Current Developments in Monetary and Financial Law, Vol. 3

CHAPTER 37 Use of a Foreign Currency Under the IMF’s Articles of Agreement

International Monetary Fund
Published Date:
April 2005
  • ShareShare
Show Summary Details

In the legal framework of the IMF’s Articles of Agreement, a number of provisions setting out the rights and obligations attached to membership imply that each member will have its own currency. However, in practice, countries have been able to accede to membership in the IMF, to exercise their rights, and to perform their obligations while using another member’s currency. Whether the issuer of that currency can object to the use of its currency by another country in its relations with the IMF or as legal tender raises questions of international law.

In 1944, when the Articles of Agreement of the International Monetary Fund were adopted at the Bretton Woods Conference, all the countries represented at the Conference had national currencies. Some even had several currencies, one for their metropolitan territories, others for their dependent territories. The possibility for a country to have different currencies for different territories was acknowledged in the Articles of Agreement.1

Using Its Own Currency

In three of the countries represented at the Conference (Dominican Republic, Liberia, and Panama), the main currency in circulation was the U.S. dollar; in two of them (Liberia and Panama), the U.S. dollar was even legal tender. However, those three countries also issued coins that were legal tender and were regarded as national currencies.

Whether a country that did not have a national currency could be admitted to membership in the IMF was not discussed at the Conference, probably because the idea that a country could decide not to exercise its sovereign power—with the attendant benefits (seigniorage)—of issuing its own currency seemed rather unrealistic. The assumption was that each country would have at least one national currency, and several provisions of the IMF’s original and present Articles of Agreement are based on a distinction between a member’s currency and the currencies of other members. In those provisions, both rights and obligations of each member are formulated in terms either of its own currency or of the relationship between its currency and other members’ currencies:

(1) When a country is admitted to membership, it is assigned a quota. The member’s subscription is equal to its quota and must be paid in full. One part is paid in the member’s currency (75 percent) and the balance in reserve assets (gold under the original Articles; special drawing rights and/or currencies of other members as specified by the IMF, under the present Articles).2 The same rules apply when a member’s quota is increased.3

(2) Under the par value system, each member had to maintain exchange rates based on the parity between its currency and the currencies of other members.4 After the de facto end of the par value system in the early 1970s, the Second Amendment of the IMF’s Articles, which became effective in 1978, allowed each member to choose between fixed and floating exchange rates.5 Therefore, a member may choose to maintain the value of its currency in terms of another currency or a basket of currencies (“pegging”). It may also decide to maintain the value of its currency in terms of the special drawing rights or another denominator (other than gold). The IMF must be notified of those exchange arrangements.6

(3) In the General Resources Account, the IMF holds the currencies, gold, and special drawing rights corresponding to its members’ quota subscriptions. These currencies and special drawing rights can be used by the IMF to provide financial assistance to its members facing a balance-of-payments problem; the gold can be sold and the proceeds used for the same purpose, except that the capital gains on those sales have to be used outside the General Resources Account for financial assistance to developing countries. The IMF’s financial assistance from the General Resources Account does not take the form of loans but rather of reversible sales of foreign exchange or special drawing rights in exchange for an equivalent amount of the purchasing member’s currency.7 These transactions are similar to the currency swaps between two central banks. Like swaps they must be reversed within a specified period; this period is determined by the policies of the IMF applicable to the facility under which the purchase was made. The repurchase consists of a payment of special drawing rights or other members’ currencies, as specified by the IMF, in exchange for a return of an equivalent amount of its currency to the member.8

(4) As long as the IMF holds a member’s currency, whether as a result of a subscription payment, purchase, or for any other reason, in the General Resources Account, the member must maintain the value of these holdings in terms of the special drawing rights.9 If the currency depreciates, additional payments to the IMF must be made.

(5) The IMF holdings of a member’s currency are not physical assets (notes and coins). They are balances held with a depository,10 or promissory notes issued by the member or its depository.11 The depository is the member’s central bank; if the member has no central bank, the depository is another institution designated by the member and acceptable to the IMF.

(6) The IMF may replenish its holdings of a member’s currency by borrowing that currency from the member. The IMF may also borrow that currency from another source. In both cases, the member’s consent is required.

Using Another Member’s Currency

Because of the importance attached by the IMF’s Articles to the distinction between a member’s currency and the currencies of other members, a clarification of how that distinction should be made could have been expected. It must have been thought by the drafters of the Articles that the distinction was self-evident, because they did not attempt to define a criterion. Given the context in 1944, the most probable assumption was that a member’s currency was understood as being issued by the member itself or by an entity (central bank) governed by the laws of the members. This definition would imply that monetary sovereignty was seen as a condition of membership, since a country cannot be admitted to membership if it is not able to perform its obligations under the Articles.

In his book on membership and nonmembership in the IMF, the then General Counsel of the IMF, Sir Joseph Gold, states that “[a]s late as February 1956 the staff [of the IMF] had been of the opinion that the existence of a domestic currency was a criterion for membership.”12 As he points out, however, “[although the norm for membership in the Fund was probably intended to be a country with a currency of its own that was subject to regulation by no other country, there were departures from that norm even among the countries listed in Schedule A.13 In addition, the Articles recognize that a member may have no central bank of its own, and this also was a characteristic of some of the countries that sent delegations to the Bretton Woods Conference.” In the same chapter of this book, Sir Joseph Gold lists a number of cases that fell outside what he called the “norm for membership.” For instance, Ghana was admitted to membership in 1957 before issuing its own currency. Botswana (1968), Lesotho (1968), and Swaziland (1969) became members while using the South African rand as their currencies. Other countries had common currencies (Syria and Lebanon; African countries within the African Financial Community (CFA) franc area; and the East African Currency Board).

From an international law perspective, these two types of cases are fundamentally different. A common currency is not a foreign currency. For example, the euro is the currency of all the members of the euro area. The issuance of a common currency may be seen as a joint exercise of monetary sovereignty.

From a practical standpoint, however, the relationship with the IMF raises similar issues: calculation and payment of separate quotas and maintenance of separate accounts either for each member of a monetary union or for the issuer as well as for the user of a currency. Thus, the IMF had to maintain separate accounts in South African rands for Botswana, Lesotho, South Africa, and Swaziland until the three foreign users of rand decided to issue their own currencies. Today, the IMF maintains separate accounts in euros for each member of the euro area. Similarly, the existence of a balance-of-payments problem and the level of access to IMF resources are assessed separately for each member country, regardless of whether it is the issuer or user of a currency or a member of a monetary union.

At this point in the analysis, it may be concluded that the IMF will regard a currency as the currency of a country either if that currency is issued by or on behalf of the country, which includes members of monetary unions, or if that currency is used by a country other than the issuer as legal tender and in its relations with the IMF.

The experience of the past 10 years shows an evolution in the use of foreign currencies toward two types of situations.

The first one is purely transitional. For example, a newly independent country is not yet able to issue its own currency and uses the currency of the country from which it has seceded or separated during this transitional phase. Thus, after the dissolution of the Soviet Union, the Russian ruble was used by the former members of the Union until they issued their own currencies. Sometimes, the newly independent country adopts a totally different currency. Thus, Bosnia and Herzegovina used the deutsche mark until August 11, 1997, when it adopted a national currency, which was pegged to the deutsche mark. In the IMF’s accounts, the ruble was temporarily used as the currency of the former Soviet Republics as their currencies. Similarly, the accounts of Bosnia and Herzegovina were temporarily denominated in deutsche marks as that country’s currency.

The second situation is the opposite of what has just been described. It consists of the replacement of a national currency by a foreign currency as legal tender and in the country’s relations with the IMF. For example, the U.S. dollar has been the currency of Ecuador since May 1, 2000.

The case of Argentina was different. The Argentine peso remained legal tender in parallel with, and pegged to, the U.S. dollar and was used by Argentina as its currency in its relations with the IMF. Therefore, the dollarization of the economy in Argentina was different from the IMF’s standpoint from the dollarization in Ecuador.

In Kosovo and Montenegro, the adoption of the deutsche mark and later the euro was decided by the UN administration (Kosovo) or the federated republic’s government (Montenegro), but neither is a member of the IMF—nor is Liechtenstein, which has adopted the Swiss franc.

In the context of Bosnia and Herzegovina’s admission to the IMF, an issue was raised concerning the adoption of the deutsche mark as legal tender. Was the consent of Germany required? The IMF staff’s position was that, in the case of a country that did not issue its own currency, it was for that country to determine what it would use as its currency in its relations with the IMF; without a currency, the new member would not be able to fulfill its obligations and exercise its rights under the IMF’s Articles. In any case, the IMF was not in a position to rule on disputes concerning the choice of a currency as legal tender, because such issues had to be resolved between the two governments. Eventually, given Bosnia and Herzegovina’s intention to use the deutsche mark only for a transitional period, the matter was settled.


Although the adoption of a foreign currency by an IMF member as legal tender and the designation of a foreign currency as the member’s currency for its relations with the IMF are two separate issues, they are somewhat related. The underlying question is whether the issuer of a currency may prohibit the use of its currency by another country.14 In many cases the question does not even arise either because the use of a country’s currency by another country is mostly beneficial to the issuer (seigniorage) or because the user’s economy is too small to have a significant negative impact on the issuer’s economy.

Nevertheless, the issue did arise in the case of Bosnia and Herzegovina and may arise again.

In support of the issuer’s right to object to the use of its currency by a foreign country, two lines of argument have been presented:

(1) The first argument is based on the duty of noninterference and the concept of tort under international law. By giving legal tender status to another country’s currency, a country would unlawfully undermine the issuer’s capacity to conduct its economic policies, thus interfering in the issuer’s internal affairs. For instance, an outflow of currency from the issuer’s territory to the foreign user’s territory would reduce the money supply in the issuer’s territory, which would require the issue of additional liquidity, with the risk of a sudden reflow, which may destabilize the issuer’s economy. Both inflows and outflows of the issuer’s currency could undermine the issuer’s monetary policies.

Regarding this argument, the following remarks may be made:

(a) The issuer of a currency may limit or even prohibit an outflow of its currency by imposing exchange controls. These controls would not restrict payments and transfers. They would require only that payments and transfers be made in foreign currencies, thus preserving the local money supply. Similarly, regulations could prevent local currency held abroad and illegally exported from being returned to be converted.

(b) The granting of legal tender status to a foreign currency does not by itself make a major difference when that currency is already widely used in payments abroad. A meaningful protection of the issuer’s economy against outflows would include a more aggressive attitude not only against the use of its currency as legal tender but also against the toleration of widespread use of that currency in foreign countries. This attitude would have to be weighed against the benefits of earning seigniorage and maintaining an open economy.

(c) Legal tender status applies to notes and coins, even though the main source of liquidity is not notes and coins but bank balances. The constitution of large bank balances in correspondent accounts of foreign banks with banks in the issuer’s territory is independent of the legal tender status of its currency in the foreign banks’ countries. Bank balances can be monitored and limited by the regulators of the local banks.

(d) Even if a foreign country’s policies have a harmful effect on a country’s economy, it does not follow that they constitute a violation of international law. Competition among countries is not in itself a breach of international law. The IMF’s Articles of Agreement impose a duty not to compete through exchange rate depreciation15 and to cooperate with respect to certain violations of another country’s exchange controls,16 which confirms the absence of a general obligation to protect the monetary system of either countries. F. A. Mann summarizes the state of international law on this matter as follows: “A State’s duty to protect the monetary systems of other States may arise from treaties or even from informal arrangements such as used to characterize the sterling area…. However, apart from treaties, it would at present not be possible to maintain that customary public international law imposes upon the State the general duty of affording protection to the monetary systems of the other members of the family of nations.”17

(2) The second argument is based on the concept of monetary sovereignty in international law. The issuance of a currency, as well as the determination of its value, legal tender status, and uses as a unit of account and medium of payment, are matters of sovereignty; they are governed by the lex monetae. Therefore, the argument goes, the issuer of a currency may prohibit the use of its currency as a legal tender by another country, whether or not this use is harmful to the issuer’s economy. The difference with the preceding argument, which was based on tort, is that there would be a breach of international law for ignoring the issuer’s objections even if no actual harm could be established.

Regarding this argument too, a few comments may be made:

(a) Asserting that a country’s monetary sovereignty means that its lex monetae is binding on other countries and that, accordingly, the use of its currency by a foreign country is a violation of the issuer’s monetary sovereignty is only a petitio principii, not a demonstration. It does not explain why one country’s lex monetae should trump another country’s territorial sovereignty.

(b) The recognition of a country’s ability to regulate certain matters under international law does not necessarily mean that foreign countries have to abide by the laws of that country. For instance, acknowledging that countries have jurisdiction over their nationals implies that each country may apply its laws to its nationals without violating international law, but it does not entail an obligation for other countries to apply these same laws to the nationals of that country residing in their territories or appearing in their courts. It is well known that rules of conflicts of laws vary from country to country (except where treaties apply), because there is no general obligation to apply foreign laws.18 Even where the applicability of a foreign law is recognized by a court of law, it may be set aside by considerations of public policy (ordre public) of the forum.19

(c) Monetary laws are public laws, which means that not only the criminal sanctions but also, in many countries, the civil sanctions of foreign monetary laws will not be recognized, as they embody the interests of a foreign sovereign. In some countries, a foreign monetary law will be recognized, but only on the basis of special circumstances, that is, as lex contractus or lex loci solutionis or loi de police. It may be noted that the Rome Convention on the law governing contractual obligations does not require the application of foreign lois de police (such as exchange controls) and allows the parties to a contract to select different laws for different parts of their contract. Even when a treaty provides for mandatory recognition of certain foreign public laws, its application is sometimes restricted through judicial interpretation. A typical example is the restrictive interpretation in the United Kingdom, the United States, and, to a certain extent, Germany of Article VIII, Section 2(b), on the recognition of foreign exchange controls.

(d) History provides examples of foreign currencies being used as means of payment. In the United Kingdom and in the United States, the courts have explicitly recognized the unfettered right of the sovereign (King or Congress) to grant legal tender status to specified foreign currencies, without requiring the consent of the issuer.20

(e) The recognition as a principle of international law of a country’s right to regulate the use of its currency abroad would give extraterritorial recognition to the lex monetae, which would have two consequences. Not only could the lex monetae prohibit the use of the issuer’s currency abroad, but it could also make it mandatory. The first consequence would go even beyond the requirements of Article VIII, Section 2(b), and would make this provision superfluous, although, as mentioned above, its interpretation has been rather restrictive because it was seen as an exception to the nonapplication of foreign monetary laws rather than the recognition of their universal application. The second consequence has not even been considered in any actual case and is contrary to well-established rules of international law. In the exercise of its sovereignty a country may prohibit the use of a foreign currency within its territory. Similarly, through the imposition of exchange controls, countries may restrict the use of foreign currencies in international payments and transfers from and to their territories, subject only to their treaty commitments.

In conclusion, and subject to future consideration that may shed more light on these matters, there is at present no basis in international law to support either the first or second argument discussed above.


IMF, Articles of Agreement of the International Monetary Fund, Article IV, Section 9 [hereinafter the “original Articles”], and IMF, Articles of Agreement of the International Monetary Fund, Article IV, Section 5, [hereinafter the “present Articles”].

Article III, Section 3, of the original Articles; Article II, Section 2, second sentence, of the present Articles. The reserve assets portion under the original Articles could not exceed either 25 percent in gold or 10 percent of the country’s net holdings of gold and U.S. dollars; in the latter case, the currency portion was increased to make up the difference.

Article III, Section 4(a), of the original Articles; Article III, Section 3(a), of the present Articles.

Article IV, Section 3, of the original Articles.

Article IV, Section 2(b), of the present Articles. All future references are to the present Articles.

Article IV, Section 2(a).

Article V, Section 2(a).

Article V, Sections 7(c) and (d).

Article V, Section 11.

Article XIII, Section 2(a).

Article III, Section 4. The promissory note “or similar obligation” is nonnegotiable, non-interest-bearing, and payable at its face value on demand by crediting the account of the IMF in the member’s depository, id.

Joseph Gold, Membership and Nonmembership in the International Monetary Fund—A Study in International Law and Organization (Washington: IMF, 1974), at 92.

That is, the countries represented at the Bretton Woods Conference (note added).

It is undisputed that the counterfeiting by a country of another country’s currency would be a violation of international law (although there are numerous examples in times of war), but using is not counterfeiting.

Article IV, Section 1(iii).

Article VIII, Section 2(b), first sentence: “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.”

F.A. Mann, The Legal Aspect of Money: With Special Reference to Comparative Private and Public International Law, 5th ed. (Oxford: Clarendon Press, 1992), at 479.

On the nonapplicability to a contract—governed by French law but denominated in a foreign currency—of the lex monetae prohibiting gold clauses, see French Cour de cassation in the Messageries Maritimes case, Oct. 29, 1964, Clunet (1965), at 637, obs. Goldman.

In the same case, an earlier decision of the Cour de cassation had held that, even if the lex monetae was also the lex contractus, the prohibition of gold clauses in international contracts could not be recognized by French courts by reason of ordre public (June 21, 1950, Clunet, at 1196).

See, in the United Kingdom, Wade’s Case, 77 Eng. Rep. 232 (K.B. 1601), and in the United States, Tyson v. United States (285 F.2d 19 (1960).

    Other Resources Citing This Publication