Printer Friendly

The U.S. national income and product accounts: revised estimates, annual 1987-89.

The U.S. National Income and Product Accounts: Revised Estimates * Annual 1987-89 * Quarterly 1987:I-1990:I

In this issue of the Survey of Current Business, the Bureau of Economic Analysis presents revised estimates of the national income and product accounts (NIPA's) for 1987-89 and the first quarter of 1990. As is usual in July, source data that are more complete, more detailed, or otherwise more appropriate than the information previously available have been incorporated into the estimates, and seasonal factors have been updated. In addition, several methodological changes have been made.

The first section of this article discusses the impact of the revisions on several measures of economic activity, the second section provides a summary of the revisions and the major source data underlying them, and the third section describes the changes in methodology made this July and summarizes the source data and methods used to prepare the NIPA estimates. Appendix A to this article shows, in current dollars, the previously published and revised annual estimates and the revisions for condensed versions of the five summary accounts of the NIPA's. Appendix B shows the revised 1989 annual estimates for the full five summary accounts. The complete set of NIPA tables follows this article; an index to the NIPA tables begins on page 109.

Impact of the Revisions

The revised estimates show that during 1987-89 the U.S. economy grew at a slower, although still moderate, pace than was indicated in the previously published estimates. From the fourth quarter of 1986 to the first quarter of 1990, the growth rate (average annual rate of increase) for real GNP was revised down 0.3 percentage point - from 3.6 to 3.3 percent (table 1). The growth rate for real gross domestic purchases was also revised down 0.3 percentage point - from 2.8 to 2.5 percent. Among major components, the average annual rates of change for personal consumption expenditures (PCE) for services, nonresidential producers' durable equipment, Federal nondefense purchases, and State and local government purchases were lower than previously estimated. [Tabular Data Omitted]

In recent quarters, the revised estimates show that the economy was considerably weaker than was previously indicated. On the revised basis, the increase in real GNP in each of the last three quarters of 1989 and the first quarter of 1990 was 1.7 percent or less; previously, the increase in only one of those quarters had been that small. The increase in PCE, particularly for services, was considerably less on the revised basis.

By either set of estimates, inflation was moderate during the 3-year period. From the fourth quarter of 1986 to the first quarter of 1990, the average annual rate of increase in GNP prices was revised down from 4.4 to 4.3 percent, and that in gross domestic purchases prices was revised down from 4.6 to 4.5 percent (table 2). Among major components, only the average annual rate of change for prices of nonresidential structures was much different than previously estimated; prices of these structures increased 3.1 percent instead of 3.8 percent. [Tabular Data Omitted]

The sections that follow provide information for several measures for which the revisions had a noticeable impact.

Personal income, outlays, and saving. - Recent increases in personal income and disposable personal income (DPI) were not as strong as previously indicated. From the first quarter of 1989 to the first quarter of 1990, the increase in personal income was revised down from 7.4 to 6.1 percent; wages and salaries, in particular, were not as strong. The increase in DPI was similarly revised - from 7.5 to 6.2 percent.

Personal outlays - largely PCE - was also not as strong according to the revised estimates. Over the four quarters, the increase in personal outlays was revised down from 7.2 to 6.5 percent. The previously published estimates showed income growing faster than outlays over the four quarters; the revised estimates show income growing more slowly than outlays.

In constant dollars, the impact of the revisions on disposable income and consumer spending was similar to that in current dollars. From the first quarter of 1989 to the first quarter of 1990, the increase in real DPI was revised down from 2.6 to 1.3 percent, and the increase in real PCE was revised down from 2.4 to 1.5 percent.

Because the downward revisions in current-dollar DPI were larger than those in outlays, personal saving and the personal saving rate (personal saving as a percentage of DPI) are both lower on the revised basis. The previously published estimates showed the personal saving rate moving from 5.6 percent in the first quarter of 1989 down to 5.1 percent in the third quarter and then up to 5.8 percent - the highest rate in nearly 5 years - in the first quarter of 1990. The revised estimates show the personal saving rate moving from 5.2 percent down to 4.1 percent and then up to 4.9 percent - a rate below that of a year earlier.

Composition of gross saving. - The revised estimates show that gross saving in the U.S. economy was higher in 1988 and lower in 1989 than previously indicated. Within gross saving, private saving and public dissaving were both smaller in 1989 on the revised basis. The revised estimates show private saving of $779.3 billion in 1989, $26.9 billion lower than previously estimated; within private saving, personal saving was $171.8 billion, $32.6 billion lower than previously estimated. Partly offsetting the downward revision in private saving, public dissaving - that is, the government deficit (NIPA basis) - was $87.8 billion in 1989, $16.7 billion smaller than previously estimated. The Federal Government deficit, at $134.3 billion, was $14.2 billion smaller on the revised basis: Personal tax and non-tax tax receipts and corporate profits tax accruals were higher than previously estimated, and purchases of goods and services and subsidies less the current surplus of government enterprises were lower. For State and local governments, the fiscal position appeared weaker than previously indicated for social security funds but stronger for "other" funds.

Productivity and related measures. - The substantial downward revisions in real gross product and in wages and salaries in 1989 led to lower Bureau of Labor Statistics estimates of productivity, compensation per hour, and unit labor costs in the nonfarm business sector. On the revised basis, real gross product per hour shows a slight - 0.3 percent - decline in 1989, the first annual decline since the current expansion began in 1982. In contrast, the previous estimate showed a 0.9-percent increase. On the revised basis, compensation per hour shows a 3.0-percent increase in 1989, the smallest increase in the current expansion. In contrast, the previous estimate showed a 5.5-percent increase, the largest increase in the current expansion. On the revised basis, unit labor costs increased 3.3 percent, the largest increase in the current expansion. The previous estimate had shown an even larger increase of 4.5 percent.

The 1989 changes in productivity and compensation per hour should be interpreted with caution because the 1988 estimates of real gross product and compensation were affected by two unusual aspects of the calendar. The year 1988 was a leap year and had an extra Friday. The leap year results in adds to the change in real gross product in 1988 and reduces the change in 1989. The extra Friday, because Friday is the most frequent payday, adds to the change in compensation in 1988 and reduces the change in 1989. The changes in productivity and compensation per hour are affected because the measure of labor input used in these calculations is not adjusted for these variations in the calendar - that is, labor input is based on a constant 52-week year. BEA was unable to estimate the impact on the change in real gross product of the extra day of production. BEA did estimate that adjusting the changes in compensation to remove the inconsistency in compensation per hour would reduce the 1988 change of 4.8 percent by about 1/2 percent and increase the 1989 change of 3.0 percent by about 1/2 percent.

Summary of the Revisions

The incorporation of newly available source data and of changes in methodology leads to revisions in current-dollar estimates and in estimates of prices. In turn, these revisions lead to revisions in constant-dollar estimates. This section describes the revisions in the current-dollar, price, and constant-dollar NIPA estimates for 1987, 1988, and 1989 and for the first quarter of 1987 through the first quarter of 1990.(1)

Annual revisions in current dollars

The level of current-dollar GNP was revised down $8.7 billion, or 0.2 percent, for 1987; down $6.9 billion, or 0.1 percent, for 1988; and down $33.2 billion, or 0.6 percent, for 1989. These revisions were about the same size as those of the last four July revisions. In contrast, the direction of the revision in GNP this July was downward for all 3 years. This July, the largest dollar revisions among the major components of GNP were in personal consumption expenditures and government purchases of goods and services.

The level of charges against GNP - that is, gross national income - was revised down $2.8 billion, or 0.1 percent, for 1987; up $11.8 billion, or 0.2 percent, for 1988; and down $40.5 billion, or 0.8 percent, for 1989. These revisions were also of comparable size to those of the last four July revisions. The largest dollar revisions among the major components of gross national income were in compensation of employees, proprietors' income, corporate profits, and net interest.

Table 3 summarizes the current-dollar annual revisions in major NIPA components. It provides a guide to the revisions by identifying the subcomponent series in which revisions were $2.0 billion or more and by listing the major source data that underlie the revised estimates. (For a list of the principal source data and estimating methods used in preparing the current-dollar estimates, see table 6.) It should be noted that newly available source data lead to a revision in the level of an estimate not only for the year into which they are directly incorporated, but often to revisions in the levels for subsequent years as well. The next sections follow the sequence of entries shown in table 3. [Tabular Data Omitted]

Personal consumption expenditures (PCE) for goods. - PCE goods was revised up $5.6 billion for 1987, $10.0 billion for 1988, and $8.0 billion for 1989. The upward revisions for all 3 years were largely accounted for by "goods other than motor vehicles and gasoline and oil." The revisions in these goods resulted from the incorporation of revised Census Bureau retail sales data; the largest upward revisions were in clothing and shoes and in "other" nondurable goods (except fuel oil and coal). Purchases of new trucks were also revised up for all 3 years, reflecting new information on shipments from the 1987 Census of Manufactures. Consumer purchases of used autos were revised down for 1988 and 1989, reflecting newly available information on sales and margins of used auto dealers from the Census Bureau 1988 Annual Retail Trade Survey.

PCE services. - PCE services was revised down $7.0 billion for 1987, $6.9 billion for 1988, and $28.9 billion for 1989. Revisions in personal business services largely accounted for the downward revisions in PCE services for 1987 and 1988 and contributed to the substantial downward revision for 1989. Within personal business services, the incorporation of newly available information from a variety of regular sources led to downward revisions in two components - in services furnished without payment by banks, credit agencies, and investment companies and in the expenses of handling life insurance.

For 1989, medical care services was revised down sharply, reflecting the incorporation of preliminary data from the Census Bureau 1989 Service Annual Survey into the estimates of physician services and of "other" professional services and the incorporation of estimates of premiums and benefits from the Health Care Financing Administration into BEA's estimates of health insurance services. In addition, recreational services was revised down for 1989, reflecting data on lottery revenues from Census Bureau surveys of State government finances. Private education and research services were also revised down, reflecting preliminary data on vocational education from the 1989 Service Annual Survey.

Nonresidential structures. - Nonresidential structures was revised little for 1987, down $0.4 billion for 1988, and up $1.3 billion for 1989. For 1989, an upward revision in "nonfarm structures other than public utilities and mining exploration, shafts, and wells" - largely industrial and commercial buildings - was largely offset by a downward revision in public utilities. Upward revisions in industrial and commercial buildings reflected revised Census Bureau value of construction put in place. The downward revision in public utilities reflected newly available data from a variety of regular sources.

Nonresidential producers' durable equipment (PDE). - Nonresidential PDE was revised up $0.7 billion for 1987, up $1.6 billion for 1988, and down $1.0 billion for 1989. Within PDE, revisions were sizable but offsetting. Information processing and related equipment was revised down for all 3 years, reflecting the incorporation of newly available shipments data from the 1987 Census of Manufactures and from the 1988 Census Bureau Current Industrial Report covering computers into the estimates for office, computing, and accounting machinery. Transportation and related equipment was revised up for all 3 years, reflecting newly available data on shipments of trucks from the 1987 Census of Manufactures and on shipments of aircraft from the 1988 Census Bureau Current Industrial Report covering civilian aircraft.

Residential investment. - Residential investment was revised little for 1987 and 1988 and down $3.6 billion for 1989. The 1989 revision was more than accounted for by a sharp downward revision in "additions and alterations, major replacements," reflecting the incorporation of information from Bureau of Labor Statistics (BLS) and Census Bureau surveys.

Change in business inventories. - The change in business inventories was revised down $1.0 billion for 1987, down $4.4 billion for 1988, and up $1.2 billion for 1989. Reflecting the incorporation of revised estimates from the U.S. Department of Agriculture (USDA), farm inventories were revised down for 1987. Within nonfarm inventories, manufacturing inventories were revised up for 1987, reflecting the incorporation of information on book values from the 1987 Census of Manufactures and of revised BEA unit labor cost indexes. The upward revision in manufacturing inventories more than offset a downward revision in "inventories other than manufacturing and trade," which reflected the incorporation of inventory data from Internal Revenue Service (IRS) tabulations of corporate tax return data for 1987. For 1988, newly available data on book values from Census Bureau surveys led to a downward revision in nonfarm inventories; manufacturing, wholesale trade, and retail trade inventories all were revised down. For 1989, "other" inventories were revised up, reflecting information from the Census Bureau Quarterly Financial Report for mining and from a variety of sources for other industries.

Net exports. - Net exports was revised down $2.1 billion for 1987, down $0.4 billion for 1988, and up $1.0 billion for 1989. For 1987 and 1988, exports were revised up less than imports; for 1989, exports were revised up slightly and imports were revised down slightly. For the most part, these revisions reflected the revisions to the U.S. balance of payments accounts released in June 1990. (For information about the revisions in the U.S. balance of payments accounts, see the technical notes of "U.S. International Transactions, First Quarter 1990" in the June 1990 Survey.)

The upward revision in exports for 1988 was in both merchandise and services - mainly in factor income. Within merchandise exports, there were sizable but offsetting revisions in the principal end-use categories for all 3 years, reflecting a change in the treatment of reexports (exports of foreign merchandise): Capital goods (except autos) and, to a lesser extent, consumer goods were revised up, and "other" goods was revised down. (See the changes in methodology section of this article.)

The upward revisions in imports for 1987 and 1988 were concentrated in services - evenly split between factor income and "other" services. For 1989, profits from foreign investment in the United States (part of imports of services in the NIPA's) was revised down, reflecting the incorporation of additional information from BEA surveys of direct investment.

Government purchases. - Government purchases was revised down $4.7 billion for 1987, $6.4 billion for 1988, and $11.0 billion for 1989. The revisions were largely in State and local government purchases, where newly available data from Census Bureau surveys of government finances led to downward revisions in "purchases other than employee compensation and structures" for all 3 years. State and local government employee compensation was revised up for 1989, reflecting the incorporation of BLS unemployment-insurance-based wage and salary data.

Federal Government purchases was revised down for 1989, as newly available USDA data led to revised estimates of net purchases by the Commodity Credit Corporation.

Compensation of employees. - Compensation of employees was revised down $3.6 billion for 1987, $2.5 billion for 1988, and $65.4 billion for 1989. The substantial downward revision for 1989 partly resulted from the regular incorporation of BLS tabulations of unemployment-insurance-based wage and salary data into the annual estimates of wage and salary disbursements; these data replace the current quarterly estimates for 1989 based on BLS monthly data on employment, hours, and earnings. An upward statistical bias adjustment that BEA used to derive the 1989 current estimates exacerbated the revision in the wage and salary estimates. (BEA had introduced this adjustment in 1988 to account for a consistent pattern of differences between the source data. For additional information, see the changes in methodology section of this article.) Moreover, the revision would have not been so large if BEA had made a downward adjustment in the 1989 current estimates to allow for the effect of having an extra Friday in calendar year 1988. (As occurs about once every 7 years, there were 53 Fridays in 1988 instead of the usual 52; because Friday is the most frequent payday, this occurrence boosted wages and salaries in that year.) For 1989, wages and salaries for all of the major industry divisions were revised down; the largest revision was in the service industries.

Supplements to wages and salaries was revised down for all 3 years. The revisions were largely in other labor income, primarily in pension and profit-sharing plans, reflecting IRS tabulations of corporate tax return data, and in group health and life insurance, reflecting revised Health Care Financing Administration estimates.

Proprietors' income with inventory valuation adjustment (IVA) and capital consumption adjustment (CCAdj). - Proprietors' income with IVA and CCAdj was revised up $11.8 billion for 1987, $26.4 billion for 1988, and $27.2 billion for 1989. The upward revisions for all 3 years were largely accounted for by nonfarm proprietors' income. For 1987, IRS corrections to tabulations of partnership tax return data led to revisions in the adjustment that BEA makes to account for interest expenses that are passed through to partners instead of being reported on the partnership return. (For information about the adjustment, which is offset in both net interest and personal interest income, see the changes in methodology section of the July 1988 Survey article on the revised NIPA estimates.) The effect of the incorporation of corrected 1987 IRS partnership tax data on the interest expense adjustment was partly offset by a downward revision in the adjustments that BEA makes to account for the misreporting of information on tax returns. For 1988 and 1989, newly available IRS tabulations of sole proprietorship and partnership tax return data for 1988 raised the estimates of nonfarm proprietors' income.

By industry, the upward revision in nonfarm proprietors' income (without CCAdj) for 1987 was more than accounted for by finance, insurance, and real estate. For 1988 and 1989, services was revised up substantially, and retail trade, mining, and construction were also revised up. Finance, insurance, and real estate was revised down for 1988 and 1989.

The CCAdj for nonfarm proprietors' income was revised up for all 3 years. (See capital consumption allowances with CCAdj for a list of the major source data incorporated into this estimate.)

Farm proprietors' income was revised up for 1988 and 1989, reflecting the incorporation of revised estimates from the USDA.

Rental income of persons with CCAdj. - Rental income of persons with CCAdj was revised up $0.3 billion for 1987, $0.6 billion for 1988, and $0.3 billion for 1989. These small revisions reflected the incorporation of newly available information from a variety of regular sources.

Corporate profits with IVA and CCAdj. - Corporate profits with IVA and CCAdj was revised up $9.6 billion for 1987, $9.0 billion for 1988, and $10.3 billion for 1989. These upward revisions were largely attributable to the incorporation of newly available IRS tabulations of corporate tax return data for 1987 into the estimates of profits before tax. By industry, profits in manufacturing, public utilities, and construction were revised up for all 3 years; profits in trade and finance were revised down for all 3 years. Profits from the rest of the world were revised up for 1989, largely reflecting the lower estimate of profits on foreign investment in the United States.

The IVA for corporate profits was for manufacturing, was revised down for 1988 and 1989, reflecting revised BEA unit labor cost indexes.

The CCAdj for corporate profits was revised down for 1989. (See capital consumption allowances with CCAdj for a list of the major source data incorporated into this estimate.)

Net interest. - Net interest was revised down $23.1 billion for 1987, $21.1 billion for 1988, and $15.7 billion for 1989. The downward revisions were more than accounted for by net monetary interest; interest paid by business was revised down substantially, and interest received by business was revised up substantially. Part of the revision in interest paid by business was traceable to revisions in the adjustment for interest expenses passed through to partners (see the section on nonfarm proprietors' income). Largely reflecting the revised estimates of interest paid and received by business, monetary interest received by persons was also revised down substantially for all 3 years. Both interest paid and interest received by government were revised down for all 3 years. (See table 6 for a brief description of the sources and methods used to prepare the estimates of net interest. For more detail, see the section on net interest in Notes on Sources of the Revision of the July 1986 Survey article on the revised NIPA estimates.)

Net imputed interest was revised up for 1987 and 1989. Newly available data led to higher estimates of imputed interest paid by life insurance carriers and private noninsured pension plans, which were partly offset by lower estimates of imputed interest paid by banks, credit agencies, and investment companies.

National income. - National income was revised down $5.1 billion for 1987, up $12.3 billion for 1988, and down $43.2 billion for 1989. These revisions reflected the aforementioned revisions in compensation of employees, proprietors' income, rental income of persons, corporate profits, and net interest.

Capital consumption allowances with CCAdj. - Capital consumption allowances with CCAdj - that is, economic depreciation - was revised up $0.3 billion for 1987, $0.7 billion for 1988, and $2.1 billion for 1989. The revisions reflected revised BEA estimates of fixed investment and prices.

Capital consumption allowances - that is, tax-return-based depreciation - was revised up considerably more than the economic depreciation measure for 1987 and 1988 and slightly more for 1989. For 1987, the upward revision largely reflected newly available IRS tabulations of corporate tax return data and revised IRS data for nonfarm sole proprietorships and partnerships; for 1988, it reflected newly available IRS tax-return-based depreciation for nonfarm sole proprietorships and partnerships. The downward revision in corporate capital consumption allowances for 1989 mainly stemmed from revised estimates of fixed investment.

The CCAdj - which is derived as the difference between the tax-return-based measure and the economic measure of depreciation - was revised up for all 3 years.

Nonfactor charges. - Nonfactor charges - indirect business tax and nontax liability, business transfer payments, and the current surplus of government enterprises less subsidies - were revised up $2.0 billion for 1987, down $1.2 billion for 1988, and little for 1989. The largest revisions were in indirect business taxes for 1988 and 1989 and in the current surplus of government enterprises less subsidies for 1988 and 1989. The downward revisions in indirect business taxes largely reflected newly available information from the Census Bureau surveys of State and local government finances. The upward revisions in the current surplus item were partly attributable to a change in the treatment of deposit insurance payouts, which lowered the operating expenses of several Federal Government enterprises. (See the changes in methodology section of this article.) In addition, the incorporation of Federal budget data for fiscal year 1989 led to lower estimates of the deficits for the Commodity Credit Corporation and the U.S. Postal Service.

Statistical discrepancy. - The statistical discrepancy - the difference between GNP and charges against GNP - was revised down $5.9 billion, to - $10.6 billion (or 0.2 percent of GNP), for 1987; down $18.6 billion, to - $28.2 billion (or 0.6 percent of GNP), for 1988; and up $7.4 billion, to - $17.0 billion (or 0.3 percent of GNP), for 1989.

Personal income and its disposition. - The pattern of the revisions in personal income - down $11.2 billion for 1987, up $6.3 billion for 1988, and down $43.0 billion for 1989 - partly reflected the previously described revisions in wages and salaries, other labor income, proprietors' income, and rental income of persons. It also reflected revisions in personal interest income, personal dividend income, transfer payments, and personal contributions for social insurance. Personal interest income was revised down for all 3 years, largely reflecting the aforementioned revisions in the estimates of net interest. Personal dividend income was revised up for 1989, reflecting the incorporation of data from publicly available corporate financial statements. Transfer payments was revised up for all 3 years, reflecting data from a variety of government sources; for 1989, most of the upward revision was accounted for by transfer payments from State and local governments.

Personal tax and nontax payments was revised little for 1987, up $5.0 billion for 1988, and up $10.3 billion for 1989. State and local government payments were revised up for 1988 and 1989, reflecting newly available Census Bureau data from annual surveys of government finances and from quarterly surveys of tax revenues. Federal Government payments were also revised up for 1988 and 1989, reflecting newly available data from the Social Security Administration and the U.S. Treasury Department.

Reflecting the revisions in personal income and in personal tax and nontax payments, disposable personal income (DPI) was revised down $11.2 billion for 1987, up $1.4 billion for 1988, and down $53.3 billion for 1989.

Largely reflecting the revisions in PCE, personal outlays was revised down $1.9 billion for 1987, up $0.5 billion for 1988, and down $20.7 billion for 1989. For 1988, interest paid by consumers to business was revised down, largely reflecting the incorporation of Federal Reserve Board data on consumer installment credit.

Personal saving - the difference between DPI and personal outlays - was revised down $9.3 billion for 1987, up $0.9 billion for 1988, and down $32.6 billion for 1989.

Annual revisions in prices

Revisions in fixed-weighted price indexes stem from the incorporation of newly available source data and of methodological changes. Source data that effect prices consist not only of price indexes, which are used for deflation, but also of current-dollar estimates or quantity data used for components for which the constant-dollar estimates are prepared by quantity extrapolation or direct valuation. (See the updated summary methodologies section of this article.)

In general, revisions in prices trend to be small, mainly because much of the source data used to derive GNP price indexes are not subject to large or frequent revisions. For example, the BLS Consumer Price Index is not routinely revised after its initial release, and the BLS Producer Price Index is typically revised only slightly; these indexes are the basic sources for price estimates of components that account for over three-fourths of GNP.

Newly available price information incorporated this July for 1987-89 includes a revised BEA computer price index, revised price indexes for foreign travel, and revised price data for national defense goods and services. Newly available current-dollar estimates affected the price estimates for services furnished without payment by banks, credit agencies, and investment companies and for brokerage charges and investment counseling. Newly available quantity data affected the price estimates for petroleum and natural gas drilling, for margins on used autos, and for State and local government compensation.

In addition, methodological changes were made in the deflation of several GNP components. For residential and nonresidential structures, a new Census Bureau deflator for houses under construction was introduced. For net exports, the use of BLS monthly export and import prices to deflate merchandise trade data was extended back to the first quarter of 1989. (See the changes in methodology section of this article.)

The level of the GNP price index (fixed weights) was revised down 0.2 index point to 118.9 for 1987, 0.2 index point to 123.9 for 1988, and 0.1 index point to 129.5 for 1989. Reflecting these small revisions in level, the annual percent increase in the index was revised down 0.1 percentage point to 3.5 percent for 1987 and was unrevised at 4.2 percent for 1988 and 4.5 percent for 1989. Revisions in the annual percent changes in the price indexes for the components of GNP were small except for nonresidential structures for all 3 years, residential structures for all 3 years, imports for 1989, and Federal nondefense purchases for 1989 (table 4). [Tabular Data Omitted]

The change in the prices of nonresidential structures was revised down 0.6 percentage point to - 0.4 percent for 1987, up 0.6 percentage point to 5.8 percent for 1988, and down 1.7 percentage points to 3.0 percent for 1989. The large downward revision for 1989 was mainly attributable to a revision in the implicit deflator for petroleum and natural gas well drilling that reflected trade source information on footage drilled. In addition, the revisions in the prices of nonresidential structures for all 3 years reflected the introduction of the new Census Bureau index for houses under construction into the deflation procedures of several components of nonresidential structures.

The increase in the prices of residential structures was revised down 0.8 percentage point to 3.7 percent for 1987, up 0.4 percentage point to 3.5 percent for 1988, and down 0.4 percentage point to 3.6 percent for 1989. These revisions were largely attributable to the introduction of the new Census Bureau index for houses under construction.

The increase in the prices of imports was revised down 0.5 percentage point to 3.6 percent for 1989. The revision was largely in imports of services, mainly in expenditures by U.S. residents for foreign travel and in purchases by the U.S. military.

The increase in the prices of Federal nondefense purchases was revised up 0.5 percentage point to 5.4 percent for 1989. The revision reflected the incorporation of newly available current-dollar data from Federal Government sources on employee compensation.

Annual revisions in constant dollars

In general, revisions in constant-dollar - or real - GNP and its components reflect (1) current-dollar revisions, (2) price revisions, and (3) "other" revisions that result from redistributions of current-dollar levels within components or from deflation at a finer level of detail. The following tabulation provides a breakdown of the percent revisions in the level of real GNP and of the revisions in the annual percent change of real GNP.
 Percent revision Revision in annual
 in level percent change
 1987 1988 1989 1987 1988 1989
Current-dollar revisions -0.2 -0.1 -0.6 -0.2 0 -0.5
Less: Price revisions -.1 -.2 -.2 -.1 0 0
Plus: Other revisions -.1 -.3 -.2 -.1 .1 0


Equals: Constant-dollar

revisions -.2 -.2 -.6 -.3 .1 -.5

The level of real GNP was revised down 0.2 percent for 1987, 0.2 percent for 1988, and 0.6 percent for 1989. As was the case for current dollars, these revisions were similar in size to, but different in direction from, those in the past four July revisions. Reflecting these revisions in level, the annual percent increase in real GNP was revised down 0.3 percentage point to 3.4 percent for 1987, up 0.1 percentage point to 4.5 percent for 1988, and down 0.5 percentage point to 2.5 percent for 1989. For 1987, current-dollar revisions and "other" revisions accounted for the downward revision in real GNP; for 1989, current-dollar revisions accounted for the downward revision. By major components, the downward revision in the increase in real GNP for 1987 was largely accounted for by nonresidential fixed investment, and the downward revision for 1989 was more than accounted for by PCE.

Revisions in components of real GNP. - The annual percent increase in PCE was revised little for 1987 and 1988 and was revised down 0.8 percentage point to 1.9 percent for 1989. The 1989 revision, which mainly reflected the current-dollar revisions, was largely accounted for by a substantial downward revision in PCE services, more than one-half of which was in medical care services. PCE goods was also revised down for 1989; the revisions were mainly in food and in furniture and household equipment.

The increase in nonresidential fixed investment was revised down 1.3 percentage points to 2.6 percent for 1987, little for 1988, and up 0.6 percentage point to 3.9 percent for 1989. The downward revision for 1987 was accounted for by PDE. Within PDE, a large downward revision in office, computing, and accounting machinery was only partly offset by upward revisions in most other PDE categories. (In current dollars, the downward and upward revisions about offset each other; the difference between the current-and constant-dollar revisions was attributable to the extremely low level of the deflator for office, computing, and accounting machinery relative to the deflators for the other PDE categories.) The 1989 upward revision in nonresidential fixed investment was accounted for by structures. Within structures, both industrial and commercial buildings were revised up, reflecting the current-dollar revisions.

The change in residential investment was revised up 0.9 percentage point to 0.4 percent for 1987, down 0.4 percentage point to - 0.8 percent for 1988, and down 1.2 percentage points to - 4.1 percent for 1989. For 1987 and 1988, the revisions largely reflected the price revisions for single-family structures. The 1989 revision was attributable both to price revisions and to current-dollar revisions for single-family structures and to current-dollar revisions for additions and alterations.

The change in inventory investment (that is, the change in the change in business inventories) was revised little at $17.2 billion for 1987, down $3.4 billion to - $0.8 billion for 1988, and up $6.2 billion to $0.2 billion for 1989. The 1988 and 1989 revisions, which largely reflected the current-dollar revisions, were accounted for by nonfarm inventories.

The increase in exports was revised little for 1987, up 0.7 percentage point to 18.3 percent for 1988, and little for 1989. The upward revision for 1988, which largely reflected the current-dollar revisions, was about evenly split between merchandise and services.

The increase in imports was revised up 0.7 percentage point to 8.2 percent for 1987 and little for 1988 and 1989. The 1987 upward revision largely reflected the current-dollar revisions in services.

The increase in government purchases was revised down 0.4 percentage point to 2.3 percent for 1987, little for 1988, and down 0.4 percentage point to 2.3 percent for 1989. The 1987 and 1989 downward revisions, which largely reflected the current-dollar revisions, were concentrated in State and local government purchases.

Quarterly revisions

July revisions in quarterly NIPA estimates come about in three major ways: (1) Adjustment of the estimates to reflect the annual revisions, (2) incorporation of new and revised source data (including the updating of seasonal factors) that are used to indicate quarterly patterns, and (3) changes in the methodology used to prepare the quarterly estimates. This July, the revisions in the quarterly estimates largely reflected the revisions in the annual estimates that were previously described. In general, the quarter-to-quarter patterns of changes in GNP, real GNP, and GNP prices were not markedly different on the revised basis (table 5). [Tabular Data Omitted]

For real GNP, the revisions in the 13 quarterly percent changes (at annual rates) averaged 0.5 percentage point (without regard to sign), slightly less than for the last four July revisions. Changes for three quarters were revised by 1.0 percentage point or more. For the third quarter of 1987, the increase in real GNP was revised down 1.2 percentage points to 4.1 percent; this revision was largely accounted for by nonresidential fixed investment. For the first quarter of 1988, the increase in real GNP was revised up 1.1 percentage points to 5.1 percent; this revision was mostly in PCE and farm inventory investment. For the third quarter of 1989, the increase in real GNP was revised down 1.3 percentage points to 1.7 percent; this revision was in PCE, farm inventory investment, and exports.

For GNP prices, the revisions in the 13 quarterly percent changes (annual rates) averaged 0.2 percentage point (without regard to sign), about in line with the last four July revisions. In only one quarter was the revision larger than 0.4 percentage point. For the second quarter of 1987, the increase in GNP prices was revised down 0.6 percentage point to 3.5 percent; price changes for most of the major components were revised down in that quarter.

Methodology

The revised NIPA estimates incorporated several changes either in the source data or in the methods used to prepare the estimates. This section of the article describes these changes and updates previously published tables showing summary methodologies for current- and constant-dollar estimates of GNP.

Changes in methodology

U.S. balance of payments revisions. - Except for a few definitional and statistical differences, the BEA balance of payments accounts provide the basis for the foreign transactions entries in the NIPA's. Of the changes in methodology that were incorporated in the June 1990 revisions in the U.S. balance of payment accounts, the following have been incorporated in the revised NIPA estimates this July: An adjustment to the estimates for travel and passenger fares to account for underrepresentation in the source data for overseas travel, a change in the procedure for estimating dividend payments to foreigners on U.S. stocks, and the reclassification of reexports to principal end-use categories for merchandise exports. (For information about these changes, see the technical notes of "U.S. International Transactions, First Quarter 1990" in the June 1990 Survey.)

Among these changes, the reclassification of reexports - that is, exports of foreign merchandise - had a substantial impact on the distribution of exports by end-use category. Beginning with 1987, reexports have been assigned to detailed end-use categories in the same manner as exports of domestic merchandise. Previously, all reexports had been included in the residual "other" category of merchandise exports. The growth of reexports in recent years has largely reflected the expansion of manufacturing activities, particularly the manufacture of capital goods and consumer goods, within foreign trade zones - secured areas legally outside U.S. customs territory - in the United States. The reclassification of reexports led to current-dollar downward revisions in the "other" category of merchandise exports of $8.8 billion for 1987, $12.1 billion for 1988, and $14.1 billion for 1989. The largest upward revisions were in capital goods (except autos) and consumer goods: Capital goods were revised up $4.6 billion for 1987, $6.4 billion for 1988, and $7.8 billion for 1989; consumer goods were revised up $2.0 billion for 1987, $2.7 billion for 1988, and $3.3 billion for 1989.

Wages and salaries. - Beginning with the estimates for January 1990, BEA has discontinued an upward statistical bias adjustment to the extrapolator used to prepare the current monthly and quarterly estimates of wages and salaries. As described in table 6, the extrapolator for most private industries is a measure derived from BLS monthly data on employment, hours, and earnings. At the time of the July annual revision, the current estimates are replaced by estimates of wages and salaries based on BLS tabulations of employees covered by State unemployment insurance (UI). For several Julys prior to 1988, the incorporation of the UI data had resulted in substantial upward revisions. This pattern had led BEA to introduce a bias adjustment to the extrapolator that raised the annual total of wages and salaries by 1 percent.

Effects of Federal tax law changes. - Federal tax law changes - especially the Tax Reform Act of 1986 - led to changes in the adjustments that BEA makes in using the IRS corporate tax return data to estimate corporate profits and, in some instances, other NIPA components. These adjustments conform the tax return data to NIPA coverage and definitions.

Two adjustments have been discontinued. (1) The adjustment to treat contributions to payroll-based stock ownership plans as a deduction instead of as a tax credit has been discontinued; these tax credits are no longer allowed (also affects other labor income). (2) The adjustment to recognize retail installment credit sales as income in the year of the sale regardless of when payment is received has been discontinued; the installment method of accounting can no longer be used.

Three adjustments have been modified. (1) The adjustment to treat capitalized construction interest payments as current-period expenses has been modified; the definition of property for which interest must be capitalized has been changed (also affects nonfarm proprietors' income and net interest). (2) The adjustment to restate profits resulting from long-term contracts so that they reflect the percentage-of-completion method of calculation instead of the completed-contract method has been modified; the use of the completed-contract method is being phased out. (3) The adjustment to define the deduction for bad debts as actual losses instead of as an amount based on the reserve accounting method has been modified; the reserve accounting method can no longer be used except by small banks and thrift institutions.

Five adjustments have been introduced. (1) An adjustment to restate profits of small business corporations is needed to reflect certain income and expense items; these items now can be passed through directly to shareholders instead of being reported by these corporations (also affects net interest). (2) An adjustment to treat 100 percent of expenses for business meals and beverages and entertainment as deductions is needed; only 80 percent of these expenses is now deductible. (3) An adjustment to restate profits of property and casualty insurance companies is needed; unearned premiums and unpaid losses are now not fully deductible. (4) An adjustment to restate profits of mutual life insurance companies is needed to reflect amounts actually paid to policyholders as dividends; these companies are now required to adjust their deductions to make their profitability equivalent to that of stock life insurance companies. (5) An adjustment to eliminate the effects of required restatements of prior years' inventories is needed; uniform capitalization of inventory expenses is now required (also affects other labor income and net interest).

NIPA table 8.13 presents a reconciliation of corporate profits as tabulated by the IRS with NIPA profits before taxes. The effects of the changes summarized above, with the exception of the modification to the adjustment for the accounting of bad debts, are included in line 2 of the table; the bad debt adjustment is shown in line 11.

NIPA treatment of the "bailout" of thrift institutions. - In December 1989, BEA changed the treatment of the expenses incurred by the Federal agencies involved in the supervision of and provision of deposit insurance for depository institutions. This July, the change in treatment has been carried back through the first quarter of 1987.

In the NIPA's, the income of Federal agencies such as the Federal Deposit Insurance Corporation and the Federal Savings and Loan Insurance Corporation (and its successors) is considered a nonfactor charge against GNP and is included as part of the "current surplus of government enterprises." The income of these agencies had been calculated as the deposit insurance premiums received from insured institutions less the agencies' operating expenses. These expenses had included net payouts to depositors in failed institutions but had excluded various types of financial assistance such as payments to a "healthy" institution to facilitate their acquisition of, or merging with, a failing one. The financial assistance was treated as an asset transfer, which is excluded from GNP and charges against GNP because it does not arise from current production. Beginning with the third-quarter 1989 NIPA estimates, BEA began treating all of the losses associated with savings and loan failures as asset transfers; under this treatment, net payouts to depositors are now excluded from the calculation of the current surplus of government enterprises. (For more information, see the "Business Situation" article in the December 1989 SURVEY.)

The carrying back of this change in treatment this July led to downward revisions in the NIPA item "subsidies less current surplus of government enterprises" of $1.0 billion for 1987, $3.5 billion for 1988, and $1.9 billion for 1989.

Deflation of residential structures. - The revised estimates of most components of residential structures reflected the incorporation of a newly developed Census Bureau deflator. BEA is now using this new index, a deflator for single-family houses under construction, to replace a deflator based on the Census Bureau price index for sales of new single-family houses. Both the new Census Bureau deflator and the sales price index are derived using a "hedonic" (or regression) method in which characteristics of specific houses are regressed against their sales prices to obtain implicit prices for the characteristics. The new deflator reflects prices associated with the size, location, and other characteristics of houses under construction in the current period; the sales price index reflects these characteristics for houses sold in the current period. Thus, the new deflator is more appropriate for the NIPA residential structures component, which is a value-put-in-place concept. Both the new deflator and the sales price index also reflect the following improvements to the regression equations: Separate equations are estimated for attached and for detached houses in each of four regions, and new housing amenity characteristics are used.

The new deflator is used for single-family and multifamily residential structures, additions and alterations, nonhousekeeping residential structures, and net purchases of existing residential structures from governments. The improved price index for sales of new single-family houses is used for brokers' commissions on the sale of new and existing houses.

Deflation of nonresidential structures. - The change in the deflator for single-family structures also affects the deflation of several components of nonresidential structures. The new Census Bureau index for houses under construction replaces the price index for sales of new homes in the deflation of nonresidential buildings, mining, and net purchases of nonresidential used structures.

Two other changes in deflation procedures affect the public utilities component of nonresidential structures. For railroads, a price index for railroad construction assembled by BEA using information from the Interstate Commerce Commission replaces a deflator that was an unweighted average of the Bureau of Reclamation Construction Cost Trend Composite Index and the Federal Highway Administration Composite Bid Price Index for highway construction. For telephone and telegraph, a new deflator assembled by the Census Bureau from weights and indexes provided by the C.A. Turner Company replaces the discontinued Handy-Whitman cost index for electric utilities.

Deflation of net exports. - In January 1990, BEA began using the BLS monthly export and import price indexes to deflate the monthly merchandise trade data for the current quarterly estimates of real GNP. (Export and import price indexes for the first 2 months of a quarter, which have been published by BLS since early 1989, are based on a subsample of the third-month sample of these prices.) Previously, Census Bureau monthly unit-value indexes had been used in the current estimates until the third-month BLS indexes became available. This July, the change in the deflation procedure has been carried back to the first quarter of 1989.

Updated summary methodologies

Table 6 identifies the principal source data and estimating methods used to prepare the current-dollar estimates of the income- and product-side components of GNP, and table 7 identifies the principal source data and estimating methods used to prepare the constant-dollar estimates of the product-side components. These tables have been updated to reflect the methodological changes introduced this July. In addition, the PCE sections of the tables have been made consistent with the descriptions in the newly published NIPA methodology paper Personal Consumption Expenditures (see the inside back cover for order information).

Current-dollar estimates of GNP. - The components in table 6 are as shown in the national income and product account (see appendix B, "Summary National Income and Product Accounts, 1989," account 1), starting on the income side and proceeding to the product side. The subcomponents in table 6, with their 1989 dollar values, are grouped according to the methodology used to prepare them.

The column in table 6 for annual estimates covers the several annual estimates in the estimating cycle; the major differences in methodology as the estimates move through the three annual revisions to a comprehensive (benchmark) revision are few enough to condense into the table. For example, for most goods in PCE (the first item on the product side), the table indicates one methodology for benchmark years and another for all other years.

The column for the quarterly estimates is a condensation in two respects. First, it refers to the advance estimate for the current quarter - that is, the estimate prepared in the first month following the end of the quarter. That one estimate, rather than all three of the current quarterly estimates, is described because more attention focuses on the "first look" at the quarter. Second, even for the advance estimate, the column does not detail how many months of source data are available nor whether the data are subject to revision by the source agency.

Table 6 lists source data referring to a variety of different economic measures - wages and salaries, premiums, expenses, interest rates, mortgage debt, tax collections, unit sales, housing stock, employment, and average price, to name a few. For most components, the source data are "value data"; that is, they embody both the quantity and price dimensions that are required for current-dollar estimates. In these cases, the methodology indicated in table 6 is the adjustment of the value data to derive estimates consistent with NIPA definitions and coverage.

When value data are not used in preparing an estimate, the table indicates the combination of data with separate quantity and price dimensions that is used to derive the required value estimate (as well as indicating any major adjustments needed to derive estimates consistent with NIPA definitions and coverage). On the product side, a "physical quantity times price" method is used for several components. For example, the estimate for new autos is prepared as unit sales times average list price. An "employment times earnings times hours" method and variations of a "stock of assets/liabilities times an interest rate" method also are used for several components.

Some of the source data shown in table 6 for the annual estimates are used to interpolate and extrapolate the levels established by source data that are viewed as final, and all of the source data shown for the advance quarterly estimates are used to extrapolate the level of the preceding quarter.(2) In addition to using indicator series, as is the case when source data are listed in the table, extrapolation and interpolation may be based on trends, as is the case when "judgmental trend" is listed in the table.

Constant-dollar estimates of GNP. - Table 7 shows which of three methods is used to prepare constant-dollar estimates and indicates the source data with which it is implemented.(3) The method used for by far the largest part of GNP is deflation. (In fact, deflation is so widely used that the term is often used to describe the preparation of all constant-dollar estimates.) In deflation, constant-dollar estimates are obtained by dividing the most detailed current-dollar components by appropriate price indexes, with the base period - at present, the year 1982 - equal to 100.

The other two methods, quantity extrapolation and direct base-year valuation, are similar to each other in that they both use quantity data. For quantity extrapolation, constant-dollar estimates are obtained by extrapolating base-year values by quantity data. For direct valuation, constant-dollar estimates are obtained by multiplying base-year prices by quantity data for each period.

The subcomponents are as shown in table 6, except where more detail is needed to highlight differences in methodology for constant-dollar estimates. For this table, the distinction between annual and quarterly methodology is far less important than it is for the current-dollar methodology, and major differences between the annual and quarterly source data are noted within the individual entries.

(2.) Extrapolation is a method of extending estimates from one period forward (or backward) in time to other periods. In simple terms, extrapolation applies a percent change - either the percent change in the indicator series or the percent change in the trend - to the level of the preceding period. Interpolation is a method of filling in estimates between two periods. Interpolation applies a more complex mathematical formula - there are several in use - to preserve the pattern of the indicator series consistent with the level of the source data viewed as final.

(3.) With few exceptions, BEA does not prepare constant-dollar estimates of income measures because price indexes cannot be associated with them, as they can be with product measures. Three exceptions are disposable personal income and, as presented in the "Business Cycle Indicators" section of the Survey, personal income and corporate profits. In these cases, the estimates are adjusted for price change by reference to the prices of the goods and services on which the income is spent. BEA derives constant-dollar net national product and national income by preparing constant-dollar estimates of capital consumption allowances with capital consumption adjustment and of the nonfactor charges and then subtracting these estimates from constant-dollar GNP.
COPYRIGHT 1990 U.S. Government Printing Office
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1990 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Publication:Survey of Current Business
Date:Jul 1, 1990
Words:9155
Previous Article:The business situation.
Next Article:National income and product accounts; national income and product accounts tables and NIPA charts.
Topics:


Related Articles
National income and product accounts estimates: when they are released; where they are available and how they are presented.
National income and product accounts tables; selected NIPA 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.
National income and product accounts tables: selected NIPA tables and reconciliation and other special tables.
National income and product accounts; national income and product accounts tables and NIPA charts.
National income and product accounts tables: selected NIPA tables and reconciliation and other special tables.
National income and product accounts: selected NIPA tables and NIPA charts.
Annual input-output accounts of the U.S. economy, 1986.
National income and product accounts.

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters