U.S. international transactions in 1989.

In 1989, for the second year in a row, the U.S. current account deficit narrowed, failing to $104 billion (excluding capital gains and losses) (chart 1). The trade deficit fell to $113 billion. However, the pace of improvement slowed noticeably during the year; part of the reason may have been a deterioration in U.S. international price competitiveness.

The overall improvement in the external accounts last year reflected a rapid expansion in exports, much of which came early in the year. Imports also grew at a fairly strong pace, however, led by an increase of nearly 30 percent in the value of oil imports. A modest improvement in the current account as the year went by stemmed from gains in net service transactions. The United States continued to enjoy positive net investment income receipts despite a large and growing net foreign debt position.

The capital account counterpart to the current deficit included large foreign private purchases of U.S. Treasury and corporate securities and a continued large net inflow of foreign direct investment.

The inflow of direct investment generated public debate about the desirability of the accumulation by foreigners of U.S. assets in this form; but it also bolstered the nation's ability to maintain the rate of capital formation in the face of a low national saving rate. In fact, the reduction of the current account deficit in 1989, both in absolute terms and as a percentage of gross national product, reflected a significant narrowing of the gap between domestic investment and domestic savings. As chart 2 shows, the improvement in the external deficit was related to an increase in the U.S. national saving rate, at a time when the share of GNP devoted to investment was trending down. (1) That increase included both a significant rise in the personal saving rate and a slight reduction in the federal budget deficit. ECONOMIC INFLUENCES ON U.S. INTERNATIONAL TRANSACTIONS The proximate determinants of the changes in U.S. trade and current account flows include movements in U.S. international price competitiveness, changes in aggregate demand or income at home and abroad, and swings in the rates of return on real and financial assets at home and abroad. The main factor in the recent slowing of U.S. external adjustment appears to have been the weakening of U.S. international price competitiveness, after its earlier gains.

A convenient measure of price competitiveness is the real exchange value of the dollar, the ratio of U.S. prices to foreign prices expressed in dollars. As chart 3 shows, the real exchange value of the dollar with respect to the currencies of the other Group of Ten (G-10) countries (weighted by multilateral trade shares) reached a low in the first quarter of 1988; that low was more than 40 percent below the peak value in early 1985. The ratio subsequently rose (and price competitiveness declined), with much of the movement coming during the first half of 1989. By the third quarter, the ratio was 13 percent above its 1988 low. In the fourth quarter and early this year, the ratio dropped back, largely because of movement in the dollar's nominal exchange rate.

The dollar rose against nearly all major currencies during the first half of 1989, under the impetus of tight U.S. monetary policy and favorable U.S. trade figures (chart 4). Political events in Japan and China also contributed to the dollar's strength. In the second half of the year, however, an easing of U.S. monetary policy contrasted sharply with further tightening of monetary policy and rising interest rates abroad, particularly in Germany and Japan. Late in the year the dollar firmed against the yen, but it declined more sharply against the mark as political developments in Eastern Europe promised to stimulate the West German economy. Despite the dollar's decline late in the year, for the year as a whole the dollar was noticeably above its average level in 1988.

In addition to the movements of nominal exchange rates, U.S. price competitiveness deteriorated because domestic prices advanced somewhat faster on average in the United States than in the other G-10 countries. This divergence was particularly evident in 1988, when approximately two thirds of the 2 percent deterioration in price competitiveness after the fourth quarter of 1987 was attributable to differences in inflation rates. By contrast, in 1989 most of the additional 5 percent decline resulted from the rise in nominal exchange rates (some, but not all, of which was reversed by the end of the year).

Vis-a-vis the developing countries that are major U.S. trading partners, U.S. price competitiveness continued to improve, about 7 percent in 1989 on average. Of the eight countries in the index of developing countries, only Malaysia experienced an improvement in price competitiveness with respect to the United States during the year as a whole, an improvement due to a superior domestic price performance. In the last three quarters of the year, however, Mexico was able to improve its price competitiveness; depreciation of the peso of approximately 16 percent over this period outweighed the difference in the inflation rates between the two countries.

Despite its loss of price competitiveness with respect to the other industrial countries, various measures of prices and costs suggest that the United States continues to enjoy on average an absolute price and cost advantage over its major trading partners. One such measure is unit labor costs in manufacturing, which in the United States are still more than 15 percent below those in other major industrial countries (see chart 5). Such a comparison is inexact, however, and whether the present gap in labor costs has yet caused a shift in production toward the United States and has had an impact on the potential for U.S. external adjustment in the longer run is uncertain. For example, although the net inflow of direct investment into the United States has been fairly strong in recent years and may have been increased by the labor cost advantage, much of this activity has involved takeovers rather than net additions to U.S. capacity; any positive effect on the rate of U.S. capacity formation could only have been an indirect one. Moreover, private fixed capital formation has been growing much more rapidly in Europe and Japan than in the United States. Clearly, relative labor cost is just one of many variables that influence investment decisions.

Typically, an important factor in underlying movements in the trade balance is the change in the relative rates of growth of the United States and its trading partners. In the past year, a slowing in U.S. growth, coupled with the maintenance of robust growth among most major groups of U.S. trading partners, has helped narrow the U.S. deficit (table 1). An exception is the estimated reduction in growth among key developing countries in 1989 to a rate approximately equal to that of the United States; this reduction is attributable to very low growth in a number of Latin American countries and somewhat lower growth in Asia. DEVELOPMENTS IN MERCHANDISE TRADE The U.S. trade deficit narrowed $14 billion to $113 billion in 1989. However, all the improvement took place early in the year; the deficit widened from $110 billion at an annual rate in the second quarter to $115 billion in the fourth (table 2). Export growth slowed dramatically in the second half of the year after a strong first half. Though the pace was less rapid than it was during 1988, total imports continued to grow steadily. This pattern resulted in part from special factors: The dollar's appreciation held down the increase in the value of imports in the first half of the year, and the strike at Boeing reduced exports in the fourth quarter. Exports The value of agricultural exports increased 9 percent for the year, although most of that increase came in the first half (table 2). A combination of factors caused a sharp drop in the quantity of exports in the third quarter (chart 6): a temporary halt in corn exports to the Soviet Union; and a decline in soybean exports because of increases in supplies from South America and a shift away from soybean feed in the European Community. Prices of agricultural exports were affected more than quantities by the drought of 1988. Prices fell continuously during the year from the drought-induced high of the third quarter of 1988.

For separate reasons, the growth of nonagricultural exports, in both value and quantity, was also bunched in the first half of the year (see table 2 and chart 6). During that period, the value of nonagricultural exports expanded at an annual rate of almost 18 percent; but during the second half, it increased at less than 4 percent. Though less dramatic, the story for the quantity of exports is similar: It grew 15 percent in the first half and 7 percent in the second. Part of the explanation for the slowdown was the Boeing strike, which reduced exports in the fourth quarter; but more significant was a cumulative, if gradual, loss of price competitiveness after the second quarter of 1988. The average price in dollars of nonagricultural exports as a whole increased only 1.5 percent for 1989 (chart 6). However, with the 8 percent appreciation of the dollar in the first half of 1989, prices in foreign currency rose substantially, imposing a considerable, if temporary, loss of price competitiveness. Estimates based on a Board staff model of the U.S. current account suggest that, had exchange rates remained at their level in the fourth quarter of 1987, eliminating the run-up in exchange rates in 1988 and 1989, the rate of growth of the value of nonagricultural exports in the second half of 1989 would have been between 7 and 8 percent instead of 4 percent. However, even this higher rate would have represented a significant slowing of export growth.

Although lower than in 1988 (and slowing in the second half), the growth in the quantity of exports by major category was broad-based (table 3). Capital goods grew a strong 9 percent for the year. The Boeing strike cut the quantity of aircraft exports somewhat when measured fourth quarter over fourth quarter; however, for the year the level was 30 percent higher than that in 1988. The quantity of computer exports grew 8 percent; given the 10 percent reduction in the quality-adjusted export price, the value of computer exports did not change in 1989. (2) Automotive exports showed virtually no growth because of the predominance of trade with Canada; that bilateral trade in turn is determined primarily by conditions in the U.S. market, which were weak in 1989. Consumer goods and industrial supplies showed healthy growth for the year.

Industry by industry, in general export prices rose much less than they did in 1988; however, as noted above, U.S. exports on average lost somewhat more in international price competitiveness in 1989 because the dollar was higher. Imports Led by a 28 percent increase in oil imports, the value of total imports increased 6.4 percent in 1989 (table 2). Non-oil imports rose a more moderate 4.3 percent.

The rise in the value of oil imports resulted from sharp increases in both price and quantity (charts 7 and 8). The quantity of oil imports jumped partly because U.S. oil production declined further, continuing the downward trend evident since 1986 (chart 8); production peaked in the first quarter of 1986 at 11.4 million barrels per day, and thereafter gradually declined to 9.6 million barrels per day in the fourth quarter of 1989. Behind this trend is a long-term decline in expenditures in the United States on oil exploration and development, which is likely to depress domestic oil production for another three to five years (chart 9). The price of imported oil responded last year to continued strong worldwide demand and a series of interruptions in supply, notably the oil spill in Alaska, accidents in the North Sea, and disruptions in the oil sectors of Romania and the Soviet Union. An increase since 1985 in OPEC's share of world oil production, as shown in chart 10, augurs for little, if any, near-term easing in oil prices.

The rise in the dollar in 1988 and 1989 actually lowered the dollar prices of many categories of non-oil imports between the fourth quarters of 1987 and 1988 (table 4). Price changes ranged between minus 2 and plus 2 percent for most categories. Through its effect on import prices, the rise in the dollar also had an initially favorable impact on the value of imports and the trade balance in the first half of 1989, the familiar J-curve effect.

The decline in the relative price of non-oil imports stimulated the quantity imported, particularly in the second half of the year. Computers, other capital goods, and foods, feeds, and beverages had robust growth.

The quantity of imports of automotive products fell 11 percent, partly because significant changes took place in the locus of production of cars sold in the United States. The value of imported automotive vehicles, as shown by the green bars in chart 11 has fallen for the last two years. The value of imported automotive parts, on the other hand, has risen steadily.

This pattern reflects significant changes in the U.S. automotive market in recent years. While the number of passenger cars imported from Canada increased between 1987 and 1989, imports from both Japan and Western Europe declined (table 5). The causes of the decline probably differ by area: In Western Europe it may be the continuing effects of the gain in U.S. price competitiveness after 1985 associated with the depreciation of the dollar against the currencies of the countries in the European Monetary System; in Japan it was largely the increase in production by U.S. subsidiaries of Japanese firms. Such transplant production rose to 1.1 million units in 1989. In fact, net sales of Japanese nameplate cars increased in 1989 because sales from transplant production rose more than imports declined. Whereas the number of units imported from Japan has been falling, the average price per unit has increased. Meanwhile, imports of automotive parts from Japan have been rising strongly. Indeed, Japan accounted for almost all of the increase in imported automotive parts in recent years (see table 6). NONTRADE CURRENT ACCOUNT TRANSACTIONS The nontrade portion of the current account balance increased $8 billion in 1989 to a net surplus (excluding capital gains and losses) of $10 billion (table 7). Capital gains of $2 billion associated with foreign direct investment in the United States lowered the surplus to $8 billion when net capital gains are included. (3) All of the $8 billion net improvement can be attributed to "other services, net." Of the many items affecting this account, significant positive changes occurred in net travel ($2.6 billion), royalties and license fees ($1.3 billion), and other private services ($4.0 billion). (4) Residents of Japan, Canada, and Western Europe accounted for most of the increase in net travel receipts. Western Europe also was the source of the largest increase in net receipts of royalties and license fees and other private services. Investment Income Receipts and Payments Net investment income remained at $3 billion in 1989. This net flow was the sum of two disparate elements: Net income from direct investment contributed $39 billion; but net portfolio income was a negative $35 billion, marking a deterioration of $6 billion in 1989. (5) Although U.S. receipts from portfolio holdings abroad increased $14 billion, this gain was more than offset by the increase in portfolio income payments to private and government investors. As discussed in more detail below, this latter is explained by increases in both the foreign holdings of U.S. portfolio assets and the rate of return earned on these assets.

Receipts from U.S. direct investors abroad, defined as dividends and interest to the parent firm plus reinvested earnings, increased $4 billion, while payments to foreign direct investors in the United States actually decreased $3 billion. This development is remarkable because the stock of foreign direct investment in the United States increased $60 billion in 1989. The reduction in payments to direct investors in the United States mirrored the general decline in U.S. corporate profits. U.S. INTERNATIONAL INVESTMENT POSITION The U.S. net international investment position declined to an estimated negative $600 billion at the end of last year (table 8), reflecting gross claims on foreigners amounting to $1.4 trillion, minus gross liabilities of $2 trillion. (6) Private portfolio investments accounted for most of the net debt position. The net private direct investment position also declined last year, to about negative $30 billion, defined as the difference between the value of U.S. assets held abroad, estimated at $359 billion, and foreign assets held in the United States, at $390 billion. The numbers for private asset holdings in table 8, particularly those for direct investment, are subject to significant error, as discussed below.

Given the magnitude of the U.S. net debt position, it may seem surprising that U.S. net investment income receipts remained positive last year. This outcome reflects significant differences in the implicit rates of return on U.S. international claims and liabilities (table 9). The estimated average rates of return on U.S. investments abroad, both direct and portfolio, are substantially higher than the rates on foreign investments in the United States.

The difference between U.S. and foreign rates of return is most pronounced for direct investment. The implicit rate of return on U.S. direct investments abroad, which was 15 percent in 1989 and averaged almost 14 percent for the period 1983-89, may be overstated substantially because the value of U.S. direct investments abroad (used as the denominator in the rates of return in table 9) is based on historical cost. At least three studies have suggested that these assets may be undervalued by as much as one half of their current market value, implying that the implicit rate of return is overstated by a factor of two. (7) However, there is no clear consensus that the error in the data is that large. (8) Even if it were, and the rate of return on direct investment were halved, the rates of return on direct investment in the United States and abroad would still differ widely. Moreover, a more realistic valuation of direct investment assets would also lower the rate of return on direct investment in the United States. But because foreign direct investment in the United States is on average more recent, the market or replacement value of these holdings is likely to be closer to historical cost than is that of U.S. direct investment abroad.

These differences in rates of return on assets may also reflect other factors. Many firms in the United States shift reported profits to foreign jurisdictions that have lower taxes. This practice tends to reduce the observed rate of return on foreign investment in the United States and to increase that on U.S. direct investment abroad. The differences may also reflect accounting practices that increase interest, amortization, and depreciation charges when U.S. companies are acquired. Since much of the recent foreign direct investment in the United States has been associated with takeovers, these practices may have artificially reduced the observed rate of return on direct investment in the United States.

Even after identifying factors that bias the observed rates of return, one cannot conclude that the consistently higher rates on U.S.-owned foreign assets are a statistical artifact. For one thing, lower rates of return on recent investments are natural during a transition period. Moreover, the rates of return may reflect the reportedly poor profits from some takeovers. The discrepancy between the rates of return on U.S.-owned and foreign-owned portfolio investments probably reflects the activities of U.S. banks, which account for a large part of both U.S. portfolio claims and liabilities: The banks make profits only as they earn more on their claims than they pay on their liabilities.

Finally, as a return to ownership or equity, the return on direct investment moves with the business cycle. In light of the weakness in U.S. corporate profits in 1989, the fall in the rate of return on foreign direct investment in the United States becomes understandable. CAPITAL ACCOUNT TRANSACTIONS In 1989, the composition of the capital account counterpart to the U.S. current account deficit shifted away from the pattern observed in recent years (table 10). The substantial buildup of official reserves in the United States in 1986-88 was reversed, as $17 billion of official capital flowed out on net. Much of this outflow represented intervention purchases of foreign currencies by the United States and other G-10 governments to limit the appreciation of the dollar.

The three major sources of capital inflows in 1989 were net foreign purchases of U.S. securities ($66 billion), net direct investment ($31 billion), and a statistical discrepancy amounting to an unrecorded inflow of $35 billion.

Securities transactions accounted for a large share of the net capital inflow in 1989. Private foreign net purchases of U.S. Treasury securities rose to almost $30 billion, half again as much as in the previous year. Purchases of U.S. government agency bonds were also strong (about $14 billion). Net foreign private purchases of U.S. corporate bonds, particularly of Eurobonds, were substantial. On the outflow side, U.S. net purchases of foreign securities reached a record $23 billion.

Inflows of foreign direct investment into the United States reached a record high of $60 billion, exceeding slightly the high level of the previous year. U.S. direct investment outflows also reached a peak, of $28 billion (excluding capital gains and losses).

The participation of foreign residents in U.S. financial markets and of U.S. residents in foreign financial markets is far greater than these net flows suggest, and it is still growing. In 1989, the sum of foreign purchases and sales of U.S. Treasury securities amounted to over $4 trillion, compared with $100 billion to $200 billion earlier in the decade. Foreign purchases and sales of U.S. stocks and bonds have also been dramatically higher; and U.S. purchases and sales of foreign stocks and bonds have increased more than ten-fold in the same period. This surge in cross-border financial transactions has paralleled a large advance in the magnitude of cross-border trade of goods and services and has been spurred by similar economic forces. Rapid technological advances, including virtually instantaneous transmission of information and sophisticated hedging techniques, have reduced the cost of managing operations around the globe and have facilitated direct, as well as portfolio, investment.

The rapid growth of foreign direct investment in the United States during the 1980s has stirred public debate over the desirability of the continued accumulation of U.S. assets by foreigners in this form. Any rapid change should, perhaps, be scrutinized, but no evidence suggests that present or foreseeable levels of foreign ownership of U.S. industry should be troublesome. Two concerns that have been emphasized are the takeover of sensitive defense-related firms and the accumulation of undesirable concentrations of market power by foreign firms. The first has been addressed by the passage of the Exon-Florio amendment to the Omnibus Trade and Competitiveness Act of 1988. As to the second, current antitrust laws should be adequate to deal with the accumulation of excessive market power, no matter by whom. In fact, as table 11 shows, the concentration of market power in key industries in the hands of foreign interests as a whole, let alone individual foreign companies, is not large. (9)

An issue that has rarely been addressed in this debate is the effect on the U.S. current account deficit of potential restrictions on foreign direct investment. Given that current and prospective U.S. trade deficits imply the necessity of a continued net accumulation of U.S. assets by foreigners, any policy action that restricts direct investment in the United States would, at a minimum, shift the composition of this asset accumulation toward portfolio capital. It also would probably provoke retaliation against U.S. direct investment abroad. Either of these outcomes would tend to increase the burden of servicing the growing U.S. international indebtedness, and it would have other undesirable effects on future current account deficits.

First, in order to attract the additional portfolio inflows necessary to offset any restriction on direct investment, rates of return paid on foreign portfolio investments would have to increase, worsening the servicing burden and the current account. Second, a shift in the composition of U.S. liabilities from the equity capital of direct investment to debt instruments would commit the United States to fixed payments, rather than payments that vary with the state of the economy. (10)

Third, the apparent differences in the average rates of return on different real and financial international claims and liabilities, discussed above and shown in table 9, suggest that any shift away from direct investment inflows as a counterpart to the current account deficit is likely to imply higher net investment income payments. For example, one outcome of restrictions that reduced the inflow of direct investment could be an equal reduction in U.S. direct investment abroad, which could result from a retaliatory imposition of restrictions by other countries. In this case, since the rate of return on U.S. direct investment holdings abroad has been substantially greater than that on foreign holdings in the United States, the loss in net direct investment income receipts could be significant. Moreover, the effects of a restriction imposed in any one year would cumulate, worsening the current account balance in every subsequent year until the capital flow was reversed.

Alternatively, restrictions on the inflow of direct investment could be offset by an increase in the inflow of portfolio investment. Since rates of payment on portfolio liabilities in the United States are currently greater than those on direct investment liabilities in the United States, such a shift would also result in an increase in net payments, even without an increase in the rate of return on U.S. portfolio assets held by foreigners.

The average rates of return shown in table 9 are subject to considerable uncertainty, and these rates of return, even if accurate, may not apply to future investment flows. Nevertheless, the differences are large enough in some cases to suggest that a shift away from direct investment inflows would have a negative effect on U.S. net investment income flows. PROSPECTS FOR 1990 The question for the year ahead is whether the recent slowing of U.S. external adjustment merely interrupts a longer-term improving trend or reverses it in a more lasting way. Two factors favor further adjustment. One is that strong growth of aggregate demand abroad is apparently continuing, while growth in the United States is moderating; this difference should be a continuing source of strength for exports relative to imports. Second, since the third quarter of 1989, the dollar has retraced part of its earlier rise against the major foreign currencies, thereby restoring some of the U.S. price competitiveness that had been lost. But several other factors cloud this prospect. With U.S. prices now rising somewhat faster than domestic prices abroad, the recent improvement in price competitiveness may be short-lived unless the U.S. inflation situation improves. Oil imports are another potential problem. With the recent rise in oil prices and the continued negative trend in domestic production, rising U.S. consumption of oil will necessitate higher imports, in both quantity and value.

The service account showed remarkable improvement in 1989, despite the continued deterioration in the U.S. net investment position. Sustained strength in receipts of net direct investment income was an important component of this performance. Whether the currently favorable differentials in rates of return will persist is not clear. Whatever future rates of return may be, the overall flow of net investment income will inevitably turn negative if the net international investment position of the United States continues to deteriorate.

At a more fundamental level, substantial further progress in reducing the U. S. external deficit seems unlikely until private and government saving rates in the United States attain and maintain higher levels. *

1. In chart 2 the external deficit is equal to "net foreign investment" in the national income and product accounts. Net foreign investment differs from the current account balance (exclusive of capital gains and losses) in the balance of payments accounts by a small amount, reflecting differences in the treatment of shipments of gold and of transactions with Puerto Rico and U.S. territories. balance (exclusive of capital gains and losses) in the balance of payments accounts by a small amount, reflecting differences in the treatment of shipments of gold and of transactions with Puerto Rico and U.S. territories.

2. Because of the rapid technical advances in computer products, it is particularly difficult to decompose changes in the value of computer exports and imports into price and quantity changes. In an attempt to account for the technical advances, the Commerce Department has adopted a quality adjusted or "hedonic" price index for the computer industry. The same index is used to deflate domestic expenditures, imports, and exports. As a result, quantity and price breakdowns for both exports and imports must be used with caution.

3. Capital gains have a negative sign on the payments side of the accounts.

4. Other private services, net exports of which reached $17 billion in 1989, include education expenses, financial services, insurance, telecommunications, and business, professional, and technical services. Important changes in the methods of collecting data for a number of these items (and for travel services) led to a substantial upward revision in the gross and net flows for the years 1982 to 1988. These improvements may have contributed to the year-to-year change in the "other services, net" line in table 7, as well as to the higher level. For a detailed discussion of the changes, see Russell C. Krueger, "U.S. International Transactions, First Quarter 1989," Survey of Current Business, vol. 69 (June 1989), pp. 50-61.

5. Because of rounding error, the net of $3 billion differs by $1 billion from the difference of the rounded flows for net direct investment income and net portfolio income.

6. As noted in the table, these estimates are based on data for the 1988 year-end investment position published by the Bureau of Economic Analysis and data on capital flows in 1989.

7. The following studies reach broadly similar conclusions regarding the undervaluation of the stock of direct investment abroad: Robert Eisner and Paul J. Peiper, "The World's Greatest Debtor Nation," Review of Economics and Finance, forthcoming; Lois Stekler, "U.S. Direct Investment Receipts and Payments: Models and Projections," International Finance Discussion Papers 140 (Board of Governors of the Federal Reserve System, Division of International Finance, 1979); Michael Ulan and William G. Dewald, The U.S. Net International Investment Position: The Numbers are Mis-stated and Misunderstood (U.S. Department of State, 1989).

For an examination of the various estimates, see Lois Stekler and Guy V.G. Stevens, "The Adequacy of Direct Investment Data," in Peter Hooper and J. David Richardson, eds., International Economic Transactions: Issues in Measurement and Empirical Research (University of Chicago Press, forthcoming).

8. In preliminary work, Walther Lederer has arrived at a much smaller replacement-cost measure of the value of direct investment abroad than have the studies mentioned above.

9. No single country accounts for a predominant portion of the holdings of 31 percent in the chemicals and stone, clay, and glass industries.

10. Part of the increase in portfolio capital would be invested in U.S. common stock, which also has a flexible return. However, assuming past proportions as a guide, a large percentage of the increase would be invested in debt instruments.