CHAPTER 14 The Retreat of Bilateralism

International Monetary Fund
Published Date:
February 1996
  • ShareShare
Show Summary Details
Margaret G. de Vries

WHEN THE MACHINERY for international monetary cooperation was being set up in 1943 and 1944, bilateral trade and payments agreements were not as extensively used as they were later to be. During the interwar period, these kinds of agreement had been employed mainly by Germany and by countries in Central and Eastern Europe.1 However, because bilateral agreements had been manipulated so as to promote exports, and hence domestic employment, and to protect agriculture and manufacturing industries against imports that threatened to increase domestic unemployment, bilateral agreements had come to be regarded as synonymous with aggressive commercial policies. Another difficulty with bilateral agreements, as revealed by earlier international experiences, was that imports might be curtailed in order to force the partner to repay past debts.2 A few economists, anticipating in the postwar period the widespread unemployment that had prevailed before the war, were fearful that bilateralism might become more prevalent.3 Consequently, one of the major purposes of the Fund, as noted in Chapter 2, is to assist in the establishment of a multilateral system of payments in respect of current transactions between members.


Rapid spread after World War II

Because virtually all non-dollar countries had balance of payments difficulties in the late 1940’s and early 1950’s, bilateral payments arrangements spread more rapidly and became much more widespread than in the 1930’s. Such arrangements provided a mechanism by which the partner countries could effect their reciprocal current settlements with a minimum use of convertible exchange and gold. In a typical case, the two central banks opened accounts in their respective currencies in each other’s name, and all permitted payments were channeled through these agreement accounts; settlements in convertible currencies or gold had to be made only for balances within limits agreed in advance, that is, for amounts specified in what were called the “swing.”4

The models for postwar payments agreements were the Belgo-Dutch (1943) and the Anglo-Belgian (1944) agreements.5 Toward the end of 1944, the United Kingdom started negotiations with a number of continental countries with a view to the signing of payments agreements. Several of these were concluded between October 1944 and May 1946, with—in chronological order—Belgium, Sweden, France, Denmark, the Netherlands, Czechoslovakia, Norway, Switzerland, and Portugal.

What happened to the number of agreements into which Switzerland entered serves as an illustration. At no time before World War II did Switzerland have more than 14 clearing systems in force; but in 1945 and 1946 alone Switzerland signed 12 new payments agreements, and by the end of May 1947 it had 19 such agreements. Other countries had similar experiences. By the end of 1946 France was already a partner to 20 payments agreements, Belgium to 17, and the Netherlands to 14. By June 1947, the number of bilateral payments agreements in effect was estimated at 200.6

Shortly thereafter came a further series of agreements by the United Kingdom. While using bilateral payments agreements after the war, the United Kingdom had at first made little use of bilateral trade agreements. However, following the failure to establish sterling convertibility in mid-1947, it concluded bilateral trade agreements with countries with which it already had payments agreements. In 1948, Germany started to enter into trade and payments agreements with Latin American countries. Also in 1948, the Supreme Commander for Allied Powers in Japan embarked on a policy of bilateralism. A number of agreements were concluded between Japan and European countries, followed in 1949 by the first series of trade and payments agreements between Japan and Latin American countries.

In addition, in the late 1940’s European countries rounded out their systems of agreements by concluding agreements with their smaller European trading partners. The establishment of the sovereign states of Israel in 1948 and Indonesia in 1949 also led to a series of bilateral payments agreements.

Similarly, the number of payments agreements of the Latin American countries gradually became larger than the number of agreements which they had had before World War II. European countries arranged a number of payments agreements with those Latin American countries that were not in the dollar area—that is, with Argentina, Brazil, Paraguay, and Uruguay. (The dollar area included the Republics of Central America and the Caribbean, and Bolivia, Chile, Colombia, Ecuador, Mexico, Peru, and Venezuela.) European countries were also anxious to enter into payments agreements with Latin American countries that were in the dollar area, as these enabled European countries to purchase, for inconvertible currencies, commodities for which they previously had been paying dollars, and facilitated their exports to these countries. The Latin American countries in the dollar area, in turn, were induced to enter into agreements with European countries in order to overcome the effects of discriminatory import licensing in countries with inconvertible currencies, and thus to recapture or to expand markets in Europe. European countries tended to discriminate against any Latin American country that did not enter into a payments agreement, particularly with regard to commodities which the former were able to obtain without financial problems from their dependent or associated territories. There were relatively few intra-Latin American bilateral trade and payments agreements.7

Not only did inconvertibility and the virtually universal existence of balance of payments deficits induce the onset of so many bilateral agreements, but bilateralism, once started, tended to generate more bilateralism. Once a number of important payments agreements had been signed, countries had an incentive to conclude additional agreements. First, because of the “swing” arrangements, bilateral agreements offered a means of obtaining or giving a tied credit. Second, countries that already had payments agreements adopted import licensing techniques and policies discriminating against the dollar area and favoring their bilateral payments and trade agreement partners; hence rather than be discriminated against, even some countries in the dollar area became partners to new agreements.

Dimension of the problem by 1954

At the end of 1954, the Fund began to take measures to reduce the use of bilateral agreements by its members. At the time, over 400 bilateral payments and similar arrangements were believed to be in force. Of these, approximately 235 were between pairs of European countries. Next in numerical importance was the network of approximately 100 agreements between European and Latin American countries. European countries also had about 60 agreements with countries in the Middle East and the Far East. Finally, a much smaller number of intraregional agreements existed within Latin America and among the Middle Eastern and Far Eastern countries. At least 18 countries were maintaining 15 or more bilateral payments agreements each: Argentina, Belgium-Luxembourg, Brazil, Denmark, Egypt, Finland, France, Germany, Greece, Israel, Italy, Japan, the Netherlands, Spain, Turkey, Uruguay, and Yugoslavia.

The number of agreements alone is not, however, a sufficient measure of their importance. The estimated amount of trade affected is perhaps even more indicative. The total value of exports in 1954 by members of the Fund with countries with which they had bilateral agreements reached $6.4 billion, which was about 8.5 per cent of aggregate world exports. For some individual countries, the significance of bilateralism for their exports was much higher. Over half of the exports of Argentina, Brazil, China, Egypt, Finland, Uruguay, and Yugoslavia were conducted with countries with which they had bilateral agreements. From one quarter to one half of the exports of Chile, Iceland, Indonesia, Iran, Israel, Japan, Korea, Spain, and Turkey, and from 10 to 25 per cent of the exports of another 15 countries—including several large ones in Western Europe (Austria, Denmark, France, Germany, Italy, the Netherlands, and Norway)—were with bilateral partners.

At the end of 1954, 41 out of the 56 members of the Fund conducted at least 10 per cent of their exports under bilateral arrangements. Much of the trade between Western European countries and the Soviet bloc in Eastern Europe was still conducted bilaterally, as well as a large part of the trade between Western Europe and Latin America and a substantial part of the trade between Latin America and Asia.

On the other hand, there was evidence that some bilateral arrangements tended to be construed less strictly. European countries had developed techniques to widen the scope of their payments arrangements. Although the United Kingdom had been a partner to many bilateral agreements, these usually gave the partner access to the whole sterling area. The existence of the sizable sterling area, the wide international use of sterling, although inconvertible, for financing trade and payments between countries unwilling or unable to finance their trade or payments in dollars, and the widening transferability of sterling, accordingly made the payments agreements of the United Kingdom much more plurilateral than bilateral.

In addition, with the establishment in 1950 of the European Payments Union (EPU),8 the payments agreements between pairs of participating European countries were left with nothing but a technical function.9 Since all member currencies had become transferable among EPU members, they were all equally “hard” (or “soft”) to any one of them. In these circumstances, while some European countries were concluding new bilateral payments agreements, the general tendency after 1953 was for the important trading countries to place a decreasing reliance upon bilateralism.

Propitious circumstances for Fund action

Not only had European countries, by 1954, decreased their reliance on bilateral agreements, but other circumstances were also propitious for the Fund to take action against bilateralism. By 1953–54, the balance of payments positions of the European countries vis-à-vis their bilateral partners outside Europe, including Latin America, the Middle East, and the Far East, had become very different from that immediately after the war. Whereas earlier the Latin American and other countries had acquired substantial balances of European currencies, especially sterling, and their inability to convert these currencies had given them a strong incentive to form bilateral arrangements with Europe, by 1953–54 accumulated balances had been used up.

Additionally, there were limits to the amount of bilateral credits that European countries were willing to grant to countries outside Europe. When these limits were reached, their bilateral partners had either to restrict imports under their agreements or to pay acceptable currencies, such as U.S. dollars. When bilateral credit margins were exhausted and settlements with the bilateral partner had to be made in dollars, the inducement on the part of countries outside Europe to discriminate against dollar imports through bilateral agreements with European countries was greatly reduced.

At the same time, the more liberal trade policies emerging in Western Europe, as described in Chapter 12, made less acceptable the artificial and noncompetitive prices which frequently had to be paid for the commodities which the European countries imported under bilateral agreements. Many creditor countries started to restrict their exports, and sometimes refused to convert export proceeds into local currency. Furthermore, the reduced scope of government purchasing and the relaxation of restrictions on imports into certain countries increased the difficulties of operating some payments agreements.10 Bilateral agreement currencies were being sold at discounts in an effort to evade the strict bilateral balancing of accounts.11

All these developments had begun to reduce the flow of trade taking place under bilateral arrangements.

At the same time, it could not be assumed that bilateralism would come to an end easily, even though convertibility might be imminent. The Japanese authorities had, for example, already suggested to the staff of the Fund that a return to convertibility might decrease rather than increase the volume of world trade; therefore, bilateral agreements would be necessary to prevent increased unemployment. The subsequent discussion in the Executive Board on the occasion of the annual consultation with Japan had revealed that Japan was not alone in such fears. In several countries bilateral agreements had become motivated more by the desire to safeguard export industries than by balance of payments considerations.


Staff initiative

In these circumstances the staff of the Fund, toward the end of 1954, decided that it was time to bring to the Executive Board the general issue of bilateral payments agreements. This staff action may have been prompted in part by the facts that the United Kingdom had, in June 1954, suggested to the OEEC Ministerial Group on Convertibility a concerted move by OEEC countries to abolish bilateral agreements and that the United Kingdom’s paper had been accepted by the Group in September 1954 as a basis for study. But a stronger reason for the decision of the Fund staff to suggest that the Fund take action against bilateralism was that, in the improving world payments situation, bilateralism seemed to be much less warranted than it had been earlier. Even more compelling was the belief, held by some members of the staff and by some Executive Directors, that a stronger line by the Fund against bilateralism might hasten the establishment of convertibility of European currencies.

The staff first suggested a three-phase policy, the timing of which would depend on the progress made by the major countries toward the convertibility of their currencies. The first phase was to extend from 1954 until the several countries then approaching convertibility had actually taken this decisive step. In this phase, the Fund would use its annual consultations to explore in greater detail than hitherto the scope and content of bilateral agreements and the reasons for their retention. Such action, it was argued, not only might help to shorten the life of some agreements, but also would build up an important body of experience, and enable the Fund to move effectively in the following phases.

The second phase would arise if not all the main currencies became convertible. In these circumstances some of the countries whose currencies were newly convertible might be faced with new restrictions by other industrial countries, with currencies still inconvertible, which continued to practice discrimination. The former might then feel the need to enter into some bilateral agreements. The Fund would have to pay careful attention to any restrictions which it was asked to approve under Article VIII by countries opting to accept the obligations of this Article. The reasons given by countries for not making their currencies convertible would also have to be carefully scrutinized. But it was recognized that this second phase might never occur.

The third phase would be reached when the largest members of the Fund—and particularly all the industrial countries—had convertible currencies. The countries whose currencies remained inconvertible might still find it useful in certain instances to conclude bilateral agreements among themselves, but these were expected to be of minor importance in world trade. During this third phase the Fund would have to ensure that, except for such minor exceptions, bilateralism would be abandoned by all countries; its jurisdiction under Article VIII would provide it with the necessary means.

The Board had no difficulty in accepting that bilateral payments agreements, having a restrictive character and an inherent bias toward discrimination, clearly fell within the Fund’s jurisdiction. However, the emphasis to be given to bilateralism was subject to considerable debate. A few Directors believed that the staff’s approach, which based the need for bilateralism on the absence of convertibility and which would intensify the Fund’s action against bilateralism as steps were taken toward convertibility, was too narrow. These Directors pointed to additional reasons why countries resorted to bilateralism. Primary producing countries used bilateral agreements because they had special problems in sustaining their basic exports or stabilizing their export earnings, or both. Other agreements were used to avoid or to repay commercial debts. Bilateral agreements also had a function for trade with countries which were not members of the Fund, such as state-trading countries.

The majority of Directors—including those from the United States, Canada, the United Kingdom, and Germany—believed, nonetheless, that the only practical course for the Fund was the pursuit of the aims of the Articles, including that of multilateralism. They advocated an even stronger line against bilateralism than that proposed by the staff. The Directors therefore agreed that the staff should ascertain the nature and extent of the bilateral arrangements then prevailing and draft a decision for the consideration of the Board.

Decision of 1955

The staff reported back to the Directors in May 1955. The Western European countries had been found to be in the process of multilateralizing their bilateral payments relationships by resort to transit trade, by acceptance of third currencies such as transferable sterling and transferable deutsche mark, and by granting permission to effect transfers to and from certain payments agreement partners through the accounts established under payments agreements with other countries. However, a significant measure of bilateralism remained in the Eastern European countries, the Latin American non-dollar countries, and the Far Eastern and Middle Eastern countries. The staff still related the proposed policy of the Fund to convertibility. With the attainment of convertibility of the major European currencies, the balance of payments justification for bilateralism would diminish, if not disappear. Another reason why the staff tied the proposed policy against bilateralism to the prospects for convertibility was its belief that countries with newly convertible currencies would require safeguards against the use of bilateral arrangements by other countries.

The ensuing discussion in the Executive Board revealed that many Directors did not accept the staff’s view that convertibility would result in the abolition of bilateral agreements by members. These Directors feared that members would still retain bilateral agreements for commercial reasons. Since members would always defend bilateralism on balance of payments grounds, would not an emphasis by the Fund on the balance of payments justification for bilateralism in effect make all bilateralism defensible? What criteria could be used to determine a “balance of payments need”? What would be accomplished if the Fund’s efforts to wipe out bilateral payments agreements were accompanied by an intensification of bilateral trade agreements? Was it not necessary for the Fund and the GATT to work together? Could not the Fund do more than merely explore with members their need to continue bilateral arrangements and the possibilities of removing these arrangements? Should not countries in surplus as well as those in deficit review their policies so as to improve the system of multilateral payments?

On the other hand, a number of Directors were concerned about too strong a push against bilateralism. They reiterated their view that the composition of the exports of a number of the less developed countries made them particularly dependent on bilateralism, and that trade with state-trading countries posed some special problems. These concerns led to the inclusion in the decision finally taken of a statement that

in its examination of the justification for reliance on such bilateral arrangements the Fund will, without excluding other considerations, have particular regard to the payments position and prospects of the members concerned.

The decision against bilateralism was adopted in June 1955.12 The Board strengthened the text suggested by the staff. Where, for example, the draft had called on members “eventually” to eliminate bilateral arrangements, the Board substituted “as rapidly as practicable.”

The essential elements of the decision were that the Fund’s policy on bilateralism was an integral part of its policy on restrictions, that the Fund urged the full collaboration of all its members to reduce, and to eliminate as rapidly as practicable, reliance on bilateralism, and that the Fund would explore with all countries which were parties to bilateral arrangements which involved the use of exchange restrictions the need for continuation of these arrangements, the possibilities of their early removal, and ways and means, including the use of the Fund’s resources, by which the Fund could assist in this process. A letter from the Managing Director transmitting the decision to members indicated that bilateral payments agreements would receive special attention in the forthcoming round of Article XIV consultations; as a preparation, members were requested to review their bilateral agreements.13


On three subsequent occasions the Fund reaffirmed that it would persist in its efforts to secure the speedy elimination of bilateral arrangements. Further ways in which this elimination might be accomplished were made explicit.

Specification of immediate goals

The first reaffirmation of the Fund’s general policy against bilateralism came in mid-1956, when the Directors reviewed the progress that had been made since the general decision of the year before. The staff was able to report that there had been considerable progress in widening the payments arrangements under existing bilateral agreements in the intervening year. The United Kingdom and Germany had, for example, facilitated the use, under their bilateral payments agreements, of certain inconvertible currencies of third (nonpartner) countries. The “Hague Club” had been formed to multilateralize the payments relations of most Western European countries with Brazil. The “Paris Club,” just being instituted, would effect virtually the same multilateralization of payments between Europe and Argentina (Chapter 12).

Nonetheless, in their consultations with the Fund several member countries were advancing a number of reasons why they had to go slowly in abolishing their bilateral arrangements. Some argued that they needed special arrangements vis-à-vis the state-trading countries of Eastern Europe. Even more common was the contention that bilateral arrangements were necessary to protect or promote exports. Some countries feared that the elimination of a bilateral agreement with a specific partner would put them at a disadvantage compared with other suppliers which still maintained preferential access to the markets of that partner. For a few countries, the inadequacy or unsatisfactory composition of foreign exchange reserves made bilateral swing credits still of some importance. For one or two others, bilateral agreements were the means for the gradual unfreezing of capital claims.

In these circumstances the staff proposed, and the Board agreed, that during the 1956 round of consultations the Fund should continue to press the issue of bilateralism. Specific goals were to be the termination of those bilateral arrangements which no longer seemed essential; the renegotiation of unnecessarily restrictive or discriminatory agreements; the replacement of bilateral swing credits by other forms of credit, including the Fund’s resources; and the abandonment of bilateralism as the use of transferable currencies became possible. It was also agreed that, where the situations so merited, the Fund would make a simultaneous approach to two or more member countries, perhaps through their Executive Directors, to end a particular agreement.

The policy further elaborated

Some three and a half years later, in October 1959, the Fund’s policy on bilateralism was again restated. The occasion was the decision on discrimination for balance of payments reasons (Chapter 13). By then bilateral payments agreements had been considerably reduced, and in the Western European countries virtually eliminated except for those with state-trading nations. It had become the Fund’s view that there was no longer any justification for bilateral agreements between countries having externally convertible currencies. However, several countries—although by no means all—were still using bilateral payments agreements for trade with Soviet bloc countries, and with other state-trading countries, most of which were not members of the Fund.

This situation called for a new elaboration of the Fund’s policy. The original decision in 1955 had not differentiated those bilateral payments agreements arranged between two members of the Fund from those agreements arranged by a Fund member with a nonmember. This distinction had now become important. The Board decision on discrimination of October 1959, therefore, clarified the Fund’s position on bilateral payments agreements between members and nonmembers. Member countries were expected to proceed with all feasible speed to eliminate discrimination against other members, including discrimination arising from bilateralism. However, some concession was thought necessary regarding the bilateral arrangements of members of the Fund with nonmembers. For such arrangements the Fund stated that it would be “prepared to consider whether balance of payments considerations would justify the maintenance of some degree of discrimination, although not as between countries having externally convertible currencies.”14

This specification represented an elaboration—not a change—in the Fund’s attitude toward bilateral payments arrangements. As expressed by the Chairman of the Executive Board at the time, the statement being made by the Fund in 1959 was that those members still engaging in bilateralism should be mindful of the fact that many other members had been able to eliminate reliance on bilateralism, and aware also of their obligation not to harm the interests of other members of the Fund by their bilateral arrangements, including those with nonmembers.

At the same time the Fund would not concern itself directly with bilateral payments agreements with nonmembers; the 1959 decision did not condemn such agreements. Rather, stress was laid on the discrimination against other members of the Fund that might be involved in any bilateral arrangements with nonmembers. The question whether bilateral agreements with state-trading countries did or did not actually involve discrimination against members remained open and, as discussed below, was to come up on various occasions in the next few years.

The third occasion on which the general problem of bilateralism was reviewed by the Executive Board was in June 1960, when the Board determined the conditions necessary for a country to accept the obligations of Article VIII (Chapter 13). Bilateralism was specifically covered by the preamble to the decision on Articles VIII and XIV, which states that

previous decisions taken by the Fund, such as those on multiple currency practices, bilateral arrangements, discriminatory restrictions maintained for balance of payments purposes, and payments restrictions for security reasons, indicate the Fund’s attitude on these matters.15

It was further agreed that bilateral payments agreements were included as “measures” in the statement in paragraph 2 of that decision:

As regards measures requiring approval under Article VIII and maintained or introduced for nonbalance of payments reasons, the Fund believes that the use of exchange systems for nonbalance of payments reasons should be avoided to the greatest possible extent, and is prepared to consider with members the ways and means of achieving the elimination of such measures as soon as possible.16

Even more significant was the consensus of the Directors that, because bilateral payments agreements involved both restrictions and discrimination, members should have eliminated all such arrangements before moving to Article VIII status, or should be prepared to get rid of them within a stated time, e.g., at the expiry of the current agreement.


The general policy of the Fund against bilateral payments agreements, like other policies of the Fund, has been implemented on a case-by-case basis. In particular, as countries consulted the Fund under Article XIV, the Fund has been able to apply the above pronouncements to specific country situations. Since 1955 the reports on the staff’s discussions with member countries have regularly contained descriptions of bilateral payments agreements, the results of the discussions pertaining to bilateralism between the representatives of the member government and the staff, and a staff appraisal of the possibilities of ending any bilateralism still remaining.

The Executive Board decisions in Article XIV consultations have also usually included recommendations as regards the member’s bilateral payments agreements. During 1964 and 1965, for example, out of 62 countries for which Article XIV consultations were concluded, 42 had some bilateral agreements; of these the Fund made recommendations on bilateralism to 27. The countries for which no recommendations were made were mainly those with only one or two agreements, sometimes involving an element of doubt about whether the arrangements could be considered as bilateral payments agreements, and sometimes covering a negligible volume of transactions. There were also a few countries which had agreements only with nonmembers and these were not important in terms of the total amount of trade.

The recommendations made to members have varied, depending on the number of agreements, the proportion of trade financed under them, the possibility of alternative outlets for exports, the degree of discrimination practiced on the import side, and the size of the accumulation of inconvertible balances. Consideration has also been given to the measures which the government was taking to deal with the problems which gave rise to the reliance on bilateralism. When there were objectionable features in the exchange systems—for example, an unrealistic exchange rate supported by heavy dependence on restrictions—the recommendations have frequently put major emphasis on these factors, and the removal of bilateralism has been linked to the need for more fundamental reform. Where progress has already been made in removing bilateral agreements, the recommendation for further action has been less forceful.

A second way in which the Fund’s general policy on bilateralism has been implemented is through the exercise of the Fund’s powers of approval under Article VIII. As countries have assumed the obligations of Article VIII, additional pressure has been applied. When the European countries began to undertake the obligations of Article VIII in the early 1960’s, the number of bilateral payments agreements which they still retained were few, and these agreements no longer reflected inconvertibility or balance of payments problems. The arrangements being maintained between Belgium-Luxembourg and the Congo, Rwanda, and Burundi, and those between the Netherlands and Indonesia, for example, were outgrowths of past relations with former overseas territories. When Belgium and the Netherlands accepted Article VIII obligations, the Fund approved these agreements.

When discussing members’ plans to move to Article VIII, the Fund also reviewed the bilateral payments arrangements maintained by some of these countries with nonmembers. Austria, Belgium-Luxembourg, France, Italy, the Netherlands, and Sweden all argued that their bilateral arrangements with non-members did not result in restrictions or discrimination against any member of the Fund. Sweden argued that its payments arrangements were, in fact, supplementary to parallel bilateral trade agreements, and that most imports from nonmembers were subject to licensing requirements while most imports of similar goods from other countries were free from such requirements. As a consequence, there was no mechanism by which the authorities could favor imports from bilateral countries to the detriment of imports from Fund members.

Still a third way in which the Fund has implemented its policy on bilateralism has been through stand-by arrangements. These arrangements have involved commitments to carry out the Fund’s policy regarding the use of bilateral agreements; however, in all but those exceptional cases where bilateral payments agreements have dominated the restrictive system, commitments concerning bilateralism have been of the second order of priority and not central elements in the program for which the stand-by arrangement was granted.17 Frequently the Fund has expected that general programs for achieving internal and external equilibrium would ensure a substantial improvement in the balance of payments and would, in due course, permit elimination of undesirable restrictive practices, including bilateral agreements.

By the end of 1965, the Executive Board had approved 167 stand-by arrangements with 50 governments; 104 of these had been concluded with 37 member countries maintaining bilateral payments agreements, some of which still had several agreements. There had also been 31 letters of intent from 17 countries, submitted in support of requests for stand-by arrangements, which contained specific policy statements regarding bilateral payments agreements. Some of these stated the government’s intent to eliminate bilateral agreements “shortly” or within a specified time period; more stated “as soon as possible,” or words to that effect. In 1965 two stand-by arrangements had “binding commitments” on bilateralism; many have included binding commitments with respect to the introduction of exchange restrictions, which have thus inhibited the introduction of bilateral agreements.


Well before the end of 1965 the use of those bilateral payments agreements which stemmed from the inconvertibility of major currencies had all but disappeared. This elimination of bilateralism had come in surges. By the mid-1950’s, European bilateralism was on the wane. The establishment of the “Hague Club” and the “Paris Club” in 1955 and 1956 led soon after to the elimination of a group of bilateral agreements between Brazil and Argentina, on the one side, and European countries on the other. By the late 1950’s other Latin American countries had also reduced their agreements. The remaining European bilateralism had greatly diminished as the countries moved to Article VIII status in 1960 and 1961.

Statistically, by the end of 1964 more than 80 per cent of the total number of bilateral payments agreements that had been in effect in 1955 when the Fund’s decision was taken had been abolished. The number of agreements abandoned by individual countries was impressively large. From 1955 to 1964 Argentina, Belgium-Luxembourg, Finland, Germany, Italy, Japan, the Netherlands, and Uruguay eliminated 15 or more agreements each; and at least 6 agreements each were given up by a number of other countries, including Brazil, Chile, Denmark, Norway, Paraguay, Spain, Sweden, and Turkey.

Looked at more generally, rather than in terms of the number of agreements that individual countries had given up, in 1955 some 165 bilateral payments agreements were maintained by those 26 countries that had, by the end of 1964, accepted the obligations of the Fund’s Article VIII; but by the end of 1964 these countries had only 16 such agreements. In other words, they had eliminated over 90 per cent of their agreements. The agreements still remaining were maintained by only 7 of the 26 countries concerned, 4 of which had agreements exclusively with state-trading countries. Taking the whole of the Fund’s membership into account, by the end of 1964 more than one third had no bilateral payments agreements at all and nearly one half had no agreements with other members of the Fund.

The value of exports being conducted under bilateral payments agreements had, of course, also fallen—by about one half between 1954 and 1964—so that only 2 per cent of world exports, instead of 8.5 per cent, was involved.

Usually a country gave up its bilateral payments agreements prior to assuming Article VIII status. But in a few instances agreements were maintained which were eliminated later. For example, in 1964 France and Japan gave up the last of their agreements and Sweden ended two of its remaining agreements.

The elimination in 1965 of an agreement between Argentina and Uruguay left only two intra-Latin American agreements. One of these, between Bolivia and Brazil, was little used in practice; however, a new one between Mexico and Argentina had come into effect.


Despite the sharp drop in the use of bilateral payments agreements, especially by Western European countries, Japan, and Latin American countries, some other less developed countries have continued to arrange bilateral payments agreements, especially among themselves. A tendency to conclude bilateral payments agreements, even though little trade was covered, has been particularly prevalent among the newly independent countries in Africa. In addition, many less developed countries still conduct much of their trade with state-trading countries under bilateral arrangements.

As a result, the total number of bilateral payments agreements in the world declined little from 1954 to 1964. While four fifths of the agreements that had been in effect in 1954 had disappeared by 1964, many new agreements had come into being, and in December 1964 there were still 322 agreements in force. The number of bilateral payments agreements within Africa (including those of the United Arab Republic) had risen from nil to 20; and the number of agreements between African countries and state-trading countries had increased from 7 to 81. No agreements had been signed between African countries and Latin American countries or the European countries that had moved to Article VIII.

As did the African countries, so less developed countries in other regions entered into new bilateral agreements. A number of such countries (for example, Ceylon, India, Portugal, Spain, and the Syrian Arab Republic) which had had no agreements with state-trading countries in 1954, subsequently concluded several. At the end of 1964, seven Fund members—Brazil, Ghana, Guinea, Mali, Spain, the United Arab Republic, and Yugoslavia—still had 15 or more bilateral payments agreements each.

The implications of these agreements for the volume of trade conducted under bilateral arrangements were not uniform among the countries concerned. Many of the countries that had joined the Fund after 1954 and that had entered into new bilateral payments agreements had previously used the payments arrangements provided by their metropolitan centers in Europe, which, except for the United Kingdom, had been using some bilateral agreements in 1954. Hence, many of the new agreements of the African countries—e.g., those of Algeria, Burundi, Cameroon, the Central African Republic, Congo (Brazzaville), Dahomey, Laos, Mauritania, Niger, Senegal, Somalia, and Tunisia—did not necessarily involve an expansion of bilaterally conducted trade.

For other countries, however, including Afghanistan, Ceylon, Ghana, Guinea, India, Jordan, Mali, Morocco, the Syrian Arab Republic, and Upper Volta, trade under bilateralism had risen sharply. Most important of all, in 1964 more than 25 per cent of the total exports of some nine countries—Afghanistan, Greece, Guinea, Korea, Mali, the Syrian Arab Republic, the United Arab Republic, Upper Volta, and Yugoslavia—were still being sold to bilateral partners.

Changed character of bilateralism

This marked change in the distribution of bilateral agreements from early 1955 to late 1964—which was to have implications for the Fund’s policies—is shown in Table 9. While there was a reduction of one half in the arrangements between Fund members, bilateral payments agreements between members of the Fund and nonmembers (primarily state-trading countries) rose by more than 50 per cent. Agreements with nonmembers, which had constituted 44 per cent of the total number of agreements in April 1955, had become 70 per cent of the total remaining agreements in December 1964.

Table 9.Bilateral Payments Agreements Maintained by Fund Members in 1955 and 1964
Number of Agreements
April 1955December 1964
Agreements with other members18995
Agreements with nonmembers146227

The important increase in bilateral payments agreements between members of the Fund and state-trading countries had to some extent been the result of new agreements with state-trading countries by Western European countries. But to a much larger extent it had been the less developed members in Africa, Asia, and the Middle East that had been partners to the new agreements. A few statistics suggest the magnitudes involved. At the end of 1964, bilateral payments agreements between European members of the Fund—which here include Cyprus—and state-trading countries totaled 74; eight European members—Austria, Cyprus, Finland, Greece, Iceland, Spain, Turkey, and Yugoslavia—each maintained more than 5 agreements apiece. Less developed countries elsewhere had some 140 agreements with state-trading countries. Several of these countries—Algeria, Brazil, Ceylon, Colombia, Ghana, Guinea, India, Morocco, Tunisia, and the United Arab Republic—still maintained a number of agreements.

Reasons for persistence of bilateralism

The Fund’s consultations with members have revealed three main reasons why the less developed countries have continued, in the 1960’s, to arrange bilateral agreements despite the absence of a currency convertibility problem such as had prevailed in the 1950’s.

First, one of the principal motivations impelling the less developed countries to enter into bilateral payments agreements has been their need to promote additional exports of primary products. Brazil, Ceylon, Colombia, Ghana, India, and the United Arab Republic, for example, have maintained agreements to boost sales of coffee, cocoa, tea, or cotton. Certain countries—for instance, Denmark, Finland, Spain, Yugoslavia, and most state-trading countries—purchased coffee under bilateral arrangements as late as 1964–65. Since many of the producing countries have exportable supplies in excess of their quotas under the International Coffee Agreement, since attempts to reach an international commodity agreement for cocoa have been unsuccessful, and since storage is costly and has inflationary implications, the pressure on exporters of primary products, such as coffee and cocoa, to enter into bilateral arrangements has been strong.

The main argument of Fund members which have been relying on bilateral arrangements with state-trading countries has been that sales to these countries must be made bilaterally because commodities such as coffee and cocoa are given low priority in the plans of state-trading nations. But the member governments concerned have also argued that to open up important new markets by fostering changes in consumption patterns in these countries will make easier of solution the structural problem of excess supplies of primary products in world markets. Moreover, imports from bilateral partners, even if obtained on unfavorable terms, are considered a net gain to the economies of the less developed countries so long as these countries are themselves unable to produce the kind of products imported. In other words, as long as the factors of production are fairly immobile, and in particular cannot be moved out of the major export industries, less developed countries have considered bilaterally conducted trade as useful for disposing of surplus exports in return for needed imports.

The magnitudes involved in such bilateral trade, although marginal, have not been unimportant. In 1964, for example, Brazil and Colombia, which together accounted for about 60 per cent of world exports of coffee, sold 9 per cent and 7 per cent, respectively, of their total exports to countries with which they had bilateral agreements; and Brazil and Ghana, which accounted for almost half of the world’s cocoa exports, sold about 20 per cent of their cocoa exports under such agreements. Exports of tea by Ceylon and India under bilateral agreements, however, involved less than 3 per cent of India’s exports of tea in 1964 and less than 2 per cent of Ceylon’s.

A second reason for prolonged bilateralism, according to the authorities of a number of Fund members, is to assist the minor exports of the less developed countries. Some less developed countries have argued that they need the special export arrangements provided by bilateralism because of the trade barriers against their products, especially nontariff restrictions, prevailing in several industrial nations of the West. The Tunisian and Jordanian authorities have, for example, used bilateral agreements for exports of phosphates. Chilean nitrates have also been exported bilaterally, as have fruits, nuts, and tobacco by a number of Mediterranean countries. A few of the less developed countries which have traditionally sold manufactured goods to markets in state-trading countries have financed that trade through bilateral agreements in an effort to avoid the difficult processes of finding alternative markets or of shifting productive factors.

Finally, the increased use of bilateral arrangements to finance mutual trade among the less developed countries has reflected the desire, partly for political reasons, for some type of regional cooperation. Not only have agreements been fostered by the central banks concerned with international payments, and by the trade ministries interested in promoting exports, but the ministries of foreign affairs of the countries concerned have often also urged bilateral agreements in order to foster political or diplomatic ties between countries. Thus, for several of the less developed countries bilateral payments agreements have been only one part of a complex of arrangements.


The Fund’s general policy decisions of 1955–60 foreshadowed the changing character of bilateralism, and hence have continued to be relevant to the altered circumstances. Thus the Fund has continued to try to reduce the use of bilateral agreements by individual members when annual consultations have been undertaken and when stand-by arrangements have been agreed.

Stressing discrimination against members

Immediately after the Board’s decision on discrimination in October 1959, the Fund’s primary concern as regards the bilateral agreements of some of its members with state-trading countries centered on the possible harm to other members. The Directors, however, had differing opinions about whether the agreements of Fund members with state-trading countries really did discriminate against other members of the Fund.

The Board’s concern with such agreements may be illustrated by its discussion in 1960 of the agreements maintained by Ceylon. These were all with nonmembers and affected mainly food items imported by the Government. The most important agreement was with mainland China; under this agreement, the two countries had undertaken to import certain items up to specified quotas and to endeavor to keep the balance of trade and payments between them as nearly equated as possible. Mr. Jayarajan (Ceylon), Alternate to Mr. Watanabe, stated that his Government believed that the bilateral agreements of Ceylon with state-trading countries did not conflict with the Fund’s objectives; in particular, these agreements did not contain any discriminatory features. Ceylon’s view was that bilateral trade agreements continued to be the only way in which Ceylon could maintain and foster trade with state-trading countries until these countries were willing to pay for their imports in convertible currencies.

Both Lord Cromer (United Kingdom) and Mr. Southard (United States), nonetheless, had reservations about the inoffensiveness of these agreements. In their view, some kind of de facto discrimination against private traders might well arise from these arrangements. Bilateral agreements of this type, they thought, gave rise to discrimination against other members of the Fund by requiring Ceylon to import from partner countries certain quantities of goods which could not therefore be imported from other countries. Mr. Southard wished to retain in the Board’s decision on the consultation with Ceylon a sentence urging Ceylon to re-examine the need for its bilateral agreements. Mr. Jayarajan argued against the inclusion of this sentence on the grounds that, as there had been no increase in the scope of bilateralism since the previous year’s consultation, and as no mention of Ceylon’s bilateralism had then been made, it was unnecessary for the Fund to say anything now. After considerable discussion, in which it was noted that Ceylon had declared its willingness to re-examine its agreements, the Directors agreed on a sentence for the decision as follows: “The Fund notes the intention of Ceylon to keep under review its bilateral trade and payments arrangements, especially as to their possible discriminatory aspects.”

Emphasizing the costs of bilateralism

In more recent years the Fund has taken special pains to point out to the individual countries concerned the economic disadvantages likely to flow from bilateral agreements. In general the Fund has conceded that, in the short run, such agreements often seem to achieve the objectives sought, which may include economizing convertible exchange, seeking more outlets for traditional exports or opening up markets for new ones (e.g., manufactures produced by emerging industries), or merely postponing settlements in convertible currencies until swing credits are overdrawn. The Fund has stressed, however, that in the longer run serious drawbacks become apparent where an attempt is made to divert trade into new channels. The most common and most obvious one is that it is difficult to find in the partner country a sufficient variety of goods of the right qualities, at attractive prices, and on acceptable delivery terms, to avoid a oneway development of the clearing account to the point where an excessive bilateral claim arises. Thus, less developed countries may find themselves granting credits which they cannot afford, rather than receiving them; alternatively, trade under the agreement may stagnate.

A less apparent drawback that the Fund has pointed out is that, in the longer run, bilateralism actually tends to harm the exports of a less developed country. Exports effected bilaterally may cause upward pressures on prices, wages, and costs; or may distort the structure of domestic production. What is more, the Fund has argued, the other party to the bilateral agreement may well experience similarly adverse economic effects, and is likely to feel obliged to maintain more discriminatory trade and exchange policies than would be required in the absence of the agreement.18

The Fund has often pointed out to members that it is doubtful that imports by other Fund members of products such as cocoa and coffee are determined by the availability of bilateral financial arrangements or by preferential access to the import markets of the producing countries. Such small amounts of trade are involved, so the Fund’s argument has run, that they could readily be handled on a convertible basis. As regards sales to state-trading countries, the Fund has also noted that the actual increase in exports by primary producers may be exaggerated; time and again evidence has been found of substantial re-exports of these commodities by state-trading countries to established markets, often with unfavorable effect on the prices of the commodities. As regards the problem of bilaterally conducted sales by member countries of minor exports, the staff of the Fund has used mainly the approach of suggesting to members alternative ways to promote minor exports.

Thus the Fund’s efforts to diminish the use of bilateral agreements, even those of members of the Fund with state-trading countries, have continued. Members have been encouraged to implement the Fund’s general policy whenever the opportunity to dispense with bilateral agreements arises.


The bilateral arrangements of the interwar period are described in Margaret S. Gordon, Barriers to World Trade, and Howard S. Ellis, Exchange Control in Central Europe.


See, for example, Henry J. Tasca, World Trading Systems, Chaps. IX and X.


See, for example, Howard S. Ellis, Bilateralism and the Future of International Trade.


The main features of postwar bilateral payments agreements, their purposes, and how they operated, are described in considerable detail by Johan H. C. de Looper in “Current Usage of Payments Agreements and Trade Agreements,” Staff Papers, Vol. IV (1954–55), at pp. 339–50.


The contrast between the type of bilateral agreements used after World War II and those used earlier is described by J. W. Beyen in Money in a Maelstrom. The genesis and spread of postwar agreements is recounted by Judd Polk and Gardner Patterson in “The Emerging Pattern of Bilateralism,” Quarterly Journal of Economics, Vol. 62 (1947–48), pp. 118–42.


Bank for International Settlements, Seventeenth Annual Report (Basle, 1947), p. 75.


Latin America’s bilateral arrangements after World War II are described by Johan H. C. de Looper in “Recent Latin American Experience with Bilateral Trade and Payments Agreements,” Staff Papers, Vol. IV (1954–55), pp. 85–112.


See Chapter 15 below.


European experience with trade and payments agreements prior to the EPU is described in William Diebold, Trade and Payments in Western Europe.


The ways in which it gradually become more difficult for countries to operate bilateral agreements are spelled out in Fifth Annual Report on Exchange Restrictions (1954), pp. 17–18.


See Barend A. de Vries and F. A. G. Keesing, “The Use of Bilateral Agreement Currencies for Trade with Third Countries,” Staff Papers, Vol. V (1956–57), pp. 170–99.


E.B. Decision No. 433-(55/42), June 22, 1955; below, Vol. III, p. 258.


The letter was published in Annual Report, 1955, Appendix I, p. 123.


E.B. Decision No. 955-(59/45), October 23, 1959; below, Vol. III, p. 260.


E.B. Decision No. 1034-(60/27), June 1, 1960; below, Vol. III, p. 260.




The provisions of stand-by arrangements are described below in Chapter 20.


See, for example, Annual Report, 1963, pp. 69–70.

    Other Resources Citing This Publication