Plant and equipment expenditures, the four quarters of 1987.

Sullivan, David F. ; Seskin, Eugene P.

Plant and Equipment Expenditures, the Four Quarters of 1987

BUSINESS plans to spend $389.1 billion for new plant and equipment (P&E) in 1987, 2.5 percent more than in 1986, according to the BEA survey conducted in July and August (tables 1 and 2, and chart 1).$f1$f Spending was $379.5 billion in 1986, 2.0 percent less than in 1985.

1. The survey covers expenditures for new facilities and for repair, expansion, or replacement of existing facilities that are chargeable to fixed asset accounts and for which depreciation or amortization accounts are ordinarily maintained. The survey excludes expenditures for land and mineral rights; maintenance and repair that are not capitalized; used plant and equipment, including that purchased or acquired through mergers or acquisitions; assets located in foreign countries; residential structures, and a few other items.

The estimates presented are universe totals of domestic P&E expenditures for all industries surveyed quarterly, which account for nearly 90 percent of capital spending by U.S. nonfarm business. Sample data are compiled from reports on a company basis, not from separate reports for plants or establishments. A company's capital expenditures are assigned to a single industry in accordance with the industry classification of the company's principal product or service.

P&E expenditures differ from nonresidential fixed investment, which is a component of GNP, in type of detail, data sources, coverage, and timing. For further information, see pages 24-25 of the February 1985 SURVEY OF CURRENT BUSINESS.

The latest estimate of planned spending for 1987 is $1.8 billion lower than that reported in June for the survey conducted in April and May. A 0.9-percent downward revision in nonmanufacturing industries more than offsets a 0.3-percent upward revision in manufacturing industries. The previous survey showed planned spending of $390.9 billion for 1987, 3.1 percent more than in 1986.(2)

2. The estimates of planned spending have been adjusted for systematic reporting biases. The bias adjustments are calculated by industry for each planning horizon. For a given time period, the bias-adjustment factor is the median of the ratios of planned to actual expenditures for that time period in the preceding 8 years. Before adjustments, 1987 planned spending was $388.29 billion for "all industries,' $150.16 billion for manufacturing, and $238.13 billion for nonmanufacturing industries surveyed quarterly.

Real spending--capital spending adjusted to remove price changes--is estimated to increase 1.4 percent in 1987. Real spending declined 2.6 percent in 1986, following an increase of 9.6 percent in 1985 (tables 2 and 3). Estimates of real spending are calculated from survey data on current-dollar spending and from estimated capital goods price deflators developed by BEA.3 The capital goods deflator for "all industries' is projected by BEA to increase 1.1 percent in 1987, following a 0.6-percent increase in 1986; the deflator declined 0.4 percent in 1985.

3. Specifically, the current-dollar figures reported by survey respondents are adjusted using implicit price deflators developed for each industry from unpublished detail on current- and constant-dollar nonresidential fixed investment in the national income and product accounts. To estimate planned real spending, the implicit price deflator for each industry is projected using the deflator's growth rate over the latest four quarters for which it is available.

The latest estimate of the real spending increase for 1987 is smaller than the 2.8-percent increase indicated in the previous survey not only because of the smaller current-dollar increase in the latest survey, but also because of revisions in P&E implicit price deflators. If restated to reflect the revised deflators, the previously reported increase for 1987 would have been 1.9 percent. (See the technical note.)

Current-dollar spending in the second quarter of 1987 increased 0.9 percent, to an annual rate of $377.7 billion, following a 3.1-percent decline in the first; second-quarter spending was 3.7 percent lower than anticipated in the previous survey. Plans reported in the latest survey indicate a 5.4-percent increase in the third quarter and a 2.1-percent increase in the fourth.

Real spending increased 0.7 percent in the second quarter of 1987, following a 3.2-percent decline in the first. Estimates indicate a 5.2-percent increase in the third quarter and a 1.8-percent increase in the fourth.

Manufacturing programs

In manufacturing, current-dollar spending increased 0.1 percent in the second quarter of 1987, to an annual rate of $140.8 billion, following a 2.4-percent decline in the first. A 2.6-percent increase in second-quarter spending by nondurable goods industries offsets a 2.4-percent decline by durables. Manufacturers plan a 6.9-percent increase in the third quarter and a 2.1-percent increase in the fourth.

For the year 1987, manufacturers plan to spend $146.4 billion, 2.6 percent more than in 1986; in the previous survey, a planned increase of 2.2 percent was reported. Manufacturers' spending declined 7.0 percent in 1986, following a 10.6-percent increase in 1985.

Durable goods industries plan a 3.0-percent increase for 1987; the largest planned increase is in blast furnaces-steel works. Increases are also planned in electrical machinery, nonferrous metals, "other durables,' and stone-clay-glass. Declines are planned in motor vehicles, machinery (except electrical), and fabricated metals. Nondurable goods industries plan a 2.2-percent increase. Planned increases in textiles, "other nondurables,' food-beverage, and paper more than offset planned declines in rubber, petroleum, and chemicals.

Real spending by manufacturers is estimated to increase 1.4 percent in 1987--2.0 percent in durables and 0.9 percent in nondurables. In 1986, real spending declined 7.8 percent--5.4 percent in durables and 9.8 percent in nondurables.

Nonmanufacturing programs

In nonmanufacturing, current-dollar spending increased 1.4 percent in the second quarter of 1987, to an annual rate of $236.9 billion, following a 3.5-percent decline in the first. Nonmanufacturing industries plan a 4.5-percent increase in the third quarter and a 2.1-percent increase in the fourth.

For the year 1987, nonmanufacturing industries plan to spend $242.7 billion, 2.5 percent more than in 1986; in the previous survey, a planned increase of 3.6 percent was reported. Downward revisions in electric utilities, gas utilities, "commercial and other,' "other transportation,' and air transportation more than offset upward revisions in railroads and mining. Nonmanufacturing industries' spending increased 1.3 percent in 1986, following an 8.4-percent increase in 1985. For 1987, planned increases in "other transportation,' air transportation, "commercial and other,' and gas utilities more than offset planned declines in electric utilities, railroads, and mining.

Real spending by nonmanufacturing industries is estimated to increase 1.5 percent in 1987; it increased 1.0 percent in 1986. For 1987, estimated increases in "commercial and other' and transportation more than offset estimated declines in public utilities and mining.

Technical Note

P&E expenditures estimates have been revised to incorporate new seasonal-adjustment factors from 1984 to date and new P&E implicit price deflators (henceforth, deflators) from 1973 to date. The revised deflators, in turn, reflect (1) revised coefficients used to transform nonresidential fixed investment (NRFI) estimates to a P&E basis and (2) recently revised estimates of current- and constant-dollar NRFI. Table 4 indicates the sources of revision in real spending for 1986 for "all industries,' manufacturing, and nonmanufacturing.

The revised transformation coefficients are based on 1977 capital flow tables, which show use of structures and equipment by industry (see pages 26-35 of the November 1985 SURVEY), and tabulations from the 1977 Census Bureau Enterprise Statistics; they affect the distribution of real spending across industries from 1973 to date, but not the levels of "all industries' spending. The revised NRFI estimates are based on the revision of the NIPA's released in July 1987; they affect the distribution of real spending across industries as well as the levels of "all industries' spending from 1984 to date.

The revised deflators have resulted in substantial revisions for some industries. Among the most widespread revisions are those associated with changes in the transformation coefficients relating to industry computer usage and with revisions in the NRFI estimates relating to computers.

An industry for which the transformation coefficients relating to computer usage increased would, ceteris paribus, have a larger proportion of P&E spending accounted for by computers. As a result, real spending for the industry would be revised up for recent years because the revised P&E deflator would reflect the increased weight being given to the relatively low NRFI deflator for computers. This effect is apparent in the 1986 revised estimates for the nonferrous metals, machinery (except electrical), printing-publishing, wholesale and retail trade, and paper industries.

At the same time, real spending for industries that purchased computers in recent years would, ceteris paribus, be revised up due to the downward revision of the NRFI deflator for computers in July 1987 (see page 15 of the July 1987 SURVEY). This effect is significant in the 1986 revised estimates for the finance and insurance, machinery (except electrical), printing-publishing, aircraft, instruments, electrical machinery, apparel, nonferrous metals, and wholesale and retail trade industries.

Printouts of revised quarterly and annual constant-dollar estimates from 1972-86 are available by writing to P&E Survey Statistics, Business Outlook Division (BE-52), Bureau of Economic Analysis, U.S. Department of Commerce, Washington, DC 20230, or by calling (202) 523-0874.

Trends in Real Capital Spending During the Current Economic Expansion

The current economic expansion is one of the longest in U.S. post-World War II history; real GNP has increased continuously for almost 5 years. Real nonresidential fixed investment (NRFI), a component of GNP, troughed in the first quarter of 1983 (two quarters later than GNP), increased at a faster rate than GNP in 1984 and 1985, and declined in 1986. For 1987, it appears likely that NRFI will show little change, possibly a slight decline. Real plant and equipment (P&E) expenditures, as estimated from the BEA quarterly P&E survey, follows a broadly similar course. P&E expenditures also troughed in the first quarter of 1983, increased at a faster rate than GNP in 1984 and 1985, and declined in 1986. For 1987, the latest survey results, including plans for the second half of the year, indicate a slight increase (chart 2).

The P&E expenditures measure of investment can be used to identify and analyze changes in industry investment patterns.4 An industry focus is useful because many of the factors that influenced investment during the current economic expansion probably had differential industry effects. For example, excess production capacity had a greater effect on investment in some industries--such as public utilities --than in others. The wave of mergers and acquisitions, the introduction of new technology, the continuing shift toward services and away from manufactured goods, changes in U.S. tax laws that affected investment incentives, high real long-term interest rates, and sharply increased foreign competition were other such factors. This section describes industry trends in actual real capital spending for 1983-86 and concludes with an update for 1987 at a more aggregate level.

4. NRFI, in contrast, provides estimates of investment by type of structure and by type of producers' durable equipment. NRFI also differs from P&E expenditures in data sources, coverage, and timing. For further information, see pages 24-25 of the February 1985 SURVEY.

1983-86.--Real capital spending increased rapidly from 1983 to 1985 in both manufacturing and nonmanufacturing; in 1986, it delined in manufacturing but increased slightly in nonmanufacturing (chart 2). Over the entire period 1983-86, there was little overall change in the importance of manufacturing investment relative to nonmanufacturing investment. From 1983 to 1985, manufacturing rose as a percentage of spending by "all industries'; it then declined in 1986, to slightly below its 1983 percentage (table 5).

In manufacturing, durable goods industries made larger contributions to the increases in real capital spending in 1984 and 1985 than did nondurable goods, while in 1986, nondurable goods contributed about twice as much to the decline. From 1983 to 1985, durables increased as a percentage of spending by "all industries'; it than declined in 1986 (chart 3). From 1983 to 1985, nondurables changed little as a percentage of spending by "all industries'; it then declined sharply in 1986.

In durables, the industry with the largest increase as a percentage of spending by "all industries' in both 1984 and 1985 was motor vehicles, an industry that was making major efforts to modernize during the expansion. Electrical machinery also showed a sizable increase in 1984, when the industry was attempting to expand capacity to meet increasing sales and to develop new products. In 1986, the largest decline in durables was in machinery (except electrical); 1986 was a year of weak demand for farm machinery, machine tools, and computers due to sluggish economic growth and increased foreign competition.

In nondurables, 1984 spending as a percentage of spending by "all industries' changed little, due to several small, offsetting changes in its component industries. In 1985, increases in several industries were nearly offset by a decline in petroleum, again resulting in little net change in nondurables. In 1986, increases in "other nondurables,' paper, chemicals, and food-beverage were more than offset by a sharp decline in petroleum. The decline in petroleum was related to the weakness in crude petroleum prices and was widespread among the industry's production, refining and petrochemicals, "other,' and transportation functions. This decline accounted for most of the decline in manufacturing spending in 1986; spending by manufacturing industries other than petroleum declined only 1.7 percent in 1986.

In nonmanufacturing, "commercial and other' made the largest contribution to the 3 consecutive years of increases in real capital spending. From 1983 to 1986, it increased rather steadily as a percentage of spending by "all industries.' Within "commercial and other,' two industry groups--finance and insurance, and wholesale and retail trade--stand out as having consistently made significant contributions to the growth in spending. In the past several years, these industry groups have increased their use of computers and other information processing equipment to automate their operations. The two other industry groups within "commercial and other' for the most part contributed less strongly. As a percentage of spending by "all industries,' personal and business services (including construction) declined in 1984 and 1985 and was little changed in 1986; communication declined in 1984, then increased in 1985 and 1986, so that by 1986 it was slightly above its 1983 percentage.

Real capital spending patterns in the three other industry groups in nonmanufacturing differed markedly from that in "commercial and other' during this period. As a percentage of spending by "all industries,' mining declined after 1984 and 1986 was considerably lower than in 1983. The 1986 decline was especially sharp and was related to the weakness in crude petroleum prices. Transportation changed little as a percentage of spending by "all industries' over the period. Public utilities declined steadily as a percentage of spending by "all industries,' reflecting the completion of many projects before 1984 and less-than-anticipated growth in demand, particularly for electric utilities.

1987.--The latest P&E survey results include estimates of real spending for 1987 based on two quarters of actual expenditures and two quarters of planned expenditures. The estimates for 1987 indicate a continuation of the general pattern of capital spending that has prevailed during the current economic expansion. Overall, estimates for 1987 indicate little change in the relative importance of manufacturing the nonmanufacturing investment; similar increases are estimated for both sectors.

In manufacturing, estimates indicate a slight increase in durable goods and a slight decline in nondurable goods as percentages of spending by "all industries.' The relatively small decline indicated in nondurables for 1987 reflects, in part, moderation of the decline in petroleum after the sharp decline in 1986.

In nonmanufacturing, estimates indicate another increase in "commercial and other' as a percentage of spending by "all industries.' The estimates indicate that transportation again changes little and that the other two industry groups decline further --mining only slightly, but public utilities quite sharply.

Table: 1.--New Plant and Equipment Expenditures by Business 1 and 2, and chart 1).(1) Spending was

Table: 2.--New Plant and Equipment Expenditures by Business in Current and Constant Dollars

Table: CHART 1 New Plant and Equipment Expenditures

Table: 3.--New Plant and Equipment Expenditures by Business in Constant (1982) Dollars

Table: 4.--Sources of Revision of Plant and Equipment Expenditures, 1986

Table: CHART 2 Real Capital Spending by Business, 1981-87

Table: CHART 3 Real Capital Spending by Major Industry Group as a Percentage of Capital Spending by "All Industries,' 1981-87

Table: 5.--New Plant and Equipment Expenditures by Business in Constant Dollars, 1981-87