Report on

U.S. Holdings of

Foreign Long-Term Securities

as of December 31, 1997,

and December 31, 1999

Office of the Assistant Secretary, International Affairs

Department of the Treasury

Division of International Finance

Board of Governors of the Federal Reserve System

April 2000

CONTENTS

Introduction...............................................................................................................................1

Chapter 1. Findings from the 1997 Survey..................................................................................3

Chapter 2. U. S. Holdings of Foreign Long-Term Securities
                  as of December 31, 1999.......................................................................................22

Chapter 3. Survey Methodology...............................................................................................26

Tables

1. U.S. Holdings of Foreign Long-Term Securities, by Country and Type
    of Security, as of December 31, 1997.....................................................................................4

2. U.S. Holdings of Foreign Long-Term Securities, by Type of Security,
    for the Top Ten Countries of Investment and the Rest of the World, as of
    December 31, 1997...............................................................................................................7

3. U.S. Holdings of Foreign Long-Term Securities for the Top Ten
    Countries and the Rest of the World, as of March 31, 1994, and
    December 31, 1997...............................................................................................................8

4. U.S. Holdings of Foreign Long-Term Securities, by Positions,
    Adjusted Transactions, and Resulting Differences,
    as of March 31, 1994, and December 31, 1997......................................................................9

5. U.S. Holdings of Foreign Long-Term Securities and Foreign Holdings
    of U.S. Long-Term Securities, 1984-97................................................................................10

6. U.S. Holdings of Foreign Equities, by Country of Investment, Amount
    Held, and Amount Held in U.S.-Traded Securities, for the Top Ten
    Countries of Investment and the Rest of the World, as of
    December 31, 1997.............................................................................................................11

7. U.S. Holdings of Foreign Long-Term Debt Securities, by Country,
    Amount Held, Amount Held Issued in the United States, and Amount
    Denominated in U.S. Dollars, for the Top Ten Countries of Investment
    and the Rest of the World, as of December 31, 1997............................................................12

- i -

8. U.S. Holdings of Foreign Long-Term Securities, by Country, Amount
    Held, and Amount Traded in the United States, for the Top Ten
    Countries of Investment and the Rest of the World,
    as of December 31, 1997.....................................................................................................13

9. U.S. Holdings of Foreign Long-Term Securities in Developed Countries
    Emerging Markets, and Caribbean Financial Centers, as of March 31,
    1994, and December 31, 1997.............................................................................................14

10. U.S. Holdings of Foreign Long-Term Securities, by Continent,
      as of March 31, 1994, and December 31, 1997..................................................................15

11. U.S. Holdings of Foreign Long-Term Debt Securities, by Currency
      of Denomination, as of March 31, 1994, and December 31, 1997.......................................16

12. U.S. Holdings of Foreign Long-Term Debt Securities, by Remaining
      Term to Maturity, as of December 31, 1997.......................................................................17

13. U.S. Holdings of Foreign Long-Term Debt Securities, by Country and
      Sector of Issuer, as of December 31, 1997.........................................................................18

14. U.S. Holdings of Foreign Long-Term Securities, by Industry,
      as of December 31, 1997...................................................................................................20

15. U.S. Holdings of Foreign Long-Term Securities, Net Transactions,
      and Changes in Percentage and Valuation, by Type of Security,
      as of December 31, 1997 and 1999....................................................................................22

16. U.S. Holdings of Foreign Securities, as of December 31, 1997 and 1999............................23

17. U.S. Holdings of Foreign Equity Securities, as of December 31, 1997
      and 1999............................................................................................................................24

18. U.S. Holdings of Foreign Long-Term Debt Securities, as of
      December 31, 1997 and 1999............................................................................................25

Appendixes

A. Statistical Tables

A.1. U.S. Holdings of Foreign Long-Term Debt Securities, by Sector of
        Issuer, as of December 31, 1997.....................................................................................30.

- ii -

A.2. U.S. Holdings of Foreign Long-Term Securities, by Total Held,
        Amount Traded in U.S. Markets, and ADRs, as of December 31, 1997............................33

A.3. U.S. Holdings of Foreign Equity Securities, by Type of Security,
        as of December 31, 1997.................................................................................................36

A.4. U.S. Holdings of Foreign Long-Term Debt Securities, by Type of
        Security, as of December 31, 1997...................................................................................40

B. Forms and Instructions

- iii -

Introduction

This report presents data and analyses regarding U.S. portfolio investment in foreign equity and debt securities.¹ The data are drawn primarily from the latest comprehensive benchmark survey of U.S. holdings of foreign long-term securities as of December 31, 1997. The survey was a joint undertaking of the U.S. Treasury Department and the Board of Governors of the Federal Reserve System. This information has been supplemented and updated with transactions data from the Treasury International Capital (TIC) reporting system to present information on U.S. foreign portfolio investment as of December 31, 1999, whenever possible.

The 1997 benchmark survey collected detailed information at the individual security level for equity and long-term debt (original term-to-maturity in excess of one year) securities. It is therefore possible to present findings from the survey in a wide variety of ways, such as by country, type of security, currency, remaining term to maturity, industry, and so forth. The TIC system, however, collects information only by country and security type (equity or debt); thus, the more recent information presented as of year-end 1999 is not as detailed as the information presented as of year-end 1997.

The 1997 survey was the second such survey conducted by the United States.² The previous survey collected data as of March 31, 1994. The United States began to conduct these surveys as the level of U.S. investment in foreign securities started to grow significantly in the 1990s. The next such survey will collect data as of year-end 2001. The United States plans to conduct portfolio asset surveys at least once every three years starting with the 2001 survey.

The 1997 survey was part of an internationally coordinated effort under the auspices of the International Monetary Fund (IMF).³ Twenty-nine countries, including most of the industrial countries, participated in the survey.⁴ This coordinated effort was undertaken primarily because there was a wide discrepancy between worldwide measured portfolio assets and worldwide measured portfolio liabilities, with reported liabilities greatly exceeding reported assets. A major

_________________

1. Portfolio investment is defined as ownership or control, by one person or an affiliated group, of less than 10 percent of the voting equity of a business enterprise.

2. The Treasury Department conducted a survey during World War II of all foreign assets owned by U.S. residents as of May 1943. Though this survey measured portfolio investment as well as other forms of investment, it is sufficiently removed in time and different in purpose that it is fair to state that the 1994 survey is the first of its kind. For the 1994 survey results, see Department of the Treasury, 1994 Benchmark Survey of U.S. Ownership of Foreign Securities, July 1998.

3. International Monetary Fund, Results of the 1997 Coordinated Portfolio Investment Survey, January 31, 2000.

4. Participating countries were Argentina, Australia, Austria, Belgium, Bermuda, Canada, Chile, Denmark, Finland, France, Iceland, Indonesia, Ireland, Israel, Italy, Japan, Malaysia, the Netherlands, New Zealand, Norway, Portugal, Singapore, South Korea, Spain, Sweden, Thailand, the United Kingdom, the United States, and Venezuela.

- 1 -

reason for the asset undercount was perceived to be that many countries had been placing greater emphasis on measuring portfolio liabilities than on measuring portfolio assets out of concern that foreign investment in their economies might expose them to influences beyond their direct control. Typically, the world's major industrial countries have begun to conduct portfolio asset surveys only during the past ten years, and many have begun their efforts as a result of the coordinated survey. The U.S. asset survey as of year-end 2001 will also be part of an IMF-coordinated effort. This survey will measure U.S. holdings of both long-term and short-term securities

The primary reason for conducting portfolio asset surveys is to determine the level of U.S. holdings of foreign securities. The TIC reporting system measures purchases and sales of foreign securities on a monthly basis. In theory, all such transactions are measured by this system, as U.S. regulations require that all major transactions in foreign securities should be reported to the TIC reporting system. However, as international communications have become more efficient and less expensive, this has enabled an ever-increasing number of U.S. investors to participate directly in foreign securities markets. As the number of intermittent, less-predictable participants has increased, the likelihood that some transactions will not be recorded also has increased. The results of the past two asset surveys seem to indicate this, as the 1994 survey measured foreign holdings some 60 percent higher than predicted based on price-adjusted and exchange-rate- adjusted TIC transactions data, and the 1997 survey measured holdings 21 percent higher. These survey findings have had the effect of reducing the estimated deficits in the U.S. balance of payments accounts and international investment position. The U.S. international investment position was revised as of year-end 1997 as a result of the survey, such that the U.S. net debtor position was decreased $293 billion. The U.S. balance of payments deficit for 1997 was reduced more than $10 billion as a result of the survey because of increased estimates of interest and dividends received from foreign securities by U.S. residents.

Chapter 1 of this report presents the 1997 survey findings. Chapter 2 combines these data with price-adjusted and exchange-rate-adjusted TIC transactions data to present estimated U.S. holdings of foreign securities as of year-end 1999. Chapter 3 discusses survey methodology. Appendix A presents a variety of statistics not elsewhere included in the report. Appendix B contains a copy of the forms and instructions used by the survey.

The Treasury Department and the Federal Reserve Board wish to express their appreciation to all survey respondents whose efforts and information have made this report possible.

-2 -

Chapter 1. Findings from the 1997 Survey

The data presented in this chapter are drawn primarily from the survey of U.S. holdings of foreign long-term securities (original term to maturity of more than one year) conducted as of December 31, 1997. Data from the previous such survey--conducted as of March 31, 1994--are also frequently included. In all tables in this report, components may not sum to totals because of rounding.

Level of Investment by Country and Type of Security

Table 1 shows U.S. foreign portfolio investment in long-term securities by country and type of investment (equity or debt). One hundred sixty-four countries attracted at least some U.S. investment. The table shows only those regions that attracted at least $500,000 in U.S. investment.

The table shows a strong preference by U.S. investors for foreign equity securities as opposed to debt securities, with $1,208 billion invested in equities and $547 billion invested in debt securities. The previous survey showed a similar preference on the part of U.S. investors, with more than 65 percent of measured investment occurring in equities. This is in marked contrast to the pattern of foreign portfolio investment in the United States as of December 31, 1997, when only 33 percent of foreign investment is in equity securities.⁵

However, U.S. investors have not always preferred foreign equities over debt securities. The survey conducted by the United States during World War II found that equities accounted for only 38 percent of U.S. foreign securities holdings. Bringing these figures forward to 1984 by using price- and exchange-rate-adjusted transactions data showed that in 1984 only 29 percent of U.S. foreign securities holdings were in equities. But after 1984, U.S. investors began to favor foreign equities over debt securities, and by 1987, the levels of debt and equity held were essentially equal.

Though U.S. portfolio investment is widespread, it is also highly concentrated, with approximately 67 percent of total investment occurring in only ten countries. Table 2 shows U.S. investment in these ten countries. Five countries attracted more than $100 billion in U.S. investment; twenty-seven countries had at least $10 billion; and fifty-four countries had at least $1 billion.

The United Kingdom was the country of choice for U.S. investors as of the date of the survey, with some $272 billion in U.S. investment. Canada finished a distant second with $177 billion in U.S. investment, followed by Japan, the Netherlands, and Germany.

_________________

5. See U.S. Department of the Treasury, Report on Foreign Portfolio Investment in the United States as of December 31, 1997.

- 3 -

Table 1.   U.S. Holdings of Foreign Long-Term Securities, by Country and Type of Security, as
                of December 31, 1997

                Millions of dollars

- 4 -

Table 1.   U.S. Holdings of Foreign Long-Term Securities, by Country and Type of Security, as
                of December 31, 1997 (Continued)

                Millions of dollars

- 5 -

Table 1.   U.S. Holdings of Foreign Long-Term Securities, by Country and Type of Security, as
                of December 31, 1997 (Continued)

                Millions of dollars

1. Algeria, Gabon, Libya, and Nigeria.

2. Bahrain, Iran, Iraq, Kuwait, Oman, Qatar, Saudi Arabia, and the United Arab Emirates.

In this report, data for Hong Kong are presented separately from those for China, and data for the islands of Jersey, Guernsey, and the Isle of Man are presented separately from those of the United Kingdom. U.S. holdings of debt issued by international organizations represent issues primarily of the International Bank for Reconstruction and Development (World Bank), the Inter-American Development Bank, the African Development Bank, and the European Investment Bank.

U.S. Portfolio Investment by Country in Descending Order

Geography obviously influences the pattern of U.S. foreign investment, with both Canada and Mexico attracting a greater percentage of U.S. investment than would be expected based solely on the size of their capital markets. The level of U.S. investment in Canadian debt securities is noteworthy, as the $107 billion figure represents approximately 20 percent of U.S. foreign debt holdings. This is strongly influenced by the tendency of Canadian institutions to issue their debt securities directly in the United States. Approximately $68 billion of the Canadian debt figure represents securities issued directly in the United States, the majority of which were denominated in U.S. dollars.

- 6 -

Table 2.   U.S. Holdings of Foreign Long-Term Securities, by Type of Security, for the Top Ten
                Countries of Investment and the Rest of the World, as of December 31, 1997

                Billions of dollars except as noted

... Not applicable.

The Netherlands also appears higher on the list than might be expected. This again is strongly influenced by the presence of U.S.-traded securities. Many of the largest American Depositary Receipt (ADR) programs represent shares of Dutch companies, and nearly half of the $107 billion in U.S.-owned Dutch equities represents holdings of ADRs.⁶

The ratio of U.S. holdings of Swiss equity ($62 billion) to Swiss debt ($1 billion) is striking. This pattern reflects the fact that relatively little debt has been issued by Swiss firms and Swiss governmental organizations, whereas the Swiss equity market is one of the largest in the world.

The United Kingdom was by far the first choice of U.S. international investors as of the end of 1997, but as table 3 shows, Japan was the country of choice at the time of the previous asset survey, on March 31, 1994.

_________________

6. American Depositary Receipts (ADRs) are receipts for shares of stock in foreign companies that are held in a custodial account by or for a U.S. bank. These receipts are traded on U.S. exchanges in U.S. dollars. ADRs entitle the owners to all dividends, capital gains or losses, and voting rights, just as if the underlying shares were owned directly.

- 7 -

Level of Investment by Country, 1994 and 1997

Table 3 shows the rapid growth of U.S. portfolio holdings abroad during the March 1994 to December 1997 period, with U.S. portfolio investment more than doubling in nominal terms over forty-five months. It also shows significant changes in the placement of U.S. portfolio assets, with Japan losing the position it held in 1994 as the country of choice for U.S. portfolio investors. In general, Europe attracted a greater share of U.S. investments in 1997, with the percentage share of total U.S. foreign investments increasing in six of the seven European countries that appear in the table. Also, the pattern of U.S. investment was more diversified in 1997, as the percentage of total U.S. investment placed in countries and regions other than the ten leading countries increased over this period from 27.5 percent to 32.6 percent.

Table 3.   U.S. Holdings of Foreign Long-Term Securities for the Top Ten Countries of
                Investment and the Rest of the World, as of March 31, 1994, and December 31, 1997

                Billions of dollars except as noted

... Not applicable.

1. The amount and percentage for "rest of world" were computed using all but the top ten countries as of the 1994 survey rather than the top ten countries as of year-end 1997.

- 8 -

The order in which countries are ranked changed quite a bit during the period between the two surveys; this was due in large part to the performance of local equity markets rather than to changes in the patterns of U.S. investments. Much of the change in rank order was due to Japan's drop from first to third position and Mexico's drop from fourth to seventh. The average share prices in Japan and Mexico decreased during this period, whereas those of Canada and the European countries in table 3 increased significantly.

The rise in the value of U.S. foreign holdings raises the question of the extent to which this increase is attributable to increased net investments rather than to changes caused by adjustments for price or exchange rate or both to existing assets. This question is somewhat difficult to answer because, as mentioned in the introduction, when the accumulated flows since the 1994 survey are adjusted for price and exchange rate and added to the 1994 survey levels, the result is a figure 21 percent lower than the level measured by the 1997 survey. Thus, either the transactions data undercount, or the survey overcounts, or both. Table 4 shows the levels of equity and debt measured by the two surveys; the measured transactions in the period between the two surveys; the applicable price and exchange rate adjustments as computed by the Commerce Department's Bureau of Economic Analysis (BEA); the resulting year-end 1997 position derived by adding the transactions and price adjustments to the 1994 survey results; and the resulting unexplained difference between the estimated and the measured 1997 figures.

Table 4.   U.S. Holdings of Foreign Long-Term Securities, by Positions, Adjusted Transactions,
                and Resulting Differences, as of March 31, 1994, and December 31, 1997

                Billions of dollars

1. Adjustments for price and exchange rate as computed by BEA.

The table shows that net adjustments to debt were negative; thus all gains in U.S. holdings of foreign debt securities were the result of net purchases during the intersurvey period. For equities, however, there was a significant upward adjustment in the value of these holdings.

As previously stated, the 1994 survey also measured significantly larger holdings of foreign securities by U.S. residents than the value-adjusted transactions data indicated.

- 9 -

U.S. Foreign Portfolio Investment Compared with Foreign Portfolio Investment
in the United States

Both foreign portfolio investment in the United States and U.S. portfolio investment abroad are growing very rapidly; however, foreign portfolio investment in the United States still exceeds U.S. foreign investment by a significant margin (table 5).

Though some of the increase in these cross-border investment figures represents price appreciation, there was also significant new net investment. With respect to foreign investment in the United States, the foreign-held share of U.S. equities increased over the March 1994 to December 1997 period from 5.6 percent to 6.3 percent of the total amount outstanding; and the foreign-held share of U.S. Treasury securities increased over the same period from 19.4 percent to 38.4 percent of the total amount outstanding. Overall, foreigners increased their ownership of total U.S. securities outstanding from 8.9 percent to 12.0 percent during this period.

Table 5.   U.S. Holdings of Foreign Long-Term Securities and Foreign Holdings of U.S.
                 Long-Term Securities, 1984-97

                Billions of dollars except as noted

- 10 -

U.S. Investment in Foreign Equity Securities

Table 6 shows the leading countries for U.S. equity investment, along with the amount of this investment represented by foreign equities that trade in the United States.

Clearly the propensity for the shares of foreign companies to trade in U.S. securities markets varies widely by country. Canada, the Netherlands, and Mexico have led the way in this respect. Canada represents very much of a special case, however, because the U.S. and Canadian securities exchanges are so closely linked as to create almost a single, integrated market.

The list in table 6 consists of the same ten countries in which U.S. investors placed the most overall portfolio investment shown in table 2, though the order has changed. Canada is fifth on this list but second overall.

Table 6.   U.S. Holdings of Foreign Equities, by Country of Investment, Amount Held, and
                Amount Held in U.S.-Traded Securities, for the Top Ten Countries of Investment
                and the Rest of the World, as of December 31, 1997

                Billions of dollars except as noted

* Less than 0.5.

1. ADR American Depositary Receipts.

- 11 -

U.S. Investment in Foreign Debt Securities

Table 7 shows the leading countries for U.S. investment in foreign debt securities, the amount of this investment represented by foreign debt securities issued in the United States, and the amount denominated in U.S. dollars.

Here we see a very different country list, with Argentina, Brazil, Australia, and international organizations appearing. Canada easily leads the list, in large measure because many Canadian debt securities are issued in the United States and are denominated in U.S. dollars. U.S. holdings of the debt of neighboring Canada and Mexico showed a much stronger tendency to be issued in the United States than was true for the other countries in this table.

U.S. investors showed a clear preference for U.S. dollar-denominated debt from the emerging- market countries. Approximately 88 percent of the debt issued by Mexico, Argentina, and Brazil combined was denominated in U.S. dollars, whereas only 41 percent of the debt held from the six developed countries in table 7 was denominated in U.S. dollars.

Table 7.   U.S. Holdings of Foreign Long-Term Debt Securities, by Country,
                Amount Held, Amount Held Issued in the United States, and Amount
                Denominated in U.S. Dollars, for the Top Ten Countries of Investment
                and the Rest of the World, as of December 31, 1997

                Billions of dollars

- 12 -

U.S. Investment in Foreign Securities: Total and Amounts Traded in the United States

Table 8 shows U.S. ownership by country of foreign securities, the amount and percentage of these securities that trade on U.S. exchanges, and the amounts of U.S.-traded equity and debt.

A substantial $506 billion, or 29 percent, of all U.S. investment in foreign securities occurred in U.S.-traded securities. For some countries, such as Canada and the Netherlands, the majority of U.S. investment was in U.S.-traded issues. For many other countries the percentage was very small.

Overall, 30 percent of U.S. holdings of foreign equities and 27 percent of U.S. holdings of foreign debt traded on U.S. exchanges.

Table 8.   U.S. Holdings of Foreign Long-Term Securities, by Country, Amount Held,
                and Amount Traded in the United States, for the Top Ten Countries of
                Investment and the Rest of the World, as of December 31, 1997

                Billions of dollars except as noted

- 13 -

U.S. Investment by Type of Market

Table 9 shows U.S. investment levels as of March 31, 1994, and December 31, 1997, in three different types of markets: developed countries, emerging-market countries, and Caribbean financial centers. All amounts are in billions of dollars.

Investment in the twenty-two developed countries accounted for the bulk of U.S. foreign investment at both dates, with 80 percent and 76 percent of the total in 1994 and 1997, respectively. The emerging markets increased their share of the total from 16 percent to 19 percent over this period. However, their share of total U.S. portfolio investment in equities decreased over the same period, from 16 percent of the total to 14 percent. As of the 1997 survey date, 73 percent of U.S. investment in developed countries was in equity securities, whereas only 50 percent of U.S. investment in emerging markets was in equities.

Securities issued through the Caribbean financial centers represent primarily activity of organizations in developed countries that issue in these countries for convenience or tax advantage.

Table 9.   U.S. Holdings of Foreign Long-Term Securities in Developed Countries,
                Emerging Markets, and the Caribbean Financial Centers, as of
                March 31, 1994, and December 31, 1997

                Billions of dollars

1. Australia, Austria, Belgium, Canada, Denmark, Finland, France, Germany, Hong Kong, Iceland, Ireland, Italy, Japan, Luxembourg, the Netherlands, New Zealand, Norway, Singapore, Spain, Sweden, Switzerland, and the United Kingdom.

2. Bahamas, Bermuda, British Virgin Islands, Cayman Islands, Netherlands Antilles, and Panama.

- 14 -

U.S. Investment by Continent

Table 10 shows a rather stable pattern of U.S. investment by continent between the two dates, with the only noteworthy change being the 8 percent increase in Europe's share over the period, largely at the expense of Asia's share. On both dates, 95 percent of total U.S. investment was accounted for by Europe, Asia, and the Americas. In Africa, the preponderance of U.S. investment was in South Africa, which attracted $13 billion of the total of $19 billion invested in the continent. The Australia and Oceania line is composed primarily of investment in Australia and New Zealand.

In Europe and Asia the preponderance of U.S. investment is in equities, whereas in the Americas, equity and debt are roughly equal.

Table 10.    U.S. Holdings of Foreign Long-Term Securities, by Continent, as of
                    March 31, 1994, and December 31, 1997

                    Billions of dollars except as noted

1. North America, South America, Central America, and the Caribbean (including Bermuda).

- 15 -

U.S. Investment in Foreign Debt Securities by Currency of Denomination

Table 11 shows the currency of denomination for foreign debt securities held by U.S. residents. The amounts presented total only to $527 billion for 1997 because accurate information about the currency of the remaining $20 billion of U.S. foreign-held debt was not available.

Table 11.    U.S. Holdings of Foreign Long-Term Debt Securities, by Currency of
                    Denomination, as of March 31, 1994, and December 31, 1997

                    Billions of dollars except as noted

In both surveys, U.S. investors showed a strong preference for foreign debt securities denominated in U.S. dollars, with the percentage increasing to 60 percent in 1997 from 48 percent in 1994. U.S. investment became increasingly concentrated in the currencies of the world's four largest economies (United States, United Kingdom, Japan, and Germany) and neighboring Canada, with more than 86 percent of the total denominated in these currencies as of year-end 1997. The total concentrated in the top ten currencies (all developed countries) rose from 92.3 percent in 1994 to 93.5 percent in 1997. In total, 97.5 percent of U.S. investment was denominated in the currencies of the twenty-one developed countries identified in table 9 (footnote 1) or in U.S. dollars.

- 16 -

Maturity Structure of U.S. Foreign Debt Holdings

Table 12 presents the maturity structure of U.S.-owned foreign-held debt securities. Data from the 1994 survey are not available at the same level of detail as that in the 1997 survey and are therefore presented only in five-year increments. Information about maturity dates was available for $504 billion of the $547 billion in foreign debt held by U.S. investors.

Table 12.    U.S. Holdings of Foreign Long-Term Debt Securities, by Remaining Term to
                    Maturity, as of December 31, 1997

                    Billions of dollars except as noted

1. Excludes perpetual bonds.

The maturity structure of U.S.-owned foreign-debt holdings was quite similar at the two survey dates, although the average maturity shortened slightly, with 74 percent of all holdings in 1997 maturing in ten years or less compared with 70 percent in 1994. Investment is distributed fairly evenly in securities that mature in ten years or less and falls off sharply beyond ten years

- 17 -

U.S. Holdings of Foreign Private and Government Debt

In table 13, "government" refers to issues of the central government as well as those of local and provincial governments. The debt of government-sponsored corporations and other quasi- governmental organizations are included in the private column.

Table 13.    U.S. Holdings of Foreign Long-Term Debt Securities, by Country and
                    Sector of Issuer, as of December 31, 1997

                    Billions of dollars

Note: The debt figures given above are less than the total amount of debt measured by the survey as only debt for which the issuer sector was known is included in this table.

U.S. investors held approximately equal amounts of foreign private and government debt issues. For the leading countries shown in table 13, the percentages did not vary significantly by region, and both developed and emerging countries showed the same general distribution. Notable exceptions were Italy and Germany, where U.S. investors held primarily government debt securities. In 1994, U.S. investors showed a stronger preference for foreign government securities, holding 62 percent of their foreign debt in that form, compared with 52 percent in 1997.

- 18 -

U.S. Ownership of Foreign Securities by Industry

Unfortunately, the survey had difficulty obtaining information about the industry classification of many foreign securities. Overall, information was available for only 60 percent of the total value of U.S.-held foreign securities. Thus, the classification presented in table 14 may well present a somewhat misleading picture, as the securities that attracted the most U.S. investment were researched with greater effort than those attracting less attention. Some industries have large, well-known companies that attract a great deal of investment and are correspondingly easy to classify. Other industries, characterized by many smaller, less well known firms, received less attention and were more likely to remain uncategorized.

Industry category information was available for 57 percent of equity securities. Of these categorized equities, more than 40 percent of the total was concentrated in four areas: oil and gas extraction ($86 billion), wired and cellular communications ($76 billion), depository institutions (primarily banks, $65 billion), and pharmaceuticals and medicines ($59 billion).

Approximately 69 percent of the total U.S.-held debt securities were categorized by industry. Of these, almost 80 percent were easily classified issues of governmental organizations. The vast majority of the government debt was issued by foreign central governments. Almost all of the unclassified debt was issued by the foreign private sector.

- 19 -

Table 14.    U.S. Holdings of Foreign Long-Term Securities, by Industry,
                    as of December 31, 1997

                    Billions of dollars

- 20 -

* Less than 0.5.

- 21 -

Chapter 2.  U. S. Holdings of Foreign Long-Term Securities
                   as of December 31, 1999

In this chapter, data from the 1997 survey have been combined with transactions data collected by the Treasury and Commerce Departments. These data have been adjusted for price and exchange rate changes to present the estimated level of U.S. portfolio investment abroad as of year-end 1999. The transactions data are available only by country, split into equity and debt; thus, many of the tables presented in the preceding chapter cannot be replicated as of year-end 1999.

Table 15 shows the overall growth of U.S.-owned foreign long-term securities as well as the change in the levels of U.S.-owned equity and debt securities.

The value of U.S.-owned foreign equities has increased significantly over the two-year period, while the value of U.S.-owned foreign debt has decreased. However, as can be seen in the table, the increase was not due to transactions by U.S. residents so much as to valuation changes in these instruments. This continues the trend experienced during the period between the March 1994 survey and the December 1997 survey, when the valuation adjustments to foreign-held equities increased strongly while the valuation adjustments to foreign debt securities decreased (see table 4).

Table 15.    U.S. Holdings of Foreign Long-Term Securities, Net Transactions, and
                    Changes in Percentage and Valuation, by Type of Security,
                    as of December 31, 1997 and 1999

                    Billions of dollars except as noted

1.    The estimated level of U.S. holdings of foreign securities as of year-end 1999 was calculated by the
       Commerce Department’s Bureau of Economic Analysis by adding price-adjusted and exchange rate-adjusted
       transactions data and valuation adjustments to the 1997 asset survey results.

- 22 -

Table 16.    U.S. Holdings of Foreign Securities, as of December 31, 1997 and 1999

                    Billions of dollars

* Less than 0.5.

The United Kingdom easily maintained its position as the country of choice for U.S. investors, while Japan moved past Canada and into second place by a wide margin (table 16). France moved from sixth to fourth position, and Sweden moved from tenth to seventh. Both France and Sweden moved up the list primarily because of the very strong performances of their equity exchanges. In fact, U.S. investors were net sellers of French securities during the period.

Only the United Kingdom, Japan, and Germany attracted significant amounts of net investment during the period. The new investment in Japan resulted from increased net portfolio purchases of Japanese equities. In the cases of the United Kingdom and Germany, much of the increase was the result of stock swaps following purchases of U.S. firms by U.K. and German corporations.

- 23 -

Table 17.    U. S. Holdings of Foreign Equity Securities, as of December 31, 1997 and 1999

                    Billions of dollars

* Less than 0.5.

U.S. holdings of foreign equities increased almost 60 percent during the two-year period, with more than 75 percent of the gain attributable to the rising value of foreign equities rather than to net acquisitions. Both the United Kingdom and Japan experienced net increases in U.S. holdings and significant gains in the value of existing securities held. On the other hand, France and Sweden attracted little or no net equity purchases, but the value of U.S. equity holdings in these countries almost doubled because of valuation gains (table 17).

- 24 -

Table 18.    U.S. Holdings of Foreign Long-Term Debt Securities,
                    as of December 31, 1997 and 1999

                    Billions of dollars

* Less than 0.5.

Valuation changes depressed the level of U.S. holdings of foreign debt in almost all countries (table 18). This was due to both a strong U.S. dollar and generally rising foreign interest rates. Only four countries (United Kingdom, Mexico, Argentina, and Brazil) showed a net increase on a valuation-adjusted basis over the period. While Canada experienced the largest absolute decline, Italy and Japan had the largest percentage decreases.

The rising value of foreign equities combined with the falling value of foreign debt securities was largely responsible for the rise of the equity-to-debt ratio of U.S. foreign holdings from 69 percent at year-end 1997 to more than 78 percent at year-end 1999. The percentage of equities in U.S. investment in Japan rose from 82 percent to 92 percent over this period.

- 25 -

Chapter 3.   Survey Methodology

This survey attempted to be as comprehensive as possible while also being cost effective and imposing a minimum burden on survey respondents. To accomplish these goals, the major sources of data on U.S. holdings of foreign long-term securities (large custodians and institutional investors) were identified and universally included in the survey. Less likely sources (smaller custodians and institutional investors) were sampled. U.S. private individuals were not directly surveyed. However, most holdings of private individuals were captured by the survey, as these investments are typically funneled through U.S. mutual funds or are entrusted to U.S.-resident custodians for safekeeping. Almost all of the survey data were obtained from the major reporters universally covered, with less than 1 percent obtained from the smaller sampled reporters.

Data for the survey were collected at the individual security level. This was done primarily to ensure accuracy, as experience has shown that data quality is frequently problematic in portfolio investment surveys, whether they measure U.S. investment abroad or foreign investment in the United States. Collecting information at the individual security level also allows for more detailed and comprehensive data analysis.

The approach taken by this survey was to collect data primarily from custodians. U.S. investors who entrusted the safekeeping of their foreign securities to U.S. custodians reported only the name(s) of their custodian(s) and the amount(s) entrusted. However, U.S. investors who entrusted foreign securities directly to foreign custodians or who kept custody of foreign securities themselves were required to submit detailed reports on their foreign holdings. Only custodians or end investors with at least $20 million in foreign securities owned or held in custody were required to report in the survey.

In total, 1,209 firms reported data in this survey, of which 346 reported actual data on foreign holdings and 863 reported the names and amounts that they had entrusted to U.S. custodians. More than 97 percent of the data collected were reported by U.S. custodians, indicating that it is uncommon for U.S. institutional investors to entrust the safekeeping of their foreign securities directly to foreign custodians or to engage in self-custody. Banks were the leading custodians, with 80 percent of the total reported, while broker-dealers reported the other 17 percent held by custodians. In total, 232 different U.S.-resident banks reported custody holdings of foreign securities, of which 79 were branches or subsidiaries of foreign banks. However, these foreign branches and subsidiaries accounted for less than 1 percent of the total custody holdings reported by banks.

A relatively small number of firms dominated the foreign custody business at the end of 1997, with the ten largest respondents reporting 75 percent of the total submitted on the survey. Three institutions reported holdings in excess of $100 billion in foreign securities held for U.S. residents, and sixteen reported at least $10 billion.

- 26 -

In some cases, foreign securities are entrusted to U.S. custodians who do not actually engage in the safekeeping of foreign securities. These custodians, in turn, entrust the securities to other U.S.-resident custodians. To avoid double-counting in these instances, the custodians who passed the foreign securities to other U.S.-resident custodians were instructed to report only the name(s) and amount(s) entrusted rather than report in detail.

Collecting data at the individual security level results in amassing a great deal of data. More than 450,000 records were submitted by survey respondents. Of these records, 370,000 concerned equity securities, and 80,000 debt securities. U.S. investors owned in total at least $1 billion of 159 different foreign securities and held in aggregate more than $10 billion in four foreign securities. The foreign security with the greatest level of U.S. ownership attracted $32 billion in total U.S. investment.

This survey was conducted under the authority of the International Investment and Trade in Services Survey Act (22 U.S.C. 3101 et seq.). Reporting on the survey was mandatory, and penalties could have been imposed for noncompliance. Data collected for holdings as of December 31, 1997, were to be reported to the Treasury Department by March 31, 1998. A copy of the survey forms and instructions are found in appendix B.

Problems Encountered

The data received were subjected to extensive analysis and editing before they were accepted as accurate. Frequent, significant problems were encountered with the data. The most common problems involved the currency of denomination for debt securities, unpriced or questionably priced securities, and the submission of data on ownership of instruments other than foreign long-term securities.

All data were requested to be submitted converted into U.S. dollars at the prevailing exchange rates on December 31, 1997. However, many respondents reported some or all of their positions in foreign currency units. In some cases, this problem was simple to detect, as holdings denominated in Italian lire (exchange rate of 1,768 to 1) or Japanese yen (130 to 1) were relatively obvious. However, in some cases there were no data submitted in currencies with a very high ratio of U.S. dollar to foreign currency unit; or some records submitted were converted to U.S. dollars, and others were in foreign currency units, presenting a greater challenge to the survey staff.

Many securities were submitted without prices. In some cases, this presented no problem because an extensive database of foreign securities had been created from commercial sources and the relevant price could frequently be determined from this database or from data submitted by other survey respondents. But there remained many unpriced securities, and a great deal of time and effort were involved in attempting to determine the price of these securities.

In other cases, securities were reported by different respondents with significantly different prices. These discrepancies had to be investigated and the best available price determined.

- 27 -

Many of the records submitted did not represent foreign long-term securities. Short-term securities, loans, and U.S. securities were also frequently reported. A total of 7,000 submitted records, with a value of $52 billion, were excluded from the survey.

Another complication was that the same foreign security could be reported with different security identification numbers by different survey respondents because many securities trade on exchanges in several countries. One respondent might thus report a security using the U.K. security coding system (SEDOL), while another might use the French security identification code (SICOVAM). To address this problem, a cross-reference database was established that contained all of the codes that a given security used on different exchanges, thus enabling the survey staff to link these records. This database aided in pricing unpriced securities and in determining the best price for a security when multiple prices existed.

Accuracy of the Survey

There is, of course, no way to know with certainty how accurate such a survey is unless one already knows what the final figures should be, in which case conducting the survey has no point. However, the accuracy of the survey can be inferred by examining the survey technique for any inherent biases, evaluating the compiled data to see if they seem reasonable, and comparing the data to any external or internal sources of information that may be available as a point of reference.

The survey as structured has two built-in biases, both of which would tend to cause an undercount. The first, as mentioned previously, is that foreign securities directly owned by private U.S. individuals or by small U.S. institutions (below the size that was sampled) that are not entrusted to U.S.-based custodians were not measured by the survey. This would occur when, for example, a U.S. individual buys Canadian equities on a Canadian securities exchange and entrusts the safekeeping of these securities directly to a Canadian bank. At present, we have no way of knowing the size of such activity but assume that it is relatively small. The United States is part of a group of countries that is exploring ways to measure these holdings. The approach would be for Canadian authorities to measure the holdings of Canadian securities by U.S. individuals and to exchange this information with their U.S. counterparts for similar data on Canadian holdings of U.S. securities by Canadian individuals entrusted to U.S. custodians.

The other downward bias results from the fact that data that should not have been reported on the survey (loans, short-term securities, U.S. securities, etc.) were excluded from the survey results, whereas any data that were not reported on the survey that should have been were typically unknown to the survey staff and thus not included. The size of this effect is unknown. However, as mentioned above, records with a combined market value of $52 billion (divided equally between equity and debt) were excluded from the survey results because the submitted data did not represent foreign long-term securities.

- 28 -

From a reasonableness perspective, the survey data were reassuring. U.S. investment by country generally increases in step with the size of a country's financial markets. At the individual security level, the largest amounts of investment are concentrated in securities with the greatest market capitalization. Total U.S. investment never reached an excessive share of the total market capitalization of a given security for the major securities on which this check was made. The custodians who were expected to report the highest levels of U.S. foreign investment did so, and the country distribution and security types held followed expected patterns. The ratio of equity holdings to debt holdings was as expected based on the previous survey and the monthly transactions data.

As discussed in the introduction, the overall level of U.S. holdings of foreign long-term securities measured by this survey exceeded by 21 percent the level that is derived by adding price and exchange rate adjusted transactions data to the level measured by the 1994 survey. However, this result was not unexpected. The prior survey measured a level 60 percent higher than the previously estimated position constructed as mentioned above and as discussed in the introduction.

The survey results also compare well with other available data sources. U.S. investors who entrusted their securities to U.S. custodians reported the name and amount entrusted to each custodian. These figures were totaled and compared with amounts reported by the custodians, and the results indicated that the amounts reported by the custodians were reasonable. The survey results were also compared with data reported to the Securities and Exchange Commission (SEC). Large institutional investment managers are required to report to the SEC their holdings of U.S.-traded equity securities. These securities can trade either as ADRs or non-ADRs. The reports were aggregated to create total U.S. holdings of ADRs and non-ADRs for each country. When compared with survey data, the results were reassuring. For ADRs, the survey results tended to be slightly higher than the SEC data, which is to be expected, since most but not all U.S. holdings would probably be reported by these institutions. In total, the survey reported much larger holdings of foreign equities than were reported to the SEC. However, again this is to be expected because many foreign equities do not trade on U.S. exchanges.

- 29 -

Appendix A. Statistical Tables

A.1.  U.S. Holdings of Foreign Long-Term Debt Securities, by Sector of Issuer,
         as of December 31, 1997

        Millions of dollars

- 30 -

A.1.  U.S. Holdings of Foreign Long-Term Debt Securities, by Sector of Issuer,
         as of December 31, 1997 (Continued)

        Millions of dollars

- 31 -

A.1.  U.S. Holdings of Foreign Long-Term Debt Securities, by Sector of Issuer,
         as of December 31, 1997 (Continued)

        Millions of dollars

Note: The figures are less than the total amount of debt measured by the survey because only debt for which the sector of the issuer was known is included in the table.

1. Algeria, Gabon, Libya, and Nigeria.

2. Bahrain, Iran, Iraq, Kuwait, Oman, Qatar, Saudi Arabia, and the United Arab Emirates.

- 32 -

A.2.   U.S. Holdings of Foreign Long-Term Securities, by Total Held, Amount Traded
          in U.S. Markets, and ADRs, as of December 31, 1997

          Millions of dollars

- 33 -

A.2.   U.S. Holdings of Foreign Long-Term Securities, by Total Held, Amount Traded
          in U.S. Markets, and ADRs, as of December 31, 1997 (Continued)

          Millions of dollars

- 34 -

A.2.   U.S. Holdings of Foreign Long-Term Securities, by Total Held, Amount Traded
          in U.S. Markets, and ADRs, as of December 31, 1997 (Continued)

          Millions of dollars