The Allocation and Distribution of SDRs

The discussion of SDR allocation opened with a session intended to solicit the views of outside experts on the cases for and against allocation under the Fund’s current Articles of Agreement (Chapter 4). For that purpose, the session began with Michael Mussa’s presentation of arguments that the IMF staff has made to the Executive Board in suggesting a rationale for SDR allocations.⁵ Mussa’s arguments then received critiques from three outside experts—Montek Singh Ahluwalia, Horst Siebert, and John Williamson—along with a fourth evaluation from IMF Executive Director J. de Beaufort Wijnholds.

As noted above, the impasse over SDR allocation can be attributed, at least in part, to a lack of consensus on what is meant by the “long-term global need … to supplement existing reserve assets.” In reflecting on why this central concept in the allocation criterion has been a challenge to interpret, Adolfo Diz noted (in Chapter 3) that: “[d]uring the discussion leading to the creation of the SDR the concept was discussed at length but no clear-cut definition could be reached. Probably the expression will remain as another striking example of the futility of seeking consensus through the creative use of ambiguity.”

Mussa provided his own perspectives of how the architects of the SDR appear to have intended the allocation criterion to operate. He then contended that allocating SDRs to meet a moderate proportion of the growth in the demand for reserves over the long run would provide meaningful economic benefits for many countries and for the global economic system. To the best of Mussa’s knowledge and judgment, “opponents of SDR allocations have never seriously challenged the analytical or factual basis of… [his line of] argument. They have mainly ignored it….” This led Mussa to the conclusion that the impasse over SDR allocation had little or nothing to do with the criterion and quantitative assessment of long-term global need, but rather reflected disagreement on “the general principle of whether the IMF should still be in the business of supplying reserves through the SDR mechanism.”

In stating the case for SDR allocation, Mussa noted that, despite the major systemic changes that have occurred since the creation of the SDR, the demand for reserves to hold had continued to grow over time, maintaining a relatively stable relationship with the scale of international trade and economic activity. He then argued that since the holding of reserves had little or no real cost for the world economy, and since in the current system all but a few very large industrial countries faced significant costs in holding reserves (as reflected in the spreads between the interest rates at which they could borrow and the rates they earned on reserve assets), SDR allocations that met a moderate share of the projected growth in the demand for reserves would provide economically meaningful benefits for most countries.

In addressing Mussa’s arguments, the papers by Siebert, Wijnholds, Ahluwalia, and Williamson provided a wide range of reactions. Siebert rejected Mussa’s conclusions, arguing that the initial raison d’être of SDR allocations no longer existed, that the expansion of international trade and international reserve holdings had not been constrained by the absence of SDR allocations, and that SDR allocations established the wrong incentives for countries with deficits. By contrast, Wijnholds saw much in Mussa’s paper with which he agreed, particularly the view that cost considerations had become the most convincing argument in favor of SDR allocation.

Ahluwalia was also in broad agreement with Mussa’s arguments. He emphasized, inter alia, that the case for SDR allocation was strengthened by considering its implications for the quality of reserves. Reserves obtained through borrowing have to be refinanced periodically, and unexpected developments that influence international financial markets, which are vulnerable to herd instincts and contagion effects, can introduce volatility in the terms and conditions under which individual developing countries can roll over their borrowed reserve holdings.

Williamson saw some validity in the main arguments put forward by opponents of SDR allocations. However, he saw little chance of future SDR allocations unless the decision-making process was “liberated from the criterion of ‘global need’ that was imposed on it in a world that has disappeared….” Noting, however, that the interpretation of words can evolve, Williamson regarded Mussa’s “reinterpretation” of the concept of long-term global need as “a linguistic veil” that would avoid the necessity of amending the Articles should a general will to resume allocations ever emerge.

Mussa’s arguments provoked comments from many other participants in the seminar. Helmut Hesse (in Chapter 6) joined Siebert in rejecting the arguments out of concern that new allocations would allow some countries to pursue irresponsible policies, and also in the faith that market forces provide appropriate assessments of creditworthiness and a good source of discipline. Peter Kenen challenged these views in Chapter 10. The first view, if pushed to its logical limit, seemed to imply that the inducements for governments to pursue responsible policies should be strengthened further by not allowing them to hold any reserve assets. Faith in market forces also pointed to logical inconsistencies. “If we can count on market forces to discipline governments, why do governments need reserves? Why should they not be made to borrow at market rates if and when they deem it appropriate” to finance payments deficits? To Kenen, the “answer is obvious. A government may require financing precisely when markets are unwilling to provide it, and markets are not always right….” An expanded but similar perception was provided by György Surányi: “A lesson from the emerging markets’ crises is that unsustainable macroeconomic policies cannot be bettered by the creation of international liquidity. Another lesson is that capital markets never respond immediately: sometimes they react too late, when social costs are too high” (Chapter 7).

Seminar participants also considered several other issues relating to the allocation and distribution of SDRs. Whereas the papers for the second session of the seminar (collected in Chapter 4) focused on the pros and cons of allocations under the present Articles of Agreement, participants in the later sessions had been asked to abstract from the present Articles in considering the possible roles that the SDR could play in the future.

Several outside experts at the seminar saw merit in the idea of regular small-scale allocations of SDRs. Some felt that allocations should be designed to maintain or gradually increase the share of SDRs in total nongold reserves, as Mussa suggested in his paper. Fabrizio Sacco-manni (speaking for himself and co-author Tommaso Padoa-Schioppa) emphasized that the challenges faced by central banks in promoting monetary stability and providing lender-of-last-resort functions would grow over time, and that it would be useful to keep the SDR available to assist, conceivably, in performing these functions. Much the same view was expressed by Camdessus and Alexander Swoboda, among others. Swoboda paraphrased Wendy Dobson in characterizing the SDR as “to some extent… a solution in search of a question”; but he also opined that something like the SDR “will likely be needed some day as a collectively created reserve asset, as a means of reaping and distributing the gains from money creation at an international level, and as a means of increasing symmetry in the international monetary system.” The argument for keeping the SDR alive was bolstered by Rhomberg, who noted that the process of creating the SDR had been very time consuming, that “internationally negotiated remedies for novel crises often tend to come too late to be helpful… [and, therefore, that] remedial machinery already in place should be kept, even if not urgently needed at the time, on the chance that the type of crisis for which it was designed may recur” (Chapter 3).

Alec Chrystal’s paper (Chapter 8) provided a conceptually different case for regular small-scale allocations. Although Chrystal did not regard allocations as important for meeting international liquidity needs, he argued that the SDR is an attractive unit of account for intergovernmental transactions and recommended that the stock of SDRs be expanded in parallel with the growth of such transactions. Still another case for regular allocations was alluded to by Dobson, who noted (Chapter 9) that over the years the Fund had spent far too much valuable time discussing the issue of SDR allocation, which might be taken to imply that automatic small-scale allocations (as well as routine allocations to new members) could be an efficient way to proceed.

Seminar participants expressed two types of concerns about the distribution of SDRs. One source of concern was that many new members of the Fund had never received SDR allocations, whereas many others had received allocations during 1979-81 but not during 1970-72 (see fn. 2, above). Camdessus, in his luncheon address, suggested that this inequity could be addressed through an amendment to the Articles that would harmonize countries’ net cumulative SDR allocations as a share of quotas. Chrystal also addressed the treatment of new members of the Fund, arguing that they should automatically receive an allocation of SDRs.

The second source of concern about the distribution of SDRs related to asymmetries in the international monetary system. As emphasized by Ariel Buira and by Muhammad Yaqub, Azizali Mohammed, and Iqbal Zaidi (Chapter 6), some countries have easy or large access to credit from international capital markets, whereas many others face limited access and relatively expensive terms. Accordingly, a number of seminar participants proposed mechanisms for shifting the distribution of allocated SDRs toward countries that would benefit the most—namely, countries with relatively limited access to international capital markets.

One such mechanism would be a scheme for voluntary postallocation redistribution of SDRs. Makoto Utsumi reminded the seminar that one variant of such a mechanism, known as the Hashimoto proposal, had been put forth by the Japanese authorities at the 1990 Annual Meetings of the Fund. Another scheme for postallocation redistribution, described in the paper by ZHU Xiaohua (Chapter 7), would involve voluntary transfers of SDRs into an account administered by the Fund in exchange for SDR-denominated shares that holders would subsequently be free to sell. The SDRs transferred to the administered account would then be used to supplement the conditional financing available in support of regular Fund programs.

As an alternative to a postallocation redistribution mechanism, SDR allocations could be targeted directly at countries with limited or no access to private credit markets. Two of the seminar papers advocated that the SDR system be amended in this way. Yaqub, Mohammed, and Zaidi proposed that allocations be focused on countries eligible for concessional financing from international financial institutions, but exclude countries that either receive investment grades by credit rating agencies or have ratios of reserves to imports that exceed some high threshold. Buira proposed that allocations be based on Fund estimates of the reserve needs of individual countries, and that the proportions of reserve needs provided through allocations could be related inversely to per capita GDPs. Under Buira’s proposal, countries could be eligible to receive such allocations only if they met Maastricht-type criteria relating to fiscal balance, external position, and other measures devised and monitored through Fund surveillance.

The idea of attaching Maastricht-type criteria or other forms of conditionality to the receipt of allocated SDRs was also favored by a number of other seminar participants, including Max Corden and Alejandro Végh. Corden (Chapter 3) suggested that SDRs be allocated directly to the Fund for use in the extension of conditional credit. Végh (Chapter 10) argued that the benefit accruing to a country that received conditional liquidity “derives in much larger degree from the conditionality than from the liquidity.”

Several seminar participants suggested that SDR allocations could be used to provide large-scale emergency financing on a temporary basis. Buira proposed that the IMF establish an emergency financing facility on which member country central banks could draw at their own initiative in exceptional circumstances, subject to paying an interest rate premium and establishing eligible collateral. Along the same lines, Marcello de Cecco and Francesco Giavazzi (in Chapter 7) advocated that the Fund be authorized to issue SDRs in unlimited quantities when it decided that interventions to resist speculative attacks were necessary and justified. They emphasized that the liquidity created in this way should be temporary, that the provision of such a safety net through the Fund would have the advantage of spreading risk widely among its many member countries, but that the effectiveness of such a safety net would require a mechanism for the Fund to come to speedy decisions regarding its responses to speculative attacks.

The proposals for large-scale emergency financing provoked some negative reactions as well. Several seminar participants worried about the moral hazard issue. Végh, for one, felt that “sometimes it is better for future discipline not to avoid default.”

Prominence of the SDR as a Reserve Asset

A number of seminar participants addressed the notion of making the SDR “the principal reserve asset in the international monetary system,” which—as noted above—is characterized as an “objective” in the Fund’s present Articles of Agreement. Although there was little support for pursuing that objective, and little expectation that it would ever be realized, the outside experts made a number of observations on the factors that had prevented the SDR from becoming a prominent reserve asset.

Many participants noted that the development of the SDR had been undermined by the breakdown of the Bretton Woods system and the growth of international capital markets, and obviously by the cessation of SDR allocations. Jonathan Frimpong-Ansah observed that the rivalry between national sovereignty and multinational bureaucratic management had rarely favored the latter: “The IMF was not enhanced as a managing institution of global liquidity, the U.S. dollar and other leading currencies were not phased out, and therefore the SDR never became the principal reserve asset.”

Seminar participants implicitly expressed a number of different views on issues relating to this observation. Rhomberg and Wijnholds regretted that the international community had failed in the several attempts it had made during the 1970s to create a substitution account. Rhomberg (Chapter 3) viewed the creation of a substitution mechanism as important for addressing the forgotten horn of Triffin’s dilemma, namely, loss of confidence in a major reserve currency. Wijnholds (Chapter 4) saw a substitution account as “the only realistic chance to have made the SDR anything close to the principal reserve asset.” Kathryn Dominguez (Chapter 5) suggested that the issue of a substitution account could arise again in the future, given the dollar’s large share in official reserve holdings and the possibility of a shift toward the euro or the yen. Hesse (Chapter 6) argued that the SDR could not be used to alleviate the shortcomings of the international monetary system and felt it was understandable that national authorities do not wish to relinquish economic policy autonomy, including unrestricted power of disposal over the greater part of their reserves. Christian Stals (Chapter 10) focused on the links between the fate of the SDR and the autonomy of the IMF: “We cannot have an international currency without an international central bank. We cannot have an international central bank unless at least the major political powers will stand behind it.” Dobson implicitly suggested that the converse of Stals’s first point did not hold. Rather, she agreed with the view that although the SDR does not appear to have an important future, the Fund does—both in responding to international crises, for which it needs a mechanism to provide emergency credit, and in preventing international crises through stronger surveillance.

Other seminar participants provided different perspectives on the SDR’s lack of prominence. Onno Ruding (Chapter 9) suggested that the SDR could be promoted by encouraging the World Bank and other official institutions to use it more widely, both as a unit of account and in their borrowing and lending operations. Christian de Boissieu (Chapter 5) argued that the official use of a currency cannot develop over a certain threshold if it is not sustained by private use of the same currency. In a similar vein, Jacques Polak, Ruding, and others expressed support for authorizing commercial banks and other private market participants to hold SDRs created by the Fund, and Utsumi saw merit in allowing for the use of the SDR as an instrument for exchange market intervention. Paul De Grauwe, however, provided a different perspective on why a private market for SDRs has not developed. Noting that “the utility of money for an individual is based on the fact that many other people use the same money” (Chapter 5), De Grauwe argued that the development of a private market for the SDR would require political decisions and collective action to promote its use, not just authorizing private market participants to hold officially created SDRs or tinkering with the characteristics of the asset to make it more attractive.

One type of collective action would be to institute an international payment system based on the SDR, as Robert Heller proposed in Chapter 8. Heller suggested that this could kill two birds with one stone, not only providing an enhanced role for the SDR but also resolving the ongoing efforts of the international community (in the context of payments system working groups convened by the Bank for International Settlements) to establish a system that would achieve final settlement of cross-currency payments in an efficient and secure manner.

Although nobody disputed the view that collective action would be needed to promote the private use of the SDR, Stals emphasized (as did others to different degrees) that the possibility of reinvigorating the SDR required renewed efforts to define clearly, inter alia, the role and functions of the Fund and the shortcomings of the international monetary system as it now operates. To Stals, the road to creating a prominent role for the SDR had to be mapped out, clearly identifying how the SDR could help solve systemic problems.

Valuation and Interest Rate Characteristics of the SDR

The Articles of Agreement authorize the Fund’s Executive Board to decide how to define or “value” the SDR. The SDR was valued initially in terms of a fixed quantity of gold (equivalent to one U.S. dollar) and was redefined in June 1974 as a basket of 16 currencies. Since 1981, it has been defined as a basket of five currencies, subject to small quinquennial revisions in the quantities of these five currencies.

In considering possible ways to enhance the attractiveness of the SDR, it is natural to ask whether its composition is optimal in some meaningful sense. Holger Wolf was asked to address this issue in Chapter 8. He considered the optimal composition of the SDR for each of two purposes. First, he defined the optimal unit of account currency basket as the basket that minimized the quota-weighted sum over all countries of the variances of bilateral exchange rates vis-à-vis the basket, after adjusting the bilateral exchange rates to eliminate the nominal exchange rate variance associated with differences in trend inflation rates. Second, he defined the optimal reserve basket in terms of how well the basket preserved real purchasing power over imports, defining an index of satisfaction for each country and averaging the index values for individual countries. Using quarterly data for 1985–94, he then computed the currency baskets (of given sizes) that proved optimal ex post on the basis of each of his definitions. The optimal unit of account basket was based on 134 sample currencies (and countries) for which data are complete; the optimal reserve basket was based on 20 candidate industrial country currencies and a sample of 125 countries for which data are complete. For each of the two types of baskets Wolf also calculated the weights that would be optimal if the basket was limited to the five currencies that compose the current SDR.

Wolf’s initial results seem puzzling at first glance. A number of currencies that are not widely traded internationally are included in the optimal unit of account baskets, and the optimal reserve basket based on the five SDR currencies would give the French franc an 82 percent share of the basket, and the pound sterling a 15 percent share. The paper made an important conceptual advance in pointing the way toward the use of explicit optimization criteria for purposes of defining currency baskets. Wolf emphasized, however, that other criteria could be chosen and would generate different results and that his empirical tests are strictly backward looking. In short, there is no clear-cut way to identify a single “optimum” currency basket that would be unambiguously superior to the SDR as presently defined.

The attractiveness of the SDR as a currency basket was also addressed in the paper by Barry Eichengreen and Jeffrey Frankel, who took a forward-looking perspective, dividing the future into the immediate, intermediate, and distant horizons. After assessing the needs for units of accounts and reserve currencies at each of these horizons, Eichengreen and Frankel concluded that “the future of the international monetary system is unlikely to entail a significantly expanded role for the SDR.” Although they did not attempt to define and identify optimal baskets, they pointed to the fact that among countries that have chosen to peg their currencies to currency baskets, the proportion that have pegged to the SDR declined from approximately 40 percent during 1979-82 to less than 15 percent in 1995. They also noted that the shares of the SDR component currencies in identified official holdings of foreign exchange reserves are considerably different from the shares of these currencies in the SDR basket. Moreover, the shares in foreign exchange reserves have changed gradually over time in a manner that appears to be predictably related to changes in the relative sizes of countries.

Despite the interesting analyses offered by Wolf and by Eichengreen and Frankel, seminar participants did not choose to focus extensively on valuation issues. Several participants commented briefly on the issue of redefining the SDR to prevent inflation from eroding its purchasing power over goods. ZHU Xiaohua spoke in favor of hardening the SDR in this way, while Chrystal saw little benefit in the low-inflation environment of the mid-1990s. A few participants also referred to the fact that the anticipated introduction of the euro as the primary currency in Europe would presumably call for a redefinition of the SDR basket.

A number of outside experts raised issues relating to interest rates on the SDR. Zhu and Siebert argued that because many countries have drawn down their SDR holdings for periods of much longer than three months, the SDR interest rate should be calculated from long-term rather than short-term rates. Mussa also expressed implicit support for a move in this direction, observing that as presently defined, the SDR interest rate “is generally quite attractive for net borrowers in the SDR system … [but] not exceptionally attractive for net creditors.”

During the general discussion at the seminar, Williamson posed the issue of introducing an auction mechanism for determining the rate of interest on SDRs. By amending the Articles to authorize private market participants to hold SDRs, the Fund could, with additional effort, catalyze the development of SDR-denominated instruments with a range of maturities, and hence the emergence of a market-determined term structure of SDR interest rates.

Several participants focused on the fact that the present Articles preclude the introduction of a spread between the rate of charge that countries must pay on their cumulative allocations of SDRs and the rate of interest that they receive on SDR holdings. Accordingly, the effective interest rate paid by net users of SDRs is no greater than that earned by net creditors. Polak noted the anomaly in the different rates of charge applied to debtor positions in the SDR Account and the Fund’s General Resources Account. The rate of charge for conditional credit under the latter is higher than the rate of charge for unconditional credit within the SDR system, Polak’s proposal to merge the two accounts (see the next section of this chapter) would eliminate this anomaly. Buira also proposed that the SDR system be changed to introduce, inter alia, a spread between the user rate and the holding rate.

Financing the Fund with SDRs

In addressing the potential for the SDR in the provision of conditional liquidity, Polak distinguished between two institutional changes that were associated with the introduction of the SDR. The First Amendment of the Fund’s Articles not only established a new way for the Fund to provide members with access to liquidity, via the allocation of unconditional drawing rights, but also introduced a new financing technique, which simply involved making entries on the Fund’s books.

Polak’s paper focused on the different financing techniques associated with the SDR system and the Fund’s General Resources Account, contrasting experiences with the two techniques and proposing that the Fund integrate all of its transactions under the SDR financing technique.⁶ He argued that this would provide a much more efficient process for financing the Fund, lead to a more economically sensible system of charges and remuneration, and introduce simplicity and transparency into the Fund’s financial balance sheet. Kenen characterized the proposal as “radical but not irresponsible,” and as “required reading for every member of the Interim Committee.” He noted that the adoption of Polak’s scheme need not represent any loosening of the control that member countries have over the financial activities of the Fund, as the extension of Fund credit would continue to be limited by Fund policies and Executive Board decisions. The Fund would issue SDRs to finance conditional programs, and would retire the SDRs upon repayment. Total issuance of SDRs would be constrained to some proportion of aggregate Fund quotas.

The implementation of Polak’s proposal would require an amendment of the Articles, but once introduced, the new scheme could reduce the subsequent need (or frequency of need) for parliamentary approval of SDR allocations or quota increases. It would also eliminate the need for countries to pay for quota subscriptions, thereby clarifying the domestic fiscal implications of Fund operations for parliaments when they consider whether to support an increase in Fund quotas.