Recent Developments in the U. S. Balance of Payments

VARIATIONS in the gold and dollar assets of foreign countries have, in recent years, been closely related to their economic transactions with the United States. The supply of dollars currently available to these countries has been directly affected by fluctuations in the U.S. demand for their exports, and the imports of these countries from the United States have tended, in turn, to show similar fluctuations, though usually with a lag of one or two quarters. Substantial disparities which appear when this relationship is tested by reference to particular areas are partly explainable by differences in multilateral dollar settlements with third areas.

Abstract

VARIATIONS in the gold and dollar assets of foreign countries have, in recent years, been closely related to their economic transactions with the United States. The supply of dollars currently available to these countries has been directly affected by fluctuations in the U.S. demand for their exports, and the imports of these countries from the United States have tended, in turn, to show similar fluctuations, though usually with a lag of one or two quarters. Substantial disparities which appear when this relationship is tested by reference to particular areas are partly explainable by differences in multilateral dollar settlements with third areas.

VARIATIONS in the gold and dollar assets of foreign countries have, in recent years, been closely related to their economic transactions with the United States. The supply of dollars currently available to these countries has been directly affected by fluctuations in the U.S. demand for their exports, and the imports of these countries from the United States have tended, in turn, to show similar fluctuations, though usually with a lag of one or two quarters. Substantial disparities which appear when this relationship is tested by reference to particular areas are partly explainable by differences in multilateral dollar settlements with third areas.

A study of the more recent changes is facilitated by tracing developments through three successive phases. During the first phase, from the middle of 1949 through July 1950, U.S. imports rose rapidly under the influence of domestic recovery from the 1948-49 recession, whereas exports continued the decline which had begun in 1947. These movements led to a substantial acquisition of U.S. gold and dollar assets by the rest of the world. In the second phase, beginning with the outbreak of war in Korea and ending with the first quarter of 1951, U.S. demand for imports continued, but there were distorting side-effects which reflected speculative reactions. The enlarged supply of dollar funds abroad caused a rise in the demand for U.S. exports, which began to increase sharply in September 1950, and the outflow of gold and dollars from the United States started to diminish. In the third phase, covering the second and third quarters of 1951, U.S. demand for imports dropped sharply—reflecting, to a great extent, a reaction against the previous build-up of prices. Exports also decreased, although not so sharply as imports. By the third quarter, the rest of the world was again drawing on its gold and dollar assets to meet obligations in the United States.

Analysis of the factors which caused these developments is complicated by the differences in timing between the two sides of the trade balance, by unusually large capital movements, and by an increase in the extent of governmental controls.

Supply and Use of Dollar Funds

From the end of 1949 through mid-1951, the supply of dollar funds currently becoming available to the rest of the world rose steadily, from $3 billion a quarter to over $4½ billion.1 Unlike the 1947-49 period, when the supply of dollar funds fluctuated mainly in response to variations in U.S. Government aid, the main source of the increased dollar supply after 1949 was larger receipts from merchandise shipments to the United States, i.e., from U.S. imports (Chart 1 and Table 1). An exception to this generalization was the third quarter of 1950, when U.S. private capital movements to Canada added about $500 million to the world supply of U.S. dollars.

Chart 1.
Chart 1.

World Sources and Uses of Dollar Funds, Quarterly,1947–51

Citation: IMF Staff Papers 1952, 001; 10.5089/9781451949391.024.A002

* “World” indicates the aggregate of all countries outside the United States.
Table 1.

U.S. International Transactions, Quarterly, 1950-51

(In millions of U.S. dollars)

article image
Source: Based on data from U.S. Department of Commerce, Survey of Current Business, June 1951 and December 1951. In the Survey for March 1952, revised figures were published for the third quarter of 1952. The major revisions were increases of $49 million in imports of unrecorded merchandise, of $84 million in government services, and of $130 million in errors and omissions.

The rise in the value of U.S. imports since 1949 is a continuation—accentuated by the special stimulus of post-Korean conditions—of an upward trend that began immediately after World War II but was temporarily interrupted in 1949. From a wartime rate of less than $1 billion a quarter, U.S. imports rose in 1948 to a rate of nearly $2 billion, as supply conditions improved abroad (primarily in Europe) and U.S. demand increased. After a drop to $1.5 billion in 1949, the rise was resumed in the spring of 1950 and the value of imports, swollen by sharp price increases, rose to a new peak of $3.2 billion in the first quarter of 1951. A subsequent decline in both volume and prices brought the value of third quarter imports down to $2.6 billion.

The increase in the supply of dollars to foreign countries since 1949 has been followed, with a lag of about half a year, by a rise in the use of dollars, mainly to purchase U.S. exports. This lag was a major factor accounting for the exceptionally large volume of U.S. gold and dollar assets acquired by the rest of the world in 1950 and early 1951.

Lag between U.S. imports and exports

The dependence of world 2 imports from the United States (U.S. exports) on prior earnings from world exports to the United States (U.S. imports) is seen clearly if the U.S. export data are adjusted to exclude amounts financed directly or indirectly by U.S. Government aid. Such an adjustment is desirable because the greater part of Government aid is “tied” (that is, payment is made in connection with shipments), and a corresponding portion of U.S. exports can move without reference to the level of U.S. imports. A comparison between world exports to the United States and world imports from the United States decreased by U.S. Government aid is presented in Chart 2 for the 1948-51 period. For convenience in this discussion, these residual imports will be called “unaided imports.”3

Chart 2.
Chart 2.

“World” Exports and Imports Vis-à-Vis the United States, Quarterly, 1948–51*

Citation: IMF Staff Papers 1952, 001; 10.5089/9781451949391.024.A002

*“World” indicates the aggregate of all countries outside the United States. “Unaided” world imports are world imports from the United States decreased by U.S. Government aid.† Balance of “unaided” imports compared with exports two quarters earlier.

From mid-1948 to mid-1951, world commodity exports to the United States generally exceeded unaided world imports from the United States.4 In the second and third quarters of 1951, however, the export balance declined abruptly.

Much of this rise and decline in the unaided trade balance of foreign countries vis-à-vis the United States is explainable in terms of a lag of U.S. export changes behind U.S. import changes. Variations in the quarterly rate of U.S. imports from the rest of the world (in the third and fourth quarters of 1948, the third quarter of 1949, the middle of 1950, and the first quarter of 1951) were followed by similar movements in unaided world imports from the United States. To be sure, the lag was not always of the same length or clearly apparent in the trade of all major areas. Nevertheless, the hypothesis that variations in the dollar supply affect the use of dollars, with a lag, has been confirmed five times in the past three years, and should therefore also be taken into account in estimating the significance of more recent movements.

When unaided world imports are compared with world exports two quarters earlier, the lagged balance that results appears to have shown no definite tendency either upward or downward from 1949 through the summer of 1951. (See the lower section of Chart 2.) The principal fluctuations were in the third quarter of 1949, when unaided imports from the United States were abnormally low owing partly to emergency restrictions imposed by the Sterling Area, and in the fourth quarter of 1950, when such imports appear to have been unhampered by the temporary check in exports two quarters earlier. The irregularity of the lagged trade balance in the second and third quarters of 1951 reflects an apparent shortening of the lag response to only one quarter. On the whole, however, changes in the value of unaided world imports from the United States have, since 1948, followed remarkably closely changes in exports to the United States one or two quarters earlier.

Until 1951 the lagged trade balance, averaging around $300 million a quarter in favor of the rest of the world, was sufficient to cover net payments to the United States for services, including the service of U.S. foreign investments. In 1951, however, the tentative indication is that the lagged trade balance dropped by about $100 million a quarter, and that the services balance also was more unfavorable, so that there was a combined deficit of around $200 million a quarter.

This analysis of the lag behavior brings out the temporary nature of the balance of payments residual during periods of rising and falling trade, and makes it easier to estimate the impact of fluctuations in U.S. imports upon the world demand for U.S. goods. It suggests the conclusion, among others, that the 1950 experience did not represent a new basic trend toward world accumulation of gold and dollars; neither, on the other hand, did 1951 developments portend a catastrophic reversal of the 1948-50 trend.

In this analysis and the conclusions which follow from it, certain implicit assumptions cannot be avoided. For example, it is assumed that the lag observed in the past will continue in the future. Insofar as the lag depends upon the time required for changes in the dollar supply to become evident and for the administrators of governmental controls to adjust their policies to these changes, such an assumption seems reasonable. The conclusion that, because of the lag, the large surplus in the U.S. trade balance will be temporary assumes also that the decline of imports shown in the 1951 recorded data will be followed by governmental action abroad which will, in a few months, bring an appropriate further reversal in the balance. (Action of this sort has already been announced by several European countries.) Apart from official measures, the prospect of such a reversal is enhanced by the likelihood that world exports to the United States will turn upward again in 1952.

U.S. Government aid

The fluctuations in U.S. Government aid from 1947 to 1951 (Chart 1) are explainable largely in terms of the three major aid programs. The sharp decline from 1947 to 1948 reflected the declining importance of additional loans under the immediate postwar programs; the rise during the second half of 1948 and the subsequent decline in 1949-50 reflected the evolution of the European Recovery Plan; and the rise since the third quarter of 1950 reflects the growth of military aid, from about 4 per cent of the total aid program in early 1950 to over 35 per cent in the third quarter of 1951.5

Government aid not only increases the amount of U.S. exports, but also has important policy implications for the direction of world trade. If dollars made available through Government purchases of miscellaneous foreign goods and services and through disbursement of troop pay abroad are included, total funds subject to U.S. Government direction averaged, according to available information, about $1¼ billion a quarter in 1951. Stockpiling expenditures for imported goods may, before their sharp decline in the last quarter of 1951, have approximated a further $100 million to $200 million a quarter. In the aggregate, all these sums amounted to about 40 per cent of the total dollar supply (as defined above) currently accruing to other countries in early 1950. Later in 1950 the proportion dropped to around 25 per cent, but with U.S. aid increasing in 1951 the proportion again rose above 30 per cent. The significance of these figures lies in the fact that about three fourths of the aid is spent directly for U.S. exports, and thus immediately affects both the origin and the destination of a substantial portion of world trade.

Private capital movements

In recent years, U.S. private capital movements have been subject to relatively wide fluctuations since, in addition to small amounts of regular capital outflow, there have been occasional large transactions. The outflow in the third quarter of 1950 was especially abnormal, identifiable items totaling $609 million.6 If these items are deducted from the total private capital outflow for that quarter (Table 2), the remainder—about $225 million—would not be out of line with the totals for adjacent quarters.7 In the third quarter of 1951 the private outflow declined sharply, most of the reduction occurring in the flow to Canada and Latin America.

Table 2.

U.S. Private Capital Movements, Quarterly, 1950–51

(In millions of U.S. dollars)

article image
Source: U.S. Department of Commerce, Survey of Current Business, June 1951 and December 1951. Revised figures for errors and omissions published in the Survey for March 1952 raise the net credit for the third quarter of 1951 to $167 million.

“Errors and omissions” in the U.S. balance of payments with the rest of the world have often represented concealed capital movements. In the autumn of 1950, they appear to have been accounted for largely by increased commercial credit balances in the United States, resulting from a more or less normal lag in clearing payments for the unusual increase in September of merchandise trade imports and outgoing capital. The sum of the two increments was close to $1 billion, and even a short lag in clearing payments could readily account for the unusual rise in “errors and omissions” in the third quarter of 1950, and their subsequent fall in the fourth quarter.

The reason for the resurgence of errors and omissions in the first half of 1951 is not yet clear. It may be significant, however, that errors and omissions and the recorded net outflow of U.S. private capital fluctuated in such a way that their combined total declined steadily, until in the third quarter of 1951 there was a small net credit.

Private donations and services

The dollar supply to other countries from private U.S. donations was fairly steady in the early postwar period; it then declined gradually, however, from about $170 million a quarter in 1947 to around $100 million a quarter in 1951. Net foreign earnings of dollars on service account improved gradually from a negative figure (indicating net U.S. earnings) of about $180 million a quarter in 1947 to a positive figure averaging $100 million a quarter in 1951, the third quarter of the year typically showing a seasonal peak reflecting tourist travel.8

Acquisition of U.S. gold and dollar assets

The persistent lag between movements in world exports to the United States and in world imports from the United States has in turn produced fluctuations in the rate at which other countries have acquired U.S. gold and dollar assets. While world receipts of dollars fluctuated irregularly from 1947 through 1949, world purchases of U.S. goods tended to decline, and thus there was a gradual slackening of the rate at which gold and other monetary assets were sold to the United States. By the end of 1949 trade balances had improved sufficiently for the residual to become positive—that is, other countries, in the aggregate, began to acquire U.S. gold and dollar assets.

During 1950 and the first half of 1951 other countries increased their holdings of U.S. gold and dollar assets by about $4.7 billion.9 By the second quarter of 1951, however, the rate of addition was much lower; the third quarter brought nearly $300 million of gold and dollar assets back to the United States, and a further movement in this direction appears to have occurred in the fourth quarter.

While changes in the trade balance produce fluctuations in monetary reserves, the magnitude of the reserves itself is also a factor that affects the volume of foreign purchases in U.S. markets. The way in which this factor has operated may be illustrated by the changes in the ratio of unaided world imports from the United States to the world’s current supply of U.S. gold and dollar assets (exclusive of funds received through U.S. Government aid). In 1948 this ratio exceeded 100 per cent, i.e., expenditures of U.S. dollars exceeded the supply of current dollars accruing to other countries. The difference was met by drawing upon gold and dollar reserves, and there was subsequently a steady decline of purchases in the United States by other countries, both in absolute terms and as a percentage of the current supply of dollars. The ratio reached a low of 50 per cent in the third quarter of 1950. The higher rates of 75 and 94 per cent in the first and second quarters of 1951 suggest that the pressure to minimize U.S. purchases in order to restore reserve positions had been by that time somewhat relieved.10

Another factor which may have influenced the use of foreign gold and dollar earnings is the rise in U.S. (and world) export prices, which has reduced the purchasing power of those earnings. From the first quarter of 1950 to the middle of 1951, U.S. unit export values rose about 20 per cent; this percentage, applied to gold and dollar holdings of the rest of the world, would represent a loss of over $3 billion in their purchasing power.

Since the impact of trade shifts on monetary reserves becomes known only with a lag, and since there is a further lag before the volume of trade responds to changes in administrative controls, the gold and dollar flow tends to move, periodically, beyond the levels desired by the various Governments. This movement creates a danger of alternating excesses of optimism and pessimism, with consequent over-adjustments in Government controls which could make monetary reserves still more unstable.

Commodity Trade and Prices

Raw materials and semimanufactured products comprise a major part of U.S. commodity imports, while manufactured products provide the bulk of commodity exports. Accordingly, the U.S. trade balance is dominated by fluctuations in the quantities traded and the prices of these groups. This is illustrated by Chart 3 where the dollar values of commodity imports and exports during 1949-51 are shown on a ratio scale. In order to indicate the extent to which rising values have reflected rising prices, the chart also shows quantity data, i.e., imports and exports valued at constant (1947-49) prices.11

Chart 3.
Chart 3.

Value of U.S. Imports and Exports by Major Commodity Groups, Monthly Averages by Quarters, 1949–51

(logarithmic vertical scale)

Citation: IMF Staff Papers 1952, 001; 10.5089/9781451949391.024.A002

Although Chart 3 has been plotted from recorded total trade,12 which includes aid-financed exports, the lag of exports behind imports is clearly discernible. But because of the continued decrease in U.S. foreign aid, the recovery of exports after the first half of 1950 was at first slower than that of imports after mid-1949. Subsequently, when aid began to rise again, the rise in exports equaled that of imports two quarters earlier. The increase in the value of total exports from August 1950 to June 1951 was about 70 per cent, which was practically the same proportion as that for the increase in the value of total imports from February 1950 through January 1951. But owing to the adverse movement in the terms of trade during this period, the increase in the quantity of exports (43 per cent) exceeded that in the quantity of imports (34 per cent).

Value and quantity of U.S. imports

On the import side, two facts are most striking: the difference after mid-1950 between the movements of import values and import quantities, and the difference between the behavior of the quantities of imports of manufactures and of nonfood raw materials.

Until the second quarter of 1951, the trend of import values was strongly upward in all commodity groups. At its peak in the first quarter of 1951, the value of total imports was 60 per cent above the value a year earlier and almost double that of the depressed third quarter of 1949. On the other hand, total import quantities rose no more than 20 per cent above the first quarter of 1950 and 50 per cent above mid-1949. Secondly, while the rate of increase of import values in 1950-51 was more than double that from mid-1949 to early 1950, the rate of increase of import quantities in 1950-51 was some 20 per cent less than that from mid-1949 to early 1950. Clearly, in contrast to the recovery after the 1949 recession, the subsequent (post-Korean) boom in import values was largely a matter of prices.

The decline by 17 per cent of import values between the first and third quarters of 1951 was accompanied by a fall of 25 per cent in import quantities. These movements brought import values by September back almost to July 1950 (i.e., to pre-Korean) levels, and import quantities were only a little greater than in September 1949 (i.e., at a level below that of full recovery from the mid-1949 dip). The collapse of the immediate post-Korean boom has meant, at least so far, a more serious recession of import quantities, which thus showed themselves to be far more sensitive to the recent decline in American demand than they had been to the post-Korean increase. These conclusions are especially applicable to nonfood raw materials. The comparison between this group and imports of manufactures, summarized in Table 3, is striking.

Table 3.

Percentage Changes in U.S. Imports of Nonfood Raw Materials and of Manufactures, Selected Periods, 1949–511

article image

Based on data from U.S. Department of Commerce, Office of International Trade, Total United States Export and Import Trade (International Trade Statistics Series), and Bureau of the Census, Foreign Trade Reports (FT930).

Several observations are suggested by an examination of Table 3. First, the changes in imports of nonfood raw materials and of manufactures over the whole year preceding the outbreak of hostilities in Korea were approximately equal, both in value and in quantity. But the total effect of the domestic U.S. recovery took place in two steps: the revival of imports of raw materials was concentrated mainly in the first, and the revival of manufactured imports in the second, half year.

Second, the post-Korean impact on imports also proceeded in two steps, again affecting the two commodity groups in the same order. During the first phase, however, rising import demand (as represented by the import value figures) drove up raw materials prices sharply, while prices of manufactures hardly moved. In the second phase, the rebound from the high import demand occurred earlier and was sharper for raw materials than for manufactures; manufactured import prices were indeed still rising in September 1951, while raw materials prices had been falling since June.

Third, the output of manufactured goods can normally be increased more rapidly than the output of raw materials, and certain industrial countries were, for the first time since the end of the war, making an effective entry into world markets. For these reasons more manufactured imports were available, and if quantity figures are accepted as an indication of the supply conditions of U.S. imports, this meant that a relatively continuous increase in import quantities could be supported at relatively moderate price rises, while tight supply conditions produced quite the contrary result for raw materials.

Thus in both the rising and falling movements of total imports, manufactured imports reacted later and showed larger upward and smaller downward changes in quantity.

The difference between the reactions of nonfood raw materials and of manufactured imports appear to be the combined result of (1) the difference in timing between the impact of increased domestic demand on the two commodity groups, and (2) the difference in their supply conditions.

Fourth, the over-all import effect of the post-Korean developments, even in value terms, has so far been relatively small; this has been in the main a temporary, and chiefly a price, boom. The adjustment to domestic post-Korean developments is, as will be shown later, still incomplete.

Imports of food raw materials are in a class by themselves. Most of their price rise (dominated by that of coffee) occurred in 1949 and the first half of 1950. Significant quantity changes, apart from seasonal movements, were almost completely absent, though the effects of the “buying spree” in the third quarter of 1950 and the first quarter of 1951 are clearly visible. Otherwise, this import group shows practically no response to domestic income changes. The general trend of food raw materials imports is only slightly upward, and their import values are almost completely determined by their price increases.

U.S. imports and domestic activity

The developments in U.S. domestic economic activity that form the background for the rapid changes in the U.S. trade balance, which have just been discussed, may be broadly summarized as follows. Recovery from the recession in mid-1949 lifted U.S. gross national product from a low of about $255 billion to $280 billion by mid-1950. The events following the outbreak of war in Korea raised it by another 16 per cent, to $325 billion by mid-1951. In the 1949-50 rise, private domestic investment, rising by 60 per cent, played the dominant role; personal consumption expenditure reacted very much less sharply, rising only 8 per cent; and there was actually a decrease of almost 10 per cent in the Government sector. In the 1950-51 phase, the roles were reversed: the Government sector increased by more than 58 per cent, private investment by 30 per cent (with a sharp decline in the third quarter of 1951, however), and personal consumption by only 5 per cent, interrupted by two temporary peaks in the third quarter of 1950 and the first quarter of 1951. The accompanying rise in industrial production was also sharper in the second phase (21 per cent) than in the first (11 per cent), while the price rise in the first phase (12 per cent) was almost twice that in the second.

Imports of raw materials may, from past experience, be expected to move in close relationship to the volume of industrial production, while imports of consumer or near-consumer goods are affected by domestic consumer expenditures. In an economy like that of the United States, which—to a larger extent than other countries—produces the same products that it imports (or close substitutes for them), differences between domestic and foreign prices would be expected to influence imports significantly. In fact, however, as the following discussion will show, only a comparatively narrow range of imports (mainly manufactures and semimanufactures) seems to possess a noticeable price sensitivity in the short run.

For imports of raw materials, the effects of the basic relationship to industrial production are overshadowed by those of other forces. The influence of these forces may be indicated, to some extent, by movements of domestic inventories, since excessive imports (i.e., excessive in relation to domestic import-absorbing activity) should add to inventories and excessive inventories should discourage future imports. Inventory movements thus result in part from previous import movements, and in part cause future import movements.

The relatively large inventories of purchased materials and goods in process which manufacturers held in the first quarter of 1949, after industrial production had already begun to decline, are probably at least a partial explanation of the subsequent drop in imports of raw materials, which proceeded at a rate considerably greater than the rate of decline for industrial output. Since the decrease in inventories did not cease until the third quarter of 1950, when industrial production had already begun to rise sharply, imports were stimulated to rise more than industrial output. The very rapid rise of imports in the second half of 1950, however, was accompanied in the late months of the year by a quite sizable increase in raw materials stocks. This increase appears to have been a factor that contributed to the decline in raw materials imports in 1951. The rise and subsequent reduction in imports of wool, rubber, and the major nonferrous metals accounted, to a considerable extent, for the movements of raw materials imports as a whole; in the markets for these commodities, either U.S. Government purchases were of major importance or U.S. domestic price ceilings and a general redirection of U.S. stockpiling policy in 1951 limited U.S. imports.

Raw materials imports might be expected to rise more sharply than industrial output in an economy with full employment which is subjected to the additional strain of armament expansion. In 1950 and early 1951, this was particularly true for semimanufactured products (which in our classification are included in raw materials), such as metal manufactures and industrial chemicals. Later, however, as domestic output of these materials rose to a level more in accord with domestic demand, imports began to fall off. This substitutive effect is an additional explanation of the reversal of the import movement in 1951.

Another cause of the sharp changes in imports—first upward, and then, after early 1951, downward—was, undoubtedly, speculative importing.

Consumers’ goods imports into the United States are dominated by crude foods, mainly coffee, cocoa, tea, and sugar. Except for the “buying sprees” in the third quarter of 1950 and the first quarter of 1951, the reaction of the quantities of crude food imports to prices and to domestic consumption expenditures on nondurable goods was quite small, so that their import values moved in close relation to rising import prices. In the case of other consumers’ goods, however—manufactured foods (excluding sugar) and manufactures other than food (excluding machinery)—importers were able to supply a larger proportion of the U.S. market; the value of each group doubled between mid-1949 and the second quarter of 1951. The increases for the two groups, which are composed primarily of marginal (or even, in part, “luxury”) imports, are not only a function of domestic income and foreign availabilities; they depend to a major degree on the success of attempts to establish marketing, distribution, and servicing facilities, since competition in U.S. markets for these goods is highly imperfect. The set-back in these two groups of imports since the first quarter of 1951 has accordingly been relatively small.

As might have been expected, the effects of price differentials, insofar as they can be measured by an over-all comparison of import values and domestic wholesale prices, were probably strongest for manufactured imports. At least through 1950 the prices of imported manufactures appear to have risen less than the corresponding domestic wholesale prices of foods and nonfood products. But as domestic prices turned downward from their peak in the first quarter of 1951, the price advantage of imports of manufactures diminished and thus must have been a contributing cause of the eventual downturn of manufactured imports.

From the foregoing discussion it appears that the first stage in the foreign trade developments which followed the decision that armament output must be more or less permanently increased consisted of a temporary, in part speculative, overimporting of raw materials followed by a reaction which appears to have gone too far. Though the quantities of raw materials imports cannot be expected to remain at their September 1951 low level, they are equally unlikely to revert to their high of the first quarter of 1951, unless domestic output rises considerably. Imports of manufactured consumers’ goods will depend on foreign supply, which may be increasingly restricted by armament output expansion abroad, on relative domestic-foreign price developments, and on the continued success of the marketing efforts of foreign suppliers in the United States. Likewise, the volume of crude food imports is almost completely dependent on foreign supply.

Delayed movement of U.S. exports

The two food components of total exports, though the smallest of the main groups in absolute amount, fluctuated much more violently than the other groups during the period 1949-51, because of the reappearance of a demand for U.S. grains in Europe and Asia, following a period when in both these areas grain imports had not been needed. There was at the same time a sharp increase in the unit values of food exports, which in 1949 had consisted to a large extent of U.S. surplus commodities, subsidized by U.S. Government agencies and sold abroad at prices below world market levels. The value of wheat and wheat flour exports rose from $129 million in the second quarter of 1950 to $319 million in the second quarter of 1951. On the other hand, the quantities of nonfood raw materials exports (raw cotton, coal, tobacco, and petroleum) fell slightly. Their rise in value terms was due to a price rise (almost 35 per cent between the first quarter of 1950 and the second quarter of 1951), which was three times that for the value of total exports and was caused mainly by a rise in cotton prices resulting from the short 1950-51 U.S. cotton crop. Since cotton prices fell by as much as 15 per cent from the middle of the second quarter of 1951 through the third quarter, there was a small decline in the value of nonfood raw materials exports in the third quarter; the quantity exported, however, actually increased between the beginning and end of the quarter. Another important factor in sustaining the value level of this export group in 1951, even against the normal seasonal downward trend, was the rise in coal and petroleum exports. As a result of increasing coal shortages in Europe, the value of U.S. coal exports rose from $88 million in the third quarter of 1950 to $195 million in the third quarter of 1951, and the increased reliance of foreign users of petroleum and its products on North American rather than Middle Eastern suppliers raised U.S. exports of petroleum and petroleum products from $125 million in 1950 to a level of $250 million in 1951.

Apart from these export groups, whose behavior can be explained by such special causes as have just been mentioned, most U.S. exports (about 70 per cent) consist of a wide variety of nonfood manufactures. The lag of exports behind imports is seen most clearly in this group. Even in the third quarter of 1950 neither the value nor the volume was significantly above the first quarter low; unit values were still falling slightly, and no pronounced increase was recorded until September. The upward movement was subsequently very rapid, paralleling the rise of imports which began in the third quarter. By the end of the second quarter of 1951, values had risen by 80 per cent, quantitities by 54, and prices by 18.

Future developments of U.S. exports are still very dependent on foreign aid policy. An increasing use of aid to finance U.S. raw materials exports will necessarily make exports of nonfood manufactures more dependent on military or quasi-military aid.

Terms of trade

During the postwar years, the terms of trade of the United States have deteriorated steadily; this means that the cost of imports in terms of exports has risen with very little interruption. Between 1936-38 and 1946, U.S. import prices rose 13 per cent more than export prices, and by the end of 1948 they were 24 per cent higher. The sharp decline in imports which accompanied the domestic U.S. recession in 1949, producing an equally pronounced rise in the U.S. export surplus, depressed import prices only slightly more than export prices. Thus, the consequent single major interruption of the worsening of U.S. terms of trade, in the first half of 1949, improved the terms of trade by no more than 2 per cent (see Table 4).

Table 4.

U.S. Export and Import Prices and Terms of Trade,1 January 1949-October 1951

(1947-49 = 100)

article image

Terms of trade = price of exports divided by price of imports.

Source: U.S. Department of Commerce, Office of International Trade, Total United States Export and Import Trade (International Trade Statistics Series).

It is remarkable that the 1949 devaluations apparently did not affect significantly the U.S. terms of trade. The continued deterioration in the last quarter of 1949 and the first half of 1950 was indeed at best arrested in November and December. Neither before nor after those two months had import prices declined noticeably, and by the end of the first quarter of 1950, the terms of trade were 6 per cent worse than before the September devaluations.

From mid-1949, the terms of trade moved against the United States for the next 26 months almost without interruption: the extent of the total deterioration was 22 per cent. To describe the causes of this deterioration, this period may be divided into three subperiods: from July 1949 through July 1950; from August 1950 through June 1951; and from July 1951 onward. It may be seen from Table 4 that the worsening of the terms of trade in the first of these subperiods was due to a rise of import prices (by 9 per cent) and a fall of export prices (by 6 per cent); in the second subperiod, both sets of prices rose, but imports (+ 33 per cent) more than exports (+21 per cent); and in the third, both fell, but by almost the same proportion (imports —3.5 per cent; exports —4.5 per cent), until October when export prices showed no change.

The impact of post-Korean developments on U.S. trade marked the beginning of the second subperiod; no imports which were ordered in consequence of the Korean outbreak could have reached the United States before August 1950. It is remarkable, therefore, that the deterioration of the U.S. terms of trade during the first subperiod was actually greater than that following the Korean invasion: from June 1949 to July 1950, the terms worsened by 14 per cent, while the additional deterioration through July 1951 was no more than 10 per cent.

The explanation is to be found in three sets of facts. First, the prices of U.S. exports, whether manufactures or raw materials, are determined by domestic rather than foreign demand, while the prices of the major U.S. imports—i.e., raw materials—follow demand for them from outside the countries of their origin. Secondly, there is a fairly regular lag of export behind domestic wholesale prices, and of import behind foreign spot prices. Thus, export prices continued their decline beyond that of domestic wholesale prices (which had fallen during the domestic inventory readjustment earlier in 1949), through the first quarter of 1950, and import prices began to show the effects of increased U.S. import demand late in 1949. Thirdly, this increase in demand concentrated on raw materials, and special supply difficulties (particularly for coffee) produced a strong and early rise in crude food import prices (see Chart 4). This rise offset the decline over most of the first subperiod in the prices of imported manufactures and semimanufactures, U.S. demand for which developed later than that for crude materials. The result of all this was a worsening of the terms of trade in the first subperiod greater than would otherwise have been expected.

The terms of trade in the second subperiod, i.e., the year immediately following the Korean outbreak, were dominated by the two forces, working in opposite directions, already mentioned (see Chart 4). On the import side, prices of imported nonfood raw materials rose by 46 per cent, while crude food prices increased 10 per cent between July and September and then stabilized; at the same time, semimanufactures prices rose by 33 per cent, and prices of finished manufactures, moving less and somewhat later, rose by 15 per cent. U.S. export prices, on the other hand, following domestic wholesale prices with a lag of 2-3 months, rose by 19 per cent. The sharp rise in export prices of nonfood raw materials (22 per cent), which accompanied the short domestic cotton crop, was an important factor in this general export price movement. The result was that the deterioration of the terms of trade, though rapid through March 1951, was slight for the last months of this period.

Chart 4.
Chart 4.

Indices of U.S. Import and Export Prices by Major Commodity Groups, Monthly, 1949-51

(1947-49 =100)

Citation: IMF Staff Papers 1952, 001; 10.5089/9781451949391.024.A002

When the movement of domestic wholesale prices and the drop in the cotton price are considered, the slight improvement of the terms of trade since July-August 1951 may be attributed mainly to falling import prices of raw materials, which represents the effect of the drop in spot prices of major U.S. import commodities earlier in the year. The sharp worsening of the U.S. terms of trade during a period when imports were expanding very rapidly and exports (in spite of substantial foreign aid) were not keeping up with imports, suggests the extraordinary degree to which U.S. trade is independent of relative price movements. It has already been pointed out that only a small sector of U.S. imports is noticeably price sensitive. In addition to price movements, it required a slowing down of the rate of increase of internal activity plus determined Government intervention in raw material import markets and increases in foreign aid to produce a noticeable reaction of U.S. imports and exports in accordance with the worsened terms of trade.

In view of the divergences between the price movements for different commodity groups, as well as between the timing of their shipments to the United States, some diversity may be expected in the terms of trade of the United States vis-à-vis different areas. Price data for exports to and imports from particular countries or regions are not readily available, but a rough notion of some of the differences may be obtained from the partial terms of trade computed for major commodity groups. A comparison of the results of such computations (Chart 5) indicates the general magnitude of the movements in the terms of trade of some areas.

Chart 5.
Chart 5.

Indices of U.S. Import and Export Prices (1947–49 = 100) and Terms of Trade by Major Commodity Groups, Monthly, 1949–51

Citation: IMF Staff Papers 1952, 001; 10.5089/9781451949391.024.A002

The upper right section of the chart shows movements in the terms of trade as between total raw materials imported into the United States and total manufactured exports. Even though U.S. export prices fell, the U.S. terms of trade, interpreted in this way, improved by almost 5 per cent in the second and third quarters of 1949, but then deteriorated steadily. In the first and second quarters of 1951 the deterioration reached a maximum of around 20 per cent, compared with a year earlier; by September, there was an improvement of only two percentage points over the June 1951 figure. These terms would indicate the balance of advantage and disadvantage of the trade between the United States and the less developed areas in the rest of the world as a whole.

Significant differences appear if separate calculations are made for food and for nonfood raw materials. Import prices of nonfood raw materials declined by about 15 per cent during 1949 and remained low in the first half of 1950 (see middle section of Chart 5). Since export prices of U.S. nonfood manufactures fell by only 7 per cent in 1949, the terms of U.S. trade vis-à-vis the suppliers of such raw materials seem to have improved in that year, as the chart shows. When import prices later rose by more than 50 per cent, the terms of trade deteriorated rapidly.

Prices of crude food imports, on the other hand, had already begun to rise in 1949 (bottom section of Chart 5). This rise, dominated by the advance of coffee prices, was the result of tight supply conditions combined with rising U.S. personal incomes. The terms on which the United States traded nonfood manufactures for crude food imports had, therefore, already worsened by the middle of 1950 by more than 25 per cent compared with a year earlier. The first post-Korean domestic buying rush moved the terms of trade against the United States by an additional 10 per cent; in October 1950 import prices began to fall and prices of U.S. manufactured exports began to catch up, and by April 1951 the terms of trade had improved by almost 15 per cent. From May 1951 onward, when U.S. export prices began to recede in conformity with domestic wholesale prices, the movement was arrested; through August the terms of trade worsened very little but in September there was some slight improvement. The terms had again reached the same level as just before the Korean outbreak, and it appeared that further improvement could be expected.

The terms of trade with the United States of Latin American countries which are important food raw materials producers must have fluctuated in the manner indicated by these price movements. A U.S. Department of Commerce tabulation13 which shows U.S. terms of trade with all Latin America deteriorating by 19 per cent from the first to the third quarter of 1950, improving 3 per cent by early 1951, and then worsening again, supports the view that the lower portions of Chart 5 may be accepted as presenting a fair picture for this area. Individual countries will have deviated from the area average presented in the Commerce figures to the extent to which they were nonfood and food raw materials suppliers.

In the same way the terms of trade for nonfood raw materials imports vis-à-vis nonfood manufactured exports may be taken as broadly representative of the position of certain sectors of the Outer Sterling Area and countries such as Indonesia and Thailand, with similar considerable variations for individual countries. For example, the terms of trade of Australia deteriorated first by 4 per cent from January to June 1950, and then by a further 62 per cent in the following 9 months; on the other hand, the terms of trade of India deteriorated 14 per cent from January to June 1950, but then improved 31 per cent in the next 9 months.

The terms of trade vis-à-vis Europe are more difficult to estimate because of the greater diversity of U.S. exports to that area. The increase in European exports to the United States from the first to the second half of 1950 was concentrated in such semimanufactured products as metals and manufactures, textiles and manufactures, and chemicals, as well as in finished manufactures. Unit values of imports of semimanufactured products rose about 20 per cent, and of manufactures 10 per cent, during the second half of 1950, and another 12 and 16 per cent, respectively, in the first half of 1951. If it can be assumed that U.S. exports to Europe, apart from manufactured items provided through U.S. Government aid, were mainly raw materials, the terms of trade of the United States vis-à-vis Europe may have improved by 1 per cent in the first of these half years and then deteriorated by almost 10 per cent in the second. The latter deterioration occurred in spite of the temporarily and artificially depressed prices of U.S. food exports in 1950 which was mentioned above.

While the over-all terms of trade moved against the United States in 1950-51, this general movement covered very diverse movements for particular countries. For countries where food raw materials are of dominant importance, the pre-Korean advantage of the supplying areas was appreciably reduced as prices of manufactured goods began to overtake prices of food. This development appears to have been in accordance with the normal time lag between fluctuations in prices of raw materials and of manufactured goods. Elsewhere, where nonfood raw materials are important, the advantage to the supplying area appears to have arisen mainly after the Korean outbreak through a bidding up of prices to a point where buyer resistance was encountered. Since the subsequent recession in world prices of such materials is reflected in trade values, these advantages enjoyed by the raw materials areas will also tend to decline. On the other hand, although prices of U.S. imports of semimanufactured and finished goods are unlikely to recede, in view of strong inflationary pressures in the supplying countries, the expected tightening of supply there is likely to cut the volume of U.S. imports of these commodities and thus reduce the more recent price advantages of their suppliers.

Area Distribution of Trade

During the first three quarters of 1950, when rising U.S. income attracted an increasing volume of imports, the sources of imports were largely determined by the supply capabilities of other countries. On the export side, the destination of the declining volume of shipments was influenced by the restrictive policies adopted by some countries in 1949, and by the reduced need for U.S. supplies which was reflected in the declining amount of U.S. financial aid. By the third quarter of 1950, U.S. imports from Europe, the Sterling Area, Latin America, and Asia had in each case reached their highest postwar point while, conversely, U.S. exports had just passed their lowest point. An important exception was Canada, where an investment boom, adequate monetary reserves, and relaxed import restrictions combined to attract larger exports from the United States. These trends were interrupted in the second half of 1950 and in early 1951 by the impact of the Korean war, which increased both U.S. imports and exports, although in varying degrees and with differences in timing.

Sources of U.S. imports14

Changes in the behavior of U.S. imports from four major areas, and also from “underdeveloped areas” (which include Latin America, Outer Sterling Area, Eastern Europe and Greece and Turkey, and the underdeveloped non-sterling areas exclusive of Japan, in Asia and Africa) are shown in Chart 6. As a result of domestic U.S. developments, imports from all areas in the second half of 1949 recovered rapidly from their mid-1949 low; this recovery was followed by a lull which extended into the second quarter of 1950. The expansion was renewed even before any increased foreign demand consequent on the Korean outbreak could have become effective. There are some remarkable differences in the reactions of imports from the various areas to domestic trends of the second and third quarters of 1950 in the United States. Imports from the ERP countries showed the most pronounced rise, a sharp advance in August indicating a response to U.S. demand more violent than for any other area. Sterling Area imports show the second largest over-all rate of increase, followed by imports from the Latin American Republics (showing the effects of the violent fluctuations in coffee and sugar imports mentioned above) and from Canada. Imports from underdeveloped areas in general (of which the Outer Sterling Area and the Latin American countries are the major components) rose much more slowly than imports from the European industrial countries.

Chart 6.
Chart 6.

Value of U.S. Imports from Major Areas and from “Underdeveloped Areas,” Monthly Averages by Quarters, 1948–51

(logarithmic vertical scale)

Citation: IMF Staff Papers 1952, 001; 10.5089/9781451949391.024.A002

*“Underdeveloped Areas” include Latin America, Outer Sterling Area, Eastern Europe and Greece and Turkey, and underdeveloped non-sterling areas, exclusive of Japan, in Asia and Africa.

The area distribution as shown in Table 5 is generally consistent with our commodity group analysis in the preceding section. The unusually sharp increases in imports of semimanufactures and, though with a slight lag, in finished manufactures explain the record of the ERP countries. Nonfood raw materials imports account for the behavior of Sterling Area imports, while imports from Latin America are dominated by the crude foods group. Nor is it surprising that the reduction of U.S. imports after March 1951 was of catastrophic proportions for the Sterling Area whose three major exports to the United States, rubber, tin, and wool, suffered most severely from the decrease in U.S. demand. A similar precipitate decline pulled Latin American exports to the United States down to pre-Korean levels. Although this was in part seasonal, the same forces were also at work here as in the Sterling Area. On the other hand, as imports of manufactures survived relatively well, and the need in the United States for imported semi-manufactures was only beginning to decline even in the third quarter of 1951, Western Europe was relatively little affected.

Table 5.

Sources of U.S. Imports, Quarterly, 1950-51

(In per cent of total imports)

article image

Includes Latin America, Outer Sterling Area, Eastern Europe and Greece and Turkey, and the underdeveloped non-sterling areas, exclusive of Japan, in Asia and Africa.

Source: Compiled from U.S. Department of Commerce data.

Western Europe: Imports from Western Europe averaged $81 million per month in 1948, then dropped to $60 million in the second and third quarters of 1949, and rose again during the first half of 1950 to a monthly rate of $80 million. After the outbreak of hostilities in Korea, there was an increase of almost 90 per cent, to an average of about $150 million a month in the last quarter of 1950, another 25 per cent rise to a peak of $187 million in March, and then a drop to a monthly average of $150 million in the third quarter.

The rise in U.S. imports during the second half of 1950 was shared by all Western European countries, and everywhere the level of U.S. purchases was higher than that recorded before the effects of the U.S. recession of 1949. In the first quarter of 1951, imports from one group of countries (Italy, Switzerland, and the Netherlands) began to decline, while imports from Germany, the “all other” group (mainly Scandinavia), France, Belgium-Luxembourg, and the United Kingdom rose further, and declined only in the second and third quarters, imports from the United Kingdom as well as from Eastern Europe (mainly Yugoslavia) being less affected by this decline than those from the other countries. Thus, the exports of countries in the first group never rose substantially above their pre-Korean levels, while the exports of the second group were, in the third quarter of 1951, above those levels, by percentages ranging from 25 (U.K.) to several hundred (Germany, France).

A further striking fact is that the seven countries which are shown separately in Table 6 increased their shipments to the United States by nearly 140 per cent, while their shipments to the rest of the world rose by only 34 per cent. Except for Belgium-Luxembourg and Switzerland, this rise in the proportion of West European exports going to the United States continued through the third quarter of 1951.

Table 6.

Principal U.S. Imports from Selected European Countries, Quarterly Totals, 1950-51

(In millions of U.S. dollars)

article image
Source: U.S. Department of Commerce, Bureau of the Census, U.S. Imports of Merchandise for Consumption (Report No. F120).

Consumer goods accounted for most of the rise in U.S. imports in the third quarter of 1950, reflecting the influence of rising U.S. income payments in previous quarters. By the last quarter of that year, however, the increase in European shipments was being augmented heavily by semimanufactures, such as steel mill products and chemicals, and these showed the largest percentage increases during the next year (see Table 6). The importance of these materials for rising U.S. defense production no doubt goes far to explain the continued high level of shipments in 1951 from the United Kingdom, France, Germany, Belgium-Luxembourg, and the Netherlands. Shipments from Italy, Switzerland, and some Scandinavian countries, on the other hand, were not so well maintained in 1951 as in the fourth quarter of 1950. But even in these countries, imports of metals, steel mill products, machinery and vehicles, and industrial chemicals were still rising when total imports had already begun to fall. In addition—and this applies especially to Germany—the continued sharp rise of shipments to the United States followed continued domestic progress outside the United States which made more goods available for export. There were also some special reasons for the continuance of high levels of certain exports, such as the French chemical exports, to the United States. Nevertheless, it appears that the peak of the post-Korean boom was passed, even for European exports to the United States, by the third quarter of 1951.

Canada: U.S. imports from Canada, after a recovery from the mid- 1949 decline which was somewhat sharper than that of imports from the other areas, showed only a very moderate post-Korean reaction. The mixed behavior of imports of Canadian origin, such as newsprint, woodpulp, lead, and zinc, furnishes an explanation. The increase of imports from Canada ceased as early as September 1950.

Latin America: Latin America, which was only slightly affected by either the U.S. recession in 1949 or the developments consequent on the Korean outbreak (Chart 6), may be subdivided either by major types of exports to the United States or according to whether dollars are the major means of international settlement. The exports of nonfood suppliers, as would be expected, were more sensitive both to the 1949 recession and to the Korean stimulus (Chart 7). These countries, plus Brazil and especially Argentina and Uruguay, i.e., the main nondollar countries, also account for the larger part of the serious decline of imports into the United States after March 1951.

Chart 7.
Chart 7.

Value of U.S. Imports from Raw-Material Supplying Areas, Monthly Averages by Quarters, 1948-51

(logarithmic vertical scale)

Citation: IMF Staff Papers 1952, 001; 10.5089/9781451949391.024.A002

* Underdeveloped areas outside Latin America and the Sterling Area.

Closer inspection of individual country data indicates considerable diversity in each group, however, and an adequate explanation of the differences would require reference to the market conditions of the particular commodities which dominate the dollar earnings of particular countries. Thus, in the nondollar group much of the sharp increase in shipments seems to be a consequence of the wool and coffee price rises that followed the rise in U.S. imports at the end of 1949 and again in the first half of 1950 (Brazil, Argentina, Uruguay, but also Colombia and Central America). The sharp reduction in U.S. imports of these two commodities that began in the second quarter of 1951 reversed the movement of imports from these countries. In the dollar group, the peak for U.S. sugar imports in the third quarter of 1950 and the subsequent decline are reflected in corresponding fluctuations in Cuba’s shipments to the United States, while the steady level of U.S. petroleum imports up to June 1951 explains the absence of any post-Korean rise in Venezuelan shipments.

Other raw materials supplying areas: Next to imports from Western Europe, those from the non-Latin American raw material supplying areas rose most rapidly in 1950-51. Imports from the “other underdeveloped areas” (i.e., underdeveloped areas outside Latin America and the Sterling Area) had in mid-1950 just exceeded their 1948 level; subsequently, sharp price increases of their major export products (rubber and tin in particular) raised their exports to the United States by 80 per cent (January and April 1951). By July, however, there had been a fall of 22 per cent, and after a temporary interruption, a further fall of 25 per cent by September. The Outer Sterling Area shows much the same record, which, however, conceals great divergencies in its component parts. Imports from the three main wool supplying countries—Australia, New Zealand, and the Union of South Africa—illustrate the rather late reaction to the pre-Korean boom; there was a short-lived but violent advance, reaching by April 1951 a peak level fully 160 per cent above the last pre-Korean month. The collapse of wool imports forced an abrupt decline in September to a level of only one third of this peak. The remainder of the Outer Sterling Area imports were by January 1951 80 per cent above those in June 1950; the subsequent movement was downward, and very much accelerated after July. The full force of post-Korean U.S. demand increases thus had its impact first on the non-sterling raw materials suppliers, next on the nonwool-supplying Sterling Area, and last on the wool suppliers. As the analysis in the preceding section has indicated, the greater part of the rise in import values of nonfood raw materials from these areas was a result of price increases, quantities at first rising only moderately, and later receding to their early 1950 levels.

Destination of U.S. exports

The general character of their economies determines the type of product imported by other countries from the United States, the industrial areas taking relatively more raw materials while the less-developed areas purchase relatively more manufactured goods. This differentiation is illustrated in Table 7, which shows the leading U.S. exports in 1950 to the major areas. Although the classification is incomplete, it correctly indicates that about half of all U.S. raw material exports go to Europe, and that most of the rest are taken by Canada and the Far East. On the other hand, about two thirds of the manufactured products go to Latin America and Canada, which are the rapidly developing areas with substantial buying power. The Far East includes both industrial countries and raw material suppliers, and accordingly absorbs both raw materials and manufactures in roughly the same proportions.

Table 7.

Principal U.S. Exports to Major Areas, 1950

(In millions of U.S. dollars)

article image

Included in “all other.”

Cotton only; tobacco included in “all other.”

Only rice, flour, and other grain and vegetables.

Machinery only; automobiles included in “all other.”

Source: U.S. Department of Commerce, Foreign Commerce Weekly, April 16, 1951. Data exclude “special category” exports purchased for cash and MDAP (Military Defense Assistance Pact) shipments. Figures shown by commodities do not represent a complete classification of such exports, since exports which are relatively minor for a particular area are included in the “all other” category.

During 1950 and 1951, the area pattern of U.S. exports was rather more stable than the area pattern of U.S. imports. As Table 8 shows, 34 per cent of U.S. exports went to Europe in the third quarter of 1951, while the proportion to Latin America was 27 per cent, both approximately the same as in early 1950. Canada’s proportion increased at first from 17 to 21 per cent, mainly because the upturn in Canadian buying began earlier than that of other areas, only to recede again in later quarters. The proportionate share of shipments from the United States to Latin American countries, seasonal changes apart, remained practically constant, dollar countries, however, losing to the nondollar countries. This parallels the difference previously noted with respect to imports from these two subgroups. The share of the Outer Sterling Area and of other countries in Asia and Africa declined during the first three quarters of 1950, and this was apparently the main factor accounting for the decline in the share of the underdeveloped group as a whole, from 46 per cent in the first quarter of 1950 to 43 per cent later in the year. The large dollar earnings of these areas in 1951 enabled them, and especially the Outer Sterling Area, to raise their share beyond the level of late 1950.

Table 8.

Destination of U.S. Exports, Quarterly, 1950-51

(In per cent of total exports)

article image

Includes Latin America, Outer Sterling Area, Eastern Europe and Greece and Turkey, and the underdeveloped non-sterling areas, exclusive of Japan, in Asia and Africa.

Source: Compiled from U.S. Department of Commerce data.

To some extent the shifts in the proportions of U.S. exports going to major areas is explainable by the directional effects of the U.S. Government aid program and export controls; they also reflect policy differences among the buying countries with respect to their use of currently accruing dollar earnings.

Directive effect of Government aid

International aid extended by the United States, whether for economic development, balance of payments, or military purposes, tends to influence the flow of international trade in two ways: it increases the proportionate share of internationally traded goods going to the aided countries, and it tends to increase the proportion of their trade supplied by U.S. markets.

From 1947 to 1950, as Table 9 shows, both U.S. exports and U.S. Government aid tended to fall. The lowest quarterly figure for exports, $2.4 billion, was recorded in the first quarter of 1950; they were somewhat higher in the second and third quarters; and subsequently they rose sharply, to $4.1 billion by the second quarter of 1951. The increases in the last quarter of 1950 and the first half of 1951 were aided by Government financed shipments, which rose from $0.9 billion in the third quarter of 1950 to $1.3 billion in the second quarter of 1951.

Table 9.

U.S. Exports and U.S. Government Aid, 1947-51

(In billions of u.s. dollars)

article image

Annual rates compiled from data for first three quarters.

Source: U.S. Department of Commerce, Survey of Current Business, selected issues.

About three fourths of U.S. Government aid over the past four years has gone to Europe, although the amounts going to the Far East also have been substantial. During this period as a whole, about 70 per cent of U.S. aid appears to have been “tied” to U.S. exports in the sense of directly financing shipments of goods and services originating in the United States.15 The “tied” aid declined from 40 per cent of total exports in 1949 to about 25 per cent in the latter part of 1950 and early 1951. As the military assistance program expands, an increasing proportion of total aid is likely to be “tied” aid going to European countries, but it will be more widely distributed within Europe than heretofore.

For some individual countries and areas, the value of their imports from the United States is directly dependent upon the amount of “tied” aid which they receive. Thus, for all ERP countries, “tied” aid provided 60 to 80 per cent of their purchases from the United States in 1950-51; the fluctuations in the two series, aid and U.S. exports, therefore were necessarily very similar in those years. The relative importance of “tied” aid declined, however, during the latter part of 1950, when the ERP countries began to make increasing amounts of unaided purchases in the United States; in the fourth quarter of 1950, “tied” aid accounted for 66 per cent of total exports, and in early 1951 for 62 per cent.

Widening area of Government control

During 1950 and 1951 the influence exercised by the U.S. Government in U.S. international transactions increased significantly. This influence took various forms which are significant for third countries as well as for the countries directly affected. Direct controls increasingly encroached on the field in which price and income differentials determine the flow of international trade. Instead of the controls being in the hands of a single Government, responsibility for some of them has recently been shared by international organizations, such as the International Materials Conference and the North Atlantic Treaty Organization.

U.S. export controls

Under the Export Control Act of 1949 (now extended through June 1953), four types of U.S. export controls have been imposed:

  • (1) An embargo which has been in effect since June 1950 on all shipments to North Korea.

  • (2) The establishment of a “positive list” of commodities which can be exported only under license. This was gradually expanded from less than 700 items in the second quarter of 1950 to about 1,400 items in early 1951, of which about two fifths apply to destinations outside the Western Hemisphere, and the other three fifths to all destinations except Canada. (Exports to Canada remain uncontrolled.) No licenses have been issued for exporting positive list commodities to China since July 20, 1950, and none for any commodity to the Chinese mainland since December 3, 1950, when all commodities for that area were brought under license control. This form of control was extended in the first quarter of 1951 to shipments for the Soviet bloc.

  • (3) Quantitative export limits for an increasing number of commodities, of which raw cotton, sulphur, the major nonferrous metals, and certain petroleum products have been the most important.

  • (4) Priority ratings for foreign orders (up to one sixth of the value of corresponding 1950 exports), beginning in May 1951, to provide friendly foreign countries with reasonable access to replacement parts for industrial equipment, and with new equipment destined for strategic production.

The effect of the embargo and the positive list has been to reduce almost to zero U.S. exports to the Chinese mainland, the U.S.S.R., and the other Eastern European countries in the Soviet bloc.

The effect of the quantitative export quotas has so far been small, except for sulphur and cotton, where the purpose and effect were to allocate available supplies on an equitable basis. The U.S. cotton crop was unusually small in 1950 and cotton exports would have fallen in any case. During the six months November 1950 through April 1951 only 2.4 million bales were exported, compared with 3.5 million in the comparable period of the previous crop year. The short crop brought higher unit prices, however, and the value of the smaller amount exported was within 2 per cent of the previous year’s larger shipment. In August 1951 the quotas were suspended when it became evident that a larger crop was in prospect.

The effects of the licensing controls can be measured only imperfectly. Published statistics show the value of controlled exports but do not adequately indicate to what extent exports may have been reduced by the refusal of licenses. Even with these qualifications, the data in Table 10 on the value of controlled exports are still significant in showing the extension of controls during 1950. From the first quarter of 1950 to the first quarter of 1951 the proportion of total exports subject to control increased from 18 to 32 per cent, but of all exports outside the Western Hemisphere more than 45 per cent were subject to control. The sharp rise in exports to Western Europe after August 1950, however, would suggest that licenses to that area have been granted relatively freely. The commodity data in the table show that chemicals, minerals, metal manufactures, machinery and vehicles (and also, though temporarily, textiles) have been subject to licensing control to a substantial extent.

Table10.

U.S. Exports Subject to Contbol, First Quarter, 1950 and 1951

article image

The published data classify exports into “total controlled exports” and “total controlled exports excluding special category classes.” A comparison of these two classifications indicates that practically all the “special category” exports are controlled.

In this table the 1951 data by area have been estimated to include “special category” exports.

The 1951 data by commodity groups exclude “special category” exports with the following exceptions: $35 million worth of “special category” exports are included in the nonmetallic minerals group, and $9 million worth of such exports are included in the nonfood animal and vegetable products group. Therefore, the sum of controlled exports by commodity groups is $269 million less than the published data for total controlled exports. The decline in the controlled export percentages between the first quarter of 1950 and 1951 may thus be due, in part, to the omission of the $269 million worth of “special category” exports–especially the omission from the metals and machinery groups.

Includes in 1951 all ERP countries and other European countries outside the Soviet bloc.

Finland and Yugoslavia are excluded from this total in 1951 but are included with Western Europe.

Source: Export Control (Quarterly Reports by the Secretary of Commerce).

The figures for the first quarter of 1951 probably understate the applicability of controls later in the year. It is understood that license applications were rejected in substantial amounts during the second quarter of 1951 because of short domestic supply. The effect of this action may be expected to show up in reports on exports for the third and fourth quarters of the year.

To sum up this discussion, the resulting U.S. merchandise balance with the major areas for which adjusted trade figures are available is presented in Chart 8. As the stimulus to U.S. trade varied among the several areas, so the merchandise trade balances also developed differently. With the Outer Sterling Area, Latin American Republics, and raw material suppliers in Asia and Africa, the previous trend toward a better U.S. import-export relationship was accentuated and a U.S. import surplus developed. In some areas, as Chart 8 shows, this import surplus was already in evidence by the first quarter of 1950, while in others it developed later. It tended to reach a peak in late 1950 or early 1951 when raw material imports and prices were generally at record high levels, and tended to narrow or even to be turned into a U.S. export surplus in the second and third quarters as U.S. purchases fell off while exports continued with less interruption. In Canada the smaller rise in prices of exported raw materials and semimanufactures did not interfere with a continued growth in exports to the United States, but imports from the United States rose even more rapidly, being financed largely by dollars received from a large capital inflow.

Chart 8.
Chart 8.

Origin and Destination of U.S. Merchandise Trade, Quarterly Totals, 1948-51 *

Citation: IMF Staff Papers 1952, 001; 10.5089/9781451949391.024.A002

* Total trade, excluding MDAP-financed exports but including other unrecorded imports and exports.

With the United Kingdom and Western Europe the progressive improvement in trade relationships continued through the third quarter of 1950, then worsened as an increase in U.S. imports in the fourth quarter was accompanied by a greater increase in aid-financed exports. U.S. imports from ERP countries had begun to rise before “unaided” U.S. exports to that area, however, and this improvement in the unaided part of the U.S.-European trade was the most marked since the inception of the European Recovery Program. The sharp decline of U.S. imports, combined with a reduction in exports, began to reverse the area balances again in the third quarter of 1951.

Disposition of Dollar Earnings by Area

In an earlier section of this paper it was shown that countries outside the United States appear (in the aggregate) to have varied their “unaided” imports from the United States according to their dollar earnings from exports to the United States one or two quarters earlier. This general relationship is discernible also when the data are classified by area, although there are substantial disparities between the areas, which are partly explainable by differences in their multilateral dollar settlements with third areas.

When all sources and uses of dollars (except merchandise imports from the United States) are summed up by area, it appears to have been the rule in 1950 that improvements in the dollar supply were not matched by a comparable rise in imports. The slowness of the import response resulted in substantial additions to the gold and dollar reserves of all areas. By early 1951, however, dollar imports of most areas had risen substantially, and the rate of acquisition of gold and dollar assets was being rapidly reduced. For raw material supplying areas, the decline in the current dollar supply was due mainly to a fall in U.S. imports of raw materials; but for the United Kingdom, the fall in the dollar supply reflected the rapid growth of U.K. payments in dollars to third areas (especially Europe and Latin America), coupled with some reduction in U.K. receipts of dollars from the Outer Sterling Area.

Dollar imports in relation to export earnings and multilateral settlements16

For the United Kingdom and the Continental ERP Area, the first two sections in Chart 9 show a general conformity between unaided imports from the United States and the respective earnings from dollar exports. For the United Kingdom, however, the trade balance has tended to vary according to the direction and size of multilateral dollar settlements. In 1948-49, when such settlements required substantial dollar payments, the United Kingdom was running a trade surplus (not counting aid-financed imports) with the United States. In the third quarter of 1950, U.K. imports began to rise about nine months after multilateral receipts had turned sharply upward; in 1951 imports gained impetus, rising above exports as the United Kingdom received a large net inflow of dollars, both from the ERP Area and from the Outer Sterling Area.17 The sharp shift to an outflow of multilateral dollars in the second and third quarters of 1951 reflected a reversal in the relationship with the ERP and Outer Sterling Areas, and an apparent outflow also to Latin America. From the chart it is evident that the sharp shift of about $700 million (quarterly) in multilateral dollar settlements during 1951 was several times larger than the loss of dollars through rising imports from the United States. The U.K. Government has announced its intention of curtailing U.S. imports, however, and previous experience, as summarized on the chart, would suggest that the quarterly rate of unaided dollar imports may decline by $100-200 million during 1952.

Chart 9.
Chart 9.
Chart 9.

Value of Exports and Imports and Multilateral Settlements of Major Areas Vis-à-Vis the United States, Quarterly Totals, 1948-51 *

Citation: IMF Staff Papers 1952, 001; 10.5089/9781451949391.024.A002

* While imports from the United States financed by U.S. aid are excluded from the data, U.S. aid payments for area imports from third countries are reflected in multilateral settlements.

For Continental ERP countries, the rate of unaided dollar imports (relative to dollar exports) appears to have been more strongly influenced by the outflow of multilateral dollar payments, both in 1948-49 when the outflow of payments was becoming greater and in 1951 when the continuing rise in imports could be supported by a favorable turn in multilateral settlements although exports to the United States were falling off. (It is not necessarily the case, of course, that the recent multilateral dollar receipts accrued to the same countries which were experiencing the dollar trade deficit.) In part the 1951 divergent movements of imports and exports may also reflect the lag hypothesis previously advanced; the upturn in dollar exports in 1950 was not followed by an upturn in imports until one quarter later.

For Canada multilateral settlements have facilitated triangular trade, as a surplus with the United Kingdom has provided dollars with which to cover a deficit with the United States. As the Canadian section on Chart 9 shows, both multilateral receipts and the trade deficit have tended to narrow over the past two years. A rise in imports from the United States began in the second quarter of 1950, about two quarters after an upturn in exports, but it tended to increase faster than the proceeds available from exports and multilateral receipts. In the third quarter of 1951, when multilateral receipts declined sharply, Canadian imports also were brought sharply down.18

The Latin American section of Chart 9 shows a fairly consistent lag of imports from the United States behind exports. During 1948-49 when this lag entailed a trade deficit, the balance was covered by large multilateral receipts. During 1949-50 when the lag entailed a rising trade surplus, the continuance of multilateral receipts added considerably to the current supply of dollars. In 1951 when exports to the United States fell off sharply, the widening margin of multilateral receipts helped to support the continuing rise of imports. Here, as in Western Europe, however, it needs to be kept in mind that the data are aggregates for a number of countries and the multilateral dollar receipts did not necessarily go to the same countries that experienced a trade deficit.

In the Outer Sterling Area, a full year passed after exports to the United States had begun to rise before dollar imports began their upturn; in the interval large multilateral dollar payments were made. In this respect the chart for the Area is complementary to that for the United Kingdom, although net payments of the Outer Sterling Area were evidently much smaller than net multilateral dollar receipts by the United Kingdom during the period. That these payments were in the nature of a residual, representing surplus dollar receipts of the Area, is suggested by the fact that they fluctuated closely with the trade surplus, even falling off sharply in the third quarter of 1951, parallel to the decline in exports.

The section of the chart representing “all other” countries includes mainly the Far East and Middle East, exclusive of Sterling Area countries, and reflects much the same raw material and import situation as that of the Outer Sterling Area. That is, the upturn in exports to the United States began early in 1950 and a sharp upturn in imports did not come until a year later. Unlike the Outer Sterling Area, however, the resulting export surplus in 1950 was not disposed of through multilateral payments. In the first half of 1951, large dollar outpayments were made by Japan—apparently to the Sterling Area—and area imports from the United States abruptly stopped rising. The available data do not indicate whether the outpayments represented a real current dollar deficit with the rest of the world, or merely a bookkeeping deficit with the United States due to the transfer of liquid balances elsewhere. In the third quarter of 1951, multilateral settlements again turned favorable.

Other elements in the balance of payments

While the fluctuations in unaided merchandise trade and multilateral settlements have been the main determinants of changes in dollar reserves for most areas, service items frequently make the difference between a steady loss or gain; for short periods direct capital movements can be of dominant importance.

Dollar receipts from U.S. capital movements, services, and donations have been sizable for Western Europe, rising from a quarterly average of under $100 million in 1948 to about $250 million in 1950. In 1951 a reversal in net movements of U.S. capital had, by the third quarter, reduced the combined balance for these three items to approximately zero, both for the United Kingdom and for Continental ERP countries.

The inflow of U.S. private capital into Canada rose irregularly from about $50 million a quarter in 1948 to $80-90 million a quarter in 1950 and 1951, and sufficed roughly to offset the net outflow of funds on service account. There was an exceptional capital inflow of over $500 million from the United States in the third quarter of 1950.

In Latin America, U.S. capital provided net receipts of about $80 million a quarter in 1948 and 1949, but the subsequent movement was smaller and more irregular; it was far outweighed by net payments of $150-200 million per quarter for the servicing of U.S. investments.

Dollar receipts from (net) U.S. capital inflow to the Outer Sterling Area provided a small quarterly balance of $15-30 million, but here again they were more than offset by service payments on U.S. investments. A similar situation has prevailed for “all other” countries in the aggregate.

Unaided imports compared with net dollars from all sources

In order to take account of all elements in the balance of payments in conjunction with fluctuations in unaided area imports from the United States, Chart 10 has been arranged to show the excess of the dollar inflow to each major area from all sources of dollars except U.S. aid over the outflow for all uses except merchandise imports from the United States. Viewed broadly, the chart shows that in four of the six major areas the increase in this excess during 1950 and early 1951 was followed by a slower and less than proportionate increase in unaided imports from the United States, so that much of the increase in the dollar supply was added to monetary reserves. The chief exceptions were Canada, where imports were already exceeding the current U.S. dollar supply in the last quarter of 1950, and the Outer Sterling Area, where additions to reserves went into the Sterling Area pool.

Chart 10.
Chart 10.
Chart 10.

Value of “Unaided” Imports from the United States, Compared with Currently Available Dollars, by Major Areas, Quarterly Totals, 1948-51 *

Citation: IMF Staff Papers 1952, 001; 10.5089/9781451949391.024.A002

*“Unaided” imports exclude imports from the United States financed by U.S. Government aid. Currently available dollars are computed as the difference between all current sources and current uses of U.S. gold and dollar funds except unaided commodity imports from the United States.

During 1951, however, the supply of U.S. dollars leveled off or declined for most major areas and failed to cover the cost of unaided imports from the United States. (On the chart, the excess of unaided imports over available dollars indicates an area loss of gold and U.S. dollar assets to the United States.) Two exceptions to this trend were the Continental ERP area and the group of “all other” countries, where available dollars in the third quarter of 1951 continued near record high levels, well above the amounts needed to pay for unaided imports.

This analysis has indicated that, while the dollar difficulties experienced by many areas during the third quarter of 1951 were attributable in part to a reduction In exports to the United States, they were in some cases much more a reflection of the delayed rise in dollar imports and in other cases the result of balance of payments difficulties with third countries. If a similar payments analysis could be made for more individual countries, it would probably show a wider range of differences of experience, attributable to causes peculiar to each country.

Note: This paper was prepared before balance of payments data for the fourth quarter of 1951 had been published. These data, now available, record a continuation of the major developments of the third quarter, although there is some indication that the next turning point in the U.S. trade balance was being approached.

*

Mr. Dirks, Acting Chief of the North American Division, is a graduate of Middlebury College, Tufts College, and Columbia University. He was formerly economist at the Board of Governors of the Federal Reserve System, Chief of the Financial Intelligence Branch and Deputy Chief of the Governmental Structures Branch in the Office of Military Government for Germany, and economist in the Office of the Secretary of Defense.

Mr. Zassenhaus, economist in the North American Division, is a graduate of the University of Bonn and the University of Bern. He was formerly Professor of Economics at Colgate University and research associate at the Twentieth Century Fund.

1

The supply of dollar funds is here taken to include proceeds from the sale of exports to the United States, plus net dollar receipts from U.S. Government grants and loans, from services (exclusive of investment income), and from private donations and private capital movements.

2

Throughout the text, “world” is used to describe the aggregate of all countries outside the United States.

3

The subtraction of all ILS. Government aid from commodity trade is not quite appropriate since part of the aid goes to pay for services. This part is ignored here because the relevant data are not readily available. The alternative method of including service items with commodity exports and imports on the chart seemed undesirable because the seasonal movement in service items would have obscured the parallelism between commodity exports and imports. The roughness of the adjustment does not affect the validity of the conclusions reached here.

4

This world export balance was in fact the major factor in the mounting outflow of U.S. gold and dollar assets.

5

The authorizations requested by the President for foreign aid in the fiscal years ending June 1952 and June 1953 would raise military aid to about 75 per cent of total aid, but the significance of these requests is qualified by the prospect of military aid including industrial supplies and equipment. The amount of actual military end items for fiscal 1952 was proposed by the President at roughly half of total aid.

6

These included an increase of $280 million in U.S. bank deposits and security holdings in Canada, an increase of $50 million in short-term claims on the United Kingdom, a private U.S. bank loan to the French Government of $225 million, and a private refunding loan of $54 million to the Canadian Government, which was not actually disbursed for retirement of the old securities until the following quarter. The first two of these items appear to have been largely speculative.

7

Other unusual transactions in the past two years include a short-term inflow of $121 million from Latin America in the first quarter of 1950, and a short-term outflow of $98 million to that area in the fourth quarter.

8

Third quarter net earnings by other countries from U.S. tourist travel amounted to about $190 million in 1950; in 1951, however, the figure was reduced to about $150 million, mainly because of greater travel by Canadians in the United States. Net income from U.S. foreign investment, on the other hand, increased steadily to a quarterly average of $370 million in 1951. Miscellaneous Government expenditures, mainly from U.S. installations abroad, have risen from under $100 million a quarter in the first half of 1950 to $160 million a quarter in early 1951; by the third quarter the rate had reached $280 million and was still rising. These sources of dollar earnings have been partly offset by net U.S. earnings for transportation and miscellaneous private services of about $150 million a quarter.

9

Nearly $2.7 billion of this was in the form of gold, and amounted to a 25 per cent addition to the gold reserves reported by these countries at the end of 1949. The remaining $2 billion was about evenly divided between short-term bank balances and U.S. securities (mostly Government issues); this amount also was about 25 per cent of the combined foreign holdings of short-term banking funds and U.S. securities (not counting foreign purchases before 1935) that were outstanding at the end of 1949.

10

The fact that in the third quarter the ratio of unaided imports from the United States to the current supply of funds (less U.S. Government aid) again exceeded 100 per cent may be explainable by the unexpectedly sharp drop in dollar earnings as U.S. buyers resisted the high prices that had been built up. Moreover, in view of the prospect that this drop might be temporary, monetary reserves could appropriately be used to maintain the level of world purchases in the United States.

11

The figures in Chart 3 are for recorded imports and exports, which include aid-financed exports. For purposes of the analysis, the published figures for merchandise imports and exports have been regrouped under the two headings, raw materials and manufactures, with subdivisions into food and nonfood. This regrouping affects imports particularly. Semimanufactures, newsprint, and jute burlap have been grouped together with crude materials to form the subclass “nonfood raw materials.” Sugar has been combined with crude foods to make up the subclass “food raw materials.” Nonfood manufactures correspondingly exclude newsprint and jute burlap, while food manufactures exclude sugar.

On the export side finished and semifinished manufactures have been combined into the subclass “nonfood manufactures.”

12

Price and quantity indices are available only for recorded trade. The figures used for the chart are therefore smaller than the (adjusted) merchandise trade figures in the balance of payments statistics of the U.S. Department of Commerce.

13

This is the only official estimate of the U.S. terms of trade with any region available by quarters. It covers all Latin American countries and does not show the terms separately for individual countries or for any subdivisions of the region.

14

Data in this section are based on recorded general imports, which differ somewhat from balance of payments data.

15

That is, military aid (MDAP), Army Civilian Supply Program payments, Export-Import Bank loans, and payments for ECA shipments other than offshore.

16

For brevity, exports to the United States and imports from the United States by other countries will be designated here as dollar exports and dollar imports. (This differs from the conventional usage, according to which dollar imports include imports from any country that customarily settles its international balances in dollars.)

17

In each period it will be noted that the difference between dollar exports and unaided dollar imports was only a small part—about a fifth—of the net outflow or inflow of multilateral dollar settlements. The smallness of the impact of dollar settlements on imports is probably attributable to the central position of the United Kingdom as a repository of balances for the Sterling Area.

18

The item “multilateral receipts” is a residual and thus includes the effect of errors and omissions in all other accounts. For Canada much of the third quarter decline may have consisted of errors and omissions, rather than true multilateral settlements.

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

    World Sources and Uses of Dollar Funds, Quarterly,1947–51

  • View in gallery

    “World” Exports and Imports Vis-à-Vis the United States, Quarterly, 1948–51*

  • View in gallery

    Value of U.S. Imports and Exports by Major Commodity Groups, Monthly Averages by Quarters, 1949–51

    (logarithmic vertical scale)

  • View in gallery

    Indices of U.S. Import and Export Prices by Major Commodity Groups, Monthly, 1949-51

    (1947-49 =100)

  • View in gallery

    Indices of U.S. Import and Export Prices (1947–49 = 100) and Terms of Trade by Major Commodity Groups, Monthly, 1949–51

  • View in gallery

    Value of U.S. Imports from Major Areas and from “Underdeveloped Areas,” Monthly Averages by Quarters, 1948–51

    (logarithmic vertical scale)

  • View in gallery

    Value of U.S. Imports from Raw-Material Supplying Areas, Monthly Averages by Quarters, 1948-51

    (logarithmic vertical scale)

  • View in gallery

    Origin and Destination of U.S. Merchandise Trade, Quarterly Totals, 1948-51 *

  • View in gallery View in gallery

    Value of Exports and Imports and Multilateral Settlements of Major Areas Vis-à-Vis the United States, Quarterly Totals, 1948-51 *

  • View in gallery View in gallery

    Value of “Unaided” Imports from the United States, Compared with Currently Available Dollars, by Major Areas, Quarterly Totals, 1948-51 *