The Identification of Capital Transfers in the Balance of Payments1/
Author:
Mr. Jack Bame https://isni.org/isni/0000000404811396 International Monetary Fund

Search for other papers by Mr. Jack Bame in
Current site
Google Scholar
Close

The separate identification of current and capital transfers was introduced for the first time in the fifth edition of the Fund’s Balance of Payments Manual (Manual), thus harmonizing with the treatment of transfers in the 1993 System of National Accounts (1993 SNA). Capital transfers are now recorded in the capital account component of the balance of payments and include debt forgiveness, migrants’ transfers, and other transfers, of which investment grants is a significant category. This paper presents the criteria for defining capital transfers and provides sources and methods of compilation, and examples of treatment, as illustrated in the Fund’s Balance of Payments Compilation Guide and Balance of Payments Textbook.

Abstract

The separate identification of current and capital transfers was introduced for the first time in the fifth edition of the Fund’s Balance of Payments Manual (Manual), thus harmonizing with the treatment of transfers in the 1993 System of National Accounts (1993 SNA). Capital transfers are now recorded in the capital account component of the balance of payments and include debt forgiveness, migrants’ transfers, and other transfers, of which investment grants is a significant category. This paper presents the criteria for defining capital transfers and provides sources and methods of compilation, and examples of treatment, as illustrated in the Fund’s Balance of Payments Compilation Guide and Balance of Payments Textbook.

I. Introduction

The separate identification of current and capital transfers was introduced for the first time in the fifth edition of the Fund’s Balance of Payments Manual (Manual). The breakdown of unrequited transfers—hitherto all treated as current—reflected the collective decision of national balance of payments compilers, representing all the constituencies of the Fund at the March 1992 meeting of compilers at Fund headquarters in Washington, D.C. The two main factors that contributed to the decision were (i) the objective of as complete harmonization as possible of the Manual with the 1993 System of National Accounts (1993 SNA), and (ii) the growing importance of debt forgiveness and its effect on current account balances of both creditor and debtor countries.

II. The 1993 SNA Criteria for Defining Capital Transfers and Classification of the Balance of Payments Accounts

Separate identification of capital transfers in the balance of payments accounts removed a major discordance between the accounts and the 1993 SNA. 2/ The change also necessitated a reorientation of the former capital account in the balance of payments. In the interest of further harmonization with the 1993 SNA, the same terminology as that used in the 1993 SNA external transactions accumulation accounts was introduced in the Manual—the capital and financial account—as the successor to the former capital account. The capital account component consists primarily of capital transfers (the other component being acquisition/disposal of non-produced, nonfinancial assets); the financial account component is roughly equivalent to the former capital account. The criteria for distinguishing between current and capital transfers in the Manual are based upon those in the 1993 SNA to the extent that the criteria are applicable for international transactions. The Manual classification within each of the two types of transfers, however, was determined on the basis of balance of payments importance and analytical needs.

1. Criteria for defining capital transfers

The following are the 1993 SNA criteria (excerpts from 1993 SNA Chapter 10, The Capital Account) for defining capital transfers and distinguishing between current and capital transfers upon which the Manual’s (refer to Chapter XV and XVII) concepts are based.

article image
article image
article image
article image

2. Classification of capital transfers in the balance of payments

The primary basis for classifying capital transfers in the Manual (in the list of standard balance of payments components) is a distinction between those of general government and those of other sectors of the compiling economy. Within general government, debt forgiveness and other transfers are specified; the breakdown for other sectors is migrants’ transfers, debt forgiveness, and other transfers. A significant category within other transfers for both government and other sectors is investment grants.

Relevant sections from the Balance of Payments Compilation Guide (Guide) and the Balance of Payments Textbook (Textbook), as to sources and methods of compilation, and examples of the treatment of capital transfers, are covered in 3. and 4. below. Concerning the distinction between sectors—general government and other sectors—general government includes transfers between the compiling economy’s government and foreign governments, and transfers between the compiling economy’s government sector and nonresident nongovernment entities. Transfers between the compiling economy’s nongovernment entities and foreign governments are classified as other sector transfers, not as those of the government sector.

3. Guidance from the Guide

Chapter XV of the Guide provides suggestions for the compilation of capital transfer items, particularly the following sections of that Chapter.

654. Table 15.1 shows the primary types of capital transfers and summarizes sources and methods that could be used to compile appropriate entries in the balance of payments. A subsequent section of this chapter provides further details on data sources and methods.

Table 15.1

Compilation of Capital Transfer Items

article image

Data Sources

655. For capital transfers involving the compiling country’s general government sector, the preferred source of information—for both debits and for credits-will typically be official records (described in chapter 8). However, surveys of embassies and international institutions and DAC data could be used, either as primary sources or as checks on official data 3/, in the compilation of certain credit entries.

656. Migrants’ transfers are among the more difficult items to record correctly in the balance of payments. (The balance of payments treatment of migrants’ transfers is outlined in paragraphs 663-665.) Often, such transfers can be measured properly only by use of information from several sources. Information on cash transferred by migrants could be obtained from an ITRS or from ES of banks and other financial institutions. Information on goods transferred by migrants could be obtained from ITS. Financial assets and liabilities (such as bank accounts retained in migrants’ countries of origin) not actually transferred at times of migration are often difficult to measure. Compilers in some countries conduct, at least on an ad-hoc basis, surveys of immigrants to measure these items. Compilers in migrants’ countries of origin may be able to review bank records and similar documents to detect changes of address indicating migration. Alternatively, balance of payments statistics of partner countries could be analyzed. 4/

657. For debt forgiveness associated with other sectors, an enterprise survey of the principals involved could be a source of information. In countries that use ES to measure transactions in, and stocks of, external assets and liabilities, a question relating to debt forgiveness could easily be included. Alternatively, information could be obtained via a supplement to an ITRS. However, use of an ITRS would presume that relevant enterprises are aware of the obligation to provide information on what constitutes noncash transactions.

658. Significant other capital transfers of the nongovernment sector are likely to be relatively few in number and fairly easy to identify. Information on those involving cash could be obtained from an ITRS or, alternatively, the principals involved in those transactions could be surveyed.

Estimation in the Absence of Data

659. For most capital transfer entries, it should be possible to obtain source data of high quality. However, for some items (most notably, migrants’ transfers), it may be necessary to develop estimates by using data models.

660. Migrants’ transfers can be estimated by multiplying numbers of migrants by per capita estimates of net worth. Information on numbers of migrants should be available from migration statistics, which are discussed in chapter 7. Data on the net worth of migrants could be obtained from periodic benchmark studies based on the types of sources described in paragraph 656. These benchmark estimates could be adjusted for inflation, exchange rate variations, and any other relevant factors. Development of a data model for estimating migrants’ transfers is comparable to that of the worker remittance data model described in paragraphs 639-640 of the previous chapter.

Extrapolations and Projections

661. When timely data are not available, estimates of transfer items will have to be extrapolated on the basis of historical trends and any other relevant information that may be available. For example, changes in numbers of migrants could be used to extrapolate migrants’ transfers. The compiler should exercise caution in extrapolating certain other capital transfer items as many-in particular, debt forgiveness-exhibit “lumpy” behavior over a period of time. In these cases, the compiler may prefer to obtain actual data on significant transactions.

662. For projections, the compiler could use historical trends and supplementary information (obtained from principals) on significant transactions that are expected. For example, the compiler should consider budgetary forecasts for projections of certain components of official transfers, such as investment grants. Also, any changes in government policy (for example, a decision to restrict the number of immigrants in future) that may affect capital transfer items should be considered when projections are developed.

Balance of Payments Treatment of Migrants’ Transfers

663. The following example illustrates the balance of payments treatment of migrants’ transfers. A resident of country B migrates to country A. Immediately prior to migration, this person’s assets and liabilities consist of:

article image

664. At the time of migration, the person transferred funds (in the form of foreign exchange) from the bank account in country B to a bank account in country A and moved the goods from country B to country A. For the period in which the migration occurred, the following entries would be required in the balance of payments of country A:

article image

665. The value of the migrant transfer entry is equal to the net worth of the migrant at the time of migration. In the balance of payments of country A, entries are recorded for the migrant’s holdings of shares in country B and bonds in country C and for the migrant’s loan from the bank in country B; the entries are recorded even though these assets and liabilities did not generate any actual transfer of funds to country A at the time of migration.

4. Textbook tips

Chapter VII of the Textbook deals with transfers. In the excerpts below, capital transfers are discussed, including examples with balance of payments entries.

a. Debt forgiveness

438. Unless the cancellation of a debt occurs by mutual agreement of debtor and creditor, the cancellation is not a capital transfer; the writeoff of debt reflects a capital loss, which is not recorded in the balance of payments.

For example, in 1988, the government of Clintonstan extended a long-term loan of 100 units to the government of Algornia and a long-term loan of 200 units to an enterprise in Bushland. In 1989, the Clintonstan government agreed to the request of the Algornian government for forgiveness of one half of the loan. In Bushland, the deteriorating economic situation led to the bankruptcy of the enterprise that had borrowed from Clintonstan’s government. The Clintonstan government subsequently recorded repayment of one half of the loan made to Algornia and wrote off, as a bad debt, the loan made to Bushland.

Clintonstan’s balance of payments statement for 1988 and 1989 would show the following entries:

article image

The forgiveness of one half of the 100-unit loan in 1989 is shown as a partial repayment of that loan, and an offsetting entry is shown under capital transfers. The write off of the 200-unit loan represents a capital loss that should not be recorded in the balance of payments statement. However, the capital loss would be reflected in the market value of the stock of loan assets shown in the IIP as of December 31, 1989.

b. Migrants’ transfers

439. In the BPM, migrants are defined as individuals (other than students; medical patients; or diplomatic, military, or similar personnel) who move to new countries and are expected to remain in the new countries for at least one year. The term migrants’ transfers refers to the household and personal effects and the financial claims and liabilities transferred by migrants from former to new countries. In the strictest sense, these transfers are not transfers between two parties, but are contra-entries to the flow of goods between economies and changes, which arise from migration, in the financial items of economies. These contra-entries are equal to the net worth of migrants.

440. The value of the household and personal effects of migrants and the movable capital goods that they actually transfer to new countries are recorded as transactions in goods, and offsetting entries are made under migrants’ transfers. Migrants’ financial claims on or liabilities to countries of former residence, (for example, bank deposits, shares, bonds, and loans extended or received) are recorded under various components of the financial account; offsetting entries are made under migrants’ transfers. Land and structures owned by migrants and located in former countries of residence and movable capital goods not transferred by migrants to their new countries are treated, according to conventions presented in the BPM, as financial investments made by migrants in notional enterprises that own these assets. In the balance of payments statements of the countries to which the migrants have migrated, these financial investments are recorded as increases in direct investment abroad, and offsetting entries are made under migrants’ transfers. A migrant’s claims on or liabilities to residents of the country to which he or she has moved are treated, in the balance of payments of that country, as if the external claims (liabilities) have been extinguished.

A resident of Domestica is migrating to Essendonia. At the time of migration, this person has the following assets and liabilities:

article image

The balance of payments statement of Domestica should reflect the following entries (Essendonia’s balance of payments statement would show reversed entries):

article image

The goods entry represents the value of the migrant’s personal effects. The net worth (assets less liabilities) of the migrant (1,030 units) is shown under migrants’ transfers. The financial account entry under direct investment abroad shows a decrease in Domestica’s direct investment capital abroad and reflects the fact that the Essendonian real estate is now owned by a resident of Essendonia rather than of Domestica. The entry under direct investment in Domestica shows an inflow of direct investment capital, because Domestican real estate is now owned by a resident of Essendonia rather than of Domestica. The equity securities entry under portfolio investment shows a decrease in external assets consisting of shares in the enterprise in Daniherland. The entry for bond liabilities of the general government indicates that bonds issued by the Domestican government are now owned by a nonresident and therefore represent liabilities of Domestica to Essendonia. The entry under other investment for the deposit assets of other sectors reflects a decrease in Domestica’s deposits with nonresident banks because these deposits are now owned by a nonresident. The entry for the loan assets of banks shows an increase in external assets and reflects the fact that a loan formerly owed by a resident is now owed by a nonresident. The entry for the deposit liabilities of banks shows an increase in external liabilities of banks.

c. Other capital transfers

442. Other capital transfers relate mainly to investment grants. Investment grants are used for adding to or financing the gross fixed capital formation of the recipient economy. For example, an investment grant could comprise financing provided through a foreign aid program for the construction of a dam. Although grants for large capital projects may be paid in installments over extended time periods, each installment is recorded as a capital transfer. General grants made to foreign governments and used for purposes other than financing capital investments are recorded as current, rather than capital, transfers. Inheritance taxes, gift taxes, other taxes on the transfer of assets, and compensation payments (other than those resulting from insurance claims) are also recorded as other capital transfers. Transfers of military equipment that also has civilian uses are recorded as other capital transfers.

III. Summary and Conclusions

1. Some degree of priority should be assigned by balance of payments compilers to improve the coverage of transfers and to distinguish between current and capital transfers.

2. Cooperation and interaction, on the national level, with national accounts compilers—for whom the distinction between current and capital transfers is important for measuring income, consumption, and savings and for analytical purposes—should be a promising approach to facilitate the separate identification of capital transfers in the international accounts.

3. Exchanges of data and discussions (to the extent feasible) between partner countries involved with capital transfers such as debt forgiveness and investment grants—and perhaps migrants’ transfers—may be helpful for identifying and valuing such transfers.

4. Data from official national agencies and from Development Assistance Committee (DAC) statistics on grants and other disbursements, together with aid flow data obtained from surveys of foreign embassies and international organizations often are valuable sources for obtaining capital (and current) transfer statistics. (The Guide provides information and model forms concerning the gathering of these data and techniques for adjusting DAC statistics to conform to balance of payments concepts and other requirements.)

5. The above suggestions may necessitate in some instances, changes in, and/or supplements to, existing data collection methods in some countries. Such changes should (of course, within resource constraints) be supported by compilers.

6. Although it may be difficult to separately identify or classify certain transfers, especially cash transfers, as current or capital, such difficulty certainly is not unique to the transfer component of the balance of payments. The 1993 SNA and Manual suggest that when there is serious doubt as to classification, the transfer should be classified as current rather than capital. However, as noted in the 1993 SNA excerpt in II,1., incorrect classifications will affect the allocation of saving between the compiling economy and the rest of the world. Therefore, “serious doubt” should not be applied as a blanket escape hatch to avoid efforts to correctly classify transfers, but rather as a last resort.

7. Hopefully, the Guide and the Textbook, together with the Manual and 1993 SNA, will assist compilers in the difficult task of distinguishing between current and capital transfers, through the clarification of the concepts involved, through suggested sources and methods of compilation, and through selected examples of proper recording of transfers in the balance of payments accounts. As is evident from the Appendix, however, there is a long way to go before practical implementation can meet conceptual requirements.

Appendix: Collecting Data on Transfers and Distinguishing Current and Capital Transfers 5/

Most respondent countries collect some data on transfers. They view the distinction between current and capital transfers, however, as often being hard to draw in practice. Nonetheless, they are exploring ways to improve their coverage of transfers and to obtain the data needed to comply more fully with the guidelines.

Among industrial countries, the Netherlands pointed out that the distinction between current and capital transfers is often difficult to make in practice because of insufficient information on the recipient’s use of the transfer. Under such circumstances, the Netherlands compilers routinely consider the transfer to be a current one. A few of the industrial countries are exploring ways to improve their coverage of transfers. Australia reported that its estimates on transfers are currently categorized under official and nonofficial transfers; nonetheless, it is examining the feasibility of designing a survey to collect information on current and capital transfers. Japan, which currently collects data on most transfers, with debt forgiveness being a notable exception, plans to seek new regulatory authority to permit the expanded coverage. Other industrial countries are reviewing methods specifically to collect information on debt forgiveness, particularly debt forgiveness involving banks and other nongovernmental institutions. Not all industrial countries receive their information directly from the parties to a transfer. New Zealand, for example, develops most of its data on transfers from modeling, although information on government transfers is obtained from survey results.

Few developing countries now distinguish current from capital transfers. Moreover, they appear divided over whether such a distinction could meaningfully be made in the near future. Some asserted that it would be particularly difficult to obtain reliable data on investment grants, migrants’ transfers, workers’ remittances, and humanitarian aid. Others suggested, however, that some of the needed data could be found in records of foreign exchange transactions and other sources. The Philippines, for example, acquires data on transfers using reports of commercial banks on foreign exchange acquisition and disposition as well as reports of donor countries. Detailed breakdowns, such as of investments grants, are not now made. Jamaican compilers reported that they obtain information on official transfers on an annual basis from government sources. In addition, they glean information on cash gifts, grants, and other private transfers from foreign exchange forms filed on a monthly basis by financial institutions. They conceded that non-cash private transfers are difficult to capture but pointed out that they survey relevant organizations annually and that additional information on gifts in kind can be obtained from customs declarations. Authorities in the Czech Republic stated that they plan to modify their transaction codes to produce more complete data on current and capital transfers. More generally, however, few developing countries appear to assign a high priority to expanding reporting on transfers.

1/

The views expressed in this paper are those of the author and do not necessarily represent those of the International Monetary Fund.

2/

Another major item of discordance was eliminated as a result of the 1993 SNA, for the first time, including reinvested earnings from direct investment as a component of property income.

3/

Surveys of embassies and international institutions are described in chapter 9, paragraphs 380-383, and DAC data are described in chapter 9, paragraphs 388-390.

4/

Because of data collection problems, many countries limit measurement of migrants’ transfers to cash and goods actually transferred. This practice is not consistent with BOP requirements. Countries are encouraged to make, even if only on an infrequent basis, more comprehensive measurements and to assess the significance of differences between limited and comprehensive measures.

5/

“Implementing the Fifth Edition of the Balance of Payments Manual - A Report on Progress Made by Selected Countries,” IMF Statistics Department, 1995, pp. 11-12.

  • Collapse
  • Expand