Back Matter

Back Matter

International Monetary Fund
Published Date:
June 1992
    • ShareShare
    Show Summary Details

    The remaining category is reserves. Reserves data are obtained from the monetary authorities of each country.

    These systems are sometimes described by other names; one such name is “direct reporting systems,”

    However, some of these 49 countries did not use ITRS as a source for any capital account items. For example, Australia used an ITRS for a few current account items only.

    In the survey, the reporting industrial countries were










    New Zealand




    United Kingdom

    United States

    The reporting developing countries were

    The Bahamas







    Costa Rica




    El Salvador




    Hong Kong













    Saudi Arabia


    Sierra Leone


    Solomon Islands

    South Africa

    Sri Lanka





    Trinidad and Tobago




    The nine reporters in this category were Belgium-Luxembourg, Denmark, Finland, Germany, Italy, the Netherlands, Norway, Japan, and Sweden.

    The five reporters in this category were Australia, Canada, New Zealand, the United Kingdom, and the United States.

    The exception was Canada.

    An enterprise group comprises enterprises related under the terms of the Corporations Law that operate within a single resident sector (such as the banking sector, the nonbank financial sector, or the trading sector).

    In practice the IBLC and the NBB collaborate closely: the NBB acts as the secretariat of the IBLC.

    Under the debtor principie, transactions in foreign securities are allocated to the country of the foreign debtor or issuer of the security.

    Marius van Nieuwkerk and Robert P. Sparling, “Discrepancies Between Netherlands and U.S. Statistics,” in The Netherlands International Direct Investment Position (Dordrecht: Martinus Nijhoff Publishers, for De Nederlandsche Bank, 1985); and Statistical Office of the European Communities, Direct Investment of the European Community 1984 to 1988, a provisional report by Christine Spanneut (Luxembourg: Eurostat, 1990).

    Table 1 also appears as Table 70 in the Report on Capital Flows. In 1987, the four countries accounted for 92 percent of reported world outward reinvested earnings and for 66 percent of reported inward reinvested earnings. Australia also compiles a detailed geographic breakdown of reinvested earnings data. However, Australian data are published for financial years (ending June 30) and cannot be directly compared with the calendar year data of the other countries.

    Bilateral comparisons for 11 countries are shown in Table 72 of the Report on Capital Flows.

    The “Benchmark Definition” was published in Organization for Economic Cooperation and Development, Detailed Benchmark Definition of Foreign Direct Investment (Paris: OECD, 1983).

    There is one exception: short-term flows between depository institutions and their direct investment affiliates presumably reflect the regular business activities of banks rather than a direct investment relationship.

    This example was originally cited in John Walker, “OECD Benchmark Definition of Foreign Direct Investment,” in Statistical News (London: United Kingdom Central Statistical Office, May 1983).

    ln addition, the benchmark definition recommends that outward earnings be compiled by the country of the indirectly held enterprise so that investing countries can see where their earnings are made.

    Suppose a U.S. company directly owns a company in the United Kingdom that owns another in Germany. If the U.S. company makes a loan to the German company, the capital flows should be allocated to Germany and not to the United Kingdom.

    These facilities include foreign-owned “finance affiliates” and “holding companies,” which are sometimes described as Special Purpose Entities.

    Draft fifth edition of the Balance of Payments Manual.

    Allen B. Frankel and Catherine L. Mann, “A Framework for Analyzing the Process of Financial Innovation,” International Finance Discussion Paper No. 283 (Washington: Federal Reserve Board, June 1986), p. 6.

    J. T. Parks, “Accounting Winds of Change—New Accounting Rules for Financial Instruments Are Coming,” Journal of Corporate Accounting and Finance (Winter 90/91), pp. 115-25.

    John Herrón, quoted in “Inside the Global Derivatives Market,” Euromoney (Supplement, November 1990), p. 6.

    William F, Sharpe. quoted in “Sluggish Wall Street Is Rushing into ‘Derivatives’,” Wall Street Journal, November 30, 1990, p. C1.

    Philip Turner, Capital Flows in the 1980s: A Survey of Major Trends. BIS Economic Papers No. 30 (Basle: Bank for International Settlements. 1991)

    This assumption is based on the Japanese data. A 75 percent ratio for bank portfolio investments would mean that over half of all Japanese bank claims on foreign nonbanks would be portfolio securities, much higher than evidently appropriate.

    In 1992, U.S. compilers began to use external sources for foreign assets of U.S. nonbanks in addition to the Treasury reports. This change raises the nonbank asset totals greatly.

    The adjustments in the Balance of Payments Statistics Yearbook, Part 2, Table 3. also include an amount for understated U.S. investment income.

    The balance of payments figures in Table 8 are the amounts published in 1991. They are revised substantially from the 1990 Yearbookversion for the United States that is used throughout the body of the Report on Capital Flows and in Tables 4 and 5 of this paper.

    The earlier years of 1984 and 1985 show greatly reduced errors and omissions when IBS flows and stocks are used.

    The treatment of U.S. bank custody accounts by other BIS reporters is a complex matter, however, with perhaps no possibility of a clear answer. At one extreme is the assumption that BIS reporters treat them as entirely nonbank. For that assumption the balance of payments amount for private nonbank assets in Table 6 would be increased from $18 billion to about $90 billion to include the custodies, a total that would still be still just one-third of the IBS total of nonbank assets. Even including the 1992 revisions leaves the U.S.version well short of the IBS total. An element of U.S. custodies that is entirely unknown is amounts invested in foreign claims for nonresident customers. These are excluded entirely from the reports filed with the U.S. Treasury. BIS reporting banks, however, include these in their positions with U.S. banks to the extent that reporters are treating custody positions as interbank. This tends to make the BIS interbank positions larger than those reported to the Treasury by unknown amounts.

    Relevant documents regarding the EMS and EMCF have been published by the Commission of the European Communities. Developments in the EMS have also been discussed in a number of occasional papers published by the IMF.

    In essence, the EMCF is a set of books maintained at the Bank for International Settlements (BIS), although the BIS itself is not a principal to EMS-member transactions.

    Members may also swap more than 20 percent of their gold and dollar reserves with the EMCF but do not do so in practice. Participants do not swap with the EMCF any nondollar foreign exchange assets in their reserves. Member countries maintain freedom to “manage” the specific dollar assets swapped into the EMCF during the swap interval.

    As noted in the Report on the Measurement of International Capital Flows, however, the EMCF does not report its transactions to the IMF, so as an “international organization” the EMCF is missing from the global system.

    EMCF swap obligations toward EMS members should be treated as liabilities constituting foreign authorities’ reserves (LCFAR) in the global accounts. LCFAR are liabilities that are treated as reserve assets by the creditor economy.

    Although it is the acquisition of domestic currency notes (change in ownership) by nonresidents and the acquisition of foreign currency notes by residents, rather than cross-border flows per se, that defines a capital flow, for convenience this paper will refer to “currency movements.” References to “currency” transactions mean physical currency: bank notes. “Currency stock” as used in this paper means currency outside deposit money banks.

    The countries that gather data usually record net inflows of currency through their banks. This may reflect the tendency for currency to be exported in large numbers of small transactions and eventually to be returned by foreign banks to the home country for deposit credit.

    If net flows of currency are entirely unmeasured by both affected countries, the omission of these capital flows might give rise to discrepancies in national balance of payments accounts, but not contribute to the global capital account discrepancy. However, the exact role played by currency movements (and the exchange of currency balances for assets that may be captured by compilers) is more complicated, so some effect on the capital account imbalance is likely.

    Potential compilation errors are not confined to the small number of identifiable key-currency countries alone. Other countries cannot measure changes in their residents’ holdings of foreign currencies. Which “holder countries” are most affected and by what amount is more conjectural.

    Officials in the United States have long been aware of the large amount of currency outstanding relative to the population. All such currency is included in the M1 measure of the money supply. Federal Reserve officials have begun to make public “guesstimates” as to the share of the currency stock that might be in foreign hands.

    The intent of the research into the “underground economy” problem usually has been to estimate biases in official national accounts statistics on income and expenditure.

    The countries included in this experiment were: Canada, France, Germany, Japan, Italy, Switzerland, the United Kingdom, and the United States. All the figures used in the calculations were taken from the IMF publication International Financial Statistics.

    Note that the reverse might apply for foreign holdings of domestic currency notes. A rise in domestic interest rates often is accompanied by an appreciation of the currency. Actual and expected appreciation might induce foreign holders of the currency to acquire more, especially if their access to other vehicles (such as deposits and securities) is constrained by local regulations.

    Among countries for which charts are not shown, smaller declines in real per capita currency balances were recorded between 1970 and 1990 in the United Kingdom. Apart from oscillations, balances did not change much in Canada, Italy, and Switzerland. An examination of exchange rate movements over the interval suggests some correlation with yen and deutsche mark currency balances.

    For instance, the exchange rate in the early 1980s was well above 220 yen per dollar. The 1990 rate used for the translation was about 145 yen per dollar.

    Austria, Denmark, France, Italy, the Netherlands, Norway, Portugal, and Sweden also participate in this arrangement, which dates to 1969.

    Some estimates of the value share of dollar notes that is outside the United States range as high as 35-50 percent of the total in circulation.

      You are not logged in and do not have access to this content. Please login or, to subscribe to IMF eLibrary, please click here

      Other Resources Citing This Publication