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

Link Between SDRs and Development Finance

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

In February 1973 the Research Department prepared a paper on the allocation of SDRs and the provision of financing for economic development—that is, the link. The paper was subsequently discussed in the Executive Board and the following slightly revised version was sent to the Committee of Twenty.

Allocation of SDRs and Financing of Economic Development

(February 8, 1973)

1. Introduction

From the time when academic proposals for issuing an international reserve asset began to appear in the late 1950s, there have been frequent suggestions that the issuance of such an asset should in some manner be made to enhance the flow of resources to less developed countries, and foster economic development.

In view of the analogy between official reserves and private money balances, the idea of a link between reserve creation and the financing of development is one which rather naturally comes to mind. Within the confines of the national state, the issuance of fiduciary money has traditionally been associated either with a transfer of resources to the sovereign or with the financing of capital expenditure within the community through the lending activities of the banking system. But those who have advocated that the creators of international liquidity, like the creators of national liquidity, should at the same time influence the distribution of real resources have had a special purpose in mind—not merely, as in the case of national banking, to reduce the imperfections of the capital market and thus promote a more productive use of capital, but also, and more specifically, to foster the development of the countries with relatively low per capita income. Such proposals reflect a dissatisfaction with the magnitude, and in some cases the “quality,” of development assistance, together with a doubt as to whether these shortcomings are likely to be overcome by the approaches thus far pursued.

When the Special Drawing Account of the Fund was established in 1969 a different view prevailed. The liquidity-creating and credit-distributing functions of a banking system are conceptually quite distinct, and it was deemed prudent, in establishing a new international reserve asset, to concentrate on the liquidity-creating function. Newly created SDRs had, of course, to be distributed somehow among participating countries, but in basing allocations on the quotas of the Fund the intention was to choose a distributive key which would roughly correspond to the effective demand for reserves and preclude, or at least confine to small proportions, any permanent transfer of resources among participants as a result of the allocations.

While the period that has elapsed since the establishment of the Special Drawing Account is a relatively short one, it is appropriate that the question of a possible link between SDR allocation and development finance should be brought up for reconsideration in the context of the general re-examination of the international monetary system that is now in progress. Consideration of this question calls for a careful weighing of the probable consequences of adopting or not adopting one or other of the possible link proposals, particularly for (a) economic development and (b) the international monetary system and the process of balance of payments adjustment. The present paper does not attempt to arrive at a judgment of these matters, but merely to present a variety of considerations germane to such a judgment.

The purpose of this paper is to assist members in making up their minds whether there should be a link between the allocation of SDRs and the financing of economic development and, if so, what the form of any such link should be.

2. Past Link Proposals

Past proposals for a link between reserve creation and development financing have generally been presented as part of specific plans for equipping the international monetary system with capabilities for generating global liquidity. The following notes are confined to suggestions that have appeared within the last ten to fifteen years.

Under the last of a series of plans put forward by Mr. Maxwell Stamp,1 the Fund would be allowed to issue certificates, which member countries would agree to receive in settlement of international obligations and to treat as reserves. The Fund would put the certificates into circulation by lending them on a long-term basis to the International Development Association (ida) for use in the ida’s normal lending. The ida would then obtain currencies required for loan disbursements by presenting the certificates to the supplying countries. Countries could if they wished refuse to accept export sales—and the associated certificates—connected with ida lending. Interest would be paid by the ida to the Fund and by the Fund to holders of the certificates.

The holding of reserves in the form of deposits with the Fund rather than in national reserve currencies was the key feature of a proposal by Professor Triffin.2 Such deposits could be used by the Fund for advances or rediscounts, undertaken at the initiative of the borrowing country, and for open market operations or investments, undertaken at the initiative of the Fund. A proportion of such investments could be channeled into development finance through purchases of International Bank for Reconstruction and Development (ibrd) bonds or other securities of similar character.

A group of experts appointed by the United Nations Conference on Trade and Development (Unctad)3 to study international monetary issues presented a link proposal under which additional international liquidity would periodically be created by the Fund issuing reserve units to all its members in exchange for equivalent amounts of their own currencies. A part or all of the currencies of the developed countries (and perhaps of some of the developing countries) so acquired by the Fund could be invested in World Bank bonds. The Bank would, in turn, use the currencies in the course of its operations in development finance.

Under Professor Scitovsky’s plan 4 also, Fund obligations would be issued against member countries’ own currencies. They would be handed over to the ida as grants for use in financing approved investment projects in developing countries, with the proviso that they could be spent, in the first instance, only in the country against whose currency they had been issued. Once the Fund obligations had been spent in the appropriate country, they would become unrestricted reserves in the hands of the latter.

Dr. I. G. Patel 5 expressed the view that the simplest way of establishing a link would be by means of an agreement whereby each act of international liquidity creation would be followed by voluntary contributions to the ida by all Part I member countries (countries that are more advanced economically and contribute to the resources of the ida). The size of the contribution would bear a certain uniform proportion to the share of each of those countries in international liquidity creation.

After the SDR scheme took form, a number of proposals for using the SDR facility to provide development finance were advanced. The proposals left open the precise institutional arrangements. For instance, the Governor of the Fund for Italy, Mr. Colombo, at the 1968 Annual Meeting, suggested that the main industrial countries should undertake to use a part of their reserves corresponding to a portion of the SDR allocation for the replenishment of the ida or for subscription to World Bank bonds.6 Mr. Colombo’s proposal found support in the Subcommittee on International Exchange and Payments of the U.S. Congress, which in 1969 called for an amendment to the Articles whereby the 18 Fund members that had previously contributed to the ida would direct that 25 per cent of their SDR allocations be retained by the Fund to finance expanded ida development assistance.7

A second Unctad study group 8 envisaged two main approaches to the establishment of a link. Both would involve contributions by the developed countries to ida in proportion to SDRs annually allocated to them—in the first approach contributions in SDRs, in the second approach contributions in national currencies. In addition, the experts referred to a third method whereby the Fund would allocate some SDRs directly to ida.

3. Nature of Proposals to Be Examined

Five main categories of proposals for a link between SDR allocations and the transfer of resources to developing countries may be distinguished. These are listed under A to E below and are described and discussed in some detail thereafter.

Before proceeding to the discussion of these various proposals, it is useful—assuming that, as a result of SDR creation in the Fund, financial resources were to become available to be used for the benefit of developing countries—to inquire also into the possibility of extending the Fund’s own programs for these countries.

One possibility to be considered would be a new facility that could be established in the Fund for the purpose of providing resources to developing countries under stand-by or similar arrangements, to be repaid over periods longer than those for which drawings are presently made available. The use of the facility would be considered in situations where countries, faced with serious imbalance, would be prepared to embark on a comprehensive program involving major shifts in policies. These programs would cover a span of three or more years; they would contain broad understandings on such matters as public revenue and expenditure policies, exchange rate management, import policy, external debt control, interest rate, and other pricing policies; and they would be subject to annual reviews.

In the past in such instances Fund resources provided under its normal policies were often accompanied by large-scale and longer-term program lending by other countries. This source of needed resources at the time when stabilization policies are put into effect has declined. The facility suggested above would fulfill the need that now exists for balance of payments assistance of longer than five-year repayment term, associated with comprehensive policy adjustments.

Beyond the use of SDRs in stabilization operations of longer duration than those currently possible under the Articles, one could also consider the possibility that the Fund might use special issues of SDRs to extend credits to developing countries in other circumstances.

Resources for any such operations would be provided in the form of special issues of SDRs made as and when required to finance lendings. SDRs received by the Fund in repayment of loans would be returned to the Special Drawing Account for cancellation. Any net creation of reserves through the facility would be taken into account in the decisions on the need for reserve creation.

Terms concerning duration, repayments, and charges under which the resources would be provided could vary. For example, the interest charged might be the normal interest on SDRs or might be reduced below that level by any of a number of ways discussed below in connection with other types of schemes. Variations in the cumulative amount of special issues would be taken into account in calculating the global need for reserve supplementation in the form of regular SDR allocations.

In what follows, the possibilities for more extensive Fund credit operations will be referred to as extended Fund facilities. Beyond whatever may be decided in this area, the various “link” proposals that have been discussed so far may be presented as follows:

A. Schemes under which the Fund would allocate SDRs directly to development financing institutions (dfis).

B. Schemes under which less developed countries would be allocated a larger share of SDR allocations than corresponded to their share in Fund quotas.

C. Schemes under which the share of developing countries in Fund quotas (and hence in SDR allocations) would be higher than would result from the criteria under which quotas are presently determined.

D. Schemes under which some or all participants receiving SDRs in allocation would contribute to dfis amounts bearing some predetermined relationship to the allocations received.

E. Schemes depending on the establishment of a reserve substitution facility as part of the reform of the international monetary system—which would provide that any excess of interest received from reserve centers on currencies held in the facility over interest paid to holders of the corresponding amounts of SDRs would be contributed to dfis or to developing countries directly, as grants, and any amortization paid by reserve centers would be allocated to dfis.

The first four of these approaches were described in Chapter VI of the report by the Executive Directors to the Board of Governors on reform of the international monetary system; the fifth, though not discussed in that report, has also received some attention.

Both the extended Fund facilities and schemes of Types A, B, and E require for their implementation changes in the present Articles of Agreement. This is not necessarily true of schemes of Types C and D.

It should be emphasized that each of these types of proposal is capable of assuming a great many variant forms, and many hybrid forms are also possible, Only a limited number of the possible variants are evaluated in the following sections. The significance of some of them will appear more clearly in the light of subsequent discussion.

TYPE A Schemes—Direct Allocation of SDRs by the Fund to Development Financing Institutions

Under the heading of development financing institutions would be included the World Bank Group and various regional institutions.9 The distribution of the allocation among dfis might be determined by the Fund on the occasion of each decision to allocate or at longer intervals, or the dfis might set up a joint organization to suballocate SDRs among themselves.

dfis might be allowed to make such use of SDRs as seemed best to them in the interests of economic development, or various understandings might be reached relating to such matters as the proportions in which these funds would have to be used for financing development in the various developing countries or groups of countries, the terms (grant, concessionary loans, etc.) on which the dfis would have to provide development assistance to countries with funds derived from the allocations.

The dfis might merge the resources derived by them from SDR allocations with those derived from other sources, or they might segregate the SDRs allocated to them and use them to finance a special type of lending. The latter alternative would tend to be required if the Fund were to impose conditions upon the dfis with respect to the use to be made of the allocations on the lines sketched forth in the preceding paragraph.

The amount to be allocated to dfis in total might be determined on the occasion of each decision to allocate in general; or the proportion of global allocations to be assigned to dfis might be laid down in revised Articles of Agreement or decided for periods longer than a single allocation period; or, finally, the amount to be allocated to dfis might be decided for periods longer than a single allocation period.

It is assumed that a substantial part of global allocations of SDRs would continue to be made directly to participating countries. This part might be distributed among participants in proportion to quotas. Alternatively, developing countries might receive, as at present, a proportion of total allocations corresponding to their share in the quotas of all participants, while such SDRs available for allocation as were not distributed to dfis or to developing countries would be distributed among participating developed countries in proportion to their quotas.

The SDRs allocated to dfis, like those allocated to countries, would sooner or later become part of the world’s active stock of reserves and provision would have to be made for their acceptance in exchange for currency. At present each participant is obliged to accept SDRs up to the point at which it is holding three times its net cumulative allocation. The dfis would not be participants but “other holders,” and would obviously not be subject to acceptance obligations. The acceptance obligations corresponding to the SDRs allocated to dfis must therefore be placed on participating countries. This obligation might be imposed (a) on all participants in proportion to net cumulative allocations, (b) on all participants in proportion to net cumulative allocations subsequent to the inauguration of the “link,” (c) on all participants in proportion to quotas, or (d) on developed country participants only, in one or other of these proportions. If, as has been suggested, limitations on acceptance obligations of participants were removed as part of the international monetary reform, this problem would disappear.

Table 1.Development Finance Institutions1(In millions of U.S. dollars, unless otherwise indicated)
afdbasdbcabeicdbeadbeibibrd/ida2idb
Beginning year of operation19661966196119701967195819461960
Paid-in capital as of end-1971804922025202502,661405
Ordinary operations
Commitments: 1966–71503532423253617410377,1061,107
Recent terms (representative)
Interest rate8 (per cent)7.07.45.5–6.098–10107.50–7.8757.258
Maturity (years)8–2012–2510–207–127–2014–3015–20
Grace period (years)33–73–52–43–74–6
Special or concessional operations
Funds available11Yes5YesNoYesYesYes
Commitments: 1966–7101071221314169152,3422,185
Recent terms (representative)
Interest rate8 (per cent)142.5–3164140.753–1417
Maturity (years)1415–30145015–30
Grace period (years)143–1014103.5–4.5
Sources: International Bank for Reconstruction and Development; International Monetary Fund, International Financial News Survey, (various issues), and annual reports of various development finance institutions.
afdbAfrican Development Bank
asdbAsian Development Bank
cabeiCentral American Bank for Economic Integration
cdbCaribbean Development Bank
eadbEast African Development Bank
eibEuropean Investment Bank
ibrdInternational Bank for Reconstruction and Development
idaInternational Development Association
idbInter-American Development Bank
Note: The United Nations Development Program (undp) disbursed $1,052 million in development assistance during 1966–71, representing undp program cost on projects and for which there are no repayments. The second European Development Fund (edf) provided $800 million in development assistance, $730 million in grants, and $70 million in concessional loans, during 1964–69. The third edf has pledged $1,000 million for 1971–75: $810 million in grants, $90 million in concessional loans and $100 million in ordinary operations of the eib. The edf does not have one independent legal identity but is a part of the European Economic Community.

In certain contexts, development banks and the following are listed as multilateral development institutions or programs: United Nations Development Program, United Nations Fund for the Congo, United Nations High Commission for Refugees, United Nations Children’s Fund, United Nations Korean Reconstruction, United Nations Relief and Works Agency, United Nations Temporary Executive Authority (West Irian), World Food Program, Mutual Aid and Guarantee Fund for the West African Entente, Andean Development Corporation, the International Finance Corporation, and the European Development Fund.

The International Development Association began operations in 1960.

1967–71.

1968–71.

The Central American Bank for Economic Integration is composed of three lending funds: Ordinary Investment Fund, Economic Integration Fund, and Housing Fund. During the 1961–71 period the distribution of loans from these sources was 45 per cent, 47 per cent, and 8 per cent, respectively.

1970–71.

Loans to Greece, other associated states, and overseas territories of the European Economic Community.

Including commissions. In many instances there is also a commitment charge attached to undisbursed balances.

Excludes any commissions.

Loans to the private sector carry 10 per cent interest rate.

In April 1972 Development Assistance Committee countries, Yugoslavia, and Brazil pledged $100 million to the Special Fund of the African Development Bank.

Lending operations from Special Funds resources commenced June 1969.

An additional $1.35 million was financed by the Canadian Agricultural Fund.

Not applicable.

Comprises Special Section loans to Turkey, and to associated states and overseas territories on behalf of the European Development Fund of the EEC. For the period 1958–71, such loans totaled $175 million and $48 million, respectively.

During 1971, two loans were made at 1½ per cent.

Furthermore, the Inter-American Development Bank has administered Special Funds from Canada. Commitment of such funds are on International Development Association terms except the service charge is ½ of 1 per cent.

Sources: International Bank for Reconstruction and Development; International Monetary Fund, International Financial News Survey, (various issues), and annual reports of various development finance institutions.
afdbAfrican Development Bank
asdbAsian Development Bank
cabeiCentral American Bank for Economic Integration
cdbCaribbean Development Bank
eadbEast African Development Bank
eibEuropean Investment Bank
ibrdInternational Bank for Reconstruction and Development
idaInternational Development Association
idbInter-American Development Bank
Note: The United Nations Development Program (undp) disbursed $1,052 million in development assistance during 1966–71, representing undp program cost on projects and for which there are no repayments. The second European Development Fund (edf) provided $800 million in development assistance, $730 million in grants, and $70 million in concessional loans, during 1964–69. The third edf has pledged $1,000 million for 1971–75: $810 million in grants, $90 million in concessional loans and $100 million in ordinary operations of the eib. The edf does not have one independent legal identity but is a part of the European Economic Community.

In certain contexts, development banks and the following are listed as multilateral development institutions or programs: United Nations Development Program, United Nations Fund for the Congo, United Nations High Commission for Refugees, United Nations Children’s Fund, United Nations Korean Reconstruction, United Nations Relief and Works Agency, United Nations Temporary Executive Authority (West Irian), World Food Program, Mutual Aid and Guarantee Fund for the West African Entente, Andean Development Corporation, the International Finance Corporation, and the European Development Fund.

The International Development Association began operations in 1960.

1967–71.

1968–71.

The Central American Bank for Economic Integration is composed of three lending funds: Ordinary Investment Fund, Economic Integration Fund, and Housing Fund. During the 1961–71 period the distribution of loans from these sources was 45 per cent, 47 per cent, and 8 per cent, respectively.

1970–71.

Loans to Greece, other associated states, and overseas territories of the European Economic Community.

Including commissions. In many instances there is also a commitment charge attached to undisbursed balances.

Excludes any commissions.

Loans to the private sector carry 10 per cent interest rate.

In April 1972 Development Assistance Committee countries, Yugoslavia, and Brazil pledged $100 million to the Special Fund of the African Development Bank.

Lending operations from Special Funds resources commenced June 1969.

An additional $1.35 million was financed by the Canadian Agricultural Fund.

Not applicable.

Comprises Special Section loans to Turkey, and to associated states and overseas territories on behalf of the European Development Fund of the EEC. For the period 1958–71, such loans totaled $175 million and $48 million, respectively.

During 1971, two loans were made at 1½ per cent.

Furthermore, the Inter-American Development Bank has administered Special Funds from Canada. Commitment of such funds are on International Development Association terms except the service charge is ½ of 1 per cent.

An important question arises with respect to the interest rate on SDRs. Interest is received on SDR holdings and equivalent charges are paid on net cumulative allocations. At present the rate of interest and of charges is 1½ per cent, but this rate will probably have to rise in future if the share of SDRs in total reserves is to be substantially increased; it will, however, probably remain somewhat below the rates payable in most countries on government securities. The question is how the interest cost corresponding to the SDRs allocated to dfis would be met. One possibility is that the cost would be met by the dfis, which would recoup themselves through the interest charged on development loans. Or the cost might be borne by all participating countries in proportion to quotas, net cumulative allocations, or some other yardstick. Or it might be met by all developed country participants only in proportion to quotas or net cumulative allocations. Finally, it is conceivable that, as part of the reform of the international monetary system the interest paid by net debtors in the SDR system (including that paid by dfis) might be at a lower rate than that paid to net creditors, or that the interest paid on allocations (including that paid by dfis) might be at a lower rate than that paid on SDR holdings—the difference, in either case, being met through assessments on participating countries, or on developed countries, or through new issues of SDRs. Whichever of these methods is adopted, it should be noted that there is no necessary connection between the way in which the burden of any interest concession to dfis is distributed, and the way in which allocations to participants (i.e., allocations other than to dfis) would be distributed.

Somewhat analogous to the problem just discussed is the question whether or not dfis should be expected to assume the contingent repayment obligations corresponding to their allocations in the event of withdrawal of a participant from the Special Drawing Account or of liquidation of that Account (Articles XXX and XXXI of the Articles of Agreement). Liquidation is perhaps a contingency too remote to require consideration. As regards a termination of participation, however, since the lending carried on by dfis on the basis of “link” resources would itself be of a revolving, albeit long-term character, it might be possible for the dfis to accept a contingent repayment obligation especially if the general resources of the institutions in question were available to meet such repayment. It would, however, be possible for this contingent repayment obligation to be transferred to participants or to developed country participants in much the same ways as the interest burden discussed above, or the acceptance obligations associated with their allocations.

Several possibilities arise with respect to the manner in which dfis might use SDRs allocated to them. Thus they might be required to use the SDRs immediately on allocation, by converting them into currencies which they would then hold; or they might retain the SDRs until such time as payment had to be made to supplying countries or to countries of investment in connection with development projects or program loans. SDRs might be transferred or converted through the ordinary designation mechanism or in accordance with a special set of rules; there might also be scope for transfers without designation.

TYPE B Schemes—Allotment to Developing Countries of a Higher Share of SDR Allocations than Corresponds to Their Share in Fund Quotas

There are two main variants of this type of scheme. Under the first variant, member countries would be divided into two categories (“developed” and “developing”) or into three or more categories based on degree of development or per capita income: within categories the ratio of allocations to quotas (in any one allocation period) would be uniform, while between or among categories the ratio would differ, being higher for the less developed and lower for the more developed countries.

Under this approach there would be a fairly obvious demarcation between the part of the total allocation to developing countries which corresponds to their share in the world need for reserve supplementation and the part which was intended to be used for development finance. As total allocations varied over time, an attempt might be made to keep the “developmental” part of the allocation at a relatively steady, or steadily growing, level.

Since the developing countries would not be expected to retain the developmental part of their allocation, but rather to spend them on goods and services, it would seem natural that the acceptance obligations corresponding to this part of their allocations should not be assumed by them but distributed in one or another of the ways suggested on pages 73 and 76 above for comparable obligations arising from allocations to dfis.

The interest burden corresponding to the developmental part of the allocations to developing countries could also be handled in one or another of the ways outlined on page 76 above in connection with allocations to dfis. Thus the burden could either remain with developing countries or be transferred to countries in the more developed categories. In the former case, the scheme would in effect be providing financing in the form of perpetual loans, in the latter case grant financing, to the developing countries. Intermediate solutions are, of course, possible, under which the developing countries would pay interest on that portion of their allocations at an especially low rate, or under which all net debtors or all recipients of allocations, respectively, would pay a lower rate of interest than that received by net creditors or holders of SDRs, respectively.

As regards the contingent repayment obligation which arises when a participant withdraws from the Special Drawing Account, the case for transferring the obligation with respect to developmental allocations from the recipients to all participants or to developed country participants is stronger with respect to Type B than with respect to Type A schemes since in the former case the ultimate recipients are in effect counting on a perpetual loan or gift and would be taken by surprise if required at short notice to make even a partial repayment.

Developing countries might, or might not, be subject to conditions, particularly with respect to the developmental use of the “additional” part of their allocations.

In the second variant of this type of scheme, the link between allocations and quotas would be entirely severed and the former would be governed by a formula calculated to produce a country distribution of SDRs more favorable than quotas to developing countries. Such a formula might assign a positive weight not only to foreign trade but also to population, or it might assign a negative weight to income per head of population. Under such an approach it would not be immediately obvious which part of the allocations of developing countries should be regarded as intended for developmental use, and if it were desired to transfer to other countries any part of the acceptance obligations, interest burden, or contingent repayment obligations corresponding to developmental allocations or to impose conditions of use with respect to them, it would be necessary to devise a formula or formulas ad hoc for these purposes. It would also be possible to lessen the interest burden of all net use of SDRs or of allocations in general along lines previously discussed.

TYPE C Schemes—Raising of the Share of Developing Countries in Fund Quotas

While the distribution of Fund quotas at any given time could not be explained in terms of any single formula, it is such as to reflect broadly such considerations as the relative economic importance of different countries in the world economy, their relative needs for conditional and unconditional liquidity, their relative ability to assume creditor positions in the General and Special Accounts of the Fund, and their appropriate participation in decision-making processes of the Fund.

It would be possible to revise the formulas that are employed on the occasion of quota revisions in such a way as to increase the share of less developed countries in Fund quotas, either by taking account of population or of some index of underdevelopment such as (lowness of) per capita income, or by classifying countries according to their degree of development and applying different percentage adjustments to the quotas of the different categories.

If this approach were followed it would presumably be with the intention that SDR allocations should continue to be based on quotas as at present.

TYPE D Schemes—National Contributions to Development Financing on the Basis of SDR Allocations

Under schemes of this type some or all of the participants in the Special Drawing Account would conclude an agreement whereby they would undertake to make finance available for development purposes in amounts in some way related to the allocations of SDRs they received from the Account. In the case of many countries such undertakings would presumably be subject to the ability of governments to obtain year by year the necessary parliamentary appropriations. In principle, such contributions, whether in the form of loans or grants, might be made either to dfis (as in Type A schemes) or to developing countries directly (as in Type B schemes). The latter variant, however, offers great practical difficulties, and references below to Type D schemes will generally relate to the former variant.

Loans or grants to dfis might be made in currency or in SDRs or in some combination of the two. If they were made in SDRs, amendment of the present Articles of Agreement of the Fund would be required to permit dfis to receive, hold, and use SDRs. Apart from this, the system of allocations and other features of the SDR system would remain unaffected by the establishment of this indirect kind of “link” with development financing.

The number of possible variants of such an indirect link between SDR allocation and development financing is probably much greater than for a direct link of the type discussed at A above. Most of the variants discussed in that section can be reproduced in an indirect link scheme, though with the latter it might be more difficult than with a direct link to distribute the sacrifice of reserves involved in contributions to dfis differently from the sacrifice of interest. For example, if contributions to the dfis took the form of grants, the participants contributing the reserves would also be those that contributed the interest.

However, there are many options under the indirect link approach which are not open under the direct link. Some of these arise from the fact that the contribution and loans may be made in currency and not, as in the direct link, necessarily in SDRs. Again, it would probably be easier under the Type D than under the Type A approach to accommodate schemes in which only a limited group of countries would take part, or in which participants contributed varying proportions of their allocations, or provided grants and loans in varying proportions and on varying terms, or in which they made commitments to participate as contributors for varying periods of time ahead.

With an indirect link of this type there would be no questions to settle as regards the distribution of acceptance or reconstitution obligations, or as regards the basis for allocations or the payment of interest. Moreover, if contributions to dfis were made in currency, dfis need not become other holders and no questions would therefore arise as regards the modalities for the use of SDRs by the dfis.

TYPE E Schemes—Transfer to Development Financing Institutions of any Excess Income or of any Amortization Accruing to the Fund as a Result of Substitution Operations

In the report on the reform of the international monetary system mention was made of the possibility of setting up a separate reserve substitution facility in the Fund whereby members would be allowed to sell reserve currencies to the Fund in exchange for SDRs specially issued for this purpose. In that event, it is possible that the interest paid to the Fund by the issuers of reserve currencies with respect to the balances of their currencies so acquired by the Fund might exceed the interest the Fund would have to pay on the additional SDRs issued in exchange. It is also conceivable that the issuing countries might gradually amortize outstanding balances held by the Fund on this account.

Fund receipts from either of these sources could conceivably be utilized to provide additional financing for economic development. There is, however, a distinction to be drawn between the two cases. In the first case, the Fund would simply transfer (not allocate) SDRs received as interest in the form of a grant, either to dfis or to developing countries directly. In the second case, the SDRs accruing from amortization would be used to cancel SDRs initially issued in exchange for the reserve currency that is now amortized. Since, however, the amortization payment involves a decline in global reserves, this might be thought to justify an additional allocation, which would then be made to dfis or to developing countries directly. In this event, whether the development financing provided could be regarded as loan or grant would depend on whether the interest cost of the allocation in question was borne by the recipients of the allocation or was taken over by developed countries.

It may be noted that under Type E schemes, resources for development come from the existing supply of reserves, while in all other schemes such resources come from net additions to the stock of reserves.

4. Criteria of Appraisal

The various possible schemes for linking SDRs and development finance have to be evaluated from two standpoints: (1) effects on development of developing countries and (2) effects on the international monetary system and the adjustment process.

From the former standpoint the most important criterion is the amount of additional resources the schemes are likely to provide, which in turn depends on the addition to development resources arising directly from the scheme in question and on the risk of offsetting reductions in other aid flows of comparable quality. Also of great importance, however, is the effect which the scheme may have on the “quality” of official development assistance. This omnibus term includes such aspects as the cost to the developing countries of the resources provided, the extent to which these resources are likely to be effectively directed to development, and the equity of their distribution among developing countries. An aspect which affects both the development objective and the international monetary system arises out of the problem of reconciling the appropriate timing of development assistance with the appropriate timing of reserve creation.

Other aspects bearing on the compatibility of different approaches with the effective working of the international monetary system include the effects on the amount of reserve creation and on the ability to take decisions on global reserve creation on their own merits, the effects on demand and demand management, and the effects on confidence in the SDR as a major reserve asset and on the mechanics of the SDR system. There may also be repercussions on the ease or difficulty of adjustment of international payments disequilibria—a topic related in some degree to the question of how the “burden” of the link is distributed among developed countries.

5. Effects on Real Resources for Development

These effects may be analyzed under three main heads: (a) the gross financial amounts transferred to developing countries; (b) the extent to which these resources are “additional,” i.e., are not offset by a reduction in development assistance in other forms; and (c) the purchasing power of the financial resources in terms of real resources. Consideration of terms under which the resources might be made available are presented in Section 6.

a. Amount of financial resources available through a link

It is impossible to guess at the amounts that might realistically be transferred under any of the schemes described above. But it may be useful to mention several relevant magnitudes and considerations. Type A schemes—those involving direct allocations of SDRs to dfis—serve as a useful starting point.

Clearly the maximum conceivable amount would be derived from such a link if the entire allocation of SDRs were made to dfis. During the first allocation period (1970–72 inclusive) SDRs were allocated to an average amount of SDR 3.10 billion per annum. No one can foresee future allocations, but clearly their amount will depend to some extent not only on the need for reserves but also on the anticipated growth of the supply of reserves in other forms. If, at some future time, through the adoption of arrangements such as are discussed in the report on the reform of the international monetary system, the accumulation of currency reserves could be effectively arrested, the need for SDR allocations might well rise to much larger figures.

Even if the entire allocation of SDRs were directed toward dfis with no country allocations whatever, it cannot be assumed that all of it would go to finance development in developing countries. First, it must be assumed that all or most of the allocation that would normally have been distributed to developing countries is required by them for the purpose of adding to their reserves; this alone might subtract, say, 30 per cent of the allocation from the net addition to development financing. Realistically speaking, one could probably not expect that the developed countries would be prepared to forgo all direct allocations of SDRs and rely entirely on their ability to earn them from additional exports to less developed countries. If the developed countries were to consent to, say, half of their normal allocations being diverted to dfis, the amount of financing accruing to the latter and through them to the developing countries would amount to not much over one third of total allocations.

What has been spoken of so far is the gross allocation to dfis to the extent that it is available for development financing in developing countries. No allowance has yet been made for various possible effects discussed in the following section.

What has been said above of the amounts that might be expected from schemes of Type A holds mutatis mutandis of schemes of Type B, namely those involving direct allocation to developing countries. On assumptions analogous to those made above, the “additional” allocations available to developing countries would amount to just over one third of total allocations.

In schemes of Type C, where quotas rather than the ratio of allocations to quotas would be altered in favor of developing countries, consideration of the implications for voting rights in the Fund and of the need to keep the General Account adequately supplied with creditor currencies would probably result in much smaller amounts of SDRs being made available for development financing than would be possible under schemes of Types A or B.

Schemes of Type D, i.e., those under which countries would agree to provide development assistance, whether to dfis or direct to developing countries, in relation to their SDR allocations, might be expected to yield rather smaller amounts than those of Types A and B, if only because participation in such schemes would probably be less extensive. However, it is conceivable that those who did participate would redirect to development financing a rather larger proportion of their SDR allocations than would be approved in general in Type A or Type B schemes under the decision-making processes of the Fund.

Schemes of Type E, under which any excess of the debt service paid by reserve centers to any reserve substitution facility over the interest cost paid by that facility would be made available for development financing, would almost certainly yield less than schemes of the types considered above, with which it might, however, be combined. Total U.S. and U.K. liabilities to official holders in dollars and sterling, respectively, amounted at end-1971 to some SDR 54 billion,10 and the proportion of these that the holding countries would be prepared to replace with SDRs under any conceivable reserve substitution arrangement is very uncertain. Even if the rate of interest payable to the Fund on balances held by it under the reserve substitution facility were to exceed that paid by the Fund on the SDR balances issued in exchange by as much as 1 per cent per annum, the amount available for development financing on account of the interest differential could hardly be large. Nevertheless, the fact that such amounts would be free of interest or repayment obligations might enhance the attractiveness of Type E schemes. If the reserve centers were to amortize their liabilities to the substitution facility, and these amortization payments became available for use in development, additional amounts would, of course, be yielded. These amounts when used might be subject to interest.

It is difficult to gauge what amounts any extended Fund facilities, involving issuance of SDRs to finance longer-term stabilization and perhaps other Fund programs for the benefit of the developing countries, would be likely to provide. These would vary from year to year depending on the number and size of eligible programs. In light of past experience, it might be estimated that longer-term stabilization programs might absorb amounts on the order of SDR 250-500 million in some years. It should further be noted that extended Fund facilities could be considered as a supplement to other types of schemes, rather than as an alternative.

It is interesting to compare the orders of magnitude of gross assistance derivable from a “link” with other magnitudes relative to the amount of development financing currently being provided to developing countries, and estimates as to the need for such financing. For example, the total flow of financial resources from Development Assistance Committee (dac) countries11 as a group to developing countries and multilateral agencies in 1971, net of amortization on earlier lending, was $18.3 billion. The private component, basically direct and portfolio investment, long-term lending, and export credits accounted for $8.4 billion. Official capital flows summed to $9.0 billion, of which $7.4 billion went to countries directly and $1.6 billion to multilateral institutions. Official development assistance more narrowly defined came to $7.7 billion, of which $1.3 billion was contributed to multilateral institutions. Grants of private voluntary agencies provided an additional $900 million. Comparative figures for earlier years are given in Table 2.12

Table 2.Flow of Financial Resources from Development Assistance Committee Countries to Less Developed Countries and Multilateral Agencies, 1969–70(In billions of U.S. dollars)
Net Disbursements1
19671968196919701971
Total Official and Private, net (I + II + III)11.413.413.715.9218.32
Total Official, net (I + II)7.17.07.28.09.0
I. Official Development Assistance, net (A + B + C)6.66.36.66.87.7
A. Bilateral grants and grant-like contributions3.63.33.33.33.6
B. Bilateral development lending and capital investments, net2.22.32.32.42.8
C. Contributions to multilateral institutions, net0.70.71.01.11.3
II. Other official flows, net (A + B)0.50.70.61.11.3
A. Bilateral30.50.70.60.91.0
B. Contributions to multilateral institutions, at market term0.30.3
III. Private flows at market terms, net4.46.46.67.08.4
IV. Grants by voluntary agencies, net(0.6)0.90.9
Total as percentage of gross national product0.740.800.750.800.83
Official development assistance as percentage of gross national product0.420.380.360.340.35
Source: Organization for Economic Cooperation and Development, Development Assistance Committee.

Gross disbursements minus amortization receipts on earlier lending.

Including grants by voluntary agencies.

Including official export credit, debt relief, and purchases of equities and bilateral assets.

Source: Organization for Economic Cooperation and Development, Development Assistance Committee.

Gross disbursements minus amortization receipts on earlier lending.

Including grants by voluntary agencies.

Including official export credit, debt relief, and purchases of equities and bilateral assets.

The data in Table 2 include flows to oil producing countries, and developing countries in Western Europe, which are much further along in the development process than most of the other developing countries. Over the 1968-70 period, however, almost 89 per cent of the net private and official receipts of developing countries from dac countries and multilateral agencies, and 96 per cent of the official receipts, went to non-oil-producing developing countries outside Europe.

The figure of $7.7 billion cited above for official development assistance from dac countries amounted to some 0.35 per cent of dac countries’ gross national product. As against this, the International Development Strategy for the Second Development Decade adopted by the General Assembly of the United Nations called for each economically advanced country to increase its official development assistance to a minimum net amount of 0.7 per cent of its gross national product by the middle of the 1970s.13 In 1971 values, this target would amount to some $15.5 billion.

b. “Additionality” of funds supplied by link schemes

If any scheme were established for providing development finance in connection with SDR allocations, it would be on the understanding that this should not be regarded as justifying any reduction in direct national assistance, present or planned, to developing countries. Nevertheless, it is possible that some decline in other flows to developing countries would in fact occur and, other things being equal, the greater such offsets the less would be the value of a link scheme.

So far as government aid is concerned, there is some reason to hope that in the case of most types of schemes, at any rate, offsetting reductions in national aid would be likely to fall significantly short of the amount of financial assistance provided under the link. In the first place, most governments have reasons for providing aid to specific recipients or for specific purposes over and above the general desire to provide development assistance to poorer nations, and these reasons would continue to operate. Second, the provision of additional aid in a “link” form might reasonably be attractive to governments and parliaments in that their own participation in such a scheme would tend to encourage the participation of other countries as well and thus exercise a multiplier effect. Third, the extent to which any country might be regarded as providing aid through the link would be difficult to measure (except for Type D schemes) and in some cases almost indefinable. On the other hand, particularly during the initial period of adjustment to the link, some developed countries might be disposed to cut the amount of the assistance they provide either because of difficulty in earning SDRs which they would no longer receive in allocations or because they feared an inflationary effect of earning these SDRs by way of additional exports.

The extended Fund facilities would be unlikely to provoke offsetting adjustments in national aid programs. As between the five different categories of “link” arrangements, schemes of Type D, where countries themselves provide development assistance on the basis of the SDR allocations they receive, would be the most likely to evoke offsetting reductions in other forms of national aid. This would be true even on the assumption that the assistance provided by countries in relation to their SDR allocations was channeled through dfis; it would be even more true if the assistance were provided to developing countries directly. It is difficult to compare Type A schemes, on the one hand, and Types B and C schemes, on the other, from the standpoint of additionality. Insofar as developed countries believe that the dfis exercise a beneficial effect on the use made of development assistance, they may be less disposed to cut their direct aid on account of link aid if the latter is routed via the dfis. The contrary may be true, however, if there were established notions as to the amount of business which dfis could usefully handle. This would give the advantage to Type B schemes, particularly those following under the second variant where the amount of aid attributable to the link would be difficult to measure. As regards Type E schemes, the main danger of offsetting reductions arises from the possibility that the countries that were paying interest and amortization on the balances held by the Fund might consider that they were financing the linked development assistance.

To the extent that the establishment of an arrangement linking SDR allocations with development financing resulted in a net increase in official (national and international) assistance on concessional terms, it might give rise to yet another partial offset in the form of a decline in private capital flows, either because dfis, having more funds from official sources, had less need to borrow on private markets or because developing countries, receiving more official assistance, directly or via dfis, borrowed less from private investors and lenders. This particular offset, however, might well turn out to be negligible, or nonexistent, if the strengthening of economic infrastructure of the recipient countries by official funds opened up, as it often does, more attractive investment opportunities for private capital.

c. Purchasing power of financial resources

All financial resources made available to developing countries under extended Fund facilities and link schemes of Types A, B, C, and E, would presumably be available for expenditure in any country, without “tying” provisions of any kind; one can be less sanguine that the same would be true of schemes of Type D.

The purchasing power in terms of importable goods of untied financial resources provided through the link might be significantly higher than is often the case with the aid provided by national governments. Generally aid-tying results in the recipient country purchasing from the donor country goods and services at higher prices than it would normally pay through international bidding. Haq, in one of the earliest studies on the cost of aid-tying,14 stated that for “Pakistan the weighted average price of … 20 projects” came “out to be 51 per cent higher from the tied sources compared to the international bids,” (p. 327) and for the total aid effort, aid-tying raised the average price of procurement, for Pakistan, by about 12 per cent (p. 331). Studies of other countries’ aid experience place the excess cost of tied aid to the recipient at between 12 and 24 per cent.15

Thus, even in the unlikely contingency that aid provided under link schemes were fully offset by reductions in national aid, the real purchasing power of total aid would be likely to increase.

6. Other Dimensions of Aid

Given the volume of aid provided under any arrangement linking SDR allocations to development finance, its value to the developing countries may nevertheless vary according to (a) the terms on which it is provided, (b) the extent to which it is put to productive use, and (c) the manner in which it is distributed among developing countries.

a. Terms of link assistance

In considering the value of financial assistance given to developing countries, the interest cost of the assistance provided is of comparable importance with the amount. Indeed, it is only if the interest charged is below the economic and social return on the investments financed and below the marginal cost at which the money could be raised on the private capital market—if it could be raised at all—that additional official assistance has any value.

Under schemes (e.g., of Type B) in which additional allocations are made to developing countries, the interest cost to the recipients would not exceed the interest payable on the use of SDRs. This is currently 1½ per cent but is not unlikely to rise as the proportion of SDRs in world reserves increases. Provided that the currency value of SDRs is reasonably well maintained, however, it may be assumed that the interest rate on what amounts to a perpetual loan in SDRs can remain somewhat below the interest rate on short-term assets in reserve currencies, and would thus be substantially below the interest rates of some 7¼ per cent to 8 per cent presently paid by developing countries on the harder type loans made by dfis (cf. Table 1).

In schemes of Type A, in which allocations are made to dfis who then use the proceeds for development lending, it is to be expected that, to cover administrative costs, the interest charged to developing countries on loans financed by SDR allocations would be slightly (say, ¾ of 1 per cent) higher than the interest payable by the dfis. This probably does not constitute a genuine difference from the charges on direct additional allocations to developing countries under Type B schemes since analogous costs (e.g., of finding and appraising projects) would in that case have to be met by the recipient countries themselves. It may be objected that dfis do not give perpetual loans, so that the debt service burden to be met by developing countries would be higher, on account of repayment, in the case of Type A than in the case of Type B financing. Once again, however, the difference is misleading, dfis can relend reimbursable loans again and again as reimbursement is made, so that, provided that all interest rates remain constant, the cumulative grant content of successive loans will be no less than in the case of a perpetual loan.

In what has been said above, it has been tacitly assumed that dfis would use SDRs allocated to them to finance a special category of development loans. This might be inconvenient for institutions such as the ibrd/ida, where ibrd-type lending is at a higher rate and ida-type lending at a lower rate than might correspond to the rate of interest on SDRs. However, if it were felt necessary to exercise a strict accounting over the use made by the dfis of link funds or to ensure that their distribution among recipient countries met certain tests, this would constitute an argument for using them to finance a separate and distinguishable category of development loans.

As was pointed out on pages 73 and 76 above, there are a variety of ways in which dfis (under Type A schemes) and developing countries (under Type B schemes) can be relieved of part or all of the interest burden on “link” assistance. Techniques which rely on lowering interest on net use of SDRs by developing countries or by all participants have the drawback of depriving the countries in question of the appropriate incentive to hold SDRs. Of the techniques which involve giving interest relief on cumulative allocations, those which confine that relief most narrowly—e.g., to allocations to dfis under Type A schemes and to the “developmental” part of the allocations to developing countries under Type B schemes—have the advantage of narrowing the amount of gross transfer required to achieve a given net transfer from developed to developing countries. Such techniques, however, are not easily applicable to schemes of Type B of the second variant. As regards methods of financing the interest relief, those which place the burden squarely on developed countries, in proportion, say, to Fund quotas, rather than on all participants, have the merit of increasing gross transfers but involve the risk that these interest transfers will be counted against the aid efforts of the developed countries. The idea of meeting the SDR cost of the transfer by reducing pro tanto new issues of SDRs incurs the objection that it puts a good deal of the burden back on to the less developed countries, who would otherwise have received higher allocations, especially under Type B schemes. Moreover, the interest relief given to developing countries, in whatever form, involves a progressive discrepancy between holdings of SDRs and cumulative allocations, and a progressive decline in the proportion of new reserves in the form of SDRs that come from new allocations.

b. The productivity of the resources used

Here, apart from the question of distribution between countries discussed in the following subsection, there are two elements to be taken into account. The first is the extent to which the resources provided to the developing countries would in fact be used for development purposes. The second considers the soundness of the additional investment projects undertaken by the recipient countries.

In view of the mutual substitutability of resources of domestic and foreign origin, and in view of the impossibility of determining how a country would have used its resources in the absence of development assistance, it is not entirely possible to ensure that all the additional resources provided would in fact be used for development purposes rather than for public or private consumption or the improvement of the country’s external position. Nor is it even entirely desirable or economic that a country should use all of any additions to its resources for additional development expenditure. Nevertheless, the rationale given for the link scheme is primarily that it will speed up the development of the recipient countries, and countries that forgo the SDR allocations that they might otherwise have expected, or bear the charges with respect to allocations provided for development, will want some assurance that the great bulk of the funds provided are used for development purposes. From this standpoint, extended Fund facilities and schemes which provide resources in the first instance to financial institutions have a certain advantage over schemes (e.g., of Types B and C) which involve direct transfers to developing countries. It might be possible to strengthen Type B schemes from this standpoint if recipients were obliged to give certain assurances to the Fund regarding the development use to be made of the allocations provided. However, it would be difficult to give economic content to such assurances, without extensive understandings on broad economic policies.

As regards the soundness of individual projects also, it will probably be felt to be an advantage of Type A schemes that the funds provided are directed toward projects that have been vetted not only by the recipient countries but also by the dfis, some of which have accumulated considerable experience in project evaluation.

c. Distribution of assistance among recipients

It would seem to be consonant with the purposes of any link arrangement, and with the circumstances conditioning it, that in determining the appropriate distribution of the funds provided among developing countries account should be taken of three principal considerations:

(i) the relative stage of development of the countries in question;

(ii) the availability of opportunities for productive investment and the likelihood that these opportunities will be exploited; and

(iii) the extent, if any, to which developing countries themselves are forgoing allocations or bearing interest costs in connection with the scheme.

The first two considerations are self-explanatory; the third consideration is the practical one that if, for example, in the case of a Type A scheme, the residual national allocations represent a lower proportion of quota for the more advanced than for the less advanced developing countries, the former would expect to receive a higher proportion of the development loans financed by SDR allocations to dfis than might naturally follow from considerations (i) and (ii) above, or than they would have expected to receive if the national allocations of all developing countries had represented the same percentage of quota.

Under Type B schemes it would be rather a simple matter, in principle, to combine considerations (i) and (iii) by providing that all developing countries received a higher proportion of allocations to quotas than developed countries; and the poorer developing countries a higher proportion than the more advanced ones. The number of categories into which developing countries could be divided for this purpose would of course be limited by statistical inadequacies and ambiguities.

In the case of Type A schemes where the aid is channeled through dfis, it would be difficult, even if it were desirable, to hew to any specified distributional scheme by countries. Not only would strict adherence to such a scheme prevent the dfis from paying due regard to consideration (ii) above, but it would also make it more difficult for them to insist on appropriate criteria for financing investment projects or development programs. Such established practices as an upper limit on the national income per head of recipients of ida loans, while it represents an effort to give due weight to consideration (i) above, would be difficult to reconcile with consideration (iii). Some of these difficulties could no doubt be overcome by cooperative arrangements among dfis to administer a special category of SDR-financed development loans.

Type B or Type C schemes thus appear to have a considerable natural advantage over Type A schemes from the standpoint of achieving a specific desired distribution of assistance, in accordance with considerations (i) and (iii). The latter, however, have a natural advantage with respect to (ii) since the dfis, especially if they act in concert, are well placed to judge the availability of investment opportunities.

A possible solution to the distributional problem could be a mixed system under which part of the SDRs available for link purposes would be distributed to developing countries directly in the form of additional allocations, while the rest would go to dfis to be distributed at their discretion.

7. The Timing Problem

In considering any link between SDR allocation and development finance, a question arises as to how far variations in such allocations, as determined by the global need to supplement reserves, could be absorbed by dfis or developing countries without detriment to development, and as to what could be done under various schemes to mitigate the effects of such variations.

The first part of this question is very difficult to answer at this point in the discussion of the international monetary reform. Whether SDR allocations can expand steadily in response to the long-term growth in the global need for reserves or, on the contrary, must vary abruptly from one allocation period to the next depends on whether the aggregate amount and growth of official reserve holdings in the form of currencies can be brought under a sufficient degree of international control and on whether credit-linked reserve growth can be kept within due bounds.

There are various ways in which these difficulties could be overcome, by adjustments on the part both of the Fund and of the dfis or developing countries.

As has already been indicated, under any type of “link” scheme other than Type C and possibly the second variant of Type B, it would be feasible to vary the proportion of SDR allocations directed to development financing in such a way that the amount assigned to such financing would develop differently from, and in general more steadily than, the amount of allocations as a whole. For example, provisions in the Articles of Agreement relating to the link might be such as to make it possible to change the proportion if need be on the occasion of each decision to allocate. The stability of developmental allocations—to dfis in the case of Type A schemes, or to developing countries in the case of Type B schemes—might be still further safeguarded if such allocations were made for periods longer than the normal allocations to countries on a quota basis.

There are, no doubt, limits to the extent to which stabilization of the amounts of SDRs assigned to development finance—at the cost of destabilizing the amounts allocated to countries on a quota basis—could be achieved. For example, if it were judged that total allocations in any period should be abnormally low, it might be impossible to maintain allocations to dfis (in Type A schemes) or “additional” allocations to developing countries (in Type B schemes) without asking developed country participants to accept some cancellation of previous allocations. Though such a situation should arise only when reserves in the world were deemed relatively plentiful, it might be very difficult to secure consent to such a step. In such circumstances, there would be a tendency either to squeeze developmental allocations or, if these had been predetermined, to expand total allocations beyond the point that would be justified on grounds of global reserve need. A general difficulty about all attempts to stabilize developmental allocations by fixing the amounts for long periods ahead is that prudence might tend to set these amounts at a figure which in the light of the secular expansion in the world economy would turn out to be low.

Some contribution to bridging any timing discrepancies could also be expected on the part of the immediate recipients of the development-linked SDRs, whether dfis in the case of Type A schemes or developing countries themselves in the case of Type B or Type C schemes. These recipients could accumulate reserves during periods whose allocations for development temporarily outstripped development expenditures, and use them during periods when development expenditures exceeded allocations. Such a buffer stocking of liquid assets, however, is not without difficulty in itself, and not without disadvantages from the standpoint of the supply of world reserves.

So far as actual development expenditures are concerned, the difficulties and the disadvantages occasioned by irregular allocations are probably greater in the short run for Type B than for some Type A schemes. The dfis are accustomed to holding large sums in liquid balances pending disbursement, and would be able to maintain the flow of disbursements associated with past commitments despite the variation in the receipts from allocations to them and from other sources. New commitments admittedly would be more severely affected by irregularity in allocations, though even here the dfis might be able to maintain a reasonable flow by calling on reserves or market borrowings. Developing countries, on the other hand, might be tempted to spend abnormally high developmental allocations on less valuable projects or even to finance expansion in consumption, and might have greater difficulty in obtaining finance from other sources when developmental allocations were low. This advantage of Type A schemes, however, would apply much less to schemes under which reserves derived from the “link” are segregated from other reserves and used by the dfis for a specific type of lending.

There is another kind of difficulty in harmonizing the timing of reserve creation and of development assistance which is specific to Type A schemes, namely that which arises from variations in the amount of SDRs that may be held by dfis at any time pending their disbursement. There are two techniques whereby dfis might be able to avoid repercussions on world reserves from such hoarding of SDRs during certain periods. One is for the dfis to use SDRs immediately on receipt and to hold the proceeds, pending their disbursement, in a variety of currencies.16 Alternatively, the Fund might regard SDRs held by the dfis as not forming part of the world reserve stock and, in determining the amount of new allocations, might look upon expected SDR disbursements by dfis rather than allocations made to them as constituting an addition to world reserves. Neither of these techniques could easily be applied in the case of direct developmental allocations to developing countries.

How far the various techniques described would be successful in overcoming the problems arising from variability in SDR allocations would depend to a considerable extent on the amount of link financing relative to other forms of financing. If dfis were financed entirely by the link, they might find it difficult to operate smoothly unless they could count on a steady stream of allocations. Also, if dfi financing absorbed too large a proportion of the amount of SDRs created in most years, it would be more difficult to maintain a steady stream of allocations when the need for reserves fluctuated. These considerations argue in favor of avoiding an amount of link financing which would constitute an excessive proportion either of total dfi financing or of the expected normal flow of reserve creation by means of SDRs.

8. Effects on the International Monetary System and the Adjustment Process

Under this head may be considered the effects of the link on (a) aggregate demand in the world as a whole, (b) decisions to allocate SDRs, (c) general confidence in SDRs and the willingness of countries to participate in the SDR system and in allocations of SDRs, (d) payments imbalances and their adjustment, and (e) the real income of developed countries.

a. Effects on aggregate demand

Normally, new allocations of SDRs are added to the reserves without entering, in the first instance, into the income stream of the recipient countries. They may affect the demand management policies of these countries by encouraging demand expansion or by mitigating the contraction that might otherwise result from balance of payments difficulties. It is, however, the intention of all “link” arrangments that amounts allocated for development purposes should, within a brief period, enter into development spending, and thus into the income stream either of the developing countries themselves or of the developed countries which supply their demands for imports. Thus, the mechanical effect of a link is more expansionary than the allocation of the same amount of SDRs to countries across the board. “Link” aid may also be more expansionary than national aid appropriated as part of budgetary expenditure insofar as countries operate on budget balancing principles, rather than use the budget to achieve a desired level of aggregate demand in their economies.

In national economies, a given increase in government expenditure is more expansionary than a tax reduction of the same amount, but a government’s choice between increasing expenditure and reducing taxation is not always made in favor of the latter, or between decreasing expenditure and increasing taxation always in favor of the former. Rather, these decisions are made on the basis of consideration both of the cyclical situation and of the importance attached to public expenditure (including the provision of resources to underprivileged individuals) as against private expenditure.

By analogy, in the international context, it would be difficult to make decisions on link financing depend from year to year on a judgment of the cyclical situation in the world, from which it does not follow, however, that that situation will at all times be tending toward excessive, rather than deficient, demand. Some sort of balance has to be kept between international objectives (including the provision of resources to low-income countries) and the needs of individual countries.

The formulation of adequate demand policies has to be carried out, in the last resort, on a national basis. These policies, in each country, have to take account not only of demands on national output arising from decisions taken in that country, such as decisions on government expenditure and domestic investment, but also of demands, including the demand for exports, arising out of decisions abroad. To be successful, demand management policies cannot be limited to budget balancing but must be responsive to expenditures arising from different sources. Among these, expenditures attributable to any link financing could add but little to the problem of demand management with which governments, in any event, must deal. The amounts potentially involved, though varying somewhat according to country, would in general constitute an infinitesimal part of the aggregate demand for the products of the countries concerned: for example, a developmental allocation of $1 billion, spent in its entirety on the exports of developed countries, would be unlikely to add more than 1 per mille to the demand for the output of any one of them. (See Table 3.)

Table 3.Exports and Gross National Product of Developed Countries

(In billions of U.S. dollars, unless otherwise indicated)

Developed

Countries*
1971Additional

Exports2
ExportsAdditional

Exports2
ExportsGNP1(As percentage of GNP)
United States44.11,050.4.2644.20.03
Canada17.792.2.01919.30.02
Japan24.0224.1.19410.70.09
Australia5.135.4.01414.40.04
South Africa2.219.0.01011.60.05
Belgium-Luxembourg12.330.1.02440.90.08
France20.5162.8.09312.60.06
Germany, Fed. Rep. of39.0216.4.12618.00.06
Italy15.1101.5.06114.90.06
Netherlands14.036.2.02538.70.07
Austria3.116.7.00918.60.05
Denmark3.619.3.00718.70.04
Finland2.411.4.00721.10.06
Iceland0.20.6.00025.1
Norway2.614.3.00718.20.05
Sweden7.435.6.02120.80.06
United Kingdom22.4135.0.12016.60.09
Total235.72,201.01.010.70.05

Part I Member Countries of the International Development Association (countries that are more advanced economically and contribute to the resources of the ida) that participate in the Fund’s Special Drawing Account.

In certain cases, gross domestic product.

Per $1 billion “link;” see Table 4.

Part I Member Countries of the International Development Association (countries that are more advanced economically and contribute to the resources of the ida) that participate in the Fund’s Special Drawing Account.

In certain cases, gross domestic product.

Per $1 billion “link;” see Table 4.

b. Effects on decisions to allocate SDRs

Article XXIV, Section 1 of the Articles of Agreement lays down the principles and considerations which should govern all Fund decisions with respect to the allocation and cancellation of special drawing rights, and specifies that in all such decisions “the Fund shall seek to meet the long-term global need, as and when it arises, to supplement existing reserve assets in such manner as will promote the attainment of its purposes and will avoid economic stagnation and deflation as well as excess demand and inflation in the world.”

Among the purposes referred to—those specified in Article I—is that of contributing to the development of the productive resources of all members, but it is clear from the context of Article I(ii) that this purpose is to be pursued by the Fund not directly through the provision of development finance but indirectly by facilitating the expansion and balanced growth of international trade.

Advocates of a link between SDR creation and development financing have generally emphasized the principle that, even after the establishment of such a link, the allocation of SDRs should continue to be determined in the light of the global need for reserves and not the need for development financing. It is assumed that this would remain the basic principle guiding the amount of reserve creation. Nevertheless, as a realistic matter, one should assess the risk that, if any link were established, its existence might influence those responsible for arriving at decisions to allocate SDRs in a manner not entirely consistent with this principle.

The existence of a link would tend to give developing countries an additional interest in larger allocations of SDRs, especially in schemes where developmental allocations constituted a constant proportion of the total: such countries, however, generally tend to favor large allocations even in the absence of a link. Any bias introduced into decisions to allocate by the existence of a link is likely to arise out of its effect on developed countries. In this respect a variety of considerations arise and the net effect, if any, which the existence of a link may have on the attitudes of developed countries toward SDR allocations is ambiguous and uncertain. Factors militating in favor of larger allocations could be, first, the desire of developed countries not to hamper the functioning of development agencies or to disappoint the expectations of developing countries and, second, the uncertainty of each developed country, under a “link” arrangement, about precisely how much of any allocation would ultimately accrue to it through additional exports. But the fact that, under a “link,” any addition to total allocations would involve an additional transfer of real resources and an addition, however slight, to inflationary pressure would normally provide an incentive in the opposite sense. Much here might depend on the nature of the scheme in question. For a given amount of link finance, schemes in which the amounts assigned to such finance were not closely geared to total allocations, and those in which the countries receiving such finance bore all or most of the interest cost involved in its use, would be less likely to predispose the developed countries to an overconservative view—or more likely to predispose them to an overliberal one—regarding the amount of total allocations.

Whatever may be the weight of these various considerations, it has to be recognized that the present method of bringing SDRs into countries’ reserves is not free from pressures either; the issue to be judged, therefore, is whether any method involving a “link” would be open to greater and more dangerous pressures than the existing one.

c. Effects on “confidence” in SDRs

“Confidence” in the SDR is a complex conception which comprises such elements as willingness to become a participant in the Special Drawing Account, willingness to support allocations when these are justified by considerations of international liquidity, willingness to accept allocations rather than to “opt out” under Article XXIV, Section 2(e), and willingness to accept SDRs and to hold them. Such confidence is of great importance if the SDR is to become the principal reserve asset in the international monetary system of the future.

Considerable importance in this connection may attach to the repercussions which a “link” scheme would have on the provisions of the Special Drawing Account with respect to interest rates. It has already been suggested that an excessive transfer of the interest burden associated with developmental allocations to the developed countries might make the latter unduly conservative in their appraisal of the global need for SDR allocations in general. It also seems possible that if participation in the Special Drawing Account involved for developed countries obligations with respect to development assistance (such as additional acceptance obligations, contingent repayment obligations in the event of the withdrawal of another participant, and the assumption of interest costs) that were large in relation to direct allocations to the countries in question, such participation might become less attractive both to present participants and to countries now nonparticipants, and might induce a larger use of the “opting-out” provisions if these provided an opportunity to avoid such obligations. Clearly this question arises mainly with respect to Type A and Type B schemes, and not with respect to Type D schemes in which participants in the Special Drawing Account may, if they wish, refuse to join.

Further effects might derive from any arrangements to impose the additional interest burdens arising out of the link on the developed countries. If this were done in proportion to the current allocations of these countries, there might be a considerable incentive for them to opt out of such allocations. If, as is perhaps more probable, these burdens were imposed in proportion to cumulative allocations, the incentive to opt out would be substantially weaker, while if they were imposed in proportion to quotas there would be no incentive to opt out at all. On the other hand, some of the arrangements which provide a relatively weak incentive to “opt out” of individual allocations provide a relatively strong incentive to countries to withdraw from participation altogether.

It is difficult to gauge the strength of such incentives. In order to eliminate all incentive for developed countries either to opt out of individual allocations or to withdraw from participation, it would be necessary either to provide that the interest burden would fall only on those countries that had undertaken to accept it—which would give schemes of Types A and B much of the character of Type D schemes—or else to refrain from imposing on developed countries any of the burden of charges arising with respect to the allocations for development purposes, in which case the developing countries would be receiving assistance in the form of (fairly cheap) loans rather than grants. Perhaps the most plausible conclusion is that in any “link” scheme that might be adopted the extent to which the charges on the allocations would be assumed by developed countries—and this, as already pointed out, is a matter of degree—would be such as to leave little likelihood that either opting out or withdrawal of participation would occur to any significant extent.

It is sometimes argued that the creation of SDRs by development spending rather than by allocation would reduce the willingness of countries to hold SDRs. If so it would be detrimental to all proposals for international monetary reform that involve a voluntary substitution of SDRs for currencies or gold in reserves. It would also tend to increase the propensity to “opt out” of SDR allocations, and might even reduce the amount of such allocations. It is, however, difficult to see why the mere fact that SDRs were put into circulation by way of development financing should in itself impair the quality of the SDR as an asset. Any deterioration in the “backing” of the asset in the event of the holding country’s withdrawal from the Special Drawing Account would be slight if the arrangements suggested at page 76 above were adopted. The main danger of unwillingness to hold SDRs might arise if the establishment of a “link” were to have the consequence that interest rates on SDR holdings were kept unduly low in order to avoid raising the charges on allocations, including developmental allocations.

d. Effects on the adjustment of payments disequilibria

The institution of a “link” scheme and any subsequent expansion of the amounts transferred under a “link,” insofar as it was not accompanied by offsetting changes in other forms of financial assistance, might be expected to produce temporary effects on the balances of payments of the countries concerned, leading to possibly longer-lasting effects on the level of their reserves.

So far as the developing countries are concerned, the effects would in most cases be such as to alleviate, at least temporarily, their payments difficulties, and to assist them in reducing the level of their restrictions and raising the level of their reserves. The last-mentioned effect might be somewhat greater under Type B than under Type A schemes. In the absence of knowledge about how “link” aid would be distributed, it is not possible to say much more about the balance of payments effects on these countries.

As regards developed countries which would be called upon to make a net contribution to aid through the “link,” it may be assumed that there would be a small and temporary negative influence on their balance of payments as a group corresponding to any increase in the reserve holdings of the developing countries. The major short-run effect on the reserves of individual developed countries, however, might be expected to arise from discrepancies between the proportions in which the several countries would forfeit direct SDR allocations and the proportions in which they would experience an additional demand for their exports. On the assumptions that the developed countries listed in Table 4 would forgo direct allocations in proportion to quota, would achieve in the aggregate an equal amount of additional exports to other countries, and would achieve individually additional exports to other countries in proportion to their existing exports to the latter, the outcome might be along the lines suggested in that table. The actual outcome would probably differ from that shown in Table 4, first because of changes in trade shares that have occurred since 1969, second because of the difference between the composition of marginal and of total demand in developing countries, and third because there might be some tendency for marginal demand for exports to be met more than proportionally from sources where there is ample excess productive capacity.

Table 4.Forgone SDR Allocation and Additional Exports Provided by Developed Countries Through a Link(In millions of U.S. dollars)
Developed Countries1Forgone

SDRs2

(1)
Additional

Exports3

(2)
Column (2) less

Column (1)

(3)
United States343264–79
Canada5619–37
Japan61194133
Australia3414–20
South Africa1610–6
Belgium-Luxembourg3424–10
France779215
Germany, Fed. Rep. of8212644
Italy516110
Netherlands3625–11
Austria149–5
Denmark137–6
Finland107–3
Iceland10–1
Norway127–5
Sweden17214
United Kingdom143120–23
Total1,0001,0000

Part I Member Countries of the International Development Association (countries that are more advanced economically and contribute to the resources of ida) which participate in the Special Drawing Account.

Direct allocations forgone in order to provide the equivalent of $1,000 million for a “link,” in equal proportion to quotas of January 1, 1972.

Calculated on the basis of shares in 1969 exports by country. It is further assumed that only the developed countries listed above would accumulate reserves arising from link-financed trade and that all other countries would offset any additional exports by additional imports.

Part I Member Countries of the International Development Association (countries that are more advanced economically and contribute to the resources of ida) which participate in the Special Drawing Account.

Direct allocations forgone in order to provide the equivalent of $1,000 million for a “link,” in equal proportion to quotas of January 1, 1972.

Calculated on the basis of shares in 1969 exports by country. It is further assumed that only the developed countries listed above would accumulate reserves arising from link-financed trade and that all other countries would offset any additional exports by additional imports.

Short-term balance of payments effects on developed countries might also be complicated by capital flows from countries where there was plenty of excess capacity to meet the additional export demand to countries where there was little excess capacity.

One further observation needs to be made when “link” schemes are considered in the context of the adjustment process. The effect of the introduction of such schemes is that reserve increases in developed countries will to a larger extent take place through current account surpluses and less through SDR allocations to them. Insofar as these countries tend to pursue current account, rather than reserve, aims in their balance of payments policies, the channeling of reserve creation through link schemes could help to relieve tensions in the adjustment process.

e. Magnitude and distribution of any “burden” of the “link”

The question examined in this section is how real income losses (if any) incurred by developed countries as a result of the “link” might be distributed among these countries after any necessary balance of payments adjustments had been carried out. It will be assumed that these adjustments would be such that each country that forwent SDR allocations as a result of the “link” would earn a corresponding amount of reserves in the form of SDRs through a surplus in the balance of payments.17

In these circumstances, link schemes—provided they did not involve the assumption by developed countries of any part of the interest cost of the amounts made available under such schemes—would affect developed countries adversely only to the extent that the interest they received on the additional SDRs they had to earn fell short of the loss of income from real investment forgone (to make room for a more positive current account) plus any net loss of interest income from abroad that may result from an adjustment of the capital account forming part of the total adjustment to the “link.” If the rate of return on real investment, the rate of interest on financial capital moving between developed countries, and the rate of interest on SDR holdings were all equal, the link would entail no burden on industrial countries, though it might still benefit developing countries where the return on capital was higher. If the rate of return on real capital in each country were equal to the international interest rate on financial capital and both were in excess of the rate on SDRs, there would be a “link” burden. The magnitude of this burden would equal the amount transferred under the link, multiplied by a factor below unity reflecting the proportional difference between the interest rate on SDRs and that on the other assets. This burden would be distributed among developed countries in proportion to the amount of SDR allocations forgone, i.e., in proportion to quotas.18 If, by contrast, the interest rate on SDR holdings were equal to the rate on international financial capital and both were below the return on real capital, and if the latter were the same for all developed countries, the “link” burden would be distributed in proportion to the amount of real capital forgone; this, in a world of internationally mobile capital, might be a distribution roughly proportional to gross national product.

In reality, the return on real capital will usually be higher than the interest rate on international financial capital, and that in turn may be expected to be higher than the rate on SDRs. This would imply a distribution of the burden somewhere between those described above.

In schemes involving the assumption by developed countries of the interest cost of developmental allocations there would, of course, be an additional burden on these countries: this burden would presumably be shared among them roughly in proportion to quotas.

9. Conclusion

A great variety of possible schemes for providing finance for development as a byproduct of the process of reserve creation through SDR allocation have been briefly considered in this paper. While no attempt has been made to work out in detail the technical aspects of any of these schemes, it appears unlikely that any of them would have to be considered technically inoperable. Their merits and demerits are to be assessed primarily on political and economic grounds.

It would be premature at this stage in the reform of the international monetary system, when it is as yet uncertain what the future of the SDR as a reserve asset will be and to what extent SDR allocations will effectively provide the primary element in the growth of world reserves, to attempt to balance out the pros and cons of “link” schemes in general or of any particular scheme. However, there are some schemes that would almost certainly reduce the cost and improve the quality, and would very probably on balance increase the quantity, of the financial assistance provided for development. On the other hand, great care would be required to ensure that using the SDR system to serve a second purpose—that of promoting development—did not impair its efficiency in the pursuit of its primary aim. The risk cannot be entirely excluded that some of the schemes, particularly those that attempted to go furthest in being helpful to developing countries, might, if applied on too ambitious a scale, impair the willingness of other countries to accumulate SDRs, accept allocations, or even continue participation in the SDR system.

Direct repercussions of “link” schemes on the aggregate demand situation in the world are ambiguous and not likely on balance to be important. The effects of such schemes on total allocations of SDRs will, in all probability, affect the timing of allocations much more strongly than their cumulative total over time.

This is part of the general difficulty of synchronization between development needs and global reserve needs—a difficulty which cannot be completely resolved but could be mitigated by flexibility on the part of the Fund in the proportion of total allocations directed to development, and by efforts on the part of the recipients to stabilize their development expenditures.

In considering individual “link” schemes, the possibility should be borne in mind of having more than one scheme in operation simultaneously or having “hybrid” schemes which combine features of the various schemes discussed above.

Among the types of link schemes that have been distinguished, Type C schemes, which would give developing countries additional quotas, and Type D schemes, which depend on the net receipts, in the form of interest or amortization, of a possible substitution account, would be likely to yield rather small amounts. The use of SDR issues to finance new longer-term lending facilities in the Fund would be likely to channel only moderate amounts to developing countries, at least initially, but would offer considerable guarantees of effective use of the resources provided and of additionality. As indicated earlier, Type E schemes or extended Fund facilities might be applied in conjunction with schemes of Types A or B.

Type D schemes, under which developed countries would themselves provide development assistance in amounts related to the SDR allocations they receive, are probably at a disadvantage compared with other types of schemes with respect to the amount of assistance likely to be available, its additionality, and the extent to which it is freely expendable in all countries.

Type A schemes (and the analogous Type D and Type E schemes) have the advantage over Type B schemes (and the analogous Types D and E schemes) in that the resources they provide are more certain to be devoted to developmental purposes, and they may be better able to absorb the inevitable variations in the amount of SDRs allocated for development without undue disturbance to actual development expenditure. Type B schemes, however, probably have an advantage over Type A schemes with respect to additionality and are certainly better able to achieve a desired country distribution of the development assistance provided. If Type A schemes are preferred, their weaknesses with respect to additionality and equity of distribution might be minimized if allocations to dfis were used to finance a new and separate form of lending. The possibility of a hybrid form of scheme, combining the control over use inherent in Type A schemes with the distributional advantages of Type B schemes, should not be overlooked.

An important distinction, which cuts across the lettered categories, is between schemes in which the interest burden arising with respect to the developmental allocations is assumed by developed countries and schemes where it is not so assumed: it is analogous to the distinction between grant and loan aid. Schemes involving grant aid are clearly more beneficial to developing countries than those involving loan aid, but it might be prudent not to go too far in this direction in order to forestall any danger that some developed countries, in their reluctance to assume the interest burden, might refuse allocations or even be tempted to withdraw from participation in the SDR system. However, if the avoidance of special arrangements for the interest payments on developmental allocations led to the maintenance of unduly low interest on SDRs in general, this might be just as dangerous to the monetary system.

Another cross-classification of schemes is between those based on fixed proportions between developmental allocations and those that provide for a steady flow or a steady rate of growth of developmental allocations from one basic period to the next. The latter arrangement—the case for which is strengthened (under Type A schemes) if the dfis are obliged to segregate the reserves demand for SDR allocations for a special type of lending—would contribute to the stabilization of development expenditure, but would call for considerable self-discipline on the part of developed countries to ensure that total allocations continue to be governed by considerations germane to the global need for reserves.

If part of SDR creation was put into circulation through Type A schemes, the allowance for such amounts in determining the appropriate magnitude of SDR allocations to countries would have to reflect not the allocations to be made to dfis during a given basic period, but the amounts likely to be disbursed by them to countries during that period. The deductions to be made for such disbursements would be comparable to the allowances now made for reserve creation in forms other than SDRs, or the allowance that would have to be made for SDRs put into circulation through extended Fund operations.

Maxwell Stamp, “The Stamp Plan—1962 Version,” Noorgate and Wall Street (London), Spring 1962, pp. 67–81.

Robert Triffin, Gold and the Dollar Crisis: The Future of Convertibility (New Haven: Yale University Press, 1960, Rev. 1961).

United Nations Conference on Trade and Development, International Monetary Issues and the Developing Countries, Report of the Group of Experts, UN Document No. TD/B/32 (Geneva, 1965).

Tibor Scitovsky, Requirements of an International Reserve System, Essays in International Finance, No. 49, Princeton University (Princeton, New Jersey: Princeton University Press, 1965).

I. G. Patel, “The Link Between the Creation of International Liquidity and the Provision of Development Finance,” in United Nations Conference on Trade and Development, Report of the Committee on Invisibles and Financing Related to Trade: Further Consideration of the Report of the Expert Group on International Monetary Issues, UN Document No. TD/B/115/Add.2 (Geneva, 1967).

International Monetary Fund, Summary Proceedings of the Twenty-Third Annual Meeting of the Board of Governors, 1968 (Washington, 1968), p. 81. See also 1969, p. 71.

U.S. Congress, Joint Economic Committee, 91st Congress, 1st Session, A Proposal to Link Reserve Creation and Development Assistance, a report to the Subcommittee on International Exchange and Payments (Washington: U.S. Government Printing Office, 1969).

United Nations Conference on Trade and Development, International Monetary Reform and Cooperation for Development, Report of the Expert Group on International Monetary Issues, UN Document No. TD/B/285/Rev. 1 (New York: United Nations, 1969).

Reproduced above, pp. 19–56.

Certain data with regard to these institutions are set forth in Table 1.

If official holdings of dollars held outside the United States were included, the amount outstanding at end-1971 would amount to some SDR 65 billion.

The membership of the Development Assistance Committee of the Organization for Economic Cooperation and Development (oecd) includes 16 of the 23 members of the oecd: Australia, Austria, Belgium, Canada, Denmark, France, the Federal Republic of Germany, Italy, Japan, the Netherlands, Norway, Portugal, Sweden, Switzerland, the United Kingdom, and the United States. In addition, the Commission of the European Economic Community sits as a full member of the Committee.

Up to the year 1971 all these figures can be converted to SDRs at the rate of $1 = SDR 1.

United Nations, General Assembly Resolution No. 2626 (October 24, 1970).

Mahbub ul Haq, “Tied Credits—A Quantitative Analysis” in Capital Movements and Economic Development, Proceedings of a Conference held by the International Economic Association, ed. by John H. Adler (New York, 1967), pp. 326–59.

See Proceedings of the United Nations Conference on Trade and Development, Second Session, Volume IV: Problems and Policies of Financing (1961), and Thomas L. Hutcheson and Richard C. Porter, The Cost of Tying Aid: A Method and Some Colombian Estimates, Studies in International Finance, No. 30, Princeton University (Princeton, New Jersey: Princeton University Press, 1972), especially p. 5.

The important point here, if a contractionary effect on world reserves is to be avoided, is that the currencies accumulated should not consist to a disproportionate extent of reserve currencies.

This disregards any small once-for-all loss of reserves to developing countries that may take place as a result of the “link.”

This follows from the fact that, assuming that each country adjusts to any “link” in such a manner that its reserves are not affected, the loss of real investment (equals the improvement in the current balance) plus the net amount of additional borrowing (the improvement in the capital account) must equal the amount of allocation forgone.

    Other Resources Citing This Publication