Abstract
Managing Editor
Shuja Nawaz
The concept of a world current account balance in economics is analogous to Newton’s Third Law of Motion in physics. For every surplus on a national current account there is, in principle, an equal deficit somewhere in the world, so that the world current account must sum to zero, with each deficit matched by an offsetting surplus. In other words, the amounts paid by international purchasers of goods and services should be matched, if all the bookkeeping is correct, by the receipts shown by the sellers of those goods and services. Yet, for the past decade or so, the global current account has shown a growing discrepancy, particularly in the accounts maintained by the International Monetary Fund. From a rough balance in the early 1970s emerged an imbalance of about $20 billion by 1978, increasing more than fivefold by 1982 to $109 billion, before gradually declining to $56 billion in 1986 (see table).
The growth and persistence of the discrepancy indicated that either the deficits of some countries were being overstated or the surpluses of others were being understated. In any case, the large size of the imbalance, relative to the recorded deficits of major regions of the world, gave rise to international concerns that it could lead to inappropriate policy actions.
To investigate the causes of this growing discrepancy so that it might be corrected, the Executive Directors of the Fund set up in 1984 a Working Party of international experts (including Fund staff) to examine, in particular, the investment income accounts and the role of offshore financial centers in the compilation of world current accounts. This Working Party, assisted by the staffs of the Fund and the Organisation for Economic Co-operation and Development, completed its work in 1986 and presented a report in early 1987. This article draws upon that report to explain the problem and its causes.
Summary of global payments balances on currant account—recorded and adjusted, 1979–86
(In billions of US dollars)
Indicates that the figure is zero or that the item does not exist.
Includes adjustments for which no geographical breakdown is available.
Summary of global payments balances on currant account—recorded and adjusted, 1979–86
(In billions of US dollars)
| 1979 | 1980 | 1981 | 1982 | 1983 | 1984 | 1985 | 1986 | ||
|---|---|---|---|---|---|---|---|---|---|
| Industrial countries | |||||||||
| Recorded | -23 | -62 | -20 | -22 | -22 | -62 | -54 | -9 | |
| Adjusted1 | -27 | -55 | -6 | — | -9 | -44 | -40 | -2 | |
| Developing Countries | |||||||||
| Recorded | 6 | 30 | -49 | -87 | -64 | -34 | -24 | -47 | |
| Adjusted | 4 | 33 | -40 | -70 | -54 | -21 | -12 | -44 | |
| Total | |||||||||
| Recorded | -17 | -32 | -69 | -109 | -66 | -96 | -79 | -57 | |
| Adjusted | -22 | -19 | -42 | -63 | -58 | -58 | -44 | -39 | |
| Other countries | -3 | -2 | -3 | 3 | 3 | 4 | 3 | 1 | |
| International organizations | 2 | 3 | 3 | 3 | 3 | 4 | 5 | 5 | |
| Total, adjusted1 | -22 | -18 | -42 | -57 | -52 | -50 | -36 | -34 | |
Indicates that the figure is zero or that the item does not exist.
Includes adjustments for which no geographical breakdown is available.
Summary of global payments balances on currant account—recorded and adjusted, 1979–86
(In billions of US dollars)
| 1979 | 1980 | 1981 | 1982 | 1983 | 1984 | 1985 | 1986 | ||
|---|---|---|---|---|---|---|---|---|---|
| Industrial countries | |||||||||
| Recorded | -23 | -62 | -20 | -22 | -22 | -62 | -54 | -9 | |
| Adjusted1 | -27 | -55 | -6 | — | -9 | -44 | -40 | -2 | |
| Developing Countries | |||||||||
| Recorded | 6 | 30 | -49 | -87 | -64 | -34 | -24 | -47 | |
| Adjusted | 4 | 33 | -40 | -70 | -54 | -21 | -12 | -44 | |
| Total | |||||||||
| Recorded | -17 | -32 | -69 | -109 | -66 | -96 | -79 | -57 | |
| Adjusted | -22 | -19 | -42 | -63 | -58 | -58 | -44 | -39 | |
| Other countries | -3 | -2 | -3 | 3 | 3 | 4 | 3 | 1 | |
| International organizations | 2 | 3 | 3 | 3 | 3 | 4 | 5 | 5 | |
| Total, adjusted1 | -22 | -18 | -42 | -57 | -52 | -50 | -36 | -34 | |
Indicates that the figure is zero or that the item does not exist.
Includes adjustments for which no geographical breakdown is available.
Why data matter
In a global economy that relies on international trade of goods and services, national economic policy is based on an analysis of data on national and international transactions. A consistent set of accounts, along the lines of similar accounts maintained by other countries, allows a country to assess its economic performance, to conduct a meaningful dialogue with its trading partners, and to adjust its policies in line with developments in its current account. These data are also important in the country’s relationship with the Fund since they are the basis for much of the consultations conducted by the Fund with its members. The Fund, through its work on the Balance of Payments Manual, has provided a uniform approach to the compilation of these data. An aggregation of these national data provides a regional and global view of developments in the international economy. This, in turn, can be used by national staff and the research staff of the Fund to conduct studies of the world economy and assess the role and performance of individual countries or groups of countries (see box “IMF global statistics”).
The emergence, in the early 1970s, of an unexplained discrepancy in the global current account balance highlighted the difficulty of making accurate analyses of global economic trends on the basis of incomplete information.
Similarly, incomplete information on the sources of imbalances affects the quality of multilateral surveillance over exchange rates of the major industrial countries, since an unexplained imbalance creates uncertainty about the accuracy of estimates of the current account of a country or countries. This, in turn, affects the analysis by Fund staff of both the imbalances and the underlying exchange rates, and undermines confidence in projections of world economic growth, such as those published in the Fund’s World Economic Outlook. If major countries adopt deflationary policies to cope with their apparent balance of payments deficits, their actions may produce a global deflationary bias. The growing discrepancy suggests that the growth of global output over 1979-82 may have been understated by as much as one percentage point.
Further, capital flows and exchange rates can be influenced by expectations about changes in balance of payments, and the expectations may themselves be guided by perceived movements in the current account. An inaccurate estimate of the current account may lead a country to take inappropriate action, such as adopting a protectionist policy to counter an expected or growing deficit. For all these reasons, it is important to understand the origins of the global imbalance on the current account, and its implications for particular countries or groups of countries.
Sources of the discrepancy
Broadly speaking, major changes in the nature and speed of international financial transactions in the past decade have been at the root of the growing statistical gaps in the income sector of the global balance of payments accounts. Rapid changes in the technology for effecting transactions and transferring enormous amounts of money across national boundaries emerged in the wake of the liberalization of the international financial system and the closer integration of financial markets. Further, political and economic uncertainties in some countries meant that investment flows emanating from those countries were concealed by investors from domestic authorities.
Large-scale financing through traditional investment centers and newer offshore centers, using indirect channels (such as foreign subsidiaries of multinational banks and other financial institutions based in tax havens) complicate the process of recording and attributing capital flows to their countries of origin. These changes, coupled with the emergence of a growing service sector in international trade, made it well nigh impossible for national authorities to track and record all transactions involving their residents, or to make reasonably accurate estimates. Traditional merchandise trade has been recorded for very long periods, partly because it is often subject to duties or regulation, so that data on such transactions are relatively well established. They happen to balance out fairly well, for the world as a whole. Service transactions by their very nature are intangible and difficult to define and measure; they do not have a long tradition of statistical enumeration. There is now an international effort to improve these data.
IMF global statistics
The Balance of Payments Statistics Year-book of the Fund serves as the repository of data reported to the Fund by member countries. Fart 1 of this Yearbook carries balance of payments statistics of individual Countries as reported by them in accordance with the procedures and categories spelt out in the Fund’s Manual for production of such statistics.
A separate volume, Part 2, summarizes the information from Part 1 and, based on further estimation and adjustment by Fund staff, presents regional and global balances. This is where the discrepancy in the global current account shows up.
Based on three comprehensive questionnaire surveys giving details of data reported by a broad cross section of national statistical agencies, detailed discussions with national compilers of balance of payments statistics, and intensive analysis of available data on international financial flows, world shipping, and unrequited transfers, the Working Party identified the main sources of the gap in the global current account.
Much of the discrepancy in global accounts resides in the failure of investment income accounts to capture transactions accurately. The emergence of a large body of “crossborder assets,” or capital held in other countries, has been recognized by most debtor countries but not by creditors in their national accounts. High interest rates after 1979 that raised the return on these investments exacerbated the inconsistency in reporting of such assets. Over the period 1977-83, for example, some $300 billion in reported net inflows (excluding reinvested earnings from these investments) did not show up in the accounts of the countries of origin of these flows. An examination of individual types of investments and related income reveals the extent and nature of the discrepancy.
Direct investment income. This includes after-tax income from foreign affiliates of multinational companies, as reflected in tax accounts of creditor and host countries-including any undistributed profits. These earnings and other activities of multinational enterprises are difficult to trace given the complicated structures of these operations and lack of detailed and consistent information from country to country. Many countries have difficulty in collecting the necessary data from these enterprises. Even when data are available it is often difficult to allocate direct investment income geographically on a consistent basis because the entity that is the source of income may be owned through a multinational subsidiary based in a third country.
Reinvested earnings. This component of income from direct investment includes portions of the net profits from overseas operations that are not declared as dividends to the home countries of investors. In 1983, some $13.9 billion in reinvested earnings were reported by investing countries to the Fund, whereas only $4.4 billion of debits were reported by countries where these investments had been made, leaving $9.9 billion unaccounted for. The main problem is that many debtor (host) countries do not collect or calculate data on such reinvested earnings, whereas the major creditor countries collect and report these credits in their accounts.
Other direct investment income. This covers dividends, branch profits, and interest paid by affiliates to the parent company. The main source of the very large discrepancy in the reporting of this item was inadequate and inconsistent reporting of such income. The information needed must come from the enterprises concerned, and, of course, the operations of multinationals are very complex and measurement and geographic allocation of source of income is somewhat arbitrary. The Working Party has suggested that the statisticians in creditor and host countries cooperate to achieve a better matching of credits and debits. There was an excess of debits of some $11.5 billion in this account in 1983, with host countries debiting some $33.8 billion in such payments while investing countries reported receipts of only $22.3 billion. The Working Party was able to identify the specific origins of about half of the discrepancy; Fund staff will make further efforts to reduce the discrepancy.
Portfolio investment income. This covers interest and dividend payments to and from banks and other financial institutions, including interest and dividends on marketable securities and other financial instruments, such as mortgages and suppliers’ credits.
A very large and rapidly growing net debit is shown when the global accounts are aggregated for this account. The Working Party traced the origins of most of this discrepancy to the fact, mentioned above, that over the years inflows of capital reported by countries receiving funds have been much larger than was recorded in the accounts of countries where the funds originated. Thus, the debtor countries recognize these growing liabilities, and are able to enter in their accounts a reasonably full record of the interest payable to the nonresident owners, whereas the countries where the owners reside have no record of the accumulating capital and therefore do not record a comparable amount of income credit.
The Working Party showed that most of the inconsistency in reporting can be identified and eliminated by using the cross-border banking data collected and published by the Fund in International Financial Statistics as a check on the income flows of individual countries. In addition to helping to eliminate the global discrepancy, the use of the detailed geographic data in the banking statistics allows an estimate of the geographic distribution of the missing income receipts.
This is the fastest growing and currently the largest item of discrepancy in the current account balance. It accounted for 43 percent of the gap in reporting of income from services in 1984. Some $42 billion more of portfolio investment income was debited in 1984 than showed up in the credits reported for that year by investing countries.
Shipping and other transportation. International shipping has an intricate network of ownership and operating responsibility. Moreover, a number of countries whose residents operate large fleets do not report freight earnings or fleet expenditures to the Fund. Consequently, a substantial part of world freight earnings and ships expenditures are not covered in the global accounts. The Working Party traced most of the problem to the incomplete coverage, together with some specific cases of incorrect reporting. Many countries also understate their earnings from port expenditures (a major portion of “other transportation”). While the global discrepancy in the shipping and transportation accounts is quite large--a net debit of over $30 billion-it is relatively steady, and therefore is less likely to lead to errors in the analysis of current trends.
Unrequited official and private transfers. These are entries in the balance of payments covering provision of goods, services, or cash by one party to another without any obligation for repayment. The main types of official transfers are development assistance, direct and indirect, including, among other things, debt relief and payments to international organizations. Private transfers cover a range of activities, most notably remittances by migrant workers.
For official transfers, donor countries showed some $14.2 billion more in debits in 1984 than was evident in the $35.2 billion of inflows indicated by recipient countries. The Working Party located part of the origins of this discrepancy-including the fact that data for international organizations were being left out-and has provided guidance for further analysis. Private transfers show an excess of $3.7 billion reported by countries as credits in 1984 over debits of some $33.6 billion reported by countries of origin of these flows.
Role of offshore centers. Traditionally, a small number of major centers, such as London, New York, Amsterdam, Zurich, and Paris, have been involved in the international movement of financial resources. With the opening up of the international financial marketplace, and in response to the demand for specialized services, many new centers opened up or expanded their operations during the past decade, making it even more difficult to follow and report accurately on all international financial flows. The major new offshore banking centers include the Netherlands Antilles, The Bahamas, Bahrain, the Cayman Islands, Hong Kong, Singapore, and Panama, accounting for a substantial share (20 percent) of foreign assets of banks that used international financial centers in 1985.
The growth of offshore banking centers was accompanied by the emergence of innovative financial instruments with considerable effect on accounting techniques. The newer instruments, together with the greater use of swaps and options, tend to obscure the ownership of assets as between residents and nonresidents--a crucial problem for international accounting. Moreover, the changes in banking practices have tended to move these instruments off banks’ balance sheets; the nonbank creditors and debtors are much more difficult to reach as sources for the data on crossborder capital and income flows, thus complicating the job of national and international authorities in compiling full and accurate data on financial flows. The existence of the offshore centers adds another layer of difficulty in measuring flows, and encourages investors and their intermediaries to attempt to conceal their activities. Moreover, the growing use of interest rate and currency swaps, and floating rate notes has created a mercurial accounting situation; as highly specific and highly liquid new instruments emerge and change ownership rapidly, their trading becomes less susceptible to recording as between residents and nonresidents in balance of payments accounts.
Removing the discrepancy
As the Working Party examined the sources of discrepancy or the accounts that reflected most of the asymmetry in the global current account, it collected additional detailed data via questionnaires and located specific examples of erroneous reporting as well as more generalized sources of error. Corrections were suggested for individual variations and gaps partly by closer examination of bilateral national accounts, and suggestions were made for changes in methodology to be applied by all countries and the Fund. The adjustments made by the Working Party reduced the estimate of the global discrepancy from $109 billion at its height in 1982 to $57 billion, and in 1986 from a reported $56 billion to some $33 billion. Broadly speaking, the Working Party found that over time a more comprehensive and uniform application of flexible statistical techniques for compiling and reporting data to the Fund, rather than dependence on information from banks on foreign exchange dealings, would resolve many of the problems leading to the imbalance in the global current account. Nevertheless, the negative discrepancy in the global balance of payments data is still quite large and further effort will be required to reduce it.
Following the release of the Working Party’s report to the Executive Directors of the Fund (see box on publication), efforts were begun to improve the collection and reporting of data on each of the items found to give rise to discrepancies in the global current account. An important part of this effort will be the updating and improvement of the Fund’s Manual for balance of payments reporting, to take into account the innovations in the global financial marketplace and the difficulties identified in tracing transactions in various sectors. The techniques used by the Working Party are expected to provide guidance for the improvement of procedures both in the Fund and its member countries to collect and publish global financial statistics.
The Final Report of the Working Party on the Statistical Discrepancy in World Current Account Balances is published by the IMF. Copies may be ordered from Publications Services, IMF, Washington D.C. 20431 USA.