Journal Issue

What Does It Really Mean?—Multiple Currency Practices

International Monetary Fund. External Relations Dept.
Published Date:
June 1966
  • ShareShare
Show Summary Details

W. John R. Woodley

THE ARTICLES OF AGREEMENT of the Fund prescribe a system in which the international value of a currency (the “par value”) is fixed in terms of gold, and no transaction in foreign exchange may differ from this rate by more than 1 per cent. But the system is not universal; many members of the Fund have separate exchange rates for different groups of exchange transactions—that is to say, they have resorted to “multiple currency practices.” These practices may be quite simple; for example, in the Philippines until recently there was one exchange rate for major exports and a single free market rate for all other transactions. Or they may be very complex, as in Colombia, where the exchange rates range from about 9 pesos = US$1 to about 18 pesos = US$1. Simple or complex, multiple currency practices are widespread. Their traditional locus is Latin America, but they frequently occur in Asia and the Middle East and occasionally elsewhere as well.

But the Fund is opposed to them; their abolition and replacement by a unified exchange system is now, as it always has been, one of the Fund’s objectives. The bias against multiple rates is based upon the lessons of history, drawn from exchange experience in the 1930’s. During the depression years, multiple rates were used, particularly in Germany and in Latin America, as an export promotion device, and this led to retaliation by other countries and to a general environment of exchange instability and uncoordinated exchange rate policies. (For a description of these conditions, see Frank A. Southard, Jr., “International Financial Policy, 1920-44,” Finance and Development, Vol. II, No. 3, pp 135-43.) The Fund issued a general statement in 1947 to all its members, expressing its attitude toward multiple rates. Since that time it has made a number of general decisions, and in these, as well as in its Annual Reports and other public statements, it has urged countries to simplify complex exchange rate systems and to make a par value system effective.

Yet in spite of its general attitude, and its efforts to persuade countries that single exchange rates are in their interest, the Fund has approved multiple rate practices for many of its members. Frequently this approval has been on a temporary basis, and the Fund has remained in consultation with the member regarding the practice. Yet the paradox remains, and calls for explanation.


The following table sets out an imaginary, but fairly typical, multiple exchange rate system. In it, the “buying rates” are the prices that the monetary authorities in the country pay for foreign exchange earned by exporters and others; the “selling rates” are the rates at which the authorities sell foreign exchange, e.g., to importers.

In order to understand such a system of exchange rates, we need a reference point or norm. Such a reference point, here assumed to be P 7.00 = US$1, is the equilibrium exchange rate—a rate which would permit exchange receipts to finance exchange payments without excessive use of foreign exchange reserves, without undue reliance on restrictions on trade and payments, and without internal policies inconsistent with the relatively low degree of unemployment.

If a country moves away from such a rate, the effects would be similar to those of taxes or subsidies, as explained below. Governments do not necessarily object to this; on the contrary, taxing or subsidizing through the exchange system often appeals to them, particularly to the governments of less developed countries which find it difficult to implement effectively other tax and subsidy devices. Multiple rates have the advantage of being enforceable through the banking system where the officials who administer them are experienced in similar work. Moreover, the rates can usually be introduced and changed by the central bank, whereas a change in import or export taxes usually needs the approval of the legislature, which is not likely to be obtained quickly or easily.

Exchange Rates(Pesos per U.S. dollar)
Buying RatesSelling Rates
Transactions to whichTransactions to which
Ratesrates applyRatesrates apply
5.00Coffee exports6.00Essential imports
7.00Other major exports8.00Semiessential imports
10.00(fluctuating market rate)10.00(fluctuating market rate)
Minor exports, invisibles (including capital)Luxury imports, invisibles (including capital)

The general way in which a multiple exchange rate system can be used to raise substantial amounts of tax revenue for the government will be apparent from the table. If the average price in local currency that the monetary authority (e.g., the central bank) pays for foreign exchange is less than the price at which it sells foreign exchange, it will make a profit in local currency which can be diverted to the government. For example, if the central bank pays P 5.00 for each dollar it buys from coffee exporters, and sells all these dollars to importers for P 8.00 or P 10.00 each, the profit will be handsome. There are, moreover, other advantages for a government in such a system of rates.


The case for taxes or subsidies varies with the commodity involved and the particular situation of the economy. In the table, the exchange rate of P 5.00 = US$1 for coffee exports involves a tax in two senses. Coffee exporters receive from the central bank P 2.00 less for each U.S. dollar they earn than they would receive under the equilibrium rate, i.e., the rate which notionally they would receive if there were no multiple exchange rate system in operation. More specifically, as the table shows, they receive fewer pesos for each dollar than do exporters of other major commodities. Governments may favor such a tax on several grounds. By giving exporters a less favorable rate of return, the tax may discourage the production of coffee. It may well pay the country to produce less coffee for higher prices. The tax may also be wanted because coffee growers are a relatively high-income group that may reasonably be called on to pay taxes. Another justification may be that the tax is easily collectible and productive of large amounts of revenue for economic development. Considerations like these underlie the penalty export rates for such products as coffee in Brazil and Colombia and wool and wool tops in Uruguay.

On the other hand, since commodities that sell for US$ 1 abroad would on average sell for P 7.00 in the imaginary country to which the table applies, exporters who receive P 10.00 for each dollar earned are being subsidized. Subsidies for certain exports—for example, the rate of P 10.00 = US$1 in the table—are advocated as a means of encouraging new industries, particularly those which process raw materials or develop nontraditional exports. One example of the encouragement of processing industries would be the use of the rate of P 10 = US$1 for exports of powdered coffee. By exporting powdered coffee, a businessman would earn (for each dollar received) twice the amount of local currency that he would earn by exporting unprocessed coffee. Obviously, he would have a powerful incentive for processing coffee beans into powdered coffee.

Countries that are themselves too dependent on exports of a few primary products which are subject to wide price fluctuations in international markets are eager to develop new export products; in those countries, special exchange rates (such as the P 10.00 rate in the table) may induce the development of new export products. They may even encourage industries that are currently producing for only the domestic market to expand their output and sell abroad; thus they may achieve the economies of large-scale production. Such considerations have influenced the granting of subsidy rates in the past to manufacturers in countries such as Colombia, Korea, Venezuela, and Yugoslavia, and they are important in Pakistan at the present time.


Similar considerations apply to the granting of special exchange rate treatment for imports. The import subsidy rate of P 6.00 = US$1 in the table might be used for particularly desirable imports—for example, raw materials or capital equipment needed to foster the expansion of local industry, or essential items in the diet of the lower income groups. Also, government purchases from abroad might be subsidized in this way in order to reduce the impact of foreign exchange expenditures on the budget. These factors were of particular importance in the Philippines prior to the exchange reform of 1959.

A tax rate for imports, such as the rate of P 10.00 = US$1 shown in the table, might be used to reduce imports of automobiles, Scotch whisky, and other luxury items regarded as undesirable in a poor country, or to increase the cost of imports competing with domestic production and thus to provide protection for domestic output. Argentina, Indonesia, Uruguay, and Yugoslavia have frequently used multiple rates for these purposes.

Capital Controls and Revenue

A number of countries, particularly in Latin America, have introduced into their multiple rate systems a special market for international movements of capital and other “invisibles” (international transactions which do not show up in customs returns; for example, personal services). Such a market now exists in Chile, Colombia, and Ecuador. In most countries having this special market, the exchange rate has been left free to fluctuate; as a result, the value of the local currency (in terms of foreign exchange) for these transactions has been less than its value for trade transactions. The market is justified as providing a “safety valve,” since both inflows and outflows of capital, which are usually more volatile than trade transactions, can occur without disturbing trading relations. If capital flight develops, the increased demand for foreign exchange in the capital market makes it more expensive to acquire exchange and so acts as a deterrent to sending funds out of the country.


Multiple exchange rates, then, have several functions that make them appear as an admirable weapon of government financial policy. However, the Fund’s long observation of the experiences of its members with such rates suggests the opposite conclusion. Multiple rate systems have, in practice, seldom operated effectively either to meet balance of payments problems or to provide the tax and subsidy effects desired by governments. Such systems have proved basically unstable, they have been difficult to administer, and they have frequently disguised problems rather than solved them.

Exchange Rate Instability

Most of the countries using complicated multiple rate systems have been subject to inflationary pressures and to a more or less rapid rise in domestic costs and prices. Such an environment greatly complicates any exchange policy, because the exchange rate has to be adjusted frequently to keep an appropriate relation between domestic and external costs and prices. The difficulties are particularly acute if multiple rates have been adopted for the purpose of providing taxation and subsidy. The underlying monetary instability continually disturbs the exchange system and destroys the function that the exchange rates are supposed to be serving. For example, many countries imposing tax rates on their principal exports, such as coffee, have found themselves paying out to producers, through price support or storage schemes, more than the amount of tax collected through the exchange system. Multiple rate systems designed to produce revenue for the government have sometimes in fact caused large government losses. In our imaginary example, this happens if the number of pesos that the monetary authority earns by selling dollars to importers is less than the number of pesos that the authority has paid to exporters to acquire the dollars. Such a situation occurs if the government, in response to inflationary pressures, depreciates the exchange rate for exports in order to keep them moving (e.g., alters it from P 5.00 = US$1 to P 8.00 = US$1), but does not adjust the exchange rate for imports. There may be a good reason for not adjusting the latter rate. Higher prices for imports could raise the cost of living; for example, to shift the exchange rate for imports from P 6.00 = US$1 to P 7.50 = US$1 would raise the wholesale price of imports by 25 per cent. One of the great drawbacks of multiple rates is that, once they are introduced, decisions about export and import rates can be taken separately. Virtually no one favors depreciation of the import rate (i.e., increasing the number of pesos that importers must pay for a dollar), even though its general effects might be highly advantageous for the long-term development of the economy. On the other hand, exporters are keenly interested in depreciating the export rate. Although problems like these are not inherent in multiple rate practices, these practices do frequently coincide with domestic inflation, and the danger of such difficulties is therefore considerable.

A further disadvantage of multiple exchange rates is that there is a high correlation between them and exchange instability. This is partly because the multiple rates are a response to payments difficulties and therefore are likely to require frequent adjustment, and partly because, if several rates of exchange exist, adjustment of one or a few of the rates is less difficult—because it has less widespread effects—than is a change in a unified exchange rate. (Adjustment of a unified rate always involves the government in a careful balancing of conflicting interests, since exporters and importers are then interested in the same rate. Exporters favor a more depreciated rate, to increase the local currency return from exports, while importers and local producers oppose depreciation because higher local currency costs will discourage sales and, at least in the short run, investment in capital equipment.) Now, instability in the exchange rate structure has inherent disadvantages. If exporters believe there is a likelihood that the export rate will be depreciated (changed in their favor), they will be tempted to withhold exports in the hope of securing better treatment. If their product is a major export and traders in this product act in concert, their speculative activities will do so much injury to the balance of payments that it may force the government to allow the relevant exchange rate to depreciate, and thus reward the speculators. Similarly, unstable exchange rates for importers make local producers of competitive goods uncertain of the degree of protection they can count on if they should invest in expanding their plants or building new ones, and thus foster the development of the economy.

Administrative Problems and Policy Dilemmas

The administration of multiple rate practices is far from easy. First, to enforce multiple rates, exchange controls must be applied to segregate various types of transactions. In the illustration that we have been using, for example, exchange controls would have to be used to force coffee exporters to sell exchange at the P 5.00 = US$1 rate. The pressure to evade such controls would obviously be strong. And in addition to evasion, there will be strong political pressures to obtain more favorable exchange rate treatment. If some exporters are receiving P 7.00 for each U.S. dollar that they sell while others are receiving P 10.00, the former group will naturally make every effort to obtain equally favorable treatment; they may easily spend more time on this than on taking steps to improve efficiency of production or to sell more. Many countries have found that the more they differentiate their exchange rates, the greater is the pressure for further proliferation of the rates. This has sometimes resulted in an exchange rate manual as big as a large reference volume.

The use of multiple rates for imports also involves the government in a number of policy dilemmas. One of the apparently more valid arguments for very depreciated rates for luxury imports is that they discourage the import of these articles. These same exchange rates, however, provide abnormal protection for domestic industry and thus stimulate scarce local capital to invest in the production of the luxury goods that would otherwise be imported. A contrary disadvantage arises if favorable exchange rates are provided for imports of essentials. While this ensures their availability cheaply as imports, it means that there is little incentive to produce such goods domestically. The harmful effects of such low rates are accentuated if they apply (as they frequently do) to imports of foodstuffs; the consequential cheapening of food represses incomes and development in the farm community where incomes are normally the lowest, and fosters migration from the farms to the towns.

Some International Implications

The use of multiple rates by a country also has implications for the international community. To use one of the examples given above, a subsidy rate for exports of powdered coffee may achieve its purpose and result in a new industry producing powdered coffee. Such an industry may, however, injure the powdered coffee industry established in another country and lead the government of that country to protest, or even to take countermeasures. In the early 1950’s, there was a lengthy international dispute over the use of different export rates for wool and wool tops in Uruguay, a number of governments arguing that the Uruguayan exchange system was destroying the wool top industries established in their countries. Changes in one exchange rate in a system of rates—for example, a more favorable rate provided for exports of cocoa—may lead other countries to protest that the country introducing this rate is seeking to obtain competitive advantages in the world market. The introduction of international commodity agreements for some commodities has meant that export rates have less effect on their sale; but the possibilities of competitive devaluation of exchange rates are still real and are of particular importance when a number of countries are trying to develop new exports. Another type of international difficulty may arise if a country grants a preferential exchange rate to imports from a second country in order to obtain preferred access to the markets of that country. In these circumstances, a third country that has export products competitive with those of the first may protest against the discrimination created by this special exchange rate because it effectively limits its own export marketing possibilities.

This outline of multiple currency practices indicates some of the many complex questions raised by a country resorting to more than one exchange rate. There have been occasions when the Fund has not entered objections to a proposal for a simple multiple rate scheme (i.e., a scheme with only two or three different rates) because the country was unable effectively to introduce a valid single rate. In such circumstances the Fund has been prepared to treat multiple exchange rates as temporary responses to presently insoluble problems, and has been willing to provide financial support to countries with multiple rates, including even fluctuating multiple rates. The Fund has found, however, that in most countries multiple rates sooner or later create as many problems as they solve and that such apparent solutions are seldom lasting ones. The aim of establishing a single rate that provides a uniform price incentive for export and import transactions is still a vital one for both the country concerned and the international community.

Other Resources Citing This Publication