Printer Friendly

Defense spending in the 1990's - the effect of deeper cuts.

Defense spending in the 1990's-- the effect of deeper cuts

In recent years, the United States has placed a strong emphasis on military preparedness and development of future weapons systems. Real defense spending climbed from $159.2 billion in 1977 to $265.2 billion in 1987, increasing the Defense Department's share of real gross national product (GNP) from 5.4 percent to almost 7 percent. The rise in defense spending as a proportion of overall Federal purchases of goods and services was even more striking, jumping from 68.7 percent in 1977 to 78.1 percent by 1987.

Combined with continuing pressure to ease the Federal budget deficit, the thaw in East-West relations and the startling political changes in Eastern Europe have led to widespread discussion of defense cuts. This arcticle offers two new scenarios for defense spending based on the moderate-growth version of the Outlook 2000 economic projections, issued by the Bureau of Labor Statistics last fall. (1)

The first scenario envisions an annual reduction of 4 percent in real defense outlays from 1989 to 2000. The second scenario assumes that defense spending will remain constant (in 1982) dollars). Five alternatives to the first scenario-- low-defense--are set forth, and three to the second scenario--high defense.

This analysis also examines detailed industry and occupational employment projections under three of the new defense alternatives. Finally, the effects of spending less on conventional arms or less on highly sophisticated weapons are assessed.

The earlier Outlook 2000 projections had assumed that defense purchases of goods and services, stated in 1982 dollars, would decline at an average annual rate of 1.3 percent, from $262 billion in 1988 to $225 billion in 2000-- an overall decrease of about 14 percent. As part of the spending decline, it was projected that the level of military forces would drop from 2.1 million to 1.9 million, Defense Department civilian employees by 14,000, and private defense-related employment by just over 1 million jobs between 1988 and 2000:

Most of the employment decline was in the private sector because, for the most part, the cuts were assumed to be accomplished by trimming purchases of goods or services, rather than by cutting the armed forces or civilian defense employment.

The increases in defense spending over the 1977-86 period occurred primarily in the areas of research and development and in material purchases. Defense Department civilian employment increased slightly during the 1980's. For the reason, most of the declines that BLS assumed for the 1990's occur not in direct employment levels (either military or civilian) but in material purchases. The effect of this cost-cutting on private sector employment is exacerbated by the fact that may of the largest spending cuts were expected to occur in manufacturing industries with projected high productivity growth:

As shown, three-fifths of the drop is attributed to lower defense spending, but over one-third is projected to result from productivity--output per hour--increases.

In 1988, 2.9 percent of total private wage and salary employment was estimated to be related to defense expenditures. (2) The estimate includes both direct defense expenditures, such as purchases of aircraft or supplies, and indirect expenditures, such as employment generated by purchases made by defense suppliers. By the year 2000, total defense-related employment was projected to decline by one-third, and, as a consequence, would be only 1.7 percent of total private wage and salary employment. Nearly 60 percent of this decline was projected in manufacturing.

The real spending cutbacks had the effect of changing defense spending from 6.6-percent share of GNP in 1988 to a projected 4.3-percent share by 2000, the lowest proportion since 1980, when defense spending accounted for only 5.1 percent of production.

New defense spending alternatives

The BLS Outlook 2000 projections illustrate one possible scenario for declining defense expenditures. Obviously, many others with either sharper or more modest declines are possible. Differing periods or differing mixes of personnel/material cuts could also be explored. This article looks at two basic scenarios covering 1989 to 2000: an upper level of defense spending derived by assuming no change in real defense spending, the high-defense scenario, and a lower level of defense spending derived by assuming a 4.0-percent annual decline in real defense spending, the low-defense scenario. This provides a projected range for real defense spending in 2000 of almost $95 billion--$260.9 billion in the high-defense scenario and $166.5 billion in the low-defense scenario. (3)

The effects of the various assumptions on GNP demand categories and on major economic indicators are presented in tables 1 and 2. In each case, the results should be viewed in comparison with the moderate-growth projections from BLS' Outlook 2000, noted in the tables as "Base 2000."

Exhibit 1 specifies the alternatives, which range from low-defense 1 to high-defense 3.

Low alternatives. Curing real defense spending by 4 percent each year results in cumulative reduction of almost $60 billion by 2000, relative to the Base 2000 projection, the moderate-growth estimate. In the context of the aggregate economic model, however, the decline lowers real gross national product by only $7.4 billion in 2000. As defense spending grows less rapidly, the loss in production generally weakens the economy, at least initially, leading to lower inflation and interest rates. These results, combined with a much larger Federal surplus, lead to lower pressure on foreign exhange rates. The exchange value of the dollar drops approximately 4.0 percent in 2000, resulting in higher exports and lower imports, both of which offset part of the defense cut. Further offsets are provided by small increases in investment as demand is spurred by the lower interest rates. The investment increases are broad-based, occurring in both business spending for plant and equipment and in new residential constructio. Personal spending on nondurable goods and services generally declines slightly. The spending cut also results in a military force level in 2000 that is 460,000 lower than the Base 2000 projection. Most of the veterans enter the civilian labor force and account for increased employment levels in the private economy (table 2). Because GNP is changing very little, the implies slightly lower labor productivity growth.

Under the low-defense 1 alternative, the so-called "peace dividend" appears as a large budget surplus in 2000 and opens the possibility of exploring alternative approaches that offset the defense spending cut. (See table 2.) One approach is to increase Federal nondefense purchases of goods and services by an amount equal to the cuts in defense spending (low-defense 2). This leads to a year 2000 economy virtually identical with that in the base run. Shifts would no doubt be seen at the industry level of detail, but the differences between what the nondefense portion of the Federal Government is buying and what the defense portion is buying are not great enough at the aggregate level to make appreciable differences in either the level or the distribution of GNP. As in the low-defense 1 alternative, however, major military reductions in force result, leading to small increases in the civilian labor force and employment and compensating small declines in labor productivity, relative to the Base 2000 projection.

Another way to absorb the "peace dividend" would be through lower taxex, offsetting defense cuts with a like cut in personal taxes (low-defense 3). Under this alternative, GNP drops slightly because defense reductions are only partially offset by increases in comsumption and investment. The balance of the higher spendable income flows into personal savings, providing a further small boost to investment. As in the low-defense 2 alternative, the Federal surplus is virtually unchanged from that in the Base 2000 projection.

Yet another approach to account for the "peace dividend" is to assume increases among several major categories of Federal civilian spending (low-defense 4 and 5). The major effect is to raise personal disposable income, and hence personal consumption spending, by increasing transfer payments, while allowing the Federal budget to shift from a $26 billion surplus in Base 2000 to a deficit in both of these alternatives. The redistribution of income from taxable sources to nontaxable transfers leads to revenue loss. Investment is virtually unchanged, as small declines in business spending are offset by increases in residential investment. Overall GNP is $14 billion lower in these alternatives than in Base 2000, as interest rates remain high, and there is no consequent boom in investment spending (table 1).

High alternatives. Allowing real defense expenditures to remain unchanged from their 1989 level, high-defense 1, puts defense spending approximately $36 billion higher than in the Base 2000 projection, but GNP rises by only $8.0 billion. The Federal deficit continues over the entire decade, ending up at $60 billion. The budget shortfalls exert more pressure on prices and interest rates, resulting in slower growth for both business and residential investment, lower export growth, and somewhat higher import growth, all offsetting the economic stimulus of high defense spending. Military force levels are 261,000 higher than the Base 2000 projection, reducing the labor force somewhat and leading to slightly lower private employment levels, especially in construction and durable manufacturing.

The higher Federal deficit is offset in high-defense 2 by a cut of Federal civilian purchases of goods and services and in high-defense 3 by an increase in personal taxes. The first alternative results in only very minor differences from the Base 2000 projection. In the second alternative, the higher personal tax rates reduce personal consumption and result in higher inflation and interest rates, as well as a less favorable foreign trade situation. Although GNP ends up slightly higher in this alternative compared to Base 2000, the costs are clear.

Military force levels. The armed forces stood at 2.1 million in 1988. The BLS moderate-growth projections to 2000 included a modest cut of 139,000 in personnel to slightly under 2 million. All of the low-defense alternatives result in a much sharper drop in military levels: 1.5 million or a cut of almost 600,000 from 1988. The resulting inflow of labor to the private sector increases the civilian labor force by almost 400,000 above the level of the moderate-growth projection. In the high-defense alternatives, holding real defense expenditures constant at 1989 levels actually results in a small increase in military force levels over the period, a rise of 120,000 to 2.2 million in 2000. The military increase in turn leads to a decline of 100,000 in the civilian labor force, compared to the moderate-growth labor force.

Although large relative to overall defense spending, the 4-percent annual reductions in 5 of the 8 alternatives remain relatively small proportions of aggregate U.S. demand. To explore the economic effects, it is necessary to carry the analysis further, to the industry and occupational level of detail.

Industry and occupational projections

The decline in defense expenditures in the original 1988-2000 BLS projections has been used to calculate future employment requirements for defense. When those calculations are performed, total defense-related employment is projected to drop by almost 20 percent between 1988 and 2000. Table 3 identifies those industries with the largest absolute declines in employment. While some industries are directly related to defense purchases, such as aircraft and missile engines and equipment, others are indirectly related but provide jobs, such as wholesale trade. Table 4 shows the industries most dependent on defense spending in 1988, ranked by projected employment decline. This grouping includes only those industries directly related to defense, such as ordnance, ships, and aircraft.

Table 5 shows occupations with the sharpest projected defense-related declines in the original 1988-2000 projections. The table lists occupations prominent in defense production, such as electrical and electronic assemblers, machinists, electrical and electronic engineers, and mechanical engineers. Employment in all of the occupations examined and 11 of the 25 industries listed in table 6 is projected to decline in absolute terms from 1988-2000.

Employment alternatives

The employment impact under three of the economic alternatives is now examined in industry and occupational detail. For each alternative the following calculations are made: (1) demand GNP was translated into detailed commodity distributions of sales to final users; (2) total output estimates at both the commodity and industry level of detail were estimated based upon inter-industry flows for 2000 from the previously published moderate-growth BLS projections and the foregoing GNP estimates; (3) the resulting industry output levels were then used to determine associated industry employment levels; and (4) the structure of occupational demand in 2000 was estimated. Defense Department expenditure distributions were patterned after the Base 2000 projection.

The effects of reduced defense spending on industry and occupational employment are viewed from two perspectives, the largest percent changes and the largest absolute differences. For industry employment, see tables 6 through 9 and for occupational employment, tables 10-13. The following discussion focuses on the employment changes associated with the low-defense 1 alternative. Generally, the apposite results and interpretations apply to high-defense 1. For example, employment rises 3.3 percent in construction and new commercial buildings, except offices, under the low alternative (table 6) but falls 3.7 percent under the high alternative (table 7).

Industry perspective. Turning first to the largest percentage job losers, we note those industries most heavily dependent upon direct defense spending, such as guided missiles and space vehicles; ordnance; ship- and boatbuilding; and aircraft. Significant job losses occur in the five most affected industries, with percentage losses tapering sharply in the other industries. The only industry among the biggest 10 job losers that may be unfamiliar is "new non-building facilities." This industry covers a myriad of facilities: ports, military base road and rail systems, and missile silo systems, to name just a few.

Because military spending inherently affects certain industries, the list of job losers presents no real surprises. Other areas of the economy benefit from the reduction in defense spending, as the deficit improves (table 7). Increasing consumer demand results in significant employment increases in the manufacture of footwear; watches, clocks, and parts; luggage and handbags; electronic home entertainment equipment; jewelry and silverware; and toys and sporting goods. Increases in the demand for producers' durable equipment lead to significant employment increases in metal mining and in office and accounting machines. Finally, rising demand for commercial buildings leads to significant employment increases in construction and in primary nonferrous metal mining. Many other industries show similar but smaller positive effects from the defense spending cutback.

The industries with the largest percentage changes in employment are either those most closely related to the Defense Department or those with relatively low employment levels. In the latter case, even a small change in employment can significantly alter the overall level. Another perspective is to examine the industries with the largest absolute changes in employment. The industries selected tend to show small percentage changes in employment.

However, a few categories also show large percent changes--Federal Government; aircraft and missile engines; aircraft; guided missiles and space vehicles; and radio and TV communication equipment. Perhaps more interesting, though, are those industries or activities which undergo relatively large job losses but which are generally not readily associated with defense spending: retail trade; eating and drinking places; research, management, and consulting services; State and local educationl and personnel supply services. The job losses in these categories are small in percentage terms but add up to almost 313,000 jobs, a not insignificant total. But the projected decline in defense-related employment does not produce absolute declines in these industries.

Turning to the larget absolute job gains, we note that increases in demand lead to increasing employment in the construction of commercial buildings and office buildings, as well as real estate--an overall increase of 42,000 jobs. An upturn in demand for producers' durable equipment creates 8,000 jobs in the electronic computing equipment industry. Employment in the remaining industries rises as a result of increasing consumer demand.

Occupational perspective. Just as the industries with the largest percentage of job losses are readily predictable, so too are those occupations with the larget percentage of cuts. Of the top 10 losers, 8 occupations are heavily and directly involved with the design, production, maintenance, or use of military hardware: shipfitters; riggers; electronics repairers; aircraft engine specialists; aircraft assemblers; aeronautical and astronautical engineers; all other motor vehicle operators; and aircraft mechanics. The two remaining occupations, procurement clerks and budget analysts, are heavily represented in the Defense Department.

Occupations with the larget percentage of employment gains tend to be in industries serving burgeoning consumer demand and demand for construction. for the most part, these occupations are in relatively labor-intensive, low-productivity areas of the economy. Further, as with industry employment, relatively few people work in these occupations, although job increases are large from a percentage point of view.

Looking at those occupations with the larget absolute losses, we tend to see support workers such as sales persons, clerical staff, and general management--occupations employed across many industries and likely, as a result, to change in line with employment.

Finally, occupations with the larget absolute job gains are relatively widespread, with no occupation accounting for a very large increase. Most of the gainers are in occupations serving the increase in demand for consumer goods and investment demand.

Alternative spending cutbacks

As noted, it was assumed that cuts in defense spending would affect all types of defense purchases in the same proportions as in the Outlook 2000 projections. The final step of this analysis examines alternative approaches to cuts among purchases of commodities with the concomitant effects on industry employment and occupational demand. Two variations of the low-defense 1 alternative were developed: cuts aimed more at conventional defense spending (alternative-distribution 1), and cuts aimed more at high-technology and research and development spending (alternative-distribution 2).

In both cases, modifications were made to the low-defense "bill-of-goods," that portion of GNP spent by the Defense Department and distributed by the commodities purchased. The redistributed GNP was then used to derive total industry and commodity output estimates, and both employment and occupational estimates were derived. The results appear in table 15 as percent changes from the low-defense 1 alternative. The effects of cuts in high-tech purchases are the opposite of those listed in table 15.

Forcing the cuts into more conventional areas such as ships and ordnance has a positive impact on employment in industries supplying strategic weapons and much of the electronics associated with such weapons. (See table 15.) Not surprisingly, highly skilled professional and technical occupations also benefit. (See table 16.)

Redirecting cuts into high-tech weaponry leads to some increases in the more traditional defense industries--ship- and boatbuilding and ordnance, along with the manufacturing sector industries which support these industries. This alternative has the further effect of raising demand for the less-skilled technical, construction, and manufacturing occupations related to these industries.

IN SUMMARY, the Bureau has explored several alternatives for future defense spending, in aggregate economic terms and in terms of employment in specific industries and occupational groups. Although the effects tend to be relatively minor at the aggregate level, they may be significant in certain industries and occupations most closely tied to the Department of Defense. While those industries and occupations may suffer from significant defense spending cutbacks, other industries and occupations may improve as a result of offsetting economic factors.

Further efforts could fruitfully be aimed at the estimation of regional effects of defense spending cuts, (4) or by estimating the employment and occupational effects of more narrowly defined cuts. (5) At this point, both the extent and timing of any possible cuts in defense spending are unknown. When the first round of budget-making for the 1990's defense establishment is completed, more narrowly defined approaches might be feasible.

(1) "Outlook 2000," Monthly Labor Review, November 1989, pp. 3-74. This series of five articles on the BLS projections to 2000 outlines the shape of the economy and detailed labor supply and demand.

(2) The estimate of defense-related employment in 1988 was derived by multiplying a 1988 employment-requirements matrix by a detailed vector of Defense Department commodity purchases. An employment-requirements matrix shows the direct and indirect employment in all industries generated by $1 of final production and is derived from a detailed total-requirements input-output matrix and similarly detailed estimates of total industry employment for the year in question.

(3) The initial calculations for each scenario assumed only the change noted in defense spending in order to determine the sensitivity of the aggregate economic model to these changes alone. The aggregate economic projections of the Bureau of Labor Statistics are performed in the context of Data Resources, Inc., Long Term Model of the U.S. Economy. For a full description of the model, refer to "The DRI Annual Model of the U.S. Economy," by Joyce Yanchar, in Data Resources U.S. Long-Term Review, Winter 1986-87, pp. 30-43.

(4) This type of regional analysis was presented in "The Peace Economy," Business Week, Dec., 11, 1989, pp. 50-55.

(5) For an example of these types of studies, which are just now beginning to appear, see Budgetary and Military Effects of a Treaty Limiting Conventional Forces in Europe, a Special Study of the Congressional Budget Office, January 1990.

Norman C. Saunders is an economist in the Office of Employment Projections, Bureau of Labor Statistics.
COPYRIGHT 1990 U.S. Bureau of Labor Statistics
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
Title Annotation:includes related articles
Author:Saunders, Norman C.; Singleton, Christopher J.; Devens, Richard M., Jr.
Publication:Monthly Labor Review
Date:Oct 1, 1990
Previous Article:Tracking America's Economy.
Next Article:An experimental price index for the computer industry.

Related Articles
The defense buildup, 1977-85: effects on production and employment.
Employment effects of the rise and fall in defense spending.
Defense-related employment and spending, 1996-2006.

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