The Place of The Defense Industry in National Systems of Innovation
The Place of The Defense Industry in National Systems of Innovation
The Place of The Defense Industry in National Systems of Innovation
CORNELL UNIVERSITY
PEACE STUDIES PROGRAM
OCCASIONAL PAPER #25
©April 2000
i
© 2000 Cornell University Peace Studies Program. All rights reserved.
ISSN 1075-4857
The Peace Studies Program was established at Cornell in 1970 as an interdisciplinary program
concerned with problems of peace and war, arms control and disarmament, and more generally,
instances of collective violence. Its broad objectives are to support graduate and post-doctoral
study, research, teaching and cross-campus interactions in these fields.
ii
TABLE OF CONTENTS
PREFACE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v
U.S. Defense Industry in the Post-Cold War: Economic Pressures and Security Dilemmas . . . 47
Kenneth Flamm
The Place of the French Arms Industry in its National System of Innovation and in the
Governmental Technology Policy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
Claude Serfati
The Place of the UK Defense Industry in its National Innovation System: Co-evolution of
National, Sectoral and Technological Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
Andrew D. James
Ideas, Identity and the Limits of European Defense Technology Cooperation . . . . . . . . . . . . . 125
Eugene Cobble
The Defense Sector as a Window into China’s National System of Innovation . . . . . . . . . . . . 168
Corinna-Barbara Francis
The Changing Role of the Defense Industry in Israel’s Industrial and Technological
Development . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 194
Dov Dvir and Asher Tishler
iii
Economic Restructuring, National Strategies, and the Defense Industry in
Newly Industrializing States . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 217
Etel Solingen
CONTRUBUTORS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 227
iv
PREFACE
Military research and development programs are a substantial component of many countries’
public spending for new technology, but their contribution to innovation processes remains a con-
tested issue. Whereas the debate used to center around the question of whether military R&D pro-
grams provided useful spin off to the civilian technology base and enhanced economic growth or,
conversely, exacted a net cost in terms of missed opportunities, the end of the cold war and the
changes associated with globalization have recast the terms of the argument. The lines between mili-
tary and civilian technology have been blurred by the increased importance of information technolo-
gies—an area in which the military lags behind the civilian sector—and by changes in government
policies that aim to increase procurement from civilian sources. In most of the major arms-producing
countries, mergers between defense firms have produced a sharp reduction in the number of major
contractors. Globalization, variously defined, has opened up the borders of national defense indus-
tries to cross-country flows of information and weapons technology that go far beyond the tradition-
al exchanges embodied in state-to-state arms transfers. These changes co-exist with (and are, in part,
a reaction to) secular trends in the complexity and costs of weapons systems, as well as declining
defense budgets.
One way of talking about these issues is to consider the place of defense industries in
national systems of innovation (NSI), where the national system is taken to be the totality of institu-
tions and practices that interact to produce and diffuse new technology. What is the effect of govern-
ment funding and regulations on the production of new technology? How do defense firms relate
to their major customer or to civilian-oriented firms in their industrial sector? What is the nature of
the transnational links and technology flows among defense firms and markets? How have the
changes outlined above affected the patterns established in the past? The value of the NSI approach
is that it focuses attention on the networks or systems that are involved in innovative activities and
at the same time problematizes the boundaries of those systems and the roles of their constituent
players.
The papers collected here are the product of a workshop on “The Place of the Defense Indus-
try in National Systems of Innovation,” held on October 16-18, 1998 at Cornell University in Ithaca,
New York. The workshop brought together specialists in defense industry issues from the United
v
States and Europe to discuss the changes taking place in the industry and to debate the utility of the
NSI approach in a world in which globalizing trends may be undermining the basis for such an ap-
proach. The authors chart various national responses to the changes affecting the defense industry
and military technology programs, and they reach different conclusions about the ability of the state
to retain control over defense technologies. One obvious source of these differences lies in the va-
lence assigned to the military sector, something that varies with a country’s role in the international
system and its individual history, as Claude Serfati’s chapter on France and Etel Solingen’s on new-
ly industrializing states make clear. Variation in the degree to which different states have embraced
market liberalization and in the strategies followed by individual firms in response to market
changes also affect the emerging configurations of state power over the industry. The case studies
range from advanced industrial economies of Europe and the United States to examples of transition
economies (Russia) and industrializing countries (China). Opinions at the workshop differed on
whether globalization has already spread to the defense sector, is certain to do so in the near future,
or will be blocked in leading countries by the state’s interest in maintaining a strong, nationally-
based defense industry.
The rich discussion and debate during the workshop are reflected in the arguments presented
in the following chapters. The authors benefitted from that discussion—particularly the contributions
of our named discussants, Susan Christopherson, Rachel Weber, and Adam Segal—in revising their
papers for publication. We have grouped the chapters under three main themes: Competing Institu-
tional Paradigms; Case Studies of Advanced Industrial States; Examples from Transition and Indus-
trializing Economies.
The workshop was sponsored by Cornell’s Peace Studies Program, with funding from an
institutional grant to the Program from the John D. and Catherine T. MacArthur Foundation. Judith
Reppy and Susan Christopherson organized the workshop, with help from Rachel Weber. Elaine
Scott and Sandra Kisner provided essential administrative support for the workshop, and Sandra
Kisner contributed significantly to the editorial task of producing this publication.
vi
I. COMPETING INSTITUTIONAL PARADIGMS
Judith Reppy
Background
The concept of national systems of innovation first emerged in the work of a network of
scholars based at the Science Policy Research Unit (SPRU) at the University of Sussex, notably
Chris Freeman’s studies of Japan’s system of innovation (Freeman 1987) and in related work by
Bengt-Åke Lundvall and his colleagues at the University of Åalborg in Denmark. Now, more than
a decade later, the concept has become well established as a leading paradigm for analyzing innova-
tion processes; it has also come under attack for being both too broad and insufficiently theorized
(e.g., Radosevic 1998). In this essay I summarize briefly the main elements of a national system of
innovation approach and discuss its utility in analyzing the role of the defense industry in
innovation.
1
See Edquist (1997), “Introduction” for an extended discussion of variants to the Freeman defini-
tion.
1
2
NSI approaches gained popularity in the late 1980s in the context of rising “techno-national-
ism” in western countries, a phenomenon that was, itself, a reaction to the increased competition in
world markets for high technology goods from Japan and the newly industrializing states of East
Asia (Nelson 1993, 3). Another factor was the end of the cold war, and with it a shift in emphasis
from military competition to economic rivalries. No longer was the East-West confrontation the
most important competition; instead, in a spate of books and reports, analysts debated the advantages
of free market strategies versus various types of state intervention.2 There was, however, consensus
about the goal: the wealth of nations, which was seen to depend on success in innovation and diffu-
sion of new technologies. This debate was largely framed in terms of cross-national comparisons
and attention to the multiple institutions influencing innovation—that is, in terms of national systems
of innovation.
2
See for example, the large literature on Japanese success in world markets (e.g., Samuels 1994;
Friedman 1988; Freeman 1987) and the numerous reports from the U.S. Office of Technology
Assessment on U.S. performance and policy options in the high technology sector (e.g., OTA 1988;
1990; 1992; 1994).
3
Standard references include Douglass North (1986); Richard Nelson and Sidney Winter (1982);
John Law and Michel Callon (1992).
3
In evolutionary theories, in particular, there is a great emphasis on learning and on search patterns
as the mechanisms that generate new solutions and bypass bottlenecks in the innovation process.
Outcomes are path dependent, and national differences may be explained in part by national histo-
ries: for example, different regulatory regimes, whether the country is a technology leader or fol-
lower, or even geographical differences can all play a role in shaping technology.4
The NSI approach is particularly well suited to analyses of technology policy, where it fo-
cuses attention on elements beyond government R&D funding and support for education, the topics
that have dominated the discourse on “policy for science” in the past. By drawing attention to the
systemic features of the innovation process and their variation across countries, the NSI approach
cautions against simple policy prescriptions that do not take into account cross-national differences
among competing systems. For example, if the strength of Japanese manufacturing lies in the atten-
tion given to continuous innovation in industrial processes, it does not follow that that success can
be replicated in other countries simply by providing funds for investment in improved manufacturing
technology. The status accorded to engineers working on the shop floor and the supply of trained
technicians may be equally important. Conversely, increased funding for basic research in Japan
may not be effective in increasing the stock of fundamental science if traditional norms of extreme
deference to senior scientists are unchanged.
The weaknesses of the NSI approach are the obverse of its strengths. It casts its net widely,
with a resultant loss of analytical bite (Radosevic 1998; Whitley 1998). Where so many factors may
play a role, assigning relative weight to particular institutions or relationships is difficult. What are
the limits to the system? Are national borders really the appropriate boundary? What about regional
systems, which may or may not lie within national borders, or the increasing globalization of major
industries? Which institutions within the innovation systems are most important? How can cross-
national comparisons be sustained, when the constituent elements of the national systems of innova-
tion may have little in common? Although attempts have been made to address these problems by
imposing more restricted definitions, or shifting the boundaries from national systems to regional
systems, business systems, or technologically defined systems or sectors, they are still at an early
stage. We must conclude that the NSI approach at present is not a formal theory, but rather a
conceptual framework (Edquist 1997, 28-9).
4
For a classic study of these factors, see Hughes (1983).
4
Even a mere framework has its uses, however, when the goal is a discussion of policy op-
tions. Although we may not be able to predict outcomes with any precision, the elements of the NSI
approach draw our attention to the important factors that should be considered in government policy
for innovation. In particular, the NSI approach suggests that interconnections among policies are
important, as are secondary impacts; that the traditional focus on governmental funding of basic
research may not be enough; and that the boundaries of the system, and hence the range of relevant
policy options, may not be easy to define.
System?
The defense industry in the United States has long been understood in systemic terms: to wit,
as a military industrial complex (MIC) or “iron triangle” (e.g., Cooling 1977; Kaufman 1972; Adams
1982). The main features of the MIC are the interlocking and self-reinforcing interests of the mili-
tary, the defense contractors, and members of Congress, who all have reasons to support high levels
of government spending on new weapons development. Although the MIC has been repeatedly
attacked by would-be reformers, it has proved remarkably robust precisely because—however costly
to the taxpayer—it has been highly functional for its constituent members. Protected by its role in
5
national security and nourished by generous budgets, the MIC has produced a continual stream of
new weapons and related innovations for close to half a century.
The institutions and practices that have held the MIC network together include supplier
chains that link subcontractors to prime contractors; the IR&D program, which combines govern-
ment funding for inhouse R&D at major contractors with a system of technical evaluations that has
provided a flow of information about new technological directions to military laboratories; the so-
called “revolving door” or movement of people back and forth between positions in the defense
department, Congress, and the defense industry; tacit understandings about how the budget game
is played; and secrecy practices that protect important aspects of military programs for new technol-
ogy from public review. In addition to government laboratories and large research divisions in
defense contractors, the network includes university researchers, federally-funded R&D centers such
as RAND’s Project Air Force, and a large number of consulting businesses that specialize in tech-
nology assessment and program analysis, all interacting with other elements in the system.
These features are well established in the United States, and—appropriately modified to
allow for differences in governance institutions and the size of the defense industry—they are pres-
ent in other countries with large military R&D budgets as well. In France, for example, the close ties
between government procurement officials and engineers working on defense contacts are exten-
sions of bonds formed during their shared experiences at the elite écoles polytechniques, which grad-
uate a large fraction of the engineers and civil servants in the French system (Chesnais and Serfati
1992, 69).
With the end of the cold war, however, an important prop supporting the MIC was removed:
it is no longer credible to argue that the United States and its allies face a powerful foe with techno-
logically advanced weapons that must be countered with continuing high levels of military spending.
The dangers of the new world disorder lie in more amorphous threats, many of them emanating from
nonstate actors. New technology may still be an important source of state military power, but it is
far from clear what form that technology should take. We can say unequivocally that the older sym-
bols of military hegemony—the tanks, battleships, and bombers—are less relevant to security than
they once were, and the technologies that are replacing them are rooted in the civilian economy
rather than the military.
In the years since the fall of the Berlin Wall, military budgets in the United States and other
NATO countries have been cut; new weapons programs have been stretched out or canceled outright
6
(the latter a violation of one of the old rules of the game, which decreed that programs in trouble
would be reshaped and/or delayed but rarely, if ever, shut down); and in the United States, at least,
the industry has undergone a massive restructuring that has reduced the number of major contractors
to only four companies: Lockheed-Martin, Raytheon-Hughes, Boeing-MacDonald, and—a distant
fourth—Northrop-Grumman. With the exception of Boeing, which has a large commercial aircraft
business, these companies have chosen to concentrate almost entirely on the defense market. They
have sold or spun off many of their civilian subsidiaries and in the process become substantially
more dependent on defense business than they were previously (see Ann Markusen’s chapter in this
publication). In addition, as a response to budget pressures and the reduced number of new pro-
grams, prime contractors have trimmed their subcontractor networks and moved to a system of pre-
certified suppliers. In short the U.S. defense industry is smaller (as measured by the number of sig-
nificant players) and more concentrated now than at any time since the start of the Korean War.
The end of the cold war and falling defense budgets have exacerbated long standing trends
of increasing complexity and cost, trends that had already sharply reduced the number of different
new weapons systems and the length of production runs. At the same time technological leadership
in important technologies, particularly electronics technology, has been shifting from the military
to the civilian sector. The Pentagon has responded with renewed efforts to reform defense procure-
ment practices by eliminating many regulations and encouraging defense procurement officers to
buy off-the-shelf commercial items rather than special purpose military products. In the first Clinton
administration there was also a substantial increase in funding for dual-use programs intended to
benefit both military and civilian users.
These changes, taken together, amount to a substantial restructuring of the MIC system. The
number of major players on the industry side is much smaller; new links to the civilian sector have
been encouraged, even as the prime contractors have become less diversified; some of the rules of
the game have been rewritten; and the involvement and attention of members of Congress is less,
now that defense business has shrunk in many members’ districts. Nevertheless, the MIC is still
recognizably a system, one that to a great extent stands on its own bottom, despite its many links to
other parts of the economy, and one that still oversees the expenditure of very large sums of money
in pursuit of new military technology.
7
National?
National boundaries are the obvious choice for delimiting an analysis of the role of the
defense industry in innovation systems. Security is the primary function of the state, and it is the pri-
mary—if not the only—justification for military R&D programs and new weapons procurement.
Thus, in the name of self-sufficiency and security of supply, countries have long protected their
defense industries against foreign competition whenever possible, even when the result was higher
costs or less advanced equipment. For example, “Buy American” clauses in the United States legis-
late against using foreign suppliers for many systems, and they are supplemented by unofficial, but
no less powerful, military preferences for dealing with U.S. companies over foreign ones. This is
one of the reasons that attempts to create a two-way street in arms flows within NATO, i.e., to in-
crease substantially U.S. purchases of weapons manufactured by its NATO allies, have been unsuc-
cessful. Similar preferences for domestically produced arms prevail in other countries and have
proved resistant to change. The Anglo-French agreement on opening up public tendering in each
country to bids from the other country’s firms, for example, has had little effect on who actually
wins the contracts (Cobble 1998).
National states also put considerable effort into controlling flows abroad of information
about military technology. They maintain elaborate systems of export controls on weapons and tech-
nological information; they supply their allies with equipment that has been “dumbed down;” they
place restrictions on mergers or acquisition of their defense companies across national lines; and
they restrict access by foreign nationals to some kinds of advanced training and employment.
Although these efforts to police transfers of military technology are not always successful, they
testify to the importance that national boundaries have for defense systems of innovation.
A case can be made for redrawing the boundaries of the defense system to allow for trends
towards globalization. In response to rising development costs and shrinking markets for major
weapons systems, there has been a marked increase in the number of strategic alliances between
defense firms across national borders (Reppy 1994). In Europe the project of a European defense
industry has been perennially discussed, although progress was for many years glacial. The example
of (and the threat posed by) the large-scale mergers in the U.S. industry, however, has propelled the
European industry toward consolidation, both at the national and European level, as exemplified by
the merger of British Aerospace and Marconi into BAE Systems and the proposed merger of Aero-
spatiale Matra and DASA to form a European Aeronautic, Defense and Space Co. (EADS). Trans-
8
atlantic partnerships are also on the rise. These new relationships challenge national borders and
governmental controls on the flow of technology because R&D and production activities may move
internationally between subsidiary units and internal transactions among the units are largely hidden
from view.5
The trend towards an increasingly globalized defense industry, while taken for granted by
many analysts, is, however, still largely prospective. To the extent that globalization is identified
with the emergence of transnational defense firms, we can note that effective consummation of a
merger requires development of a shared corporate culture; thus we can expect that transnational
mergers will face special difficulties (Sparaco 2000). Short-term collaborations around specific
projects are less ambitious, but even here difficulties abound (Malone 1980). On the one hand, cross-
national links between some firms and between some governments have existed for a long time—for
example, the special relationship between the United States and Britain has always included
cooperation on defense technology, and several of the European consortia have stayed together
through multiple versions of a weapon systems.
Globalization involves more than foreign trade, international collaborations, or transnational
corporations and capital mobility, however: it speaks to a deracination of consumers and markets
as well as producers. I would argue that in the defense industry the essentially national character of
the national industries remains quite strong for the United States and most European countries, and
that this is generally the case where defense industries have been important components of national
systems of innovation.6 Thus, whether the current restructuring moves of defense firms will create
a de facto global industry remains speculative at this point. An alternative future might be “Fortress
America” countering “Fortress Europe” or even a retreat to nationalism and job protection if eco-
nomic conditions turn sour.
5
A recent report from the DOD’s Defense Science Board (1999) underlines the tension between the
assumed inevitability of the globalization process and security concerns raised by that process.
6
The same can be said for industrializing countries, which for better or worse often give defense
technology a leading role in strategies to promote the development of a national technology base
(Ball 1988; Molas-Gallart 1998; see also the chapter by Etel Solingen, this publication).
9
Innovation?
What role does a nationally-based defense industry play in innovation? With this question
we are on the familiar ground of debates over spin-off, spin-on and spin-away (Samuels 1994), not
to mention the justification that national security gives otherwise liberal states to pursue mercantilist
policies. To many observers, the obvious explanation for U.S. dominance of high technology mar-
kets in the post-World War II period was the cross-subsidization of its civilian technology by invest-
ments in military R&D. Aircraft design, space technology, nuclear power, and solid-state electronics
are examples of areas that benefitted from large-scale military spending, either for R&D or procure-
ment or both. We can also trace management and accounting innovations to developments within
the U.S. defense sector, especially the techniques for managing large-scale, complex programs. The
Soviet Union, France, and Britain also pursued national policies based on investment in military-
related technologies, whereas Japan nourished a strong civilian technology base that has had spin-on
benefits for military technology.
Whether we regard the outcomes as a by-product of security policies that emphasized tech-
nological supremacy or as the result of a conscious policy to base industrial policy on investments
in military technology, the privileging of the defense sector produced innovation systems that were
simultaneously enlarged by high levels of government spending and distorted by the demands of
military procurement. Mary Kaldor, for example, has argued that the defense innovation systems
of the United States and Britain are biased towards trend innovation, ever more baroque and ever
more isolated from civilian needs and markets (Kaldor 1981). These defense innovation systems do
not, therefore, present models that other countries would be wise to emulate. There is little doubt,
however, that they have produced technological innovations and that over time many of these inno-
vations have found their way into civilian products.
National innovation systems that rely heavily on the defense sector provide another kind of
resource for their governments when the political environment is hostile to government intervention
in the economy. Thus, during the Thatcher government in Britain and the Reagan-Bush years in the
United States, industrial policy initiatives for the civilian sector were routinely disparaged as
doomed to failure and ideologically incorrect. Government spending for innovation under the rubric
of national security, however, was exempt from these criticisms, and government programs such as
Sematech and the Very High Speed Integrated Circuits (VHSIC) program in the United States—
10
programs clearly intended to bolster the civilian micro-electronics industry in its competition with
Japanese firms—were funded from the DOD budget.
Even under the Clinton administration, which came into office with a declared intention of
revitalizing the U.S. industrial base, the practice of linking technology programs to the DOD budget
continued, and dual-use programs have emphasized military applications of civilian technologies
rather than the reverse (Stowsky 1999). Technology transfer programs designed to transfer defense
technologies into the civilian economy have had only limited success.7 Thus, the importance of the
defense sector to the national innovation system goes beyond its technological contributions, per se,
to include the shaping of structural and rhetorical dimensions of government policy.
Conclusion
The case for a national system of innovation approach in analyzing defense technology is
a strong one, at least for those countries that have maintained large military R&D programs. In the
defense sector the difficulties with defining appropriate boundaries and naming constituent elements
that bedevil some other applications of the NSI approach are diminished by the strong identification
of military innovations with state security and the interest of the state in maintaining a fence around
military technology. Any analysis of the national systems of innovation for these defense-intensive
countries must give pride of place to the military programs that absorb so many budgetary and real
resources. At the same time, the language of systems draws our attention to the interlocking
elements and processes that make up the defense innovation network, including their connections
to the rest of the economy, national and international. While it clearly would be desirable to have
more refined theoretical propositions to employ in our analyses of the role of the defense industry
in national systems of innovation, the current NSI approach provides a workable framework in
which to begin our task.
7
For example, in Ham and Mowery’s (1998) study of five cooperative R&D agreements (CRADAs)
between Lawrence Livermore National Laboratory and commercial firms, the most unambiguously
successful CRADA was one in which the principal customer for the technology was the laboratory
itself.
11
References Cited
Ball, Nicole. 1988. Security and Economy in the Third World. Princeton, NJ: Princeton University
Press.
Cobble, W. Eugene. 1998. “Identity, Ideas and the Limits of European Defense Integration: The
Failure of the Anglo-French Reciprocal Purchasing Agreement.” Paper presented at the
Peace Studies Program Research Seminar, Ithaca, N.Y., 18 February.
Chesnais, François and Claude Serfati. 1992. L’armement en France: genèse, ampleur et coût d’une
industrie. Paris: Nathan.
Dosi, Giovanni, Christopher Freeman, Richard Nelson, Gerald Silverberg, and Luc Soete, eds. 1987.
Technical Change and Economic Theory. New York: Pinter.
Edquist, Charles, ed. 1997. Systems of Innovation: Technologies, Institutions and Organizations.
Washington: Pinter.
Ergas, Henry. 1987. “Does Technology Policy Matter?” In Technology and Global Industry, ed.
Bruce R. Guile and Harvey Brooks. Washington, DC: National Academy Press.
Freeman, Christopher. 1987. Technology, Policy, and Economic Performance: Lessons from Japan.
New York: Pinter.
Friedman, David. 1988. The Misunderstood Miracle: Industrial Development and Political Change
in Japan. Ithaca, NY: Cornell University Press.
Gummett, Philip and Judith Reppy, eds. 1988. The Relations between Defence and Civil Technolo-
gies. Dordrecht: Kluwer.
Ham, Rosemary and David C. Mowery. 1998. “Improving the Effectiveness of Public-Private R&D
Collaboration: Case Studies at a US Weapons Laboratory.” Research Policy 26, 6: 661-75.
Hughes, Thomas P. 1983. Networks of Power: Electrification in Western Society 1880-1830.
Baltimore: Johns Hopkins University Press.
Kaldor, Mary. 1981. The Baroque Arsenal. New York: Hill and Wang.
Law, John and Michel Callon. 1992. “The Life and Death of An Aircraft: A Network Analysis of
Technical Change.” In Shaping Technology/Building Society, Wiebe Bijker and John Law,
eds. Cambridge: MIT Press.
Lundvall, Bengt-Åke, ed. 1992. National Systems of Innovation. London: Pinter.
Malone, Daniel K. 1980. Roland: A Case for or Against NATO Standardization? National Defense
University Monograph Series, no. 80-5. Washington DC: The National Defense University.
Nelson, Richard R., ed. 1993. National Innovation Systems. New York: Oxford University Press.
Nelson, Richard R. and Sydney Winter. 1982. An Evolutionary Theory of Economic Change. Cam-
bridge: Harvard University Press.
North, Douglass. 1986. “The New Institutional Economics.” Journal of Institutional and Theoretical
Economics 142: 230-37.
Radosevic, Slavo. 1998. “Defining Systems of Innovation: A Methodological Discussion.” Tech-
nology in Society 20: 75-86.
Reppy, Judith. 1994. “Defense Companies’ Strategies in a Declining Market: Implications for Gov-
ernment Policy.” Peace Economics, Peace Science, and Public Policy 1, no. 2: 1–8.
Samuels, Richard. 1994. “Rich Nation Strong Army”: National Security and the Technological
Transformation of Japan. Ithaca, NY: Cornell University Press.
Sparaco, Pierre. 2000. “EADS Foresees Cultural Hurdles.” Aviation Week & Space Technology 152
(21 February): 42-3.
12
Stowsky, Jay. 1999. “The History and Politics of the Pentagon’s Dual-Use Strategy.” In Arming the
Future: A Defense Industry for the 21st Century, ed. Ann Markusen and Sean Costigan. New
York: Council on Foreign Relations.
U.S. Congress, Office of Technology Assessment. 1988. Commercializing High-Temperature
Superconductivity, OTA-ITE-388. Washington, DC: U.S. Government Printing Office.
———. 1990. Arming Our Allies: Cooperation and Competition in Defense Technology, OTA-ISC-
449. Washington, DC: U.S. Government Printing Office.
———. 1992. Building Future Security: Strategies for Restructuring the Defense Technology and
Industrial Base, OTA-ISC-530. Washington, DC: U.S. Government Printing Office.
———. 1994. Assessing the Potential for Civil-Military Integration: Technologies, Processes, and
Practices, OTA-ISS-611. Washington, DC: U.S. Government Printing Office.
U.S. Department of Defense. Office of the Under Secretary for Acquisition and Technology. 1999.
Final Report of the Defense Science Board Task Force on Globalization and Security. Wash-
ington DC: Office of the Undersecretary for Acquisition and Technology (December).
Whitley, Richard. 1998. “Innovation Strategies, Business Systems and the Organisation of
Research.” Paper prepared for the Sociology of Science Yearbook meeting, Krusenberg,
Sweden, 24-27 September.
The Defense Industry as a Paradigmatic Case of “Actually Existing Globalization”
John Lovering
Globalization?
Globalization is held to be a world-embracing process with several dimensions—social, cul-
tural, political, and economic (Waters 1995). In most summaries, the emphasis is on the economic
dimension, which in turn is generally presented as a historic culmination of the drive to efficiency
in the use of resources. This is seen as a consequence of the increased mobility of resources, espe-
cially capital, consequent upon the reduction of national protectionism, signified not least by the
demise of the autarkic “Soviet model” and the emergence of an international economic policy con-
sensus in all but a shrinking minority of “deviant” states. Some commentators add an emphasis on
the revolutionary—and irreversible—nature of new technologies, especially those based on elec-
tronics, and on biotechnologies (Castells 1997).
13
14
This portrayal of globalization—as an irresistible historic process giving rise to new geo-
graphical outcomes—is widely endorsed by official agencies, most notably the World Bank (1995,
1997). Numerous social analysts have drawn out the implications for national socio-economic
change and for policy. In Britain, Dahrendorf (1995, 38), for example, has argued that Western
governments have no choice but to seek to raise “national competitiveness,” and that this means they
have to choose between social cohesion and political freedom. In Europe they have mainly opted
for the latter (while the appointment of Putin in Russia is seen as signifying a turn towards the
former). Dahrendorf is one of a large cohort of sociologists for whom the Western orthodoxy of
political freedom plus globalization is posing new problems of managing widening economic
inequalities. To tackle these, governments are advised to consider a “Third Way” between traditional
social democracy, with de-facto protectionism built in through the effects of public subsidy, etc., and
the equally traditional “laissez-faire” model (Giddens 1998). Quite what this Third Way amounts
to remains obscure, although enthusiasts claim that traces can be seen in the United States and in
Britain.
This conception of globalization as unavoidable, but eventually benign, is founded on both
abstract arguments and empirical observations. The theoretical arguments tend to invoke an econo-
mistic conception of the way economic forces operate and a dualistic conception of the relationship
between the “global” and the “local,” in which the former is the domain of world-historical eco-
nomic forces, symbolized by the transnational corporation and international financial flows, while
the latter is the domain of responsivity, adaptation, or resistance, more often than not resulting in
failure (see Lovering 1996). The running is made by the Global, where economic forces reign
supreme—or at least should be allowed to do so.
As a description of the decision-making processes that lie behind globalization this is
seriously inadequate. Giddens may insist that “Globalizing processes have transferred power away
from nations and into depoliticized global space” (1998, 141). But the idea that globalization is
somehow free of politics is quite mistaken. That globalization is deeply political, albeit often at the
“micro-political” rather than formal electoral State political levels, is the gist of some of the Critical
Globalization literature that has exploded over the past few years (Gordon 1988; Glyn 1995; Eatwell
1995; Amin 1997; Hirst and Thompson 1996; Weiss 1998). “Actually Existing Globalization”—I
choose the term to hark back to debates about the difference between what Socialism might have
been, and what it really was—is shaped by, and reproduces, shifting configurations of power. Within
15
these, traditional nationally accountable entities, such as elected polities, have declining influence.
The defense sector is an outstanding case in point.
For this reason, I suggest that Chomsky’s (1998) recent breezy overview is pertinent here.
Noting that only perhaps 15 percent of all trade is truly “free” (p.19), and that 40 percent of U.S.
imports are actually within-firm transfers, Chomsky writes (p. 27) of globalization as the creation
of:
. . . an international political economy which is organized by powerful states and
secret bureaucracies whose primary function is to serve the concentrations of private
power which administer markets through their own internal operations, through net-
works of corporate alliances, including the intra-firm transactions that are mis-
labelled ‘trade.’ They rely on the public for subsidy, for research and development,
for innovation and for bail-outs when things go wrong . . . In such ways, they seek
to ensure that the ‘prime beneficiaries’ . . . are the right people; the smug and
prosperous [minority of] ‘Americans’. . . and their counterparts elsewhere [notably
in the corporate elites of Britain, and soon Germany, Italy, and even France—JL]
Outside the United States, and to a degree within it, the defense industry is being “de-
nationalized” according to the priorities and strategies of a tiny group of decision makers. While
many economists, defense scholars, and writers on science policy continue to talk of the defense
industry as an arm of the national state (e.g., Sandler and Hartley 1995; Edgerton 1991, 1998), this
is becoming an increasingly misleading description of the way it really works, especially in Europe.
The industry is increasingly becoming privatized, and even where it remains in formal public owner-
ship, its behavior is increasingly “marketized” (Gansler 1995; Lovering 1998a, b, 1999a). It is also
increasingly interwoven with finance capital—this is new. “The City” or “Wall Street” played only
a minor role in the Cold War years, when the industry was essentially financed by the state. The
globalization of defense is not evidence of the much-discussed “decline of the nation state” (Ohmae
1993), but rather of its reorientation and increasingly closed nature and partisan behavior.
Globalization or “Americanization”?
Most of the “globalization” in the defense industry is in fact “internationalization.” It is
premised on the survival, rather than the transcendence, of national states. Since the key state on any
dimension is the United States, this means that “globalization” in defense is to a large degree a
matter of “Americanization.” The simplest way to indicate the overwhelming influence of the United
States is in terms of spending on military R&D. As shown in Figure 1, U.S. spending is roughly
17
equivalent to the combined total of the rest of the world, and if company spending is added in to the
equation, almost certainly exceeds it.
This is but one dimension of the “Americanization” of the world’s defense industries. It is
perhaps more importantly evident in the conscious strategic choices of all major defense companies
to get into the American market, which is both enormous in itself and a key stepping-stone to ex-
ports to third party customers. All the major British and Germany defense companies, for example,
have U.S. sales at the top of their list of priorities. Companies such as British Aerospace (BAe) have
made it clear that their aim is to become a “global” company, but only in the sense that many U.S.
transnational corporations are global: “[W]e wish to appear as British company in Britain, a Japa-
nese one in Japan, and an American one in the USA’ (British Aerospace Strategy Director 1997).”
The substance of the push to globalize, as interpreted by key actors, is thus profoundly
shaped by the uniqueness of the U.S. defense market, and the unique scale of its major defense cor-
porations. There is another dimension to this, just as there is in the earlier successful globalization
of U.S. civilian companies, amongst which Coca-Cola is perhaps a paradigmatic case: in order to
be constructed as a market for U.S. products, buyers in other countries must adopt American styles
of consumption. We can thus observe an “Americanization” of defense markets around the world,
symbolized by the ubiquity of American-style military uniforms, the influence of U.S. (and British)
military technical advisors, and not least the adoption of the characteristically American emphasis
on air power as the key military technology. This is having some bizarre effects. In Central-East
Europe, for example, three relatively poor countries are being asked to invest huge sums in pur-
chasing new aircraft as part of the “ticket” for entry to NATO and the European Union.
This in turn has profound implications for the political-economy of the defense industry
within Europe, and the circuits of flows through which it operates. The pressure for Globalization-
Americanization in the defense industry varies in detail from country to country, but some common
themes are evident. The defense sector is becoming “de-nationalized” not only through international
trade (arms exports), but also and more importantly through the proliferation of cross-border collab-
orative development and production deals of various forms. Current examples in which European
producers are involved include just about every sub-sector, including ships, radar, combat aircraft,
troop-carrying aircraft, helicopters, missiles, and guidance systems. The ammunition sector has per-
haps been the most “autarkic,” and even this is changing, with the prospective merger of British and
18
French capacities (Royal Ordnance and—subject to clarifying the role of state ownership and the
costs of de-manning—GIAT).
Cross-border collaborations, joint ventures, etc. are not merely technical industrial arrange-
ments, they are also inextricably political because they involve the construction of enduring net-
works that tie in companies, government departments, and armed services over many years. The ties
consist not only in the financial penalties built into contracts, but also the weightier if less tangible
constraints of foreign policy agreements. One important effect of these arrangements, and the new
networking circuits they involve, is thus to reduce the susceptibility of a program to national polit-
ical influences. Once established, as a British Minister approvingly noted at the beginning of this
decade, collaborative programs imply “less and less political surveillance over defense production”
(cited in Hayward 1990, 39). They help to insulate the companies involved from what they see as
the disinterest, short-term thinking, or active opposition that characterizes politicians’ attitudes to
defense spending in the post-Cold War era. In so doing they create a new domain within which
corporate decision-makers exert sway unfettered by the kind of interference to which the Cold War
defense industry was subject. This is presented as an efficiency gain, but it is clearly has another
side. In a nutshell, it creates a new arena, one that is chronically prone to corporate capture. And
from the point of view of companies, this indeed is precisely the point (see Lovering 1999a).
For these reasons the defense sector is, I suggest, an outstanding example of the kind of polit-
ically engineered Globalization described in Chomsky’s conception cited earlier. This bears little
connection to either the “abstract” models set out by neo-classical economists or the “inevitablist”
scenarios of sociologists such as Dahrendorf and Giddens.
The NSI literature brings together in a new paradigm a number of observations that have
long been familiar in the defense field. It is useful in that it focuses attention on the relations be-
tween publicly-provided or underwritten capacities and the defense sector in a comparative perspec-
tive. However, the globalization tendencies noted above suggest that the NSI approach is severely
limited in that—at least to date—it lack an adequate historical dimension.
The changes associated with Globalization-Americanization in defense suggest that the
defense sector is in general ceasing to be such a straightforward exemplar of a National System of
Innovation. Put another way, they show that the NSI must be seen as historically as well as geo-
graphically specific. The other side of the coin of Globalization-Americanization in defense is that
the relationship between the defense sector and national science and technology investment is
changing. In short, defense is ceasing to be a major window through which national innovatory
capacities are projected into the national economy in those states that increasingly pattern their
militaries on American technology.
I suggest that this is the predictable consequence of the privatization of the defense industry
that accompanies “Actually Existing Globalization.” In Europe—where since 1997 governments and
industry have been discussing ways to form a cross-border defense giant as the next convulsive
phase of restructuring—the benefits spilling over from the defense sector to other industries, firms,
and employees within the nations are shrinking. Subcontracting linkages are becoming closer, and
the number of firms involved is declining, but this does not follow a national (much less a regional)
pattern. In sum, the net benefits of national inputs of skills, public resources, etc. to defense are
being channelled more exclusively into the private (corporate) networks outlined above.
lavish public funds to pursue the kind of “blue skies” research in which it engaged in the past (result-
ing in some important technological gains, such as vertical take-off and landing aircraft and infra-red
technology, which were then transferred to British companies and remain vital to their profitability
and competitive footholds). The decline of public funding for DERA has been a cause célèbre in the
British defense industry, which although now privatized, still wants to be able to draw upon R&D
inputs provided by the public sector. In 1995 the effects of the rundown were such that it was
claimed that necessary defense aerospace R&D was making unsustainable demands on companies,
which were spending 130 percent of their profits on R&D (OST 1995; Lovering 1995). Defense
R&D fell four times as rapidly as civilian R&D did. The new Labour government has begun to
address this problem, with the Strategic Defense Review of 1998 promising more public spending
on demonstrator programs guided by the companies. Meanwhile, the proportion of R&D in the
British defense industry accounted for by “overseas” resources continues to rise.
For DERA to fulfil its mission of providing expert advise to the MOD, it needs to adopt an
impartial posture. But, as its spokespersons have pointed out, they are increasingly dependent on the
companies themselves for the requisite technical expertise. Public defense research, in short, is in-
creasingly bound up with the construction of “specialist knowledge” in which private interests are
increasingly dominant actors. Globalization, together with neo-liberal public policies, have created
an almost insuperable dilemma for the official repositories of the public interest.
1
For example, see Braczyk, Cooke, and Heidenreich (1998). Two of the authors are European
bureaucrats, while the third is an academic who has worked closely with the Commission, and with
regional agencies.
21
However, the regionalized versions of the NSI literature is much less substantial than its
National elder brother. Indeed it relies on assertion, generalization from abstract propositions, and
extremely loose empirical work. It shares many of these weaknesses with the wider new literature
on “Regional Resurgence,” which is coming under increasingly critical scrutiny (see Lovering
1999b). There is virtually no empirical content to the claims of the RSI writers. Insofar as there is
any evidence of the regional flows of R&D within the United Kingdom, for example, it suggests that
there exist several “islands of innovation” in UK aerospace that are as closely embedded in networks
stretching overseas as in networks indigenous to the United Kingdom (Hickie 1991). Similarly, such
evidence as is available on R&D connections within the United Kingdom shows a national rather
than a regional pattern (Goddard et al. 1994). In short, the notion of the RSI, at least in Britain, is
best described as not very accurate.
The “globalization” of the defense sector would seem to be living proof that if they exist at
all, regional systems of innovation have little influence on the leading edge of modern high-technol-
ogy industry. The impact of “globalization” on the geography of the European defense industry
would seem to be clear in general terms: the consolidation of high-level activities within a shrinking
number of existing established locations (such as Warton, Lancaster—military aircraft; Munich,
Bavaria—military aircraft; Hengelo, Netherlands—naval electronics; Cherbourg, France— naval
ships), alongside a radical decline of employment in lower-tier companies and in labor-intensive
production and assembly activities. A sign of the future in this context is British Aerospace’s recent
agreement with PZL Mielec of Poland to make parts of Hawk airframe, parts that at present are
made at Hull and at Hamble. While some high-level professional jobs remain in Western Europe,
and others decline, the companies are also engaged in proliferating small “buy-local” initiatives that
help to sustain their local image and legitimacy as key employers and important actors in the new
regional structures of economic governance, especially in the labor market (e.g., BAe Warton). The
real action, however, is elsewhere, as a result of increasing de facto corporate and labor market inte-
gration linking sites and elite labor forces in British and other European towns to their collaborators
and intended customers in St Louis or Seattle. BAe has long had most of its software provided by
wire from India. The search for wider markets and lower labor costs is leading to the rapid establish-
ment of new outposts in Capetown, Beijing, and elsewhere. In short, the phrase “globalization-
localization” in the defense sector entails very different effects under each of its terms.
22
References Cited
Amin, Samir. 1997. Capitalism in the Age of Globalisation. London: Zed Books.
Braczyk, Hans-Joachim, Philip Cooke, and Martin Heidenreich, eds. 1998. Regional Innovation
Systems. UCL Press: London.
Castells, Manuel. 1997. The Rise of the Network Society (The Information Age; Economy, Society
and Culture, vol. 1). Oxford: Blackwell.
Chenais, François. 1993. “The French National System of Innovation.” In National Innovation
Systems: A Comparative Analysis, ed. Richard R. Nelson, 192-229. New York: Oxford Uni-
versity Press.
Chomsky, Noam. 1998. “Power in the Global Arena.” New Left Review 230: 3-27.
Dahrendorf, R. 1995. “Preserving Prosperity.” New Statesman 15/29 (December 1995): 36-41
Eatwell, John. 1995. “The International Origins of Unemployment.” In Managing the Global Econ-
omy, ed. Jonathan Michie and John Grieve Smith, 271-86. Oxford: Oxford University Press.
Edgerton, David. 1991. England and the Aeroplane. London: Verso.
——. 1998. “Tony Blair’s Warfare State. New Left Review 230: 123-130.
Gansler, Jacques S. 1995. Defense Conversion: Transforming the Arsenal of Democracy. Cam-
bridge: MIT Press.
Giddens, Anthony. 1998. The Third Way: The Renewal of Social Democracy. Oxford: Polity Press.
23
Goddard, J.C., D. Pike, A. Potts, and G. Bradley. 1994. “Universities and Communities.” Report of
the Committee of Vice Chancellors and Principals. London: CVCP.
Glyn, Andrew. 1995. “Social Democracy and Full Employment.” New Left Review 211 (May/June):
33-55.
Gordon, David. 1988. “The Global Economy: New Edifice or Crumbling Foundations?” New Left
Review: 168 (March/April): 24-64.
Gummett, Philip and Josephine Stein. 1994. European Defense Technology in Transition: Issues for
the UK. London: Science Policy Support Group.
Hickie, D. 1991. Archipelago Europe—Island of Innovation: The Case of the United Kingdom.
Brussels: FAST (Forecasting and Assessment in Science and Technology) Prospective
Dossier No1 FOP303, Commission of the European Communities.
Hirst, Paul and Grahame Thompson. 1996. Globalization in Question. Cambridge: Polity Press.
Lovering, John. 1997. “Global Restructuring and Local Impact.” In Britain’s Cities—Geographies
of Division in Urban Britain, ed. Michael Pacione. London: Routledge.
———. 1998a. “Labour and the Defence Industry: Allies in ‘Globalisation’.” Capital and Class 65
(Summer): 9-20.
———. (1998b) “Opening Pandora’s Box: De Facto Industrial Policy and the British Defense
Industry.” In Manufacturing in Transition, ed. Rick Delbridge and James Lowe, 151-68.
London: Routledge.
———. 1999a. “Which Way to Turn? The European Defense Industry After the Cold War.” In
Arming the Future, ed. Ann Markusen and Sean Costigan. New York: Council on Foreign
Relations Press.
———. 1999b. “Theory Led by Policy?: The Inadequacies of the ‘New Regionalism’ (Illustrated
from the Case of Wales)” International Journal of Urban and Regional Research 23 (June):
379-95.
Lundvall, Bengt-Åke, ed. 1992. National Systems of Innovation. London. Pinter.
Nelson, Richard R., ed. 1993. National Innovation Systems: A Comparative Analysis. New York:
Oxford University Press.
Ohmae, K. 1995. The End of the Nation State. New York: Free Press.
OST (Office of Science and Technology). 1995. Technology Foresight 12: Defense and Aerospace.
London: HMSO.
Sandler, Todd and Keith Hartley. 1995. The Economics of Defense (Cambridge Surveys of
Economic Literature series). Cambridge: Cambridge University Press.
Scott, Allen J. 1998. Regions and the World Economy. Oxford: Oxford University Press.
Storper, Michael. 1995. “The Resurgence of Regional Economies, Ten Years Later.” European
Urban and Regional Studies 2: 161-221.
Walker, William. 1993. “National Innovation Systems: Britain.” In National Innovation Systems:
A Comparative Analysis, ed. Richard R. Nelson, 158-91. Oxford: Oxford University Press.
Waters, Malcolm. 1995. Globalization. London: Routledge.
Weiss, Linda. 1998. The Myth of the Powerless State. London: Verso.
World Bank. 1995. Workers in an Integrating World: World Development Report 1995. Oxford:
Oxford University Press for the World Bank.
———. 1997. The State in a Changing World: World Development Report 1997. Oxford: Oxford
University Press for the World Bank.
Should We Welcome a Transnational Defense Industry?
Ann Markusen*
Introduction
In the 1990s, as defense budgets plummeted, nations scrambled to rationalize their defense
industries, with less than salutary results. In the United States, for instance, the Pentagon twice
altered its policy towards defense industry restructuring. The Bush administration maintained a Cold
War tradition of staunch opposition to full-scale mergers among large defense firms. The Clinton
administration encouraged the implosion of dozens of contractors into just four firms, only to reverse
itself by opposing the Lockheed Martin bid for Northrop Grumman in 1997. The policy did not
produce the savings anticipated, and surprisingly few production lines have been closed down. More
troubling, mergers diminished competition in some weapons lines, undermined defense conversion
and civil/military integration, and produced fewer, more politically powerful defense-specialized
firms.
Now weapons-producing nations face a new development—the proliferation of transnational
mergers and alliances. American contractors, emerging from a decade of deep and abrupt domestic
military spending cuts, are actively seeking foreign, chiefly European, partners. European firms,
long disadvantaged by the small size of their national markets, are trying to merge with each other,
while their governments engage in a politically-embattled privatization process to pave the way. The
defense industry is belatedly doing what its commercial counterparts have done for decades—gone
global—not just by exporting arms but by establishing design, production, and marketing operations
in foreign locales.
But the defense industry is not just like any other industry. Its character and operations pose
technological, economic, political and security problems not present in a Chrysler/Daimler-Benz
*
The author would like to thank members of the workshop and the Council on Foreign Relations
Study Group on the Transnationalization of the Defense Industry and Arms Exports, especially
Michael Brzoska, Anna Giunta, David Gold, Larry Korb, Stephanie Neuman, Michael Oden, Erik
Pages, Judith Reppy, Fred Roggero, and Jeff Roncka for their comments on earlier drafts, and David
Lewis and Harpreet Mann for research assistance. A shorter version of this paper, “The Rise of
World Weapons,” appeared in Foreign Policy, Spring 1999.
24
25
merger. Traditional issues of cost, quality, and innovation in a market already marked by consider-
able concentration will become even more formidable. Governments already find it difficult to
adequately oversee the industry, and coping with transnational suppliers will be just that much more
difficult. In addition to these micro-economic concerns, it will become more difficult to ensure that
government investments in military R&D result in spin-offs that are captured within the national
economy. Transnational defense firms, in other words, pose large challenges to the notion of a
national system of innovation.
There are also security issues. Nations might have to compromise on weapons design, as
other buyers loom larger in the strategic plans of transnational companies. Supply lines could be
more easily disrupted in times of crisis. Sophisticated weapons technologies could move more easily
from country to country, quickening the pace of proliferation. Within the confines of a single firm
or strategic alliance, people, ideas, and technologies, rather than weapons, would move more readily
across national borders, making it difficult for governments to monitor cost, pricing, possession, and
re-export of arms. Lead nations could risk their competitive edge in weaponry altogether, as know-
how diffuses to other centers of expertise.
The military industrial conundrum demonstrates just how deeply economic imperatives have
become intertwined with security policy. There are good economic and political reasons to encour-
age transnational mergers and partnerships. They could speed the elimination of redundant capacity
and lower the cost of designing and producing weaponry. They might also undermine the pork-
barreling that keeps military spending high and crowded into inappropriate activities. But these gains
must be weighed against a fundamental fact—that a transnational defense industry would entail
fewer sellers facing a greater number of buyers, shifting market power on balance from governments
toward private industry.
In this chapter I review these multiple aspects of the transnational defense industrial chal-
lenge. The analysis is based on twelve years of interviews and discussions with defense firm man-
agers, investment bankers, government defense industrial base managers, policymakers, and schol-
ars. It has been subjected to the “vetting” of three years of scrutiny by a broad-based expert study
group at the Council on Foreign Relations in New York. I also draw upon my own published empiri-
cal work and that of others. The discussion focuses on the United States, but the implications are
quite similar for other weapons-producing countries. I speculate briefly at the end on the conse-
quences of transnational mergers for weapons-poor countries.
26
I conclude that global industry restructuring should be welcomed, but only if nations coordi-
nate their defense industrial and arms export policies, matching the global reach of transnational
defense firms. Such coordination will be easier to the extent that cooperative security strategy and
multilateral arms export controls supplant “go-it-alone” security postures. Without the former, polit-
ical pressures will mount on individual governments to support their own domestic companies, even
when owned by or paired with transnational firms, and to compete in export markets, speeding
proliferation. Ironically, then, economic realities may force a rethinking of security policy. Citizens
of all countries—both weapons producers and weapons buyers—deserve full debate on the econom-
ic and security consequences of transnational military industrial alliances before they become a fait
accompli.
damaging the industry’s ability to design and build weapons and without unduly eliminating cost-
containing competition (Gansler 1995). For some, the challenge was also to preserve the govern-
ment’s historic ability to ensure American civilian technological leadership through a quiet industrial
policy at the Pentagon (Alic, Branscomb. et al. 1992). For yet others, the strategic sea change offered
the possibility for large-scale conversion of redundant military industrial capabilities and a definitive
shift to a civilian-led technology policy (Markusen and Yudken 1992; Stowsky 1999). The peace
dividend was also seen as an opportunity to fund infrastructure, education, and social spending
priorities that had taken a back-burner to the Reagan military build-up (Bischak 1991).
Table 1
Military Expenditures, 1985, 1990, 1996
Selected countries (Index, 1996 = 100)
Responding to the challenge proved difficult, not the least because of conflicting agendas.
Even within the Pentagon, controversy arose over the distribution of spending cuts. One option was
to shield research and development contracts from the budget knife while cutting operations and pro-
duction contracts. But military leaders concerned with readiness (i.e., funds for personnel and opera-
tions) fought the modernizers, who pressed for spending on new weapons systems. Meanwhile, the
defense industry pushed for procurement contracts (especially for existing systems), closing bases,
and privatizing depot and other functions. Peace groups, environmentalists, and other constituencies
pressed for lower defense budgets altogether, citing studies that suggested the United States could
mount a credible defense for only a fraction of current spending (Forsberg 1992; Bischak 1999).
28
Those defending military budgets found it impossible to hold the line on Reagan era spend-
ing. It was hard to defend new R&D spending, slated mainly for Cold War systems, as it became
clearer that security policy was itself in question. Threats are more diffuse, while terrorism, biolog-
ical and chemical warfare, and nuclear proliferation pose greater risks than those anticipated in cur-
rent modernization plans. Then, too, negotiated settlements, peacekeeping, and preventive strategies
have emerged as important new foreign policy activities, with quite different implications for
defense policy and thus for procurement. The upshot has been that American military R&D funding,
while less hard hit than orders for existing weapons systems, fell 26 percent in real terms since 1987.
The significance of overall military spending cuts of nearly 40 percent in the United States is still
quite controversial. The industry laments procurement cuts of 60 percent or more, but critics point
out that procurement in real terms is still higher than in the cold war 1970s and accounts for a larger
share of the defense dollar now. Furthermore, they argue, the need for high levels of defense spend-
ing has diminished dramatically.
In the 1980s, when defense dollars were easy to come by, the Pentagon had little reason to
fret about industrial structure and performance. But faced with the squeeze from binding budgets
on the one hand and concern with readiness on the other, the Pentagon had to redouble its efforts
to get the best return on its shrinking modernization dollar. It crafted three new defense industrial
base approaches in the 1990s—a dual-use strategy to break down barriers between military and
civilian sectors, a merger policy to right-size the industry, and a liberalized arms export policy. What
did each of these achieve?
The Bush administration instituted procurement reforms in this vein, and President Clinton
accelerated them, also funding a $2 billion plus Technology Reinvestment Program (TRP) with
grants to firms who would move technologies either way across the border between defense and
civilian sectors. Procurement reform has proceeded slowly. It confronts a deep structural problem:
because innovation is the basis of military superiority, a fair degree of secrecy and close Pentagon
oversight over research, design, and production of leading edge weapons will remain essential, dis-
couraging civilian/military integration, especially at the systems integration level. But the Pentagon
is buying more off-the-shelf commercial components. The TRP, despite a promising start, was cut
heavily by the Republican Congress elected in 1994, as part of its general repudiation of technology
programs and it disappeared in the FY1997 budget request, to be replaced by the Dual-Use Applica-
tions Program (DUAP) (Stowsky 1999; Oden, Bischak, and Klock-Evans 1995).
The merger wave undercut incipient diversification efforts. Companies had begun using
accumulated cash reserves and defense-bred technologies to enter commercial markets, a develop-
ment anticipated in the trade press and congruent with dual-use policies (Oden 1999). Diversifying
companies like Hughes, Raytheon, Rockwell, Texas Instruments, and TRW were subsequently pres-
sured to spin off their military from their civilian operations and/or to merge, and by 1997 all but
TRW capitulated. The preoccupation with finding mates and consolidating marriages crowded out
diversification, as retained earnings were diverted to cover purchases prices and retire new merger-
related debt. Wall Street investment banking firms played an important initiating role in this process,
as they will in the international acquisitions now in the planning stages (Markusen 1998).
The consolidation slide was greased with Pentagon dollars. In an unprecedented policy deci-
sion, the Pentagon permitted contractors to write off their costs of realizing the mergers, plus a rate
of return, against current contracts, on the presumption they would save the government money in
the future. Although the GAO has confirmed savings in some cases of completed mergers, these
have been far below the levels promised and in the aggregate, below the threshold allowed by Con-
gress (U.S. General Accounting Office 1995).
Surprisingly little capacity was actually retired in the wake of the mergers. MIT’s Harvey
Sapolsky and Eugene Gholz, in a careful census of actual production lines, confirm the retention of
most of them, albeit at reduced levels of capacity utilization (Sapolsky and Gholz 1999). University
of Texas economist Michael Oden has shown that most mergers were of the market-extension type,
in which the largest firms expanded their portfolios of offerings to the Pentagon. Most of the savings
resulted from real estate liquidation and layoffs of relatively well-paid, unionized workers, some of
whom were replaced by lower-paid workers in the subcontracting sector and others by relocation
to cheaper, more military-friendly Congressional districts in the South and Southwest (Oden 1999).
Disappointed with progress on dual-use policy and alarmed at the persistence of defense-
specialized giants and the diminution in their numbers, Defense Secretary William Cohen and
Acquisitions Chief Gansler sharply reversed the merger policy in 1997. Although late in the game,
they opposed the absorption of Northrop-Grumman by Lockheed-Martin, on grounds it would
eliminate effective competition in a number of weapons lines. Stunned, the industry protested, but
to no avail. Persistent opposition convinced Lockheed-Martin to drop its merger bid in July of 1998,
ending five years of industry-altering mergers and leaving four large system-integrating defense
contractors in the American market (Figure 1).
31
played a supporting role. Early in the decade, the industry leaders argued they needed to export more
units of existing weapons systems to compensate for shrinking domestic orders. International sales
would help lower the cost of each fighter craft, communications satellite, or missile to the American
military, saving taxpayers money and/or releasing funds for military modernization. They would
also keep production lines “hot” in lieu of domestic orders. The industry pressed for an arms export
policy which would explicitly take defense industrial base issues into account.
The Clinton administration was sympathetic. Election year competition between Bush and
Clinton had already scotched promising conventional arms export control talks. Bush, under pres-
sure from Clinton, overruled the State Department and approved the sale of F-16s to Taiwan, a move
that angered the Chinese. Arms control advocates began to hear from State Department officials that
although certain sales might be problematic on other grounds, economic factors justified them. Early
in 1995 the White House issued two policy briefs stating that industrial base concerns would hence-
forth be weighed in the permit process for arms exports (1995a, 1995b). Liberalized exports to coun-
tries like Indonesia and more recently, Latin America, are credited to this policy shift. President
Clinton appointed an Arms Export Task Force chaired by Brookings’ Janne Nolan, but ignored its
emphatic conclusion that economic concerns should not govern export policy. An all-out competi-
tion among allies to serve remaining growth markets—chiefly the Middle East and East Asia—
ensued.
The Clinton administration backed its new mercantilist arms export policy with hefty finan-
cial commitments, as did its major competitors—Russia, France, the UK, and Germany. It redoubled
its efforts to market American weapons through its trade attachés and armed forces participation in
military air shows around the world. In a move that undercut the cost savings rationale, it eliminated
fees on arms sales that had previously recouped the development costs footed by the American
taxpayer. The World Policy Institute’s William Hartung (1996) estimates that annual American arms
export subsidies reached into the billions. United Nation’s economist David Gold (1999) concludes
that on economic grounds alone the gains from arms sales do not appear to justify the public costs
of promoting them. The enthusiasm for arms exports as an industrial base policy is perplexing,
especially when the Pentagon’s own study (U.S. Department of Defense 1994) showed that arms
were unlikely to make much of a dent in American excess capacity. Indeed, U.S. arms exports fell
by 10 percent in real terms between 1989 and 1996 even though U.S. contractors’ world market
share rose by nearly 50 percent (Table 2).
34
Table 2
Arms Exports: Britain, France, U.S.
millions of 1990$, 1989-1996
The American liberalized arms export policy is costly and contradictory. On the one hand,
it does not seem to have produced the economic results promised. It has kept production lines open,
as in the F-16, but the United States is not going to be ordering any more anyway. It has produced
and retained few jobs, since to sell overseas, defense contractors must agree to offsets sometimes
as high as 100 to 150 percent. These offsets take the form of either final assembly or component pro-
duction in the buyer country, or they displace American workers in other sectors. Wisconsin Senator
Russell Feingold was outraged to find that to meet its offset commitments McDonnell Douglas was
marketing European paper at a discount, forcing a Wisconsin firm into bankruptcy. The export pol-
icy enabled defense contractors to make higher profits in the 1990s, unprecedented in a period of
deep military spending cuts (Trevino and Higgs 1992; Pint and Schmidt 1994; Oden 1999).
On the other hand, liberalized exports have undoubtedly decreased national security. Sophis-
ticated weapons are now in the hands of some regimes that may prove to be unstable in the longer
run—Saudi Arabia, Egypt, Indonesia among others, as well as those to whom the French, Russians,
and Chinese have sold under the worldwide free-for-all. And they have created pressures for costly
new weapons innovation. The Federation of American Scientists’ Lora Lumpe (1999) has shown
how Lockheed-Martin used the possession of F-16s by potentially unstable regimes as a rationale
35
in lobbying for the F-22. American arms export policy, in other words, has set off a parody of the
Cold War arms race in which the nation is engaged in an arms race with itself.
Christian Democrat Helmut Kohl recommitted in 1998 to joint development of the pricey
Eurofighter, an aircraft analysts believe will be inferior to American offerings.
Yet European firms have no viable choice but to team up with someone. European nations
cut procurement spending abruptly and deeply in the 1990s. With budgets already much smaller than
that of the U.S.—Britain, France and Germany spend 6 percent, 12 percent and 23 percent respec-
tively of the American level—the Europeans know they must abandon their autarkic defense indus-
tries. Many already buy equipment from the United States and from each other. Europeans viewed
the American mergers, which exacerbated the trans-Atlantic scale gap considerably (Table 3), with
alarm and in 1998 redoubled their efforts to create a pan-European industry.
Table 3.
Major Defense Contractors by Nation, Sales 1995
But intra-European partnerships have proved difficult. The French, with more of their
science and engineering workforce tied up in defense sectors than any other country, are reluctant,
despite palpable failure, to give up a decades-long industrial policy grounded on defense innovation.
37
Tiring of waiting for the French, British firms began to partner with Americans; British Aerospace,
for instance, has paired up with Lockheed-Martin on the Joint Strike Fighter. The disarray in the
European military industrial complex beckons to American firms, and some European officials
believe this may be the only way to overhaul a poorly configured and highly inefficient defense
industrial base.
The impulse towards a transnational industry is unlikely to wane, and it will reach beyond
the United States and Europe. American firms already help the Taiwanese, Koreans, and Turks
assemble fighter jets, and they partner with Japanese firms on important high-tech components.
Lockheed Martin has bought up the remains of the Argentinean aircraft industry, transforming it into
maintenance facilities for a military fleet comprised of American craft, and Russian capacity is being
selectively bought out by U.S., European, and Japanese firms for commercial as well as military
purposes. Although design and innovation will remain the province of the rich countries able to pay
for it—principally the United States and Europe—military production will increasingly replicate the
international division of labor of commercial sectors.
weapons procurement may serve as a convenient means for some buying regimes to command and
control a hefty increment of jobs within their own economies, used to political advantage. (When
asked recently why Chile, friendly to its neighbors, would want to buy expensive new fighter air-
craft, the Chilean Finance Minister is reputed to have exclaimed, “because we have the money!”)
The result will be a patchwork of specialization spread across the globe. A future generation
fighter, for instance, might be designed in the United States, its prototype built in Britain or Italy,
and the first units tested in France. Once in production, it might be assembled in the United States,
Turkey, Korea, and Taiwan, as the F-16 now is, with unique high-tech componentry from Japan,
Germany, Russia, and Israel and more cost-sensitive and commercially available components from
Spain, Poland, Brazil, and South Africa. A world weapon could look a lot like a world car on a wall
map, but its implications are far, far different.
A Lopsided Market?
Despite the fact that buying nations will be able to extract jobs and chunks of weapons-
producing capacity with every weapons purchase, the implosion in the number of major weapons
makers and their increasingly international orientation constitutes a shift in the balance of power in
the arms market, away from governments and toward contractors. Less than a decade ago, the U.S.
Pentagon, for instance, could count on three or more firms to compete in most weapons lines. Today,
only two domestic competitors survive in most large weapons lines, and in some, only one. Eco-
nomic theory predicts that heightened oligopolistic power on the supply side will curtail innovation
and raise prices above actual costs, concerns that induced Secretary William Cohen and his acquisi-
tions chief Jacques Gansler to hold firm on their opposition to the Lockheed/Martin Northrop/Grum-
man merger. One solution would be to buy from the Europeans, an option that Gansler cautiously
favored in his book, Defense Conversion (1995). Buying foreign would reverse decades of “buy
America” habits and would undoubtedly attract strong Congressional opposition. Furthermore, trans-
national mergers and partnerships would whittle down this option.
For sellers in the arms market, meanwhile, things are improving structurally—there are now
more buyers, and any one country, the United States included, accounts for a smaller portion of the
big contractors’ sales. To be sure, the Pentagon is still by far the single largest customer for Ameri-
can contractors, as their own governments are for European firms. The share of exports in American
firms’ output rose only modestly in the past decade, from 14 percent in 1989 to 18 percent in 1997
39
(Table 2). But if transatlantic mergers occur on anything like the scale of domestic mergers in the
1990s, we could witness an altered market in which a relatively few international firms sell to
dozens of major buying nations.
Under these circumstances, the ability of any one government, even the United States, to
develop weapons appropriate to its security needs, enjoy technological superiority, or limit the diffu-
sion of technology would be greatly curtailed. Other nations have suffered under these circum-
stances for some time. The French Air Force, for instance, was impelled to compromise on its fighter
jet designs to satisfy Middle Eastern customers, whose purchases were needed to cover development
costs (Kolodziej 1987). The U.S. Air Force is now facing the same prospect. For its next-generation
fighters, it has been asked to specify which other countries might be expected to buy the craft and
whether those countries’ defense needs might require alterations in the design.
Private Arsenals?
The trend toward private sector provisioning of arms poses new problems that impinge on
national security and require creative solutions, especially as the defense industry goes global. Out-
sourcing has made slow but steady inroads on public arsenals, from the rise of the private naval
shipbuilder Vickers in the 1880s, described in Mary Kaldor’s The Baroque Arsenal (1981), to the
momentous decision of the incipient American Air Force to rely on private aircraft companies rather
than government facilities for the fighters, bombers, transports, and ballistic missiles that trans-
formed twentieth century warfare. A larger share of the military dollar now goes for procurement
from private sector firms (and less for military manpower, bases, and DOD civilian employees) than
the Cold War average (Figure 2). This shift reflects the rising significance of remotely-delivered de-
structive power, reliant on aircraft carriers, fighters, bombers, missiles, precision-guided munitions,
and the paraphernalia required to gather intelligence, communicate it, coordinate decisionmaking,
and provide for battlefield management from afar (Markusen and Yudken 1992, ch. 2).
There were good reasons for going this route in the past. The competition among youthful
aircraft companies served the Army Air Corps and subsequent Air Force well, as they churned out
a wide array of designs that either stood or failed the market test. The relative superiority of the Air
Force ballistic missile program in the 1950s, spread among competing firms, over the Army’s in-
house effort at Redstone Arsenal confirmed the advantages of competition and an arms-length rela-
tionship between a public sector buyer and private sector suppliers (Markusen, Hall et al. 1991).
40
But as the number of suppliers shrinks, the Pentagon and its European counterparts may be
confronting what Sapolsky and Gholz (1999) call “private arsenals”—huge firms with more or less
monopoly positions in various weapons lines. As limited liability corporations, these organizations
operate with entirely different motivations than do the armed forces, the Pentagon, or public sector
agencies in general. Their loyalties are first and foremost to shareholders, whose priorities increas-
ingly stress short-term returns (Weber 1998).
Governments have always relied on two features to keep private sector contractors in line:
competition among firms, and regulation to curb potential and punish actual abuses (McNaugher
1989). But today domestic competition is considerably diminished, and regulation is under attack,
as budget-conscious legislators and privatization advocates favor a smaller, more business-friendly
Pentagon. As contractors become transnational and form international oligopolies, the oversight
challenge will become considerably more difficult. There is no substitute for well-informed, vigilant
government oversight of the private sector defense industry and for a coordinated policy that antici-
pates and shapes international restructuring of firms.
450
400
350
Military Pay
300
250
200
150
Military Purchase
100
50
0
70 7 2 7 4 76 7 8 8 0 82 8 4 86 8 8 9 0 92 9 4 9 6 98 0 0
1 9 19 19 19 19 19 19 19 1 9 19 19 1 9 19 19 1 9 20
hold in the longer run. Dual-use technological applications have already made national borders quite
permeable, and a number of important partnerships and acquisitions have already taken place.
Furthermore, fiscal pressures clash with the extravagant cost of the Two Fortress outcome. Restive
under the budget knife, procurement chiefs in many countries have strong incentives to buy, from
whomever, the best weapon at the most affordable price. The British under Margaret Thatcher did
just this in the 1980s, cutting military spending faster, with less debilitating effects on readiness,
than other Cold War protagonists.
We are apt to see a transnational industry develop regardless of government reticence. Of
the security issues posed, the third—proliferation of conventional weapons—is by far the most wor-
risome to the world community, and the fourth—loss of technological leadership—serious enough
to induce countries to rethink their security strategies. To date, defense contractors principally
design and make weapons within their home countries’ borders and sell some portion of them
abroad, affording their respective ministries of State and Defense ample opportunity to monitor the
appropriateness of foreign sales. But as transnational or allied firms, they will increasingly allocate
their design, development, and production activities among nations in response to buyers’ demands
for offsets. Personnel, ideas, and technologies will move more freely from one corporate unit to
another heedless of national borders. Under these circumstances, governments will have a more
difficult time monitoring cost, pricing, and proliferation. But their role as overseers will be more
important than ever.
Such fluidity would also undermine security policies rooted in the possession of a clear tech-
nological lead in weapons and delivery systems. For purely technical reasons, maintaining techno-
logical superiority is increasingly difficult, because the edge is shifting away from platforms per se
towards sophisticated electronics, communications, and guidance capabilities, heavily rooted in
“dual-use” sectors of the economy. While arms control advocates contend that it is still possible to
slow the spread of these technologies, economists increasingly argue that it is impossible to do so
and oppose efforts to restrain dual-use exports. If transnational firms supplant domestic firms, it will
be just that more difficult to ensure pre-eminence, even though certain governments—the United
States, the Europeans—will remain the largest investors in military R&D.
How would a transatlantic defense industrial base strategy work? First, initiating nations—at
a minimum the United States, Britain, France, and Germany—could agree to create a common data
and knowledge base about their defense industries, firms, and production lines. This would include
43
analyses of existing and potential economies of scale, an evaluation of excess capacity, and profiles
of the technical and business strengths and weaknesses of major contractors. This may sound “ho-
hum,” but it is amazing how little is known about the shape of the defense industrial base, even in
the most sophisticated and “open” nations (cf. Kudrle and Bobrow 1998).
Second, these nations along with other suppliers would jointly develop an industrial base
strategy that would ensure sufficient competition to keep prices close to costs but ensure innovation
where it is desirable. The strategy would distinguish between desirable and unacceptable transna-
tional partnerships among firms, and it would target particular segments for closing. In imple-
menting the latter, partner states would chip in to support worker adjustment programs and alterna-
tive economic development strategies for exiting sectors and regions, much the way the EU Konver
program has done for European military base and shipyard closings.
Third, governments would engage in joint procurement programs where domestic and trans-
national firms are invited to compete and where R&D in particular is jointly agreed upon, with
provisions for disseminating the results so that additional firms might compete for production
contracts. When buying already developed systems, nations would be guaranteed a modicum of eco-
nomic activity corresponding to their level of ongoing financial and military commitment, though
this need not be in the defense industry. Governments, rather than firms, then, would negotiate offset
policy.
Such a transnational industrial base strategy would work best if it were linked to a coopera-
tive security policy along the lines suggested by Kaufman, Steinbruner, and Forsberg (Steinbruner
1994; Kaufman and Steinbruner 1991; Forsberg 1992). Indeed, the pursuit of unilateral military
strategy and the ability to count on one’s own military superiority seem incompatible with an evolv-
ing transnational industry. A clear and shared vision of the nature of contemporary threats, appro-
priate military response, burden-sharing in readiness and war, stabilizing forms of deterrence in the
conventional weapons sphere, effective arms export restraint, and “fair sharing” of the costs of mili-
tary industrial downsizing would facilitate the fashioning of an affordable and efficient defense
industrial base.
It would be foolish to underestimate the obstacles facing such a coordinated industrial base
strategy: parochialism, legitimate fears of loss national sovereignty, unwillingness to countenance
job loss, and high start-up costs, to name a few. But anything short of this will be suboptimal on both
economic and security fronts. The United States could lead such an effort, as the world’s largest
44
weapons buyer and major exporter. America’s lively history of relatively transparent, arms length
relationships between contractors and the Pentagon is much admired in Europe. So is its historic
commitment to competition among multiple suppliers for designing and producing weapons systems.
And American data on defense contracts and arms exports are better than most. A relatively more
peaceful albeit confusing security environment, fiscal austerity, and a troubled world economy
should propel allies into partnering on military industrial base strategy. The payoff would be
substantial.
References Cited
Alic, John; Lewis Branscomb, Harvey Brooks, Ashton Carter, and Gerald Epstein. 1992. Beyond
Spinoff: Military and Commercial Technologies in a Changing World. Boston: Harvard
Business School Press.
Bischak, Gregory, ed. 1991. Towards a Peace Economy in the United States: Essays on Military
Industry, Disarmament, and Economic Conversion. New York, St. Martin’s Press.
———. 1999. “Contending Security Doctrines and the Military Industrial Base.” In Arming the
Future: A Defense Industry for the 21st Century, ed. Ann Markusen and Sean Costigan. New
York: Council on Foreign Relations.
Brömmelhürster, Jörn and Ann Markusen. 1999. “Contrasts in Defense Industrial Conversion: Ger-
many and the United States.” Working Paper, Project on Regional and Industrial Economics,
Rutgers University.
Brzoska, Michael, Peter Wilke and Herbert Wulf. 1999. “The Changing Civil-Military Production
Mix in Western Europe’s Defense Industry.” In Arming the Future: A Defense Industry for
the 21st Century, ed. Ann Markusen and Sean Costigan. New York: Council on Foreign
Relations.
Coffman, Vance. 1998. “The Future of Transatlantic Industrial Partnerships.” Remarks delivered
at the 1998 Wehrkunde Conference, Munich, Germany, February 7. Mimeo, Lockheed
Martin Corporation.
Dowdy, John. 1997. “Winners and Losers in the Arms Industry Downturn.” Foreign Policy 107
(Summer): 88-101.
Flamm, Kenneth. 1999. “Redesigning the Defense Industrial Base.” In Arming the Future: A
Defense Industry for the 21st Century, ed. Ann Markusen and Sean Costigan. New York:
Council on Foreign Relations.
Forsberg, Randall. 1992. “Defense Cuts and Cooperative Security in the Post-Cold War World.”
Boston Review 17, no. 3-4 (May-August): 5-9.
Forsberg, Randall, ed. 1994. The Arms Production Dilemma: Contraction and Restraint in the
World Combat Aircraft Industry. Cambridge: MIT Press.
Gansler, Jacques. 1995. Defense Conversion: Transforming the Arsenal of Democracy. Cambridge:
MIT Press.
Gold, David. 1999. “The Changing Economics of the Arms Trade.” In Arming the Future: A
Defense Industry for the 21st Century, ed. Ann Markusen and Sean Costigan. New York:
Council on Foreign Relations.
45
Gonchar, Ksenia, Yevgeny Kuznetsov and Alexander Ozhegov. 1995. Conversion of the Post-Soviet
Defense Industry: Implications for Russian Economic Development. Bonn: Bonn Interna-
tional Center for Conversion.
Hartung, William D. 1996. Welfare for Weapons Dealers: The Hidden Costs of the Arms Trade.
New York: World Policy Institute.
Kaldor, Mary. 1981. The Baroque Arsenal. New York: Hill and Wang.
Kaufmann, William and John Steinbruner. 1991. Decisions for Defense: Prospects for a New Order.
Washington, DC: Brookings Institution.
Kolodziej, Edward. 1987. Making and Marketing Arms: The French Experience and Its Implications
for the International System. Princeton: Princeton University Press.
Korb, Lawrence. 1996. “Merger Mania.” Brookings Review (Summer): 22-25.
Kovacic, William and Dennis Smallwood. 1994. “Competition Policy, Rivalries, and Defense
Industry Consolidation.” Journal of Economic Perspectives 8, no. 4: 91-110.
Kudrle, Robert T. and Davis B. Bobrow. 1998. “Foreign Acquisition of Defense-Related U.S. Firms:
Concentration, Competition and Reality.” In Foreign Ownership and the Consequences of
Direct Investment in the United States: Beyond Us and Them, ed. Douglas Nigh and Douglas
Woodward. Westport, CT: Quorum Press.
Kurth, James. 1973. “Aerospace Production Lines and American Defense Spending.” In Testing the
Theory of the Military-Industrial Complex, ed. Steven Rosen. Lexington, MA: D.C. Heath.
———. 1993. “The Follow-on Imperative in American Weapons Procurement, 1960-1990.” In Eco-
nomic Issues of Disarmament: Contributions from Peace Economics and Peace Science, ed.
Jurgen Brauer and Manas Chatterji. New York: New York University Press.
Lovering, John. 1999. “Which Way to Turn? The European Defense Industry after the Cold War.”
In Arming the Future: A Defense Industry for the 21st Century, ed. Ann Markusen and Sean
Costigan. New York: Council on Foreign Relations.
Lumpe, Lora. 1999. “A Framework for Limiting the Negative Consequences of Surplus U.S. Arms
Production and Trading.” In Arming the Future: A Defense Industry for the 21st Century,
ed. Ann Markusen and Sean Costigan. New York: Council on Foreign Relations.
Markusen, Ann. 1998. “The Post Cold War Persistence of Defense Specialized Firms.” In The
Defense Industry in the Post-Cold War Era: Corporate Strategies and Public Policy Per-
spectives, ed. Gerald Susman. Oxford: Elsevier.
Markusen, Ann, Peter Hall, Scott Campbell and Sabina Deitrick. 1991. The Rise of the Gunbelt: The
Military Remapping of Industrial America. New York: Oxford University Press.
Markusen, Ann and Claude Serfati. 2000. “Remaking the Military Industrial Relationship: A
French-American Comparison.” Defence and Peace Economics 11, no. 3: 1-29.
Markusen, Ann and Joel Yudken. 1992. Dismantling the Cold War Economy. New York: Basic
Books.
McNaugher, Thomas. 1989. New Weapons, Old Politics. Washington, DC: Brookings Institution.
Moran, Ted. 1993. “Foreign Acquisition of Critical U.S. Industries: Where Should the United States
Draw the Lines?” Washington Quarterly 16, no. 2 (Spring): 61-71.
Muradian, Vago. 1997. “Kaminski Institutes Vertical Integration Warning System.” Defense Daily
195, no. 27 (July 5).
Oden, Michael. 1999. “Cashing In, Cashing Out, and Converting: Restructuring of the Defense
Industrial Base in the 1990s.” In Arming the Future: A Defense Industry for the 21st
Century, ed. Ann Markusen and Sean Costigan. New York: Council on Foreign Relations.
46
Oden, Michael, Gregory Bischak and Christine Evans-Klock. 1995. “The Technology Reinvestment
Project: The Limits of Dual-Use Technology Policy.” New Brunswick, NJ: Rutgers Univer-
sity, Project on Regional and Industrial Economics.
Pages, Erik. 1996. Responding to Defense Dependence: Policy Ideas and the American Defense
Industrial Base. Westport, CT: Praeger.
Pint, Ellen and Rachel Schmidt. 1994. Financial Condition of U.S. Military Aircraft Prime Con-
tractors. Santa Monica: Rand, Project Air Force.
Sapolsky, Harvey and Eugene Gholz. 1999. “Private Arsenals: America’s Post-Cold War Burden.”
In Arming the Future: A Defense Industry for the 21st Century, ed. Ann Markusen and Sean
Costigan. New York: Council on Foreign Relations.
John Steinbruner. 1994. “Weapons Production and Force Structure Planning as a Problem of Collab-
oration.” Paper prepared for a Brookings US-Russian Conference on Arms Production and
Trade, February 5.
Stowsky, Jay. 1999. “The History and Politics of the Pentagon’s Dual-Use Strategy.” In Arming the
Future: A Defense Industry for the 21st Century, ed. Ann Markusen and Sean Costigan. New
York: Council on Foreign Relations.
Trevino, Ruben and Robert Higgs. 1992. “Profits of US Defense Contractors.” Defense Economics
3, no. 3: 211-18.
United States General Accounting Office. 1995. Defense Downsizing: Selective Contractors
Business Unit Reactions. GAO/NSIAD-95-144, May. Washington, DC: US Government
Printing Office.
U.S. Department of Defense, Office of the Undersecretary of Defense (Acquisition and
Technology). 1994. World-Wide Conventional Arms Trade (1994-2000): A Forecast and
Analysis. Washington, DC: Department of Defense.
U.S. Department of Defense, Office of the Secretary of Defense. 1997. Final Report of the Defense
Science Board Task Force on Vertical Integration and Supplier Decisions. Washington, DC:
Office of the Secretary of Defense. May.
Weber, Rachel. 1998. “The State as Corporate Stakeholder: Governing the Decline of the Military-
Industrial Complex.” Ph.D. dissertation, Cornell University.
White House, Office of the Press Secretary. 1995a. “Conventional Arms Transfer Policy.” Fact
sheet, February 17.
White House, Office of the Press Secretary. 1995b. “Criteria for Decisionmaking on US Arms
Exports.” Fact sheet, February 17.
II. THE PLACE OF THE DEFENSE INDUSTRY
IN ADVANCED INDUSTRIAL ECONOMIES
Kenneth Flamm
During the last 50 years, the single largest category of discretionary spending for many West-
ern governments was the national defense. That allocation came about in the course of the Cold War
in a context in which rapid technical development and immediate preparedness were considered
more important than economic efficiency. Redundant capabilities and inefficient practices were
tolerated. The analysis of efficient defense production was not intensively pursued.
In the very different circumstances that now prevail, the industrial infrastructure (the
“defense industrial base”) that currently builds the systems used by Western military establishments
is in the midst of a painful transition. In the United States, deliveries of major defense systems are
currently, in real terms, about sixty percent of the mid-1980s cold war peak.1 A determined push to
reform the acquisition process is slowly gaining momentum. If successful it will lower barriers that
have sheltered those willing to invest in mastering the very arcane rules of the defense acquisition
process. In technologies once driven by defense, commercial applications now lead the process of
technical innovation. Recent American policy has been encouraging the use of commercial technol-
ogies and products in defense systems wherever possible. This initiative promises to erode the
boundaries between defense and commercial production even more and to open an already shrinking
market to a larger set of commercial competitors.
In this new environment, major investments in systems and technology may be developed
and demonstrated in the form of prototypes. They will have to be produced and deployed on a lim-
1
See Kenneth Flamm, “Defense Industry Consolidation in the 1990s,” in G. Susman and Sean
O’Keefe, eds., The Defense Industry in the Post-Cold War Era, (New York: Pergamon, 1998) for
a more thorough discussion of real spending levels. Unless otherwise noted, all statements made in
this chapter are drawn from a research project that will be published by the Brookings Institution.
47
48
ited scale to work out operational issues. Even where it is clear that new large scale weapons plat-
forms must be procured, the production and fielding of such systems may stretch out over years,
even decades, into the future.
The U.S. defense industry has some formidable strengths as it faces these necessary adjust-
ments. It is the premier global producer of a variety of defense systems, the clear world leader in key
systems integration skills and specialized technologies. It accounts for roughly half of world exports
of military equipment. Its domestic market is by far the largest in the world—more than triple that
of Japan which has the second largest procurement budget among the Western allies. The United
States’ weapons procurement budget exceeds the combined total of all its European NATO allies.
As a consequence, the formidable volume of U.S. military equipment exports amounts to only about
20 percent of procurement by the Department of Defense (DOD), and in few cases is the fundamen-
tal viability of a U.S. defense industry critically dependent on success in export markets. Further-
more, despite a continuing decline in military budgets around the globe, U.S. foreign defense sales
have remained roughly constant, and U.S. export market share has therefore increased.
The same cannot be said for the U.S. allies’ defense industries. Japanese companies are pro-
hibited from exporting military equipment, and flat defense budgets—combined with uneconomic
volumes and fantastically high costs—have thrust Japan’s defense industry into a deepening crisis.
In response, both Japan’s industry and its promoters within the Ministry of International Trade and
Industry (which controls defense procurement) have begun to publicly advocate a controversial
legalization of military exports. European governments have long recognized the importance of ex-
ports to the survival of their defense industries: in France, for example, exports exceeded 40 percent
of arms industry sales in the mid-1980s. Despite aggressive promotion efforts, European export sales
dropped with the continued decline in global defense spending (in France, by the early 1990s, to 25
percent of a falling industry output), and European defense industries today are in a state of turmoil.
Responses to these developments—initially championed by France, and supported strongly by Ger-
many—have included the formation of a Western European Armaments Group under the political
aegis of the Western European Union, the first embryonic steps toward a European defense procure-
ment agency, proposals to restrict European defense procurement to European-only suppliers when-
ever possible, and a renewed emphasis on export promotion. But despite repeated public calls for
consolidation of a fragmented European defense industry into a smaller number of producers
49
(responding to a major consolidation that has occurred with U.S. defense industry), until recently
there had only been slight movement toward real industrial restructuring in Europe.2
For the United States, in contrast, the issue has not been industrial survival, but how and
where to reduce the industrial capability primarily dedicated to defense to a level appropriate to a
new, post–cold war force structure. Despite an ebbing wave of mergers and acquisitions in Amer-
ica’s defense industries, existing capacity in the current defense industrial base may still exceed the
investment requirements of tomorrow’s military forces. What those investment requirements are,
and what set of industrial capabilities are needed to meet them, is, unfortunately, a complex set of
issues that so far has largely defied a crisp conceptualization or a clear answer.
With defense downsizing in full swing around the globe, every major producer of high-tech
armaments other than the United States now faces a virtual economic crisis in its defense industries.
The economics of defense production, this chapter will argue, are dominated by various flavors of
economies of scale—in assembling and sustaining essential design capabilities, in systems R&D,
in start-up costs, in lumpy production capacity, in learning curves. This means that unless they are
willing to give up maintenance of a national capability to produce advanced military systems as a
national security objective (exceedingly unlikely), America’s allies will be pushed to (a) close off
their national markets to foreign-built systems and (b) dramatically increase exports. Both develop-
ments would raise significant problems within the coalition of allies built during the Cold War.
Thus economic pressures seem destined to push the defense industry against restraints on
the proliferation of advanced conventional military capabilities. In many cases, the best customers
for sophisticated capabilities will be in precisely those regions where sales are likely to feed existing
tensions, where old disputes may blossom into new wars. In other cases, now dormant rivalries
could be exacerbated by the aggressive salesmanship of strapped companies fighting to stay com-
petitive.
If higher walls around national defense markets are a component of the response to these
economic pressures, a different but equally painful source of divisions within the alliance will
emerge. The allies are still highly dependent on the liberal transfer of American defense technology
2
The 1999 merger of British Aerospace and Marconi Electronic Systems into BAE Systems and the
projected establishment of the European Aeronautic, Defense and Space Co.(EADS) in June 2000
from Aerospatiale Matra, DASA and CASA have changed the European picture dramatically. These
two large firms will rank among the five largest defense firms in the world.
50
to their defense industries. Barred from access to foreign markets, the American taxpayer—who
ultimately foots the bill for these technology investments—is likely to ask why a policy of easy tech-
nology transfer is being continued, and express those doubts through the U.S. political system.
Finally, if economic pressures spawn easy access to advanced military capabilities by mar-
ginal customers, the United States is certain to press even harder to accelerate its “Revolution in
Military Affairs” (RMA) to maintain its current technological advantage over potential adversaries.
Large U.S. investments in new technology are already beginning to spur talk that a “technology gap”
between the U.S. and its allies exists, and threatens the interoperability of American forces with
those of its friends.3 The likely response is greater pressure on the allies to buy advanced information
and sensor systems like those fielded by the Americans. Given the enormous costs of developing
these systems, already tight European and Japanese defense budgets are unlikely to have room for
either domestic production of compatible indigenous versions or purchase of American equipment.
The predictable outcome of these developments would seem to be rising dissent within the
Alliance. While it might not be sufficient to rupture the fabric of our security ties, it will certainly
stress them and leave us less prepared to deal swiftly and decisively with crises as they develop.
This chapter proposes to dissect these economic forces in some detail, discuss their implica-
tions for Alliance security, and examine some possible responses. We first look at the dynamics of
changes in defense procurement budget. We then define the concept of a “defense industry,” and
examine the contours of defense industry in the United States. Looking at the economics of design-
ing and building an advanced fighter aircraft illuminates the economic pressures driving consolida-
tion within the defense industry. A final section of the chapter analyzes likely outcomes and possible
solutions.
3
See Bryan Bender, “U.S. Worried by Coalition ‘Technology Gap’,” Jane’s Defence Weekly (29
July 1998). The problems NATO faced in Kosovo have increased these worries. See John D.
Morrocco, “Transatlantic Links a Crucial but Elusive Goal,” Aviation Week & Space Technology
150 (24 May 1999): 28-29.
51
that there is excess capacity in the industry, and that—given pervasive scale economies—this excess
capacity is manifested in too many firms competing for too few defense dollars. Therefore, recent
consolidation within the U.S. defense industry has been a much-needed step toward rationalization.
At first glance this is a plausible chain of logic. But when scrutinized more closely, there are
a number of points where the assertion is misleading, if not incorrect. To begin, while procurement
appropriations may indeed have declined by close to 70 percent from a Cold War peak, procurement
outlays—disbursements of those funds to industry, which may take a decade or more, smoothing
out budget peaks and valleys—has declined somewhat under 60 percent. Even this may be a mis-
leading figure. The Defense Department procures all sorts of goods and services, from paper clips
and toilet seats to fighter jets and tanks. Arguably, when discussing the condition of the defense
industrial base, we are less concerned about the paper clips and more concerned with the fighter jets
—and other specialized, defense-unique products and systems that cannot be quickly and easily pro-
cured from commercial suppliers. DOD investment in military equipment—aircraft, ships, vehicles,
missiles, electronics, and other defense equipment—might be considered a better measure of trends
in government purchases from the industries that we tend to think of as making up the “defense
industrial base.”
The national income accountants at the U.S. Bureau of Economic Analysis have constructed
an index of real spending on defense equipment investment measured in 1992 dollars. This index
shows a substantial drop from a calendar year 1987 peak—about a 40 percent decline. But this is
considerably reduced from the 70 percent drop in procurement budgets with which we started. And
while total procurement outlays sank well below a previous 1976 low before stabilizing around
1995, real equipment spending in 1995 remained at a much higher level—above 1982 levels.
The considerably less steep decline in equipment investment may in part reflect the fact that
sophisticated defense systems take much longer to produce and deliver than the more ordinary goods
and services that DOD also procures. Cutbacks would therefore be stretched out in outlays over a
longer time period, and the industrial impacts felt in a more gradual fashion. But it is also true that
in relative terms, the U.S. military has spared the specialized systems and capabilities it is most con-
cerned about protecting and maintaining—sophisticated, high-tech equipment—from the sharpest
relative cuts in the procurement budget. There was considerable unevenness in the way this overall
40 percent decline was distributed. Some sectors faced large cutbacks, others small adjustments, so
the impact on industry may have been enormous, economically, in some cases. Nonetheless, on aver-
52
age, defense equipment spending in real terms was about where it was in 1982. How does this relate
to the argument for consolidation?
Pressures to consolidate in the U.S. defense industry can be viewed through two alternative
lenses. First, one could think of defense as a business like any other, and argue that competitive pres-
sures force change. Given significant economies of scale in defense systems (based on the econom-
ics of fixed investments in R&D and plant, as well as learning economies), fewer competitors will
be able to “fit” into a smaller market in the long-run and maintain a market rate of return on their
capital investments, as they compete with one another. The “natural” pressures of competition as
the market shrinks will therefore force some out.
The alternative view is that the usual story about competitive markets does not fit defense
well. In the real, cost-plus world of administered pricing of defense contracts by a single large custo-
mer bearing the risk of designing and building products to meet its unique needs, the defense depart-
ment will simply pay whatever overhead costs exist within those defense-oriented firms it chooses
to maintain through the awarding of contracts. Therefore, it is the defense department itself, in its
quest to achieve greater bang per buck of procurement spending, that must create pressure on its
contractor base to reconfigure to reduce overhead. The pressure can be in the form of positive carrots
—like agreements to share overhead savings with contractors when they take steps to reduce costs
through mergers and acquisition—or negative sticks—like letting companies go bankrupt when they
don’t receive sufficient contract awards to keep themselves profitable. In both views of the world,
government will play a key role in determining the defense industry configuration, through both its
policies toward mergers and acquisitions and its contracting practices.
There is little doubt that there was major shift in American policy toward defense industry
mergers when a new crop of top officials entered the Pentagon in 1993. The Bush Administration
had effectively signaled its willingness to block major defense mergers when it torpedoed a big
consolidation deal in the ammunition and ordnance industry in the early 1990s. Under Defense
Secretary William Perry’s leadership in 1993, this course was unequivocally reversed.
But did shrinking budgets really made a new policy toward the defense industry inevitable
in 1992? The data suggest that in 1992 we were at about the same level of real defense output as in
53
the early 1980s. Contrary to what an economist might presume,4 the evidence suggests there was
effectively no increase in the numbers of competing defense producers during the massive Reagan
defense buildup of the 1980s, and even an increase in concentration in a small number of defense
sectors (like aircraft). This raises an important question. If the amount of output and the number of
producers involved in manufacturing it were about the same in 1992 as in 1982, why was a reduction
in the number of companies producing this output newly desirable? Or to put it another way, if the
problem was so apparent in 1992, why was there no pressure to undertake (or at least discuss) con-
solidation and restructuring of the defense industry back in 1982? One logically consistent answer
would be to argue that the industry was facing an imminent crisis back in 1982.5 Only the knowledge
that an incoming Reagan administration was likely to vastly increase defense procurement, it might
be argued, kept restructuring of the defense industry off the policy agenda. The problem with this
argument is that it flies in the face of the facts of profitability in the defense industry. Simple calcu-
lations of returns on equity or investment for aerospace firms6 or more complex methodologies
designed to estimate economic profits in defense firms7 show that the early 1980s were a period of
high levels of profitability for defense companies. Paradoxically, defense contractors’ rates of return
and profitability fell steadily after 1983 through the remainder of the decade, even as defense pro-
curement soared.
A more compelling argument hinges on an increase in the soaring costs of R&D for new,
higher-tech military platforms. A typical fighter aircraft R&D program, for example, ran about $6
4
Based on the notion that, all else being equal, with some fixed entry costs even imperfect competi-
tion will generate larger equilibrium numbers of firms in an industry in response to an increase in
demand.
5
In fact, this was precisely the answer that Bernard Schwartz, former CEO of defense giant Loral
(now for the most part absorbed into Lockheed-Martin) gave when asked this question in October
1997, at a presentation at the Johns Hopkins University’s School of Advanced International Studies.
Schwartz’s prepared remarks, calling for prophylactic policies to discourage further vertical integra-
tion within the defense industry, have been published as “Defense Industry Consolidation: Where
Do We Go From Here?” (brochure, ca. 1998).
6
For example, using data from the Census Bureau’s Quarterly Financial Survey of Manufacturing.
7
See James Lewis Field, “Economic Profit in United States Government Defense Contracting:
Theory and Practice of a Fair and Reasonable Return.” Ph.D. diss., Graduate School of Business
Administration, Harvard University, 1993, pp. 264-68.
54
to $8 billion 1990s vintage dollars in the late 1970s and early 1980s.8 Today, the price tag on R&D
for a new stealth fighter has roughly tripled, to $20 billion or more (current estimates for the F-22
and Joint Strike Fighter, for example).9 This makes it much more expensive to have a large number
of programs. With a decline in the number of new defense programs also comes a reduction in the
number of producers that can be sustained on an ongoing basis. Exploding price tags for investments
in new technology needed to develop advanced platforms—not a reduction in the overall size of the
procurement pie, but an increase in the size of the minimum slice needed for a new system—seem
a more persuasive argument for why 1992 was so different from 1982.
Note, incidentally, that this argument—which focuses on having too many firms shouldering
the fixed costs required to maintain the capability to design and develop a new defense system—is
a little different from what is normally meant, at least by some, when they refer to “excess capacity,”
i.e., underutilized factories and equipment. Data from U.S. guided missile and space vehicle facto-
ries, for example, suggest that over a period in which there was little change in industrial concentra-
tion and output was dropping sharply, defense managers nonetheless removed excess capacity and
sharply improved capacity utilization. Suggestions that no Cold War production lines have been
closed down in the United States are not only wrong on the facts—they also err in implying that
excess production capacity was the main problem for U.S. defense industry efficiency. From the per-
spective of the firm, the real “excess capacity” is the cost of maintaining and sustaining a minimum
critical mass of skilled people with design skills and experience, not buildings and tools.10
Thus conventional wisdom most likely comes to the right conclusion, but for the wrong rea-
sons. It is a massive increase in the R&D investments required to design leading edge military equip-
8
In the 1970s, R&D costs for 3 fighter jets with R&D started in that decade (in FY 91$) were for
the F-16 ($3.3 billion), the F-15 ($8.2 billion), and the F/A-18 ($7.4 billion). See J.A. Drezner, G.K.
Smith, L.E. Horgan, C. Rogers, and R. Schmidt, “Maintaining Future Military Aircraft Design Capa-
bility,” Rand Report R-4199-AF (Santa Monica: Rand, 1992), p. 24.
9
See Congressional Budget Office, A Look At Tomorrow’s Tactical Air Forces (Washington, DC:
Congressional Budget Office, 1997), pp. 83-87.
10
Indeed, much of the physical plant and tooling used in at least some U.S. defense production lines
continues to be owned by the government, supplied to the contractor at little or no cost, and operated
by the contractor. There would seem to be few incentives for a contractor not to use these “free
goods,” even if the number of firms operating these plants were to be reduced.
55
ment (and possibly management overhead), not excess factory capacity, that leads one to focus on
reducing the number of producers.
Whatever its rationale, it is certain that the process of consolidation unleashed in the United
States in the early 1990s has had a profound impact on the contours of industry. Table 1 displays
the reduction of the number of producers supplying selected types of weapons systems from 1990
to 1998. (The numbers in brackets show the number of producers that would have existed had the
proposed Lockheed Martin-Northrop Grumman merger not been blocked.) The numbers are striking:
in tactical missiles, from thirteen to four companies; in fixed wing aircraft, from eight to three firms.
Space launch has gone from six to two, and satellites from eight to five. In tracked combat vehicles,
strategic missiles, and torpedoes, three producers dropped to two. In helicopters, the military has
gone from four to three suppliers, and in tactical wheeled vehicles, from six to four. In surface ships
we now have five contractors, compared with eight in 1990.
Table 1
Prime Contractors in U.S. Defense Market Sectors
These counts of companies make it clear that, at least among prime contractors for major
weapons platforms, the process of mergers and acquisitions within U.S. defense industries is nearing
its logical end point. (One glaring exception is in surface ships, where pork barrel politics has forced
the Navy to continue to support a number of shipyards that by all accounts is wildly mismatched to
plans for future ship procurement.) For cost is not the only issue when the Defense Department
56
contemplates its industrial base. The number of producers available to bid on new procurement may
have at least some competitive impact on the margins negotiated by contractors, and ultimately, the
price actually paid by DOD. And competition is also presumed to have an impact in providing con-
tractors with an incentive to innovate, to aggressively introduce new ideas and technology.
11
For this and a variety of other reasons, there is little obvious basis in economic theory for choosing
three as a magic number. The number may be greater than five: see Louis Phlips, Competition
Policy: A Game-Theoretic Perspective (New York: Cambridge University Press, 1995), ch. 2;
Stephen Martin, Advanced Industrial Economics (Oxford: Basil Blackwell, 1993), ch. 5.
57
In any event, it is clear that the United States is in most sectors now approaching the mini-
mum number of producers it can choose to sustain, yet still claim it is benefitting from competition
in defense procurement. It is small wonder that the defense department opposed the merger of Lock-
heed Martin and Northrop Grumman, which would have dropped tactical aircraft producers to two,
and missile makers to three. (Issues of concentration in electronic warfare gear and radar were also
raised.) A theme we shall return to below is whether it is possible to introduce additional competi-
tion into defense procurement without supporting additional numbers of national producers, through
international cooperation or the creation of “virtual” companies.
Some argue that even though the process of consolidation among prime contractors may be
reaching its limits in the United States, an equally needed consolidation among subcontractors has
yet to unfold, and is the logical next step in rationalization of the defense industrial base. While the
enthusiasm of Wall Street dealmakers for this argument is understandable, its empirical underpin-
nings seem weak. At least one survey of subcontractors to U.S. defense primes found that for the
vast bulk of these subcontractors, defense sales are only a relatively small part of their business.12
These firms are mainly focused on commercial sales, with defense as a sideline, and any drive to
consolidate in a quest for efficiencies should clearly be primarily driven by non-defense factors. To
the extent that DOD worries about its subcontractors, it is a very small group of specialized pro-
ducers of defense-unique items that are of concern.
One final point that emerges from Table 1 is the relatively large numbers of primes that con-
tinue to produce satellites (five) and wheeled vehicles (four). These sectors are not obviously bene-
fitting from high levels of political interference. Because their products share much in common with
commercial items, this raises the interesting question of whether these are now truly defense sectors,
or instead commercial industries that happen to produce some particularly important, specialized
products for defense applications. Put another way, are satellites really a defense industry today, and
if not, where do they fit into our concerns about the defense industrial base?
12
See Maryellen R. Kelly and Todd A. Watkins, “Are Defense and Non-Defense Manufacturing
Industries Really All That Different?” in Susman and O’Keefe, eds., The Defense Industry. Fewer
than one in five plants doing defense work in 1991 sold more than 50 percent of their output to DOD
or primes. Interestingly, a 1994 LMI survey of small prime contractors to DOD found only 30
percent of these firms did more than half of their business with the defense department. D. Peterson,
C. Webster, E. Gentsch, and M. Myers, “Capital Availability for Small Businesses with Dual-Use
Applications,” Report EC404R1, Logistics Management Institute, 1994.
58
a defense industrial policy in such circumstances might sensibly take action to ensure needed access
to an industrial capability that was primarily commercial in nature.
Thus, if advanced steel available to our military need not be significantly better than that
available to our rivals, the problem of access to advanced steel is really not qualitatively different
than the problem of access to oil. Problems of access to capabilities and resources that are either
commercial, or defense-unique but “commodities” (in the sense we have just defined) we shall
define as a dependency or general industrial base issue, and distinguish it from a defense industrial
base issue.
If on the other hand, the availability of special, advanced steel is critical to important perfor-
mance advantages of U.S. submarines, it is what we are calling a strategic differentiator. Even if
only a small part of a mainly commercial industry’s sales are involved, our taxonomy will label the
item a defense industrial base issue. Heuristically, we make this distinction on the grounds that
industry’s ability to supply this product makes a unique and important contribution to a military’s
strategic or technological advantage against potential adversaries.
The second approach to defining a defense industry applies this label to sectors that sell (or
historically have sold) a large share of their output to military users. This definition is also not en-
tirely satisfactory as it stands, because some products that are arguably commodities—like uniforms,
or anchor chain—may be given special protection by political authorities that effectively creates a
domestic industry that can only compete and sell within a protected military market. In some coun-
tries, a state-owned enterprise may even exist for the sole purpose of producing a commodity for
military use. Intuitively, we should not want to let pork barrel politics shape our analytical notion
of a defense industry.
Therefore, we shall define a defense industry as an industry that sells a large share of its out-
put to military users, and produces at least some products or services that are strategic differentia-
tors. Thus, according to the taxonomy we have just created, problems in access to commodity
products or services—whether commercial or defense-unique—may create a dependency issue, but
will not be a defense industrial base issue. Problems in access to strategic differentiators, on the
other hand, will be labeled a defense industrial base issue. That subset of the defense industrial base
that largely or mainly sells to military customers is what we will call the defense industry.
A final definitional issue about which we will not worry much is at what level of statistical
disaggregation we define an industry. Is aircraft a defense industry, or is a piece of that—military
60
aircraft—the defense industry? We propose to finesse this question by simply taking the lowest level
observational units that are available—whether factories, divisions or groups, or enterprises—and
applying the general concepts outlined above to the available data.
13
We have estimated military output in search and navigation equipment by counting all defense-
unique products (reconnaissance and surveillance electronics, IFF, proximity fuzes, sonar and ASW,
electronic warfare equipment, and underwater navigational systems) plus radar, which is dual use,
but where the military accounts for the bulk of sales.
61
Table 2
Military Output in U.S. Defense-Related Industries
selected years, 1982-1992
Military Products, % of sales Military Product, value $x1000
1982 1987 1992 1987 1992 note
Heavy Ammo 100.0 100.0 100.0 1
Other Ordnance 100.0 100.0 100.0 1
Ship Building & Repair 62.1 83.7 84.0 6985 8732 2
Aircraft 57.9 61.5 38.3 22,138 21,969 3
Aircraft engines NA 46.8 36.6 8812 7541 4
Aircraft parts 49.8 60.5 229.6 11,796 6552 5
Missiles and Space 71.2 76.5 85.0 12,255 11,754 6
Vehicles
Space Propulsion 74.1 76.5 58.4 2652 3162 7
Space Vehicle Equipment 70.1 70.7 68.3 2372 3086 8
Tanks & Tank Components 100.0 100.0 100.0 3017 2503 1
Search & Nav Equipment 79.0 85.0 86.3 28,898 29,708 9
Optical Inst. & Lens 35.5 37.7 34.4 750 787 10
1. All products assumed military.
2. Self-propelled ship building and ship repair for military as share of total for all customers.
Share of covered products in total industry sales: 1982, 85.7 1987, 95.3 1992, 91.6
3. Aircraft sold to or designed for military customers, R&D for military customers or designs, conversion or overhaul
of military aircraft, other services for military aircraft as share of totals for conversion or overhaul of military aircraft,
other services for military aircraft as share of totals for both military and commercial.
Share of covered products in total industry sales: 1982, 99.9 1987, 98.9 1992, 96.6
4. Military engines (built for military aircraft or to military specs), R&D on military engines, other services for military
engines.
Share of covered products in total industry sales: 1982, NA 1987, 97.0 1992, 91.7
5. Mechanical power transmission, hydraulic subassemblies, pneumatic subassemblies, landing gear, R&D on aircraft
parts, other subassemblies for military aircraft as share of totals for military and commercial.
Share of covered products in total industry sales: 1982, 94.5 1987, 91.5 1992, 87.3
6. All guided missiles and services assumed military. Guided missiles plus space vehicles, R&D on space vehicles,
other services for space vehicles for U.S. government military customers as share of totals for all customers. (1987 only:
R&D on complete space vehicles allocated to U.S. military and all other customers on basis of 1982 ratios.)
Share of covered products in total industry sales: 1982, 100.0 1987, 99.8 1992, 89.9
7. Complete missile and space vehicle propulsion units, R&D on propulsion units, other services on propulsion and other
propulsion unit parts for U.S. military customers as share of totals for all customers.
Share of covered products in total industry sales: 1982, 100.0 1987, 98.8 1992, 74.8
8. Airframe, space capsule and other space vehicle, and other parts and accessories, and R&D, for U.S. military
customers as share of totals for all customers.
Share of covered products in total industry sales: 1982, 71.9 1987, 98.1 1992, 88.9
9. Search, detection, navigation, and guidance equipment (SIC 38122) share of all search, detection, navigation, and
guidance equipment and aeronautical, nautical, and navigational instruments x share “military or mainly military” in
38122.
Share of separately classified products in total industry sales: 1982, NA 1987, 97.5 1992, 97.5
Share of “military and mainly military” in SIC 38122: 1982, 85.4 1987, 91.2 1992, 86.3
10. Sighting, tracking, and fire-control equipment share of all optical instruments and lenses.
Share of classified products in total industry sales: 1982, NA 1987, 95.3 1992, 96.4
62
and over half of computer R&D in that decade was paid for by the U.S. military. As late as the mid-
1960s, over half of the integrated circuits sold went to American defense and space users.
Subsets of these same industries probably pose more of a dependency issue today than a
defense industrial base problem. In general, microelectronics and computers—using the definitions
developed above—probably belong in the defense industrial base, given the centrality of specialized
systems based on customization of advanced information technology to the much-heralded RMA.
But DRAM chips or Windows-based personal computer systems—if access issues were to devel-
op—would probably qualify today only as dependency issues.14
Beyond the surprisingly large share of defense output accounted for by electronics, three
other trends are worth noting. First, through the late 1980s, military sales accounted for over half
of total sales of aircraft by U.S. producers. It wasn’t until 1992 that the military accounted for less
than half of sales, and the U.S. aircraft industry finally made the transition from a defense industry
to a primarily commercial sector embedded within the defense industrial base. Second, the ship-
building industry has suffered a continuing decline that, today, leaves it essentially a ward of the
U.S. navy. Little commercial industry remains in the United States. Because it has strategic capabil-
ities—stealth designs, nuclear-powered propulsion, submarines, etc.—as well as military customers,
shipbuilding is clearly a defense industry in the United States. Finally, with the explosive growth
in space-based commercial communications systems over the last decade, one might have expected
space systems to be edging toward the same sort of commercial transition as in aircraft. Indeed, we
see precisely that sort of trend in space propulsion, i.e., the launchers used to boost space-based pay-
loads. Surprisingly, however, the military share of missiles and space vehicles has increased substan-
tially in recent years. The most likely explanation for this contrarian trend is an increasing role for
large, sophisticated, space-based sensors and communications systems in the U.S. force structure.
To answer a question raised earlier, satellites remain primarily a defense industry.
14
In the mid-1980s, the Pentagon’s Defense Science Board argued that memory chip dependency
was a defense industrial base issue, mainly because of the key role of DRAMs in developing leading
edge semiconductor production technology. Today, in the age of the mass-market microprocessor
(logic) chip, it is clear that the most advanced chip manufacturing technology can now be sustained
without having any presence at all in mass-market memory.
63
15
Because of a fragmented Russian industrial structure, these data certainly underestimate the Rus-
sian presence in global defense markets. The data include both goods and services.
16
Based on data in International Institute for Strategic Studies, The Military Balance 1997/98
(Oxford: Oxford University Press, 1997).
64
make it possible for our allies’ defense industries to acquire some of the key components of high-
tech weapons systems at prices that may approach their marginal cost of production. In this sense
the U.S. national system of innovation spills over to our allies as a matter of national policy.
Figure 1
Share of Defense Sales by Top 100 Defense Firms Worldwide
by National Origin of Company
Russia (0.18%)
India (0.48%)
Israel (1.58%)
Canada (0.40%)
Australia (0.22%)
Japan (4.47%)
Non-Nato Europe (1.70%)
US (58.84%)
last batch of 125 of the last vintage fighter are replaced with the last production lot of the current
fighter, and R&D on the next generation aircraft is finally completed. In year 25, the first batch of
the new fighter is delivered, replacing the oldest models of the fighter model closed out in year 24,
R&D on a still newer fighter is begun, etc.
I assume that the cost structure for this archetypal fighter looks a lot like the numbers now
being vetted for the Joint Strike Fighter (JSF): $21.5 billion to develop, $134.5 recurring production
costs on a production run of 3300, $11.2 billion in non-recurring costs.17 I also assume a 10 percent
learning curve,18 and a simplified set of pricing rules that provide an approximation to how contrac-
tors’ output is actually priced and paid for in the U.S. acquisition system. Absent from these calcula-
tions is the possible benefit from increased competition among additional producers.
How do unit production costs change when our total fighter requirement of 3000 is divided
up among varying numbers of producers? With all production concentrated in a single producer, the
recurring cost per fighter, averaged over the life of the program, is about $45 million. Adding in
non-recurring costs, the total “flyaway” cost is a little less than $49 million. With 2 producers build-
ing 1500 aircraft, costs rise, but not dramatically: the recurring unit cost is $46 million, total flyaway
cost is $54 million. Increasing the number of contractors to 6, with a production run of 500, how-
ever, is most definitely not cheap: recurring unit costs only inch up to $49 million, but the total fly-
away cost for the jet is now over $71million! The non-recurring costs—the fixed costs of tooling
up to produce any aircraft at all—simply kill the affordability of the aircraft, when spread over
significantly fewer producers.
This last comparison is especially significant because 500 is close to the numbers that are
being projected for total program size for European aircraft like Rafale and Eurofighter. If only 300
are produced, total flyaway cost in our hypothetical example jumps to a horrific $88 million.
Another interesting calculation involves the total annual cost (including R&D) of main-
taining the procurement program producing a steady-state force of 3000 fighters. The cost of a single
contractor producing all these aircraft is about $7 billion annually—not insignificant, even for the
17
Estimates of non-recurring costs for the JSF are a much larger share of total production cost than
is typical for a fighter program. Some cost analysts have suggested to me that, because this is a
multi-role fighter, specialized mission equipment not standard with every aircraft is being lumped
with non-recurring costs in data given to Congress.
18
That is, every doubling in cumulative output drops recurring production cost by 10 percent.
66
U.S. procurement budget. Adding an additional producer, with whom production is shared equally,
adds a whopping $1.5 billion, an incremental cost that remains roughly constant if even more pro-
ducers are added.
Some have suggested that “design houses” could maintain industrial capabilities, and com-
petitive pressure for innovation, by simply designing, building, and flying prototypes, then part-
nering with an established airframe integrator to form a “virtual contractor” if a design is produced.
RAND analysts estimate that to sustain a full development and prototyping effort would cost about
$500 million annually, split evenly between airframe integration, and avionics and engine work.
Others have countered that the full experience of moving a design through from development into
actual production is needed to maintain needed technical skills, and a company working only on up-
grades and prototypes lacks needed experience. If the design bureau/upgrade specialist could some-
how be combined with a small number of full service aircraft producers to generate greater competi-
tive and innovative pressure, however, the potential savings could be large: the difference between
$250 to $500 million annually for a design house, and $1.5 billion to add an additional producer.
Another cut at the same issue is to ask how much is saved when the steady-state force (and
production) is halved, from 3000 to 1500. With only a single producer, total program costs drop by
a little less than 40 percent; with two producers, costs drop by about a third. With five producers,
expenditures fall by only 23 percent. Keeping a large number of firms going winds up consuming
most of the possible “peace dividend” from reducing force sizes. Viewed from this angle, the consol-
idation kicked off by the Clinton administration in 1993 might be interpreted as a way of guaran-
teeing that any future cuts in forces would wind up resulting in real savings, rather than being
absorbed in an oversized and inefficient industrial base.
The message of this exercise is that maintaining an industry that designs and produces only
small numbers of advanced aircraft is going to yield a product that is virtually unaffordable, and at
a substantial cost disadvantage when exported. Huge economies of scale will push smaller pro-
ducers—in Europe and Japan—to merge, and even after that, to export just as much as is humanly
possible. Given their severe cost disadvantage relative to higher volume American producers, smal-
ler competitors will also be forced to turn to those export markets where cost—and possibly perfor-
mance—is less important. Such “marginal” customers are likely to be located in “problem” regions,
where sales may generate frictions with the United States.
67
national security. This last idea has gone by various names—a suppliers’ cartel, an “inner circle,”
etc.—and is probably best viewed as an experiment to be pursued, rather than a crystal clear vision
of a particular endpoint. One of the implications of such a system would be that transnational tech-
nology sharing in the defense sector would continue to be governed largely by the state rather than
being simply the product of globalizing market forces.
to guarantee the economic viability of European defense capabilities, the only alternative will be
relatively indiscriminate exports to dubious buyers.
An additional advantage of the inner circle approach is that it can be defined and refined
incrementally. In the beginning, its domain could be quite narrow, negotiated on a case-by-case
basis. For example, we could initially experiment with this approach in very specific systems—for
example, ballistic missile defense systems, or Stealth cruise missiles, or Stealth radar—with our
most trusted allies. Given a track record of initial success, it could be then expanded to cover addi-
tional types of systems, and eventually, perhaps, become an integrated framework covering produc-
tion and export of a whole range of advanced weapon systems.
Incremental expansion could also cover new categories of membership, so that instead of
having a single inner circle, the system could be more like concentric circles. Close allies would
share the greatest degree of access, and accept the greatest degree of restraint. The outermost circle
would include virtually everyone, but also be associated with the least forceful restraints—an
expansion and elaboration, perhaps, of current global agreements covering international export of
sensitive military technologies and weapons of mass destruction. Intermediate levels of inclusion
and restraint between these two limits could be negotiated where it made sense: bringing Russia into
the fold, for example, or covering sensitive but somewhat more widely diffused advanced military
technologies mastered by a larger number of players. In short, the inner circle idea could be viewed
as an experiment rather than an endpoint: a graduated and progressive construction of an interna-
tional regime blending restraint and cooperation in military weapons systems production and sales,
using pragmatic and selective principles for inclusion of participants and technologies.
Some might argue that this is an impractical and utopian approach that would never survive
in the rough and tumble of the real world. In fact, however, “impractical” restraints on export of
components and systems for missiles and weapons of mass destruction—though far from perfect—
currently serve us well in reducing the dangers from proliferation of these systems. And there are
real examples where an incremental “inner circle” type approach has shown itself to be practical.
When the United States, Germany, France, and Italy sat down and agreed to pool funding and tech-
nologies in a cooperative development program for a common theater missile defense system—
MEADS, the Medium Extended Air Defense System—in 1995, all four partners agreed that no
export sales could take place without common assent. True, France later dropped out of this program
when it finally felt the pressures of defense budget cuts in 1996. But MEADS showed that export
70
restraints linked to a sharing of funds, technology, and production in a common acquisition program
can be negotiated, and with strong-willed and independent partners. Out of such incremental first
steps, an inner circle of armaments cooperation and export restraint can gradually be built, and later
expanded.
One thing is certain. Weapons exports are at least partly driven by economics. Creating
incentives to not engage in irresponsible proliferation of advanced capabilities requires global coop-
eration in constructing a regime which reconciles national security interests in maintaining defense
establishments and curbing uncontrolled proliferation with the economic realities of a high-tech
defense industry inexorably driven to seek the widest possible market. Security concerns provide
a powerful argument for maintaining the special character of defense technology in the national
system of innovation.
The Place of the French Arms Industry in its National System of Innovation
and in the Governmental Technology Policy
Claude Serfati
Introduction
This chapter addresses the role of military R&D and defense-related innovation networks
in the French national system of innovation. We find that this role has remained highly significant
so far, despite profound changes in the geopolitical and world economy setting, despite strong do-
mestic budgetary cuts in procurement and military R&D, and despite an outright decline of military
innovation as a propeller for commercially oriented innovations. The resilience of the French arms
industry can be explained by institutional and organizational factors and mechanisms. As we have
long argued, the arms industry is organized as a “mesosystem,” with strong and interactive relations
existing between the different “players”—companies, technical agencies, and the procurement agen-
cy, the Délégation générale pour l’armement (Chesnais and Serfati 1992). The priority given to mil-
itary objectives for five decades has resulted in a pre-eminent role for defense innovation networks
in the national system of innovation, as well as the pervasive role of the French mésosystème de
l’armement (FMSA) in the government’s technology policy.
First, we present statistical data on French expenditures for military R&D and related major
technology programs. An outstanding feature of these data is the strong concentration of expendi-
tures on a small number of sectors, regions, and companies. Focusing our attention on the latter, we
suggest that a handful of major defense contractors are at the nexus of R&D funding networks and
of the national system of innovation. This situation can be traced back to the early 1960s, when mili-
tary objectives were given top priority in the governmental technology agenda. That this situation
persists illustrates that governmental policy is “path-dependent,” a fact that can be explained only
if the role of structures, institutions, and organizations are taken seriously. Analysis of the restruc-
turing process of the French industry reveals how attractive arms production remains. It also under-
lines the extent to which beefing up the arms industry has been at the top of the industrial policy
agenda, giving no room for a debate on the future of commercially-oriented high-tech activities.
Now, with the “europeanization” of the arms industry as the “new horizon” for the French govern-
71
72
ment and companies alike, the challenge is to reconcile the setting-up of a European industry with
preserving the FMSA.
as well. From the 1980s on, in absolute terms, the increase in military equipment expenditures has
been much higher than expenditures for civilian equipment (e.g., transport infrastructure, hospitals,
etc.). Despite strong cuts in the military equipment budget since 1992, it remains at a higher level
than its civilian counterpart. This trend is not for France only—in most OECD countries cuts in
DBRDE have been no greater than in CBRDE (OECD 1998). The downward trend in the French
defense budget experienced during the decade was, moreover, reversed in 1999 compared to 1998,
with an increase in the total defense budget (+2%), in procurement (+6.1%), and in R&D (+7.6%)
(MacKenzie 1998).
Table 1
Defense and Civilian Budget R&D allocated to Business
R&D category 1985 1986 1987 1988 1989 1990
DBRDE (1) 1.455 1.413 1.615 1.670 1.828 2.120
CBRDE (2) 821 965 884 878 870 934
TBRDE (2+1) 2.276 2.378 2.500 2.547 2.698 3.054
(1)/(2+1) ratio 63.94% 59.42% 64.62% 65.54% 67.77% 69.42%
1
This section draws upon a 1997 research program, with C. Carpentier, for the Observatoire des Sci-
ences et des Techniques (OST) (Carpentier and Serfati 1997).
2
The exchange rates used throughout this chapter are the purchase parity power (PPP) exchange
rates, 1US$ = 6.5 FRF. Figures are rounded.
75
on Defense Technological Programs (DTP), while the civilian ministries’ spending relates mainly
to “Major Technological Programs” (MTP) in aeronautics and space, nuclear, telecommunication
($1.55 billion). The remaining funding ($0.3 billion) is less programs- and more incentives-oriented.
Even though they are labeled civil-oriented, MTPs offer significant similarities to DTPs. They are
very high in terms of costs, their lead time spans over years—sometimes decades—and in both cases
the government procurement market accounts for a significant share of the firms’ total output. Final-
ly, by some standards, MTPs could be labeled as “strategic” in the triple meaning put forward by
Freeman and Soete (1997, 341): technological (with learning and dynamic increasing returns); trade
(with support policies dictated by comparative or potential comparative advantage); and industrial
(with active industrial policies “best described with reference to the French notion of filières”).
In a study based on MoR data, Carpentier and Serfati (1997) found that not only are public
funds heavily concentrated on a few firms (120 out of the 3500 firms declaring an R&D activity),
but also that the firms benefitting from military R&D contracts also receive the bulk (94 percent!)
of the $1.55 billion of R&D contracts allocated for MTP by civilian ministries. Public R&D con-
tracts (military and civilian) are concentrated in two industrial sectors, the aerospace and the equip-
ment sector (the latter being a broad category that includes telecommunications and defense elec-
tronics). The aerospace sector receives 46 percent of the military R&D contracts and 72 percent of
the civilian contracts, while the equipment sector receives 42 percent of the military and 16 percent
of the civilian contracts. Finally, geographical distribution is also strongly concentrated, with the
“Greater Paris” region receiving 52 percent of the total amount of R&D contracts, while accounting
for only 16 percent of the total domestic industrial output and 32 percent of the French arms produc-
tion (de Penanrose and Serfati 2000). At the firm level, R&D contracts are also concentrated on a
few “industrial groups” as discussed in the next section.
depart from the neoclassical view that competition is driven by market forces and thus fell short in
not giving due account of the market power conferred upon the corporation by virtue of its size and
organization (Peterson 1989). This is, obviously, not the case in the “Institutionalist” tradition,
which, following Veblen’s analysis (1963), has a long record in studying how power matters and
how big corporations shape the economy. This stream has been particularly useful as far as the role
of military institutions is concerned (Melman 1997).
The Institutionalist tradition gives great weight to the fact that ownership is separated from
control in large enterprises—see Veblen (1963) on “absentee ownership”—with opposition between
“nominal” and “effective” ownership (the former referring to the right to receive an income in return
for risking one’s wealth, while the latter reflects the actual ability to control the corporate assets)
(Berle and Means 1932). It is fair to say, however, that case studies of industrial countries have
found that the economic development of capitalism has been to some extent rather different from
the picture painted by those asserting a strong opposition between managers and shareholders. Cor-
porate directors and executives have been disproportionately numbered among the large personal
shareholders (Scott 1997), a situation that has been reinforced in recent years by agency theory,
which argues that if potential conflicts between managers and shareholders are to be overcome,
stock options should be an important part of the top manager’s pay (Jensen and Meckling 1976).
Another theoretical school that has a longstanding interest in power and control issues is
finance capital theory. Here, the emphasis is on the way that ownership control and market power
can be brought together, thanks to the intercorporate networks created through interweaving of share
participation. Based on a comprehensive investigation, Scott (1997, 16) has found that extensive net-
works of intercorporate capital relations have formed in all of the advanced capitalist economies,
with, however, distinctive national patterns of development. This approach allows us to understand
what we mean by industrial activities-oriented “groups.” These organizations are often designated
in the Anglo-Saxon literature as “firms”—even when they comprise tens and indeed over a hundred
affiliates and are owned and managed by holding corporations, which are the heart of the contempo-
rary transnational corporation (TNC). These groups have dominant activities in given manufacturing
and/or service industries, but their activities are managed as a particular type of asset, namely, the
rate of return (RoR), which is measured against that offered by other types of assets—first and fore-
most, bonds and stocks. These industrial groups are becoming more and more what has been called
in France “financial groups with industrial activities.”
77
Applying this analytical reading, it can be said that in France defense contractors rank at the
top of the groups involved in industrial and R&D activities. While this chapter mainly focuses on
the industrial and technological activities by defense contractors, the last section will comment on
how the globalization process, in particular the changes related to the triumph of global finance,
makes defense contractors act more and more as financial groups.
3
This does not mean that only one percent of the R&D contracts from MoD goes to small and
medium-sized enterprises (SMEs). SMEs that are subsidiaries of major defense contractors are also
recipients. The “indepenent” firms mentioned here are firms having military activity that are not
owned by the main defense contractor groups, but by commercially-oriented industrial groups.
4
It seems that the self-funded R&D-to-sales ratio of U.S. contractors is much lower than that of their
French counterparts (Hitchens and Finnegan 1998) and European (EC 1996) counterparts.
78
talized sub-systems (Chesnais 1993; Serfati 1992). It is hierarchical because the core of innovation
policy is “mission-oriented” and is set in motion by the State. This impetus is carried out by govern-
mental agencies, which are in charge of procuring the programs, orienting technological trajectories,
etc. The industrial leadership then goes down to defense contractor groups, whose role is to set up
the network of hundreds of subcontractors (as seen above, most of those are subsidiaries of defense
contractors) and to perform the work of the programs.5
Another feature of the FSI is that it is compartmentalized, and this for a series of reasons.
Each governmental agency is keen not to lose its core competences to another agency; manager
recruitment in each agency comes mainly from the different and rival Polytechnique high schools’
grands corps (e.g., Ingénieurs des mines for CEA, Ingénieurs des télécommunications for CNET,
Ingénieurs de l’armement for DGA, etc.) (Serfati 1995); and finally, the core of the FSI is located
in the aerospace, nuclear, and arms industries. These industries, which receive the bulk of govern-
mental R&D funding, have rather poor records so far as their capabilities in terms of intersectoral
technological diffusion are concerned.
5
Changes are under way in the procurement and R&D rules that could lead to an increase in the role
and responsibility by defense prime contractors.
80
with the consequences of behavior of people in a market—a term with at least five distinct and
sometimes opposed meanings, as listed by Boyer and Hollingworth (1997)—while the analytical
background of this chapter is the study of innovation as a part of the process of production and
reproduction of value. In that sense, our approach is in phase with Dalum et al., when they state that
“the specific characteristics of each NSI are, so to speak, the ‘superstructure’ of its production
system” (1995, 307).
Defining the state is important if one wants to depart from the understanding of governmental
technology policy conceived as a corrective for “market imperfections.” First, the state is a political
entity: it reflects a form of power organization in human society and can wield the “monopoly of
legitimate violence” (Weber 1959 [1919]). This dimension of state responsibility is particularly rele-
vant here, since as observed by Ergas (1987, 51), “Nation-states have long been major consumers
of new products, particularly for military uses.” Second, the state should not be instrumentalized as
a tool for managing externalities (as in the fundamentalist neoclassical approach) or, antithetically,
as a simple agent used to solve “overaccumulation” difficulties (as in a fundamentalist marxist
approach). The state is made up of institutions and organizations6 that have gained momentum and
power by crystallizing definite types of socio-economic relationships and that have, over time, been
able to gain momentum to develop on their own and to acquire autonomy. In a nutshell, the role and
place of state institutions and organizations in any nation is strongly determined by history.
The variety of institutional configurations assumed by the state in the main industrialized
countries is a major factor for explaining the diversity of their technology policies. To illustrate this
point, we can refer to the new relationships emerging from WW II among science, technology, and
the economy and the arrival of what has become known as “big science.” Even if these changes were
in part the result of “technology- and economic-push,” few doubt that in a broad way they were
“organization-pulled.” For example, Freeman and Soete (1997) remind us that overemphasis on “big
science and technology” is in a large part due to the self-interest of the R&D and military establish-
ments.
That the institutional setting plays a decisive role in the shaping of technology policy is illus-
trated by comparing the very dissimilar objectives and records in governmental policies for the same
6
Here it should be mentioned that differentiating between the two categories is a daunting task (see
Edquist 1997).
81
technological domains in different countries. The first case is the very large differences in the out-
comes of governmental policies related to laser technology in France and Germany, policies
launched at the same time and with the same level of public funding commitment (Wolter et al.
1996). The second case relates to the biotechnology industry in the United States and Japan. A simi-
lar strong commitment by both governments has resulted in very different technology trajectories,
because of the differences in their respective institutional settings (Callan 1997).
To sum up: a) the modalities, and not only the degree, of state involvement results in dif-
ferences in governmental technology policy and outcomes; b) governmental involvement cannot be
explained simply as aimed at correcting market imperfections—rather it is the result of active inter-
vention by state organizations in the process of creating technological innovation, at least as regula-
tors, producers, and users (Gregersen 1995).
Institutional Features of the French Model of Technology Governmental Policy: The Core Role of
the Mesosystem of Armaments
Considering history is useful when one investigates to what extent science, technology, and
the state have been intimately linked with each other in France (Papon 1978; Chesnais and Serfati
1992; Chesnais 1993). The type of technology policy set in the late 1950s reflects the lasting role
of the state. The “major programs” launched were military- or strategic-oriented. Meanwhile, gov-
ernmental agencies were created or strengthened, combining in an unusual way administrative,
procurement, and technological responsibilities.
As explained above, we argue that major features of French technology policy and—beyond
it—of the national system of innovation can be made more clear if this set of interrelations among
organizations is taken into consideration. Regarding France, the pervasive role of state institutions
and organizations should not be accounted for by an appeal to “market failure,” but rather by “entre-
preneurship failure.” There is no space here to further develop the set of cultural and political factors
explaining this kind of strange “failure” in a capitalist system. In any case, what began as “colbert-
ism,” a term referring to the key role played by the Louis XIV’s prime minister in developing
nascent capitalism, has lasted for over three centuries and gained further momentum after the second
world war, a time when capitalism has become more than “mature.”
Hence, the large place of DBRDE and MTP-related funding in the national system of R&D
that has been presented above is only the visible part of the iceberg. Behind the overconcentration
82
of resource allocations stands a handful of organizations that wield power and have the technical
expertise to carry out those programs. All of these programs, and all of the organizations in charge
of them, are fully or partly defense-related. They mainly include the DGA (Directorate General for
Armaments), the state technical agencies (CEA, CNES, CNET, ONERA, etc.), and major industrial
groups that are at the same time MoD prime contractors (Thomson-CSF, Alcatel, Aerospatiale,
Dassault, Snecma, Matra, etc.). Extending the concepts of filière and mésosystème developed in the
1980s by French economists, we have proposed the term of mésosystème de l’armement (French
Mesosystem of Armaments, FMSA) to describe the set of organizations, bound up with each other
through market and—possibly more important in the defense industry—through non-market rela-
tionships (Serfati 1992; Chesnais and Serfati 1993). The cohesion of the FMSA is reinforced by the
core role played by DGA’s Engineers of Armaments (one of the grands corps), the pervasive influ-
ence of which is all the stronger because most of the grands corps are represented in the top jobs
in the defense contractors. A noted American scholar has characterized the grands corps as the glue
that holds the entire defense productions system together (Kolodziej 1987).
The electronic sector is a case in point. The Juppé government decided to proceed with a
total separation within Thomson S.A. (the parent company) between Thomson-CSF (military activi-
ties) on one hand and Thomson Multimedia (electronic appliances) and SGS-Thomson (electronic
components) on the other. The government also announced that it would dispose of Thomson Multi-
media by giving it free (allegedly because of its level of indebtedness) to the Korean conglomerate
Daewoo. (This decision was called off by the Privatization Commission some months later.) The
government’s strategy reveals how large the gap is between what is happening in reality—i.e.,
erecting a “wall of separation” between military and commercial capabilities—and the official dis-
course made at length on the need to strengthen the links between military and commercial technolo-
gies, promote “dual-use technologies,” etc.
After years of indecision—it took over 30 months and an unexpected change in the parlia-
mentary majority for the government to decide—and as a result of uphill lobbying from the firms
concerned and their allies in the FMSA, Thomson-CSF and Alcatel’s military businesses were
merged in September 1997, with Alcatel becoming the biggest corporate shareholder in Thomson-
CSF. Not surprisingly, the split of Thomson S.A. between its military and commercial divisions was
confirmed, and no governmental comment on the future of Thomson Multimedia and SGS-Thomson
was made by the Jospin government. Following its merger with Alcatel, Thomson CSF has speeded
up foreign acquisitions, clearly keen to become a global company on its own. The company holds
a significant defense market share in Europe, ranking number two in July 1999 before the Matra
Aerospatiale-DASA merger. Since 1997 its strategy has been to focus on defense, or near-defense,
electronics.
For Alcatel, gaining the Thomson-CSF acquisition bid for Matra-Lagardère marked an unex-
pected comeback in the defense industry. In 1983 an agreement, nicknamed by the parties “the Yalta
of the electronic sector” (Yalta de l’électronique) was struck under the auspices of the government
between Thomson S.A. (parent company) and Alcatel (parent company). This deal aimed at covering
the whole scope of electronic products. A major point of this agreement was to evenly divide the
state-oriented markets. Alcatel was to become the sole supplier for the state-dominated market in
civilian products (public switching boards, telecommunications equipments, etc.), and Thomson-
CSF (a branch of Thomson S.A.) was to become the sole supplier for the state-dominated market
in military products (defense electronics, weapon systems, military telecommunications etc.) and
commercial electronic business.
84
Alcatel’s 1997 comeback in the arms industry through the take-over of Thomson-CSF
reflects how attractive and lucrative the defense industry remains to French firms. The attraction of
military R&D and procurement markets has been a driving force in the French restructuring process.
Similarly, the military market proved attractive to dealmakers in the United States, raising concerns
that the U.S. megamergers (Boeing-McDonnell, Lockheed-Martin, Raytheon-Hughes) could lead
to further isolation of the defense industry in the economy (Markusen 1997). The comparison should
not be overdrawn, however, as there are strong differences between the two countries: the role of
financial markets in the restructuring process, the promotion of start-up companies, the importance
and role of SMEs, and—obviously—the leading role played by the United States in international
relations (Markusen and Serfati 2000).
In July 1998, a year after Alcatel acquired Thomson-CSF, the Jospin government announced
that the state-owned Aerospatiale company would be privatized and merged with Lagardère’s Matra
Hautes Technologies. Then, in October 1999, Aerospatiale, Matra, and DASA announced the crea-
tion of EADS (European Aeronautics, Defense and Space Group). The new group will rank first in
Europe, and third worldwide. Thus, the European aerospace industry will be dominated by two
major companies of roughly equal size ($22 billion). EADS and the new British Aerospace Systems
are confronted with two options: merging their activities and risking the accusation that they are
erecting “fortress Europe” or looking for transatlantic links with the risk of opening further avenues
in Europe to American companies.
There is no doubt that the consolidation to two “national champions” (Thomson-CSF and
Aerospatiale-Matra HT) is aimed at fortifying the French industry in order to prevents its marginali-
zation in the restructuring process of the European aerospace and defense industry. The next section
examines what is at stake if the French mesosystem of armaments is to be preserved.
Thompson 1996; Kleinknecht and ter Vengel 1998). These analysts are right to put the actual pro-
cess in a long-term perspective but, in our view, they underestimate the dramatic changes that have
taken place in recent years. Using comparisons in the levels of Foreign Direct Investment to GDP
or trade-to-GDP ratios between the pre-WWI and the current period to conclude that nothing has
changed since the beginning of the century is hardly convincing, and does not pay sufficient atten-
tion to the key role actually played by TNCs, whose grip on the world economy is a compelling
fact.7 Still more sweeping is the momentum gained by “global finance” and the re-emergence of a
new kind of finance capital, not made up of rentier families,8 but of a variety of new institutional
investors (pension, mutual, hedge funds) controlling over 10 trillion dollars.
Our understanding of the process is that this “global finance”—encapsulated as the “manic
logic of capitalism” (Greider 1997)—is the very driving force of globalization and accounts for a
large part of the difficulties of the current economic situation (Henwood 1998). Since the early
1980s, the high level of leverage by the borrowing9 and financial innovations that were relentlessly
introduced on the markets have contributed to an unprecedented amount of credit creation, making
it more and more plain that a large part of the capital swirling around the world is “fictitious” capital
(Guttmann 1994) or to put in more conventional terms, “paper wealth” (Business Week 1998). For
sure, owing to the Asian crisis this paper wealth “has already began to vanish,” at least for Asian
companies and households holding financial assets. More damaging, what is sometimes too narrowly
called a “financial crisis” in emerging countries was transformed into a dramatic economic crisis on
the “real side” of economy with resulting social backlash (Business Week 1998).10
Despite strong concerns, the American and, to a lesser extent, western European countries
have escaped the devastating consequences of the economic crises that have been plaguing most of
7
Estimates are that the 200 top TNCs have a total turnover equivalent to over 28 percent of the
world aggregate GDP and account for two-thirds of world trade.
8
At the turn of the century, 5000 British families and 20,000 continental European families provided
the bulk of the capital on the bond and equities market.
9
See Plender 1998: “[I]t is not uncommon for hedge funds to borrow five or six times their inves-
tor’s funds in pursuit of high returns.”
10
The International Labour Organization estimates that the number of underemployed will be around
one-third (750-900 million people) of the world’s workers. See Taylor 1998.
86
the world since 1997. At the turn of the century, “once again, North America and Europe are the
global anchors of prosperity and stability, while the rest of the world struggles in economic limbo”
(Warner 1999). At the close of 1999, Wall Street’s Dow Jones industrial average was 60 percent
higher than in August 1996, a time when the Federal Reserve Board’s chairman, Mr Greenspan, said
that the bubble reflected “an irrational exuberance.” That the U.S. economy has been enjoying good
macroeconomic growth rates and low unemployment since 1993 is unchallengeable, as are the sky-
rocketing foreign indebtedness ($2 trillion at the end of 1999) and domestic debt used to finance
household consumption and share buy-back by non-financial business.11 And in the United States
as well as in Europe, the pressures exerted by the financial markets through the reinforcing of “cor-
porate governance” rules have negatively affected the level of firms’ capital accumulation along
with the growth and orientation of their R&D expenditures (Chesnais and Serfati 2000). Financial
market constraints, whether exerted at the macroeconomic level on social-oriented public expendi-
tures or at the microeconomic level, have also meant increased social inequalities and poverty, as
flexibility in labor markets translated into serious deterioration for workers in the United States
(Appelbaum and Berg 1996) and Europe (O’Sullivan 1998). Once distributional issues and the situa-
tion of workers have been taken into account, it should become clear that the “Atlantic prosperity”
is restricted to the upper income levels of the population.12
Even though the French industry was able, albeit with difficulty, to cope during the 1980s
with the internationalization of production and trade (through a dramatic increase of FDI and techni-
cal interfirm agreements), the pressures exerted by global finance throughout the 1990s have posed
a major threat to the French model of capitalism, including the French mésosystème de l’armement
(FMSA), which is one pillar of this model.
11
Over half of the $400 billion debt increase by non-financial businesses was used to finance share
buy-backs (The Economist 1999).
12
Just to give an example, 86 percent of the Wall Street gains between 1989 and 1997 were captured
by 10 percent of the households (Economic Policy Institute 1999). In France, reversing the three
post-WWII decades, the 1980s saw in increasing inequalities: between 1979 and 1996, the income
of the bottom 10 percent of the population fell by 3 percent, as the one of the top 10 percent in-
creased by 9 percent.
87
Table 2
Degree of Capital Held by Foreign and Anglo-Saxon Institutional Investors
(% of company’s capital)
Company Foreign shareholders Anglo-Saxon institutional investors
Alcatel 49 30
Lagardère 47 36.7
Thomson-CSF 14.6 11.1
Source: L’Expansion, November 4-17, 1999
The restructuring of France’s arms industry should be analyzed in this context. There is no
doubt that the arms industry remains in France, as in other major arms producing countries, a lucra-
tive activity. The consolidation process has been used by the major companies to reinforce their
13
The number of managers recruited from the grands corps has continuously increased between
1986 and 1995, despite the privatization of firms and more free–market oriented policies, a fact
construed by sociologists as the reinforcement of non-market links between government and firms.
14
Estimates are that foreign (mainly American and British) pension funds would control between
30 and 40 percent of the capital of the French “blue chip” companies.
88
presence in this sector. The privatization of ownership of major contractors—a process that has
somewhat paradoxically been more active under the left-wing coalition government (la gauche
plurielle, associating the Socialist, Communist, and Green parties) than between 1993 and 1997,
under the right-wing government of Mrs. Balladur and Mr. Juppé—is expected to reinforce their
competitiveness and free-market orientation, as privatization did in the United Kingdom (Dunne and
Smith 1992). Meanwhile, another major objective of the restructuring process is to preserve the
“cohesion” of the FMSA, through setting up capital cross-participation among the leading firms
(Aerospatiale, Alcatel, Dassault, Matra Thomson-CSF). Figure 2 displays the financial links as of
the end of 1998.
The redefinition of the DGA’s mission, which began in 1997, will probably weaken its role
in arms production and the R&D process, by refocusing its administrative responsibilities and tech-
nical expertise and moving toward dropping its production activities (in shipyards). On the other
hand, DGA could increase its influence, since it will be in charge of long-term strategy through the
“foresight thirty-year plan,” aimed at identifying future technological breakthroughs. Also, the
DGA’s arms engineers, one of the grand corps of the Polytechnique schools, possess unique exper-
tise in defense technology, expertise that will remain critical in the design of complex weapons
systems.
All in all, the industrial restructuring and financial reshuffling of the FMSA that has taken
place may be seen as aimed at “locking in” control of the system and preventing would-be foreign
institutional investors from having a leading role on the boards of the defense contractor groups.
This strategy may be feasible in the arms industry, where protecting the capital ownership from for-
eigners can be grounded in sovereignty and security motives, but it seems more difficult to carry out
in other industries. Consequently, the current restructuring process could be seen as an endeavor to
preserve the arms mésosystème as one of the last building blocks of the French model of capitalism.
But with europeanization of the arms industry is the new horizon for the French government and
companies alike, how it can be reconciled with preserving the FMSA is anything but clear.
Figure 2
Cross Participation in the French Mesosystem of Armaments
of the arms industry. The role of the United States and of the State (in the broad sense defined
above) remain critical if we are to understand how the restructuring process is proceeding in other
(mainly European) countries. The particular features of the U.S. arms industry create unique compet-
itive advantages for U.S. contractors, including the size of their domestic market, firmly sheltered
by procurement rules from the “winds of liberalization”; the momentum of geopolitical factors in
the arms trade; and the strong commitment of the Pentagon to military R&D. Finally, even though
Wall Street has been active in the consolidation of the industry, the role played by the DoD when
confronted with budget cuts in the post–cold war era (e.g., the “last supper” meeting) should not be
overlooked. Whereas it was possible to consolidate the U.S. industry at a fast pace, for years the
consolidation the European industry has been a daunting task, haunting the “long dinners” that took
place between industrialists and governments (let alone defense or civilian European- or Commu-
nity-level bodies) in Europe.
Basically, the obstacles to further consolidation of the European industries come from the
difficulty of disentangling the complex set of cross-links among the companies, public institutions,
90
ministerial departments, and political parties that have been at the very foundation of national
defense industry-related networks for many decades. Even though TNCs have a major role in the
globalization process, we should not imagine that globalization puts an end to the role of the state,
conceived of in this paper as a set of institutions and organizations that are anything but disappear-
ing. Even for those companies with a strong commitment to shareholder value, the government’s
involvement remains highly critical (Serfati 2000).15 This is why the strategy of preserving the
domestic base should not be seen as only part of a “gallic” tradition. It is why, until recent years, the
consolidation process in all countries was mainly based on mergers between companies of the same
nationality.
That the role of governments in preserving the domestic base has remained critical is evident
in the way the restructuring process has been proceeding. Even after the formation of national cham-
pions in the United Kingdom (British Aerospace and GEC-Marconi) and Germany (DASA), strong
pressures are present in these countries to prevent domestic companies left out of these giant com-
panies from falling into foreign hands.16 GEC-Marconi met opposition from the German government
in its bid to take over the simulator manufacturer STN Atlas, and was allowed to acquire only 49
percent of the capital. Siemens’ defense electronics facilities have been split between DASA and
British Aerospace, which each acquired Siemens’ defense assets in their respective countries. In the
UK ground weapons industry, the three major firms (Alvis, GKN, Vickers) agreed on one thing: that
consolidation would have to begin at home, a fact confirmed by the decision made by GKN and
Alvis to merge their fighting vehicle lines. A similar concern exists in Germany, where armored
vehicle firms are consolidating on a national basis (Krauss-Maffei and Wagmann). Similarly, the
friendly cash offer launched by Saab for Sweden’s other large manufacturer is aimed at creating a
Big Nordic defense firm (Latour and Michaels 1999). So far, what is labeled European mergers have
15
See, for example, the BAe’s chief executive’s testimony before the Committee of the French Na-
tional Assembly: “The challenge for the European industry will be to maintain a fair balance
between its international dimension and the concerns of each nation to maintain its influence on the
industry.” (“Le défi à relever pour l’industrie européenne sera de trouver un équilibre entre son
caractère international et les soucis de chaque nation de garder une influence sur l’industrie”)
(Quiles and Chauveau 1997).
16
See, for example, the VSEL acquisition by GEC-Marconi. The European Commission was refused
any say by the British government on the grounds of Article 223, as the fight to acquire VSEL was
between the two leading British companies.
91
mainly concerned bi-national mergers, as if this kind of process offered guarantees to both govern-
ments and major contractors: to the former that they would be able to retain some control on their
national firms and to the latter that their domestic market would be protected while they remained
the main recipients of public funding of R&D (Lovering 1998).
It is in this context that the creation of a single European aerospace and defense company
should be analyzed. More often than not, the obstacles to the creation of such a company are pre-
sented as epitomizing the opposition between globalization-minded firms and governments alike,
and—the others. Table 3 suggests a more complicated story, with significant divergences among
major players on several issues as to how the constitution of this company should be carried out.
Such a company, if created, would be only a segment—albeit an important one—and not the entire
defense industry, since shipyards, ground weapons, chemicals and nuclear establishments, all sectors
with strong national vested interests, are left aside. Finally, some commentators are concerned with
the monopoly situation that would be created if a Single Corporate Entity were to be set up. The
recent creation of two European companies in this field (Matra Aerospatiale DASA and BAE
Systems), the finalization of which has still to be confirmed, raises in new, not easier, terms the issue
of the consolidation of the European aerospace and defense industry.
Table 3
The consolidation of the European aerospace and defense industry: a long way off?
Source: C. Serfati
92
Were this trend confirmed, it would mean that a new kind of military-industrial system,
“leaner and meaner” through consolidation and jobs downsizing, more internationalized, and with
institutional investors holding a key role within it, will emerge in coming years. Keeping at the
leading edge on technology innovation will continue to be the priority on the defense agenda in the
coming years. Probably sourcing an increased portion of its technology in the commercial industries,
these military-industrial systems will remain a core component of national systems of innovation
in the countries with strong procurement and R&D budgets.
References Cited
Appelbaum, E. and P. Berg. 1996. “Financial Market Constraints and Business Strategy in the
USA.” In Creating Industrial Capacity: Towards Full Employment, ed. J. Michie and J.G.
Smith. New York: Oxford University Press.
Bartlett, R. 1994. “Power.” In Institutional and Evolutionary Economics, ed. G.M. Hodgson, W.J.
Samuels, and M.R. Tool. Northampton, MA: Edward Elgar.
Bauer, M. and B. Bertin-Mourot. 1995. L'accès au sommet des grandes entreprises françaises 1985-
1994. CNRS/Boyden.
Berle, A.A. and G.C. Means. 1932. The Modern Corporation and Private Property. New York:
Macmillan.
Boyer, R. and J.R. Hollingsworth, eds. 1997. Contemporary Capitalism: The Embeddedness of Insti-
tutions. New York: Oxford University Press.
Business Week. 1998. “Asia: ‘Social Backlash’” (special issue), August 17.
Callan, B. 1997. “Why Production Technology is Not a Measure of Competitiveness in the Biotech-
nologies.” Public and Science Policy 24, no. 3 (June).
Carpentier, C. and C. Serfati. 1997. “R&D militaire, grands programmes et capacités technologiques
nationales: caractérisation et indicateurs.” Paper prepared for the Observatoire des Sciences
et des Techniques (OST).
Chesnais, F. 1993. “The French National System of Innovation.”' In National Innovation Systems:
A Comparative Analysis, ed. R.R. Nelson. New York: Oxford University Press.
Chesnais, F. and C. Serfati. 1992. L'armement en France, Genèse, ampleur et coût d’une industrie,
Luçon: Nathan.
———. 2000. “L’evolution des dépenses de R&D et la gestion de l’innovation dans le régime
d’accumulation a dominante financière.” In Technologie et Connaissance dans la Mondial-
isation, M. Delapierre, P. Moati, and E.M. Mouhoud. Paris: Economica.
Christensen, J.L. 1992. “The Role of Finance in National Systems of Innovations.” In National
Systems of Innovation: Towards a Theory of Innovation and Interactive Learning, ed. B.A.
Lundvall. London: Pinter.
Dalum, B., B. Johnson, and B.A. Lundvall. 1992. “Public Policy in the Learning Society.” In
National Systems of Innovation: Towards a Theory of Innovation and Interactive Learning,
ed. B.A. Lundvall. London: Pinter.
94
de Penanrose, R. and C. Serfati. 2000. “Regional Conversion Under Conditions of Defense Industry
Centralization: The ‘French Case’.” International Regional Science Review 22, no. 1
(January).
Dunne, P. and R. Smith. 1993. “Thatcherism and the UK Industry.” In The Economic Legacy 1979-
1992, ed. J. Michie. San Diego: Academic Press.
Economist. 1999. “Trapped by the Bubble,” September 25.
Economic Policy Institute. 1999. “Economic Snapshots.” March 24, Washington, DC.
Edquist, C., ed. 1997. Systems of Innovation: Technologies, Institutions, and Organizations. Lon-
don: Pintner.
Ergas, H. 1987. “The Importance of Technology Policy.” In Economic Policy and Technological
Performance, ed. P. Dasgupta and P. Stoneman. Cambridge: Cambridge University Press.
European Commission. 1996. The European Aerospace Industry, 1996, Trading Position and Fig-
ures, Directorate - General III, Industry, Belgium.
Freeman, C. 1987. Technology Policy and Economic Performance: Lessons from Japan. London:
Pinter.
Freeman, C. and L. Soete. 1997. The Economics of Industrial Innovation. Cambridge: MIT Press.
Gally, R. and M. Teubal. 1997. “Paradigmatic Shifts in National Innovation Systems.” In Systems
of Innovation: Technologies, Institutions, and Organizations, ed. C. Edquist. London:
Wellington House.
Gompert, D.C., R.L. Kugler, and M.C. Libicki, 1999. Mind the Gap: Promoting a Transatlantic
Revolution in Military Affairs. Washington, DC: National Defense University Press.
Gregersen, B. 1992. “The Public Sector as a Pacer in National System of Innovation.” In National
Systems of Innovation: Towards a Theory of Innovation and Interactive Learning, ed. B.A.
Lundvall. London: Pinter.
Greider, W. 1998. One World Ready or Not: The Manic Logic of Global Capitalism, 2nd ed. New
York: Penguin Books.
Guttmann, R. 1994. How Credit-Money Shapes the Economy: The United States in a Global System.
Armonk, NY: M.E. Sharpe.
Henwood, D. 1997. Wall Street: How It Works and for Whom. London: Verso.
Hirst, P. and G. Thompson. 1996. Globalization in Question, Cambridge: Polity Press; Cambridge,
MA: Blackwell.
Hitchens, Theresa and Philip Finnegan. 1998. “One on One: Denis Ranque.” Defense News 13, no.
36 (September 7-13).
Jensen, M.C. and W.H. Meckling. 1976. “Theory of the Firm: Managerial Behavior, Agency Costs
and Ownership Structure.” Journal of Financial Economics 3, no. 4.
Journal Officiel. 1996. October 16.
Kline, S.J. and N. Rosenberg, 1986. “An Overview of Innovation.” In The Positive Sum Strategy:
Harnessing Technology for Economic Growth, National Academy Engineering. Washington
DC: National Academy Press.
Kleinknecht, A. and J. ter Vengel. 1998. “The Myth of Economic Globalization.” Cambridge Jour-
nal of Economics 22.
Kolodziej, E.A. 1987. Making and Marketing Arms: The French Experience and its Implications
for the International System. Princeton: Princeton University Press.
Latour, A. and D. Michaels. 1999. “Saab Offers Euros 579.7 Million for Defense Firm Celsius.”
Wall Street Journal, November 17.
95
Lovering, J. 1998. “Labour and the Defence Industry: Allies in ‘Globalization’.” Capital and Class
65 (Summer).
MacKenzie, C. 1998. “French Budget Plan Will Boost Defense Procurement.” Defense News,
September 28-October 4.
Markusen, A. 1997. “How we Lost the Peace Dividend.” American Prospect, July-August.
Markusen, A. and C. Serfati. 2000. “Remaking the Military Industrial Relationship: A French Amer-
ican Comparison.” Defence and Peace Economics 11, no. 3: 1-29.
Melman, S. 1997. “From Private to State Capitalism: How the Permanent War Economy Trans-
formed the Institutions of American Capitalism.” Journal of Economic Issues 31, 2 (June).
Morin, F. 1998. Le Modele Français de Détention et de Gestion du Capital. Rapport au Ministère
des Finances. Paris: Éditions de Bercy.
Mulholland, D. 1998. “Research and Development Spending.” Defense News, July 26–28.
Nelson, R.R. and N. Rosenberg. 1993. “Technical Innovation and National Systems.” In National
Innovation Systems: A Comparative Analysis, ed. R.R. Nelson. Oxford: Oxford University
Press.
OECD (Organisation for Economic Co-operation and Development). 1998. Main Science and Tech-
nology Indicators. Paris: OECD.
O'Sullivan, M. 1998. “Sustainable Prosperity, Corporate Governance, and Innovation in Europe.”
In Globalization, Growth, and Governance: Creating an Innovative Economy, ed. J. Michie
and J.G. Smith. New York: Oxford University Press.
Papon, P. 1978. Le pouvoir et la science en France. Paris: Le Centurion.
Peterson, W.C. 1989. “Market Power: The Missing Element in Keynesian Economics.” Journal of
Economic Issues 23, no. 2 (June).
Plender, J. 1998. “Revisiting of a Deadly Disease.” Financial Times, September 21, p. 23.
Quiles, P. and G.M. Chauveau. 1997. L'industrie française de défense: quel avenir? Assemblée
Nationale, Commission de la défense, Rapport d'information no. 203.
Sandler, T. and K. Hartley. 1999. The Political Economy of NATO: Past, Present, and Into the 21th
Century. New York: Cambridge University Press.
Scott, J. 1997. Corporate Business and Capitalist Classes. New York: Oxford University Press.
Serfati, C. 1992. “Le méso-système de l’armement et son impact sur le système productif de la
France.” Ares 13, no. 4.
———. 1995. Production d'armes, croissance et innovation. Paris: Economica.
———. 1999. “From One War (Irak) to the Next (Serbia): The New Look Arms Economy.” Paper
prepared for the COST A 10 workshop: Defense Restructuring and Questions of Identity,
Copenhagen, 10-11 September.
Serfati, C., ed. 2000. State-Companies Relationships in the Arms Industry: Between Change and
Continuity, Cost series, 2000.
Taylor, R. 1998. “Market Fallout Will Lift Jobless Total.” Financial Times, September 24, p. 8.
Veblen, T. 1963. The Engineers and the Price System, New York: Harcourt, Brace & World.
Walsh, V. 1995. “The Evaluation of Design.” International Journal of Technology Management 10,
no. 4/5/6.
Warner, J. 1999. “The Atlantic Century?” Business Week, February 8.
Weber, M. 1959 [1919]. Le savant et la politique. Paris: Plon.
Wolter, A., T. Putz, and H. Schmied. 1996. “Le développement inégal de la technologie des lasers
industriels en France et en Allemagne: Quelques enseignements pour la politique technolo-
gique.” Revue d'économie industrielle, no. 76, 2nd trimester.
The Place of the UK Defense Industry in its National Innovation System:
Co-evolution of National, Sectoral and Technological Systems
Andrew D. James*
Introduction
A purely national perspective is no longer appropriate for the study of the UK defense indus-
try. Indeed, it probably never has been. The institutions and relationships that contribute to the
generation, application, and commercialization of technologies used in the UK defense industry are
increasingly transnational in character, and they are likely to become more so in the decade ahead.
This chapter will argue that the UK defense industry is at the intersection of a number of dis-
tinct but overlapping innovation systems. The National Innovation System remains a key factor in
the strategy and competitive performance of the defense industry. However, the co-evolution of
National, Sectoral and Technological Systems increasingly shapes the context for innovation in the
UK defense industry. This chapter will consider the place of the UK defense industry at the
intersection of these innovation systems.
*
This chapter draws on on-going discussions on Systems of Innovation and the Defence Industry
within the METDAC network. METDAC (Managing European Technology: Defence and Competi-
tiveness Issues) is a thematic network funded by the Commission of the European Communities
under its Targeted Socio-Economic Research (TSER) program. Thanks go to my colleague Philip
Gummett for allowing me to use parts of our recent co-authored work as the basis of some sections
of this chapter. I would also like to thank Stan Metcalfe for his extremely useful comments on earlier
drafts and colleagues at the ESRC Centre for Research on Innovation and Competition, University
of Manchester for their thought-provoking work on the innovation systems perspective.
96
97
Such an approach places particular emphasis on three principal components of the system:
the institutions and organizations that contribute to the production, diffusion, and use of new knowl-
edge, which may include firms (including suppliers, users and competitors), universities, public and
private research organizations, and the government. The relationships among those elements and
their interdependence and interaction within the system, which may be formal or informal and may
be based on market or non-market means. The boundary of the system, which may be national but
may equally be regional or international in scope.
Three major variants of the systems of innovation literature can be identified (Andersen et
al. 1998); National Innovation Systems (NIS); Sectoral Innovation Systems (SIS); and Technologi-
cal Systems (TS). I will (briefly) consider their main characteristics.
Sectoral Systems
In response to some of the concerns about the national focus of much of the early innovation
systems literature, another perspective has begun to emerge that focuses on Sectoral Systems
98
(Breschi and Malerba 1997; Andersen et al. 1998). Breschi and Malerba define a Sectoral Innovation
System as:
. . . that system (group) of firms active in developing and making a sector’s products
and in generating and utilizing a sector’s technologies; such a system of firms is
related in two different ways: through processes of interaction and cooperation in
artefact-technology development and through processes of competition and selection
in innovative and market activities (Breschi and Malerba 1997, 131).
Note that, while the NIS approach takes the geographical boundary of the system as given,
the SIS perspective sees it as endogenous and emerging from the specific conditions of each sector.
Consequently, different sectors may have different boundaries that are not necessarily national but
may spill over national boundaries through processes of cooperation, competition, and transnational
knowledge transfer.
Technological Systems
A third approach focuses on specific technologies rather than the national or sectoral systems
in which they may develop. Carlsson and Stankiewicz state that:
A technological system may be defined as a network of agents interacting in a spe-
cific economic/industrial area under a particular institutional infrastructure or set
of infrastructures and involved in the generation, diffusion and utilization of technol-
ogy. Technological systems are defined in terms of knowledge/ competence flows
rather than flows of ordinary goods and services. They consist of dynamic knowledge
and competence networks (Carlsson and Stankiewicz 1995, 49).
As such, this Technological Systems approach is principally concerned with the knowledge and
competence networks surrounding a particular technology, rather than the broader innovation pro-
cess in which that technology is turned into commercial products. The boundaries of these knowl-
edge and competence networks are likely to vary depending on the particular technology in question.
However, with the increasing internationalization of scientific and technological knowledge, a
Technological System may well be transnational in scope.
99
1
The notion of co-evolution as it might apply to innovation systems is discussed by Andersen et al.
(1998).
100
sequently, Andersen et al. (1998) call for more comparative and historical work on similar Sectoral
Systems operating in different national contexts.
The defense sector would seem an appropriate choice. It has framed the context for innova-
tion in many countries in the post-War period. At the same time, we already know that there is con-
siderable institutional variety between National Innovation Systems (Gummett and Stein 1997). The
METDAC Network funded by the European Commission is engaged in some comparative work of
this kind compiling a series of case studies of the defense industries in Europe, the U.S. and Japan.
The Cornell Workshop is to be welcomed for taking the process of transnational comparison a stage
further. This is clearly a rich vein for comparative empirical work.
At the same time, the innovation systems approach may contribute to an improved under-
standing of the nature of the defense industry and its relationship with National Systems. The
perspective raises a number of interesting questions: How does the division of labor operate in the
generation and application of knowledge for defense-related objectives? What are the relationships
between National, Sectoral, and Technical Systems in the generation and application of such
knowledge? What is the boundary around the system, and is a national perspective still appropriate
when analyzing the defense industry?
2
The drivers of this consolidation process and the main features of the U.S. defense industry struc-
ture that has emerged are described in greater detail in James (1998).
101
defense or exited the business altogether. The consolidation process has led to major changes in the
structure of the U.S. defense industry and the industry structure that has emerged is characterized
by considerable concentration. Thus, consolidation has seen the emergence of three giant defense
companies—Lockheed-Martin, Boeing, and Raytheon—whose overall size and breadth of business
activities dwarf their competitors in the United States and—until recently—Europe (see Table 1 ).
Table 1
Top 10 Firms in 1998 by Defense Sales (U.S. unless otherwise stated)
3
In this classification, General Dynamics and Northrop Grumman present some problems. While
considerably smaller than the three leading U.S. companies in terms of total defense sales, they are
much larger than the next ranked U.S. company (Litton Industries). Indeed, it will be noted that in
a previous FOA report by this author, Northrop Grumman was classified alongside Lockheed
Martin, Boeing, and Raytheon (James 1998). Changes in the U.S. industry structure since that report
mean that Northrop Grumman now has more common with the medium-sized companies than the
three giants.
102
giants and the new BAe Systems, but they are considerably larger than European companies such
as Finmeccanica, Smith Industries, and Saab-Celsius. In the defense industry supply chain, these
medium-sized companies act as prime contractors and platform manufacturers, as well as tier-one
suppliers of sub-systems.4 Thus, United Technologies manufactures helicopters and Northrop Grum-
man is a manufacturer of military aircraft. Litton Industries, Newport News, and General Dynamics
manufacture naval combat ships. General Dynamics is also the only U.S. manufacturer of main
battle tanks. Indeed, it is worth emphasizing at this point that these companies tend to be diversified
defense contractors with a broad range of capabilities, and very few of them focus exclusively on
defense electronics. Below these medium-sized companies are a range of smaller second- and third-
tier companies who supply components and sub-systems to the prime contractors and tier-one
suppliers.
4
By tier-one companies is meant those who supply complete systems and integration skills directly
to the prime contractors. Tier-two companies tend to supply sub-systems and components rather than
integration services. Their customers are both the tier-one companies and the prime contractors. For
a discussion of the defense industry supply chain, see James et al. (1998).
103
Background
The UK’s defense industrial capabilities mean that it is one of very few countries that can
design, manufacture, integrate, and market complete sea, land, and air-based systems: fixed and
rotary wing aircraft, aero-engines, warships and submarines, air-to-air and air-to-surface missiles,
low level air defense, field guns, and military land vehicles.
Since the 1980s, the defense industry has been privately owned. The two leading defense
companies are British Aerospace and GEC. British Aerospace is the leading defense and aerospace
company in Europe. A manufacturer of military aircraft, missiles, and munitions, it is the fourth
largest defense company in the world, although considerably smaller than the leading U.S. com-
panies. GEC, a large diversified industrial group, has major interests in defense electronics and war-
ship and submarine building. In addition, the UK has a strong and diverse group of companies sup-
plying sub-systems and components. These include aero-engine manufacturer Rolls-Royce as well
as a sizeable number of smaller suppliers and sub-contractors (James et al. 1998).
The defense industry continues to represent an important part of UK manufacturing activity.
In 1996, the industry employed 310,000 people (Bonn International Center for Conversion 1998).
This represents roughly 7 percent of the UK manufacturing workforce and does not include indirect
employment effects, which are estimated at about 300,000 jobs. The defense sector also makes an
important contribution to UK exports. Exports in 1997 increased 10 percent on the previous year to
stand at £5.5 billion (6.7 billion ECU) maintaining the UK’s position as the second largest defense
exporter after the United States (Jane’s Defence Weekly 1998). Sixty percent of aerospace produc-
tion is exported—the highest export ratio of any UK industrial sector—and British Aerospace, Rolls-
Royce and GEC are consistently in the top five UK exporters (Cabinet Office 1995).
A Wall Of Separation
Historically, scientific and technological development for defense purposes has been organi-
zationally (and often physically) separate from civil science and technology, both in terms of its
institutions (with dedicated defense research establishments and defense divisions of multi-business
104
While most UK defense companies are in fact defense divisions of larger multi-business
companies, military and civil production has historically been organizationally and physically sepa-
rate. The “wall of separation” that has characterized many U.S. defense companies has been just as
high in the UK (Markusen and Yudken 1992). Technology transfer between defense and civil divi-
sions has been limited by security concerns, differences in organizational culture, but also (particu-
larly in the case of GEC) internal accounting and budget procedures that have reduced incentives
for interdivision knowledge sharing. The situation with regard to smaller suppliers is rather different,
and there is considerable evidence of dual-use technological activity among first and second tier
suppliers (James et al. 1998).
5
The influence of the scale and organization of UK defense technological activities on the innova-
tive and competitive performance of the national economy has been the subject of considerable (and
unresolved) debate. This is not the focus of this chapter. However, it is worth noting that the terms
of the debate have been similar to that in the United States, with concerns about “crowding-out,” the
extent (or lack) of spill-overs from defense to non-defense sectors and the implications for the
international competitiveness of defense-dependent companies and sectors. A series of reports
during the 1980s, including the 1983 Maddocks Report to the Electronics Economic Development
Council, the 1986 report of the Council for Science and Society, and the 1989 report by the Cabinet
Office’s Advisory Council on Science and Technology (ACOST), expressed concern at the limited
civil benefit obtained from defense R&D spending. The ACOST report estimated that less than 20
percent of the Ministry of Defence’s R&D expenditure was likely to have any applicability in the
civil sector, which was attributed to the defense-specific nature of many of the technologies funded,
combined with the low spin-off potential of much of the work conducted during the costly
engineering development of military systems albeit carried out under the heading of R&D. At the
firm-level, the literature on the defense industry has long noted that the relationship between defense
companies and their customers has had an important influence upon the character of those
companies and, in particular, their competitiveness in non-defense sectors.
105
1990s, with criticism of the Ministry of Defence for its failure to recognize its responsibility to
ensure the health of the defense industrial base.
One of the key legacies of the Thatcher era, as far as the UK defense industry is concerned,
was the privatization of state-owned defense companies. I have already noted that the leading
defense companies were privatized during the 1980s, making the UK defense industry very different
from many others in Europe where defense companies remain in state ownership. There is little
doubt that this has placed the UK defense industry in a strong position in the European context.
Privatization has meant that UK companies have been freed of many of the political constraints on
national level rationalization that have limited restructuring in some other European countries.
Consequently, restructuring of the UK defense industry began rather earlier than in most other
European countries. Most significantly for European restructuring, in France, the consolidation
process has been inhibited by the structure of the defense sector and in particular the role of social,
industrial, and economic considerations. This has resulted in what Hébert has described as
“contained conversion,” with the French state seeking to delay key decisions on the defense industry
for as long as possible (Hébert and de Penanros 1995).
At the same time, privatization decisions and competition in the procurement process have
also made UK defense industry more competitive than some other parts of Europe, and UK defense
contractors such as British Aerospace are generally regarded as being at the leading edge of organi-
zational and process innovation in such areas as supply chain management. Thus, a 1998 Aviation
Week and Space Technology survey identified British Aerospace and Smiths Industries as being
among the best managed companies in the industry (Velocci 1998).
The UK has also been a prime mover at the government level in encouraging the consolida-
tion of the European defense industry. In part, this is based on a belief that the combination of profit-
able private sector companies, strong export performance, and access to both European and U.S.
markets, means that the UK is well placed to benefit from European restructuring.
To some extent, the dynamics of European rationalization match UK market power against
the political power of France through its relationship with Germany and the portrayal of the UK de-
fense industry as a “Trojan horse” for U.S. interests. Certainly, as a 1998 policy document from the
Society of British Aerospace Companies makes clear, UK defense companies recognize the impor-
tance of the UK government in ensuring the outcome of the rationalization process is to the benefit
of UK companies and in blocking any emergence of a “Fortress Europe” that would inevitably
107
damage UK companies with their strong transatlantic relationships (Society of British Aerospace
Companies 1998a). At the same time, the UK defense industry is also pressing hard for government
support for European Union level funding of aerospace technology and the development of a com-
mon European weapons acquisition process and a more open and competitive European defense
equipment market.
The UK government has long supported defense exports both through financial support and
diplomatic efforts. Such support increased markedly in the 1980s. However, the election of the new
Labour Government in 1997, with its commitment to an “ethical foreign policy” and stricter controls
on the arms trade, has caused some tensions with the defense industry. While the new Government
has made it clear that it continues to support arms exports, some defense exporters have claimed that
new regulations have slowed the export licensing process and threaten UK competitiveness in some
export markets.
reform of the Ministry of Defence’s procurement process and, in particular, a shift towards more
competitive Commercial Off-The-Shelf (COTS) procurement of defense equipment.
Consequently, the 1995 Defence and Aerospace Foresight Report noted that declining gov-
ernment and industry R&D expenditure threatened UK competitiveness and called on industry and
government to reverse the decline in UK spending on research and technology demonstration.6
6
This report was published as part of the wider UK Foresight exercise that sought, through sectoral
Panels comprised of representatives of government, industry and the scientific and engineering com-
munities, to inform the Government’s decisions and priorities in the area of science, engineering and
technology. The fact that Defence and Aerospace had its own Panel is significant in itself, and says
much about their importance in the UK National Innovation System. The inclusion of defense and
aerospace as a specific area of attention made the UK Foresight exercise different from its German
and Japanese counterparts (James and Gummett 1998).
109
result was a toughening of the competitive environment for defense firms, and a growing volume
of criticism of the Ministry of Defence by industrialists for its alleged failure to realize the need for
partnership and stability in the long-term development and maintenance of technological capabil-
ities.
Whether such criticism was “special pleading” by an industry long used to a cosy relation-
ship with a monopsonistic customer or reflected real problems within the defense innovation system
is difficult to test empirically. However, it is clear that, in the perception of industry, the old rela-
tions of trust and cooperation between themselves and the Ministry of Defence had been replaced
by a more straightforwardly commercial relationship. Criticisms have been raised that this emphasis
on “value-for-money” paid little attention to concerns about its implications for UK defense indus-
trial capabilities.
A series of official reports during the mid 1990s reflected industry concerns about the pro-
curement system, raising two broad concerns. First, that the procurement system defined defense
considerations and interpreted value-for-money too narrowly and failed to give sufficient weight to
the wider technological and industrial implications of its procurement decisions. Thus, the Defence
and Aerospace Foresight Report warned that the UK’s leadership in the trend towards more off-the-
shelf purchases posed real threats to the medium to long-term capability of the UK defense technol-
ogy base (Cabinet Office 1995). Similarly, the Ministry of Defence’s lack of formal responsibility
to foster the strength of the defense industrial base or wealth creation, was noted by the 1995 joint
report of the House of Commons Defence and Trade and Industry Select Committees (House of
Commons 1995).
These reports also expressed concerns that foreign (for this, read American) firms had unfair
advantages over UK firms in the UK procurement process. They argued that changes in government
R&D funding meant that UK firms were expected to fund an increasing proportion of R&D expendi-
ture, while their major competitors in France, Germany and the United States were benefitting from
increasing government R&D funding. The size of the U.S. market meant that U.S. companies could
benefit from the lower unit costs of longer production runs and benefit from government R&D sup-
port. The Foresight Report warned that this meant that equipment based on UK technologies would
be replaced by equipment where UK companies would act as partners to U.S. companies, putting
at risk the UK’s technological capability and its ability to participate in future collaborative projects
or compete in the world market (Cabinet Office 1995).
110
In the context of these mounting concerns, the Ministry of Defence has, in recent years, sig-
nalled a shift towards a more explicit and public recognition of its role in the maintenance of na-
tional defense industrial capabilities. This changing attitude has been reflected in recent statements
on procurement policy. Take, for example, the 1996 Statement on the Defence Estimates which
comments:
We recognise the need to take defence industrial factors fully into account in our
decision-making. . . . Where relevant, we consider the defence case for seeking to
retain particular United Kingdom-based defence industrial capabilities. . . . (MoD
1996).
important part of the UK National Innovation System. Approximately one-third of the government’s
defense R&D budget is spent in these research establishments, and the Defence Evaluation and
Research Agency is the largest research organization in Europe, with more than 10,000 employees.
These defense research establishments have experienced considerable changes during the 1990s both
in their internal organization and their relationships with the Ministry of Defence and defense
companies.
The Ministry of Defence’s own research establishments comprise the Atomic Weapons
Establishment (AWE) and the Defence Evaluation and Research Agency (DERA). AWE has respon-
sibility for the design, development, and manufacture of the UK’s independent nuclear deterrent.
DERA supplies scientific and technical services to Ministry of Defence in the areas of strategic
research, applied research, operational assessments and studies, project support, the formulation of
operational requirements, and equipment testing services and quality assurance. It does this partly
through its own laboratories and test facilities, and partly by placing contracts with industry and uni-
versities. It also provides scientific and technical services to other government departments and to
other public and private sector customers where this supports the achievement of its main objectives.
Both organizations have experienced significant organizational changes. In 1993/94 manage-
ment of the Atomic Weapons Establishment became the responsibility of a private contractor—
Hunting-BRAE. DERA was established (in its earlier form as DRA) in 1991 as an Executive Agency
of the MoD. This was part of a wider initiative to move civil service organizations into a more com-
mercial relationship with government, while still leaving them under ministerial control. The aim
was to improve performance by introducing a more commercial style of management, freed from
traditional civil service constraints. In 1998 the government announced its intention to explore the
possibility of some form of Public Private Partnership for parts of DERA, which would introduce
private sector participation in the ownership of the organization and would mark an even more
fundamental shift in its relationship with the Ministry of Defence (MoD 1998). These privatization
plans have run into significant political difficulties, however. The proposals were heavily criticized
in a report by the House of Commons Defence Select Committee. At the same time, the U.S. Depart-
ment of Defense has expressed serious reservations about its implications for the long established
relationships between UK and U.S. government defense research establishments.
The changes in DERA have altered its relationship with defense companies as well as with
the Ministry of Defence. The defense research establishments and large UK defense companies have
112
traditionally had a close working relationship. DERA subcontracts work to industry and academic
institutions worth between 30 percent and 50 percent of the organization’s annual research funding
and in 1996-97 such extramural research cost DERA £162 million (House of Commons 1998). This
is DERA’s principal mechanism for involving industry in defense research and an important mech-
anism for transferring the benefits of its work to the defense industry.
The move towards greater commercialization of DERA’s activities has generated some ten-
sion in this relationship. A 1995 joint report by the House of Commons Defence and Trade and
Industry Select Committees noted that companies had three broad concerns about DERA: whether
enough of its research was contracted out to industry; whether it duplicated industry’s own research;
and the availability of DERA’s research to companies (House of Commons 1995). Indeed, DERA’s
Chief Executive himself noted in 1996 that “There is deep unease at the working level in UK de-
fense industry about DERA and its objectives” (DERA 1996, p. 7). Such concerns remain, and were
the subject of comment in the House of Commons Defence Select Committee’s 1998 report on
DERA. The Report noted that a recent DERA survey had found that most industry representatives
considered that the organization’s pursuit of non-Ministry of Defence income had interfered with
its relations with industry, and a fifth of the respondents considered that relations with DERA had
been severely damaged by the introduction of competition between the Agency and the industry
(House of Commons 1998). Clearly, any problems in the relationship between a key technological
resource and companies has potentially important implications for the technological performance
of the UK defense industry.
there is the possibility of a gradual loss of market share as its technologies are sur-
passed by competitors and ageing products fail to be replaced. The damage would
be done before it became obvious, and would probably be irreversible . . . there could
be relative or even absolute decline in the British aerospace industry which reflected
not market forces but the fact that competitors received greater assistance from their
governments (House of Commons 1993, paras. 130-31).
Similarly, the 1995 report of the Technology Foresight Defence and Aerospace Panel con-
tains a strong warning about the UK’s technological competitiveness. The Foresight Panel observed
that the success of the defense and aerospace industry derived largely from the previous levels of
investment by industry and government in research, development, and procurement in the 1970s and
1980s, and the Panel report identified
[T]he concern that the UK today is consuming its technological inheritance. Current
UK success derives from past farsighted R&D programmes, particularly in govern-
ment establishments and in industry, which have yielded world class technology. The
UK must continue to make sufficient investment in key technologies and in technol-
ogy demonstrators for continued success. This will require a national strategy and
concerted investment in new technology (Cabinet Office 1995, pp. 18-9).
Many of these concerns have been framed in terms of the implications for the competitive-
ness of the UK defense industry in UK and export markets and as a partner in future collaborative
projects. Thus, considerable emphasis has been placed on the importance of the UK sustaining and
enhancing the key technological capabilities necessary to win a leading position in future collabora-
tive programs, whether those are with U.S. or European partners. In this respect, above all others,
the National Innovation System remains a key factor in the strategy and competitive performance
of the defense industry.
However, a purely national perspective is no longer appropriate to the study of the UK de-
fense industry. Indeed, it probably never has been in the modern age. The remainder of this chapter
considers the role of the UK defense industry in transnational Sectoral and Technological Systems
and how the interaction of National, Sectoral and Technological Systems influence the context for
innovation in the UK defense industry.
try are increasingly transnational in character. In this respect, the UK is little different than many
other countries (including the United States), as countries have become more interdependent in
defense-related technologies (Edmonds et al. 1990).
While the UK defense industry has a broad range of defense industrial and technological
capabilities, the UK accepted in the 1950s that self-sufficiency was beyond the resources of the
country (Edmonds et al. 1990). Subsequent policies have encouraged weapons collaboration with
European partners and close relationships with the United States in some technologies. At the same
time, the UK has long imported some weapons systems, particularly from the United States. In this
vein, the Defence and Aerospace Foresight Report noted that the UK could not lead across the whole
spectrum of defense and aerospace, but needed to focus its strengths in the sub-sectors that are
important for national security or where there is significant potential for wealth creation (Cabinet
Office 1995).
A number of economic, technological and political factors are driving this trend towards
growing interaction and interdependence:
First, the cost of developing weapons systems and the technologies that underpin them is es-
calating; few countries (with the exception perhaps of the United States) have military markets that
are sufficiently large for the cost-effective design, development, and manufacture of such systems.
Second, technological change and complexity means that individual countries find it increas-
ingly difficult to keep abreast with technological developments across the whole spectrum of mili-
tary equipment.
Third, technologies are increasingly dual-use: advanced technologies of civil origin may be
required in defense equipment, and scientific and technological knowledge is increasingly interna-
tional, and the innovation process has an increasingly distributed character.
Finally, political factors are encouraging closer integration within Europe, both in research
and technological development and (more controversially) in the fields of defense and security. The
Western European Union/Western European Armaments Group is seeking to encourage increased
European defense industrial cooperation and a European equipment market, and has responsibility
for the EUCLID cooperative research and technology program. At the same time, OCCAR (Organi-
sation de Coopération Conjointe en Matiere d’Armement) is seeking to develop common procure-
ment methods for Europe that will take both less account than in the past of juste retour and more
account of managerial effectiveness, and also an institution that may ultimately be merged into a
115
European armaments agency. However, the significance of many of these actions is as much sym-
bolic as anything else; there are formidable political barriers at the national government level to
closer cooperation in the fields of defense and European security.
Similarly, various efforts by the European Union have been constrained by national political
factors. The European Commission has expressed its opinion on matters related to the restructuring
of the defense-related industries in two recent Communications (Commission of the European Com-
munities 1996 and 1997). At the same time, while the Community’s Framework Programme for
Research and Technological Development is aimed at civil objectives, it can support the develop-
ment of civilian applications of technologies of defense origin through technology transfer and R&D
cooperation between civil and defense-related organizations. Defense-related organizations such as
British Aerospace, Rolls-Royce, DASA, and DERA are active participants in Community programs,
and the European Commission has estimated that between one-quarter and one-third of Framework
Programme funding could be described as dual-use.
7
In the remainder of this chapter, I follow the distinction drawn in parts of the innovation systems
literature between Sectoral Systems (with their emphasis on that group of firms and other organiza-
tions active in developing and making a sector’s products and utilizing its technologies) and Tech-
nological Systems (with their emphasis on dynamic knowledge and competence networks). In prac-
tice, the distinction is rather hard to draw in some circumstances. Empirically, the content of rela-
tionships is not always clear. Theoretically, the definitions overlap in some respects, not least
because the specific technologies that are the focus of the Technological System perspective will
also be utilized within the Sectoral System. However, bearing these caveats in mind, the distinction
seems worth drawing, not least because it emphasizes that the innovation process involves much
more than technology alone, and that knowledge/competence flows may be rather different from the
flow of goods (in this case, defense equipment).
116
ness as a Sectoral System comes from its product—defense equipment—and its customers—the
Ministries of Defence of national governments. The Sectoral System gains its coherence through the
inter-relationships of firms through both competition and collaboration.
Note that I regard defense-related firms to be the main institutional components of the Sec-
toral System. These firms may be prime contractors, but may equally be suppliers of sub-systems
and components. These firms may be defense specialized but (especially at the supplier level) they
are also increasingly dual-use (James et al. 1998).
Competition as well as cooperation structures the Defence Sectoral Innovation System. Thus,
competition among defense firms influences the context for innovation and the strategies of individ-
ual firms. Clearly, a central concern for UK companies is how they should respond to the growing
dominance of U.S. defense companies. Of course, what has emerged from the restructuring process
is a defense sector where a fragmented European industry faces a number of very large U.S. com-
panies. Once post-merger integration is completed, these companies are likely to achieve competi-
tive advantage from economies of scale and scope with regards to production and the development
of technology. Notwithstanding the barriers to integration at the European level, the threat posed by
the U.S. defense industry is increasingly driving European companies to seek to consolidate. At the
same time, UK companies more than any others in Europe, have sought to collaborate with the U.S.
giants by building on existing transatlantic relationships.
UK defense companies face competition not only in export markets but also in their domestic
markets. There is competition, as I have already noted, in the procurement system for defense equip-
ment in the UK and this drives the strategies of UK defense companies. Competition has forced
them to reduce costs and also to seek collaborative relationships with companies who can provide
access to off-the-shelf technologies, thus reducing development costs. In practice, that frequently
means U.S. companies.
Bilateral and multilateral consortia to develop major weapons systems have been a common
form of transnational relationship for a number of decades. European consortia have tended to arise
out of intergovernmental agreement and have typically been based on the juste retour principle of
work sharing among partners. In military aircraft, the UK and France cooperated from the 1960s to
develop the Jaguar ground-attack aircraft; in the 1970s a UK-German-Italian consortium developed
the Tornado; and Eurofighter is a consortium of companies from the UK, Germany, Italy and Spain.
In naval systems, the Horizon next-generation frigate is being developed by a UK, French, and
Italian consortium. At the same time, UK companies have close relationships with U.S. companies.
British Aerospace has had a long standing cooperation agreement with McDonnell-Douglas
(Boeing) and is teamed with Lockheed-Martin on the Joint Strike Fighter program.
European firms have encountered important barriers to transnational mergers and acquisi-
tions. These barriers have included state ownership of some of Europe’s largest companies, varying
corporate legal frameworks, and significant differences in production costs between national defense
industries and defense contractors. At the same time, defense continues to be viewed as a sector of
strategic significance by national governments, who therefore actively scrutinize foreign interven-
tions (Commission of the European Communities 1997; Gummett and Stein 1997; Lewis and Starr
1997).
Nevertheless, the situation is changing. The December 1997 joint statement by the govern-
ments of the UK, France, and Germany, which asked their national aerospace and defense industries
to draw up plans for restructuring, suggested that national governments, having long lagged com-
panies regarding these matters, had accepted the importance of such changes (Lewis 1997). The im-
minent formation of EADS, along with a revised corporate organization for Airbus Industries, is the
outcome of this new found national will to achieve restructuring of the European aerospace industry.
UK companies already have some limited stakes in other European firms. British Aerospace
recently took a one-third stake in SAAB as an expansion of an earlier marketing and production ar-
rangement with the Swedish company. In 1997, Matra-British Aerospace Dynamics, a joint venture
between France’s Lagardere and British Aerospace, took a 30 percent stake in DASA’s LFK missile
subsidiary, and British Aerospace and DASA acquired the defense electronics assets of Siemens.
UK companies also have significant interests in the United States. GEC has acquired Tracor,
the U.S. defense electronics company, to add to its existing U.S. defense electronics interests and
has made approaches to Northrop-Grumman. Rolls-Royce acquired the Allison Engines Corp. in the
118
mid-1990s. Indeed, a recent industry survey estimated that UK-owned aerospace assets overseas
(defense and civil) amounted to £2.3 billion, employing more than 25,000 people. This makes the
UK aerospace industry one of the most globalized in the world (Society of British Aerospace Com-
panies 1998b).
At the same time, it is worth noting that there is a small but significant degree of foreign
ownership of UK defense companies. Issues related to the integrity of the defense industrial base
have led the government to restrict the share of foreign ownership in major privatized defense con-
tractors such as British Aerospace, Rolls-Royce and VSEL. Foreign firms have, however, been al-
lowed to buy into the UK defense industry. Examples include Bombardier of Canada, which controls
Shorts; Raytheon Systems; and, most controversially at the time, United Technologies, which took
control of Westland Helicopters in 1986 (subsequently selling the business to the UK’s GKN).
The considerable political constraints on cross-border ownership means that, in Europe at
least, joint ventures have so far been the main mechanism for consolidation. Thus, Matra-British
Aerospace Dynamics is a joint venture between France’s Lagardere and British Aerospace. GEC has
formed a joint venture (Matra Marconi Espace) with Matra over space interests, with Thomson of
France over sonar (GEC Thomson Sonar), and with Italy’s Finmeccanica. Many of these relation-
ships between UK and European defense companies have been driven by concerns about work-
sharing or economies of scale in production. However, it is also worth noting British Aerospace’s
technology demonstration program with Dassault as an example of a collaboration driven principally
by technological motives and to emphasize that many relationships established for other motives
develop an increasing technology component (for example, the British Aerospace-SAAB relation-
ship).
UK firms also participate in the wider Sectoral System through supply chain relationships.
Thus, Edmonds et al. (1990) have noted the extent of interdependence between countries through
the supply of specialized systems and components. UK prime contractors rely heavily on U.S. sup-
pliers of specialized sub-systems, although the extent of this dependence is hard to quantify.
Certainly, the UK defense electronics industry is largely dependent upon overseas companies for
components, and the Ministry of Defence places few restrictions on the use of foreign components
and sub-systems in “British” defense equipment, preferring instead to devolve decisions to prime
contractors with the key emphasis being on cost-competitiveness and not nationality of ownership
(Taylor and Hayward 1989). Interestingly, the impression is that these supply chain relationships
119
are perhaps better developed than those within Europe, where there is surprisingly little evidence
of a single market in defense-related components (James et al. 1998).
Technological Systems
The focus of this section will be on Technological Systems, those dynamic knowledge and
competence networks involved in the generation, diffusion, and utilization of specific technologies.
These technologies are used within the Defence Sectoral Innovation System in the design, develop-
ment, and manufacture of defense equipment. Such Technological Systems may be comprised pri-
marily of actors from the Sectoral System, but the emergence of dual-use technologies means that
they are increasingly likely to be of civil origin.
The UK defense industry has long participated in transnational knowledge and competence
networks. In part, this is a result of the recognition on the part of the UK government that certain
classes of defense technologies were simply too costly to develop alone. More recently, it reflects
the emergence of dual-use technologies that have their origins outside the Defence Sectoral Innova-
tion System.
Ministry of Defence policy sees international collaboration as attractive in reducing costs. The Min-
istry of Defence currently has collaborative research programs with 20 countries to which it commits
10 percent of its research resources, and effort is being directed to increasing this proportion. Inter-
national collaboration is seen by the Ministry of Defence as a good way of promoting technology
exchange, leveraging greater returns on investment, and providing wider benefits from exposure to
multinational debate on defense research needs and key technologies as well as issues surrounding
arms and export controls.
Among the bilateral and multilateral research programs in which the Ministry of Defence
participates are the Anglo-French Defence Research Group, the Technical Cooperation Programme
(with the United States, Australia, Canada and New Zealand), the NATO Defense Research Group,
and the Western European Armaments Group’s European Cooperation for the Long-term in Defence
(EUCLID) program. In addition, the Ministry of Defence has also collaborated with the United
States since 1985 on technologies related to Ballistic Missile Defense (DTI/OST 1996a; MoD 1996).
In a recent initiative, the United States, the UK, France, and Germany have established committees
to consider new ways of cooperating on international arms research, development, and procurement.
The UK is heading a panel on sea systems and the United States a committee on space and com-
mand, control, communications, computers, and intelligence/surveillance. France chairs a committee
on air systems, and Germany leads on land systems. The goal is to prioritize future military needs
in each area and determine how best to cooperate in developing systems to meet those needs. If
successful, the initiative may be expanded to include the participation of other NATO countries
(Jane’s Defence Weekly 1997). Reflecting the changes in the international environment, there are
on-going discussions on defense technology with South Africa and a number of Eastern European
states as well as a recently signed defense research and technology agreement with Sweden.
Dual-Use
Leadership in the development, design, and manufacturing of many key technologies with
defense applications now resides not only with the government or the traditional defense industrial
base, but also in commercial industry. With the emergence of these dual-use technologies and the
decline in defense-related R&D expenditure, technology acquisition, especially at the sub-compo-
nent level, is increasingly coming from the commercial sector. Defense companies have retained in-
house design and production capabilities for the most sensitive and defense-specific components,
121
but they increasingly find themselves part of wider Technological Systems that extend beyond the
defense sector. Thus, most semiconductors used in UK defense systems are sourced from commer-
cial suppliers and many are imported, from the United States or South East Asia (Edmonds et al.
1998). The Foresight Defence and Aerospace Report noted that in technological terms the UK is lag-
ging in areas, such as off-the-shelf electronic components, in which off-shore suppliers dominate
the market (Cabinet Office 1995).
At the policy level, collaboration with the civil sector is seen as increasingly attractive by
the Ministry of Defence, as it can encourage the use of dual-use technologies in defense applications.
Interestingly, the Government’s recent discussion (Green) paper on its proposed Defence Diversifi-
cation Agency within DERA suggests that it will pay as much attention to the scope for the spin-in
of technology from the civil sector as it will the spin-out of defense origin technologies to new civil
applications.
Procurement Reform
The changes in the procurement system brought about by the Levene reforms provides a
good example of co-evolution. In this case, change in the National Innovation System altered the
competitive landscape of the Defence Sectoral Innovation System. UK defense companies, as we
have seen, were forced to adapt their behavior to the new environment. They sought to reduce their
production costs, increase dual-use technology acquisition, and develop collaborative bid strategies
with foreign partner companies. At the same time, the emphasis on value-for-money and open com-
petition presented new competitive opportunities for non-UK (principally U.S.) defense contractors.
122
Dual-use Technologies
A final example is provided by the emergence of dual-use technologies. As we know, the rate
of technological change in some Technological Systems dominated by civil users has meant that
technological leadership in some fields has shifted from defense to civil-origin technologies. The
consequence of these changes within Technological Systems has been to prompt changes in the
National Innovation System as the Ministry of Defence and DERA have sought to spin-in technol-
ogies of civil origin. At the same time, companies within the Defence Sectoral Innovation System
have sought to adapt their approach to technology acquisition.
Inevitably, these examples provide a rather crude “story” of what are complex issues. My
objective has been to try to show that the UK defense industry is at the intersection of three distinct
but overlapping systems. Each system influences the context for innovation in the defense industry
and the strategy and competitive performance of defense companies. I believe that this perspective
has some merit, although it requires further development.
References Cited
Advisory Council on Science and Technology. 1989. Defence R&D: A National Resource. London:
HMSO.
Andersen, B., J.S. Metcalfe, and B.S. Tether. 1998. “Innovation Systems as Instituted Economic
Processes,” Paper prepared for the Thorstein Veblen Conference on Institutional Analysis
of the Economy Today, Oslo, June.
Barrie, D. and T. Hitchens. 1998. “Stealth Talks May Hold Key to Britain’s FOAS.” Defense News
(August 31-September 6): 3-44.
Bonn International Center for Conversion. 1998. Conversion Survey 1998. Oxford University Press:
Oxford.
123
Breschi, S. and F. Malerba. 1997. “Sectoral Innovation Systems: Technological Regimes, Schumpe-
terian Dynamics, and Spatial Boundaries.” In Systems of Innovation: Technology, Institu-
tions and Organizations, ed. C. Edquist. London: Pinter.
Cabinet Office. 1995. Progress Through Partnership, 12: Defence and Aerospace. London: HMSO.
Carlsson, B., ed. 1995. Technological Systems and Economic Performance, Kluwer: Dordrecht.
Carlsson, B. and R. Stankiewicz. 1995. “On the Nature, Function and Composition of Technological
Systems.” In Technological Systems and Economic Performance, ed. B. Carlsson. Kluwer:
Dordrecht.
Commission of the European Communities. 1996. “The Challenges Facing the European Defence-
Related Industry: A Contribution for Action at European Level.” Communication from the
Commission, COM(96) 10 final. Brussels: Commission of the European Communities.
Commission of the European Communities. 1997. “Implementing European Union Strategy on
Defence-Related Industries.” Communication from the Commission, COM(97) 583 final.
Brussels: Commission of the European Communities.
Cook, N. 1998. “BAe Seeks Delay to Future UK Aircraft Decision.” Jane’s Defence Weekly, 2
September, p. 3.
Council for Science and Society. 1986. UK Military R&D. Oxford: Oxford University Press.
Defense News. 1999. “Defense News Top 100.” Defense News, July 26, p. 8.
DERA. 1996. Defence Evaluation and Research Agency Annual Report and Accounts 1995/96
London: HMSO.
Department of Trade and Industry, Office of Science and Technology. 1996a. Forward Look of
Government-funded Science, Engineering and Technology 1996, Cm 3257-I. London:
HMSO.
———. 1996b. Science, Engineering and Technology Statistics 1996, Cm3257-II. London: HMSO.
The Economist. 1997. “A Shrinking Arms Market.” November 22, p. 20.
Edmonds, M., M. Uttley, and G. Hayhurst. 1990. “Defence Interdependence: UK and US Depen-
dence on Foreign Technology in Defence Research and Development.” Science and Public
Policy 17, no. 3: 157-69.
Edquist, C., ed. 1997. Systems of Innovation: Technology, Institutions and Organizations. London:
Pinter.
Freeman, C. 1988. “Japan: A New System of Innovation.” In Technical Change and Economic
Theory, ed. G. Dosi, C. Freeman, R. Nelson, G. Silverberg, and L. Soete. London: Pinter.
Gummett, P. and J. Stein, eds. 1997. European Defence Technology in Transition. Amsterdam: Har-
wood Academic Publishers.
Hébert, J.-P. and R. de Penanros. 1995. “French Defence Industry Conversion.” Defence and Peace
Economics 6, pp. 212-13.
House of Commons. 1993. The British Aerospace Industry, Third Report of the Trade and Industry
Committee. London: HMSO, HC 563-I. Also Government Observations on the Third Report
from the Trade and Industry Committee (Session 1992-93) on the British Aerospace Indus-
try. London: HMSO, HC 945.
———. 1995. Aspects of Defence Procurement and Industrial Policy, First Report of the Defence
Committee and the Trade and Industry Committee. London: HMSO.
———. 1998. The Defence Evaluation and Research Agency, Sixth Report of the Defence Com-
mittee. London: HMSO.
124
James, A.D. 1998. Post-merger Strategies of the Leading US Defence Aerospace Companies. FOA-
R-98-00941-170–SE. Report to the FIND Programme. Division of Defence Analysis,
Defence Research Establishment. Stockholm.
James, A.D., H. Cameron, and P. Gummett. 1998. Transfer and Civil Use of Defence-Related Tech-
nologies and Diversification of Defence-Related SMEs, Final Report to the European Com-
mission, July. Manchester: Policy Research in Engineering, Science and Technology.
James, A.D. and P. Gummett. 1998. “Defence.” In Science and Technology in the United Kingdom,
2nd ed., ed. P. Cunningham. Bath: Cartermill International.
James, A.D. and T. Watts. 1996. Government Policy Towards the Defence Industry, MBS Working
Paper Series, No. 346. Manchester: Manchester Business School.
Jane’s Defence Weekly. 1997. “Siemens Puts Defence Electronics Up for Sale.” Jane’s Defence
Weekly, 5 March, p. 25.
Jane’s Defence Weekly. 1998. “UK Exports up 10% to $8.8b.” Jane’s Defence Weekly, 18 March,
p. 15.
Lewis, J.A.C. 1997. “Euro Defence Groups Ordered to Rationalise.” Jane’s Defence Weekly, 17
December, p. 17.
Lewis, J.A.C. and B. Starr. 1997. “A Landmark Year for Crucial Consolidation.” Jane’s Defence
Weekly, 17 December, pp. 20-22.
Lundvall, B.-A. 1992. National Systems of Innovation: Toward a Theory of Innovation and
Interactive Learning. London: Pinter.
Maddock Report. 1983. Civil Exploitation of Defence Technology. Electronics Economic Develop-
ment Council. London: NEDO.
Ministry of Defence. 1996. Statement on the Defence Estimates 1996, Cm 3223. London: HMSO.
———. 1998. The Strategic Defence Review, Cm 3999. London: HMSO.
Markusen, A. and J. Yudken. 1992. Dismantling the Cold War Economy. New York: Basic Books.
Nelson, R.R., ed. 1993. National Innovation Systems: A Comparative Analysis. New York: Oxford
University Press.
Nelson, R.R. and N. Rosenberg. 1993. “Technical Innovation and National Systems.” In National
Innovation Systems: A Comparative Analysis, ed. R.R. Nelson. New York: Oxford Univer-
sity Press.
Samuels, R.J. 1994. Rich Nation, Strong Army: National Security and the Technological Transfor-
mation of Japan. Ithaca: Cornell University Press.
Society of British Aerospace Companies. 1998a. “Global Rationalisation and the UK Aerospace
Industry: An SBAC View.” August. London: SBAC.
———. 1998b. UK Aerospace Statistics 1997. London: SBAC.
Taylor, T. and K. Hayward. 1989.The UK Defence Industrial Base: Development and Future Policy
Options. London: Brassey’s.
Velocci, A.L. 1998. “Efficiency Gains Elude Biggest Merger Survivors.” Aviation Week and Space
Technology, 10 August, pp. 45-52.
Walker, W. 1993. “National Innovation Systems: Britain.” In National Innovation Systems: A Com-
parative Analysis, ed. Richard R. Nelson. New York: Oxford University Press.
Ideas, Identity and the Limits of European
Defense Technology Cooperation
Eugene Cobble
Among students of European integration there exists a pervasive, or rather a perverse, ten-
dency to see Europe in realized terms. Too often, we become enamored with the rhetoric of manipu-
lative national actors and self-promoting Europeanists who speak of a Europe that is, proclaiming
the existence or sanctity of nearly everything European—a European technology base, a European
aerospace industry—or of an effective and vibrant, collective European identity. Reality, of course,
is not nearly so rosy. The states of the European Union (EU) are today closer toward a consensual
political confederation and economic union than at any other time in their history, bound together
under a dense web of organizations and initiatives that have brought greater harmonization, coopera-
tion, and even incremental integration. Nonetheless, despite forty years of progressive reinterpreta-
tions of state sovereignty and of institutional innovation that have propelled the integration move-
ment forward, “Europe” remains a potentiality—a whole that is less than the sum of its parts. There
is no single defense industry, or technology policy, or for the moment, even a common currency;
there are instead more than a dozen distinct state actors who admittedly too often acknowledge a
European interest only when it is compatible with their national interests.
In the absence of radical change, the dynamic between existing state identities and a nascent
European identity is hard to gauge. Numerous Eurobarometer polls and other surveys conducted
over the last thirty years indicate that a discernible European identity now coexists along with fifteen
separate national identities.1 Although this collective identity is clearly the weakest facet within a
hierarchy of local, regional, and state identities, its existence raises the question as to its ability to
influence state policy. Thomas Risse argues that without substantive ideational transformations
among state actors—the formation of a Deutschian community founded upon the convergence of
interest, mutual sympathies, and a “we-feeling”—European integration will not proceed.2 The
1
For a focused review of popular attitudes concerning “European” solutions toward national security
issues see Werner J. Feld, The Future of European Security and Defence Policy (Boulder, CO:
Lynne Rienner, 1993), ch. 5, 11.
2
Thomas Risse, “Identity Politics in the European Union: The Case of Economic and Monetary
Union,” unpublished paper (March 1997), p. 6; Karl Deutsch, et al., Political Community and the
125
126
“collaborations of nations” that are the foundation of Europe will remain just that, and the necessary
integrations of policies and markets that would mark the transition of the EU into a more state-like
entity will not occur.
To mark the extent of a transnational European identity, therefore, is to denote the outer fron-
tier of the union project. In an association of still sovereign states, however, analysis requires one
to be sensitive to nuances of behavior that may betray a sensitivity to what Risse labels as “ideas of
a collective Europeaness.”3 Here, one might expect a greater propensity for diffuse reciprocity, or
in-group solidarity, or asymmetrical distributive outcomes, or just a simple willingness to aggres-
sively pursue intra-regional cooperation in areas that have been traditionally dominated by the im-
peratives of state sovereignty. I contend that recent efforts to create a European system of innovation
in both civil and defense technologies offer an excellent example of the limits of a collective
European identity in shaping state behavior. Arguably none of the prominent examples of initiatives
to promote multilateral cooperation in Europe such as BRITE or ESPRIT are themselves exemplars
of “community spirit.”4 Nonetheless, differences in their organization and execution demonstrate
the boundaries of European solidarity.
This chapter is a study of the failure of Western European states to institutionalize and multi-
lateralize their cooperation in defense R&D within the European Cooperative Long-term Initiative
in Defense (EUCLID) program, using a comparison between it and its nearest civilian counterpart:
the European Research Coordination Agency (Eureka) initiative. Both are intergovernmental in na-
ture; both stress the pursuit of national interest and oversight; both are organizationally minimalist,
having been explicitly designed to produce as small a bureaucratic footprint as possible; and both
focus upon enabling technologies, either at the precompetitive or intermediary stages of develop-
ment. Yet here, the substantive similarities end. Eureka and EUCLID differ in terms of technological
focus and outcome. The former is implicitly civilian; the latter is the only effort of its kind in the
defense realm in Western Europe.
Eureka is a qualified success, promoting the europeanization of civilian R&D in Europe and
the development of research contracts worth billions of ECU. Moreover, as Margaret Sharp notes,
it has helped set the foundations for a regional technology base by creating “a complex network of
horizontal and vertical collaborations [pulling] together the capabilities and resources of the Com-
munity across national boundaries and across old institutional divides.”5 EUCLID is, on the other
hand, little more than a footnote—yet another failed or faltering effort among many to create a single
European defense market. It suffers from the profound neglect of the same states who initiated it
eight years ago to ease the national fragmentation of their collective defense effort. EUCLID has
been under-funded and under-utilized in favor of national defense R&D programs and more limited
interstate cooperative ventures.
That this discrepancy exists is a puzzle. After all, Europe enjoys the highest level of interna-
tionalized R&D of any region in the international system. Moreover, whereas this process has solidi-
fied with the emergence of the single European market in the early 1990s, it emerged first in the
defense field nearly forty years ago in the late 1950s. At that time, the burdens of rapid technological
change and spiraling development costs first began to compel European states to pool their resources
and to share both the expense and risks of defense technological innovation.6 Today, Europeans face
increasing pressures to embrace multinational defense procurement, as inter-generational enhance-
ments have made some high technology weapons systems too costly for even medium-sized powers
to develop unilaterally.
Further complicating the situation is the transformed relationship between the production of
civilian and defense technologies, particularly the former’s new significance as an engine of techno-
industrial development. In the 1950s, 1960s, and early 1970s, the civilian technology base benefitted
from higher levels of military demand and military R&D financing of high technology. Today the
reverse is true: the defense technology base is not only dwarfed by its civilian counterpart, it is
dependent upon it as a source of innovation for materials, components, and complete systems in the
form of dual-use goods. Consequently, the divisions between civilian and military technologies have
blurred. The significance of this transformation in the European context cannot be overstated.
5
Margaret Sharp, “The Community and New Technologies,” in The European Community and the
Challenge of the Future, ed. Juliet Lodge (New York: St. Martin’s, 1993), p. 219.
6
Interview, British Aerospace, Spring 1997.
128
Thirty-five years ago, multinational R&D collaboration in Western Europe was principally and ex-
plicitly defense-oriented; in fact, while cases of civilian pairings were exceptional, the military divi-
sions of firms like British Aircraft Corporation, Dornier, and Aérospatiale cooperatively produced
families of commercially and technologically successful weapons, albeit on a limited scale.7 Since
that time, however, there has been a diversification of interstate technology cooperation privileging
civilian production, one that has been partially guided by state and supranational actors. Yet as state
governments and the institutions of the European Union recognize this transformation and seek to
capitalize from it, defense production remains politically valued in a way that limits the potential
for deepening interstate cooperation—even as upwards to 50 percent of the technologies produced
through explicitly civilian collaborations have potential military applications.8
Europeans can and do collaborate to produce technologies with military value. Yet, they pre-
fer the indirect and ad hoc creation of defense technologies through civilian channels to the coordi-
nation of a dedicated defense forum. I contend that EUCLID’s failure is neither technological nor
organizational, but is instead ideational. Europeans have been remarkably adept at recognizing their
collective technological failings and then acting in concert to remedy them. In both civilian and mili-
tary domains, they have created bi- and multilateral groupings to solve problems of common interest
in areas ranging from integrated circuit design, to refining the land-based rearing of sea bass,9 to the
development of helicopter gunships. The technologies under EUCLID’s purview have been no more
radical than these; indeed, EUCLID projects are not dissimilar to some conducted within the major
civilian-oriented collaborative initiatives. Further, parallels between EUCLID and Eureka do not
stop at technology. EUCLID’s founders initiated the program with the successes of organized inter-
state civilian cooperation in mind. Consequently, they structured it as a coordinating organization
along lines similar to that of the technologically and commercially successful Eureka program.
7
Interview, British Aerospace, Spring 1997.
8
European Parliament, European Armaments Industry: Research, Technological Development and
Conversion, Final Report (STOA/GRIP/IA 1993), p. 49. For a discussion of changing state attitudes
toward the civilian technology base, see Philip Gummett, “West European Defense Industrial Policy
After the Cold War,” in Military R&D after the Cold War: Conversion and Technology Transfer in
Eastern and Western Europe, ed. Philip Gummett, Mikhail Boutoussov, Janos Farkas, and Arie Rip
(Dordrecht: Kluwer Academic Publishers, 1996).
9
Eureka Project E!1376
129
I argue that EUCLID has failed to develop because of its symbolic character. The differences
between it and ongoing efforts to institutionalize and Europeanize technology innovation in the ci-
vilian field were more visible than real. Nonetheless, in Europe as elsewhere, the links between the
state and the provision of defense remains too strong to permit the smooth functioning of fixed trans-
national military R&D regime. The connection between state identity and armaments has imposed
a fundamental barrier on the evolution of a “European” defense technology base through dedicated
means, just as it has done in other areas of Western European procurement collaboration. In the fol-
lowing sections, I explore the nature of this relationship, and frame the larger discussion within
International Relations Theory. I illustrate my argument with an empirical discussion detailing the
evolution of technology cooperation in Western Europe since the early 1980s, paying particular
attention the relative fortunes of the Eureka and EUCLID initiatives.
Theoretical Foundations
To understand European integration one must first understand and appreciate change: chang-
ing environmental pressures, changing state behavior, changing ideas of appropriateness, changing
interests, and importantly, changing conceptions of state identity.10 Traditional analytical approaches
within International Relations theory, namely neo-realism and neo-liberal institutionalism, do not
emphasize interest, ideas, or identity—except, of course, to dismiss them as predetermined, as meth-
odologically unapproachable, or as epiphenomenal. The study of European defense collaboration,
however, requires that we consider all three, because in this area both conventional wisdom and gov-
ernmental proclamations maintain that they are in flux, with an emerging regional identity, shifts
in the value and extent of defense technological innovation, and finally a transformation in state in-
terest that arises from these two developments. Any analysis of military R&D cooperation therefore
requires a theoretical approach that deems identity and interests as mutable and as endogenous to
observed behavior.11
10
Ronald Jepperson, Alexander Wendt, and Peter Katzenstein, “Norms, Identity, and Culture in
National Security,” in The Culture of National Security: Norms and Identity in World Politics, ed.
Peter Katzenstein (New York: Columbia University Press, 1996), p. 51.
11
Alexander Wendt, “Anarchy is What States Make of It: The Social Construction of Power Poli-
tics,” International Organization 46 (Spring 1992): 392.
130
This need is compounded by the special character of defense technology in the international
system. Weapons are unlike any other commodity. They are uniquely defined by their lethality and,
as the “distinguishing emblem of the modern nation,” by an almost organic tie to state identity.12
Arms provide military protection, techno-industrial benefits, and status in a world system where
“normative definitions of statehood” and of political efficacy are dependent upon the ability of states
to produce, possess, and project military force.13 Consequently, if a supranational identity is indeed
emergent and significant, multinational defense procurement is an ideal test of the value of new
approaches in international relations. A new identity would need to be very potent to produce new
definitions of state interests and new behaviors in an issue-area as historically bound to conventional
visions of sovereignty and state identity as defense procurement.14 We cannot even assess this
possibility, however, unless we are attuned to the roles that identity and the non-material aspects of
armament play in determining how states behave towards defense technologies.
Constructivism, one the newer additions to International Relations theory is well suited to
addressing armaments collaboration because it acknowledges both identity formation and shift.15
Like Realism and neo-liberal Institutionalism, modernist Constructivist logic, as pioneered by
Wendt, Emmanuel Adler, Peter Katzenstein, and others, recognizes the primacy of state actors in
the international system and the anarchic nature of that system. Unlike the mainstream theories,
however, Constructivism offers a socially based view of the material resources that constitute social
12
Cited in François Heisbourg, “Public Policy and the Creation of a European Arms Market,” in The
European Armaments Market and Procurement Cooperation, ed. Pauline Creasey and Simon May
(New York: St. Martin’s, 1988), p. 86; Mark O’Connell, “Putting Weapons in Perspective,” Armed
Forces and Society 9 (Spring 1983): 450; Mark Suchmann and Dana Eyre, “Military Procurement
as Rational Myth: Notes on the Sociological Perspective on Weapons Proliferation,” Sociological
Forum 7 (March 1992): 151.
13
Dana Eyre and Mark C. Suchmann, “Status, Norms, and the Proliferation of Conventional Weap-
ons: An Institutional Theory Approach,” in The Culture of National Security: Norms and Identity
in World Politics, ed. Peter Katzenstein (New York: Columbia University Press, 1996), p. 92.
14
Holly Wyatt-Walter, The European Community and the Security Dilemma, 1979-1992 (New York:
St. Martin’s, 1997), p. 253.
15
Nevertheless, it has rarely been applied to security issues. See David Segal, Mady Segal, and Dana
Eyre, “The Social Construction of Peacekeeping in America,” Sociological Forum 7 (March 1992):
125.
131
structures.16 A Constructivist analysis of interstate collaboration, for example, would note that state
identity (e.g., sovereign, capitalist, advanced, French, etc.) determines the potential for either coop-
eration or competition, based upon initial conceptions of the self and other.
Arms production is driven by material and ideational incentives that establish the potential
for interstate collaboration. The military significance of defense technology is perhaps the most
easily understood of these ideals, and it is thus the most readily overemphasized. After all, weapons
are the ultimate tools of statecraft: replacing diplomacy with purposive violence when the former
fails. These systems allow states to control their own populations, resist outside incursions, and
apply controlled force in the advancement of foreign policy. Armaments procurement provides the
tools needed to preserve a state’s political autonomy and to protect its territorial integrity. The
politics of weapons production is therefore intertwined with the very existence of the state itself.
The relationship between defense technology and the state, however, extends far beyond the
provision of military power. Arms industries sit at the nexus between security and economics—
between power and plenty.17 The potential material benefits from a domestic defense industrial base
(DIB) and defense technology base (DTB) are enormous, not only for a country’s force posture but
for its economic well-being as well. An indigenous production capacity offers secure, dependable
access to the means of defense without reliance on foreign sources of supply that may either deny
the availability of war goods or perhaps use a dependence relationship as a source of political
leverage. Defense industries also provide economic side-benefits: jobs for the civilian labor market
through direct employment and potential technological spin-off to the civilian sector. Policy-makers
in the West and elsewhere regarded military technological innovation as an engine for growth that
could pull an entire national economy with it, particularly in high-return, “sun-rise” sectors based
on emerging technologies.18
16
Alexander Wendt, “Constructing International Politics,” International Security 20 (Summer
1995): 73.
17
Ethan B. Kapstein, “International Collaboration in Armaments Production: A Second-Best Solu-
tion,” Political Science Quarterly 106, no. 4 (1991-92): 658.
18
For an example of this phenomenon in the French case, see Edward Kolodziej, Making and Mar-
keting Arms (Princeton: Princeton University Press, 1987), p. 138. See also Michael Borrus and John
Zysman, “Industrial Competitiveness and American National Security,” in The Highest Stakes: The
Economic Foundations of the Next Security System, ed. Wayne Sandholtz, Michael Borrus, John
132
While economic and security incentives certainly may explain much of state behavior here
and elsewhere, they cannot provide a complete picture. States do not procure arms simply to defend
their territories or protect their labor markets. Weapons also have socio-cultural value. As Dana Eyre
and Mark Suchmann note:
. . . technology is never just technology . . . every machine has a socially constructed
meaning and a socially oriented objective and that the incidence and significance of
technological developments can never be fully understood or predicted outside their
social context.19
Armaments are positional goods: they convey status and prestige in an international system, in
which a country’s “membership in modernity” is defined by the sophistication of its force posture.20
Some symbols, however, are more potent than others. Within the armaments field, status adheres
not just upon the possession of weapons, but also upon the ability to design and to locally produce
them—and furthermore, not just any weapons, but sophisticated, high-technology goods. As states
face normative incentives to modernize their economies, a high-tech weapons capacity denotes both
a great nation and a great national economy capable of considerable technological innovation. These
weapons are consequently “loaded with meaning,” symbolizing sovereignty, technological advance-
ment, strength, and political efficacy. In the early 20th century, the battleship epitomized how tech-
nology of this type could become tightly embedded in a state’s self-perception, and, as important,
how that state wished to be regarded by others. Michael Howard writes:
The Battleship was indeed a symbol of national pride and power of a unique kind;
one even more appropriate to the industrial age than armies. It embodied at once the
technological achievement of that nation as a whole, its world-wide reach and, with
its huge guns, immense destructive power. It was a status symbol of universal valid-
ity, one which no nation conscious of its destiny could afford to do without.21
Zysman, Ken Conca, Jay Stowsky, Steven Vogel, and Steve Weber (New York: Oxford University
Press, 1992), pp. 29-31.
19
Eyre and Suchmann, “Status, Norms,” p. 86.
20
Alexander Wendt and Michael Barnett, “Dependent State Formation and Third World Militari-
zation,” Review of International Studies 19 (1993): 337.
21
Michael Howard, “War and the Nation-State,” Daedalus 108 (1979): 104.
133
Today, other technologies fill this role: the air-superiority fighter, the chobham-armored
main battle tank, and the ballistic missile submarine among others. The social value of these technol-
ogies is tightly intertwined with their material attributes. In many cases, however, symbolism can
outweigh any objective criterion. For example, defense-seeking states require both artillery shells
and ground-attack jets if they are to maximize their military potency. Both can be important tools
of statecraft, but when judged solely from the historical record, artillery is a far more effective
means of either killing or disrupting enemy forces. Nonetheless, because ground-attack planes are
carriers of national prestige due to their status as symbols of techno-economic sophistication, this
technology receives undue attention in decision-making fora.
Consequently, the social incentives of weapons production can, and sometimes clearly do,
outweigh the material incentives of procuring those items. These possibly divergent motivations can
have a significant impact upon the conduct of multinational defense collaboration between different
classes of defense commodities. In Western Europe, multinational procurement has evolved over
the last four decades. From a limited and ad hoc beginning, it now involves increasingly complex
interactions producing many kinds of military hardware, including sizable inventories of high-tech,
high value-added weapons.
What is rarely recognized, however, is that this ideational component of defense procurement
can shape cooperation for enabling technologies and dual-use goods as well. Armaments are more
than the end-user items that one finds on a parade ground or airplane pylon. They are the sum of pro-
duction processes and sub-assemblies, some of which may not have an explicitly military focus. For
example, the same techniques and machine tooling used to manufacture a howitzer barrel can also
be used to produce a naval propeller shaft or the axle of a railroad car. The development of rotary
forges may seem, at first glance, a poor symbol of state sovereignty. Yet, when they are produced
within a distinctly defense context, any objective banality disappears and the technologies become
contested by states, who strongly resist sharing them with their allies. This phenomenon is most
clearly seen with a comparison of civilian and defense collaborative schemes in which similarities
in technology have nonetheless led to wildly disparate outcomes. The EUCLID and Eureka
programs demonstrate how the interplay between identity and defense can produce such an effect.
134
EUREKA
Eureka predated EUCLID by nearly five years, having been initiated in 1985 as a French-led
response to the perceived techno-industrial challenge posed by the United States’ Strategic Defense
Initiative. Specifically, it reflected a growing awareness among European elites that national tech-
nology assets could be best protected and expanded in an increasingly competitive global environ-
ment through interstate collaboration. In the early 1980s, the European Commission had initiated
a series of moderately successful cooperative programs, most notably ESPRIT, which promoted
regional collaboration in information and telecommunications technologies at the pre-competitive
level. Eureka’s founders sought to capitalize upon this development as well as to reassert state inputs
into the collaborative process. European governments hoped to promote and guide industrial coop-
eration while concurrently preempting the europeanization of technology policy through the supra-
national influence of the European Community.22
The initial French proposal, while vague in detail and requiring no formal treaty, nonetheless
envisioned a “top-down” framework in which intergovernmental committees determined R&D
policy and state governments provided substantial program funding.23 The only notable difference
between the original Eureka plan and EUCLID was the former’s implicitly civilian orientation, even
though its initial technological emphasis was very similar to that pursued under the Strategic
Defense Initiative, and later under EUCLID: remote sensing, optoelectronics, high-speed computing,
materials, lasers, and communications systems among others. As Eureka coalesced during 1985 and
1986, the involvement of additional state actors and the largest information technology firms in
Western Europe pushed the program away from its initial, state-driven conception and toward a
framework that was at the time unique in the region. Industry insisted that “Eureka should be prod-
uct and market-oriented.”24 Germany and Britain, along with other states in consultation, refused any
22
Wayne Sandholtz, High-Tech Europe: The Politics of International Cooperation (Berkeley: Uni-
versity of California Press, 1992), p. 295.
23
David Dickson, “EUREKA! Cooperative Research & Development Program in Europe,” Tech-
nology Review 91 (August 1988), available from http://web.lexis-nexis.com/universe; accessed 16
December 1998.
24
“Declaration of Common Intent” by GEC, Siemens, Philips and Thomson. Quoted in John Peter-
son, “Technology Policy in Europe: Explaining the Framework Programme and Eureka in Theory
and Practice,” Journal of Common Market Studies 29, no. 3 (March 1991): 278.
135
mandatory state funding requirements. Moreover, they rejected the creation of a centralized deci-
sion-making secretariat in favor of ad-hoc industrial control complemented by minimal state and
intergovernmental oversight.
Eureka’s Declaration of Principles, signed in late autumn 1985 by representatives of seven-
teen West European states, codified all of these demands. Eureka officially became a “near-market”
initiative intent upon collaborative technology development and commodification. In practice, how-
ever, it represented more of a “technological potluck” in both emphasis and technical content. The
Eureka label has been attached to collaborative projects as far afield as biotechnology and infrastruc-
ture development, and to schemes at nearly every stage of development from the pre-competitive
level onwards.25 The initiative’s civilian orientation was retained; both to set it apart as a distinct
alternative to SDI, and to win the support of European neutrals and the publics of a number of allied
states, namely Germany.26
Eureka’s brilliance lay not its technological portfolio, however, but rather in the way that it
promoted cooperative development. The Declaration stressed that commercial interests should pre-
vail over political imperatives whenever possible in shaping collaboration and defining program pri-
orities. To this end, the framework established a “bottom-up” program in which firms chose objec-
tives of collaboration and the means to achieve it. Indeed, although the initiative was intergovern-
mental in structure, its operation was explicitly industry-led. There was no governing organization
or common funding. Eureka states invested national firms with the responsibilities of defining proj-
ects and arranging industrial partnerships. To attain Eureka status, all that was required was for two
or more firms from at least two Eureka member states to declare an intention to collaborate upon a
project of their choosing.27 This information would be then transmitted to the respective national
governments involved, and if approved, would permit the possible allocation of state matching
funds, which served as a cooperation subsidy. Manufacturers not only proposed tentative ventures
for Eureka status and support, they also reserved the right to either invite or reject other firms as
25
Sandholtz, High-Tech Europe, p. 283.
26
Dickson, “EUREKA!”
27
Henry Durand, “Building Up a Common European Science and Technology Policy,” Technology
in Society 13 (1991): 374.
136
28
John Peterson, High Technology and the Competition State: An Analysis of the Eureka Initiative
(London: Routledge, 1993), p. 32.
29
Juste retour has been a feature of most European industrial collaborations in defense, and also in
civilian industry before the emergence of widespread industry-initiated interstate partnerships in the
mid-1980s. The practice of juste retour allowed state governments to predetermine the level of
involvement of their national firms within any interstate collaborative project; and moreover, to do
based solely upon the level of financial support that the state or its leading firm contributed to the
collaboration. For example, in the Anglo-German-Italian Tornado fighter-bomber project, the
participating states held a 42.5/42.5/15 percent respective cost-share/work-share. This meant that
Britain provided 42.5 percent of the total R&D costs and, in return, was permitted to produce 42.5
percent of the value of the airframe, engine, and avionics packages. The UK also received 42.5
percent of the total production run of planes. See Peterson, High Technology, p. 72.
137
could then strategically apply state subsidies to indigenous companies as an incentive for them to
forge transnational relationships in other European countries to better achieve the desired ends.30
These mechanisms provided both states and firms with a mutually beneficial cooperative
environment. The former could use public funds to coax select firms to establish cross-border part-
nerships for advanced technology research, and thus help vitalize their national industrial and tech-
nology bases; moreover, they enjoyed the participation of other national governments who could
potentially provide support to these same projects, thus spreading the financial burden and ensuring
a minimum of national expenditure by any given country.31 The firms, on the other hand, enjoyed
limited state subsidies on collaborative projects that in some instances might not have otherwise
occurred. Eureka offered both sides exactly what they wanted from interstate R&D cooperation and
without any pronounced faults in organization or implementation that might taint its appeal. Indeed,
the record shows that both government and firms throughout the Eureka community were quite
quick to exploit the tool that they had created for themselves. The framework quickly became the
most successful program of its type in Western Europe. By 1996, Eureka had grown to 1250
approved projects across all technology domains with a total investment of over 18 billion ECU.32
Between 1986 and 1992, 505 separate projects worth more than 8 billion ECU had been sanctioned
by Eureka officials. Of these, 107 were authorized during the first year at a value of 3.2 billion
ECU.33
While the statistics of Eureka’s success were unmatched by any another regional collabora-
tive R&D initiative, and in fact, remain unparalleled, one must not forget that these impressive fig-
ures were made possible by the Eureka framework’s singular ability to harness diverse national in-
terests and integrate them into a diffuse sense of European solidarity. John Peterson argues that the
30
State financial support for Eureka projects varied according to state interest in promoting a given
technology or industrial sector. Among Eureka member countries, this support ranged from 30 per-
cent to 50 percent of the funding supplied by national industrial contributions, and averaged approxi-
mately 30 percent for the Eureka community as a whole. Durand, “Building,” p. 375.
31
Peterson, High Technology, p. 101.
32
FBIS Report. FBIS-EST-96-013. 4 July 1996. “EUREKA 1996: Positive Results for France and
Belgian Presidency,” AFP Sciences, 4 July 1996, p. 1.
33
Peterson, High Technology, p. 33.
138
program’s variable geometry of state-state and state-industry contacts allows national governments
to pursue their particular policy agendas for technology innovation through subsidies and indirect
guidance, but nonetheless to act such that national policies converge within a European framework.34
Through Eureka, states could attain their parochial objectives by encouraging private actors to effec-
tively “Europeanize” their behavior by instituting the inter-state, industrial relationships within the
community that are a prerequisite for eventual economic union. Thus, European decision makers
could enjoy the best of both worlds: pursue the national self-interest, while also acting in concor-
dance of the loftier ideals of European integration, and in such a way that brought these ideals closer
to fruition. Even Margaret Thatcher, while certainly no advocate of European union, argued in June
1986 that “Eureka is a key element in Europe’s industrial strategy . . . Through Eureka, European
firms can help us identify the steps to open markets which will most help them.”35
Proclamations such as this that have surrounded the Eureka program since its initiation must
be embraced cautiously. Eureka did not emerge from some radical shift in state perceptions and atti-
tudes. It both reflected and promoted state interests, albeit enlightened interests, that acknowledged
the legitimacy of transnational modes of technology innovation, were shaped by the interests of
others pursuant to some commonly valued goal, and significantly, were made compatible with the
grander aspirations and processes of community-building and of regional integration. Eureka
demonstrated the permissive potential of a nascent European identity—the “idea of collective
Europeaness”—to redefine state interests toward a communal good. While such an identity was not
strong enough to overwrite national prerogatives and, in this case, facilitate the creation of a unified
system of innovation, it at least supported the harmonization of those interests.
EUCLID
While Eureka might show the promise of a European identity, when viewed comparatively,
it also demonstrated the fundamental weakness of that identity and its limited applicability to other
areas of European activity—even those that overlap with Eureka’s mandate. By the end of the 1980s,
Eureka’s success was undeniable, with nearly 300 approved collaborative projects totaling nearly
34
Ibid., 201.
35
Cited in Peterson, High Technology, p. 70. Emphasis added.
139
6.5 billion ECU by January 1, 1990.36 With a record like this, Eureka became a model to be emu-
lated, and Europeans were keen to do so elsewhere in hopes of attaining the same level of perfor-
mance. The troubled regional defense industry was quickly seen as the ideal area to apply the lessons
learned in the civilian arena. Here was an entire industrial sector with considerable existing collabo-
rative activity, but under pressure due to falling defense budgets and a legacy of interstate coopera-
tion that rewarded inefficiency and hindered any meaningful rationalization of the regional defense
industrial base.
In order to understand better the EUCLID project, one must first appreciate the environment
in which it was crafted. Since the founding of the European alliance system defined by NATO and
the Western European Union in the late 1940s and early 1950s, armaments collaboration has been
a core objective of the collective military effort. In fact, multinational procurement had become well
established in some high-cost areas such as aerospace and defense electronics. Systems such as
Roland, Jaguar, Tornado, and Gazelle all reflect the considerable energy invested into procurement
cooperation, to say nothing of the billions of marks, francs, pounds, and lira. This collaboration,
however, was not the same as integration. Europeans have been very keen to retain maximal influe-
nce over their defense industrial bases and to receive maximum material and security returns from
them. Moreover, as Rae Angus asserts, they structured their cooperative endeavors so as to move
“the shortest possible distance from autarky.”37 This was cooperation that served the economic and
military pretenses for collaboration while preserving sovereignty and rigid national control over the
procurement process. Indeed much of traditional European armaments collaboration occurred
through processes best characterized by petty nationalism, industrial bickering, and the waste that
these factors produce in bulk—a situation that did more to hinder integration than to promote it.
Defense cooperative schemes since the 1960s—and continuing into the present day—exhib-
ited four principal disintegrative traits. First, collaboration was de-marketized. Defense Ministries
regarded the outright purchase of foreign arms to be, at best, a “measure of last resort” undertaken
36
Durand, “Building,” p. 375.
37
Rae Agnus, “The Tornado Project,” in International Arms Procurement: New Directions, ed.
Martin Edmonds (New York: Pergamon Press, 1981), p. 166.
140
under the most extreme economic or political constraints.38 Only the smaller allies, with their re-
stricted defense industries and tiny defense budgets, had to rely upon the direct purchase of compet-
itively-priced foreign defense technologies. The medium-sized and large states, with home industries
to protect and money to spend, insisted on maximum domestic production, even when they osten-
sibly bought weapons from abroad. Through offset compensation and licensed production, states
either distorted defense trade or evaded it by selecting national firms to produce up to 100 percent
of the contract value within a given program. This took the form of either the domestic production
of subsystems, with local firms functioning as subcontractors to foreign primes under offset pack-
ages, or the complete assembly under licensed production.
The choice for national production did not reflect any cost or competence criteria but instead
represented the political desire to channel technology transfers and work orders to specific actors
within national defense industries. This principle was also prevalent in the multinational ab initio
collaboration based on arms co-development that occurred principally among the larger, and more
self-contained, defense producers such as France and the UK. Here, procurement cooperation would
begin through inter-governmental agreements to control rising defense technology costs by pooling
national resources to attain common equipment goals. States again adhered to the principle of “no
money across borders.” Countries would share the total costs of development and/or production, but
all national funds would go exclusively to national research centers and firms, even as this guaran-
teed dis-economies through higher transportation costs, and duplicated administrative and assembly
nodes.
Second, the division of responsibilities underlying co-development/co-production projects
reflected the practice of le juste retour, or fair return. States demanded an immediate and exact
workshare equal in value to their contribution to a given project’s development cost. These figures
were meticulously measured and were often changed to reflect currency fluctuations or revised na-
tional military requirements. In some cases, states calculated cost-share/work-share to the hundredth
decimal point in order to extract some techno-industrial benefit from every last cent of expenditure.
The third feature of traditional European armaments collaboration has been its deliberately
uninstitutionalized and ad hoc nature. European governments denied international organizations any
38
James Moray Stewart, “The European Defense—Principles and Policies,” NATO’s Sixteen Nations
(December 1989/January 1990), p. 21.
141
real control over the conduct of their collaborative projects, and continue to do so. The European
Community, for example, has been flatly denied competence, while dedicated procurement-harmo-
nizing groups, such as NATO’s Council of National Armaments Directors or the Independent Euro-
pean Programme Group (IEPG), could only rubber-stamp projects after states had already estab-
lished cooperation guidelines. This type of limited collaboration permitted states to micro-manage
procurement collaboration. Moreover, it allowed them to maximize potential gains to their defense
base through access to technology and to production skills without the meddling of a long-standing,
potentially efficiency-seeking institutional arrangement—organizations that, as Walker and Gum-
mett note, might
. . . [compel] governments to submit to a form of arbitration which diminishes their
individual command over industrial assets most tightly embedded in notions of
sovereignty—a tall order in the best of times.39
Collaboration thus assumed an a la carte character as states engaged in strategic alliances on partic-
ular projects and then dissolved those partnerships once the desired technologies had been obtained.
Throughout the 1960s, 1970s and 1980s there were a host of joint venture programs, such as Tor-
nado or Jaguar, in which states united behind a co-development scheme and then parted company
after final production.40 This behavior ensured that European decision-makers, as well as the actual
producing firms, did not enjoy a learning-curve in cooperation and thus could not easily improve
their collaborative strategies across programs.41
39
William Walker and Philip Gummett, “Nationalism, Internationalism and the Future of the
European Defence Market,” Chaillot Paper of the Institute for Strategic Studies (Western European
Union, 1993): 4.
40
One must note that the experience of cooperation is not forgotten after project completion. Accu-
mulated trust and knowledge of former partners has prompted states and their firms to renew con-
tacts for future programs. PANAVIA Tornado exemplifies this phenomenon given its “evolution”
into Eurofighter EFA. Nonetheless, the terms of cooperation have changed from project-to-project
as states renegotiate their partnerships.
41
A good example of this phenomenon is the relationship between the Tornado Eurofighter projects.
Among the same three states and three national aerospace industrial champions, lessons learned in
the earlier Tornado project were not applied to Eurofighter. This led to considerable waste of time
and resources until the participating firms established greater contact between their Tornado and
Eurofighter divisions several years into the Eurofighter R&D phase. Interview, 19 February 1997.
142
The fourth and final feature of traditional European armaments cooperation was its pervasive
tendency away from multilateralism. Although the West Europeans fielded twelve national militaries
for the collective defense, the bulk of their procurement collaboration has been limited to dyadic or
triadic groupings. Even cooperation in defense-related research has followed this general pattern,
with over 70 percent of such collaborations occurring bilaterally among France, Germany or the
United Kingdom.42 These limited partnerships have guaranteed that administrative costs of coop-
eration have been kept to a minimum and thereby assured that the desired cost-savings of collabora-
tive procurement vis-à-vis solely national production could be attained. Although “minilateral”
cooperation facilitated the development of collaboration by making the process financially viable,
it has not served to bring defense producers together on a regional level.
EUCLID reflected the recognition that defense collaboration in Western Europe could be
done better, at least within the area of research and development. The Eureka model, or more pre-
cisely its methodology, and its theoretical applicability to this area seemingly offered a ready-made
solution. France was the first to argue that such a conversion take place, and to stress the point,
assigned its Eureka national program coordinator, Yves Sillard, in 1989 to the French delegation of
the Independent European Programme Group (IEPG) to promote Eureka’s extension into the defense
field.43 By the summer of that year, IEPG national armaments directors had agreed that pushing
collaboration as far upstream as possible into the initial stages of development would significantly
enhance armaments cooperation.44 They hoped that such an initiative would offset the national frag-
mentation of R&D assets within Europe by promoting cross-border linkages between armaments
producers. These relationships, coordinated under the soon-named EUCLID program, would pool
research funds, thus minimizing defense R&D duplication among states, as well as providing a
larger potential funding base per project than might be available within a purely national effort.
Eurico del Melo, the Portuguese minister of defense in 1990, noted that EUCLID’s principal
aims were to increase Europe’s developmental capacity in critical technology areas and to create the
42
Assembly of the Western European Union, “WEAG: The Course to be Followed,” Document
1483, 6 November 1995, fn. 88.
43
Peterson, High Technology, p. 53.
44
Assembly of the Western European Union, “European Cooperation on the Procurement of Defense
Equipment—Lessons Drawn From the Symposium,” Document 1587, 4 November 1997, para 28.
143
foundation for what later observers labeled a “defense research Europe”—the cornerstone of what
was to be a European community of technology.45 To this end, the initiative was structured around
eleven, later fifteen, Common European Priority Areas (CEPA) of technologies deemed strategic
to the Allied defense effort and to the continued competitiveness of the regional defense industrial
base (Table 1).
Table 1
List of CEPA Technologies
CEPA 1 Advanced Radar Technology
CEPA 2 Microelectronics
CEPA 3 Materials and Advanced Structural Technologies
CEPA 4 Modular Electronics
CEPA 5 Electric Gun§
CEPA 6 Information Processing/Artificial Intelligence
CEPA 7 Stealth§
CEPA 8 Optoelectronics
CEPA 9 Satellite Surveillance
CEPA 10 Underwater Detection and Associated
Technology
CEPA 11 Human Factors – Simulator Technologies
CEPA 12 Aerothermodynamics
CEPA 13 Biological and Chemical Defense Technology
CEPA 14 Energetic Materials
CEPA 15 Missile Control and Guidance Technology
§
Inactive
Source: Carol Reed, “Sharing Out the Cost of R&T,” Jane’s Defense Weekly 10 June, 1995, p. 50
Each CEPA had a steering committee composed of academic, industrial, and state defense
representatives from IEPG countries, who selected specific research and technology projects (RTP)
to be pursued within each domain. These projects typically sought to support the enabling technol-
ogies that would facilitate the later development of future equipment deemed necessary by the steer-
ing committees. In fact, it was hoped that the development of precursors, such as high speed analog-
digital converters, “ruggedized” microelectronics assemblies, combinatorial algorithms, solid-state
45
David Fouquet, “IEPG Hopes to Copy Success of Eureka Research Project,” Jane’s Defense
Weekly, July 8, 1989, p. 12.
144
lasers, and computer workstation design, would provide the foundation for collaborative weapons
production between two more member states at some future date.46
For each RTP, the state pledging the largest amount of financial support received lead nation
status and assumed responsibility for attracting additional state participants, performing program
management, and selecting an international industrial consortium to conduct the given research con-
tract. Although the EUCLID framework officially encouraged state participants to consider techno-
industrial competencies and the competitive awarding of research contracts, in practice neither goal
was assured because they were left to the discretion of the member states. Countries—beginning
with the choice of lead nation—literally bought admittance into advanced technology research.
Participant states could subsequently select their respective national firms to comprise the consortia
that actually performed the research. Even those countries lacking basic technological capacities
within certain fields, namely the developing defense industry countries (DDI), could join as equal
partners by way of their financial contribution to a given RTP.
Such behavior did very little to promote research efficiency, and as we shall soon see, it was
a contributing factor to the EUCLID framework’s eventual failure. Nonetheless, its structure offered
the prospect of more diverse national participation within the regional defense technology base, as
firms throughout the IEPG community began tentative steps toward cooperative development in
technology fields once dominated by the major arms producers located in the larger defense markets.
EUCLID multilateralized European defense R&D collaboration to an extent hereto unmatched.
Some research projects had as many as eight national participants, with an average of four per RTP
in the defense electronics fields.47
EUCLID’s success was largely limited to symbols such as this. By mid-1996, 57 RTPs had
been approved by EUCLID member states with a total contract value of nearly 250 million ECU.48
While this figure may seem impressive, it fell far short of the founders’ stated ambitions. It was
46
Mark Stenhouse, “The Future of Collaborative Production and Procurement,” International
Defense Review 26 (January 1, 1993): 19.
47
Assembly of the Western European Union, “The Euclid Programme and Cooperation Between
European Defense Electronics Industries,” Document 1524, 14 May 1996, Appendix II.
48
Of these, eight had been completed, 24 were under contract, 11 were in preparation, and 14 had
reached the implementation stage.
145
hoped initially that EUCLID would eventually consume 30 percent of total IEPG R&D spending
each year, valued at approximately 500 million ECU.49 By Spring 1997, however, EUCLID projects
had averaged roughly a tenth of this figure per annum.50 Throughout its first seven years, the initia-
tive consistently failed to live up to expectations. In fact, it had encountered significant problems
from the outset. EUCLID’s objective was to bring greater coordination to Western Europe’s col-
lective defense, but the very states that launched the initiative and claimed to support its mandate
remained wedded to existing modes of collaborative behavior. More significantly, they continued
to privilege the national production of defense technology to such an extent both within and without
the EUCLID framework that they effectively stifled the framework.
Consider, for example, the overwhelming state control over EUCLID’s execution. Strict
intergovernmentalism colored nearly all of the program’s conduct. A committee of national defense
ministerial representatives meeting semi-annually within the IEPG possessed primary program over-
sight authority. All decisions, ranging from the nomination of EUCLID projects to the organization
of the industrial consortia assigned to work on them, required consensus among the participating
states.51 Even though national financial contributions determined lead nation and subcontractor
status under juste retour rules, these relationships still required the explicit approval of all member
governments. While industry was invited to offer suggestions and was obliged to provide 50 percent
of project funding, ultimate control remained firmly vested in state actors.
The desire for maximal state control over the innovation process actually extended beyond
guidance and had far more immediate consequences. First, state involvement at nearly every stage
of decision-making significantly slowed the speed of cooperation. With upwards to thirteen states
on each CEPA selection committee and the need to appeal to each national government for project
approval and funding contributions, the time required for a project to move from initial examination
49
FBIS Report, FBIS-EST-97-002, 6 December 1996; Christel Tardif, “EUCLID Corrects the Errors
of its Youth,” Air & Cosmos/Aviation International, 6 December 1996, p. 24.
50
FBIS Report, FBIS-WEU-97-120, 8 May 1997; S. Brosselin, “European Military Research in
Trouble,” Paris Electronic International Hebdo, 8 May 1997, p. 16.
51
Claude Serfati, “Reorientation of French Companies,” in Restructuring of Arms Production in
Western Europe, ed. Michael Brzoska and Peter Lock. (New York: Oxford University Press, 1992),
p. 103.
146
to implementation could exceed 14 months.52 During this period, firms often lost interest, as the pro-
posed research ceased to be topical and the potential returns from cooperation no longer warranted
the required financial contribution.
Second, and far more critical from the defense industry’s perspective, was the issue of intel-
lectual property rights. While industry performed the research, the draft memorandum of understand-
ing (MoU) that founded EUCLID gave the participant states full and free access to the technologies
that it produced. These countries were then entitled to use the information as they desired, either to
apply it immediately in a production program, to withhold it, or possibly even to transfer it to other
manufacturers not involved in the original innovation process.53 Not surprisingly, industry rejected
this arrangement and held the initiative in limbo until the final MoU was revised, in September
1990, to give manufacturers shared control over their own R&D discoveries. This issue has yet to
be adequately resolved to the satisfaction of all members of the European defense industrial commu-
nity. In any case, the fear that states may transfer technology to third parties and thus possibly
undermine the commercial interests of producing firms played a significant role in impeding
EUCLID’s initial development. By the Fall of 1992 only 37 RTPs had been approved and only seven
contracts signed.54
The defense industry’s reluctance to participate in EUCLID was paradoxically matched and
strengthened by the detached behavior of the member states themselves. Even as state governments
shaped the initiative to maximize their presence in the collaborative process, they nonetheless re-
mained uncomfortable with it as a vehicle for cooperation. In 1990 EUCLID signatories pledged to
provide a total of 120 million ECU through national contributions to selected RTPs. These promises
quickly proved hollow, however, as annual state funding failed to exceed half that level. According
to a European Defense Industries Group report, governments were declining to allocate moneys even
to those projects for which they had declared an interest.55 Moreover, when they contributed funds,
52
Assembly of the Western European Union, “The Euclid Programme,” para 20.
53
“Defense Research Programme is Hit by Dispute Over Intellectual Property Rights,” Reuter Text-
line Engineer, August 9, 1990.
54
Walker and Gummett, “Nationalism,” p. 52.
55
Assembly of the Western European Union, “The Euclid Programme,” para 21.
147
the amounts were meager when compared to the support for national research programs. EUCLID
projects in 1994 received only 1 percent of the French and United Kingdom defense research spend-
ing, 2 percent of German R&D expenditures and between 5 to 50 percent of that of the other mem-
ber states.56
A report produced by the Western European Union noted that in some technology sectors,
collaboration within EUCLID was at best marginal. For example, the modular avionics CEPA was
an active component of EUCLID from its inception. Yet during its first half-decade, the member
states approved only one, two-year research project in this area, while concurrently devoting tens
of millions of ECU to R&D collaboration in existing minilateral procurement projects involving
similar technologies, most notably Eurofighter and Tiger.57 States consistently kept the bulk of their
collaborative activities outside the EUCLID framework, and by the mid-1990s the initiative had be-
come saddled with an undisclosed, but reportedly significant, number of inactive RTPs as both gov-
ernments and industry either refused to pursue new contracts or withdrew from existing ones.58 Chief
Inspector De Beauchene, member of France’s Délégation Général pour l’Armament and former
EUCLID coordinator, argued that this
. . . collapse was foreseeable partly because the small countries dragged EUCLID
down and partly because the two or three nations committing themselves to
binational or multinational cooperation programs among themselves never helped put
EUCLID across. The program was actually complementary to the mechanisms
concerned with major cooperation programs, but never replaced them.59
In 1994, the EUCLID governments declared their intention to correct this situation and revi-
talize the institution. They created a permanent secretariat in the form of the 7-person Research Cell.
This was appended to the secretariat of the IEPG, now renamed the Western European Armaments
56
Germany, France, and the United Kingdom are the dominant defense markets in Western Europe,
providing upwards to three-quarters of all military expenditures within the EU. Thus higher levels
of relative support by the smaller states actually denote marginal amounts of absolute funding.
Assembly of the Western European Union, “The European Armaments Agency—Reply to the
Thirty-ninth Annual Report of the Council,” Document 1419, 19 May 1994, para 69.
57
Assembly of the Western European Union, “The European Armaments Cooperation After
Maastricht,” Document 1419, 23 October 1992, para 68.
58
Assembly of the Western European Union, “The Euclid Programme,” para 21.
59
Brosselin, “European Military Research in Trouble,” p. 16.
148
Group. The Cell sought to coordinate the CEPA steering committees and to assume management
over new and existing research and technology projects. Moreover, by early 1996, member states
pledged to grant The Cell the responsibility of assigning research contracts—a move that EUCLID
officials hoped would speed up the process and thus motivate greater industrial participation in the
program. In fact, overcoming industry’s reluctance to participate was a major component of
EUCLID’s renovation. The states no longer demanded that industry provide at least 50 percent of
research financing, and under the EUROFINDER procedure overseen by the Research Cell, firms
could now offer unsolicited bids on approved EUCLID contracts.60
It is unclear as whether these institutional innovations produced the desired effect. The most
recent available data, however, suggest that they did very little move the program forward. Decision-
making procedures improved, with the waiting times between preparation to implementation of RTP
contracts falling to approximately nine months. Further, by May 1997, the number of approved of
RTPs had increased by 14 new projects since spring 1996, to a total of 71. Nonetheless, the total
contract value remained unchanged at 250 million ECU, and the average value of EUCLID projects
actually declined from 5 million ECUs to 4 million ECUs per RTP contract.61 Moreover, state
contributions to EUCLID projects continued to stagnate at a level between 50 million and 60 million
ECU per annum—a figure that is less than 3 percent of all yearly IEPG-Europe defense research
spending and perhaps a third of total resources devoted to collaborative R&D projects not included
in weapons platform development.62
Philippe Girard, chairman of the European Defense Industries Group, argued in 1996 that
after nearly seven years, EUCLID’s objectives remained largely “hypothetical.”63 While some work
had been accomplished, arguably benefitting the participating states and firms, the initiative never
achieved a critical breakthrough that would denote sincere interest on the part of both states and
industry to use it as an effective forum for collaborative defense research. Firms remained reluctant
60
Tardif, “EUCLID Corrects Errors of its Youth,” p. 24.
61
Assembly of the Western European Union, “The Euclid Programme,” Appendix I; fn. 4.
62
Carol Reed, “Sharing Out the Cost of R&T,” Jane’s Defense Weekly 10 June 1995, p. 49;
Brosselin, “European Military Research in Trouble,” p. 16.
63
Tardif, “EUCLID Corrects Errors of its Youth,” p. 24.
149
to invest substantial funding at any level to projects not wholly embraced by their own governments
and, moreover, accompanied by foreign industrial partners who might possibly be present only by
government diktat and unable to provide any substantive contribution to the innovation process. The
member states, on the other hand, although eventually accepting some limits on program oversight,
continued to insist on patent ownership rights while treating the initiative with an almost benign
neglect.
Defense economist Andrew James at the University of Manchester contends that EUCLID
never had a chance to develop; indeed, was never given the chance. It has been a relative non-entity,
largely undiscussed in European defense analysis circles and barely remembered elsewhere now at
the end of the decade.64 This is unsurprising in an environment in which 50 percent of all multina-
tional cooperative R&D occurs bilaterally between just two states, France and Germany, and well
outside any multilateral framework. EUCLID never represented a viable alternative to the status quo.
Its founding states crafted a program that offered them exactly what they wanted—access to cooper-
atively funded, advanced defense research—but they engineered it in such a way as to ensure that
they received little that they desired; indeed, not only did the initiative alienate those who would
develop the technology, EUCLID states did not move with any force to either exploit its potential
or to correct its glaring faults.
EUCLID represents an example of what one German official has identified as the pervasive
European habit of placing symbolism over substantive action in matters of collaborative defense.65
It was a showpiece institution intent upon europeanizing defense technological innovation but put
forward in an environment in which the desire for such a transformation was weak. Its failure, how-
ever, was more remarkable than it might seem at first glance. As we have seen, EUCLID was not
the only attempt to create the foundations for a regional technology base in Western Europe during
the 1980s, but it was the least successful. EUCLID research projects rarely focused upon the extra-
ordinary; exotic technologies, such as Stealth or hypervelocity weapons, though officially under the
EUCLID mandate, were not explored at all. In fact, most RTPs dealt with technologies that were not
64
Workshop on The Place of the Defense Industry in National Systems of Innovation, Cornell Uni-
versity. 17-19 October 1998.
65
FBIS Report, FBIS-WEU-98-020, 20 January 1998; Michael Peter and Friedrich Thelen, “The
Last Screw,” Duesseldorf Wirtschaftwoche 11 December 1997, p. 22.
150
inherently defense oriented in function. For example, the first research project conducted under
EUCLID’s Optoelectronics CEPA involved study of light-weight thermal imaging sensors. This
particular technology is as easily adapted for use in the firing control systems in tanks and aircraft
as it is to less lethal applications such as terrain mapping, firefighting, and commercial security
devices. When pursued within the EUCLID framework, however, collaboration was distorted and
endangered by states intent upon securing maximum return from its production.
Conclusion
EUCLID has often been referred to as the “Military Eureka.”66 Like Eureka, its structure was
intergovernmental, projects pursued within its framework received a combination of public and pri-
vate funding, and much of the technology research was dual-use. Most important, its founders ex-
plicitly praised the Eureka model as a template for action that they would apply to the defense field.
Yet after more than ten years, and a handful of research projects, EUCLID governments have failed
to fully embrace the methodology that they had praised and from which—in the civilian area—they
had prospered. EUCLID not only failed to approximate Eureka’s performance, it also did not seri-
ously affect existing state defense research practices. Whereas Eureka projects such as Eurimas,
FAMOS, Prometheus, and COSINE have become important parts of the European techno-industrial
scene, EUCLID’s RTPs were nearly unrecognizable—much like EUCLID itself, in fact.
The question remains why many of the same states that enjoyed Eureka’s achievements were
both unwilling and unable to translate this success into the defense domain. This difference in for-
tunes between the two frameworks is best understood in the context of the symbolic relationship
between defense technology and the state. In many instances, the technologies explored by the two
programs differed by designation only.67 Eureka’s EUROFAR project offers an excellent case-in-
point.68 EUROFAR is a six-firm scheme to research tilt-rotor aircraft design with the stated intent
of satisfying potential demand in civilian air transportation. This particular technology, however,
66
Fouquet, “IEPG,” p. 12.
67
The example of EUCLID’s thermal imaging research and technology project is particularly salient
here. The Eureka project database lists 5 projects that are either underway or have been completed
that are devoted to infrared sensing applications. See http://Eureka.belspo.be.
68
Dickson, “EUREKA!”
151
has immediate military applications, as it can provide capable combat transport and assault plat-
forms—a fact that has not gone unnoticed by the United States Marine Corps, which has actually
weaponized the technology.
Work such as this, when produced under Eureka, is expressly “civilianized” by both govern-
ment officials and industrial representatives; potential defense uses are rarely discussed for fear of
unsettling certain state elites or national publics, even when some of the participant firms may be
defense-oriented companies or corporate divisions. Significantly, however, the conduct of Eureka
does not change during these instances. States continue to allow their firms to conduct interstate
cooperation within a depoliticized environment and according to commercial and technological cri-
teria that served their parochial economic interests.69 Under the EUCLID umbrella, on the other
hand, the rules are completely different. State involvement is strict, “top-down,” and ultimately self-
defeating; and moreover, states behave in this manner for technologies and processes that are far
more banal than innovative airframe designs. Computer workstation design (CEPA 6, RTP 1) or
“training system concepts” (CEPA 11, RTP 1) will not in themselves, carry 40 fully armed soldiers
or deliver a payload of iron bombs. Yet when pursued in a purely defense framework, these assume
subjective values that make them as potent and as desirable to state decision-makers as any complete
weapons system.
The symbolic character of defense technology cannot be overemphasized, as the link be-
tween its possession and the exercise of state sovereignty is extremely strong. States are socialized
to covet this technology unlike any other. It not only provides a means for self-protection in a violent
international system, it also becomes a component of state identity: the embodiment of national
grandeur and autonomy. The evolution of European union invites us to assess the degree that these
ancient ideals are changing in the face of new collective understandings of self and interest.
EUCLID was a symbol of European solidarity in defense—but one that did not exemplify a seamless
convergence of national interest or the subtle effects of a collective regional identity. It instead high-
lights the extraordinary calculus that state decision-making elites will engage in whenever they are
faced with the acquisition of defense technology, one of the last barriers to the deepening of Euro-
pean integration.
69
Peterson, “Technology Policy in Europe,” p. 279.
152
Europeans can collaborate in the development of advanced technology, and do so quite effec-
tively—albeit not necessarily for the most European of motives. Eureka, after all, is a mechanism
by which states strive to accomplish national goals, e.g., protecting home industries and technolog-
ical competencies, through poly-national means. It contains no formal device for promoting intra-
regional competition or opening national markets;70 and it could never, in the words of Pierre-Henri
Laurent, “fuse all national high-tech businesses into one intertwined transstate giant.”71 Nonetheless,
the fact remains that Europeans under the Eureka banner cooperate with considerable zeal, and these
collaborations have pushed forward the internal market. States freely indulge in cross-border
industrial linkages and mutual dependencies that both reflect and create what one might recognize
as the foundation of a common European identity. EUCLID, on the other hand, forcefully
demonstrates that, despite the ideational transformations that may have occurred in Western Europe
over the last forty years, an act of definition as “defense” can determine the perimeters of European
cooperative behavior, because that designation evokes traditional conceptions of national identity
and interest that are inimical to the establishment of any effective collaborative regime.
70
Guy de Jonquieres, “European High Technology; Fresh Impetus for Collaboration,” Financial
Times, 30 June 1986, SURVEY, p. 1 [newspaper on-line]; available from http://web.lexis-
nexis.com/universe; Internet; accessed 16 December 1998.
71
Pierre-Henri Laurent, “Eureka, or the Technological Renaissance of Europe,” Washington Quar-
terly 10 (Winter 1987) [journal on-line]; available from http://web.lexis-nexis.com/universe;
Internet; accessed 16 December 1998.
III. THE PLACE OF THE DEFENSE INDUSTRY IN TRANSITION
AND INDUSTRIALIZING ECONOMIES
Judith Sedaitis
153
154
of which path restructuring takes, it is clear that to survive, former Russian innovators need to learn
how to shift their strengths and findings in military research to civilian production and new product
development.
Two processes capture the myriad of new institutional arrangements for restructuring Rus-
sian science and technology: the integration of research units with production, and the opposite pro-
cess, a fragmentation of enterprises and institutes into smaller, quasi-separate private firms. After
a brief overview of the organization and collapse of the former Soviet innovation system, this chap-
ter marshals theoretical and empirical evidence on both processes of fragmentation and integration.
Survey data collected from 100 research firms and their spin-off new firms are used to test which
strategy, if any, is more conducive to supporting innovation and the transfer of technology. Finally,
the findings of this research will be reviewed to suggest that, in contrast to official Russian policy
and much Western advice (Schneider 1994; Schweitzer 1996), the nascent sector of private R&D
firms stands poised to become an important new driver of national innovation. In particular, their
combination of access to state technologies without the “responsibilities of largeness,” helps the new
and parasitical firms transfer technology and innovation away from their former military focus to
commercial applications.
Table 1 indicates, in response to the precipitous crisis their formal numbers grew steadily, from 1762
in 1990 to 2200 in 1994! This ostensible growth was one of fragmentation. Unable to pay their
workers, institutes have allowed them to set up their own institutes in order to increase the trickling
flow of state subsidies.
Table 1
Russian R&D Organizations (1991 - 1994)
Change
R&D Organizations 1990 1991 1992 1993 1994 1991-94
Scientific Institutes 1762 1831 2077 2150 2240 +27.1
Engineering-design 937 930 865 709 589 -37.1
bureaus
Engineering- 593 559 495 395 353 -40.1
construction bureaus
Experimental 28 15 29 17 20 -28.6
factories
In-house 449 400 340 299 302 -32.7
Design bureaus were differentiated as those that designed structures or designed new
products and processes. The latter began their work after receiving the initial research and develop-
ment phase from scientific institutes. Design bureaus specified the ideas in the form of working
drawings and other technical documents that were then sent either to an experimental factory for
prototype testing or directly to the end-user production facility. In the early 1970s there were
approximately 4000 design bureaus. As Table 1 indicates, the number of design bureaus dropped
sharply between 1991-1994. Construction design bureaus in particular were hit hard by the transi-
tion, and their numbers fell by about 41 percent. It is unclear, however, to what extent the decrease
in number of design bureaus is the result of mergers with production facilities.
Finally, Soviet research was also carried out in the small in-house research division of Soviet
enterprises. These were always the smallest, most incrementally-oriented organizational sector of
the R&D establishment and numbered approximately 449 in 1990. Following the experience of East-
ern Europe, these in-house units have the greatest prospects for the future in sectors that experience
growth. Gokhberg and Kuznetsova (1998) argue that most innovation is currently driven by these
156
in-house R&D units in the Russian mining and chemical sectors, specifically. They are currently
among the most competitive and hence most able to invest in R&D. However, others find that only
about 6 percent of all in-house R&D units introduced any new product or process over the early
transition years (Gaponenko 1995). Their number fell by about 33 percent to 302 units in 1994 and
continued to fall through the 1990s.
In Russia, these advantages point to the greater efficiency and control of integrating the for-
merly stand-alone R&D sites with production facilities through share-holding (Schneider 1994).
Thus, organizational integration should help former R&D facilities retain the better part of their
resources and staff, while making it more efficient to adapt their capabilities to manufacturers’ needs
and new product development.
Consolidation of former defense enterprises along the same production chain is clearly sup-
ported by certain groups of Russian policy-makers as well. The efficiencies of vertical integration
were a large part of the rationale given by the Ministry of Defense for its official program supporting
the creation of so-called “financial-investment groups.” Beginning in 1993, the financial-industrial
group (FIG) form was promoted by Deputy Defense Minister Kokoshin as the best Russian approx-
imation of the successful industrial groups in Southeast Asia, such as the kereitsu in Japan or the
chaebol in South Korea (FBIS-SOV June 11, 1993; FBIS-SOV July 27, 1993). However, subsequent
legislation aimed at defining and regulating official FIGs disallowed cross-shareholding and limited
block shareholding to no more than 10 percent ownership. These stipulations thwarted any strong
investment or governance role for banks, which had been a key motivating factor behind their initial
popularity. As a result, only 5 FIGs were officially registered by September 1995, including one in
the defense sector. Unofficially, however, the tendency to consolidate ties among former defense
enterprises or to form new groups has continued, especially in the aviation sector.1
Two of the largest of these conglomerations include the Yakovlev Design Bureau. It teamed
up with two separate aviation companies to create a group of cross shareholders that together cover
the whole stream of aircraft production (Sanchez-Andres 1995). Another world-renowned design
lab, the Krunachev KB, designer of the Energia rocket systems, became the lynchpin of the new
Energia Association, an effort to join several design, diagnostic, and manufacturing firms in Mos-
cow and in other cities across European Russia. Although both groups claim that the majority of
their output is now civilian, the state has retained a controlling block of shares in these and other
industrial conglomerations of former defense enterprises (Drugov 1995).
The second factor of continued or increased role of the state in Russia’s future science base
involves the importance of state funds in support of the difficult re-tooling and conversion process.
1
For the purposes of this research, consolidated enterprises are defined as those in which dominate
ownership is held by outside, domestic organizations (excluding state agencies).
158
After serious neglecting the research sector in the initial throes of market reform, the Russian
government began planning to increase subsidies for technology transfer to commercial usage. In
April 1995, a special decree on state support for the development of science announced that future
civilian R&D investment would be increased to at least 3 percent of the federal budget in 1996
(FBIS-SOV April 28, 1995, p. 24). Similarly, the new conversion fund decree earmarked 7.3 trillion
for the new 1995-97 conversion program, up from the 1.4 trillion promised in 1994-95 (FBIS-SOV
September 20, 1995, p. 38; FBIS-SOV January 22, 1996, p. 49). The new conversion program is
centered on facilitating import substitution, similar to current Chinese policy and to earlier Israeli
policies. The plan also includes support for civilian output worth 41.1 trillion 1995 rubles with the
goal of absorbing thousands of former defense sector specialists.
Certainly state funding can ease the difficulty of market transition, and the option of integra-
tion may be appropriate for some industries, such as the competitive aviation sector in Russia. How-
ever, the program of further state control and integration generally faces several structural obstacles
and political criticisms. Close Western observers suggest that integration in Russia may be less a
new attempt to rationalize and de-militarize Russian R&D than the preservation of the former
military industrial complex (Blank 1995). Consolidation may support a hawkish political agenda of
increasing Russian military presence in the region or even reintegrating the new nations of the CIS
into a singular military production structure (Blank 1995). In addition, the legacy of the centralized
command structure of Soviet R&D makes this model both difficult and inappropriate.
Most Russian enterprises are specialized, stand-alone facilities whose rationale depended on
administrative links and a political system that are now defunct. In the Soviet Union, research insti-
tutes and design bureaus were often physically separated by vast distances from their production
sites (Berliner 1983). In addition, the culture of secrecy surrounding their work meant that units in
the same development chain were administratively organized into separate ministries and their
researchers put into competitive relationships with one another. These security concerns seriously
precluded the possibilities of collegial exchange among specialists and limited critical feedback from
downstream the production flow (Evangelista 1988).
As a result, the level of Soviet technological innovation was generally poor. In addition to
the limits of their segregated organization, R&D units suffered the same irrational constraints of the
administrative command system that stymied risk-taking and personal initiative across the Soviet
economy. Instead of supporting innovation, therefore, the Soviet R&D system evolved into one less
159
geared towards creating innovations itself than at finding and adapting innovations made elsewhere,
including and especially military technologies (Holloway 1983). As such, the R&D units bear little
resemblance to the Japanese industrial groups or U.S. research consortia that the Ministry of Defense
purports to emulate. Without major internal revisions, the simple integration of former research units
could risk preserving traditional Soviet management practices and dependence on state subsidies that
will hardly foster economic competitiveness (Kuznetsov 1994).
ment with new product development at low cost. A new, shell entity within the institute was desig-
nated as the “holding company,” which was also the repository of the shares of the new private satel-
lite firms.2 The extent of shares owned and control exercised by the state lab varied depending on
the nature of the technology involved. Unlike the spin-off process in U.S. labs, however, the Russian
state lab gets back a portion of the profits from successful transfer attempts and is more inclined to
encourage rather than constrain access to the potentially most practical technologies. Ideally, the
state-owned core retains control over sensitive research and gives scientists an environment dedi-
cated to the pursuit of scientific knowledge, unpressured by demands for practicality. At the same
time, those same employees who are seeking opportunities to apply their research are able to do so
without draining the lab of talent or putting themselves at great financial risk. Organizational de-
coupling allows for control where it is necessary and market forces where it is not. As a result, the
spinning off of new satellite firms is sweeping across Russia (Sabel and Prokop 1996; Schweitzer
1996).
A hallmark of Soviet collectivity was the wealth of tacit knowledge built up over the many
years that Russian research teams typically stayed together. Spin-offs could ostensibly help convert-
ing enterprises hold on to their talented staff by providing them a safe haven for releasing their crea-
tive and commercial capabilities (Sabel and Prokop 1996). They offered the scientific entrepreneur
access to the often sophisticated tools and infrastructure of the enterprise for a fraction of the over-
head costs. At the same time, satellites were not encumbered with the range of social and state obli-
gations, such as providing social welfare and guaranteed employment for their workforce. They were
more free to engage in the high risk activities of adapting innovations to new products and to make
the maximum use of the slack and extensive resources at extant, parent institutes. In this way, satel-
lites could access the positive attributes of the old-style Soviet R&D firms, while sloughing off the
rest.
Despite the potential of satellite firms as the genesis of a new Russian technology base, the
fragmentation of formerly large military science institutes is not without its potentially negative
2
Before privatization, a “spin-off” is a new firm whose capital stock and premises are controlled by
a parent firm. After privatization, spin-off refers to those firms whose simple majority share belongs
to a “founding” organization. In Russian legal terms, a spin-off is a “dependent” firm when the par-
ent company owns at least 20 percent of its stock and is a “daughter” firm when the parent company
owns at least 50 percent.
161
consequences as well. Russian policy-makers feared that satellite firms would exacerbate the hemor-
rhaging of large institutions who were losing their top scientific personnel or that they would give
foreign companies the opportunity to “cherry pick” the best of what a parent state institute had
developed without full compensation. The phenomena became so wide-spread and worrisome that
the formal creation of daughter companies at R&D organizations was legally forbidden in 1994
without the approval of the relevant ministry (Kayukov and Silliman 1997). Satellites might also
increase the risk of lapsed security and easy access to dangerous nuclear or chemical materials that
are housed in these institutes (Marten-Zisk 1994). Or their commercial fervor could help facilitate
the sale of Russian arms and technology to unfriendly, rogue states (Shlykov 1995; Sapir 1997; von
Hippel 1995; Cheung 1993).
The goal of this research was essentially to consider both the causes leading up to satellite
creation and the ramifications of integration versus fragmentation. In particular, the effect of having
created daughters was used to explain the exodus of staff from the parent company, the parent’s
income stream, and the parent’s attempts to transfer technologies to new product development. The
first of these variables was measured as the difference between the number of people employed in
1995 and the number of people that used to work at the same enterprise in 1990. Income was
measured as the firm’s yearly sales turn-over in millions of rubles for 1994. Finally, the success of
technology transfer was operationalized as a weighted index based on the response to twelve con-
crete mechanisms gleaned from the literature and compiled by Gibson and Rogers (1994, 354).
These mechanisms ranged from passive methods, such as publishing scientific articles, to more
active and interactive methods, such as arranging special site visits for showcasing technologies to
potential users and investors.
Fragmentation was construed as the process of creating satellite firms. This concept was
operationalized as an indicator variable that singled out companies that had created at least one
daughter firm. Conversely, integration was measured as an indicator variable that separated com-
panies that had reliable ties to suppliers from those that did not.
Two extraneous elements must be controlled for in the regression analysis. One is the role
of the state, which is crucial in the early transition phase. Links to the state were conceptualized as
both financial and political. Financial links were measured as the percentage of a company’s work-
ing capital coming directly from state funds. Political links were identified with the strength of ties
to state administrators and operationalized as an index measuring enterprises’ lobbying activities
with respect to the executive branch of the federal government. The second important control vari-
able was the size of the firm. In addition, another control variable that measured the firm’s retention
of scientific employees was included in the regression analysis. The size of the enterprise was
measured as the natural log of current staff size excluding temporary employees. The retention of
scientific staff was calculated as the change in the relative proportion of scientific staff to total staff
between May 1994 and January 1991.
First, however, the antecedents to satellite creation were considered. These include the pro-
portion of scientific output of a firm (i.e., drawings, prototypes, computer code, analytic/diagnostic
services) to total output and the ratio of consumer products to total output. Ownership and profita-
bility were also important considerations of whether a firm would be likely to fragment and allow
163
satellite firms, in particular the ownership structures over intellectual property. Where a parent has
clarified intellectual property rights or feels secure in their control, the creation of satellites raises
fewer risks to that control and should be more likely. Finally, as noted above, the size of the firm
was controlled for by using a natural log of the original size of the firm in 1990 as one of the
independent variables.
Findings
Which types of Russian R&D firms were the most likely to create satellites? Interestingly,
our preliminary research points to a link between firms that self-reported themselves as profitable
as one of the strongest predictors. Those that were profitable in 1994 appeared more likely to also
be those firms that created daughter satellite companies, even when controlling for the important
variable of original firm size. The two other variables that were significant in the logistic regression
were the importance of intellectual property rights holdings by the parent firm and the proportion
of scientific products in the output of the firm (see Table 2). As expected, those that felt secure in
their control as the originators of innovation were also those more likely to allow for fragmentation.
Similarly, satellite creation was also associated with those parents with a greater proportion of scien-
tific production. The results of the regression, taking the number of the satellite companies created
as the dependent variable, are very similar. The effects of profitability are somewhat weaker, as the
relevant coefficient is significant only at the 10 percent level. But the security of intellectual prop-
erty rights and the proportion of scientific goods in total output remain highly significant for the
process of satellite firm creation.
Our preliminary findings on the consequences of creating satellites is consistent with those
tracing the antecedents leading to satellite creation. Overall, the creation of daughter firms at former
Soviet R&D enterprises appears more important than integration. Table 3 indicates the effect of
these relationships on staff exodus and firm income streams, and on the firm’s attempts to actively
engage in technology transfer. In terms of staff exodus, daughter satellite firms lost the most. Parents
of satellites also showed a positive relationship to staff exodus, but it was only significant at the 10
percent level. Interestingly, this model indicates that those firms that had lost the larger percentage
of their technical staff were more likely to retain their other, non-technical staff. This finding coin-
cides with other anecdotal data on the difficulty facing the large state institutes, whose best and
164
brightest have left for more lucrative settings, leaving their institutes burdened with the non-creative
staff to support.
Table 2
Characteristics Associated with the Creation of Satellite Companies
Satellite creation showed a positive relationship to firm income. Just as firms reporting prof-
its were more likely to create satellites, they also showed more positive income streams. The satel-
lites themselves did not, however. Interestingly, the politicized firms with strong ties to state offices
showed a significantly negative relationship to income.
Finally, in terms of technology transfer, both integration and fragmentation evidenced strong
links to active technology commercialization attempts. Being the parents of satellite firms was help-
ful, but so were strong ties to supplier firms as well as state financial subsidies. Most interestingly,
those strongly linked to active technology transfer were the small, new satellite firms themselves.
165
Table 3
Regression of key organization variable to staff exodus, firm profitability,
and fostering technology adaptation
Discussion
Given the size of this sample and the difficulty of gathering objective data in the FSU, the
findings on satellite creation can be only suggestive at best. Nonetheless, they point in the opposite
direction from both the American organization of effective industrial R&D and the image of satellite
firms in the popular Russian press. Contrary to the fears of Russian policymakers (Kayukov and
Silliman 1997), satellites seem not to drain away manpower and profits at the expense of adjustment
by extant parent organizations. Instead, those firms that created satellites show more income than
extant firms that did not and more income than the satellites themselves. This may reflect one feature
of satellite creation criticized by Russian economists, that they allow parents greater flexibility in
behavior, but also for record-keeping and tax evasion, which is especially important to rule-bound,
state-owned enterprises. In turn, the satellites themselves appear the most focused on the project of
channeling older technologies in new, market-oriented directions and, as such, have higher staff
turnovers and lower income streams. The Russian case of technology adaptation stands in stark
166
contrast to the Western experience, where concerns about the personal enrichment of private-sector
firms using public research have constrained the support for entrepreneurial activity at U.S. federal
labs.
At the same time, the organizational approach does not so much oppose the market model
as it seeks to explain the dynamics that drive it. As such, the fragmentation process of creating new,
satellite firms is not mutually exclusive of state support, but rather indicates how the firms used the
important breathing room that federal funding made possible. State subsidies while firms attempted
conversion were helpful to this process. Similarly, the rational integration of firms within their own
supply chains also gained support in the research, and those firms with stronger ties to other, down-
stream producers were also those more actively seeking to transfer their technologies. While the new
satellites themselves may be the harbinger of Russia’s future technology sector, therefore, this future
is still distant.
References Cited
Alic, J.A.. 1997. “Science, Technology and Economic Competitiveness.” In Commercializing High
Technology: East and West, ed. J.B. Sedaitis. Lanham, MD: Rowman and Littlefield.
Alic, J.A., L. Branscomb, H. Brooks, A. Carter, and G. Epstein. 1992. Beyond Spinoff: Military and
Commercial Technologies In a Changing World. Boston: Harvard Business School Press.
Berliner, J.S. 1976. The Innovation System in Soviet Industry. Cambridge: MIT Press.
Bergson, A. and H.S. Levine, eds. 1983. The Soviet Economy: Toward the Year 2000. London:
Allen and Unwin.
Blank, S.J. 1995. “Reform and Revolution In Russian Defense Economics.” The Journal of Slavic
Military Studies 8(4): 691-717.
Branscomb, L.M. 1993. “National Laboratories: The Search for New Missions.” In Empowering
Technology, ed. L.M. Branscomb. Cambridge: MIT Press.
Cheung, T.M. 1993. China’s Buying Spree. Far Eastern Economic Review 8: 24-26.
Coombs, R., P. Saviotti, and V. Walsh. 1987. Economics and Technological Change. London:
MacMillan.
Drugov, Y., A. Danilenko, S. Cherpakov, A. Kalin, A. Kokorev, D. Korobkov, A. Ponomarev, Y.
Simachev, A. Svinarenko, and Y. Vinslav. 1995. Reforms in Russia: Industrial Complex,
Part I. Moscow, Russia: Kniga-Service.
Evangelista, M. 1988. Innovation and the Arms Race. Ithaca: Cornell University Press.
Florida, R. and M. Kenney. 1990. The Breakthrough Illusion: Corporate America’s Failure to Move
from Innovation to Mass Production. New York: Basic Books.
Gibson, D.V. and E.M. Rogers. 1994. R&D Collaboration on Trial. Boston: Harvard Business
School Press.
Gaponenko, N. 1995. “Transformation of the Research System in a Transitional Society: The Case
of Russia.” Social Studies of Science 25: 685-703.
167
Gokhberg, L. and I. Kuznetsova. 1998. Technology Innovation in Russia. Moscow: Centre for Sci-
ence and Research Statistics.
Holloway, D. 1984. The Soviet Union and the Arms Race. New Haven: Yale University Press.
Kayukov, E. and E. Silliman. 1997. “New Company Formation in Russia: Legal Regulation.” In
Commercializing High Technology: East and West, ed. J.B. Sedaitis. Lanham, MD: Rowman
and Littlefield.
Kuznetsov, Y. 1994. “Adjustment of Russian Defence-Related Enterprises in 1992-1994: Macroeco-
nomic Implications.” Communist Economies and Economic Transformation 6, 4: 473-513.
Marten-Zisk, K. 1995. “Arzamas-16: Economics and Security in a Closed Nuclear City.” Post-
Soviet Affairs 11: 57.
Mowery, D.C. 1992. The U.S. National Innovation System: Origins and Prospects for Change.
Research Policy 21(1): 125-44.
———. 1994. “The Changing Structure of U.S. Industrial Research: Implications for R&D Organi-
zation in the Russian Federation.” International Journal of Technology Management 9(5-7):
89-102.
Piore, M.J. and C. Sabel. 1984. The Second Industrial Divide: Possibilities for Prosperity. New
York: Basic Books.
Powell, W., K. Koput, and L. Smith-Doerr. 1994. “Technological Change and the Locus of Innova-
tion: Networks of Learning in Biotechnology.” Paper presented at the Academy of Manage-
ment annual meeting, Dallas, TX.
Sabel, C.F. and J.E. Prokop. 1996. “Stabilization Through Reorganization: Some Preliminary Impli-
cations of Russia’s Entry into World Markets in the Age of Discursive Quality Standards.”
In Corporate Governance in Central Europe and Russia, ed. R. Frydman, C.W. Gray, and
A. Rapaczynski. Washington, DC: World Bank.
Sanchez-Andres, A. 1995. “The First Stage of Privatization of Russian Military Industry.” Com-
munist Economies ~ Economic Transformation 7, 3: 353-367.
Sapir, J. 1997. “Defense Conversion and Restructuring in the Russian High-Technology Sector: Is
There an Alternative to Uncontrolled Exports?” In Commercializing High Technology: East
and West, ed. J.B. Sedaitis. Lanham, MD: Rowman and Littlefield.
Schneider, C.M. 1994. Research and Development Management: From the Soviet Union to Russia.
Heidelberg: Physica-Verlag.
Schweitzer, G.E. 1996. Moscow DMZ: The Story of the International Effort to Convert Russian
Weapons Science to Peaceful Purposes. Armonk, NY: M.E. Sharpe.
Sedaitis, J.B. 1995. “Comparative Analysis of International High Technology Alliances in Russia.”
Paper presented at the Pacific Sociological Association Meetings, Seattle, March.
———. 1997. “Commercializing State-Owned R&D: A Russia-U.S. Comparison.” In Commer-
cializing High Technology: East and West, ed. J.B. Sedaitis. Lanham, MD: Rowman and
Littlefield.
Shlykov, V.V. 1995. “Economic Readjustment within the Russian Defense-Industrial Complex.”
Security Dialogue 26,1: 19-34.
von Hippel, E.A. 1978. “Users as Innovators.” Technology Review 80 (January): 31-39.
von Hippel, E.A. and M.J. Tyre. 1995. “How Learning is Done: Problem Identification in Novel
Process Equipment.” Research Policy 24,1: 1-12.
Williamson, O.E. 1981. “The Economics of Organization: The Transaction Cost Approach.” Ameri-
can Journal of Sociology 87: 548-77.
The Defense Sector as a Window into China’s National System of Innovation
Corinna-Barbara Francis
Introduction
China’s approach to market reform has allowed and encouraged individual public institutions
and government agencies to engage directly in business.1 Governmental offices and departments,
educational, cultural, religious, scientific, and other types of public institutions actively engage in
the market economy by operating their own profit-making companies and enterprises,2 and they
form joint ventures with other agencies. In other cases entire agencies or institutes are transformed
into semi-independent corporations. While China’s private sector still remains relatively small, the
sector of “quasi-public” firms affiliated with government agencies and institutions has been domi-
nant in China’s market. This approach to market transition raises a host of questions. What precisely
is the nature of this government entrepreneurship? How does it work? What is its impact on the
workings of the market, on the functioning of government, and on the integrity of public institu-
tions?
China’s military-defense institutions, like most other governmental and public agencies, have
been actively engaged in commercial activities since the initiation of market reforms in the late
1970s. This chapter explores the key features of military entrepreneurship in China, using the mili-
tary to illustrate key features of China’s evolving national system of innovation and emerging mar-
ket system. China’s military (despite having distinct qualities) reflects key aspects of China’s evolv-
ing national system of innovation. First, it reflects the degree to which entrepreneurship has penetrat-
ed to the core of the state. China’s military’s vast business empire reflects the fact that virtually no
1
Marc Blecher, “Development State, Entrepreneurial State: The Political Economy of Socialist
Reform in Xinju Municipality and Guanghan County,” in Gordon White, ed., The Chinese State in
the Era of Economic Reform: The Road to Crisis (London: Macmillan, 1991); Jane Duckett, “The
Emergence of the Entrepreneurial State in Contemporary China,” The Pacific Review 9, no 2 (1996):
180-98; Milton Mueller and Zixiang Tan, China in the Information Age: Telecommunications and
the Dilemmas of Reform (Washington, DC: Praeger Publishers, 1997).
2
Corinna-Barbara Francis, “Bargained Property Rights in China’s Transition to a Market Economy:
The Case of the High-Technology Sector,” in Andrew Walder and Jean Oi, eds., Property Rights
and Economic Reform in China (Stanford: Stanford University Press, 1999).
168
169
niche within the public sector has been immune to the pressure to “jump into the sea,” i.e., to engage
in commerce. Second, military entrepreneurship illustrates the fragmented structure of ownership
and management within this quasi-public sector associated with government entrepreneurship. Far
from being firmly controlled by “the state” as a whole, or even by large macro-institutions and re-
gional entities within the state such as the military, control over enterprises is decentralized to myr-
iad low-level units. We cannot even say that the top command within “the military” exercises firm
control over its sprawling business empire. Rather, control is fragmented among myriad sub-units
of the military.
Third, military entrepreneurship illustrates both the negative and positive features character-
istic of China’s approach to market transition. The last section of this chapter examines the impact
that military entrepreneurship has had on China’s economy—its contribution to the relatively more
efficient use of resources, increased revenue, and other positive impacts on China’s market transition
and economic modernization on the one hand, and the rise of corruption, the anti-market forces, and
the erosion of the integrity and sanctity of the military as an institution on the other hand.
The concept of a national system of innovation is used here in the broadest sense to refer to
the distinct characteristics of the process of research and development, production, and diffusion of
economic goods (including but not restricted to technology) as it operates within a nation-state.3 This
concept has been criticized on a number of grounds, including the choice of the nation-state as the
critical unit of analysis, which is said to overstate the distinction of national institutions and to
downplay the impact of global networks and the global impact on national economic processes.4
However, the concept is useful in the Chinese context, due in part to the relative isolation of the Chi-
nese economy from the global economy for many decades prior to the post-Mao economic reforms
and to the continued impact of China’s distinct institutions on economic processes.
3
Richard Nelson, ed., National Innovation Systems (New York: Oxford University Press, 1993);
Charles Edquist, ed., Systems of Innovation: Technologies, Institutions and Organizations (Washing-
ton, DC: Pinter Press, 1997).
4
Slavo Radosevic, “Defining Systems of Innovation: A Methodological Discussion,” Technology
in Society, 1998: 75-86.
170
5
According to one estimate the number doubled from 10,000 in 1985 to 20,000 several years later.
Cheung, “The Chinese Army’s New Marching Orders,” in Mixed Motives: Uncertain Outcomes:
Defense Conversion in China, ed. Jörn Brömmelhörster and John Frankenstein (Boulder: Lynne
Rienner, 1997), p. 183.
6
Ibid.
7
In 1992 Deng Xiaoping took a well-publicized trip to China’s southern region, visiting the Shen-
zhen economic zone, among other places. This trip was used to signal his intention to revive eco-
nomic reforms following the slow-down after the 1989 Tiananmen Square incident, and it sparked
a rapid nation-wide resurgence of market activities in China.
171
as suggested by the military’s inclusion in the “Four Modernizations,” the military-defense budget
declined steadily as a portion of the state budget during the 1980s, dropping from 16 percent in 1980
to a low of 7.52 percent in 1993.8 Despite small nominal annual increases the military budget expe-
rienced a steady decline in real economic terms due to inflation.9 This financial pressure has forced
the military, as it has other institutions like universities, research institutes, cultural organizations,
etc., to find its own financial solutions and to generate more of its own revenue.
Second, the military illustrates how government policies have directly encouraged public
organizations to engage in commercial activities. Specifically, the policy of military conversion that
was promoted in the late 1970s and early 1980s was similar to policies that the government pro-
moted in other sectors.10 It had similar goals—to push institutions in a profit-oriented direction and
to enhance the use of specialized technology and resources (in this case military technology) for
broader use in the civilian and commercial arenas. As officially stated, military conversion was to
“reform and convert the past unified military product system into an integrated military-civilian
national defense scientific research and military-industrial production system.”11 This reflects poli-
cies applied to science and technology (S&T) sectors, which have sought to enhance the utility of
technology and scientific resources for commercial production.12 Finally, military entrepreneurship
8
Paul H.B. Goodwin, “Estimating China’s Military Expenditure,” in Chinese Economic Reform: The
Impact of Security, ed. Gerald Segal and Richard H. Yang (New York: Routledge, 1996). By
contrast, the total portion of the state budget allocated to culture/health/education/S&T increased
slightly over the same period. In real economic terms, according to Chinese government reports, the
military budget declined at a rate of 5.8 percent in the 1980s.
9
The government’s new international security policy in the 1980s, which stressed the peaceful
nature of China’s international environment, resulted in sharp drops in production orders for military
enterprises, thus putting added financial burden on the military.
10
For further discussion of these factors in the high-technology sector see Francis, “Bargained Prop-
erty Rights in China’s Transition to a Market Economy.”
11
Commission on Science, Technology and Industry for National Defense (COSTIND), Science and
Technology Intelligence Agency, ed., Shijie junshi gongye gailian [Survey of World Military Indus-
try] (Beijing: National Defense Industry Publishing House, 1990), p. 107. Also quoted in Mel
Gurtov, “From Swords to Market Shares,” China Quarterly, No. 134 (June 1993): 213-241.
12
For an overview of reform policies in the S&T sector see Shulin Gu, A Review of Reform Policy
for the S&T System in China: From Paid Transactions for Technology to Organizational Restruc-
turing (United Nations, Institute for New Technologies, Maastricht, The Netherlands, UNU/
172
also reflects the simple role of greed—given the opportunity, organizations and individuals have
acted to enrich themselves, a motivation going far beyond simple institutional survival.
Military Conversion
The policy of military conversion promoted by the Chinese state in the late 1970s and early
1980s has been part of a broad economic strategy of shifting public resources to more efficient and
productive uses in order to serve economic growth and modernization.13 In the Chinese context de-
fense conversion has entailed not only the transfer of military technology to civilian use, but as im-
portantly the commercialization of defense production under the continued supervision of the mili-
tary. Policies encouraging the shift of military production to commercial markets are reflected in
other sectors—educational, cultural, and scientific—in which public institutions have been pushed
to be financially more self-sufficient. These policies have aimed at relieving the state’s budget bur-
den and have been key to China’s overall modernization strategy. The greater financial self-suffi-
ciency of public institutions is expected to contribute to their eventual modernization. The revenue
generated by the military’s commercial activities is expected to contribute to military moderniza-
tion.14
The clear distinction between civilian and military products makes it easier to assess the
degree of commercialization within the military, as opposed to civilian industries. One can assume
that production of civilian goods or provision of services by the military has been largely driven by
profit motivation.15 Conversion has been most complete in particular areas within the military, such
as high technology, satellite technology, computer and electronics, microelectronics, nuclear energy,
INTECH Working Paper No. 16, January 1995) and Gu, Spin-off Enterprises in China: Channeling
the Components of R&D Institutions into Innovative Businesses (United Nations, UNU/INTECH
Working Paper No. 16, Maastricht, The Netherlands, December, 1994).
13
Jörn Brömmelhörster and John Frankenstein, eds., Mixed Motives: Uncertain Outcomes: Defense
Conversion in China (Boulder: Lynne Rienner, 1997).
14
This policy had the additional goals of absorbing excess personnel and production capacity, pro-
viding job opportunities for public sector employees, and putting to greater use the resources often
hoarded by public institutions.
15
In order to more completely assess the shift to for-profit production one would also have to add
production of arms and military equipment for sale.
173
aerospace, lasers, and more.16 However, it has affected all sectors to a degree. At the beginning of
China’s economic reforms in the late 1970s around 8 percent of military output was in civilian
goods. By the mid-1990s these figures were nearly reversed; it is estimated that civilian production
accounted for around 80 percent of total military output.17 According to a People’s Daily article, in
1994 77.4 percent of gross output value of the military-defense complex was in civilian products.18
In 1993 only about 10 percent of China’s defense production capacity was being used for defense
production.19 The figure varies between military industrial sectors. The electronics industry expe-
rienced one of the most rapid conversions, with civilian production reaching 97 percent in 1992.20
The Ordnance Ministry stated that in 1994 90 percent of its production was civilian products.21
16
Mel Gurtov and Byong-Moo Hwang, China’s Security: The New Roles of the Military (Boulder:
Lynne Rienner, 1998), p. 155.
17
According to the 1995 OECD Report on Chinese defense conversion, about 80 percent of the
Chinese military-industrial complex had been involved in conversion by 1995. Reported in John
Frankenstein and Bates Gill, “Current and Future Challenges Facing Chinese Defence Industries,”
China Quarterly, no. 146 (June 1996), p. 419, fn. 60.
18
“Conversion of Military Technology to Civilian Use Discussed,” FBIS, 25 March 1994, p. 35.
Also cited in John Frankenstein, “China’s Defense Industry Conversion: A Strategic Overview,” in
Mixed Motives, ed. Brömmelhörster and Frankenstein, p. 3.
19
“Making a modern industry,” Jane’s Defense Weekly, 19 February 1994, p. 28. Also quoted in
Frankenstein and Gill, “Current and Future Challenges,” p. 396.
20
Frankenstein, “China’s Defense Conversion,” p. 3. Cited in China Electronics Industry Trading
Delegation catalog, 1993 Hong Kong Conference on International Cooperation to Promote Con-
version from Military to Civilian Industry, Hong Kong, 7-11 July 1993, p. 1.
21
“Arms Maker Produces More Civilian Goods,” FBIS-CHI, 9 June 1994, p. 36. Cited in Franken-
stein, “China’s Defense Conversion,” p. 3.
174
ation, hierarchical level, and region. It thus mirrors the fragmented structure typical of China’s gov-
erning institutions.22
At the highest level China’s military-defense business empire is divided between the Peo-
ple’s Liberation Army—the name given to China’s combined army, navy, and air force—and its
defense industries, such as the Ministry of Nuclear Industry and the Ministry of Ordnance, which
have overall responsibility for the research, development, and production of military equipment and
technology.23 This institutional separation has been in place since the 1950s when captured Guomin-
dang defense factories were placed under the control of newly established defense ministries rather
than under the PLA.24 The two components of the military-defense complex are now supervised
through separate channels. The Central Military Commission (CMC), which has both a party and
a government component, has been the highest military authority. It oversees the PLA through three,
co-equal, departments—the General Staff Department (GSD), the General Political Department
(GPD), and the General Logistics Department (GLD). The State Council supervises the defense in-
dustries through a variety of organizations, including the State Planning Commission, the now dis-
banded State Economic Commission, and the Ministry of Finance, which also have responsibilities
for non-defense related production.25
22
Kenneth Lieberthal and Michel Oksenberg, Policy Making In China: Leaders, Structures, and
Processes (Princeton: Princeton University Press, 1988).
23
Between 1978 and 1987 the 6 defense ministries were as follows: 1) The Ministry of Nuclear In-
dustry; 2) the Ministry of Astronautics Industry, including space rocketry satellites, telecommunica-
tions, and microelectronics; 3) the Ministry of Aeronautics Industry, including aviation, metallurgy,
computer-controlled machinery, and electronics; 4) the Ministry of Ordnance, including munitions,
guns, cannons, tanks, and optics; 5) the Ministry of Electronics Industry, including consumer and
military electronics and microelectronics; and 6) the China State Shipbuilding Corporation, in-
cluding civilian and military ocean-going ships, submarines, oil-rigs, etc. For further discussion see
Paul Humes Folta, From Swords to Plowshares? Defense Industry in the PRC (Boulder: Westview
Press, 1992). The term military defense industry is somewhat misleading, as these ministries have
always also produced some civilian goods, although prior to the reforms their output was primarily
military goods.
24
As a consequence, since that time there has been a separation between the ultimate end-users of
military technology and its producers.
25
Folta, From Swords to Plowshares.
175
The defense industrial bureaucracy has gone through important restructuring over the years.
In the early 1960s China’s industries were under eight machine-building ministries (which were
numbered for secrecy), six of which were involved in armaments production.26 A major reform
towards the goal of military conversion was implemented in 1982, when the defense industries were
placed under civilian control—i.e., under the sole supervision of the State Council. In 1988 the vari-
ous defense industries were restructured and reduced to three—the Ministry of Energy Resources
(MER), the Ministry of Machine Building and Electronics Industry (MMBEI), and the Ministry of
Aerospace Industry (MAS).
A third major component of China’s military-defense complex is the set of organizations
whose primary task has been to coordinate between the PLA and the defense industries and to coor-
dinate military R&D and production among the defense industries. The Commission on Science,
Technology, and Industry for National Defense (COSTIND) was established in 1982 through the
merging of the National Defense Science and Technology Commission (NDSTC) and the National
Defense Industry Organization (NDIO), a move aimed to overcome the longstanding conflict
between the military research and development sector and production sector and to better coordinate
these tasks and organizations.27 COSTIND was put in charge of supervising China’s military con-
version and given supervisory authority over the commercial ventures of the defense industries.
26
In the early 1960s, coordination of weapons production was vested with the National Defense
Industry Office (NDIO). The National Defense Science and Technology Commission (NDSTC) was
responsible for military R&D, funding, training, and weapons testing. The NDIO and the NDSTC
were supervised by both the CMC and the State Council.
27
For further discussion see Benjamin Ostrov, Conquering Resources: The Growth and Decline of
the PLA’s Science and Technology Commission for National Defense (Armonk, NY: M.E. Sharpe,
1991).
176
ate their own commercial ventures and exercise the key managerial control over these companies.
In some cases companies themselves operate with little effective supervision. PLA enterprises can
be divided into three categories: enterprises managed at the highest level of the PLA; enterprises
operated by regional and provincial-level military units; and enterprises managed by unit-level PLA
entities.
In the first category we can differentiate enterprises operated by the general departments of
the PLA (the GSD, GLD, and GPD), the three armed services departments (the army, air force, and
navy), and other top-level components of the PLA such as the People’s Armed Police (PAP). In this
group the enterprises affiliated with the PLA’s general departments have tended to be the largest and
most profitable. The General Staff Department of the PLA operates two of the largest military-affil-
iated corporations—the Poly Group Corporation and the China Huitong Corporation. Poly was first
established as the foreign weapons trading arm of the GSD’s Equipment Department in 1983 [al-
though it has assumed a civilian identity as a unit of China International Trust and Investment Cor-
poration (CITIC)]. In 1992 it was converted into a nominally independent enterprise group. The
General Political Department also operates some large corporations including the Kaili Corporation
(Carrie Corporation) and China Tiancheng Corporation. Kaili illustrates the diversification of mili-
tary-affiliated corporations. It began as a trading company, later becoming active in real estate devel-
opment and audiovisual products. It is also said to run an iron-ore mining company in Australia and
a bank in Cook Islands, has a range of businesses in Hong Kong, and is involved in international
weapons sales.28
The General Logistics Department, the third general department in the PLA, also operates
some large corporations. One of the largest is the China Xinxing Corporation, which was founded
in 1984 and now has more than 100 enterprises and a range of subsidiaries in a variety of indus-
tries.29 Xinxing was initially established to promote the commercial sale of products that GLD fac-
tories had earlier produced for the army. It then gradually diversified and subsequently became the
28
Cheung, “The Chinese Army’s New Marching Orders,” p. 186-87.
29
According to one estimate, the Xinxing Corporation has around 140,000 workers. Cheung, “The
Chinese Army’s New Marching Orders,” p. 187.
177
GLD’s core company engaged in weapons sales.30 Another GLD business group, the 999 Enterprises
Group, was formed in early 1992 in order to bring together under unified management all the GLD
companies operating in the Shenzhen Special Economic Zone. According to one report, within its
first 10 months of operation the 999 Group generated 157 million yuan in profits in industrial sectors
as diverse as pharmaceuticals, import-export, electronics, real estate, clothing, food, stock and secu-
rities, and more. By the mid 1990s this group had 34 enterprises and fixed assets of 1.6 billion
yuan.31 In 1993 the foreign trade of the Group was reported to be worth nearly $12 million, involv-
ing trade with Russia, Sudan, Singapore, Hong Kong, Egypt, and Qatar.32
China’s armed services departments—its army, navy, and air force—also operate their own
commercial ventures. The air force has been estimated to have around 430 enterprises and mines,
160 air bases, nearly 400 farms, and to run its own commercial airline (United Airlines). By 1992
this airline had established 39 domestic routes.33 The air force has eight airport construction teams
that undertake commercial contracts to build civilian and military airports. China’s navy uses its
shipping fleet to engage in commercial ventures. According to one account, units of the South China
Fleet are involved in more than 460 construction companies in the special economic zones in south-
ern China.34 The China Songhai Industrial and Commercial Corporation is the navy’s main business
conglomerate. China’s Second Artillery, the PLA’s main infantry force, has not been left behind in
the rush towards commercialization. Its Shanhaidan Enterprise Group runs a range of commercial
ventures, that have established a particularly strong position in the pharmaceutical industry.35
30
Xinxing is primarily involved in the sale of logistics equipment. For further discussion see Ellis
Joffe, “The PLA and the Economy: The Effects of Involvement,” in Chinese Economic Reform, ed.
Segal and Yang.
31
Zhongguo Xinwen She, 29 November 1992; FBIS, 9 December 1992, p. 33. Cited in Joffe, “The
PLA and the Economy,” p. 21.
32
Jiefangjun bao [Liberal Army Daily], 10 June 1993; FBIS, 21 September 1993, p. 43. Cited in
Joffe, “The PLA and the Economy,” p. 22.
33
Jiefangjun Bao, 4 October 1993; FBIS, 14 October 1993, p. 20. Cited in Joffe, “The PLA and the
Economy,” p. 21.
34
Beijing, Xinhua [Domestic Service], 9 April 1993; FBIS, 20 April 1993, p. 22. Cited in Joffe, “The
PLA and the Economy,” p. 21.
35
Cheung, “The Chinese Army’s New Marching Orders,” p. 188.
178
The military’s business holdings are further divided according to regional military com-
mands. China’s military regions and provincial military districts operate their own enterprises and
business interests. The Shandong Dongyue Corporation oversees the 600 army-owned enterprises
in the Jinan Military Region. The Sichuan Military District set up the Sichuan Bashu Enterprise
Group in 1992 to oversee its businesses.36 Differences in profitability between military business
groups are influenced by a range of factors, including their geographical location, resources, tech-
nology, infrastructure, etc. Many of the most successful military-affiliated corporations are in the
southern Guangdong area. Some of the special economic zones (SEZ) in the south were established
and are operated by regional military units, including the Shantou SEZ, which is operated by the
Guangzhou military region. Military enterprises that operate in more remote areas, such as in the
Third Front, are at a much greater disadvantage. The Southwest Great Wall Economic Development
General Corporation, which oversees the enterprises in the Chengdu military region, covering
remote areas in Tibet, Sichuan, and Yunnan, faces considerable disadvantages compared to those
in the Guangdong area.37
The decentralization of the military’s business structure extends to the lowest level of the
military hierarchy, with group armies and individual departments within higher level commands and
regional armies also operating their own business ventures. The businesses operated by these lower
level units tend to be smaller and less profitable than those operated by higher level units. In-land
military groups in general have fewer commercial opportunities than coastal groups. The Luyan
Enterprise Group, operated by a group army in Shaanxi province, is more typical of an in-land type
group, consisting of only ten enterprises, mostly factories and mines.38 There are, however, excep-
tions. The 42nd Group Army that is based in Huizhou, Guangdong province, operates the Chang-
cheng (Great Wall) Huihua Industrial Corporation, one of the largest and most successful business
36
Ibid., p. 190.
37
The Third Front refers to the military-industrial complex set up Mao in the 1960s as a means of
spreading China’s industrial capability over diverse geographical areas. These enterprises are geo-
graphically isolated and cut off from resources and adequate infrastructure. They have a difficult
time retaining skilled personnel and suffer from old technology, outdated equipment, and over-
staffing.
38
Jiefangjun Bao, 10 January 1992, p. 1. Cited in Cheung, “The Chinese Army’s New Marching
Orders,” p. 190.
179
conglomerates in southern China. Changcheng has over 90 enterprises that operate throughout the
southern coastal area, including in special economic zones such as Shenzhen and Shantou.39
The range of industries in China’s military-defense complex is evident from the above dis-
cussion. A look at aggregate figures confirms the important position of military-affiliated businesses
in a wide range of economic sectors. Military-defense production accounts for about 20 percent of
China’s annual passenger car production. There are nearly seventy automobile plants owned by the
military, with a total annual capacity of 50,000 vehicles and 500 million yuan worth of vehicle
parts.40 The PLA alone operates around 400 pharmaceutical factories, which were estimated in the
early 1990s to account for around 10 percent of China’s annual output in this sector.41 The PLA
continues to operate the farms and enterprises that traditionally were intended to make it more self-
sufficient in food, supplies, and spare parts. In 1993 the PLA’s military farms earned it extra revenue
of around 700 million yuan.
39
Jiefangjun Bao, 1 December 1988, 26 January 1992. Cited in Joffe, “The PLA and the Economy,”
p. 22.
40
Zhongguo Jingying Bao, 28 October 1994, p. 7. Cited in Cheung, “The Chinese Army’s New
Marching Orders,” p. 190.
41
Zhongguo Junzhuanmin Bao, 8 January 1991, p. 1 Cited in Cheung, “The Chinese Army’s New
Marching Orders,” p. 192.
42
Gurtov and Hwang, China’s Security, p. 155.
180
armed forces—not to produce high-technology equipment or engage in military R&D. PLA factories
have therefore had a history of producing low-technology goods, which support the army directly,
including quartermaster goods, textiles, food production, etc.
Most of the defense industries have rapidly redirected their production towards the commer-
cial market. The aerospace industry is a good example of this. The Ministry of Space Industry, con-
verted from the old 7th Ministry of Machine Building in 1982, has been quite successful in utilizing
its technology to shift into civilian/commercial production. In 1980 civilian products accounted for
16 percent of this industry’s output value, but this increased to 75 percent by 1988.43 The industry
operates 16 plane-manufacturing companies and has contracts with numerous foreign companies,
many in Southeast Asia, to manufacture aircraft parts. By 1990 the industry had a total workforce
of 800,000, and its sales abroad were U.S. $294 million. It operates three export-promoting corpora-
tions to handle its sales abroad. Despite its advantages, it still took this industry time to become
commercially successful. It greatest foreign currency earnings come from its sale of tactical ballistic
missiles to the Middle East.44 And the industry has faced considerable difficulties, such as numerous
disastrous satellite launches in the mid-1990s.
A second major defense industry that illustrates the defense industries’ rapid and relatively
successful shift to civilian and commercial production is the nuclear industry. This industry employs
about 300,000 workers in 200 enterprises. Like other defense industries it has been restructured a
number of times.45 Production of nuclear power at two nuclear power plants for the civilian market
43
SPC (State Planning Commission) and COSTIND, eds., Zhongguo guofang gongye junminjiehe
shizhounian zhuanji, Special Edition on the 10th Anniversary of Military Conversion in China’s
Defense Industries (Beijing: Ordnance Industry Publishing House, 1990), p. 13. Cited in Gurtov and
Hwang, China’s Security, p. 155.
44
John Lewis and Hua Di, “China’s Ballistic Missile Programs: Technologies, Strategies, Goals,”
International Security 17, no. 2 (Fall 1992), p. 34.
45
Originally the nuclear industry was part of the secret cluster of numbered machine-building indus-
tries responsible for the atomic bomb construction. In 1982 it was converted into the Ministry of
Nuclear Industry, which reflects the new importance placed on civilian nuclear power. Then in 1988
it was transformed into the China Nuclear Energy Industry Corporation and placed under the super-
vision of the Ministry of Energy Resources. Finally, in 1993 it was converted into the China
National Nuclear Corporation (CNNC). See John Lewis and Litai Xue, China Builds the Bomb
(Stanford: Stanford University Press, 1988), pp. 54-59. See also Gurtov and Hwang, China’s
Security, p. 156.
181
has been a key source of this industry’s profits.46 Its other commercial products include nuclear tech-
nology and radioisotope products, fire alarms, rare-earth products, nonferrous and precious metal
products, metallurgical-chemical products, and machine instruments. Between 1989 and 1996 the
proportion of the industry’s civilian production increased from 42 to 80 percent of its output value,
and in 1996 it reported annual growth of 20 percent.47 With the proportion of nuclear power in
China’s total electricity output increasing steadily, this industry can probably count on steady growth
for some time.48 There is a large potential international market for China’s nuclear industry, includ-
ing the building of nuclear power plants, not to mention the sale of weapons-making capability that
has been such a concern to the United States.
COSTIND Businesses
A third component of the military-defense business empire are the commercial ventures
directly affiliated with the institution that coordinates between the PLA and the defense industries—
COSTIND. This illustrates how even the organizations responsible for regulatory activities have also
“jumped into the sea.” While COSTIND is responsible for supervising and regulating the commer-
cial activities of the defense industries and for coordinating overall military R&D and production,
it has simultaneously set up its own business companies.49 Its larger business entities include the
Xinshidai Corporation, which is engaged in marketing and publications among other things; the
Xiaofeng Technology and Equipment Corporation, which is involved in computers and other high-
technology ventures; and the Yuanwang Group. It has up to a hundred other companies in addition.50
46
According to one report the production of civilian nuclear power was responsible for a fourfold
increase in the industry’s profits from 1984 to 1986. Gurtov and Hwang, China’s Security, p. 156.
47
Gurtov and Hwang, China’s Security, p. 156.
48
In 1995 nuclear power accounted for 1.2 percent of China’s electricity output, but the percentage
is expected to rise to 6 by 2020. “Nuclear Power Plays Important Electricity Role,” China Daily,
January 6, 1996. Cited in Gurtov and Hwang, “The New Military Economy,” p. 156.
49
One indication of the important status of this organization is the fact that its head is typically a
member of the Central Military Commission.
50
Frankenstein and Gill, “Current and Future Challenges,” p. 400-01.
182
According to official Chinese statistics, the defense sector’s share in the total production of
various goods include automobiles, 9 percent; motorcycles, 60 percent; and container trucks, 26 per-
cent.51 Upwards of 90 percent of the production in the electronics industry is estimated to be done
by defense-affiliated enterprises. Among the eight big automobile manufacturing entities in China,
three belong to the defense industrial system.52
Earnings
There is considerable variation in earnings among the various components of the military-
defense sector. Many military enterprises run in the red. According to a common estimate, one half
to two-thirds of the PLA’s industrial factories run a deficit. Military industrial enterprises have had
the worst record, and the industrial enterprises of the Third Front are among the worst performers.
It would appear that larger enterprises have performed better than smaller ones owing to advantages
in terms of bank loans, access to capital and technology through domestic and foreign sources, gov-
ernmental preferential treatment, employment of skilled personnel, etc. Three of the PLA’s corpora-
tions—the Poly Group, Shenzhen’s 999 Group, and the Xinxing corporation—are said to account
for about one third of the PLA’s total business earnings.53
Declared profits from PLA businesses have been around 10-15 percent of the official defense
budget. In 1992 PLA businesses declared 5 billion yuan in profits, out of a total defense budget of
37 billion yuan.54 Western estimates of military profits have tended to much higher, between U.S.
$5 to U.S. $20 billion.55 However, only rough estimates of the earnings of military-affiliated busi-
nesses can be made. Official figures almost certainly under-report profits, for several reasons. First,
51
Feng-cheng Fu and Chi-Keung Li, “An Economic Analysis,” in Mixed Motives, ed. Brömmel-
hörster and Frankenstein, p. 48.
52
Beijing Jeep, Chongqing Changan, and Guizhou Aviation. See Fu and Li, “An Economic Analy-
sis,” p. 48.
53
Cheung, “The Chinese Army’s New Marching Orders,” p. 194.
54
Loc. cit.
55
Richard A. Bitzinger, “China’s Defense Budget,” International Defense Review (February 1995):
36. Some have put the figure between U.S.$1-2 billion. See Cheung, “The Chinese Army’s New
Marching Orders,” p. 196.
183
there are a considerable number of PLA businesses that are illegal or whose legality is borderline,
as discussed further below. The PLA is heavily involved in the booming prostitution business,
contraband, manufacturing of pirated CDs and other products, smuggling of foreign products into
China, and more. These illegal activities could account for large-scale under-reporting of its profits
by the PLA.56 Second, like most business entities, the PLA is interested in minimizing the taxes its
subsidiaries have to pay, and may not therefore push for thorough accounting and auditing of its
enterprises. The underdeveloped financial system further contributes to the state’s difficulty of effec-
tively regulating China’s burgeoning businesses. Finally, top military commanders have had their
own difficulty in monitoring the commercial activities of their own subordinate entities. A self-audit
conducted in 1993 by the auditor general of the PLA discovered billions of yuan in unauthorized
spending by military units.57 Reportedly, a review of more than 130 receipts, bills, and invoices
brought to light 3.25 billion yuan in economic benefit to the PLA. One can guess that only a portion
of the hidden economic activities was discovered by the PLA auditors. This means that at least
billions of yuan are going unreported to the PLA by its subordinate units annually.
56
The military is perfectly positioned to engage in such activity because of its traditional operational
autonomy and the difficulty which any state police force would have in effectively supervising its
activity.
57
Beijing, Xinhua [Domestic Service], 18 August 1993; FBIS 20 August 1993, p. 21.
184
In the first category, a common source of new, profit-oriented enterprises in the military has
been restructured Third Front enterprises, which are often relocated from in-land to coastal areas.
Third Front industry enterprises are typically located in remote areas with difficult access to energy
sources, infrastructure, capital, human resources, etc. One solution adopted has been to restructure
and relocate resources from these in-land enterprises to the coastal areas, or to establish entirely new
enterprises on the coast with the military’s resources. This type of enterprise is referred to as a “win-
dow” for the third-line military sector, as it offers a connection to an area in which capital, resources,
infrastructure, personnel, etc. are more available. This has been a restructuring strategy adopted by
in-land provinces such as Shaanxi and Sichuan that have a high concentration of Third Front enter-
prises.58
The military also illustrates the second type of business form, which has resulted in what one
commentator refers to as “ministerial-cum corporations.”59 The aim of this transformation has been
to turn ministries, or portions of them, into semi-private business corporations responsible for their
profits and losses.60 While nominally independent, these new corporations still operate under the
supervision of a military superior, either COSTIND or a ministry. For instance, the former Ministry
of Aerospace was broken up into the China Aerospace Corporation (CASC) and the Aviation Indus-
tries of China (AVIC), both of which are supposed to operate as independent business entities under
the guidance of COSTIND. The old Ministry of Ordnance was converted into the China North In-
dustries Group (NORINCO), now one of China’s largest arms trading companies. These “ministry-
cum corporations” typically remain large and highly diversified, with a large number of subsidiaries.
AVIC includes more than 200 trading companies and enterprises, employs more than 500,000
58
For example, this became a key strategy in Shaanxi province. National level officials met in 1986
and decided that because more than 50 percent of the country’s third-line enterprises were located
in remote areas, that they would advocate the restructuring and relocation of some of these enter-
prises to the coastal areas. In many cases, entirely new, non-military managers were appointed to
these new enterprises. Gurtov and Hwang, China’s Security, p. 155.
59
Frankenstein, “China’s Defense Industry Conversion.”
60
China State Council Information Office, “China: Arms Control and Disarmament,” Beijing
Review, no. 48, p. 16.
185
workers, and has more than 30 affiliated research institutes and six universities and colleges.61 The
subsidiary enterprises of these corporations can themselves be quite large.
Through the 1980s the structure of the military-defense business empire fairly closely mir-
rored the military’s bureaucratic structure. However, new commercial organizations have prolifera-
ted that cut across the old structure, giving rise to new business forms. For instance, military con-
glomerates have been established that operate across a number of regional territories. The Nanfang
Industrial and Trading Corporation, which reported exports of $117 million in 1993, operates across
5 provinces.62 Like other public agencies, the military has also become a partial owner in a variety
of joint ventures with domestic and foreign entities.
61
Frankenstein and Gill, “Current and Future Challenges,” p. 399.
62
Cheung, “The Chinese Army’s New Marching Orders,” p. 189.
186
exist for the People’s Armed Police, although they are quite similar to those operating in the other
divisions of the PLA. Enterprises directly under the PAP headquarters are supposed to turn over 70
percent of their profits to the PAP logistics department. Enterprises operating under provincial level
PAP corps or colleges are expected to hand over 20 percent of profits to PAP headquarters, and 30
percent to the corps or college, and may retain the remaining 50 percent.63 The specific formula and
allocation of profit may vary from branch to branch within the military, and from enterprise to
enterprise, but a basic system exists in which profits are divided three ways among the enterprise
itself, its immediate supervising unit, and the higher level unit. Despite guidelines such as these, the
evidence suggests that in practice the distribution of profits of commercial ventures is subject to
negotiation by the parties involved and is affected by the degree of leverage that each exercises, a
pattern that is also evident in other public sectors.64
63
Zhongguo junshi caiwu shiyong daquan [Complete Practical Guide to Chinese Military Finance]
(Beijing: Jiefangjun chubanshe, 1993), pp. 32-33. See also Arthur S. Ding, “China’s Defense
Finance: Content, Process and Administration,” China Quarterly, no. 146 (June 1996): 428-42.
64
For the role of bargaining in the distribution of profits from high-technology companies operated
by universities, research institutes, and other educational institutions see Francis, “Bargained Prop-
erty Rights: China’s High Technology Spin-off Firms.”
187
for the purchase of modern weapons. Comparisons between the conditions of the Chinese and the
Russian military help to put this positive function in perspective.65
Military entrepreneurship has contributed to China’s economic reforms by encouraging
greater and more efficient use of military resources, including technology, capital, plant facilities,
human resources, etc. Military conversion, for example, has also ameliorated the problem of over-
capacity by encouraging the shift of underutilized plant facilities, resources, technology, etc. to civil-
ian and commercial uses. China’s military business activities also demonstrate how governmental
entrepreneurship may contribute to the level of competition in the economy in the shift away from
a centrally-planned economy. The fragmented structure of control and property rights within the
military’s business empire contributes to the level of competition. The enterprises supervised by
diverse organizations within the military’s institutional framework must compete with each other.
Competition may even exist within a single institutional group. While the air force operates its own
airline, a trading company set up by the air force also set up an airline, which could potentially come
into competition with the one operated by the air force itself!66
Military entrepreneurship illustrates how the generation of revenue from commercial activi-
ties can create a feed-back mechanism which in turn promotes modernization of that sector. The
transfer of technology and resources from public sectors (including the military, education, etc.) to
civilian/commercial uses has generated revenue that is then used to modernize those sectors. Mili-
tary conversion has been used to modernize military technology, to pay for additional military R&D,
to purchase foreign technology and weapons, and more. For instance, the revenue generated by the
military’s nuclear industry’s conversion to production of civilian nuclear power has helped to fund
an active research program in nuclear fusion and large reactors.67
65
See Jörn Brömmelhörster, “Semantic Differences: Comparing Defense Conversion in China and
Russia,” in Mixed Motives, ed. Brömmelhörster and Frankenstein.
66
The China Business Review (Sept.-Oct. 1989), p. 30. Cited in “The PLA and the Economy,” p. 21.
67
Xinhua (Beijing) domestic broadcast, January 10, 1996, in FBIS-CHI 96-016, January 24, 1996,
pp. 23-24. Cited in Gurtov and Hwang, China’s Security, p. 156.
188
Negative Consequences
The Chinese military also illustrates negative or ambiguous aspects of China’s approach to
economic reform. First, the direct engagement by public institutions in commercial activities can
create a conflict over their fundamental identity and integrity. They must simultaneously seek to
maximize their profits and respond to political directives from the state. This type of conflict was
reported by managers of China North Industries Corporation (NORINCO) plants, whose commercial
plans reportedly clashed with the State Council’s effort to impose certain military production tar-
gets.68 While it is difficult to assess the extent to which this may threaten national security, evidence
of the tension is clear. At a practical level, this tension is evident in the simple question of supplies.
As an organization committed to national security, military institutions would be expected to main-
tain adequate supplies on hand in the event of a security conflict. On the other hand, in the role of
profit-seekers they would not want to maintain excess inventory. Military entrepreneurship could
also affect the attitude of the military towards the outbreak of war if the military had business hold-
ings with the countries involved. Given that the military was involved in more than 300 joint ven-
tures with foreign companies in 1995, including some from countries with which China has potential
national security issues, such as Taiwan and Japan, the chance of that should be seen as consid-
erable.
Commercial activities can damage the military’s institutional professionalism and integrity
in other ways. It can create tension between treating its soldiers as a labor force versus treating them
as a professional force. Soldiers are hired as laborers in many of the military’s factories. This creates
a tension in the military’s interest in paying them low wages and benefits in order to maximize prof-
its, and its interest in maintaining their professional status. Using soldiers as workers risks the mo-
rale of the troops, many of whom may find themselves working in low-skilled, low paying positions
that have little to do with the role of the soldier. It also conflicts with the goal of professionalizing
the military that has been advocated by the state during the reform period.
There is evidence that the focus on profits has damaged professional standards within the
military. According to some Chinese press reports, military units sometimes cut short military train-
68
Frankenstein and Gill, “Current and Future Challenges,” p. 399.
189
ing with the excuse that they lack the funds to carry them out.69 In response to this problem in 1993
the government imposed a resolution that prohibited military enterprises from hiring active service
military personnel in their enterprises. Military entrepreneurship appears to have contributed to the
decline of ethical standards within the ranks. Promotions, awards, sought-after transfers, recruitment
of new party members, etc. are reported to be increasingly contingent on bribing the officers in
charge. To get the necessary funds to bribe his officers for a desired promotion one soldier in Beijing
was reported to have become a thief.70
Commercialism appears to have weakened the institutional coherence of the military through
an erosion of financial control. Commercial activities have exacerbated the decentralization of
control over revenue and expenditures. Finance departments within the military were weak to begin
with, lacking control over a number of important expenditures, including equipment (which accounts
for one third of the military’s published expenditures), scientific research, and capital construction.71
The reform process has further weakened their authority. Extra-budgetary revenue generated through
the military’s commercial activities are typically not turned over to finance departments.72 Finance
departments have not been given managerial authority over profits from production operations or
other forms of extra-budgetary revenue. As a consequence, the increased revenue from these sources
has further decentralized financial management within the military.
Anti-market Forces
The military’s commercial activities illustrates the anti-market potential of government entre-
preneurship. The military’s monopoly over certain resources can inhibit competition in certain
69
Jiefangjun Bao, 16 March, 1990. Cited in Cheung, “The Chinese Army’s New Marching Orders,”
p. 197.
70
Li Xusheng, Li Quanmo and Tao Ke, “Xin shiqi gongfu de meili” [“The charm of being a public
servant in the new era] , Jiefangjun bao [Liberation Army Daily], 14 December 1992. Cited in David
Goodman, “Corruption in the PLA,” in Chinese Economic Reform, ed. Segal and Yang, p. 44.
71
Ding, “China’s Defense Finance,” p. 439.
72
Arthur Ding argues that the fact that finance departments are part of the military’s logistics system
is an additional disincentive against military-defense units turning their earnings over to them, since
they are perceived as having their own distinct interests to protect. See Ding, “China’s Defense
Finance,” p. 440.
190
sectors, even as there is overheated competition in other sectors. For instance, in some localities the
military exercises a near monopoly over commercial transportation and shipping because of its large
fleet of vehicles, aircraft, and ships, and its vast transportation network of railways, harbors, airports,
and roads that are often for its exclusive use. This is evident in the coal industry, where in many
remote areas the military exercises a monopoly over transportation, in addition to owning numerous
coal mines.73 While this may be good for the military’s business, it is deleterious for the market
system.
Military commercialism illustrates the persistence of soft-budget constraints towards govern-
ment-affiliated enterprises. From one-half to two-thirds of military enterprises are estimated to be
operating in the red.74 The military, like other state sectors, has resources to keep failing enterprises
going, and they are under political pressure to do so in order to minimize unemployment and main-
tain social order. Commercial activities, ironically, may to an important extent facilitate the subsi-
dizing of failing enterprises. Continued soft-budget constraints may undercut the role of competition
in weeding out poorly performing enterprises. Furthermore, military enterprises have access to a
wide range of subsidized goods and commodities and some of them enjoy special tax and customs
breaks, as well as privileges in importing and exporting.75
These factors can lead to irrational market fragmentation, with an excess number of firms
competing in a limited market. When a new product or industry is “hot,” the military, like other gov-
ernment agencies, rushes to enter this sector, creating a glut. However, the relatively soft budget
constraints characteristic of government-operated companies weakens the role of the market mecha-
nism in weeding out less efficient companies. The result is that a glut of companies in a particular
industry does not always get weeded down through market competition. This occurred in the real
estate industry, electronics, home appliances (refrigerators, televisions), etc. The result is often a
failure to consolidate and restructure in the style of Western markets. Rather, the Chinese military
73
Min Chen, “Market Competition and the Management Systems of PLA Companies,” in Mixed
Motives, ed. Brömmelhörster and Frankenstein, p. 212.
74
Cheung, “The Chinese Army’s New Marching Orders,” p. 194.
75
Ibid., p. 188.
191
mirrors the continuing tendency for government agencies to subsidize their loss-making enterprises
from their profitable ones.76
Corruption
The military’s commercial activities illustrates the potential corruption associated with gov-
ernmental entrepreneurship. Because of its status and its role in national security the military has
probably been one of the worst perpetrators of corruption in China. It has been close to immune from
inspections and investigation. For example, the simple fact that the cargo being hauled in its military
transport vehicles does not get inspected the same way as cargo in civilian commercial transport
vehicles opens up a whole arena of potential corruption. Furthermore, military corruption cases have
been treated as classified. For the most part the military has regulated corruption within its ranks,
and as of January 1, 1993 cases of corruption within the military were put under the sole jurisdiction
of military tribunals and subject to military secrecy.77 Economic pressures are also a source of
corruption within the military. Declining living standards of PLA soldiers have made the military
an unattractive career—activities that generate extra revenue to counterbalance this trend have thus
become an institutional necessity.
Corruption within the military reflects the various forms of corruption flourishing in China,
including bribery, smuggling, fraud, and illicit use of public resources, personnel, and equipment
for personal or collective gain. Military agencies, like other state agencies, accept kickbacks and
commissions in exchange for contracts, sell state-subsidized goods in the private market at below-
market prices, and engage in other forms of bribery. Embezzlement, the outright theft and sale of
state property, and the illegal or illicit use of public resources are particularly severe in the military
because of its vast resources. Within a three month period between February and April 1991 a
political instructor of a unit of the PLA Navy was reported to have stolen more than four tons of fuel
76
Richard J. Latham, “A Business Perspective,” in Mixed Motives, ed. Brömmelhörster and
Frankenstein.
77
“Junshi fayuan shenli jingji jiufen anjian de ruogan wenti” [“Several questions on handling cases
of economic disputes by military tribunals”] in Jiefangjun bao [Liberation Army Daily], 4 January
1993; and Geng Renwen, “Junnei jingji shenpan yin zhunxun weihu junshi liyi yuanshe” [“The trial
of economic cases in the army should follow the principle of protecting military interests”] in
Jiefangjun bao [Liberation Army Daily], 31 May 1993. Cited in Goodman, “Corruption in the PLA,”
p. 39.
192
from a military oil terminal, which he sold for 4,400 yuan.78 The PLA’s Liberation Daily reported
that the theft of weapons and military equipment had become a common practice among soldiers.79
The extent and types of corruption are evident in the regulations aimed at prohibiting these activities.
The “ten nos” issued in 1989 sought to prohibit the PLA from activities such as setting up businesses
without permission; buying goods illegally, fixing prices, and profiteering; producing and selling
fake goods; using military equipment or vehicles to engage in smuggling, speculation, etc.; leasing
or selling of military equipment, vehicles, bank accounts, blank invoices; using officers and men on
active duty to run enterprises and to engage in trade; and exploiting the positions of military men
for business purposes.80
Smuggling has become a specialty of the military because of its access to transportation vehi-
cles such as helicopters, ships, and gunboats, and its ability to transport goods with little risk of in-
spection. PRC Customs anti-smuggling teams need special approval from Beijing in order to inves-
tigate a military unit.81 One of the best known smuggling cases was the Hainan province case of
1985, in which the Hainan local government used its special status to buy foreign luxury cars that
were then resold to buyers on the mainland. Both the PLA Navy South China Sea Fleet and the army
cooperated with the local government to smuggle the cars into China using their helicopters and
gunboats. In another big case, 38 gun-boats, torpedo boats, escort vehicles, and submarine chasers
that were being used in smuggling were captured in Guangdong in 1992. In this case the military
defended itself by claiming that these vehicles had already been retired and that the military was thus
not involved.82 Because military personnel and units enjoy special privileges, there has been an
78
Zheng Jingjie, “Zhiquan shengliao tan moushi de jieti” [“Power becomes a tool for him to seek
personal gain”] in Jiefangjun bao [Liberation Army Daily], 14 December 1993. Cited in Goodman,
“Corruption in the PLA,” p. 44.
79
Gao Laifu and Liu Xiangiang, “Jiaqiang dui junren weifa fanzui shehui youyin de fangfan”
[“Strengthen guard against temptation for soldiers to commit crimes”] in Jiefangjun bao [Liberation
Army Daily]. 17 May 1993. Cited in Goodman, “Corruption in the PLA,” p. 44.
80
Hong Kong Zhongguo Tongxun She, 25 April 1989; FBIS, 2 May 1989, p. 116. Also cited in Joffe,
“The PLA and the Economy,” p. 23.
81
Goodman, “Corruption in the PLA,” p. 46.
82
Yan Changjiang, Guangdong Da Lieban [The disintegration of Guangdong] (Jinan University
Press, 1993), p. 242. Cited in Goodman, “Corruption in the PLA,” p. 46.
193
explosion of fraudulent use of military official seals, registration plates, and military registration
forms that enable non-military individuals and organizations to pass themselves off as belonging to
the military.
Conclusion
The Chinese military’s entrepreneurship reflects key features of China’s evolving national
system of innovation—the emergence of state entities and public institutions as dominant market
actors, the fragmentation of ownership within this quasi-public sector, increased economic dyna-
mism coupled with heightened corruption, and more. On the positive side, military entrepreneurship
has contributed, along with other forms of public entrepreneurship, to the overall dynamism of
China’s market economy. It has helped to shift public resources to more efficient and profit-oriented
enterprises. It has enhanced the use of military technology for civilian production. And the revenue
from military businesses has been critical to sustaining the military institutionally, helping to pay
for the purchase of advanced weapons as well as subsidizing institutional maintenance expenditures.
On the other hand, military entrepreneurship illustrates the potential pitfalls of public
entrepreneurship. It has helped to extend the practice of soft-budget constraints to new, nominally
market-oriented firms. It has allowed the emergence of sectoral and regional monopolies and has
in certain ways undercut the role of competition in weeding out inefficient enterprises. Finally, the
military’s commercial activities show how public entrepreneurship can erode the integrity and
cohesion of public institutions and potentially undermine their professionalism.
The Changing Role of the Defense Industry in Israel’s Industrial
and Technological Development
Introduction
Since the end of the cold war, national defense budgets have shrunk drastically while produc-
tion capacity has not changed much. Consequently, export markets have become more competitive,
and national defense industries, which until several years ago were the dominant suppliers in their
home markets, are no longer fully protected from competing imports. Moreover, despite the in-
creased competition and the use of cheaper off-the-shelf commercial components instead of specially
designed military ones, the prices of new weapons and defense systems seem to be rising inexorably.
Economic necessity is beginning to wear away the defense industry’s segregation, forcing compa-
nies and governments to cooperate as well as to compete across borders. The outcome of this consol-
idation has been the emergence of a small group of defense giants in the United States and Europe.
Size, it seems, is a crucial factor in the defense industry (Economist 1997).
Though the Israeli defense industry, like defense industries elsewhere, suffers from excess
capacity and, hence, inefficient operation, myopic government policies and lack of long-run plan-
ning, have prevented it from following the world-wide trend of consolidation and mergers across
borders. Clearly, this isolationism cannot be sustained much longer. The Israeli defense firms, some
of which are world leaders in various defense products, must change, or risk default.
In this chapter we describe the Israeli defense industry, emphasizing its role in Israel’s indus-
trial development. In particular, we show how the effect of the defense industry on the economy
depends primarily on the following factors: (1) the defense needs of the country; (2) the overall eco-
nomic situation of the country and the size of its defense industry relative to the civilian industry;
and (3) the stage of technological development of the country.
*
We are grateful to M. Balch, I. Ben Israel, C. Serfati, J. Reppy and the participants in the Work-
shop on The Place of the Defense Industry in National Systems of Innovation, October 16-18, 1998,
Cornell University, for valuable comments and suggestions.
194
195
Our analysis shows that judging the defense industry on economic performance alone is mis-
leading, since it does not account for the threat from enemies and disregards the defense vision of
the nation, thus ignoring Israel’s need to conduct defense research and development (R&D) for its
own exclusive use.
The limited available data suggest that the Israeli defense industry was instrumental in trans-
forming the country’s civilian industry into a successful high-tech sector. The entrepreneurial spirit,
the problem-solving approach, and the system-oriented approach, which are characteristic of most
of the successful high-tech firms in Israel (Kaplan 1998), originated in Israel’s military and the
defense industry. Moreover, the defense sector is still a very important source of new technological
know-how and experienced human resources for the civilian high-tech industry.
By the mid-1960s, the defense industry workforce had grown to about 15,000 from some
5000 during the 1950s. The rapid growth of the Israeli economy that followed the 1967 Six Day War
prepared the ground for the subsequent continuous growth of the defense industry over the next two
decades. The number of employees in the defense industry tripled between 1967 and 1975, and
increased by a further 50 percent between 1975 and 1985. The rapid growth of the 1967-1975 period
was due mainly to the rapid growth in internal demand (especially after the 1973 Yom Kippur War).
The growth in 1976-1985, by contrast, was based mainly on a tenfold increase in defense exports
during this period (Lifshitz 1999; Tishler and Rotem 1995).
The slow-down in the growth of the defense industry, which started in the mid-1980s, turned
into a severe crisis at the beginning of the 1990s, following the termination of the cold war and the
signing of peace treaties between Israel and several of its Arab neighbors.
The path of technological development along which the Israeli defense industry has pro-
gressed is similar to that of other developing countries. At the beginning, before the establishment
of the state of Israel and until the mid-1950s, the young defense industry concentrated primarily on
the production of light arms and ammunition and the reconstruction of surplus equipment. The sec-
ond period, after the 1956 Sinai Campaign, was characterized by production under license, especial-
ly from French firms. A notable example is the production of the Fuga Magister, a light jet training
aircraft, which was produced under license from a French aircraft manufacturer. In the third phase,
the industry started to modify and improve weapon systems produced under license or purchased
from other countries. During this phase, for example, the Fuga Magister aircraft was converted into
a fighter plane by adding guns and rocket launchers. The expertise gained during these improvement
programs was used later on to produce new platforms such as the “Eagle,” an Israeli version of the
Mirage 5. A new era opened in the late 1960s and early 1970s when the local industry was called
on to develop entirely new weapon systems. Since then, the Israeli defense industry has developed
a fighter plane (the “Lavi”), unmanned airborne vehicles (UAVs), main battle tanks such as the
Merkava, missile boats, various types of air-to-air, air-to-ground and ground-to-ground missiles, and
even communication and intelligence spacecraft.
Currently, Israel’s defense industry consists of about 150 firms. The ten largest firms account
for 78 percent of the defense industry workers, 82 percent of its total sales, and 87 percent of its total
197
exports.1 More than 75 percent of the sales of the defense industry are exported. Defense products
and systems account for 32 percent of Israeli industrial exports.
Israeli defense firms can be divided into three groups. The first group includes the three large
government defense organizations, IAI, TAAS and Rafael, which mainly develop and produce
defense systems. The second group consists of privately-owned large and medium size firms. Three
of the firms in this group—ELOP, Elbit Systems and Elisra—concentrate almost entirely on defense
products. The other firms in this group, ECI and Tadiran, produce mainly civilian products (commu-
nication equipment), but have defense systems divisions. The third group consists of relatively small
privately-owned firms, each producing a narrow line of defense products. For example, BVR devel-
ops computerized aircraft simulators, Astronautics manufactures command and control systems,
International Technologies produces laser designators, and Rokar develops navigation equipment
(MOS 1996/7). Beside the three groups mentioned above there are several large refurbishment and
maintenance centers that are part of the army’s Division of Technology and Logistics. These centers
maintain armored vehicles, aircraft, communication equipment and other support devices used by
the military forces. One large refurbishment center is devoted to the Merkava battle tank.
The largest defense firms not only rate among the largest industrial firms in Israel; they are
also included among the 100 largest defense firms in the world. Only eight out of the 100 largest
defense firms in the world are from developing countries, and five of them are Israeli firms.
The defense industry in Israel incorporates a vast array of technology, from computers and
electronics to electro-optics, aeronautics, mechanical design and metal works, chemical engineering,
software engineering, and many other areas. Table 1 provides information on the total sales, the
number of employees and the areas of expertise of Israel’s six largest defense industry firms. The
areas of expertise of the smaller defense firms are similar to those of the larger firms. However,
some of these smaller firms specialize in specific areas. For example:
Ordan: heavy metal casting, including tank armor;
Soltam: mortars and cannon barrels;
Beit Shemesh Engines: refurbishment and fabrication (under license) of jet engines;
Elisra (a subsidiary of Tadiran): naval, airborne and ground EW systems;
1
See Lifshitz 1999, for a detailed description and analysis of Israeli defense firms by size, ownership
and expenditure on R&D.
198
Table 1
Largest Defense Firms in Israel, 1995
The Israeli government is heavily involved in the defense industry, being the owner of some
of the main defense organizations and, at the same time, acting as the industry’s main customer
(through the military). The government also controls defense export via a special division in the
MOD called “Sibat,” which is in charge of authorizing export of classified products.
199
By virtue of their size, the management and workers unions of government-owned defense
firms have accumulated considerable “political” power, which is sometimes used to promote “pri-
vate” interests. Nevertheless, there is a consensus on the importance of the local defense industrial
capability and wide public support for its needs.
The relations between the military and the defense firms are very close. The small size of
Israel and its economy, the common background of military service of almost all citizens, and the
small number of engineering schools have created the basis for open communication between the
military professional staff and the industry. Over the years, these close relations have enabled the
shortening of development time, the cutting of development costs, and the development of some
unique weapon systems suitable to the conditions in the Middle East and to the special needs of the
IDF.
For the last ten years, in addition to its excess capacity problems, intense competition in the
worldwide market for defense products, and the chronic shortage of experienced and skilled man-
power in electronics and computers, the Israeli defense industry has been suffering from several
other specific problems. First, the Israeli defense firms are small in comparison to the large Ameri-
can and European defense firms (Economist 1997; Lifshitz 1999). Currently, large firms have con-
siderable advantages in developing and marketing the platforms, large systems, and expensive com-
ponents that are also produced by Israeli firms. Second, Israel’s target markets are geographically
far away. Third, several Israeli firms with similar technologies are competing fiercely among them-
selves for the same international markets. Fourth, Israeli labor unions in the government defense
firms are very influential. They intervene in the daily operations and managerial decisions of their
firms, thus preventing the necessary adjustments to the ever-changing, shrinking, and more com-
petitive world market for defense products.
In the following sections we analyze the economic and structural issues related to the above-
mentioned problems, laying the basis to the changes that, it is to be hoped, will result in better use
of the potential of the Israeli defense industry and improve its performance and competitiveness.
1992). However, except for the large government-controlled defense organizations such as Rafael,
IAI, and TAAS, data on the production and sales of the defense-related products and services of
most of the firms on these lists are unavailable. Moreover, many Israeli firms develop, produce, and
sell the same final goods, services, semi-finished goods, and materials to both the civilian and the
defense sectors (for example, silicon wafers, voice recognition systems, satellite equipment, laser
devices, etc). Thus, it is not a simple matter to classify these products as civilian or defense, particu-
larly when sales data are not classified according to customers (see Lifshitz 1999, for a literature
review on this issue). For the purpose of this work, firms are included in the defense industry if they
produce platforms, finished products and/or systems for the IDF or other armies. Thus, firms that
produce components for defense and civilian use are not considered here.2
The effect of the defense industry on an economy depends primarily on the following factors:
(1) the defense needs of the country; (2) the overall economic situation and the size of the defense
industry relative to the civilian industry; and (3) the stage of technological development of the
country.
2
See the somewhat different definition proposed by Kenneth Flamm in his chapter in this publica-
tion.
201
A somewhat similar, though less extreme, approach is adopted by countries that feel they
need to maintain a well-developed army in order to intervene when their allies are threatened (the
United States and, possibly, France, Britain, and Russia).
The Overall Economic Situation and the Size of the Country’s Defense Industry Relative to its
Civilian Industry
The local purchases of Israel’s Ministry of Defense were 8 percent (5 percent) of GDP in
1985 (1990). Thus, the defense sector constitutes a sizeable proportion of the country’s industry,3
and may have macroeconomic repercussions on the economy as a whole. Clearly, the government’s
policy toward the defense industry depends on its defense expenditures. Generally, the local demand
for defense products depends on the country’s perception of the threat from its enemies in both the
short and the long run, though the economic conditions of the country serve as a constraint on this
demand. Since the defense sector normally plans for the long run, we expect that the activities of
the defense firms will be somewhat counter-cyclical; that is, government purchases of locally pro-
duced defense products will have some, but not necessarily a strong, correlation with the country’s
long-run economic performance (GDP or government expenditure). Hence, we expect that the de-
fense industry will be perceived as a larger burden on the country during slow-downs or recessions
(as was the situation in Israel during 1996-8). On the other hand, expanding demand for locally
produced defense products can be very helpful during the process of getting out of a recession (as
was the situation in Israel during 1967-1968).
3
See Lifshitz (1999) for comparison to other countries.
202
developed (as has been the case since the mid-eighties), the defense firms may lose their role as
technology leaders, except in a number of very specific defense-related applications.
Table 2
Technology Transfer from Defense Industries to Commercial Use
Many of the high-tech firms in Israel are headed or owned by entrepreneurs who started their
career as officers or professionals in the defense sector in Israel. Table 3 lists several examples. In
addition, most of the military engineers, computer software personnel, and other professionals from
the intelligence, air force, and communications core are employed in the civilian industry after they
retire from the military, mostly by high-tech firms. Their defense sector background, particularly
their entrepreneurial approach and experience in project management, is highly valued.
Finally, Israel’s overall expenditure on defense R&D in 1997 has been estimated at more
than US$800 million, which is about 0.9 percent of GDP (for comparison, civilian R&D was about
2.3 percent of GDP and outlays for education amounted to 9 percent of GDP). A substantial portion
of the defense R&D is spent within universities and civilian research institutes (some of these out-
lays are for maintaining “knowledge centers”), thus contributing directly to both the defense and the
civilian sectors.4
4
See Teubal (1993) on the role of Israeli universities in defense and overall R&D in Israel.
204
Table 3
Firms Managed or Initiated by Personnel Previously Employed
in the Defense Industry or Military Organizations
5
See, for example, Levi-Faur (1998) on the liberalization of the Israeli telecommunication industry.
6
See, however, Williamson (1998) on the limitations of current economic theory and practice in the
analysis of the performance of existing institutions.
205
It seems that the markets for new platforms or very large integrated systems (such as an
advanced fighter plane, battleship, integrated anti-aircraft radar system, or ballistic missiles) are best
organized as a natural monopoly. This argument is justified on the grounds that the fixed cost of
production of such platforms or systems (which includes the relevant R&D, the set up of the pro-
duction line, etc) is very high relative to the (also high) marginal cost of production (see Flamm, this
publication, on the case of fighter aircraft). The quantity demanded of these products is relatively
low for two reasons. First, their maintenance and operating costs are high. Second, their purchase
price is high because their marginal production cost is high and the producers are, generally,
monopolies that charge a price that is higher than the marginal production cost (see, however, Econ-
omist 1998b). Flamm shows that “an industry that designs and produces only small numbers of ad-
vanced aircraft is going to yield a product that is virtually unaffordable, and at a substantial disad-
vantage when exported.” Consequently, it seems reasonable that in the future only large interna-
tional or American firms will have the vast amounts of capital need to develop and produce these
expensive platforms and very large integrated systems (see Flamm, James, Lovering, Markusen,
Reppy, Serfati, all this volume; Lifshitz 1999; Economist 1997, 1998b). Since the cancellation of
the Lavi (the Israeli fighter plane) project in 1987, Israel has realized that it should not attempt to
develop and produce such platforms or very large integrated systems alone (Lifshitz 1999; Teubal
1993).
Two questions remain: (1) what should be developed and produced by Israeli defense firms,
and (2) how should the defense sector in Israel be organized?
The answers to these questions depend, among other factors, on the threats facing the coun-
try, and the resulting Israeli defense vision. It is important to note that the process of adjusting the
Israeli defense industry to its desired structure (one that is consistent with the answers to the two
questions posed here) depends on the industry’s current structure.
The “defense vision” is a set of rules according to which the country plans its long-run
defense policy and actions. It can be derived by maximizing the country’s long-run security, subject
to various constraints. The constraining factors are: the long-run economic resources that are ex-
pected to be available to the country, the political situation in the region and elsewhere, the quality
and quantity of its human resources during the relevant time horizon, the evolution of the country’s
social fabric over time, the available and expected technology, and other variables that may influ-
ence the security of the country. For obvious reasons, the official Israeli defense vision is not
206
available to the public. However, its principles can be inferred from the studies of Ben Israel (1998),
Halperin (1988), Lifshitz (1995, 1999), and the Peled Committee (1999). Briefly, these principles
are based on the following assumptions:
C Israel cannot afford a long war or many casualties. That is, it should defeat its enemies
swiftly and fully in an all-out war.
C Israel’s economic resources are relatively small (Israel’s GDP in 1998, about US$100 mil-
lion, is somewhat larger than the GDP of its four immediate neighbors combined. It is, how-
ever, much smaller than the GDP of all the countries in the Middle East that are, potentially,
Israel’s opponents).
C The real prices of new platforms and modern defense systems increase at an annual rate that
is much higher than the long-run average rate of growth of the Israeli economy (Lifshitz
1999).
These three assumptions and the state of development of the military forces of Israel’s potential
enemies imply the following three conclusions.
C In the long run Israel will not be able to sustain a sufficiently large arsenal of modern plat-
forms (fighter planes, tanks, guns, armored vehicles, ships, etc.) to enable a swift and full
defeat of its potential enemies, who are equipped with similar platforms.
C Israel’s advantage in the long run is in its technological know-how and the quality of its
human resources. Modern technologies in the hands of motivated, educated, enterprising,
and well-trained soldiers can be transformed into “power multipliers.” Therefore, Israel must
base its main weapon systems on state-of-the-art technologies and ensure that some of these
weapons and systems are unknown to its potential enemies (Ben Israel 1998; Lifshitz 1999).
Moreover, it is instrumental for Israel to excel in Information Warfare (see Ben Israel 1998;
U.S. Army 1997, Toffler and Toffler 1995). Provided the know-how and skilled human
resources are available, this may well be the least costly policy to execute and the most
effective in both full-scale and low-intensity war. Combining the element of surprise with
high-tech is probably sufficient, even with a relatively small army, for Israel to inflict a swift
and full defeat on its enemies.
C Due to the highly destructive nature of the weapon systems that are available to its potential
enemies, its small size, and its economic inability to sustain a large regular army, Israel has
to detect and predict any forthcoming war or major attack in an accurate and timely manner
207
(unlike the experience of the Yom Kippur war in 1973). Hence, the continuous development
of modern and highly sophisticated intelligence system—an area in which the Israeli defense
industry is very active—is required. To avoid surprises, at least some of these systems
should be unknown to Israel’s potential enemies; they must therefore be developed and
produced solely for the IDF.
The necessity to develop and maintain weapon systems that are unknown to its potential
enemies (to be denoted Israeli Specific Weapon Systems, ISWS) plays a key role in the design and
operation of Israel’s defense industry. Clearly, because of the need for absolute secrecy in the devel-
opment and production of ISWS, competitive markets for these types of products cannot exist. Obvi-
ously, the R&D cost of these products and possibly but not necessarily their production and main-
tenance must be borne and controlled by the government. Private Israeli firms could produce ISWS,
but, since these products cannot be sold elsewhere, the government has to shoulder most of the risk
and cost of their production (part of the risk may be borne by firms bidding for their production).
addition to those that can be purchased in the world market. In this case, public policy toward
defense R&D should depend, among other things, on whether, and to what degree, these “country
specific” weapon systems are shared with other countries (through exports and/or joint ventures).
Hence, it is helpful to classify defense R&D into three main categories, depending on how much
spillover of knowledge (technology) is allowed: (i) R&D that is country specific, (ii) R&D aimed
at developing technologies that will be shared with a small number of allies, and (iii) R&D aimed
at developing technologies that will be available to almost all interested buyers (via exports or joint
ventures).
In the extreme case of R&D for country-specific technologies when no R&D spillover is
allowed, all of the defense R&D risk and cost should be borne by the government, and defense R&D
expenditure on these technologies should be determined almost exclusively by the country’s defense
needs and resources (summarized in the country’s defense vision). At the other extreme, category
(iii), the optimal public policy toward R&D—the type and amount of subsidies the government
should allocate to defense R&D and the level of government regulation of the monopolistic competi-
tion in this market—should be determined according to economic models such as those in Romer
(1990), Leahy and Neary (1997) and Segerstrom (1998). R&D in category (ii) is a hybrid of the
other two categories, and should therefore be determined on a case by case basis. The R&D of the
Arrow (anti-ballistic missile) project, for example, is a joint venture between Israel and the United
States. Its technology will not be available, for sometime at least, to other countries. Hence, it should
be viewed as category (ii) R&D. The Eurofighter, which is being developed as a joint venture by
several European countries, will be available for sale to many but probably not all countries in the
world. The Eurofighter R&D can be thought of as a category (iii) R&D, but with relatively smaller
than average spillover.
Following the analysis in Section 4 and bearing in mind that Israel lives under a constant
threat to its survival, we suggest analyzing the country’s R&D according to the three categories
described above. Israel’s R&D for ISWS is country-specific, category (i), and should be wholly
financed by the Israeli government. R&D for categories (ii) and (iii) should be determined according
to economic models similar to those in Romer (1990), Leahy and Neary (1997), and others.
209
7
For a survey and analysis of the existing, feasible, and desired interactions among the government,
the Department of Defense, and defense suppliers in the United States, see Lichtenberg (1988),
Kovacic and Smallwood (1994), and Rogerson (1990, 1994).
210
traditional modern weapon systems, and their purchase is normally a simple procedure (Tishler and
Rotem 1995; Economist 1997; Flamm, this publication). Moreover, it is likely that a viable private
Israeli defense industry, in the form of at least two major private firms, will continue to operate suc-
cessfully in Israel and abroad, serving as a shock absorber when required. Thus, we conclude that
the Israeli government should gradually transfer the ownership of all government defense organiza-
tions, other than those specializing in R&D or production of sensitive warfare materials, to the pri-
vate sector.
The third issue—the globalization of the world market for defense, as well as civilian, plat-
forms and complex integrated systems (Economist 1997, 1998a, 1998b; Lifshitz 1999)—poses a
severe constraint on the Israeli defense sector. The recent wave of mergers and buy-outs in the world
defense industry has decreased the number of producers of specific high-tech products, transforming
their R&D and production into monopolistic markets. Besides the power that monopolistic producers
exert on their customers, the added pressure of governments supporting their own defense industry
is also becoming apparent. The emergence of large U.S. firms and European conglomerates is evi-
dence of these political and economic pressures (Melman and Raviv 1994; Economist 1997, 1998a,
1998b; Louscher et al. 1998). The complications of marketing in an environment that is governed
by large monopolies and governments suggest that in the future only large organizations will be able
to compete in the markets for expensive modern defense products, particularly integrated systems.
Smaller firms, however, will be able to survive, and even prosper, in the components market or very
specialized products.
The wave of mergers and consolidation of defense firms in the United States began with the
rapid acquisition of second tier contractors by Loral, which emphasized financial rather than engi-
neering priorities, and was accelerated by the Pentagon’s policies (and financial incentives) follow-
ing Perry’s “last supper” in 1993 (Markusen, this publication). The American consolidation trig-
gered a flurry of similar though initially unsuccessful efforts in Europe, whose defense firms per-
ceived themselves as too small to compete against American giants such as Lockheed-Martin,
Boeing-McDonnell Douglas, Raytheon and others.8 It is interesting to note that these mergers and
8
James (this volume); for another view see Louscher et al. (1998).
211
joint ventures may not be justifiable by economic theory.9 They also contradict conventional
political wisdom insofar as they lead to the proliferation of high-tech weapon systems to marginal
customers in problem regions where sales are likely to feed existing tensions and local disputes
perceived as harmful to the West (Flamm, this volume). Nevertheless, the consolidation efforts in
Europe and the cross-Atlantic mergers and joint ventures between American and European defense
firms are continuing and will, most likely, be successful in the near future.
If Israel chooses to ignore the negative and unfortunate consolidation trend and reject merg-
ers and/or joint ventures of Israeli defense firms with foreign defense firms, opting unilaterally to
continue to support a few, relatively small, local defense firms, it will find itself increasingly iso-
lated. This isolation may have further adverse effects on Israel if the so-called “Inner Circle”
approach is adopted by the United States (Flamm, this volume). If successful in jointly producing
platforms and high-tech weapon systems and jointly controlling their diffusion around the globe, the
“Inner Circle” of a handful of producers, such as the United States, France, Britain, and Germany,
will be able to prevent the proliferation of high-tech weapon systems in “problem regions and coun-
tries.” This may, however, be detrimental to Israel’s security if it leaves Israel outside of the main
circle.
9
See Leahy and Neary (1997) on the negative welfare effects of lax competition, and Kovacic and
Smallwood (1994) on the incentives and positive aspects in forming these mergers.
212
effort should be considerably larger than it is at present (including the R&D that takes place within
the IDF). Moving from the current inefficient defense industry structure to a more efficient and
mostly private one should free substantial resources for investment in R&D.10
The government-controlled defense organizations, other than those in R&D, should be
merged into two privately-owned firms.11 One firm, for example, could be based on the electronic,
aerospace, and missiles divisions of Rafael, IAI, and TAAS. A second firm could include the
remainder of TAAS, the IDF’s maintenance and refurbishment bodies, the unit producing the Israeli
tank (the Merkava), and several other smaller units of the IAI and Rafael. The strength of the labor
unions of these organizations should be taken into account during the consolidation and privatization
process (Lifshitz 1999; Peled Committee 1999), and the workers should be consulted and possibly
offered partial ownership in the newly formed firms. However, the process itself must be carried out
quickly and be based on a massive reduction in the number of workers and restructuring of wages.
The financial structure of the newly formed firms should be changed (particularly, their own capital
should be expanded) to allow them to operate independently in the local and international markets.
Clearly, the two firms should be, unconditionally, private. Once the restructuring and privatization
processes have been completed, the two firms should be free to form international alliances. The
Israeli government should let the newly formed firms and the already existing private defense firms
decide whether, with whom, and how to merge or form joint ventures.12
The Israeli government should encourage mergers of private Israeli defense firms with other
Israeli or foreign firms. It can do this by suitably adjusting the structure and size of its tenders for
future production to the local market and the contracts that the national laboratories sign to test and
later produce newly developed products.13
Finally, forming national R&D laboratories is not enough. Israel should develop an R&D
policy that also supports the private Israeli defense firms (Romer 1990; Leahy and Neary 1997; and
10
For a possible priority list of areas of R&D see Ben Israel (1998) and Peled Committee (1999).
11
For the size and other details of the major defense firms in the international markets, see Lifshitz
(1999), Flamm (this volume), James (this volume) and the Economist (1997).
12
For analysis and justification of joint ventures between Israeli and foreign defense firms see
Teubal (1993).
13
For similar incentives to American firms see the Economist (1997).
213
Goolsbee 1998). As Goolsbee (1998) notes, most of the expenditure on R&D consists of salary pay-
ments for R&D workers, and the supply of this scientific and engineering talent is quite inelastic
(Hall 1996; Grossman and Helpman 1994; and Flamm, this volume). Goolsbee shows that U.S.
government increases in R&D spending through subsidies or direct provision brought about signifi-
cant increases in the wages of engineers and scientists in areas that directly benefitted from these
increases, particularly aeronautical, mechanical, metallurgical, and electrical engineers, as well as
physicists. Federal R&D spending (85-95 percent of which goes to defense, space, and energy) has
no significant effect on the salaries of mining, civil, industrial, and chemical engineers, or scientists
in agriculture, biology, and geology). At the same time, U.S. government spending on R&D has had
an insignificant effect on the quantity of inventive activity (measured by hours of work of workers
in R&D). Goolsbee also shows that, through the wage increases, the U.S. government spending on
R&D directly crowds out private spending on R&D (similar crowding out effects were observed in
Israel at the peak of its defense investment on the Lavi project during 1982-87).
Thus, Israel’s public policy should determine the level of spending on defense R&D, as well
as monitor and properly use incentive schemes to ensure the desired allocation of R&D spending
to wage increases and to increasing the quantity of inventive activities.
Summary
In this chapter we have described the Israeli defense industry, emphasizing its role in Israel’s
industrial and technological development. In particular, we show how the effect of the defense
industry on the economy depends, among other things, on the following important factors: (1) the
defense needs of the country; (2) the overall economic situation of the country and the size of the
defense industry relative to the country’s civilian industry; and (3) the stage of technological devel-
opment of the country.
The limited available data suggest that the Israeli defense industry was instrumental in trans-
forming Israel’s civilian industry into a successful high-tech industry. Moreover, the defense sector
is still a very important source of new technological know-how and experienced human resources
for the civilian high-tech industry. Other major conclusions of the paper are as follows:
C Israel must base its main weapon systems on state-of-the-art technologies, and ensure that
some of these systems are unknown to its enemies. Moreover, Israel has to excel in informa-
214
tion warfare, which may be the least costly to execute and the most effective in both full-
scale and low-intensity conflicts.
C The Israeli defense industry must continue to develop modern and highly sophisticated intel-
ligence systems. To avoid surprises, at least some of these systems should be unknown to
Israel’s potential enemies; hence, they must be developed and produced solely for the IDF.
C Israel alone should not attempt to develop and produce by itself platforms or very large inte-
grated systems that can be purchased elsewhere. Adhering to this policy may also be helpful
in avoiding excessive crowding out of civilian R&D by defense R&D (see Teubal 1993 for
a similar argument).
The analysis that we present here calls for extensive, government-operated, national labora-
tories to provide most of the necessary defense R&D. This conclusion is not new—see Ben Israel
(1998) for a possible priority list of areas of R&D, and Peled Committee (1999). Clearly, these labo-
ratories should exploit the capabilities of academic and other research organizations in Israel and
abroad. Additionally, in our opinion, Israel’s defense R&D should be considerably expanded.
C The government-controlled defense organizations, other than those engaging in R&D, should
be merged into two private firms at most. The consolidation process must be carried out
quickly and be based on a massive reduction in the number of workers and restructuring of
wages. The two firms should be unconditionally private. Once the restructuring and privati-
zation processes are completed, the two firms should be free to form international alliances.
The Israeli government should also encourage joint ventures and mergers of private Israeli
defense firms with other Israeli or foreign firms.
C Almost all of the technology transfer from the defense to the civilian sector may be attributed
to individuals (engineers, scientists, managers and officers) moving from the defense indus-
tries or the military to the civilian sector. The experience that these individuals gain during
their military service, often followed by a long service in the army reserves, is an important
factor in their overall education and attitudes.
C Finally, Israel’s welfare and security depend on its defense forces which, in turn, depend on
state-of-the-art technologies and defense R&D. Though only anecdotal statistical data are
available on the Israeli defense industry, it is possible to analyze its past and present struc-
tures and plan and suggest better ones, relying on economic theory, the experience of other
nations and Israel’s defense vision. The non-availability of data is crucial, however, in
215
understanding the processes in which the defense and civilian industries interact. Specifi-
cally, it is important to better understand the channels of technology transfer between the
two sectors and the effect of the Israeli draft system on defense and civilian R&D. We plan
to investigate these processes in future research.
References Cited
Ben Israel, I. 1998. “Defense, Technology, and the Future Battle Field.” Directorate of Defense
Research and Development. Tel Aviv: “MAFAT” [Hebrew].
Dvir, D., A. Hauptman, Y. Sharan, A. Shenhar, M. Sokolov, and A. Tishler. 1998. “Commercial-
ization of Military Technologies in Israel” [Hebrew] Economic Review 2/98: 358-70.
Dvir, D., S. Lipovetsky, A Shenhar, and A. Tishler. 1998. “In Search of Project Classification: A
Non-Universal Approach to Project Success Factors.” Research Policy 27: 915-35.
Economist. 1997. “Linking Arms: A Survey of the Global Defense Industry” (June 14).
Economist. 1998a. “Aerospace on the Defense” (September 12).
Economist. 1998b. “Lets Do Launch” (September 12).
Evans, G.W., S. Honkapohja, and P. Romer. 1998. “Growth Cycles.” American Economic Review
88, no. 3: 495-515.
Goolsbee, A. 1998. “Does Government R&D Policy Mainly Benefit Scientists and Engineers?”
American Economic Review, AEA Papers and Proceedings 88, no. 2: 298-302.
Greenwald. 1992. Scientific Research in Israel, 9th ed. Jerusalem: Ministry of Science and Tech-
nology.
Grossman, G.M. and E. Helpman. 1994. “Endogenous Innovation in the Theory of Growth.”
Journal of Economic Perspectives 8, no. 1: 23-44.
Hall, B. 1996. “The Private and Social Returns to Research and Development.” In Technology,
R&D, and the Economy, ed. B. Smith and C. Barfield, 1-14. Washington, DC: Brookings
Institution.
Halperin, A. 1988. “The Defense Industry and Economic Growth” [Hebrew]. Jerusalem: Jerusalem
Institute for Israel Studies.
Hougi, S., A. Shenhar, D. Dvir, A. Tishler, and Y. Sharan. 1998. “Understanding the Defense Con-
version Dilemma.” Technological Forecasting and Social Change 59, no. 3 (November):
275-89.
Israel Association of Electronics Industries (AEI).1998. Israel Electronics Industries—Profile. Tel
Aviv: Israel Association of Electronics Industries.
Kaplan, G. 1998. “Israel: A High-Tech Haven.” IEEE Spectrum 35, no. 5 (May): 22- 32.
Kovacic, W.E. and D.E. Smallwood. 1994. “Competition Policy, Rivalries, and Defense Industry
Consolidation.” Journal of Economic Perspectives 8, no. 4: 91-110.
Leahy, D. and J.P. Neary. 1997. “Public Policy Towards R&D in Oligopolistic Industries.”
American Economic Review 87, no. 4: 642-62.
Levi-Faur, D. 1998. “The Dynamics of Liberalization of Israeli Telecommunications.” Centre for
European and Asian Studies at the Norwegian School of Management, Norway.
Lichtenberg, F.R. 1988. “The Private R&D Investment Response to Federal Design and Technical
Competitions.” American Economic Review 78, no. 3: 550-59.
216
Etel Solingen*
*
This paper was written under the generous support of a Social Science Research Council-
Macarthur Foundation Fellowship on Peace and Security in a Changing World.
1
For an extended formulation of this argument, see Solingen 1998.
217
218
Internationalist Coalitions
The grand strategy of internationalist coalitions amalgamates economic liberalization at
home, cooperation and stability in the region, and the maintenance of secure access to global mar-
kets, capital, investments, and technology. The economic programs of these coalitions give primacy
to macroeconomic stability and to the discipline induced by international competition. Macro-
economic stability reduces uncertainty, encourages savings, and enhances the rate of investment
(including foreign). Why are these coalitions cooperative with their neighbors? Conflict-prone pos-
tures need to be backed up with the internal mobilization of resources for potential military conflict.
Such mobilization often contributes to many of the ailments afflicting these countries’ domestic
political economy (from the standpoint of internationalist coalitions): the expansion of state power,
the maintenance of unproductive and inflation-inducing military investments, the protection of state-
owned enterprises under a mantle of “national security” considerations, and the perpetuation of rent-
seeking patterns. In principle, therefore, internationalist coalitions resist this syndrome, in an attempt
to avoid inflated military budgets that increase governmental and payments deficits, raise the cost
of capital, curtail savings and productive investment, deplete foreign exchange coffers, induce over-
valued exchange rates, currency instability and unpredictability, and distort the humanpower base.
The increasing high-technology content of modern weapons multiplies these effects, rendering the
trade-offs imposed by military investments more evident than ever before. In sum, in light of the
high opportunity costs of military expenditures, internationalist coalitions are often less predisposed
to extract and mobilize societal resources for external conflict, because such extraction threatens im-
portant macroeconomic and macropolitical objectives they endorse.
This is not to say that internationalist coalitions do not invest in weapons at all, but that when
they do so, two conditions are likely to hold:
(a) Their levels of military expenditures do not fundamentally endanger their internationalist
strategy or shatter the fiscal discipline essential to their political-economic agenda. This approach
to military expenditures is underpinned by a primary concern with sustained economic growth, and
is thus compatible with both the capital formation model and the export-led growth model of the
impact of military expenditures on economic performance, as discussed by Chan (1992). The capital
formation model stresses private investment as the key determinant of economic growth. The export-
led growth model argues that military expenditures tend to deprive the most dynamic sectors—those
involved in exports—of important resources and skills. The consequent decline in international com-
219
petitiveness thus leads to a weaker currency, structural unemployment, chronic trade deficits, and
a less attractive environment for international investments—all outcomes that are anathema to
internationalist coalitions. Where military expenditures are kept at a level that averts such outcomes,
hard political choices between guns and butter can be deferred, as was the case in some East Asian
states until the 1997 crisis.
(b) The second condition likely to hold is that military investments by internationalist coali-
tions are incurred as an insurance policy, particularly against backlash adversaries in the region or
against generalized uncertainty (of the kind unleashed, for instance, by the end of the Cold War).
Broader foreign policy patterns reflect this defensive posture as well. The essential ingredient of an
internationalist grand strategy is economic access, not military prowess. Yet, politically successful
internationalist coalitions are able to persuade their domestic and foreign audiences alike that they
can provide defense, growth, and welfare at the same time, as the 1996 re-election of Taiwan’s
President Lee Teng-hui suggests.
These two conditions are key to understanding the role of military expenditures in states
ruled by internationalist coalitions, a point frequently muted (and most often missed) by aggregate
accounts of military investments.
Backlash Coalitions
The grand strategy of backlash coalitions seeks to preserve statist and ancillary military-
industrial complexes, to uphold a regional context of insecurity and competition, and to resist inter-
national regimes that threaten those objectives. First, these coalitions have an inherent affinity with
import-substituting models of industrialization and classical populist programs involving a strong,
active government controlling prices, overvaluing the currency to raise wages and profits in non-
traded-goods sectors, protecting state enterprises and the military-industrial complex, allocating
credit at low interest rates, and dispensing rents to protected private industry. Arms-importing and
arms-producing military establishments are often adversely affected by adjustment programs, as is
the military as an institution frequently addicted to heavy budgetary transfers. We are all familiar
with the proliferation of military and security agencies—with overlapping jurisdictions—competing
for budgets throughout the industrializing world. External threats are used to legitimize their
existence, yet domestic repression and the ruling coalition’s survival are their most common mis-
sion, as in Iraq and Syria. Military-industrial complexes are also beneficiaries of indirect rents, via
220
state subsidies of important inputs, including raw materials and energy, as in the Russian Republic.
Such complexes (and rents) are often justified on the basis of their positive impact, or spinoffs, on
the development of a modern economic infrastructure. Under backlash coalitions, the military has
thus captured a lion share of rationed foreign exchange.
A second factor accounting for the strong association between backlash coalitions and the
military-industrial complex is that the resolution of regional conflicts has a detrimental impact on
the military-industrial complex. The waning of external conflict weakens the military institutionally
(through its contraction in size and mission) and personally (top and middle echelons undergo radi-
cal trimming in material advantages and prerogatives).
A third factor is the threat that an internationalist grand strategy posits for the military-indus-
trial complex, in the form of emerging global security regimes that restrict the rationale for military-
industrial complexes and undermine notions of territorial sovereignty that are central to the military
as an institution. An internationalist grand strategy, by definition, implies an acceptance of certain
strings—political and strategic—that constrain the military and its industrial complex.
In sum, generally speaking, internationalist coalitions may invest in military capabilities but
prevent such investments from overwhelming domestic reform, regional stability, or global access.
Backlash coalitions, instead, create and maintain a Wehrwirtschaft (war economy) that functions as
their core political pivot.
International Institutions
More open, liberal, international economic structures and regimes have undermined the
viability of military-industrial complexes and military establishments. Structural adjustment efforts
often threaten these complexes, as does the very demand for greater budgetary accountability and
transparency in intergovernmental relations. Whatever disagreements on guns-versus-butter trade-
offs there may be in the scholarly community, it is rather clear to the relevant political actors that
economic stagnation exacerbates such tradeoffs. The tradeoff is most salient for developing coun-
tries facing severe financial constraints (onions versus weapons in India). International investors
understand such tradeoffs well, and the World Bank has began addressing, more directly than ever,
the size and transparency of military budgets. The fact that the interlocutors of these international
institutions are mostly Central Banks and finance ministries and not military agencies, is a double-
edged sword. On the one hand it can undermine the domestic legitimacy of such agencies, accused
221
of taking their cues from foreign institutions. On the other hand it can be used to shift the blame for
downsizing “national” symbols—such as the military sector—to external actors, while highlighting
the positive economic, social, and political outcomes of that process.
Preliminary Findings
I have examined this general argument quantitatively (Solingen 2000) through an historical
analysis of four industrializing regions (the Middle East, East Asia, Latin America, and South Asia),
with many states and many coalitions within them. This study surveyed a sizeable part of the indus-
trializing world longitudinally, allowing a dynamic assessment of older forms as well as more cur-
rent responses to internationalization. This led to a wide domain of cases and to a deeper historical
probe than would be possible if I had concentrated on the post-1990 era alone. After all, the
conceptual framework has roots in earlier periods, even if it has gained sharper definition at the end
of the twentieth century.
This overview of coalitional behavior regarding economic openness, military investments,
and regional and international security regimes suggests that there is more in common within coali-
tional categories than across them. Thus, entrepreneurs coalescing an internationalist coalition are
more prone to deepen their country’s trade openness, expand exports, attract foreign investments,
tame profligate military-industrial complexes, eschew weapons of mass destruction, defer to interna-
tional economic and security regimes, and strive for regional cooperative orders that reinforce those
objectives. Although pristine and coherent grand strategies are hard to find, the links between a com-
mitment to an internationalized economic strategy and regional cooperation and stability are evident.
The links are even thick with respect to nuclear policies, where the avoidance of a nuclear compe-
tition is expected to have positive externalities at home (downsizing a major backlash constituency),
regionally, and internationally. This link transcends differences in the depth of security dilemmas
across regions, and is found from South Korea to Egypt to Latin America’s Southern Cone.
Examples of internationalists’ performance in military investments include South Korea,
where external and internal conditions weighed in favor of an expansive military complex. Yet,
despite such circumstances Park—a military ruler—subdued supporters of a large statist military-
industrial complex that threatened his grand strategy (Amsden 1989). South Korea’s GDP grew by
10 percent (1965-1989 average), whereas military expenditures as a percentage of GDP remained
222
largely constant, declining to 3.6 percent by the early 1990s.2 Even during the period of consolida-
tion of its export-led strategy (1962-72), South Korean expenditures for economic development were
higher than for defense.3 The defense burden absorbed on average 4 percent of the GDP before 1975,
although U.S. direct military assistance was extensive between 1953 and 1973. Beginning in the
early 1970s U.S. military grants declined drastically (ceasing completely in 1978), raising South
Korea’s average military expenditures to near 6 percent of GDP initially, decreasing to 5 percent by
1985 and 3.6 percent by the 1990s. Taiwan’s military expenditures/GDP was 8 percent (on average
for 1961-1987), declining by the 1970s as internationalization took root (Chan 1992). Averages for
internationalizing Southeast Asia by 1990-1991 were 2.8 percent.
Backlash entrepreneurs were found to restrict and reduce trade openness and reliance on
exports, curb foreign investment, build expansive military-industrial complexes, spearhead weapons
of mass destruction programs, challenge international security and economic regimes, and exacer-
bate civic, religious, and ethnic nationalist differentiation within their region through an emphasis
on territoriality, sovereignty, and self-reliance. Even when states attempted to avoid what might be
self-defeating wars and arms races, the risks and externalities of these policies pushed toward them.
Backlash entrepreneurs have used chemical weapons and have spearheaded an overwhelming major-
ity of the wars fought since World War Two, in many cases prodding reluctant superpower benefac-
tors: for example, North and South Korea in the 1950s or Egypt in the 1960s and 1970s. At the same
time, the Cold War era provided a more supportive global structure for the objectives pursued by
backlash coalitions, from economic protection to militarization and regional conflict. This structure
may also explain the relative scarcity and weakness of internationalist coalitions and entrepreneurs
during that era (relative to backlash cases) and their diluted grand strategies.
The record of backlash coalitions regarding military expenditures contrasts dramatically with
that of internationalists. The high incidence of backlash coalitions in the Middle East helps account
for particularly high regional averages (nearly 19 percent of GDP in the 1970s-1980s), over three
2
Data on military expenditures are from Stockholm International Peace Research Institute (SIPRI)
(1975-1996), International Institute for Strategic Studies (IISS) (1995-96, 266-67), and World Bank
(1998, 279).
3
On how military investments came to compete with economic developmental objectives by the
1970s, with the onset of nearly full employment of human and fiscal resources, see Wolf (1981, 82).
See also Byung Chul Koh (1984, 210-11), Moon and Hyun (1992), and Ball (1988, 54).
223
times higher than the industrializing world’s average.4 Iraq’s military spending reached over 50
percent of GDP (1973-1985), ten times the global mean. Military expenditures/GDP under Nasser
and Sadat’s backlash period (1970-1973) reached 24 percent, and Syria’s mean under Hafez al-Asad
was 16 percent (1973-1985), declining with incipient liberalization to less than 10 percent after
1988.5
Hybrid instances were common, and, as expected, straddled the grand strategies of their
purer types, intermittently striving for economic openness, the contraction of the military complex,
and cooperative regional and international policies, albeit less forcefully or coherently. As this last
point suggests, notwithstanding coalitional commonalities, the relative domestic strength of a coali-
tion and the degree to which entrepreneurs must logroll across disparate sectors create significant
differences even within coalitional categories, compelling different entrepreneurs to package their
grand strategy differently, maintaining some aspects while discarding others. For instance, in Israel
Netanyahu’s hybrid coalition slid towards more backlash constituencies after 1996, reversing the
decline in military spending under the internationalists, and this despite the most favorable regional
conditions ever. Netanyahu proclaimed a strategic decision to increase military spending and
“strengthen security” as the first budgetary priority.
In Argentina, President Raúl Alfonsín’s hybrid coalition contracted military budgets to some
extent between 1984 and 1989 but retained the air force’s Condor II program in 1985, maintained
relatively high levels of military expenditures as a percentage of GDP (over 3 percent), and sus-
tained Argentina’s opposition to the NPT and its right to peaceful nuclear explosions, as well as its
refusal to ratify the regional Tlatelolco treaty (Franko 1994, 37-74). By contrast, President Carlos
S. Menem’s internationalist revolution drove military expenditures to all-time lows, both relative
to past military expenditures (1 percent of GDP in 1992) and relative to health and education (51
percent), the lowest in three decades. The military’s total size shrank by 60 percent from the 1970s
in absolute terms, and by 70 percent relative to population. Employment in military industries
declined by 80 percent, the officer corps was dramatically reduced, compulsory military conscription
ended (1996), and a pillar of the military-industrial complex (DGFM)—the largest drain on the
4
The average ratio of military expenditures over GDP for developing countries in the 1970s and
1980s was between 4 and 6 percent (West 1992, 25, 31).
5
SIPRI warns its figures for regimes such as Iran tend to underestimate military expenditures.
224
federal budget—was privatized.6 Part of Menem’s strategy for neutralizing the Argentine military’s
domestic presence was to engage them in UN peacekeeping activities, an endeavor that only en-
hanced Argentina’s new internationalist credentials.
Military investments were high overall during the Cold War, particularly in regions more
directly affected by it. Hence, one might argue that the dice are somewhat loaded against finding
effective differences across coalitional variants. Furthermore, a number of internationalist coalitions
—notably in East Asia—were particularly engulfed by Cold War threats and their regional corol-
laries. Nonetheless, the evidence does reveal clear contrasts across coalitional variants. While invest-
ing in military capabilities internationalist coalitions have largely prevented them from overwhelm-
ing domestic reform, regional stability, or global access. By contrast, backlash coalitions reflect the
hypothesized penchant for extensive military expenditures as a percentage of GDP. Not every
backlash coalition exhibits such high levels, although they do reflect military investments that are
higher than their internationalist counterparts in the same country. Hybrid regimes facing powerful
backlash constituencies at home and in the region exhibited, as expected, higher levels of military
expenditures as a percentage of GDP and an essential inability to contract them.
Conclusions
Important sectors within the military have opposed internationalization in many cases,
including Argentina, Brazil, Uruguay, Egypt, and Thailand, at least initially (and witness current
winds in Venezuela and Ecuador as of 2000). By contrast, in Chile and South Korea, the leading
political entrepreneurs (generals in both cases) coalesced their respective armed forces behind a
strategy of integration into the global economy. In Algeria and Turkey, the military transformed
itself from custodian of the post-independence statist-nationalist (or backlash) project to the main
line of defense against recent backlash and confessional onslaughts on economic liberalization. It
is important to recognize that the interests of military factions can differ and that highly fragmented
armed forces have led to factions supporting competing coalitional arrangements. Under such
conditions, political entrepreneurs within or outside the armed forces have sometimes been able to
impose a leading grand strategy over military factions (Presidents Park and Menem are examples).
6
DGFM’s cumulative deficit was estimated at over $700 million and its assets at $5.6 billion in
1990 (Manzetti 1993, 197).
225
Which faction will prevail is never easy to predict ex ante. As economist Díaz Alejandro (1983, 45)
concluded prophetically in his study of international economic openness, “the nature and laws of
motion of the collection of men in uniform are the darkest black boxes in Latin American social
science, but one may conclude that the attitude of the armed forces toward economic openness has
been neither unambiguous nor steady.”
The findings reported here suggest that, on the whole, military institutions seem more likely
to join backlash—rather than internationalist—coalitions, thus turning dependencia-style arguments
about a basic alliance between global capitalism and local military establishments on their head. De-
pendency theory has traditionally thrown the military into the “globalizing” camp, as the chief polit-
ical executor of “external” designs, brutally whipping production for the global economy. The his-
torical affinity between inward-looking statism and military regimes challenges this assumption.
Integration into the global economy does not require military-industrial complexes, while inward-
looking statist-nationalism has more often than not overlapped with the expansion of such com-
plexes. The domestic allies of global capitalism can be the military’s most powerful political adver-
saries. In broad terms, the fact that the explosion of economic liberalization (and the rise of its polit-
ical bearers) has been associated with a dramatic collapse in budgetary allocations to the mili-
tary—and in the latter’s political leverage—deals a rather serious blow to dependency-deterministic
theories of military institutions and their associated industrial complexes in the industrializing world.
References Cited
Amsden, Alice H. 1989. Asia’s Next Giant: South Korea and Late Industrialization. New York:
Oxford University Press.
Ball, Nicole. 1988. Security and Economy in the Third World. Princeton: Princeton University Press.
Chan, Steve. 1992. “Military Burden, Economic Growth, and Income Inequality: The Taiwan
Exception.” In Defense, Welfare, and Growth, ed. Steve Chan and Alex Mintz, 163-78. New
York: Routledge.
Díaz Alejandro, Carlos F. 1983. “Open economy, closed polity?” In Latin America in the World
Economy: New Perspectives, ed. Diana Tussie, 21-54. Aldershot, England: Gower.
Franko, Patrice. 1994. “De Facto Militarization: Budget-Driven Downsizing in Latin America.”
Journal of InterAmerican Studies and World Affairs 36, no.1 (Spring): 37-74.
International Institute for Strategic Studies (London). 1995. The Military Balance 1995/96: 266-67.
Jeon, Jei Guk. 1995. “Exploring the Three Varieties of East Asia’s State-Guided Development
Model: Korea, Singapore, and Taiwan.” Studies in Comparative International Development
30, 3 (Fall): 70-88.
226
Koh, Byung Chul. 1984. The Foreign Policy Systems of North and South Korea. Berkeley: Univer-
sity of California Press.
Manzetti, Luigi. 1993. Institutions, Parties, and Coalitions in Argentine Politics. Pittsburgh: Uni-
versity of Pittsburgh Press.
Moon, Chung-in and In-Taek Hyun. 1992. “Muddling Through Security, Growth, and Welfare: The
Political Economy of defense Spending in South Korea.” In Defense, Welfare, and Growth,
ed. Steve Chan and Alex Mintz, 137-62. New York: Routledge.
Solingen, Etel. 1998. Regional Orders at Century’s Dawn: Global and Domestic Influences on
Grand Strategy. Princeton: Princeton University Press.
———. 2000. “Mapping Internationalization: Domestic and Regional Impacts.” Ms. University of
California, Irvine.
Stockholm International Peace Research Institute (SIPRI). Yearbook 1975, 1984, 1986, 1989, 1990,
1995. Armaments, Disarmament and International Security. Oxford: Oxford University
Press.
UN Development Program. 1991-1996. Human Development Report. New York: Oxford University
Press.
West, Robert L. 1992. “Determinants of Military Expenditure in Developing Countries: Review of
Academic Research.” In Military Expenditure and Economic Development: A Symposium
on Research Issues, ed. Geoffrey Lamb with Valeriana Kallab. World Bank Discussion
Paper No. 185. Washington, DC: World Bank.
Wolf, Charles Jr. 1981. “Economic Success, Stability, and the ‘Old’ International Order.” Interna-
tional Security 6, 1 (Summer): 75-92.
World Bank. 1998. World Development Indicators. Washington DC: World Bank.
227
Contributors
Eugene Cobble is a Ph.D student in the field of Government at Cornell University. He is currently
employed by the Center for Naval Analysis in Alexandria, VA.
Dov Dvir works in the Management of Technology Department, Holon Center for Technological
Education, Israel.
Kenneth Flamm is a Professor and Dean Rusk Chair in International Affairs at the Lyndon B.
Johnson School of Public Affairs, University of Texas at Austin, and a Senior Fellow at the
Brookings Institution, Washington, DC.
Andrew James is a Research Fellow at PREST (Policy Research in Engineering, Science &
Technology) and CRIC (ESRC Centre for Research on Innovation & Competition) at the University
of Manchester, UK
John Lovering is a Professor in the Department of City and Regional Planning, Cardiff University,
Wales, UK where he researches and teaches regional development and defense restructuring issues.
Ann Markusen is Professor of Planning and Public Affairs at the Humphrey Institute, University of
Minnesota, and Senior Fellow at the Council on Foreign Relations, New York.
Judith Reppy is a Professor in the Department of Science & Technology Studies and Associate
Director of the Peace Studies Program at Cornell University.
Judith Sedaitis is an organizational sociologist and Russian area specialist currently teaching at
Columbia University. She has published several articles and edited volumes on restructuring and
technology transfer in post-socialist economies.
Etel Solingen is Professor of Political Science at the University of California, Irvine and Senior
Fellow at UCLA's Center for Pacific Rim Studies and UCLA's Gustav von Grünebaum Center for
Near Eastern Studies.
Asher Tishler is a member of the Faculty of Management, Tel Aviv University, Israel.