Latin American Medium-Term Import Stabilization Policies and the Adequacy of Reserves
Author: Bruno Brovedani

BOTH THE LEVEL of the international monetary reserves, however defined, held by a country at any given point of time and the variations in reserves over a period of time reflect a combination of changing circumstances and a certain pattern of economic policy decisions made by the monetary and exchange authorities.1 Conversely, a broad range of policy decisions is influenced by both the existing level of monetary reserves and the level which it is desired to establish in the future. A country with large reserves has a choice between a program of overinvestment or of overconsumption which results in balance of payments disequilibrium and a depletion of reserves, and policies which maintain reserves, or even increase them further—policies which imply that some consumption and/or some investment has for the time being been foregone. On the other hand, if a country insufficiently endowed with reserves engages in large programs of investment and/or of consumption, these programs can be sustained only so long as export receipts remain high. If such a country is confronted with a drastic fall in its export income, it may have to impose a large cut in imports and in employment in order to maintain some degree of equilibrium in its balance of payments. Thus the range of interactions between policy decisions and the level of reserves is very broad and full of complexities.

Abstract

BOTH THE LEVEL of the international monetary reserves, however defined, held by a country at any given point of time and the variations in reserves over a period of time reflect a combination of changing circumstances and a certain pattern of economic policy decisions made by the monetary and exchange authorities.1 Conversely, a broad range of policy decisions is influenced by both the existing level of monetary reserves and the level which it is desired to establish in the future. A country with large reserves has a choice between a program of overinvestment or of overconsumption which results in balance of payments disequilibrium and a depletion of reserves, and policies which maintain reserves, or even increase them further—policies which imply that some consumption and/or some investment has for the time being been foregone. On the other hand, if a country insufficiently endowed with reserves engages in large programs of investment and/or of consumption, these programs can be sustained only so long as export receipts remain high. If such a country is confronted with a drastic fall in its export income, it may have to impose a large cut in imports and in employment in order to maintain some degree of equilibrium in its balance of payments. Thus the range of interactions between policy decisions and the level of reserves is very broad and full of complexities.

BOTH THE LEVEL of the international monetary reserves, however defined, held by a country at any given point of time and the variations in reserves over a period of time reflect a combination of changing circumstances and a certain pattern of economic policy decisions made by the monetary and exchange authorities.1 Conversely, a broad range of policy decisions is influenced by both the existing level of monetary reserves and the level which it is desired to establish in the future. A country with large reserves has a choice between a program of overinvestment or of overconsumption which results in balance of payments disequilibrium and a depletion of reserves, and policies which maintain reserves, or even increase them further—policies which imply that some consumption and/or some investment has for the time being been foregone. On the other hand, if a country insufficiently endowed with reserves engages in large programs of investment and/or of consumption, these programs can be sustained only so long as export receipts remain high. If such a country is confronted with a drastic fall in its export income, it may have to impose a large cut in imports and in employment in order to maintain some degree of equilibrium in its balance of payments. Thus the range of interactions between policy decisions and the level of reserves is very broad and full of complexities.

As was pointed out in the Fund report on the adequacy of monetary reserves presented to the Economic and Social Council in 1953,2 any general discussion of the adequacy of reserves can be fruitful only if it is related to some general objective of economic policy. This paper has the limited purpose of analyzing, for countries whose “real exports” (i.e., exports of goods and services adjusted by reference to an import price index) are liable to sharp variations, the relationship between adequacy of reserves and the policies which may be adopted with a view to ensuring a medium-term stabilization of the volume of their imports.

Variability of Real Exports

The international trade position of many countries in Latin America is generally characterized by two outstanding features: (1) exports are a very important determinant of income and of investment; (2) fluctuations in real exports are very pronounced.

The ratio of exports to net monetary income, which is generally high in Latin America, indicates the importance of exports as a determinant of income and of investments. Their importance in these countries is, indeed, even greater than an examination of this ratio might suggest. With very few exceptions the major industries are export industries; since their products are not sold domestically in any significant proportion, the effects of sudden fluctuations in export income cannot be mitigated by opposite variations in home consumption.

Export incomes accrue, to a larger extent than in countries at a more advanced stage of development, to the most influential economic groups, including the government, which control a very large proportion of investments and of imports. Thus the most active sectors of the economy regulate their decisions with respect to investment and imports to a large extent on the basis of changes in export income.

A marked degree of variability in export values is generally the result of lack of diversification and of complex circumstances relating to the production and marketing of the commodities involved. When there is a large number of diversified exports, the normal changes in prices abroad or in production trends at home often tend to offset each other; variations in the general price and/or quantum index of exports tend to be less pronounced than the variations in single price or volume indices.

The most important Latin American raw material exports are either mining products, such as copper, tin, and petroleum, or agricultural and livestock products, such as coffee, cocoa, sugar, cotton, and wool. Practically all of these commodities are subject to special conditions which often produce a marked degree of instability in world demand for them and in their world market prices.

The income elasticity of demand for the mining products exported by the countries of Latin America is comparatively high; the same is probably true of their agricultural and livestock products, at least in some important markets. The rather high income effect on demand is often aggravated by the adoption in consuming countries of commercial policies which, in order to prevent an undue fall in employment at home during a recession, may lead to the elimination or reduction of imports.

The low elasticity of supply of many agricultural products leads to wide fluctuations in their prices in response to small changes in demand. Unpredictable changes in world supply conditions also may cause wide fluctuations in prices. Although the output of minerals can generally be adjusted more readily than the output of agricultural and livestock products to changes in demand, the wide fluctuations in demand for minerals that are often due to political events, such as war or danger of war, frequently result in price instability.

The impact of this intense price variability on the economy of the exporting country is generally not softened by any variations in an opposite direction in the volume of their exports but, on the contrary, is aggravated by variations in the same direction.

In exporting countries which are responsible for only a small part of total world output of the commodity involved, there may be a tendency, often encouraged by the government, for producers or exporters to accumulate stocks (or, in the case of mining products, to reduce production) during periods of price decline, and to deplete stocks (or to increase the exploitation of mines) when prices are rising. But also countries which produce a major part of the world output of the commodity involved sometimes try to slow down a fall in international prices, or at least to mitigate its impact on domestic producers, by policies of stock accumulation during periods of price decline and of stock depletion during periods of price rise.

As a consequence of these policies, the prices of exports and the quantities exported often tend to fluctuate in the same, rather than in opposite, directions; the fluctuations in total receipts are, consequently, wider and steeper than the fluctuations in either prices or quantities exported. Since these fluctuations in the value of exports are generally much wider than those in the prices paid for imports, “real exports” as defined above, i.e., the purchasing power of exports over imports, also tend to show wide fluctuations. An aggravating factor is the tendency, which might indeed be expected, for the net proceeds from long-term foreign investments or loans to move in the same manner as the proceeds from exports, the main reason for this being that long-term investments and lending are more profitable and less risky during periods of prosperity and of boom in the country of destination.

Recent Experience of Variability of Real Exports

The recent experience of six selected Latin American countries in relation to fluctuations in real exports is illustrated in Charts 1 and 2, which show the fluctuations of export values, adjusted by an import price index, around their mean level during the period 1946–53. (Exports of services have been neglected as being of small importance.) The selection of this period as a basis for comparison is perhaps open to some criticism: it was hardly normal, because of the short-lived dip of 1949 and, particularly, of the raw material boom in the first stages of the Korean war. The degree of variation was in fact at its maximum during this period. However, any search for a “normal” period would carry the investigation so far into the past as to get out of touch with current conditions; provided that its limitations are understood, 1946–53 may be accepted as providing a useful basis for investigation.

Chart 1.
Chart 1.

Variability of “Real Exports” of Ecuador, Paraguay, and Uruguay, 1946–531

(1946–53 = 100)

Citation: IMF Staff Papers 1955, 001; 10.5089/9781451960150.024.A003

1 Indices of “real imports” given in Table 1.
Table 1.

Variability of “Real Exports” of Selected Latin American Countries, 1946–531

article image

From International Monetary Fund, International Financial Statistics. Estimates for 1953, which for many countries were not available at the time of writing, were prepared on the basis of information available to the Fund up to February 1954.

Ratio of index of value of exports to the index of import prices for Latin America (weighted average of export price indices for the United States, continental EPU countries, and the United Kingdom) as published in International Monetary Fund, International Financial Statistics. For the period covered here, the import price indices used are as follows: 1946, 100; 1947, 119; 1948, 127; 1949, 123; 1950, 109; 1951, 127; 1952, 127; 1953, 124.

Billion cruzeiros.

Chart 2.
Chart 2.

Variability of “Real Exports” of Bolivia, Brazil, and Chile, 1946–531

(1946–53 = 100)

Citation: IMF Staff Papers 1955, 001; 10.5089/9781451960150.024.A003

1 Indices of “real imports” given in Table 1.

Chart 1 covers three countries where there have been pronounced fluctuations, with short periods of expansion or of contraction. The real value of exports of Ecuador rose steeply, from 42 per cent below average in 1949 to 31 per cent above average in 1950; from 1952 to 1953 there was a fall of about 13 percentage points. Equally pronounced were the fluctuations in the real exports of Uruguay, which expanded from 19 per cent below average in 1948 to 34 per cent above in 1950 and declined to a few percentage points below average in 1952. In Paraguay the maximum change occurred from 1947 to 1950, when real exports rose from 30 per cent below average to 21 per cent above. (See also Table 1.)

Chart 2 covers three other countries for which fluctuations in real exports are a little less pronounced than those of the countries in Chart 1. During 1946–50 the real exports of Bolivia, Brazil, and Chile were below the average for the period as a whole; they expanded in 1951 and subsequently tended to decline. The maximum deviations were 26 per cent below average in Bolivia in 1947 and 32 per cent above average in Chile in 1952. (See Table 1.)

These changes during 1946–53 may be usefully compared with the wartime fluctuations in real exports (Table 2). From 1939 to 1946 fluctuations also were large, especially in Uruguay, where the value of real exports rose from 36 per cent below average in 1942 to 41 per cent above average in 1946; in Brazil, where it rose from 12 per cent below average in 1943 to 20 per cent above average in 1946; in Peru, from 24 per cent below average in 1943 to 53 per cent above average in 1946; and in Cuba, from 37 per cent below average in 1942 to 23 per cent above average in 1944.

Table 2.

Variability of “Real Exports” of Nine Latin American Countries, 1939–461

(1939–46 = 100)

article image

Indices of “real exports” as described in footnote 2 of Table 1. Based on data from International Monetary Fund, International Financial Statistics.

For these fluctuations there were several reasons, which, as a rule, tended to produce their effects at the same time. Political conditions abroad often strongly affected the prices of raw materials to a much larger extent than they affected the general level of prices. During 1942–46 and 1950–51, real exports of nearly every one of these countries rose rapidly, and there also were pronounced fluctuations in the volume of exports.

Whatever the reason, the pronounced variability in real exports, which is the most outstanding characteristic in every country studied, is a powerful obstacle to any kind of economic programing of investment, imports, public expenditures, etc. As neither the volume nor the prices of exports can be easily regulated, the possibility of reducing such variability is slight. A degree of variability in real exports sufficient to create substantial problems for the raw material countries exposed to these variations seems likely to continue, whatever its immediate causes may be.

The Search for Short-Term Balance of Payments Equilibrium

The tendency toward pronounced variability in real exports creates some particularly acute problems. If the rate of real imports is more or less maintained, there is a serious risk of disturbance to the balance of payments equilibrium; or, if measures are taken to keep the balance of payments as close to equilibrium as possible, wide fluctuations in imports are likely, which can have disturbing effects on the economy. It is generally recognized that the maintenance of equilibrium in the balance of payments is one of the most important objectives of economic policy. Such equilibrium may emerge as the result of spontaneous responses to changing conditions. It is inevitable that equilibrium must be attained in one way or another, for the demand for exchange cannot indefinitely exceed the supply which becomes available. Equilibrium, however, has to be interpreted with reference to a time period. Since the current analysis is confined to the problems which arise for countries whose real exports are liable to marked variations as those countries endeavor to determine the periods within which they may safely permit temporary disequilibrium in the balance of payments, equilibrium is examined here only from the point of view of its short-term and medium-term aspects.

Very few of the countries whose exports show strong seasonal variations attempt to ensure strict equilibrium for a period shorter than one year. Fluctuations in export receipts will not, as a rule, be followed immediately by corresponding fluctuations in imports, for central and commercial banks are usually equipped with such modest reserves as are necessary to overcome merely seasonal disequilibria between receipts and payments. “Short-term” equilibrium may therefore be defined as involving the attainment of a balance between receipts and payments over a period of about a year. Where the reserve position is sufficiently strong, a country may be content to allow a longer period to elapse before equilibrium is established. Such “medium-term” equilibrium may be defined as covering a period of up to eight years.

The monetary authorities who wish to assure balance of payments equilibrium may act on exports, on imports, or on other items in the balance of payments, and their actions may be combined in a wide variety of ways. For obvious reasons, imports appear to be the most flexible item in most of these countries, and intervention is usually concentrated upon them. An expansion of imports from one year to another is easy in countries where the exchange rate is overvalued and there is a general system of restrictions. A relaxation of restrictions nearly always results in a sudden expansion of imports, especially when the income level is rising. If imports have to be reduced drastically, the desired result can be achieved by strengthening controls or by suddenly devaluing the exchange rate or by using both instruments. In all cases, when the aim of policy is short-term equilibrium in the balance of payments, very effective instruments of control have to be applied to imports, the most important and most flexible item in the balance of payments. Clearing agreements require an automatic adjustment of imports to the varying level of exports, especially when the agreements do not provide for reciprocal granting of credit.

If economic policy aims at the maintenance of a medium-term balance of payments equilibrium, as arbitrarily defined here, any short-term disequilibrium has to be covered by the use of reserves; exchange restrictions can then be less stringent than is necessary if short-term equilibrium is the objective.

The foreign trade policies of a country are not always consistent; they vary extensively in response to political and economic changes. In general, their objectives fall in an intermediate position between the short-term and the medium-term equilibrium described above. Countries with fairly large monetary reserves, or whose exports do not vary so widely, may aim at maintaining some degree of stability in the volume of imports. This does not imply precise uniformity from year to year in the volume of imports; rather, it implies the prevention of recurring broad fluctuations. On the other hand, many countries for which the variability of real exports is a constant problem may, because of inadequate reserves or the difficulties of forecasting, or for other reasons, follow policies that aim at attaining short-term equilibrium in the balance of payments. This often implies a degree of variability in imports as large as, if not even larger than, the variability of exports.

In most countries, rather severe forms of import control, e.g., rigid systems of quantitative restrictions or cost restrictions imposed through multiple rate policies, have been the instruments most commonly used to ensure short-term equilibrium. The exchange budget based on estimated foreign exchange receipts is a useful instrument for achieving this purpose. The maintenance of overvalued rates for at least a proportion of imports implies the existence of a large backlog of demand that can be satisfied whenever a sudden rise in export receipts permits a relaxation of restrictions. Moreover, the close association in some countries between export income and government revenue means that government imports fluctuate more or less directly with variations in export earnings. Other balance of payments items, such as services and capital movements, are less flexible and less easy to control by direct methods. An attempt is often made to revise legislation in order to promote a larger inflow of capital during periods of falling exports, but the effort is often frustrated by the very existence of those conditions which the inflow of foreign capital should relieve. The attempt to reduce the outflow of investment income through controls may result in a decline in the inflow of new capital, so that there is no benefit for the balance of payments.

Some Aspects of Short-Term Equilibrium Policy

An analysis of the economic implications of policies designed to ensure short-term equilibrium in a country’s balance of payments raises a number of important issues. When real export income is rising as a result of a substantial improvement in the terms of trade or for reasons relating to production, both the increase in national income and the improvement in the reserve position tend to lead to a rapid increase in imports. An improvement in export earnings affects the income of both the private and the public sector. In the private sector—first in the export industries, and then in other industries as the income expansion radiates through the rest of the economy—both consumption and investment will increase. Spending may increase more rapidly than income, as a general wave of optimism pervades the economy and a less cautious attitude on the part of the banks makes possible an expansion of credit in order to finance expanded demand.

The public sector may respond in a somewhat similar manner. In several Latin American countries, the methods most convenient for the collection of government revenue create a direct link between fluctuations in those revenues and fluctuations in export income. In fact, whenever export and government revenues rise, the government’s consumption and investments also expand. In such periods of prosperity, large deficits often occur, which indicate that a purely temporary movement has been interpreted as a permanent improvement. When the general programing of public expenditures is based on overoptimistic considerations, there is likely to be substantial overexpansion.

As a result of these responses in the private and the public sectors, the stimulation provided by higher exports may lead to an even larger increase in demand for imports. A comparative abundance of foreign exchange resources may induce the government to liberalize import controls greatly, before the full effects of the expansion of internal demand on imports have become evident. As a result of a chain of events of this nature, it not infrequently happens (as the figures given below indicate) that a country shows a balance of payments deficit during periods of export expansion.

If real exports decline owing to a deterioration in the terms of trade or to unfavorable production trends, imports eventually have to be curtailed, to maintain equilibrium in the balance of payments. Imports may be reduced by even larger percentages than the fall in exports, to permit repayment of foreign indebtedness, frequently accumulated during periods of prosperity. This process of adjustment to a lower level of imports is painful and often wasteful. A sudden return to a more moderate level of consumption of imported commodities is difficult to achieve except by the adoption of drastic forms of control. Some investment projects may have to be abandoned, and economic activities may have to be reduced. A reduction of imports of raw materials and of machinery may result in large disturbances to the productive process, in production bottlenecks, and in some unemployment.

As indicated above, a substantial fall in export income may result in a pronounced decline in government revenue. However, a curtailment of public expenditures is not easily achieved, and the resulting deficits may be financed by resort to central bank credit. Such an expansion in the money supply at a time when goods are less readily available has a strong inflationary impact. This chain of reactions weakens the economy considerably; in an unstable environment, the programing of development becomes extremely difficult if not impossible.

Variability of Real Imports

The fluctuations in real imports associated with fluctuations in real exports are shown for six selected Latin American countries (including five whose export fluctuations are discussed above) in Charts 3 and 4. As in the charts for exports, the import values, adjusted by reference to an import price index, are plotted as percentages of the average for 1946–53.

Chart 3.
Chart 3.

Variability of “Real Imports” of Brazil, Paraguay, and Uruguay, 1946–531

(1946–53 = 100)

Citation: IMF Staff Papers 1955, 001; 10.5089/9781451960150.024.A003

1 Indices of “real imports” given in Table 3.
Table 3.

Variability of “Real Imports” of Selected Latin American Countries, 1946–53

article image

From International Monetary Fund, International Financial Statistics. Estimates for 1953, which for many countries were not available at the time of writing, were prepared on the basis of information available to the Fund up to February 1954.

Ratio of index of value of imports to the index of import prices for Latin America (weighted average of export price indices for the United States, continental EPU countries, and the United Kingdom) as published in International Monetary Fund, International Financial Statistics. For the period covered here, the import price indices used are as follows: 1946, 100; 1947, 119; 1948, 127; 1949, 123; 1950, 109; 1951, 127; 1952, 127; 1953, 124.

Billion cruzeiros.

Chart 4.
Chart 4.

Variability of “Real Imports” of Chile, Ecuador, and Peru, 1946–531

(1946–53 = 100)

Citation: IMF Staff Papers 1955, 001; 10.5089/9781451960150.024.A003

1 Indices of “real imports” given in Table 3.

Brazil shows the widest range of fluctuations (Chart 3). From a level 36 per cent below average in 1946, Brazilian real imports rose to 44 per cent above average during 1951 and 1952, and then declined below average in 1953 (Table 3). Real imports of Paraguay also moved erratically within a very broad range: they were 13 per cent below average in 1947, 8 per cent above average in 1949, 17 per cent below average in 1950, and 14 per cent above average in 1952. For Uruguay, real imports were 17 per cent below average in 1946, 3 per cent above in 1947, 17 per cent below in 1949, 41 per cent above in 1951, and roughly 9 per cent below in 1953.

The fluctuations (shown in Chart 4) for Chile, Ecuador, and Peru are much less pronounced, especially if a rising line rather than a simple average is taken to represent the trend. Imports of these countries rose steadily; short-term fluctuations from one year to another were the exception rather than the rule.

An examination of data for the war period, on the other hand, throws little light on the import policies of these countries at that time. During the first few years of the war (roughly 1939–42), Latin American countries maintained a reduced volume of imports, owing to difficulties of transportation and of production in foreign countries. At that time, monetary reserves were accumulated, which permitted a steady and large expansion of imports in the later period.

Assurance of Greater Stability to Imports

If the serious disadvantages resulting from sharp and frequent fluctuations in the volume of imports are to be avoided, a country has to be in a position where there is less need for concern over the maintenance of short-term balance of payments equilibrium. A country with a relatively easy reserve position can afford to take measures tending toward a stabilization of income in the face of wide export fluctuations, and it can recurrent short-term balance of payments disequilibria and from the wide fluctuations in monetary reserves. However, there is likely to be some offset to these disturbances.

In periods of falling export receipts, domestic income generally tends to decrease in both the private and the public sector. However, imports could not be maintained without the adoption of appropriate fiscal and monetary policies which would also offset the general decline in income. The deflationary impact of a balance of payments deficit could be offset by the expansionary impact of fiscal and monetary policies. Likewise, during periods of export expansion monetary and fiscal policies could be influential in the opposite direction in maintaining imports at a more or less constant level. The demand for credit would be kept under control and increased revenue would strengthen the financial position of the government. Therefore, through the actions of banks and the fiscal policies of the government, a reasonable degree of stability could be maintained in the economy in the face of recurrent disequilibria in the balance of payments.

A clear illustration of the compensatory impact of banking policies is provided by data for Uruguay. A comparison of the annual expansion in the outstanding volume of bank loans in Uruguay with the annual changes in the banks’ holdings of gold and foreign exchange, both expressed as percentages of the money supply, is given in Chart 5 and Table 4. From the diverging trends of the two curves it is clear that whenever a decline in the banks’ holdings of gold and foreign exchange had a deflationary impact, the expansion in the banks’ loans tended to be large; when an increase in the banks’ holdings of foreign exchange had an inflationary impact, the expansion in the banks’ loans tended to be very small.

Chart 5.
Chart 5.

Factors of Monetary Expansion in Uruguay, 1946–521

(Annual changes as per cent of money supply)

Citation: IMF Staff Papers 1955, 001; 10.5089/9781451960150.024.A003

1 For description of data, see Table 4.
Table 4.

Factors of Monetary Expansion in Uruguay, 1946–52

article image

Data as of beginning of year. Money supply as published in International Monetary Fund, International Financial Statistics, plus government deposits with the Bank of the Republic.

Derived from data in the consolidated balance sheet of the commercial banks of Uruguay and the balance sheet of the Bank of the Republic.

The chart shows, therefore, that in Uruguay the banking system tended, on the whole, to mitigate the impact of export fluctuations and not, as may frequently happen elsewhere, to reinforce it by credit expansion in boom years and by credit contraction in years of depression. As a result, when export incomes declined, imports were maintained by the extension of credit to exporters and importers, or for development purposes. The institutional structure of the banking system compelled it to follow corresponding offsetting policies in the years when exports were high and reserves were accumulating. In those years the central bank had to reduce its lending for developmental purposes, because it was not permitted to issue new currency against the greater part of its foreign exchange balances (i.e., those held in inconvertible currencies). Its cash was, in effect, “frozen” through the purchase of foreign exchange balances.

To one familiar with the economic structure of most Latin American countries, however, the mechanism of the second correcting factor, i.e., fiscal policy, appears more important. International trade produces income for the central government from (a) customs duties and other indirect taxation of imports, (b) taxation of exports, which is related to export prices and volume, and (c) exchange rate differentials, which result from maintaining selling rates generally higher than buying rates.

Whenever export income falls drastically and imports are correspondingly reduced in order to achieve equilibrium in the balance of payments, government revenue falls. Thus a deficit will appear with a compensatory impact, which may offset in part the deflationary trends in the economy. The opposite may happen during periods of expansion if an adequate degree of caution is exercised.

The adoption of policies of import stabilization would presumably change this picture, but not very drastically. It is likely that during periods of depressed export income government revenue would decline less than in the previous case, since imports, production, and investment in the private sector would be maintained; at the same time, government expenditure, especially on public investment, is less likely to be curtailed if imports are kept more or less constant.

During periods of export expansion the increase in government revenue would presumably be less if imports were maintained constant. But since the public program of investment expenditure, so closely linked to imports in these countries, would not respond so quickly to fluctuations in exports, there is a possibility of an improvement in the budgetary position, which may offset, in this sector, the expansion originating from the increase in exports and from the over-all surplus in the balance of payments.

If they are to be completely effective, such policies of income stabilization and import stabilization require that substantial monetary reserves should be available for use during periods of depressed exports. Monetary reserves might fluctuate extensively from year to year, but they would remain roughly unchanged over medium-term periods of, say, eight years. Countries with very large reserves may permit an overall deficit in the balance of payments for a medium-term period; if reserves are small, medium-term stabilization would still be possible if imports were kept at a somewhat lower average level, which would permit some long-run accumulation of reserves.

The difficulties of pursuing these policies may, however, be so great as to offset the benefits that can be derived from them. A country normally confronted with a high degree of variability in its real exports should attempt to adopt policies of import stabilization only when it can forecast with some accuracy the future trend of its exports and it possesses adequate monetary reserves. These two requirements are closely interrelated.

An approximate forecast of trends of real exports for a medium-term period can be based on an appraisal of production trends for the main export commodities and on considerations affecting fluctuations in income abroad (business cycles) and prices. Production trends can be more readily appraised by countries that export only a few commodities, though forecasting errors are more likely to cancel out if there is a considerable diversity of exports.

The range of fluctuations in the past, measured by reference to the ratio of average real exports during the period selected for consideration to real exports in the base year, may be illustrated from the experience of 1946–53. This experience suggests, in retrospect, the considerations that would be applicable to a judgment regarding the adequacy of reserves at the beginning of the period, i.e., at the beginning of 1946, for a policy of medium-term import stabilization. For the period 1946–53, the real value of exports averaged 28 per cent higher than in 1946 for Chile, 14 per cent higher for Uruguay, 10 per cent higher for Brazil, and 5 per cent lower for Paraguay. These figures show that there was, at least in a few countries, a strong rising trend during the period under consideration. A projection of these figures, however, cannot yield an accurate forecast of import policy. Large errors may be made in estimating the equilibrium level of imports. Monetary reserves would increase if the estimate is too low, and they would decline if the estimate is too high.

A country with large reserves can afford to be optimistic in its projections, but the opposite is true for countries insufficiently provided with reserves. In any event, the estimates are tentative projections; from time to time there may have to be some deviations from the policy of import stabilization, especially if unusual events occur. For instance, a deep business depression in foreign markets, which might result in a tremendous decline in exports, is difficult to foresee. If there should be such a depression, the equilibrium level of imports should be immediately lowered, to avoid a great loss of monetary reserves. Conversely, if unusual events, such as the boom resulting from the Korean war, should cause exports to rise sharply, an upward revision of imports could be contemplated. The exchange control authorities should be very cautious in interpreting as an indication of a permanent change a movement, whether in exports or in reserves, which may be merely temporary. The experience of the 1946–53 period shows that fluctuations may be very pronounced; the very high export levels reached in 1951, for example, could not be maintained in the following years. To base a program of investment, of development, and of public consumption on an exceptional situation involves serious dangers.

A Review of Policies

The juxtaposition for individual countries of their export and import curves makes possible a more detailed examination of the general characteristics of national monetary reserve and import policies in the period 1946–53. The two curves are shown together in Charts 610 for four of the countries to which reference has already been made, and also for El Salvador where, in contrast to the other countries which have been examined, direct government control of foreign trade has been insignificant. All items in the balance of payments other than exports and imports (i.e., services and capital movements) are neglected because in most cases their net amount is not large and their real variations are not significant; trade fluctuations are the dominant influence in determining the balance of payments of these countries.

Chart 6.
Chart 6.

Variability of “Real Exports” and “Real Imports” of Chile, 1946–531

(1946–53 = 100)

Citation: IMF Staff Papers 1955, 001; 10.5089/9781451960150.024.A003

1 Indices of “real exports” and “real imports” as given in Tables 1 and 3.
Chart 7.
Chart 7.

Variability of “Real Exports” and “Real Imports” of Uruguay, 1946–531

(1946–53 = 100)

Citation: IMF Staff Papers 1955, 001; 10.5089/9781451960150.024.A003

1 Indices of “real exports” and “real imports” as given in Tables 1 and 3.
Chart 8.
Chart 8.

Variability of “Real Exports” and “Real Imports” of Ecuador, 1946–531

(1946–53 = 100)

Citation: IMF Staff Papers 1955, 001; 10.5089/9781451960150.024.A003

1 Indices of “real exports” and “real imports” as given in Tables 1 and 3.
Chart 9.
Chart 9.

Variability of “Real Exports” and “Real Imports” of Brazil, 1946–531

(1946–53 = 100)

Citation: IMF Staff Papers 1955, 001; 10.5089/9781451960150.024.A003

1 Indices of “real exports” and “real imports” as given in Tables 1 and 3.
Chart 10.
Chart 10.

Variability of “Real Exports” and “Real Imports” of El Salvador, 1946–531

(1946–53 = 100)

Citation: IMF Staff Papers 1955, 001; 10.5089/9781451960150.024.A003

1 Indices of “real exports” and “real imports” as given in Table 7.
Table 5.

Monetary Reserves of Nine Latin American Countries, 1946–531

(In millions of U.S. dollars)

article image

Gross holdings of gold and foreign exchange of central banks as of beginning of year.

Source: International Monetary Fund, International Financial Statistics.
Table 6.

Monetary Reserves as a Percentage of the 1946–53 Average of the Real Value of Exports of Nine Latin American Countries, 1946–531

article image

Monetary reserves data are those in Table 5. The 1946–53 averages for the real value of exports were as follows: Bolivia, $92 million; Brazil, $1,072 million; Chile, $272 million; Cuba, $573 million; Ecuador, $46 million; El Salvador, $129 million; Paraguay, $25 million; Peru, $150 million; Uruguay, $174 million.

Table 7.

Vabiability of “Real Exports” and “Real Imports” of El Salvador, 1946–53

(In millions of U.S. dollars)

article image

From International Monetary Fund, International Financial Statistics.

For description of this series, see Table 1, footnote 2.

For description of this series, see Table 3, footnote 2.

Chile

The record shows that the Chilean authorities placed considerable emphasis on the search for short-term equilibrium in the balance of payments. During the period 1946–53 there were fluctuations in reserves, expecially the decline of $13 million in 1947 and of $6 million in 1949, and the increase of $12 million in 1952 (Table 5). These movements are a result of variations in imports and exports shown in Chart 6. However, the fluctuations reflect, for the most part, the difficulty of achieving from year to year a perfect adjustment of imports to the changing level of exports. In general, there was a tendency for the two curves to move in the same direction. On the other hand, there was substantial equilibrium in the balance of payments for the period as a whole, since reserves did not change much from 1946 to 1953. The expansion of exports was matched by a similar expansion of imports. The average of real exports, for the period as a whole, was 28 per cent higher than in 1946, while the average of real imports was 22 per cent higher.

Strict import controls were maintained to assure short-term equilibrium in Chile’s balance of payments. The controls included quantitative import restrictions, multiple rates, and, in general, a yearly programing of foreign exchange payments on the basis of estimated foreign exchange receipts, through the preparation of an exchange budget at the beginning of each year.

The fluctuations in real exports were not very sharp; a continuous expansion at least through 1952 is noticeable, in large part resulting from an improvement in the terms of trade, following an upward movement in the prices of copper and nitrates, Chile’s main exports. The policies of the banking system and of the Government were not greatly affected by fluctuations in reserves, although they were strongly influenced by variations in export income. The large expansion in export receipts during 1949–52 resulted in a recrudescence of the inflationary forces which have for a long time undermined the Chilean economy. The Government expanded its outlays much in excess of the growth in revenues resulting from the prosperous foreign trade position. Large deficits in the public sector contributed to a steady growth in the money supply. The banking system did not pursue compensatory or stabilizing policies; its lending expanded in line with the growth of the credit base.

Monetary reserve holdings of the Central Bank, which amounted to $110 million at the beginning of 1946, appear to have been “inadequate” for the adoption of policies tending to maintain stability in Chile’s imports. The ratio of monetary reserves at the beginning of 1946 to the average value of real exports during the period 1946–53 was 40 per cent, while the sum of the annual percentages by which real exports in 1946–50 fell short of the average for the whole period 1946–53 (as indicated by the shaded area in Chart 6) was 60.

Uruguay

The monetary authorities of Uruguay, in a manner similar to that of Chile, also aimed at short-term equilibrium in the balance of payments (Chart 7). However, the availability of large reserves made possible a greater degree of flexibility in the adjustment of imports to the changing level of exports. There were large balance of payments disequilibria, especially during years of sharp changes in exports, as in 1947, 1950, and 1951. There were marked fluctuations in central bank holdings of gold and foreign exchange: a decline of $36 million in 1947, an increase of $96 million in 1950, and a decline of $113 million in 1951 (Table 5).

Uruguay’s real exports showed pronounced variations, especially from 1947 to 1950 (an increase of 56 percentage points) and from 1950 to 1952 (a decline of 39 percentage points). This resulted from sharp price variations for major exports (wool, meat, etc.), as well as from a reluctance to export during periods of price decline. For the entire period 1946–53, real exports averaged 14 per cent above such exports in 1946; for imports the expansion was larger, aggregating 20 per cent.

Import controls in Uruguay included quantitative restrictions, multiple exchange rates, and the preparation of exchange budgets. As a result of the time lag in effecting the necessary adjustments to imports, there were large deficits or surpluses in the balance of trade in the years immediately following drastic changes in export trends. For this reason, the variability from year to year in real imports was even greater than that for exports.

The ratio of monetary reserves at the beginning of 1946 to the average value of Uruguay’s real exports during the period 1946–53 amounted to 145 per cent (Table 6). The sum of the annual percentages by which real exports in 1946–49 fell short of the average for the whole period was 63. Thus reserves in Uruguay appear to have been adequate for purposes of import stabilization.

Ecuador

Ecuador’s real exports fluctuated extensively during the period under consideration (Chart 8), as a result of sharp changes in the terms of trade and in production, and of the export control policies designed to assure adequate domestic supplies of staple foodstuffs, such as rice. The import and export curves showed divergent movements. As indicated by Chart 8, there was some degree of import stability from 1946 through at least 1952; it is apparent that the monetary authorities had considerable success in stabilizing imports, at least up to 1952, in the face of sharp variations in the level of exports. In 1953, a rise in income and an expansion in investment and in bank lending resulted in a steep rise of imports. For the entire period, real imports averaged 34 per cent above those in 1946. The expansion in real exports during this period was small, the average being only 8 per cent above the value in 1946.

The lending policies of the banking system were compensatory, and fluctuations in monetary reserves did not affect significantly domestic monetary trends. Rough stability in government expenditure was maintained, even during periods of large expansion in income from exports.

The monetary reserves of the Central Bank rose only slightly, from $33 million at the beginning of 1946 to $39 million at the end of 1953. The ratio of monetary reserves at the beginning of 1946 to the average of real exports during the period 1946–53 was 72 per cent. The sum of the annual percentages by which real exports in 1946–49 fell below the average for the whole period was 88. Reserves do not appear to have been quite adequate to ensure import stabilization throughout the entire period.

There were large deficits in Ecuador’s balance of payments during the period 1947–49, when imports were maintained despite a sharp fall in the value of exports. The decline between 1946 and 1949 in the Central Bank’s gross reserves amounted to $10 million. In 1950 there was an increase of almost $11 million, since a large expansion in export income was not counteracted by a similar expansion of real imports.

Brazil

The sharp fluctuations in the real exports of Brazil during the period under consideration (Chart 9) resulted from changes in export prices or production trends or from the accumulation of stocks of export commodities during periods of low international prices. In the period 1946–49, imports were maintained at a level commensurate with the level of exports, the balance of payments was roughly in equilibrium, and there was little change in reserves. The monetary authorities attempted to maintain short-term equilibrium in the balance of payments during those four years. In 1950 imports were kept low, notwithstanding the sharp expansion in exports during that year, and stringent quantitative controls on imports permitted the achievement of a large surplus. During 1951, however, imports were allowed to expand much more than exports; and in 1952 they were maintained at the same high level notwithstanding the substantial decline in export income in that year.

The movements of the two curves in Chart 9 indicate the adoption in Brazil of policies with objectives other than import stabilization; large balance of payments deficits occurred during periods of record exports, and imports fluctuated widely around their 1946–53 average.

The large deficits in the trade and payments balance resulted in a depletion of monetary reserves and in an accumulation of short-term indebtedness to foreign exporters.

The overexpansion of imports for the period as a whole is indicated by the fact that real exports averaged 10 per cent higher than the comparable value for 1946, while the average for real imports was 56 per cent higher than in 1946. The substantial expansion in imports during 1951–52 was the result of a liberalization of controls, which, however, were still rather stringent, and of the maintenance of a stable rate of exchange, despite the considerable decline in purchasing power of the Brazilian currency during this period of strong inflation. The fact that inflation continued in 1951 and 1952, notwithstanding the huge balance of payments deficit, indicates that the monetary impact of the balance of payments was of small significance and was more than completely offset by other factors. The variations in imports had a strong impact on the Brazilian economy and the productive structure.

If Brazil’s large short-term indebtedness is included in the calculation of reserves, the reserve position in 1953 was probably negative. The immediate problem for Brazil is the re-establishment by means of balance of payments surpluses of its previous reserve position.

The ratio of monetary reserves at the beginning of 1946 (gross holdings of the Bank of Brazil) to the average of real exports during the period 1946–53 was 62 per cent (Table 6). Since the sum of the annual percentages by which real exports in 1946–49 fell short of the average for the whole period was 54, reserves appear to have been barely enough in 1946 for purposes of import stabilization.

El Salvador

The record for El Salvador contrasts, in many respects, with that of the other four countries examined in this study. El Salvador is a country in which conditions closely approximate those of a free economy, where economic adjustments occur automatically without interference by any of the purposive controls adopted in most of the other countries.

Coffee is El Salvador’s most important export. Despite some sharp variations in production and in the volume of exports, El Salvador’s real exports rose steadily from 1946 to 1953 and real imports tended to rise at much the same rate (Table 7). There was little effort to accumulate reserves, and export earnings were spent very nearly as they were earned.

As shown by Chart 10, fluctuations around the trend line were negligible. However, this trend should not be projected into the future. A considerable flattening out of both curves may be expected, but fluctuations from year to year may appear again, if export prices remain stable while there are strong variations in production and the volume of exports.

Summary and Conclusions

With a few exceptions there was a substantial degree of variability in the real exports of most Latin American countries during 1946–53, which may be regarded as a medium-term period. Such variability has particular importance in view of the great sensitivity of these economies to the impact of international factors. Their economic history has, to a large extent, been dominated by the actions they have had to take in response to sudden increases and decreases in export income. There are reasons for believing that, over a medium-term period, a substantial degree of variability in real exports is a permanent feature of these economies; this is due chiefly to lack of export diversification, marked variations in prices and production of major exports, and the adoption of commercial policies of “price defense,” which have meant the accumulation of commodity stocks during years of price declines for sale during periods of price increases.

If adequate reserves are not available, a great variation in export income will have serious effects on the smooth development of an economy. In that situation the government may be forced to pursue policies designed to adjust imports each year to variations in exports. This would lead to frequent revisions of the levels of public and private investment and consumption in countries, such as those of Latin America, where such investment and consumption are largely based on imported goods. As a result, internal programing of investment and of consumption becomes almost impossible when the country cannot rely on stability of imports.

Furthermore, from the adoption of policies of short-term balance of payments equilibrium, there frequently arises the need to make drastic revisions in the import program. This presupposes the maintenance of efficient systems of import control to be strengthened whenever the need arises, for example, because of sudden decreases in export income.

Policies designed to secure greater stability of imports can be based only on long-term projections and forecasts of economic trends. They also imply the availability of rather large monetary reserves to be used as “buffer stocks” for the maintenance of a high level of imports when real exports drastically decline. These policies would have a beneficial effect since they would remove the major disadvantages of fluctuating real imports. The destabilizing effects on income of balance of payments disequilibria are not likely to be of great importance.

The adequacy of reserves for the pursuance of medium-term import stabilization requires consideration from three standpoints: (1) the degree of export variability; (2) the country’s position at any point of time in relation to the trade cycle; (3) accuracy of forecasts.

Degree of export variability

The greater the degree of variation of real exports, the greater the need for reserves. A study of past trends can indicate only the order of magnitude of past fluctuations, and any projections or estimates for the future based on such study must be interpreted with great caution.

The variability of real exports in the countries examined in this paper was roughly indicated by summing the percentages by which real exports fell short each year of the average during periods of consistently depressed exports. The highest total (88) was for Ecuador. A plausible conclusion is that reserves higher than, say, 90 to 100 per cent of average real exports would be roughly adequate for purposes of medium-term import stabilization.

Relation to the trade cycle

In order to judge the adequacy of monetary reserves, it is necessary to estimate the position which the country has reached at any point of time in relation to the trade cycle. The maximum need for reserves is in the year or two just before exports fall below their estimated average. In periods of export expansion, it is necessary to provide for a continuous growth of reserves.

Accuracy of forecasting

Any projection of average real exports for a medium-term period is difficult, and large errors are likely. The greater the degree of possible error, the greater the need for reserves. Particularly when reserves are not large, estimates should be cautious and imports should be established at a level lower than that indicated by the export forecast.

Even if reserves should be inadequate to assure perfect stability of imports, they may be sufficient to permit the avoidance of sharp and strong variations in real imports. The more limited goal of reducing the degree of instability in imports can be achieved by establishing, with extreme caution, the level of future imports and by making appropriate revisions of programs during the period.

*

Mr. Brovedani, economist in the Latin American (South) Division, was educated at the University of Venice, Italy, and at the Catholic University of Washington, D.C. He was formerly a member of the staff of UNRRA and of the U.S. Economic Cooperation Administration in Italy, and has served as economic advisor to the Government of El Salvador and Professor of Economics in the San Salvador University.

1

This paper was presented to the Fourth Meeting of Technicians of Central Banks of the American Continent, held in Washington, D.C., May 1954.

2

“The Adequacy of Monetary Reserves,” Staff Papers, Vol. Ill, No. 2 (October 1953), pp. 181–227.

IMF Staff papers: Volume 4 No. 2
Author: International Monetary Fund. Research Dept.
  • View in gallery

    Variability of “Real Exports” of Ecuador, Paraguay, and Uruguay, 1946–531

    (1946–53 = 100)

  • View in gallery

    Variability of “Real Exports” of Bolivia, Brazil, and Chile, 1946–531

    (1946–53 = 100)

  • View in gallery

    Variability of “Real Imports” of Brazil, Paraguay, and Uruguay, 1946–531

    (1946–53 = 100)

  • View in gallery

    Variability of “Real Imports” of Chile, Ecuador, and Peru, 1946–531

    (1946–53 = 100)

  • View in gallery

    Factors of Monetary Expansion in Uruguay, 1946–521

    (Annual changes as per cent of money supply)

  • View in gallery

    Variability of “Real Exports” and “Real Imports” of Chile, 1946–531

    (1946–53 = 100)

  • View in gallery

    Variability of “Real Exports” and “Real Imports” of Uruguay, 1946–531

    (1946–53 = 100)

  • View in gallery

    Variability of “Real Exports” and “Real Imports” of Ecuador, 1946–531

    (1946–53 = 100)

  • View in gallery

    Variability of “Real Exports” and “Real Imports” of Brazil, 1946–531

    (1946–53 = 100)

  • View in gallery

    Variability of “Real Exports” and “Real Imports” of El Salvador, 1946–531

    (1946–53 = 100)