Chapter 2: Developments in the International Monetary System

International Monetary Fund
Published Date:
September 1984
  • ShareShare
Show Summary Details

This chapter deals with salient aspects of the present international monetary system and its recent evolution. The first part of the chapter reviews exchange rate arrangements and policies of industrial and developing member countries; it concludes with a discussion of developments in the Fund’s surveillance over members’ exchange rate policies. The second part of the chapter takes up the question of international liquidity, beginning with an analysis of recent changes in official international reserve holdings and proceeding to a description of developments in private international credit markets; it ends with a discussion of the adequacy of international reserves and the role of the Fund in providing liquidity to its members.

Exchange Rates and Surveillance

Throughout 1983 and the first half of 1984, the pattern of exchange rates, especially the weakness of many of the major currencies against the U.S. dollar, remained a source of concern. While the reasons for this pattern are not fully understood, it seems that divergences in economic and financial conditions—in particular, the strength of the economic recovery in the United States compared with that of other countries and the renewed rise in U.S. interest rates—were important contributing factors. Within the European Monetary System (EMS), the persistence of relatively high inflation rates in a number of countries led to a new period of tension, followed by a realignment of central rates in the first quarter of 1983. Since then, however, the adjustment efforts of the countries with relatively high inflation have led to an abatement of the tension.

Most of the non-oil developing countries experienced a marked reduction in their current account deficits in 1983 as a result of policy measures taken in these countries and, to a lesser extent, of the economic recovery in industrial countries. In many instances, these policy measures included a significant devaluation of the exchange rate supported by a comprehensive program to reduce the fiscal deficit and the rate of credit expansion, improve the allocation of public investment, and, more broadly, restore an adequate system of interest rate and price incentives. In a number of countries, however, progress toward reducing the external deficit was brought about by programs relying mainly on import restrictions and domestic demand cuts, especially cuts in domestic investment, without adequate incentives for a switching of economic resources into the export sector.

The developments during 1983 and the first part of 1984 have demonstrated once more that any move toward a stable system of exchange rates requires domestic policies that foster the achievement of stable domestic economic and financial conditions, especially in the major industrial countries. A policy stance consistent with low inflation rates is important in this context, but it is not enough. It is also important that monetary and fiscal policies be mutually supportive to avoid undue pressures on interest rates. In its surveillance activities, the Fund has continued to stress the need for such a balanced configuration of policies. In the many developing countries with major external and internal adjustment problems, the Fund has stressed the need for comprehensive longer-run adjustment policies aimed at an expansion of exports and a restoration of the basis for sustained growth, in addition to the short-run policy measures of demand restraint that are often unavoidable.

Exchange Rate Issues in Industrial Countries

Concerns about the levels of exchange rates among the major currencies intensified during 1983 and the first few months of 1984, in particular because of the continued strength of the U.S. dollar (Chart 12). Except for the Japanese yen, which appreciated sharply against the U.S. dollar toward the end of 1982 and then maintained a relatively stable dollar value during 1983, the pattern of an appreciating dollar generally persisted throughout 1983. The dollar then depreciated somewhat in nominal effective terms during the first quarter of 1984, but rose again in the second quarter, approaching the level reached in January.1 These exchange rate movements raised the real value of the dollar to about 15 percent above its average value for the decade 1973-82 in relation to the Japanese yen and the pound sterling, some 40 percent in relation to the deutsche mark, and nearly 50 percent in relation to the French franc.

Chart 12.Selected Major Industrial Countries: Nominal and Real Bilateral Exchange Rates, 1980-First Quarter 1984

(Indices, 1973-82 = 100)

1 Nominal exchange rate multiplied by the ratio of normalized unit labor costs in manufacturing for the two countries. Normalized unit labor costs are calculated by dividing an index of hourly labor costs by an index of potential output per man-hour.

A country’s average real exchange rate over the past decade is, of course, not necessarily sustainable and appropriate under present circumstances. Changes in underlying economic and financial conditions often call for changes in real exchange rates. For example, the very sharp real appreciation of the pound sterling from 1978 through the first quarter of 1981 can be explained in part by the development of North Sea oil at a time of rapidly rising oil prices, while the subsequent reversal was aided by the softening oil market in 1982 and 1983. Also, shifts in private saving and investment patterns or in market confidence may induce changes in real exchange rates in order to generate a balance between the current and capital accounts. The demand for U.S. dollars, in particular, has occasionally been stimulated by the traditional role of the United States as a haven for financial investments during periods of turmoil in other countries. During the past year and a half, this demand may also have been stimulated by the positive effects of the strong economic recovery and the 1981 tax package on after-tax corporate profits and the expected rate of return on investment.

In a narrow sense, one can even say that, in the absence of intervention by the authorities in foreign exchange markets, exchange rates are always at equilibrium levels, since they are simply a reflection of the preferences and expectations of market participants engaged in free and open trading based on information available to them. But that observation does not imply that serious misalignments cannot occur in terms of the relative prices at which international trade takes place. Whenever stable domestic economic and financial conditions are absent, developments in financial markets can lead to swings in exchange rates that, while reflecting the free play of forces in the foreign exchange market, may not be consistent with the proper functioning of the adjustment process in the goods markets. This does not mean that the movements in exchange rates per se are unjustified but that the international interest would be well served if both the unstable domestic conditions and the accompanying movements in exchange rates were eliminated. For example, a lack of balance between monetary and fiscal policies may lead to a rise in real interest rates, and an appreciation of the exchange rate to the point of jeopardizing a country’s international competitive position. Such a development would be detrimental not only to that country but also to its trading partners because it would have to be reversed sooner or later, with all countries obliged to bear the cost of moving factors of production first in and then out of certain sectors. Moreover, in the interim it may result in the introduction of protectionist measures that may be difficult to remove later.

The main reason for the current concern about exchange rates is that they appear to have exerted strong pressure on trading patterns and current account positions of the major industrial countries. For example, the rise in the exchange value of the U.S. dollar during the past several years has contributed to a sharp weakening in the U.S. current account balance (including official transfers) from a surplus of $5 billion in 1981 to a deficit of $39 billion in 1983, which, if continued, could have disquieting implications for the international allocation of world private saving. On the other hand, the U.S. current account deficit has had beneficial effects on economic recovery in other countries.

Economic and Financial Convergence

It is becoming increasingly apparent that the smooth functioning of the exchange rate system requires stable underlying conditions in the major industrial countries, including low rates of inflation, sustainable growth of real incomes, and a stable pattern of financial conditions. In the absence of such conditions, exchange markets tend to generate large swings in, and unsustainable levels of, exchange rates as market participants shift funds among currencies in response to actual or expected changes in relative economic and financial conditions.

Some important improvements in economic performance occurred during 1983 and the first part of 1984. In particular, there was a continued trend toward improved cost and price performance, with most of the major industrial countries recording reduced inflation in 1983, compared with the preceding three years. The average rise in the gross national product (GNP) deflators of the seven largest industrial countries, which had surpassed 9 percent in 1980, fell to 4½ percent from the fourth quarter of 1982 to the fourth quarter of 1983 (Chart 13). Furthermore, the spread between the highest and lowest inflation rates was reduced somewhat, especially in comparison with 1980 and 1981.

Chart 13.Seven Major Industrial Countries: Indicators of Economic Performance1

1 The dashed horizontal line in each panel represents the mean value of the plotted data, weighted by the relative size of each country’s gross national product for the 1980-82 period.

2 Yields on government bonds with maturities ranging from 7 to 20 years deflated by a weighted average of the rate of inflation in the current quarter and the next two quarters, with the deflator of private final domestic demand serving as the price index. Staff projections of this deflator are used for the most recent data.

3 Averages of daily rates on money market instruments of about 90 days’ maturity, except for Japan, where the discount rate on two-month (private) bills is used. These rates are deflated by the same index used for long-term interest rates.

4 Money stocks used are M2 for the United States, Canada, France, and Italy; M2 plus certificates of deposit for Japan; sterling M3 for the United Kingdom; and central bank money for the Federal Republic of Germany.

A similar improvement was recorded in the achievement of real economic growth. The previous three years (1980-82) had been characterized by cyclical swings in economic activity that were sizable and not synchronous; five of the seven major countries—Japan and France being the exceptions—experienced a decline in real GNP in at least one of those years. In 1982, the rate of change in real GNP in the major countries ranged from a 4 percent decline in Canada to a rise of over 3 percent in Japan, with an average decline of ½ of 1 percent. In contrast, all of these countries showed positive growth on average during 1983, although the rates of growth still ranged from less than 1 percent to almost 7 percent.

As regards real interest rates, the evidence suggests an increased dispersion of real long-term rates around a higher average level in 1983 than in the years 1980-82, although real short-term rates moved down somewhat in most countries (Chart 13). Real interest rates are difficult to evaluate because they involve estimating market expectations of inflation. These difficulties are particularly pronounced at times when inflation is changing rapidly, as it was during 1982 and 1983. Nonetheless, most plausible estimates, including the estimates shown in the chart that reflect interest rates adjusted by a three-quarter moving average of actual inflation rates, indicate that in 1983 the high real long-term interest rates in the United States and Canada were in marked contrast to the rates in the other large countries. In the United States, the estimated real long-term rate of 8 percent was about 6 percentage points higher than the average that prevailed during the 30-year period 1950-79. Part of this difference, however, may reflect a greater persistence of expected inflation in the United States and Canada than is assumed in these estimates.

The balance between monetary and fiscal policies also continued to differ substantially among major industrial countries during 1983. Monetary policy has been even more difficult than usual to assess during this period, owing to shifts in the demand for money at a time of financial deregulation in the United States and a number of other industrial countries, and pressures associated with the EMS realignment in March 1983. On the whole, however, monetary growth seems to have been at least moderately constrained in most countries (Chart 13). Less success has been achieved in controlling fiscal positions. In particular, fiscal deficits in Canada and the United States were significantly larger in relation to GNP in 1983 than the average for the three preceding years. By contrast, in Japan and the Federal Republic of Germany, where the recovery was less advanced, the deficit in 1983 was equal to, or smaller than, the average deficit in the three preceding years. The following analysis discusses the effects of fiscal policies on exchange rates, as well as the broader issue of the role of convergence of economic performance to sustainable levels among countries with floating exchange rates and among countries having pegged rates or participating in cooperative exchange arrangements.

Countries with Floating Exchange Rates

The role of convergence of economic performance in the smooth functioning of the exchange rate system may at one time have appeared to be less crucial for countries with floating exchange rates. The pressures that arose from divergent policies and conditions in the late 1960s and early 1970s furnished a primary impetus in the evolution of the current system. It was frequently argued at the time that floating exchange rates would enable countries to pursue domestic monetary objectives with a greater degree of independence than was possible under fixed exchange rates. In the event, divergent conditions—especially with respect to inflation—have proved to be a serious problem for countries with floating rates as well. Indeed, the experience of the instabilities associated with high inflation made it clear that a generalized reduction of inflation was a prerequisite for creating the climate of confidence that was needed to underpin a smoothly functioning exchange rate system.

Part of the problem associated with the high rates of inflation experienced by many industrial countries during the 1970s and early 1980s was their unpredictability, which added to the risks of investing in financial assets denominated in those countries’ currencies. In addition, the failure of some countries to control inflation led to a decline of confidence in the sustainability of their policies. Market participants came to expect inflationary policies to be reversed sooner or later, but the magnitude and timing of those reversals were subject to substantial uncertainty. The riskiness associated with financial investment during periods of high inflation reduced the confidence with which expectations were formed, weakening the tendency toward stabilizing speculation in foreign exchange markets.

A further problem was that inflation detracted from the smooth adjustment of floating exchange rates, even among countries with similar inflation rates, because the relative prices of individual goods, services, and assets tended to be less stable and to have more divergent patterns than in periods of greater overall price stability. In turn, this instability led to increased uncertainty about movements in relative prices. To the extent that these uncertainties pertained to prices of internationally traded goods, they contributed directly to increased volatility and misalignment of exchange rates.

A related aspect of convergence that proved to be important is the maintenance of a steady stance of demand management policies. The cyclical pattern of economic growth during the 1970s led to frequent changes in these policies, which were at times counterproductive and added to uncertainty with regard to the future course of the world economy. Even more important, undue delays in the adoption of firm measures to put an end to the rise in inflation made it ultimately necessary to have a marked shift in policies in 1979-80, with a very painful adjustment period during 1980-83.

The problems that are raised for the smooth functioning of the exchange rate system by large policy-induced cyclical swings in economic activity occur whether or not those cycles are synchronous among countries. If the cycles are not synchronous, as was the case in the recessionary period of 1981-82, then the countries undergoing relatively rapid expansion of domestic demand will tend to have relatively weak external payments positions, and conversely. On the other hand, as the experience in 1972 and 1973 demonstrated, synchronous cyclical movements among major countries can create large swings in prices of primary commodities. Such swings can seriously disrupt normal trading relationships and can add to uncertainties about relative prices, thus contributing to exchange rate instability. The implementation of steadier demand management policies would contribute to economic and exchange rate stability by fostering an environment in which market participants could have greater confidence in the persistence of current policies and the stability of underlying conditions.

Changes in circumstances do at times require short-run adjustments in economic policies. For example, large shifts in the demand for money resulting from institutional or regulatory innovations or from shifts in preferences for holding various assets may occasionally need to be accommodated by the authorities through temporarily higher or lower monetary growth rates. This is particularly true from the vantage point of the international financial system when changes in the demand for money involve shifts among assets denominated in different currencies. Monetary authorities that have gained considerable credibility through past actions will normally be able to make the shortrun adjustment without significant effects on inflationary expectations. However, monetary authorities with a less well established credibility may have to exercise greater caution, especially if the circumstances are such that the short-run adjustment could easily be misinterpreted by economic agents.

Much progress was achieved in 1983 and the early months of 1984 in the abatement of inflation and the restoration of real growth. What became apparent, however, was that this progress was not sufficient for a return to a sustainable pattern of exchange rates. A further requirement was that progress should also be achieved in the avoidance of abnormally high or low real interest rates. This objective is intrinsically important, inasmuch as abnormally high real interest rates hinder private capital investment and create major hardships for developing countries with substantial external indebtedness; on the other hand, negative real interest rates give rise to inflation and encourage excessive indebtedness. In addition, this objective is important because abnormally high real interest rates in an industrial country with broad financial markets can have a major effect on its exchange rate.

The emergence of very high real interest rates in the United States after 1980 induced increases in other countries that were generally somewhat smaller, resulting in sizable real interest rate differentials favoring the U.S. dollar over other major currencies (Chart 14). These differentials apparently gave rise to a substantial flow of capital into dollar-denominated assets and consequently may have contributed to the appreciation of the U.S. dollar in real terms. In addition, interest rates were quite volatile during 1980 and 1981 in relation to historical patterns, especially in the United States but in a number of other countries as well, contributing to the volatility of exchange rates. During 1982, the real interest rate differentials in favor of U.S. dollar investments over those in other major currencies tended to subside, as did the volatility of nominal and real interest rates. Since the beginning of 1983, however, these differentials have widened again as a result of continued progress against inflation in the United States, as well as the rise in nominal U.S. interest rates. To a major degree, this renewed rise in U.S. interest rates reflects the strength of the recovery in the United States, which has had beneficial effects on the prospects for economic growth in other countries; but it is likely that the large U.S. fiscal deficit was also a contributing factor. Since convergence of real interest rates could promote a more stable pattern of exchange rate developments, it would be desirable for fiscal and monetary policies to foster such convergence by taking adequate account of differences among countries in the flow of saving relative to the demand for private investment.

Chart 14.Seven Major Industrial Countries: Monthly Average Real Long-Term Interest Rates, January 1980-May 19841

(In percent per annum)

1 Yields on government bonds with maturities ranging from 7 to 20 years deflated by a weighted average of the rate of inflation in the current quarter and the next two quarters, with the deflator of private final domestic demand serving as the price index. Staff projections of this deflator are used for the most recent data

Countries with Managed Exchange Rates

The problems associated with a lack of convergence of economic performance are also apparent in countries with pegged exchange rates and among those that participate in cooperative exchange arrangements. For these countries, divergences of inflation rates have necessitated periodic adjustments in central rates and have sometimes subjected the currencies involved to bouts of destabilizing speculation. In addition, because interest rates have tended to be relatively high in countries with higher inflation rates, investors in some circumstances have been able to take advantage of the relative exchange rate certainty that follows a change of parities by placing funds temporarily in those higher-yielding assets. A divergence of nominal interest rates has then placed upward pressure on the high-inflation country’s currency for a time, followed by downward pressure once the competitive effects of the country’s high inflation rate have begun to be felt.

The recent functioning of the EMS illustrates these difficulties and the ways that countries participating in a joint floating arrangement have been able to cope with them. There has been a substantial divergence in inflation rates among the EMS member countries since the inception of the current system in March 1979. This pattern of divergence, along with other factors affecting competitive relationships, has resulted in several periods of tension and has necessitated a total of seven realignments over the past five years. The cumulative nominal depreciation of the Italian lira visa-vis the deutsche mark during this period has amounted to some 27 percent, roughly matching the cumulative inflation differential; similar, though somewhat smaller, net movements have taken place for the French franc and the Irish pound.

Adjustment of the EMS parities has sometimes occurred in an atmosphere of pressure associated with anticipation of an impending realignment and with official intervention in support of existing parities. To reduce these pressures and to promote convergence of economic performance, most EMS realignments have been accompanied by the introduction of restrictive budgetary or monetary policies in the high-inflation countries. In particular, the most recent realignment, in March 1983, was accompanied by the implementation of a package of restrictive measures in France involving both budgetary and monetary policies as well as foreign exchange controls. Subsequently, monetary growth was successfully reduced in both France and Italy, and the real short-term interest rate differential between France and the Federal Republic of Germany switched from approximately 1 percentage point in favor of assets denominated in deutsche mark just before the realignment to about the same margin in favor of French francs a year later. In addition, France, Ireland, and Italy have all made further efforts during the past year to moderate inflationary pressures by controlling government expenditure and reducing fiscal deficits.

The recent experiences of the other industrial countries that have managed their exchange rates in terms of a composite of currencies—comprising Austria, Australia, Finland, Norway, and Sweden—also help to illustrate the importance of convergence of economic performance for the smooth management of exchange rates. The Austrian economy, for example, is closely integrated with that of the Federal Republic of Germany, and there has been in general a very high degree of convergence between the two countries. Consequently, for several years the Austrian authorities have been able to maintain a rather stable relationship between the schilling and the deutsche mark in both nominal and real terms.

The Australian dollar was, until recently, managed in relation to a basket of currencies, with an exchange rate posted daily in terms of the U.S. dollar (the intervention currency for the authorities).2 During the late 1970s and early 1980s, Australia’s cost and price performance was close to that of its major trading partners. This situation, together with a flexible approach to exchange rate management, permitted a reasonable degree of stability in the real effective exchange rate for the Australian dollar, relative to that observed in the preceding period. However, in 1982-83, prices in Australia rose at a substantially more rapid rate than in its major trading partners, making it more difficult to maintain international competitiveness. Moreover, in the first part of 1983, exchange market pressures were exacerbated by heavy speculative capital flows around the time of the national elections. In these circumstances, the nominal exchange rate was allowed to depreciate quite sharply from the fourth quarter of 1981 to the first quarter of 1983, culminating in a 10 percent devaluation with reference to the trade-weighted basket in March 1983. Until a wages pause began to take effect in the early part of 1983, much of the depreciation fed back into wage inflation through indexation. Consequently, the real effective exchange rate, measured on the basis of normalized unit labor costs, declined only slightly from the high level of late 1981, and the current account deficit remained about 4 percent of GNP.

Finland, Norway, and Sweden provide a further example of the importance of convergence of economic performance for the management of exchange rates. These three countries are all close trading partners and competitors in international markets, and each one pegs its exchange rate to a trade-weighted currency basket. During the late 1970s, inflation rates were similar both within this group of countries and between them and their other major trading partners, and interest rate policies were fairly effective in curbing speculative pressures and maintaining exchange rate stability, with occasional adjustments of parities relieving shifts in competitive positions. In the early 1980s, however, inflation rates in these countries became higher than in their partner countries. By the autumn of 1982, the resulting loss of competitiveness led to a series of devaluations by all three countries. Since then, all of these countries have implemented policies intended to promote convergence by reducing inflation to rates closer to those prevailing abroad. The average rate of inflation in consumer prices in these three countries, which had reached nearly 10 percent during 1982, declined to 7½ percent for the 12 months through January 1984.

Exchange Rate Policies in Developing Countries

For most developing countries, exchange rate policies since 1982 have been dominated by three major features in the economic setting: the pressing need to carry out adjustments in the current account of the balance of payments; the domestic economic situation, which in many instances was conditioned by high and rising rates of inflation; and large medium-term swings in exchange rates among the major currencies. These topics will be discussed after an examination of recent exchange rate developments.

Exchange Rate Developments

During 1983, the real effective exchange rates of the non-oil developing countries, calculated as an average across countries weighted by gross domestic product (GDP), depreciated by over 6 percent, bringing the cumulative depreciation for the period 1982-83 to about 13 percent (Chart 15); this change followed a cumulative appreciation of 8 percent over the previous three years.3 Nevertheless, the currencies of a large number of these countries continued to appreciate in real effective terms, as demonstrated by the fact that an unweighted average of real effective exchange rate indices for these countries appreciated by 2 percent in 1982 and depreciated by less than 1 percent in 1983.4

Chart 15.Developing Countries: Real Effective Exchange Rates, 1977-831

(Indices, 1977 = 100)

1 These indices measure the evolution of a country’s prices relative to those of its trading partners, adjusted for exchange rate changes. Prices are measured by the average annual consumer price index, with indices of partner countries averaged by using import weights, and exchange rates are measured by an import-weighted index of average annual effective exchange rates. Group indices are GDP-weighted averages of country indices.

The appreciation of real effective exchange rates in 1978-81 was accompanied by a deterioration in current account balances, supported by heavy external borrowing, and the real effective depreciation of 1982-83 by many of these countries played a significant role in their efforts to improve their current account balances and overall external positions (see Chapter 1). For example, the adjustment in the average real effective exchange rate of the 25 major borrowers among developing countries 5 amounted to a cumulative depreciation of about 12 percent over 1982-83, following a cumulative appreciation of nearly 10 percent over the years 1978-81 (developments were similar for unweighted averages).

By contrast, the major oil exporting countries tended to experience an appreciation of their real effective exchange rates, with a cumulative appreciation for the group as a whole of 16 percent over the two-year period 1982-83,6 despite relatively low domestic inflation rates for most of the countries in this group. This development can be explained in large part by the policy of a number of these countries of maintaining relatively stable exchange rates between their currencies and the U.S. dollar, which appreciated over this period against most other currencies.

Among the non-oil developing countries, the low-income countries experienced a slight depreciation of real effective exchange rates in both 1982 and 1983 when averaged using GDP weights (Chart 16), but an unweighted average showed a significant appreciation in both these years. The authorities in a number of these countries did not take exchange rate action to remedy a decline in international competitiveness, perhaps because they did not expect such action to have an immediate impact on volumes of primary product exports. The opposite was true for most other non-oil developing countries, in which the manufacturing sector plays a more important role in exports and import substitution. For instance, major exporters of manufactures, other net oil importers, and net oil exporters all experienced substantial depreciations in their real effective exchange rates in 1983, when averaged with GDP weights (although these changes were less marked for unweighted averages). For a number of countries in these categories, the depreciation exceeded 10 percent, notably for Brazil, Chile, Ghana, Mexico, and Yugoslavia.

Chart 16.Non-Oil Developing Countries: Real Effective Exchange Rates by Analytical Subgroups of Countries, 1977-831

(Indices, 1977 = 100)

1 These indices measure the evolution of a country’s prices relative to those of its trading partners, adjusted for exchange rate changes. Prices are measured by the average annual consumer price index, with indices of partner countries averaged by using import weights, and exchange rates are measured by an import-weighted index of average annual effective exchange rates. Group indices are GDP-weighted averages of country indices.

2 Within the group of “non-oil developing countries.”

Some sharp regional differences in exchange rate trends emerged among the non-oil developing countries in 1983 (Chart 17). In Africa, after a tendency toward appreciation of real effective exchange rates over much of the previous decade, there was a depreciation of 5½ percent in 1983, using a GDP-weighted average of national indices. On the same basis, the average depreciation of real effective exchange rates was sharper among non-oil developing countries in the Western

Chart 17.Non-Oil Developing Countries: Real Effective Exchange Rates by Region, 1977-831

(Indices, 1977 = 100)

1 These indices measure the evolution of a country’s prices relative to those of its trading partners, adjusted for exchange rate changes. Prices are measured by the average annual consumer price index, with indices of partner countries averaged by using import weights, and exchange rates are measured by an import-weighted index of average annual effective exchange rates. Group indices are GDP-weighted averages of country indices.

Hemisphere (12 percent) and Europe (9½ percent). In Asia, where changes in real effective exchange rates have tended to be more moderate over the past half decade than in the other regions, there was on average only a slight depreciation in 1983. In the small Middle Eastern group (dominated by developments in Egypt and Israel), there was an average appreciation of 14 percent.7

Exchange Rate Policies and External Imbalances

The relationship between exchange rate movements and balance of payments developments over the past five years was complex. Sharp changes in both current account balances and in the direction and magnitude of capital movements were closely associated with exchange rate movements in many developing countries. The large net inflows of lending received by these countries in 1978-81 were cut back sharply in 1982-83. Those major borrowing countries that rescheduled their debt in 1982-83 experienced an especially sharp swing between the appreciation of real effective exchange rates between 1978 and 1981 and the depreciation of 1982-83, while for major borrowers with less severe debt problems there was a mild degree of appreciation of real effective exchange rates over the period 1980-83, after some depreciation in 1978 and 1979. However, the experience of the rescheduling countries—a period of appreciation of real effective exchange rates through 1981 followed by depreciation thereafter—was the more typical one for the non-oil developing countries as a whole (Chart 18).

Chart 18.Major Borrowing Developing Countries: Average Real Effective Exchange Rates, 1977-831

(Indices, 1977 = 100)

1 This group consists of the 25 developing countries with the largest total external debt at the end of 1982. These are, in order of the amount of their debt, Brazil, Mexico, Argentina, Korea, Indonesia, Venezuela, Israel, India, Chile, Egypt, Yugoslavia, Turkey, Algeria, the Philippines, South Africa, Portugal, Nigeria, Thailand, Malaysia, Peru, Pakistan, Morocco, Romania, Colombia, and Hungary.

The indices measure the evolution of a country’s prices relative to those of its trading partners, adjusted for exchange rate changes. Prices are measured by the average annual consumer price index, with indices of partner countries averaged by using import weights, and exchange rates are measured by an import-weighted index of average annual effective exchange rates. Group indices are unweighted averages of country indices.

The interpretation of these developments is complicated by the interdependence of exchange rates and capital flows. In countries where exchange rate determination is allowed some degree of sensitivity to market pressures, the impact of capital flows on exchange rates is a direct one. But even in countries where the market does not influence these rates directly, the exchange rate policies of the authorities have been strongly affected by the availability of external finance. This is indeed suggested by the widespread—in some cases, dramatic—exchange rate changes after 1981, when drying up of financing flows made the need for external adjustment evident. It may be argued, however, that the appreciation of the real effective exchange rate creates a greater demand for external finance through its impact on the current account. The large capital inflows experienced in some countries prior to 1982 were in part a response to the growth of current account deficits, which in turn reflected, among other factors, real effective exchange rates that had become increasingly appreciated.

It is nevertheless difficult to determine the relative contribution of several concurrent policies to the current account improvement achieved by the non-oil developing countries. In addition to exchange rate adjustments, the fall in incomes and output, as well as trade and exchange restrictions, played key roles. The available evidence shows that countries that acted to maintain their international price competitiveness— in other words, to avoid substantial real effective appreciation of their currencies—during the period 1978-81 were, on the whole, better able to keep the deterioration of their current accounts in check than were countries whose international price competitiveness, as measured by their real effective exchange rates, had been allowed to decline. Moreover, the former countries were more successful in adjusting to the difficult external situation prevailing in 1982-83 than were the latter countries. These tendencies can be illustrated by comparing the experience of those non-oil developing countries whose currencies depreciated in real effective terms over the years 1978-81 with those whose currencies appreciated. While the average ratio8 of the current account deficit to total exports of the appreciating group steadily rose from 1978 to 1982, with only a slight reversal in 1983, that of the depreciating group increased only slightly over the same period and declined sharply in 1982 and 1983, showing by the latter year a substantial decline from 1978. Although real effective exchange rates are not a perfect measure of international competitiveness and the latter is certainly not the only factor determining the sign and size of the current account, this evidence suggests that exchange rate policies contributed to external adjustments carried out by developing countries during this period.

In the period 1979-83, the authorities of many developing countries encountered several dilemmas with regard to exchange rate policies.9 A major problem was whether to allow exchange rates to reflect fully, partially, or not at all the impact on the balance of payments of the salient external developments occurring during this period—the increase in petroleum prices in 1979-80, the rise in interest rates after 1979, the recession in the industrial countries, and the cutting back of international lending by the commercial banks. Difficult judgments had to be made, in an environment of considerable uncertainty, as to whether these developments were of a temporary or permanent nature and, if temporary, of how long a duration. It was not implausible to believe that at least some of the developments were temporary, but, when the availability of external financing declined, the need for immediate action was apparent. At first, some countries sought to avoid exchange rate action by imposing trade and exchange restrictions. By 1982, however, it was evident that a serious balance of payments crisis had developed and that it was necessary to increase external competitiveness to achieve viable positions with regard to the balance of payments, the level of debt, and the size of debt service payments. Hence, substantial exchange rate adjustments were carried out in 1982-83 by a number of countries.

Nevertheless, as indicated earlier, many countries with current account positions that had deteriorated markedly up to 1982 still have not undertaken adequate exchange rate adjustments. These countries, therefore, have had to carry out adjustments in their balance of payments positions principally by means of measures to reduce aggregate demand and of restrictions on imports and external payments. A large number of non-oil developing countries introduced or intensified import restrictions in 1982 or 1983—although, to be sure, certain restrictions were reduced or eliminated, sometimes in connection with Fund programs. While restrictive measures were considered necessary to meet the immediate balance of payments crisis, they cannot in the long run provide an adequate substitute for measures to improve the price competitiveness of the sectors producing traded goods in these countries. Such improvement will depend not only on exchange rate policies and the removal of price-distorting subsidies and controls but also on the success of controlling the level of aggregate demand, especially the demand of the public sector.

While there can be little doubt that exchange rate adjustments improved the external positions of the countries undertaking them, there is reason to believe that the impact of these adjustments on the current account was less marked than it would have been under different circumstances. First, the fact that so many countries were simultaneously attempting to improve their external competitiveness tended to dampen the effect of exchange rate measures on the current account position of individual countries. In addition, the scope for improved export performance for the developing countries, as a result of exchange rate adjustments, was limited by the existence of protective tariffs and quantitative restrictions in the industrial countries. Finally, the effects of exchange rate adjustments undertaken so far may not yet have been fully realized, because of the lags between the initial adjustment and the expected increase in market shares, and also because the initial adjustments occurred while the industrial economies were still in recession.

Global developments over the past several years have had a significantly different impact on countries exporting oil10 than on oil importing developing countries (see Chapter 1), although there have been striking contrasts within the former group. For those oil exporting countries that had financed a rapid rise in government expenditure after 1980 with heavy borrowing from abroad, the decline in the price of oil since 1981, which was accompanied by lower oil export volumes and higher interest rates, created an especially difficult situation. In the 1970s, these oil exporting countries had already faced the problem of choosing an appropriate exchange rate, particularly when demand for oil was high, since the short-run equilibrium rate was not always consistent with the objective of the authorities of these countries to expand their non-oil traded goods sectors. In many countries that export oil, real appreciation of the effective exchange rate was permitted to occur during the 1970s; subsequently, when some of these countries (e.g., Mexico, Indonesia, and Venezuela) were faced with serious balance of payments strains, exchange rate adjustments were then used to help remedy the situation. In Indonesia, the long-run danger of allowing the competitiveness of the non-oil traded goods sector to weaken was seen at a relatively early stage, and a major exchange rate devaluation was carried out in 1978, as well as in 1983.

In centrally planned economies, as noted in previous Annual Reports, exchange rate policies may not play the same role as in market economies. This role varies among planned economies depending on the extent to which individual units of production make their decisions on the basis of price incentives rather than quantitative targets and direct allocations of inputs from the central economic authority. In a number of member countries with planned economies, steps have been taken to expose individual enterprises to changes in world prices, to unify exchange rates (thereby better reflecting the set of relative prices prevailing in the world economy), and to adjust rates in the direction suggested by changes in the underlying balance of payments position.

Exchange Rates and Domestic Policies

An important factor determining the course of exchange rate policies, including the choice of exchange arrangement, is the domestic policy setting. Domestic considerations often weigh heavily in decisions to allow the domestic currency to appreciate or depreciate. In particular, the movement in the real effective exchange rate of the currency depends on whether the nominal exchange rate is adjusted in line with differences between domestic and foreign rates of inflation. At the same time, changes in the nominal exchange rate are often perceived as an important influence on the level of domestic prices.

If every country adjusted the external value of its currency exactly in line with differences between domestic and foreign inflation rates, real exchange rates would remain constant and there would consequently be no difference in the real effective exchange rate performance of low-inflation, medium-inflation, and high-inflation countries.11 In fact, however, there were striking differences among these groups for certain periods during the past decade (Chart 19).

Chart 19.Developing Countries: Average Real Effective Exchange Rates for Low-, Medium-, and High-Inflation Countries, 1977-831

(Indices, 1977 = 100)

1 These indices measure the evolution of a country’s prices relative to those of its trading partners, adjusted for exchange rate changes. Prices are measured by the average annual consumer price index, with indices of partner countries averaged by using import weights, and exchange rates are measured by an import-weighted index of average annual effective exchange rates. Group indices are unweighted averages of country indices.

The real effective exchange rates of high-inflation countries appreciated, on average, substantially between 1975 and 1980, and then sharply depreciated (correcting the previous overvaluation) between 1980 and 1983. For many of these countries, the downward adjustment of exchange rates began only in 1982, after a period of heavy borrowing and real appreciation of exchange rates in the period preceding 1982. A few countries (Argentina and Uruguay) experimented with schemes to exert a downward influence on inflationary expectations by keeping the exchange rate overvalued. Obviously, the substantial real depreciation after 1981 was made necessary by the previous appreciation.

For medium-inflation countries, the pattern was quite different, with the real effective exchange rate gradually depreciating over the period 1974-79 and thereafter steadily appreciating. The tendency for exchange rate adjustments to lag behind inflation after 1979 was especially strong in countries where inflation significantly exceeded the historical norm and was therefore regarded as a temporary phenomenon (e.g., many African countries). In some of these countries, also, the fear that devaluation would further stimulate inflationary pressures may also have hindered timely exchange rate adjustments.

For low-inflation countries, changes have in general been more muted, with a gradual real appreciation of about 8 percent occurring from 1978 to 1983. This appreciation was in part related to the fact that a number of these countries pegged their currencies to the U.S. dollar (see section on Exchange Arrangements) and did not choose to alter the level of the peg during this period, thereby allowing their currencies to appreciate with the U.S. dollar against non-dollar currencies.

The implementation of required exchange rate adjustments has been complicated by the high rates of inflation prevalent in many of the adjusting countries. For example, in 28 member countries12 inflation rates averaged 20 percent or more during the period 1979-83. In such economies, the direction and implementation of exchange rate policy became a major concern of the authorities and necessarily posed more complicated choices than in countries with relatively low rates of inflation.

One approach that has been taken in recent years in a number of countries—industrial as well as developing—is to gear exchange rate policies, at least in part, to domestic price objectives. Such use of exchange rate policy has taken different forms. In some countries, the authorities deliberately maintain a fixed exchange rate that overvalues the domestic currency: this policy of “cheap imports” is expected both to keep domestic prices down and to weaken upward pressures on wages. In other countries, a depreciation of the domestic currency is simply avoided or delayed because of its anticipated impact on domestic prices. The concerns of the authorities in this regard influenced exchange rate policies in Chile, Mexico, and numerous African countries, as well as in some oil exporting countries, during the inflationary surge following the increase in oil prices in 1979-80.

A related technique used in several countries since the mid-1970s (Argentina, Chile, Portugal, and Uruguay) is to preannounce a schedule of future exchange rates, which may, at least initially, be somewhat overvalued. This policy, which has tended to be associated with relatively high rates of inflation, is intended not only to dampen domestic price increases but also to influence expectations. When accompanied by appropriate fiscal and monetary policies, such a policy might contribute to a gradual decline in the rate of inflation, until it fell below the preannounced rate of devaluation, and hence to future exchange rates that were no longer overvalued. Nevertheless, this policy has in some cases also led to overvaluation of the domestic currency, a deteriorating current account, the distortion of relative prices, and foreign exchange earnings that not only fell but were also funneled into black markets; any slowing of inflation thereby achieved tended to be temporary.

Countries with high or accelerating inflation had difficulty in coordinating exchange rate policies with other policies affecting the money supply, wages, and controlled prices. While the adverse effects of permitting the domestic currency to become overvalued were evident, there was a countervailing fear of entering into an inflationary spiral in which an effectively indexed exchange rate played a key role. In some countries (notably Chile) exchange rate developments tended to have an asymmetric, or ratchet, effect on real wages. While an appreciation of the rate automatically raised real wages, the corresponding depression of real wages caused by a depreciation tended to be resisted by wage earners, resulting in a rise in unemployment to the extent that such resistance was successful.

The task of coordinating exchange rate and interest rate policies also proved to be especially complicated in an inflationary setting, particularly if it was uncertain whether inflation would accelerate or decelerate. Almost all countries rejected the approach of permitting both rates to be determined in the free market because of the thinness of the markets involved. When the authorities set both rates, the problem arose that expectations created by the rates might not be consistent with the goals of the authorities. For example, in countries where capital outflows were permitted (or could not be successfully controlled), an overvalued exchange rate tended to provoke expectations of a devaluation, which in turn led to capital outflows: this was the experience of certain Latin American countries in 1981-82. When interest rates were raised—sometimes to rates exceeding in real terms those prevailing in international markets—in order to induce the placement of financial savings within the country, the result was either to raise the public’s expected rate of inflation (because officially determined interest rates were regarded as a signal of the authorities’ own expectations) or, when private expectations were not affected in this way, to discourage borrowing for private investment. Conducting exchange rate and interest rate policies that avoid such outcomes continues to be one of the most difficult tasks facing the authorities in countries suffering from both external payments problems and high inflation.

The ability of governments to deal with problems of the sort just described has generally been enhanced by a firm control over the fiscal balance. When the authorities have been seen to be unable to prevent the fiscal deficit from rising or to reduce it when it had already become excessive, expectations with respect to both the exchange rate and inflation have tended to become more volatile, and the authorities’ margin for maneuver with regard to their exchange rate and interest rate policies has been narrowed.

Impact of Exchange Arrangements

An analysis of movements in real effective exchange rates by type of exchange arrangement reveals, not surprisingly, that currencies pegged to the U.S. dollar have appreciated substantially (about 25 percent) since 1980 (Chart 20). In most of the countries affected, exchange rates have not been adjusted downward against the dollar to offset the influence of the effective appreciation of the dollar itself. Currencies pegged to the SDR, as well as those pegged to a composite, have also appreciated over this period, but by less than those pegged to the U.S. dollar. It may be noted that all three of these groups include some oil exporting countries. In 1983 there was continued depreciation (again, since 1980) of currencies pegged to the French franc and of currencies classified in the category of other exchange arrangements (including those countries moving exchange rates according to a set of indicators).

Chart 20.Developing Countries: Real Effective Exchange Rates by Exchange Arrangements, 1977-831

(Indices, 1977 = 100)

1 These indices measure the evolution of a country’s prices relative to those of its trading partners, adjusted for exchange rate changes. Prices are measured by the average annual consumer price index, with indices of partner countries averaged by using import weights, and exchange rates are measured by an import-weighted index of average annual effective exchange rates. Group indices are unweighted averages of country indices.

The large changes in exchange rates that have taken place among the major currencies over the past five years have had a substantial impact on the developing countries, not only on their real effective exchange rates (as shown above) but also on such variables as commodity prices, the real value of international reserves, and external debt. In particular, the dollar appreciation tended to depress commodity prices in U.S. dollar terms while nominal magnitudes of debt service (which was mainly in dollars) remained unchanged, thereby raising the real burden of debt service. To the extent that the currency composition of reserves differs from that of payments, the real value of reserves varies with changes in exchange rates among the currencies in which reserves are held.

The impact of changes in exchange rates among major currencies on the variables just mentioned cannot be prevented by choosing a particular exchange rate regime. The effect of such changes on commodity prices in terms of one of the internationally traded currencies cannot be offset by an exporting country unless it possesses, alone or together with other exporters, a certain degree of monopolistic power in its export markets. Fluctuations in the real values of international reserves and external debt resulting from foreign exchange rate movements can be reduced by management of the currency composition of external assets and liabilities but not by pegging to a particular currency composite.

Some effects of exchange rate movements, however, can be offset, at least partially, through the choice of exchange arrangement. For example, when the currency composition of import payments differs from that of export receipts, or when trade is conducted in more than one foreign currency, a change in exchange rates among major currencies will alter the balance of trade in domestic currency terms, thereby affecting the level of aggregate income and demand; this impact can be muted through choice of an appropriate currency composite to which to peg the local currency. The same type of exchange arrangement can be used to reduce, on average and over time, the exchange risk faced by individual traders because contracts are denominated in a foreign currency other than that to which the domestic currency is pegged, or to dampen variations in any key economic variables, such as external competitiveness, that are measured by real effective exchange rates. The choice of currency, or currencies, to which to peg the local currency has sometimes also been used to reduce the extent to which inflation is transmitted from major trading partners through increases in the prices of their exports.

It is also generally true, however, that no exchange arrangement can entirely offset all these effects, because an arrangement that is especially well suited to smooth the effects of foreign exchange rate fluctuations in one respect (e.g., variations in average external competitiveness) may deal less effectively with other types of effects (e.g., those on the domestic currency value of the balance of trade). Hence, the ideal exchange arrangement, taking into account all relevant criteria will generally involve a compromise among a variety of objectives. Nevertheless, for medium-and high-inflation countries, the objective of maintaining external competitiveness is a dominant one, requiring frequent adjustment of exchange rates more or less in line with the difference between domestic and foreign inflation.

The response of developing countries to the exchange rate environment they face is shown by the evolution of their exchange arrangements (Table 9). Although no radical shifts occurred in the pattern of such arrangements in the 12 months ended June 30 1984 there has been over the past five years a gradual increase in arrangements involving either a peg of the domestic currency to a composite of other currencies or flexible exchange arrangements, paralleled by a decrease in arrangements under which the domestic currency is pegged to a single currency. The fact that there are still 51 of the latter type of arrangements suggests that in many countries the single currency peg is either highly advantageous in terms of convenience, especially when the bulk of external transactions is conducted in a single foreign currency, or is perceived as a cornerstone of domestic financial policies and, therefore, also of confidence in the national currency.

Table 9.Developing Countries: Exchange Rate Arrangements, End of June 1978-841(Number of countries)
Pegged to a single currency62615856565451
U.S. dollar41414038383633
French franc11141414131313
Pound sterling4311111
Other currency3333444
Pegged to composite28273232343535
Other composite13141718192124
Flexible arrangements23292832353639
Adjusted according to a set of indicators5434456

Based on midyear classifications; excludes Democratic Kampuchea, for which no current information is available. For classification of countries in the group shown here, see Chapter 1. Table 3. footnote 1.

This category comprises the following categories used in Table 10: “Flexibility limited vis-à-vis single currency,” “Other managed floating.” and “Independently floating.”

Surveillance Over Exchange Rate Policies

Against the background of an improving but still difficult world economic and financial situation, the Fund has continued to attach great importance to the exercise of its surveillance function. An important aspect of this function is the need to ensure that the exchange rate policies of Fund members are guided by the three principles enunciated in the 1977 document “Surveillance Over Exchange Rate Policies,” which enjoins members to avoid manipulating exchange rates or the international monetary system, to intervene in the exchange market if necessary to counter disorderly conditions, and to take into account in their intervention policies the interest of other members.13 Much more broadly, however, the Fund’s surveillance activities are concerned with the obligations of its members under Section 1 of Article IV of the Articles of Agreement. This section not only proscribes the manipulation of exchange rates but also obliges members to collaborate with the Fund and other members to promote a stable system of exchange rates. In particular it refers to the need for each member to direct its economic and financial policies toward the objective of fostering orderly economic growth with reasonable price stability and to seek to promote stability by fostering orderly underlying economic and financial conditions. In this wider dimension, surveillance embraces, in addition to exchange rate policies and policies adopted for balance of payments purposes, policies that are undertaken primarily for domestic reasons but also have consequences for exchange rates and for the international monetary system as a whole.

In March 1984, the Executive Board conducted both its biennial review of the 1977 document and its annual review of the implementation of surveillance.14 The Board’s conclusion was that the experience in the implementation of surveillance did not call for a revision of the principles and procedures set out in the document, but that it did call for more active implementation. In this context, the Board stressed that, while the assessment of exchange rate policies was a complex task, the Fund had to take a position on the appropriateness of members’ exchange rate policies, irrespective of the exchange arrangements chosen by an individual member and of the member’s need for Fund financial support. The Board also stressed that, to have firm and effective surveillance, it was necessary that Fund members give active and broad support to the positions taken by the institution.

A further point noted during the 1984 surveillance review was that many of the international economic difficulties of recent years have been associated with the pronounced swings in exchange rates between major industrial countries, and with the repercussions of the prevailing low levels of economic activity and high interest rates in these countries on the rest of the world. There was a broadly shared view that to some extent these developments resulted from domestic policies in major industrial countries that did not sufficiently promote the evolution of stable economic and financial conditions and also failed to take account of the implications for other countries and for the international monetary system as a whole. Therefore, the Executive Board concluded that it was incumbent on the Fund to form a view on the domestic policies needed to foster the smooth working of the system and to attempt to persuade its members to follow such policies.

Issues in the Implementation of Surveillance

The restoration of stable domestic economic and financial conditions in the major industrial countries remained one of the central issues in the implementation of Fund surveillance during 1983 and the first half of 1984. The nature of foreign exchange markets is such that it is unrealistic to assume that much progress can be made in the direction of reducing exchange rate instability among the major currencies as long as domestic conditions are unstable in some fundamental ways. The existence of an inflation problem, in particular, is bound to be accompanied by exchange rate instability because economic agents will be uncertain as to the future level of inflation, the impact on the balance of payments, the corrective financial measures that will be used, and the degree of success of these measures. Stable domestic conditions, however, involve much more than an absence of inflation. They require a monetary policy that is viewed by economic agents as insuring them against a return of inflation. They also depend on a fiscal policy that is adequately supportive of the monetary stance and is consistent with the financing of private capital formation without undue pressures on real interest rates.

While there is generally no disagreement with respect to the need for a restoration of stable domestic conditions, policymakers at times differ as to how this aim is best achieved. In recent years, it is mainly in the fiscal area that opinions have tended to differ. In Japan, the Federal Republic of Germany, and the United Kingdom, the first priority of fiscal policy was to reduce the fiscal deficit, if only because this was the best way to foster a reduction of interest rates and a recovery of private investment. On the other hand, in the United States the authorities were convinced that the first priority of fiscal policy was a reduction of the tax burden imposed on the private sector; if this reduction led to an excessive fiscal deficit, then government expenditures must be reduced. If this reduction turned out to be a long and arduous task, it would still be better—in the view of the U.S. authorities—to have a deficit that was high by historical standards for a number of years than to abandon the plan to reduce the tax burden.

The authorities of various countries have also assessed differently the importance of the relationships among the fiscal deficit, interest rates, and the exchange rate. The main problem in this area is that the relationships are often obscured by a number of factors, including the level and composition of government spending and taxes, the extent to which government debt is perceived to be wealth by investors, the ratio of government debt to GNP, and expectations. Expectations are important in several respects. First, expectations are an important element influencing the effect of fiscal expansion on interest rates. For example, fiscal expansion may lead to higher long-term interest rates, even during a period of relatively low private demand for credit, if economic agents have reasons to expect that the expansion will persist when the private demand for credit recovers. Second, expectations also influence the relation between interest rates and exchange rates. The expectation that the fiscal expansion is to be accommodated by an expansionary monetary policy often will be associated with anticipations of higher inflation, in which case the combination of policies could lead to a rise in nominal domestic interest rates associated with a depreciation of the exchange rate. In contrast, the expectation that the fiscal expansion is not to be accommodated may contribute to a rise in domestic real interest rates and an appreciation of the exchange rate. This complexity of the relationship between fiscal policies and exchange rates explains the lack of a systematic correlation, either over time or across countries, between fiscal policies and exchange rates. It also explains why, in any specific case, it is often difficult for the authorities in different countries to arrive at a common judgment of the contribution of fiscal policy to exchange rate developments.

In the monetary area, the problems have been essentially technical rather than related to any fundamental differences of view. In all major industrial countries, increased reliance has been placed over the past several years on the control of monetary growth as the principal way to set monetary policies in the medium term. More recently, however, large changes in velocity have obliged a number of central banks to adopt a flexible and judgmental approach to the determination of monetary growth. Part of the problem is temporary, associated with the rather sudden decreases in inflation and in nominal interest rates that have characterized the past two years. But part of it may be longer lasting, being associated with the development of close substitutes for money balances as traditionally defined and with the increase in substitution among major currencies on the international level. For all these reasons, it has become particularly difficult to determine the appropriate rates of growth of monetary aggregates, and, therefore, to assess whether specific exchange rate developments are the result of excessively low or high rates of monetary growth.

While the complexities involved in assessing the source of exchange rate problems and other problems resulting from international interdependence are all too real, they should not be exaggerated. In some instances, there is little doubt with respect to the basic source of the problem, even though it may be difficult to give a precise description of the relevant relationships. The Fund has then a clear mandate to press the country concerned to improve its policy stance. At times, the process of distinguishing the various issues involved can help to reduce divergences of view. It is essential for the success of surveillance that, whenever the exchange rate implications of certain policies are not well understood, a serious attempt be made at a very early stage to work toward a common understanding of the processes at work. This is why the Fund has endeavored to develop its analysis in this domain in recent years, and why it has sought through its various publications, through informal contacts between the Managing Director and member countries, through participation in meetings of the Group of Ten and other country groups, and through Article IV consultations to present the conclusions of its work to its members.

More specifically, during the past year and a half the Fund has forcefully advocated the adoption of policy measures that would help to foster a favorable convergence of financial conditions among its industrial members. The progress that many industrial countries have made toward restoring noninflationary sustainable rates of economic growth is encouraging, not only for its own sake but for its salubrious effects on the exchange rate system. The consolidation and furthering of these gains over the next few years should go far toward enhancing the stability of exchange rates among the major industrial countries and reducing pressures on trading relationships and on the economic prospects of other countries. But the one area where progress is so far lacking in a number of countries is also one of those most directly important for the smooth operation of the exchange rate system: the attainment of balanced and consistent macroeconomic policies so as to reduce the incentive for the shifting of financial assets among currencies in search of relatively high returns. Only by making progress in this area will countries be able to provide financial markets with the means to formulate stable expectations based on a reasonable certainty that policies as well as economic performance will be sustained over the medium term.

The serious economic problems faced by developing countries were the other central issue in the implementation of Fund surveillance during 1983 and the first half of 1984. The Fund has continuously stressed that these problems are partly the result of unsatisfactory economic conditions in developed countries. Over the past year and a half, developing countries have benefited from the strong expansion of economic activity and imports in the United States. Nevertheless, on the whole the external environment remains unfavorable to them, especially because of the high interest rates in international capital markets and the still relatively low level of economic activity in many industrial countries. Thus, further improvements in the underlying economic and financial conditions in developed countries—particularly the major industrial countries—would greatly ease the adjustment problems of developing countries by fostering an enlarged demand for their exports and a reduction in the interest rates on their foreign debt.

At the same time, the Fund has stressed that developing countries must further strengthen their adjustment efforts to restore a sustainable external position and to renew the process of economic growth. They cannot afford to wait for better conditions in industrial countries. Furthermore, in many instances, economic problems can also be traced to domestic economic policies. Ultimately, the developing countries have no choice but to harness their economic resources to expand exports and replace certain imports with due regard to comparative advantage, while providing adequate incentives for the investment of national saving at home rather than abroad and for foreign equity investment. To achieve these objectives, it is equally essential to keep aggregate demand under control through appropriate fiscal and monetary policies and to improve domestic resource allocation through a rationalization of the structure of incentives.

In both respects, exchange rate policies can be expected to play a key role. The experience of a number of developing countries over the past two or three years, including Indonesia, Mexico, Pakistan, Somalia, Turkey, and Uganda, has demonstrated once more that exchange rate flexibility, accompanied by a removal of controls on domestic prices and the implementation of appropriate fiscal and monetary policies, can contribute to external adjustment in developing as well as developed countries. Because of national differences in economic structure, the contribution of exchange rate flexibility to the adjustment process is greater in some countries than in others; and that contribution may in some respects also be limited by the global nature of the current adjustment problem. But there is no doubt that exchange rates do influence the aggregate volume of products, especially manufactures, that the industrial countries import from developing countries; that in all countries the efficiency of allocation of foreign exchange among domestic economic agents is greatly enhanced when the price of foreign exchange is permitted to reflect its true scarcity; that the process of lowering unduly high real wage rates is often facilitated by an exchange rate adjustment; and that a realistic exchange rate is ultimately necessary to avoid a flight of domestic capital. Furthermore, the use of trade and exchange restrictions is a more costly form of adjustment in the long run because of its detrimental effect on the efficiency of resource allocation.

Exchange rate flexibility and the removal of controls on domestic prices should not, however, be viewed as policies taken in isolation from others. Reduced fiscal deficits, for example, are required for improved control over the growth of monetary aggregates. With inadequate control of monetary growth, exchange rate flexibility will restore overall balance of payments equilibrium, but at the cost of runaway inflation and, possibly, a dislocation of the domestic economy. The Fund in its surveillance activities has therefore continued to stress that improvement in economic performance requires the implementation of domestic financial policies that provide appropriate support for exchange rate adjustment. Given the magnitude of the problems faced by many developing countries, adjustment also requires a longer-run strategy aimed at improving the overall efficiency of the production system, increasing private saving and investment, and exploiting developments in a country’s comparative advantage. This is also a crucial aspect of the adjustment policies that the Fund, in cooperation with the International Bank for Reconstruction and Development, has been advocating in its discussions with member countries.

Throughout 1983 and the first half of 1984, the Fund has continued to encourage the commercial banks to provide adequate financing in support of Fund-assisted adjustment programs. In many cases, the debt burden of the country, in particular the indebtedness to banks, is such that a continuation of the support of the banks is absolutely necessary if the adjustment program is to proceed with a reasonable chance of success. When the country has shown a serious willingness to adjust, commercial banks have generally become more forthcoming and have provided a critical level of financing. For the smaller low-income developing countries, in particular many of the African countries, that continue to have limited access to financial markets or cannot afford the interest rate prevailing in these markets, the level of external financing depends more heavily on foreign aid, and hence the lack of real growth in aid often undermines the success of adjustment efforts in these countries.

Possibly the most detrimental effect of the economic problems faced by developed and developing countries alike over the past few years is the increase in trade restrictions, especially restrictions taking the form of nontariff barriers negotiated on a bilateral basis outside the framework of the General Agreement on Tariffs and Trade (GATT). There are many reasons for this growth of protectionism. Some are related to the unwillingness to accept the consequences of changes in comparative advantage. Others are more directly related to the exchange rates currently prevailing among the major currencies, the desire on the part of many developing countries to limit the depreciation of their exchange rates despite the persistence of extremely high inflation rates, and the belief in countries suffering from unemployment that import restrictions may safeguard jobs. Whether related to exchange rate issues or not, the proliferation of trade barriers has now reached a stage at which the effectiveness of exchange rate adjustments and the efficiency of the multilateral trading system are in jeopardy.

As was indicated in the 1983 Annual Report, the Fund has intensified its collaboration with the GATT and has placed greater emphasis on issues of protectionism in all of its surveillance activities. So far, however, progress in restraining the rise of protectionism has been very disappointing. Not only have there been no concrete policy actions to resume progress toward an open multilateral trading environment, but, in fact, protectionist measures have been intensified. One especially worrisome consequence of this evolution is that it hinders the adjustment process of developing countries with high external indebtedness. These countries must expand their exports of manufactured and agricultural products at a rapid rate over the next few years if they are to restore an adequate external position while sustaining a moderate rate of economic growth. This, however, will be possible only if there is a reversal of the current protectionist tendencies.

Procedures for Implementing Surveillance

The Fund carries out its surveillance activities both in a multilateral context and through its consultations with individual members. These activities are supported by a continuous monitoring of developments in exchange rates and exchange arrangements.

The world economic outlook exercise remains the main framework for multilateral surveillance. The comprehensive analysis of the world economy and the exchange rate system that is undertaken in the world economic outlook papers prepared at least twice each year provides a basis for the reviews by the Executive Board and the Interim Committee of the central surveillance issues, including those that are related to the global effects of the economic policies of major countries. In recent years, these papers have emphasized a medium-term approach. In particular, medium-term scenarios have been developed to highlight the policies needed to achieve a sustainable noninflationary recovery in the industrial countries and to analyze the adjustment and debt problems of developing countries. Aside from these papers, the staff also prepares Executive Board papers on particular surveillance issues. During 1983 and early 1984, papers on particular issues included “Issues in the Assessment of the Exchange Rates of Industrial Countries,” and “The Exchange Rate System: Lessons of the Past and Options for the Future.”15

Another channel for multilateral surveillance is the participation of the management of the Fund in various multilateral meetings of the major industrial countries. The Managing Director often participates in the meetings of the ministers and governors of the Group of Ten and, since the 1982 economic summit meeting, in the regular meetings of the ministers of the five countries whose currencies are part of the SDR basket. On these occasions, the Managing Director has reviewed recent economic developments and has underlined the need for cooperation in order to put in place policies designed to achieve convergence toward a lasting reduction in inflation, higher employment, and sustainable economic growth. More broadly, the Managing Director has stressed the need for major countries, in framing their domestic policies, to give weight to the impact of their decisions on exchange rates and on economic conditions in other countries.

One subject that has continued to be given particular prominence in the multilateral surveillance activities of the Fund in 1983 and early 1984 is the problem of external indebtedness. The importance attached to this subject was underscored by a number of Executive Board meetings on questions related to Fund policies in the context of external debt-servicing problems and developments in international capital markets. Moreover, a number of projects were initiated to further improve the monitoring, presentation, and analysis of external indebtedness. New tables on international banking statistics have been incorporated in the Fund’s International Financial Statistics beginning with the January 1984 issue. Because of the seriousness of the debt problem, cooperation with official and private creditors has intensified. Members of the Fund staff cooperate closely with official creditors in the Paris Club, and adjustment programs supported by Fund resources in the upper credit tranches remain a prerequisite for the consideration of rescheduling requests. Furthermore, when private flows are impaired and countries are seeking comprehensive packages of financial assistance from commercial banks, the Fund now often finds it necessary to adopt a more active role in securing confirmation from both official and private creditors that the external financing assumptions of economic adjustment programs are realistic. In this connection, the Fund staff has also been prepared, at the request of a debtor country, to discuss the outlook for the country’s external payments with private financial institutions, as well as with official agencies.

The main channel for the Fund’s exercise of surveillance in a bilateral context is the Article IV consultation. Even though the basic procedures of these consultations have remained unchanged in recent years, there is a continuing evolution in Fund practices with respect to the subject matters covered. The range of topics has been enlarged particularly in respect of international interdependence, external debt, and protectionism. Consultation reports for the major industrial countries now include a section on the international repercussions of their domestic policies. Furthermore, there has been an increasing emphasis in consultation reports on the elaboration of medium-term scenarios concerning the external debt position, which often includes a sensitivity analysis with respect to the major assumptions. This analysis is now viewed as an analytical tool that should be part of the set of elements used in any assessment of the sustainability of the balance of payments situation of all countries regardless of their debt situation. There has also been an increasing emphasis on trade policy matters. Of special importance is the emphasis in the 1984 surveillance review on the need to cover in consultation reports the protectionist measures and export subsidies adopted by members as a result of their participation in customs unions or other regional arrangements. Moreover, to the extent possible the economic costs of protectionist measures taken by individual countries or groups of countries since the last Article IV consultation should be quantified, and the impact of protectionism on domestic adjustment should be examined where relevant. More generally, there is now to be greater emphasis on following up on the conclusions of previous consultations in evaluating current policies.

The timing and timeliness of Article IV consultations are of considerable importance. In principle, consultations are to take place annually, but in practice even the operational guideline that had been adopted—of covering three fourths of the membership annually—was not met in the early 1980s. This slippage led the Executive Board on the occasion of the 1983 surveillance review to decide on the implementation of a stricter adherence to an annual consultation cycle for most members—including in particular those whose policies have a substantial impact on other economies, those that have Fund-supported programs, and those for which there are substantial doubts about the medium-term viability of the balance of payments situation. Furthermore, it was decided that at the conclusion of each consultation a date would be set by which the Executive Board discussion of the next consultation was expected to be concluded. These procedural changes have led to a marked increase in the frequency of consultations during the past year and a half, with membership coverage rising to 80 percent during 1983. The improvement in coverage was particularly marked for countries with Fund-supported programs; all of them were covered during 1983 or the first few months of 1984.

Another decision that was taken during the 1983 review and implemented in the course of 1983 was the initiation of a system whereby the Executive Board is notified regularly of all sizable changes in real effective exchange rates. This system supplements existing procedures for notification of changes in member countries’ nominal exchange rates. Currently, all except 35 Fund members are covered by the new system. For the remaining members, the unavailability of reliable price or other data has so far made the computation of a meaningful index of the real effective exchange rate difficult, but efforts are continuing to make the system comprehensive. Under the new system, a large change in a country’s real effective exchange rate since the last Executive Board discussion of the country’s exchange rate policies results in the issuance of an information notice to the Board. This notice includes a brief factual discussion of developments in the member’s exchange rate and its costs and prices, together with a statement of how these were related to changes in the balance of payments and to other developments in the economy.

Exchange Arrangements

The international exchange rate system has continued to evolve toward greater flexibility of arrangements in recent years. A major aspect of this evolution has been the substitution of various forms of managed floating for pegged arrangements—in particular, arrangements involving a peg to a single currency. Another feature of the movement away from single currency pegs has been the increased use of currency composites, including the SDR, which has helped to reduce short-run fluctuations in effective exchange rates. Of the 18 members that changed their exchange arrangements in 1983 and the first half of 1984, 5 ceased to peg their currencies (including de facto peg) and adopted more flexible exchange rate arrangements. Another member that previously maintained its currency with limited flexibility vis-à-vis the U.S. dollar adopted a regime of managed floating. Four other members whose currencies were previously pegged to the U.S. dollar adopted currency composites as pegs (2 of them choosing the SDR), while 3 other members gave up pegging to the SDR basket and pegged instead to a currency composite tailored to reflect more closely their particular trading patterns. Only one member changed from a more flexible exchange arrangement to pegging its currency to that of another member.

Developing countries made increased use of inflation differentials, alone or in combination with other factors, as an indicator for automatic or quasi-automatic adjustment of an otherwise pegged arrangement. As of the end of the first quarter of 1984, 8 members had incorporated in their exchange arrangements a policy of periodic adjustment of the exchange rate to maintain international competitiveness according to such an indicator, while 5 other members undertook such a policy as part of a program supported by the use of Fund resources in 1983 and in the first quarter of 1984.

In sum, the number of Fund members with “more flexible” exchange rate arrangements increased from 33 at the end of 1982 to 38 at the end of June 1984 (of a total membership of 146).16 The number of countries pegging to a single currency declined from 56 to 51 over the same period, while the number of those pegging to a composite of currencies increased from 38 to 39. Arrangements in the intermediate group between pegged and flexible arrangements (labeled “flexibility limited”) declined from 18 to 17. Since the inception of the present classification system, at the end of 1981, the proportion of Fund members with “more flexible” arrangements has risen from 21 percent to 26 percent, and those pegging to a currency composite from 25 percent to 27 percent. The use of the SDR as a peg declined, however, from 10 percent to 7½ percent of the membership, as composites tailored to individual trade patterns were preferred.

At the end of June 1984, the currencies of 51 members were pegged to a single currency—33 to the U.S. dollar, 13 to the French franc, 2 to the South African rand, and 1 each to the Indian rupee, the pound sterling, and the Spanish peseta (Table 10). Eleven currencies were pegged to the SDR and 28 to other currency composites. In all, therefore, 90 members had exchange arrangements classified under the “pegged” category. Seventeen members maintained exchange arrangements classified in the “flexibility limited” category. Within this group, 9 currencies were in the subclassification “single currency” (all vis-à-vis the U.S. dollar) as a result of having their exchange rates fluctuate within the equivalent of margins of 2¼ percent or less against an identifiable single currency of another member; the other 8 currencies were those of countries maintaining cooperative arrangements under the EMS. Thirty-eight members maintained exchange arrangements in the “more flexible” category; of these, 6 adjusted their exchange rates according to a set of indicators, 24 managed floating rates, and the currencies of 8 members floated independently.

Table 10.Exchange Rate Arrangements as of June 30, 19841
Flexibility Limited vis-à-vis
a Single Currency or
Group of Currencies
More Flexibh
Pegged toAdjusted
to a set of
U.S. dollarFrench
Antigua and
CameroonBurundiAustriaBahrain 4DenmarkChile 3Costa
Rica 3
Bahamas 3Central
Guinea 3Bangladesh ‘GhanaFranceColombiaJapan
Republic of
BotswanaGuyanaGermany,Peru 3Ecuador 3Lebanon
BoliviaChadChina 3Qatar 4Republic ofSomalia 5
DjiboutiCongoThe Gambia
JordanCyprusSaudi Arabia 4IrelandGuinea-
Ivory CoastRwandaFinland6United Arab
Emirates 4
Luxembourg 3IcelandUnited States
São Tomé
El Salvador 3TogoSwaziland
EthiopiaUpper VoltaViet NamMalaysia6Korea
GrenadaMaltaMexico 3
HaitiMauritiusNew Zealand
IraqPapua New
Lao People’s
RomaniaSri Lanka
Libyan Arab
Nicaragua 3SolomonUganda 3
St. LuciaTunisiaZaïre
St. Vincent
and the
Sierra Leone
Sudan 3
Republic 3
Trinidad and
Venezuela 3
Yemen Arab
Yemen, People’s
Republic of

No current information is available for Democratic Kampuchea.

All exchange rates have shown limited flexibility vis-à-vis the U.S. dollar.

Member maintains dual exchange markets involving multiple exchange arrangements, shown is that maintained in the major market.

Exchange rates are determined on the basis of a fixed relationship to the SDR, within margins of up to ± 7.25 percent. However, because of the maintenance of a relatively stable relationship with the U.S. dollar, these margins are not always observed.

The exchange rate is maintained within overall margins of ± 7.5 percent around the fixed shilling/SDR relationship; however, the exchange rate will be re-evaluated when indicative margins of ± 2.25 percent are exceeded.

The exchange rate is maintained within margins of ± 2.25 percent.

Margins of ± 6 percent are maintained with respect to the currencies of other countries participating in the exchange rate mechanism of the European Monetary System.

The exchange rate is maintained within margins of ± 5 percent on either side of a weighted composite of the currencies of the main trading partners.

International Liquidity, Reserves, and Capital Markets

The current recovery from the world recession and the response of financial market participants to the external payments difficulties of many developing countries had a major impact on the expansion and distribution of international liquidity during 1983. Many countries pursued a policy of rebuilding their reserve holdings in order to offset the losses of foreign exchange reserves in 1982. This expansion of international reserves encompassed diverse patterns of growth in its components of official holdings of gold, foreign exchange, SDRs, and reserve positions in the Fund.

The remainder of this chapter examines a number of aspects of the recent movements in international reserves and liquidity. First, there is a review of the recent growth, composition, distribution, and sources of international reserves, non-gold reserves, and foreign exchange reserves. Second, an examination of the roles of private international financial markets and official agencies in the provision of international liquidity and the financing of the adjustment process is undertaken. In addition to considering the implications for international liquidity of reduced access for many countries to financial markets, this discussion describes the efforts of debtor countries, banks, national agencies, and international organizations to develop a smooth and orderly adjustment process in the period since the emergence of large external payments difficulties in 1982. Next, the issues involved in determining the adequacy of international reserves are reviewed. Finally, there is a discussion of the role of the Fund in the provision of international liquidity through the extension of conditional credit and SDR allocation.

Recent Evolution of Official Reserve Assets

In 1983, total international reserves were affected by renewed accumulation of non-gold reserves and a decline in the market value of official gold holdings. The growth of non-gold reserves reflected larger holdings of both Fund-related reserve assets and foreign exchange reserves. While all the major country groups increased their holdings of Fund-related reserve assets, the expansion of foreign exchange reserves was the result of accumulations in industrial countries and non-oil developing countries that outweighed reduced holdings in the oil exporting countries. A decline in the market price of gold reduced the market value of official gold holdings by 12 percent. The reduction in the market value of official gold holdings in 1983 was slightly larger than the increase in non-gold reserves, and total international reserves, including gold valued at market prices, declined by 2 percent.

Non-Gold Reserves

Non-gold reserves increased by 10 percent in 1983 to SDR 364 billion at the end of the year (Table 11). Although this rate of growth of non-gold reserves was lower than the average rate of growth experienced during the period from 1973 to 1980, it represented a recovery from 5 percent growth in 1981 and a decline of 2 percent in 1982.

Table 11.Official Holdings of Reserve Assets, End of Selected Years 1978-83 and End of March 19841(In billions of SDRs)
All countries
Total reserves excluding gold
Fund-related assets
Reserve positions in the Fund14.811.816.821.325.539.140.0
Subtotal, Fund-related assets22.924.228.637.743.253.554.6
Foreign exchange224.2249.9293.1299.0287.8310.1310.5
Total reserves excluding gold247.1274.1321.8336.7331.0363.6365.1
Quantity (millions of ounces)1,036.8944.43952.4951.5947.1945.7945.5
Value at London market price179.9367.1440.2325.0392.3344.6345.2
Industrial countries
Total reserves excluding gold
Fund-related assets
Reserve positions in the Fund9.67.710.713.517.125.625.9
Subtotal, Fund-related assets16.017.119.625.531.137.137.7
Foreign exchange127.2136.1164.7159.6153.2167.8170.1
Total reserves excluding gold143.1153.2184.3185.1184.4204.9207.8
Quantity (millions of ounces)884.2789.13787.9787.6787.3786.6786.4
Value at London market price153.4306.7364.2269.0326.1286.6287.1
Oil exporting countries
Total reserves excluding gold
Fund-related assets
Reserve positions in the Fund4.
Subtotal, Fund-related assets4.
Foreign exchange40.151.062.866.659.854.950.7
Total reserves excluding gold44.955.
Quantity (millions of ounces)36.336.740.141.742.042.242.2
Value at London market price6.314.218.514.217.415.415.4
Non-oil developing countries
Total reserves excluding gold
Fund-related assets
Reserve positions in the Fund0.
Subtotal, Fund-related assets2.
Foreign exchange56.962.765.672.874.987.489.7
Total reserves excluding gold59.065.969.377.578.191.193.0
Quantity (millions of ounces)116.4118.7124.5122.3117.8116.9116.9
Value at London market price20.246.157.541.848.842.642.7
Source: International Monetary Fund, International Financial Statistics.

“Fund-related assets” comprise reserve positions in the Fund and SDR holdings of all Fund members and Switzerland. Claims by Switzerland on the Fund are included in the line showing reserve positions in the Fund. The entries under “Foreign exchange” and “Gold” comprise official holdings of those Fund members for which data are available and certain other countries or areas, including Switzerland. For classification of countries in groups shown here, see Chapter 1, Table 2 and Table 3, footnote 1.

One troy ounce equals 31.103 grams. The market price is the afternoon price fixed in London on the last business day of each period.

The decrease recorded in the quantity of countries’ official gold holdings from the end of 1978 to the end of 1979 reflects mainly the deposit by the members of the European Monetary System of 20 percent of their gold holdings with the European Monetary Cooperation Fund. The European currency units issued in return for these deposits are shown as part of the countries’ official foreign exchange holdings.

The expansion of non-gold reserves in 1983 by SDR 33 billion represented increased holdings of both Fund-related assets and foreign exchange. Fund-related reserve assets grew by SDR 10 billion (24 percent), and foreign exchange reserves expanded by SDR 22 billion (8 percent). Since Fund-related reserve assets grew at a more rapid rate than foreign exchange reserves, the share of Fund-related reserves in total non-gold reserves continued to rise to 15 percent in 1983.

The larger stock of non-gold reserves at the end of 1983 reflected additional holdings by industrial countries (SDR 21 billion) and non-oil developing countries (SDR 13 billion) and lower holdings by oil exporting countries (SDR 0.9 billion). As a result of these changes, industrial countries held 56 percent of total non-gold reserves, the oil exporting countries 19 percent, and the non-oil developing countries 25 percent.

Foreign Exchange Reserves

Although the increase of 8 percent in foreign exchange reserves in 1983 was lower than the average annual growth rate of 14 percent during the period from 1973 to 1980, it represented a considerable expansion of foreign exchange reserves when compared with the growth of 2 percent in 1981 and the decline of 4 percent in 1982. Foreign exchange reserves increased in 1983 for the industrial countries and the non-oil developing countries but declined for oil exporting countries (Table 11). Foreign exchange reserves in industrial countries, after declining for two consecutive years, increased by SDR 15 billion during 1983. The experiences of the various countries in the group, however, were not uniform. The largest expansions in foreign exchange reserves occurred in Italy (SDR 6.3 billion) and France (SDR 4.0 billion), as a result of significant improvements in their current and capital account balances following the devaluation of their currencies in the realignment of central rates in the European Monetary System in March 1983. On the other hand, the largest contraction of foreign exchange reserves occurred in the United States (SDR 3.3 billion). For oil exporting countries, movements in foreign exchange reserves have been closely associated with changes in their oil export earnings. Following the oil price increases of 1979-80, oil export earnings increased sharply from 1978 to 1980, but then declined in the three subsequent years because of falling world oil consumption and oil prices. As a result, the current account surplus of oil exporting countries fell from $111 billion in 1980 to $53 billion in 1981, and turned into deficits in 1982 ($12 billion) and 1983 ($16 billion). At the same time, foreign exchange reserves, which had grown by almost SDR 27 billion during 1979-81, fell by nearly SDR 12 billion during 1982-83.

Non-oil developing countries increased their holdings of foreign exchange reserves by SDR 13 billion in 1983. This accumulation of reserves during a period of limited access to international financial markets was brought about in part by significant improvements in their current account positions. Since the various groups of non-oil developing countries faced different external financing constraints and undertook diverse adjustment programs, their experiences regarding the accumulation of foreign exchange reserves also differed. Many Asian countries retained access to international financial markets and reduced their current account deficits in 1983, and they accumulated almost SDR 10 billion of foreign exchange reserves, with China expanding its holdings significantly. In contrast, Middle Eastern countries, whose current account deficit increased from $9.3 billion in 1982 to $12 billion in 1983, saw their holdings of foreign exchange reserves decline by SDR 0.8 billion. African, European, and Western Hemisphere countries, which on the whole improved their current account balance in 1983, accumulated some modest amounts of foreign exchange reserves. African countries increased their holdings by SDR 0.6 billion, European developing countries by SDR 1.1 billion, and Western Hemisphere countries by SDR 1.9 billion, with a large increase in Mexico and significant declines in Argentina and Colombia.

At the end of 1983, 54 percent of foreign exchange reserves were held by the industrial countries, 18 percent by the oil exporting countries, and 28 percent by the non-oil developing countries. In 1973, in contrast, the shares were 64 percent for the industrial countries, 10 percent for the oil exporting countries, and 26 percent for the non-oil developing countries. This redistribution of foreign exchange reserves, essentially from the industrial countries to the oil exporting countries, did not occur at a uniform pace throughout the decade. While the share of the oil exporting countries increased by 17 percentage points in 1974, it declined by 9 percentage points in 1978. Although variations in the shares of the three major groups have been less pronounced since 1978, the share of the oil exporting countries in foreign exchange reserves declined by 5 percentage points in the past two years.

Holdings of Fund-Related Reserve Assets

In 1983, holdings of Fund-related reserve assets increased by 24 percent to SDR 54 billion, but this development encompassed sharply contrasting behavior for reserve positions in the Fund and holdings of SDRs. While reserve positions in the Fund increased by SDR 14 billion, SDR holdings by member countries declined by SDR 3.3 billion. These changes were associated with the increase in quotas under the Eighth General Review of Quotas. Since member countries paid 22 percent of their quota increase in 1983 in SDRs, their holdings of SDRs declined and their reserve positions in the Fund increased simultaneously. These SDR payments were made in December 1983 and amounted to SDR 6.0 billion. There was no SDR allocation during 1982 or 1983.

Fund-related reserve assets increased in 1983 for all country groups, but the larger holdings of industrial countries and oil exporting countries accounted for most of the total increase of SDR 10 billion. Industrial countries increased their holdings by SDR 6.0 billion, with the United States experiencing the largest individual accumulation, SDR 4.2 billion, mainly as a result of the large purchases of U.S. dollars from the Fund by member countries. Oil exporting countries increased their holdings by SDR 4.0 billion. Saudi Arabia raised its holdings by SDR 4.2 billion, largely as a result of additional Fund borrowing from that country. Non-oil developing countries added SDR 0.4 billion to their holdings. At the end of 1983, industrial countries held 69 percent of total Fund-related reserve assets, oil exporting countries held 24 percent, and non-oil developing countries held the remaining 7 percent.


In 1983, the movements in the market value of gold reserves reflected a small change in the physical amount of official holdings of gold, accompanied by a relatively large change in its market price. Throughout the period since 1973, physical holdings of gold reserves have remained quite constant, apart from a decline by 9 percent in 1979. This decline was largely associated with the deposits by the members of the European Monetary System of 20 percent of their gold holdings with the European Monetary Cooperation Fund in exchange for European currency units (ECUs). The distribution of gold holdings among the various country groups has also remained fairly constant. At the end of 1983, the total physical amount of gold held as international reserves was 946 million ounces, of which industrial countries held 83 percent, oil exporting countries 5 percent, and non-oil developing countries 12 percent.

The market price of gold has shown large changes during the past decade, ranging from SDR 90 an ounce at one point in August 1976 to SDR 639 an ounce at one point in January 1980. Given the rather fixed physical stock of gold reserves, fluctuations in the market value of official gold holdings mirror changes in the market price of gold. During 1983, a decline of 12 percent in the market price of gold reduced the market value of official gold holdings to SDR 345 billion.

The share of gold valued at market prices in total international reserves has been strongly influenced by the sharp fluctuations in the market price of gold during the past decade. The share declined from 52 percent in 1974 to 38 percent in 1977, and then increased to 58 percent in 1980. The decline in gold prices in 1983, together with the increase in non-gold reserves, reduced the share of gold in total international reserves from 54 percent in 1982 to 49 percent in 1983. In that year, the share of total reserves held by the industrial countries was 58 percent, the share of oil exporting countries equaled 19 percent, and the share of the non-oil developing countries was 32 percent.

Developments in First Quarter of 1984

In the first quarter of 1984, non-gold reserves rose by SDR 1.5 billion. There was a small increase (SDR 0.4 billion) in the foreign exchange component and a somewhat larger rise (SDR 1.1 billion) in the holdings of Fund-related reserve assets, with holdings of SDRs growing by SDR 0.2 billion and reserve positions in the Fund by SDR 0.9 billion. The increase in non-gold reserves during the first quarter of 1984 was not shared by all major country groups. While the holdings of oil exporting countries fell by SDR 3.4 billion, continuing the decline first evident in 1982, the industrial countries and the non-oil developing countries increased their holdings by SDR 2.9 billion and SDR 1.9 billion, respectively.

Since the market value of official gold holdings also increased slightly during the first quarter of 1984 as a result of a small increase in the market price of gold, total reserve holdings rose by SDR 2.1 billion during the same period. The total physical amount of official gold holdings and the distribution among the major country groups remained constant.

Currency Composition and Sources of Foreign Exchange Reserves

This section examines the currency composition of foreign exchange reserves, the effects of exchange rate movements and market transactions on official holdings of major currencies, and the sources of foreign exchange reserves.

Currency Composition

The growing diversification of currency reserves that was evident in the period 1977-80 has not continued. Since the diversification of the late 1970s was in part undertaken to minimize the risk of large capital losses owing to the depreciation of the U.S. dollar, the sharp appreciation of the U.S. dollar and high real returns on U.S. dollar assets that emerged in the early 1980s increased the attractiveness of U.S. dollar reserve assets and limited the incentives for further diversification (Table 12). When the SDR value of ECUs issued against gold is not counted as part of foreign exchange reserves, the share of U.S. dollar assets (including ECUs issued against dollars) in total official holdings of foreign exchange declined from 78 percent in 1977 to 67 percent in 1980 but then recovered to 69 percent in 1983. A slightly different picture emerges if ECUs, which were introduced in 1979 and accounted for 14 percent of total foreign exchange reserves in 1983, are treated separately. Under this alternative treatment of ECUs, the share of the U.S. dollar declined from 78 percent in 1977 to 55 percent in 1980. It then increased to 58 percent in 1981 and remained stable at that level during 1982 and 1983.

Table 12.Share of National Currencies in Total Identified Official Holdings of Foreign Exchange, End of Selected Years 1976-831(In percent)
ECUs Treated
All countries
U.S. dollar76.577.975.672.866.769.468.569.158.8
Pound sterling1.
Deutsche mark9.
French franc1.
Swiss franc2.
Netherlands guilder0.
Japanese yen2.
Unspecified currencies5.
Industrial countries
U.S. dollar87.
Pound sterling0.
Deutsche mark6.
French franc0.
Swiss franc1.
Netherlands guilder0.
Japanese yen1.
Unspecified currencies1.
Developing countries3
U.S. dollar68.867.862.362.756.963.564.166.566.5
Pound sterling2.
Deutsche mark11.412.814.915.615.813.813.111.311.3
French franc2.
Swiss franc3.
Netherlands guilder1.
Japanese yen2.
Unspecified currencies7.
Sources: Various Fund publications and Fund staff estimates.

The detail in each of the columns may not add to 100 because of rounding. Starting with 1979, the SDR value of European currency units (ECUs) issued against U.S. dollars is added to the SDR value of U.S. dollars, but the SDR value of ECUs issued against gold is excluded from the total distributed here. For classification of countries in groups shown here, see Chapter 1, Table 2 and Table 3, footnote 1.

This column is for comparison and indicates the currency composition of reserves when holdings of ECUs are treated as a separate reserve asset, unlike the earlier columns starting with 1979 as is explained in the preceding footnote. The share of ECUs, amounting to 13.5 percent for the total and 25.0 percent for the industrial countries, respectively, has been added to that of unspecified currencies.

The calculations here rely to a greater extent on Fund staff estimates, especially for the oil exporting countries, than do those provided for the group of industrial countries.

The currency composition of the increase in the SDR value of foreign exchange reserves in 1983 can also be decomposed into price and quantity changes. While total foreign exchange reserves increased by SDR 22.3 billion in 1983, U.S. dollar reserves rose by SDR 14.4 billion and ECU reserves by SDR 4.1 billion (Table 13). Changes in the SDR value of official holdings of the other major currencies were much smaller. The increase in the SDR value of U.S. dollar reserves in 1983 resulted from both a larger quantity of officially held U.S. dollars (up by SDR 5.5 billion) and the effects of the appreciation of the dollar in terms of the SDR (adding SDR 8.9 billion). This contrasts with the experience of 1982, when a large drop in the quantity of officially held U.S. dollars was only partially offset by the appreciation of the U.S. dollar in terms of the SDR, resulting in a significant decline in U.S. dollar reserves.

Table 13.Currency Composition of Official Holdings of Foreign Exchange, End of 1978-End of 19831(In millions of SDRs)
U.S. dollar
Change in holdings12,211-14,0545,84112,496-5,85514,420
Quantity change23,362-12,5882612-3,200-14,8315,476
Effect of price change-11,151-1,4675,22915,6968,9768,945
Year-end value169,424155,369161,211173,706167,852182,272
Pound sterling
Change in holdings2786923,004-1,599498929
Quantity change3133732,395-6901,1851,294
Effect of price change-34319609-910-688-365
Year-end value3,7824,4747,4785,8796,3777,306
Deutsche mark
Change in holdings6,0844,1089,477-2,671-2,852367
Quantity change4,5743,01712,549-806-2,8333,153
Effect of price change1,5091,091-3,071-1,865-19-2,786
Year-end value24,55628,66438,14135,47032,61732,985
French franc
Change in holdings1784021,263-485-311-368
Quantity change503201,59215581160
Effect of price change12882-329-640-392-528
Year-end value2,7053,1074,3703,8853,5743,206
Swiss franc
Change in holdings2759642,088-611-522-389
Quantity change-3468772,631-1,086-141-144
Effect of price change62287-542475-380-245
Year-end value5,1566,1208,2097,5987,0766,687
Netherlands guilder
Change in holdings293380899-159-502-353
Quantity change1453291,1279-470-109
Effect of price change14851-227-168-33-244
Year-end value2,0162,3963,2963,1372,6352,281
Japanese yen
Change in holdings2,4207802,802242105705
Quantity change1,7962,410942115257-20
Effect of price change623-1,6311,860128-152725
Year-end value7,1077,88710,68910,93111,03711,742
European currency units
Change in holdings32,70614,952-4,727-5,0074,074
Quantity change327,295-1,545-2,143-1,4601,195
Effect of price change5,41116,497-2,584-3,5472,879
Year-end value32,70647,65842,93137,92541,999
Sum of the above
Change in holdings21,73925,97840,3282,486-14,44619,386
Quantity change29,89422,03420,303-7,646-18,21211,006
Effect of price change-8,1553,94320,02510,1323,7668,380
Year-end value214,745240,723281,052283,537269,092288,478
Total official holdings4
Change in holdings20,40625,68343,2685,890-11,19122,274
Year-end value224,172249,855293,123299,013287,822310,095
Source: Fund staff estimates.

The currency composition of foreign exchange is based on the Fund’s currency survey and on estimates derived mainly, but not solely, from official national reports. The numbers in this table should be regarded as estimates that are subject to adjustment as more information is received. Quantity changes are derived by multiplying the change in official holdings of each currency from the end of one quarter to the next by the average of the two SDR prices of that currency prevailing at the corresponding dates (except that the average of daily rates is used to obtain the average quarterly SDR price of the U.S. dollar). This procedure converts the change in the quantity of national currencies from own units to SDR units of account. Subtracting the SDR value of the quantity change so derived from the quarterly change in the SDR value of foreign exchange held at the end of two successive quarters and cumulating these differences yields the effect of price changes over the years shown.

Reflects mainly deposits of U.S. dollars by members of the European Monetary System (EMS) in the European Monetary Cooperation Fund.

Quantity changes in European currency units (ECUs) issued against dollars are evaluated by applying the SDR price of the U.S. dollar on the swap date to the estimated change in dollar holdings. Similarly, quantity changes in ECUs issued against gold are determined by applying the SDR price of the ECU on the swap date to the ECU price of gold used by the EMS and multiplying by the change in the number of ounces.

Include a residual whose currency composition could not be ascertained, as well as holdings of currencies other than those shown.

The increase in the SDR value of official holdings of ECUs in 1983 also reflected positive quantity and price changes. ECUs are issued by the European Monetary Cooperation Fund to the central banks of the members in exchange for the deposit of 20 percent of the gold holdings and 20 percent of the gross U.S. dollar holdings of these institutions. These swaps are renewed every three months, and changes in the members’ holdings of U.S. dollars and gold, as well as in the market price of gold and in the value of the U.S. dollar, affect the amount of ECUs outstanding. Quantity and price changes in the SDR value of ECU holdings, therefore, depend on the evolution of its two components, gold and U.S. dollars. In 1983, the positive quantity effect, SDR 1.2 billion, resulted solely from increases in the quantity of dollars deposited at the European Monetary Cooperation Fund, since the quantity of gold deposited remained virtually unchanged at 85.7 million ounces during the year. The positive price effect, SDR 2.9 billion, resulted from both the appreciation of the U.S. dollar with respect to the SDR and the increase in the swap price of gold in terms of SDRs during the year.17 The significant increase in the SDR value of ECU holdings that resulted from positive quantity and price changes in 1983 contrasts sharply with the two previous years, when negative quantity and price changes produced significant declines in the SDR value of ECU holdings.

The other major currencies have also been affected by the process of diversification of international reserves. The share of the deutsche mark, the second most important currency in official portfolios after the U.S. dollar, has declined steadily from 15.1 percent in 1980 to 11.9 percent in 1983. A similar, although less pronounced, trend is observed in the shares of the Swiss franc, which fell from 3.2 percent in 1980 to 2.4 percent in 1983, and the Netherlands guilder, which declined from 1.3 percent in 1980 to 0.8 percent in 1983. While the share of the pound sterling has shown a slightly upward trend in the past three years, the shares of the French franc and the Japanese yen have not changed significantly.

For these other major currencies, the quantity changes in the SDR value of the official holdings in 1983 was positive for the pound sterling, the deutsche mark, and the French franc and negative for the Swiss franc, the Netherlands guilder, and the Japanese yen. The price changes reflected relative exchange rate movements. While the price change was positive for holdings of Japanese yen owing to the appreciation of the yen with respect to the SDR, it was negative for all the other currencies except the U.S. dollar, reflecting their depreciation with respect to the SDR.

When all the major currencies are taken together, the change in the SDR value of foreign exchange reserves in 1983 showed a sharp contrast to the experience of the previous year. While a positive price change was outweighed by a larger negative quantity change in 1982, a positive price change of SDR 8.4 billion was reinforced by a positive quantity change of SDR 11.0 billion in 1983.

The pattern of currency diversification differs between industrial countries and developing countries (Table 12). Historically, the industrial countries have held a greater proportion of their reserves in U.S. dollars than have the developing countries. As a result, the share of reserves accounted for by the other major currencies has been larger in developing countries than in industrial countries. Although this relationship also held in 1983, there has been a different tendency between the two country groups during the past few years. While the share of the U.S. dollar in total foreign exchange reserves remained at about 77 percent in industrial countries from 1980 to 1983, it increased from 57 percent to 67 percent in developing countries during the same period. This shift to U.S. dollar reserves in developing countries was reflected in a reduction in the shares of each of the other major currencies, in particular the deutsche mark, whose share declined from about 16 percent in 1980 to 11 percent in 1983.

Expected rates of return, together with considerations of liquidity and risks, exert an important influence on the allocation of international reserves among assets denominated in different currencies. Since expected rates of return cannot be directly observed, realized rates of return in short-term investments denominated in the major reserve currencies can provide some indication of the yields on reserve holdings as well as the uncertainty that gives rise to portfolio diversification. During the past decade, the rates of return, in terms of SDRs, on investments in the five currencies that currently comprise the SDR basket diverged substantially.18 Although interest rate differentials explain part of this divergence of yields, the dominant factor has been the differences among the rates of appreciation or depreciation of each currency with respect to the SDR. For example, a U.S. dollar-denominated investment would have produced the lowest rate of return among investments denominated in the different currencies in 1978, when there was a large depreciation of the U.S. dollar with respect to the SDR; and it would have produced the highest rate of return in 1981 and 1982, when the U.S. dollar appreciated substantially with respect to the SDR. In 1983, an investment denominated in Japanese yen had the highest rate of return (22 percent) and the highest rate of appreciation with respect to the SDR. In contrast, investments denominated in French francs and pounds sterling had the lowest rate of return (1 percent) and the highest rates of depreciation with respect to the SDR. At the same time, the U.S. dollar appreciated with respect to the SDR and produced a rate of return of 12 percent on U.S. dollar investments, while the deutsche mark depreciated with respect to the SDR and produced a rate of return of 2 percent on deutsche mark investments. The uncertainty regarding the realized rate of returns, which reflects mainly the size and duration of unexpected exchange rate changes, provides incentives for the diversification of international reserves among investments denominated in different currencies.

Placement of Foreign Exchange Reserves

Increased official claims on residents of the United States and ECUs accounted for much of the growth of foreign exchange reserves in 1983 (Table 14). While foreign exchange reserves increased by SDR 22 billion, official claims on residents of the United States grew by SDR 12 billion and ECUs by SDR 5 billion. Other holdings of foreign exchange reserves expanded by more modest amounts. Official claims on residents of countries other than the United States in the debtor’s own currency grew by SDR 2 billion, and official Eurocurrency deposits grew by SDR 3 billion. Since official Eurodollar deposits did not change from the previous year, the entire increase in official Eurocurrency deposits is accounted for by official deposits in other currencies.

Table 14.Placement of Official Holdings of Foreign Exchange Reserves, End of Year 1976-831(In billions of SDRs)
Liabilities of residents of the United States to foreign official institutions79104120109123139149163
Items not included in reported official U.S. dollar holdings2-8-10-7-13-22-35-50-52
Reported official U.S. dollar claims on residents of the United States71941139610110499111
Reported official claims on residents of other countries denominated in the debtor’s own currency1719273041393840
Identified official holdings of Eurocurrencies
Other currencies1319212534333033
European currency units3348433843
Total official holdings of foreign exchange162204224250293299288310
Sources: International Monetary Fund, International Financial Statistics; U.S. Treasury Department, Bulletin; and Fund staff estimates.

Official foreign exchange reserves of Fund members and certain other countries and areas, including Switzerland. Beginning in April 1978, Saudi Arabian holdings exclude the foreign exchange cover against a note issue, which amounted to SDR 4.3 billion at the end of March 1978.

Mainly U.S. dollars deposited with the European Monetary Cooperation Fund in connection with the issuance of European currency units, U.S. obligations to official institutions in countries not reporting to the International Monetary Fund, and U.S. obligations that are not classified as foreign exchange reserves in the reports provided to the International Monetary Fund by the holders.

Part of this residual occurs because some member countries do not classify all the foreign exchange claims that they report to the Fund. Includes identified official claims on the International Bank for Reconstruction and Development, on the International Development Association, and the statistical discrepancy.

Over the past four years there has been no significant change in the composition of the holdings of foreign exchange reserves. At the end of 1983, official claims on residents of the United States accounted for 36 percent of total foreign exchange reserves, official claims on residents of other countries for 13 percent, official Eurocurrency deposits for 29 percent, and ECUs for 14 percent. Official Eurodollars amounted to 63 percent of total official Eurocurrency deposits.

The holdings of the various reserve currencies have been influenced not only by official portfolio preferences and foreign exchange market intervention but also by external payments developments and changes in the access of countries to international financial markets. The U.S. dollar has remained the principal intervention currency for most authorities. The other reserve currencies have served much more as investments, hedges against exchange risk, or working balances. In this regard, the relatively large decline in the quantity of U.S. dollar holdings in 1981 and 1982 reflected to some degree the foreign exchange market intervention undertaken by the authorities in a number of industrial countries intended to limit the depreciations of their currencies relative to the U.S. dollar.

Changes in official holdings of U.S. dollars have at times paralleled movements in the U.S. current account balance. For example, the U.S. current account deficits of $12 billion in both 1977 and 1978 were accompanied by a large increase in official holdings of U.S. dollars, even though the share of U.S. dollars in total holdings of foreign exchange reserves fell from 78 percent to 76 percent as a result of a depreciation of the U.S. dollar relative to the SDR. A U.S. current account surplus in 1979 was associated with a decline in official claims on residents of the United States. Continued surpluses in 1980 and 1981, however, were accompanied by rising official U.S. dollar claims. In 1982, the U.S. current account deficit was accompanied by larger liabilities of U.S. residents to foreign official institutions but a decline in reported official U.S. dollar claims on residents in the United States. This divergent behavior reflected, in part, the foreign exchange market intervention undertaken in the industrial countries and the extensive use of foreign exchange reserves by many developing countries to meet external payments when new credits could not be readily obtained from international financial markets. In 1983, however, the larger U.S. current account deficit ($39 billion) was accompanied by an increase in official holdings of U.S. dollar claims and a larger share of U.S. dollars in total foreign exchange reserves.

The Role of International Financial Markets in the Provision of Liquidity and the Adjustment Process

Throughout the 1970s and early 1980s, international financial markets came to play an increasingly important role in the provision of international liquidity and the financing of the adjustment process. This section reviews these developments since 1973, but places emphasis on the period 1982-83. Although a major proportion of international lending involves loans to borrowers in industrial countries, the discussion focuses on lending to developing countries in order to examine the relation between current account imbalances, debt service commitments, conditions in the world economy, and disturbances in financial markets.

Since mid-1982, activity in international financial markets has been strongly affected by the external payments difficulties of a number of developing countries. During the period January 1983 to April 1984, 17 developing members of the Fund underwent multilateral debt restructurings involving an estimated $94 billion in bank debt. In addition, there were official multilateral debt renegotiations to reschedule payments on debt owed to, or guaranteed by, the governments or appropriate agencies of creditor countries. Despite these difficulties, an unprecedented degree of cooperation among international banks, the developing country debtors, and national and international agencies made possible an orderly balance of payments adjustment process based on a case-by-case approach involving the implementation of a Fund stand-by program, the restructuring of amounts in arrears and of debt falling due, and, in some important instances, the arrangement of new credits from international banks. The more limited access to international financial markets by many developing countries has nonetheless affected the availability of international liquidity. In the late 1970s and early 1980s, access to the international banking markets allowed many countries to economize on holdings of actual reserves, and a number of countries increased their international liquidity through establishing credit lines that would allow them to borrow reserves when needed. However, events since August 1982 have shown that the amounts of borrowed reserves available to individual countries, as well as to groups of countries, could be subject to abrupt change. Thus, even though actual reserves of developing countries increased in 1983, their more limited access to the international markets may well have resulted in a decline in the overall international liquidity available to this group. Since the disturbances in international financial markets in 1982 and 1983 are intimately related to economic and financial developments in earlier years, this section first reviews the experience in the years 1973-81 before considering developments in 1982 and 1983.

Developments in 1973-81

During the period 1973-81, the role of capital markets in the international adjustment process expanded sharply as private lending became an increasingly important source of financing for large external payments imbalances. The first major increase in private international lending occurred in the period 1973-75, which was characterized by the need to recycle the current account surpluses of the major oil exporting countries that had grown from $7 billion in 1973 to $69 billion in 1974 before declining to $35 billion in 1975. During this period, the current account deficit of the non-oil developing countries grew from $11 billion in 1973 to $37 billion in 1974 and to $46 billion in 1975. Despite these current account deficits, this group of countries was nonetheless able to increase its holdings of non-gold reserves by more than $10 billion over the years 1973-76, in part owing to the sharp inflow of private credits. As the share of net external borrowing from private sources relative to the aggregate of current account deficits and reserve accumulations of the non-oil developing countries rose from 27 percent in 1973 to 51 percent in 1975, non-oil developing countries almost tripled their external debt from $130 billion to $337 billion. This large-scale borrowing, which in part constituted an attempt to produce a more gradual adjustment to the sharp changes in international price structures, was facilitated by the low average real cost of borrowing19 of ½ of 1 percent per annum during 1974-78. In some of these countries, however, the borrowing was used to finance not only an adjustment to external developments but also extensive investment and government spending programs.

While there were a number of sources of funds to finance this increased lending, deposits from oil exporting countries with international banks were especially important. Such deposits increased by $30 billion in 1974 and by about $11-13 billion in each of the following three years.

Movements in external payments imbalances in the period immediately following the second major increase in the price of oil in 1979-80 were similar to those experienced in the period after 1974. The current account deficits of the non-oil developing countries rose from $42 billion in 1978 to $88 billion in 1980 and $109 billion in 1981. The holdings of non-gold reserves by non-oil developing countries nonetheless rose by nearly $21 billion, as international bank lending again became a major source of financing balance of payments deficits and reserve accumulation. The external debt of non-oil developing countries grew from $395 billion in 1979 to $633 billion in 1982 and $669 billion in 1983.

Although the pattern of payments imbalances and private capital inflows of the early 1980s was similar to that of the middle 1970s, there were differences in macroeconomic policies pursued by the industrial and developing countries, as well as in general economic conditions in the world economy. While monetary and fiscal policies in the mid-1970s had been directed in a number of countries toward promoting a recovery from the 1975 recession, the presence of high and rapidly rising inflation rates in 1979-80 resulted in a determined effort to control inflation in many industrial countries. Inflation (as measured by the change in GNP deflators) in the industrial countries declined from 9 percent in 1980 and 1981 to 5 percent in 1983. The rate of growth of output also fell sharply in both industrial and developing countries. One aspect of the response of the financial markets to these changes in policies and economic developments was the emergence of high real rates of interest, reaching an average of 7 percent in 1981 and 1982.

These developments had a pronounced effect on the relation between financial markets and major borrowers among the developing countries. In the course of the 1970s, the composition of the debt of many developing countries had changed toward liabilities to private lenders based on commercial terms and toward shorter maturities. These changes in the composition of their external debt made the debt-servicing capacity of developing countries more vulnerable to rising interest rates and falling exports. The shift in the composition of the external liabilities toward debt obtained on commercial terms meant that the developing countries experienced a rapid increase in their debt service as interest rates rose during 1979-81. In addition, the shortening of the maturity of bank debt also implied that changing creditor perceptions of lending risks could quickly create major refinancing problems.

These developments were reflected in the rise of the ratio of external debt to GDP of non-oil developing countries from 122 percent in 1981 to 144 percent in 1982 and 150 percent in 1983, as well as in the increase in the external debt service ratio20 from 27 percent in 1981 to 32 percent in 1982 and 37 percent in 1983; both of these ratios had remained virtually unchanged from 1977 to 1980. As a result of the high real cost of bank loans, worsening terms of trade, and shrinking export volumes, borrowing by many non-oil developing countries continued to expand in 1981 and 1982.

Developments in 1982-83

The emergence of external payments problems for a number of developing countries during the second half of 1982 reduced sharply and rapidly the access of many member countries to international financial markets. National monetary authorities, international agencies, debtor countries, and creditor banks coordinated their efforts to prevent a widening of the crisis and a severe and possibly prolonged interruption of international credit and trade. Since the approach to these problems was to deal separately with external payments difficulties of individual countries, adjustment programs supported by the use of the Fund’s resources have become a central element in the current approach to the resolution of debt crises.

While there were four debt restructuring agreements21 with commercial banks involving an estimated total of $1.6 billion in 1982, there were 15 agreements involving 17 countries (including two nonmembers) for an estimated $67 billion in 1983. In addition, there were also six debt reschedulings through the Paris Club in 1982 and 16 in 1983, involving total rescheduled amounts of $629 million and $11 billion, respectively. The terms on rescheduled debt varied, but there are some broad similarities. Bank debt rescheduling usually covered from 80 to 100 percent of principal falling due on medium-term debt that had not previously been rescheduled. The grace period on rescheduled debt extended from two to three years and the new maturities averaged between eight and nine years. The interest spreads above the London interbank offered rate (LIBOR) on rescheduled debt ranged from 1¾ to 2¼ percent, and rescheduling fees averaged to 1 to 1½ percent of the amounts restructured. Generally, the best terms were received by countries undertaking strong adjustment programs. In almost half of the reschedulings, short-term debt was also restructured. The treatment of interbank deposits generally involved the greatest difficulties and, in some cases, restoring such deposits was not entirely successful. Several recent debt renegotiations were complicated by the existence of substantial external obligations of the private sector, often without public guarantee. The lack of foreign exchange forced some governments to assume, and then to reschedule, the obligations of private debtors, which were able to service debt in local currency but unable to obtain the requisite amount of foreign exchange. Multilateral rescheduling through the Paris Club of debt owed to or guaranteed by official creditors also required, as a precondition, agreement between the debtor country and the Fund on the use of Fund resources in the upper credit tranches.

While banks have generally been unwilling to reschedule future interest payments and almost always have insisted on the elimination of interest arrears, they have in several instances, including most of the large borrowers, committed new financing in advance of scheduled interest payments. Such commitments for additional financing totaled $15 billion in 1983.

Debt restructuring in 1983 had a major impact on the debt service payments of non-oil developing countries. These payments were reduced by $8 billion in 1982 and by $19 billion in 1983—by 2 percent and 4 percent of exports, respectively. Furthermore, the restructuring of debt resulted in a decline of short-term debt by $20 billion in 1983. The reserve holdings of many developing countries were also sharply affected by reduced access to international financial markets and the need to service external payments. For the first time since 1978, the non-gold reserves of oil exporting countries declined in 1982. Although the reserves of non-oil developing countries grew marginally in 1982, most of this increase reflected larger holdings by certain Asian countries, with other regional groups (especially in the Western Hemisphere) showing sharp declines.

The adjustment efforts of many developing countries and the decreased willingness of the banks to extend credit to certain countries were reflected in a relatively small increase in the total debt of developing countries, from $725 billion in 1982 to $768 billion in 1983.

International bank lending declined from $162 billion in 1982 to $138 billion in 1983 (Table 15). Lending to borrowers in industrial countries, which continued to account for the bulk of this cross-border lending (68 percent in 1983), declined from $117 billion in 1982 to $95 billion in 1983, reflecting low private credit demands and greater reliance on bond issuance rather than bank loans. New lending to non-oil developing countries fell from $41 billion in 1982 to $27 billion in 1983. More than half of the growth of banks’ claims on non-oil developing countries in 1983 took the form of coordinated lending to four Latin American countries and Yugoslavia in conjunction with bank debt restructurings and Fund-supported programs; lending to Mexico alone accounted for over $4 billion. Loans to oil exporting developing countries were $5 billion in 1983, compared with $9 billion in 1982.

Table 15.Net External Bank Lending and Deposit Taking, 1982-831(In billions of U.S. dollars)
Borrowing by2
Industrial countries615698611795
Offshore centers3419-1132312
Developing countries32189235032
Oil exporting54-2795
Deposit taking from5
Industrial countries5659158311598
Offshore centers31297192126
Developing countries564191123
Oil exporting-3-43-3-1
Change in net liabilities of6
Industrial countries5-3-632-3
Offshore centers3-810-8-62-14
Developing countries271254399
Oil exporting8424126
Sources: International Monetary Fund, International Financial Statistics; and staff estimates.

Data on lending and deposit taking are derived from stock data on liabilities and assets, including a correction for valuation changes owing to exchange rate movements. For classification of countries in groups shown here, see Chapter 1, Table 2 and Table 3, footnote 1.

As measured by differences in the liabilities of borrowing countries, defined as cross-border interbank accounts by residence of borrowing bank plus international bank credits to nonbanks by residence of borrower.

Consisting of the Bahamas, Bahrain, the Cayman Islands, Hong Kong, the Netherlands Antilles, Panama, and Singapore.

Including centrally planned economies that are not Fund members and certain international organizations.

As measured by differences in the assets of depositing countries, defined as cross-border interbank accounts by residence of lending bank plus international bank deposits of nonbanks by residence of depositor.

Lending to minus deposits taken from.

These reduced lending flows reflected not only the adjustment efforts of developing countries but also the attempts of international banks to improve the risk-adjusted return on their assets. In part, this has involved more comprehensive consideration of country risk factors and some emphasis on directing new lending toward lower-risk activities such as bond underwriting and trade financing rather than large syndicated loans. However, in 1983, lending opportunities in many markets were limited because of reduced external payments imbalances as well as low levels of economic activity. As a result, the international capital market became segmented, with little or no expansion in new spontaneous lending to many developing countries, while many borrowers from industrial countries and a limited number of developing countries obtained credits on highly competitive terms.

The increased segmenting of borrowers according to perceived lending risks also contributed to the historically high level of activity in the international bond markets in 1982 and 1983. Although total international bond issues reached $76 billion in 1982 and $77 billion in 1983, few developing countries had access to these markets; and bond issues of developing countries fell by $1.4 billion from 1982 to 1983. Increased bond placements have generally been confined to high-quality borrowers from industrial countries and selected European and Asian developing countries. Floating rate notes issued in large volumes by the most creditworthy borrowers were considered by many— but not all—financial institutions as more liquid and less risky securities. Issues of floating rate notes rose from $15 billion in 1982 to $20 billion in 1983, when they constituted 25 percent of total international bond issues. International banks have often been willing to absorb floating rate notes at a lower yield than syndicated loans in order to improve the risk and liquidity features of their balance sheets.

The sources of funds for this international bank lending also changed in 1983. New deposits from industrial countries declined from $115 billion in 1982 to $98 billion in 1983, with banks from the United States reducing their new deposits from $80 billion in 1982 to $34 billion in 1983. In contrast, new deposits from both oil exporting and non-oil developing countries increased, reflecting in part a buildup of gross official reserves.

As a result of these developments in lending and deposit taking in 1983, private and public agents in industrial countries more than doubled their net borrowing from the international banking system (Table 15); and agents in the United States switched from being net lenders of $34 billion in 1982 to being net borrowers of $8 billion in 1983. The net flow to non-oil developing countries from banks in the rest of the world declined sharply to small amounts.

Adequacy of Reserves and the Role of the Fund

The recent changes in reserve holdings and capital market conditions described in the two preceding sections are important factors that affect the adequacy of the stocks of international reserves and liquidity. This section considers the impact of these changes and other macroeconomic developments and also examines the role of the Fund in the provision of international liquidity.

Adequacy of International Reserves

An adequate stock of reserves is one that is consistent with the smooth functioning of the international monetary system, an expansion of world trade, and the absence of persistent inflation or deflation. An adequate level of reserves may be larger or smaller than the desired holdings of reserves under a particular set of asset prices and economic conditions. For example, an effective demand for reserves that reflects a depressed level of international trade and high interest costs could be considerably smaller than the amount of reserves that would be held under more favorable conditions. In assessing the adequacy of the stock of international reserves, current economic conditions and, in particular, the factors currently influencing the supply of reserves must be evaluated. The effective demand for reserves is influenced by such factors as the level and variability of trade flows, the willingness of the authorities to adopt timely stabilization programs aimed at reducing payments imbalances, the speed with which external imbalances respond to policy changes, the opportunity cost of holding reserves, the exchange rate regime, the size and nature of the domestic and external shocks that affect a country, and the extent to which a country has access to international capital markets. Empirical studies of the demand for reserves in industrial and developing countries suggest that the reserve demands in the world as a whole and in each of these two country groups were reasonably stable during the 1970s. Although it had been anticipated that there would be a significant reduction in the demand for reserves in the period following the end of the Bretton Woods par value system and the move toward a greater degree of exchange rate flexibility, it has generally been found that the demand for reserves for all countries taken together has continued to grow roughly in accordance with the pattern observed earlier under the par value system. In part, this continued growth in reserve holdings reflected the greater variability of trade flows, interest rates, and other macroeconomic variables that was evident in the 1970s and early 1980s. The growth of world trade became significantly more variable in the decade 1974-83 than in earlier years. Using the standard deviation of the rate of growth of world imports as a measure of variability, the import variability increased from 10 percent in 1965-73 to 19 percent in 1974-83. This greater variability quite likely stimulated the precautionary demand for reserves.

Over the long term, total reserves have, in fact, grown approximately in proportion to the value of imports. For example, the ratio of non-gold reserves to imports for all countries has had an average value of 21 percent and has remained within the narrow range from 20 to 25 percent between 1974 and 1983; for the industrial countries, it has fluctuated between 14 and 19 percent; and for the non-oil developing countries, it has moved between 18 and 27 percent (Table 16). In contrast, the ratio of the oil exporting countries has fallen from 110 percent in 1974 to 50 percent in 1983. Apart from periods involving major changes in exchange rates or financial market conditions, these reserve ratios have also been relatively stable over longer periods. The ratio of reserves to imports for all countries had average values of 17 percent for the period 1959-63 and 20 percent during 1979-83. Thus, while the ratios of reserves to imports do not reflect all of the factors influencing the demand for reserves, they may provide an indication of likely trend movements in the effective demand for reserves.

Table 16.Ratio of Non-Gold Reserves to Imports, End of Year 1970-831(In percent)

The annual rate of imports in the fourth quarter is the divisor of the stock of reserves at year-end. For classification of countries in groups shown here, see Chapter 1, Table 2 and Table 3, footnote 1.

Even if countries adjust their actual holdings of reserves to their effective demands, there is still the issue of whether the supply of reserves is adequate for the smooth functioning of the international monetary system. The supply of reserves is affected by the monetary and fiscal policies of the reserve currency countries as well as by the state of their balance of payments, movements in the market price of gold, and changing access to international capital markets. Policies of monetary restraint in the reserve center countries will tend to slow the growth of foreign exchange reserves, both those held as direct claims on reserve centers and those placed in offshore financial markets. The monetary restraint observed in many industrial countries since the early 1980s has contributed to the slowing in the growth of international reserves.

Movements in the price of gold have appreciably affected the reserves of countries holding a large proportion of their reserves in the form of gold. While the rise in the market price of gold from 1973 to the early months of 1980 sharply raised the market value of official gold reserves, that value has receded since 1980 because of a general downward trend in the market price of gold. For the past two years, the market price of gold and the market value of official gold holdings have stabilized at a level not quite halfway between the levels at the beginning of the 1970s and a decade later. For many countries, particularly the developing economies, this influence was not very strong, since their gold holdings were relatively small.

Of much greater importance for many developing countries was their reduced access to international capital markets. In the 1970s and early 1980s, borrowing in international financial markets became an increasingly important source of international reserves. As the external payments problems of a number of major debtor countries emerged in the period after August 1982, a large number of countries were unable to obtain new credits from international financial institutions and had to use existing reserves to meet a variety of external payments. Reduced access to international financial markets may therefore have been a factor contributing to the decline in international reserves in 1982 for many countries and, even more, to the adequacy of international liquidity, which reflects access to international credit as well as actual reserve holdings. During 1983, non-gold reserve holdings of non-oil developing countries increased sharply, reflecting, in part, significant improvements in their current account positions and sustained flows of official transfers and credits. However, a significant part of this accumulation was accounted for by a small number of Asian countries.

The impact of the sharp changes in the growth of international reserves during 1982 and 1983 on the adequacy of international reserves is subject to different interpretations. On the one hand, the decline in reserve holdings in 1982 experienced by many countries could have reflected adjustment to a lower effective demand associated with reduced values of international trade and payments and adverse financial market conditions. The higher reserve holdings for a number of non-oil developing countries since 1983 would, in turn, reflect improving economic conditions associated with the recovery of world output and the volume of trade. On the other hand, reserve holdings could have declined relative to both the effective demand for reserves and an adequate stock of reserves.

The reduced access of many developing countries to international financial markets, which has made it difficult to replenish depressed reserve holdings except through current account surpluses, may have slowed the pace of adjustment of actual reserve holdings to the level effectively demanded. This reduction in market access may have reflected changing perceptions of market participants regarding the ability of individual countries to service their external obligations over an extended period. Such credit reputations can be influenced by both external factors (e.g., high real international interest rates) and internal developments (e.g., inappropriate domestic adjustment policies). For many countries, reduced access to financial markets has raised the cost of adjusting reserve positions, as measured either by real borrowing rates or by forgone imports. When reserve holdings are constrained by effective demand, reserves will rise only as the effective demand for reserves grows with international trade and payments. By contrast, in a situation where supply constraints prevent adjustment of reserve holdings to effective demand, an increase in the supply of reserves can reduce the real cost of generating reserves and thereby lead to a rise in actual holdings.

These two interpretations may be valid for different groups of countries. Countries with ready access to international credit markets and strong balance of payments positions can readily ensure the growth of their reserve holdings, which will, therefore, fully reflect their effective demand for reserves. Adverse conditions of reserve supply would neither depress the reserve holdings of these countries below the amounts they feel are needed in the conduct of their external economic relations, nor would their policies be affected in any appreciable way by changes in the conditions under which reserves are being supplied. On the other hand, countries with little or no access to international capital markets and a constrained potential for affecting their holdings of reserves through exchange rate policies are likely, in a period of growing values of international trade and settlements, to hold smaller reserves than required for the efficient conduct of their international transactions. For countries facing such a situation, the receipt of additional reserve assets may well lead to greater reserve holdings rather than additional spending. Such behavior seems evident in the efforts of many developing countries to rebuild their reserve holdings even in the presence of serious external payments problems.

The Role of the Fund in the Provision of Liquidity

Recent developments in holdings of Fund-related reserve assets—SDRs and reserve positions in the Fund—have been reviewed earlier in this chapter. While SDRs are created through allocation, members’ reserve positions in the Fund come into existence as claims on the Fund that are the counterpart of members’ subscriptions paid in other reserve assets and of credit extended to the Fund. At the end of 1983, the most important source of these Fund-related reserve assets was the credit made available to the Fund by its members, which at SDR 31 billion accounted for almost three fifths of the total. The resources lent to the Fund are used to provide temporary financial support to members carrying out balance of payments adjustment programs. The largest part of these credits to member countries is extended in conjunction with programs of economic policy agreed between the members in question and the Fund, with certain specified features of the programs being considered conditions for the continuation of the phased financial support. Access by Fund members to this type of international credit constitutes an extension of international liquidity beyond that provided by reserve holdings and access to private international credit markets. In comparison with these other sources of liquidity, the Fund’s credit is characterized by its conditionality, which gives it an important role among the assets and availabilities that constitute international liquidity.

At the end of 1983, the Fund’s resources rose as a result of an increase in Fund quotas from SDR 61 billion to SDR 90 billion.22 In addition, borrowing arrangements totaling SDR 6 billion with the Bank for International Settlements, the National Bank of Belgium, Japan, and Saudi Arabia have been concluded. These commitments are intended to supplement the regular quota resources of the Fund and make it possible in 1984 to continue providing access to Fund resources within specified limits to members with economic adjustment programs that merit support on a more substantial scale than would be permitted by quota resources alone. Under certain circumstances, the Fund may also obtain access to additional resources as a result of the enlargement of the General Arrangements to Borrow (GAB) from SDR 6.4 billion to SDR 17 billion (reflecting full participation by the Swiss National Bank) and the conclusion of a borrowing arrangement with Saudi Arabia in association with the GAB for an amount of SDR 1.5 billion.

Although recent developments in international liquidity, the world economy, and international financial markets have led to extensive discussions of the role of the SDR and the possibility of an SDR allocation, there have been no allocations since January 1, 1981. Because of the continuing growth of other non-gold reserves, the share of cumulative SDR allocations declined from 6.5 percent of non-gold reserves at the end of January 1981 to 6.0 percent at the end of March 1984. Allocations are made on the basis of proposals by the Managing Director, concurred in by the Executive Board, and approved by the Board of Governors by an 85 percent majority of the total voting power. Although the question of further SDR allocations has been kept under continuous review by the Executive Board during the past two years, it has not been possible to make a proposal for a new allocation that commands the required support of members with 85 percent of the total voting power in the Fund. At its April 1984 meeting, the Interim Committee agreed that the Executive Board should continue its examination of the issues involved before the next meeting of the Committee in September 1984, and that the Managing Director should then present a report on the outcome of the Board’s discussions.

Recent exchange rate developments are described in more detail in Chapter 1.

On December 12, 1983, the Australian authorities announced that they were abandoning the administered exchange rate system in favor of independent floating. This decision followed a period of exceptional net foreign exchange inflow that was creating substantial difficulties for the management of both domestic monetary policy and the exchange rate.

Indices of real effective exchange rates measure the evolution of a country’s prices relative to those of its trading partners, adjusted for exchange rate changes. Prices are measured by the average annual consumer price index, with indices of partner countries averaged by using import weights, and exchange rates are measured by an import-weighted index of average annual effective exchange rates.

GDP weights were used for computing these averages for major groups of developing countries (Charts 15-17), in order that overall trends for a group might reflect the relative importance of countries within the group. Nevertheless, unweighted averages are also reported, because in some cases the experience of a few large countries in a group might give a misleading impression of developments in other countries in the group. Later in this section (Charts 18-20), only unweighted averages are presented, in order to focus on the association between exchange rate developments and a special common characteristic of countries within each group.

This group consists of the 25 developing countries with the largest total external debt at the end of 1982. These are, in order of the amount of their debt, Brazil, Mexico, Argentina, Korea, Indonesia, Venezuela, Israel, India, Chile, Egypt, Yugoslavia, Turkey, Algeria, the Philippines, South Africa, Portugal, Nigeria, Thailand, Malaysia, Peru, Pakistan, Morocco, Romania, Colombia, and Hungary.

Similar developments are shown when using unweighted averages across countries in this group.

For all these regions, unweighted averages of national indices changed in the same direction as the GDP-weighted averages, although in most cases the changes were smaller.

Unweighted average of ratios of countries in each group.

Problems arising from the role of exchange rates in the management of domestic demand and from the task of managing exchange rates in a world of floating rates among major currencies are topics taken up in the following subsections.

The reference to oil exporting countries in this paragraph includes both the group of major oil exporters and that of net oil exporters among the “non-oil developing countries.”

For the purpose of this grouping, “high inflation” was defined as an average of more than 31 percent over the period 1979-83; “medium inflation” as an average of 11 to 31 percent; and “low inflation” as an average of less than 11 percent.

Out of 104 developing countries (both oil exporting and non-oil developing countries) for which data were available.

See Executive Board Decision No. 5392-(77/63) adopted April 29 1977 Selected Decisions of the International Monetary Fund and Selected Documents, Tenth Issue (Washington, 1983), pages 10-14 (hereinafter referred to as Selected Decisions).

See Executive Board Decisions Nos. 7645-(84/40) and 7646-(84/ 140), adopted March 12, 1984 (reproduced in Appendix II).

Published as Occasional Papers Nos. 29 and 30 (Washington: International Monetary Fund, June 1984).

No information is available on Democratic Kampuchea.

Although the SDR market price of gold showed a declining trend during 1983, the swap price of gold in terms of SDRs increased as a lagged response to the increase in gold price during the second half of 1982. According to the method used for valuing gold against ECUs, the ECU swap price is equal to the lower of the two prices: the average of the prices recorded daily at the two London fixings during the previous six calendar months and the average price of the two fixings on the penultimate working day of the period. For any given date, the swap price in terms of SDRs is obtained by multiplying the SDR/ECU exchange rate on that date by the ECU swap price on the latest swap date.

See Annual Report of the Executive Board for the Financial Year Ended April 30, 1982 (Washington: International Monetary Fund, 1982), page 67 (hereinafter referred to as Annual Report, 19—), for a description of the procedure followed to calculate the rates of return that are compared in this section.

The real cost of borrowing is measured as the London interbank offered rate (LIBOR) on three-month U.S. dollar deposits less the rate of change of the GDP deflator in the United States. Alternatively, if the rate of change of the export unit value of the non-oil developing countries is used as a measure of inflation, then the implicit real interest rate was a negative 2.5 percent per annum.

Ratio of interest and amortization payments to the total exports of goods and services.

Bank debt restructuring is defined here to cover either the rescheduling or the refinancing, or both, of debt service payments in arrears and/or of future service payments on the short-term and medium-term debt, or both. Rescheduling is a formal deferment of debt service payments over a period exceeding one year, with new maturities applying to the deferred amounts. Refinancing either is a straight rollover of maturing debt obligations or involves the conversion of existing debt service payments into new mediumterm loans.

See Chapter 3 for a more detailed discussion of these developments.

    Other Resources Citing This Publication