This item is only available as the following downloads:
THE NEW INTERDEPENDENCE:
MARKET COMPETITION RECONFIGURING THE INTER-STATE SYSTEM
NATHAN A. ADAMS. IV
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
Nathan A. Adams, IV
I dedicate this dissertation, first, to Yahweh, second, to my
wife whom I adore, and, third, to my parents and the memory
of my Grandfather. Without the love, encouragement, and
inspiration that they have shown me I would not be who I am,
and this dissertation would never have been written. All that
it is, and all that I am I owe to Jesus Christ.
All things were made by him; and without him was not any
thing made that was made. In him was life; and the life was
the light ofmen.
In writing this dissertation, I benefited greatly from the help of many professors. First, I
want to thank the members of my dissertation committee, Professors Leann Brown. David
Conradt, Alfonso Damico, Gary Goertz. and Winston Nagan. One-time committee members,
Professor Walter Rosenbaum and Roy Crum, were also more than helpful. Professor Gary Goertz
went far beyond the call of duty in scrutinizing drafts of this dissertation and providing
constructive comments even after he left the University of Florida for the University of Toronto.
I owe him the most thanks. But Professor Brown has also been painstaking in her critique of
chapters, so I thank her: and I thank Winston Nagan for his thoughtful reflections on the
implications of The New Interdependence for international law, my new love.
At the University of Florida, I was privileged to be among an outstanding group of
scholars. I would like to thank the Department of Political Science and its former Chairman,
Kenneth Wald, and former Graduate Coordinators. Alfonso Damico and Michael Martinez. for
supporting me throughout my doctoral studies. Also, special thanks are due to the Department's
secretaries, Debbie Wallen and Marty Swilley, for their technical support and constant good
Since I completed this dissertation at the University of Texas and University of Virginia.
I owe library and technical staff at those institutions my gratitude, as well. More importantly. I
should acknowledge the continued mentorship of professors from Wheaton College in Wheaton,
Illinois, who introduced me to my fascination and love for political science: Professors Mark
Amstutz and Lyman Kellstedt. In particular, I thank Professor Mark Amstutz, whose primary
field is international politics, for the wise guidance that he has provided me throughout my
undergraduate and graduate training in my field of interest and in many other areas of life.
The University of Florida Graduate School provided me with a fellowship in 1992-93.
which enabled me to finish most of the research for this dissertation. I am grateful for this
financial support and for the Graduate School's editorial assistance. I have received only patient
help and knowledgeable guidance from the Graduate School. Hence, I also gratefully
acknowledge the many people in the Graduate School who helped me administratively.
Last, but most importantly, my wife, Heather L Adams, and parents, Nathan A. Adams.
Ill and Eleanor Adams, have helped me though the most challenging days of this project. They
have been long-suffering encouragers and my wife. an absolutely essential helpmate. Many was
the night after a long day of work that she gave up her own pursuits to type a Figure or Table to
add to the many in this dissertation. Thus, by rights, I consider this achievement, ours. To be
fair, it is also partly my dad and mom's. They have been wonderfully supportive of everything
that I have done. Thus. with love in my heart. I acknowledge my family, wife, and God, and
dedicate this work to them
TABLE OF CONTENTS
TABLE OF CONTENTS ............
LIST OF TABLES .. .. ....
LIST OF FIGURES ....... .........
ABSTRACT ... ... ...
1. INTRODUCTION .....................
Understanding Old and New Interdependence ...... .
Methodology ....... .. ..
2. STATE OF THE THEORY ......... ....
The Interdependence Approach .. ...
Convergence........... .. ...
Collective Goods Problems .. .. ... .....
Lowest Common Denominator Agreements .....
Framework for the Interdependence Approach ...
Independent Variables .
M obility of Capital ................
Technological Innovation .. ....
A Contradiction, Not Compromise .... .
The Competition State .....................
Sum mary .......... ... ..O. .
3. WEAPONS PROCUREMENT POLICY ..................
Convergence ............ .. ....
Collective Goods Problem .. ... ...
Lowest Common Denominator Agreements ..... ..
Weapons Cooperation in Western Europe ..........
Capital Mobility and Technological Innovation ......
. .. .. vi
... ...... ......... 10
. .... 17
.. .. .. 27
. 2 8
. . 2 9
. . 3 4
. 3 5
. 3 9
. . 40
. . 4 6
. 4 7
. 5 1
. . . 60
. . 70
... . 7 6
. . 8 2
. . 8 8
Contradiction in NATO Procurement ..........
Conclusion .. ..........
4. MONETARY POLICY .....................
Collective Goods Problem ................ .
Lowest Common Denominator Agreements .....
The European Monetary System .... ......
Capital Mobility and Technological Innovation ....
The Contradiction of Embedded Liberalism .......
Conclusion ... ............
5. INTERNATIONAL TRADE ....... ..... ...
Collective Goods Problem ..................
Lowest Common Denominator ...........
Capital Mobility and Technological Innovation ....
Embedded Liberalism in Trade ........ .
.. . 9 3
. 9 5
. . 10 4
. . 10 5
. . 1 15
. . 12 1
. 12 7
. . 132
.. . .. 134
. . 13 8
. . 14 9
. . . 150
. . . .. 15 7
. . 16 8
. .. 178
S. .. . 18 7
C conclusion .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. 19 3
6. ENVIRONMENTAL POLICY .... ..... .. ................... ..... .207
Convergence ................................ ................ 208
Collective Good ... .... ... ... .. ..... ........... 221
Lowest Common Denominator ......... ......................... .... 229
Capital Mobility and Innovation .. ................ 235
Contradiction ..... .... ........................... .. 245
Conclusion ................ .. ............... 248
7. CONCLUSION ......... ........................................ 257
Confirming the Hypotheses. ...................................... .... 258
Policy Responses to the New Interdependence .................... ... ... 278
APPENDIX A: POPULATION OF NON-REGIONAL, MULTILATERAL ENVI-
RONMENTAL TREATIES ................... ......... .. 283
APPENDIX B: POPULATION OF REGIONAL, MULTILATERAL ENVIRON-
MENTAL TREATIES .......... ............. 285
REFERENCES........ ............. ............... .......... .... 287
BIOGRAPHICAL SKETCH ....................................... ....... 305
LIST OF TABLES
1-1 Country Stock Market Correlations with the World Stock Market ... .......... 20
1-2 Growth in Unregulated Capital Markets, 1970-88 ................. .... .... 21
1-3 Relative Size of the Eurocurrency Market ............................. ... 22
2-1 Correlation Matrix of Measures of Innovative Activity. ................... .... 54
3-1 Number of Major Weapons Systems in NATO ..... ........................ 99
3-2 Relative US/USSR Technology Level in Deployed Military Systems (late-1980s) .100
3-3 Members of Systemic Weapons Standardization Groups ......... ......... 101
3-4 The MLRS: An Example of the Family of Weapons Approach ........... .... 102
4-1 Financial Liberalization .... .. ....... .......... ....... ... .... 142
4-2 Periods When the Collective Good of International Financial Stability was
Provided and its Results ......... ... ..... ..................... 143
4-3 The EMU Indicator ......... ................. ...... 144
4-4 Innovation in International Financial Markets, 1950-80 ....... ........ 145
4-5 Correlation of National Stock Markets ......... .................. ... 147
4-6 Major International Financial Agreements and Events, 1962-92................ 148
5-1 Trade Reform in Latin America ..... .................................. 198
5-2 Liberalization in Regional Trade Blocs. ................. ....... ... 199
5-3 Income Convergence Coefficients in Postwar, by Group ..................... ..200
5-4 Long-Term Income Convergence Coefficients, by Group ...................... 201
5-5 Average Intra-Industry Trade in Selected EC Countries. ................... .. 202
5-6 Unfair Trade Practices Cited Under Sect. 301. ........ .................. 203
5-7 Assessing the Strength of the GATT's Normative Structure. ................. 204
6-1 Comparisons of Actual Air Quality Control Costs to Estimated Least Costs ....... 252
6-2 Comparisons of Actual Water Quality Control Costs to Estimated Least Costs .... .253
6-3 Summary of Regional and Non-regional Multilateral Treaties, Appendices 6-1
and 6-2 ......... .. .. ............. .254
6-4 Capital Expenditures of Nonbank Majority-Owned Foreign Affiliates as
Percentage of Total Manufacturing. ...... ........ ................... .255
6-5 Trade in Hazardous Substances as a Percentage of Total Annual
Merchandise Trade ................... ..................... .256
LIST OF FIGURES
Conceptualizing Interdependence .......................
General Framework for the Interdependence Approach ......
Eroding Autonomy of the Stale .. .. .............
Case-study Variance .................................
Interdependence Approach .....................
Technological Innovation in Information Technologies. ......
General Framework for the Interdependence Approach ........
The Co-Produced F-16 .. ..................
The Trilateral Commission -- Trade and Investment Flows ....
......... ...... .23
. .. .. 24
.. 2 5
. 2 6
.. .5 5
S . .. 56
. .. .. 5 7
.. 10 3
. ....... 206
Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
THE NEW INTERDEPENDENCE: MARKET COMPETITION RECONFIGURING THE INTER-
Nathan A. Adams, IV
Chairman: Alfonso Damico
Major Department: Political Science
This dissertation develops and tests a new approach to the study of interdependence in the
international political economy that marries the insights of structural-realism with liberal political
economy. The Interdependence Approach. Three dependent variables define the approach: (1)
convergence in public policy toward a liberal, global equilibrium; (2) the transformation of domestic
regulatory problems into international collective goods problems; and (3) the tendency of the market to
restrain states acting multilaterally from enforcing regulations that exceed a lowest common denominator.
The two independent variables which primarily account for these processes are (1) the increasing mobility
of capital and (2) technological innovation.
Together, these variables illuminate the contradiction which an interdependent political economy
poses in the 1990s. for a state which is naturally inclined to govern autonomously, but which must
coordinate its regulatory apparatus with other states to solve problems effectively, if at all. The author
calls this new stage of interdependence, in which even the most economically influential countries are
limited: The New Interdependence. He contends that, contrary to the assertions of neo-realists.
technological innovation and capital mobility best account for the failure of multilateral regulatory
systems like Bretton Woods and the European Monetary System (EMS). Countries which try to regulate
the market more than others or which pursue macro-economic policies at variance with the norm in other
industrialized countries experience capital flight and declining industrial competitiveness.
Thus, the trend since the 1970s has been for the state to regulate the market less and, in fact, to
facilitate market competition. This dissertation finds this to be the case in four case studies that deal with
some of the primary functions of the state: defense procurement policy, monetary policy, trade policy, and
environmental policy. Moreover, it finds the same trends in the context of both Western European
integration and industrialized countries, in general. Each case study tests a series of five hypotheses and
three corollaries, and reaches the conclusion that the modern welfare state is being transformed into a
I believe that we are undergoing a major reorganization of
international life at the present time which will result in drastic
modification of the world order system that has prevailed since the
Peace of Westphalia in 1648.
(Richard Falk, 1975)
How can we explain the fact that, increasingly in international politics, even advanced
industrialized states appear constrained in their ability to pursue both their international and domestic
objectives? The assumption is now widespread that the world has become significantly more
interdependent and, thus, the state, less autonomous than before. Yet the way scholars have modeled
this difference has not always been persuasive. Hence, many structural-realists have come to the
conclusion that interdependence is just a new word for the same international politics in which states
have always participated.
Three processes accelerated in the 1980s, which cast doubt on this conclusion. I believe that
they are dramatically changing the character of the international political economy. When considered
together, these processes constitute what I call the Interdependence Approach. Those variables are (1)
convergence in public policy toward a liberal, global equilibrium; (2) the transformation of domestic
regulatory problems into international collective goods problems; and (3) the tendency of the market to
restrain states acting multilaterally from enforcing regulations that exceed a lowest common
Scholars have traditionally associated none of these three processes with interdependence,
either theoretically or empirically. Instead, they have generally related it to so-called international
regimes in the 1990s. Despite the merits of this association, I am convinced that it has been a major
oversight for scholars to overlook the three variables of the Interdependence Approach. They may
provide the best index of how the state is changing in the last decade of the Twentieth Century. In the
case-studies that follow I provide evidence that the state is shedding its Keynesian and, now, even
monetarist heritage to become a competition state designed to snatch the rents from imperfect
competition which modern trade in technologically innovative products and transnational investment
stimulates. This new state is not primarily a market regulator, so much as simply a market taker and
The most important variable in the Interdependence Approach is convergence toward a liberal
global equilibrium in public policy. By "convergence." I mean (in the words of Alex Inkeles 1975,
472) a process, "Where social units start from diverse positions on some scale of organization,
sociopolitical structure, or culture, and then move toward some more connmon form on the given
dimension." More generally, if we imagine that a continuum extends from independence to integration
as in Figure 1-1, interdependence would be relevant only between the limits. Convergence would
occur in the zone beyond the median of the continuum to its end-point. The end-point represents
"integration." Thus, for our purposes, complete convergence is the same thing as integration.
Convergence is a critical process, because it suggests that industrialized countries which try to
regulate the market more than others will face substantial repercussions. Most devastating, capital will
tend to flow out of the state which regulates it excessively in search of higher rates of return.
Likewise, industry will prove non-competitive in the long-term in the overly regulative state. Finally,
public policy, itself, will prove too costly for the state to enforce. In the end, the state will have to
abandon the very regulations that cost it so dearly.
As for the transformation of domestic problems into international collective goods problems,
this process is denying the state its former control over substantive issue-areas. Now, effective
solutions to many regulatory problems that the state used to solve by itself require multilateral
solutions. Even such classically low politik concerns as hazardous waste management and monetary
policy now require coordinated attention at the highest levels of government.
More disturbing is evidence that not even the multilateral community can solve problems that
states formerly took care of individually. If states address collective goods problems at all, they are
likely to take largely symbolic actions to demonstrate their normative belief that they should cooperate
to resolve the problem, rather than to effectively regulate the problem. This posturing is likely to
center around the response least costly for the most powerful actors involved to take. For this reason,
I call these policy responses lowest common denominators (LCDs). As we shall see, the conflict and
unresolved problems this leaves us with are a powerful reason why some expect supra-national
governance in the future.
The two independent variables which I believe account for the acceleration of the three
processes of the Interdependence Approach in the 1980s are (1) the mobility of capital and (2)
technological innovation. Figure 1-2 should begin to make the outline of this causation obvious. The
mobility of capital is the primary independent variable. It refers to the ability of both liquid and fixed
assets--that is, money and investment--to change locations. The more mobile capital is, the more likely
the three processes of interdependence are to accelerate together.
The second independent variable, technological innovation, includes "not only technical
changes in production processes and equipment but also the discovery of new resources, the
development of new types of organization, and new methods of management" (Schumpeter cited in
Weber and Handfield-Jones 1954, 115). The primary technological innovations which are of interest to
me are in information technologies, e.g. computers, telecommunications, software, and micro-
The most important way in which technological innovations influence the three dependent
processes that I have described is indirect. Technological innovation in information technologies allows
capital to be more mobile; that is, it creates a more conductive medium through which capital can
move. However, technological innovation can also have an important independent effect on the
variables of the Interdependence Approach, since its increased pace makes research and development
(R&D) more important to firms that want to remain competitive in the global market. This R&D has
become more costly with time, encouraging firms to join multinational consortia to develop new
products. Obviously, this increases interdependence and the tendency toward convergence in product
Thus, capital mobility and technological innovation can co-vary, but not necessarily. Clearly,
some innovations have no effect on the mobility of capital. Likewise, the availability of capital does
not always advance technological innovation. Scarcity of ideas plays an independent role in the
innovation process. Yet, as R&D costs rise, scarcity of capital is becoming a more critical restraint on
Why is this so important? Capital became mobile enough and technological innovation rapid
enough in the 1980s, that the three dependent processes of the Interdependence Approach began
seriously limiting the autonomy of even the largest industrialized states. To be sure, prior to the
1980s, these processes prompted a number of crises in the international political economy (like the
collapse of Bretton Woods), but they seriously bound only the autonomy of industrialized and
developing countries with relatively small national incomes in issue-areas where capital was highly
liquid (usually, areas related to low politics). The revolutionary change which began in the 1980s was
that the dependent processes began to bind even economically influential industrialized countries in the
area of high politics. This new, advanced stage of convergence, far to the right on Figure 1-1, I call
the New Interdependence. Since it is completely different from past periods of interdependence, I
believe that it deserves special exploration.
Understanding Old and New Interdependence
There are two primary theoretical contributions this research makes to the study of
interdependence that are important to distinguish between, as well as to distinguish from classical
(1) The "Interdependence Approach" is the name I have given to the new approach to
interdependence which I take. It is concerned primarily with three processes:
a. convergence in public policy toward a liberal, global equilibrium
b. the transformation of domestic regulatory problems into international collective goods
c. the tendency of the market to restrain states acting multilaterally from enforcing
regulations that exceed a lowest conmlon denominator (LCD).
(2) The "New Interdependence" refers to an empirical phenomenon--a new, advanced stage of
interdependence that is binding even the most economically influential industrialized countries,
in the areas traditionally regarded as sacrosanct to the state, i.e. high politics. Always
advancing, the three processes of the Interdependence Approach caused the level of
interdependence to increase and traverse the continuum in Figure 1-1 from left to right, until at
last we reached a level of interdependence-the New Interdependence--which suggests the
outlines of a radically new inter-state system.
Taking up the second of these themes, first, Figure 1-1 shows where the old interdependence
would fit in relation to the New Interdependence. If a y-axis were added to this continuum
(representing autonomy), Figure 1-3 would be the result. Here, as interdependence increases it is
likely to limit the autonomy of, first, small states in the area of low politics and, last, large developed
states in the area of high politics. Thus, whereas in the 1970s, interdependence forced northern
European states to remain attached to the German Mark in foreign currency markets, France, Britain,
and Italy detached themselves from it. In the 1980s, in contrast, interdependence forced even the most
economically powerful European states to follow the lead of the Bundesbank.
Overall, what is most important about the empirical condition of interdependence today is that
even the most influential industrial countries cannot pursue policies penalizing capital for long, no
matter the issue-area concerned. In this sense, the New Interdependence stands for an unprecedented
situation where the autonomy of even influential industrialized countries is limited by the market. For
example, in the 1970s, whereas the US could simply "close the gold window" and not worry about the
implications for its own economy, today it cannot significantly influence the value of its currency over
the long-term. Regulating investment is also a problem. When Germany decided in October 1987 to
authorize a 10% withholding tax on interest income (effective January 1989), German capital poured
into deutsch mark Eurobonds and other offshore assets. In the end, Germany had to retract the tax,
three months after it went into effect. Indeed, in an environment characterized by the New
Interdependence monetary policy has to be liberalized.
To demonstrate more generally how "new" interdependence is today, I will concentrate for a
moment on my primary variable (convergence) and steal a page from the literature on international
finance. According to that literature, as long as developed countries are fairly independent of one
another, their stock markets should reveal randomly diverging average yields. In contrast, as the
markets of developed countries become linked or, in other words, converge, stock market yields should
begin to correlate. Table 1-I shows a few examples of how dramatic this increasing correlation is. On
average, the correlation in yields between the hypothetical world market (the average of all stock
markets) and the individual G-5' stock markets increased from .58 in the late-1970s to .82 in the
1990s. Chapter Four discusses these trends in more detail (see infra, Table 4-5).
Another variable which dramatically underscores how new contemporary interdependence is in
the 1980s is the mobility of capital. Table 1-2 reveals dramatic growth in the eurocurrency and
eurobond markets.' Within the eurocurrency market, eurodollars tripled in size from the time they
were invented in 1959 to the late-1960s. Consistent with our expectations, the primary reason had to
do with restrictions that the US put on foreign investment and interest rates in the 1960s to preserve the
value of the dollar. Table 1-3 shows the relative importance of this trend. By 1980, the net
eurocurrency market exceeded the total value of every major country's individual time deposits (TDs),
including the United States'. Thus, for the first time in the 1980s, the largest block of currency
floating around in the world economy was unregulated and extra-territorial.
Yet, neither this trend toward increasing capital mobility, nor toward convergence are
emphasized as important variables in the old interdependence literature. Keohane and Nye defined
"complex interdependence" in terms of three characteristics: (1) multiple channels connect societies
(including informal ties between governmental and nongovernmental elites and transnational
organizations), (2) there is no hierarchy among issue areas: hence, military security does not
consistently dominate the agenda, and (3) governments do not use military force against other
'The G-5 countries include the U.S., Japan, Germany, Britain, and France.
2Although eurocurrency originally got its name from the fact that die market originated in Europe,
"Euro-" now properly refers simply to "external." That is, the only restriction on location in this currency
market is that it must be situated outside the country of the currency in which the claim is denominated.
For the most part, this market is a wholesale one that deals in sums exceeding $1 million and extends credit
primarily to well-known corporations, banks, or governments. Its most important advantage for the private
sector is the absence of public regulation.
governments within the area covered by the term. Although innovative, this approach has not led the
search for a compelling model of interdependence out of the wilderness of critiques by structural-
In general, the interdependence literature of the 1970s was hard to falsify because scholars
often reached conclusions deductively. They illuminated the burgeoning role of non-state actors in
international politics (Keohane and Nye 1971, 1977; Morse 1970; Taylor 1984; Inkles 1975),
increasing interconnectednesss" of the world community (Inkles 1975; Keohane and Nye 1971), the
merger of foreign with domestic politics and "low" with "high" politics (Cooper 1972; Morse 1970;
Keohane and Nye 1977), and the declining autonomy of the state (Morse 1970; Cooper 1972; Wagner
But scholars often were not clear on how to operationalize these changes in the international
political economy. They usually failed to distinguish between the theoretical approach they used and
the empirical reality that they observed. This is why I have carefully maintained a distinction between
the "Interdependence Approach" (theory) and the "New Interdependence" (the new reality).
Another shortcoming of old interdependence scholarship is that many of the trends it pointed to
were really not new, nor perhaps as qualitatively different as they might have imagined, at least until
the late-1970s. For example, Kenneth Waltz observed (1979. 158) that prior to World War I, trade as
a percentage of gross national product (GNP) was higher than it is today. Furthermore, the
transnational corporation (TNC) had at least an early relative established in 1600, the English East
India Company. Finally, the first international organization can be traced to the Delian League of the
ancient Greek city-states.
These dated facts underscore why it is useful to operationalize the Interdependence Approach
as a process with unique stages leading to more and more convergence, rather than as an
epiphenomenon specific to a particular time period. The behavioral variables that I have chosen to
define the Interdependence Approach can be scaled and, for this reason, have inductive value. Once
again, those variables pictured on Figure 1-2 are: (I) convergence in public policy toward a liberal,
global equilibrium, (2) the transformation of domestic regulatory problems into international collective
goods problems, and (3) the tendency of the market to restrain states acting multilaterally from
enforcing regulations that exceed a lowest common denominator (LCD). All three variables tend to
move together and are really just different aspects of the same Interdependence Approach. The most
powerful of the variables is convergence toward market liberalization for reasons which I will explain
Unfortunately, only two approaches in international relations deal specifically with the three
dependent variables of the Interdependence Approach: regime theory and transactionalism.
Transactionalism is far and away the most significant influence on this research, as should be evident
from the fact that transactions are the units of analysis, rather than the institutions upon which most
interdependence scholars dwell. Nevertheless, regime theory has also influenced me in very specific
Keohane and Nye (1977) first associated international regimes with interdependence. The
association is now so strong in the minds of most that few can examine the merits of either apart. Yet,
by operationalizing the Interdependence Approach without reference to regimes I hope that it is clear
that I intend to divorce this research from what at least many realists perceive to be a theoretical
muddle. Nevertheless, a specific element of regime theory has caused me to think deeply. Regimes
are theoretically a form of convergence. According to their most widely accepted definition,
international regimes are "sets of implicit or explicit principles, norms, rules, and decision-making
procedures around which actors' expectations converge in a given area of international relations"
(Krasner 1983, 2) (emphasis added). Convergence is the central variable in this study.
But, in practice, regime theory has stressed the process of convergence less than institutional
forms of cooperation. I treat the process of convergence much more seriously, because my focus is
upon transactions rather than institutions. Thus, in Chapter Four, we examine theories in international
finance (Arbitrage Pricing Theory) and, in Chapter Five, theories in international trade (Factor-Price
Equalization Theorem) that explicitly model convergence. The regime literature never considered these
more exacting explanations of convergence. Still. I owe an intellectual debt to regime scholars, who
first pointed out the importance of interdependence to the world community.
In addition, regime scholars were among the first to relate the provision of collective goods--
also called public goods--to interdependence, usually conceptualized as regimes. Regime scholars,
usually neo-realists, also interested me in the notion of international "hegemons" and hegemonic
decline. The grandfather of this literature, Charles Kindleberger (1973), argued that only a hegemon
could be counted on to pay the price of providing collective goods. Expanding on this work, Robert
Gilpin (1987) contends that regimes are likely to be underprovided when the power of hegemons
erodes. I argue that die more compelling explanation for the transformation of domestic regulatory
problems into international collective goods problems is the new mobility of capital and technological
However, more critical to the Interdependence Approach than regimes is transactionalism.' In
fact, the primary reason that I can develop the idea of convergence to a greater extent than other
interdependence scholars have is due to transactionalism's emphasis upon exchange as a measure of
integration. Like Deutsch, I do not deny the independent causal importance of institutions, principles,
and norms on interdependence. I just believe that it is most useful to concentrate upon transactions.
Gabriel Almond (1990) pointed out that the disadvantage of focusing upon institutions is that they vary
cross-nationally; that is, whereas the functions which die state performs are everywhere the same, the
institutions that they develop to perform those functions vary. In contrast, transactions occur in
discrete units that can be measured and compared meaningfully, wherever their location. For example,
$1 has the same value to a banker in Europe as it does to one in the US.
'The integration which Deutsch studied is simply die end-point toward which the process of
interdependence naturally leads. Hence, Deutsch's (1957) ideas should apply to my framework, not just
to European integration studies.
So, is the Interdependence Approach just transactionalism writ large? It is not, because I (1)
define the Interdependence Approach in terms of three processes--the most important leading to policy
convergence-which Deutsch never emphasized; and (2) limit my focus on transactions ostensibly to
capital (liquid and fixed) and technological innovation in information technologies. In contrast, Deutsch
never explicitly referred to technological innovation. Once more, he liked to consider all kinds of
transactions, which I do not take into account; for example, mail service, exchanges through media,
phone calls, exchange students, and tourism. Clearly, the Interdependence Approach contributes
theoretically and empirically to the field of international political economy.
One of the shortcomings of the old interdependence literature was its methodology. Over-
determination in regime theory, for example, is one of the reasons that the approach is in crisis. When
we agree with Keohane and Nye (1977) that since 1920 an oceans regime (x) has ensured the free
navigation of the seas (y), we are begging the question of what the absence of this regime (-x) would
have implied (-y) and whether the latter implication is even plausible. It would not be plausible to
insist, for example, that the absence of an oceans regime would have meant a complete inability to
navigate the seas. It might be reasonable, on the other hand, to argue that an absence of a regime
would have made navigation harder. But how do we know at what point the number of restrictions is
so great as to verify the absence of a regime?
No answer is necessarily better than another, a problem which has justifiably led many to
doubt the utility of interdependence theory. The methodology with which I have chosen to illuminate
the New Interdependence does not suffer from these same methodological problems. To be sure, I use
a case-study methodology, based upon historical analysis like Keohane and Nye do,4 and I employ
statistical techniques like correlations and cross-tabulations only in limited ways to support my case-
4This is because, first, the case-studies in which I am interested are simply too limited to permit
large-n analysis; and, second, trend data related even to these case-studies is often tenuous or unavailable.
study analysis; however, in contrast to Keohane and Nye's work, I preserve variance on the x and y-
axes of my analysis by distributing the case-studies that I consider along the curve of Figure 1-4. All
of the case-studies represent issue-areas, or primary functional areas in which the state is involved.
Whether I put convergence or either of my two other dependent processes on the y-axis makes no
difference, since all three processes basically move together. The processes are just three aspects of
the same Interdependence Approach. Monetary policy and environmental policy are at opposite ends of
die curve because capital is the most liquid and innovative in the former instance and die least, in the
latter. I will explore these differences and the other case-studies in more detail below. For now, die
point that I want to emphasize is that the Interdependence Approach succeeds where the old
interdependence literature failed by maintaining case-study variability.
The approach I use to analyze my case-studies comes from comparative politics: comparative
historical analysis (Skocpol 1979). Alternatively, Lijphart (1971; 1975) calls the sane basic approach
the "controlled comparison" method; Harry Eckstein (1975). the "disciplined-configurative" mode of
analysis; and Alexander George (1974, 1979), the "method of structured, focused comparison." The
advantage of this method is that it allows the researcher to structure his inquiry selectively around the
variables of interest to him or her and then to examine them cross-sectionally, cross-temporally, or
both. Theda Skocpol (1979, 36) explains that there are two primary ways of proceeding with this
method: (1) try to establish that a number of cases have variable y in common, as well as a set of x
variables or (2) contrast cases in which variable y and a set of x variables are present with cases in
which both are absent. In practice, she continues, it is best to combine these two comparative logics.
This is precisely what Ruth and David Collier (1991) do. They combine Adam Przeworski
and Henry Teune's (1970) "most similar" and "most different" systems approach into one research
design. I will do the same. The Colliers consider a set of eight Latin American countries that are
roughly matched on a number of variables and then proceed to analyze pairs of countries that are as
different as possible in specific issue-areas. The countries that I will consider are primarily
industrialized countries. Thus, in respect to their connmercial and political development, they are fairly
matched. This is good for convergence. In other respects, however, the industrialized countries that I
study are different. Most importantly, some are part of regional integration efforts (the EC) and others
are not. Once more, only one industrialized country is a hegemon in the classic sense that
Kindleberger (1973) used the term.
The hegemon is especially important to isolate in tie case-studies that follow, because I claim
that a primary reason that the New Interdependence represents a departure from the past is that now
even the most influential industrialized countries are subject to convergence. Of course, the hegemon
is the most important of these powers that could be so affected. In addition, by isolating die hegemon
we can determine what its special influence is, if any, on providing collective goods. I mentioned
above that, whereas neo-realists contend that the explanation for the break down of many multilateral
coordination efforts (i.e. Bretton Woods) is declining hegemony, I argue that it is capital mobility and
technological innovation. One way to segregate the hegemon's influence is to consider it in a forum
where it is active and one where it is not; i.e. in monetary policy, the G-7 and EC.
The third important difference among industrialized countries for my purposes Ias to do with
their state-society relationship. John Zysman and (1983) and Jeffrey Hart (1992) point out that most
European countries and Japan have pursued corporatist strategies to deal with the challenges of the
international political economy. In contrast, the US and Great Britain retain fairly liberal economies,
lacking strategic industrial and trade policies.
The last important distinction that I make in this dissertation relates to the division scholars
have raised between "low" and "high" politics. I think that this division is artificial, but in order to
overcome the rift between structural-realists and the liberal school of political economists in research on
interdependence, I think that it is important for me to deal with the division on its own terms. Hence, I
want to evaluate at least one issue-area that scholars have traditionally associated with high politics and
another with low politics.
The only issue-area which scholars universally recognize as part of high politics is national
defense. Thus, it represents a "crucial case" study, according to Eckstein's terminology (1975). If I
can show the New Interdependence in the context of national defense, then it is likely that it will also
be apparent in issue-areas in low politics to a more considerable extent. I cannot consider national
defense as a whole, however, without spending my entire dissertation upon it. Hence, I will consider
one aspect of national defense: weapons procurement policy.
The last methodological point that I should make before moving on to discuss the case-studies,
refers to the obvious gap in Figure 1-4 between weapons procurement policy and environmental policy.
That is, I consider no case-studies which fall into the low innovation/high capital mobility block of the
x-axis. This is because this block represents the null set. I cannot imagine any function which the state
performs that is characterized by low innovation and high capital mobility. This makes sense, too,
because of my previous argument that innovation and capital mobility tend to co-vary. In most
situations, it is difficult to have high capital mobility, premised upon low innovation.
Weapons procurement policy. Consistent with my expectation, we shall see that the New
Interdependence poses a thorny dilenmia for North American Treaty Organization (NATO) members.
Although NATO members want to ensure their defensive autonomy (and, hence, own national security)
and provide for international security (the collective good), they cannot do both. In an age of "smart"
or technologically advanced weapons, defensive autonomy cannot be achieved fully even by the largest
industrialized countries. Moreover, the effort to be self-sufficient requires states to spend millions
trying to keep up with each other in similar R&D and production efforts. Much duplication occurs,
which, if eliminated and invested in NATO, would make NATO much more effective. Instead, an
increasing gap opened with the Warsaw Pact in the 1980s. NATO members could buy fewer and
fewer weapons with the same defense budget, causing them to "structurally disarm" and NATO to lose
some of its deterrent effect.
NATO and, separately, Europe tried to respond in earnest to this problem in the mid-1970s,
but with little overall effect at tie systems level of weapons procurement. This is because the
standardization programs which they tried preserved state autonomy. For example, co-production and
co-licensing agreements that guaranteed domestic defense contractors business, also eliminated many of
the economies-of-scale that cooperation was supposed to bring about. In addition, the end of the Cold
War reduced considerably the pressure on NATO to produce weapons efficiently. In the 1990s, most
NATO members are simply permitting structural disarmament to take place unimpeded.
While this goes on at the systems level of weapons procurement, though, the market is
undermining the defensive autonomy of industrialized countries at the components level of weapons
systems. The state's inattention and, in many cases, inability to counteract this interdependence does
not square with the "realistic" understanding of world politics. However, it does square with the
Interdependence Approach. Standardization at die components level is proceeding rapidly, because
defense contractors must purchase tie lowest cost and highest quality parts that they can. Eventually
this trend, together with the increasing multinationalization of defense contractors, must take-over
standardization at the systems level, too.
Monetary policy. The next case-study which I consider is monetary policy, which it is fair to
say has traditionally been treated as part of low politics. Although I think that this characterization
makes little sense now, I put monetary policy in this category to remain true to the structural-realist's
understanding of it. Not surprisingly, monetary policy deals with capital that is highly liquid and
financially innovative (see Figure 1-4). I have discussed its mobility already by referring to the
tremendous growth in the eurocurrency and eurobond markets. Its innovativeness stems from
government regulation and the risks associated with cross-border transactions. The market is constantly
inventing unregulated capital instruments to get around regulated ones and trying to minimize its
vulnerability to exchange rate and interest rate fluctuations.
Overall, the combination of high innovation and capital mobility in monetary policy leads me
to expect that the New Interdependence will be the most obvious in this case-study. And this is
precisely so. Liquid capital began to bind small states as early as 1971, when Bretton Woods
collapsed. Beginning in the 1980s, though, exchange rate intervention, interest rate controls, and
money targeting on the part of even large central banks became less effective. Since 1971, states have
tried to restore a degree of international financial stability by coordinating currency values, but only in
fits and starts and based upon LCD agreements. The glaring exception to this rule is the regional
coordination effort in Europe, the European Monetary System (EMS). Even it nearly collapsed in
September 1992, however, when the unwillingness of states to coordinate their fiscal policies, in
addition to monetary policies, prompted massive speculation.
In general, monetary interdependence was advanced enough by the 1980s, that industrialized
states could not sustain both financial autonomy and international financial stability--the so-called
"compromise of embedded liberalism" (Ruggie 1982). The reason had nothing to do with the decline
of die hegemon, either. Rather, the reason was that technological innovation linked national financial
Trade policy. The third case-study that I consider belongs somewhere between high politics
and low politics, albeit it leans more toward low politics. There is no more consensus than this in the
literature about what kind of politics trade policy represents. What is clearer is that trade generally
deals with fixed capital, rather than liquid capital. Hence, it belongs in Figure 1-4, to the right of
monetary policy. Even so, since trade per se deals with goods that are often innovative and flow more
easily between borders than many of the fixed investments in weapons manufacturing facilities, it
belongs to the left of weapons procurement policy. Thus, there should be considerable convergence in
trade policies. In fact, there is.
Convergence toward commercial liberalization in international trade policy is evident in two
respects: (1) most states have liberalized their trade policies since the late-1970s--developing countries
because of their efforts to restructure national debt; and (2) factor-price equalization has occurred
among industrialized countries in keeping with neo-classical trade theory. Even industrialized countries
continue to cut trade barriers in (1) areas historically associated with trade and (2) within regional trade
groupings like the EC and NAFTA. Another phrase for this type of liberalization is "shallow
integration"--"shallow" because it does not embrace so-called "competition policies" like antitrust
policies and domestic industrial subsidies.
But competition policies, which are outside of the framework of the General Agreement on
Trade and Tariffs (GATT) and, now, the World Trade Organization (WTO) are the source of a great
deal of trade tension among industrialized countries in the 1990s. Because industrialized countries are
driven to obtain the monopoly rents that imperfect competition promises in the 1990s, they are adopting
so-called industrial targeting and strategic trade policies that are largely GATT-consistent. because they
are extra-GATT in character. The challenge for the future will be to harmonize these competition
policies with the trend toward increasing liberalization of tariff and non-tariff barriers. In addition,
states will have to decide on whether to standardize their health, labor, environmental, and human
rights expectations, since these also affect trade.
Environmental policy. The last case-study which I consider was difficult to choose. To make
my research agenda manageable, I had to eliminate two of the possible six case-studies I could have
examined. To my knowledge, there are just six basic functions which the state performs. Besides
national defense, monetary policy, trade policy, and environmental policy, they are energy policy and
social welfare policy.
To maintain the variance evident in Figure 1-4, the most obvious case-study that I should
eliminate is one that overlaps on the curve with another. Energy policy would have overlapped with
environmental policy, since energy policy deals with fixed investments that have not changed
dramatically since the 1970s. I eliminated energy policy, first, instead of environmental policy,
because much has already been written about energy policy and because environmental policy is a new
issue-area of growing importance to the state. This left me with a choice between social welfare policy
or environmental policy. I chose to eliminate social welfare policy for an obvious reason:
comparatively speaking, it has the least to do with international relations.
Scholars generally agree that environmental policy is a low politics concern. I have positioned
it far to the right on Figure 1-4, because, first, the capital that environmental policy deals with is fixed
in the sense that traded goods like hazardous waste are fixed and foreign investment in so-called
"pollution havens" like tie Mexican-Anmerican maquillidora corridor is fixed. Second. technological
imnovation has so far not been a big factor in environmental policy, overall. One reason is that so-
called commladan-and-control environmental regulations eliminated the incentive that firms have to
innovate in cutting their pollution emisssion levels throughout the manufacturing process. This situation
has only recently begun to change. Thus, in general, I would expect convergence to be the least in
environmental policy of any of the issue-areas that I have so far discussed. Again, what I expect holds
Consistent with the novel character of environmental problems and environmental regulation,
convergence in environmental policy is nuanced. First, although the new tendency is for states to adopt
environmental regulations that are market-friendly and consensual in nature, rather than rigid and
inefficient, another tendency affecting environmental policymaking since the 1970s is for the state to
regulate a broader variety of environmental problems. Whereas the first of these trends appears
consistent with the Interdependence Approach, the second does not. However, in reality, the regulation
of a broader variety of environmental problems reflects the artificially low level of environmental
regulation which states enforced until the 1970s. Until even the 1980s, we were not aware of some of
the most serious environmental problems fundamentally at odds with basic human needs. Either that,
or we could not imagine alternatives to them.
As this situation changed in the 1970s, industrialized countries first responded by passing
regulations that were insensitive to the market incentive structure. Then, in keeping with Hypothesis
One, as the costs of enforcing this regulation escalated and industries in the most heavily regulated
sectors became less competitive globally, the state was forced to look for market-oriented and
consensual regulatory tools to accomplish its new regulatory function.
Is there something dramatically different about the international political economy in the
1990s? There is. Three empirical processes accelerated in the 1980s, that began to bind even the most
influential industrialized countries. The most powerful of these is convergence in public policy toward
a liberal, global equilibrium. Increasingly, it is convergence around the market's preferred equilibrium
which dominates. This is because the tendency of many goods that states previously provided on their
own to become collective goods problems means that more and more problems can only be effectively
addressed with a multilateral approach. The trouble is, each state acting on its own has an incentive to
defect from multilateral cooperation, so that capital will flow its way.
This is not a trend limited to low politics, either. For example, states now have no choice but
to purchase weapons dependent upon parts sourced from around the world. They must set monetary
policies without resorting to exchange rate intervention, interest rate controls, and even money
targeting. Thus, their monetary policies and even fiscal policies compete with that of other states vying
for the same increasingly unregulated flow of capital.
Likewise, in international trade industrialized countries are now competitively pursuing the
initial monopoly rents that technological ilnovation leading to imperfect competition causes. Not even
the US can afford any longer to ignore this race. Last, there is increasing pressure for convergence in
conservation efforts to deal with multilateral environmental problems. The trouble is, the market
creates strong incentives for some states to be less than environmentally conscious.
For reasons like these, it appears that no less than a revolution is occurring in world politics.
I have called this empirical reality, wherein even the most influential industrialized countries are bound
to universal market imperatives, the New Interdependence. The theoretical approach I have taken to
understand interdependence I call simply the Interdependence Approach. Its heuristic contribution to
the field of international political economy is to take three variables never before concentrated upon to
measure nothing less than an awesome leap in the level of interdependence.
How different is the organization of the international political economy under the New
Interdependence from prior periods in world politics? If we accept that one of the primary functions of
the modern welfare state has been to prevent and counteract market failure and to remove some
activities from the marketplace, the importance of the shift away from these functions should not be
underestimated. Increasingly, die state looks more like a market-taker than regulator. In this kind of
"Brave New World" the choices that we are left-with are challenging, if not unsettling.
Indeed, unless we choose to decentralize and to use applied science, not as the end to
which human beings are to be made the means, but as the means to producing a race
of free individuals, we have only two alternatives to choose from: either a number of
national, militarized totalitarianisms, having as their root the terror of the atomic
bomb and as their consequence the destruction of civilization (or, if the warfare is
limited, the perpetuation of militarism); or else one supra-national totalitarianism.
called into existence by the social chaos resulting from rapid technological progress in
general and the atomic revolution in particular, and developing, under the need for
efficiency and stability, into the welfare-tyranny of Utopia. You pays your money
and you takes your choice.
(Aldous Huxley 1946, xiv)
Table 1-1: Country Stock Market Correlations with the World Stock Market
Sauce Morgan Stanley, 1977-1992.
Note Here, the "world market" is the average of all national stock markets.
It is only a hypothetical market.
Table 1-2: Growth in Unregulated Capital Markets, 1970-88 21
Gross Annual Absolute
Size Rate of Size % of World
Year $ bill. Growth $ bill. Bond Issues
1970 110 n.a. 3.0 65
1974 395 .20 2.1 31
1978 950 .22 14.1 41
1982 2146 .10 51.6 66
1984 2359 .04 79.5 74
1986 3632 .22 188.7 83
1988 4493 01 119.8 80
Source: Morgan Guaranty Trust Company of New York. 1970-90.
Table 1-3: Relative Size of the Eurocurrency Market 22
(billions of dollars)
End of Eurocurrency US Japan German
Period Market (Net) TDs TDs TDs
1965 17 141 42 30
1975 255 432 254 157
1980 755 739 677 286
1985 945* 901 1086 309
1990 n.a. 1305 2793 701
Source: International Monetary Fund, International Financial Statistics, 1965-90.
Note: TDs stand for time deposits; I have converted them into dollars using the
annual exchange rate listed in International Financial Statistics; stands
for net eurocurrency market figure for 1983.
Continuum of Interdependence
Zone of Convergence
Affects Small States
Affects Big States
Figure 1-1: Conceptualizing Interdependence
I Collective Goods LCD ,
I Problems Regulations
I -- tr
-Il-- -- -- -- -- -- -- -- -- -
L- Technological Innovation
Figure 1-2: General Framework for the Interdependence Approach
N1t : The thickness of lines indicates the importance of the causal relationship. Dashed
lines indicate a causal relationship that is not constant.
Zone of Convergence
Small States Large States
Low Politics High Politics
Figure 1-3: Eroding Autonomy of the State
Maximum -Monetary Policy (L)
Ce Trade Policy (L)
> Weapons Procurement Policy (H)
U0 Environmental Policy (L)
.o .'4 .4 -o o
5. i 0 |t
Figure 1-4: Case-study Variance
N~te: "L" stands for low politics and "H" stands for high politics.
STATE OF THE THEORY
Neither the modernists nor the traditionalists have an adequate framework for
understanding the politics of global interdependence. Modernists point correctly to the
fundamental changes now taking place, but they often assume without sufficient
analysis that advances in teclmology and increases in social and economic transactions
will lead to a new world in which states, and their control of force, will no longer be
important. Traditionalists are adept at showing flaws in the modernist vision by
pointing out how military interdependence continues, but find it very difficult
accurately to interpret today's multidimensional economic, social, and ecological
interdependence. (Keohane and Nye 1977, 4)
Time has only exacerbated the problem Keohane and Nye point out in this statement. For the
most pan, interdependence scholars and structural-realists talk past one another, one concentrating on
so-called "high politics" and the state, and another, on "low politics" and non-governmental actors.
Very often, the level of analysis each chooses is also different. Structural-realists prefer the system
level of analysis and interdependence scholars, the sub-system levels of analysis. Hence, although the
claims of interdependence scholars and structural-realists are not fully reconcilable, neither are they
always clearly at odds. After all, their claims are about different phenomena in world politics.
This analysis tries to move beyond the stalemate between interdependence scholars and
structural-realists by developing a new approach to interdependence which takes both high and low
politics into account and which explains a qualitatively new level of interdependence which has
heretofore not been adequately explored. The Interdependence Approach refers to a new theoretical
paradigm for conceptualizing interdependence, based upon three substantive processes:
1. convergence in public policy toward a liberal, global equilibrium
2. the transformation of domestic regulatory problems into international collective
3. the tendency of the market to restrain states acting multilaterally from enforcing
regulations that exceed a lowest common denominator (LCD).
Since these processes have advanced together to the point that the market can bind even large
industrialized states now in the area of high politics, I believe that a new, advanced stage of
interdependence has been reached that deserves exploration. 1 call this empirical stage of
interdependence the "New Interdependence." It represents a stage close enough to integration on the
convergence continuum shown in Figure 1-1 to begin to suggest the outlines of a radically new inter-
state system. Interest in the cause and character of this reshaping is what drives this research.
In this chapter, my objective is to emphasize how the Interdependence Approach relates to the
old interdependence literature by structuring it around my dependent and independent variables.
Second, I will propose the hypotheses in this chapter that I intend to pursue in each of my case-studies.
Last, I will present a general framework for the Interdependence Approach and show how it can be
useful in illuminating changes in the international political economy.
The Interdependence Approach
Deutsch argued in the 1950s (1978, 258-9) that the real difference between the interdependence
which existed in the early-1900s and interdependence during his generation had to do with the greater
"range" and "scope" of interdependence during Deutsch's time. That is. Deutsch argued that
interdependence in the 1960s more profoundly affected all of the goods, services, and activities it
influenced in 1913, and a larger array of new activities (greater scope). Second, he argued that the
costs and benefits of interdependence in the 1960s were larger than in 1913. Deutsch said that states
could not only help each other more in his time through programs like the Marshall Plan, but they
could also damage each other more with devices like the hydrogen bomb.
If "range" and "scope" were useful analytic tools, we could make the same argument about the
difference between interdependence since 1980, and the interdependence that Deutsch observed (1950-
79). But like many other terms in the old interdependence literature, Deutsch's are difficult to
operationalize. They do not fully capture interdependence, either, since Deutsch represents them with
static events, rather than processes. I have defined the Interdependence Approach explicitly in terms of
three processes, the first of which is the most important: convergence in public policy toward a liberal,
Convergence is a process wherein "social units start from diverse positions on some scale of
organization, sociopolitical structure, or culture, and then move toward some more common form on
the given dimension" (Inkeles 1975, 472). More generally, on a continuum of interdependence
extending from complete independence to full integration, convergence occurs on the right half of the
continuum as in Figure 1-1. It is more complete the further right that we move, since it eventually
embraces both small and big states and low and high politics. Ultimately, convergence leads to
In the post-World War II period, die most fundamental form of convergence has been the
closing (in absolute terms) productivity and income gap among developed countries. Gross domestic
product (GDP) per person employed in the US was about twice that of the major European countries in
the 1950s. In contrast, by the early 1980s, West Germany and France and, to a lesser extent, Japan
had GDPs per person about equivalent to the United States'. Data on patent applications and grants,
the percentage of youth attending college, R&D expenditures as a percentage of GNP, and the
percentage of trained scientists and engineers show a similar pattern (see Nelson in Heiduk and
Yamamura 1990), with the US and Japan the consistent outliers (the US because of its generally poor
performance and Japan, its excellent performance).
Nevertheless, from the last chapter we know that only one literature in international relations
relates interdependence to convergence: regime theory. In theory, regimes are a form of convergence,
whether it be in principles, norms, rules, or decision-making procedures (Krasner 1983). In practice,
regimes appear to be more like institutions than a process of convergence. In fact, the word
1In this dissertation, convergence is a process identical with "harmonization" and
"standardization." All of these words connote movement toward a single equilibrium. NATO prefers the
concept "standardization," though,; and the EC more frequently uses ite term "harmonization" than
convergence. In the latter context, Iarmonization generally deals with eliminating unfair cost advantages
in commerce for some member states. On the other hand, standardization usually arises in relation to
NATO's Rationalization, Standardization, and Interoperability (RSI) Program, which is stimulating the
development of "NATO" weapons. I will discuss this more in the weapons procurement case-study.
"convergence" never arises as a substantial theoretical construct in any of the major areas to which
regime theory has been applied.
In contrast, two theories outside of the field of international relations nake the process of
convergence a central concern: the Factor-Price Equalization (FPE) Theorem in economics and the
Arbitrage Pricing Theory (APT) in finance. According to the Factor-Price Equalization Theorem of
the neo-classical Hecksher-Ohlin Trade Model, where there is perfect competition and no transportation
costs, trade between two countries completely equalizes the factor prices in both. Likewise, according
to the APT, barring government impediments to capital movements, foreign exchange fluctuations, and
individual "irrationality," capital markets will become perfectly integrated and asset prices will be
perfectly correlated, regardless of the location at which capital is traded. Table 1-1 illuminated die
dramatic extent to which this is really occurring. But convergence is also occurring in many other
areas, besides finance and economics; for example, in law, weapons procurement policy, social welfare
This is despite the obvious invalidity of an assumption at least the economics and finance
theories I have mentioned rest upon: that the state will not interfere in the market. On the contrary,
the state jealously guards its autonomy from the market and often asserts a "right" to delink itself from
world processes of convergence at a political level. In the words of Edmund Dell:
It is of the nature of the state that its principal responsibility, in economic matters as
in military, should be self-defense. There is nothing in economic theory that could or
should persuade a prudent govenmnent to abandon that priority. .. [This leads tol
one of the few reliable economic laws .. the law of instability in international
economic relations. Governments will strive for national economic security.
They will not be dissuaded by the thought that in an interdependent world such an
objective can only be chimerical. They will shrug-off such defeatist teaching. They
will do their duty and try. (1987, 13)
Nevertheless, in die 1990s, states often fail. The market gives powerful incentives for states
to advocate the equilibrium for public policy which it prefers: one characterized by fiscal discipline
and oriented around the four capitalist criteria--cost, quality, performance, and delivery time. This is
not a regulatory agenda, such as we associate with the Keynesian state. Rather, it is the basis for a
competition state, maneuvering strategically for connnercial advantages over other states.
The worst penalty that a state faces for not being competitive is capital flight. Excessive
regulations encourage capital to look elsewhere for higher rates of return. Because of modern
technological advances, all else being equal, capital can evaluate and take advantage of differential rates
of return, worldwide. Therefore, these rates of return equalize across states over time. Again, Table
1-1 made this point with regard to national stock markets.
Likewise, literature in finance (Grubel 1968; Levy and Sarnat 1970: Agmon 1972; Solnik
1974; Grubel and Fadner 1971) underscores the attractiveness of international portfolio diversification
to investors. Benefits include, not only higher rates of return, but also risk-reduction. In comparative
terms, Donald Lessard (1974) judges the risk-reduction possible from cross-national diversification even
greater than from cross-industrial diversification within the same country.
Theoretically, the same benefits should exist for those interested in direct foreign investment
(DFI). The literature suggests (see Adler and Dumas 1983; Collins 1990) that DFI results from
imperfections or segmentation in either the goods or money market. In reality, findings suggest
(Collins 1990, 284) that, in net terms, stockholders benefit from a company's decision to diversify only
into select countries. Diversification away from purely domestic operations to developed-country
operations, results in higher rates of return (279). But diversification into investments in developing-
countries reduces overall returns. This is why most DFI flows between industrialized countries.
This finding of bias in capital mobility is not surprising, since not "all else is equal" in capital
markets. In fact, the finding of bias relates to many others in finance (Allan 1982; Adler and Dumas
1983; Errunza 1983; Van Agtmael and Errunza 1982) which show that the correlations between the
equity markets of the US and those of developing states are much lower than those between the US and
developed states. For example, Van Agtmael and Errunza (1982) report correlations of .64 between
the US and Canada in 1976 and 1980 and -.033 between the US and Brazil during the same time
period. The conclusion we should draw is that the New Interdependence is a concept which applies
primarily to industrialized countries. Labor, land, infrastructure, and markets are different in every
country, but the comparisons are most dramatic between developed and developing countries.
In sum, the pressure on developed countries to minimize the inter-state spread in their
marginal rate of return on capital (liquid and, to a lesser extent, fixed) is substantial. In addition, states
have an incentive not to "overly" regulate capital, because regulations can make industry non-
competitive. In the final analysis, the market can even make public policy (i.e. exchange rate
intervention), itself, too costly for die state to enforce. For example, Germany abandoned a 10%
withholding tax on interest income after paying dearly in capital flight for its regulation effort. These
sorts of situations are the more likely, the further right that we move on the continuum of
It is this movement which separates the New Interdependence from die old, too. In the 1980s,
we crossed a critical point on the interdependence continuum, such that smaller economic powers and
large ones and low and high politics are all the victim of convergence, now. Of course, smaller
economic powers are more subject to it than big ones,. and low politics more than high politics. As
we shall see in the case-studies that follow this chapter, convergence is occurring not just in economics
and finance now, but in a non-random representative sample of issue-areas and among a non-random
representative sample of states. This is a dramatic departure even from the 1970s, when economically
influential industrialized countries could still basically chart their own course in most issue-areas.
Thus, the first hypothesis and corollary that I will examine in my case-studies deals with this change.
6Keohane and Nye were the first to point out that interdependence can be asymmetric. In fact,
they distinguished (1977, 12-19) two genres of asymmetrical interdependence: "sensitivity" and
"vulnerability." According to them, a state is "sensitive" to another when it rapidly experiences the costs
of another's change in policy before it has the chance to adjust to the new environment. In this sense, the
state sensitive to another is asymmetrically interdependent. A state "vulnerable" to another finds it difficult
to adjust even over the long-term to a new, externally-imposed order. The costs of adjustment are simply
too exorbitant for it to pay.
Structural realists like Waltz insist (1979) that, because interdependence can be asymmetrical.
interdependence really does not exist. Waltz insists, for example, that "|al world in which a few states
can take care of themselves quite well and most states cannot hope to do so is scarcely an interdependent
one" (1979, 159). But this argument does not seem to be a particularly damaging one. Just like structural-
realists refer to states as being "very" powerful or "somewhat" powerful, it seems reasonable to refer to
states as being "very" interdependent or "somewhat" interdependent. The one important theme Waltz and
even dependency theorists leaves us with, however, is that interdependence in the international political
economy can always increase without affecting underlying power relationships.
Hypothesis One: Every state and every issue-area, whether in high or low politics, is subject
to a pressure to converge toward a global equilibrium which the market prefers: commercial
Corollary One: Convergence more substantially influences countries with relatively small
national incomes and issue-areas in which capital is highly liquid (usually in low politics).
In keeping with this hypothesis and corollary, I will demonstrate below that even in the
"critical" case study on national defense, the pressure on industrialized countries to liberalize the
international sourcing of their parts and equipment is severe. The reason is that weapons have become
technologically more advanced, necessitating greater economies-of-scale for their production than even
die US market can deliver. States must collaborate in weapons procurement to obtain these economies-
of-scale or face structural disarmament, a term which stands for the decreasing weapons a constant and
even increasing military budget can supply without an offsetting increase in the quality of the weapons.
Smaller military powers within NATO and non-aligned countries have experienced this
problem the most substantially, as Corollary One predicted. In contrast to other case-studies, though,
the conclusion of the Cold War has allowed NATO the luxury of inefficient weapons procurement.
NATO members go on wasting millions on unnecessary R&D and production efforts and its defensive
readiness suffers, as a result. Nonetheless, even as ite state continues to protect its self-sufficiency at
this level, at tde components level of weapons systems convergence is accelerating rapidly. Ultimately,
convergence at this level will overtake regulation of interdependence at die systems level of weapons
This difference between the system level and sub-system level is not so obvious in monetary
policy. In keeping with Corollary One, far more dramatic convergence is occurring in this issue-area
at both the system and sub-system level than in any other. This is because capital is the most liquid on
the currency exchange market and in domestic banking operations. We shall see that this liquidity
enables the market to punish those states which venture away from realistic exchange rates and
disciplined fiscal policies. Capital in the monetary policy issue-area flows in the direction which is
least regulated and has the highest return, thereby encouraging financial liberalization in countries that
would otherwise starve for capital. The result is two-fold: (I) states are being forced to liberalize their
markets to a similar extent and (2) the standard for liberalization is being liberalized itself. I call this
shift in monetary policy, convergence toward "instrumental ineffectiveness."
Collective Goods Problems
The second process which makes up ile Interdependence Approach has to do with the
expansion of collective goods problems.' The expansion is occurring because innovation and capital
mobility are (1) driving many goods that states formerly considered it their autonomous duty to provide
to become collective goods and (2) creating many new problems that only the world community can
effectively address. What are some of these new problems? They usually relate to the so-called
"negative externalities" that the market generates; for example, pollution and proliferation--proliferation
because, as we shall see in Chapter Three, pressure on the state to expand trade makes controls on
sensitive military technologies difficult to maintain.
Environmental problems, like many new collective goods problems, only came on to the
international political agenda in the 1970s and 1980s. Scientists, for example, did not discover the
"hole" in the ozone layer until the early-1980s. Once more, the problem of cross-border flows in
hazardous waste only began to touch developing countries in die late-1980s (see chapter six). The list
could go on, but the point is the sane: many problems which we now call collective goods problems,
were not even recognized as problems until recently.
Even more important for my purposes than new negative externalities, however, are older
problems that the state formerly regulated by itself, but only recently gave up as collective goods
problems. This foregone authority is a more significant limitation upon the state than simply depriving
the state of new areas to regulate. This is especially the case, since collective goods which formerly
were provided by the state tend to focus not on negative externalities, but regulations upon capital. For
7Collective goods problems are characterized by (1) "jointness of supply" or "nonrivalness of
consumption" and (2) "nonexclusion." In other words, a state must be able to consume a given product
without diminishing the quantity available to another (nonrivalness); and every state must be able to enjoy
the product, whether or not the state helped to "pay" for it (nonexclusion).
example, the maintenance of international financial stability is now a collective goods problem. Such
stability would "simultaneously benefit all states" in a way that "consumption could not degrade,"
whether or not the states paid part of the cost of enforcing the regulation. This is the case, so long as
some minimum number of states cooperated to enforce the regulation (i.e. a universal reserve
requirement) and make it meaningful.
Hypothesis Two: Goods that states formerly provided on their own are being substantially
transformed into collective goods problems.
I will demonstrate tlat. in fact, domestic security is becoming a collective goods problem,
because the weapons procurement needs of NATO and EC members can be met only as a result of
multilateral cooperation in high-technology areas. In addition, international financial stability now
depends on multilateral collaboration in monetary and fiscal policymaking. For that matter, a liberal
trade order depends increasingly on restraint in the competitive industrialization policies to which
industrialized countries have resorted. Last, many problems to which environmental regulations have
been addressed have always been collective goods problems, hut there are a number of problems like
solid waste management and biodiversity conservation that the state once tried to regulate on its own.
Now, these problems have also become collective goods problems.
Lowest Common Denominator Agreements
Related, but independent of the tendency of domestic goods to be transformed into collective
goods problems, is the new tendency of public policy coordination to center on a lowest common
denominator (LCD). The opposite of a LCD regulation is an effective regulation. Effective
multilateral regulations would plug the flow of capital to the most liberal states (in a regulatory sense),
because they would provide for a level playing field; that is, they would ensure that every state was
equally liberal (or illiberal); thus, capital would have no incentive to migrate.
The trouble is, each state acting on its own has an incentive to defect from the multilateral
equilibrium, so that capital will flow its way. "The consequence is that the multilateral arrangement
tends to be cast in terms of least common denominators ILCDs] so far as concessions are concerned,
and most elevated standards so far as demands are concerned" (Rubin in Rubin and Hufbauer 1983,
15). In general, rather than agree to and enforce effective multilateral regulations, this approach saves
face and costs little. The permissive multilateral agreements which result, I call LCDs. They represent
hardly more than symbolic steps to address problems, centered on the response least costly for the most
powerful actors involved to take. At the same time, LCDs have the advantage of underscoring the
signatories' normative conviction that they should cooperate to resolve a given problem.
This is a strategy consistent with the new competitiveness of the state. The imperfect
competition that technological innovation prompts by stimulating investment and trade in R&D-intensive
products encourages the state to offensively pursue fleeting commercial advantages. The powerful
temptation states now have not to effectively regulate capital is no better illustrated than by the $4,493
billion (in 1988 gross terms) Euromarket. In the words of some finance scholars:
S. imposing reserve requirements presents major practical difficulties. In order for
them to be effective, they would have to be universally adopted. If just one country
refused, the Euromarket would relocate to that country and carry on business as usual.
Many countries do not have domestic reserve requirements and would be reluctant to
impose a control on the international pan of their banking system that they have not
tried in their domestic sector. (George and Giddy 1983, 31)
In short, every state would prefer to be a "free rider"' on the regulations of every other. The
regulation of capital or, for that matter, the conservation of biodiversity has become a prime instance of
what Garret Hardin (1968) calls the "tragedy of the commons": although it may be in the collective
interest of the world community to conserve biodiversity, it may not be in the individual state's interest
to do so. Brazil, for example, may prefer to cut down the Amazon Jungle, rather than to preserve
species. In general, the tragedy of the commons becomes more insoluble as convergence tightens
around a multilateral equilibrium--undermining the argument of some functionalists that successful
multilateral coordination in the past necessarily leads to more in the future.
To explain, consider a world in which every one of 180 states has a different regulation on
airline companies. In this setting, the gain to a state that reduces its regulations to a point less
8A free rider will be understood throughout this dissertation in its classic sense: a free rider is
any state which benefits from the initiative of another state, but does not help pay for it. In essence, it
free-loads on the other state.
restrictive than all other states, so that it might attract airline carriers, might not be worth the costs in
lost lives, lost tax dollars, etc. In contrast, the gain to a state of reducing its regulations one notch
below a standard agreed to by 180 states might be substantial. Great Britain, for example, has lesser
regulations on airline traffic than the rest of the EC's hubs, which is why it is cheapest for Americans
to pass through London on their way to the continent. Great Britain is unlikely to give up this
advantage, except insofar as something else that it values more highly is threatened, e.g. its
membership in the EC.'
Not surprisingly, LCDs are more prevalent now that the number of actors in inter-state politics
that matter for reaching meaningful multilateral agreements has increased. This includes state and non-
state actors. David Hume was one of the first scholars to point out the hazard of multiple actors for
reaching a multilateral equilibrium. As early as tle eighteenth century, he wrote:
.. two neighbors may agree to drain a meadow, which they possess in conmion,
because it is easy for them to know each other's mind; and each must perceive that
the immediate consequences of his failing in his part is the abandoning of the whole
project. But it is very difficult, and indeed impossible, that a thousand persons should
agree in any such action; it being difficult for them to concert so complicated a
design, and still more difficult for them to execute it; while each seeks a pretext to
free himself of the trouble and expense, and would lay the whole burden on others. .
(cited in Frohlich 1971, 4)
More recently, neo-realists have sounded this theme, too. They contend that international
coordination is always more problematic in a multi-polar and multi-actor world (Gilpin 1987). Even
Duncan Snidal (1985, 611), a critic of neo-realist literature, admits that N-person game theory
demonstrates time and again that when the power of individual actors and coalitions of intermediate size
is small, coordination is the most viable. In addition, Bruno Frey (in Vaubel and Willett 1991, 15)
demonstrates that moving from the Group of Five (G-5) countries (US, U.K., Germany, France, and
91n general, homogenous cultural and economic groupings are likely to be the most effective at
enforcing a multilateral equilibrium. But since these groups are primarily regional, they are likely to leave
out other powerful economic competitors, ie. Japan and the US. As such, capital will still have the ability
to penalize non-competitiveness. Furthermore, groupings, themselves, are likely to continue experiencing
centrifugal forces, such as my airlines example, the near-failure of the European Monetary System, and
trade and foreign policy differences in the EC demonstrate (New York Times 1 July 1993, A:4).
Japan) to the Group of Seven (G-7) countries (add Canada and Italy) to die Group of Ten (G-10)
countries (add Belgium, the Netherlands, and Sweden) and, finally, to the OECD (including thirteen
more countries and, now, Mexico) dramatically increases the marginal cost curve of coordination and
diminishes the amount of benefits that each country can receive from coordination. Overall, since a
multilateral equilibrium requires more parties to make it effective and faces severe free rider problems,
I add a fourth hypothesis.
Hypothesis Three: Public policy coordination designed to regulate the market is increasingly
likely to center around agreement on a lowest common denominator (LCD).
Indeed, I will show that this is so in every issue-area that I consider. In weapons procurement
policy, LCDs are obvious in two principle areas: (1) NATO's standardization program and (2)
CoCom's export restraints on sensitive technology. NATO's standardization program has produced
little results, overall, at the systems level of weapons procurement, because designed into the program
is an inconsistent insistence on self-sufficiency. Through mechanisms like co-production and co-
licensing agreements, many of the potential gains from standardization at the systems level are lost. In
addition, the NATO Projects which result meet only LCD performance expectations.
As for CoCom, it has been divided into two camps, one more restrictive than the other. The
less restrictive camp--the free riders--have spoiled export controls for the more restrictive group.
Hence, overall, CoCom has had little effect in limiting the proliferation of sensitive military technology
to the East. But both CoCom and NATO's standardization program have served a symbolic function,
if nothing else.
Likewise, in monetary policy industrialized states were satisfied to agree to a series of ad hoc
accords following Bretton Woods (i.e. the Jamaica Agreement, the Plaza Agreement and Louvre
Accord) that merely ratified the monetary situation which the market had already created or that
managed to deceive the market for, at best, a few weeks. The agreements were more symbols of
resolve and a concern that the value of the dollar was "inequitable" than effective regulatory tools.
Also, in environmental policy the vast majority of environmental agreements are LCDs. But,
consistent with the exceptional character of environmental policymaking, there are a few, notable
exceptions to this trend, involving highly publicized, potential or actual environmental disasters (e.g.
the Bhopal tragedy) that stimulate public concern and even moral outrage.
Framework for the Interdependence Approach
In sum, the three processes which make up the Interdependence Approach are diagrammed in
Figure 2-1. They each represent different aspects of the same revolution in world politics. To the
extent that convergence increases, collective goods problems and LCD regulations generally proliferate,
as well. This is because convergence around a multilateral equilibrium (e.g. a standard regulation for
airline companies-see supra) ironically contributes to a free rider problem and makes certain defections
from the equilibrium more likely. Second, convergence toward a universal market standard
collectivizes problems that states could formerly address on their own.
But there are some environmental treaties that violate my expectation about the trend toward
LCDs even when the prevalent convergence appears to be toward more liberalization; for example, the
Montreal Protocol. Whereas the dominant tendency in environmental policy is now for the state to
liberalize the regulation of a greater variety of environmental problems, some effective multilateral
regulations are being negotiated. These agreements are not LCDs; and they help to resolve
international collective goods problems.
Examining in reverse the linkage of collective goods problems and LCDs with convergence
shows that causality is not bi-directional within die Interdependence Approach; in other words, more
collective goods problems and LCD regulations do not necessarily lead to more convergence. For
example, the fact that more collective goods problems exist has no necessary bearing on whether or not
states will decide to solve them by agreeing to similar policies (convergence). These problems can
simply persist. On the other hand, convergence necessarily makes formerly national problems,
international ones that require collective action to resolve. Thus, where a state could formerly restrict
the proliferation of defense technologies by regulating its armed forces and defense contractors, now
the defense contractors are multinational corporations with multinational suppliers that can only be
effectively regulated multilaterally.
In addition, whereas convergence in the expected direction toward commercial liberalization
generally leads (with some notable exceptions like the Montreal Protocol) to LCDs, because states
prefer to maintain a competitive advantage and to leave regulation to another, it is not obvious that
LCD agreements necessarily prompt further convergence. Usually. LCDs simply ratify post hoc a
nominal public policy convergence that already exists. LCDs rarely inspire new regulatory steps.
Agreements that would inspire new steps would probably not be meaningless agreements; thus, they
would not be LCDs.
Thus, whereas convergence toward commercial liberalization always drives collective goods
problems and usually leads to LCD agreements, vice-versa is not necessarily the case. Moreover,
convergence in the opposite direction front what Hypothesis One predicts, i.e. toward more effective
regulation that is not market-friendly, definitely leads to effective regulation, not to LCDs, and to the
resolution of collective goods problems. What causes all three dependent variables of the
Interdependence Approach to increase are the factors to which I turn next.
The independent variables which I believe drive interdependence as I have operationalized it
are the following: (1) the mobility of capital and (2) technological innovation. These variables tap the
intuition of many scholars (Kuznets 1966; Heiduk and Yamamura 1990; Cerny 1990: Stopford and
Strange 1991) about what makes contemporary interdependence unique. In general, technological
innovation in this analysis refers to "not only technical changes in production processes and equipment,
but also the discovery of new resources, the development of new types of organization, and new
methods of management" (Schumpeter cited in Weber and Handfield-Jones 1954. 115).
The primary innovations which are of interest to me are in information technologies (IT), i.e.
computers, telecommunications, software, and micro-electronics. Information technologies are the only
one of five generic technologies that emerged in the 1970s and 1980s (including information
technology, biotechnology, materials technology, energy technology, and space technology) that a
special report of the OECD says are inspiring an on-going technological revolution. The report states
that information technologies are having "such pervasive effects on the economy as a whole that they"
are changing "the 'style' of production and management throughout the system" (Freeman in OECD
The diffusion of information technologies was evident in the 1970s, "but they have become
more accentuated in the 1980s" (UNCTC 1988, 18). Diffusion, not incidentally, is a necessary element
of innovation. Many inventions are never widely adopted; hence, they have little affect on the
international political economy and, thus, do not rise to the level of an innovation (see Van Duijn 1983,
175). It is important to keep in mind this distinction between inventions and innovations, as we begin
As for the "mobility of capital," it is a phrase which refers in the following chapters to the
ability of both liquid and fixed assets to change location. For obvious reasons, liquid capital is likely to
be more mobile than fixed capital. In general, capital can become more mobile under two conditions:
(1) if it has a more conductive medium through which to move and (2) if it can take advantage of a
substantial cost-saving or profit-increasing opportunity. Technological innovation especially in
information technologies (IT) creates the medium for capital mobility. Entrepreneurs do the rest.
Therefore, under many (but not all) circumstances the mobility of capital and my secondary
independent variable, technological innovation, co-vary. In fact, the primary way in which
technological innovation affects the process of convergence is by facilitating the mobility of capital and
diffsing services and R&D. The important independent effect on interdependence that technological
innovation can have is by making R&D more important to firms and states that want to maintain a
competitive edge in the global market. This R&D has become more expensive with time, encouraging
firms to join in multinational consortia to develop and market new products. Hence, in the automobile
industry, for example, competition is no longer centered around cross-national competitors; for
example, GM-Ford-Chrysler, Toyota-Nissan-Honda-Mazda, and Volkswagen-Mercedes-BMW. It is
also between international consortia; for example, GM-Opel-Saab-Isuzu-Suzuki and Ford-Mazda-Jaguar-
Even so, not all technological innovations influence the mobility of capital. Clearly, some
innovation never lubricates capital. The flip-side of this relationship, the extent to which capital is
necessary for innovation, is disputed in the innovation literature. For example, Van Duijn doubts that
finance acts as much of a constraint at all on innovation. He writes,
Financial constraints rarely form a bottleneck factor in the innovation process.
Research and development have to be financed, of course, but many innovations reach
the market with only a moderate financial input. The financial requirements for
innovation have certainly increased over time, but here, too, the priority of market
demand reveals itself: it is recognition of demand potential for the innovation to be
developed which induces financiers to back a project. (1983, 130)
In contrast, Arnold Heertje contends,
If no venture capital market is available, many possible technical breakthroughs are
forced to die before they have a chance to prove their commercial profitability. (1988,
To my knowledge, the only reliable empirical research which has been conducted on this
subject in the 1990s is surmnarized on Table 2-1. Table 2-1 is based upon 1977 company R&D
expenditures from the Federal Trade Conmmission's Line of Business Survey and upon the total number
of innovations and patented inventions between June 1976 and March 1977 (for methodological details
see Acs and Audretsch 1991, 4-10). Most significantly, what it shows is a fairly strong relationship
between company R&D expenditures and total innovations (.75). This would suggest that Herrtje's
conclusion is the most accurate one. Clearly, scarcity of ideas can play an independent role in the
innovation process. But rising R&D costs and the growing importance of worldwide market access
mean that even large TNCs can no longer go it alone. McDonnell Douglass, for example, may go
bankrupt now that it cannot find a suitable partner to develop a new commercial jet, the cost of which
exceeds its net value.
The second important finding which Table 2-1 returns our attention is that innovation and
invention are not ite same thing. The correlation Table 2-1 shows between company R&D
expenditures and patents is just .44 and between total innovations and patents, only .47. The reason is
that, whereas the relationship between firm size, innovative input, and R&D expenditures is roughly
proportional, that between firm size and patent activity is less than proportional (Scherer in Acs and
Audretsch 1991). In some cases, the relationship between firm size and R&D expenditures is more
than proportional (ld.). Therefore, smaller firms may have an advantage in patenting products and
services, but many of these inventions are not widely adopted, probably because small firms do not
have the capital to manufacture and market them effectively. Hence, they never become innovations.
This summarizes the positive relationship between capital mobility and technological
imlovation, but, some might ask: What effect would a decrease in either capital mobility or
technological innovation have on die three dependent variables of the Interdependence Approach?
Fortunately, this has not occurred in die past few decades, so I do not need to spend much effort
speculating about this possibility. However, if the mobility of capital was to contract, it is likely that
divergence from the multilateral equilibrium for public policy would result. States that could
effectively regulate capital because of segmented markets would reassert their authority; we would
move back in the direction of the Keynesian regulatory state. However, my expectation is die opposite,
as the next hypothesis makes clear.
Hypothesis Four: As the mobility of capital increases, die three dependent variables of the
Interdependence Approach will necessarily increase. And as technological innovation
continues, convergence is likely to increase.
In each of the case-studies that follow, more capital mobility and technological innovation
promote interdependence. In the chapter on weapons procurement policy, innovation in information
technologies is making weapons "smarter" (more technological advanced) and, therefore more costly.
Countries are being forced to share the burden of R&D. Once more, capital is becoming more mobile
in the area of defense, because of two trends: (1) defense contractors can now search globally for the
best quality and lowest cost parts; and (2) to obtain economies-of-scale and access to new technologies
defense contractors are shifting toward New Forms of Direct Foreign Investment (NDFI), i.e. teaming,
subcontracting, joint ventures, consortium, acquisition, and stock options. NDFI is a form of the new
mobility of capital which is standardizing weapons from the components level up.
Second, in the case-study on monetary policy, innovations in computers and
telecommunications make fixed exchange rate systems less sustainable. Innovations in information
technologies even make fiscal undiscipline very costly, because they allow the market instantaneously to
draw money away from economies pursuing macro-economic policies not to their liking. This explains
why speculators were able to splinter the EMS in 1992. "Ultimately the whole structure of banking.
financial services and distribution is likely to be transformed as a result of the introduction of
'electronic funds transfer'" (Freeman in OECD 1987, 134).
Mobility of Capital
The mobility of capital is the primary factor which I contend accounts for the new level of
interdependence that we reached in the 1980s. It is not a variable upon which the old interdependence
literature reflects in any great detail. Even so, my use of it is consistent with a theory important to the
integration literature: transactionalism. In general, the father of transactionalism. Deutsch, argued
(1957, 58) that the best way to measure the integration of a community is to examine the volume of its
transactions or, in other words, exchanges between countries. The transactions that he counted
important were mail service, trade, exchanges through media like television and movies, phone calls,
exchange students, tourists, and international publications.
Although the only transactions I focus upon are financial and technological ones, I find
Deutsch's general emphasis upon exchange as a measure of interdependence more persuasive and
theoretically useful than the functionalist (Mitrany 1966, 1975) and neo-functionalists' (Haas 1958;
Haas and Schmitter 1966; Nye 1971) emphasis on the proliferation of international organizations (IOs).
Even functionalists (Mitrany 1966, 1975) admit that IOs are likely to vary cross-nationally, even when
the functions that they perform do not. That is, as Almond put it (1990), different institutions are
likely to develop to serve the same functional ends around the world. Transactions, in contrast, occur
in discrete units that can be measured and compared meaningfully, whatever their location.
Furthermore, we know in retrospect that it is not necessarily the case that institutions naturally
grow stronger and drive a community of states automatically toward integration like functionalists
insisted (Mitrany 1966; Soroos 1986). Rather, many supra-national institutions persist without states
transferring significant, new powers to them. Nor is it obviously the case that institutions are necessary
to bring about integration. The lack of any significant institutional framework for the North American
Free Trade Agreement (NAFTA) demonstrates this point. Finally, we know that significant integration
efforts have begun in developing areas that failed completely, never mind multiplied in functional
This is not to devalue the importance of institutions. Many criticized Deutsch (Mitrany 1975,
xiv) for this very thing, despite this statement:
integration requires, at the international level, some kind of organization, even
though it may be very loose. We put no credence in the old aphorism that among
friends a constitution is not necessary and among enemies it is of no avail. The area
of practicability lies in between. (1957, 6)
I agree and will make reference throughout this manuscript to IOs that are of fundamental
importance to the reconfiguration of world politics. Still, like Deutsch, I find it more useful to
emphasize the influence that transactions are having on the direction of the international political
economy than institutions.
Transactionalism even appears to have been a more important influence on the old
interdependence literature than functionalism. Consider Keohane and Nye's definition of "complex
interdependence" (1977). It has three main characteristics: (1) multiple channels connect societies
(including informal ties between governmental and nongovernmental elites and transnational
organizations); (2) there is no hierarchy among issue-areas (hence, military security does not
consistently dominate the agenda); and (3) governments do not use military force against other
governments within the area covered by the term. "Multiple channels connecting societies" is just
another way of describing transactions.' And Keohane and Nye's focus on pacific relations between
10This is die more evident when we dissect another term that Keohane and Nye added to the old
interdependence lexicon, "transnational relations." According to Keohane and Nye, transnational relations
has to do with interactions with non-governmental actors (NGOs) and governmental subunits -- in other
words, "all of world politics that is not taken into account by the state-centric paradigm" (383). The
"global interactions" (1971) that interested them are revealing for their transactional character, though:
interdependent communities matches Deutsch's characterization of integrated "security-communities."
The second independent factor which I evaluate in the case-studies, technological innovation, is
also not integral to the old interdependence literature. The one literature in international relations
which deals explicitly with imlovation as a variable is the long-cycle school. Within this school, the
most important cycle to which scholars relate innovation is the fifty-year Kondratieff cycle (1984)."
Most agree that technological revolutions or "'creative gales of destruction'" are at the heart of these
long waves (Freeman in OECD 1987). The most important innovations, according to Schunnpeter
(1939), have been:
1. cotton textiles and iron (1780s-1817)
2. railroads, steam, and steel (1840s-1875)
3. electricity, industrial chemistry, and die internal combustion engine (1890s-1920)
Rosecrance (1987) and Thompson (1990) argue that in the post-World War II period, the most
important innovations have been automobiles, plastics, electronics, atomic fission, aerospace, and
biotechnology. However, in the 1980s and 1990s, it is information technology (IT), in general, which
is revolutionizing production and management throughout the economy. Among the most important
characteristics of IT are that it is: (I) continuing to experience high rates of technical change, leading
to rapid obsolescence in succeeding "generations" of end-products; (2) contributing greater flexibility
and speed to manufacturing processes, making "just in time" inventory systems feasible, (3)
encouraging a rapid rate of product and process change and more intense technological competition in
other industries; (4) integrating the design, manufacture, procurement, sales, administration, and
technical service functions of enterprises; (5) merging the "manufacturing" economy with the "service"
movements of information, money, physical objects, people, or other tangible or intangible items across
iOther cycles include the following: (1) die Kitchin or inventory cycle, which lasts 3-5 years,
(2) the Juglar or investment cycle, which lasts 7-11 years, and (3) the Kuznets or building cycle, which
lasts 15-25 years. Schumpeter (1939, 173-74) contended that all of these cycles were linked (I Kondratieff
= 3 Kuznets cycles = 6 Juglars = 12 Kitchins). Others disagree.
economy; (6) contributing labor-saving, materials-saving, and energy-saving manufacturing techniques;
and (7) accelerating the international transfer of technology and capital (Freeman in OECD 1987, 132-
There is a seemingly constant flow of innovations in information technology. Figure 2-2
shows some of these innovations in IT and how they have increased with time. The opportunity which
IT has given modern TNCs to design "world" products and to manufacture them simultaneously in
multiple locations around the globe means that the stages of Raymond Vernon's product life-cycle
(1966) are no longer clearly demarcated. They are fused and sped-up.
Recall that according to life-cycle theory, products are (1) initially designed and manufactured
in developed countries for the home market, (2) exported to foreign markets, then manufactured in
foreign, developed countries, and, (3) manufactured in developing countries and exported back to the
home and other developed markets. In contrast, the current manufacturing process best implemented
by the Japanese circumvents this entire cycle by bringing goods to market and allowing them to mature
in the same location with the cost-saving advantage of automation. This process happens more rapidly
than ite former product life-cycle, too. Hence, states must work harder now at presenting attractive
investment opportunities that will keep the competitive advantage within their borders.
Computers are perhaps the best case in point of rapid innovation. Every two years new
models render computers with less memory, slower central processing units (CPUs), and, oftentimes,
inferior data storage systems. In this environment, it is much more important for a firm to be the
product innovator. Competitors may not have time to develop clones before the cloned products are
outdated by a second generation of new products. Companies that fail to innovate lose-out on the
initial windfall profits that product innovation can bring and often fall behind the front-runners without
any hope of catching-up.
A Contradiction, Not Compromise
In this environment of intense technological innovation leading to increased capital mobility, a
contradiction is emerging in nearly every functional area of the state between, on the one hand, the
state's desire to pursue autonomy and, on the other hand, its need to ensure international stability. I
argue that the state cannot achieve both at the same time in the 1990s. This is a quite different thesis
from John Ruggie's (1982), which states that the inter-state system was designed around a "compromise
of embedded liberalism." I believe that this so-called compromise was more of a latent defect.
Embedded liberalism could persist only so long as the economic liberalism that the inter-state
system was organized to facilitate remained adolescent. That is, embedded liberalism could be
sustained so long as economies remained segmented. When this condition changed because of
technological innovation and increasing transnational flows of capital, embedded liberalism failed.
Thus, embedded liberalism was not so much a compromise between, on die one hand, economic
autonomy, and, on the other hand, international financial stability, as it was a basic contradiction.
This contradiction between autonomy, on the one hand, and international stability, on the other
hand, is not restricted to economic affairs, either. The contradiction is an inherent part of almost every
functional area in which the state is involved, whether national defense, trade, or the environment. It
seems that the architects of the post-World War II inter-state system wanted to "have their cake and eat
it, too." At the same time as they agreed that states should work together cooperatively to ensure
international stability in a host of areas, they likewise insisted that the state should be autonomous
enough to set its own agenda. This is as evident in the Charter of tie United Nations as it is in GATT
and Bretton Woods. For. the UN Charter enshrined autonomy in the form of national sovereignty and
the right to self-defense with national means, as well as international stability in the form of a ius
cogens against unilateral aggression and the collective peace-keeping function of the Security Council.
Furthermore, to demonstrate that the contradiction between autonomy and international stability
is caused by capital mobility and technological innovation, not waning hegemony, I will illuminate the
contradiction in a context where the hegemon is excluded: the EC." Whereas neo-realists contend
that eroding American hegemony accounts for the collapse of Bretton Woods and the GATT-- the two
12By hegemonn" I mean what Kindleberger (1973) did: a global superpower like Pax Britannica
and Pax Americana.
principle institutions embodying Ruggie's "compromise of embedded liberalism" (1982)--1 hypothesize
that there were more rudimentary forces undermining these international institutions, mobile capital and
Hypothesis Five: A contradiction will be evident in every issue-area that I research between
the state's preference to preserve its policy autonomy and its desire to preserve international
stability in the given area.
Corollary Two: But capital mobility and technological innovation, not declining hegemony,
accounts for the failure of multilateral regulatory systems (e.g. Bretton Woods) to control the
The four basic contradictions that I will illuminate are between: (1) economic autonomy and
international financial stability. (2) self-sufficiency and efficiency in weapons procurement, (3) the
reciprocity norm of the GATT and its nondiscrimination norm, and (4) between environmental
autonomy and international environmental stability. Besides examining these contradictions, we shall
see that, in general, capital mobility and technological itmovation, not waning American hegemony,
best account for the failure of multilateral regulatory systems like the GATT, Bretton Woods, and the
International Tropical Timber Agreement.
Specifically, in trade policy, whereas neo-realists assert that eroding American hegemony
precipitated the collapse of Bretton Woods in 1971, it was (1) a decade after America's share in world
trade and production stabilized that the trade system came into jeopardy and (2) a decade after that
before America's willingness to absorb trade discrimination dissipated. Most of the decline in
America's share of world production and trade took place as the natural result of Europe and Japan's
reconstruction following World War II. Moreover, the GATT's most important successes occurred in
the 1970s, well after neo-realists claim they should have come.
The Competition State
Another consequence of the speed of technological innovation in the 1990s, and the new
mobility of capital, is that states have reason to fear the loss of competitiveness and of the initial
monopoly rents from imperfect competition. Thus, states are trying to out-liberalize one another in
monetary policy and to subsidize innovation related to trade policy in the hopes of capturing fleeting
commercial advantages and investment. Philip Cerny is worth quoting at length in support of this
proposition (1990, 230):
In effect, the economic and social activities characteristic of the welfare state, as its
name implies, still conformed to the image of the state as an authoritative actor (or set
of actors), the main effect of the intervention of which was to take certain activities
out of the marketplace--especially to attempt to prevent and counteract market failure--
and to 'socialize' them. The state, in this sense was still primarily a
decommodifying agent in the wider economy. But what we are seeing in the world
today, the main impact of increasing openness on the state and on state actors, is the
reverse--the re-emergence of the state as a commodliging agent, a role which it was
often seen to play in the emergence of capitalism itself. (emphasis original).
This new competitiveness of die state relates to one implication of long cycle theory: the
notion that leadership in the international political economy depends upon technological innovation.
Building on Kuznet's work on "epochal innovations" (1966, 286), Kenwood and Lougheed (1970. 14-
16) argue that, since innovation creates fresh opportunities for trade and financial flows, shifts in the
center of innovation importantly influence the international political economy."
In a period as volatile and rapid as the present when it comes to innovation, the possibility for
sustained political upheaval in the inter-state system is significant. It is not my intent in the remaining
chapters to investigate how this possibility translates into system war-proneness as most cycle theorists
do (see Goldstein 1985; Toynbee 1954; Modelski 1978, 1981, 1985: Modelski and Thompson, in
Midlarsky 1989: Thompson and Zuk 1982; Thompson and Vescera 1992; Organski 1958; Organski and
Kugler, in Midlarsky, 1989; Doran and Parsons 1980; Gilpin 1981). Rather, I am primarily interested
in exploring what implications the increased speed of technological innovation in information
technologies in the 1990s, has for the structure of the international political economy. I propose to
treat this investigation as a corollary which I will refer to often throughout the case-studies that follow.
"William Thompson and Lawrence Vescera echo these arguments with what they contend is a
summary of the long cycle literature's treatment of innovation: (1) economic growth is discontinuous and
dependent on spurts in die development of radical imiovations that produce new products or technologies;
(2) states and their economies are organized hierarchically along a technological gradient, and radical
innovation is virtually monopolized at the top; (3) the sources of new products or technologies tend to be
spatially concentrated during a given period. The spatial concentration contributes to conferring lead
economy status on a single state as the principal source of new products or technologies (1992, 517-18).
Corollary Three: A defining characteristic of the New Interdependence is that states behave
more competitively in the new period of interdependence in their pursuit of commercial
The case-studies which follow corroborate this hypothesis. Each state is trying to out-
liberalize the other in monetary policy in order to gain a fleeting advantage by making investment more
attractive. Moreover, in trade policy, many industrialized countries are turning to competition policies
and industrial targeting to snatch from each other the rents from imperfect competition. The state is
least competitive in environmental policy, albeit even here states are concerned about regulating the
environment more rigorously than others and, thus, subjecting themselves to a commercial
Figure 2-3 is the general framework which illuminates how each of my variables interacts with
every other. Convergence is on the top of the dependent processes of the Interdependence Approach to
signify its primary importance. Likewise, capital mobility is positioned above technological innovation
to show that it is the most important independent variable. Advanced capital mobility is the primary
reason that I expect to find a pressure to converge in every issue-area that I examine. This is why the
primary causal arrow in Figure 2-3 extends between capital mobility and convergence. I also expect
technological innovation to increase convergence, but not as directly or unambiguously. This is why a
dashed arrow extends from technological innovation to convergence in Figure 2-3. Finally, dashed
arrows also point in two directions between capital mobility and technological innovation to show that
the two can co-vary, but not necessarily.
The independent variables also point toward collective goods problems and LCD regulations,
because 1 predict that advanced innovation and capital mobility makes it more likely that (1) goods
which states could formerly provide upon their own will change into collective goods problems and (2)
the multilateral agreements which states will reach will tend to be LCDs. Arrows point in only one
direction from convergence to collective goods problems and LCD agreements for two reasons. First,
as I explained, convergence in public policy toward a liberal, global equilibrium makes defection from
coordination more attractive. Second, it collectivizes problems that states could formerly address on
their own. This linkage does not work backwards; that is, collective goods problems do not necessarily
bring about a convergence in public policy to solve them. Neither do LCD regulations necessarily
cause further convergence, since they, in effect, ratify convergence in public policy that has already
taken place. If LCDs prompted more steps toward agreement, they would be meaningful regulations
on the market and, thus, not LCDs.
The large arrow pointing to the New Interdependence in Figure 2-3, from the three processes
that make-up the Interdependence Approach underscores the new reality of interdependence which the
three dependent processes have advanced. Always advancing, the three processes caused the level of
interdependence to increase and traverse the continuum in Figure 1-1 from left to right, until at last we
reached an unprecedented level of interdependence in the 1990s. It was unprecedented because the
market began to limit the autonomy of even large industrialized states in issue-areas historically
sacrosanct to the state, e.g. in high politics.
Therefore, in this sense the three processes I have tapped in my theoretical approach have
created a new environment of interdependence wherein no issue-area nor state is safe from its reach:
the New Interdependence. The next chapter is my "crucial case" study (Eckstein 1975), which will
demonstrate that high politics and industrialized states are caught up in the New Interdependence. The
Interdependence Approach, therefore, suggests a solution to the initial problem with which I began this
chapter--the contradiction interdependent and realistic frameworks pose for understanding international
politics. The solution is to marry both perspectives, using the Interdependence Approach.
In general, I expect that in every issue-area that we consider, the Interdependence Approach
will illuminate a growing contradiction between the state's preference to preserve its autonomy and its
desire to preserve international stability. Innovation and capital mobility make these objectives
dichotomous and encourage the state to act more competitively to promote its own commercial
advantage, even at the expense of odter states. This strategic action and the dichotomy from which it
arises in no way relates to declining hegemony, either. Rather, the de-segmentation of markets and
substantial realization of the liberalizing objectives of Bretton Woods, the GATT, and other post-World
War II multilateral institutions explains this shift. De-segmentation and liberalization are the
consequences of innovation and capital mobility, not waning hegemony.
This will be obvious in each of the case-studies which follow. Chapter Three deals with
weapons procurement policy, my "crucial case" study. It is innovation which is making defensive
autonomy unaffordable and market competition which has made COCOM, an organization designed to
stem the flow of sensitive technology, largely ineffective. Chapter Four evaluates monetary policy and
finds that convergence is the most substantial in this issue-area. Capital mobility and technological
innovation in the context of monetary policy has undermined both Bretton Woods and the EMS. In
Chapter Five, I deal with trade policy and find that liberalization of tariff and non-tariff barriers is
proceeding, even while extra-GATT or WTO competition policies and industrial targeting are becoming
attractive to industrialized countries. In Chapter Six, I examine environmental policy and find that the
tendency is for the state to liberalize regulations on a broader variety of environmental problems.
Finally, in chapter seven, I close with some over-arching observations and expectations for the future.
It should become obvious that my foremost impression is that the market is radically
reconfiguring the inter-state system, leading to the development of competition states in the short-term
and probably to supra-national governance in the long-term. In the words of E.H. Carr,
The conclusion now seems to impose itself on any unbiased observer that the small
independent nation-state is obsolete or obsolescent and that no workable international
organization can be built on a membership of a multiplicity of nation-states.
Table 2-1. Correlation Matrix of Measures of Innovative Activity
Large Small Total Company
Total Firm Firm R&D R&D
Innov. Innov. Innov. Expend Expend.
Large firm innovations .92
Small firm innovations 92 .70
Total R&D expenditure .48 .53 .38
Company R&D expenditure .75 .74 .67 .76
Patents .47 .40 .38 .33 .44
Source Acs and Audretsch 1991. 8.
Figure 2-1: Interdependence Approach
Figure 2-2: Technological Innovation in Information Technologies
Source: Freeman in OECD 1987, 133.
The New Interdependence
r- -----_>- ------ I
Collective Goods LCD
Problems Regulations .
-1- -- -- -- -- -- --- -- -- -- -- -
Figure 2-3: General Framework for the Interdependence Approach
Note The thickness of lines indicates the importance of the causal relationship. Dashed
lines indicate a causal relationship that is not constant.
WEAPONS PROCUREMENT POLICY
In the coming decade, one of the most significant decisions facing European defense
planners, as well as their US counterparts, will concern the extent to which their
defence industries will be internationalized (Moravcsik 1990, 65) If large
corporate alliances and takeovers, such as the Siemens-GEC raid on Plessey, are
successful, governments may be left with little practical alternative but to permit the
international integration of military electronics. (Id. at 70)
Self-reliance is a theme Americans from Ralph Waldo Emerson on have advocated. Yet, in
matters of defense it is now difficult to ensure, as the Gulf War demonstrated. The Gulf War was the
first in which silicon chips played a decisive role. At one level, silicon enabled the Joint Deployment
System to manage the movement of material for all US armed services; the Airborne Battlefield
Command Control Center (ABCCC III) to coordinate air, land, and sea operations; and the Modern
Age Planning Program (MAPP) to simulate battle scenarios for Schwarzkopf's Central Command. At
another level, silicon was at work in laptops, helping pilots to plot their missions; in magnesium-
encased portables, enabling artillerymen to plot trajectories; and in the Aegis Air Defense System,
permitting the Navy to protect itself.
The list could go on, but the point is that in the modern armory "smart" weapons--in other
words, weapons dependent on semi-conductors and other advanced micro-electronics--are a fact of life.
Consequently, foreign-sourcing of the components for these weapon systems is common, either because
they cannot be obtained domestically or because obtaining parts from overseas is more cost effective.
The unique questions this raises include: (1) Are the weapon systems of states sufficiently independent
to enable them to sustain a war effort against their component suppliers; and (2) Can a state reasonably
afford and even develop a viable national defense based upon strategic autonomy? Astonishingly, I
argue that the answer is no if present trends continue. Although the state still manages to limit the
extent to which it cooperates in building entire weapons systems," defense contractors are procuring
components internationally and, thus, standardizing weapons from the ground up. In addition, defense
contractors are forming international alliances through New Forms of Direct Foreign Investment
(NDFI) that promise to integrate the defense industry even further (OTA 1990).
This conclusion is a direct challenge to the structural-realists, who refuse to elevate
interdependence to the realm of "high" politics. Since I do just that in this chapter, weapons
procurement policy represents a "crucial case" study (Eckstein 1975) for my thesis that the market is
reconfiguring the international political economy at the end of the Twentieth Century. I find that all of
the variables that characterize the New Interdependence are as evident in weapons procurement policy
as in any other functional area of the state.
The structure of this chapter follows the order of the hypotheses presented in Chapter Two,
but beginning with those related to the dependent processes that characterize the Interdependence
Approach. Hence, I start with Hypothesis One, and examine the convergence taking place in the
weapons procurement policy area. I shall demonstrate that it is less pronounced at the systems level
than components level of weapons procurement; and that convergence is affecting small industrialized
countries the most substantially, in keeping with Corollary One. Second, I turn to Hypothesis Two
which states that goods states formerly provided on their own are being substantially transformed into
collective goods problems. Here, I will show that, although states have made self-sufficiency in
weapons procurement their primary objective in tie post-World War II period, they have been forced to
accept that their procurement needs can be met only by collaborative programs. Finally, I verify tie
last hypothesis dealing with my dependent variables, Hypothesis Three, by demonstrating that NATO's
systemic weapons cooperation and CoCom's export controls--both efforts to regulate the market's
natural inclination to liberalize commerce--represent LCD agreements.
I41n this chapter a "weapons system," as distinct from a sub-assembly or component, refers to the
entire weapon in question, e.g. the F-16 or M-l Abrams Tank. In contrast, a component of a weapons
system can be, for example, a semi-conductor or dram. A sub-assembly is an organized collection of
components that performs a particular purpose, e.g. the avionics equipment for the F-16.
In the last half of this chapter I take up, first, Corollary Two, which challenges the notion that
declining hegemony accounts for the failure of multilateral regulatory systems to control the market. I
show that, in fact, NATO and pan-European weapons collaboration failed because of the tendency of
the state to preserve its self-sufficiency. This was the case even though, arguably, the US retains a
dominant position in the defense of the West. Second, 1 examine my independent variables found in
Hypothesis Four and discover that, in fact, the mobility of capital and technological innovation is
increasing the three variables of the Interdependence Approach. Weapons are becoming "smarter" and
New Direct Forms of Investment is expanding the mobility of capital in defense. Last, I show that in
keeping with Hypothesis Five, a contradiction is evident in weapons procurement policy between self-
sufficiency and efficiency in weapons production, leading either to collective insecurity or security,
depending on how the contradiction is solved. Increasingly, it is being solved in favor of efficiency.
Hypothesis One states that every country and every issue-area, whether in high or low politics,
is subject to a pressure to converge toward a global equilibrium which the market prefers: conunercial
liberalization. Convergence in weapons procurement programs took longer to develop and has been
less complete than in the other case-studies that we shall examine, particularly monetary policy where
capital is highly liquid. Europe has pursued standardized equipment the most wholeheartedly. This is
in keeping with Corollary One, which states that convergence more substantially influences countries
with relatively small national incomes and issue-areas in which capital is highly liquid. In this case,
since even middle-range military powers like Switzerland and France could not achieve ecnonoies-of-
scale, they eventually tried cooperation. In contrast, the sheer size and technological sophistication of
the US economy allowed it to efficiently produce virtually everything necessary for defense until the
This was the case until the Ford Administration. Then, the US also pursued convergence in
weapons procurement through NATO projects. The term NATO uses to describe this trend is
Rationalization, Standardization, and Interoperability (RSI). In more specific terms, RSI stands for
efforts to achieve the closest practicable substitutability and interchangability of weapons equipment and
components, the most efficient use of R&D and production resources, and the most compatible
technical and operational procedures possible."
"Rationalization" is the overarching term which describes any action that makes more rational
use of defense resources. I will not refer to this term on its own, because of its over-inclusiveness. I
prefer "standardization," which refers to a process whereby military equipment, doctrine, and
procedures in NATO are made identical. NATO uses the term exactly as I do "convergence." Hence,
in this chapter "standardization" is the same as "convergence." Finally, "itteroperability" refers to
steps taken to make different equipment more compatible. Interoperability permits interface with other
weapons systems (i.e. use the same radio frequency), but stops short of further standardization.
Usually, it is retrofitted on to equipment. Thus, interoperability reflects convergence, but only to a
Structural disarmament is the primary reason NATO inaugurated the Rationalization,
Standardization, and Interoperability program. Although I will discuss it in more detail later, structural
disarmament, in brief, describes the decreasing quantity of arms without an equivalent and offsetting
increase in their quality, which a stable and even increasing defense budget can purchase over time.
As weapons have become "smarter," obtaining economies-of-scale in production has required a larger
market than even the US offers. This provides powerful incentive for NATO members to standardize
weapons systems, so that they can be procured in optimal lots and the cost of national defense can be
lowered. However, we shall see that RSI has not been as effective a program as it potentially could
be. This is because states jealously protect the well-being and self-sufficiency of their defense
contractors. This prevents the market from liberalizing the production of weapons systems to the extent
"This definition of RSI is a variant of the Pentagon's, albeit the Pentagon's is even broader. The
Pentagon argues that RSI includes efforts to procure common or compatible operational, administrative,
and logistics procedures, as well as conunon or compatible tactical doctrine (Taylor 1982. 8). Although
these are clearly important elements of standardization, they are not related directly to weapons acquisition.
Thus, they are outside the purview of this chapter.
possible. Many more weapons manufacturers are in business than the market would permit without
On the other hand, standardization is occurring rapidly in defense-related components, where
the state has had little success regulating the market. In effect, the market is standardizing weapons
systems from the ground-up. It is also beginning to influence the form RSI is taking in the direction of
greater standardization, as a result of New Forms of Direct Foreign Investment (NDFI). As we shall
see, NDFI is beginning to link defense contractors globally, despite the state. This process probably
would have accelerated, except for the end of the Cold War. Its end has decreased substantially the
need NATO members feel to produce weapons efficiently. Nevertheless, the market continues to find
cost and quality incentives to standardize weapons components. Below, I will begin this chapter by
examining Hypothesis One in the light of RSI and offshore components sourcing.
NATO Procurement Policy
The first NATO agency to promote weapons standardization emerged at the same time NATO
did: the Military Production and Supply Board (MPSB). Neither this agency nor six reorganizations
of it had much effect. At first, this was because NATO equipment was in fact standardized. It took a
decade for Europe's defense industry to recover after World War II. The US supplied Europe's
arsenal until then. However, by 1959 the situation had changed and divergence was the norm. One
We still have not managed to obtain any worthwhile standardization of our equipment
in NATO. With one or two minor exceptions, we have not succeeded in properly
apportioning armament production tasks among the allies. We repeat experiments in
one country which have already been concluded in another; we insist on reinventing
what has already been invented; and we refuse to trust our friends with secrets which
have been known to the enemy for a long time (Spaak cited in Taylor 1982, 20)
This situation is not completely different even in the 1990s. This is because capital is far less
fungible in arms markets than money markets, and because states have more effectively guarded against
the loss of self-sufficiency in die production of weapons systems than in the governance of domestic
money markets. But, beginning in the late-1970s, market incentives encouraging the standardization of
weapons systems even caught up with the US and thus, NATO. The role that NATO's Conference of
National Armaments Directors (CNAD). a successor to the MPSB, began to play in the mid-1970s,
reflects this best.
CNAD was as feckless as its forebears at weapons standardization until roughly 1975.
However, in that year CNAD's fortunes improved slightly when NATO members agreed to widen its
role and support it more strongly. CNAD began to deal with the future weapons needs of the Alliance
through an agency called the NATO Armaments Planning Review. At this stage, the impetus to
standardization was limited to co-production and co-licensing agreements. Co-production agreements
are government-negotiated arrangements in which the industries of two or more countries take part in
the production of a single weapon system, acquired by all of them (US DOD 1987). Co-licensing
arrangements, are really a subset of co-production agreements, except that, whereas co-producers may
also be co-developers, licensees are never co-developers; licensees must produce weapons according to
government regulations (US DOD 1987). The theoretical benefit of both of these approaches is that
they permit weapons convergence, at the same time as they guarantee local manufacturing content and
Five key weapons programs are associated with CNAD's first stab at weapons standardization:
the multirole combat aircraft (MCRA), the Roland short-range air defense missile, the MAG-58 armor
machinegun, the US/German main battle tank competition, and the F-16 fighter aircraft. Only the last
of these was an unqualified success." Moreover, this is probably because the F-16 became a NATO
"As for the Roland Air Defense System and MAG-58 Machinegun, DOD purchased them in large
part to demonstrate that the US was committed to buying weapons from Europe and, in the case of the
MAG-58, as an incentive for Belgium to purchase F-16's. American producers alleged that their systems
(the Maine-built M60E2 and Chaparral Missile System) were superior. Whether or not this is true, DOD
purchased only about $30 million worth of MAG-58's and installed them on M-l Abrams Tanks after
DOD's plans survived a court challenge by American manufacturers. The purchase contract for the MAG-
58's provided for coproduction in the US of all machineguns in excess of the initial 10,000, in the event
the US preferred this. For its part. the Roland Air Defense System was cancelled in 1979. Rather than
buy Roland as is, the US insisted on "Americanizing" it and co-producing the defense system under license.
Thus, whereas tie French and German Armies began receiving operational Roland missiles in 1976, die
US version was still undergoing reengineering when Congress got tired of delay and canceled the program.
The final cost of the program was $1.9 billion.
Last, the US/German main battle tank competition was solved in favor of politics and
destandardization. It was supposed to be a winner-take-all competition within NATO for main battle tank
project only after it was a demonstrated success in the US Air Force.
The F-16 remains the most successful co-production program within NATO to date. Initially,
the program was meant to produce 348 European and 650 US Air Force multi-role fighters. But by
1986, it had produced 515 planes for Europe, 2,821 for the US, and 756 for eight other countries--and
the number was still increasing (Nunn 1986, Nexis Service). By 1990, the US had licensed or entered
into co-production agreements with 17 countries, including seven NATO members (OTA 1990, 43-44).
Figure 3-1 breaks down the multinational F-16 fighter in its present multinational configuration. In
comparison, the RCRA or Tornado Fighter-Bomber, the next most successful of the NATO projects
announced in the 1970s, was co-produced in Italy, West Germany, and Britain. Canada, Belgium, and
the Netherlands withdrew from the Tornado Progran because of cost over-runs, performance
disagreements, and delays.
Despite the success of the F-16, "minimal" is a fair assessment of the standardization which
the first five key NATO projects achieved in the 1970s. Coordination evolved in an ad hoc fashion,
and had limited overall effect on standardization in NATO. The primary reason was that participation
in the NATO projects was limited to only a few allies. Usually, development and production lead time
on these projects was long, too; and cost overruns, damaging.
The summary lesson of these experiences may be that when the goal of
standardization has conflicted with other interests-such as differing concepts of
military requirements, contracts for domestic firms, and national pride--the
compromises that have been made have tended to increase costs while increasing
standardization incrementally. (Committee on International Relations 1977, 26)
This is not to say that the programs had no effect. At the conclusion of the Tornado Fighter-
Bomber and F-16 procurement programs begun in the 1970s, NATO had plans to operate potentially
eight less fighter aircraft than in 1975. This would have left NATO with 18 basic fighter aircraft,
replacements (about 10,000 over the next couple of decades). This would have represented substantial
standardization in the Alliance. But DOD was more committed to RSI than Congress in the late-1970s.
Congress demanded an "American" tank. DOD finally agreed to this demand after winning some
concessions on interoperability.
rather than 26. Once more, the NATO projects demonstrated new thinking about standardization. The
fact that they progressed as far as they did was unprecedented.
Failure may have also encouraged bureaucratic learning. CNAD shifted from an ad hoc
approach to standardization to a more forward-looking strategy in the mid-1980s. Rather than catalog
the weapons needs of countries and promote a procurement dialog once the countries were ready to
buy, CNAD began to take the initiative in developing weapons specifications to meet future Alliance
needs. The US paid for this shift in 1986, by setting aside $250 million for NATO cooperative R&D
projects. In addition, DOD increased its collaborative R&D expenditures from 3 to 25% of total R&D
expenditures that year. This funding represented a much more substantial commitment to
standardization than DOD's support of co-production and co-licensing agreements. After all,
standardization through R&D eliminates the need to retrofit interoperability and integrates weapons
systems at the design stage.
The trouble is many of the NATO projects begun under the banner of R&D cooperation were
abandoned in the late-1980s and early-1990s. There are some exceptions. Many of these proved
critical to the successful prosecution of the Gulf War; for example, the NATO Navstar Navigation
System, coalition minesweeping forces, interoperable anmunition, interchangeable chemical warfare
protection suits, and interoperable maritime helicopter flight decks. But key joint R&D standardization
programs like the NFR-90 Frigate and Advanced Short-Range Air-to-Air Missile (ASRAAM)/Advanced
Medium-Range Air-to-Air Missile (AMRAAM)--the symbols of the new era of weapons cooperation
that NATO held up--were cancelled. Critics charged that the weapons had technical problems and
would be too expensive for post-Cold War defense budgets. In truth, in these instances efficiency and
structural disarmament proved less compelling interests than the jobs domestic political coalitions
argued would be lost.
Hence, in the 1990s, despite the emergence of strong incentives for weapons standardization in
NATO, the state continues to protect, albeit not totally successfully, the self-sufficiency of its defense
contractors in the production of weapons systems. Recession and, more importantly, the end of the
Cold War made this decision easier. Because of a reduced security threat, Alliance members seem
willing to bear the cost of inefficient weapons procurement policies--most notably, structural
disarmament. Thus, although the benefits of CNAD's shift in the mid-1980s to a more forward-
looking standardization process should have accrued in the early-1990s, they did not. Funds
contributed to joint R&D programs for prototypes of weapons were wasted. Table 3-1 shows that the
fairly steady growth in numbers of major weapons systems in NATO since 1962-63 continued through
1990. For example, although there were ten tactical aircraft in NATO in 1962-63, there were 42 in
1990; and, whereas there were 2 helicopter models in NATO in 1969-70, there were 38 in 1990.
Thus, contrary to Hypothesis One, divergence in the procurement of weapons systems
continues to be the rule in NATO, despite its commitment to the opposite. This would seem to
undermine the Interdependence Approach, except that standardization of weapons systems is not the
only type of convergence important in defense procurement policy. Most importantly, divergence can
occur at the systems level, while convergence takes place at the components level of weapons systems.
Ultimately, the twain shall meet, since the convergence of parts of a system must lead to convergence
of the system itself. This is how I shall verify Hypothesis One in this chapter. Interestingly,
convergence at the components level is occurring in spite of public policy. This is similar to the
pattern we shall see in Chapter Four, where despite the decision of the G-7 not to cooperate in setting
convergent monetary policies, convergent monetary policies are the result of financial innovation and
dramatic capital mobility. We turn to standardization at the components level of weapons next.
Offshore Component Sourcine
One of the oddest things from a neo-realistic viewpoint about the standardization of weapons
components is that the state has largely ignored it. Standardization is occurring by default in the
Alliance because many weapons components are now purchased from offshore sources--either from
other NATO members (major subsystems and built-up components) or Japan (semiconductors). This is
despite the obvious threat to national security foreign dependence even at this level causes. Although
there is no data on the precise extent of overall foreign component sourcing, informed estimates put the
number at 20% of the parts in US weapons systems--and growing (Vernon and Kapstein 1991). This is
compared to basically no foreign components sourcing just two decades ago. In general, European
members of NATO have had a higher level of components dependence upon the US, especially in
aviation and micro-circuits. Roughly 30% of the Tornado's components, for example, are American-
sourced (Moravcsik 1990, 78).
Of course, in some instances dependence on offshore component sourcing is much more severe
than 20 or 30%. In the US, for example, the single domestic supplier of ceramic package
manufacturing, one of four elements essential for micro-circuits used in defense applications, is
Kyocera America, Inc., a subsidiary of a Japanese firm. Its sales represent about 13% of sales in the
US ceramic packaging market. The rest of the market is controlled by overseas firms.
Three primary factors explain why this kind of interdependence is growing at the components
level: (1) defense contractors are becoming multinational, (2) "dual-use" technology, or technology
that has both a commercial and military application, is more common, and (3) capital can now search
internationally for optimal cost, quality, performance, and delivery time criteria. I will save the first
point for later. Evidence for the second became apparent in 1981, when the Reagan Administration
attempted to tighten export controls on technology potentially useful in military applications. The
Administration discovered that so-called "dual-use" technologies had grown so much that "it threatened
the entire control process and therefore the national security of the US (today, 90% of defense needs
are dependent upon dual-use technology)" (Parks 1990, 78). Dual-use technology permits access to
parts for which states could not otherwise achieve economies-of-scale. Indeed, during the 1970s and
1980s, the most rapid increase in international trade took place in products closely linked to battlefield
capabilities like data processing equipment, transistors, telecommunication equipment, engines,
measuring instruments, and artificial resins (Vernon and Kapstein 1991, 9).
Dual-use technology is the most pronounced in micro-electronics, a field critical to "smart"
weapons. It was also a field at one time dominated by DOD, but now far more commercial than
defense-related. The military contributed handsomely to AT&T's initial research on semiconductor
technology and later encouraged Bell Labs to disseminate technical data to its licensees. Military
orders provided early markets for new products and spurred American firms on to technological
leadership. As late as 1963, defense-related purchased accounted for over 35% of US semiconductor
output and over 90% of the market for integrated circuits. In contrast, by the late-1970s, defense-
related purchases of semiconductors amounted to just 10% of US output (Ziegler 1992. 157).
In the 1970s, military procurement exerted less and less leverage over the semi-conductor
industry. Whereas earlier DOD could count on receiving state-of-the-art chips specially designed for
its weapons systems, the new, defense-related specialty chips had low priority in the industry and were
often obsolete by the time weapons systems were deployed. This trend especially concerned DOD
when American domination of semi-conductor manufacturing and the production of equipment for
fabrication of chips shifted to Japan in the early-1980s. By 1981, Japan claimed 70% of the US market
for memory chips (DRAMs). Likewise, between 1984 and 1988, the share of world exports of semi-
conductor fabrication equipment generated in the US slipped from 66% to 50%. In NATO,
convergence in the weapons-related procurement of DRAMS and semi-conductor fabrication equipment
is now around the exports of primarily Japanese manufacturers.
It will be difficult to reverse this trend, as some American companies learned in the late-
1980s. Semi-conductor fabrication equipment and DRAM production depends on the geographical
proximity of so many factors "that the task of returning all production to US territory had become
technically, as well as financially overwhelming" by the late-1980s (Id. at 173-74). Nonetheless, DOD
determined to protect America's still strong presence in the integrated circuit industry. Thus, it
abandoned its preoccupation with only defense-related research on semi-conductors and contributed to
tie founding in 1988 of Sematech, an association of US-based integrated circuit manufacturers designed
to disseminate technology. In so doing, DOD embraced Japan's strategy of encouraging commercial
success in the industries that indirectly benefit national security. DOD thought that Sematech would
rejuvenate specifically American semi-conductor manufacturers. It has, but, as we might expect based
on the New Interdependence, it has also benefited foreign producers through the joint research and
manufacturing agreements that American firms have, ie. IBM-Siemens and Texas Instruments-Toshiba.
Moreover, it was still the case that in 1989, 30% of the semiconductor purchases of the US military
were foreign-sourced (Id. at 174). The figure for European members was dramatically higher, at 80%.
As this cursory look at the semi-conductor industry shows, it is often the case that even if a
state decides that it wants to become self-sufficient in the manufacture of a particular weapons
component, it cannot. Either the state may not have manufacturers capable of producing the component
or if it does, the manufacturers may not be able to produce it at competitive prices--my final point as to
why interdependence in components-sourcing is growing. Consider some of the foreign-sourced
components on the M1-AI Abram Tank: optic quality glass, MAG-58 machine guns, 120-millimeter
smooth-bore canons, trimer, and semiconductors. Optical glass for it can be produced in the US, but
at a cost two to three times that of foreign sources. Similarly, trimer, an ingredient in the plenum seal
connecting the engine and the air intake system of the MI Abram tank, is a product which the US has
the technology to deliver. But a new domestic production facility would need to be built at a cost of
$1.5 million. DOD considered building the facility, but determined that because it could not guarantee
more than $1 million in purchases, it probably could not he sustained over the long-term (GAO 1991,
DOD has also found, on occasion, that even when it could purchase the manufacturing
capability to produce a weapon component, that was not enough. For example, uider Congressional
pressure to find a domestic alternative to the British-built F/A-18 fighter ejection seat, DOD negotiated
a co-production agreement between Martin Baker Aircraft Co. of the U.K. and East West Industries in
New York. Despite the deal, East West still purchases elements of the seat from Britain. As DOD
officials put it, having a complete technical data package for the seats does not guarantee that they can
be entirely produced here. The other essential component is capability (GAO 1991, 22).
Thus, despite the overall failure of NATO to standardize its weapons procurement at the
systems level in the 1980s, in keeping with Hypothesis One the internationalization of defense-related
components procurement is standardizing them, anyway. This is happening the most dramatically in
the electronics industry and advanced aviation, but it is also occurring in less sophisticated fields like
the production of ejection seats. Standardization of parts is convergence not designed purposefully into
weapons, but as a result of market incentives. Hence, even when NATO fails to respond to the
inefficiency obviously leading to the structural disarmament of the Alliance, convergence in weapons
procurement policies continues at the components level.
Furthermore, this convergence is essentially irreversible for three primary reasons: (1)
essential components are simply not available domestically or, at least, not at a competitive price, (2)
the increasing cost of weapons R&D and production makes international consortia of defense producers
necessary, and (3) the increasing availability of dual-use technologies makes the purchase of specialty
defense components inefficient. In addition, the sheer number of parts involved in a weapons system
and the multinationalization of defense contractors that has already occurred makes state regulation and
tracking of interdependence at this level virtually impossible.
Collective Goods Problem
In fact, it is (1) the very inability of the state to avoid interdependence at the components level
of weapons systems and (2) the choice the state faces between structural disarmament and coordinated
weapons procurement which allows me to verify Hypothesis Two. It states that goods which countries
formerly provided on their own are being substantially transformed into collective goods problems. In
this case, it is national security which is being transformed into a collective goods problem for even tie
most powerful industrialized countries. Since World War II, industrialized countries have tried to be
self-sufficient in weapons production and, thus, in some measure to guarantee their own national
Until the 1970s, the US in particular could have severed all contact with the rest of the world
and still met its weapons procurement needs. European countries, especially non-aligned ones, would
not have been quite so successful at achieving autarky, but they could have come close with second-
class weapons. Beginning in the 1970s, however, dependence on offshore components and the need to
counter structural disarmament undermined even the superpower's most concerted efforts to remain
self-sufficient in armaments production. Now, even the US has had to give-up the goal of autarky and
depend on multilateral cooperation to meet its procurement objectives.
Structural-realists have argued that interdependence in defense-related equipment is not unique
to the late-Twentieth Century. Although this is true, a brief review of the interdependence in weapons
procurement in the past shows that the extent and nature of interdependence in weapons procurement
today is unprecedented. To begin with, during the 1800s, industrialized countries were, for tie most
part, able to attain the goal espoused by mercantilists like Alexander Hamilton and Friedrich List: self-
sufficiency in armaments production. The largest exceptions had to do with imports of essential raw
materials. This situation changed only in the late-1800s and early-1900s, as interdependence increased
and trade in weapons expanded again. Germany's Krupp, for example, sold to both sides of the
Franco-Prussian War; and on the eve of World War I, filled Russian, French, and British weapons
orders. However, prior to World War 1, there was no ideological split between the major powers.
The shift toward ideological warfare, together with an era of unprecedented government intervention in
weapons research and procurement during the Cold War, forever changed tie character of
The fixed costs of new weapons, particularly aircraft and advanced naval vessels, increased
after World War II. However, exponential increases in their cost did not occur until the 1970s. Only
huge markets for these weapons could offset the tremendous investment they required. In the 1970s,
"fixed costs placed autarky far beyond the means of any single country in Europe" (Moravcsik 1992,
40), while the US remained self-sufficient for a decade. Never before had interdependence affected the
great powers so asymnuetrically. Moreover, the New Interdependence of the late-1970s was not
primarily in raw materials, but electronics and advanced composites which require advanced technical
knowledge to manufacture. Knowledge and economies-of-scale, in other words, not natural resource
endowments, is the key to military interdependence in the 1990s.
Now, not even the superpower can provide for its own weapons procurement needs. Not only
that, but it too is undergoing structural disarmament, as a result of inefficient procurement practices
that could ultimately endanger its own national security. The problem is not acute, because of the end
of the Cold War. However, the trend in the late-1970s and 1980s, was for the Warsaw Pact to outstrip
NATO in almost every category of major weapons systems production. Most startling, this was despite
the fact that the Alliance had a collective wealth worth 2.5 times the GNP of the Warsaw Pact, a
population 1.5 times larger, and a defense budget roughly equivalent to the Warsaw Pact's.
To reverse this trend would have required the collective effort of the Alliance to standardize
weapons. The technical knowledge and economies-of-scale necessary to produce and procure weapons
efficiently now depends on multilateral cooperation, since no individual state in the Alliance can deliver
either. As we saw in the last section, though, it was difficult even during the Cold War to convince
Alliance members not to "buy national" for political reasons. Most NATO projects involved few
Alliance members. The ones that failed usually were canceled because states withdrew from them.
Even NATO projects that were completed inevitably finished with fewer participants than they began
with. This is the free rider problem at work. The two conditions of a collective goods problem are
also present: (1) participants in NATO projects could not prevent other Alliance members from
enjoying the standardization and strategic advantage they provided (nonexclusion) and (2) NATO
project participants obviously could benefit from standardization without diminishing the level of
standardization from which other Alliance members could benefit (nonrivalness). It is no surprise.
therefore, that Rationalization, Standardization, and Interoperability failed and the gap between East and
Between 1987 and 1988, the size of the Soviet military machine grew almost 50%, whereas
NATO's forces grew just 10% (Committee on Foreign Relations 1988, 122). The Warsaw Pact also
dramatically reduced its qualitative inferiority throughout the 1970s and 1980s, in pan because it could
deploy new technology faster than NATO (Moodie 1986. 24). Table 3-2 shows that in some weapons
technologies the Warsaw Pact may have been even superior to NATO. The consequence of this shift
was that in 1987, Secretary of Defense Weinberger reported that tie Pact's advantage in-place ground
force combat power had increased from around 1.5 to I in 1965 to 2.2 to 1 (Committee on Armed
Services 1988, 67). The index he referred to combines a measure of the quantity and quality of
opposing forces. It suggested that the cushion NATO members would have in the event of Soviet
aggression was reduced from a projected two weeks to maybe just two days before NATO would have
to resort to nuclear weapons.
Readiness reports also indicate that in a crisis, NATO forces probably could not have
effectively reinforced one another, because they fired different projectiles, serviced different equipment,
and even communicated on different frequencies. All of this was the resort of divergent weapons
procurement at the systems level in NATO. The gathering convergence at the components level had
not yet caught up with the weapons system as a whole. In the late-1980s, NATO had six main battle
tanks, each of which fired different ammunition. Even NATO's multinational F-16 had unique weapon
carrying systems that were not compatible between national forces. Hence, if an F-16 landed at an
Allied base, most likely it could not have been fitted with bombs; the plane would have been grounded.
Other NATO jets could not even be serviced at Allied bases; hence, even in good conditions, they
might be grounded on the basis of technicalities. Most astonishing, NATO forces often could not even
communicate with one another.
Warsaw Pact forces were substantially more standardized in the 1980s, because they had a
dominant supplier: the Soviet Union. Ironically, socialist countries, which tried to disconnect
themselves from the global economy, took better advantage of capitalist economies-of-scale in defense
procurement than did NATO. Of the approximately $23 billion NATO spent in 1981 on R&D, as
much as 28% of the total (Europe's combined R&D expenditures) was redundant (Callaghan in
Conunittee on Armed Services 1988). More recently, DOD's Office of the Inspector General estimated
that the Defense Department could save $10 billion between FY1992 and FY1997 if it would take full
advantage of the R&D programs of American allies and their contractors (Defense & Aerospace
Electronics 1992, 2). Likewise, the Grace Commission estimated that by ignoring the market's signals
to standardize NATO R&D, the US wasted $2.3 billion annually through the early-1990s.
NATO wasted far more money by ignoring the market's incentives to coordinate actual
weapons acquisition, too. A rule of thumb in the aerospace industry is that the unit price declines by
15 to 20% for each doubling of production, simply because firms learn (Moravcsik 1990, 67).
Likewise, Callaghan estimates that the US could save 10% of its procurement expenditure and Europe,
25%, if economies-of-scale were realized (Taylor 1982, 51). In May. 1987, a Senate sub-committee
During the last five years, weapons have been produced at just 50% of efficient
production capacity. A fourth of the major acquisition programs during the past five
years were manufactured at a rate below the minimum economic rate for those
systems. Half of the largest 20 weapons systems were stretched-out, compared to last
year's plans.'7 (Committee on Armed Services 1988, 90)
Stretch-outs increase the proportion of overhead to military product. A longer stretch-out
means less cost effective production, since each dollar buys more overhead and fewer weapons. In
contrast, an optimum funding level is the minimum amount of money required to produce the maximum
number of units on one eight-hour shift, five days a week. To make this concept more tangible,
consider the now famous Patriot missile system. The Army estimates that stretching-out this missile
system in die 1980s, cost $1 billion, a sum that could have purchased 1,760 additional Patriot missiles.
Likewise, the Air Force claims that stretching-out the F-15 for three years in the 1970s cost $2 billion,
a sum equal to the purchase price of 83 F-15's. Instead, the money purchased overhead.
All of this inefficiency, the result of divergence in NATO's weapon procurement policies, is
the root of structural disarmament: states are spending more and more dollars to get fewer and fewer
weapons. In the words of the former Supreme Allied Commander of Europe (SACEUR) Gen. Bernard
Rogers, "Every time we replace a system, the cost of the newer one has increased by a factor of from
three to five. Consequently, we procure fewer and fewer systems as replacements each time. ."
"So-called stretch-outs increase the proportion of overhead to military product. Every funding
level below an optimum level buys more overhead and fewer weapons. An optimum production level is
the minimum amount of money required to produce the maximum number of units on one eight-hour shift,
five days a week" (Committee on Armed Services 1988, 94).
(Committee on Armed Services 1988, 125). Rogers's successor, SACEUR Gen. John Galvin echoed
In 1988 prices, the cost of building a new F4 Phantom would be US$9 million; the
F14 Tomcat that replaced it is priced at US$32 million. Similarly, an M60 tank could
be fielded for US$1.4 million (again at 1988 prices); but its successor--the M --cost
US$2.7 million. A type 23 frigate is 3.5 times as expensive as the Leander which it
replaces. (Galvin 1988, 18)
Once again, the same budget bought the US 3.000 aircraft per year in the 1950s, 1,000 per
year in the 1960s, 500 per year in the 1970s, and roughly 300 per year in the 1980s (Committee on
Armed Services 1988, 124). This was not primarily because we got a "bigger bang" for the dollar,
either. Declining quantities of weapons are not being equivalently offset by increases in quality.
Andrew Moravcsik reports that the cost of a given unit of increased performance (the cost/performance
ratio) of new fighter aircraft has been steadily increasing for three decades at a real annual rate of
between 0.5 and 1.2%.'8
Structural disarmament has most severely affected small NATO members and non-allied
industrialized countries. This is as we might expect, based upon Corollary One, which states that
convergence more substantially influences countries with relatively small national incomes. Middle-
range military powers are forced to seek large export markets for weapons to maintain a semblance of
defense autonomy. Hence, they tend to be less favorable toward restrictions on the proliferation of
defense technology, too.
For example, France, a middle-range military power has always strongly resisted limitations
on the export of defense technology. Exports to developing countries are a natural strategy for France
to pursue because of its many former colonies. French firms like Avions Marcel Dassault, a military
aircraft manufacturer, exported roughly 70% of its production in the 1970s and 1980s (Moravcsik
1990, 68). In 1984 this prompted the French Defense Minister to acknowledge, "If France didn't
"This deflation was more severe in defense than the civilian sector, too. Between 1947 and 1987,
whereas the cost of commercial manufactured items increased only tenfold in the US, the cost of military
manufactured items increased 100-fold (Conimittee on Armed Services 1988, 90).
export, it couldn't buy its own weapons from its own producers. The cost would be too high ...
and we would have to buy more weapons abroad" (Committee on Armed Services 1988, 125). The
same is true for non-aligned Sweden. Sweden's so-called "national champion" in defense, the Celsius
Industries Corporation, has 50% of its weapon sales overseas. Celsius does most of its naval business
outside of the country, yet Celsius is still tremendously subsidized--to the point that Sweden is
considering privatizing the corporation (Hitchens 1992, 22).
Thus, in review, the divergent defense procurement patterns of NATO members in the decades
prior to the 1970s, could not be sustained without structural disarmament occurring and a decline taking
place in Alliance readiness. Formerly, states could achieve self-sufficiency or nearly so in weapons
production and, therefore, provide for national security. They might still be beat on the battlefield, but
it need not have been for lack of economies-of-scale in weapons production. However, in keeping with
Hypothesis Two, which states that goods states formerly provided on their own are being substantially
transformed into collective goods problems, adequate weapons production now depends on technical
knowledge and economies-of-scale that can be obtained only multinationally. Provision of adequate
weapons for defense has become a collective goods problem. Not even the superpower can prevent its
own structural disarmament. Smaller allies are even less capable of meeting these criteria, as Corollary
One would have predicted. In fact, the asymmetry apparent in a country's ability to provide for
national security is more dramatic now than at any time previous.
Lowest Common Denominator Agreements
To reverse the trend toward structural disarmament, the state must liberalize policies for
procuring weapon systems. Alternatively, if the state intends to try to control the multinationalization
of weapons components, it will have to take some strong steps to pull back from the international
economy in ways likely to be violative of the GATT. Either way, it will have to act decisively. So
far, industrialized countries have taken primarily symbolic steps, in keeping with Hypothesis Three,
which states that public policy coordination designed to regulate the market is increasingly likely to
center around agreement on a LCD. For example, rather than coordinate standardization effectively
and provide mechanisms whereby the Allies could share die political costs of not "buying national." the
NATO Rationalization, Standardization, and Interoperability Program is itself a LCD. The RSI
Program suggests that NATO is serious about procuring weapons efficiently, but it actually protects the
right of the Allies to be self-sufficient. As a result of weak enforcement regulations and co-production
and co-licensing agreements, RSI has not altered the overall inefficiency in weapons procurement.
Weapons produced collaboratively under the RSI framework, are separately manufactured, thus
eliminating important economies-of-scale.
NATO projects have failed not only to stem the tide of structural disarmament in net terms,
but also to satisfy the performance needs of most participants in the Projects. In the worst instance, a
weapon that reflects a compromise between multiple national missions may be unsuited to perform any
of them effectively. For example, disagreement over performance expectations and cost ultimately
undermined the NFR-90 Frigate project and Advanced Short-Range Air-to-Air Missile
(ASRAAM)/Advanced Medium-Range Air-to-Air Missile (AMRAAM) program. Alternatively.
differences have led members in NATO projects to modify the base NATO project model with
customized options and to demand coproduction rights that eliminated many of the economies-of-scale
RSI was supposed to guarantee. For example, the European Fighter Aircraft (EFA) planned for the
1990s, will have four models, each manufactured separately in Britain, Spain, Italy, and Germany.
None of these states could decide which of them should benefit from the revenue and jobs creation the
European Fighter Aircraft program would bring if completed as a single aircraft.
In general, the problem of dividing the costs and benefits of standardization equally is a root
cause of LCD NATO projects. Real economies-of-scale in the production of a NATO weapon depend
upon the ability to depreciate overhead in a single location. Efficiency requires optimal production runs
in one country. But that means that many other Project members cannot benefit from technology
transfer and revenue that particular weapons systems generate, except insofar as they source
components for the main weapon system. The problem is quite similar to that analysts encounter in
asymmetric economic communities like those the developing world has tried.
For example, investment in the Andean Pact flowed to the strongest economies and virtually
overlooked the weakest. But the weakest needed investment the most: and the fact that investment
bypassed them meant that they fell farther behind the strongest economies. Likewise, a NATO project
which maximized the gains from RSI would locate primary production in the most technologically
advanced country relative to that product; therefore, contributing to that country's leadership in the area
in question and widening the technology gap between it and other Alliance members. Rather than be
part of the "out" group and obtain a weapon not quite suited to a particular national need, states choose
to co-produce NATO projects or to back out of NATO projects. From a strictly utility-maximizing
perspective this makes sense, too. Free riding on the standardization efforts of others in an Alliance
guarantees the free riders basically the same level of national security available to those committed to
Rationalization, Standardization, and Interoperability.
Recall that free riding, in general, is intimately wrapped-up in the trend toward LCD
agreements. Since every state prefers to be a free rider on die regulations of others, no individual state
has an interest in providing the regulation. So, the regulation oftentimes is not provided and statements
of resolve or agreements that have at best short-term efficacy--in other words, LCD agreements--
replace them. This is true especially as convergence tightens around a multilateral equilibrium (the
airline regulation example from Chapter Two). CoCom, the Coordinating Committee on Multilateral
Export Controls, represents such an equilibrium for controlling the multinationalization of weapons
components and technology." It has confronted one primary question since its inception in 1949:
How does a state prevent the transfer of sensitive military technology capable of undermining its
national security, while not impeding legitimate scientific, technical, or commercial exchange?
The Allies have not answered this question the same way. CoCom has always been composed
of two groups: the strict export controls bloc and the loose export controls bloc. The US has always
been in the former bloc, in part because its large defense market made exports of technologically
1"COCOM is composed of NATO Member states minus Iceland, but including Japan.
sensitive weapons less crucial to obtaining economies-of-scale. Most other Allies have favored looser
export controls, since they need to trade to protect their defense industry and because their ideological
opposition to communism is less pronounced. This disagreement in CoCom only became critical in the
late-1970s. Until then, CoCom fairly effectively regulated the diffusion of technology with obvious
But the shift toward knowledge-intensive weapons in the late-1970s, made economies-of-scale
more critical and "dual use" technology more of a problem. Dual use items confused what should be
regulated. In addition, the number of states vying for sensitive technology increased. Until the 1970s,
CoCom concentrated primarily on denying military technology to the Soviet bloc, but afterwards it had
to worry about developing countries, as well. Trends which concerned the Alliance most were nuclear
weapons proliferation, as a result of advances in "civilian" reactor technology, and the diffusion of
ballistic missile-related technology to countries like China, North Korea, India, Brazil, Israel, Iraq, and
Libya, as a result of advances in space and defense programs (McCall 1992).
Changes like these, together with the Warsaw Pact's numerical advantage in military
equipment in the late-1970s, spurred the US to pass even tougher domestic laws against sensitive
exports (the 1979 Export Administration Act) and to press for stricter guidelines in CoCom. But
CoCom members preferred a LCD approach. CoCom adopted three categories of restricted weapons:
the Direct Military Use List. International Atomic Energy List, and Dual Use List). It limited
technology on the Dual Use List the least. Furthermore, since agreements reached in CoCom depend
upon implementation by members states, widely varying enforcement of CoCom's restricted lists has
been the rule (Parks, 1990). Not even the invasion of Afghanistan and martial law in Poland created a
consensus in favor of a more restrictive multilateral trade policy and effective law enforcement. As a
result, American manufacturers lost business in the East to Allied manufacturers. The National
Academy of Sciences (NAS) estimated those losses at $9.3 billion in 1985 (Parks 1990, 99). Individual
industries like the computer and telecommunications industry and the oil and natural gas industry claim
much higher losses.
Moreover, despite these losses, tie US did not stem the proliferation of sensitive military
technologies. This is in keeping with Hypothesis Two on the transformation of goods that states could
formerly provide on their own into collective goods problems. It also verifies that CoCom has been no
more than a LCD, since the period of the New Interdependence began. Analysts disagree about the
extent of proliferation under the not so watchful eye of CoCom, but a DOD Assistant Secretary for
International Security Policy characterized it as a "hemorrhage of technology" to hostile nations
(Feldman 1985, Nexis Service). Likewise, Sam Nunn testified,
Inhibitions within this Alliance on technology transfer are supposed to slow the Soviet
military flywheel; but I submit that the Soviets are stealing almost everything they
want and that the machinery we've set- up to restrict our own technical dialogues are
self-defeating, because they are slowing us down, and that's why we deploy front-line
weapons so much more slowly than the Soviets. (Nunn 1986, Nexis Service)
Although the US tried to extend its regulatory powers extraterritorially and to foreign
companies with operations in the US to stem this flow by itself, Canada, France, and the U.K.
introduced blocking legislation. Many members of CoCom did not even enforce adequately the few
export limitations to which they had agreed.
The 1987 Toshiba-Krongsberg Affair illuminated quite dramatically the consequences unequal
and lax enforcement of CoCom export controls could have for NATO. Some contend that the propeller
refining technology these companies transferred to tie USSR advanced its propeller development
program seven to ten years, and would have made $30 billion in countermeasures necessary had the
Cold War not ended. This amount may be small, however, compared to the costs of technology
transfers that took place through channels that COCOM did not monitor during the Cold War. The
FBI estimates that as much as 90% of the defense technology which the Soviets obtained through
espionage was supplied in this manner (Feldman 1985. 65).
The Toshiba-Krongsberg scare did have one positive outcome. It led to a rare display of unity
on strategic export controls in CoCom (Parks 1990, 89), which might have propelled CoCom out of the
ranks of LCD's toward an effective regulatory system. In response to the Toshiba-Krongsberg affair,
CoCom members agreed to shorten the list of controlled technologies, but also to toughen enforcement
procedures and, thus, level the regulatory playing field. This unity did not last long after the Iron Wall
fell. CoCom fell back into its LCD rut. The US stood alone in 1989. insisting on controls on sensitive
weapons sales to Eastern Europe and even stricter controls on sales to the ailing Soviet Union. In June
1990, the US finally gave in to a more liberal standard, as the sales American companies lost to Allied
countries mounted. In the first of many reductions to come, CoCom agreed in 1990, to end controls
on 30 of 116 "protected" technology categories (Auerbach, Washington Post, 8 June 1990, F: ).
Debate continues on how far this process should go, with Germany and the US the most conservative,
and Japan and France, the least (Behr, Washington Post, 27 August 1993, B:1). But for all intents and
purposes, the shift of CoCom toward an almost completely symbolic function is nearly complete.
In review, Hypothesis Three states that public policy coordination designed to regulate the
market is increasingly likely to center around agreement on a LCD. This section verified this shift in
three separate ways: (1) NATO projects designed to bring about standardization and reverse the trend
toward structural disarmament, yet not totally abandon self-sufficiency objectives through co-production
and co-licensing arrangements, have failed substantially to stem the proliferation of weapons systems in
the Alliance: (2) NATO projects, themselves, represent a compromise on performance, cost, and
delivery time expectations for weapons that fully satisfy none of their participants; and (3) CoCom has
evolved from being a fairly effective and non-controversial organization that controlled the diffusion of
technology to the Soviet bloc to an organization fulfilling the largely symbolic function of listing a
decreasing number of knowledge-intensive and dual-use technology that China, North Korea, and a few
other radical developing countries ought not to have access to. For the most part, these states can still
obtain access to the technology, because there is always at least one free rider in CoCom eager to make
a dollar profit or generate political influence. LCD agreements can never effectively achieve the
Rationalization, Standardization, and Interoperability, performance, or non-proliferation objectives of
NATO. The trend toward them may not be so pronounced in a regional integration community,
though. We shall examine this question next.
Weapons Cooperation in Western Europe
In Western Europe the temptation for states to free ride on the regulations of one another is
less than among industrialized countries as a whole, because of the institutional supports which exist.
A primary purpose of these supports is to level the playing field in Europe and ensure that the
regulations enacted by one Western European country will not be undercut by another. Hence, we
might expect that in specifically Western European forums efforts to address structural disarmament
and the proliferation of sensitive technologies would proceed further than in NATO--especially as
American domination of NATO has waned. This is not the case, though.
The same tradeoffs which affected NATO between self-sufficiency and efficiency in weapons
procurement, and between stemming proliferation and promoting trade, influence Europe. Hence, in
keeping with Corollary Two, which states that capital nobility, not declining hegemony, accounts for
the failure of multilateral regulatory systems to control the market, I find in this section that declining
American hegemony is not an adequate explanation for structural disarmament, failed standardization
programs, and weak export controls on military technology. It is innovation which is making defensive
autonomy unaffordable and market competition which has made CoCom largely ineffective. In other
words, the Interdependence Approach better explains structural disarmament, failed RSI, and weak
export controls on military technology than the neo-realistic preoccupation with declining hegemony.
To prove this I will show in this section that every hypothesis which I verified in the previous sections
on NATO using the Interdependence Approach, can also be verified in the context of pan-European
weapons cooperation. Yet, in Europe American domination of NATO is, if anything, as secure as it
ever has been.
As in NATO, pan-European defense cooperation emerged when the European Union did. The
most powerful symbol of that cooperation, the European Defence Conmunity (EDC), was designed to
procure weapons for the entire Community. This would have been a strong standardizing influence.
But the French Assembly narrowly defeated the EDC in 1954. In response, Great Britain convinced
Europe to replace the EDC with a less powerful organization, formally independent of the EC. Great
Britain hoped thereby to justify Germany's rearming and admittance to the Alliance. So, in 1955
Europeans established the Western European Union (WEU). In addition, two other institutions
emerged independently in the 1970s, also to support pan-European weapons procurement: the
Independent European Programme Group (IEPG) and Eurogroup. Table 3-3 provides a summary look
at which military powers belong to which organizations involved in weapons procurement collaboration.
Like CNAD, all three pan-European procurement organizations were moribund until late in their life.
In fact, in Europe they hit full stride after NATO's CNAD did--not until the late-1980s.
The Western European Union awakened from its slumber first, but like the other pan-European
procurement organizations has had limited effect, overall. In 1984 Senator Samuel Nunn's effort to
arm-twist Europe into increasing its defense outlays for NATO under threat of an American troop
withdrawal had the unintended effect of giving the Western European Union a new lease on life. The
reason for their new enthusiasm, Europeans (except France) insisted publicly, was "to make the WEU
into the European pillar of the Atlantic Alliance" (Financial Times 1984, 1:14). However, in Europe
the most widely cited reason was America's continued domination of the European weapons market
(OTA 1990, 56) and structural disarmament. Regarding the former point, not declining American
hegemony, but its continued strength, prompted a backlash in Europe. Although the ratio of European
weapons imports to US imports dropped from 8:1 in the late-1970s, it remained at approximately 2:1 in
1990 (Moravcsik 1990, 78).
Structural disarmament was also a growing problem in Western Europe in the 1970s, that
created severe pressure for liberalization in national weapons procurement programs. This is in
keeping with Hypothesis One, which states that every country and every issue-area, whether in high or
low politics, is subject to a pressure to converge toward a global equilibrium which the market prefers:
conunercial liberalization. States resisted the market's demand for economies-of-scale in the form of
free access to multiple anms markets at the cost of national insecurity. The same stretch-outs and
diminishing "bang for the buck" evident in the US struck European countries, but with more force.
European countries responded initially with two strategies: (1) mergers and subsidization and (2)
exports (Moravcsik 1992, 34). Mergers aggregated the number of independent contractors into one or
two commercial units; thus, increasing potential economies-of-scale. Subsidization guaranteed a
minimum income for the contractors. Finally, exports expanded the potential market of a weapons
For example, France consolidated its aircraft contractors and sub-contractors into two "national
champions," Aerospatiale and Dassault. Similarly, in Britain five main aircraft contractors in the
1960s, were reduced to one in 1990, British Aerospace. All of these national champions had to export
to survive. This is the fundamental reason for opposition in CoCom to effective export restraints on
sensitive military technologies. France has been the most liberal, as well as the most determined in its
pursuit of export markets. For example, beginning in the 1970s, France concentrated on designing and
exporting aircraft to the Third World. Exports constituted roughly 70% of production in the 1970s and
1980s (Moravcsik 1990, 68). The trouble is tie aircraft that resulted were not adequate for Europe's
central theater, as even the French Navy complained (Moravcsik 1992, 34). Furthermore, the end of
the Cold War created a buyer's market for weapons systems. In net terms, the market diminished;
and, worst of all. the French failed to capture the loyalty of their Third World clients. Consequently,
"the French arms industry, it is widely agreed, is in crisis" (Moravcsik 1992, 34).
Based on these facts, it is no surprise that Europeans finally embraced collaboration and. thus,
standardization in the late-1980s. No longer able to guarantee their own weapons procurement needs
through strategies like building national champions and exporting weapons, European countries had to
face the fact that national security was being transformed into a collective goods problem, as
Hypothesis Two predicted. While export markets held-up, the shift toward collaboration was slow.
However, when the export market collapsed, collaboration became more important
Thus, until the mid-1980s. the WEU. Eurogroup, and Independent European Programme
Group were like NATO's CNAD--no more than LCDs that symbolically demonstrated resolve to
standardize weapons, but realistically accomplished little else. For its part, the WEU concentrated with
unknown effect on strategic airlift and intelligence and surveillance satellites (Coeme 1991, 19)--
expensive projects not easily exportable. Likewise, the Eurogroup specialized in coordinating primarily
software-oriented military needs, i.e. training, logistics, communications, military medicine, and
operational concepts (Eurogroup 1988, 5). As for the Independent European Programme Group, it did
nothing noteworthy until the issuance of the Vredeling Report in 1986. This led to the endorsement of
an "Action Plan" in 1988, which designated the Independent European Programme Group as the major
organization for coordinating European defense industrial cooperation and proposed a program for
establishing a "common European arms market."
Not coincidentally, the collapse of the Iron Curtain occurred shortly thereafter, destroying
Europe's export objectives and other means of softening the impact of structural disarmament. The
Vredeling Report called for a standardized reporting system for cross-border contracts; aid for the
defense industries of Greece, Turkey, and Portugal; and open competition within Europe for contracts,
subject to the principle of just retour. Perhaps most importantly, the Plan created a European military
research program named EUCLID (endowed with only $140 million to begin with), modeled after die
European Research Coordinating Agency (EUREKA), to advance the objective of pan-European
weapons coordination. Finally, in 1991, the demand of some EC members (primarily France,
Luxembourg, Belgium, and Spain) for a coordinated defense policy became at least a topic of serious
"2Building on the precedent of the Single European Act of 1987, the Treaty on European Union
finalized in Maastricht on 9-10 December 1991, declares as one of its five objectives "the implementation
of a common foreign and security policy including the eventual framing of a common defense policy, which
might in time lead to a common defense." Furthermore, in a declaration attached to the Treaty, WEU
members state that they aim "to develop WEU as the defense component of the European Union." Three
provisions of the Single European Act of 1987 related to weapons cooperation: (1) The preamble pledges
signatories to act with consistency and solidarity for the purpose of protecting their cotmnon interests more
effectively and to make their own contribution to the preservation of peace and security, (2) Article II
broadens the basis for Community-wide weapons procurement programs within the aegis of a common
industrial policy, and (3) Article III promises that signatories will more closely coordinate their political
and economic policies which affect security and maintain adequate technological and industrial conditions
to defend the security of Western Europe (Runmnel 1989, Lexis Service).
All of this suggests that Western Europe made a significant commitment to cooperation in the
early-1990s. The list of collaborative weapons systems that Europe is developing is growing. The
most celebrated of these weapons, the European Fighter Aircraft (EFA), was die choice of Britain,
Spain, Italy, and Germany over the US Advanced Tactical Fighter (ATF) Program. In the early-1990s,
WEU members even set up a "Euro-corps," an extension of the Franco-German brigade of the late-
1980s. Thus, late in life, European arms cooperation looks even more substantial than that taking place
in NATO, as a whole.
The impression outside observers gained in the summer of 1992 was of a shift of
attitude in all major European capitals, away from the rhetoric that had characterized
so many previous West European defense initiatives, towards a concern to build an
effective structure for defense policy coordination, with at least some elements of
military integration. (Menon, Forster, and Wallace 1992, 117)
So far, though, the fruits of collaboration have been minimal. Great Britain and France still
produce, respectively, 75% and 80% of their major weapons systems domestically (Moravcsik 1990,
66). Excessive defense industrial capacity in Europe remains high, too--roughly 27% of total European
defense spending in 1987 (Moravcsik 1992, 36). Furthennore, individual collaborative programs have
received mixed reviews. The European Fighter Aircraft, for example, is rumored to have the same
shortcomings that the Tornado Fighter Bomber did: it performs many roles adequately, but none
deftly. Compromise has eliminated many of the economies-of-scale that the European Fighter Aircraft
promised, too. This is because the European Fighter Aircraft is now not one aircraft, but four.
Britain, Spain, Italy, and Germany plan to test and manufacture their version of the EFA separately.
Thus, the EFA is another good example of LCD weapons procurement cooperation.
More substantial coordination is likely to result in the 1990s, if the Independent European
Prograrmne Group's projects of the late-1980s are carried through. The future of cooperation through
the WEU looks more clouded, though. If die weakness of the EMS (next chapter), and, more
importantly, the divisions in the European Conmmunity which war in former-Yugoslavia made obvious
are a harbinger of things to come, a pan-European common defense may be a long time off. In any
event, further Europeanization of defense faces the steadfast opposition of Britain and the Netherlands,
countries which are strong Atlanticists (Menon. Forster, and Wallace 1992, 105-116).
On the other hand, convergence in the sourcing of weapons components is likely to go on,
regardless. This is the level to which the state has so far not effectively reached and, thus, left the
market to its own liberalizing tendencies as Hypothesis One predicted. Convergence in the sourcing of
components is occurring because of market incentives like economies-of-scale, cost, and quality.
Overall, Europe is even more dependent than the US or Japan on imports of defense components,
especially in electronics and aviation. And this is as it should be, according to Corollary One, which
predicts that convergence will more substantially influence countries with relatively small national
incomes--in this case, middle-range military powers--than large. Hence, in 1989, whereas 30% of
semiconductor purchases were foreign-sourced in the US, 80% was foreign-sourced in Europe; and
whereas less than 20% of most American fighters were foreign-sourced, 30% of the Tornado Fighter
Bomber was American-sourced.
In review, although theoretically Western European countries should be able to address
structural disarmament and, thus weapons standardization more effectively than NATO, they have not
yet to any appreciable extent at the systems level. Structural disarmament has always been a problem
for European countries because of their small defense markets. However, subsidies, concentration, and
exports permitted European countries a degree of self-sufficiency in arms until the late-1980s, when
export markets dried-up and innovations in "smart" weapons continued. Then, the objective of
maintaining an adequate arsenal evolved almost completely into a collective goods problem, as per
Western European countries have responded to this challenge through the WEU, Independent
European Programme Group, and Eurogroup, and tried to coordinate weapons procurement more
closely. But since even European countries have insisted on co-production arrangements that keep their
country's defense contractors in business, the market has not been able to liberalize production to the
extent it naturally would. This situation may be changing; however, so far liberalization and
xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID E3IW8PFD3_OEES4S INGEST_TIME 2013-04-08T20:36:42Z PACKAGE AA00013574_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC