special factor affecting borrowing abroad by manufacturing companies also was applicable for mining companies (particularly in the oil and gas extraction industry), some of the foreign holdings of debt of the transportation and public utilities group may reflect the residual effects of historic foreign interest in the debt financing of these industries. The foreign borrowing of the banking industry consists to a large extent of long-term certificates of deposit.

Distribution by Type of Debt

The distribution of private foreign investment in U.S. debt instruments by type of debt and by industry is shown in Appendix Table A-10. Official holdings of U.S. debt are excluded from this table; as shown elsewhere (Table 6), the overwhelming proportion of such holdings consisted of U.S. Government issues.7

Of the private holdings, about half of the total consisted of marketable bonds, mainly the off-shore Eurobonds. About two-thirds of the total of $8 billion of bonds consisted of such issues. In addition, loans from foreign banks amounted to $4.9 billion; to an unknown extent these also reflected borrowing to finance direct investments abroad in excess of the quotas allowed by the foreign direct investment controls.

The remaining $3.1 billion included $600 million of long-term certificates of deposit, representing funds placed with banks and other financial institutions. The remainder represented miscellaneous borrowing abroad by U.S. residents. Since only the nature and not the purpose of the credit was reported, one can only speculate on

the latter. Long-term trade credits could be an important factor as well as borrowing to meet the Foreign Direct Investment Program requirements from foreign financial institutions other than banks.

Finally, it may be noted that the extremely small amount of foreign holdings of "domestic" (as distinguished from "Euro") bonds of U.S. corporations may result in part from reporting deficiencies. There was no practical way to obtain information on foreign-held U.S. bearer bonds-private and Government-if the bonds were physically held abroad. Only if the bonds were held in custody accounts with domestic reporting institutions such as banks and brokerage firms could information regarding foreign holders be obtained.

Manner in Which Securities are Held

Of the total of $23.7 billion of private foreign holdings of U.S. stocks,8 about $7.4 billion was directly registered under a foreign address and about $16.3 billion was held in custody or nominee accounts with U.S. banks, brokers, or other persons acting as custodians or nominees for foreign owners (see Appendix Table A-6). Foreign persons, whether holding these securities directly or through U.S. custodians or nominees, in turn frequently were also acting in a nominee capacity. Thus there could be two or even more nominees between the issuing corporation and the ultimate owner.

Even more so than U.S. investors, foreigners tend to leave their securities with banks and brokers for convenience in safeguarding their securities, collecting dividends and interest, and executing buy and sell orders.

The most notable shift occurred in the percentage decline of the Latin American Republics and other Western Hemisphere countries, and the rise of Asian investors as holders of U.S. portfolio assets. In Asia, the disproportionate growth in holdings of U.S. securities resulted from the emergence of Middle East oil exporting countries as international investors, and the rise of Japan as an economic power.

From the earliest days of the nation up to World War I, foreign portfolio holdings were channeled primarily into railroad stocks and bonds, and issues by some of the states. That war caused major investing countries to liquidate most of their U.S. portfolio holdings to help pay for war costs. Subsequent to World War I, foreign portfolio investment activity turned toward investment in U.S. manufacturing and service industries. While in 1919 railroad securities still accounted for about threequarters of foreign portfolio holdings of U.S. securities, by 1929 this proportion had declined to less than 1 percent of the total.

From the late twenties to the present, manufacturing industries have been in the foreground of attention of foreign investors, and by 1949 they accounted for more foreign investment than all other U.S. industries combined. No information on the industrial distribution of foreign portfolio investment in the United States was obtained in the 1941 benchmark survey. Data from the current benchmark study show that stocks of manufacturing firms held by foreign investors amounted to 60 percent of all U.S. stocks held abroad, or about $14.8 billion out of $24.7 billion.

Total reported in survey

The remaining difference is probably due in part to the more complete coverage in the survey of foreign official holdings of marketable U.S. Treasury bonds and notes, as compared with the Treasury Bulletin data. The Bulletin data on foreign holdings of these assets are calculated by applying monthly transactions reports to a 1971 Treasury survey of amounts held for foreigners by banks and brokers reporting on the Treasury Foreign Exchange forms, which was less complete than the new survey. A part of the remainder may also arise from definition and valuation differences.

Comparison with Prior Data

While the value of foreign portfolio investments in the United States grew substantially in the 33-year interval since the 1941 benchmark survey, as shown in Table 7 and Appendix Tables A-19 and A-20, there was little change in the relative geographical distribution of holders. European and Canadian holdings continued to predominate, accounting jointly for 80 to 90 percent of the total foreign investment in both benchmark years.

Note: Data for the years 1914-1934 are taken from America's Stake in International Investment, by Cleona Lewis, Brookings Institution; data for 1941 are taken from Census of Foreign-Owned Assets in the United States, U.S. Treasury Department, Government Printing Office, 1945; data for 1949-1974 (old) are taken from various issues of the Survey of Current Business, Department of Commerce; data for 1974 (new) are the result of the new Treasury benchmark Survey of Foreign Portfolio Investments.

In the early part of the century, two-thirds to threequarters of private foreign portfolio investments in the United States were in U.S. bonds. But, beginning with the stock market boom of the 1920's, foreign holdings of bonds declined and holdings of stocks increased so that by 1929 less than 20 percent of foreign portfolio holdings in the United States were in bonds. There was little change in this relationship until the mid-sixties when, in response to deficits in the U.S. balance of payments and to U.S. capital control programs, private U.S. borrowers placed more debt issues abroad. Many of the private issues placed abroad in that period were in the form of debentures convertible into common stocks, but until the end of 1974, at least, the amount of conversions had been relatively small.

Dividend and Interest Payments

The survey questionnaire required reporters to provide information on dividend and interest rates on their stock and debt issues outstanding. Information was also obtained on interest rates on Government securities, and on other payments of income to owners or creditors of non-corporate business firms. Application of these payments to the shares and long-term debt held by foreign owners as of December 31, 1974, gives an annualized income to foreign portfolio investors of $4,196 million ($1,154 million for dividends and $3,042 million for interest).

Heretofore, estimates of investment income payments in the balance of payments have been based on an assumed yield applied to estimated foreign portfolio investment. For the year 1974 these income payments (on comparably defined investments) were estimated at $3,779, about 10 percent less than the survey totals.

Income payments derived from the survey indicated an overall yield of nearly 6.3 percent, about 4.7 percent on foreign holdings of stocks and 7.2 percent on foreign holdings of long-term debt. By comparison, current balance of payments yields on total foreign portfolio investments equalled about 6.7 percent. Thus the higher income amounts shown in the Treasury survey are not based on notably different yields applied to foreignowned assets, but rather on the higher amount of foreign investment established in the survey.

Detailed data on dividends and interest are shown in Tables 8, 9, and 10 and Appendix Tables A-17 and A-18. The overall yield on foreign investment in U.S. stocks, 4.7 percent, was virtually the same as that calculated by Moody's Investors Service for all dividends but less than the 5.5 percent on the 500 common stocks in the Standard and Poor's index. This yield showed no significant variation by area, thus indicating that the composition of foreign portfolios probably also does not vary much from

area to area.

However, an analysis of dividend yields by industry was more revealing. In the industry with the highest average dividend yield, transportation and public

utilities (nearly 7.5 percent), foreign holdings were only 2.8 percent of the total stock outstanding, while foreign holdings in the much lower yielding manufacturing and mining industries were 4.4 and 5.4 percent, respectively. This would support the view expressed by the respondents in the survey of the securities industry (see chapter 3) that foreign investors in general are more interested in growth and appreciation considerations than in yields.

The yield (rate of interest divided by market value) on all debt was 7.2 percent and on privately held debt it was 8.3 percent. There are no domestic series with which these figures can be directly compared, particularly the holdings of non-marketable debt. But the data are reasonably in line with yields prevailing in U.S. financial markets at the end of 1974.

As in the case of dividends, there was no significant divergence by geographic area; observed differences between areas were related to the differences in the type of debt held by investors in the area. Thus, for example, the high yield of 10.4 percent observed for Latin America is due to the heavy proportionate holdings by Latin American investors of high yielding certificates of deposit (36 percent of Latin American holdings of U.S. debt); overall, foreign holdings of these certificates comprised only 3.7 percent of total debt holdings.

An examination of yields by major industry shows a spread from 6.3 percent to 10.1 percent. The low of 6.3 percent, however, was applicable to Government securities, which tend to have lower yields than private issues. Government securities aside, the yields by industry fall in the 8-10 percent range. Considering the differences in the holdings of various types of debt for various industries, and the variations in rates of return, the spread does not indicate significant industry preferences among foreign investors in debt issues.

particularly in U.S. stocks. Net foreign purchases of U.S. stocks totaled $4.5 billion in 1975, and $1.5 billion in the first quarter of 1976.

The value of foreign holdings of U.S. stocks was also substantially increased by the sharp recovery in U.S. stock prices in 1975 and early 1976. The addition to the value of these holdings on account of price increases is estimated at $8.1 billion for 1975, and $5.2 billion in the first quarter of 1976, which, along with the net inflows of $4.5 billion and $1.5 billion during those periods, raised the total of foreign investment in U.S. stocks to about $37 billion by the end of 1975, and $43.9 billion by March of 1976. (See Table 11.) The latter figure was 78 percent above end-1974 holdings.

The increasing interest of the Middle East oil producing countries in U.S. equities is evident in these figures: those countries accounted for a third of the total inflow in 1975-1976. Practically all of the remainder was accounted for by the same countries which have been the major purchasers of U.S. equities in the past-Canada and a few European countries.

The total of $4.5 billion of net foreign purchases of stocks for 1975 was a record annual high, but it was probably affected by the unusually small inflow in 1974 when market prices were declining and net foreign purchases of U.S. stocks were only $0.5 billion. Taking 1974 and 1975 together, the total inflow was no greater than the total for 1972 and 1973.

It should be noted, however, that purchases by European residents in 1974-1975 were considerably less than in 1972-1973. While the oil producing countries are obviously an important new element in foreign demand for U.S. stocks, it remains to be seen whether purchases by these countries will continue to match the amounts of 1975 and early 1976, part of which probably represented one-time shifts from short-term to long-term investments in the United States.

The relative lack of foreign interest in U.S. corporate bonds continued to be evident in 1975 and early 1976. Middle East oil producing countries purchased $1.7

Area

billion of these instruments in 1975 and the first quarter of 1976, but, as in the case of their purchases of U.S. stocks, this may have been mostly a one-time shift in order to balance the U.S. portfolios of these countries with more long-term assets. International institutions redeemed or sold $1 billion of U.S. corporate bonds (the term includes U.S. agency issues) in 1975 and the net purchases of all other countries were about $0.1 billion. At the end of 1975 foreign holdings of these bonds were $11.1 billion, and at the end of March, 1976 they were $11.5 billion.

As noted elsewhere in this report, foreign interest in U.S. corporate bonds has not been significant during the past 60 years, except during the mid-sixties and early seventies when U.S. Government balance-of-payments programs caused U.S. companies actively to sell Eurobonds to foreigners at attractive interest rates or with special convertibility features. With the ending of these programs in early 1974, U.S. corporate bonds appear to be returning to their previous position of relative unimportance in the U.S. portfolio of foreigners.

The continued dominance of U.S. stocks in this portfolio is clearly seen in the estimates of the total amounts outstanding for foreign portfolio investment in