2 The International Monetary Issues of the Bretton Woods Era Are They Still Relevant?1
- James Boughton
- Published Date:
- December 2004
The key question addressed in this chapter can be formulated as follows: do we revisit the intellectual battlefields of the 1960s and 1970s, merely to reminisce about a temps perdu? Or do we find in this experience much that is of value to us in facing the problems of today’s international monetary system?
There is no doubt, as Robert Solomon has pointed out recently, that there have been major shifts in the features of the international monetary system that preoccupy officials (Solomon 1991, pp. 67–77). We think differently about adjustment now than we did in the 1960s. We hardly think any more about international liquidity, which was our major preoccupation in the 1960s. Market fluctuations of 15 per cent in the exchange rates of the major currencies, though they make most of us feel somewhat uncomfortable, are pretty well tolerated by all concerned, in contrast to the four months of crisis it took in 1971 to reach agreement on more modest exchange rate changes. Of late, we have even developed an indifference to current account imbalances that would have given us apoplexy less than a decade ago.2 Solomon attributes many of these changing perspectives on the international monetary system to the enormous increase in the international mobility of capital. In a broad sense, that characterization is surely correct. But I would want to register two important reservations.
First, even though today’s problems may look radically different from yesterday’s, that does not make yesterday’s solutions irrelevant to today’s policymakers. Let me refer to some examples that relate to exchange rate regimes. At the end of the ‘reform exercise’ of 1972–4 most observers would probably have agreed with what Williamson wrote in 1977 ‘that the principal and crucial intellectual error of the Committee of Twenty (C-20) lay in the decision to opt for restoration of the adjustable peg at a time when this system had ceased to be viable because of the development of capital mobility’ (Williamson 1977, p. 181). Yet, two years later most members of the European Community (EC) readopted the adjustable peg. By 1991, an enlarged European Monetary System (EMS), with freedom of capital movements for its members, is generally accepted as a striking success (I am not overlooking the special—and hence limiting—political factors that were necessary to this success). Not only the Bretton Woods experience, but also the gold standard experience, now seems quite relevant to the understanding and the design of the Economic and Monetary Union (EMU). To cite a quite different case, I find that what Nurkse wrote fifty years ago on the difficult exchange rate choices facing some sterling area countries is highly pertinent to the current problems of Australia and New Zealand (League of Nations 1944, pp. 134–6). The problems that populate the time-space universe in which economists roam belong to only a limited number of species. Old friends and enemies, or their close relatives, pop up all the time.
The other side of the relevancy question is that one should not assume that, at any time, officials are focusing their attention on issues that are relevant even at that time. As seen from a safe distance, the 1960s and 1970s—and I am sure many other decades as well—reveal a considerable amount of lost motion, pursuit of solutions to phantom problems, or of ill-considered solutions to real problems. I shall come to some interesting instances in the course of my chapter. Here, I shall just briefly recall a single major misdirected effort of the 1970s. In 1974–5, the Organization for Economic Cooperation and Development (OECD) and the Group of Ten (G-10) spent an intense six months in working out an SDR 20 billion ‘Financial Support Fund’ proposed by the US authorities and the Secretary General of the OECD. This facility, to be attached to the OECD, was designed ‘to back up private capital markets and other financing mechanisms’.3 The developed countries did not need this third line of defense; in fact, since 1977, they have not used the second line, the International Monetary Fund (IMF). When the US Congress refused to put up the US share of the money, the plan collapsed without a trace and with hardly a tear.4
On the question of relevancy I would put my intermediate conclusions as follows:
1. Even though the system has changed, much of the experience of the 1960s and 1970s remains relevant. Many of the outlines of that experience have become blurred by the events of the past quarter century. They deserve to be brought back to light.
2. The same period also saw some major preventable misdirections of international efforts. They too deserve our attention—first, as a general matter, to maintain a healthy skepticism with respect to the collective wisdom of officials, and second, more specifically, because in economics bad ideas seem to have almost as many lives as cats. (My favourite example of this proposition is the topic of ‘commodity currency’.)
Guided by these two thoughts, I have selected five topics for discussion, all of them very much in the area in which Ossola played a leading role. Each contains some elements of theory that deserve to be brought out, some aspects of history that need straightening out and perhaps even some modest lessons to be learnt. The topics relate to the adjustment process, liquidity creation, the interest rate on the Special Drawing Rights (SDR), a substitution account, and objective indicators for exchange rate adjustment. I will not reveal in advance which of these belongs in category one (the past unjustly forgotten) and which in category two (if only we had been brighter). And, as will become clear, where I criticize ‘officials’, I do not exclude those in international organizations.
I realize that my selection of topics leaves out what some might regard as the most significant failure of financial officials in the period: the long and agonizing death of the par value system, lasting (at least) from the November 1968 Bonn Conference to the second devaluation of the dollar in February 1973. That topic deserves a full-scale study of its own. I shall leave it here with a question. Can one fault the official family (ministers, central bank governors, senior officials) for their unwillingness to administer euthanasia to a beloved system while there was still a glimmer of hope that it might survive?
There is a widespread impression that, in the period we are here considering, there was too much emphasis on financing payments disequilibria and too little on adjustment. Up to a point, this impression is correct: note, for example, the fact that the two industrial countries that used the Fund’s oil facility, the United Kingdom and Italy, soon thereafter had to come to the Fund for standby credits. But it would be quite wrong to say that the need for adjustment was neglected.
The first G-10 Deputies Report (1964) on ‘the monetary system and its probable future needs for liquidity’ started out with a long chapter on the importance of adjustment and suggested that Working Party 3 (WP-3) of the OECD prepare a special study on this. It did not take too many meetings of the WP-3 to prepare this study, which was published in 1966 (OECD 1966). The members of the committee knew their Tinbergen and Meade, and they could readily agree that ‘the adjustment process is therefore essentially a question for governments as to how to achieve a wide range of aims of an economic, political, and social nature with the limited number of policy instruments actually or potentially available to them’ (para. 16). The report notes wide agreement on the policies appropriate in ‘pure demand cases’ (para. 45). It is recognized that it is more difficult to deal with cases of ‘imperfectly adjusted competitive positions’: surplus countries don’t want to inflate and deficit countries don’t want to accept prolonged periods of stagnant demand. Exchange rate changes are mentioned (for deficit countries only!), but with considerable reservation. It should be remembered that in the mid 1960s, exchange rate changes were a near taboo subject among the G-10 countries, especially for the United States and the United Kingdom.5
Looking at this 1966 OECD report from today’s perspective, there are at least two interesting and I would say endearing characteristics that have tended to get lost in the succeeding period. They are:
1. Countries are, on the whole, responsible for their own payments imbalances. This presupposes, of course, that aggregate demand in the world is maintained on a broadly satisfactory level; but that condition was reasonably fulfilled in the 1950s and 1960s. Any country that had a large deficit or surplus could, in this setting, be expected to take remedial measures.
2. The appropriate measures depended on the cause of the imbalance. This was accepted doctrine, following Meade.
Much of the harmony in WP-3 in the 1960s can be explained in terms of these two propositions. They led, in principle, to clear prescriptions as to which country should do what. In practice, they left room for discussion on the facts, on the urgency of action or on its political acceptability. But if deficits or surpluses persisted, a strong consensus that included the country under consideration was likely to evolve. Neither of these two propositions is still present as a strong element in the adjustment process as it currently works (or does not work) among the main industrial countries. The second proposition—that there are objective criteria for the choice of appropriate adjustment policies—succumbed first. This happened in the reform exercise of the C-20. Considering how many rules that exercise tried to impose on countries (asset settlement, reserve composition, indicators, etc.), and anxious to avoid the criticism that it treated every imbalance beyond a certain size as calling for an exchange rate change, the Committee thought it wise to leave countries the choice among possible adjustment policies. Its rule on that subject was brief and vague (IMF 1974, p. 8): ‘Countries will take such prompt and adequate adjustment action, domestic or external, as may be needed to avoid protracted payments imbalances. In choosing among different forms of adjustment action, countries will take into account repercussions on other countries as well as internal considerations.’
It is not so easy to pinpoint when the first proposition was abandoned, among other reasons because it is sometimes difficult to give to it a concrete content. Ex post statistics show only surpluses and deficits. In the 1960s and early 1970s, these figures at least added up to about zero for the world as a whole, but even that is no longer the case. To find causes, to find which countries are ‘at fault’, one has to dig below the surface. In some instances, the imbalances may be attributable in part to surplus, in part to deficit countries. This was certainly the case in the 1971 situation, which called for, and in the end produced, multilateral adjustment. But it is one thing to acknowledge the difficulties in assigning responsibilities for adjustment, and another to abandon the effort altogether and to replace it by a political mechanism that makes external adjustment among the major countries simply their ‘shared responsibility’, helped along by ‘a regular dialogue at the political level’ and ‘peer pressure’. This is how a US Treasury Report to Congress describes the ‘structured but judgmental framework’ of the Group of Seven (G-7), which is taking the place of the ‘automatic techniques [that] failed in the past’ (US Treasury 1988). (It is not clear what failing automatic techniques or what past the authors of this report may have had in mind).
The point here is not to invoke nostalgic feelings about the coordination of adjustment measures as seen in the 1960s. There is no question that changes in the world financial system, among other factors, have made the design of adjustment rules much more difficult. But before we abandon the fundamentals of the approach of that period (and, at the same time, the rationale for the IMF’s consultations with individual countries) we would do well to know what we put in its place.
3 Liquidity Creation
In the 1960s the liquidity question took up much more time (elapsed time and meeting time) than the adjustment question. This was essentially because it raised intellectually far more difficult issues. Senior officials found themselves virtually without guideposts on the design of a system of reserve creation. In addition, coming from central banks and treasuries, they were naturally hesitant to engage in an activity that amounted to the creation of money ex nihilo. Steps in that direction were possible only if presented as ‘contingency planning’, to deal with a potential liquidity shortage. But beyond these very real intellectual and psychological difficulties, the process that finally culminated in the design of the SDR was held up by strongly negative feelings on the part of the US, the UK, and French authorities. In particular in the first year of the discussions (1963–4), the United States and the United Kingdom were concerned about the possible competition that a new international reserve asset would bring to the two main reserve currencies. Even after this reservation had been overcome, the United States continued a rearguard action to keep the interest rate on the SDR as low as possible. The French continued throughout the discussions to ride their own hobby horses: to enhance the role of gold and to deprive the reserve centres—in particular the United States—of their ‘exorbitant privileges’; when they did put forward a plan for reserve creation it was in fact a thinly veiled plan to raise the price of gold (Solomon 1982, p. 76). The patience of Ossola, as Chairman of the Study Group on the Creation of Reserve Assets, was sorely tested as large sections of his report had to be taken up by the discussion of two plans that never had a chance of broad support: the French Collective Reserve Unit (CRU) scheme and the British Mutual Currency Account (MCA) scheme (G-10 1965). A further retarding factor was the political ineptitude of the G-10, once they got around to designing a system for reserve creation, to limit the benefits of this system to themselves (‘liquidity creation of the Ten, by the Ten, for the Ten’, as Pierre-Paul Schweitzer characterized it), with at best a second window dispensing comparable amounts of credit to the rest of the Fund membership.
Against this background, the time and effort taken to design an agreed SDR system were not exorbitant; indeed, it is still somewhat of a miracle that officials and governments found the courage to agree to a novel mechanism, designed to protect the international monetary system against a threat that was only potential.
But, one may ask from today’s vantage point, were these efforts necessary? Why all this commotion about a potential shortage of international liquidity? The last fifteen years have pretty convincingly shown that such a shortage is a non-problem. With free capital markets, creditworthy countries—like creditworthy corporations—can borrow the amount of liquidity they need. The supply of liquidity is a global one, to meet the total world demand of liquidity from private and official holders combined. That supply is set by the monetary policies of the major industrial countries whose currencies are used in reserves. These policies may, at any moment of time, be unduly tight or unduly lax by some sort of world standard, as judged for example by the weighted votes of the governors of the IMF. But the possibility that the sum of the monetary policies of a small number of key countries may not be optimal does not create the need for a collective power to create or destroy money, such as that provided by the SDR mechanism. If the main central banks agreed that more money was needed, they would supply it themselves; if not, they would have the votes to block SDR creation in the Fund.
It is easy to project this view of the liquidity question backwards and to pass the judgment that the question was not relevant in the 1960s either. I think this would be a wrong inference from the available evidence. At that time, capital markets were not yet free enough and deep enough to satisfy governments’ legitimate needs for reserves. A few small countries (Norway, Australia) could satisfy their reserve needs in that market. Larger countries could not. It took the petrodollar deposits of the mid 1970s to get the commercial banks into the sovereign lending business on a world scale.
By the late 1960s, this stage had not yet been reached and it would be hard to fault officials for not foreseeing it. What they did see was, instead, a certain tightening of international liquidity (gold and dollars) and emerging symptoms of a reserve shortage that their analysis had suggested as a possibility.6 More detailed analysis of the 1969 decision to allocate SDRs would certainly be desirable. Were officials unduly influenced by the short-term tightening of credit in the United States and the resulting return flow of capital? Was there a certain itch to testdrive the new mechanism that had been put together with so much effort over the preceding six years? Perhaps there was a bit of both. But, mostly, this was a case where unpredictable events in the world economy turned a highly relevant problem into a marginal problem within a decade, creating a situation where it would be helpful to low-income countries, but hardly essential to the system, to continue SDR allocations (Polak 1988, pp. 175–90).
Could the problem come back—if not on the scale feared in the 1960s, but large enough to require the resumption of allocations? A good case can be made that it has come back. The countries making up the Commonwealth of Independent States (CIS) are making their entry in the international monetary system with negligible reserves. For one of them, Russia, the G-7 have announced a willingness to marshal a reserve fund of $6 billion, and commensurate action will no doubt prove necessary for the others. The G-7 and the G-10 have also acknowledged this as a system-threatening problem by announcing their willingness to activate the General Arrangements to Borrow (GAB) for this purpose.7 But why invoke the GAB, which is designed for short-term lending, when what is needed is a permanent addition to the reserves of a major group of members and to the world reserve total?
To achieve that objective, the SDR mechanism is clearly the more appropriate one. This is not the place to enter into a discussion on how precisely that mechanism could best be used for this purpose, given the fundamental rule that SDR allocations are made to all members in proportion to quotas. Quite likely, the best technique would be a moderate-sized allocation, combined with an agreement that the main industrial countries would on-lend their allocations (perhaps through the Fund) to the CSI members and to the, equally needy, developing countries, as a second addition to all these countries’ reserves.
4 The Rate of Interest on the SDR
One of the puzzling aspects of the long discussion (1963 to 1967) on the creation of the SDR was the almost total disregard of the interest rate on this asset. Indeed, that discussion can only be understood on the implied assumption that the SDR would carry a zero interest rate and that hence, in the absence of mandatory reconstitution, SDR allocations were equivalent to 100 per cent grants. At one of the early G-10 meetings, the Canadian Deputy, Plumptre, shocked others by making this assumption almost explicit when he suggested that decisions to distribute reserve assets ‘would presumably be taken just before Christmas’. But the response, then and later, was not to call for a proper interest mechanism to ensure that these were not gifts, but to argue that the benefits of receiving SDRs as gifts would be balanced by the potential ‘burdens’ to which participants in the scheme would be exposed.
In the 1960s, the Fund staff—I am sorry to say—was as much at fault as anyone else in treating SDR allocations as pure grants. This common misconception was responsible for much of the confusion in the semipublic debate that Ossola and I had in the fall of 1965 on the benefits of SDR allocations.8 In the absence of a market interest rate on the SDR, the obligation to accept SDRs for a payments surplus would indeed be ‘a burden’, and a case could thus be constructed for limiting the distribution of SDRs to countries whose balances of payments could be expected to oscillate between deficits and surpluses. It then took only two (counterfactual) assumptions—that all G-10 countries qualified under this test and that the developing countries were not in the habit of building up reserves—to find a logical base for restricting the new reserve system to the G-10. That proposition was, in the end, abandoned by the G-10 but (as the Ossola correspondence shows) not because of a proper design of the SDR system with a market interest rate, but because it became untenable politically.
The Ossola Report dismissed the question of the SDR interest rate in a simple paragraph: ‘The opinion has been expressed that payment of interest on all or some reserve claims would enhance their attractiveness as reserves, and some would make the rate variable in order to improve the flexibility of the scheme. Others, however, considered that, where a gold value guarantee is envisaged, the asset should be non-interest bearing’ (G-10 1965, para. 146).
The initial interest rate on the SDR was set at 1.5 per cent per annum, which was copied from the interest rate set in 1962 under the GAB. In 1962, that had not been a negligible interest rate for a gold-guaranteed asset, but that was no longer so by the end of the decade.9 By that time, many of the theoretical issues concerning the SDR had been clarified at a Fund-sponsored conference (IMF 1970) and the logic of moving the SDR interest rate to the market rate was recognized, at least by the technicians.
Once the SDR was severed from gold in 1974 and linked to a basket of currencies, the rate of interest was first raised to 5 per cent (about half the market rate of the basket) and then gradually to the market rate by 1980. This helped to clarify the benefits of SDR allocations. Countries with ready access to the world’s capital markets at the going rate now derive negligible benefits from SDR allocations—most of these countries have turned against further allocations, which has brought the SDR system to its current impasse. But countries with no access to capital markets, or that have to pay high premiums in such markets or for suppliers’ credits, still benefit from allocations.
But on another contentious aspect of collective reserve creation, viz. the link to development financing, disregard of the interest rate question continued to produce needless controversy. A large part of the developing countries’ case for the link rested on a zero SDR interest rate. As Ossola pointed out in 1972, ‘if SDRs carry a market type rate of interest, a link would entail a long-term loan rather than an outright grant’ (Ossola 1972). One of the very few points on which all participants in the debate on the link in the framework of the 1912–4 reform exercise could agree was that ‘the overall benefit of link allocations to developing countries might not appear very large’ at market interest rates, although these countries would at least get long-term credit at short-term rates without a risk premium (IMF 1974, p. 108). The developing countries—in particular those that were eligible for the zero interest credits of the International Development Assistance—could have taken the clue at this point and given up the struggle for the link. In fact, it took them until 1986 to do so.10
The muddle on the SDR interest rate is not an isolated example. On the contrary, it seems to fit into a pattern to use the rate on official assets and liabilities to pursue extraneous objectives. You will recall the proposal by Keynes (of all persons) to charge interest not only on debit but also on credit balances in the Credit Union in order to push creditors towards adjustments,11 a proposal repeated to great acclaim by the French in the C-20 (Solomon 1982, p. 254).
Subordination of interest rates to other considerations is not entirely a thing of the past—it has recurred in the Fund (in a less extreme form) in recent years. It had taken the Fund about forty years to achieve (near) market interest rates: a credit rate equal to the SDR rate and a debit rate slightly below the SDR rate (the small element of concessionality made possible by the fact that the Fund had sold about one third of its gold in 1976–80, adding the book value of SDR 35 per ounce to its usable resources). But, as some countries fell into arrears in the 1980s, the Fund, anxious to keep up its income and to build up some additional reserves started adding basis points to the rate of charge and subtracting basis points from the credit interest rate, thus increasingly running the risk of pricing itself out of the market on two sides at the same time.
5 A Substitution Account?
Three times in the past twenty years the world’s financial officials have laboured over the design of a substitution account—an account into which countries could deposit dollars (and perhaps other currencies) from their reserves against new SDRs or claims expressed in SDRs. The idea was studied as part of the 1972–4 effort to reform the international system, in 1978, as the counterpart of the decision to allocate SDRs, and again in 1979–80 in a period of widespread concern about a ‘dollar overhang’. All these efforts remained unsuccessful. The reform effort at substitution disappeared from the scene together with the other hobby horses of that failed exercise: ‘stable but adjustable par values’, asset settlement, the link, and reserve indicators. In 1978, it turned out that agreement on an SDR allocation could after all be reached without reduction in the dollar component in reserves through substitution. And the last, the most determined, and nearly successful, attempt to create a substitution account failed in April 1980 when no agreement could be reached on the ultimate distribution of the risks of such an account and, perhaps more important, when the dollar began its ascent that lasted until early 1985.
Against this background of failure, it is important to put on record the single successful instance of substitution in the Fund. It occurred well before there were SDRs and it was Ossola’s idea. Specifically, during the meeting of the G-10 Deputies in Rome in May 1966, Ossola, together with Governor Dewey Daane of the Federal Reserve System, approached me to explore whether a Fund transaction could be designed to solve the following problem. Italy considered that it had excess dollars and would like to deposit $250 million in the Fund. The United States was agreeable to the transaction. The operation was to be kept outside of the GAB, adopted in 1962, which would require agreement among the G-10. My reply was that this could indeed be handled in the Fund, provided that the transaction was executed by means of a series of steps allowed under the Fund’s Articles. Italy would lend the Fund an amount of lire equivalent to $250 million. The United States would buy this amount of lire from the Fund, and then use the lire to purchase $250 million in US dollars from Italy. These three operations would give Italy a transferable gold-guaranteed claim on the Fund instead of its $250 million in dollars;12 the United States would have a corresponding reduction in its reserve tranche (then still called ‘gold tranche’) in the Fund. While the Fund would describe the operation as a loan, it offered to find language that enabled Italy to describe it as ‘depositing dollars with the Fund’. It took a few more months to tie down the details of the operation, which took place in August 1966.13 In mid 1970, when Italy had a payments deficit, it used its claim on the Fund as a reserve asset by transferring it to Japan.
For the modest amount involved, thus, the attempt to use the mechanism of the Fund as a substitution account was entirely successful. It could have set an important precedent—although a more general use of a substitution facility would have had to overcome formidable difficulties. The United States had less than $500 million left for gold tranche drawings; would it have agreed to enter into a Fund standby arrangement in the credit tranches to accommodate other ‘depositors’? What would have been the Fund’s terms for such an unusual standby arrangement? Would it have been phased? None of these questions came up. The potential importance of the arrangement was barely noticed, and no other Fund member asked for the same treatment.14
But another feature of this substitution transaction is to be noted. Not only did Italy get a gold-guaranteed claim; the United States also accepted a gold-guaranteed liability. If the United States had agreed to the equivalent provision in later discussions on a substitution account (a liability expressed in SDRs), the negotiations in 1979–80 on such an account would have been materially simpler. As it was, the US insistence that its liability to a substitution account be expressed in dollars rather than SDRs made the problem of the financial balance of that account insoluble, thus posing the politically unmanageable problem as to how to distribute among the members any loss that such an account might incur.
It does not seem likely at present that the concept of a substitution account will again come to the fore as a conceivable answer to a real problem—or might it, when the European Central Bank (ECB) finds that the merged dollar holdings of the EC central banks far exceed its reserve needs? In any event, one lesson deserves to be carefully stored for possible future use, namely that expressing the claims on such an account in SDRs—the arrangement which the United States absolutely rejected at the time—would come very close to providing the magical 50/50 distribution of risk between creditors and the United States as debtor.15
6 ‘Objective Indicators’ for Exchange Rate Adjustment
The trials of the Summer and Fall of 1971 had brought home to the United States the asymmetry of an adjustment system based on convertibility only: the deficit country is forced to take adjustment action, when reserves run out; the surplus country, on the other hand, while it may suffer an inflationary impact from the inflow of excessive reserves, is not put under the same necessity to adopt radical policy measures. Keynes had made the point in 1943; by 1972 the United States was ready to act upon it (Solomon 1982, p. 242).
In the first available forum for reform of the system—the preparation by the executive directors of the IMF, at the request of the governors, of a report ‘on the measures that are necessary or desirable for the improvement or reform of the international monetary system’—the US Director proposed a symmetrical system triggered by objective reserve indicators. Under this system, internationally agreed upper and lower limits for reserves would be established for each country to serve as ‘objective indicators of par value changes’. If the reserves reached one or the other limit, ‘this would be regarded as an indication, or as prima facie evidence, that a parity change was needed’.16
The reaction of the Executive Board to this proposal was overwhelmingly negative—not to the concept of symmetry, but to its implementation by reserve indicators. The Fund itself had an unsatisfactory experience with reserve statistics as a determinant of a far less important obligation of member countries, viz. the amount of past drawings they would have to repurchase. In the original Articles of Agreement this was linked to reserve levels and reserve increases, but members that wanted to delay repurchase had found it possible to manipulate reserve statistics. As part of the second amendment, the entire automatic provision was dropped and replaced by more judgmental rules.
More generally, the discussion made clear that most countries would not want to attach a decision to change their exchange rate, even presumptively, to reserve movements that might reflect short-term rather than fundamental developments, although they had to admit that experience with the alternative to indicators—’balanced assessment’—had not been impressive.
When the reform discussion next moved to the Deputies of the C-20, the indicator proposal was again put on the table (C-20 1973, pp. 160–74). By now, some measure of flexibility had been introduced: the indicators would not give automatic, ‘but strong presumptive’, signals for adjustment, and some room was made for adjustment by measures other than changes in the exchange rate.
Even in its modified form, the American proposal could hardly be expected to find wide support. But an Italian counterproposal did much to avoid a head-on clash on indicators as such (Ossola and Palumbo 1972). In a paper circulated two days before the US paper, the Italian Deputies accepted the need for indicators. As they put it: ‘The discretionary assessment of fundamental disequilibria has been tried, and it has been found wanting.’ But they rejected reserve indicators as ‘being too easily manipulated by the monetary authorities’, giving rise to friction over their interpretation, and subject to random disturbances. Instead, they suggested the cyclically adjusted basic balance as ‘the indicator least subject to official tinkering and most likely to provide correct information on the underlying trend of the balance of payments’. And they added that any information provided by indicators would need to be supplemented by analysis of the current and prospective situation.
The Italian paper had the beauty of integrating the American proposal into a wider range of indicator plans with differing degrees of automaticity and judgment. A Fund staff paper comparing the two indicators from a variety of angles pushed further in this direction (IMF 1985b, pp. 57–67). Conceivably, if other elements of the reform exercise had fallen into place, this could have led to the adoption of some kind of judgmental indicator component as part of the reformed system. But the reserve indicator as a trigger of exchange rate changes could never have made it.17 That, I am sure, was clear to Rinaldo Ossola in 1972. And a twenty-year jump to the present will provide confirmation, if any were needed. Article 104c of the Maastricht Treaty describes the consequences of an EC member state exceeding the fiscal objective indicators of that treaty. Although the EC members are pursuing an ‘ever closer union’, the punishment of fiscal transgression even in stage three is restricted to reports, the need to publish more information before issuing bonds, a possible cutoff from the European Investment Bank, possible fines and (our old standby since the days of Keynes) mandatory non-interest-bearing deposits! (Kenen 1992, pp. 73–4).
7 A Moral?
Is there a moral to be drawn on the basis of these snapshots of the international monetary system, and of those who laboured at it, in the 1960s and 1970s? One lesson no doubt is that there was much lost motion, in part because of a true lack of understanding (the nature of the SDR, the role of interest rates), in part also because countries pursued political objectives rather than improvements of the system or even recognizable national economic or financial interests. The specific problems caused by these shortcomings are no longer relevant; at best, recognition of them may help future monetary negotiators to lose a little less time in dead-end streets. But the fundamental issues of adjustment and liquidity are still with us—and the constructive labour that Ossola and his colleagues performed in past decades will continue, not only to inspire us, but also, in a practical way, to guide us.
Committee of Twenty. 1973. ‘The US Proposals for Using Reserves as an Indicator of the Need for Balance of Payments Adjustment’Document of Deputies (27 November 1972) reprinted in Council of Economic Advisers Economic Report of the PresidentJanuary.
CordenW.M.1991. ‘Does the Current Account Matter? The Old View and the New’inJ. A.Frenkel and M.Goldstein (eds.) International Financial Policy Essays in Honour ofJ.J.PolakWashington DCIMF.
de VriesM.G.1976a. The International Monetary Fund1966–71vol. IWashington DC.
de VriesM.G.1976b. The International Monetary Fund1966–71vol. IIWashington DC.
FlemingJ.M.1971. ‘The SDR: Some Problems and Possibilities’IMF Staff PapersMarch.
Group of Ten1964. Ministerial Statement of the Group of Ten Annex prepared by DeputiesParisAugust.
Group of Ten1965. Report of the Study Group on the Creation of Reserve Assetsn.p.,May.
HirschF.1971. ‘SDRs and the Working of the Gold Standard’IMF Staff PapersJuly.
HorsefieldJ.K.1969. The International Monetary Fund1945–65vol. IWashington DC.
International Monetary Fund. 1970. International Reserves—Needs and AvailabilityWashington DC.
International Monetary Fund. 1974. International Monetary Reform Documents of the Committee of TwentyWashington DC.
International Monetary Fund. 1985a. ‘Reform of the International Monetary System’Report by the Executive Directors to the Board of Governors (18August1972) reprinted inM.G.de Vries (eds.) The International Monetary Fund1972–78 Cooperation on Trialvol. III.
International Monetary Fund. 1985b. ‘Reserves and Basic Balances as Possible Indicators of the Need for Payments Adjustment’ Document of DeputiesCommittee of Twenty, (10January1973) reprintedinM.G.de Vriesibid.
International Monetary Fund. 1986. ‘Communique of the Interim Committee’ IMF Survey21April.
KenenP.B.1992. EMU after MaastrichtWashington DC,Group of Thirty.
League of Nations. 1944. International Currency ExperiencePrinceton.
LindbeckA.1970. ‘Comment on Sohmen’inInternational Monetary FundInternational Reserves—Needs and AvailabilityWashington DC.
MarrisS.1985. Deficits and the Dollar: the World Economy at RiskWashington DC,Institute for International Economics.
Organization for Economic Cooperation and Development. 1966. The Balance of Payment Adjustment Process Working Party No. 3 of the Economic Policy CommitteeParis,August.
Organization for Economic Cooperation and Development. 1976. Activities of OECD in 1975 Report by the Secretary GeneralParis.
OssolaR.1972. Statement in the Second Meeting of the Deputies of Committee of Twenty25November.
OssolaR. and PalumboS.1972. Adjustment Process and Exchange Rate Mechanism Document of Deputies Committee of Twenty25November.
PolakJ.J.1988. ‘The Impasse Concerning the Role of the SDR’ inWietzeEizenga et al. (eds.) The Quest for National and Global Economic StabilityDordrecht.
SohmenE.1970. ‘General Reserve Supplementation: Some Central Issues’ inInternational Monetary FundInternational Reserves—Needs and AvailabilityWashington DC.
SolomonR.1982. The International Monetary System1945–81New York.
SolomonR.1991. ‘Changing Perspectives on the International Monetary System’ in
J.A.Frenkel and M.Goldstein (eds.) International Financial Policy: Essays in Honour ofJ.J.PolakWashington DC,IMF.
US Treasury. 1988. Report to the Congress on International Economic and Exchange Rate Policy,15October.
Reprinted with the permission of Cambridge University Press, from Peter B. Kenen, Francesco Papadia, and Fabrizio Saccomanni (editors), The International Monetary System: Proceedings of a Conference Organized by the Banca d’ltalia [in honor of Rinaldo Ossola], (Cambridge University Press, 1994), pp. 19–34.
Editor’s note: The subtitle “Are They Still Relevant?” was in the original paper presented at a conference in July 1992, but it was omitted in the published proceedings.
OECD 1976, pp. 17–19. At the time this was known as the Kissinger Plan. The total of quotas at SDR 20 billion about equalled the sum of the quotas of the OECD members in the IMF.
No reference to the plan can be found in Solomon (1982).
The first G-10 Deputies Report (chaired by Robert Roosa) had managed only barely, in a backhanded way, to recognize the exchange rate as an instrument of adjustment. The report did not include it in its listing of six instruments of economic policy, but then added: ‘Such instruments must be employed with proper regard for obligations in the field of international trade and for the IMF obligation to maintain stable exchange parities which are subject to change only in cases of fundamental disequilibrium’ (G-10 1964, para. 6).
‘Proposal by the Managing Director on the Allocation of Special Drawing Rights for the First Basic Period’ (1969), reprinted (IMF, 1970, pp. 491–509).
Use of GAB resources for a non-participant requires a finding of ‘an exceptional situation … that could threaten the stability of the international monetary system’ (GAB decision of 24 February 1983, para. 21 (b)).
Polak’s letter to Ossola, 3 November 1965 and Ossola’s letter to Polak, 24 November 1965 (documentation for the first joint Meeting of the IMF Executive Directors and the G-10 Deputies, Washington DC, 28–30 November 1965, unpublished).
This is evident from the consensus among economists who discussed the matter at the time. Sohmen described credit at the SDR rate as ‘almost without charge’ (Sohmen 1970, p. 18). Lindbeck speaks of ‘gratis reserves’ (Lindbeck 1970, p. 36). Fleming describes the SDR allocation mechanism as ‘reserves [being]rationed out among countries virtually as a gift in predetermined proportions’ (Fleming 1971, p. 36). Hirsch sees the full fruition of the SDR at risk from ‘an artificially low interest rate’ (Hirsch 1971, pp. 245–6).
When they agreed to para. 7 in the April 1986 Communique of the Interim Committee which stressed ‘the monetary character of the SDR, which should not be a means of transferring resources’ (IMF 1986, p. 116).
‘Are we to love, honour, cherish, and thank or to kick in the bottom the blokes who hold bancor?’ (Cited in Horsefield 1969, vol. I, p. 19).
The transfer was subject to Fund consent, as under the GAB, but Italy had every reason to expect the Fund to give its consent when asked.
The text of the exchange of letters setting out the condition of Italy’s loan is found in de Vries 1976 b, vol. II, pp. 211–13.
The Fund’s history reports the operation in a brief paragraph that misses entirely its systemic significance: ‘On only one occasion did the Fund engage in bilateral borrowing. In August 1966, when the United States sought to draw $250 million in Italian lire, the Fund’s holdings of lire were down to $70 million. The Fund thereupon borrowed the $250 million in lire from Italy under a special arrangement similar to, but outside of, the GAB’ (de Vries 1976 a, I, p. 376).
With the weight of the dollar at 40 per cent of the SDR basket, the effect of expressing the claims of the account, like those of its liabilities, in SDRs would 13 have meant that the United States would have borne 60 per cent, and the creditors of the account 40 per cent, of the exchange risk.
Report by the IMF Executive Directors to the Board of Governors, 18 August 1972 (reprinted in IMF 1985a, vol. HI, pp. 19–56). For the discussions leading up to the report, see de Vries 1976a, vol. I, pp. 130–7. Note that Solomon gives no inkling that at least this part of the ‘Volcker plan’ got an airing well before the famous Shultz speech at the 1972 Annual Meeting and the submission of the plan to the Deputies of the C-20 in November 1972.
As Williamson points out, that indicator also suffered from the handicap that it would, in a most unwelcome way, advertise impending par value changes (Williamson 1977, p. 182).