular from divergent short-term interest rates. The question arises whether more deliberate and more continuous coordination of monetary and interest rate policies could usefully be undertaken with a view to limiting disequilibrating short-term arbitrage flows. It is desirable to persevere with efforts at such harmonization as far as is practicable without sacrifice of domestic objectives. Most countries may, however, find it impracticable to rely predominantly on fiscal policies for the achievement of domestic objectives and to employ monetary policy primarily for external objectives. They may therefore be unable to make more than limited use of monetary policy as a means of influencing short-term flows.

In countries where forward exchange markets exist and their facilities are widely used, changes in the forward premium or discount of a currency achieved by official intervention in the forward market may be a useful supplement (and sometimes alternative) to interest rate policies in discouraging short-term flows or in encouraging desired movements of the interest-arbitrage type. Its use to counter speculative movements, though not excluded, may in some circumstances encourage further private speculation and will be expensive in cases where the original exchange relationships cannot be maintained.

c. Administrative control of capital movements

A wide range of possible techniques is available for administrative control of capital movements. Measures regulating the net or gross foreign positions of banks, or the interest payable on, or reserve requirements applicable to, their liabilities to nonresidents in domestic currency are relatively easy to administer, though frequently unwelcome to the business and financial communities. These measures are, however, narrow in scope and may lead to bypassing of banks by direct borrowing or lending abroad by the nonbanking sector. Comprehensive capital restrictions can in some cases be more effective, and can be used in a selective way, but require a considerable apparatus of administration and supervision. Even the most comprehensive controls, however, are subject to evasion and may be relatively ineffective against the operation of “leads and lags” in the payments and receipts associated with permitted transactions.

Capital controls have been clearly effective in some countries, at least in the short term, but experience is not sufficiently uniform to warrant the conclusion that countries that do not have comprehensive controls would be well advised to institute them. Moreover, some countries have strong reservations based both on principle and on practical considerations as to the desirability of controls. Countries which do not have available an apparatus of control might however consider developing, on a contingency basis, a set of relatively simple and flexible instruments that could be activated promptly to obstruct at least the more important channels for disequilibrating short-term flows. It has been suggested that foreign currency banking, which in most countries is not subject to extensive regulation, should be subjected to some form of coordinated control. The object would be to limit the extent to which disturbing short-term flows could be financed through these markets. The potential advantages and disadvantages involved in possible methods of regulating foreign currency banking would require considerable further study.

d. Separate markets for capital transactions

A dual exchange market in which there is a separate floating exchange rate for capital transactions may provide an efficient means of buffering capital flows since it takes advantage of the price mechanism and encourages flows in one direction and discourages them in the other. While a dual market requires a framework of exchange controls, this framework is simpler than that required for the purpose of administrative or quantitative restrictions. On the other hand, a separate financial market shares, albeit in lesser degree, some of the defects of quantitative capital restrictions, particularly as regards the possibilities created for evasion. Moreover, the effectiveness of a separate market may be reduced or offset if the authorities intervene frequently to prevent a substantial divergence between the rates for capital and for current transactions, or if the emergence of a large spread induces evasion of the controls.

e. The switching of currencies held in official reserves

Although transfers of officially held reserves from one currency to another have been less volatile and on a smaller scale than transfers of privately held balances, such official transfers and, to some extent also, transfers of currency balances between the country of issue and Euro-markets, have some of the same effects as private capital flows. In particular, they expand or contract the reserves of the countries whose currencies are held, and possibly also affect global reserves. Such transfers, if carried very far, could exacerbate and extend to new reserve centers the problems associated with existing reserve currencies. They might also tend indirectly to impair the attractiveness of the SDR as a reserve asset. The possible arrangements which are discussed in Chapter IV for the substitution of SDRs for reserve currencies would diminish the potential scale of the problem, and several of the possible techniques for asset financing which are examined in Chapter III would automatically limit transfers of officially held balances. In any event, it would be desirable to persevere through international cooperation with efforts to check as far as possible the emergence of new reserve currencies, and to limit the holding of official balances in the Euro-currency markets. Another view is, however, that countries should retain the maximum possible degree of freedom in determining the composition of their reserves.

4. External financing versus measures to influence capital flows

Conflicting views are held on the relative roles of measures to finance, and measures to influence, disruptive capital flows. On the one hand, although possibilities for domestic offsetting and special facilities for financing cushion the impact of such flows, both may involve problems which some countries will wish to avoid insofar as the flows can be headed off directly. Some countries in deficit positions will seek to restrict short-term outflows to the maximum extent possible in order to avoid indebtedness and the interference with national policy that borrowing may entail. Again, some surplus countries may prefer to use controls, or to urge their use by others, rather than to have to cope with the difficulties of monetary offsetting. On the other hand, satisfactory provision for the official financing of disruptive flows, particularly if arranged on a broad multilateral basis, may itself improve confidence and diminish speculative pressure. Moreover, it may be more appropriate for temporary flows to be accommodated by temporary official credits rather than by controls, which may disturb international commercial transactions and may tend to become built into the system.

As mentioned earlier, countries differ substantially in their attitudes toward capital movements, in the difficulties that such movements could pose for the conduct of their domestic policies and in their willingness and ability to control such movements and to use other means to shield their economies from them. Accordingly, different countries are likely to respond differently to the problem of short-term flows. It will, however, be necessary in the context of reform to establish some modus vivendi on the respective roles of measures to influence such flows and to finance them. First, decisions will be needed on the future role of swap facilities, on the extent if any to which they should be subject to international surveillance, and on the possible establishment of a Fund facility that could take their place or supplement them and that would be available to all members on a nondiscriminatory basis. If there were agreement on such a facility in principle, a number of aspects would require consideration, such as the size of any such facility, the circumstances in which it could be drawn upon, the provisions with respect to interest rates and repayment, the financing arrangements, and the impact which its operations would have on the stock of reserves. Second, it would be useful to clarify attitudes on the extent to which short-term flows should be controlled and on the appropriate means of influencing them. In this connection, it should be considered whether the powers of the Fund in this area should be extended in order to achieve improved coordination.

CHAPTER VI International Monetary Reform and the Developing Countries

1. The developing countries and the international monetary system

Two aspects of the framework within which, as mentioned in Chapter I, the reform has to be viewed are of particular relevance to the developing countries: “the development of the productive resources of all members” and “the expansion and balanced growth of international trade.” The rate of growth in the developing countries is still not sufficient to narrow the gap between them and the industrial countries, and their share in world trade has declined while that of the developed countries has risen. While these unsatisfactory circumstances have been partly attributable to basic economic and other factors in these countries, they have been compounded by difficulties of access to markets and an inadequate flow of capital to developing areas. The Fund should continue to take all steps open to it, within the purview of its objectives, to seek to improve this situation. Reform of the international monetary system provides an opportunity to make further progress in this direction; a sound international monetary system should play its part in the development of the resources of its members. Measures directly addressed to the needs of the developing countries that should be considered in the context of the reform are discussed in Section 2 of this chapter.

Financial and economic developments in the second half of 1971 illustrate the proposition that all countries are affected when the international monetary system ceases to function smoothly. The events that followed the actions taken by the United States on August 15, 1971 affected the developing countries in a number of ways. Some of them were affected directly by the 10 per cent surcharge on dutiable imports into the United States. For all of them, the relative fluctuation in the exchange rates for major currencies posed new problems and uncertainties with respect to their own exchange rate and reserve management policies, the prospects for their terms of trade, and even their development plans. Moreover, the uncertainties in the world economy which followed from the monetary upheaval appear to have reinforced the effects of the slow pace of growth of several industrial economies in depressing the prices of primary products in the last few months of 1971.

Some of these difficulties have now been overcome, and the renewed expansion in the U.S. economy and in other developed countries has improved the export prospects of the developing countries in general. But the experience underlines the major interest of the developing countries in a sustained economic expansion in the industrial countries and in the effective functioning of the system. The developing countries would have much to gain from a steadier growth of the economies of the industrial countries. The particular policies employed by the industrial countries are also relevant in this respect. In recent years the developing countries have been unfavorably affected by the higher cost of credit in international financial markets, attributable to some extent to the considerable emphasis by industrial countries on monetary policy in programs of financial restraint. Furthermore, there has been a sluggishness in the overall flow of official capital and aid, related in part to preoccupations in many industrial countries with problems of inflation, budgetary pressures, and balance of payments difficulties, as well as to noneconomic factors. One result of this may have been a tendency for developing countries to resort to shorter-term debt on relatively expensive terms. These adverse developments have been accompanied by a certain slackening of progress in other directions. Negotiations for the untying of aid have fallen into the background and the plan for preferential treatment of the exports of developing countries has been accomplished only in part. Moreover, progress in the negotiation of international commodity agreements continues to be slow, and restraints on imports from developing countries continue seriously to impede their trade prospects.

The vital interest of developing countries in the functioning of the international monetary system requires the substantive participation of their representatives in the international formulation and negotiation of proposals to reform it and of any associated proposals relating to international trade. This interest has found an expression in the support on the part of the developing countries for the establishment of the Committee of the Board of Governors, which is to concern itself with reform. It is also reflected in the creation of the “Group of 24” and in the considerable attention devoted to monetary matters by the Third United Nations Conference on Trade and Development.

A number of aspects of the world’s trading and monetary system are of particular importance to the developing countries. These include freer access for their products to the markets of the developed countries on more satisfactory terms, a steady and enlarged flow of official foreign aid, and adequate access to capital markets accompanied by measures on the part of both developed and developing countries to encourage, rather than limit, the flow of private development capital in various forms and on mutually beneficial terms.

The developed countries for their part have a direct interest in the economic expansion in the developing world. The rise in incomes and purchasing power generated by such expansion provides opportunities for an increased and mutually advantageous expansion of trading and investment relationships between the two groups of countries. If, through appropriate policies, such expanding relationships could help to maintain high levels of trade and economic activity, this should also assist the adjustment process and contribute to the stability of the international monetary system itself.

With respect to international monetary arrangements in general, the developing countries have for the most part expressed a strong preference for the par value system, with narrow rather than wider margins.13 Although recognizing the disadvantages to them of delays in needed par value adjustments in developed countries, and the risk of exchange crises in a system that lacked sufficient flexibility, developing countries have laid particular stress on the difficulties that could be caused to their own economies by fluctuations in exchange rates for major trading currencies. As regards their own exchange rate policies, many of the developing countries have attached high priority to the maintenance of a par value for their currency. Others, less able to control domestic inflation, at least in the short run, have made frequent downward adjustments in the external value of their currencies to avoid as far as possible the internal and external distorting effects of rapid inflation.

The developing countries have tended to hold their reserves predominantly in foreign exchange rather than in gold, considering the interest income on the currency balances held as an important offset to the opportunity cost involved in holding reserves. A decision in the framework of any future reformed system to raise the interest rate on SDRs—as a means of encouraging increased holdings of SDRs—would have special implications for the developing countries. These countries have, on balance, been substantial net users of the SDRs allocated to them and may be in a similar position in future: thus a rise in interest rates on SDRs would increase the interest costs that these countries bear. This consideration lends further importance to any measures that could be taken in the context of reform that would be directly addressed to the needs of the developing countries.

2. The international monetary system and the supply of financial resources to developing countries

One major aspect in which the needs of the developing countries taken as a group are being inadequately met is in the provisions of development finance. This is a matter of widespread and growing concern because this constraint on their economic growth inhibits balanced growth in the world economy as a whole. The question has been posed as to whether the total supply of development resources, from national and international, official and private sources could be supplemented by the provision of financial resources for development through the international monetary system, and whether such action would be compatible with the effective functioning of the system.

That question has been posed for some years now as that of a “link” between SDR creation and development finance; but in the context of the reform the question can be approached more broadly, to encompass a number of possible approaches, each of which would be intended to provide additional financial resources to the developing countries, either directly or through development agencies. A number of possibilities are listed below:

(1) The allocation of SDRs by the Fund to development agencies, to be used by these agencies in their operations on the basis of internationally established rules. This could be done on the basis of a new statutory provision which stipulated that a given proportion of any SDRs created should be made available to development agencies, with the remainder being allocated to participants on the basis of quotas. This approach would entail fluctuations in the flow of SDRs to development agencies in accordance with variations in the decisions on the need for global SDR creation. Alternatively, the amount to be put at the disposal of development agencies could as far as possible be kept stable in the short run, with allocations to participants absorbing any variations in amounts created.

(2) Maintenance of the principle that SDRs would be allocated only to participants, but with an agreement among the countries with high per capita incomes to transfer to development agencies either (a) part of the SDRs allocated to them or (b) the equivalent in currencies.

(3) The adoption of a new formula for the allocation of SDRs among countries, differing from the present one in that it would not be strictly proportional to quotas, but would channel to the developing countries a larger share of total allocations than corresponds to their share in Fund quotas.

(4) An adjustment of the Fund’s quotas that would raise the share of the developing countries in aggregate quotas. This would raise these countries’ participation in SDR allocations without a change in the distribution formula and, in addition, would increase the other benefits of their membership in the Fund, including access to conditional liquidity in the General Account and their proportion of voting strength.

The first approach would be the most direct way of contributing to development assistance, but it would also be likely to have the greatest impact on the working of the SDR mechanism. Technical and policy issues arising in relation to SDRs as a result of this and other possible changes that might be incorporated in the reform are reviewed briefly in Section 3 of Chapter IV. The approaches described under 2(b) and 4 are the only ones that would not require amendment of the Articles.

To make a full appraisal of the relative merits of the various possibilities that have been listed, certain economic and financial as well as technical and institutional aspects need to be considered. Of these, the three following would seem to be the most important:

(a) The compatibility of different approaches with the effective working of the international monetary system, and in particular with the role of the SDR as a major reserve asset. A key issue in this context is to what extent a given approach would permit decisions on global reserve creation to be taken on their own merits.

(b) Whether these approaches would make a net addition, and which of them would be likely to make the largest net addition, to the provision of resources for development from other sources and would involve the minimum risk of being substantially offset by reductions in other aid flows of comparable quality.

(c) Which of the approaches would contribute most to an improvement in the average quality of official development assistance, including cost aspects. This would depend importantly on the rate of interest charged on net use of SDRs, in particular if the SDR interest rate were increased as discussed in Section 3 of Chapter IV.

A study of these and a number of related questions has been undertaken and will be transmitted to the Board of Governors by the Executive Directors as soon as it has been completed.

ANNEX 1 RESOLUTION 26-9 Adopted by the Board of Governors of the Fund on October 1, 1971

Whereas the present international monetary situation contains the dangers of instability and disorder in currency and trade relationships but also offers the opportunity for constructive changes in the international monetary system; and

Whereas it is of the utmost importance to avoid the aforesaid dangers and assure continuance of the progress made in national and international wellbeing in the past quarter of a century; and

Whereas prompt action is necessary to resume the movement toward a free and multilateral system in which trade and capital flows can contribute to the integration of the world economy and the rational allocation of resources throughout the world; and

Whereas consideration should be given to the improvement of the international monetary system and the adjustment process; and

Whereas the orderly conduct of the operations of the International Monetary Fund should be resumed as promptly as possible in the interest of all members; and

Whereas all members of the Fund should participate in seeking solutions of the aforesaid problems;

Now, therefore, the Board of Governors hereby resolves that:

  • I. Members of the Fund are called upon to collaborate with the Fund and with each other in order, as promptly as possible, to

    • (a) establish a satisfactory structure of exchange rates, maintained within appropriate margins, for the currencies of members, together with the reduction of restrictive trade and exchange practices, and

    • (b) facilitate resumption of the orderly conduct of the operations of the Fund.

  • II. Members are called upon to collaborate with the Fund and with each other in efforts to bring about

    • (a) a reversal of the tendency in present circumstances to maintain and extend restrictive trade and exchange practices, and

    • (b) satisfactory arrangements for the settlement of international transactions which will contribute to the solution of the problems involved in the present international monetary situation.

  • III. The Executive Directors are requested:

    • (a) to make reports to the Board of Governors without delay on the measures that are necessary or desirable for the improvement or reform of the international monetary system; and

    • (b) for the purpose of (a), to study all aspects of the international monetary system, including the role of reserve currencies, gold, and special drawing rights, convertibility, the provisions of the Articles with respect to exchange rates, and the problems caused by destabilizing capital movements; and

    • (c) when reporting, to include, if possible, the texts of any amendments of the Articles of Agreement which they consider necessary to give effect to their recommendations.

ANNEX 2 [Resolution No. 27-10 on the Establishment of a Committee on Reform of the International Monetary System and Related Issues that was attached as Annex 2 is reproduced below, pp. 151–53.]

The full text of the Resolution [No. 27-10] of the Board of Governors under which this Committee was established is reproduced below, pp. 151–53.

The Role of Exchange Rates in the Adjustment of International Payments: A Report by the Executive Directors, International Monetary Fund (Washington, 1970). [Reproduced in Margaret Garritsen de Vries, The International Monetary Fund, 1966–1971: The System Under Stress, Vol. II, Documents (Washington: International Monetary Fund, 1976), pp. 273–332.]

Ibid., p. 70. [Page 324.]

The term “par value,” or “par,” is used in this Report to indicate the value of a currency in terms of gold or SDRs; the term “parity” denotes the relationship between two currencies, determined by the ratio of their par values.

Issues involved in the movement of market rates within wider margins and the appropriate width of margins are discussed in Section 5 of this chapter.

The relevance of the balance between devaluations and revaluations to the attractiveness of the SDR as a reserve asset is discussed in Section 3 of Chapter IV.

The intervention and settlement arrangements that are related to this discussion are discussed in Section 5 of Chapter III.

In principle, intervention could be undertaken directly in SDRs, but this would require the extension of SDR holdings into private hands.

For the purpose of some of the calculations connected with Fund operations and the maintenance of the value of the Fund’s assets, the market values of currencies in terms of gold (and of SDRs) are normally determined by taking the value of the U.S. dollar to be its par value, and that of other currencies to be calculated from their market value in U.S. dollars. If it were so desired it would be possible for these calculations to be undertaken on a different basis, with the place of the U.S. dollar taken, for example, by an appropriate combination of currencies.

The authority to change the official price of gold, which is referred to in the Articles [that is, prior to the Second Amendment] as the power to make uniform proportionate changes in the par values of the currencies of all members, is a reserved power of the Board of Governors and its exercise requires a majority of 85 per cent of the total voting power of members (Article IV, Section 7; Article XII, Section 2(b)(iii)).

Each participant is required to accept SDRs from another participant if designated by the Fund and if the acceptance of these SDRs would not increase its total holdings above 300 per cent of its net cumulative allocation.

Abrogation of the reconstitution obligation—which does not require amendment of the Articles and would therefore not have to await other action on reform—would remove a further reason for the maintenance of a certain volume of transactions subject to designation.

The effects of wider margins on developing countries and other primary producing countries were referred to in Section 5 of Chapter II and discussed in considerable detail in the 1970 Report on the Role of Exchange Rates (cited in footnote 2, p. 26 above), pp. 60–63. [Pp. 318–21.]

    Other Resources Citing This Publication