CHAPTER 10 Exchange Restrictions: The Setting

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

The second part of the code of conduct implicit in the Articles of Agreement (as described in Chapter 2) concerns exchange restrictions and convertibility of currencies. When the Articles were signed at the end of 1945, the flow of international commerce and payments was very much restricted. The preceding fifteen years had seen the introduction and extension of fairly rigid controls over international trade and payments by all except a few countries, mainly in the Western Hemisphere. The depression of the 1930’s had forced most countries to abandon the gold standard. Faced with drastically reduced supplies of foreign exchange, countries had had to take measures to curb the demand for exchange. Many adopted exchange controls.1


By 1945, European countries could be classified in three groups. (1) A first group of countries had formed a “sterling bloc.” In September 1931 Denmark, Finland, Norway, Portugal, and Sweden devalued along with the United Kingdom, and thereafter their currencies were pegged to the pound sterling, the rate for which fluctuated. They retained their free markets and, with minor exceptions, did not impose exchange controls. (2) Another group of countries—Belgium, France, the Netherlands, and Switzerland—had constituted a “gold bloc.” 2 These countries in general maintained both the pre-1931 parities for their exchange rates and the convertibility of their currencies into gold until 1936, after which they departed from the gold standard but avoided introducing exchange controls.3 (3) A third group of countries in Central and Eastern Europe had taken the exchange control route. Germany adopted controls in August 1931, and shortly thereafter Austria, Bulgaria, Czechoslovakia, Greece, Hungary, Rumania, and Yugoslavia did likewise. Italy held out until 1934, and Poland until 1936, when exchange controls were adopted by those two countries also.4

Exchange controls had also been adopted by countries outside Europe. Japan resorted to controls in July 1932; and in 1931–32 several Latin American countries, notably Argentina, Bolivia, Brazil, Chile, Colombia, Paraguay, and Uruguay, found it necessary to restrict transactions in foreign exchange. The exchange depreciation already undertaken by most of these Latin American countries had failed to alleviate sufficiently the acute shortages of foreign exchange brought about after 1929 by severe declines in the demand for their exports and a marked deterioration in their terms of trade.

With the outbreak of World War II, practically all countries had been compelled to bring their international payments and reserves under strict regulation. In September 1939 the United Kingdom had set up a comprehensive system of controls over foreign exchange transactions. Moreover, what had been the informal and voluntary arrangements of the “sterling bloc” had become the formal and legal arrangements of the “sterling area.” Regulations similar to those of the United Kingdom had been put into effect in the British Dominions (except for Canada and Newfoundland) and throughout the British colonies, protectorates, and mandated territories. Thus exchange controls had been put in force in what were then—or were to become after the war—Australia, Burma, Ceylon, Egypt, India, Iraq, Ireland, Israel, Jamaica, Jordan, Kenya, Kuwait, Malaysia, New Zealand, Nigeria, Pakistan, Sierra Leone, South Africa, the Sudan, and Uganda. Furthermore, as Germany had occupied the continent of Europe, the system of German controls and clearing arrangements had been extended. By early 1942, some seventeen countries, including Belgium, Denmark, France, the Netherlands, and Norway, had been linked to the German arrangements.5

Exchange controls and quantitative restrictions on trade had become major instruments of national policy, and their manipulation had become a plausible new alternative to exchange rate adjustment when a country’s balance of payments fell into disorder. Moreover, exchange controls had become not only a method of adjusting the external financial position of a country, but also a way in which countries could pursue internal monetary policies in their own interests independent of external considerations. Behind a protective wall of controls, countries could, for example, employ expansionary monetary policies aimed at stimulating full employment.6

It was hoped by many economists and government officials that, once the war was over, most of these controls would be removed. They favored a return to some mechanism which would again permit freedom of international transactions on a multilateral rather than a bilateral basis. To this end, the convertibility of sterling was looked upon as a key element. The Anglo-American Financial Agreement of 1945 had, therefore, envisaged an early restoration of sterling convertibility: under the terms of this Agreement, sterling was to be made convertible by July 16, 1947.

The problem of eliminating exchange restrictions—and of subsequently maintaining freedom of international payments—was, however, inextricably bound up with the question of how reconcilable was the pursuit of domestic full employment with the maintenance of free and multilateral international trade and payments—a question discussed in Chapter 2. Countries agreed to the code of conduct of the Fund in the realization that much harm had been wrought both to the system of international payments and to their own domestic economies by the wide use of exchange controls. But they were by no means confident that they could live without such controls. It was evident that the Fund’s task in this field would not be an easy one. Controls would not wither away merely because World War II had come to an end, or because the Fund’s Articles had come into effect.


This was the environment in which the Fund came into being. One of its major purposes, as noted in Chapter 2, is “to assist in the establishment of a multilateral system of payments in respect of current transactions between members and in the elimination of foreign exchange restrictions which hamper the growth of world trade” (Article I (iv)). Beyond that, its Articles spell out several obligations that members must assume. Article VIII defines the general obligations of members; Sections 2, 3, and 4 specify those regarding exchange restrictions and convertibility of currencies. These are, in part:

  • Sec 2. Avoidance of restrictions on current payments.—(a) Subject to the provisions of Article VII, Section 3 (b), and Article XIV, Section 2, no member shall, without the approval of the Fund, impose restrictions on the making of payments and transfers for current international transactions.7
  • (b) Exchange contracts which involve the currency of any member and which are contrary to the exchange control regulations of that member maintained or imposed consistently with this Agreement shall be unenforceable in the territories of any member….8
  • Sec 3. Avoidance of discriminatory currency practices.—No member shall engage in, or permit any of its fiscal agencies … to engage in, any discriminatory currency arrangements or multiple currency practices except as authorized under this Agreement or approved by the Fund….
  • Sec 4. Convertibility of foreign held balances.—(a) Each member shall buy balances of its currency held by another member if the latter, in requesting the purchase, represents
    • (i) that the balances to be bought have been recently acquired as a result of current transactions; or
    • (ii) that their conversion is needed for making payments for current transactions.

However, these obligations did not have to be fulfilled immediately. Article XIV, Sections 2, 3, 4, and 5, provided for a transitional period:

  • Sec 2. Exchange restrictions.—In the post-war transitional period members may, notwithstanding the provisions of any other articles of this Agreement, maintain and adapt to changing circumstances (and, in the case of members whose territories have been occupied by the enemy, introduce where necessary) restrictions on payments and transfers for current international transactions…. members shall withdraw restrictions maintained or imposed under this Section as soon as they are satisfied that they will be able, in the absence of such restrictions, to settle their balance of payments in a manner which will not unduly encumber their access to the resources of the Fund.
  • Sec 3. Notification to the Fund.—Each member shall notify the Fund before it becomes eligible under Article XX, Section 4 (c) or (d), to buy currency from the Fund, whether it intends to avail itself of the transitional arrangements in Section 2 of this Article, or whether it is prepared to accept the obligations of Article VIII, Sections 2, 3, and 4….
  • Sec 4. Action of the Fund relating to restrictions.—Not later than three years after the date on which the Fund begins operations and in each year thereafter, the Fund shall report on the restrictions still in force under Section 2 of this Article. Five years after the date on which the Fund begins operations, and in each year thereafter, any member still retaining any restrictions inconsistent with Article VIII, Sections 2, 3, or 4, shall consult the Fund as to their further retention.
  • The Fund may, if it deems such action necessary in exceptional circumstances, make representations to any member that conditions are favorable for the withdrawal of any particular restriction, or for the general abandonment of restrictions, inconsistent with the provisions of any other article of this Agreement. The member shall be given a suitable time to reply to such representations. If the Fund finds that the member persists in maintaining restrictions which are inconsistent with the purposes of the Fund, the member shall be subject to Article XV, Section 2 (a).
  • Sec5. Nature of transitional period.—In its relations with members, the Fund shall recognize that the post-war transitional period will be one of change and adjustment and in making decisions on requests occasioned thereby which are presented by any member it shall give the member the benefit of any reasonable doubt.

Not only did these Articles specify the obligations of members, but they made clear that despite the considerable powers of the Fund, there were certain limitations on its objectives and on its jurisdiction in the field of exchange restrictions.


Restrictions on payments

The first point to be noted about the obligations of Article VIII is the phrase in Section 2 (a), restrictions on the making of payments and transfers (referred to in this volume as exchange restrictions). This phrase has the effect of demarcating the Fund’s authority.

At the time of the Bretton Woods Conference in 1944, it was expected that two agencies would be established in the international economic field: one, a trade organization, to deal inter alia with restraints on international trade, such as tariffs, quotas, import controls, and the like; the other, the International Monetary Fund, to deal with restrictions on international payments and transfers. Later, in its first Annual Report, the Fund stressed that it “was conceived as one element in a many-sided approach to the task of re-establishing a functioning world economic system,” and pointed reference was made to the plans for a parallel International Trade Organization.9

For a variety of reasons, the trade organization was not in fact ever set up. But most of the matters that were to have been under its jurisdiction eventually came to be handled through the Contracting Parties to the General Agreement on Tariffs and Trade (GATT). The GATT, accordingly, has jurisdiction over trade and import restrictions. Many economists have argued that distinctions between the various types of obstacles to international transactions—and in particular between restraints labeled “import restrictions” and others called “exchange restrictions”—have little economic significance. Nonetheless, the Fund’s competence was delimited by the specific words of Article VIII, Section 2 (a).

In the Fund’s early years there were repeated discussions among both the staff and the Executive Directors as to the extent of the Fund’s jurisdiction. Frequently, such internal discussions focused on the questions of just how an exchange restriction should be defined and how it should be differentiated from a trade restriction. Clearly, something called restrictions on trade—including import licensing—were to be the concern of the GATT, while restrictions on payments—including payments for imports through an exchange licensing procedure—were the proper province of the Fund. It was to prove difficult in practice, however, to distinguish trade restrictions from payments restrictions. Differences between the exchange control systems of one country and another often seemed to result more from variations among countries’ legal requirements and administrative procedures than from the economic intent of the controls. In some countries—including those of the sterling area—goods could not be imported without import licenses; and such licenses were not freely issued. However, once an import license had been obtained, foreign exchange could be automatically obtained for the permitted import. On the surface of things, this type of regulation seemed to constitute a restriction on trade but not a restriction on foreign exchange payments. The converse, a restriction on foreign exchange payments but not on trade, seemed to exist in other countries, where it was the foreign exchange which was subject to license and where such exchange licenses were limited in issue but, once obtained, conferred an automatic right to import or to obtain an import license.

Hence, important questions had to be answered: Was there an essential difference between a situation where an importer was unable to import a commodity because he could not obtain an import license, and one where he was denied the foreign exchange with which to purchase it? Was the point at which a restriction was applied—i.e., on imports or on exchange—an adequate basis for deciding which countries would have to consult the Fund or obtain the Fund’s approval for their restrictions? If so, were many of the Fund’s principal members that applied controls to imports, such as those in the sterling area, to be left free of the Fund’s code of conduct as regards restrictions?

If the form alone of a restriction did not constitute a sufficient basis for differentiating an exchange restriction from a trade restriction, it quickly became evident that neither did the purpose for which a given restriction might be applied. Licensing of imports could easily be used as a means of restricting payments in foreign exchange—that is, import restrictions could be introduced to restrain deficits in the balance of payments. In fact, the GATT charter explicitly recognized the possibility of “quantitative restrictions for balance of payments purposes.” Similarly, restrictions on payments originally applied in order to conserve foreign exchange could be retained, after the need to conserve foreign exchange no longer existed, for the benefit of domestic industry that had grown up under its protection. Thus, a division of the jurisdiction between the Fund and the GATT using the purpose of the restrictions as a guide was not feasible.

Debates within the Fund over how to define and characterize an exchange restriction became especially intense early in 1952, just before the first consultations under Article XIV were to take place. For it was then that it had to be decided what types of restrictive and other policies of members would be made subject to consultation. Although an explicit definition of an exchange restriction was not agreed until 1960,10 the Directors, the member countries, and the staff worked out arrangements by which, in the meantime, the Fund could discuss, even if it did not comment formally on, a wide range of members’ restrictions.

Although on other occasions the problem of a clear differentiation between the jurisdiction of the Fund and of the GATT has been a subject of discussion by the Executive Directors, the Fund and the GATT have not had any actual disputes over jurisdiction; on the contrary, cooperation has been close and substantive.

The European Payments Union

Shortly after its establishment, the Fund found itself operating in a world that contained not only the GATT but also various regional groupings which, like the Fund, concerned themselves with payments arrangements across national boundaries. The most important of these was the European Payments Union (EPU), which was founded in Paris in September 1950 under the auspices of the Organization for European Economic Cooperation (OEEC), also relatively new. The EPU led to the creation of a new monetary area known as the “EPU monetary area,” the main features of which were that currencies of EPU members (as well as currencies of the countries in the members’ own monetary areas) held by residents of other EPU members became, in practice, transferable, i.e., convertible within the EPU area. Intra-EPU payments hence became “multilateralized.” Furthermore, members of the EPU were committed to liberalize their restrictions on intra-European trade and payments. Trade liberalization ratios were established which applied to each member’s total trade with the whole area rather than to trade with each member separately.

The Fund’s task of achieving liberalization of exchange restrictions and convertibility of currencies was thus made both more complex and easier. It was more complex because there was now another formal organization in the same field, and the loyalties of countries that were members of both organizations might be divided. Initially, the establishment of the EPU monetary area gave rise to anxiety among the staff, the management, and some of the Executive Directors, who feared that the EPU might become a permanent monetary area and that discrimination against outside countries, especially those in the dollar area, might be prolonged. At the same time, however, the achievement of the Fund’s objectives was facilitated by the corresponding efforts of the OEEC in liberalizing the restrictions of the members of the EPU and in restoring the convertibility of their currencies.

Current transactions

In addition to the limitations described above, the Fund’s regulatory functions apply only to certain kinds of international transactions, some kinds being deliberately left out. The Articles recognize that members might find it necessary to control capital movements. The obligation of members under Article VIII, Section 2, with respect to freedom of exchange transactions is explicitly limited to the avoidance of restrictions on payments and transfers on current international transactions. To underscore the point, Article VI, Section 3, explicitly provides that “members may exercise such controls as are necessary to regulate international capital movements,” although it goes on to provide that, with certain exceptions, “no member may exercise these controls in a manner which will restrict payments for current transactions or which will unduly delay transfers of funds in settlement of commitments….”

As the Fund’s Articles were drafted against the background of the disturbing capital movements that had taken place during the 1930’s, there was an understandable desire to prevent movements of “hot” money and to minimize the risk that inadequate foreign exchange reserves would be depleted by more or less panic-inspired capital transfers. Hence, it was thought that controls over capital movements might be necessary and beneficial.

The freedom of the members of the Fund to exercise such controls as are necessary to regulate capital movements for any reason, without prior Fund approval, was restated in a decision by the Executive Board interpreting Article VI, Section 3.11

Transitional arrangements

Finally, the Articles provided that the Fund was not to pursue to the full its objectives in the field of restrictions during a “transitional period.” This limitation affected the Fund’s activities in a very important way in the early years, and has continued to be relevant until the present time for the majority of the Fund’s members. As quoted above, Article XIV provides for the retention of restrictions during this transitional period. After World War II, when supplies of foreign exchange nearly everywhere were running far short of probable demands for imports, it was obviously impossible to expect any country to abandon at once the elaborate control machinery which had been built up; and it was certain that, at least during an interim period whose length no one could predict, countries where the whole economic structure had to be painfully reconstructed after enemy occupation would feel obliged to impose strict controls upon their international trade and payments.

No period of time was set by the Articles for this transitional period. No definitional criteria of any kind were given. The Fund was merely admonished to recognize that the “post-war transitional period” would be a period “of change and adjustment” and in making any decisions to “give the member the benefit of any reasonable doubt.” Two periods were mentioned in Article XIV—three years between the time the Fund started operations (i.e., March 1947) and the time when it was required to make a report to its members on any exchange restrictions still in force, and five years between the time the Fund began operations and the start of consultations on restrictions maintained under Article XIV. But neither of these periods was intended to define the transitional period.

The provision for a transitional period has had several implications for the formation of the Fund’s policies on restrictions. First, for several years, the Fund hesitated to take much, if any, action concerning the exchange restrictions maintained by its members. Indeed, prior to the publication of the first Exchange Restrictions Report, in 1950, the only time that the Executive Directors undertook to make any general pronouncements on exchange restrictions was in the course of agreeing on their Annual Reports to the Board of Governors. Second, inasmuch as the Fund had already subjected multiple currency practices to its review (as early as 1947), it seemed to many members, especially those in Latin America, that there was an unfair dichotomy in the Fund’s policies. With other forms of control so prevalent and the Fund doing little about them, harping on the elimination of multiple rates which seemed to many relatively minor was a source of irritation to a lot of members. Third, most members were to continue to take advantage of the transitional arrangements for a number of years. Before 1961, only Canada, Cuba, the Dominican Republic, El Salvador, Guatemala, Haiti, Honduras, Mexico, Panama, and the United States had accepted the obligations of Article VIII.

Therefore, from time to time the question arose whether the Fund should not declare the transitional period ended. The Legal Department, in 1959, advised that the Board had no power to terminate the transitional period. By 1965 the issue was no longer much debated, although all but 27 of the Fund’s members still took advantage of these arrangements.


In addition to the limitations imposed by the Articles, Article VIII defines a general and ultimate objective. It does not catalog the particular types of restrictions that usually make up an exchange control system. No differentiation is made, for example, between payments made by a member’s residents, and payments received by residents, e.g., payments made by nonresidents.

Of course, when the Fund began to take action concerning restrictions on payments and transfers, it had to seek to apply these general principles to individual restrictions. Moreover, it became necessary to formulate separate lines of action in respect of certain particular forms of restrictions or payments arrangements. Additionally, as the Fund specified its goals from time to time, it had to formulate a lesser objective than the total elimination of restrictions. Only gradually could its ultimate objectives be attained.

Three particular aspects of exchange restrictions were to become the focus of the Fund’s attention during the years 1945 to 1965: partial (or external) convertibility, discrimination, and bilateralism.

Partial convertibility

In the exchange control regulations with which the Fund was faced when it undertook consultations on restrictions in 1952, various monetary or currency areas were usually differentiated. Payments by residents of one monetary area to those of another were subject to restrictions of varying degrees. For instance, countries in the sterling area applied much stricter rules to payments to countries in the dollar area than to payments to other sterling area countries.

After the Fund had been in operation for some eight or nine years, the concepts of external and internal convertibility became regularly used to distinguish different degrees of convertibility. Such concepts had not been part of the Articles. When convertibility of sterling was attempted in 1947, total convertibility, not partial, was the aim. Yet as time went on, it became evident that currencies would be made convertible by degrees, and not all at once. Hence the distinction between external and internal convertibility became a useful one.

The payments system of a country whose currency is externally convertible has two important characteristics. First, all holdings of that currency by nonresidents are freely exchangeable into any foreign (nonresident) currency at exchange rates within the official margins. Second, all payments that residents of the country are authorized to make to nonresidents may be made in any externally convertible currency that residents can buy in foreign exchange markets. On the other hand, if there are no restrictions on the ability of a country’s residents to use their holdings of domestic currency to acquire any foreign currency and hold it, or transfer it to any nonresident for any purpose, the country’s currency is said to be internally convertible. Thus external convertibility alone is tantamount to partial convertibility whereas total convertibility involves both external and internal convertibility.

The Fund members that initially opted for Article XIV and that subsequently moved to Article VIII (mainly countries in Western Europe), in general established external convertibility first and internal convertibility later. Only a few countries—Germany, for example—established the two simultaneously. Both types of convertibility were achieved gradually by most of the European countries, and not without setbacks. Formal recognition was usually eventually given to the de facto convertibility established by a series of administrative acts which had in successive stages liberalized the country’s system of restrictions. In this process, a number of stages of less than total convertibility were reached.


The second distinction of which the Fund has had to take account is that between discriminatory and nondiscriminatory restrictions. Prior to the establishment of external convertibility, there was considerable discrimination in the restrictions applied to different monetary or currency areas. Again, as an illustration, the tighter restrictions imposed on transactions in dollars than on transactions in sterling constituted discrimination against the dollar area. In general, discrimination refers to a situation where a country treats its international transactions differently depending on which foreign countries, or which currencies, or both, are involved.

The issue of discrimination in international trade and finance has been a long-standing one, extending far beyond the payments restrictions falling under the jurisdiction of the Fund. Discrimination may be involved, for example, in trade restrictions, in regional arrangements, in customs unions, and in customs tariffs.12 Discrimination in payments was for many of the years reviewed here especially pronounced and prolonged against countries in the dollar area; transactions with the dollar area were, as a rule, the last to be freed from restrictions.

Thus, the field of discrimination was a particular aspect of exchange controls that the Fund also had to tackle. Indeed, discrimination was one of the first problems pertaining to exchange restrictions considered by the Fund; the question of discrimination arose as early as 1949 in connection with South Africa’s restrictions. The problem of discrimination against the dollar area persisted to some extent even after the establishment of the external convertibility of European currencies at the end of 1958. Hence, in 1959 the Executive Board took a special decision pertaining to discrimination in exchange restrictions.


Another feature of restrictive systems consists of bilateral payments agreements, i.e., agreements between two countries on how payments between them will be settled. Under a typical bilateral agreement, payments between the partner countries have to be made through the bilateral accounts held by the countries’ central banks with each other. The outstanding balances on these accounts are generally subject to limits which represent the amount of credit each country is willing to extend to the other; when one of the limits is reached, further amounts accruing to the creditor are to be settled in gold or convertible currencies. The Fund has often used the term bilateralism to mean the use of such agreements.

Although the Articles of Agreement do not refer specifically to bilateral payments agreements, they do specify as one of the aims of the Fund the attainment of a multilateral system of payments. Consequently, the Executive Directors have had no difficulty in accepting the view that bilateral payments agreements fall within the Fund’s jurisdiction. In 1955 the Board took a special decision stating the Fund’s policy on bilateralism.


Against this background, the next several chapters have been organized. Chapter 11 explains in detail how the Fund carries on its consultations with individual members. In addition, there are described the reasons why the present procedures were agreed, and the way in which the Fund resolved the question of which restrictions should be the subject of consultation. The chapter also explains how the consultations, which started as consultations on exchange restrictions under Article XIV, evolved into full-blown discussions on many other topics, even with countries not under Article XIV.

Chapter 12 traces developments in the Fund’s policies on restrictions from 1945 to the end of 1958 when European countries established external convertibility for their currencies. This chapter interweaves the related developments in world economic conditions during these years and the process by which restrictions were gradually liberalized. Chapter 13 explains these same developments from 1959 to 1965, as well as some special problems of the 1960’s, such as those of the less developed countries and some newly emerging difficulties with regard to capital movements among industrial nations. Chapter 14 relates the separate story of bilateral payments agreements. The Fund’s relations with the EPU are briefly described in Chapter 15. The discussion of exchange restrictions is then concluded with a resume of the Fund’s relations with the GATT in Chapter 16.


The term exchange controls in this volume is used to designate a broad course of action and an entire apparatus of control, which usually includes exchange restrictions.


Italy was part of the gold bloc until 1934, and Poland until 1936.


Belgium introduced controls in March 1935 but rescinded them soon afterward.


Details of the introduction, extent, and nature of controls in Europe during the 1930’s can be found in League of Nations, Report on Exchange Control.


Details of the extension of exchange controls as a result of World War II can be found in Raymond F. Mikesell, Foreign Exchange in the Postwar World, pp. 3–21.


The details of how countries had learned to use controls and of their harmful consequences are given in League of Nations, International Currency Experience; in Margaret S. Gordon, Barriers to World Trade; and in Howard S. Ellis, Exchange Control in Central Europe.


Article VII, Section 3 (b), refers to the special situation when the Fund has formally declared its holdings of a given currency to be scarce; see Vol. I, p. 193. Article XIV, Section 2, provides for a postwar transitional period. For the full text of the Articles, see below, Vol. III, pp. 185–214.


For a discussion of the interpretation of this provision and of the legal proceedings involved in the courts of member countries in enforcing this provision, see Joseph Gold, The Fund Agreement in the Courts.


Annual Report, 1946, p. 15.


E.B. Decision No. 1034-(60/27), June 1, 1960; below, Vol. III, pp. 260–61.


E.B. Decision No. 541-(56/39), July 25, 1956; below, Vol. III, p. 246.


For a study of discrimination in these broader aspects covering the same years as this volume, see Gardner Patterson, Discrimination in International Trade.

    Other Resources Citing This Publication