Printer Friendly

The business situation.


REVISED (45-day) estimates show that real GNP increased at an annual rate of 4 1/2 percent in the fourth quarter of 1987; the preliminary (15-day) estimates had shown a 4-percent increase (table 1 on page 16). The increase in the GNP price index (fixed weights) was unrevised at an annual rate of 3 1/2 percent.(1)

(1) Quarterly estimates in the national income and product accounts are expressed at seasonally adjusted annual rates, and quarterly changes in them are differences between these rates. Quarter-to-quarter percent changes are compounded to annual rates. Real, or constant-dollar, estimates are expressed in 1982 dollars.

The small upward revision in real GNP was more than accounted for by personal consumption expenditures and nonresidential fixed investment. In personal consumption expenditures, nondurable goods were revised up $4 1/2 billion, reflecting revised retail sales data for November and December. In nonresidential fixed investment, producers' durable equipment was revised up $4 billion, reflecting revised November data and newly available December data for manufacturers' shipments of equipment. Largely offsetting these upward revisions was a downward revision in net exports; imports were revised up $4 1/2 billion, reflecting newly available merchandise trade data for December.

Despite the size of the revisions in some of the major components of GNP, the fourth-quarter direction of change for all of them was the same as in the preliminary estimates. The broad picture of the economy as sketched in last month's "Business Situation" was not much altered. However, a somewhat different view did emerge within two of the major components. In business inventories, the fourth-quarter accumulation in both manufacturing and wholesale trade durables was not nearly as strong as had been estimated last month. In contrast, the accumulation of retail trade inventories -- especially autos and nondurables -- was substantially stronger than estimated earlier. In imports, the fourth-quarter increase in nonpetroleum merchandise imports was considerably stronger than estimated earlier. In contrast, petroleum imports fell even more sharply than estimated earlier.

Leading Indicators

BEA's composite index of leading indicators declined 0.6 percent in January, according to estimates based on 9 of its 11 component series, after increasing 0.3 percent in December. The preliminary estimate for December (based on 9 components) had shown a decline of 0.2 percent; both of the series that were unavailable at that time -- change in credit and change in inventories -- made positive contributions to the revised December index.

Average weekly initial claims for unemployment insurance made the largest negative contribution to the January change. The stock price index, which had declined 26.8 percent over the previous 4 months, increased 4 percent and made the largest positive contribution. The contributions of these and other components of the leading index are given in table 1.

The leading index has now declined in 3 of the 4 months since September 1987, and for January the index was 1.4 percent lower than for September. The January decline, of course, increases the probability that September's level will eventually be identified as a cyclical peak in the leading index. It would be premature to identify it as such now, however, if only because data revisions could alter the path of the index over the last several months.

In the cyclical indicators approach, the direction of the leading index's movement is obviously important in drawing inferences. Less obviously important, perhaps, is the extent to which its direction of movement is shared by its component series. For example, a decline of given depth and duration may be judged more ambiguous if it is the result of declines in only 1 or 2 components than if it is the result of declines in 10 or 11.

It will be recognized immediately that taking the leading index apart to examine component movements is somewhat at odds with the rationale for constructing a composite index. The incongruity, however, merely highlights the pragmatic nature of the indicator approach and the frequently repeated reminder that the leading index must be used in conjunction with other analytical tools.

The extent to which the components move in the same direction as the leading index may be investigated in several ways. One is a diffusion index, which show what percentage of the components is increasing. BEA has published a diffusion index for the leading index for as long as it has published the leading index itself. (2) This diffusion index tends to be quite volatile on a month-to-month basis. Nevertheless, near a peak in the leading index, the diffusion index typically falls from a value of more than 50 -- showing that more than one-half of the components had been increasing -- to a value of less than 50. This pattern has manifested itself recently: The diffusion index dropped below 50 in November and stayed there in December and January (based on the 9 available components). However, the fact that the diffusion index remained above 50 in October -- 1 month after the high point of the leading index -- is not at all typical. In the past, the diffusion index was always below 50 in the month following a peak in the leading index.

(2) In calculating the diffusion index, a component that increases is counted as 1; a component that declines, as zero; and a component that is unchanged, as 1/2. The diffusion index for the leading index is shown in Business Conditions Digest.

While a limitation of the leading index itself is that it conveys no information about how many of its components are moving in a given direction, a limitation of the diffusion index is that it conveys no information about the magnitude of changes in the components. One way to get a sense of both concerns (following peaks) is to ask: What is the smallest number of components that would have to be removed in order to eliminate the decline in the index? If, say, 10 components would have to be removed, then the weakness is obviously very widespread; if, in contrast, only 1 or 2 would have to be removed, then the weakness is much less widespread.

Declines in the leading index in the 4 months following September 1987 would be eliminated by removing 1, 2, 0, and 2 components, respectively. According to this gauge, then, the weakness does not seem to have been very widespread in any of these months. The pattern over the 4 months as a whole describes weakness that was somewhat less widespread than was typical of past peaks: The typical number of components that would have to be removed to eliminate declines in the leading index in the 4 months following past peaks were 2, 3, 4, and 2, respectively. (These "typical" figures are medians, based on the 9 previous peaks in the leading index after 1948 -- including 2 that were not followed by business cycle peaks -- and are based on the same 11 components that currently constitute the index.)

A related approach would be to ask: How would the leading index behave if the component responsible for the single largest negative contribution were removed? In the 4 months since September 1987, removing the single largest negative contribution would result in an index that increased 1/2 percent or more in October and December and that declined slightly in November and January. Following past peaks in the leading index, removing the single largest negative contribution typically resulted in an index that was flat in the first month after a peak and that declined slightly in each of the next 3 months.

These three approaches paint similar pictures of the extent to which the movement in the leading index is shared by the components of the index: Since September 1987, the scope of the weakness in the leading index has been somewhat more moderate than was typical in earlier episodes, although not so different as to look out of place in the context of those earlier episodes.

Looking Ahead...

Local Area Personal Income. A comprehensive revision of county and metropolitan area estimates of personal income is nearing completion. Total and per capita personal income for 1984-86 will appear in the April SURVEY. More detailed estimates of personal income by type and industry, 1969-86, will be available as of May 4.

Gross State Product. Estimates of gross State product will be introduced in an upcoming issue of the SURVEY. The estimates are by industry for 1963-86.

Business Statistics. The 25th edition of this biennial volume is available from the Government Printing Office. It contains historical data and methodological notes for all series in the "S" (or blue) pages of the SURVEY and for selected BEA series. See page 28 for more information.

Table: Table. -- Net Contributions of the Individual Components to Changes in the Leading Index

TABLE: 1987 1988

TABLE: Component Feb. Mar. Apr. May June July Aug. Sept. Oct. Nov. Dec. Jan.
COPYRIGHT 1988 U.S. Government Printing Office
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1988 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:fourth quarter of 1987
Publication:Survey of Current Business
Date:Feb 1, 1988
Previous Article:National income and product accounts estimates: when they are released; where they are available and how they are presented.
Next Article:National income and product accounts tables; selected NIPA tables; reconciliation and other special tables.

Related Articles
The business situation.
The business situation.
The business situation.
The business situation.
The business situation.
The business situation.
The business situation.
The business situation.
The business situation.
Business situation.

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