Linkages Between External Debt Data and Balance of Payments, Government Finance and Monetary Statistics
Author: Ms. Jan Bové1
  • 1 0000000404811396 Monetary Fund

This paper examines the linkages between key macroeconomic statistics: external debt, balance of payments, government finance, and monetary statistics, which enable analysts and policy makers to monitor economic developments. The paper reviews statistical compilation issues and concludes that, while considerable progress has been made in recent years to account for external debt and related flows, further progress must be made. It makes recommendations, such as improved communication among compilers of related statistics, that may help compilers to achieve greater consistency between statistics concerned and thereby enhancing their usefulness for analysts and policy makers. The need for increased resources to develop statistical systems is also recognized.


This paper examines the linkages between key macroeconomic statistics: external debt, balance of payments, government finance, and monetary statistics, which enable analysts and policy makers to monitor economic developments. The paper reviews statistical compilation issues and concludes that, while considerable progress has been made in recent years to account for external debt and related flows, further progress must be made. It makes recommendations, such as improved communication among compilers of related statistics, that may help compilers to achieve greater consistency between statistics concerned and thereby enhancing their usefulness for analysts and policy makers. The need for increased resources to develop statistical systems is also recognized.

I. Some Macro-Relationships Involving External Debt, the Balance of Payments, Government Finance, and Monetary Accounts

Macroeconomic policy making requires that the authorities take a global view of the economy in order to achieve their growth, employment, and other policy objectives, while avoiding imbalances in the economy that might endanger these objectives over the long-run. Such imbalances may typically arise, inter alia, in financial markets, where excessive supply of money might increase the pace of inflation, or in foreign exchange markets, where excessive demand for foreign exchange might put downward pressure on the exchange rate and the willingness of foreigners to invest in the country. Imbalances are bound to occur from time to time, either because of shortcomings in the design of economic policies, or because of various exogenous shocks that may cause the economy to deviate from its balanced long-term growth path.

It is important, therefore, that the authorities have access to a set of macroeconomic statistics, which are an essential tool to alert them as imbalances occur in various markets, and to allow them to take appropriate action to bring the economy back on track. In order to be useful, macroeconomic statistics need to be timely, accurate, and consistent in their methodology. This paper explains the conceptual relationships between key macroeconomic and financial concepts, and what is required to maintain consistency between the various sets of accounts.

The starting point for this analysis is the relationship between the national economy and the rest of the world. It can be represented by an equation stating that the current account balance, excluding capital transfers, is equal to the gap between gross domestic saving and investment, namely:



article image

This conceptual identity, which has its equivalent in the balance of payments and national accounts data, shows that an increase (decrease) in domestic saving vis-à-vis investment, must necessarily be matched by strengthening (weakening) the current account position of the country.

Let us now explore some of the linkages between the government’s spending and taxing decisions, and the external position of a country. If we denote government revenue by Rg and government expenditure as the sum of Cg (i.e., consumption expenditure) and Ig (i.e., capital expenditure), and if we assume, as is mostly the case in practice, that the government has virtually no capital revenue, then we can represent government deficit as Rg - Cg - Ig, and government saving (or dissaving in the case of deficits) as follows:


If we identify within the relationship represented by equation (1) a government sector, then that equation can be rewritten as:


with subscripts p and g denoting the private and government sectors, respectively.

This equation, which relates the external sector to, inter alia, government finance, shows that the budgetary position of the government may be an important determinant of the current account of the balance of payments. Thus, a sustained current account deficit may reflect a persistent excess of government spending over receipts, and this can only be maintained through sustained autonomous capital inflows. The latter condition follows from the balance of payments identity stating that a current account deficit (surplus) must, per definition, be financed by net capital inflows (outflows) and/or a decrease (increase) in reserves, or:



NKA = Net capital account

CRA = Transactions in reserve assets

since we know from equation (1) that

CAB = S - I, it follows that


The latter shows that, if capital accumulation exceeds domestic saving, this will necessarily require (net) capital inflows from abroad. Since capital flows lead to changes in the stock of foreign financial claims and/or liabilities, a current account deficit must give rise to a decrease in foreign assets or an increase in external debt, i.e., external liabilities for which there is a contractual obligation to repay principal and/or pay interest. This relationship between the current account of the balance of payments and the accumulation of external debt may give rise to a destabilizing situation where the current account progressively deteriorates unless effective policy action is taken to generate sustained autonomous capital inflows.

Using the identity shown in equation (3), equation (5) may be rewritten as:


Equation (6) implies that, assuming unchanged private sector behavior, an increase in government investment that is not matched by an increase in domestic saving will cause the current account to deteriorate, which will necessarily lead to an increase in external debt or other financial liabilities (net of any reserve or other financial assets that can be disposed of). The increase in external debt may reflect direct financing of the new investment projects through bank loans from abroad. Alternatively, the increase in external debt may be the result of capital inflows in response to an increase in domestic real interest rates relative to those abroad which would, given sufficient investor confidence, induce such capital inflows. This scenario could occur, for example, if the additional investment is financed from domestic bank loans, while the monetary authorities contain the resulting expansionary effect on the money supply through non-accommodating monetary and financial policies.

In another scenario, excess investment demand (by the private and public sectors combined) may be generated by an expantionary monetary policy, with a rapidly expanding money supply. In this case., the concomitant inflationary environment may adversely affect confidence of foreign investors, which in combination with relatively low real interest rates, will not be conducive to spontaneous capital inflows, and the authorities may be forced to finance the rising current account deficit with reserve assets or through an increase in external debt. In either case, the long-term sustainability of the current account deficit, which may reflect higher imports of capital goods as well as the increased debt servicing burden, will ultimately depend on whether new investment leads to an increase in the productivity of the economy, which should generate the income required to service the additional debt burden.

II. Some Key Macroeconomic Statistics and their Linkages

1. Principal issues in consistency between statistical systems

An important function of the Fund is to give macroeconomic and financial policy advice to member governments. The Fund has, therefore, a longstanding interest in the key macroeconomic statistics on which such policies should be based. Over the years, the Fund has gained considerable experience with those statistics and has been assigned the responsibility for developing methodologies in balance of payments, government finance, and money and banking statistics; these methodologies are described in the fourth edition of the Balance of Payments Manual (BPM), A Manual on Government Finance Statistics, and A Guide to Money and Banking Statistics in International Financial Statistics. Care has been taken to make these methodologies generally consistent among themselves and with the United Nations’ A System of National Accounts (SNA), which is in the process of being revised. To the extent that full consistency with the SNA could not be assured (often because of practical considerations and because each system is designed to meet special needs), attempts were made to identify elements in the systems that permit them to be linked to the SNA. In the preparation of the fifth edition of the BPM, particular attention is being given to enhancing the consistency between the balance of payments and the draft SNA. Although the Fund relies on external debt statistics in its work, much of the methodological work in this area has been developed by the World Bank in the context of its Debtor Reporting System (DRS) and the Organization for Economic Cooperation and Development (OECD) in the context of its Creditor Reporting System (CRS). The Fund collaborates on external debt methodology issues with the Berne Union, the OECD, the Bank of International Settlements (BIS), as well as with the World Bank, in the International Working Group on External Debt Statistics (IWGEDS). In developing its methodology on external debt statistics, the World Bank has adopted many of the underlying principles of the Fund’s balance of payments methodology, with a view to assuring consistency between such data sets. The issue of statistical consistency is also a principal concern of the IWGEDS.

All the systems mentioned above are broadly consistent, with some exceptions, in terms of the underlying principles, namely:

(i) For the purpose of compiling statistics, the coverage of a national economy is determined by the residency criterion (rather than a nationality or currency criterion). Specifically, the data should cover transactions and assets and liabilities of economic units whose center of economic interest is in a country’s economic territory, which is defined to consist of an area administered by a government within which persons, goods, and capital circulate freely. These economic units are considered to be residents of the economy concerned. Among residents of an economy one identifies general government, non-profit organizations, enterprises, and individuals. External positions or transactions measured by economic and financial statistics involve relationships between residents and nonresidents of an economy.

(ii) The time period in which transactions must be recorded in the statistics is the period in which transactions (involving changes in the ownership of real and financial assets) actually take place. In the fourth edition of the BPM, debt service (i.e., interest and principal repayments) are to be recorded when the payment is due. In the draft fifth edition of the BPM it is proposed, in conformity with the draft SNA, that interest be recorded as it is accrued (e.g., if interest on a bond is due yearly on March 31, then three fourths of the interest needs to be recorded in the balance of payments for the preceding calendar year). It is noteworthy that the methodology espoused in the Manual on Government Finance Statistics differs from the other Fund systems, in that it records transactions on a cash basis, i.e., at the time money changes hands, rather than on due for payment basis. However, with the gap between cash and due for payment data increasing very significantly in recent years, mainly because of the phenomenon of payments arrears, the Fund now routinely presents government finance data also on a due for payment basis for its operational work.

(iii) The valuation of transactions and positions is based on the market prices underlying the transactions, while the stock of external assets and liabilities are valued with reference to the relevant end-of-period market prices or proxies thereof. In this regard, the external debt methodology tends to deviate from the other methodologies, as debtors tend to record their foreign liabilities at face value rather than at market prices. If conversion of currencies into a single currency (or unit of account) is required to aggregate data in a single set of statistics, as is often the case when external positions or transactions are concerned, the generally accepted principle is to use the market exchange rates that were applicable at the time that the transaction occurred (in practice, one takes usually an average rate for the period during which several transactions took place or the time for which a position is recorded (e.g., external debt outstanding at the end of the year).

(iv) Some key economic statistics are stock concepts: mainly financial assets and liabilities or real assets outstanding at one point in time. The most important examples are data on national wealth (which also includes nonfinancial or “real” items), international investment position (IIP) data, external debt data, and banking or financial survey data. Economic analysis often requires data that permit one to gauge dynamic relationships, namely flow or transaction data. Important examples are production and income accounts, government finance statistics, and the balance of payments. In many cases, stock data need to be transformed into flow data, in order to analyze them in conjunction with other flow data. However, changes in stock data between two points in time do not necessarily represent transactions in the assets or liabilities concerned during the intervening period, which is usually the object of flow data; there may also be changes in stock data that are the result of “nontransaction factors”, such as variations in the market value of the liabilities or assets concerned, or changes in the exchange rates at which they have been converted to a common unit of account, or liabilities that are unilaterally repudiated, etc. None of these “nontransaction factors” involve new contractual relationships between the parties concerned.

2. Links between external debt, balance of payments, government finance, and monetary statistics

As was mentioned above, methodological consistency is a necessary condition if macroeconomic statistics are to form a reliable basis for economic analysis and policy making. Additionally, if these statistics are to be used together to provide a global view of economic conditions, it is important that the compilation systems be compatible, and that the classifications provided in each of the systems permit variables in one system to be related to the other systems.

Let us first look at the linkages between the external debt statistics and the balance of payments. These two statistical systems are closely related to each other, albeit not in a direct way.

Gross external debt has been defined by the IWGEDS as the amount, at any given time, of disbursed and outstanding contractual liabilities of residents of a country to nonresidents to repay principal, with or without interest, or to pay interest, with or without principal.” 3/ Therefore, external debt can be viewed as a subset of the IIP of a country, because the latter is defined as a statistical statement, as of a given date, of the value and composition of the stock of an economy’s financial claims on and liabilities to the rest of the world, plus monetary gold and special drawing rights (SDRs). The IIP will be discussed in depth in the fifth edition of the BPM. The IIP is directly related to the capital account of the balance of payments via an analysis, which decomposes variations in the IIP, during a period, into transactions recorded in the balance of payments and valuation changes or other adjustments. It is worth noting that the proposed standard classification of the IIP in the fifth edition of the BPM, will be fully compatible with the standard classification of interest payments or earnings and of the capital account of the balance of payments, although it may involve some regrouping of balance of payments items.

Although the IIP is conceptually closely related to external debt’ statistics, its coverage is wider. Firstly, because it includes claims as well as liabilities, and secondly, because it includes all financial instruments, instead of just those that involve a contractual obligation to pay interest, with or without principal or to repay principal with or without interest. The additional coverage concerns essentially equity type liabilities, classified either under direct investment (only the nonequity components of direct investment are included in foreign debt) or under portfolio investment. Although most debt instruments are rather easy to identify, there are some less evident cases, most importantly financial leasing and financial derivatives. It is generally accepted that cross-border financial leasing, while it does not, strictly speaking, involve a change in ownership and creates a liability to repay, is—as far as its purpose and financial consequences are concerned—akin to a debt instrument; as a result, an imputation should be made accordingly, both in the IIP (and the balance of payments) and in external debt statistics.

The components of the IIP and the balance of payments are shown in Attachments I and II, respectively. The standard IIP items marked with an asterisk in Appendix I are generally accepted to fall within the bounds of the “core” definition of external debt mentioned above. It should be noted that these, as well as the entire text of the draft fifth edition of the BPM, are subject to further revision before its publication.

Since users of external debt statistics, are especially interested in the way debt is created and how it is serviced, stock data presented in debt statistics are normally accompanied by information on debt flows. This information then allows, in principle, analysts and compilers of balance of payments statistics to make the link between the two data sets; additionally, for compilers, it facilitates the use of data coming from debt offices as an input for balance of payments statistics. For the purpose of the current account of the balance of payments, transactions data needed from debt statistics are interest payments and the incurrence of arrears thereon. Any commission charges should be identified separately from interest charges, because the former represent services rather than income and should be recorded as such in the balance of payments (in the current account). For debt securities, any discounts from, or premiums over, the face value (“par”) at issue, represent in fact interest and should be recorded as such in the balance of payments at the time that the securities are redeemed. However, in the draft fifth edition of the BPM it is proposed to record all interest, including discounts or premiums on the face value, on a full accrual basis (see above). Moreover, in the draft fifth edition of the BPM, a much finer classification of interest payments has been included than is the case for the current edition of the BPM; this will permit one to relate interest more closely to the categories of debt on which the charges were accrued. In debt statistics, interest arrears, when recorded, are mostly shown under short-term debt or as a memorandum item, but the debt service data, being on a cash basis, do not include unpaid interest. In the balance of payments, increases in interest arrears are recorded, as accrued, with an offsetting amount shown under short-term liabilities in the capital account.

For the capital account, transactions information is required with the appropriate breakdown by instrument (direct investment, portfolio investment, use of Fund credit, drawings on and repayments of long-term loans, short-term loans, deposits, and other liabilities), by maturity, and by domestic sector. Data on bonds (included under portfolio investment) and long-term loans are normally (or should be) readily available in official external debt monitoring units (which, in turn, may obtain these data from administrative records or other sources); collection of data on direct investment, which may have a debt component in addition to an equity component, usually requires special surveys; data on deposits are drawn from balance sheets of banks (which requires the derivation of transactions data from stock data); data on other short-term accounts (e.g., trade accounts) are the most difficult to collect in any compilation system. Reconciliation of debt data with the balance of payments also requires information on the accumulation of payments arrears, which, in external debt statistics, tend to be left untouched as part of long-term debt (if the original maturity exceeded one year), while for balance of payments purposes, a long-term repayment is recorded, offset by an increase in short-term liabilities. Information on payments of arrears should moreover, be monitored separately for interest and principal in order to permit a reconciliation of debt and balance of payments statistics. Any delayed payment of interest that was not paid when due but allowed to fall into arrears, tends to be shown as interest payments in debt service data of external debt statistics, as opposed to short-term capital repayments in the balance of payments, at the time that actual cash payments are made. Any debt restructuring agreed with creditors following external payments problems tend not to be reflected in external debt statistics, except in the stock data when the debt is assumed by a government agency or the central bank, in which case the amounts involved are reclassified accordingly. On the other hand, any contractual agreement that replaces another one, such as in debt restructuring, is deemed to represent a transaction in balance of payments methodology and should, therefore, be recorded in the balance of payments. Consequently, any detailed information on such restructuring should be collected to assure the links between the external debt and the balance of payments statistics.

Let us now look at the linkages between external debt and the budget accounts or more precisely, government finance statistics. Because government finance data are compiled in terms of flows, the linkages are established with reference to borrowing and debt service payments, either directly or via balance of payments statistics.

It should be noted that government finance statistics may differ in coverage from budget accounts. The latter are defined in the budget laws of countries, leading to widely different interpretations of what constitute government operations. On the other hand, government finance statistics, as well as other macroeconomic statistics, include in the government sector the transactions of all institutions that carry out public policy through the production of non-market services primarily for collective consumption and the transfer of income, financed mainly by taxes. Such a definition can be uniformly applied in all countries. Strict adherence to this statistical definition is important for the compilation and the comparability of, especially, government finance and external debt data, partly because governments may route external borrowing and/or debt service transactions through off-budget accounts which does not, however, reduce the economic impact of such transactions. On the other hand, operations of government units engaged in financial, commercial or industrial activities (whether they are covered in the budget or whether they are undertaken by corporate or quasi-corporate enterprises owned or controlled by government) should not be included in the government sector in macroeconomic statistics, but rather in a more broadly defined public sector. External debt statistics often adopt the public sector as a basis for classification, whereas for balance of payments classifications a more narrowly defined sector is used. 4/ While government finance statistics are established separately for central and general government, for the purpose of reconciliation with external debt transactions data, only central government statistics are usually relevant, as lower levels of government are generally not empowered to borrow abroad.

In government finance statistics, external debt operations are recorded as in the balance of payments, in part, under current expenditures (namely, interest payments), where they affect the savings of government and, in part, under financing (namely, borrowing and amortization). As is the case for the balance of payments, any commission charges are recorded separately from interest as payments for financial services. Similarly, discounts from and premiums over for on debt securities are recorded under interest at the time that the securities are redeemed. In contrast with the balance of payments, interest is recorded when actual cash payments are made, rather than when due and not paid. 5/ As mentioned before, with the increasing incidence of arrears, most analysts now record arrears in government finance statistics whenever these occur. Furthermore, in contrast with balance of payments statistics, interest paid on the defaulted debt of others (when the government has extended a guarantee on the debt concerned) is only recorded under interest if the government has actually assumed the debt (i.e., when it has become the principal obligor). Otherwise, it is recorded as an advance made to the defaulting party (i.e., under “lending minus repayments”) if the government expects to recover the sums involved, or as a current transfer payment, if it does not. With regard to arrears, the same observations made earlier on interest, apply equally to financing transactions.

Government finance statistics provide for the classification of borrowing and repayment by type of debt holder and debt instrument. These classifications are shown in Attachment III. The proposed breakdowns are broadly consistent with those in the balance of payments and do not normally require further details to link them to external debt statistics flows.

Let us now turn to the monetary or banking survey, which, inter alia, relates the assets of the banking system, in particular domestic credit and net foreign assets, with its liabilities, in particular, the money supply of the country. As was mentioned before, the banking survey is usually presented in terms of stock data as are the external debt statistics. The linkages between the banking survey and external debt statistics are rather limited as they concern only the external liabilities on the balance sheets of domestic banks, including the monetary authorities. It should be noted that information on any external long-term borrowing or debt securities issued by domestic banks and held abroad, should be identified separately by external debt monitoring units even if these are classified with such borrowing and securities of other sectors of the economy in the debt statistics, in order to permit reconciliation with the banking survey data.

III. Compilation Issues

While macroeconomic linkages are well established at the theoretical level and considerable efforts have been made to achieve methodological consistency between the principal macroeconomic statistics systems, in practice, it is rare that full consistency is achieved by the compilers of each of these data bases. It is, therefore, unfortunately often up to the analyst to impose consistency on these data so that a coherent analysis of economic conditions can be made. Such adjustment “after the fact” can, however, only be an approximation and are not a good alternative for well designed compilation systems that produce coherent data on a continuous and timely basis.

A set of tables for a hypothetical country, based broadly on relationships typical for Africa, illustrates the linkages that were discussed above, is presented in Attachment IV, Let us first review the linkages between these tables as they bear on external debt and debt flows. The critical time series have been highlighted in each of the tables.

A first external debt linkage is found in national accounts where interest on foreign debt is included under “net factor income from abroad” (Table 1), Interest is also recorded (on a gross basis) in the current account of the balance of payments (Table 2); in the external debt data on debt service (Table 5a, which, only covers public sector debt) 6/ and in the central government finance data (Table 6, the coverage, again, is smaller than that of the entire public sector debt) 7/. The fact that the government some times falls behind in making interest (or other) payments to foreign (or other) creditors, and, at times, makes up for earlier delays, shows up in the difference of the government deficit on a “commitment” and on a “cash” basis. The monetary survey data do not provide a link in respect of interest payments, because they reflect balance sheet positions rather than income statements of banks.

Similar links can be traced between these data sets with regard to borrowing and repayments of debt. 8/ Table 3 shows balance of payments data on disbursements of long-term official borrowing and amortization of official long-term debt on a due for payments basis. The differences with repayments on a cash basis are recorded in Table 4, as exceptional balance of payments financing, under net accumulation of arrears and rescheduling of arrears, and in Table 5, which provides details on external debt flows for the public sector with respect to the creditor status of the debt in the context of rescheduling arrangements. Data on external borrowing and repayments are also shown in Table 6, as they relate to central government debt.

Short-term capital flows, which cover both liabilities and assets are shown as a residual item together with errors and omissions in the balance of payments (Table 3). As is suggested by this presentation, these data tend to be highly tentative as few countries actually measure short-term capital movements and, even less, short-term capital stocks. These data are, therefore, not reflected in other tables, except to the extent that they are, in principle, included in government finance data (in the residual item for financing in Table 6 and in the banking survey in Table 7, under foreign assets and liabilities).

As was mentioned above, the original source material underlying this set of tables lacks in coherence in many respects, which requires the analyst to make ad hoc adjustments on the basis of independent research of sources that may not have been tapped by the compilers of the original data, or to opt for what he believes to be the most reliable among conflicting data, which are then substituted in the statistics where data are believed to be of inferior quality.

Following are some of the shortcomings that are often encountered in the basic national statistics (including those presented in the attached tables) that give rise to these inconsistencies.

Frequently, exchange rates used to convert debt data from the currency of denomination into the national currency or another currency unit (often the U.S. dollar) differ from those used by balance of payments or government finance statistics compilers, especially (but not only) when data need to be transformed from stock debt data to flow data. At times, the methods used to transform stock into flow data when foreign currency denominated liabilities are involved, are inappropriate and /or differ between the compilers of the various data sets. A recommended procedure is shown in Attachment V.

Differences in coverage are a frequent occurrence and are often due to data being compiled from different original sources, such as administrative records used for government finance and external debt stock data, as opposed to central bank exchange rate records used by balance of payments compilers.

Typical cases are drawings on project loans, which may not be recorded in the Treasury ledgers used for government finance statistics, but information on which may be available to the debt monitoring office and the balance of payments compiler from the creditor records or records maintained by the executing agencies concerned. Alternatively, prepayments of commercial or repayments of military loans are sometimes not adequately captured by debt monitoring offices, but are included in the balance of payments, thanks to information obtained in exchange records. The linkages between the data sets are often adversely affected by differences in the revision status. The frequency of adapting or revising the various data sets often differs in many countries.

Differences in data can frequently be traced to timing differences. The Treasury accounts may make temporary imputations of debt related expenditures before debiting the appropriate final expenditure accounts from which government finance statistics are compiled. The exchange records, which may be used for the balance of payments, will often identify the payment as it is made to the foreign creditor.

Differences very often stem from inadequately developed data sources (or, more broadly, statistical systems). This often leads to more or less tenuous estimates that are frequently made independently by the various compilers for their own purposes. In some data sets, the gaps may not even be filled with estimates. Typical cases are data relating to private non-guaranteed debt and short-term debt.

Differences can also be traced to inadequate knowledge of statistical methodologies by compilers, which gives often rise to errors in classification, coverage and valuation of debt or debt related transactions. Evidence of this is often found, e.g., in the confusion between debt owed to nonresidents and debt denominated in foreign currencies. Also, the principles for recording of transactions related to debt restructuring arrangements (e.g., reschedulings, debt/equity swaps, debt for nature swaps) are very often not well known and implemented. This tends to compound the problem that not all compilers concerned are equally informed on the terms and implications of debt restructuring.

Even though significant problems remain in the compilation of external debt related statistics in general, and the linkages in this regard between the various data sets concerned, there is a general recognition within the Fund and elsewhere that very substantial progress has been made in recent years in improving the measurement of external debt transactions in Africa and in other regions. A principal reason for progress in this area has been the multitude of external debt restructuring arrangements, which has forced compilers to collect and present reliable data for negotiations with creditors.

However, further progress clearly needs to be made. Experience of Fund and World Bank economists as well as national experts seems to indicate a few crucial ways in which improvements could be made—often at a relative low cost.

Further improvements in the data as well as the linkages between the related data sets hinge in most countries, first and foremost, on improved communications among the compilers of related statistics and between the compilers and the users of the data. Improved contact with users should lead to greater awareness of better designed data systems that permit the joint analysis of data to obtain a global view of the economic situation. Better communications among compilers could at least begin to address many of the problems enumerated above, such as the coexistence of conflicting ad hoc estimates to fill data gaps (this is especially useful for estimating the most current data, for which some compilers (sources) may have a comparative advantage over others), and the tracing of shortcomings of some data sources through verification against alternative sources used by other compilers. Special mention needs to be made in this context to communication between those responsible for negotiating debt restructurings and the various statistical compilers.

Improved linkages between the macroeconomic data sets also depends crucially on better knowledge of, and stricter adherence to, internationally accepted methodologies, in particular, those developed by the United Nations, the Fund, and the World Bank, in close collaboration with national experts.

These low-cost remedies may, however, not suffice in all cases: where basic data sources are deficient, additional resources may need to be devoted to the development of new data sources and statistical procedures. The strategies needed in this regard will evidently depend on the problems of individual countries.

Appendix I

International Investment Position: Standard Components

article image

The standard components proposed in the draft fifth edition of the BPM call for a breakdown of this item into the following domestic sectors: Monetary authorities, General government, Banks, and Other sectors.

Denotes item that is generally accepted to fall within the bounds of the “core” definition of external debt.

Appendix II

Balance of Payments: Standard Components

article image
article image
article image


I. External Financing by Type of Debt Instrument

Long-term bonds

  • Issues

  • Amortization

Short-term bonds and bills

Long-term loans not elsewhere classified

  • Drawings

  • Amortization

Short-term loans and advances not elsewhere classified

Other liabilities

Changes in cash, deposits, and securities held for liquidity purposes

II. External Financing by Type of Debt Holder

From international development institutions

  • Drawings

  • Amortization

From foreign governments

  • Drawings

  • Amortization

Other borrowing abroad

  • Bank loans and advances

  • Supplier credits

  • Other borrowing abroad not elsewhere classified

Change in cash, deposits, and securities held for liquidity purposes

Appendix IV

Table 1.

Sample Country—Gross Domestic Product by Expenditure Category at Current Prices, 1981–90

(In millions of domestic currency)

article image
Source: Staff estimates.
Table 2.

Sample Country—Balance of Payments: Current Account, 1985–90

(In millions of U.S. dollars)

article image
Source: Staff estimates.

The exclusion of official transfers from the current account reflects a working assumption that those transfers represent capital transfers.

Table 3.

Sample Country—Balance of Payments: Capital Account, 1985-90

(In millions of U.S. dollars)

article image
Source: Staff estimates.

The inclusion of official transfers in the capital account reflects a working assumption that those transfers are not used for current expenditures.

Table 4.

Sample Country—Balance of Payments: Financing, 1985-90

(In millions of U.S. dollars)

article image
Source: Staff estimates.
Table 5a.

Sample Country—External Debt Service of the Public Sector, 1990

(In millions of U.S. dollars)

article image
Table 5b.

Sample Country—Public and Publicly-Guaranteed Debt Stocks, 1989-90

(In millions of U.S. dollars)

article image
Source: Staff estimates.
Table 6.

Sample Country—Central Government Finance Operations, 1986–90

(In millions of national currency)

article image
Source: Staff estimates.

Grants are normally not a financing item, but are at times treated as such for some analytical purposes.

Table 7.

Sample Country—Monetary Survey

(In millions of national currency; end of period)

article image
Source: Staff estimates.

Appendix V: Evaluation of Claims and Liabilities in Foreign Currencies

Valuation changes may stem from either or both price changes in the asset in terms of the transaction currency and changes in exchange rates. This feature and the relationship between stocks, transactions, and valuation changes may best be illustrated with the following example. Assume that at the beginning of the accounting period, an economic entity held ten units of U.S. Treasury securities, valued at $2 million dollars each, and that during the period under discussion five units were sold at a market value of $2.2 million each. Assume also that the remaining five units have further appreciated and that their market price in terms of U.S. dollars at the end of the accounting period was $2.5 million per unit. The exchange rate of the U.S. dollar vis-à-vis the unit of account is assumed to be US$1 = 1.1 units of account at the beginning of the accounting period, US$1 = 1.3 units of account at the time of the transaction, and US$1 = 1.5 units of account at the end of the accounting period. The value of transactions, stocks, and valuation changes of these securities expressed in millions of U.S. dollars and in millions of units of account would appear as follows:

article image

It should be noted that the value of the opening and closing stock as well as the transaction value in terms of the unit of account, have been derived by converting the values in terms of U.S. dollars at the exchange rates indicated in the table, whereas the total value of the valuation changes in terms of the unit of account has been derived as a residual, As shown, the total consists of two components (1) the valuation change reflecting the change in the price of the financial claim in terms of the transaction currency, which in the example has been derived separately by converting its value in terms of the transaction currency at the exchange rate prevailing at the beginning of the accounting period, and (2) the valuation change reflecting the change in the exchange rate of the transaction currency vis-à-vis the unit of account, which can only be derived as the residual.

Another observation concerning the total valuation changes, either in terms of U.S. dollars or in terms of the unit of account, is that they are a combination of both realized and unrealized capital gains. In terms of the transaction currency, the realized and unrealized capital gains amount to US$1 million and US$2.5 million, respectively; in terms of the unit of account, they are UA 3.30 million and UA 7.75 million, respectively.


I wish to thank Mahinder Gill and Kevin O’Connor of the Statistics Department and Thomas Klein of the World Bank for useful comments.


This equation assumes that there are no errors and omissions, i.e., the balancing item which, in practice, assures equality between credit and debit entries in the balance of payments.


World Bank, International Monetary Fund, Bank for International Settlements, Organization for Economic Cooperation and Development, External Debt, Definition, Statistical Coverage and Methodology, Paris 1988.


In the current edition of the BPM, the basis of the sectoral classification is the “official sector,” encompassing both government and the monetary authorities; in the draft fifth edition of the BPM it is proposed to distinguish between these two sectors.


Note the change to full accrual accounting of interest payment proposed for the balance of payments in the draft fifth edition of the BPM discussed in section III.1. above.


External debt of other sectors tends to be small for most African countries.


Government often carries the main burden of foreign interest payments, especially in countries where the government has taken over the debt service burden of public enterprises.


Borrowing and repayments are recorded in the capital account of the national accounts, which is, however, not represented in this set of tables.


The standard components proposed in the draft fifth edition of the BPM call for a breakdown of this item into the following domestic sectors: Monetary authorities, General government, Banks, and Other sectors.


Includes suppliers’ credits.


Includes utilization/redemption of suppliers’ credits.


Includes utilization/redemption of suppliers’ credits.