The World Bank Methodology on external debt
The World Bank is the sole repository for statistics on the external debt of developing
countries on a loan-by-loan basis. The Debtor Reporting System (DRS), set up in 1951 to
monitor these statistics, is maintained by the staff of the Financial Data Team (FIN),
part of the Development Data Group of Development Economics.
Methodology
for aggregating data
Using the DRS data, in combination with information obtained from creditors through the
debt data collection systems of other agencies such as the Bank for International
Settlements (BIS) and the Organization for Economic Cooperation and Development (OECD),
the staff of the Financial Data Team estimate the total external indebtedness of
developing countries. The data are also supplemented by estimates made by country
economists of the World Bank and desk officers of the International Monetary Fund (IMF).
Converting to a
common currency
Since debt data are normally reported to the World Bank in the currency of repayment,
they have to be converted into a common currency (usually U.S. dollars) to produce summary
tables. Stock figures (such as the amount of debt outstanding) are converted using
end-period exchange rates, as published in the IMF’s International Financial
Statistics (line ae). Flow figures are converted at annual average exchange rates
(line rf). Projected debt service is converted using end-period exchange rates. Debt
repayable in multiple currencies, goods, or services and debt with a provision for
maintenance of value of the currency of repayment are shown at book value. Because flow
data are converted at annual average exchange rates and stock data at year-end exchange
rates, year-to-year changes in debt outstanding and disbursed are sometimes not equal to
net flows (disbursements less principal repayments); similarly, changes in debt
outstanding including undisbursed debt differ from commitments less repayments.
Discrepancies are particularly significant when exchange rates have moved sharply during
the year; cancellations and reschedulings of other liabilities into long-term public debt
also contribute to the differences.
Public and
publicly guaranteed debt
All data related to public and publicly guaranteed debt are from debtors except for
lending by some multilateral agencies, in which case data are taken from the
creditors’ records. These creditors include the African Development Bank, the Asian
Development Bank, the Central Bank for Economic Integration, the IMF, the Inter-American
Development Bank, and the International Bank for Reconstruction and Development (IBRD) and
the International Development Association (IDA). (The IBRD and IDA are components of the
World Bank.)
Starting with the 1988–89 edition of World Debt Tables (as this book was
previously titled), all data pertaining to World Bank loans from 1985 onward are recorded
at their current market value. Starting with the 1991–92 edition, all data pertaining
to Asian Development Bank loans from 1989 onward are recorded at their current market
value as well.
Private nonguaranteed
debt
The DRS was expanded in 1970 to incorporate private nonguaranteed long-term debt.
Reports, submitted annually, contain aggregate data for disbursed and outstanding debt,
disbursements, principal repayments, interest payments, principal and interest rescheduled
for the reporting year, and projected payments of principal and interest. Data are usually
presented in dollars and currency conversion is not necessary. A few reporting countries
choose to provide data on their private nonguaranteed debt in the loan-by-loan format used
for reporting public and publicly guaranteed debt. In those cases the currency conversion
and projection methodology just described is used.
Although the reporting countries fully recognize the importance of collecting data on
private nonguaranteed debt when it constitutes a significant portion of total external
debt, detailed data are available only in countries that have registration requirements
covering private debt, most commonly in connection with exchange controls. Where formal
registration of foreign borrowing is not mandatory, compilers must rely on balance of
payments data and financial surveys.
Thirty-four countries report their private nonguaranteed debt to the DRS. Estimates are
made for twenty-eight others that do not report but for which this type of debt is known
to be significant.
For private nonguaranteed debt that is not reported, the standard estimation approach
starts from a calculation of the stock of debt outstanding, using data available from
creditors. Figures on guaranteed export credits, obtained from the OECD’s Creditor
Reporting System (CRS), are supplemented by loan-by-loan information on official lending
to private borrowers and by information on noninsured commercial bank lending to the
private sector.
Disbursements and debt service payments for private nonguaranteed debt are more
difficult to estimate. Amortization is estimated by making an assumption regarding the
proportion of debt repaid each year and then applying these ratios to generate a first
approximation of annual principal repayments. Disbursements are then estimated as a
residual between net flows (equal to the change in the stock of debt) and estimated
amortization. Interest payments are estimated by applying an assumed average interest rate
to the stock of debt outstanding.
Data on the balance of payments flows provide useful guidelines in the process of
building a time series because private nonguaranteed debt can be treated as a residual
between total net long-term borrowing and net long-term borrowing recorded in the DRS for
public and publicly guaranteed debt.
Short-term debt
The World Bank regards the individual reporting country as the authoritative source of
information on its own external liabilities. But for short-term debt, defined as debt with
an original maturity of one year or less, accurate information is not widely available
from debtors. By its nature, short-term debt is difficult to monitor; loan-by-loan
registration is normally impractical, and most reporting arrangements involve periodic
returns to a country’s central bank from its banking sector. Since 1982 the quality
of such reporting has improved, but only a few developing countries have figures available
for short-term debt.
Where information from debtors is not available, data from creditors can indicate the
magnitude of a country’s short-term debt. The most important source is the BIS’s
semiannual series showing the maturity distribution of commercial banks’ claims on
developing countries. Those data are reported residually. However, an estimate of
short-term liabilities by original maturity can be calculated by deducting from claims due
in one year those that had a maturity of between one and two years twelve months earlier.
There are several problems with this method. Valuation adjustments caused by exchange
rate movements will affect the calculations, as will prepayment and refinancing of
long-term maturities falling due. Moreover, not all countries’ commercial banks
report in a way that allows the full maturity distribution to be determined, and the BIS
data include liabilities only to banks within the reporting area. Nevertheless, combining
these estimates with data on officially guaranteed short-term suppliers’ credits
compiled by the OECD gives what may be thought of as a lower-bound estimate of a
country’s short-term debt. Even on this basis, however, the results need to be
interpreted with caution. Where short-term debt has been rescheduled, the effect of lags
in reporting and differences in the treatment of the rescheduled debt by debtors and
creditors may result in double counting if short-term debt derived from creditor sources
is added to long-term debt reported by the country to obtain total external liabilities.
Some of the short-term debt estimates published are drawn from debtor and creditor
sources, but most are from creditor sources. Only for a few countries can the data be
regarded as authoritative, but they offer a guide to the size of a country’s
short-term (and, hence, its total) external debt. The quality of these data is likely to
improve.
Back to top
Use of IMF credit
Data related to the operations of the IMF come from the IMF Treasurer’s Department
and are converted from special drawing rights (SDRs) into dollars using end-of-period
exchange rates for stocks and average over the period exchange rates for converting flows,
as described earlier. IMF trust fund loans and operations under the structural adjustment
and enhanced structural adjustment facilities are presented together with all of the
Fund’s special facilities (the buffer stock, compensatory financing, extended fund,
and oil facilities).
Treatment of arrears
The DRS collects information on arrears in both principal and interest. Principal in
arrears is included and identified in the amount of long-term debt outstanding. Interest
in arrears of long-term debt and the use of IMF credit is included and identified in the
amount of short-term debt outstanding. If and when interest in arrears is capitalized
under a debt reorganization agreement, the amount of interest capitalized will be added to
the amount of long-term debt outstanding and the corresponding deduction made from the
amount of short-term debt outstanding.
Treatment of debt
restructurings
The DRS attempts to capture accurately the effects of the different kinds of
restructurings on both debt stocks and debt flows, consistent with the circumstances under
which the restructuring takes place. Whether a flow has taken place is sometimes difficult
to determine.
In compiling and presenting the debt data, a distinction is made between cash flows and
imputed flows. Based on this criterion, rescheduled service payments and the shift in
liabilities from one financial instrument to another as a result of rescheduling are
considered to be imputed flows.
The imputed flows are recorded separately in the Revised External Debt (RXD) system,
but these debt restructuring transactions are not evident in the main body of the debt
data—only the resulting effect of these transactions is reflected.
Changes in creditor and debtor status that can result from debt restructuring are also
reflected. For example, when insured commercial credits are rescheduled, the creditor
classification shifts from private sources to official sources (bilateral). This reflects
the assumption of the assets by the official credit insurance agencies of the creditor
countries. The debts to the original creditors are reduced by the amounts rescheduled, and
a new obligation to the official creditor agencies is created. This shift also applies to
private nonguaranteed debt that is reduced by the amounts rescheduled, which in turn are
included in the public and publicly guaranteed debt owed to official creditors. On the
debtor side, when a government accepts responsibility for the payment of rescheduled debt
previously owed by private enterprises, the DRS registers a change in debtor categories in
the DRS. Similarly, when short-term debt is included in a restructuring agreement, the
rescheduled amount is shifted from short-term to long-term debt.
Back to top
Methodology
for projecting data
An important feature of the RXD system of the DRS is its ability to project future
disbursements of unutilized commitments and future debt service payments.
Undisbursed debt
Projections of disbursements help underpin future capital
requirements in the implementation of externally financed projects. In addition, they help
determine the interest portion of projected debt service. Future interest payments are
based on projected debt outstanding that is itself determined by projected disbursements
and repayments. The underlying assumptions of these projections are that loan commitments
will be fully utilized and that the debtor country will repay all sums due. Future
disbursements and debt service refer only to existing debt and do not reflect any
assumptions on future borrowing.
Disbursement projections use two methods:
- Specific schedules. Debtor countries are requested to submit a calendar of future
disbursements, if available, at the time individual loans are first reported. Country
authorities are in a better position to provide estimated disbursement schedules when
there is a solid public sector investment program in place.
- Standard schedules. In the absence of specific schedules, the RXD system projects
disbursements by applying a set of profiles to the last actual undisbursed balance of
individual loans. The profiles are derived under the assumption that specific sources of
funds have some common characteristics that cause them to disburse, in the aggregate, in
some observable pattern. Accordingly, some thirty profiles have been derived that roughly
correspond to creditor type. Profiles exist for concessional and nonconcessional loans
from official creditors. For bilateral lending, profiles have been developed for the
Development Assistance Committee, the Organization of Petroleum-Exporting Countries
(OPEC), and other creditor groupings. For multilateral lending, specific profiles are
available for major international organizations. An estimating equation for each profile
is derived by applying regression analysis techniques to a body of data that contains
actual disbursement information for more than 100,000 loans. Although these standard
profiles are reestimated from time to time, under the best scenario they can only
approximate the disbursement pattern of any single loan.
Future debt service
payments
Most projections of future debt service payments generated by the
RXD system are based on the repayment terms of the loans. Principal repayments
(amortization) are based on the amount of loan commitments, and the amortization profile
of most loans follows a set pattern. Using the first and final payment dates and the
frequency of the payments, the system calculates the stream of principal payments due. If
future payments are irregular, the RXD system requires a schedule.
Projected future interest payments are calculated similarly.
Interest is based on the amount of debt disbursed and outstanding at the beginning of the
period. Again, using the first and final interest payment dates and the frequency of
payments, the system calculates the stream of interest payments due. If interest payments
are irregular, the RXD system requires a schedule.
The published figures for projected debt service obligations are
converted into U.S. dollars using the end-December 1996 exchange rates. Likewise the
projection routine for variable interest rate debt, such as commercial bank debt based on
the London interbank offer rate (LIBOR), assumes that the rate prevailing at the end of
December 1996 will be effective throughout.
On definitions
On debt indicators
Back to Research Methods
On population projection On
social indicators On economic timeseries
On Classification of economies
|
