Debt records are normally kept on a loan-by-loan basis. To prepare statistics, these records must be aggregated by various criteria, such as short-term and long-term, type of borrower or type of creditor.
A myriad of details in external debt accounting records must be organized by classification criteria in order to create summary statistics suitable for analysis.
Two fundamental distinctions are those between (i) domestic and external debt, and (ii) short-term and long-term debt. Data on long-term public and publicly guaranteed debt is normally available to the statistician on a loan-by-loan basis, thus, it can be arranged according to any number of criteria. Because short-term debt can be obtained on a highly aggregated basis only, the possibilities of classification are more limited.
Other possible ways of aggregating debt data are by type of borrower or by type of creditor. Among borrowers, the direct debt of the central government should be shown separately. Within private sector debt, it is useful to identify separately debt that is guaranteed by government, as such debts are in effect contingent budgetary liabilities. (Guaranteed debt of public enterprises and of other fiscally-autonomous agencies must also be identified.) Creditors can first be broken down into official and private lenders. Official creditors, in turn, can be multilateral or bilateral lenders. Private lenders are commonly grouped into bond holders, commercial banks, suppliers and other. A cross-creditor scheme often used is export credits. Private export credits consist of suppliers credits (credits extended by producers or contractors to finance their products) and buyers credits (loans by commercial banks to finance an export order by their customers). There are also official export credits: loans by specialized agencies like the U.S. Export-Import Bank to finance individual purchases of goods. These may be shown as one category of official flows or combined with private export credits to give a total picture of this type of finance.

Furthermore, data on flows like loan disbursements, debt service payments, debt restructuring and the accumulation of arrears should be shown for each category separately.

|