IMF History (1972-1978) Volume 3

Sketch of a Reformed System

International Monetary Fund
Published Date:
February 1996
  • ShareShare
Show Summary Details

In March 1972 a paper prepared in the Research Department under the direction of the Economic Counsellor and Director of the Research Department, J. J. Polak, which presented a sketch of the scope and content of a reformed international monetary system was circulated to the Executive Board. This paper is published as (A) below.

In May 1972, after the Executive Directors had discussed this staff sketch, two supplementary notes to the sketch, also prepared in the Research Department, were circulated. These supplementary notes were aimed at answering two questions that Executive Directors had raised about features of a reformed system: (1) the possible financing by the Fund of short-term capital movements and (2) the possible purchase by the Fund of gold offered to it by its members. These supplementary notes are published as (B) and (C) below.

(A) Reform of the International Monetary System A Sketch of Its Scope and Content

(March 7, 1972)

I. Introduction—The Need for Reform

Reform of the international monetary system means something quite different now from what it meant before August 15, 1971. Earlier, reform was concerned with improvement of a system—frequently referred to as the Bretton Woods system—that showed certain specific weaknesses but whose basic suitability was not generally questioned; now, the task of reform must address itself to the central fact that the interim arrangements developed since last August do not reflect economic and political realities and must, therefore, be replaced by a truly viable system. If this was not done, there was a risk that these arrangements would develop, in an unplanned way, into some new system—probably one that would involve the separation of the world into a number of currency blocs, with the attendant economic and political hazards of such blocs. Clearly, it is now a matter of urgency that serious attention be devoted to the problem of international monetary reform.

The present situation is characterized by notable contradictions. The changes in relative economic positions among major countries during the past 25 years have meant that the United States is no longer in a position, as it was in the early years of the Fund, to assume ultimate and virtually sole responsibility for the functioning of the system. Correspondingly, the United States has felt an increasing need for access to the same facilities for international financial adjustment to changing circumstances as have been available for other countries; and in these conditions, the special position of the U.S. dollar at the center of the international system has become increasingly anomalous, for the United States and for other countries. In spite of this the system is more dollar centered now than it was 25 years ago. The decision to create a special drawing rights facility in the Fund reflected an expressed desire that the share of reserve currencies in total reserves should decline; yet that share has since increased to an all-time high, and there is no assurance that it will not increase further. Although the immediate cause of the crisis of August 1971 was the unwillingness of other countries to accept a dollar standard, the present arrangements are more dollar centered than ever before. A minor additional point, but an interesting symptom: when the gold value of the U.S. dollar was unquestioned, parities were nevertheless expressed in gold; now the central rates of many currencies are expressed in U.S. dollars.

The reform of the system, if it is to be worthwhile, will have to be radical. It will have to involve a quantum jump into a new system, not merely attempts at improvement of certain features of the present system. A reform of so wide a scope is, of course, difficult to envisage. It will entail changes in existing practices, the need for which may not seem obvious in itself but that are necessary from the point of view of the system as a whole. The reform will also have to take into account the diverse positions and interests of different groups of Fund members.

In a context in which countries may have to accept certain changes that may strike them as both radical and inconvenient, every effort should be made not to overload the reform with changes that are not essential. The package of reforms should include all ingredients considered necessary for a more effective functioning of the system for the next 15 or 20 years—and nothing more. This has particular relevance to what is proposed with respect to both gold and SDRs. At this stage, it is believed, no useful purpose would be served by decisions or arrangements to demonetize gold or by extending the holding of SDRs beyond official circles.

The central characteristics that distinguish the system sketched below from the Bretton Woods system as it operated in the 1960s are the introduction of greater exchange rate flexibility and a much reduced role for reserve currencies. While that role was a natural development of the system and contributed greatly to its success in the postwar period, it has become increasingly clear that it leads to serious difficulties for the adjustment process and is not compatible with international control over the level of global reserves.

The four subsections that constitute Section II of this paper discuss certain major components of the reform that need to be considered:

1. improvement of the exchange rate mechanism;

2. reduction of reserve currencies in other countries’ reserves, primarily by means of consolidation into SDRs;

3. introduction of equivalence of the dollar to other currencies in the financing of balance of payments deficits and surpluses; and

4. possible replacement of dollar intervention by symmetrical multicurrency intervention.

Section III then provides an analysis of the interrelations among the elements discussed, sets them against the background of the adjustment process as a whole, and introduces certain comments on certain other aspects of the system.

Beyond the core of the reform, certain changes in a number of other areas will also be required if the reform is to represent a consistent and acceptable whole. These are discussed in Section IV.

Many of the subjects dealt with here have been analyzed in greater detail in previous staff papers. The most directly relevant of these are referred to in the appropriate sections. Since the staff papers issued on reform subjects since last August formed part of a process of development of ideas, the present paper does not follow in all respects the lines of earlier papers. It also sets out some preliminary thoughts on aspects that have not been discussed previously.

II. Discussion of Major Reform Components

1. The Exchange Rate Mechanism

Reform in the exchange rate mechanism and in its operation will have to be more fundamental than was generally envisaged at the time of the 1970 report by the Executive Directors.1 There is general agreement that the par value system is to be maintained. But if that system is to be less crisis-prone than in the past, members must adjust parities before the evidence has become overwhelming that a change is necessary. That is to say that they must not only, negatively, abstain from par value changes where there is no fundamental disequilibrium; they must also, positively, make par value changes whenever there is a fundamental disequilibrium.

Prompter parity changes will, on the whole, also entail changes smaller than those in the past. This, in conjunction with wider margins—which it is assumed will become a permanent feature—will curb anticipatory speculation. There was substantial agreement on this in the 1970 report. What has become clear since is that the implementation of a more active exchange rate policy requires considerably more initiative by the Fund than existed before August 15, 1971—and also on the part of all members adequate domestic arrangements to bring about parity changes without delay. One further important factor in the new situation is that parity changes of the U.S. dollar have become an acknowledged part of the exchange rate mechanism.

A condition for a reasonable and consistent par value policy by members is a solid common basis in terms of the relevant facts and members’ approach to the facts, in particular with respect to

(a) the extent to which prevailing parities at any moment of time are reasonably compatible with payments balance over the cycle as a whole;

(b) the approximate magnitude of the relative exchange rate changes necessary to bring about the desired results; and

(c) the manner in which any important discrepancies from positions of fundamental equilibrium are to be eliminated, in particular the relative role of devaluations and revaluations.

In the absence of this basis, members might well be at a loss to know what specific action would be expected from them under the general rubric of “prompter and smaller parity adjustment,” and the Fund could equally find it difficult to advise members about desirable exchange rate action—indeed to exercise meaningful authority in response to proposals by members for par value changes.

A beginning toward a common basis of understanding has been created by the calculations and consultations in the context of the 1971 realignment. The realignment decisions of December 18, 1971 were possible only when a sufficient measure of working agreement on the three aspects mentioned had been reached.

It may be hoped that if in the future individual parity adjustments are made without undue delay, multicurrency parity adjustments on the scale of the 1971 realignment will not reoccur. But adjustments of the parities of one or two currencies will require fundamentally the same approach if they are to be carried out reasonably promptly.

The calculations and consultations in connection with the 1971 realignment show that in principle this approach is possible. It provided the basis for a multilateral decision, in circumstances where it was clear that the adjustments needed were relatively large and the reaching of a decision was felt by all to be imperative. Technically, however, the approach still leaves much to be desired, e.g., in the definition of operationally relevant national policy objectives, in the allowances to be made for cyclical abnormalities, and in the calculation of future effects of past exchange rate and price movements. As a consequence of these and other weaknesses, including perhaps some of which one is not yet aware, the approach can at this stage yield at best only rather vaguely defined equilibrium zones. These weaknesses cannot be expected to be removed in a short period of time. More is required than intensive analytical and statistical work; the international community will also have to live with these problems through a variety of circumstances before it can feel confident to have mastered them. This entire subject will have to be left to growing understandings among members, which may be reflected in increasingly precise general decisions on this subject by the Executive Directors.

A considerable advance over the present situation could be achieved be a clear exposition (presumably in a report accompanying the amendment) of the broad aims of the new exchange rate regime, the role to be played in this regime by Fund initiative, and the way in which members would be expected to collaborate with the Fund to implement the principles of that regime.

The amendment itself can be—and will have to be—narrower in scope. The general approach under which par value changes should be related to disequilibria of a basic or more than transitory character at prevailing par values continues to be valid. This approach does not fit formula changes in par values of the “automatic crawling peg” variety; but there does not appear any likelihood that members would want to take recourse to this device. It is primarily a matter of judgment whether the adjective “fundamental” should continue to be used to describe disequilibria for the correction of which par value changes are judged to be appropriate within the framework of a more flexible parity system.

The amended Articles should further provide for a permanent regime of wider margins—and preferably one that would be compatible with the practice of multicurrency intervention.

Finally, in the consideration of amended exchange rate provisions, the Fund will have to come to terms with the fact that situations are likely to arise in which a member cannot maintain an existing par value nor make the immediate transition to a new par value—although such situations should be more exceptional than in the past if par value changes were smaller and margins wider. The Fund will have to decide whether to create provisions under which it can approve temporary deviations from par values, or whether it will continue to treat such deviations as lapses from the legal regime.

2. Consolidation

Any consolidation arrangements for official holdings of reserve currencies will, in the main, have to provide the holders of these currencies with an alternative immediately usable asset. There is no other convenient asset for this than the SDR (with suitable modifications); the creation of a different consolidation asset that would have to coexist with the SDR, or the establishment of two different SDRs, would be impracticable.

In order to bring about consolidation, the Fund would therefore have to have the power to make special issues of SDRs against balances of dollars (and other reserve currencies) tendered by members.

If intervention in the exchange markets continues to be based on the use of reserve currencies, consolidation could not extend to members’ working balances in foreign exchange—a concept that could be defined more or less broadly. This latter question is related to the choice between voluntary consolidation and consolidation that would in some manner be mandatory for members of the Fund or participants in the Special Drawing Account. The general issue of the scope of consolidation will be discussed in Section III below. In any event, however, consolidation will to a large extent have to rely on the willingness of countries to hold their reserves to the agreed extent in the form of special drawing rights. Official reserves and other official or semiofficial assets are close substitutes for each other, and if reserves were to be kept in a form that the authorities considered substantially inferior to that available for external assets held by others than monetary authorities, the system would risk being seriously undermined. This has important consequences for the rate of interest on SDRs.

Insofar as the total reserves of some countries are at present excessive, consolidation could take the form of bilateral funding of dollars with the United States. This would minimize the amount of SDR creation for consolidation in the short run and would thus reduce the burden on the SDR system; it is not obvious that it would have other important benefits to the system, since the United States could hardly be expected to redeem these credits more rapidly than the dollars the Fund would acquire as a result of consolidation in SDRs. Thus, unless the dollar loans were of very long duration—which might not suit the creditors—or the United States ran some very large surpluses, bilateral funding might be little more than an interim arrangement pending replacement of these dollars by SDRs. If all foreign official dollar holdings beyond working balances were included in a consolidation arrangement, countries with excessive reserves could make bilateral loans to the United States in SDRs rather than in dollars. The resulting increase in reserve assets held by the United States would contribute to its ability to undertake the conversion obligations discussed below.

The technical aspects of consolidation are not complicated in principle, although the precise terms would raise many difficult issues. The gold or SDR value of the balances of reserve currencies thus acquired by the Fund would be maintained and the issuer would undertake to make annual service payments on these balances. Service payments would be related to the interest paid by the reserve center on the consolidated balances in a recent reference period or to the current interest rate on similar assets in its market; the second alternative would probably be preferable. The service charges would serve in the first instance to enable the Fund to pay interest on the equivalent special issue of SDRs, with the balance being available for amortization of balances of reserve currency held by the Special Drawing Account and/or for some other agreed purposes.

The consolidation facility should apply to all reserve currencies, and measures should be taken to discourage the growth of official balances in such currencies. This would involve abolishing guarantees on such balances and perhaps agreement that, insofar as reserves continued to be held in reserve currencies, Fund members should refrain from shifting from one reserve currency to another and would hold currency balances only in the reserve center concerned, as distinct from Eurocurrency markets.

3. Financing of Balance of Payments Deficits and Surpluses of Reserve Centers

Convertibility—in the sense of the right attached to balances in a given currency held by monetary authorities to be converted into reserve assets upon demand without any obligation on the holder to exercise this right—has served the system well in the past, until it led to an undue expansion of U.S. dollar balances. It is no longer in itself suitable either for the dollar—for which it no longer exists—or for certain emerging reserve currencies where a potentially dangerous expansion of reserve liabilities already threatens to begin.

In present circumstances the need is for conversion arrangements which would ensure that further increases in reserve currency balances are avoided (except where some modest increase is considered necessary to raise working balances). Such arrangements would require (annual or quarterly) compulsory conversion by the reserve center of any increase in balances of its currency that might arise from its overall payments deficits. For this, the reserve center would use its holdings of reserve assets (SDRs, gold, or reserve positions in the Fund) or draw on negotiated credits. The countries that would have to exchange currency balances for these reserve assets could be determined by a method administered by the Fund, such as designation, or each country could convert any increases that had taken place in its holdings of reserve currencies. Balances existing at the beginning of the overall plan would not be subject to conversion by the reserve center; however, the holder would have the option (and perhaps up to a point the obligation) to turn these balances into SDRs through the consolidation arrangements.

A counterpart to financing by the reserve center of any deficits with reserve assets would be its acquisition of reserve assets when it was in surplus. In part this might arise naturally as other countries had to use reserve assets to finance their deficits. The remainder—as measured by the extent to which the amount of reserve currencies had shrunk as a result of net payments to the reserve center—could be met by the reserve center being entitled to present this amount to the consolidation facility for exchange into SDRs. To this extent at least, the consolidation facility would have to be open-ended for the amount of any currency outstanding at the time the arrangement between the reserve center and the Fund was concluded.

4. Symmetrical Multicurrency Intervention

The central place of reserve currencies, in particular the U.S. dollar, in the present system derives from their use as intervention currencies. In certain respects, such as exchange rate adjustment and the financing of balance of payments imbalances, symmetry can be promoted along the lines indicated in preceding sections. Full symmetry, however, even in these respects might require equating the dollar with currencies of a substantial number of other members in the intervention process. Under a system of multicurrency intervention each of these members would buy or sell the currency of each other member on exchange markets, under established arrangements, with a view to influencing their relative exchange rates, the transactions being settled more or less simultaneously by a direct transfer of primary reserve assets between the countries in question.

Under this method of intervention the relative exchange rates for any two of the currencies concerned could deviate from their parities by the same margin, in contrast to the difference in margin that now applies between non-dollar currencies and the U.S. dollar, on the one hand, and vis-à-vis each other, on the other hand. Intervention could be carried out (a) by deficit countries acquiring the currencies of other countries in exchange for primary reserve assets, and then selling them in the market for their own currencies, (b) by surplus countries buying the currencies of other countries on the exchange market and selling them for primary reserve assets to the issuers, or (c) by the Fund.

Any such system would need to be confined to currencies which were actively traded in exchange markets. The issuers of other currencies would probably find it convenient to continue to peg them somewhat narrowly to one or other of the principal currencies.

An important concomitant feature of any symmetrical intervention arrangement would be an undertaking on the part of the countries concerned not significantly to vary, except in the very short run, their holdings of currency reserves. This would provide a strong incentive for countries to exchange substantial amounts of such balances, under consolidation arrangements, for newly created SDRs.

III. Interrelations Among Elements of the Reform

This section discusses certain interrelations among the various elements briefly analyzed above, as well as between them and certain other, more general, aspects of the system. It also explores the extent to which the reform measures mentioned are to be considered essential for the reform as a whole to be successful. While some of the questions considered appear to have rather clear-cut answers, this cannot be said for all of them. When one contemplates the reasons for the successful functioning of the international monetary system in the 1950s and 1960s, one realizes the fundamental importance that attached to the attitudes of members—the willingness to cooperate, the desire to make a reality of the existing international institutions—compared with the more formal features of “the system.” Without doubt, much the same considerations will be of overwhelming importance in determining the viability and the success of any future system.

1. The introduction of greater exchange rate flexibility than existed in the past is essential for a more effective functioning of the system. In addition, exchange rate flexibility for the U.S. dollar would appear to be a necessary condition for effective convertibility of the U.S. dollar in the sense defined above: The United States could not reasonably undertake a commitment for asset financing of deficits unless it could count on exchange rate policy as one means of curing deficits.

2. The link between convertibility as here defined and consolidation of dollar balances is not primarily technical. Asset financing of U.S. deficits does not require consolidation; it may be recalled that the Managing Director recommended a U.S. policy to that effect at the Annual Meeting in Copenhagen within the framework of the then prevailing system. Furthermore, asset financing of a U.S. surplus requires only the existence of a facility that provides the United States with reserve assets to the extent that there is a net reduction in dollar balances outstanding. Nevertheless, it is questionable whether, in the future, U.S. dollar convertibility, in the sense here defined, can be a solid element of a system in which members continue to hold the bulk of their reserves in dollars. An important point to bear in mind in this connection is that the present inconvertibility of the U.S. dollar is not comparable to the inconvertibility of the currencies of other industrial countries before 1958. The latter found in the course of the 1950s that the acceptance of market convertibility of their currencies for all nonresidents was a precondition to their effective participation in the international financial mechanism, and to their currencies playing an active role in international money markets. Of course, the links to foreign money and capital markets have not proved an unmixed blessing. But these countries would not want to go back from convertibility to the financial isolation that preceded it. By contrast, inconvertibility of its currency into reserve assets has not condemned the United States to financial isolation, because the U.S. dollar continues to enjoy market convertibility.

Thus, while other countries embraced convertibility to attain immediate and direct benefits, the benefits to the United States of convertibility of the dollar—as has become clear in recent months—while in no sense smaller, are less direct; they consist in a better working of the system and the more general economic and noneconomic benefits that derive from this. In a world in which dollar balances are the predominant form of reserves, asset financing of its deficits by the United States may easily appear as an act of self-denial, to be undertaken for some theoretical benefit to the adjustment process or the stability of world reserves, with the inherent risk of such an arrangement being abandoned again in the event of a large U.S. deficit. A system in which other main countries were to refuse to make large-scale use of an existing consolidation facility because they preferred to hold dollars, and yet to insist on the conversion by the United States of marginal additions to dollar balances, might lack the strength that derives from a consistency of approach.

3. Large-scale consolidation would also help to facilitate exchange rate flexibility for the U.S. dollar, in the sense that it would further weaken countries’ natural assumption of maintaining the exchange rate between their currencies and that of their reserve center except when they were themselves in serious imbalance. That assumption has, however, already been weakened substantially by the 1967 experience with sterling, when few sterling area countries followed the devaluation of the pound, and by the recent dollar devaluation, when a number of countries with very large reserves in dollars appreciated vis-à-vis their reserve currency.

4. The introduction of multicurrency intervention would completely eliminate the special position of the dollar for those countries whose currencies would participate in such a regime, and would thus facilitate exchange rate flexibility for the U.S. dollar. It would bring about automatic asset financing for the countries involved and for the United States, and would leave no other practical option than consolidation for the whole of these countries’ holdings of dollars. Thus it would most adequately serve the ends of the reform as indicated. It would also, however, constitute a very sharp break from past practice. Further consideration will be required to see whether this is attainable at this stage. If not, effective reform might proceed without it, provided sufficient advance was made in the other elements.

5. Asset financing of the payments imbalances of all countries would ensure that the stock of reserves would no longer be affected by fluctuations in the amounts of reserve currencies outstanding. The burden of regulating the level of reserves would then fall far more heavily on the SDR facility—as indeed it had essentially been agreed that it should when the creation of that facility was considered some five years ago. There would be a corresponding need to ensure that the decision-making process of the SDR facility produced the best possible decisions.

6. Special problems arise in connection with the contentious issue of short-term capital movements, which have been responsible for the strongest pressures on the par value system in recent years. These problems, and alternative measures to deal with them, have been discussed in the Fund’s 1971 Annual Report2 and, in greater detail, in a related staff paper. It is not necessary to enter here into the alternative means available to countries to restrain excessive capital flows. What is of immediate interest is that in part the solution to the problem of capital flows will probably lie in accommodating such flows—which means that facilities must exist to finance them in one way or another. Financing without effect of global reserves could be effected by official borrowing from private capital markets abroad. Beyond this, arrangements or understandings are needed on the flows of official financing, and these arrangements are immediately linked to the control over the supply of reserves and to the convertibility rules for reserve centers. What is to be the future role of swaps, and how are they to be controlled? Should reserve creation by the Fund be in some way responsive to the needs for reserves caused by ebbs and flows of capital movements? Should fluctuations in dollar balances—with the implied derogation from the principle of asset financing of overall deficits—be allowed to accommodate such flows?

These questions will have to be resolved as part of the reform. To provide the necessary answers to these questions, a modus vivendi will have to be agreed among the attitudes (now sharply conflicting) of members with respect to the relative roles of measures of restraint (whatever form they may take) and measures of financing capital movements. In this context the powers of the Fund in connection with capital movements may also have to be reviewed.

7. In the broadest sense, the viability of the system will depend on the working of the adjustment process as a whole. The new arrangements with respect to the holding of reserve currencies are intended to provide better incentives for the adjustment process, and greater exchange rate flexibility will permit a more effective working of that system in one particular respect. Whether the working of the system will in fact improve will, of course, depend on countries’ willingness to make the system work and to use to this end the whole range of policy tools at their disposal. The role to be played in this connection by the Fund, both in its consultative and in its lending functions, will continue to remain essential.

Generally speaking, the rules of the adjustment process would require countries to refrain from policies that would lead to severe balance of payments disequilibria, and to take early corrective action if such disequilibria nevertheless materialized. With respect to some particulars, the rules of the adjustment process are laid down in legal obligations of members, such as those relating to the conditions in which a member may propose a change in its par value. In some further areas, a consensus is gradually being built up as to the appropriate policies for members from the international point of view; perhaps the most important of these are the conditions in which members should propose changes in par values. But there are also wide areas in which a member’s duties, and, particularly in situations of conflict, members’ respective duties, under the adjustment process are by no means established. Improvement in this area can hardly be seen as a precondition for reform of the system; rather, it is a task that will remain to be worked at before the reform, after the reform—and after the next reform.

8. This paper does not discuss sanctions or remedies to be applied by the Fund to improve the working of the adjustment process. The omission is believed justified on the ground that any contribution of such measures can at most be ancillary to the behavioral rules which they are intended to enforce, and that their consideration is better delayed until such rules have been devised. One of the main reasons adduced for the introduction of additional or alternative remedies in the Articles relates to the achievement of a greater participation by surplus countries in the adjustment process. The suggestions made above with respect to decisions to be made on the relative role of devaluations and revaluations in the exchange rate mechanism would appear to provide the most effective substantive approach to this question.

9. The enlargement of the functions of the Fund, as well as the increased importance of some of its present functions, that would be inherent in many of the elements of reform discussed here suggest the desirability of a new look at the decision-making process of the Fund. One particular suggestion in this area is discussed in a recent staff paper.3

IV. Other Aspects of the Reform

1. The Numeraire of the System

Gold is the numeraire of the present system—the standard in which par values are expressed. It is also the standard in terms of which the value of the Fund’s assets is to be maintained.

In the context of a much enlarged role for the SDR, consideration should be given to making the SDR, rather than gold, the numeraire of the system.

In a sense, the SDR is already the de facto numeraire, jointly with gold as the de jure numeraire. The value of the unit of special drawing rights is stated to be 0.888 671 gram of fine gold (Article XXI, Section 2), with a deliberate absence of any provision for change in the equivalence. Hence par values are 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.

Thus a change to make the SDR the de jure numeraire of the system—e.g., by a provision in the Articles expressing gold in terms of SDRs rather than SDRs in terms of gold—would be essentially presentational in nature, unless it were accompanied by some loosening of the link between the two. However, such a loosening would tend to increase the risk of speculation on price changes of gold versus SDRs as reserve assets, and should therefore be avoided.

Four related points need to be made in this connection:

(a) Whether the SDR is formally made the numeraire of the system or not, and whether gold is maintained in that position or not, the quality of the SDR must rest on the characteristics attached to it by the Articles (including in particular the acceptance obligation) and the wisdom of the decisions taken with respect to it pursuant to the Articles—and not merely on the definition of its equivalent in terms of gold or currencies.

(b) Changes over time in the value of the SDR in terms of currencies in general will depend in the first instance on the balance between devaluations and revaluations, although they could also conceivably be brought about by application of the provisions on uniform changes in par values.

(c) The extent to which the SDR is usable among participants or between them and the General Account is unrelated to its status as a numeraire.

(d) Perhaps more important, the role of gold in the Fund and the practices of members with respect to gold would continue to be of concern to the Fund whatever was done with respect to the role of gold as the numeraire.

2. Gold

There are many fundamental issues about the role of gold in the system, but not all of these need to be resolved in the context of the present reform. In particular, there does not seem any need to tackle on this occasion the question of the demonetization of gold, or to take special measures to induce members to replace their holdings of gold by holdings of SDRs.

It is necessary, however, as a minimum to provide for certain contingencies, even though it may not be widely agreed that they will materialize.

The first contingency is that, at a higher interest rate than at present, SDRs would become so much more attractive than gold that substantial amounts of gold would be sold to the Fund under Article V, Section 6(a), either directly for SDRs or for currency that would be converted into SDRs under the consolidation facility. As a result, the General Account would lose income on its holdings of SDRs or have to pay remuneration on the increased creditor positions. At the same time, it might find that its gold holdings were less usable than before to buy needed currencies because, if the preference for Fund positions or SDRs over gold was rather general, members would not be prepared to have the Fund sell gold for currencies. One way to meet this contingency might be a provision under which the Special Drawing Account would be authorized to acquire gold from the General Account against SDRs created for this purpose.

The second contingency, closely related to the first, is that the Fund may want to invest such gold holdings of the Special Drawing Account, so as to cover the interest payments on the matching SDRs outstanding, and may, as part of the investment operation, be prepared to sell the gold so invested in the free market or to see it sold by the investee. Any profit on the sale of gold above its SDR value could accrue to the issuer of the securities or to the Fund. It might be—although it is futile to attempt to predict in this area—that arrangements for the sale of gold in some collective manner such as the one described would prove valuable to keep down the premium in the free market and to continue a policy under which individual members would refrain from selling gold in the free market.

3. The Rate of Interest on SDRs

The attractiveness of SDRs compared with other reserve assets depends—apart from certain immaterial aspects that would be difficult to evaluate in general—on the expected future value of SDRs in terms of these other assets,4 and on interest rate considerations. Assuming that the exchange rate system of the future would be guided to balance, in a rough way, revaluations and devaluations of individual currencies, the value of SDRs would do about as well as that of currencies in general; with perhaps some bias in the system toward devaluations or depreciations, SDRs might do a little better than currencies in general. However this may work out exactly in practice, it would seem that when SDRs form a large proportion of total reserves the equilibrium interest rate on them cannot be far below the rate that can be earned on balance of currencies in general.

This could have a number of probable consequences.

1. The interest rate on SDRs, which has been constant at 1 ½ per cent so far, would probably have to be adjusted at intervals to reflect interest tendencies in market rates in major centers.

2. The service payments by reserve centers on consolidated liabilities would probably have to follow the same market rates so as to keep the Fund’s service receipts in reasonable balance to its interest payments on the SDRs issued in consolidation.

3. While remuneration on creditor positions in the Fund can deviate to some extent from the interest rate on SDRs, the similarity of the assets would make large differences undesirable; thus remuneration, too, may have to fluctuate more than it has done in the past.

4. This in turn could have consequences for the Fund’s debit interest rate (charges on balances of a member’s currency held by the Fund in excess of its quota).

5. Raising the SDR interest rate and the corresponding rate of charge would reduce the attractiveness of SDR allocations to countries that tend to be persistent net users of SDRs—mostly the developing countries with low reserves.

All these considerations suggest that the new system may be subject to strains arising from, on the one hand, the need to pay a sufficiently high interest rate on members’ reserves in the form of SDRs and, on the other hand, the desire to avoid unduly high levies on members drawing on the General Account, participants making net use of SDRs, and the reserve centers whose currency liabilities are to be consolidated. One should consider whether it would be possible to make a contribution to diffusing these strains by using some part of newly created SDRs to raise interest rates on SDR holdings, prior to any allocation on the basis of quotas.

4. Other Changes in the SDR System

In the context of the reform it will be necessary to review many features of the SDR system. The requirements of an asset that serves as the main reserve instrument are different from those of an asset that constitutes only a marginal supplement to other reserve assets. The necessary changes in operational provisions will obviously depend a good deal on the choices made among alternative aspects of a reformed system such as have been discussed above.

Compared with present provisions, the need for a number of changes is apparent. The provisions on Other Holders should be broadened to include a wider group of official bodies; there should be some broadening of the categories of the transactions in SDRs that are permitted among holders; and the range of potential transactions in SDRs with the General Account should be extended.

In a system in which SDRs are created other than by allocation to participants, acceptance obligations in excess of allocations can no longer be limited to twice the amounts of such allocations. Either a new formula will have to be devised to ensure adequate room for the acceptance of all outstanding SDRs or, more logically in the new system, the limit on acceptance obligations should be abolished, an equitable system of acceptance remaining assured by the rules of the designation system and/or the rules relating to holdings of reserve currencies in connection with convertibility commitments. If these latter rules were of a rigid kind, e.g., specifying that all incremental holdings of reserve currencies were to be presented to the issuer for conversion into SDRs, this might in principle supersede the system of designation, and also the requirement of need. If greater leeway were to be left to countries individually on their incremental foreign exchange holdings, rules for designation and the requirement of need would in any event have to be maintained, though the former would need to be revised to accommodate very much larger designation plans at the same time that participants would on the whole have smaller balances of convertible currencies. The present rules on reconstitution would lose most of their present rationale in conditions where countries held their reserves predominantly in SDRs. In the context of the adjustment process there would be reason for legitimate concern, however, if countries exercised an undue drain on other countries’ resources by a large and persistent use of reserves in general and of allocated SDRs in particular.

As mentioned earlier, the extension of the holding of SDRs beyond official holders does not appear to be a subject that requires consideration as part of the present reform.

5. The International Monetary System and Development Finance

Given the agreed view that present methods of marshalling resources for international development agencies, in particular the International Development Association (ida), leave much to be desired, the question arises whether substantial improvement in this area could be achieved by the provision of financial resources to development agencies through the international monetary system, and whether such action would be compatible with the effective functioning of this system. The question may receive increased emphasis in the context of the reform of the system because that reform may, as already mentioned, involve a number of aspects which, taken by themselves, may be seen by developing countries as potentially disadvantageous to them, such as the effect of replacement of holdings of reserve currencies by SDRs on countries with large reserves and the effect of the increase of the SDR interest rate on countries with low reserves.

The question was originally posed as that of a link between SDR creation and development finance, but in the context of the reform it could be considered more broadly under the following four headings, either as alternatives or in possible combination.

(1) “The link,” i.e., the channeling of some part of the SDRs created by the Fund (or the equivalent in currencies) to development agencies, with the transfers to be made either by the Fund itself or by the high-income countries upon receipt of their allocations.

(2) The use for the same purpose of some part of the service payments made by the reserve centers to the Fund on currency balances held by it as a result of consolidation. (Section II, paragraph 2, above.)

(3) The investment in development agencies of gold acquired by the Special Drawing Account. (Section IV, paragraph 2, above.)

(4) The investment of the Fund’s general reserve (a question already under active consideration in the Fund).

While the potential amount involved under (4) is not large and that under (3) dependent on developments that cannot be appraised now or probably for years to come, it would appear that the order of magnitude of the financial resources that could be provided by (1) and (2)—assuming that one or both had been judged a suitable source for development finance—might well be comparable to the current annual rate of ida contributions.

(B) Reform of the International Monetary System A Sketch of Its Scope and Content Supplementary Note on Possible Fund Financing of Short-Term Capital Movements

(May 2, 1972)

In Section III, paragraph 6 of the sketch, the point was made that there would be a need for arrangements or understandings on the financing of capital movements insofar as such movements would not be restrained by various measures, because the resources normally available to members in the form of reserves and credit tranche facilities in the General Account would not be likely to be adequate to enable members to handle such exceptionally large capital flows as might from time to time occur. One question that was raised was whether reserve creation by the Fund should in some way be responsive to the needs for reserves caused by the ebbs and flows of capital movements. Some of the issues that would arise if one were to consider Fund financing of such flows would be the following:

1. The members to which such a facility, if it were created, should be available—i.e., the members exposed to short-term capital movements that were relatively large compared with their flows of basic payments. Should the facility in principle be available to all members, to members with Article VIII status, or perhaps those Article VIII members that met certain additional criteria?

2. The size of the facility, in terms of individual members’ quotas and overall. This question would depend to a large extent on the expectation of the effects of wider margins and other measures on the size of capital movements, on the understandings with respect to the future role, if any, of the swaps now in existence, on the degree of adequacy of the supply of reserves and Fund credit, etc.

3. The connection between the use of such a facility by a member and conditions set by the Fund, or consultations to be held with the Fund, on measures to reduce the flow of capital.

4. Features of the facility—e.g., in terms of interest rates or short repayment terms—to minimize the risk of its use to meet the financing of a basic deficit.

5. Financing arrangements to ensure that the Fund would have the means to meet promptly even very large drawings under the facility and would be able to ensure in turn the usability of the assets acquired by the members that had made the financing available to the Fund.

(C) Reform of the International Monetary System A Sketch of Its Scope and Content Supplementary Note on Gold Purchases by the Fund

(May 2, 1972)

In Section IV, paragraph 2 of the sketch, mention was made of the possibility that SDRs would become so much more attractive than gold that substantial amounts of gold would be sold to the Fund under Article V, Section 6(a). In the discussion of this question it was observed that, in the context of a general amendment of the Articles, it would be necessary to clarify the uncertainty that now exists with respect to members’ rights and obligations under this Article. It has been concluded by the staff that Article V, Section 6(a) imposes a duty on the Fund to purchase the gold that members are obligated to sell to the Fund under that provision, but this conclusion has been questioned; so far, this issue has not been settled by the Executive Directors. On December 30, 1969, the Fund decided, “as a matter of policy,” that it would purchase gold from South Africa when South Africa offered gold for sale in certain specified circumstances.

In any amended version of Article V, Section 6(a), the Fund would want to make it clear (a) in what circumstances a member would be entitled to sell gold to the Fund either generally or with the permission of the Fund; and (b) in what (if any) circumstances a member would be obliged to sell gold to the Fund and the Fund would then be held to buy such gold from the member.

Without entering into the substance of what may be decided on these issues, it can be seen that the question posed in the sketch could, in any event, arise, namely that members would want to sell gold to the Fund on a large scale which the Fund would either be obliged to purchase or, having a discretion, would want to purchase. The Fund might want to purchase the gold either to prevent members from selling the gold in the free market or, if the free market price was below the official price, to avoid putting members in the position of being unable to sell the gold legally. Thus, it would appear that some provision to deal with this contingency will, in any event, have to be considered as part of the reform.

The question of the conditions in which the Fund can sell gold to members would also deserve consideration in the same context.

The Role of Exchange Rates in the Adjustment of International Payments: A Report by the Executive Directors (Washington: International Monetary Fund, 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. 273–330.]

Annual Report of the Executive Directors for the Fiscal Year Ended April 30, 1971 (Washington: International Monetary Fund, 1971), pp. 14–16.

“An Advisory Committee of the Board of Governors—Outline of an Illustrative Plan,” January 24, 1972 [reproduced below, pp. 129–41].

Since it is assumed that the firm link between gold and SDRs of the present Articles [that is, prior to the Second Amendment] will remain, this is the same concept as the expected implicit value of the gold value maintenance of the SDR.

E. B. Decision No. 2914-(69/127), adopted effective December 20, 1969. See Selected Decisions of the Executive Directors and Selected Documents, Sixth Issue, September 30, 1972 (Washington: International Monetary Fund, 1972), pp. 57–62, and Margaret Garritsen de Vries, The International Monetary Fund, 1966–1971: The System Under Stress, Vol. II, Documents (Washington: International Monetary Fund, 1976), pp. 203–206.

    Other Resources Citing This Publication