Chapter

Chapter 2 Developments in the International Monetary System

Author(s):
International Monetary Fund
Published Date:
September 1980
Share
  • ShareShare
Show Summary Details

Certain trends that have characterized the evolution of the international monetary system in recent years, particularly in the period since 1978, persisted during the period under review. In the determination of exchange rates, the monetary authorities of a number of industrial countries maintained or increased efforts to influence the course of exchange rate movements. This tendency has manifested itself, inter alia, in some increase since 1978, compared with earlier years of the period of floating exchange rates, in the level of official intervention in foreign exchange markets and in the pursuit of monetary policies tending to stabilize exchange rates. The variability of exchange rates among major currencies in 1979 appears to have declined somewhat from that observed on average in the first several years of the period of floating. The question arises whether measures by the industrial countries to influence exchange markets, as well as the greater stability of exchange rates, can on balance be expected to assist or hamper the pursuit of their macroeconomic objectives.

Surveillance over members’ exchange rate policies exercised by the Fund under Article IV of the Articles of Agreement is designed to assist members in avoiding both excessive variability of exchange rates, which can be disruptive without serving an economic purpose, and undue rigidity of rates, which can cause distortions in the economy and hinder the international adjustment process without improving economic stability. Procedures for conducting this surveillance have evolved further during the year under review, especially with respect to discussion of developments in the pattern of exchange rates between major currencies in the context of the world economic outlook. These topics are treated in the first part of this chapter.

Developments in the area of international liquidity are closely linked to countries’ exchange rate policies and, in particular, to intervention in the exchange markets. More generally, these developments are affected by the nature and severity of influences disturbing balance of payments equilibrium in the world economy, the modalities of financing external imbalances, and the functioning of the adjustment process. Preferences of central monetary institutions with respect to the form in which their reserves are held can influence not only the asset composition of global reserves but also their aggregate amount. These matters and related developments are discussed in the second part of this chapter, which also treats the Fund’s role in providing conditional liquidity and in supplementing the global supply of international reserve assets.

Exchange Rate Arrangements and Policies

Previous Annual Reports have described both the contribution made by floating exchange rates to the adjustment process and the problem of excessive exchange rate variability. As noted in the 1979 Annual Report, efforts were intensified, beginning late in 1978, to deal with excessive exchange rate movements between the major currencies. These efforts, which were in some countries related to endeavors to stem inflationary pressures, have continued through 1979 and the first quarter of 1980; they are discussed below, with special attention paid to the experience of the European Monetary System (EMS) in its initial year and to the implications of the more active efforts to prevent excessive exchange rate fluctuations for the successful pursuit of domestic economic objectives and external adjustment. A further section treats the role of exchange rate adjustments in dealing with the large payments imbalances presently prevailing. These questions are of special significance to the Fund because of its responsibilities for surveillance over exchange rate policies. The first part of the chapter concludes with a review of the Fund’s experience with surveillance.

Policies Influencing Exchange Rates Between Major Currencies

Since the onset of widespread floating of exchange rates, the industrial countries have been faced with the difficult question of the degree to which the determination of exchange rates should be left to the free play of market forces. Periods of relative stability of rates alternated with phases of instability, during which exchange rate fluctuations were perceived as excessive in relation to the adjustments required to offset divergences in underlying price and current account developments. The desire to deal effectively with occasional disorderly market conditions was strengthened by concern about the impact of apparently unjustified exchange rate movements on domestic inflation and external competitiveness. This section summarizes recent policy developments influencing exchange rates, the effect of those developments on exchange rate variability, and their implications for overall economic conditions in the countries affected.

Policy Developments

A number of developments in the world economy since 1978 have contributed to underlying strains in exchange markets. Among these developments have been continued substantial divergences among the major industrial countries in rates of inflation and economic growth, the differential impact of oil price increases on their economies, and uncertainties as to the future of monetary and fiscal policies.1 Nevertheless, the variability of exchange rates was on the whole lower in 1979 and the first quarter of 1980 than it had been in 1978. The shift in current account positions of the Federal Republic of Germany, Japan, and the United States that occurred in 1979 (see Chapter 1.) was one factor calming foreign exchange markets. The reduction in exchange rate variability may also have been related to the tendency on the part of some industrial countries to intensify efforts to influence exchange rates. These efforts were perhaps a reaction to the instability in foreign exchange markets in the latter half of 1978 and also a consequence of growing concern about the impact of foreign exchange movements on the domestic economy. They were manifested by more active intervention policies by the industrial countries and, among certain countries, a greater coordination of monetary policies in the light of developments in exchange markets.

As a result of the interplay of these developments, there were fewer sharp changes in exchange rates in 1979 than in 1978, when major exchange rate realignments took place (Chart 11). With the important exceptions of Japan and the United Kingdom, and, to a lesser extent, Switzerland, effective exchange rates tended to remain close to levels prevailing at the end of 1978 (as for Canada, France, Italy, and the United States) or to proceed along their respective historical trends (as for the Federal Republic of Germany until the latter part of 1979).

In the aftermath of the monetary measures announced on November 1, 1978, the United States pursued a policy of monetary restraint designed to deal with inflationary pressures, accompanied by active exchange market intervention to correct excessive exchange rate movements, which had contributed to these pressures. This policy, in combination with a greatly improved current account balance, led to a shift in market sentiment regarding the dollar. This sentiment, fortified by the strong impact that the oil price increases were expected to have on some of the other major industrial countries, accounted for the relative strength of the dollar in the first half of 1979.

On March 13, 1979, the EMS was formally implemented. With regard to exchange rates, its principal effect was to include France, Ireland, and Italy in a common margins arrangement with EEC members of the former “snake.”2 On the whole, exchange markets were reasonably stable until the third quarter, and there were few notable policy shifts during this period. By mid-June, the dollar came under heavy selling pressure and depreciated despite a large-scale intervention program coordinated by the authorities of the United States, the Federal Republic of Germany, and Switzerland. In July, all three of these countries increased their discount rates, maintaining restrictive monetary policies. The decline of the dollar slowed. Meanwhile, strains among the currencies of countries participating in the EMS had appeared. These strains were dissipated in September by the revaluation of the deutsche mark and the devaluation of the Danish krone.3

Chart 11.Eight Industrial Countries: Indices of Effective Exchange Rates, January 1975-June 1980

(Monthly average of daily rates, 1975= 100)1

1 The weights for calculating effective exchange rates are derived from the Fund’s multilateral exchange rate model.

Several important policy changes were announced in October. Most industrial countries raised their discount rates, and the United States initiated a policy package designed to ensure better control over the expansion of credit and money, to help curb speculation in the financial, foreign exchange, and commodity markets, and to dampen inflationary pressures. The U.S. measures included an increase of 1 percentage point in the discount rate, the imposition of a marginal reserve requirement of 8 per cent on managed liabilities (including borrowings from the Eurocurrency markets), and a shift in emphasis from the federal funds rate to the growth of bank reserves as the operating target of monetary policy. The U.S. measures elicited strong expressions of support by central banks in other industrial countries.

The first half of 1980 saw a quick rise and a subsequent decline in the exchange rate of the U.S. dollar vis-à-vis most of the other major currencies. The sharp increases in U.S. interest rates and the decline in many commodity prices, as well as the anti-inflation program announced on March 14, 1980, combined to strengthen the dollar substantially by the end of the first quarter. When U.S. interest rates turned down, while those of other industrial countries remained at higher levels or declined to a lesser extent, the dollar adjusted in an orderly manner.

The two major currencies that experienced large movements in their exchange rates were the pound sterling, which appreciated during 1979, and the Japanese yen, which depreciated. The rise in sterling can be attributed to the sharp increases in oil prices, given the United Kingdom’s near self-sufficiency in oil, as well as to the strong anti-inflationary measures adopted by the newly elected government. The Bank of England intervened fairly heavily at times in the first half of 1979 to moderate the appreciation of the pound and raised its discount rate substantially in order to contain the growth in the money supply within the official target range.

The Japanese yen experienced strong downward pressure throughout 1979. The authorities intervened very strongly to moderate the yen’s decline during the first half of 1979, over which period Japan’s foreign exchange reserves fell by $8 billion, and at the same time raised the discount rate and encouraged capital inflows into Japan. As the pressure on the yen continued, the authorities followed monetary policies that discouraged capital outflows and continued heavy intervention in exchange markets, as a result of which, foreign exchange reserves fell further by $4.5 billion in the latter half of 1979 and by $1.8 billion in the first quarter of 1980. The yen stabilized in March 1980, in part because of a new program designed to attract capital, and since mid-April has substantially appreciated. In April and May, foreign exchange reserves rose by $2.7 billion.

Except for an initial weakness early in 1979, the Canadian dollar remained stable with respect to the U.S. dollar. To some extent this stability can be ascribed to Canada’s current and anticipated future position as a net energy exporter in a period of increasing oil prices, as well as to a view that external competitiveness had been restored over the previous two years. It was, however, in part a result of actions by the authorities to prevent Canadian short-term interest rates from diverging too far from those in the United States. These actions were also in line with Canadian domestic policy objectives.

To obtain an overall view of how the authorities responded to exchange rate pressures over the last year, it is useful to examine some measures of official intervention and monetary policy. The changes in international reserves excluding gold, used as a rough measure of official intervention,4 suggest that the authorities have stepped up their activity in the foreign exchange markets since the latter part of 1978. As measured by the sum of the absolute values of monthly net changes in non-gold reserves, expressed as a percentage of current exports, total official activity in the foreign exchange market of eight industrial countries has been increasing steadily since the mid-1970s (Table 11). The increase from 1977 to 1979 was most pronounced for Canada, France, the Federal Republic of Germany, Japan, and the United States. For these countries, with the exception of France, intervention appears to have risen sharply from the earlier years of the floating period. For the three other countries, there was a decline in intervention between 1977 and 1979, which was especially marked for the United Kingdom. Nonetheless, the sharp increase in intervention by the three largest industrial countries has no doubt had a particularly significant effect on exchange markets in general.

Table 11.Eight Industrial Countries: Sum of Monthly Changes in Official Reserves Excluding Gold Relative to Exports, March 1973-April 19801(In per cent)
Mar. 1973-Dec.19762197719781979Jan.-Apr.

19802
Canada4.385.3610.397.019.46
France6.941.494.893.8819.66
Germany, Fed. Rep. of9.066.3812.4013.2811.91
Italy15.3415.7311.9311.3829.55
Japan8.228.0115.0614.907.82
Switzerland63.5572.5056.5550.5442.83
United Kingdom8.8927.498.168.8011.54
United States32.091.481.925.664.58
Eight industrial countries8.7410.0010.5510.6912.60

Monthly changes are summed without regard to sign for the periods shown. Changes in reserves are here defined as the total of changes in foreign exchange holdings, reserve positions in the Fund, and holdings of SDRs, minus changes in use of Fund credit and cumulative SDR allocations. In calculating these changes, foreign exchange holdings are valued in U.S. dollars (except for the United States, as explained in footnote 3), and all changes in the SDR-denominated reserve components are converted into U.S. dollars at average monthly exchange rates. The figures for 1979 and 1980 have been adjusted by excluding the value of ECUs issued against gold holdings of EMS members.

Annual rate

For the United States, foreign exchange holdings were valued in SDRs for the purpose of calculating changes, which were then converted into U.S. dollars at current exchange rates.

Monthly changes are summed without regard to sign for the periods shown. Changes in reserves are here defined as the total of changes in foreign exchange holdings, reserve positions in the Fund, and holdings of SDRs, minus changes in use of Fund credit and cumulative SDR allocations. In calculating these changes, foreign exchange holdings are valued in U.S. dollars (except for the United States, as explained in footnote 3), and all changes in the SDR-denominated reserve components are converted into U.S. dollars at average monthly exchange rates. The figures for 1979 and 1980 have been adjusted by excluding the value of ECUs issued against gold holdings of EMS members.

Annual rate

For the United States, foreign exchange holdings were valued in SDRs for the purpose of calculating changes, which were then converted into U.S. dollars at current exchange rates.

A comparison of changes in total non-gold reserves with changes in exchange rates for these eight countries indicates that intervention during 1978 and 1979 moved, in general, counter to the direction of market pressures (Chart 12). The extent to which this was the case was probably greater than indicated by Chart 12, which shows only net intervention for the month as a whole.

Chart 12.Eight Industrial Countries: Monthly Changes in Official Reserves Relative to Exports and in Exchange Rates, January 1978-June 1980

(In per cent)

1 For each country, the change in reserves is expressed as a percentage of average monthly exports for the period 1978-79. The change in reserves is defined in Table 11, footnote 1.

2 The exchange rate at the end of each month is measured in U.S. dollars per unit of domestic currency for all countries except the United States, for which the effective exchange rate derived from the Fund’s multilateral exchange rate model is used.

The proxy for intervention used in Table 11 has serious limitations as an indicator of the volume of official activity in exchange markets: for example, it takes account neither of intervention activity in different directions within the same month nor of official compensatory borrowing. Where figures measuring such activity are available, they indicate that the reserve changes shown in Table 11 on the whole underestimate both the volume of intervention and its recent increase. For example, gross intervention by the U.S. authorities, measured by total sales and purchases of foreign exchange, rose from approximately $19 billion in 1978 to $23 billion in 1979. These amounts were in sharp contrast to gross intervention prior to 1978, when the cumulative total from March 1973 to December 1977 was only half of the 1978 figure. By comparison, the sum of monthly changes in official reserves without regard to sign was about $3 billion in 1978 and $10 billion in 1979. Intervention statistics published by the Federal Republic of Germany indicate that total intervention by the Bundesbank was larger in 1978 and 1979, and increased by greater amounts after 1977, than reserve changes alone would indicate. Against this, U.K. data show that when compensatory official borrowing is included in the definition of intervention, total intervention was somewhat greater before 1978, and somewhat less in 1978 and thereafter, than indicated by reserve figures alone.

In addition to official intervention, monetary policies appear to have played a part in the dampening of excessive exchange fluctuations since late in 1978. There have been episodes of coordination of monetary policies since 1978, but the degree of coordination is difficult to assess, in part because coordination does not necessarily imply that monetary policies of participating countries move in the same direction, nor does a parallel movement necessarily imply intentional coordination. It is nevertheless noteworthy that there were several instances when monetary policies moved in tandem—for example, discount rate increases in most industrial countries in the latter half of 1979—and that these parallel movements tended to prevent excessive exchange rate fluctuations by limiting the size of, and changes in, interest rate differentials. Movements in the latter are indicated in Chart 13. The growth in money supply (also shown in Chart 13), which provides another measure of monetary policy, indicates similar parallel movements during the second and last quarters of 1979, when the authorities of France, the Federal Republic of Germany, Japan, Switzerland, and the United Kingdom pursued a more restrictive monetary policy. In large part, these episodes reflected a congruence of objectives, particularly with respect to the common effort to reduce inflation, and the effects of these policies on exchange markets were in considerable part a fortuitous by-product. Among the EMS participants, however, efforts to coordinate monetary policies were more explicit. For this and other reasons the experience of the EMS in its first year is worthy of special attention.

Developments Within the EMS

The EMS, which was established to create a greater measure of monetary stability within the European Community, represents a major modification of exchange arrangements and international policy coordination. It differs from the snake not only through the inclusion of France, Ireland, and Italy in a common margins arrangement but also through the introduction of innovative features, such as the establishment of the European Currency Unit (ECU), plans for the establishment of a European Monetary Fund, and the divergence indicator.5 The fact that participating countries in the EMS conduct 44 per cent of their external trade with each other (compared with 30 per cent for the members of the previous snake arrangement) implies that stable exchange rates among the participants would go far toward stabilizing their effective exchange rates.

Chart 13.Eight Industrial Countries: Short-Term Interest Rates and Changes in Narrow Money, January 1978-June l980

1 The interest rates for France, the Federal Republic of Germany, Italy, and Japan are the call money rates; for the United States, the federal funds rate; for Canada and the United Kingdom, the three-month treasury bill rates; and for Switzerland, the three-month interbank deposit rate.

2 Three-quarter moving average of quarterly percentage changes (at annual rates) in narrow money, defined as currency plus demand deposits.

The relative calm of exchange markets in the first half of 1979 facilitated the successful functioning of the EMS in its early months. Within the EMS, interest rate differentials seem to have been important in determining exchange rate movements. The French franc was a beneficiary of capital inflows, largely from the Federal Republic of Germany and partly in response to higher interest rates in France. The lira remained at the top of the band around its central rate, partly because of relatively high interest rates in Italy. When the authorities in the Federal Republic of Germany tightened monetary policy, the authorities of the other EMS currencies quickly raised their interest rates to maintain the differentials against the deutsche mark. At the same time, the authorities of all EMS countries engaged in heavy intervention in order to dampen the depreciation of their currencies against the U.S. dollar.

The technical features of the EMS were reviewed in mid-September and no changes were made, although tensions had arisen within the EMS as a result of disparities in current account positions and inflation rates among the participating countries. On September 23, 1979 the deutsche mark was revalued by 5 per cent against the Danish krone and by 2 per cent against the other participating currencies. The strains within the EMS subsided following this realignment, but on November 30, 1979, the Danish authorities announced a further devaluation of 4.76 per cent of the krone against all other participating currencies. The U.S. dollar strengthened against the deutsche mark and other currencies in the first quarter of 1980, and intervention was necessary on most days to support the Belgian franc and the Danish krone. In March 1980, the authorities of the Federal Republic of Germany intervened heavily to moderate the depreciation of the deutsche mark against the dollar in order to avoid excessive changes in the rate, in view of the expectation that the tendency for the deutsche mark to depreciate would soon be reversed, as it was, in fact, in April.

A coordinated policy vis-à-vis the U.S. dollar has broadly been followed by the EMS participants. Thus, while there has been an increase in intervention by participants using European currencies to support bilateral rates—which was relatively rare under the “snake”—intervention in U.S. dollars has continued on a large scale, mainly to reduce fluctuations of EMS currencies vis-à-vis the U.S. dollar or to carry out necessary adjustments. For example, the ECU appreciated by about 7 per cent against the U.S. dollar between mid-March 1979 and the end of the year.

As a guide to intervention policy and occasional realignments of bilateral central rates among EMS participants, the divergence indicator has to date signaled the need for adjustment only for the weaker currencies. Intervention limits were frequently reached before the divergence thresholds of the currencies concerned were breached. The realignments of EMS currencies that have taken place so far have been carried out in a smooth and orderly manner.

Variability of Exchange Rates Among Major Currencies

In assessing the impact of more active official intervention and of increased orientation of monetary policies toward exchange rate objectives, it is worthwhile to compare recent exchange rate movements with those in previous years. These movements result from closely related developments in underlying economic factors, official policies designed to influence exchange markets, and shifts in expectations in those markets. In some respects, exchange markets tend to behave like other financial markets: at times, they go through periods of unusual volatility, and it is not always evident, even with the benefit of hindsight, which economic forces have induced the increased volatility. For these reasons, shorter-term fluctuations—i.e., movements that reverse themselves within a day, a week, or a month—cannot always be distinguished, at the time of their occurrence, from the longer-term movements of exchange rates spanning several quarters or perhaps years. A large irreversible shift in the exchange rate may occur within a short period of time, and long-term readjustments are often accompanied by large short-term fluctuations.

A useful measure of short-term variability of exchange rates is the average divergence of the monthly percentage change in the effective exchange rate from the average of monthly changes during the period examined. This measure of variability suggests that for the industrial countries exchange rates were substantially more stable during 1979 than during 1978 and the average of the years 1973-77 (Table 12). For developing countries, the reduction in the variability of the effective exchange rate in 1979 paralleled that in the industrial countries, but was less marked.

Table 12.Average Monthly Variability of Effective Exchange Rates, April 1973-December 19791(In per cent)
1973-762197719781979
Industrial countries1.411.131.440.96
Canada0.871.141.401.21
France1.620.501.610.64
Germany, Fed. Rep. of1.820.981.330.64
Italy1.980.600.730.68
Japan1.361.683.132.03
Switzerland1.421.733.380.91
United Kingdom1.450.971.612.16
United States0.740.420.940.60
Other industrial countries1.411.211.220.87
Developing countries31.600.981.671.48
Pegged to U.S. dollar1.470.561.571.53
Pegged to SDR1.930.631.992.58
Pegged to French franc0.810.250.700.37
Other developing countries1.871.631.991.53

Monthly import-weighted effective exchange rates are used in these calculations. Variability is defined here as the standard deviation of monthly percentage changes in the effective exchange rate about the average percentage change during each year. Figures for groups of countries are unweighted averages of those for the countries in each group.

Annual average, with the period April-December used for 1973.

Countries are classified according to their exchange arrangements as of June 30, 1980.

Monthly import-weighted effective exchange rates are used in these calculations. Variability is defined here as the standard deviation of monthly percentage changes in the effective exchange rate about the average percentage change during each year. Figures for groups of countries are unweighted averages of those for the countries in each group.

Annual average, with the period April-December used for 1973.

Countries are classified according to their exchange arrangements as of June 30, 1980.

The experience of individual industrial countries confirms the general increase in exchange rate stability already noted (Table 12). France, the Federal Republic of Germany, Japan, Switzerland, and the United States, as well as the group of “other industrial countries,” all show a sharp reduction in the variability of exchange rates in 1979 compared with both 1978 and the earlier years of floating. The reduction from 1978 to 1979 was more moderate for Canada and Italy than it was for the other countries. Italy has maintained a remarkably steady and low level of effective exchange rate variability for the lira for the past three years, in contrast to the earlier years of floating. The variability of the effective exchange rate of the pound sterling declined during 1978, but increased again in 1979 to a level above the average for the floating period as a whole.

While the authorities in industrial countries have taken measures to limit excessive exchange rate fluctuations, the private sector has increasingly sought to reduce the impact of exchange rate variability by diversifying exchange risk in the forward exchange markets, which have become wider and more differentiated in recent years, and also in the financial markets, e.g., by hedging through trade credits in the invoice currency. In addition, there seemed at times to have been large portfolio shifts between financial and commodity markets. For example, there were occasions when sharp exchange rate movements were reflected in gold prices, as funds shifted out of one currency were placed not only in other currencies but also in nonfinancial assets, including gold.

Implications of Greater Efforts to Influence Exchange Rates

The foregoing discussion has outlined the recent tendency toward increased official efforts to influence developments in exchange markets. These efforts have been intensified since the last quarter of 1978 because of the large and disruptive exchange rate movements occurring at that time, the growing realization that the market cannot always be relied upon to perform the function of dampening fluctuations in exchange rates, and perceptions regarding the harmful effects of excessive fluctuations. The policy of smoothing such fluctuations involves measures aimed at both maintaining orderly conditions in exchange markets and counteracting the tendency for exchange rates to overshoot the mark in moving to new equilibrium levels. Such measures are complicated by the need for the authorities to judge whether market-determined movement in the exchange rate signals a longer-run change justified by underlying economic conditions rather than a temporary fluctuation based on ephemeral causes. Hence, while official intervention can assist the process of adjustment by preventing exchange rate movements unjustified by underlying factors, it may also at times prevent movements in rates that would have fostered adjustment.

Moreover, the adverse effects of excessive exchange rate fluctuations must be seen in proper perspective. In particular, they need to be compared with the harmful consequences of protracted deviations of exchange rates from their medium-term equilibrium levels. The policies pursued by different countries reflect their perceptions regarding the balance between the economic costs of excessive exchange rate fluctuations, on the one hand, and those of protracted deviations from equilibrium rates, on the other. While the consequences of such deviations are well known, the effects of excessive fluctuations are less clear and bear some discussion.

A principal reason for taking measures to stabilize exchange rates is the fear that exchange rate variations apparently unrelated to underlying economic conditions would have perverse effects on patterns of trade and production, thereby impeding desired adjustments in current accounts. After some years of experience with floating rates, the private sector might be expected to be more aware of the often temporary nature of changes in price competitiveness induced by exchange rate fluctuations and accordingly respond only to those changes in relative wages and employment opportunities that are regarded as permanent. There may nevertheless be good reason for concern that overshooting of exchange rates could lead to unnecessarily large fluctuations in prices, wages, and employment, because the initial changes in exchange rates may induce reactions in prices and wages that are subsequently not easily reversed.

Another reason for official actions to dampen exchange rate fluctuations has been the belief that such fluctuations in exchange rates have been damaging to the growth of foreign trade and investment. Although statistical studies and business surveys have yet to show clear evidence on this point, it is admittedly difficult to come to a judgment on this question. Such a judgment implies a comparison with more stable exchange rates, but greater stability might not have been achievable during this period without at times suppressing rate movements that were justified by underlying economic developments, and this in turn could have discouraged foreign trade and investment. Moreover, in view of the various contemporaneous developments that have had adverse effects on both economic growth and price stability since 1973, there is some question whether the occasional tendency toward excessive fluctuations of exchange rates has been one of the symptoms of these developments or an additional cause of slow growth and inflation.

A further reason for official actions to influence exchange markets is that market-determined movements in exchange rates may, at times, hamper anti-inflationary policies. It has, moreover, been argued that the more freely floating are exchange rates, the weaker is the political support for anti-inflationary policies. There is indeed some evidence that under the par value system the contractionary demand-management measures taken by countries with balance of payments deficits tended to be stronger than the expansionary measures taken by countries with surpluses of comparable magnitude. While this tendency alone would have produced a marked deflationary bias in the world economy, there were also many instances during that period when countries with higher than average inflation rates resorted to an increased use of exchange restrictions or exchange rate action rather than to anti-inflationary monetary and fiscal policies. The effect of floating exchange rates in this respect would appear to be a neutral one. To be sure, floating rates permit countries to maintain inflation rates above the world average without changes in the price competitiveness of their traded-goods sectors. Equally, however, market-determined exchange rates assist countries with relatively low rates of inflation in insulating their economies from the impact of higher rates of inflation in the rest of the world. Whether, under market-determined exchange rates, there is any systematic tendency for low-inflation or high-inflation countries to adjust to the average is uncertain: while low-inflation countries may at times resist appreciation of their currencies in order to avoid harming their export sectors, there is also pressure on the authorities of high-inflation countries to avoid inflation, since the depreciation of their currencies tends to reinforce domestic price increases by raising import prices. On balance, the resistance of a government to inflationary pressures appears to have much less to do with the extent of exchange rate management than with the country’s past history of inflation and unemployment, its industrial and labor union structure, and the effectiveness with which policies can be implemented through its monetary and fiscal institutions.

A related concern is that some countries might be drawn into a vicious circle of inflation and currency depreciation, based on the feedback effects of movements in exchange rates on import prices, money wages, domestic prices, and export prices, and the fact that the effects on prices and wages are more rapid than the corrective effects on the current account. The extent to which exchange rate changes lead to a vicious or virtuous circle depends on the characteristics of an economy, such as its size and openness as well as the degree of wage and price indexation that is present. There is evidence for the industrial countries suggesting that the smaller and more open is a country, the greater is the extent to which domestic price effects of a depreciation are apt eventually to offset much, if not all, of the competitive advantage gained by the depreciation. The authorities of these economies face especially great difficulties in resisting the inflationary pressures generated by a depreciation of their currencies. Nevertheless, exchange rate depreciation and domestic inflation both typically reflect an excessive rate of money expansion, and the domestic currency cannot depreciate indefinitely unless accompanied by an accommodating monetary policy that is excessively expansionary relative to that of other countries.

A further reason why large, frequent fluctuations in exchange rates may raise rates of inflation above what they would otherwise be has to do with the supposition that effective depreciation of a country’s currency tends to accelerate the domestic rate of inflation while an appreciation does not have a corresponding dampening effect on inflation because of the downward inflexibility of wages and prices. Empirical evidence shows that import prices do fall after exchange rate appreciations and that domestic prices do not seem to react in a significantly different manner to positive and negative changes in import prices. Nevertheless, money wages tend in general to be highly rigid in a downward direction. There is thus some basis for concluding that wage and price effects of depreciations and appreciations are somewhat asymmetrical. There is also an additional worry for the authorities in countries with deeply embedded inflationary expectations that those expectations would be immediately exacerbated by a depreciation of the domestic currency while they may respond more sluggishly to an appreciation.

While there are often good reasons for active official intervention in exchange markets and other policy measures to influence exchange rates, such actions can at times be at the expense of attaining domestic monetary targets. This is true both of intervention, to the extent that its monetary impact cannot readily be offset by other monetary measures, and of monetary policies employed to influence exchange rates. The benefits of measures to stabilize exchange rates must be weighed against their domestic monetary impact.

The balance between fostering orderly market conditions and allowing appropriate exchange rates to be established in the market is important, not only for the industrial countries involved but also for the rest of the world. Large exchange rate fluctuations for the major currencies unfavorably affect the conditions under which most of world trade is transacted. It is, however, in the interest of all members to allow those exchange rate movements to take place that foster international adjustment, and this concern must continue to be a prime consideration in framing exchange rate policies.

Exchange Rates and the Adjustment Mechanism

Over the last decade, exchange rate movements have contributed to required adjustments in the current account, but this contribution was subject to a number of limitations. For the industrial countries, short-run fluctuations in exchange rates have not always reflected underlying economic conditions, and this has led to fears that rate changes may often fail to transmit to the market the signals required to stimulate the desired adjustments. Moreover, when exchange rates have moved in the required direction, the impact on the current account has typically been slow in taking hold and, in the short run, even perverse, because of low short-run price elasticities of import demand and export supply.

The appropriate use of exchange rate policies to help in correcting excessive payments imbalances differs according to the prevailing world economic situation and the circumstances of the country concerned. While the aggregate current account surplus of oil exporting countries, and the corresponding deficit of the rest of the world, is not likely, for reasons discussed below, to be amenable to adjustment through changes in exchange rates, such changes may nevertheless be justified for certain individual countries. In formulating and assessing exchange rate policies, the widely varying circumstances of each country must be carefully considered. In doing so, there are somewhat different sets of factors to take into account for industrial, non-oil developing, and oil exporting countries.

Industrial Countries

Major progress was achieved in 1979 in reducing or eliminating excessive current account imbalances among the major industrial countries, and exchange rate changes appear to have played a significant role in accomplishing this change. The appreciation of the effective exchange rates, relative to inflation rate differentials, that occurred in 1977 and 1978 in the Federal Republic of Germany, Japan, and the United Kingdom seems to have been in part responsible for the subsequent decline in their current account balances. For the Federal Republic of Germany and Japan, the oil price increases also played a major role in this decline, and for all three countries current account balances were affected as well by cyclical and structural factors. The improvement in the current account of the United States appears related to the effective depreciation of the U.S. dollar in 1977 and 1978. The relative current account performance of individual industrial countries in 1979 thus appears to have been influenced by movements in exchange rates and relative prices, even though the relative cyclical evolution of domestic demand has also been a significant factor.6

While exchange rates have played an important role in reducing current account imbalances, the question arises of whether, and to what extent, measures should be taken to influence exchange rates so as to facilitate correction of prospective imbalances of industrial countries created in large part by the oil price increases in 1979. Industrial countries as a group can reduce their current account deficit, consistent with an orderly international adjustment process, only by reducing their oil imports or increasing their exports to the oil exporting countries. An important consideration in this regard is the expectation that the rate of increase of imports of oil exporting countries is likely to be smaller than it was after the 1973-74 oil price increases and that the import expansion of these countries will in large part be independent of exchange rate policies in the industrial countries. The depreciation of the currency of an individual industrial country could, nevertheless, be a helpful contribution to international adjustment if its balance of payments position were judged to be weak relative to that of other industrial countries. Such a judgment would depend, inter alia, upon the impact of relative cyclical positions on both the capital and current accounts, as well as upon the effects on domestic saving and investment of the changes in trade and capital flows induced by the new level of oil prices. In view of the importance of oil imports, judgments as to the usefulness of exchange rate movements would also have to take into account differences among the industrial countries in their capacity to exploit domestic energy resources, in the available scope for additional energy conservation, and in the ease with which funds can be attracted from oil exporting countries to finance productive investment. Such judgments would also have to be made in the light of the feasibility and desirability of reducing divergences in growth and inflation trends among countries by means of cooperative policy action.

Non-Oil Developing Countries

Difficult choices are faced by the non-oil developing countries in determining how to adjust to the increased oil prices. This adjustment effort is hampered by the slowing in the pace of industrial activity in the rest of the world, as well as by protectionist barriers to certain types of their exports to the industrial countries. As explained elsewhere in this Report, financing of payments imbalances will be required on a large scale. Nevertheless, the aim of the non-oil developing countries will be to achieve structural adjustments in their economies that permit the eventual attainment of sustainable current account deficits—by export expansion to oil exporting and industrial countries and by avoiding too rapid a rate of growth of imports from those countries. They will attempt to pursue this aim without sacrificing their fundamental objective of a satisfactory pace of economic development. For some of these countries with large external disequilibria, exchange rate policies can play an essential complementary role in this adjustment effort.

In the short run, significant expansion in the value of exports is often difficult for developing countries to achieve because, in many cases, supply elasticities at home and demand elasticities abroad are low. In the longer run, supply elasticities are larger and there is greater scope for export expansion, especially for exporters of manufactured goods. For the latter, in particular, active exchange rate policies have often proved helpful in maintaining their price competitiveness. While for individual countries that export primary products such policies may prove useful, for exporters of a particular primary product as a group, attempts to achieve rapid increases in such exports will, in most cases, simply lead to lower overall export revenues, because the price and income elasticities of demand for many primary goods tend to be low. For most non-oil developing countries, however, as for others, export expansion could be facilitated by the removal of existing trade barriers in the industrial countries.

While more rapid growth of the developing countries normally requires larger imports of investment goods, there is considerable scope for eventually reducing dependence on certain types of imports. Import substitution in the manufacturing sector can be achieved on a significant scale in a number of countries but is often limited by the dependence on essential imports of capital equipment and raw materials, as well as by the character of resource endowments. The capacity for efficient import substitution can also be hindered by price distortions arising from government policies. For countries importing foodstuffs, a gradual movement toward import substitution is often possible, since most developing countries have possibilities for substantially raising the productivity of agricultural resources. Another type of import substitution, which is of special interest in present circumstances, is the development over the long run of domestic energy resources, for which there is considerable scope in many countries. To be sure, the investment requirements for both agricultural and energy development are large and will normally entail substantial foreign assistance. Some scope may also exist for import substitution for the group of developing countries as a whole through the expansion of trade within the group, which can be facilitated by regional trade and payments arrangements.

In programs designed to achieve the types of structural adjustments just described, exchange rate action plays an important role where such adjustments require changes in relative prices facing producers and consumers or involve the elimination of exchange and trade restrictions. Exchange rate action, to be effective, needs to be accompanied by supporting policies that encourage efficient resource allocation and investment and prevent an excess of aggregate demand over supply from offsetting, through general inflation, the initial changes in the relative prices of traded and nontraded goods. Additional measures are often required to accompany exchange rate action, for instance, the removal of controls and restrictions, public investment in certain areas, or changes in the structure of taxes and subsidies.

Oil Exporting Countries

In many oil exporting countries, the process of adjustment initiated by the rise in oil revenues has been accompanied by a deterioration in their non-oil current accounts, domestic inflation, a rise in the prices of non-traded goods relative to those of traded goods, and a tendency toward corresponding changes in the structure of domestic production. These developments have been very important in bringing about balance of payments adjustment on current account. At the same time, however, they have also been regarded as interfering with the efforts of these countries to create a more diversified base of non-oil production.

The question of what is the correct exchange rate policy for the oil exporting countries to follow under these circumstances does not have a simple answer. Normally, a country with a persistent balance of payments surplus would be expected to consider an appreciation of its currency as an appropriate element in a mix of corrective policies. In the special conditions in which the oil exporting countries presently find themselves, however, such a step would have little relevance for the major decisions to be made by these countries in the coming years with respect to the rate of petroleum production, the pace of import expansion and domestic economic development, and the allocation of their foreign investments. These decisions are largely independent of exchange rate policies, since they are affected mainly by the rates of return on investment abroad, the real terms of trade between petroleum and the goods imported by oil exporting countries, and the absorptive capacity of their economies. Moreover, in some oil exporting countries, the dominance of government involvement ensures that the structure of domestic production is determined largely by government decisions without regard to market incentives. In these instances, the effects of exchange rate changes are likely to be much smaller than in most non-oil developing countries.

Nevertheless, in a number of oil exporting countries the market plays a significant role; and in these countries another objection is raised against a policy that results in an appreciation of the national currency, namely, that it discourages diversification of the economy by reducing the competitiveness of the non-oil traded goods sector. The desirability of encouraging the growth of this sector, and the correct timing of such encouragement, varies considerably among economies, depending on the projected time span and size of oil earnings, as well as on the immediate development needs of each country.

Where encouragement of the non-oil traded goods sector is desired, suitable incentives may include special government assistance to certain activities and changes in the structure of taxes and subsidies. The use of exchange rate policy to foster a more diversified non-oil sector—as has sometimes been suggested—would in general be a less effective way of encouraging this development than would policies aimed directly at reducing the growth of consumption expenditure or at augmenting the share of expenditure devoted to investment. This is especially so in countries that already have large balance of payments surpluses, since in these circumstances exchange rate depreciation tends to result in increases in wages, prices, and money supplies that offset the initial effect of the depreciation on relative prices and domestic expenditure.

Surveillance

This section provides a general review and analysis of the Fund’s first two years of experience with the principles and procedures for surveillance over the exchange rates of its member countries. The Second Amendment of the Articles of Agreement gives Fund members freedom to choose their exchange arrangements, except for a prohibition against maintenance of the value of a currency in terms of gold. At the same time, each country undertakes, under Article IV, “to collaborate with the Fund and other members to assure orderly exchange arrangements and to promote a stable system of exchange rates,” and the Fund is required to engage in firm surveillance over members’ exchange rate policies to ensure that they are consistent with this broad objective. In order to perform this function, the Fund adopted, in April 1977, a set of principles for the guidance of members’ exchange rate policies, as well as principles and procedures for Fund surveillance over these policies, which became the basis for surveillance as soon as the Second Amendment entered into force on April 1, 1978.

Principles and Procedures of Surveillance

Three principles for the guidance of members’ exchange rate policies were devised as a means of ensuring that each member country follows exchange rate policies compatible with its general obligations under Article IV:

  • A. A member shall avoid manipulating exchange rates or the international monetary system in order to prevent effective balance of payments adjustment or to gain an unfair competitive advantage over other members.

  • B. A member should intervene in the exchange market if necessary to counter disorderly conditions which may be characterized inter alia by disruptive short-term movements in the exchange value of its currency.

  • C. Members should take into account in their intervention policies the interests of other members, including those of the countries in whose currencies they intervene.7

The principles of Fund surveillance over exchange rate policies provide a list, which is not exhaustive, of developments that might indicate the need for discussions between the Fund and a member country. These developments include protracted large-scale intervention in one direction in the exchange market; an unsustainable level of official or quasi-official borrowing or lending for balance of payments purposes; various kinds of restrictions or incentives affecting current transactions or capital flows; abnormal encouragement or discouragement to capital flows through financial policies for balance of payments purposes; and exchange rate behavior that appears to be unrelated to underlying economic and financial conditions.

The procedures for surveillance oblige each member to notify the Fund promptly of any changes in its exchange arrangements and to hold regular consultations under Article IV. They require the Executive Board to engage in periodic reviews of broad developments in exchange markets and to undertake an annual review of the general implementation of surveillance. They also provide that the Managing Director is to keep in close contact with members concerning their exchange arrangements and stand ready to discuss, at the member’s initiative, any contemplated changes in these policies. If the Managing Director considers that a member’s exchange rate policies may not be in accord with the exchange rate principles, he is to initiate special discussions with the member and, if necessary, report the matter to the Executive Board.

Experience with Surveillance Procedures

In practice, the Fund’s surveillance over exchange rate policies is conducted on two distinct but complementary levels. The first, surveillance at the global level, is concerned with ascertaining whether or not there is a broad international consistency in members’ balance of payments and exchange rate policies. This concern is grounded in the Fund’s role of assistance to members in achieving international economic and financial stability. Global surveillance is carried out in the context of the regular discussions of the world economic situation and outlook that take place both in the Executive Board and at the meetings of the Interim Committee. These discussions are the primary means by which the Fund reviews major exchange rate developments affecting the world economy, as well as the consequences that arise from the policy stances of those countries that are important for the international adjustment process.

The assessments reached in these discussions are taken into account by the Fund in implementing the second aspect of surveillance, which is that of identifying, and encouraging the correction of, inappropriate exchange rates and exchange rate policies of individual member countries. This task is carried out largely within the framework of the regular Article IV consultations with members. These consultations take place annually, whenever possible, and include consideration of the observance by a member of the principles of surveillance and of its general obligations under Article IV. These consultations are completed by conclusions reached in the Executive Board, which take the form of a summing-up by the Managing Director of the views expressed by Executive Directors during the discussion of the staff report on the Article IV consultation.

In the interval between Article IV consultations, the Managing Director may initiate special discussions with a member if he considers that the member’s exchange rate policies may not be in accord with the exchange rate principles. In January 1979, in response to concern that these standard procedures might give rise to uneven treatment of members according to the type of exchange arrangement being maintained, the Executive Board decided to institute a supplemental surveillance procedure. Under the supplemental procedure, the Managing Director would initiate informally and confidentially a discussion with a member before the next regular consultation under Article IV whenever he considered that a modification of a member’s exchange arrangements or of its exchange rate policies, or the behavior of its exchange rate, may be important or may have important effects on other members. This supplemental procedure permits the Fund to look into exchange rate developments or situations of importance without the presumption that the member in question has not complied with its obligations under Article IV.

Changes in Exchange Arrangements

When the Second Amendment of the Articles of Agreement entered into force, all countries were required to notify the Fund of the exchange arrangements that they intended to apply after April 1, 1978. They were also required to notify the Fund promptly of any changes in their exchange arrangements. Since April 1978, many countries have revised their views about what constitutes an appropriate exchange rate regime. As has already been noted, several industrial countries have altered their intervention policies in a significant way, and another group has recently formed the currency area of the European Monetary System. Some adjustments in exchange arrangements and policies have also taken place in many developing countries over the past two years as their circumstances have changed. These adjustments are shown in Table 13, in which the exchange rate regime that each country adhered to on June 30, 1980 is compared with the one that was first notified to the Fund in April 1978. Since some countries have notified the Fund of adjustments on more than one occasion, column 2 of Table 13 indicates the number of times that the Fund’s classification of a country’s exchange arrangements has changed during this period. During the period under review in this Report, from the beginning of 1979 to the end of June 1980, 24 members notified the Fund of changes in their exchange arrangements resulting in a reclassification under the categories shown in Table 13. To a considerable extent, the changes in exchange arrangements that occurred during this period, as well as earlier, were motivated by the desire to prevent excessive fluctuations in members’ effective exchange rates.

Table 13.Exchange Arrangements of Member Countries on April 1, 1978 and June 30, 19801
Arrangements on June 30, 1980
Currency pegged to
Arrangements on April 1, 1978Number

of

Changes
U.S.

dollar
Pound

sterling
French

franc
Other

currency
SDROther

currency

composite
Cooperative

exchange

arrangements
Exchange rate

adjusted according

to a set of

indicators
Other

arrangements
Industrial countries
Currency pegged to
Pound sterling
Ireland1
Other currency composite
Austria
Finland
New Zealand1
Sweden
Cooperative arrangements
Belgium2
Denmark2
Germany, Fed. Rep. of2
Luxembourg2
Netherlands2
Norway1
Other arrangements
Australia
Canada
France1
Iceland
Italy1
Japan
Spain
United Kingdom
United States
Developing countries
Currency pegged to
U.S. dollar
Bahamas
Barbados
Bolivia1
Botswana1
Burundi
Costa Rica
Djibouti3
Dominica3
Dominican Republic
Ecuador
Egypt
El Salvador
Ethiopia
Ghana1
Grenada
Guatemala
Guyana
Haiti
Honduras
Indonesia1
Iraq
Jamaica2
Korea1
Lao People’s Democratic Republic
Liberia
Libyan Arab Jamahiriya
Maldives1
Nepal
Nicaragua
Oman
Pakistan
Panama
Paraguay
Romania
Rwanda
Somalia
South Africa1
St. Lucia3
St. Vincent3
Sudan
Suriname
Syrian Arab Republic
Trinidad and Tobago
Venezuela
Yemen Arab Republic
Yemen, People’s Democratic
Republic of
Pound sterling
Bangladesh1
Gambia, The
Seychelles1
Sierra Leone2
French franc
Benin
Cameroon
Central African Republic
Chad
Comoros
Congo
Gabon
Ivory Coast
Madagascar
Mali
Niger
Senegal
Togo
Upper Volta
Other currency
Equatorial Guinea2
Guinea-Bissau1
Lesotho
Solomon Islands31
Swaziland
SDR
Bahrain1
Burma
Guinea
Jordan
Kenya
Malawi
Mauritius
Säo Tome and Principe
Tanzania1
Uganda
United Arab Emirates1
Zaïre2
Zambia
Other currency composite
Algeria
Cape Verde 3
Cyprus
Fiji
India1
Kuwait
Malaysia
Malta
Mauritania
Morocco
Singapore
Thailand
Tunisia
Western Samoa1
Exchange rate adjusted according to a set of indicators
Argentina1
Brazil
Colombia
Portugal
Uruguay1
Other arrangements
Afghanistan
Chile1
Greece
Iran1
Israel
Lebanon
Mexico
Nigeria
Papua New Guinea1
Peru2
Philippines
Qatar
Saudi Arabia
Sri Lanka
Turkey
Viet Nam1
Yugoslavia

No current information on the exchange rate system of Democratic Kampuchea is available. The table also excludes the currency of the People’s Republic of China, since notification of exchange arrangements has not yet been received from the Chinese authorities.

Prior to March 13, 1979, the category “cooperative exchange arrangements” refers to the European common margins arrangement; since that date it refers to the European Monetary System.

The following countries joined the Fund after April 1, 1978: Cape Verde (November 20, 1978); Djibouti (December 29, 1978); Dominica (December 12, 1978); St. Lucia (November 15, 1979); St. Vincent (December 28, 1979); and Solomon Islands (September 23, 1978).

No current information on the exchange rate system of Democratic Kampuchea is available. The table also excludes the currency of the People’s Republic of China, since notification of exchange arrangements has not yet been received from the Chinese authorities.

Prior to March 13, 1979, the category “cooperative exchange arrangements” refers to the European common margins arrangement; since that date it refers to the European Monetary System.

The following countries joined the Fund after April 1, 1978: Cape Verde (November 20, 1978); Djibouti (December 29, 1978); Dominica (December 12, 1978); St. Lucia (November 15, 1979); St. Vincent (December 28, 1979); and Solomon Islands (September 23, 1978).

Current Issues Relating to Surveillance

A basic precept of the exchange rate system set forth in the amended Articles of Agreement is that exchange market stability should be achieved through the pursuit by member countries of stable underlying domestic economic and financial conditions. To the extent that members’ policies and performance regarding growth and inflation differ, they should allow their exchange rates to adjust in an orderly manner, rather than attempting to maintain inappropriate exchange rates through such measures as extensive official intervention, large accommodating official capital movements, or restrictions on trade or payments.

Within this general framework, the Fund’s approach to surveillance over members’ exchange rate policies is flexible and permits an evolution of principles and procedures in accordance with the accumulation of experience regarding various exchange arrangements and policies, and in response to relevant developments in the international monetary system. The experience of the last few years suggests that it is far from easy to attain a smooth and steady evolution of exchange rates when economic objectives and policies differ markedly among countries. Over this period, broad trend movements in exchange rates have generally been in the right direction in the sense that the currencies of low-inflation countries have tended, on balance, to appreciate while those of high-inflation countries have depreciated. Nevertheless, the evidence indicates that, especially during periods of high domestic instability, market forces sometimes tend to result in exchange rate movements that do not reflect underlying economic and financial developments. This has led to policy initiatives in two areas. First, there has been an increasing tendency to intervene more extensively in exchange markets and to form wider currency blocs. Second, national authorities have made a greater effort, in part through discussions within the Executive Board and the Interim Committee, to formulate the broad outlines of economic policy in the light of developments in the world economy.

A number of members have also called for the Fund to exercise firmer surveillance over the exchange rate policies of both surplus and deficit countries. The concern about rising worldwide inflationary pressures, the large fluctuations in the values of major currencies, and the existence of serious external adjustment difficulties in many countries have given rise to a widespread feeling that progress would be desirable in improving the international adjustment process and in strengthening confidence in the present exchange rate system. These concerns have led to a wide-ranging discussion by the Executive Board of the Fund’s evolving practices in this area, with the objective of enhancing the contribution of surveillance to international monetary stability. In this discussion, a number of general aspects of recent practice were noted.

In the first place, Fund surveillance activities have in many cases been concerned with situations in which unstable domestic economic conditions and policies of excessive rigidity of exchange rates led to overvalued currencies, rather than with situations in which balance of payments surpluses called for the revaluation of a pegged exchange rate. This asymmetry may be exacerbated by the current world economic situation. In the immediate future there is only limited scope for the large current account surpluses of certain oil exporting countries to be adjusted through exchange rate changes. While some part of the aggregate current account deficit of oil importing countries is borne by countries in a reasonably strong overall external position, another portion may be borne by countries with exchange rates that are out of line with underlying conditions and are maintained only through unsustainable levels of compensatory borrowing or use of trade and exchange controls, all of which have harmful economic effects both on the countries themselves and on others. For such countries, in view of the increasing urgency of adopting corrective policies as time passes, the Fund has encouraged timely adjustment through exchange rate action and supporting measures. In order to soften the deflationary impact of measures taken to restore equilibrium in deficit countries, however, the Fund must also direct surveillance at countries that are in an overall surplus position, while recognizing the limitations, already noted, of exchange rate policy in adjusting the current surpluses of some oil exporting countries.

It is also important to avoid limiting surveillance to extreme cases, since its main purpose is to prevent the emergence of severe external imbalances that would eventually make disruptive adjustment measures unavoidable. The Fund has, therefore, also focused its attention on cases of incipient imbalances that could become severe at a later stage if no corrective measures were taken. The main difficulty with this procedure is the differentiation between countries in which an overall stabilization strategy has a good prospect for success even without a change in the exchange rate and countries in which a stabilization program could not be expected to succeed without depreciation of the currency.

The Fund has also made use of its wider authority to oversee the compliance of members with their obligations to pursue domestic economic policies that will promote a stable system of exchange rates. This became necessary in many instances where the behavior of the exchange rates for a number of currencies was a source of concern as a result of domestic economic policies. The causes of “excessive variability” of exchange rates have often been wide divergences in inflation rates and the difficulties encountered by monetary authorities in controlling the growth of monetary aggregates. It is only by dealing directly with these factors that exchange rate variability can be checked.

The Fund strives for greater international coordination of domestic policies among its members to limit excessive exchange rate volatility in order to promote the steady expansion of trade and output. Coordination does not, however, mean that countries with a lower than average rate of domestic price increase should have to compromise domestic stability for the sake of limiting exchange rate movements. Furthermore, it is not always feasible or desirable to harmonize cyclical movements in activity levels in major industrial countries. In many cases, it is not so much coordination as an improvement in demand management policies that is required. In particular, it has become more crucial for member countries to abide by the obligations undertaken by them to seek stable underlying domestic economic conditions, not only for their own sake but also because unstable domestic conditions are disruptive to the exchange and trade system as a whole.

When the principles and procedures of surveillance were adopted in April 1977, it was recognized that it would not be possible to produce a comprehensive set of rules applicable to all situations that might arise in the future. Instead, the Fund chose to take a pragmatic approach that would allow the practices of surveillance gradually to emerge from the experience accumulated in dealing with individual cases. In line with these considerations, the decision establishing the principles of surveillance also specified that they should be reviewed at two-year intervals, or more frequently if the need arose. April 1, 1980 marked the end of the first period, and this review has just been completed by the Fund.

In their review, Executive Directors reiterated that because of the considerable freedom of exchange arrangements permitted under the present system, firm surveillance under Article IV is an essential element in ensuring orderly and stable international monetary relations. Furthermore, they expressed the view that surveillance must continue to be directed not solely toward the exchange rate policies of members but also toward their general economic policies and the overall consistency of these policies across countries. As regards the implementation of procedures for surveillance, Executive Directors stressed that the effectiveness of the Fund’s role depends not so much on formal or rigid procedures as on the quality and candor of the dialogue between the Fund and each of its member countries. In implementing surveillance, the Fund should therefore rely as much as possible on persuasion, rather than on prescription. The importance of confidentiality and flexibility was also emphasized.

In general, Executive Directors felt that the procedures that were established by the original decision on surveillance were adequate to the task. This decision provides a basis for the evolution of practices to deal with the kinds of exchange rate problems that have been encountered over the past two years. It recognizes that market forces alone cannot always ensure that exchange rates reflect underlying economic conditions while acknowledging the dangers of excessive exchange rate rigidity. It also reflects the view that, since exchange rate problems are too complex to justify sole reliance on statistical indicators for judging the appropriateness of countries’ exchange rate policies, the Fund should continue to adhere to a judgmental approach.

For the time being, Executive Directors concluded, effective implementation of existing procedures is more important than the introduction of additional procedures. They considered, moreover, that it is important to achieve uniform implementation of surveillance with respect to all countries—whether surplus or deficit, large or small, developed or developing. To this end the Fund should continue to monitor closely the policies of countries where exceptionally large payments imbalances occur or appear likely to occur. The Fund should also expand its efforts to monitor the policies pursued by major industrial countries and their international consistency in the context of its regular analysis and discussion of the world economic outlook. Finally, the Managing Director of the Fund should initiate supplemental surveillance procedures in a discreet and confidential manner whenever he considers it appropriate to do so.

International Reserves and Liquidity

Developments in international liquidity are intimately linked with exchange rate movements and with the monetary and exchange rate policies that countries pursue in view of their macroeconomic objectives. A country can use its reserves, together with official borrowing or lending, to finance temporary payments imbalances—deficits or surpluses—and thereby modify the exchange rate movements that would otherwise take place. The global stock of reserves is influenced in a number of ways by developments in the exchange markets and by the response of monetary authorities to temporary imbalances. First, the authorities of many countries may at times intervene in the same direction, thereby increasing or reducing aggregate holdings of the principal reserve currency. Second, the SDR prices of reserve currencies may change and cause a change in the value of foreign exchange reserves measured in SDRs. Third, movements in the rates at which interest can be earned on short-term investments in reserve currencies affect the growth of these currency holdings.

The following sections deal with the growth of international reserves and with changes in their composition. After a description of the most recent developments, factors contributing to the change in the SDR values of foreign exchange reserves and Fund-related assets are analyzed for the past seven years of floating among major currencies. This review of international reserves is extended in a later section to conditional and unconditional liquidity provided by the Fund and in Eurocurrency markets to furnish a broad basis for assessing the adequacy of international reserves. The role of the SDR is discussed in the final section of this chapter.

Recent Developments in International Reserves

Total reserves excluding gold, as reported in International Financial Statistics, increased by 11 per cent (SDR 26 billion) in 1979, compared with annual growth rates of 8 per cent in 1978 and about 19 per cent in 1977 and 1976 (Table 14). As in past years, the growth of foreign exchange reserves in 1979 accounted for most of the rise in non-gold reserves. However, in contrast to earlier years, official holdings of national currencies rose little (SDR 5 billion). Rather, the 1979 increase in foreign exchange reserves resulted in large part from the issuance, by the European Monetary Cooperation Fund, of European Currency Units (ECUs) equivalent to some SDR 20 billion against deposits of a portion of EMS members’ gold holdings. Since March 1979 the members of the EMS have deposited 20 per cent of their official holdings of gold and U.S. dollars with that Fund. The ECUs issued against gold add to the total of recorded foreign exchange reserves, while the ECUs issued against dollars change only the composition, and not the size, of such reserves. Changes in the currency composition of official foreign exchange holdings will be discussed in greater detail below. The total of Fund-related assets increased by SDR 1.3 billion. This change resulted from an increase of SDR 4.4 billion in holdings of special drawing rights (somewhat more than the annual allocation of SDR 4 billion), which was partly offset by a decline of SDR 3.1 billion in reserve positions in the Fund resulting from the reduction by SDR 2.2 billion in the Fund’s net borrowing under the oil facilities and from the repayment of an outstanding credit tranche drawing by the United Kingdom of SDR 0.8 billion. The volume of gold held in countries’ official reserves declined by almost 10 per cent, mainly as a result of the deposits of gold by members of the EMS in the European Monetary Cooperation Fund. The market value of official gold holdings nevertheless doubled between the end of 1978 and the end of 1979, since the decline in volume was much more than offset by a rise of 124 per cent in the market price (measured in SDRs).

Table 14.Official Holdings of Reserve Assets, End of Years 1973-79 and End of May 19801(In billions of SDRs)
1973197419751976197719781979May

1980
All countries
Total reserves minus gold
Fund-related assets
Reserve position in the Fund6.28.812.617.718.114.811.812.1
Special drawing rights8.88.98.88.78.18.112.516.2
Subtotal, Fund-related assets15.017.721.426.426.222.924.228.3
Foreign exchange101.6126.5137.3160.3200.3221.12246.0258.7
Total reserves excluding gold116.6144.2158.7186.7226.5244.12270.2287.0
Gold3
Quantity (millions of ounces)1,0201,0181,0171,0131,0151,0229304934
Value at London market price94.9155.1121.9117.4137.8177.3361.4381.5
Industrial countries
Total reserves minus gold
Fund-related assets
Reserve positions in the Fund4.96.27.711.812.29.67.77.8
Special drawing rights7.17.27.27.26.76.49.311.9
Subtotal, Fund-related assets12.013.314.919.118.916.017.119.6
Foreign exchange65.764.968.773.7100.0127.2135.9143.3
Total reserves excluding gold77.778.383.792.7118.9143.1153.0162.9
Gold3
Quantity (millions of ounces)874874872872881884789 4789
Value at London market price81.3133.1104.5101.2119.6153.4306.7322.1
Oil exporting countries
Total reserves minus gold
Fund-related assets
Reserve positions in the Fund0.31.94.35.45.44.43.03.1
Special drawing rights0.30.30.30.30.40.51.01.5
Subtotal, Fund-related assets0.62.24.65.85.74.94.04.6
Foreign exchange10.235.042.449.155.240.1251.057.6
Total reserves excluding gold10.837.247.154.961.045.0255.062.2
Gold3
Quantity (millions of ounces)3434353734363739
Value at London market price3.15.24.24.34.76.314.215.8
Non-oil developing countries
Total reserves minus gold
Fund-related assets
Reserve positions in the Fund0.90.70.60.50.50.91.01.2
Special drawing rights1.41.41.21.11.11.22.12.8
Subtotal, Fund-related assets2.32.11.81.61.62.13.14.0
Foreign exchange24.925.725.336.244.052.858.156.8
Total reserves excluding gold27.227.827.137.845.654.961.260.8
Gold3
Quantity (millions of ounces)1101081081019899102104
Value at London market price10.216.512.911.713.217.239.642.6
Source: International Financial Statistics.

“Fund-related assets” comprise reserve positions in the Fund and SDR holdings of all Fund members. 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 the Netherlands Antilles, Switzerland, and Fund members except the People’s Republic of China, for which data are not published. Figures for 1973 include official French claims on the European Monetary Cooperation Fund.

Beginning with April 1978, Saudi Arabian holdings of foreign exchange exclude the cover against the note issue, which amounted to SDR 4.3 billion at the end of March 1978.

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 nine member countries of the European Monetary System of 20 per cent of their gold holdings with the European Monetary Cooperation Fund. The European Currency Units (ECUs) issued in return for these deposits are shown as part of the countries’ official foreign exchange holdings.

Source: International Financial Statistics.

“Fund-related assets” comprise reserve positions in the Fund and SDR holdings of all Fund members. 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 the Netherlands Antilles, Switzerland, and Fund members except the People’s Republic of China, for which data are not published. Figures for 1973 include official French claims on the European Monetary Cooperation Fund.

Beginning with April 1978, Saudi Arabian holdings of foreign exchange exclude the cover against the note issue, which amounted to SDR 4.3 billion at the end of March 1978.

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 nine member countries of the European Monetary System of 20 per cent of their gold holdings with the European Monetary Cooperation Fund. The European Currency Units (ECUs) issued in return for these deposits are shown as part of the countries’ official foreign exchange holdings.

Total reserves excluding gold increased for all three major country groups. For the industrial countries, this total rose proportionately less (by 7 per cent) than the average for all countries, and it would have declined if a part of the gold stock of EMS member countries had not been converted into ECUs; for the major oil exporting countries, it increased at a rate twice (22 per cent) the average rate of increase; and for the non-oil developing countries it grew by little more than average (12 per cent). However, developments differed greatly among countries, particularly those in the last group, with a single country (Argentina) accounting for more than half of the SDR 6.3 billion growth in non-gold reserves for the entire group of non-oil developing countries combined. The reversal in 1979 of the sharp decline in the reserves of oil exporting countries that had occurred during the previous year resulted chiefly from the increase in petroleum prices, which strengthened the external payments position of these countries. This development, together with the relatively strong external payments position of the United States, may also explain the slower growth of reserves excluding gold in the two other country groups in 1979 compared with the preceding year. The market value of gold holdings of industrial countries rose by SDR 153 billion during 1979. The corresponding increases for oil exporting countries and non-oil developing countries were SDR 8 billion and SDR 22 billion, respectively. However, the significance of such changes is limited on account of the instability of gold prices and other factors discussed later.

The London market price of gold, which had risen steeply during the first few weeks of 1980, was little higher at the end of May than it had been at the end of 1979. While the market value of official gold holdings thus rose only moderately in the first five months of 1980, gold valuation adjustments continued to play a major role in raising official holdings of ECUs. The European Monetary Cooperation Fund values the gold deposited by member countries (20 per cent of their holdings) at the lower of (1) the average market price over the six preceding months and (2) the average market price on the penultimate working day preceding the swap period for which the ECU price of gold holdings is to be established.8 According to this formula, the price of gold used by the EMS rose from ECU 165 per ounce for the first swap period started March 13, 1979 to ECU 211 for the third swap period ended January 8, 1980. During 1979, the resulting revaluation of the stock of gold deposited, whose volume remained essentially unchanged at 85 million ounces after July 1979, amounted to almost SDR 5 billion. Further gold revaluations, to ECU 371 per ounce, added almost SDR 15 billion to foreign exchange reserves during the first five months of 1980. A small depreciation of the ECU relative to the SDR prevented that figure from being even larger.

The depreciation of the ECU against the SDR coincided with appreciation of the U.S. dollar against the ECU from the start of the fourth (January 8, 1980) to the start of the fifth (April 9, 1980) swap period. By itself this would have caused more ECUs to be issued against dollars. However, the sales of U.S. dollars by foreign official holders in the face of upward pressures on the dollar were so large during the fourth swap period as to lower the ECUs issued against 20 per cent of the dollar holdings of EMS members at the start of the fifth swap period. On balance, the SDR value of ECUs issued against U.S. dollars thus declined from a peak of SDR 14 billion at the end of the third quarter of 1979 to SDR 11 billion at the end of May 1980, with changes in the SDR value of U.S. dollars contributing very little to this outcome. Since the SDR value of ECUs issued against gold doubled from SDR 17 billion to SDR 34 billion over the same period, ECUs issued against gold accounted for three fourths of all ECUs outstanding at the end of May 1980. At that time, ECUs valued at almost SDR 46 billion were available for the settlement of multilateral balances that may arise from intervention operations between members of the EMS.9

Gold price changes and the ensuing valuation adjustments have therefore dominated changes in non-gold reserves via the ECU in recent months, just as they did during much of 1979. The additional ECUs issued, some SDR 13 billion, exceeded the increase of SDR 11 billion in official holdings of foreign exchange for all countries over the first five months of 1980, and the foreign exchange holdings of industrial countries (including ECUs) grew by only SDR 7 billion. Fund-related assets increased by about SDR 4 billion. SDR holdings rose by SDR 3.7 billion on account of the second allocation of SDR 4 billion in the third basic period being made at the beginning of 1980. Reserve positions in the Fund increased by SDR 0.3 billion from a five-year low point reached toward the end of 1979, mainly because repurchases of sterling in January and April 1980 had raised the reserve position of the United Kingdom to SDR 0.4 billion at the end of May 1980. On the other hand, continued reductions in the Fund’s net borrowing under the oil facilities were not quite matched by increased net borrowing under the supplementary financing facility. Altogether, non-gold reserves rose at an annual rate of 15 per cent in the first five months of this year, but this rate of increase is expected to decline as the year progresses.

As explained in more detail later in this chapter, the demand for non-gold reserves may be affected by the rise in the market value of official gold holdings. Even though that market value is subject to large fluctuations and its future evolution is therefore subject to considerable uncertainty, its present high level may have some depressing effects on the demand for non-gold reserves.

However, national authorities appear to treat gold more often as a permanent store of wealth than as a source of funds for conducting transactions and market intervention. For this reason it remains appropriate to concentrate on the demand for non-gold reserves.

To relate developments in reserves excluding gold to the growth of international transactions—taken as a proxy for payments imbalances—it is useful to take a long-term approach. During the first seven years of the floating exchange rate regime, from the end of March 1973 to the corresponding date in 1980, official holdings of foreign exchange increased by a factor of 2.5, from SDR 98 billion to SDR 248 billion while world exports of merchandise, measured in SDRs, almost quadrupled, rising by a factor of 3.6 over the same period. If world export prices (the export unit value index converted to SDRs), which rose by a factor of 2.7 over the same period, were applied to deflate the SDR value of the foreign exchange component of international reserves, the purchasing power of these reserves over world exports would be found to be about 4 per cent lower in March 1980 than it had been seven years earlier. On this same basis, the “real” value of SDRs and reserve positions in the Fund would each have fallen by over one fourth. Although non-gold reserves need not grow at the same rate as the value of trade, and changes in the availability of borrowed reserves also matter as explained later, the reduction in the “real” value of non-gold reserves over the past seven years is remarkable in view of the 35 per cent growth in the volume of world exports. During a period marked by cyclical and price instability, supply shocks, and other factors contributing to payments imbalances, the world as a whole thus appears to have reduced its reserve cushion against further imbalances.

Foreign Exchange Reserves

This section describes the sources and characteristics of the growth in official holdings of foreign exchange, valued in SDRs, over the past seven years of floating among the currencies of the major industrial countries. It shows how the composition of official foreign exchange reserves has changed as a result of changes both in the quantities of holdings of national currencies and in their SDR prices. It analyzes how official holdings of currencies would have grown if they had earned the corresponding national short-term market rates of interest. Rates of return in a common unit of account on short-term investments in various currencies are calculated, and the hypothetical growth of the SDR value of these currency holdings is compared with the notional growth of the SDR itself that would have occurred if it had earned interest at the “combined market rate,” i.e., at the weighted average of the five national money market rates that enter into the calculation of the SDR rate of interest.

Currency Composition

The currency composition of the increase in foreign exchange reserves in 1979 differed from that of previous annual increments (Table 15). In the years 1973-77, the largest part—on average, three fourths—of the increase in foreign exchange holdings consisted of U.S. dollars. A change in this pattern was already observable in 1978. In that year, the U.S. dollar component of SDR 10 billion was less than one half of the total increment, with much of the remainder taking the form of additions to holdings of deutsche mark (SDR 5 billion) and Japanese yen (SDR 3 billion). In 1979, the principal non-dollar currencies held in official reserves—the deutsche mark, the Japanese yen, the Swiss franc, and the pound sterling—again accounted for an appreciable portion (SDR 9 billion) of the total rise in foreign exchange holdings of SDR 25 billion. Official holdings of U.S. dollars, however, declined by SDR 17 billion. Because the SDR price of the U.S. dollar, which had fallen by 7 per cent during 1978, declined by only 1 per cent in the course of 1979, almost the entire reduction in the SDR value of U.S. dollar holdings resulted from quantity changes, rather than from price changes. The largest part of this reduction was the result of deposits of U.S. dollars equivalent to SDR 13 billion by EMS member countries in the European Monetary Cooperation Fund against an equal amount of ECUs issued to the depositors. These ECUs, as well as those issued against gold deposits, are counted as part of countries’ foreign exchange reserves, while the U.S. dollars (and the gold) depositee with the European Monetary Cooperation Fund are not counted as part of countries’ reserves. Even if the effects of this substitution and of the decline in the SDR value of the U.S. dollar on official holdings of U.S. dollars were left out of account, these holdings would have decreased somewhat during 1979, with sales of U.S. dollars in support of the Japanese yen not fully compensated by net acquisitions by the authorities of other countries. As already mentioned, this leaves only a small net increase (SDR 5 billion) of national currencies held in foreign exchange reserves, with four fifths of the total increase of SDR 25 billion in these reserves stemming from ECUs issued against gold. If one identified the ECUs issued against U.S. dollars as dollars and excluded the ECUs issued against gold from foreign exchange reserves, only a small decline in official holdings of U.S. dollars and a small increase in foreign exchange reserves would remain. In the official statistics, how-ever, the total amount of ECUs outstanding at the end of 1979, almost SDR 33 billion, accounts for 13 per cent of aggregate foreign exchange reserves and 12 per cent of total reserves excluding gold (compared with a share of 9 per cent for total Fund-related assets).

The nominal SDR value of official foreign exchange holdings is affected by changes both in the quantity held and in the SDR price of each component currency (Table 15).10 During the first four years of the period of floating exchange rates, the value of the U.S. dollar in terms of SDRs either remained constant (as during the 15 months preceding the introduction of the first SDR basket in mid-1974) or increased. The SDR value of official holdings of U.S. dollars thus rose not only as a result of the growth in the volume of these holdings but also, although to a much smaller extent, as a consequence of the appreciation of the U.S. dollar against the SDR. In the following three years, 1977-79, the U.S. dollar depreciated against the SDR, and the SDR value of official U.S. dollar holdings decreased as a consequence of these exchange rate effects by a total of SDR 18 billion, which offset almost one half of the increase in the volume of U.S. dollar reserves of SDR 40 billion over this three-year span. As already mentioned, in 1979 a small reduction in the SDR value of official U.S. dollar holdings stemming from a depreciation of the dollar against the SDR by about 1 per cent was reinforcing a decline in the volume of these holdings by SDR 16 billion, which was in large part the result of a substitution of ECUs for U.S. dollars in the reserve accounts of countries that are members of the EMS.

Table 15.Quantity and Price Changes Affecting the SDR Value of Official Holdings of Foreign Exchange, by Currency and in Total, over Contiguous Periods, End of First Quarter 1973-End of 19791(In millions of SDRs)
1973:1to 1974:111974:11 to 1975:1V1975: IV to 1976:IV1976: IV to 1977:1V1977: IV to 1978:1V1978:IV to 1979:IV1973:1 to 1979:IV
U.S. dollar
Starting value75,15186,007105,144122,145151,885161,43775,151
Quantity change10,85615,97616,32336,32820,137−16,03483,586
Price change03,161678−6,588−10,585−1,325−14,659
Total change10,85619,13717,00129,7409,552−17,35968,927
Pound sterling
Starting value6,2366,2335,0523,0113,2673,1746,236
Quantity change199−243−1,29641−69675—693
Price change−202−938−745216−24294—1,399
Total change−3−1,181−2,041257−93969−2,092
Deutsche mark
Starting value5,1926,7488,20810,39115,18520,2035,192
Quantity change1,0541,4721,1623,8063,7902,45313,737
Price change502—111,0219881,2289444,672
Total change1,5561,4612,1834,7945,0183,39718,409
French franc
Starting value9091,0521,6261,4211,4841,973909
Quantity change232479−6648401751,169
Price change−8995—13915885222
Total change143574−205634891271,191
Swiss franc
Starting value1,0751,6252,0672,2293,8624,0411,075
Quantity change45714711,093—3632,0353,370
Price change93295161540542791,710
Total change5504421621,6331792,1145,080
Netherlands guilder
Starting value287413743758809988287
Quantity change94336—5827107483989
Price change32−673247230225
Total change12633015511795131,214
Japanese yen
Starting value7151,1142,0624,903
Quantity change7093637222,5323,6507,976
Price change....636226309−1,255−678
Total change7153999482,8412,3957,298
ECU
Starting value00
Quantity change227,69127,691
Price change4,8184,818
Total change32,50932,509
Sum of above
Starting value88,850102,078123,555141,069178,554196,71988,850
Quantity change12,89218,87616,42942,06526,53521,028137,825
Price change3362,6021,085−4,579−8,3703,637−5,289
Total change13,22821,47817,51437,48618,16524,665132,536
Total official holdings 3
Starting value98,285114,400137,347160,330200,294221,13498,285
Total change16,11522,94722,98339,96420,84024,891147,740
Ending value114,400137,347160,330200,294221,134246,025246,025
Source: Fund staff estimates.

The currency composition of foreign exchange is based on the IMF currency survey (see Table 18, footnote 4) 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 then yields the SDR value of the quarterly price change for each currency. All changes are summed over several quarters to yield cumulative changes over the periods shown.

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 European Monetary System 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.

Source: Fund staff estimates.

The currency composition of foreign exchange is based on the IMF currency survey (see Table 18, footnote 4) 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 then yields the SDR value of the quarterly price change for each currency. All changes are summed over several quarters to yield cumulative changes over the periods shown.

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 European Monetary System 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 only other substantial reduction in the volume of official holdings of any currency during the period under review was that of sterling in 1975 and 1976. This decline in volume was reinforced by the depreciation of sterling against the SDR in those years. For the deutsche mark, the Swiss franc, and the Japanese yen, appreciation of the currency against the SDR tended to reinforce the growth in the volume of official holdings, with the decline in the SDR price of the yen by almost 20 per cent in 1979 constituting the only major departure from this pattern.

Measured over the entire period characterized by floating exchange rates of major currencies, the growth in the volume of holdings of separately identified currencies (excluding ECUs) of SDR 110 billion was offset by a decline in the SDR value of currency holdings resulting from exchange rate changes by SDR 10 billion, or 9 per cent of the volume growth. If ECU holdings are included, about half of the decline in the SDR value of foreign exchange reserves stemming from exchange rate movements is offset by the issue of additional ECUs of almost SDR 5 billion induced by the rise in the market price of gold, so that the net decline in holdings resulting from exchange rate and other price changes is about SDR 5 billion.

These observations indicate that, while the SDR value of a particular currency held in international reserves can rise or decline significantly as a result of exchange rate movements even in the absence of changes in the volume of the currency being held, exchange rate effects are not nearly as important in the evolution of foreign exchange reserves as a whole. The determination of the value of the SDR by the value of a bundle of specified amounts of 16 currencies ensures that the average change in the SDR price of these currencies, weighted by the respective currency amounts in the SDR basket, must always be zero. Hence, exchange rate changes can affect the total of foreign exchange holdings measured in SDRs only to the extent that the composition of these holdings differs from that of the SDR itself.

These differences in composition have, in fact, been considerable (Table 16). The share of the U.S. dollar in the SDR value of foreign exchange reserves identified by currency declined from a peak of 87 per cent at the end of 1976 to 82 per cent at the end of 1978. It then fell further to 65 per cent at the end of 1979, with a third of this decline resulting from the substitution of ECUs for U.S. dollars, as already described. If ECUs issued against U.S. dollars were added to dollar holdings and ECUs issued against gold eliminated from the total in the denominator, the share of U.S. dollars in total foreign exchange reserves identified by currency would be 78 per cent, rather than 65 per cent, at the end of 1979—a decline of 4 percentage points from the end of 1978. Official intervention designed to slow the appreciation of the U.S. dollar against some currencies, particularly the Japanese yen, contributed to this outcome. Even at 65 per cent, the recorded share of the U.S. dollar in foreign exchange reserves was about twice as large as its initial weight of 33 per cent in both the first and second SDR baskets, which went into effect on July 1, 1974 and July 1, 1978, respectively. The shares of other identified currencies that are also included in the SDR basket are less than their shares in the value of the SDR, although for the deutsche mark the difference in the two shares is small. Hence, whenever the SDR price of the U.S. dollar declines and the SDR prices of these other currencies rise, the net effect on the SDR value of foreign exchange reserves is negative. The reason why exchange rate changes had very little net effect on the evolution of the SDR value of the total of official holdings identified by currency is therefore to be found in the behavior of the SDR value of the U.S. dollar, which fell only moderately, by about 8 per cent, from the end of March 1973 to the end of 1979, although there were wide swings within this period.

Table 16.Share of National Currencies in SDR Value of Total Official Holdings of Foreign Exchange, Compared with Shares in SDR Valuation and Interest Rate Baskets, End of Selected Quarters, 1973-791(In per cent)
SDR Basket3
1973:11974:111975:IV1976:IV1977: IV1978:IV1979: IV1979: IV

Excluding

ECU2
Valuation4Interest rate5
U.S. dollar84.684.385.186.685.182.165.1677.833.049.0
Pound sterling7.06.14.12.11.81.61.92.17.511.0
Deutsche mark5.86.66.67.48.510.310.711.712.518.0
French franc1.01.01.31.00.81.00.91.07.511.0
Swiss franc1.21.61.71.62.22.02.83.1
Netherlands guilder0.30.40.60.50.40.50.70.75.0
Japanese yen0.60.81.22.53.33.67.511.0
ECU14.76
Total100.0100.0100.0100.0100.0100.0100.0100.073.0100.0
Sources: Various Fund publications and Fund staff estimates.

The detail in each of the first eight columns may not add to 100 because of rounding

In this alternative calculation, the SDR value of European Currency Units (ECUs) issued against U.S. dollars (SDR 12,784 million at the end of 1979) is added to the SDR value of U.S. dollars, but the SDR value of ECUs issued against gold (SDR 19,725 million) is excluded from the total distributed here. This lowers the total identified by currency to SDR 201,659 million at the end of 1979 and reduces the total change therein during that year from SDR 24,665 million (Table 15) to SDR 4,940 million.

The shares shown for the SDR valuation basket apply precisely only on the date of its inception (July 1, 1978), while the shares in the SDR interest rate basket have been held fixed since that date. The weights of particular currencies in the SDR valuation basket change with the SDR prices of these currencies.

Apart from the six currencies included in this column, specified amounts of ten other currencies are included in the second SDR basket, which accounted for the remaining 27 per cent of the value of the SDR on July 1, 1978. Currency weights in the SDR valuation basket are determined by the shares of issuing countries in total exports of goods and services, except for the weight of the U.S. dollar, which was set at a higher level (33 per cent) to take account of its financial importance.

The weights of the five currencies in the interest rate basket are approximately equal to the initial weights of the five currencies in the second SDR valuation basket adjusted so as to sum to 100. The weight of the U.S. dollar was 2 percentage points lower and that of the pound sterling 2 percentage points higher during the period of the first basket (July 1, 1974 to June 30, 1978).

The share of U.S. dollars would rise by 5.8 percentage points and that of the ECU would fall by the same amount if ECUs issued against U.S. dollars were treated as U.S. dollars in foreign exchange reserves.

Sources: Various Fund publications and Fund staff estimates.

The detail in each of the first eight columns may not add to 100 because of rounding

In this alternative calculation, the SDR value of European Currency Units (ECUs) issued against U.S. dollars (SDR 12,784 million at the end of 1979) is added to the SDR value of U.S. dollars, but the SDR value of ECUs issued against gold (SDR 19,725 million) is excluded from the total distributed here. This lowers the total identified by currency to SDR 201,659 million at the end of 1979 and reduces the total change therein during that year from SDR 24,665 million (Table 15) to SDR 4,940 million.

The shares shown for the SDR valuation basket apply precisely only on the date of its inception (July 1, 1978), while the shares in the SDR interest rate basket have been held fixed since that date. The weights of particular currencies in the SDR valuation basket change with the SDR prices of these currencies.

Apart from the six currencies included in this column, specified amounts of ten other currencies are included in the second SDR basket, which accounted for the remaining 27 per cent of the value of the SDR on July 1, 1978. Currency weights in the SDR valuation basket are determined by the shares of issuing countries in total exports of goods and services, except for the weight of the U.S. dollar, which was set at a higher level (33 per cent) to take account of its financial importance.

The weights of the five currencies in the interest rate basket are approximately equal to the initial weights of the five currencies in the second SDR valuation basket adjusted so as to sum to 100. The weight of the U.S. dollar was 2 percentage points lower and that of the pound sterling 2 percentage points higher during the period of the first basket (July 1, 1974 to June 30, 1978).

The share of U.S. dollars would rise by 5.8 percentage points and that of the ECU would fall by the same amount if ECUs issued against U.S. dollars were treated as U.S. dollars in foreign exchange reserves.

Rates of Return on Major Currencies

To analyze one of the factors involved in currency diversification that may help to explain pressures that are at times exerted on exchange markets as adjustments of currency portfolios are undertaken or attempted, it is useful to compare the rates of return, expressed in a common unit of account, that would have been earned on short-term investments in major currencies. International investors, public and private, arrange their portfolios in the light of a number of criteria, including liquidity, risk, and the expected use of resources denominated in various currencies in the future in addition to expected rates of return. Although the returns that are expected to prevail in the future are not observable, these expectations are formed partly on the basis of past experience. A review of ex post rates of return on short-term investments in major currencies and in the SDR can thus shed light on important factors affecting the portfolio decisions of international investors, particularly private investors. If such investors attempt to alter the currency composition of their portfolios rapidly, the authorities of major industrial countries may feel obliged to intervene in the other direction to smooth exchange rate changes because for official holders yield considerations may be less important than stabilization objectives.

SDR rates of return on investments in each of five major currencies over a given time span are calculated by converting SDR 1 into one of these currencies at the exchange rate prevailing at the beginning of the period, investing the proceeds in a suitable short-term instrument denominated in that currency, and converting both principal and interest back into SDRs at the exchange rate prevailing at the end of the period. The own rate of return on the SDR that is used for comparison is equal to the weighted average of the yields on money market investments in the U.S. dollar, deutsche mark, sterling, French franc, and Japanese yen.11

Realized yields on these short-term investments in terms of the currency in which the investment is denominated (“national yields”) differ systematically among currencies primarily because of differences in expected rates of inflation. Rates of return on instruments with equal liquidity, terms to maturity, and default risk calculated in any common unit of account, such as SDRs, should in principle be equalized through a matching of expected exchange rate changes and expected differences in inflation rates. International interest parity, with full forward cover of the exchange risk, does in fact tend to hold ex ante on such instruments when they are traded in free markets. When viewed ex post, however, this matching tends to be incomplete, at any rate in the short run and in some instances also in the long run (Table 17). For instance, the rate of return on the deutsche mark in terms of SDRs was higher than the return on the dollar in five of the past seven years. The average annual SDR rate of return over the first seven years of floating, starting at the end of the first quarter of 1973, was 12.0 per cent for the deutsche mark, compared with 6.5 per cent for the U.S. dollar. Returns on the three other major currencies also exceeded those on U.S. dollar investments, but by margins (1¾ per cent to 3 per cent) that were considerably smaller than the margin for the deutsche mark.

Table 17.National and SDR Rates of Return Realized on Money Market Investments in Specified National Currencies,1 Annually from End of First Quarter, 1973-80, and Period Averages(In per cent)
Average2
1973:11974:11975:11976:11977:11978:11979:11973:11975:1
tototototototototo
1974:11975:11976:11977:11978:11979:11980:11980:11980:1
Short-term assets
National Rates of Return
denominated in
U.S. dollar7.218.085.845.075.387.3910.457.056.81
Deutsche mark12.7010.275.104.324.443.756.856.734.89
Pound sterling9.6712.1010.9912.138.299.0314.2910.9110.93
French franc9.1913.788.169.019.588.429.819.699.00
Japanese yen7.3513.1411.097.145.804.385.997.806.86
Weighted average38.7410.177.196.496.096.689.657.857.21
SDR Rates of Return
U.S. dollar7.214.5214.184.82−1.223.2013.606.496.75
Deutsche mark25.3115.984.4210.5815.598.005.6912.028.79
Pound sterling5.969.08−4.780.439.5416.7823.118.248.53
French franc4.0824.355.352.1811.4311.078.369.347.62
Japanese yen3.402.7817.4915.4323.746.59−8.638.2110.32
Weighted average49.669.109.356.197.146.9010.138.347.93
Source: Fund staff estimates

The yields on these investments are assumed to be the same as the interest rates in the SDR interest rate basket. Three-month treasury bill rates are used for the United States and the United Kingdom, three-month interbank deposit rates for the Federal Republic of Germany and France, and the call money market rate (unconditional) for Japan.

The fixed annual rate is shown that would have yielded the same growth from the beginning to the end of the period indicated, with annual compounding at the variable rates actually available.

Since the weights applied are the same as those in the SDR interest rate basket, the resulting series is approximately equal to the interest rate officially earned on SDR holdings in excess of allocation if such interest were paid at 100 per cent of the weighted average of the five national rates in the SDR interest rate basket.

While the weights are the same as those in the SDR interest rate basket, they are applied to SDR rates of return and not to the corresponding national rates in the basket. Since the SDR price of the 5 currencies in the interest rate basket has changed relative to that of the 11 additional currencies in the SDR valuation basket, the two weighted averages are not the same. The discrepancy arises from the lack of correspondence of the two baskets.

Source: Fund staff estimates

The yields on these investments are assumed to be the same as the interest rates in the SDR interest rate basket. Three-month treasury bill rates are used for the United States and the United Kingdom, three-month interbank deposit rates for the Federal Republic of Germany and France, and the call money market rate (unconditional) for Japan.

The fixed annual rate is shown that would have yielded the same growth from the beginning to the end of the period indicated, with annual compounding at the variable rates actually available.

Since the weights applied are the same as those in the SDR interest rate basket, the resulting series is approximately equal to the interest rate officially earned on SDR holdings in excess of allocation if such interest were paid at 100 per cent of the weighted average of the five national rates in the SDR interest rate basket.

While the weights are the same as those in the SDR interest rate basket, they are applied to SDR rates of return and not to the corresponding national rates in the basket. Since the SDR price of the 5 currencies in the interest rate basket has changed relative to that of the 11 additional currencies in the SDR valuation basket, the two weighted averages are not the same. The discrepancy arises from the lack of correspondence of the two baskets.

These calculations are very sensitive to the choice of time period: average annual rates of return calculated over the last five of the seven years covered in Table 17 are highest for the yen, in spite of the fact that its dollar price declined at a rate that exceeded the level of Japanese interest rates during the past year. Investments in instruments denominated in deutsche mark showed average annual rates of return that were only 2 percentage points higher than the rate of return on the dollar over this more recent period, and returns on investments in other major currencies exceeded the return on U.S. investments by 1 to 2 percentage points per annum.

That rates of return, measured in SDRs, on short-term investments denominated in major currencies diverge substantially over shorter periods is illustrated in Chart 14. An investor who placed one SDR’s worth of dollars at the end of March 1973 (US$1.20635) in U.S. Treasury bills would have dollars worth SDR 1.55 seven years later, at the end of March 1980. If he had instead invested the same amount in instruments denominated in deutsche mark, he would have SDR 2.21 at the end of the seven-year period, or almost 43 per cent more than for the U.S. dollar investment. Since the deutsche mark appreciated by almost 46 per cent against the U.S. dollar over this period, only a small fraction of the exchange rate change was offset by lower interest rates in the Federal Republic of Germany than in the United States. Returns on sterling were low, and at times negative, during the first three years of the period examined but have risen sharply during the most recent period of 3½ years. Returns on investments in other major currencies have also varied substantially over time, but the cumulative return on each of them exceeds that on the U.S. dollar over the seven-year span examined.

Chart 14.Growth of One SDR Invested in a Specified National Currency or in SDRs with Interest Compounded Quarterly, End of First Quarter, 1973-80

Large and persistent differences in the rates of return realized on particular currencies may be a factor bearing on changes in international currency portfolios. They may also by themselves bring about shifts in the composition of foreign exchange holdings measured in SDRs, if interest receipts on national currency balances are simply reinvested. SDR rates of return have tended to be highest on currencies whose relative importance in both official and private holdings of foreign exchange has been increasing. Furthermore, in 1979, when rising rates of domestic inflation caused rates of return, weighted by the importance of reserve currencies in foreign exchange reserves, to approach double digits, simple accumulation of interest receipts would have caused foreign exchange reserves to grow by about 10 per cent. Official foreign exchange holdings, excluding ECUs issued against gold, actually grew by only 2 per cent in 1979. This indicates that countries were content to allow their foreign exchange reserves to grow by less than the interest receipts on these holdings.

Fund-Related Assets

From the end of March 1973 to the end of March 1980, the share of Fund-related assets in international reserves excluding gold fell from 13.2 per cent to 10.2 per cent. The relative importance of Fund-related assets has thus declined by almost one fourth in spite of renewed SDR allocations at the beginning of 1979 and 1980 and of increases in the drawing facilities of member countries. Even after the last of the three consecutive annual allocations of SDR 4 billion in the third basic period takes place on January 1, 1981, the share of SDRs in total international reserves excluding gold will remain lower than it was at the end of March 1973, and far lower than the peak share of over 10 per cent registered after the last allocation in the first basic period early in 1972. In addition, reserve positions in the Fund equaled only 4.3 per cent of non-gold reserves at the end of March 1980, compared with 5.5 per cent seven years earlier. The four-year program (1976-80) under which the Fund disposed of one third of its gold holdings subtracted about SDR 2 billion from reserve positions in the Fund but increased the liquidity of members both through distribution and through the establishment of the Trust Fund, financed through the profits from gold sales by the Fund and designed for low-interest loans to eligible members.

Official holdings of SDRs by all member countries largely reflect past allocations of SDRs. At the end of March 1980, only about 6 per cent of allocated SDRs were held outside of the group of initial recipients, chiefly by the Fund’s General Resources Account. There have, however, been shifts in SDR holdings between major country groups. For instance, SDRs have been transferred from the non-oil developing countries to the oil exporting countries. At the end of March 1980, the amount of SDRs held by industrial countries was almost exactly equal to their cumulative allocations and that held by oil exporting countries equaled 132 per cent of their allocations. Non-oil developing countries had meanwhile reduced their holdings to 68 per cent of their allocations. Not all developing countries were net users of SDRs, however; notably Brazil and Argentina, among the larger countries, held SDRs in excess of allocations. This contributed to SDR holdings for the non-oil developing countries in the Western Hemisphere remaining much closer to allocations (90 per cent) than those for non-oil developing countries in other regions—Middle East, 31 per cent; Europe, 32 per cent; Africa, 46 per cent; and Asia, 78 per cent. There were also striking differences among industrial countries. At the end of March 1980, both the Federal Republic of Germany and Japan held about twice the amount of SDRs originally allocated to them, while the United Kingdom and the United States had both made use of more than one fourth of their allocations. The amount of net use of SDRs by these two countries exceeded the shortfall of holdings from allocations for the entire group of non-oil developing countries combined.

Changes in aggregate reserve positions in the Fund stem from one of three sources: quota subscription payments in assets other than members’ own currencies, drawings other than those in member’s reserve tranches, and net borrowing by the Fund under the oil and supplementary financing facilities. Reserve positions in the Fund fell from a peak of SDR 19.0 billion in mid-1977 to SDR 11.8 billion at the end of 1979, the lowest value since the third quarter of 1975. Most of the decline occurred in the reserve position of industrial countries (SDR 5.3 billion). In particular, the U.S. reserve position in the Fund decreased by SDR 3.3 billion, mainly as a result of heavy reserve tranche drawings conducted in November 1978. The reserve positions of the oil exporting countries also declined steeply—by SDR 2.5 billion—on account of repayments of oil facility drawings by members, which, in turn, led the Fund to reduce its borrowing under these facilities, mainly from oil exporting countries. On the other hand, reserve positions in the Fund increased by SDR 0.6 billion for the non-oil developing countries from the middle of 1977 to the end of 1979, with little further change recorded in the first few months of 1980. The increase in members’ quotas by 50 per cent, which has been decided but has not yet become effective, will raise the reserve positions of most countries, since 25 per cent of the increase is to be paid in SDRs. However, Fund-related assets as a whole, including the SDR holdings of members, can rise as a result of quota increases only when additional credit tranche drawings ensue.

Reserve Transactions with Banks

Composition of Official Holdings

As in recent years, central banks and other official institutions have continued to hold a significant proportion of their foreign exchange reserves in the Eurocurrency markets. In the period 1973-77, the rise of SDR 113 billion in foreign exchange reserves resulting from transactions (i.e., excluding valuation changes) encompassed an increase of SDR 47 billion in identified Eurocurrency deposits (Table 18). In 1978, the increase in direct claims of SDR 28 billion far exceeded that of Eurocurrency holdings of SDR 2 billion; in 1979, by contrast, direct claims fell by SDR 6 billion, while Eurocurrency holdings again showed a small increase of SDR 2 billion. The significant increase in direct claims during 1978 was the result of large U.S. deficits on both current and capital account. The small rise in identified official holdings of Eurocurrencies in part reflected the fact that much of the increase in reserves in 1978 was concentrated among members of the Group of Ten, which have generally refrained, for most of the period since 1971, from redepositing reserve accruals in the Eurocurrency markets. In addition, oil exporting countries, some of which hold a substantial proportion of their official non-gold reserves in Eurocurrency assets, experienced a substantial decline in total reserves excluding gold—by SDR 12 billion (adjusted). In 1979, declining exchange market pressures on the dollar accompanied the reduction in the U.S. current account deficit. The sharp fall in the capital account deficit reinforced the swing from accumulation to decumulation of direct claims on the United States, even though the exchange of dollar claims for ECUs accounted for about one third of the difference in accumulation.

Table 18.Composition of the Change in Official Holdings of Foreign Exchange Reserves, 1973-791(In billions of SDRs)
1973197419751976197719781979
Official claims on countries
United States4.78.54.211.330.224.5−11.4
Other countries1.83.60.31.43.35.0
Subtotal6.512.14.211.631.627.8−6.4
Identified official holdings of
Eurocurrencies
Eurodollars3.713.54.67.310.3
Other currencies2.00.11.7−0.24.22.22.0
Subtotal5.713.66.37.114.52.22.0
European Currency Units32.5
Residual231.9−0.4−3.23.2−1.33.8−3.4
Total change arising from
transactions414.125.37.321.944.833.824.7
Valuation change4−8.4−0.53.61.1−4.8−8.70.2
Total35.724.810.923.040.025.1524.9
Sources: International Financial Statistics and Fund staff estimates.

More detailed information on changes in official holdings of foreign exchange reserves can be found in the IMF Survey, Vol. 9 (June 3, 1980), page 167.

Includes identified official claims on the International Bank for Reconstruction and Development, on the International Development Association, and on the European Monetary Cooperation Fund, except ECUs, and the statistical discrepancy.

Small differences between the changes in the IMF Survey and those shown here are due to the inclusion of Romania in the present table

In the corresponding table published in the IMF Survey, monthly changes in national currencies held in foreign exchange reserves are converted into SDRs through use of the average monthly SDR price of the currency involved, while the average of two adjoining end-of-quarter SDR prices was applied to quarterly quantity changes in Table 15. (See footnote 1 to that table.) As a result of this refinement and because the entire change in the residual and in the ECUs issued in 1979 is attributed to quantity change in the present table, there are small differences between this table and Table 15 in the way in which the total change in foreign exchange reserves is decomposed into changes arising from transactions (quantity changes) and valuation changes owing to price effects. More important differences in the changes in holdings identified by currency, particularly the U.S. dollar, arise from a difference in data sources. This table uses U.S. balance of payments statistics on official claims on the United States to identify such holdings, while Table 15 is based on the survey on the composition of monetary authorities’ gross claims on foreigners conducted by the Fund.

The decrease of SDR 4.3 billion in coverage of Saudi Arabia’s foreign exchange holdings in 1978 is not reflected in this total or in any of its components.

Sources: International Financial Statistics and Fund staff estimates.

More detailed information on changes in official holdings of foreign exchange reserves can be found in the IMF Survey, Vol. 9 (June 3, 1980), page 167.

Includes identified official claims on the International Bank for Reconstruction and Development, on the International Development Association, and on the European Monetary Cooperation Fund, except ECUs, and the statistical discrepancy.

Small differences between the changes in the IMF Survey and those shown here are due to the inclusion of Romania in the present table

In the corresponding table published in the IMF Survey, monthly changes in national currencies held in foreign exchange reserves are converted into SDRs through use of the average monthly SDR price of the currency involved, while the average of two adjoining end-of-quarter SDR prices was applied to quarterly quantity changes in Table 15. (See footnote 1 to that table.) As a result of this refinement and because the entire change in the residual and in the ECUs issued in 1979 is attributed to quantity change in the present table, there are small differences between this table and Table 15 in the way in which the total change in foreign exchange reserves is decomposed into changes arising from transactions (quantity changes) and valuation changes owing to price effects. More important differences in the changes in holdings identified by currency, particularly the U.S. dollar, arise from a difference in data sources. This table uses U.S. balance of payments statistics on official claims on the United States to identify such holdings, while Table 15 is based on the survey on the composition of monetary authorities’ gross claims on foreigners conducted by the Fund.

The decrease of SDR 4.3 billion in coverage of Saudi Arabia’s foreign exchange holdings in 1978 is not reflected in this total or in any of its components.

There are a number of reasons, however, for interpreting these reserve changes with caution. First, since central banks or other designated authorities can borrow from the Eurocurrency markets and from each other, the reported figures do not reveal the potential access to international liquidity of an individual country. Second, it is possible for monetary authorities to sell a portion of their foreign exchange holdings to their commercial banks under repurchase agreements; as a result, these holdings would not be included in the reported reserve figures.12 Finally, some central banks have utilized foreign exchange market intervention techniques (especially in the forward markets) and swaps that have altered the time path for their reserves. During periods of domestic liquidity shortages caused by capital outflows, the central banks have used these swaps to encourage foreign borrowing; and, as a result, the stock of official international reserves has been temporarily enlarged. Such transactions can be quite large and can have a major impact on the monetary base of the domestic economy as well as on the stock of reported international reserves. For instance, central bank money is temporarily reduced if the monetary authorities sell foreign exchange to domestic banks under a repurchase contract. Conversely, central bank money is increased if they buy foreign exchange in the spot market and sell it forward, even though this will not produce a lasting effect on international reserves.

Regulation and Supervision of the Eurocurrency Markets

The growing importance of the Eurocurrency markets has led to some further discussions about the level of official regulation and supervision of these markets that would be appropriate in light of the aim of maintaining the soundness and stability of the international banking system. Last year’s renewal of the 1971 agreement, under which central banks of the Group of Ten refrained from depositing reserve accruals in the Eurocurrency market for a period of several years, meant that central banks will not contribute further to the growth of that market. A more ambitious proposal, that the central banks of the major industrial countries jointly impose a reserve requirement against Eurocurrency deposits of their domestic commercial banks and their Eurocurrency branches, has received little support. Such reserve requirements would reduce the cost advantage of offshore banking relative to domestic banking operations, but successful implementation would require broad participation among the world’s central banks. The central bank Governors of the Group of Ten countries reaffirmed in April of this year the cardinal importance that they attached to the maintenance of sound banking standards—particularly with regard to capital adequacy, liquidity, and concentration of risks. Such wide agreement has led the authorities of many countries to focus on more active supervision of commercial banking activity in the Eurocurrency markets. To this end they place high priority on bringing into full effect the initiatives already taken by the Committee on Banking Regulations and Supervisory Practices with regard to the supervision of banks’ international business on a consolidated basis, improved assessment of country risk exposure, and the development of more comprehensive and consistent data for monitoring the extent of banks’ maturity transformation. All the steps taken or contemplated so far recognize the important role played by banks in recycling large surpluses that have arisen during the last few years and are likely to persist in coming years.

International Liquidity and Adjustment

Recycling

A major issue confronting the world economy in 1980 and beyond is whether international capital markets will be able to play the same role in recycling the surpluses of the oil exporting countries as they did in 1974-76. The increase in crude oil prices from an average of approximately US$13 a barrel in December 1978 to more than US$30 in 1980 is expected to lead to an increase of about US$160 billion in the revenue from oil exports this year compared with 1978, thereby contributing to an estimated current account surplus of oil exporting countries of US$115 billion in 1980. For comparison, the current account of the oil exporting countries was nearly in balance in 1978. The modalities of recycling these surpluses will be especially important for the non-oil developing countries, whose current account deficit is expected to increase from US$53 billion in 1979 to US$70 billion this year and to rise even further in 1981.

There have been suggestions to the effect that the solution of the recycling problem is virtually automatic. Since the oil exporting countries will tend to invest their additional reserves in liquid form, at least initially, the resulting deposits will provide the international banking system with a matching amount of funds to lend to oil importing countries. While this recycling process might work once again as smoothly as it has in the past, there are certain factors that could create difficulties.

In the period 1974-76, much of the recycling took the form of Eurodollar deposits by the oil exporting countries being used to finance U.S. dollar loans arranged through large U.S. and European banks. Since that time, however, the currency composition of official deposits and bank loans has become more diversified, with almost half of the increase in official holdings of Eurocurrencies identified in Table 18 invested in currencies other than the U.S. dollar. Furthermore, risk-averse surplus countries may continue to diversify their reserve holdings by both currency and location to reduce political and economic risks on investments placed abroad. For banks, greater concern about loan quality may be stimulated by the possibility of a worldwide recession and by the need to protect their increasingly slim capitalization base. If international banks, as a group, should attempt to “upgrade” their portfolios by reducing foreign exposure and directing more of their net lending toward domestic markets and low-risk instruments in major industrial countries, some of the non-oil developing countries with the most pressing current account imbalances might find it difficult to obtain funds from private sources. This could happen even though, in the aggregate, the supply of funds from earned surpluses matched the requirements for financing corresponding deficits. The margins between loan and deposit rates may thus widen appreciably for some non-oil developing countries.

Margins may widen to some extent for all countries if banks regard strengthening their capitalization base as imperative and falling interest rates and growing risk aversion facilitate the widening of spreads as they did from 1974 to 1975. While this spread is the net cost relevant for the holding of borrowed reserves, declining lending rates reduce the nominal cost of foreign exchange borrowed for other purposes. Whether this also implies a fall in real interest costs, for example, on U.S. dollar loans to non-oil developing countries, must be assessed by referring to the expected increase in the dollar price of the exports of these countries and not just to inflation prospects in the United States. If the terms of trade of non-oil developing countries should decline, their real borrowing costs may rise and this, combined with a cyclical decline in the demand for their exports and with non-gold reserves that have already fallen from 29 to 24 per cent of merchandise imports (c.i.f., at an annual rate) from the first to the last quarter of 1979, could lead to growing difficulties in financing the large current account deficits projected for this group of countries.

The Provision of Liquidity by the Fund

During recent years, the Fund has taken initiatives to increase the accessibility of its existing resources and to expand the availability of balance of payments financing. In providing such financing, the Fund plays a role complementary to that of private international financial institutions by seeking to reach understandings that members will follow macroeconomic policies aiding their balance of payments adjustment. As adjustment programs are devised by its members, the Fund applies the same basic criteria in consultations with all its members while paying due regard to the domestic social and political objectives, the economic priorities, and the circumstances of the member. Although the normal period for stand-by arrangements is one year, the period may be extended to up to three years. The Fund encourages members to adopt corrective measures, which can be supported by the use of the Fund’s facilities, at an early stage of the development of balance of payments difficulties or, indeed, as a precaution against the emergence of such difficulties.

The Fund has augmented its resources through increases of members’ quotas and through borrowing for the establishment of temporary facilities, such as the oil facilities of 1974 and 1975 and the supplementary financing facility.13 Although Fund liquidity is currently adequate, the size of the projected recycling problem could lead to heavy utilization of existing resources for a longer period than has been usual in the past, since the increase in quotas under the Seventh General Review of Quotas (from SDR 39 billion to SDR 59 billion) has not yet come into effect. This may be especially true if the Fund is to be able to support appropriate medium-term programs, particularly for non-oil developing countries that are attempting to adjust to a substantial change in the relative price of oil.

To enable members to utilize existing liquidity more effectively in support of long-range programs of stabilization and structural change, the Fund has increased the maximum repurchase period under the extended Fund facility from eight years to ten and reduced the number and frequency of repurchase installments associated with this facility. The combined effect of these measures will increase the average life of a drawing under the extended Fund facility by almost one fifth. There have also been discussions of closer coordination between the Fund and the World Bank in designing adjustment and development programs for member countries. While such adjustment programs will necessarily continue to include demand management policies to avoid overconsumption in relation to available resources, they will also place greater emphasis on the supply side of the economy. This will require, inter alia, increased importance being attached to long-term energy programs providing for the development of alternative sources of energy and energy conservation. Such structural programs can be put in place only over a longer period than has been typical in past Fund standby programs. At its meeting in Hamburg in April 1980, the Interim Committee recommended that the Managing Director begin discussions with possible lenders to the Fund on the terms and conditions under which the Fund could borrow to increase its resources, if and when the need arose.

The Adequacy of International Reserves

The preceding discussion of recent developments in international reserves and liquidity provides a broad background for assessing the adequacy of international reserves. Such an appraisal can begin by examining the ratio of international reserves (excluding gold) to annual merchandise imports. For all countries combined, this ratio has fallen from about 26 per cent in 1973 to about 22 per cent in 1979. If the United States were excluded—on account of the special status of the dollar as principal reserve currency—the corresponding figures would be 31 per cent and 25 per cent. This decline does not necessarily mean that reserves have become less adequate. The real value of international reserves demanded appears to grow proportionally less than the volume of trade. The growing diversification of production and trade of many economies tends to reduce payments imbalances, and thus reserve needs, relative to the value of external trade. Indeed, the fall in the ratio of reserves to imports over the years 1973-79 follows a period of almost two decades in which this ratio showed a declining trend interrupted only briefly during the period 1970-72.14 More recently, the sharp rise in the price of primary commodities relative to the price of manufactures in international trade has also contributed to lowering reserves-to-trade ratios in the importing countries, since the price elasticity of demand for such commodities tends to be low. Until recently, however, high rates of inflation that were accompanied by rapid expansions of money and credit and low real costs of borrowing provided persuasive evidence against a global inadequacy in the supply of reserves. Rather, ample borrowing opportunities and lines of credit made it less pressing to bring actual reserve holdings up to the level that would provide a satisfactory cushion against contingencies even in the absence of new loan commitments. Finally, although loans backed by gold are the exception rather than the rule, the ability of countries to borrow by pledging gold as collateral at a moderate discount from the market price has been greatly enhanced by the more than tenfold increase in the SDR price of gold over the past decade. The dramatic increase in the market value of gold reserves may have had some depressing effect on the demand for international reserves excluding gold.

National and international liquidity and credit are now rather closely linked. In recent years, high rates of domestic credit expansion in the major financial centers were accompanied by high rates of expansion of credit to a large number of developing countries. Although this situation may change, the elasticity of supply of international reserves has remained sufficiently high to make global reserve holdings determined largely by the effective demand for reserves. Apart from the facilities of Eurocurrency markets, a number of arrangements made in an atmosphere of international cooperation—notably the increased opening of national markets to foreign issues, negotiated investments by the authorities of some surplus countries in the official obligations of deficit countries, and the broadening of swap arrangements and lines of credit between central banks—have contributed to maintaining a high elasticity of supply of international liquidity.

This picture of global reserve adequacy should not obscure the fact that both the distribution of reserves and the terms at which reserves can be acquired are less satisfactory for many non-oil developing countries than they appear for the world as a whole. Even though the ratio of reserves to imports for this group of countries was close to the average for all countries excluding the United States at the end of 1979, current and prospective payments imbalances and the variability of export proceeds are proportionally greater for non-oil developing countries than for industrial countries, implying greater reserve needs. Furthermore, non-oil developing countries are frequently unable, under current conditions, to increase their reserves in any way other than through official foreign borrowing at interest rates that are generally higher than those charged to other countries.

The SDR and Its Uses

In recent years, concern about disturbances that may arise through attempts by private and some official holders of foreign exchange to alter the currency composition of the assets and liabilities in their portfolios has intensified. Under present conditions of floating exchange rates among major currencies, relatively unrestricted international banking transactions and capital flows, and large payments imbalances in the world economy, disturbances resulting from changing views on desirable portfolio compositions are potentially disruptive. For this reason, there has been a growing recognition that a more central role of the SDR—as the principal reserve asset in the international monetary system, as an investment vehicle, and as a contract unit of account—would reduce the incentive to seek exchange risk diversification by other, more costly and potentially destabilizing, means.

The Fund has recently taken a number of decisions to make the SDR more useful in serving these purposes, including the prescription of a number of official entities as “other holders” of SDRs and the prescription of additional ways in which SDRs can be used. These decisions are described in detail in Chapter 3. An important step in the direction of making the SDR a more useful asset was the decision, discussed in last year’s Report, to raise the interest rate on the SDR from 60 per cent of the combined market rate to 80 per cent. Strengthening the role of the SDR in the international monetary system requires, among other things, the maintenance of a sound financial basis as reflected in the overall yield of the asset—composed of nominal return and changes in the capital value resulting from exchange rate changes—in relation to the yields of other important assets in the system.

Further attention was drawn to these matters by exploratory work conducted by the Executive Board in connection with the possible establishment of a substitution account, administered by the Fund, that would accept deposits of U.S. dollars in exchange for an equivalent amount of SDR claims. In the period between the Annual Meetings in Belgrade in September 1979 and the meeting of the Interim Committee in Hamburg, the Executive Board studied possible features of such an account to design a plan for its establishment along the lines requested by the Interim Committee in its communique issued in Belgrade. The Committee had concluded that such an account, if properly designed, could contribute to an improvement of the international monetary system and could constitute a step toward making the SDR the principal reserve asset in the system. As the Committee had noted, in order for such an account to achieve widespread participation on a voluntary basis and on a large scale, it should fulfill certain requirements: inter alia, it should satisfy the needs of depositing members, both developed and developing; its costs and benefits should be fairly shared among all parties concerned, and it should contain satisfactory provisions with respect to the liquidity of the claims, their rate of interest, and the preservation of their capital value.

In the course of the work of the Executive Board between the two meetings of the Interim Committee, provisional agreement was reached on a large number of features that a substitution account might possess in order to conform to the requirements set forth by the Committee. A number of important issues, however, still remain to be solved, including arrangements for the maintenance of financial balance in the account, the maximum initial target size of the account, the determination of the rate at which the United States would pay interest on the account’s U.S. dollar holdings, the sharing of any profits or losses between the United States and depositors, and the relative voting power of the United States and other participants. The Interim Committee expressed its intention to continue work on this subject.

On a related issue already mentioned, the Interim Committee endorsed the intention of the Executive Board to continue its examination of the SDR valuation and interest rate baskets with a view to simplifying and enhancing further the attractiveness of the SDR. The Committee expressed the view that it would be desirable for the interest and valuation baskets to be identical. This proposed change, which is further discussed in Chapter 3, would substantially facilitate the wider use of SDR-denominated assets and liabilities in financial markets, and in international transactions generally, and would thus remove a severe obstacle to the establishment of a central role for the SDR in the international monetary system.

For a description of these developments, see Chapter 1.

The members of the EMS are Belgium, Denmark, France, the Federal Republic of Germany, Ireland, Italy, Luxembourg, the Netherlands, and the United Kingdom. The United Kingdom does not participate in the common margins arrangement. Italy’s margins are 6 per cent on either side of its bilateral central rates, compared with 214 per cent for the other participants.

Developments within the EMS are discussed in greater detail in the following subsection.

The change in the value of reserves can be no more than a rough measure of intervention in the foreign exchange market, since it does not take into account valuation changes owing to exchange rate variations, swap operations between the monetary authority and commercial banks, and other indirect forms of intervention.

The divergence indicator is a measure of the divergence of the market rate expressed in ECUs for each EMS currency from its ECU central rate. When the indicator reaches one of the thresholds established for each member, the authorities at their option may respond by undertaking official intervention, central rate changes, monetary policy measures, or other domestic policy measures.

Executive Board Decision No. 5392-(77/63), adopted April 29, 1977. See AnnualReport, 1977, pages 107-109

The dual approach to gold valuation reflects concern about distortions arising from the instability of gold prices. By contrast, only the second method is used to value the dollar holdings of the members of the EMS, i.e., current rather than average exchange rates are applied for that purpose exclusively. Neither the gold nor the dollars against which ECUs are issued are transferred permanently to the European Monetary Cooperation Fund, and both remain under the administration of the countries making these “deposits,” with the interest on dollar balances accruing to the depositors.

The United Kingdom, which holds about 8 per cent of all ECUs issued, does not participate in the exchange rate mechanism

Since only about 90 per cent of all holdings of foreign exchange reserves are identifiable by currency for the period under review, the cumulative change during this period of almost seven years in total foreign exchange reserves of SDR 148 billion exceeds the change in holdings identified by currency of SDR 133 billion by about 11 per cent. The periods examined in Table 15 are annual, except for the first two periods, which are somewhat longer to accommodate both a starting point at the beginning of the period of floating exchange rates and a break at the date, in the middle of 1974, when the value of the SDR first came to be determined on the basis of a currency basket.

The national yield series employed in all calculations are the same as those contained in the SDR interest rate basket: the market yields for three-month U.S. Treasury bills and for three-month U.K. Treasury bills, the three-month interbank deposits rate in the Federal Republic of Germany, the three-month interbank money rate against private paper in France, and the call money market rate (unconditional) in Japan. For a description of this SDR interest rate basket and the 16-currency SDR valuation basket, see Table 16. Not all of these instruments have always been available for investment by foreigners, but comparable, although generally somewhat higher, yields have been available on Eurocurrency investments in all five currencies in the most recent years. Problems arising from the lack of correspondence between the SDR interest rate and valuation baskets are mentioned in the last section of this chapter.

In this situation, the central bank would have a domestic claim on the commercial banks; and the commercial banks would have a claim on a foreign financial institution.

See Annual Report,1979, pages 54-55, for a detailed discussion of these facilities

The historical pattern is shown in AnnualReport, 1975, page 40.

    Other Resources Citing This Publication