The Myth of the Specialized Military Contractor
Maryellen R. Kelley and Todd A. Watkins
Technology Review April 1995, v. 98(3) p. 52-8.
Contrary to the beliefs of analysts from both the right and the left, U.S. manufacturers already combine production for military and civilian markets. Would-be designers of industrial policy can readily build on this situation for the post-Cold War era.
Most people think of companies that manufacture the world's most advanced military machines as inhabiting a separate industry highly dependent on the Pentagon. To both critics who want to shrink the defense budget and supporters who want to raise it, wholesale conversion of such firms to serve nonmilitary customers seems fanciful at best. Even the more modest goal of integrating commercial and military manufacturing looks formidable. But our research indicates that the image of a few highly specialized defense contractors occupying an enclave walled off from commercial manufacturing is largely a myth. In extensive interviews with managers in the "machining-intensive durable goods" (MDG) sector, which accounts for half of all durable goods purchased for defense in 1990 and about one-fourth of all U.S. manufacturing output, we found that the vast majority of defense contractors serve both military and civilian customers. What's more, strengths developed under the umbrella of national security are being tapped to benefit firms' commercial work, and vice versa. This is true among not only the major contractors that serve the Pentagon but the thousands of subcontractors that actually perform a substantial share of military manufacturing. This outcome stems partly from the fact that in the industries we studied, many of the features thought to be peculiar to the defense contracting relationship - low volumes, custom orders, a high degree of dependence on a few customers, and a small number of rivals - apply just as strongly to commercial work. A manufacturer custom-builds a new conveyor system for the United Parcel Service or a massive, high-speed paper-cutting machine for a daily newspaper, for example. And managers in the MDG sector reported reaping an average of 60 percent of their revenues from their three largest customers, while some 50 percent cited six or fewer potential or actual competitors, whether or not their companies performed defense work.
Our findings suggest that fears of a defense industrial base jeopardized by the decline in U.S. military spending - from a post-World War II peak of 6.5 percent of GDP in 1986 to 3.5 percent by 1997 - are overblown: strong links between military and civilian manufacturing mean that most firms in this sector will remain capable of supplying the needs of the Department of Defense (DOD) in the coming decade. But our revisionist view does suggest that the kinds of benefits the United States has reaped from its defense industrial policy will require counterparts in various civilian-based initiatives if such strengths are to be maintained.
What Wall of Separation?
In Beyond Spinoff, John Alic of the congressional Office of Technology Assessment and Harvey Brooks, Lewis Branscomb, and Ashton Carter of Harvard's Kennedy School of Government report that the 67 largest publicly traded prime contractors - the major firms responsible for producing military airplanes, tanks, missiles, submarines, and weapons - derived only 9 percent of their total revenues from military work during the peak years of the 1980s buildup. Only a handful of those firms truly specialized in defense work, gaining at least 50 percent of their income from DOD. Today even large defense companies such as Rockwell and TRW take in only about a third of their income from the Pentagon.
Far from responsible for most of the nation's military manufacturing, these major defense contractors stand at the top of diverse and deep supply structures. AlliedSignal, consistently among the top 25 prime contractors, depended upon materials and components from 7,500 to 10,000 suppliers who accounted for 60 percent of its defense contract costs in 1993, for example. Similarly, some 60 percent of the dollar value of Pratt & Whitney's military aircraft engines goes to suppliers. To build the inertial guidance system for the MX missile, Northrop relied on more than 500 subcontractors to make 19,000 parts. And at Lockheed/Fort Worth, subcontracts consume 75 percent of the cost of making the F-16 aircraft.
This supplier base encompasses a significant percentage of all U.S. manufacturing companies. In a 1991 survey of firms in 21 durable goods industries, as well as an analysis of 1988 data gathered by the Census Bureau, we found that fully half of all plants make parts, components, or materials for military equipment. Fewer than 1 in 10 prime contractors and their suppliers in this sector depended on DOD for more than 80 percent of their output in 1990. Clearly a vast number of companies perform both defense and civilian work.
Still, high-profile cases such as Lockheed's Skunk Works, incubator for the U-2 and SR-71 "Blackbird" spy planes and more recently the F117A Stealth fighter, and Maine's Bath Iron Works, which builds Navy destroyers, have helped solidify the view that whatever the overall importance of defense contracts to their total business, companies almost always perform military work in facilities dedicated to defense. But our 1991 survey revealed that most companies in the MDG sector in fact make commercial and defense products on the same machinery with the same people.
For example, General Electric has been among the top 10 defense contractors for more than 35 years. But at GE Aircraft Engines, which derives more than half its revenues from civilian sales, commercial and military production share management, R&D facilities, and manufacturing workstations. In a joint venture with the French firm Snecma, the company produces CFM56-2 jet engines, the technical core of which powers the B-1 bomber, for DC-8 commercial airplanes as well as the Air Force's KC-135R tanker aircraft. Hewlett-Packard's Microwave Semiconductor Division and major producers of satellites such as Hughes integrate military and civilian production, while Pratt & Whitney sells the PW-2037 engine for both commercial and military use.
Examples among medium-to-large firms that supply these prime contractors abound, especially those serving the aerospace industry. The Lord Corp., a leading supplier of rubber-to-metal adhesives and computerized vibration-control equipment, uses a single division and the same engineering group to work on the Boeing 737, 757, and 767 aircraft as well as the Black Hawk helicopter and the Osprey tilt-wing transport. The Casting and Forgings Division of Wyman Gordon Co. employs the same people, processes, and equipment in supplying special alloy castings to GE Aircraft Engines, Pratt & Whitney, Boeing, and McDonnell Douglas. Menasco, one of the world's leading suppliers of landing gear, makes equipment for the military's Apache helicopter in the same plant where it produces titanium parts for the new Boeing 777. And at Vought Aircraft, fabrication and assembly of the wing for the Gulfstream V corporate jet occurs in the same facility as work on the C-17 military transport.
Our visits to smaller subsystem and component manufacturers also revealed completely integrated facilities. Electroid, a specialty manufacturer, produces high-performance equipment that engages or disengages mechanical power, or stops and locks moving parts, in packaging machinery, photocopiers, industrial robots, the Apache helicopter, and the turret on the M1 tank. Electroid's fail-safe brakes also latch bay doors in position on NASA's space shuttle. The company's major customers include AlliedSignal, Boeing, General Electric, Goodyear Aerospace, McDonnell Douglas, Scientific Atlanta, and Westinghouse. According to a division president, the company has purposely prevented its military business from exceeding 20 percent of annual revenues since it began doing defense work in 1981.
A division of Valcor Engineering, Electroid has been in business for 35 years, 17 in its present manufacturing facility, which employs 100 people. This facility dedicates no equipment or employees to military production: a worker might spend one hour on a military job and the next hour on a product for an industrial robot. Engineers, too, design products for both commercial and military customers. The only feature unique to military goods identifies all production for the stringent "nuclear, aerospace, and military" sector: NAM work is placed in blue plastic tote bins while pieces for non-NAM customers go in tan-colored bins. This color coding alerts employees to follow specific written instructions at each workstation, since inspection procedures as well as designs, materials, and tolerances must be checked at each stage for NAM customers, whether civilian or military.
Tecknit, another small company that supplies military contractors, makes an array of products for shielding electrical equipment from electromagnetic-radiation interference. These products include conductive elastic polymers (similar to rubber), adhesives, paints, and greases, as well as shielding screens, coated windows, and air-vent panels. At the height of the 1980s defense buildup, the firm's U.S. operations depended on military contracts for more than 70 percent of sales, but this share shrank to 50 percent by 1993.
Focusing on a core technical area enables the company to serve a wide variety of customers: its largest include major defense contractors such as Westinghouse, Rockwell, Raytheon, Boeing, Hughes, and Martin Marietta, as well as AT&T and computer companies such as Apple, DEC, Dell, and IBM. Production is low-volume and labor-intensive - much of the assembly work is still done by hand. At this Company no special labeling distinguishes defense products from commercial goods: the technology, manufacturing equipment, process flow, labor, and engineering are identical. The only differences occur in the more stringent documentation required by DOD of the final inspection and testing.
Delroyd Worm Gear, manufacturer of standard and customized products for both military and civilian markets, makes large speed-reducing gears for conveyors, printing presses, oil drilling pumps, and cranes, as well as canal locks (including some used in the Panama Canal) and aircraft-hangar doors. The company depends on the Pentagon, specifically the Navy, for only 5 to 10 percent of overall sales. At the plant we visited, which employs fewer than 100 people, the company does not distinguish its work for the Navy in any way, including during testing and inspection - the military simply places an order identifying the part it wants Delroyd to build. In the past, orders for replacement parts usually had to match the specifications on the original order, meaning that the Navy sometimes bought gears made with older, and thus inferior and more expensive, materials. But recent procurement reforms have made it easier to change the specifications on purchase orders, thus allowing the Navy to take advantage of new commercially available technologies and materials.
The Advantages of Defense Contracting
Overall we saw little evidence that manufacturers in the MDG sector see defense work as a contaminating drag on their civilian work. Indeed, we found that firms in the defense contracting network are actually better prepared than their strictly commercial counterparts for the collaborative relationships often credited with providing a competitive advantage. That's because support for interfirm learning is particularly strong among companies that do defense work. More than 60 percent of the defense contractors in our survey provided technical assistance to their subcontractors, compared with fewer than 44 percent of strictly commercial manufacturers. Military contractors in the MDG sector are also more likely to collaborate with subcontractors in developing new products and to provide financial aid. One manager told us his company became a defense subcontractor specifically to take advantage of the technical assistance the prime contractors provide insetting up new quality-control systems. Another manager attested that because companies often have access to advanced technology, "being involved in defense work ... keeps your engineers sharp."
In fact, firms that do defense work are way ahead of the curve in applying information technology to manufacturing. By 1991, about two-thirds of the plants with defense contracts we surveyed used programmable machine tools, compared with less than half of the strictly commercial plants. (We estimate that such tools reduce production time per unit of output by an average of 30 percent.) Firms with defense work invested 31 percent more per employee and tended to adopt newer technology. We also found that defense contractors have been more innovative in applying computer-aided design, computer-aided manufacturing, and computerized quality control.
Government initiatives directed at the defense industrial base are at least partly responsible for these significant results. DOD supported the development of advanced technologies and process techniques through its ManTech program during the 1980s, for example. The program disbursed $150-$200 million annually throughout the decade, exceeding all state spending on technical assistance during that period. Although Man-Tech aided a relatively small number of contractors, DOD sponsored annual conferences and workshops to highlight the lessons that early adopters of new manufacturing technologies had learned.
The Department of Defense also financed technical assistance to subcontractors from 1982 to 1992 through its Industrial Modernization and Incentives Program. The aircraft industry was a major beneficiary - all large prime contractors had their own IMIP, and hundreds of subcontractors benefited. The most extensive effort, run by General Dynamics for F16 subcontractors, provided technical assistance to more than 60 firms involved in 200 projects.
Toward New Collaborations
Whether or not the few extremely defense-specialized large contractors are able to survive the downsizing of Cold War-driven demand, the large majority of companies in the military production network clearly are already poised to make the adjustment. Not only have most been using the same workers and equipment to make military and commercial products for years, but even companies highly dependent on defense contracts during the 1980s have diversified their customer base.
Yet if the problems these firms face have almost nothing to do with their fitness to serve commercial markets, our findings do suggest the need to continue federal support for interfirm collaborations aimed at developing and diffusing new technologies and enhancing the abilities of supplier firms, on which the production of major items like automobiles, aircraft, machine tools, and computers depends. The need is especially great given that large firms are demanding that suppliers invest in new technologies and quality control while also cutting prices. Unfortunately, lead companies are much less likely to accompany demands on suppliers outside the defense network with technical aid or funding. Without systemwide support for new technologies and revamped management and shop-floor organizations to support them, suppliers' efforts to cut costs are likely to undermine their long-term viability.
Partly to meet this need, programs such as the Clinton administration's Technology Reinvestment Project (TRP) explicitly support the kind of industrial networks we have found among defense contractors (the average TRP-funded program includes 7.5 organizations) and encourage their extension to new commercial ventures. Although established under the auspices of DOD, TRP aims to stimulate both commercial and military strengths in 11 high-priority areas, including information infrastructure, design and manufacture of advanced electronics, vehicle technology, and health care and environment. Dedicated to speedy proposal review (the program received 2,763 applications in 1993) and requiring virtually no new bureaucracy, TRP's three-pronged approach includes bringing together consortia of companies and universities to develop, demonstrate, and apply dual-use technologies; helping small companies improve their use of advanced manufacturing technology; and enhancing individuals' dual-use engineering and technical skills, especially among those displaced by defense cutbacks. TRP also emphasizes interagency cooperation: the president's National Economic Council coordinates the participation of DOD's Advanced Research Projects Agency, the National Institutes of Standards and Technology, the National Science Foundation, NASA, the Department of Energy, and the Department of Transportation. TRP will award $1.5 billion for 212 projects in 1994-95, with the government contributing 40 percent of the overall costs and other groups, including states and nonprofit organizations as well as industry participants, providing the rest.
The National Institute of Standards and Technology's Manufacturing Extension Partnership Program (MEP) has the potential to serve civilian needs even more directly. For example, an MEP center in Michigan provides a forum where companies making autos and office equipment can learn about suppliers' problems, while a center in Minnesota is helping large firms develop common standards for their subcontractors. The federal government now funds this $90 million program, but states and the private sector are supposed to pick up the tab within a few years. Taxing industry members would be one way, with precedent in sectors such as apparel, to ensure long-term funding for these centers.
To build on such efforts, the bipartisan 41-member Manufacturing Task Force in the U.S. House of Representatives recently proposed a broad package aimed at diffusing new technology, largely to small and mid-sized businesses. Besides expanding the Manufacturing Extension Partnership Program, the task force recommends that the Department of Commerce develop teaching factories and fund state pilot programs to help firms evaluate their technology and training needs.
Institutions that encourage collaborative design, development, and problem solving among large companies, suppliers, and third parties such as government agencies already define the state of the art in European and East Asian industrial policy. In the wake of the Cold War, the United States actually possesses most of the critical ingredients for pursuing these best practices. All that remains is to shed our ill-informed fears that U.S. defense contractors are so isolated from market forces that they are not up to the job, and forgo our ideological opposition to the idea of government and business actively working together to strengthen the commercial industrial base.
Maryellen R. Kelley is with Carnegie Mellon University, and a visiting associate professor in MIT's Political Science Department and its Industrial Performance Center during the 1994-5 academic year. Todd A. Watkins is with the College of Business and Economics at Lehigh University.