Printer Friendly

The U.S. national income and product accounts: revised estimates.

The U.S. National Income and Product Accounts: Revised Estimates

ESTIMATES of the national income and product accounts (NIPA's) for the years 1984-86 and the quarters from the first quarter of 1984 through the first quarter of 1987 have been revised. As is usual in July, source data that are more complete, more detailed, or otherwise more appropriate than data previously available have been incorporated, and seasonal factors have been updated. As well, several methodological changes have been made.

The revised estimates are shown in the complete set of NIPA tables that follow this article. An index to the NIPA tables, presented for the first time in this issue, begins on page 87. Also in this issue is an article that summarizes the source data and estimating methods used to prepare GNP.

On the whole, the July 1987 NIPA revisions did not significantly alter the picture of the U.S. economy. The growth in real GNP was stronger on the revised basis, although the quarterly pattern of change remained much the same (chart 4). The increase in GNP prices was about the same on the revised basis; the quarterly pattern of increase also remained much the same (chart 5). Following a brief discussion of the impact of the revisions on several aspects of economic activity, the first section of this article provides a summary of the revisions and the major source data underlying them, and the second section provides notes on the methodological changes that were incorporated.

Economic growth and inflation.-- The revision period covered a major portion of the current expansion, which began following the trough in real GNP in the third quarter of 1982. Over the period from the fourth quarter of 1983 through the first quarter of 1987, the increase in real GNP was revised up $32.8 billion; the average annual rate of increase in real GNP was revised up 0.3 percentage point-- from 3.3 percent to 3.6 percent (table 1). (Over the period from the GNP trough through the first quarter of 1987, the average annual rate of increase was revised up 0.2 percentage point--from 3.9 percent to 4.1 percent.)

For most of the major components of real GNP, the average annual rates of change over the revision period were not greatly altered. On the revised basis, consumption was somewhat stronger (increasing 3.8 percent instead of 3.3 percent) and investment was somewhat weaker (increasing 4.8 percent instead of 5.4 percent).

Inflation, as measured by GNP prices, was unrevised over the period. The average annual rate of increase in the GNP price index (fixed weights) was 3.3 percent by either set of estimates (table 2). Price changes for the components of GNP were about the same as previously published; residential investment prices were up slightly less, and import prices were down slightly less.

Income, consumption, and saving.-- The revised estimates present a somewhat different picture of the relationships of personal income, consumption, and saving. Compared with the previously published estimates, the revised estimates of personal income, disposable personal income (DPI), and personal consumption expenditures (PCE) were stronger in 1985, and, especially for income, in 1986. The annual percent increase in real DPI was revised up from 2.3 percent to 2.9 percent in 1985 and from 2.9 to 4.0 percent in 1986, and the increase in real PCE was revised up from 3.5 to 4.6 percent in 1985 and from 4.1 to 4.2 percent in 1986. Thus, in 1986, the rates of increase in income and consumption were about the same on the revised basis; as previously estimated, income had increased considerably less than consumption.

The pattern of personal saving in 1985 and 1986 differed in the revised estimates. Compared with the previously published estimates, the revised estimates of personal saving fell more sharply in 1985 and then leveled off in 1986 rather than continuing to drop. The revised personal saving rate was 4.5 percent in 1985 and 4.3 percent in 1986, rather than 5.1 and 3.8 percent, respectively.

By either set of estimates, real DPI and real PCE were weak in the second half of 1986 and the beginning of 1987. As well, by either set of estimates, the personal saving rate was low during that period.

Summary of the NIPA Revisions

The incorporation of new source data and changes in methodology lead to revisions in the current-dollar estimates. In general, revisions in these estimates, as well as revisions in the price estimates, affect the constant-dollar estimates.

Annual revisions in current dollars

The revised and previously published annual estimates for current-dollar GNP, charges against GNP, and their major component series, which comprise the national income and product account, are shown in table 3. The revisions on the product side of the account primarily stemmed from the incorporation of new annual data for 1985 from four Census Bureau surveys--Annual Retail Trade Survey, Service Annual Survey, Annual Survey of Manufactures, and surveys of State and local government finances. The revisions on the income side of the account stemmed from a variety of source data, including such major ones as the Internal Revenue Service (IRS) tabulations of corporate tax return date for 1984 and of sole proprietor and partnership tax return data for 1985, Bureau of Labor Statistics (BLS) tabulations of State unemployment insurance wage data for 1986, U.S. Department of Agriculture farm data for 1986, and Federal agency reports on assets and interest of financial institutions for 1986.

On the product side of the national income and product account, the level of GNP was revised up $7.2 billion, or 0.2 percent, in 1984; up $12.2 billion, or 0.3 percent, in 1985; and up $28.9 billion, or 0.7 percent, in 1986. The revisions in the year-to-year change in GNP--$7.2 billion in 1984, $5.0 billion in 1985, and $16.7 billion in 1986-- were comparable in size (without regard to sign) to those in recent July revisions. However, because each year-to-year revision was in the same direction, they cumulated to a somewhat larger-than-usual revision in the level of GNP in the most recent year--1986. The upward revisions in GNP were largely traceable to PCE-- both goods and services. Downward revisions in nonresidential fixed investment were a partial offset.

On the income side of the account, the level of charges against GNP was revised down $0.1 billion in 1984, up $12.3 billion in 1985, and up $36.8 billion in 1986. The upward revisions in charges against GNP were largely traceable to wages and salaries, farm proprietors' income, and net interest. Downward revisions in supplements to wages and salaries and in corporate profits were a partial offset.

Largely reflecting revisions in some of the components of charges against GNP, personal income was revised down $1.5 billion in 1984, up $12.5 billion in 1985, and up $48.6 billion in 1986. Revisions in DPI were in the same direction as, and slightly larger than, those in personal income.

Table 4 provides a guide to the July NIPA revisions by identifying the subcomponent series in which the revisions were concentrated and by listing the major source data that underlie the revised estimates.1 Using the format in table 4, the following sections discuss revisions in NIPA components, focusing on the revisions that are notable either because of their size or content or because they involve methodological changes.

1. The years listed with the sources in table 4 are the ones for which new or revised data became available. In addition to revisions in those years, subsequent years were affected. For example, the incorporation of data from the 1985 Annual Retail Trade Survey, which raised the level of PCE goods in 1985, carried through to raise the level of PCE goods in 1986 as well.

PCE goods.--PCE goods was revised up substantially in 1985-86, partly reflecting the incorporation of data from the 1985 Annual Retail Trade Survey and partly reflecting revised estimates of used auto purchases. The survey data raised spending estimates for nearly all categories of durable and nondurable goods.

The revisions in used autos, which, on balance, raised PCE and lowered producers' durable equipment (PDE), stemmed from several sources. A correction in the valuation of 1983 model year cars raised the estimated value of used autos sold by business to consumers in subsequent years. Also, newly available data on fleet ownership from trade sources indicated that a higher volume of used cars had been sold by business and purchased by consumers. In addition, an adjustment for quality change in the valuation of used auto purchases in current dollars was introduced. The effect of this adjustment, which lowered PCE and raised PDE, was to partly offset the other revisions. (A description of this methodological change, which affected current-dollar and price estimates but not constant-dollar estimates, is in the next section of this article.)

PCE for gasoline and oil was revised down in 1986, largely reflecting the incorporation of data on motor gasoline supply from the Energy Information Administration. Beginning with the first quarter of 1987, these data have replaced Census Bureau retail sales data in the quarterly procedure for estimating PCE for gasoline and oil.

PCE services.--The upward revision in PCE services in 1985-86 largely reflected the incorporation of newly available data from a number of regular data sources, particularly the service annual survey. The upward revisions in services were concentrated in recreation, medical care, personal business, and personal care. (See, in addition to the article on sources and methods in this issue, a description of the estimating procedures for PCE services in "Notes on Sources of the Revisions' in the July 1986 SURVEY article on the revised estimates.)

The downward revisions in PCE for housing services in 1985-86 were largely due to the incorporation of information from the current population survey on the distribution between owner- and tenant-occupied housing; a higher portion of tenant-occupied units--which have a lower average rental--led to reduced estimates of rent. (Data from the 1985 American Housing Survey, which would normally have been incorporated in the estimates of PCE for housing services, were not yet available.)

The revised estimates of PCE for recreational services in 1984-86 included first-time estimates of expenditures on video cassette rentals. (A description of this change is in the next section of this article.)

Nonresidential structures.--The revisions in structures largely reflected the incorporation of newly available source data for public utilities for 1984-86. The remainder of the downward revision in structures in 1986 was largely accounted for by commercial buildings.

Nonresidential PDE.--The substantial downward revisions in PDE in 1985-86 largely reflected the incorporation of newly available source data, particularly shipments data from the 1985 Annual Survey of Manufactures. The survey data lowered estimates for most PDE categories; an exception was office, computing, and accounting machinery. The revisions in used auto purchases, described under PCE goods, also contributed to the downward revisions in PDE. In addition, beginning with the first quarter of 1986, the quarterly commodity-flow procedure for estimating PDE uses more detail on exports and imports.

Residential investment.--Revisions in residential investment were concentrated in the category nonfarm additions and alterations, and major replacements. The revisions, which reflected newly available data on expenditures, primarily affected additions and alterations.

Change in business inventories.-- Revisions in business inventories reflected newly available information from several Census Bureau surveys and from other sources. The largest revisions were downward in manufacturing inventories in 1985, upward in retail trade inventories in 1985-86, and upward in other inventories in all 3 years.

Net exports.--In general, the revisions in net exports reflected the revisions to the BEA balance of payments accounts released in June 1987. The upward revision in merchandise exports in 1986 was largely due to the incorporation of information from the annual reconciliation of merchandise trade data for the United States and Canada. In the reconciliation process, the two countries reach mutually agreed estimates of their bilateral trade, primarily by using each country's import statistics. In the revised estimates, exports of industrial supplies and materials, capital goods (except autos), and autos were higher in 1986 than previously estimated. Exports and imports of services other than factor income were revised up in 1986. These revisions largely reflected changes made in June 1987 that improved coverage of commissions on securities transactions in exports and in imports and improved coverage of medical services in exports. (See the next section of this article for a description of these improvements.)

Government purchases.--The upward revisions in government purchases largely reflected information from the Census Bureau surveys of State and local government expenditures for fiscal year 1985 and preliminary information on State government expenditures for fiscal year 1986. The information raised estimates of purchases other than employee compensation and structures for all 3 years.

Compensation of employees.--Private wages and salaries were revised up, largely due to the incorporation of BLS tabulations of State unemployment insurance wage data. The revision was concentrated in wages and salaries in the distributive and service industries. In supplements to wages and salaries, other labor income was revised down due to lower estimates of business contributions for pension and profit-sharing plans and for group health insurance.

Proprietors' income.--Farm proprietors' income was revised up sharply in 1986. The revision reflected the incorporation of newly available data from the U.S. Department of Agriculture (USDA) survey on farm costs and returns, which led to lower estimates of farm expenses, and of revised USDA farm output data.

Rental income of persons.--Rental income was revised up, largely due to information that lowered estimated expenses for maintenance, repairs, and other costs. The downward revision in rent that was mentioned under PCE services was a partial offset.

Corporate profits with inventory valuation adjustment and capital consumption adjustment (CCAdj).--About one-third of the downward revision in corporate profits in 1986 was traceable to profits before tax and two-thirds to the CCAdj--the difference between IRS tax-return-based depreciation and NIPA economic depreciation. The downward revision in profits before tax was concentrated in manufacturing, financial, and rest-of-the-world profits. See capital consumption allowances with CCAdj for a discussion of the revisions in the CCAdj.

Net interest.--As has been the case in recent July revisions, revisions in interest were large in the most recent year--1986. The schedule of availability of the source data makes these estimates particularly subject to revision; in particular, information on interest paid and received by business is only available on an annual basis. (See, in addition to the article on sources and methods in this issue, a description of the estimating procedures for net interest--and personal interest income--in "Notes on Sources of the Revisions' in the July 1986 SURVEY article on the revised estimates.)

National income.--National income was revised down in 1984, up in 1985, and up substantially in 1986. These revisions reflected the aforementioned revisions in compensation of employees, proprietors' income, rental income, corporate profits, and net interest.

Capital consumption allowances with CCAdj.--Capital consumption allowances (IRS tax-return-based depreciation) was revised down for all 3 years, and capital consumption allowances with CCAdj (NIPA economic depreciation) was revised up for all 3 years. Thus, the CCAdj--the difference between the two--was revised down for all 3 years. The bulk of the revisions in the depreciation measures was in the corporate estimates. The downward revision in corporate capital consumption allowances in 1984 mainly reflected the incorporation of data on depreciation from tax return tabulations. The downward revisions in 1985-86 largely reflected revised BEA projections of tax-return-based depreciation from the lower 1984 estimates. The small upward revisions in corporate capital consumption allowances with CCAdj largely reflected a higher allocation of new investment to corporate business.

Nonfactor charges.--The upward revisions in nonfactor charges--business transfer payments, indirect business tax and nontax liability, and current surplus of government enterprises less subsidies--in 1984-85 were due to all three components. In 1986, revisions in these components were offsetting.

Personal income.--The upward revisions in personal income and DPI in 1985-86 largely reflected the revisions in wages and salaries, farm proprietors' income, and rental income. In addition, the revisions in the interest estimates led to a large upward revision in personal interest income. Transfer payments were also revised up; newly available data led to higher estimates of transfers to persons from the Federal Government and from State and local governments.

Annual revisions in constant dollars

Revisions in real, or constant-dollar, GNP largely reflected the revisions in current-dollar GNP. The following tabulation provides a decomposition of the percent revisions in the level of real GNP into three parts: (1) The part that is due to the current-dollar revisions; (2) the part that is due to revisions in prices; and (3) the part that is due to other revisions, primarily revisions in current-dollar component detail that, in effect, reweight the constant-dollar component detail.

The remainder of the discussion of constant-dollar revisions, and that of price revisions, will focus on revisions in annual percent changes. In general, revisions in annual percent changes are very close to year-to-year differences in percent revisions in level. For example, the annual percent change in real GNP was revised up 0.4 percentage point in 1984, 0.3 point in 1985, and 0.4 point in 1986; the year-to-year difference in the percent revisions in the level of real GNP, as calculated from the above tabulation, were 0.4 percentage point in 1984, 0.2 point in 1985, and 0.4 point in 1986.

Revisions in constant dollars for the major components of real GNP were largely attributable to the current-dollar revisions already discussed. In particular, the annual percent increase in PCE was revised up 1.1 percentage points to 4.6 percent in 1985; the change in nonresidential fixed investment was revised down 2.5 percentage points to a 6.8-percent increase in 1985 and down 1.3 points to a 2.3-percent decline in 1986 (table 5). Revisions in annual percent changes for the remaining components, except for residential investment in 1985 and 1986, were less than 1.0 percentage point.

Annual revisions in prices

For the most part, revisions in prices were small and tended to be downward. The annual percent change in the GNP price index (fixed weights) was unrevised in 1984 and was revised down by 0.1 percentage point in 1985 and 1986. The revisions stemmed mainly from price estimates that come from sources other than the BLS Consumer Price Index (CPI) and Producer Price Index (PPI); the BLS indexes are used for components that account for well over three-fourths of GNP.

Among the major components of GNP, the revisions in residential investment prices reflected the incorporation of a modified price index for new houses from the Census Bureau. The modified index use the methodology of the Price Index for New One-Family Houses Sold, but is based on current rather than fixed weights. The upward revision in import prices in 1986 reflected newly available price information for travel, transportation other than travel and passenger fares, and military sales. In PCE prices, a newly available CPI component for new trucks was substituted for the CPI component for new cars, which had been used to deflate consumer purchases of trucks.

The revised estimates also incorporated improved price estimates for computers, which affected PCE, PDE, net exports, and government purchases. In addition, for the first time, prices of personal computers are included in the computer price estimates. (A description of this change is in the next section of this article.) On the revised basis, computer prices fell even more sharply in 1984 and 1985 and dropped somewhat less in 1986 than previously estimated.

Quarterly revisions

July revisions in quarterly NIPA estimates come about in two major ways: The adjustment of the quarterly estimates to reflect the annual revisions; and the bringing in of new and revised source data, including the incorporation of updated seasonal adjustment factors, that are used to indicate quarterly patterns.

This July, the quarterly revisions in GNP, real GNP, and GNP prices largely reflected the revisions in the annual estimates and did not sharply alter the quarter-to-quarter pattern of changes (table 6). For real GNP, the revisions in the quarterly percent changes at an annual rate averaged 0.6 percentage point (without regard to sign), somewhat less than in the three preceding July revisions. Three of the four largest upward revisions were in the first quarters of 1984, 1985, and 1986; these were quarters in which the previously published estimates had already shown relatively strong increases in GNP. For GNP prices, the revisions in the quarterly percent changes at an annual rate averaged 0.3 percentage point (without regard to sign), about in line with recent July revisions. The largest downward revisions in prices were in the fourth quarters of 1984, 1985, and 1986.

Notes on Changes in Methodology

Changes in methodology are made to incorporate new source data and improved estimating techniques. The following is a list of the major methodological changes that were made in this July revision: In PCE, an adjustment for quality change in net used auto purchases (affected current-dollar and price estimates only), first-time estimates of PCE for video cassette rentals, and use of different source data for quarterly estimates of gasoline and oil purchases; in PDE, use of more detailed exports and imports in the abbreviated commodity-flow method, and the adjustment for quality change in used auto purchases (affected current-dollar and price estimates only); and, in net exports, added coverage of commissions on securities transactions and of medical services. Several changes were also made in deflation procedures: In PCE, the substitution of the CPI for new trucks for the CPI for new cars in the deflation of consumer truck purchases; in residential investment, the use of a price index for single-family houses based on current rather than fixed weights; and, in PCE, PDE, net exports, and government purchases, the inclusion of personal computers in the estimates of computer prices.

These methodological changes have been mentioned earlier in the article in conjunction with revisions in the NIPA components. Several of them are described more fully in the following paragraphs.

Net purchases of used autos.--A new price series was used to prepare the revised estimates of current-dollar net purchases of used autos. The current-dollar estimates are derived by multiplying a direct estimate of contant-dollar net purchases by a price series, previously the CPI for used cars. In the revised estimates, a new, quality-adjusted price series was used to reflate the constant-dollar estimates. Unlike the CPI for new cars, the CPI for used cars does not include an adjustment for quality change. BEA estimated a quality adjustment for the CPI for used cars based on information on the quality adjustments that the BLS makes in the CPI for new cars. From December 1983 to December 1986, the CPI for used cars increased 0.5 percent, while the new series declined 6.5 percent.

The effect of the new procedure was to lower the value of used auto purchases in current dollars. Because consumers are net purchasers of used autos and businesses are net sellers, PCE was lowered and PDE was raised, by $0.3 billion in 1984, $1.0 billion in 1985, and $1.6 billion in 1986. The new procedure will be carried back to the late 1960's in the next comprehensive NIPA revision.

Video cassette rentals.--The revised estimates for PCE for recreational services included first-time estimates of expenditures on rentals of video cassettes for home viewing. These expenditures raised PCE services by $1.0 billion in 1984, $2.9 billion in 1985, and $3.2 billion in 1986. The estimates were developed primarily from household surveys on cassette rentals conducted in 1984-86 by a private market research firm and from trade data on sales of video cassette recorders.

Previously, cassette rentals from establishments mainly engaged in retail sales (e.g., convenience stores) had been implicitly included in PCE goods estimates; rentals from establishments mainly engaged in the rental of cassettes were not included in PCE at all. The latter establishments are covered--but not separately identifiable --in the Census Bureau service annual survey in the category for business equipment rental and leasing; this category is not used in estimating PCE.

In the revised estimates, video cassette rentals are estimated separately as part of PCE services. The assumption is made that the overstatement in PCE goods due to cassette rentals included in retail sales data is offset by the lack of an adjustment in PCE goods for sales of video cassettes, recorders, etc. by establishments mainly engaged in cassette rentals.

Net exports of services.--The revised estimates for net exports incorporated the changes that were made in June 1987 to the U.S. balance of payments accounts that improved coverage of commissions on securities transactions in exports and imports and of medical services in exports.

The estimates of commissions on securities transactions are made by multiplying the value of securities transactions by estimated average rates for commissions and adding other fees. Transactions are based on U.S. Department of Treasury data on gross security transactions between U.S. and foreign residents and BEA estimates of new securities issues and redemptions. Average rates for commissions and other fees are based on Securities and Exchange Commission data for the United States and on published commission schedules of foreign security exchanges and discussions with security dealers for transactions made abroad. The added coverage of commissions on securities transactions raised exports of services by $0.6 billion in 1984, $0.8 billion in 1985, and $1.5 billion in 1986. Imports of services were raised by $0.6 billion in 1984, $1.3 billion in 1985, and $1.9 billion in 1986.

The estimates of medical services receipts cover services performed at nonprofit hospitals and at State and local government hospitals for foreigners who travel to the United States for medical treatment; estimates cover only hospital in-patient expenditures and associated physicians' fees. The estimates are based on hospital admission information supplied by hospital administrators, primarily at major medical centers and university hospitals and at hospitals in major foreign visitor destinations. (See the technical notes in "U.S. International Transactions, First Quarter 1987' in the June 1987 SURVEY for more detail on these revisions.) The added coverage of medical services raised exports of services by $0.4 billion in 1984, $0.5 billion in 1985, and $0.5 billion in 1986.

Computer prices.--The revised estimates incorporated improved price estimates for computers. In addition to updating the price information and hedonic indexes used for large- and medium-scale processors, printers, and displays, price information for personal computers (PC's) has been incorporated for the first time. The new PC price index decreased at about the same rate as the index for large- and medium-scale processors from 1983 to 1986 (about 15 percent per year), but decreased more rapidly in 1984 and 1986 and less rapidly in 1985. The PC index is a matched-model index based on IBM PC's, judgmentally adjusted by BEA to take into account quality changes associated with the introduction of new models and to take into account models of other manufacturers. Research is underway at BEA to develop an improved PC index using hedonic techniques that covers a wider selection of computer models.

BEA used the new information on computer prices in the deflation of PCE, PDE, exports, imports, and government purchases. For PCE, changes in the PC index were substituted for changes in the combined index covering processors, printers, and displays, which had been used previously. For the other components, the PC index was incorporated into the combined index using shipments weights.

Table: CHART 4 Real Gross National Product

Table: CHART 5 GNP Price Index (Fixed Weights)

Table: 1.--Revisions in Real GNP and Other Major Aggregates Over the Period 1983:IV-1987:I

Table: 2.--Revisions in the GNP Price Index (Fixed Weights) and the Price Indexes of Other Major Aggregates Over the Period 1983:IV-1987:I

Table: 3.--National Income and Product Account

Table: 4.--NIPA Revisions: Selected Component Detail and Major Source Data

Table: 5.--Revisions in Real GNP and Price Indexes (Fixed Weights)

Table:

Table: 6.--GNP, Real GNP, and GNP Prices: Change From Preceding Quarter

Table: A.--SUMMARY NATIONAL INCOME AND PRODUCT ACCOUNTS, 1986

Account 1.--National Income and Product Account

Account 2.--Personal Income and Outlay Account

Account 3.--Government Receipts and Expenditures Account

Account 4.--Foreign Transactions Account

Account 5.--Gross Saving and Investment Account
COPYRIGHT 1987 U.S. Government Printing Office
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1987 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Publication:Survey of Current Business
Date:Jul 1, 1987
Words:4760
Previous Article:The business situation.
Next Article:National income and product accounts tables.
Topics:


Related Articles
National income and products accounts tables.
National income and product accounts tables.
National income and product accounts tables.
National income and product accounts tables.
National income and product accounts tables.
National income and product accounts tables; selected NIPA tables.
National income and product accounts tables; selected NIPA tables and reconciliation and other special tables.
State personal income, revised estimates for 1958-96.
Appendix B: Suggested Reading.
Annual Revision of the National Income and Product Accounts.

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