Printer Friendly

Replacing GNP: the updated system of national economic accounts.

GNP has been replaced by GDP as the primary measure of U.S. production. The rationale for this shift is provided, as well as an explanation of what it entails. Beginning with the Economic Statistics Initiative, this paper also discusses the changes ahead for the U.S. economic accounts. Focusing on the National Income and Product Accounts, the United Nations System of National Accounts is identified and compared to the present set of U.S. accounts, highlighting some changes it suggests. The paper concludes with an indication of the plans to modernize and extend the U.S. economic accounts.

GNP REPLACED? Can it really be that the Bureau of Economic Analysis (BEA) has replaced its focus on gross national product (GNP) with a focus on gross domestic product (GDP) as the primary measure of U.S. production?(1) Yes, it's true, and NABE members have probably picked up other, more subtle, signals of change ahead in the U.S. economic accounts developed by BEA:

1. The Administration's FY 1992 Economic Statistics Initiative supported a modernized and extended set of economic accounts that would follow the United Nations revised System of National Accounts.

2. The January 1991 issue of Business Economics carried an article discussing something called "satellite accounts" that are to be part of a modernized and extended set of U.S. economic accounts.

Where is all this leading? It is useful to start with the Economic Statistics Initiative. Understanding the serious problems facing the U.S. statistical system in keeping up with the rapid pace of change in today's economy, a Working Group on improving the quality of economic statistics was established in April 1989, chaired by Michael J. Boskin, Chairman of the Council of Economic Advisors. The Working Group developed a package of high priority recommendations for improving economic statistics, which the President approved on November 25, 1989. To implement these recommendations, the Administration's FY 1992 budget proposed a multi-year Economic Statistics Initiative.(2) The Initiative's goal of improving economic statistics provides the broad context from which I will focus my remarks.

One aspect of this improvement for BEA is the creation of a modernized and extended set of economic accounts. Although BEA's international accounts -- balance of payments and underlying investment data -- are also involved in this project, I will focus today on BEA's national accounts, including the national income and product accounts (NIPAs). With respect to the national accounts, the Initiative said:

National economic accounts that follow the United Nations revised System of National Accounts (SNA) would be introduced in the comprehensive revision . . . in the mid-1990s. The principal new features will include (1) an integrated set of current and capital accounts . . . , and (2) satellite accounts . . .

These statements frame my topic. In discussing it, I will identify the SNA, discuss its integrated current and capital accounts, highlight some changes they suggest for BEA's NIPAs, and indicate BEA's plans. I will omit a discussion of satellite accounts and instead refer you to an earlier issue of Business Economics.(3) However, I would like to mention that BEA's work on a research and development satellite account is moving forward, and the FY 1992 budget contains funding for us to begin work on a second satellite account dealing with natural resources.


The SNA is a set of international guidelines for economic accounting. It is "international" in that it originated with the United Nations, and it serves as a guide for the economic accounts of many countries. Of the major industrial countries, the United States alone does not follow the SNA. Further, Central and Eastern European countries are abandoning their Marxist-based Material Product System and adopting the SNA, or at least doing SNA accounting in parallel.

But while the United States may be the odd ball now, historically, the SNA and the NIPAs had much in common. Some of the people that helped shape the NIPAs were instrumental after World War II in developing the United Nations SNA. The first SNA, in 1953, was similar to the five-account summary of the NIPAs, which is the framework within which BEA still works.

However a major extension of the SNA in 1968 encompassed input-output, flow of funds, and, in principle, balance sheets. What emerged as the 1968 SNA was intended to be a comprehensive, integrated framework for recording the stocks and flows of the economy. The SNA is now undergoing yet another revision,(5) and thus the phrase "revised System of National Accounts" in the Statistics Initiative. This revision of the SNA has two goals. First, it is to bring the SNA up to date, incorporating developments in economic accounting and changes in the institutions within countries around the world since the 1968 revision. Second, it is to harmonize the SNA with other international statistical guidelines, such as the IMF's Balance of Payments Manual, which are also being revised.(6)

In the forty years since the SNA and the NIPAs shared a common origin, the SNA has been revised twice while the NIPAs are just now facing their first major revision.


Broadly speaking, the SNA is more comprehensive and more integrated than are the U.S. accounts. This characterization requires some explanation.

"Comprehensive" refers to a full recording of the stocks and the flows in the economy. For each sector that makes up the economy, you have an opening balance sheet and a closing balance sheet -- the stocks of the economy. And for each sector, you have accounts to record all the transactions and other changes that "explain" the differences between the balance sheets -- the flows of the economy. Structurally, this full explanation begins with a set of current accounts that show production, income earned in production, redistribution of income, consumption, and saving, similar to what we now have in the NIPAs. Then you move to a set of what the SNA calls accumulation accounts. They include capital and financial accounts that show saving that is invested, including the intermediation that redistributes saving from sectors with a surplus to sectors that need to finance investments; a revaluation account to record both general changes and relative changes in the price level for the various assets; and finally an account to record other changes in assets, including natural disasters and depletion.

Even at the broadest level, the United States does not have a full set of accounts. Specifically, we do not have revaluation accounts nor balance sheets for all sectors. The Federal Reserve Board (FRB) prepares these accounts, but does not have them for all of the sectors.

"Integration" refers to internal consistency across all the accounts, including balance sheets. Consistency covers several dimensions: definitions of sectors, definitions of transactions, time of recording, valuation, etc.

The sets of U.S. accounts are not fully integrated. There is not complete consistency in the dimensions just mentioned. To illustrat: the difficulties of reconciling saving as it appears in BEA's NIPAs and saving as it is in the FRB's flow of funds are in part due to differing definitions used in the two accounts.

Our goal in moving toward the SNA is a comprehensive and integrated set of national accounts for all sectors in the economy covering what we now have as the national income and product accounts, input-output accounts, flow of funds accounts, and balance sheets for some sectors of the economy.

As we move from a broad view of the two systems toward more specific, operational differences, we return to replacing GNP. The SNA features GDP as the measure of production in an economy. The accompanying box mentions the increased international comparability that results from using the SNA featured measure as one of the reasons for replacing GNP, and explains the main differences between GNP and GDP.

However, just replacing GNP with GDP doesn't go the full distance to the SNA. The SNA-basis GDP does not equal NIPA-basis GDP because the classifications underlying the SNA differ somewhat from NIPA classifications. Economic accounts are, when you get right down to it, systems of classifying, or grouping, economic transactors and transactions (and resulting assets) in meaningful ways. The classification of transactions separates them into, for example, consumption and investment, or intermediate consumption and final consumption.

To illustrate some of the differences between the SNA and the NIPAs, the SNA classifies government purchases of nonmilitary(7) durable goods and structures as investment and counts services provided by that capital as consumption by government. The NIPAs treat all government purchases of durable goods and structures as consumption; by NIPA convention, governments do not invest.(8) It should be apparent that this basic classification differences has far-reaching effects throughout the accounts. Adopting the SNA treatment of government purchases is likely to be a feature of the modernized U.S. accounts. Other classification differences at present include the treatment of imputed charges for bank services and the treatment of pensions of Federal Government employees. The basic point to note is that underlying classifications of transactions will also be examined in updating the U.S. accounts.

So far, I have been concerned with differences between the SNA and the NIPAs. Differences should not be overemphasized, however. For instance, if you compare SNA-basis GDP and NIPA-basis GDP, quantitatively the difference is not great. SNA-basis GDP is about 0.6 percent less than NIPA-basis GDP in 1990.

Users of BEA's estimates may well ask: If I open the quarterly news release from BEA in five years or so, will I be lost if the United States is following the SNA? Let's use Canada as a case study. Canada over the years has built a set of accounts that, as Statistics Canada describes it, "in broad outline . . . has a close relationship to the SNA." A Canadian news release leads off with GDP (Canada switched from GNP in 1986), and it features components of demand, as in a BEA release. However, some obvious differences appear in the tables: what we know as government purchases is divided into current expenditure, government investment in fixed capital, and government investment in inventories. As well, there are some not so obvious differences: exports and imports do not include factor income from abroad. But from the Canadian experience we can see that changes in the way the accounts are constructed will not necessarily change the way the accounts will be presented. I expect that users will not feel at all lost.


The advantages of the SNA can be grouped under three broad headings: analytical power, international comparability, and system organization. First, the SNA provides the enhanced analytical power of a comprehensive, integrated framework.(9) Particularly for analyses based on a mix of measures (of production, income, saving and investment; of domestic and foreign activities; of financial and non-financial transactions; of stocks and flows) comprehensiveness and integration of estimates from a system such as the SNA would seem to be a strong advantage.

Two examples will illustrate the analytical power of such a framework. The first is about investment. To date, most studies and models of fixed nonresidential investment have concentrated on short-term effects of the demand for investment, typically based on neoclassical theory. With a set of accounts based on the revised SNA, the effects of financial flows and balance sheets can be investigated in a conceptually and statistically integrated framework. The information will not only include that for non-financial corporations, but also the balance sheets of financial intermediaries whose lending is likely to have both cyclical and long-term effects on business. In addition, the integrated rest-of-the-world accounts should permit the tracing of international capital flows that support domestic investment.

The second example is about the measurement of natural disasters, such as the Loma Prieta earthquake and Hurricane Hugo in 1989. Typically, the Nation's production, and thus GDP as a measure of production, does not register substantial direct effects from natural disasters. Capital consumption allowances are increased by the amount of damage to private capital; rental income of persons, proprietors' income, and corporate profits are affected by lost income and the effects of insurance payments. But the revised SNA registers additional effects more directly. It registers capital losses, including losses to government capital (such as the highways, bridges, etc., affected by the earthquake), and provides a sharper picture of the intersector income flows (such as insurance payments) resulting from the disaster. The capital losses are carried through to affect the balance sheets.

The second advantage is the increased international comparability of a system closer to the SNA. International comparability is important because policy is increasingly formulated in an international setting. This advantage is realized, of course, because most other countries follow the SNA. Take, for example, the saving rate. Several studies have shown that a big part of the difference in the saving rates of the United States and Japan stem from differences in definition. These, in turn, can be traced in large part to the fact that Japan, following the SNA, treats government purchases of durable goods and structures as investment.

The third advantage is the strengthened organizing, or integrating, framework for the U.S. statistical system provided by a comprehensive set of accounts such as the SNA. Such a set of accounts would provide a framework within which cost and benefits of source data improvement could be weighed. Several other countries, most notably Canada, explicitly use their accounts in this way.


BEA's target for introduction of modernized accounts along the lines of the SNA is the mid-1990s. The introduction will probably be tied to a comprehensive, i.e., benchmark, revision when BEA traditionally introduces changes in definitions, classifications, and table presentations. However, we will introduce some features as we go along, for example, as supplementary tables.

At this point, we have made a start, but have a long way to go. As already noted, beginning with this benchmark revision, BEA is focusing on GDP as the primary measure of production. One of the criteria applied in deciding on changes in definitions and classifications for the revision was that the changes should not move counter to the SNA. In addition, there is now a small staff evaluating the SNA accounting structure and definitions. This same staff is developing BEA's first satellite account -- on R & D expenditures.

Of course, funding will determine how fast and far we can move. Our general strategy is along the following lines. We are asking, "Is the SNA approach really an improvement for the U.S. accounts?" Our presumption is that the SNA, as the system followed by other major industrial nations, has much to offer. In particular, the ongoing revision represents a major investment in intellectual capital, seeking to update the SNA and increase its usefulness by harmonizing it with other international statistical guidelines. However, no set of international guidelines such as the SNA can exactly fit a particular country's economy or statistical system. Earlier I cited Statistics Canada stating their set of accounts "in broad outline . . . has a close relationship to the SNA." The United Kingdom and Australia use similar wording.

To implement this strategy, BEA is mounting a substantial evaluation effort. Earlier we had a consultant with an SNA background raise our consciousness, so to speak, and help us develop an inventory of issues that are likely to come up in moving toward the SNA. Now we are asking, "Why do we do it this way? How can we do it better?" This effort builds upon the substantial work begun in 1990 to upgrade the existing procedures for a range of NIPA estimates, including construction, GDP by industry, and services. Further, it will benefit from closer cooperation with the Federal Reserve Board -- cooperation manifested at present by joint exploration of estimates for a nonprofit sector. Finally, it must be stressed that BEA intends to involve our users in the evaluation effort. One way may be by the establishment of an advisory committee, and we are exploring that possibility.


(1)See "Gross Domestic Product as a Measure of U.S. Production," Survey of Current Business, vol. 71, no. 8 (August 1991), p. 8.

(2)The Working Group's recommendations were summarized in the Survey, vol. 70, no. 2 (February 1990), p. 2. The Economic Statistics Initiative was summarized in the Survey, vol. 71, no. 3 (March 1991), pp. 4-5. In addition, the NABE Statitics Committee provides updates on the Statistics Initiative's progress in the "NABE News."

(3)See Carol S. Carson and Bruce T. Grimm, "Satellite Accounts in a Modernized and Extended System of Economic Accounts," Business Economics, vol. xxvi, no. 1, January 1991, pp. 58-63.

(4)For an overview of the SNA, see Carol S. Carson and Jeanette Honsa, "The United Nations System of National Accounts: An Introduction," Survey, vol. 70, no. 6 (June 1990), pp. 20-30. Reprints are available from Bureau of Economic Analysis, Public Information Office, 1401 K Street, NW, Washington, DC, 20230.

(5)Although the current revision of the SNA is not expected to be formally adopted by the United Nations Statistical Commission until February 1993, its main features are already clear.

(6)In revising the various international guidelines, the concern underlying the goal of harmonization among the systems is both to make them analytically more powerful when taken together and to conserve the resources within individual countries needed to put them together.

(7)The revised SNA will also treat some military purchases as investment.

(8)Although it should be noted that government purchases of durable goods and structures are included in BEA's capital stock figures.

(9)It should be mentioned that the idea of a comprehensive integrated system is not a new idea. Several individuals have outlined such systems, and it has been discussed within the Federal Government: The National Accounts Review Committee (set up by the National Bureau of Economic Research at the request of the predecessor of the Office of Management and Budget) in the mid-1950s recommended integration of various parts of the economic accounts.
COPYRIGHT 1992 The National Association for Business Economists
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1992 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Carson, Carol S.
Publication:Business Economics
Date:Jul 1, 1992
Previous Article:Investment, valuation, and dynamics: a case study in the law of supply and demand.
Next Article:International perspectives: the United States and Latin America: toward a theory of interdependence and full exchange.

Terms of use | Copyright © 2016 Farlex, Inc. | Feedback | For webmasters