The IMF's Statistical Systems in Context of Revision of the United Nations' A System of National Accounts
Chapter

18 Overall Relationships Between the IMF's Government Finance Statistics and A System of National Accounts

Author(s):
Vicente Galbis
Published Date:
September 1991
Share
  • ShareShare
Show Summary Details
Author(s)
Jonathan Levin

Although the detailed relationships between items in the United Nations' A System of National Accounts (SNA) and the IMF's government finance statistics (GFS) have been worked out over the past decade in the bridge tables appearing in A Manual on Government Finance Statistics (GFSM) and the UN Handbook on public sector statistics1 and are explored further in Chapter 19 of this volume, it may be helpful to explore briefly the larger relationships between the two data systems and the relative significance of particular differences for their reconciliation. A useful starting point for this purpose may be the origins and purposes of the GFS system.

In both respects, the GFS system is designed to measure and forecast behavior on the basis of observed relationships of motivation and effect. The classifications of government transactions in the GFS system have sought to reflect current thinking in the field of public finance as a means of fitting the analysis and planning of government policies into overall national programs for stabilization and growth. Thousands of Fund country missions have concluded that the strategic role and motivational structure of government require an analytical and planning framework of government operations different from that for any other sector. This framework, over the years, has become the GFS system.

In essence, this framework has called for comprehensive coverage of a government's transactions with the rest of the economy and the world, measuring primarily government payments and receipts and, secondarily, any unpaid accruing obligations. Because it focuses on past and expected future behavior, GFS is based on behavioral or motivational, rather than institutional, relationships. For the sake of such analytical, and predictive, relationships, GFS departs from institutional “integrity” when all or parts of an institution carry out economic functions that differ from its legal or institutional classification. To focus on analysis of government behavior, moreover, GFS classification of some transactions may have to differ from the perceptions of some transactors with whom the government deals.

Within its overall requirements of comprehensive coverage and the gross measurement in real time of government payments and receipts, the heart of the GFS system is the classification of all government transactions from the government's point of view by the nature of what flows. Because government accounting systems are unable to keep track of all flows of accruing liabilities and resources to and from government, these systems are based on money flows. The GFS system utilizes government accounting data, rather than estimates or imputations, of money flows, classified by the nature of what flows in the opposite direction, to construct its analytical framework of government operations.

This GFS analytical framework is represented graphically in Figure 1, which divides all flows between payments and receipts, with an undivided balancing cell at the bottom for changes in cash balances. Flows of receipts and payments are divided horizontally by what flows in the opposite direction: goods and services or resources, nothing, others' liabilities to government, and government liabilities to others. Money flows in exchange for goods and services (or nothing) are further divided by whether or not they involve capital goods—either in direct purchase or sale or in transfers for their acquisition by the recipient. Flows in exchange for others' liabilities to government are divided between those undertaken for public policy purposes—that is, government lending or equity acquisitions undertaken for the same types of goals as government spending—and those undertaken for the management of government liquidity. A further horizontal division among government receipts for which nothing flows in exchange separates grants from other governments or international organizations from flows of taxes or other transfers from nongovernmental sources.

Figure 1.Analytical Framework of the IMF's Government Finance Statistics (GFS)

Figure 2 illustrates the classification of various government transactions within this analytical framework. Figure 3 identifies the principal aggregates of the GFS system fitted into this analytical framework, with shading for each. As may be seen from Figure 3, Revenue plus Grants, less Expenditure and Lending Minus Repayments, equals the Deficit/Surplus, which, in turn, is equal to Financing.

Figure 2.Location of Various Transactions in GFS Analytical Framework

Figure 3.Identification of Major GFS Aggregates in GFS Analytical Framework

To provide an approximate comparison between the GFS and the SNA, a number of SNA components are superimposed on this GFS analytical framework in Figure 4. An approximation of the SNA accounts, with shading for each account, is superimposed on the GFS framework in Figure 5, with notation also of the balancing items-operating surplus, saving, and net lending—carried from each account to the next. Examination of Figures 4 and 5 permits identification of several components that are classified differently in the GFS and SNA and that contribute to differences between major aggregates in the two systems. One such example is estate and gift taxes, which are classified as capital transfers in the SNA and as taxes, and hence current, in GFS, resulting in differences between the two systems' concepts of taxes, current receipts, capital receipts, and saving.

Figure 4.Components of SNA Accounts in GFS Analytical Framework

Figure 5.SNA Accounts and Balancing Items in GFS Analytical Framework

Overall comparison of the GFS and SNA in the GFS analytical framework serves also to point out similarities and differences between the two systems. The similarities are quite strong in the identical classification of most, though not all, components by the nature of the flows involved. Organization of the two systems differs, clearly, in the single, comprehensive account used by the GFS in contrast to the several accounts used in the SNA to measure separately production, income and outlay, and capital accumulation and finance. The single account structure of the GFS calls for each flow to appear only once, whereas the multiple account structure of the SNA requires the repetition of some components and the insertion of balancing items along the way. Restriction of the nonfinancial major GFS aggregates to either the receipt or payments side contrasts with the combination within some SNA components to include both receipts and payments in a single net concept.

Whatever conclusions are reached about the feasibility of closer future relations between the two data systems, it is evident from this examination of their principal features that they have developed different formats and organizational structures to serve their different objectives: the GFS for overall evaluation of fiscal performance and its impact on financial conditions, and the SNA for measurement of government production, income and outlay, capital formation, and finance as a component with other sectors in the calculation of national aggregates. The primary parameter in any prescription for possible future changes must not divert either system from its primary purpose but should strengthen each as a result of its close relation to the other.

United Nations, Handbook of National Accounting: Public Sector Accounts, Studies in Methods, Series F, No. 50 (New York, 1988).

    Other Resources Citing This Publication