CHAPTER 6 Multiple Exchange Rates

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

Multiple exchange rates were one of the first problems that faced the Fund in 1946, and have probably been its most common problem in the field of exchange rates. An impressive number and diversity of countries in the last twenty years have experimented with one form or another of what the Fund has called multiple currency practices, at least for a few, if not for most, of their transactions.1 And, although the use of such practices has been much reduced in recent years, a succession of countries that had not previously tried multiple rates has been experimenting with them. In the mid-1960’s, some new suggestions for multiple exchange rates were put forward. Economists, especially those outside official circles, have continued to propose exchange rate schemes as a means of assisting primary producing countries. Moreover, a few have suggested that subsidies-cum-tariffs—in a sense a form of multiple rates—or two exchange rates, one a floating rate for capital, might be possible compromises between fixed and fluctuating rates for the industrialized nations. The problem of multiple rates, then, never seems entirely at an end.


The interwar period

Multiple rates originated in the exchange control mechanisms of the 1930’s. First used by Germany in 1934 in dealings with its bilateral agreement partners, multiple rates allowed considerably more exchange rate manipulation than did ordinary clearing agreements with a fixed exchange rate.2 German exports to a particular country could enjoy the benefits of sudden exchange depreciation. Through such devices, Germany succeeded in expanding its trade with certain countries in southeastern Europe and with a number of Latin American countries, especially those which had their own exchange controls, for example, Brazil, Chile, and Peru. Brazil’s imports from Germany rose from 13 per cent of its total imports in 1929 to 25 per cent in 1938, and its exports to Germany rose from 9 per cent to 19 per cent. In the same period, Peru’s imports from Germany rose from 10 per cent to 20 per cent of the total.3 Although Germany’s methods met with increasing resistance, Latin American countries tended to encourage bilateral transactions with Germany, even at unfavorable exchange rates, when raw material prices were low and exports to free exchange countries were depressed.

Multiple rates also began to be employed by other countries during the Great Depression. After 1932, the majority of European countries with exchange controls resorted eventually to some form of currency devaluation. Although some countries, like Czechoslovakia and Italy, devalued in one step, other countries passed through a period of gradual depreciation, usually through the use of multiple rates. Black markets developed, and later became legal free markets: Austria, Hungary, Yugoslavia, and Rumania were among the countries in Europe which, after a time, gave official recognition to black markets and thus to the multiple exchange rates prevailing in them.

Several Latin American countries similarly embarked on programs of multiple rates, essentially as exchange control devices. The purposes in Latin America, however, differed from those in Europe in two respects. Although the initial purpose of exchange controls in Europe was to control capital movements, this was taken less seriously in Latin America: several countries there permitted capital transactions through free markets. Secondly, in Latin America one of the main objectives of differentiated rates of exchange was to secure, at preferential rates, foreign exchange to service the government debt. Exchange depreciation had increased the cost of this service in terms of domestic currency, and the compulsory surrender of part of export proceeds at official rates operated in effect as a special tax to meet that cost. The obligation of exporters to surrender exchange receipts was sometimes confined to one or a few of the chief export products, to make it easier to enforce. In European countries, on the other hand, generally because of their creditor status, debt service was not an important stimulus to exchange control.

What precipitated international concern about these practices was that the trend in most countries with multiple rates was toward discrimination. The discrimination was seldom explicit. More often the allocation of foreign exchange among different countries became subject to administrative discretion, and in these circumstances some form of discrimination, or “preferential allotment,” became virtually inevitable. The objective of discrimination was usually to evade the most-favored-nation clause of trade agreements and to force trade into bilateral channels.

Free exchange countries, therefore, when concluding trade agreements with exchange control countries, frequently insisted on the insertion of various formulas to ensure a “fair and equitable share” in any allotment of foreign exchange. However, because such formulas were difficult to prescribe in such a way as to give real equality of treatment, most of them had in practice little real meaning.

Multiple rates at Bretton Woods

Against this background the delegations at Bretton Woods treated multiple currency practices in the same way as they treated quantitative exchange restrictions. In the Keynes Plan no specific mention had been made of multiple currency practices. But in the White Plan, one of the stated purposes of the Fund was

  • 5. To reduce the use of such foreign exchange restrictions, bilateral clearing arrangements, multiple currency devices, and discriminatory foreign exchange practices as hamper world trade and the international flow of productive capital.4

Clause IX of the Joint Statement by Experts on the Establishment of an International Monetary Fund, paragraphs 2 and 3, under the heading The Obligations of Member Countries, contained statements to the effect that each member country is

  • 2. Not to allow exchange transactions in its market in currencies of other members at rates outside a prescribed range based on the agreed parities.
  • 3. Not to impose restrictions on payments for current international transactions with other member countries (other than those involving capital transfers or in accordance with VI, above) or to engage in any discriminatory currency arrangements or multiple currency practices without the approval of the Fund.5

“VI, above” referred to the provisions for allocating a scarce currency.

In June 1944, the scope of these provisions was interpreted by the U.S. Treasury in Questions and Answers on the International Monetary Fund. Question 20 was: Would differential rates of exchange for different classes of imports and exports (visible and invisible) be permitted by the Fund? The answer suggested that what was considered most objectionable in multiple currency practices was the uses to which such devices might be put, rather than the multiplicity of rates itself. Multiple rates should be avoided because they might so easily become the means for discrimination in trade relationships, and because they usually involved control of the exchanges so complete as to offer a strong temptation to restrict transactions on current account. Nonetheless, it was made clear that, conceivably, the Fund might decide that a member’s multiple currency practices did not conflict with the principles of the Fund. Moreover, where conflict did occur, the Fund would allow adequate time for the abolition of multiple currency practices.6

As finally drafted, Article VIII of the Articles of Agreement obliges members (Section 2 (a)) not to “impose restrictions on the making of payments and transfers for current international transactions,” nor (Section 3) to

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. If such arrangements and practices are engaged in at the date when this Agreement enters into force the member concerned shall consult with the Fund as to their progressive removal unless they are maintained or imposed under Article XIV, Section 2, in which case the provisions of Section 4 of that Article shall apply.

Article XIV makes no explicit reference to multiple currency practices.

Prevalence of multiple rates in 1946

By the time the Fund was organized, multiple rates were in use not only in the six countries of Latin America that had used them during the 1930’s (Argentina, Brazil, Chile, Ecuador, Peru, and Venezuela), but also in Bolivia, Colombia, Costa Rica, Nicaragua, and Uruguay; and Cuba and Honduras had introduced small exchange taxes that had somewhat the same effect as multiple rates.7 The use of multiple rates in Iran suggested that the problem might not remain confined to Latin America.

The countries adopting multiple currency practices had problems which did not seem to be solvable in any other way. Mr. (now Professor) Robert Triffin, then with the Board of Governors of the Federal Reserve System, had in fact, when helping a number of Latin American countries to overhaul their monetary, exchange control, and banking systems, suggested policies which, in his own words, were “highly unorthodox … for the newly born International Monetary Fund.” 8

The problem to which Mr. Triffin was seeking a solution was that of a primary producing country facing balance of payments disequilibrium due to a cyclically depressed world market for raw materials or for its own particular exports. This situation, in which over-all exchange depreciation was regarded as ineffective, was distinguished from that of fundamental disequilibrium, where an over-all exchange rate adjustment was required. For the cyclical problem, Mr. Triffin advocated two markets—a normal market and an “auction” market. Exchange proceeds from exports or other easily controllable sources would be channeled into the normal market, and be available for essential imports and current invisibles. Exchange proceeds from less controllable transactions, such as capital movements, would be sold in a free or “auction” market; purchasers could use these proceeds to make payments for nonessential imports or for outgoing capital. Such a dual exchange system would, Mr. Triffin argued, prevent the emergence of a black market and the need for administrative allocations, and would provide an escape valve: capital movements could take place without upsetting the markets or rates for normal transactions.9

These arguments were not merely theoretical; they expressed the objectives of countries’ actual policies. They explained, for example, the initial establishment of multiple rates in Argentina and of dual markets in several other countries, e.g., Chile, Costa Rica, Ecuador, and Paraguay.


Initial questions for the Fund

When the Fund opened its doors, 13 of its 39 members were using multiple exchange rates. Specific questions began to arise almost immediately. Even before concluding membership formalities, Venezuela (in September 1946) asked for assurance from the Fund that adherence to the Articles would not prevent it from taking a suitable period to eliminate its multiple currency practices. When the establishment of initial par values was considered at the end of 1946, the question arose as to what rates could be used as par values where multiple rates existed. When, in the early months of 1947, the Fund asked countries to state whether or not they were going to take advantage of the transitional arrangements, the significance of multiple rates again came to the fore. Cuba indicated that it would avail itself of the transitional arrangements primarily because of its small exchange tax.

Questions of definition hampered consideration of these practices. What precisely constituted a multiple currency practice under the Fund Agreement? Could the small exchange taxes of Cuba and Honduras be so defined? Did an unusually large difference between the effective buying and selling rates for a currency constitute a multiple currency practice? Were multiple currency practices to be regarded as a restriction and therefore subject to the Fund’s transitional arrangements, or were these practices to be considered solely as exchange rates?

More basic were questions pertaining to the Fund’s authority in this field. Could a member maintaining a multiple currency practice under the transitional arrangements introduce a new multiple rate without Fund approval? Could a member with a multiple currency practice change the classifications of commodities subject to different multiple rates without prior consultation with the Fund? Could a member which did not have any multiple currency practices, but which took advantage of the transitional arrangements, introduce a multiple currency practice without the Fund’s approval?

Important questions of policy also had to be answered: What was the economic significance of these practices? What should be the Fund’s attitude toward the multiple rates already in existence? Under what conditions might these rates be temporarily retained? What should be the view of the Fund toward new practices?

The December 1947 letter

During its first year, the Fund studied these questions at great length. Moreover, a review of individual country situations in 1947 made it clear that the extent of multiple currency practices warranted early discussion between members and the Fund; the Board agreed that most of these discussions should be held in the territory of the members concerned.

In order that a detailed examination could be made of the economic, legal, and procedural issues raised by all the questions on multiple rates, the Board set up an Ad Hoc Committee on Spreads and Multiple Currency Practices in the middle of 1947. The members of the committee were Mr. Mladek (Czechoslovakia), Chairman; Mr. Bruins (Netherlands); Mr. Joshi (India); Mr. Luthringer (United States), Alternate to Mr. Overby; and Mr. Martínez-Ostos (Mexico), Alternate to Mr. Gómez. This committee held many meetings and undertook a careful assessment of the extent of the multiple rate problem, of the Fund’s powers in this field, and of possible policies. In its report to the Board the committee urged moderation in the Fund’s approach to multiple rates. It emphasized that the Fund was especially interested in multiple rates because they were not simply restrictive practices but rates of exchange as well, and hence could be a primary source of exchange instability. Exchange rate stability, the report stressed, was one of the Fund’s main objectives. In order to make a start toward this objective the Fund had already, late in 1946, decided to set initial par values. While the mere existence of multiple exchange rates might not endanger exchange stability, frequent changes and extensions of such rates could well undermine the whole par value system. Moreover, multiple rates might also endanger the Fund’s objective of the avoidance of competitive depreciation.

A multiple currency practice was defined as any practice which, as a result of official action, gives rise to an effective buying or selling rate differing from the par value by more than 1 per cent. Such a practice was explicitly recognized to include any of a series of fixed exchange rates, exchange taxes, exchange surcharges, free markets, and auction systems. Although on several future occasions debates were to occur between a member and the Fund, and even within the Fund, as to whether particular exchange systems did in fact constitute multiple currency practices, this definition proved to be sufficiently comprehensive to cover the vast majority of cases for the next twenty years.

After deliberation, the Board adopted the committee’s recommendations unanimously. It agreed with the committee that the Fund had broad powers over the introduction of new practices and the adaptation of old ones, even during the transitional period; and decided that member countries should, as a minimum, consult the Fund before introducing a multiple currency practice, before making any change in a multiple exchange rate, before reclassifying transactions subject to different rates, and before making any other significant change in their exchange systems.

The Board also spelled out the policies to be followed. It recommended that early steps should be taken toward the removal of multiple currency practices that were not necessary for balance of payments reasons, although ample time should be provided for members to take these steps and to install appropriate substitutes where necessary. The Fund would encourage members engaging in multiple currency practices for balance of payments reasons to establish, as soon as possible, conditions which would permit the removal of these practices, with the general objective of seeking this removal not later than the end of the transitional period. Where complete removal by the end of the transitional period proved impossible, the Fund would assist the members concerned to eliminate the most dangerous aspects of their multiple currency practices and to exercise reasonable control over those retained.

As a consequence, there was circulated to the members in December 1947 a letter and memorandum outlining the Fund’s powers and policies in the matter of multiple rates. This letter and memorandum, which together came to be known as “the December 1947 letter on multiple currency practices,”10 laid the basis for relations with members in this field, although there have subsequently been modifications in the policies and procedures which it set forth.

A few Directors had had some difficulty with the terms of the letter. Mr. de Largentaye (France), among other points of disagreement, was of the opinion that members which had been occupied by the enemy and which were availing themselves of Article XIV could independently (i.e., without authorization from the Fund) introduce and modify multiple currency practices. In his view, such a member’s only obligation was to consult the Fund, And this obligation did not mean that the Fund could object. Some weeks later (in January 1948) this difference of opinion led to the vexing difficulties between France and the Fund described below.

Mr. Martínez-Ostos, while not disagreeing with the committee’s conclusions, pointed out two factors not mentioned in its analysis which he thought would have to be taken into consideration when specific courses of action were considered: the importance of the monetary management (nontrade) functions of multiple rates, and the negligible amount of world trade conducted by a few Latin American countries which maintained multiple rates.

It is noteworthy, in retrospect, that despite the importance that other Executive Directors attached to exchange rate stability during as well as after the transitional period, Mr. Martínez-Ostos stressed at this time that fixed multiple rates were not to be preferred over multiple rate systems with a floating rate (that is, a fluctuating or free market rate). He believed the opposite: since floating rates did not freeze particular rates into the economy, members would have less difficulty in moving to a single rate later on. Thus, floating rate mechanisms tended to be self-liquidating. This point is the more memorable because some ten years later, as noted below, this was to be the road by which many members were to move to unitary rates and a gradual change in Fund policy became necessary.

Disagreement with France

Immediately after the December 1947 letter—that is, in January 1948—the first example of a multiple currency practice was brought to the Board and led to disagreement with a major member. France proposed a devaluation of the franc from 119.107 francs to 214.392 francs per U.S. dollar. In addition, France proposed a free market for certain convertible currencies—namely, U.S. dollars and Portuguese escudos; Swiss francs were added subsequently.

The French authorities were willing to maintain orderly cross rates between all inconvertible currencies and between all convertible currencies. Thus, there would be only one case of disorderly cross rates—between the official rates of all currencies and the free rates of the convertible currencies. Mr. de Largentaye adduced two arguments in support of the French proposal. Legally, Article XIV, Section 2, permitted France, as an enemy-occupied country, to introduce a multiple currency practice. Economically, the free market in certain currencies would help to meet the deficit in the French balance of payments. A premium on exports to the dollar area would encourage such exports, while the penalty on dollar imports would encourage the substitution of European imports for dollar imports. In addition, the repatriation of French assets held abroad would be encouraged, and tourist receipts and other invisibles would be diverted from black markets into a legal free market.

The Executive Directors objected to the free market. In the first place, they thought that it was an introduction of a multiple currency practice which was not necessary to achieve the trade objectives sought by France. In the second place, many Directors thought that the free market would be liable to have adverse effects on other members. There would be scope for competitive depreciation. Trade could be distorted by French traders purchasing goods in soft currency areas and selling them for hard currencies. France would be able to buy raw materials for soft currencies and sell finished goods for dollars, thereby giving French manufactures a substantial price advantage. This would hamper the eventual achievement of multilateral trade.

As an alternative to its proposal, France was prepared to institute a free market for all currencies, but this was also objected to by the Fund, as in effect destroying the whole system of fixed and stable par values. The Directors proposed that the free market should be limited to nontrade transactions, but this was not acceptable to the French authorities. A considerable effort was made by the Executive Directors to work out some alternative acceptable to all countries, but without success. France proceeded with its measures, and the Fund declared France ineligible to use its resources.

These difficulties between the Fund and France were, however, short-lived. In October 1948 France made changes in its exchange system which restored orderly cross rates for trade transactions. In September 1949, following the devaluation of sterling, the French Government consulted the Fund on a proposal to unify its exchange system at the exchange rate for the U.S. dollar prevailing in the free market—350 francs per U.S. dollar—which involved a further depreciation of the franc against the dollar. The Fund welcomed the modifications proposed by the French Government. Between September 20, 1949 and May 1950, the range of fluctuation in the franc-dollar rate was less than 0.3 per cent. Thus France had introduced a stable unitary exchange rate, although not a new par value.

Implementing the December 1947 letter

In the next few years, a flood of multiple exchange rates was proposed for consideration and action by the Executive Board. There were, for example, some twenty occasions in 1948 and some twenty-five in 1949 on which broad changes in their exchange systems were proposed by members, and several smaller changes of rates or shifts of commodities from one rate to another.

Ways of implementing the general policies laid down in the December 1947 letter had to be worked out to deal with individual cases. Formulas had to be found for approving or disapproving specific practices. In addition, some procedure had to be developed for dealing with countries—although these were a minority—which did not consult the Fund at all.

The member’s action in one or two instances seemed almost like deliberate flouting of the Fund. But in most cases the lack of consultation was due to misunderstanding. Changes were sometimes made after informal discussion with the staff, or after a visit to the member country of a staff mission, which the member had regarded as “consulting” the Fund. Other changes represented progress toward a unitary rate structure, and some members believed that steps of this sort should not be subject to prior consultation and approval. In instances of nonconsultation, the Directors either communicated with the member in an informal way, calling the member’s attention to its obligations and requesting an explanation, or took a decision drawing the member’s attention to the need for prior consultation.

A question related to nonconsultation arose where the member did not allow adequate time for the Fund to consider a measure before it went into effect: What constitutes effective consultation? In December 1949, the Board sent a letter to all members carefully explaining the need for time for the staff to prepare economic and legal analyses, as well as time for consideration by the Executive Directors, and specifically requesting members to notify the Fund at least seven days before multiple rate changes were to go into effect.

From time to time the details of a particular practice had to be examined by both the economic and the legal staff in order to determine whether or not a multiple currency practice as defined by the Fund existed. Many different practices, in addition to straightforward fixed or free market rates, were soon uncovered, some of which gave rise not to explicit but to implicit rates. Among these practices were “mixing” rates, import surcharges, compensation arrangements, aforo techniques, negotiable and nonnegotiable exchange certificates, a variety of export bonus and import entitlement schemes, and “cheap currency” schemes. Of these, “mixing” rates were by far the most common. A “mixing” rate occurs when a specified proportion of exchange proceeds is sold at the official rate and the remainder is sold in a free market, resulting in an effective rate of exchange which is a weighted average of the two rates at which the exchange has been sold.11

Because of the number of multiple exchange rate changes it had to consider, the Executive Board, early in 1949, requested the staff to study and recommend a procedure by which at least some changes in multiple currency practices—for example, “minor” ones—might be acted upon simply. Several Directors, however, were uneasy lest a procedure might be adopted which would result in changes in multiple rates being made without Directors having adequate time to consult their governments, or without their even being informed of the changes. How, they asked, would it be decided what was a “minor” change? What appeared to be a minor change on the surface might be of material significance to a particular country exporting a given commodity. And certainly it must not appear to members that the Board was relinquishing any authority in this field.

Nevertheless, a procedure for dealing with minor changes in multiple rates had finally been worked out by May 1951. When a member country proposed to the Fund a change in its multiple currency practices which the Managing Director (with the assistance of the staff) regarded as minor, the Managing Director was to inform the Executive Directors, stating that he did not intend to place the matter on the agenda unless a Director so requested by a specified date. In the absence of any such request, the minutes of the next Board meeting would record the proposal as being approved. This gradually became known internally in the Fund as the “minor change procedure” or the “lapse of time procedure.”

This procedure was similar to one that had been worked out earlier for China’s multiple rates. While still on the mainland, the Chinese authorities began to maintain a series of multiple exchange rates which were, in the midst of civil war and severe inflation, almost continuously being altered. In view of the unusual difficulties that confronted the exchange authorities in China, a special procedure for dealing with them was evolved in 1948 by which China would inform the Fund of changes in advance where possible, or in any case at the earliest possible time, and the Managing Director would reply to China promptly on behalf of the Fund, without a Board meeting, unless the Managing Director preferred to raise the matter with the Board.

In addition to problems of procedure, issues of substance, even more intractable, concerned the Board. Member countries, and also economists who were closely concerned with the subject, came to realize that cogent arguments might be advanced for multiple rates. Some of these were economic, others administrative. It was contended, for example, that multiple currency practices represented an attempt to use the exchange rate mechanism to adjust the balance of payments when it was not politically possible to change the par value.

There were even times when selective exchange rate devaluation was preferable to general devaluation—when general devaluation would be inflationary, or would raise the cost of living of essential items, or would bring excessive windfall profits to exporters but would not help significantly to solve the balance of payments deficit. In regard to the last mentioned, much was made of the price inelasticity of both the demand for and the supply of primary products. It was argued that these inelasticities made multiple rates necessary so that import demand might be restrained without the export rate being depreciated. In these circumstances, use of an exchange spread—that is, maintenance of the exchange rate for exports but depreciation of the effective rate for imports—might be preferred to the depreciation of a single rate.12

Multiple rates were also defended as merely another form of exchange control, like quantitative restrictions. Here it was important that they were easier to administer than quantitative controls. A multiple rate system could be handled by a small group of trained bankers who bought and sold foreign exchange, whereas quantitative controls required a comprehensive system of licensing, exchange budgeting, and exchange allocation. It was argued, in addition, that if a choice had to be made, multiple rates were preferable to quantitative restrictions because they distributed foreign exchange on a basis of price rather than of ability to obtain a license. They constituted a mechanism by which the arbitrary decisions of exchange control authorities, inherent in quantitative restrictions, were thought to be avoided. They imposed less interference on consumer choice

Another rationale of multiple rates was that they might have monetary effects which would alleviate inflation. Multiple rate systems usually yield profits in local currency to the exchange authorities: exchange is sold to importers at higher rates (that is, more units of local currency per unit of foreign exchange) than the authorities pay out to the exporters surrendering foreign exchange. If these profits are spent by the authorities, no money is withdrawn from circulation. But to the extent that the authorities refrain from spending these profits—that is, if they freeze them—a decline in money occurs. This anti-inflationary effect of a multiple rate system has repercussions on the balance of payments of the country concerned additional to the effects of devaluation. Both help to restrain import demand.

In these early years the objections of the Fund to multiple rates were of several types. Most frequent were the fear of harmful effects on other countries, especially through broken cross rates, and, although to a lesser extent, the fear of competitive depreciation. Another was that multiple exchange rates tended to perpetuate themselves, as vested interests in their continuation developed: complicated exchange systems had not, in practice, led eventually to a unitary rate; multiple rates had, in fact, proliferated.

Another objection to multiple currency practices was that, by making official rates nominal, multiple rates tended to undermine par values. Because most of the countries using multiple rates were subject to strong inflationary pressures, their exchange rates had to be adjusted frequently. Experience had demonstrated that, shortly after a country instituted multiple rates, its official exchange rates or its agreed par value no longer applied to many transactions.

Many Directors objected to multiple currency practices on yet another ground: they did not help to overcome the underlying economic problems of the countries employing them. Devaluation via multiple rates was usually inadequate and partial. Resort to multiple rates in many instances had not avoided the need for quantitative controls to curb import demand. When used for their anti-inflationary effects, multiple rates were insufficient to restrain inflationary pressures; primary reliance still had to be placed on the usual monetary and fiscal policies.

The Fund was, in the next few years, to learn even more about the adverse consequences of multiple rates, as well as the circumstances in which they might be useful; these are discussed below.

Adaptation of policies

As the arguments about the relative merits of multiple rates continued, it became evident that multiple rates usually resulted from balance of payments deficits, which in turn were most often caused by inflation. Until inflation was brought under control, the removal of multiple rates would necessitate, in their place, the introduction or tightening of quantitative restrictions on imports and payments. Since quantitative restrictions were themselves a form of control, and gave rise to many difficulties, the replacement of multiple rates by such restrictions would achieve little. (The Fund’s first Exchange Restrictions Report, dated March 1, 1950, used the term cost restrictions for multiple currency practices and referred to quantitative restrictions as supply-type restrictions. These expressions did not subsequently become common Fund usage, although they occur in unofficial discussions of multiple rates.)

There was also increasing evidence of the widespread use of multiple rates for purposes not concerned with the balance of payments, such as taxation and protection. Therefore, the Fund realized that insistence on the immediate discontinuation of multiple rates might compel members to put into effect hastily devised tax substitutes, different in form but no better in practice than the existing systems of multiple rates. Preferable substitutes would require comprehensive reforms of fiscal systems and customs tariff schedules, and these had first to be carefully designed and then had to go through prolonged legislative procedures.

In these circumstances the general principles outlined in the 1947 letter had to be adapted to deal with specific problems. This adaptation took several forms.

One, rather than insisting on the immediate elimination of multiple rates, the Fund came to put stress on the achievement of general economic conditions which would facilitate this elimination as soon as practicable. Most important among these conditions was domestic financial stability. Thus, as part of its early consideration of multiple currency practices, the Fund came to concentrate heavily on the domestic inflation of its members.13

Two, there was gradually evolved what was in effect a case-by-case approach: each member’s request for approval of multiple rates was decided on its own merits. In certain inflationary conditions, for example, members’ exchange difficulties were considered to be related to the inadequacy of their import rates for restraining excessive import demand. Among the recommendations which the Fund made to members, along with anti-inflationary measures, were higher penalty selling rates for nonessential imports. At times the further restriction of import demand—including, occasionally, tightened quantitative restrictions—was advised.

Members were also encouraged to avoid the use of many particular practices—such as auction systems, compensation arrangements, and “mixing” rates, especially for exports—which might unduly complicate their exchange rates or harm other countries. “Mixing” systems were considered to be unduly susceptible to pressure from exporters for further rate multiplicity, as well as to administrative juggling. Compensation arrangements also gave rise to extensive multiplication of rates. In order to keep exchange rates as stable as possible, the Fund urged that free market rates be limited to balance of payments items other than trade, such as capital or invisibles. This sort of “code of fair practices,” it was thought, would keep continuing systems within tolerable limits.

Despite these difficulties, the Fund still hoped for progress in the elimination of multiple rates. Not wishing to disapprove multiple rates or to approve them outright, the Board gave approval to those proposed by individual countries as “temporary measures,” and remained in consultation with the members concerned.

During these early years, the Fund worked intensely with its members on multiple rate problems. Innumerable informal contacts were made, many staff missions were sent out, sundry reports were written, and frequent discussions took place in the Board. All this activity was almost entirely concerned with the Fund’s Latin American members, especially Chile, Colombia, Costa Rica, Ecuador, Honduras, Nicaragua, and Paraguay.

In many respects, these years were frustrating ones. The Fund was criticized at times for too much and at times for too little attention to multiple rates. Some of the Latin American countries began to ask whether the Fund, in a world of extensive exchange controls and the continued inconvertibility of key currencies, was not devoting excessive attention to the multiple rates of a few countries that contributed little to total world trade. On the other hand, some economists, both in academic and in some official circles, anxious for evidences of successful international monetary cooperation within the new organization, frequently lamented the lack of progress, even in the field of multiple rates, which was thought to be easier than that of other controls.

Reassessment of policies

In January-February 1950 these problems were brought to a head in the Executive Board, which undertook a series of informal sessions on multiple currency practices. In these sessions it was stressed that the Fund had been willing to approve multiple rates only for balance of payments reasons. Much discussion centered, therefore, on the other reasons for multiple rates, including the use of such rates for taxation and subsidization and for protection.

To assist in formulating a new policy, Directors considered a model system of multiple rates. This model was designed to take account of balance of payments deficits without making concessions to the other purposes of multiple rates. The model had one import and one export rate with a fixed spread in excess of 1 per cent, and a free market as a safety valve for excessive import demand. The various advantages and disadvantages of the model were debated. But the basic question remained, and was left unanswered: Would such a model imply the permanent use of multiple rates? No firm consolidation of views among Directors took place even after these informal sessions.


Second phase in the Funds approach

The beginning of consultations under Article XIV in 1952 led to what may be seen as a second phase in the Fund’s policies on multiple exchange rates. This phase continued until about the end of 1955. Although the legal obligation of members to consult the Fund in advance of multiple rate changes, set forth in the December 1947 letter, continued to apply, the Board’s operating procedures were altered in important respects. The Directors no longer reviewed changes in multiple exchange rates when such changes were made. Instead, they gave temporary approval to a proposed change in a multiple rate system without immediate review of a particular practice, pending the forthcoming annual Article XIV consultation. In the course of that consultation a review of the total exchange rate system was undertaken and a final decision reached on the particular rate changes made.

This change in procedure had two important advantages. First, it enabled the Fund and the member country to consider the reasons for, and possible alternatives to, multiple rates in a broad, regular (annual) context, and eventually to work together toward the elimination of multiple rates. Second, as such consultations came to be held with practically all members, including the major European ones, the dualism in the Fund’s policy between multiple exchange rates and other types of exchange restrictions seemed more or less at an end. After 1952, the consultations became the main activity of the Fund, in contrast to the previous situation, when multiple currency practices had been the great generator of the work of the Fund.

This switch in procedure—from considering multiple currency practices as they were introduced to considering them more fully in the course of the Article XIV consultations—culminated in a decision of the Executive Board, taken in February 1956, to use much more extensively the so-called lapse of time procedure for changes in multiple rates. The changes approved by the Managing Director under this procedure were, of course, more fully examined at the time of the Article XIV consultations. However, it was understood that this procedure would not be used for multiple rate changes involving shifts of payments from an official market to a new or narrow free market or to a free market in which the prevailing exchange rates were far out of line with the official rate at which most exports were handled. The Directors were particularly sensitive to multiple rate changes that jeopardized fixed exchange rates or that might in effect subsidize exports.

During this second phase, it was still evident that multiple currency practices could not be eradicated easily or quickly, and that the case-by-case approach would have to be continued for some time. A search was therefore begun for some new policy that could be implemented on a case-by-case basis.

Spread of multiple rates

The question of what policy to apply had been made even more difficult by an increasing divergence in practice between the Fund’s ultimate goal of unification of exchange systems and the existing situation. Despite the Fund’s efforts, multiple currency practices continued to spread.14 Several countries that were already employing multiple exchange rates joined the Fund: Afghanistan, Indonesia, Israel, and Korea. Many countries that were already members introduced multiple currency practices: China, Egypt, Finland, Greece, Iceland, the Philippines, Turkey, and Yugoslavia. Even a number of Western European countries, although retaining unitary fixed rates for the bulk of their transactions, felt compelled to open free markets for the sale of certain bilateral or EPU currencies, or for certain capital flows. Belgium, for example, established a free market for EPU currencies; Germany introduced a free market for “Brazilian accounting dollars.” Multiple rates were thus no longer confined to Latin America; they had spread to Asia, the Middle East, and Western Europe. Table 6 at the end of this chapter, which provides a detailed chronological summary of the principal developments in multiple rates, shows this extension.

Table 6.Principal Developments in Multiple Exchange Rates, 1946–65
Sept.Venezuela asked Fund assurance that membership would not prevent it from taking suitable time to eliminate multiple currency practices.
Nov.The question arose whether an initial par value could be agreed if multiple rates prevailed.
Jan.-Feb.The Board discussed several matters pertaining to multiple rates. Do differences exceeding 2 per cent between buying and selling rates constitute a multiple currency practice? Are multiple rates subject to Article XIV? How prevalent are they?
MayThe Board set up an Ad Hoc Committee on Spreads and Multiple Currency Practices to study the application of the Articles to multiple rates.
June-Aug.The Board considered the recommendations of the Ad Hoc Committee.
Dec.A letter setting forth the decisions on policy and procedures for multiple currency practices was sent to all members.
Jan.France devalued the franc and instituted a multiple currency practice, to which the Fund objected.
JuneThe Fund agreed to a special simple procedure for changes in China’s exchange system because of special problems in the case: the Managing Director approved in writing changes proposed by the Chinese authorities, and circulated the correspondence to the Board for information, Board consideration of the change not being first required.
Oct.-Nov.The Fund reviewed Latin American exchange problems and policies. No Board decision was taken but it was stressed that continuing inflation had been found to be a major cause of exchange difficulties and that, in view of the diversity of conditions in Latin America, proposals by each country would have to be considered separately.
Feb.During a review of the exchange position of one country, the question first arose whether a member could properly maintain multiple rates if they were not necessary for balance of payments reasons.
Apr.Instructions given to a Fund mission to one country required it to investigate the possibility of substituting nonexchange measures for the existing exchange surcharges in order to restrain imports.
Jan.-Feb.The Board held a series of informal sessions on multiple currency practices.
Mar.A separate Multiple Currency Practices Division was created in a staff reorganization.
Mar.The Philippines imposed a 17 per cent tax on sales of foreign exchange.
Mar.-Apr.Paraguay devalued the guarani, removed taxes on sales of exchange, and transferred various transactions to the free market in an attempt to meet objections of the Fund to its exchange system.
MayAn internal Fund procedure for minor changes in multiple rates was established to lessen the need for consideration by the Board of all changes in multiple rate systems.
Sept.Costa Rica abolished surcharges, reducing the complexity of its multiple rate structure; the abolition was facilitated by upward revisions in its tariffs. It was agreed by the Board that the staff should look into the general problem of surcharges versus tariffs.
Jan.After consideration of a staff paper on tariffs versus surcharges, the Board decided that no general line either favoring or disapproving elimination of multiple rates through upward tariff revisions could be taken and that each case must be considered on its merits.
Jan.-JuneYugoslavia, in a series of steps, introduced a complex system of multiple rates.
Feb.Ecuador simplified its exchange system by abolishing its compensation system.
Sept.Germany introduced a free market for “Brazilian accounting dollars” (used in payments transactions under the payments agreements between Germany and Brazil).
Feb.Brazil inaugurated a free market for invisibles and most capital transactions and for certain trade items.
Apr.Greece unified its exchange structure and devalued the drachma.
MayAustria unified its exchange structure and set an initial par value.
Aug.Brazil reduced the list of export products subject to a “mixing” arrangement, but a complex system remained.
Jan.Israel (not yet a member) substituted a single official rate for three fixed official rates, but for certain transactions either one of two premiums or a surcharge applied.
JulyThe Netherlands, Belgium, and Luxembourg, in connection with arrangements to facilitate capital movements between them and other EPU countries, extended the free market for capital to residents of other EPU countries; previously the free market had been confined to transactions between their own residents.
Aug.Paraguay depreciated to a new par value and made other substantial alterations in its exchange system, but a complex system remained.
Israel (now a Fund member) extended its surcharge to all imports.
Jan.-Feb.Brazil established export bonuses.
Feb.-MayColombia abolished a differential rate for coffee exports and introduced a broad free market; imports were made subject to stamp taxes.
Mar.China introduced a system of negotiable exchange certificates.
MayChile eliminated the special exchange rate for copper companies.
JulyNicaragua removed surcharges and depreciated its par value but retained an exchange spread.
Belgium’s free market was opened to capital transfers to all destinations.
July-Oct.Indonesia considerably revised its system in an attempt at simplification, but a wide range of rates, especially on imports, continued.
Sept.-Dec.Uruguay announced a special premium for exports of wool, including wool tops.
Jan.Thailand completed unification of its exchange system; all transactions were to be effected at a fluctuating free market rate.
Mar.Paraguay depreciated its par value and greatly simplified its complex multiple rate system. A broad internal monetary stabilization program was introduced.
The Board agreed to more extensive use of the lapse of time procedure for multiple rates: it could now be used for all changes that did not, in the opinion of the Managing Director, seem to require immediate discussion by the Board.
Apr.Chile considerably simplified its multiple rate system, abolished the import licensing system, and instituted a fluctuating free market for all commodity transactions.
Dec.Bolivia undertook an exchange reform and comprehensive economic stabilization measures, replacing the complex multiple rate system with a single fluctuating rate.
Mar.The Board began consideration of a staff paper on a change in policies for multiple currency practices which would put an end to Fund approval of complex systems and would encourage genuine and substantial simplification.
MayIran, as a final step in unifying its multiple rates, changed its par value to what was the de facto unitary exchange rate.
JuneThe Board reached a second general decision on multiple currency practices.

The Fund notified countries that it would no longer approve complex systems unless the countries maintaining them were making reasonable progress toward simplification.
Aug.Paraguay gave up all transactions at the par value, eliminated all multiple rates and quantitative restrictions, and adopted a freely fluctuating rate.
France introduced a 20 per cent surcharge and premium on most transactions, which was extended to all transactions shortly thereafter.
Sept.Finland depreciated its par value and abolished its multiple currency practices.
Nov.Nicaragua applied the official rate to exports of cotton and minor exports, thus moving toward unification.
Jan.-Dec.Uruguay made various changes in its rate structure, making it more complex.
Apr.China simplified its exchange system, abolishing its 20 per cent defense tax and other rates. There were now two effective rates.
MayIceland introduced three exchange premiums on export proceeds, replacing the assistance previously granted to exports through different forms of subsidization. Discrimination as to country destination was terminated.
JulyYugoslavia continued to simplify its complex rate structure by reducing the number of export and import coefficients.
Aug.Turkey replaced its complex system with a simpler one.
Oct.Nicaragua extended the par value rate to coffee exports; thus by the end of the crop year all trade transactions were unified at a rate based on the par value.
Dec.Argentina, as part of an economic stabilization program, abolished multiple rates and introduced a single fluctuating rate.
Jan.Chile combined its two fluctuating exchange markets.
Pakistan introduced an export bonus scheme.
JulySpain set a par value and abolished all multiple rates.
Aug.China established a single fluctuating rate for all transactions except government payments.
Indonesia abolished its exchange certificate system and introduced other multiple rates.
Turkey modified its export premiums so that a single rate applied to all transactions except exports of two commodities.
Feb.Iceland eliminated its multiple rate system, and depreciated its par value.
Apr.The Philippines introduced a fluctuating free market for most transactions, with a “mixing” rate for exports.
MayPeru unified its two exchange markets.
Dec.The special procedure for changes in China’s exchange system arranged in June 1948 was discontinued.
Jan.Yugoslavia abolished its multiple rate system and introduced a unified fixed rate for all transactions except receipts from tourism.
May-JulyBrazil introduced new exchange arrangements for coffee export proceeds and transferred to the free market all imports previously at preferential rates.
China introduced a fixed rate for all transactions.
Ecuador replaced a complex multiple rate system with a new par value applying to most transactions; only a minor free market with a fluctuating rate was retained for some transactions in invisibles and unregistered capital.
Costa Rica set a new par value and except for temporary arrangements abolished all multiple rates.
Jan.Chile again instituted two exchange markets.
The Philippines introduced a fluctuating rate for all transactions except merchandise exports, to which a “mixing” rate would apply.
Feb.Israel set a new par value and eliminated multiple rates.
MayThe United Arab Republic replaced its multiple rates with a single rate.
Aug.Costa Rica removed its last multiple rate.
Oct.Indonesia reclassified imports in its complex rate structure.
Mar.Nicaragua transferred the remaining items from the free to the official market, so that a single rate of exchange applied to all transactions.
Afghanistan devalued its official rate, set an initial par value, and greatly simplified its complex rate structure.
Jan.Laos devalued its official rate and established a legal free market for imports other than aid goods, and for tourism, capital transfers, and other payments.
Venezuela considerably simplified its exchange system through a series of steps worked out with the Fund.
Mar.-JulyBrazil undertook several simplifications of its exchange system.
MayKorea established a unitary rate which was only nominally fluctuating, and abolished multiple rates.
Jan.-Dec.Brazil changed the effective rates applicable to coffee exports on several occasions throughout the year.
Mar.Korea established an exchange market in which the rate was allowed to fluctuate, and substantially liberalized its restrictive system.
Nov.The Philippines unified its exchange system, and set a new par value of
3.90 = US$1.
Dec.Somalia introduced a temporary 3 per cent exchange tax for revenue purposes.

Reasons for extension of multiple rates

The increased use of multiple currency practices during these years reflected two distinct kinds of problem, each of which gave rise to different practices. One problem was what was commonly known at the time as the world dollar shortage. The currencies of the world had become divided into “hard” currencies, mainly the U.S. and Canadian dollars, which were convertible, and “soft” currencies, including the pound sterling, the French and Belgian francs, the deutsche mark, and the guilder, which were inconvertible. Although a country might have achieved equilibrium in its aggregate balance of payments, a surplus in an inconvertible currency could not be used to settle deficits where convertible currencies were needed.

This situation had led, in Western Europe as elsewhere, to some multiple currency practices—such as limited free markets and broken cross rates in broader free markets—and to retention quotas, which aimed at increasing a country’s supply of “hard” currencies and decreasing its holdings of “soft” ones.

The trend to such practices began to be reversed in 1954–55 as a result of the marked general improvement in the international payments situation which had started to take place. On January 23, 1954, the Report of the Randall Commission of the U.S. Government noted, for example, that the conditions prerequisite to currency convertibility were more nearly in prospect at that time than at any previous time since the war.15

By the spring of 1954 the Fund was able to discern a general tendency toward the removal of barriers to trade and payments; restrictive practices had been considerably reduced or modified.16 The Exchange Restrictions Report for 1955 called attention to continued—although more gradual—progress in relaxing restrictions and, in particular, to increased facilities for nonresidents to convert inconvertible currencies into dollars.17

The second reason for the spread of multiple rates, applying to the less developed countries, was a more difficult one to counter. Multiple exchange rates, originally introduced to alleviate the depression-based payments problems of the 1930’s, now seemed useful instruments to support economic development efforts, particularly in the economic and institutional environment of developing countries.

The problems of establishing or maintaining internal monetary stability were often intensified by the pressing claims of development programs. Multiple rates could, as described above, have an anti-inflationary effect if exchange profits were frozen. Dual exchange markets could isolate speculative capital from trade transactions. Alternatively, dual markets could separate exports of primary products, which might not benefit from devaluation, from exports of manufactured goods, where an exchange rate incentive might induce new export industries, so essential to development. On the import side, multiple rates might offer protection to some domestic industries; or they might provide subsidies to essential consumer goods, thereby keeping down the cost of living. Additional arguments for multiple rates as an aid to development were also advanced by member countries.

Multiple rates thus persisted in developing countries. Several countries, dissatisfied with the results of quantitative controls, switched to multiple rates. The Annual Report of the Executive Directors for 1955, after noting that world trade and payments were less subject to restriction than they had been a year earlier, commented that much less progress had been made in the elimination of multiple currency practices.18

At the same time, the problem of multiple export rates had become of added concern to countries not employing them. The less developed countries had begun to increase their use of multiple rates for exports, and keener competition in international markets made countries with unitary rates increasingly fearful that competitors using multiple rates might derive unfair advantages in export markets. It seemed certain that tension between the Fund’s members concerning multiple rates might again increase, as it had in the Fund’s first days.

In these circumstances both the staff and the Board began to consider even more intensively the economic functions and effects of multiple rates, especially the relation between multiple rates and economic development. The conclusion was that the use of exchange restrictions and multiple rates to foster economic development would be unnecessary if substitute measures were introduced. Since multiple rates were, for example, often used in place of customs duties or taxes, they could be eliminated if equivalent revenue was raised in another way. Similarly, multiple rates that provided protection to domestic producers could be abolished if protective tariffs were appropriately raised. Increasingly the devising of acceptable alternative measures for multiple rates became a main problem for the Fund. Meanwhile, the questions facing the Fund were these: Should it condone, or at least not object to, these uses of multiple rates for reasons other than the balance of payments? Should it recommend specific alternatives?

As background for the Board’s consideration of these problems, the staff also prepared studies of the revenue-raising functions of multiple currency practices 19 and of the desirability of substituting increases in tariffs for exchange surcharges. In the latter study, a series of propositions was advanced as to the economic circumstances for which tariffs and surcharges, respectively, were most appropriate. Essentially, surcharges were considered best suited to meet temporary balance of payments deficits. Consequently, when surcharges were used to help cope with temporary payments deficits, they should not be incorporated in the customs tariff schedule. When, however, surcharges were serving a longer-run balance of payments purpose, the study deemed it advisable for the country concerned to introduce anti-inflationary measures or to adjust its exchange rate rather than to revise its customs tariff rates. On the other hand, the study concluded that tariffs were preferable to surcharges as a long-run source of revenue and for protection.

Consideration of these staff suggestions showed that Directors held widely differing views. After much discussion the Board came to the conclusion that generalizations on alternative measures for multiple rates were difficult and dangerous, and that it was preferable to consider the individual circumstances of each country.

By the end of 1955, membership of the Fund had grown to 58. But, despite some lessening of multiple rates, there were still 36 members that had some kind of practice which the Fund considered a multiple currency practice. Many of these practices, especially in Europe, were minor, and most transactions were still conducted at exchange rates based on par values. But there remained very complex systems with many different rates in many countries throughout the world.


Toward the end of 1955, a third phase in the history of the Fund’s dealings with multiple rates began. In the Exchange Restrictions Report for 1956, the Fund was able to report considerable progress toward the lessened use of multiple rates:

  • In the past year, many countries maintaining multiple exchange rates and similar practices made some changes in these practices, which varied considerably in character and significance. The countries that did not change their multiple currency practices are, for the most part, those in which the practice is only a minor element in the exchange rate structure. The number of countries in which multiple currency practices were simplified appreciably exceeded those in which there was an increase in multiple rates.20

The intensive postwar investment programs, especially in Western Europe, had borne fruit, and production was rising rapidly. Increased supplies of goods, together with the improved domestic financial and monetary positions of Western European countries following the realignment of exchange rates in 1949, brought the world dollar shortage to an end. Hence, from 1956 onward, the main industrial countries relaxed many of the barriers to trade and payments and made their currencies convertible. Multiple currency practices, as well as discriminatory quantitative restrictions, which distinguished between “hard” and “soft” currencies could be, and were, quickly removed. By 1958 the free markets in which some but not all transactions took place, broken cross rates, retention quotas, and other special devices of a number of Western European countries had all been discontinued.

Toward the end of 1955 and in 1956 events occurred which were, in some ways, even more unexpected. Three less developed countries—Bolivia, Chile, and Paraguay—which had maintained multiple rates since the 1930’s and had believed them imperative for primary producing economies, undertook comprehensive exchange reforms, eliminating multiple rates at a single stroke and introducing unitary fluctuating rates. Thailand also achieved a unitary fluctuating rate, but gradually. In addition, several other multiple rates, such as an exchange tax in the Philippines and the application of an official rate to oil company transactions in Syria, were eliminated.

New Fund approaches

The time was ripe for an intensification of the Fund’s efforts to eliminate multiple rates. The countries employing them now were less convinced of their usefulness than they had been several years before. Greater emphasis was being placed on the practical difficulties of eliminating multiple rates than on a preference for them over unitary rates.

The experiences of the Fund in the previous ten years had demonstrated that, for most countries with complex systems, unification of the exchange system required the mapping out of a special course of action. The elimination of multiple rates could not be expected to come about automatically as internal monetary conditions were brought satisfactorily under control, or even as balance of payments positions were strengthened. Moreover, in many countries, unification at a realistic par value was not likely to be attained in one step: what was required was either a process of simplification to a few fixed rates or to one fixed and one free rate, or, alternatively, unification at a fluctuating rate.

The Fund had also learned that it was much easier for a country to simplify its exchange rate structure than to unify it. The necessary degree of stringency in internal monetary policies was significantly smaller for simplification than for unification, and fewer reserves were required if a simplified system included a fluctuating rate. Over the years it had become evident that floating rate mechanisms were often more self-liquidating than a group of fixed rates. Another advantage of simplification as against unification was that an exchange spread, such as often characterized simple multiple rate systems, might absorb part of any internal monetary expansion; such absorption was not possible with a unitary rate.

Moreover, it had been learned that it was the complexity of multiple rates that was especially at fault. Complex exchange systems brought into economic decisions many more elements of arbitrariness and uncertainty. Extensive differentiation of rates among export and import commodities was particularly likely to distort domestic prices and output and made it more difficult for the authorities to resist pressures for special treatment of a given commodity. Further, an excessive number of rates increased the possibilities of hidden subsidization and encouraged the continuation of overvalued rates. But perhaps the most important condemnation of complex rate systems was that they tended to perpetuate themselves until the countries using them experienced crises which might well have been avoided.

For several months in 1957 the Board carefully considered the possibility of making new and intensified efforts to reduce multiple rates. All the Directors were anxious to reach a decision that would be widely supported, especially by the developing countries that maintained multiple rates. In June 1957 this consensus was reached, and the Board decided to send to the members a new communication on multiple currency practices.21 This memorandum urged countries using such techniques to simplify considerably their existing rate structures. Such a simplification meant more than a mere reduction in the number of existing rates: emphasis was placed on limiting the system to two or three rates which would be sufficiently realistic to maintain a satisfactory balance of payments with a minimum of restrictions. A simpler system could include, where necessary, the temporary use of a widened free market or of a fluctuating rate as a step toward a stable unitary rate.

The Fund undertook to help countries to plan specific exchange systems. Technical assistance could also be made available to cover the formulation, where appropriate, of substitute fiscal arrangements. Most important of all, where the exchange reform was sufficiently broad and was accompanied by a domestic stabilization program, the Fund was willing to consider permitting members to use its resources in support of exchange reforms. Meanwhile, the Board would be reluctant to approve changes in multiple rate systems which would make them more complex.

While not insisting on the immediate unification of exchange rates, this communication sounded a warning that the Board would no longer approve complex systems except in certain circumstances.

Accelerated elimination of multiple rates

For several years after the issue of the June 1957 memorandum, there was a strong trend toward elimination of multiple exchange rates by the developing countries. Significantly, the countries abandoning multiple rates included Argentina (1958), which in the 1930’s had been one of the first Latin American countries to use them, and Yugoslavia (1961), which had had in the mid-1950’s an exchange system often regarded as the most complex of all. Table 6 at the end of this chapter gives additional details of the decline in the use of multiple rates.

Some countries devalued, eliminated multiple rates, and set a new fixed rate, all at the same time. Generally these were the European countries, but they included Israel. This was the pattern in Finland, Spain, and Turkey. Many of these countries also set new par values. Others—especially the less developed countries—reached a par value by stages, first eliminating multiple rates and introducing a unitary fluctuating rate, then gradually stabilizing the fluctuating rate, and eventually converting it into a par value. Costa Rica, Iran, Jordan, Korea, Nicaragua, the Philippines, Saudi Arabia, and Thailand attained fixed rates (or even par values) via fluctuating rates, at least for some transactions.22

A few countries eliminated multiple rates when they were enjoying a favorable balance of payments; usually the strong payments situation was the result of peak production or high world prices for exports, or both. More frequently, however, countries reshaped their exchange rate structures as part of a broad program to meet balance of payments crises. These general reforms of the exchange system were usually accompanied by exchange devaluation and by internal policies designed to remove price distortions and to restrain inflation.

Frequently also tariff adjustments, usually upward, were undertaken simultaneously with exchange reform. In some countries, comprehensive fiscal reforms were carried out. Other countries, introducing a single fluctuating rate, temporarily applied taxes or surcharges to some imports. Apart from whether or not these taxes and surcharges were different from multiple rates insofar as the Fund’s Articles were concerned, their economic function was different. Because they were now superimposed on unitary exchange rates that were unrestricted and hence presumably realistic, these taxes and surcharges were regarded purely as fiscal devices; previously, because they comprised part of the exchange system, they had been considered at least in part as balance of payments measures. The main reason why countries used taxes and surcharges, after exchange reform, was that alternative revenue sources could not be quickly developed. Some countries proposed explicitly to undertake tariff revision at a later date, at which time these taxes and surcharges were to be incorporated into the tariff structure.23

In sum, between 1955 and 1962, not only did several Western European countries eliminate their minor exchange practices, but major systems of multiple rates were unified in many other countries as well, in every geographic region. By the end of 1962, only 15 Fund members out of a total of 82 had multiple rates of any type, in contrast to 36 out of 58 members seven years earlier.

Explanations of decline in multiple rates

One of the principal reasons for the decline in multiple rates was the adverse experiences with these practices that many countries had had. Although some countries had been able to use them to advantage, the experiences of most had been disappointing and frustrating. In order to achieve any of the professed purposes of multiple exchange rates—that is, to tax a major export, to restrain imports of luxury commodities, to raise revenue, to alleviate inflation, to stimulate minor exports, or to isolate capital movements from trade transactions—countries had found that certain conditions had to prevail.

First, it was essential that the exchange system be simple; that is, there should be only a few exchange rates rather than many. Complex systems worked out much less well; they rarely achieved several objectives concurrently. Finding a rate structure suitable for such diverse purposes as holding down external deficits, redistributing internal income, and checking inflation seemed to be virtually impossible, even where several rates existed. Seeking to make multiple rates perform both balance of payments and taxation functions, especially under conditions of chronic inflation, led to complex systems and the dissipation of the initial objectives.

Second, periodic adjustments had to be made in the exchange system or else fluctuating rates had to be used in order to keep at realistic levels the exchange rates applicable to most exports and imports; otherwise, there was a danger that the exchange rates prevailing would become overvalued.

Third, a reasonable amount of domestic monetary stability had to be maintained.24 Under conditions of inflation, traditional export industries were harmed by being subjected to penalty rates of exchange at overvalued levels. New and potential export industries were neglected, even where special exchange rates for so-called minor exports existed. Multiple import rates had adverse effects on the structure of domestic production and investment.

These conditions for successful use of multiple rates had not always, or even frequently, been fulfilled in the countries that had employed multiple rates, and multiple rates had then proved worse than practicable alternatives.

Following the memorandum of June 1957, the Fund vigorously encouraged countries to eliminate multiple rates. On many occasions the Fund urged countries not to be unduly cautious in moving toward simplified exchange systems. It drew attention to specific circumstances where, in the Fund’s view, greater simplification was feasible and even essential. Frequently it singled out for comment to a member exchange rates that were considered especially troublesome. The Fund provided technical assistance to work out specific exchange reforms and internal stabilization programs. Finally, there were stand-by arrangements to bolster countries’ reserves.

Once some countries had eliminated complex systems of multiple rates, the Fund could point to the favorable aftermath. In their Annual Report for 1961 the Directors had this to say:

  • In arguing the advantages of [the removal of multiple rates], the Fund has been able to point to the generally favorable experience of several members that have eliminated complex multiple exchange rate systems in the recent past. In general, these countries have found that their economies are better able than previously to deal with both domestic and balance of payments problems. In some countries, this improvement occurred despite lower prices for major export products upon which they largely depend for foreign exchange. In several of the countries which adopted freely fluctuating exchange rates when they abolished their multiple exchange [rate] systems, these rates remained virtually unchanged for long periods. For some, moreover, the initial, sometimes difficult, period of adjustment to the new conditions that inevitably follows a major economic reform has been succeeded by a period of renewed progress, accompanied at the same time by monetary and exchange rate stability and an increasingly satisfactory reserve position.25


Developments in multiple exchange rates in the eighteenth to twentieth years of the Fund’s history were of three types.

(1) Countries which had long made use of multiple rates continued to join others which had considerably simplified or unified their exchange rates. In 1964–65 Venezuela and Brazil simplified their exchange systems in a series of steps. The Philippines, which had introduced multiple rates with a fluctuating rate in 1962, eliminated them and set a new par value in November 1965. Also in 1965, Korea replaced its multiple rates with a unitary fluctuating rate.

(2) A few countries, including Afghanistan, Colombia, Indonesia, and Viet-Nam, continued to maintain significant multiple rate systems, although some of these became somewhat less complex.

(3) On the other hand, certain countries where multiple rates had been eliminated or very much simplified reintroduced them or reverted to complex systems. And in some countries which previously did not have them, multiple rates appeared. Chile, for example, following a virtual unification of its exchange rate in 1956, introduced in 1962 a dual market system which still continued at the end of 1965. Multiple rates were adopted in several countries where they had not been used previously. Pakistan, to assist exports of all except its major primary products, had in 1959 introduced an export bonus scheme patterned along the lines of the retention quota employed by Germany in the mid-1950’s. This system was still in use at the end of 1965, and its scope and complexity had been increased. A few new African countries had begun to use devices reminiscent of those used in Latin America in the 1940’s.

Some other devices, such as import surcharges and advance deposit requirements, which are in many respects similar to multiple currency practices, were instituted by several countries to deal with recurrent payments or fiscal deficits. Among these countries were Argentina, Brazil, Ceylon, Colombia, Ecuador, India, Iran, Ireland, Japan, Pakistan, and Uruguay. Moreover, among the measures introduced by a number of industrial countries to reduce the outflow of capital were special taxes or markets with premiums for investment currencies, as in the United Kingdom.26

The problem of multiple exchange rates may never be entirely solved. The nature of and reasons for the practices used differ, their locale shifts, and there is a constant need for vigilance by the Fund. Despite some continued and some newly emerging problems, however, the extent and complexity of multiple rates were clearly less at the end of 1965 than they had been twenty years earlier.


The established technical term of the Fund for any multiple rate is multiple currency practice. In what follows, the terms multiple exchange rate and multiple currency practice are used interchangeably.


Further details of the interwar experiences with multiple rates can be found in Margaret S. Gordon, Barriers to World Trade, Chapters V and VII; and in League of Nations, International Currency Experience. A detailed description of European exchange controls in the interwar period is in Howard S. Ellis, Exchange Control in Central Europe. League of Nations, Report on Exchange Control, briefly examines the economic effects of exchange controls, especially on the domestic economies of the countries concerned.


League of Nations, World Economic Survey, 1938/39, p. 202.


Vol. III below, p. 86.


Vol. III below, p. 135.


Vol. III below, pp. 160–61.


Argentina was not an original Fund member; the other Latin American countries mentioned were.


Robert Triffin, The World Money Maze, p. 141.


Ibid., pp. 142–77.


Reproduced below, Vol. III, pp. 262–65.


Detailed descriptions of all these various arrangements can be found in the individual country surveys of the Annual Report on Exchange Restrictions, especially for the years 1950–55, and in the monthly issues of International Financial Statistics.


These economic objectives of multiple exchange rates are explained by E. M. Bernstein in “Some Economic Aspects of Multiple Exchange Rates,” Staff Papers, Vol. I (1950–51), pp. 224–37.


Annual Report, 1948, p. 28.


Fourth Annual Report on Exchange Restrictions (1953), pp. 40–50.


U.S. Commission on Foreign Economic Policy, Report to the President and the Congress, p. 470.


Annual Report, 1954, p. 74.


Sixth Annual Report on Exchange Restrictions (1955), pp. 2–6.


Annual Report, 1955, p. 76.


W. John R. Woodley, “The Use of Special Exchange Rates for Transactions with Foreign Companies,” Staff Papers, Vol. III (1953–54), pp. 254–69; Joyce Sherwood, “Revenue Features of Multiple Exchange Rate Systems,” ibid., Vol. V (1956–57), pp. 74–107.


Seventh Annual Report on Exchange Restrictions (1956), p. 5.


Reproduced below, Vol. III, pp. 265–66.


These alternative approaches have been described in Margaret G. de Vries, “Fund Members’ Adherence to the Par Value Regime,” Staff Papers, Vol. XIII (1966), at pp. 516–20.


A discussion of the techniques of unification of multiple rates is to be found in F. d’A. Collings, “Recent Progress in Latin America Toward Eliminating Exchange Restrictions,” Staff Papers, Vol. VIII (1960–61), pp. 274–86.


A fuller account of the economic experiences of countries with multiple rates can be found in Margaret G. de Vries, “Multiple Exchange Rates,” Staff Papers, Vol. XII (1965), pp. 282–313.


Annual Report, 1961, p. 117.


See Seventeenth Annual Report on Exchange Restrictions (1966), p. 6.

    Other Resources Citing This Publication