- International Monetary Fund
- Published Date:
- September 1987
© 1987 International Monetary Fund
Library of Congress Cataloging-in-Publication Data
International Monetary Fund. Working Party on the Statistical Discrepancy in World Current Account Balances.
Final report of the Working Party on the Statistical Discrepancy in World Current Account Balances.
Includes bibliographical references.
1. Balance of payments—Statistical methods.
HG3882.I57 1987 381.1’7 87-26196
Address orders to:
External Relations Department, Publication Services
International Monetary Fund, Washington, D.C. 20431
Telephone: (202) 623-7430
List of Tables
Chapters and Appendices
The following symbols have been used throughout this report:
… to indicate that data are not available;
— to indicate that the figure is zero or less than half the final digit shown, or that the item does not exist;
– between years or months (e.g., 1983–84 or January–June) to indicate the years or months covered, including the beginning and ending years or months;
/ between years (e.g., 1983/84) to indicate a crop or fiscal (financial) year.
“Billion” means a thousand million.
Details may not add to totals shown because of rounding.
The analysis and views expressed in this report are the responsibility of the Working Party as a whole, but there were some specific contributions by members of the Technical Staff and others that deserve mention. Overall direction of the research program and the preparation of the Report was assigned to Samuel Pizer. Other members of the Technical Staff participated jointly in many aspects of the study, but most members concentrated on particular topics, as follows:
|Robert L. Sammons||Direct investment income, shipping, and transportation|
|Dietrich Hartenstein||Portfolio income|
|Chester L. Callander||Compilation systems and practices|
|Edna E. Ehrlich||Offshore financial centers and financial innovation|
|Stephen P. Taylor||International banking data and computer applications to questionnaire and other data|
|D. Keith McAlister||Unrequited transfers, international organizations, and liaison with the Fund’s Bureau of Statistics|
In addition we would like to acknowledge the assistance of the staff of the Organization for Economic Cooperation and Development (OECD), in particular Dr. Erwin Veil; the many specialists in the Bureau of Statistics and the Research Department of the Fund who provided essential data and contributed to the analysis in the Report; and a number of persons in financial institutions, the shipping and insurance industries, and in private economic research groups who provided very helpful insights into many topics covered by the Report.
We would especially like to express our sincere appreciation to the many economists and researchers in statistical offices around the world who generously cooperated in this project by completing special questionnaires and responded to specific inquiries of all kinds. We hope this study will prove helpful to the persons who are directly responsible for producing the basic data in this field.
Terms of Reference of the Working Party
The Working Party will investigate the principal sources of discrepancy in global balance of payments statistics, consider various ways in which statistical practices might be amended, and make recommendations.
It is understood that the principal focus of the group’s activities will be the Investment Income and Financial Services accounts, and that particular attention will be given to the role of the offshore centers. In carrying forward its work in this area the group will be assisted by a technical staff, of up to five professionals, that will be provided by the Fund and will be based in Washington.
The Working Party may also consider other sources of discrepancy in balance of payments accounts, if these appear to be of significant importance and amenable to investigation. In undertaking work in these areas, the Working Party may call on the assistance of the Fund staff, the OECD secretariat, or other agencies, within the limits of the resources available.
The Chairman of the Working Party will determine, in consultation with other members, the program of work and the timing of meetings. The final report of the Working Party will be presented to the Managing Director no later than December 1986, and an interim report will be presented no later than December 1985.
Members of the Working Party
|Chairman:||Mr. Pierre Esteva, Ministry of Finance, Paris, France|
|Members:||Dr. Gunter Baer, Bank for International Settlements, Basle, Switzerland|
|Mr. Max Baltensperger, Swiss National Bank, Zurich, Switzerland|
|Mr. Andrew Crockett, IMF, Washington, D.C.|
|Mr. Werner Dannemann, IMF, Washington, D.C.|
|Mr. Piero Erba, Statistical Office, Eurostat, Luxembourg|
|Mr. Michael Feiner, Organization for Economic Cooperation & Development, Paris, France|
|Dr. Mohammed Haider Ghuloum, Central Bank of Kuwait, Safat, Kuwait|
|Dr. Lin See-Yan, Bank Negara Malaysia (Central Bank of Malaysia), Kuala Lumpur, Malaysia|
|Mr. Marius van Nieuwkerk, De Nederlandsche Bank N.V., Amsterdam, Netherlands|
|Mr. Samuel Pizer (Director, Technical Staff), IMF, Washington, D.C.|
|Dr. Kurt Senff, Deutsche Bundesbank, Frankfurt, Federal Republic of Germany|
|Mr. Jack Wells, Central Statistical Office, London, England|
|Mr. Yoneyoshi Yasugi,* Bank of Japan, Tokyo, Japan|
|Mr. Ernesto Zedillo, Director General de Ficorca, Mexico City, Mexico|
|Rapporteur:||Mr. D. Keith McAlister, IMF, Washington, D.C.|
Members of the Technical Staff
Chester L. Callander
Edna E. Ehrlich (Federal Reserve Bank of New York)
Dietrich Hartenstein (Deutsche Bundesbank)
Samuel Pizer (Director)
Robert L. Sammons
Stephen P. Taylor
D. Keith McAlister (Liaison with Bureau of Statistics, IMF)
Martha J. Haldeman
Succeeded Mr. Kozo Tsukagoshi of the Bank of Japan in July 1985.
Executive Summary and Major Recommendations
1. ORIGINS OF STUDY
In the period after 1979, the available statistics on the world current account began to show a large negative discrepancy, indicating that either the deficits of some countries and areas were being overstated, or that surpluses were being understated. Concern that such discrepancies could lead to inappropriate policy reactions was heightened in 1982, when the excess of reported debits exceeded $100 billion—a very large amount when scaled against the recorded deficits of major world areas in that year. Responding to this concern, the Executive Directors of the Fund determined in 1984 to establish a Working Party to investigate and recommend procedures to improve statistical practices. The terms of reference directed the Working Party to focus principally on the investment income accounts and the role of offshore centers.
This report contains the results of the study by the Working Party that was organized early in 1985 and was assisted by the small Technical Staff and supported by contributions from the staffs of the Fund and the Organization for Economic Cooperation and Development (OECD). The Working Party has been able to identify the sources of most of the discrepancy in the investment income accounts, and to suggest procedures for reducing it. The Working Party also initiated analyses of the other large discrepancies in the current account—shipping, other transportation, and official and private unrequited transfers—and laid the groundwork for reduction of these discrepancies in the future. In addition the Working Party examined the role of offshore financial centers and financial innovations in the widening of the discrepancy after 1979 and suggested some measures to limit the erosion of international financial statistics.
Conduct of these analyses required more detailed information than is usually available in national balance of payments accounts as supplied to the Fund. Consequently, the Working Party carried out three comprehensive questionnaire surveys and also engaged in intensive discussions with national compilers. The results of these surveys are included in the Report and should be valuable for future research.
The Working Party emphasizes that improving the world’s data on current account transactions will be a formidable task, especially in an environment where the capacity for statistical measurement is challenged by rapid changes in the technology and forms of international transactions and by budgetary constraints. If the findings of the Working Party are to have a lasting influence, the Fund and other international institutions, as well as the governments of member countries, must assign a high priority to improving these data and ensuring that deterioration does not recur.
2. MAJOR CONCLUSIONS
The Working Party explored a fairly wide range of explanations of the growth of the discrepancy in the global investment income account, but reached the conclusion that the overriding factor was the emergence of a large body of cross-border assets recognized by the debtor countries but not by the creditors, coupled with higher interest rates after 1979. Examination of the reported data on capital flows for the 1977–83 period (Table V) shows an excess of reported net inflows of nearly $300 billion (apart from reinvested earnings). Explanations of the discrepancies in the shipping and unrequited transfer sectors of the current account generally refer to relatively specific and persistent statistical deficiencies that can be corrected with patient effort.
Even though all of the statistical problems cannot be fully overcome, considerable gains can be achieved reasonably soon by applying the estimating techniques and other suggestions contained in the Report. This conclusion applies especially to the investment income accounts, where improved techniques can reduce the discrepancy to minor amounts from the debit of $33.6 billion recorded in 1983 and the even larger debit of $47.5 billion recorded in 1984. The solution for the persistent excess of debits in the shipping and transportation accounts (averaging over $35 billion annually) depends mainly on improved national accounting and partly on efforts by the Fund to generate data for the large unreported sector of the world fleet. In the case of the persistent excess of reported debits under the heading of official unrequited transfers (averaging about - $16 billion since 1979) the Working Party has shown where certain large gaps exist and where further research should be directed. These transaction categories accounted for over 85 percent of the negative balance reported for the current account, excluding merchandise trade, in 1983–84.
A further major conclusion of this study is that the indicated adjustments would tend to add net credits to the current accounts as now recorded for most world areas, but the additions would not be so concentrated in any single country or group of countries as to invalidate the basic thrust of analyses drawn from the uncorrected figures.
a. Investment Income
The statistical record of the current account discrepancies (Table I) shows the widening of the overall discrepancy from about minus $20 billion in 1978–79 to a peak of minus $114 billion in 1982. According to the Fund tabulations, the overall discrepancy remains at high levels, but below the peak. By far the largest and persistently rising discrepancy appears in the investment income accounts. The Working Party was able to identify the sources of this discrepancy and to establish the basis for adjustments that nearly eliminate it, as is shown in summary form in Table II.
|Reinvested earnings on direct investments||6.7||11.8||11.2||10.4||7.5||9.9||5.8|
|Other direct investment income||-4.6||0.1||-7.6||-10.7||-11.3||-11.5||-11.7|
|Other investment income||-6.2||-7.3||-11.2||-22.3||-35.9||-32.0||-41.6|
|Other official transactions||-4.0||-9.6||-11.4||-18.3||-24.0||-18.2||-20.5|
|Other private transactions||9.6||6.4||6.2||0.4||-0.4||5.1||1.8|
|Current account (excluding official transfers)||-2.1||-3.0||-14.0||-50.1||-99.1||-62.2||-81.6|
|Current account (including official transfers)||-19.7||-19.4||-34.7||-69.0||-113.9||-75.1||-95.8|
|Memorandum item: Service balance as a percentage of service payments||5.8||5.4||7.1||10.4||12.8||10.9||12.7|
Does not include estimates, which are reported in the Fund’s World Economic Outlook, of certain current transactions of the U.S.S.R. and other countries of Eastern Europe that were not Fund members. International organizations do not supply comparable data.
|Total||Reinvested Earnings||Other Direct Investment Income||Portfolio Investment Income|
|As reported by Fund1||-33.6||9.9||-11.5||-32.0|
|Adjustments as compiled by Working Party||27.7||-10.4||5.4||32.7|
|Income after adjustment||-5.9||-0.5||-6.2||0.7|
As reported in International Monetary Fund, Balance of Payments Statistics Yearbook, Vol. 36 (Washington, 1985).
With respect to direct investment income, the Working Party found that detailed geographic data compiled by the major creditor countries were the primary basis for locating the discrepancies and reducing them over time. In the case of reinvested earnings, the excess of reported credits mainly reflects the fact that most debtor countries do not compile this information. In worldwide compilations, therefore, it will be necessary either to insert an adjustment for the missing debits (computed from the creditor side) or to omit the reinvested earnings entirely from such summations. However, efforts should be maintained to encourage all countries to compile this essential information.
With respect to other direct investment income (dividends, interest, and branch profits), the Working Party was able, by matching the data for individual countries, to identify several sources of inconsistency that could be corrected but did not account for the entire discrepancy. However, the remaining excess of reported debits, about $6 billion, can be traced to relatively few countries and probably can be cut back by further research and consultations. In general, the experience of the Working Party indicates that the discrepancies in the direct investment income accounts can be substantially reduced by exploiting the detailed data compiled by some major creditor countries, and by encouraging national compilers to share their information in this area directly or via the Fund.
By far the largest discrepancy shows up in income flows related to portfolio investments, including interest on positions held with or by banks, as well as returns on cross-border holdings of securities, trade credits, and miscellaneous assets. Income flows related to official reserves are included under this heading because they cannot be segregated consistently in the records of the creditor and debtor countries. After considerable experimentation with estimation methods, the Working Party determined that the key to the problem was to establish an independent basis for estimation. The principal basis capable of providing consistent figures across countries and between creditors and debtors is information on the stocks of the main types of cross-border assets and liabilities. Fortunately, a large part of that basis already exists in the International Banking Statistics (IBS) compiled by the Bank for International Settlements (BIS) and the Fund, and the Working Party was able to create consistent estimates of the income associated with banking positions using that base. Of course, much additional work was done to effect country-by-country corrections (including revisions supplied by national compilers on the questionnaire issued by the Working Party), and further adjustments were made to deal with special cases—such as the international institutions themselves—and to incorporate the income related to portfolio holdings of securities. Carrying through the same types of adjustments applied to 1983, the Working Party was able to reduce the discrepancy on world portfolio income to minor amounts for each year since 1979. The detailed description of these procedures, and the results obtained, which are presented in this report, should provide adequate guidance hereafter to national compilers and to the Fund for the preparation of improved estimates of this type of income.
b. Shipping and Transportation
The negative discrepancy in the shipping and transportation accounts is quite large, but, unlike the income discrepancy, it shows no tendency to increase over time. (See Table I.) The Working Party has been able to develop a body of information on these transactions. It believes its analysis points to the main sources of error, and it has established a basis for more comprehensive work on this topic. In the case of the shipping account, which primarily relates to expenditures and earnings associated with the carriage of international trade, the large excess of reported debits results mainly from the fact that whereas all countries are able to compile fairly readily the amount paid to foreign-operated carriers to cover the freight on imports, several economies with large maritime interests (notably those of Greece, Hong Kong, and Eastern Europe) do not report the corresponding freight earnings of their fleets. There are a number of complicating factors, but the discussion in Chapter VII indicates that progress can be made through a combination of improved reporting by maritime countries and preparation of estimates, where necessary, by the Fund staff.
A smaller discrepancy in the “other transportation” account reflects offsetting factors: the lack of data on the port expenditures of the “missing” fleet (debits) and underestimates by some countries of their receipts from the port expenditures of foreign ships (credits). Here also, the analysis by the Working Party indicates where the gaps are largest, and suggests a procedure for correcting the data as now reported. As with other suggestions, this would require coordination between national compilers and the Fund staff, in order to—among other things—permit the development of estimating techniques that would be consistent across countries.
c. Unrequited Transfers
Finally, the Working Party has undertaken a detailed analysis of the reasons why there is a large negative discrepancy in official unrequited transfers, offset in part by an opposite discrepancy for private transfers. This work has been aided by the cooperation of the Development Assistance Committee (DAC) of the OECD. As noted above, the findings of the Working Party show where some immediate sizable corrections can be made and will also help to guide the additional research leading to basic improvements in the data.
d. Geographic Allocations
An important object of this study was to determine whether the discrepancies were associated disproportionately with particular countries or areas, resulting in serious errors in the appraisal of their external positions. Therefore, the Working Party has attempted, as far as is possible at this time, to allocate its suggested adjustments to geographic areas. This can be carried out fairly well for the investment income accounts (see Table III), but since there are large stocks of assets that cannot be allocated geographically (not to mention cross-border assets that are hidden and do not appear in any available statistics), it cannot be assumed that the result for any country or area represents a full tally of assets or income. Nevertheless, it seems reasonable to conclude that most of the missing income credits can be assigned to industrial countries and lesser amounts, to most other country groups. These modifications of the reported income data seem to imply that there is a need for only a moderate revision of the broad picture of the experience of the major country groupings: the industrial countries are shown to have considerably larger net investment incomes; the Middle Eastern oil exporters also emerge with larger net incomes; and the other developing countries remain large net payors of investment income, though the amount paid is somewhat reduced.
|As Reported||Adjustments||Adjusted Data|
|Industrial countries||-1.7||+ 18.3||16.6|
|Middle Eastern oil exporters||18.5||+ 6.0||24.5|
|MOB centers2||0.4||+ 2.3||2.7|
|Other developing countries||-60.6||+ 6.2||-54.4|
|Eastern European countries3||—||-3.7||-3.7|
|International organizations||—||+ 3.1||3.1|
|Unallocable by country||—||+ 5.9||5.9|
Includes “other” direct investment income and portfolio income, but excludes an allocation of the adjustment for omitted direct investment reinvested earnings The latter adjustment, for any given country or area, would also automatically require an equal offsetting entry in the capital account, with no net effect on the overall balance of payments.
Major offshore banking centers; Hong Kong, Singapore, Bahrain. The Bahamas, the Cayman Islands, the Netherlands Antilles, and Panama.
Includes those Eastern European countries that were not Fund members.
Geographic allocations of the adjustments needed for the shipping accounts are less well established at this time. It seems clear that there are sizable missing net credits attributable to ship operators associated with Greece, Hong Kong, and Eastern Europe, and there may also be a generalized understatement of receipts from ships’ expenditures in port. There may also be a widespread overstatement of shipping costs, but this would be matched in principle by an understatement of merchandise imports on an f.o.b. basis in the overall current account.
In the case of the excess of debits in the official transfer accounts, about one third can be traced to the lack of reporting by international institutions themselves and to a few specific situations. However, it will require further analysis by the Fund, together with the OECD, to establish a firm basis for allocating the remaining discrepancy between donor and recipient countries. A tentative and illustrative set of allocations is shown in Table IV and is discussed in Chapter IX.
|Current Account Balance as Reported1||Indicated Adjustments2||Adjusted Balances|
|Industrial countries||-18.5||+ 14.0||-4.5|
|Developing countries||-56.6||+ 31.2||-25.4|
|Major offshore financial centers||-0.9||+ 2.9||2.0|
|Middle Eastern oil exporters||-13.3||+ 11.5||-1.8|
|Other developing countries||-42.5||+ 16.8||-25.7|
|International organizations||4||+ 3.1||3.1|
International Monetary Fund, Balance of Payments Statistics Yearbook, Vol. 36 (Washington, 1985), Part 2.
Includes only countries that were not Fund members.
Not included in BOPS Yearbook data, adjusted balances reflect only indicated adjustments.
e. Future Prospects
The main causes that have contributed to the emergence and growth of a global discrepancy in current account balances of payments will still exist in the future. It is not likely that the enlarged flow of capital across national borders will be reduced and the trend toward integration of financial markets will probably continue. The shipping or transfer account discrepancies will probably not decline if present statistical practices remain unchanged. Therefore, national compilers and the Fund will continue to be confronted with the challenge of reconciling differences in reported balance of payments figures, even though there will be improvements as countries and the Fund effect the changes we recommend.
The chapters of the Report devoted to direct investment, portfolio investment, shipping and transportation, and official and private unrequited transfers include specific suggestions designed to fill some of the gaps that have been detected at the various stages of the study. Some of these recommendations are also intended to encourage a diversification of compiling and estimating methods with a view to achieving a more consistent and comprehensive coverage of cross-border flows. But in addition, the Report presents some suggestions that are wider in their scope and should contribute in the future to more effective analysis of international economic developments.
3. MAJOR RECOMMENDATIONS
The Working Party believes that considerable progress could and should be made in reducing the discrepancies both at the level of global aggregation and at the level of the statistics prepared in individual countries. It is also convinced that lasting improvements can only be made if close cooperation and effective interchange of views and information characterize relations between the Fund and the national authorities responsible for preparation of the international accounts. Moreoever, senior policymakers must express their interest and their strong support for these recommendations if they are to have their intended effect.
(a) National statistical agencies, with the assistance of the Fund staff, should implement the specific suggestions for adjustments given in the Report, modifying them when necessary to meet individual circumstances, but aiming at overall international consistency.
(b) The Fund should give a high priority to the furtherance of the improvements suggested in the Report. The Bureau of Statistics should prepare a plan of action to implement these recommendations in the period immediately ahead. Over the longer run, consideration should be given to establishing a core of experts in the Bureau of Statistics to pay close attention to each of the types of international transactions that is important and for which the quality of the data shows a tendency to deteriorate. The Working Party recommends enlisting the support of experienced national compilers, through either general meetings of experts or more limited consultations with specialists in such fields as shipping or financial innovation, as continuing reference points for questions that arise in the field of balance of payments accounting.
(c) With respect to direct investment income, the Working Party urges countries with inadequate data on either reinvested or distributed earnings to initiate direct inquiries to parent companies and their affiliates. Countries should coordinate their work in this area, including the publication (or exchange) of as much country detail as possible to facilitate bilateral comparisons. The Fund should support such collaborative efforts by assisting in the exchange of data and the design of statistical inquiries.
(d) With respect to portfolio income, national compilers should take advantage of all the information available on the cross-border assets and liabilities of their residents, especially the comprehensive International Banking Statistics compiled by the Bank for International Settlements (BIS) and the Fund and published regularly by the Fund. Those institutions should continue to improve and broaden these statistics, making further efforts to obtain more complete geographic allocations of banking positions. Efforts should continue to obtain better information on cross-border holdings of securities and to compile statements of international investment positions to complement the flow data in the balance of payments accounts.
(e) With respect to the shipping and other transportation accounts, the Fund should actively pursue the suggestions for closing the, gaps given in the Report, including the collection of relevant statistics for that industry and intensive consultations with countries whose shipping accounts are either missing entirely or appear to be inadequate.
(f) National compilers should be encouraged by their authorities and by the Fund to use flexible statistical estimating processes where it is shown that the results of present data gathering and reporting techniques are not credible. This applies particularly to data collection limited to officially authorized external transactions. Where necessary, statistical offices should be provided with explicit official support to impress on transactors the importance of providing essential information.
(g) Preparation of the balance of payments accounts should be done in close coordination with all the national agencies, including central banks, that can contribute either relevant data or analytical support. Data collection and analytical work should be done in a mutually supportive manner.
(h) The Balance of Payments Manual of the Fund is the cornerstone of the established accounting principles in this field. In the view of the Working Party, it is now time to review and update the current 1977 edition, keeping in mind the need to simplify wherever possible and to give greater priority to the practical application of the principles. Also, the Fund should examine critically the process by which national statistics are supplied to the Fund and recast into global aggregates. Many changes have occurred since these data flows were started, and many countries still provide inadequate data that nevertheless enter into the published record.
(i) The OECD and the Fund should seek to coordinate their work on official unrequited transfers and should draw on the survey by the Working Party to help reduce inconsistencies and discrepancies. Countries should also be asked to achieve greater consistency in reporting on these transactions to these institutions.
(j) The Research Department of the Fund, in collaboration with the OECD, should evaluate the results of the Working Party and consider how best to use these results in revised statistical summations and analyses of world international transactions.
4. STATISTICAL ANNEX
|Cumulated 1964–76||Cumulated 1977–831|
|Of which: Investment income2||…||-170|
|Capital movements (including reserve transactions)2||34||297|
|Errors and omissions||4||111|
Adjusted to exclude reported reinvested earnings because they are recorded only partially and introduce an arbitrary net credit entry in the income accounts (with an offsetting discrepancy in the capital account).
Balances of reported transactions, which in principle should be zero for the world as a whole.