The Supply Side of Security: A Market Theory of Military Alliances 9780804798594

The Supply Side of Security conceptualizes military alliances as contracts for exchanging goods and services. At the int

159 76 3MB

English Pages 256 Year 2016

Report DMCA / Copyright

DOWNLOAD FILE

Polecaj historie

The Supply Side of Security: A Market Theory of Military Alliances
 9780804798594

Citation preview

The Supply Side of Security

se rie s e ditor s Amitav Acharya, Chief Editor American University

David Leheny, Chief Editor Princeton University

Alastair Iain Johnston Harvard University

Randall Schweller The Ohio State University

inte rnational board Rajesh M. Basrur Nanyang Technological University

Brian L. Job University of British Columbia

Barry Buzan London School of Economics

Miles Kahler University of California, San Diego

Victor D. Cha Georgetown University

Peter J. Katzenstein Cornell University

Thomas J. Christensen Princeton University

KhongYuen Foong Oxford University

Stephen P. Cohen The Brookings Institution

Byung-Kook Kim Korea University

Chu Yun-han Academia Sinica

Michael Mastanduno Dartmouth College

Rosemary Foot University of Oxford

Mike Mochizuki The George Washington University

Aaron L. Friedberg Princeton University

Katherine H. S. Moon Wellesley College

Sumit Ganguly Indiana University, Bloomington

Qin Yaqing China Foreign Affairs University

Avery Goldstein University of Pennsylvania

Christian Reus-Smit Australian National University

Michael J. Green Georgetown University

Varun Sahni Jawaharlal Nehru University

Stephan M. Haggard University of California, San Diego

Etel Solingen University of California, Irvine

G. John Ikenberry Princeton University

Rizal Sukma CSIS, Jakarta

Takashi Inoguchi Chuo University

Wu Xinbo Fudan University

Studies in Asian Security Amitav Acharya, Chief Editor David Leheny, Chief Editor

The Studies in Asian Security book series promotes analysis, understanding, and explanation of the dynamics of domestic, transnational, and international security challenges in Asia. The peer-reviewed publications in the Series analyze contemporary security issues and problems to clarify debates in the scholarly community, provide new insights and perspectives, and identify new research and policy directions. Security is defined broadly to include the traditional political and military dimensions as well as nontraditional dimensions that affect the survival and well being of political communities. Asia, too, is defined broadly to include Northeast, Southeast, South, and Central Asia. Designed to encourage original and rigorous scholarship, books in the Studies in Asian Security series seek to engage scholars, educators, and practitioners. Wide-ranging in scope and method, the Series is receptive to all paradigms, programs, and traditions, and to an extensive array of methodologies now employed in the social sciences.

The Supply Side of Security a market th e ory of m i l itary alli anc e s

Tongfi Kim

stanford unive r sity pre ss S t a n f o rd , C a l i f o r n i a

Stanford University Press Stanford, California © 2016 by the Board of Trustees of the Leland Stanford Junior University. All rights reserved. No part of this book may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying and recording, or in any information storage or retrieval system without the prior written permission of Stanford University Press. Printed in the United States of America on acid-free, archival-quality paper Library of Congress Cataloging-in-Publication Data Names: Kim, Tongfi, author. Title: The supply side of security : a market theory of military alliances / Tongfi Kim. Other titles: Studies in Asian security. Description: Stanford, California : Stanford University Press, 2016. | © 2016 | Series: Studies in Asian security | Includes bibliographical references and index. Identifiers: LCCN 2015034987 | ISBN 9780804796965 (cloth : alk. paper) Subjects: LCSH: Alliances—Case studies. | United States—Military relations—Japan. | Japan—Military relations—United States. | United States—Military relations—Korea (South) | Korea (South)—Military relations—United States. | International relations. Classification: LCC JZ1314 .K55 2016 | DDC 355/.031—dc23 LC record available at http://lccn.loc.gov/2015034987 ISBN 9780804798594 (electronic) Typeset by Thompson Type in 10.5/13.5 Bembo

I dedicate this book to my parents, who fostered my interest in international politics.

Contents

Acknowledgments xi   1 Introduction

1

  2 A Market Theory of Alliances

28

  3 A Systemic Theory of the Alliance Market

49

  4 Alliance Politics and Polarity

77

  5 A Domestic Theory of Intra-Alliance Bargaining

112

  6 Case Studies of Domestic Politics and Alliances

123

  7 Conclusion

166

Notes 177 References 203 Index 225

Acknowledgments

I would like to express my gratitude to a number of people for assisting me throughout this project. Most of the research for this book was conducted at Ohio State University and Griffith University. I owe the most heartfelt thanks to Bear Braumoeller, Richard Gunther, Andrew O’Neil, Brian Pollins, Jason Sharman, Randall Schweller, Alexander Thompson, and Daniel Verdier for their advice and encouragement. At various stages of this project, I also benefited from the comments of Quintin Beazer, Daniel Blake,Yuko Kawato, Seth Goldstein, Dane Imerman, Brett Ashley Leeds, Autumn Lockwood Payton, Jeniffer Dabbs Sciubba, Byungwon Woo, and Andrew Yeo, as well as participants of various conferences and workshops. I conducted interviews in Japan and South Korea for this project, and the interviewees gave me not only answers to my questions but also fascinating questions to pursue. In addition to those whose identity I keep anonymous, I thank the following people for their insights and help: Nisuke Ando, Masahiro Akiyama, Robert Eldridge,Yasuhiro Izumikawa, Narushige Michishita, Satoru Mori, Koji Murata, Hiroshi Nakanishi, Tsuneo Nishida, Masashi Nishihara, Tetsuya Nishimoto, Joji Okuhira, Ryo Sahashi, Kazuya Sakamoto, Yasuyo Sakata, Noriyuki Shikata, Naoto Tsuzaki, Thomas Wilkins, Shunji Yanai and Takeshi Yuzawa in Japan, and Cheon Seong-whun, Choi Jong-kun, Choi Kang, Chun Chae-sung, Chung Jae-ho, Kim Ki-jung, Kim Sung-han, Lee Chung-min, Lee Ji-yong, Lee Sook-jong, Moon Chung-in, and Park Joonseong in South Korea.

xii acknowledgments

David Leheny’s advice and warm encouragement guided me throughout the difficult revision process. I am also grateful to Amitav Acharya and the editorial committee for the Studies in Asian Security series; Geoffrey Burn, Jenny Gavacs, and James Holt at Stanford University Press; and Margaret Pinette at Thompson Type for their support and assistance for the book. The anonymous reviewers for the manuscript gave me meticulous and extremely helpful suggestions for revisions. I thank Heiwa Nakajima Foundation, the Mershon Center for International Security Studies, Griffith University, and East-West Center for their financial support of my research. Finally, on the most personal note, I would like to thank my wife, Line, who is the most difficult and helpful reader of my writing and our newborn daughter, who patiently listened to me reading the copyedited manuscript of this book.

The Supply Side of Security

1

Introduction

What are the factors that explain the bargaining power of a state in a potential or existing alliance? This book offers answers to the question by explaining how systemic polarity and domestic politics affect the supply of military commitments (that is, the quantity of military commitments the sellers are willing to provide at various prices) and other concessions through alliances. Looking beyond existing alliance literature that pays attention predominantly to the demand side of the dynamics, this study demonstrates that both demand and supply of military commitments are important in the alliance market and inf luence the bargaining power of states.1 As the alliance literature demonstrates, military threats, both external and internal, create demand for protection through alliances (Liska 1962; Walt 1987; Barnett and Levy 1991; David 1991; Snyder 1997). If no actor faced threats (and there was no demand for military commitment), there would be no military alliance. Military threats have therefore been the central variable in alliance literature. At the same time, however, if no actor was willing and capable of providing military commitment (meaning that there was no supply), there would also be no military alliance. This book sheds light on the supply side of the alliance market by explaining how the structure of the international system and the domestic politics of allies systematically affect the willingness of states to make commitments and concessions to their military allies. Including analysis of the supply side enables a more comprehensive understanding of the relationship between allies and explains recent developments that traditional notions of military alliances cannot adequately explain. Although I attempt to develop a general theory of military alliances in this study, I am most interested in the theory’s implications for the U.S. alliances with Japan and the Republic of Korea (ROK/South Korea). These alliances have been important throughout the Cold War and post–Cold War eras, and their significance will not diminish in “America’s Pacific Century” (Clinton 2011). These alliances face not only North Korea, which itself is a major security threat, but also China, the most important rival of

2 introduction

the United States in the foreseeable future. Although North Korea’s provocative behavior and the rise of China have drawn attention to the role of military threats (a demand-side factor), this study will shed light on the supply side, to gain a more complete picture of the relationship between allies.

U.S. Alliances in East Asia and Beyond On April 29, 2015, Japanese Prime Minister Shinzo Abe (2006–2007, 2012–, Liberal Democratic Party) became the first Japanese leader to address a joint session of the U.S. Congress. Calling the U.S.–Japan alliance “an alliance of hope,” Abe declared that his government would achieve sweeping reforms in security affairs by the coming summer. Two days before Abe’s speech, the two governments had revealed new “Guidelines for U.S.–Japan Defense Cooperation,” which were originally approved in 1978 and revised in 1997. As an anonymous senior U.S. defense official commented, the new agreement “loosens the restriction on what Japan can do” militarily, and the United States and Japan will “be able to do a lot of things globally” that in the past they could do only in the defense of Japan.2 As with Abe’s reinterpretation of Japan’s pacifist constitution in 2014 and new security laws introduced in 2015, these policy changes are interpreted by many as ref lections of Japan’s increased need for U.S. military protection, due to the rising external threats from countries such as China and North Korea. The increased demand for U.S. military protection, however, is only part of the big picture for the U.S.–Japan alliance. In terms of intra-alliance bargaining, Japan has been working harder to accommodate this powerful ally since the end of the Cold War, when military threats decreased considerably. In the 1990s, Russian power significantly declined, and threats posed by North Korea and China were much lower than their current level. Yet, Japan contributed US$13.5 billion to the U.S. efforts in the Persian Gulf War (1990–1991) and also began to increase its military commitment to the United States (Nishihara and Tsuchiyama 1998; Sotooka, Honda, and Miura 2001; Akiyama 2002; Okamoto 2002). By 1996, the U.S.–Japan alliance was announced as “redefined” (Cronin 1996). The early post–Cold War origin of the alliance transformation is puzzling from the perspective of traditional alliance theories that focus on threats. Japanese experts on the alliance whom I interviewed generally agree that Japan now has to work harder to maintain the decades-old alliance.3 As they explain, Japan has to cooperate with the United States in global security issues such as the Iraq War and the War on Terror in exchange for



introduction 3

American support in regional security issues, such as the North Korean threat and the rising power of China. Before the Iraq War began in March 2003, for example, then Prime Minister Junichiro Koizumi (2001–2006) expressed unequivocal support for U.S. policy, stating that it was not in Japan’s national interests to hurt the credibility of the U.S.–Japan alliance.4 This reasoning is based on a demand for U.S. protection. More important from my perspective, Japanese experts also point out that American interests at the global and regional levels were better aligned with those of Japan during the Cold War, and Japan could take U.S. cooperation for granted at the regional level. This close alignment of interests, experts argue, changed after the end of the Cold War. I theorize this change as a shift in the supply of U.S. military protection. Analyses based on national interests of allies implicitly touch on both demand and supply sides of the alliance market, but they obscure the difference between demand and supply. Convergence and divergence of national interests are certainly important, but they vary by issue and f luctuate frequently. This book instead focuses on shifts of bargaining power that are caused by factors beyond mere changes in particulars, namely, systemic polarity and certain elements of domestic politics. Post–Cold War developments in the U.S. alliances need to be understood in the context of both global trends and links between domestic politics and foreign policy. In South Korea, conservative and supposedly pro-American Park Geun-hye assumed the presidential office in February 2013, succeeding another pro-American president, Lee Myung-bak (2008–2013). As a U.S. analyst put it, “One could almost hear a sigh of relief from Washington,” because many “U.S. Korea experts were concerned about a progressive victory” (Revere 2012). At first glance, it would seem that f luctuations in military threats and leaders’ ideology explain the ups and downs of the U.S.–South Korea alliance. Under the presidencies of Kim Dae-jung (1998–2003) and Roh Moohyun (2003–2008), who each had a leftist political base, South Korean perception of the North Korean threat has significantly declined, and Roh Moo-hyun was criticized for weakening the alliance. In contrast, South Korea under the conservative President Lee Myung-bak (2008–2013) experienced the sinking of the South Korean navy ship Cheonan in March 2010 and North Korea’s bombardment of Yeonpyeong Island in South Korea in November 2010; the alliance is said to have then been reinforced against the North Korean threats. It is misleading, however, to explain the alliance

4 introduction

politics of the United States and South Korea simply through the threat of North Korea and ideological divides in South Korea. Since the mid-1990s, South Korea has been wary of becoming entangled in an American conf lict with North Korea, although it still fears being abandoned by the United States. Because Seoul’s alliance policy is no longer solely characterized by the fear of abandonment, we should expect South Korea to take some distance from the United States. Anti-Americanism in the country ref lects this change, but the South Korean elite are nevertheless eager to strengthen the alliance. Former President Roh Moo-hyun, for instance, was widely considered anti-American, but even he made many sacrifices for the alliance. In part, the South Korean government has made concessions to have inf luence on U.S. policy toward North Korea, thereby reducing the risk of military entanglement. There remains the question, however, of why South Korea’s bargaining power seems so limited. Granted, Pyongyang is still unpredictable and bellicose, but South Korea is now clearly less dependent on the United States than it was during the Cold War. The key is to understand the decreased dependence of the United States on its lesser allies. Despite sharing the conservative and pro-American support base with her predecessor, Park Geun-hye’s warm relationship with China so far has made a stunning contrast with that of Lee Myung-bak, who had poor relations with Beijing (Hwang 2014). Personal factors, North Korea’s behavior, and even historical disputes with Japan seem to play significant roles. As I explain in Chapters Three, Four, and Seven, however, the possibility of systemic changes due to the rise of China provides an important context. Moreover, my argument and findings in Chapters Five and Six suggest that domestic politics of South Korea, especially whether Park becomes politically vulnerable, will have a large impact on the country’s relations with the United States and China. There is much to be said about the specifics and particulars of East Asian international relations during the Cold War and the post–Cold War era. This book, however, focuses on more general and theoretical topics of military alliances and then uses the findings to explain the past and to predict the future of the region. For instance, how did the shift from Cold War bipolarity to American unipolarity affect the supply of American military protection in the alliance market? Answers to this question can help us address questions about the future, such as how the dynamics of alliance politics might change if the international system develops into a Sino– American bipolarity or a multipolar system with three or more military



introduction 5

great powers. On the effects of domestic politics, this book offers a theoretical framework to analyze how domestic opposition to cooperation in a state and the ally’s perception of the state’s leader systematically affect the ally’s willingness to make concessions. This is a book about bargaining between military allies. For the operational definition of alliances, I use that of the Alliance Treaty Obligations and Provisions (ATOP) dataset: Alliances are “written agreements, signed by official representatives of at least two independent states, that include promises to aid a partner in the event of military conf lict, to remain neutral in the event of conf lict . . . or to consult/cooperate in the event of international crises that create a potential for military conf lict” (Leeds et al. 2002, 238). This book does not deal with some relationships that are colloquially referred to as “alliances” because they do not entail any of the obligations previously listed. According to the ATOP dataset, for example, Israel ceased to be a formal ally of the United States on December 26, 1991. Based on the definition of the ATOP dataset, many of the “major non-NATO allies” are not military allies of the United States.5 Although military commitment from at least one party is required for a military alliance, functions of alliances have many nonmilitary aspects because they facilitate exchanges of various concessions among states. I conceptualize military alliances as contracts in which states pledge a continuous exchange of goods and services, at least one of which is an obligation to facilitate the military success of the alliance partner(s) in the event of military conf lict. It is important to note that such exchanges can involve goods other than military obligations, and not every ally needs to provide military commitment.

Puzzles of Post–Cold War Alliances The nature of alliance bargaining is inf luenced by both international and domestic factors, but the initial motivation for my theory of alliances comes from puzzles at the system level. The United States currently dominates world security affairs as the only global military power, but alliances formed after the Cold War have been neither initiated by nor directed against the sole military superpower. Additionally, despite pessimistic predictions by some scholars (for example, Mearsheimer 1990; Waltz 1993), many alliances formed during the Cold War have outlasted bipolarity. Meanwhile, although some scholars predict Sino–American bipolarity (Ross 1999; Dempsey 2012; Yan 2011, 2013, 2014), the Chinese alliance policy has not

6 introduction

matched its rising power. China currently has only one alliance agreement with a defense obligation (with North Korea since 1961), and its other two agreements (a bilateral one with Russia and the multilateral Shanghai Cooperation Organization starting in 2001) entail only the obligation of consultation.6 These developments are puzzling for several reasons. First, they contradict the core predictions of the two classic theories of international relations, hegemonic stability, and balance of power. According to hegemonic stability theory (Kindleberger 1973; Gilpin 1981; Keohane 1984), the hegemon maintains global peace and prosperity by providing public goods that sustain a stable international economic and security environment. Chief among these public goods are security guarantees in the form of bilateral or multilateral alliances with the hegemon.7 Therefore, the theory predicts either: (1) a decrease in alliances because hegemonic peace and global governance make them unnecessary; or, conversely, (2) if we observe a high rate of alliance formation, a significant proportion of the new alliances should be initiated by the hegemon to promote systemwide and regional stability. Yet neither of these predictions has come true. A large number of alliances have been formed since the end of the Cold War. Not one of these new alliances was with the United States, although the United States enlarged the North Atlantic Treaty Organization (NATO).8 Balance of power theory fares no better. Its core claim is that states form alliances to prevent or check preponderance, such that none among them can coerce all the rest put together.9 For this to happen, states must align against threatening accumulations of power. Accordingly, if alliances emerge under unipolarity, we would expect some if not most would be directed against the unipole. Yet, there has been little evidence of external balancing against the United States. Those alleging that balancing against the United States is taking place have resorted to the concept of “soft” balancing, conceding that “hard” balancing against the unipole has not occurred to this point.10 Second, many Cold War alliances persist more than twenty years after the Soviet collapse, violating the corollary “balancing” proposition that alliances break apart when the threat disappears.11 According to the ATOP dataset (version 3.0), of the 95 Cold War alliances that were active into the 1980s, 24 ended due to the passing of the Soviet Union, East Germany, and the former North and South Yemen. Of the remaining alliances, more than half (36 of 71) were still active as of December 31, 2003.12 So much for the dictum that “success is the greatest enemy of alliances.” Moreover, if one



introduction 7

considers the number of alliances terminated for each active alliance in a given year, the end of the Cold War did not have much impact on alliance termination (Figure 1.1).13 Third, the dramatic increase in alliance formation also contradicts common expectations about the post–Cold War era. Out of the 538 alliances formed in the period between 1815 and 2003, 163 were formed between 1990 and 2003 (Figure 1.2). Indeed, the rate of alliance formation remains at one of the highest levels in history, even when controlling for the substantial increase in the number of states and politically relevant dyads (Figure 1.3). Again, if alliances are responses to military threats, then the relatively peaceful post–Cold War environment should have led to a decrease in alliances. Even at the subsystemic level, where the concerns of many states may be related only to their close neighbors, the capability aggregation model at the core of balance-of-power or balance-of-threat logic does not explain the surge in alliance formation.14 After all, threats of interstate wars are declining in most regions of the world (Mueller 2004). These trends simply do not square with the traditional view of military alliances as a method of power aggregation to balance and deter the enemy or, if that fails, to fight and defeat it on the battlefield. Fourth, these anomalous empirical patterns are further complicated by an important and conspicuous exception to the current U.S. trend of “no new” alliances: NATO has not only survived but expanded, admitting twelve Central and Eastern European states after German reunification in 1990 and the end of the Cold War.15 Although the new members bring some military power and strategic bases to the alliance, this scarcely offsets the significant costs imposed on the old members by these new military commitments. As Michael Altfeld points out, “Some alliances will add nothing at all to a nation’s security or even reduce its security by placing it in a more vulnerable position than it was in before it chose to join the alliance” (1984, 525). From a military security perspective, new members of NATO are not considered strategic assets but rather burdens (that is, they supply much less military value than they consume).16 The economic and military resources of the new members are limited, and the financial cost and the military risk of the enlargement for the old members are substantial, even when we ignore the opportunity costs incurred in aggravating tensions with Russia.17 We must infer, therefore, that the value the United States has sought to advance through NATO expansion must be something other than increasing its own security through capability aggregation.18 From the perspective

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1815 1819 1923 1827 1831 1835 1839 1843 1847 1851 1855 1859 1863 1867 1871 1875 1879 1883 1887 1891 1895 1899 1903

1987

1983

1979

1975

1971

1967

1963

1959

1955

1951

1947

1943

1939

1935

1931

1927

1923

1919

Figure 1.1.  Number of alliances terminated per number of active alliances in the system, 1815-2003.

1999

1995

1991

1915

1911

1907

note: For Figure 1.1, the number of active alliances in a given year equals the number of active alliances in the previous year plus the number of new alliances formed throughout the current year, minus the number of alliance terminations in the previous year. I do not count alliances formed before 1815, as very few of those remained active beyond 1815, according to the ATOP codebook. Gibler (1999) reports only three such cases.

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

2003

0

5

10

15

20

25

30

35

40

1815 1819 1923 1827 1831 1835 1839 1843 1847 1851 1855 1859 1863 1867 1871 1875 1879 1883 1887 1891 1895 1899 1903

0

5

10

15

20

25

30

35

40

1907 1911 1915 1919 1923 1927 1931

Figure 1.2.  Number of alliances formed, 1815-2003.

1935 1939 1943 1947 1951 1955 1959 1963 1967 1971 1975 1979 1983 1987 1991 1995 1999 2003

0

0.005

0.01

0.015

0.02

0.025

0.03

0.035

0.04

0.045

0.05

1816 1819 1822 1825 1828 1831 1834 1837 1840 1843 1846 1849 1852 1855 1858 1861 1864 1867 1870 1873 1876 1879 1882 1885 1888 1891 1894 1897 1900 1903 1906

1990

1987

1984

1981

1978

1975

1972

1969

1966

1963

1960

1957

1954

1951

1948

1945

1942

1939

1936

1933

1930

1927

1924

1921

1918

1915

Figure 1.3.  Number of alliances formed per number of politically relevant dyads in the system, 1816-2003.

1999

1996

1993

1912

1909

note: I used EUGene’s (version 3.204) default option for the number of politically relevant dyads (http://eugenesoftware.org/). See Bennett and Stam (2000) for EUGene.

0

0.005

0.01

0.015

0.02

0.025

0.03

0.035

0.04

0.045

0.05

2002

14 introduction

of this study, the goal was to prevent the emergence of another supplier of security in Europe. As the Defense Planning Guidance circulated within the George H. W. Bush administration put it, the United States in the post– Cold War era aimed to “maintain the mechanisms for deterring potential competitors from even aspiring to a larger regional or global role.”19 Finally, China—a rival of the United States—has a post–Cold War alliance policy that is surprisingly limited, given its elevated status in the international system and the increasingly competitive Sino–American relations. China has been traditionally distancing itself from formal military alliances, and it is the least active in alliance politics among the five permanent members of the United Nations. One simple reason for the inactivity is that few, if any, countries would be willing to ally with China (Zhang 2012, 140). However, it is interesting to ask why other states are unwilling to ally with China and, more important, why China itself seems reluctant to pursue more allies. The U.S.–Soviet bipolarity was associated with the divisions between two large groups of alliances; one wonders what alliance politics under Sino–American bipolarity might look like. I explain these puzzling developments by means of a market theory of military alliances that focuses on exchanges and contracts between allies. The reconceptualization of military alliances leads to further questions and findings about different aspects of alliances. I examine my argument using the Alliance Treaty Obligations and Provisions dataset (ATOP, version 3.0) and the cases of America’s alliances with Japan, South Korea, and Spain.

Theoretical Framework: The Alliance Market Exchange in the Alliance Market This study develops a market theory of military alliances that treats alliances as contracts pledging a continuous exchange of goods, at least one of which is a military obligation, but the others need not be.20 The model bases alliance politics in the demand for and supply of various goods; important examples include protection, military bases, and political and economic concessions. The capability aggregation model of alliances comprises a subset of this theory, with the exchanged goods limited to military force. The central assumptions of my theory are that states and domestic actors participate in military alliances for various goals and that the common function of all alliances is to generate efficiency gains. With these assumptions, my theory explains how systemic polarity affects state bargaining power (for example, U.S. intra-alliance bargaining power increased after the Cold War



introduction 15

as the number of security suppliers declined) and how domestic politics affect intra-alliance bargaining (for example, the relationship between a leader’s political vulnerability and state bargaining power). As with international trade, alliances assist states in obtaining efficiency gains from exchanges of goods based on comparative advantage (Boyer 1993) and economies of scale (Hartley 2006).21 For example, many criticize the U.S.–Japan alliance for its asymmetric assignment of responsibilities. Yet both the United States and Japan have benefited greatly, with the United States offering military protection and Japan providing military bases and economic contributions. Japan can supply security for itself, but the gains would be outweighed by the domestic political costs and adverse effects on Japan’s relationships with its neighbors. Thus, efficient exchange has been continuing for more than six decades. Moreover, this division of labor is common among traditional military alliances as well: States have always formed alliances for efficiency gains (Palmer and Morgan 2006, 142).22 In the alliance market, the most prominent good is military protection, and its price relative to other goods is a major source of bargaining power for a security exporter. As Joseph Nye points out, “Provision of protective force can be used in bargaining situations” (Nye 1990, 31). Some aspects of security production from alliances can be explained by a public goods model (Olson and Zeckhauser 1966); military commitment, however, is actually a good that is rivalrous (subtractable) and excludable. Military force committed to the defense of a state reduces the amount of force available to others, and the provider of the force has the freedom to withhold its support to its allies. Military commitment, therefore, is better treated as a private good in the international system.23 Although subsidy treaties, in which one state loans its troops to another state for money, have become rare, states still actively exchange military for nonmilitary goods.24 Figure 1.4 shows demand and supply curves for security. The demand for military protection, which derives from military threats confronting states, increases with the number of buyers. The supply of military protection derives from the state’s own military capability and the expected support from other states; such supply increases with the number of suppliers.25 When demand for security increases (from D to D′), ceteris paribus, the relative price of security increases (P0 to P1, the security exporter’s bargaining power increases; the security importer’s bargaining power decreases). Similarly, when the supply of security—the quantity of military force available to the buyers at various prices—decreases (from S to S′), ceteris paribus, the price of security increases (P1 to P2).

16

introduction

P2

Relative price of security

P1 S’

P0

S

D’ D

Quantity of security produced relative to nonsecurity goods

Figure 1.4. Demand and supply of security.

In other words, increased demand or decreased supply of military protection improves the terms of an alliance for an ally that exports military protection. With the Sino–American rapprochement during the Nixon era, for example, American bargaining power increased vis-à-vis Japan. Military threats to Japan declined in the period, but demands for American military protection arguably increased as China sought a quasi-alliance with the United States against the Soviet Union (Zhang 2012, 129). Meanwhile, the supply of military protection for Japan was perceived to be declining because of the Nixon Doctrine.26 Over the last two decades, both the North Korean threat and China’s rising presence increased the demand for U.S. military protection and improved the bargaining position of the United States vis-à-vis Japan.27 The North Korean nuclear threat has become more substantial, with nuclear tests in 2006, 2009, and 2013. China’s military expenditure in 2014, estimated to be 8.6 times larger than in 1994, is 3.2 times Japan’s military expenditure in 2014 (SIPRI 2015, in constant 2011 U.S. dollars). In the last several years, the U.S. bargaining position further improved because of tensions between Japan and China over the Senkaku/Diaoyu Islands. For instance, many believe that the Japanese government refrained from cutting its financial support to the U.S. forces in Japan in 2011 because of the crisis resulting from an incident involving a Chinese fishing boat and Japanese coast guard vessels in September 2010 (Heginbotham, Ratner, and Samuels 2011,



introduction 17

145). However, I argue that long before these developments the supply-side dynamics had already increased U.S. bargaining power, with the end of the Cold War. Despite continuing tensions with North Korea, South Korea’s need for U.S. military protection decreased after the end of the Cold War. Granted, North Korea’s nuclear weapons development is a significant factor, especially in more recent years, but North Korea’s conventional military capability has declined, and Seoul’s relationships with Beijing and Moscow have significantly improved (Oberdorfer 2001). These developments on the demand side of the alliance dynamics lead us to expect increased bargaining power for South Korea, which receives American military protection. South Korea’s bargaining position should also have improved in the supply side, as the country’s economic growth has increased its domestic supply of military force. As will be discussed in Chapters Three and Four, however, other things, especially the international supply of military commitments, were not held equal over time. When the number of suppliers who produce a good decreases, the supply of the good declines and its price increases. The effects of reduced competition among suppliers are perhaps best illustrated by North Korea’s predicament after the collapse of the Soviet Union.28 North Korea’s economic collapse in the 1990s is fundamentally due to its own economic mismanagement, but withdrawal of material support from the two patron states brought its problems to the surface.29 Before then, North Korea reaped bargaining advantages from the competition between the two neighboring communist giants.30 With the Soviets gone, however, the Chinese had few reasons to worry about abandonment by North Korea. As a result, the value of North Korean political concessions declined. Given that the material concessions Pyongyang could make for Chinese support were always limited, Pyongyang lost most of its bargaining leverage, except for its nuclear weapons development and the increasing threat of its total collapse that would produce regional chaos. China’s support to North Korea was significantly reduced in the early 1990s (Oberdorfer 2001, 243–248), and the credibility of their military alliance has also declined. Although Article 2 of their 1961 alliance treaty guarantees automatic assistance and intervention, “China conceives itself to have the right to make an authoritative interpretation of the principle for intervention” (Choo 2008, 370). In the context of local international politics, phenomena discussed in this study may sometimes appear to be caused simply by shifts of military threats. The term threat is often related to both the demand and supply sides of military goods, but I exclude the supply side from the calculation of

18 introduction

threat. A country could feel less secure because of either increased capabilities of potential enemies or reduced support from allies, but only the former is related to the demand for military protection. When threat is used as an umbrella term, it hides more than it uncovers about what actually motivates certain behaviors. Once we exclude the supply of military protection from the assessment of military threats, the effects of systemic polarity are distinct from the effects of these threats. Whereas demand for security is, for most states, largely determined by local situations, the availability of an alternative security supplier is more affected by systemic changes, that is, the rise and fall of great powers. As will be discussed later, the underlying assumption is that, in most situations, great powers are the only effective suppliers of military protection.31 The alliance market is far from perfectly competitive: A few important players enjoy disproportionate inf luence. Accordingly, I treat great powers as the only effective security suppliers in the alliance market. Just as the number of sellers affects commercial transactions, the number of great powers or systemic polarity (Waltz 1979; Snyder 1997) inf luences how goods are exchanged among states.32 Although the quality of goods and the locations of suppliers matter as well, the number of suppliers systematically affects the level of competition among them and thus their willingness to provide military commitments. Because polarity plays a significant causal role in the structure of dependence among potential and actual allies, changes in international structure generate predictable variations in intra-alliance bargaining power and general alliance patterns such as formation, duration, and mode of termination.33 The end of the Cold War changed the global distribution of military threats and commitments. This shift had two contradictory effects on the bargaining power of the United States: (1) The demand curve shifted to the left (look again at Figure 1.4), reducing the bargaining power of a security supplier, but (2) the supply curve also shifted to the left, increasing the price of security. When the former Eastern Bloc collapsed, its militaries had neither the intent nor the capabilities to remain a threat to the West. Meanwhile, in regions outside Europe, communist regimes and rebels lost their most important ally, and actual and potential U.S. allies lost their leverage toward the United States afforded by the competition between two superpowers. Where demands for military protection decreased (for example, Western Europe, South Korea), the intra-alliance bargaining power of a security exporter, such as the United States, should have



introduction 19

decreased. At the same time, however, the shift from bipolarity to unipolarity increased the bargaining power of a security exporter by decreasing the supply of security. I argue that the leftward shift of the supply curve was more significant than the leftward shift of the demand curve: hence, the increased rather than decreased bargaining power of the United States after the Cold War. As the unipolar power became more reluctant to supply security, states that import military protection preserved their existing alliances or formed new ones. A reduction in the demand and supply of military protection, however, should lead to a decline in the overall production of military protection relative to other goods in the alliance market. Chapter Four demonstrates that this has indeed occurred in the current unipolar system, despite the large number of new alliances. The concepts and problems discussed in this study build on the literature on exchanges in alliances. Drawing on Altfeld (1984), Morrow (1987, 1991, 1993, 2000) employs the concept of exchange and argues that states exchange security and autonomy in alliances.34 Unlike in symmetric alliances, smaller partners in asymmetric alliances cannot contribute much to capability aggregation. Hence, Morrow argues, it is common that smaller partners concede autonomy to their larger partners, whereas the more powerful states supply security to the smaller partners. Boyer (1993) analyzes the roles of nonmilitary goods and comparative advantage in exchanges between allies, although his model was designed to criticize the public goods model of alliances and did not directly engage the rest of the alliance literature. Building on these earlier works on exchanges between allies, the market theory of alliances extends the literature to a systematic analysis of intra-alliance bargaining. Nonmilitary Goods in the Alliance Market To understand the rationale for asymmetric alliances and the persistence of alliances after dramatic international structural change, we need to take into account nonmilitary goods in the alliance market.35 In 38.8 percent of allied dyad-years between 1816 and 2003, one alliance partner was more than ten times stronger than the other in terms of the Composite Index of National Capability (CINC) score of the Correlates of War project (version 4.0)—a disparity that casts doubt on the capability aggregation model of alliances.36 These asymmetric alliances make sense, however, once we relax the assumption that alliances are limited to the exchange of military force

20 introduction

and allow for the exchange of heterogeneous goods. Such exchanges are not confined to the military field (for example, military protection in exchange for military bases), and the provision of military bases alone cannot explain them, as only a limited number of weak states are so strategically located that the value of their military bases can be used to extract costly commitments from stronger states. Because states pursue various goals, they are sometimes willing to exchange military commitments for concessions in nonmilitary fields. An imbalance in military capabilities between states does not mean that they cannot find a mutually acceptable exchange as the basis of an alliance. In a similar vein, alliances do not always collapse after the threats that motivated them disappear. Even when the capabilities and strategic goals of allies change, exchanges of heterogeneous goods allow them to keep benefiting from the alliance. Exchanges of military goods facilitate nonmilitary cooperation, and vice versa.37 For example, “One reason the United States and Japan were able to weather the trade disputes of the 1980s with their relationship intact was the underlying foundation of the security alliance” (Finnegan 2010, 13). Nonmilitary cooperation, in turn, is expected to promote military cooperation. A major motive of and purpose for the South Korean government in signing the Free Trade Agreement (FTA) with the United States in 2007 was to reinforce the U.S.–South Korea alliance (Heo 2008; Sohn and Koo 2011). As Japanese Prime Minister Abe put it, the Trans-Pacific Partnership (TPP) “goes far beyond just economic benefits. It is also about our security. Long-term, its strategic value is awesome.”38 Nonmilitary goods include economic concessions in trade and finance as well as political concessions in diplomatic and domestic affairs. Some alliances in earlier periods were blatant exchanges of money and troops; for example, Portugal literally bought its alliance with England in 1661 (Belcher 1975); through subsidy treaties, England, in turn, paid the monarchs of continental powers (for example, Russia and Prussia) to hire their troops to fight for England’s interests (Schweizer 1989). Similar economic incentives have motivated the formation of many modern alliances. During the interwar period (1920–1936) for instance, France offered commercial concessions to gain an alliance with Belgium (Long and Leeds 2006); Nazi Germany used its economic power to ally with Bulgaria, Hungary, and Romania (Hirschman [1945] 1980). Economic goods played important roles even in alliances that seem to be based on common threats, such as the Franco–Russian alliance before World War I (Viner 1929, 410–411) and the Anglo–Japanese alliance of the early twentieth century (Davis 2009).



introduction 21

With respect to political goods, every alliance entails some exchange of political loyalty, which serves as the foundation of military cooperation. Political concessions made by allies, of course, are never perfectly symmetrical, and stronger allies tend to have inf luence over their weaker ­partners’ domestic and foreign policies. During the Cold War, the two superpowers provided military protection and economic aid to their respective junior partners, and the latter conformed their foreign and domestic policies to those of the bloc leaders.39 U.S. Major General Henry Stackpole pointed out that the U.S.–Japan alliance served as “the cap in the bottle,” preventing a resurgence of Japanese military threats, but the alliance did much more. The bilateral alliance made sure that Japan would follow U.S. policy and support the East Asian regional order that the United States designed (Pyle 2010, 6). South Korea, with its even deeper dependence on the United States, shaped its domestic and foreign policy in accordance with the U.S. policy, although the process was by no means frictionless (Oberdorfer 2001). Effects of Polarity As already mentioned, polarity inf luences how goods are exchanged among states.40 In a multipolar system, a security importer can satisfy its demand for military protection through many alliance options.41 In a bipolar system, a threatened nonpolar power can seek help from either or both of the two superpowers and, if refused by one of them, can turn to the other for security (albeit with less bargaining power than if the other superpower was also an option).42 As will be explained in more detail in Chapter Three, a great power’s bargaining power vis-à-vis a lesser power is expected to be stronger under multipolarity than under bipolarity; the two great powers under bipolarity can export military commitments only to their lesser allies, whereas great powers under multipolarity can ally with their fellow great powers. Gowa (1989) argues that the cost of exit from an alliance is higher under bipolarity than under multipolarity, but the increase in the cost is steeper for great powers than for nongreat powers. In a unipolar system, states have less bargaining power vis-à-vis the most likely supplier of military force, the unipolar power, which enjoys a virtual monopoly in the security market. From the perspective of this study, the absence of traditional balancing against the United States is not particularly puzzling or surprising (cf. Ikenberry 2002). The shift from bipolarity to unipolarity gave the United States huge advantages in the alliance market by increasing its relative capability and reducing the attractiveness

22 introduction

of ­alternative allies; the lack of balancing against the United States simply ref lects its vaulted position within the new, unipolar structure of this market. Even if the United States was not too strong to balance against (cf. Wohlforth 1999), other states are likely to pursue an alliance with the United States because of the benefits provided through the exchanges. These advantages explain the increased use of coalition building in U.S. foreign policy. To return to the marketplace metaphor, the use of ad hoc coalitions in lieu of formal alliances resembles the use of part-time workers in place of full-time employees. When the alliance market favors the employer (the United States), it can attract employees ( junior partners) without making formal alliance commitments. Thus, we observe neither the formation of counteralliances against the United States nor new alliances being initiated by it. Domestic Politics My argument starts with an assumption of states as unitary rational actors, but, in reality, preferences of inf luential domestic actors can diverge significantly from the aggregated interest of the people within the state. For instance, according to the Japanese daily newspaper Asahi Shimbun, U.S. diplomatic cables released by WikiLeaks have shown that there are bureaucrats, politicians, businesspeople, and academic experts in Japan who have a strong preference for cooperation with the United States. Japanese political scientist Hideki Kan calls this group “Nichibei Anpo Mura (the village of the US–Japan security treaty).”43 This book does not use WikiLeaks materials, but the analysis of these documents by Asahi Shimbun suggests a strong conf lict of interest within the Japanese government over negotiations with the United States.44 Some Japanese bureaucrats advised American negotiators not to be soft toward the leaders of the Democratic Party of Japan (DPJ), who took power in 2009.45 Similar conf licts were present in South Korea’s Roh Moo-hyun government, where the leftist president’s foreign policy stance made the foreign policy establishment anxious. From international factors, we can estimate a range of bargaining outcomes that are beneficial to all alliance partners conceived as unitary actors. Where within (or without) that range a specific bargaining outcome lands is mostly determined by the allies’ domestic politics and their interests regarding specific alliance-related issues. Because the latter varies issue by issue, a generalizable theory is better served by looking in the former for systematic



introduction 23

patterns that affect the willingness to concede in intra-alliance bargaining.46 It is important to state that the argument I develop is not a mere description of proalliance forces strengthening an alliance and less cooperative groups obstructing alliance cooperation. Among others, my domestic theory of alliance bargaining proposes that politically vulnerable procooperation leaders and politically nonvulnerable anticooperation leaders have advantages in intra-alliance bargaining. The domestic politics of a state naturally affect its own alliance policy, but because of the nature of the alliance market—which creates the expectation of repeated cooperative interactions—domestic politics of a state also inf luence the ally’s willingness to make concessions. I argue that three factors are particularly important: domestic opposition to the alliance cooperation and both the state’s leader’s vulnerability and his or her attitude toward the alliance (discussed in Chapter Five). Changes in regime type and leadership are so consequential in alliance politics (cf. Leeds and Savun 2007; Leeds, Mattes, and Vogel 2009), partly because they affect the three variables. In addition to these factors, state resources such as military capability and wealth also affect alliance terms. Specifically, resource scarcity may engender credible domestic opposition that inhibits the weak state’s ability to grant alliance concessions; here, weakness in terms of material power translates into bargaining strength. Conversely, an abundance of resources may weaken the materially rich state’s bargaining power over allies. Multiple Aspects of the Alliance Market Although my market theory of alliances is motivated by puzzles at the systemic level, the theory can be extended to the analysis of the contents of alliance contracts and domestic politics. Because efficient exchanges between allies require the protection of alliance contracts, the market theory should deal with contractual aspects of alliances. Bargaining discussed in this book is closely related to how actors design and negotiate alliance agreements. Because I have partially addressed the issues surrounding the design of alliance agreements elsewhere (Kim 2011), this book touches on the subject only brief ly. My “domestic” level of analysis includes both first (individual) and second (state) image variables (Waltz 1959). When we examine transactions in the alliance market, we see that the actors who pursue benefits from military alliances are subnational groups and individuals. British historian Ian Nish, for example, describes the Anglo–Japanese Alliance of 1902–1922

24 introduction

as a secretariat alliance, formed by particular interests of certain ministries within the two governments (Nish 1972). Similarly, “Washington’s mutual defense agreements with Franco’s Spain and Chiang Kai-shek’s Taiwan were created by and for particular interests within the armed services and powerful legislators on Capitol Hill” (Dingman 1979, 249). Moreover, internal politics and leaders’ attitudes affect not only a state’s alliance policy but also that of its ally because the ally adjusts its policy according to its perception of these variables. The key point is that alliance bargaining is shaped by factors from different aspects of the alliance market: (1) the systemic variables (such as polarity, demand and supply of military protection), (2) contents of the alliance contracts, and (3) the domestic factors (for example, regime type, leaders’ attitudes toward specific alliances). My market theory of alliances connects the three aspects with the common theme that alliances are contracts of exchange by which actors pursue their interests. An explanation based solely on one aspect of alliance politics is partial at best. For example, the fear of abandonment and entrapment (Snyder 1984, 1997) is affected by systemic polarity, but it is mitigated or amplified by the designs of alliance agreements and also by domestic political climates of allies (for example, proalliance groups tend to worry about abandonment, whereas neutralists worry about entrapment or entanglement). The future of America’s alliances will depend on international factors (such as military threats, American unipolarity), the contents of the current alliance agreements, and domestic political developments in the United States and its allies. In short, variables from different aspects of the alliance market interact with one another, although they are separately addressed in the present study for analytical purposes. Systemic factors affect the conditionality of alliance agreements, and they also have a significant impact on the domestic politics of alliance members.47 The specific contents of alliance contracts, because they determine the benefits and risks of such agreements, inf luence how leaders and the public view the alliance and can sometimes override the effects of systemic variables. Domestic political variables affect intraalliance bargaining and the contents of alliance contracts; they may also have long-term effects on systemic factors because changes in capabilities and threats often originate in domestic politics.48 Therefore, this study aims to develop a theory of alliances that cuts across levels of analysis as well as the divide between the fields of international security and international political economy.49



introduction 25

Cases and Methods I use the ATOP dataset to show trends of alliance politics since 1815, as well as to test some of my arguments. My focus is explicitly on formal military alliances, though I argue that they have nonmilitary aspects as well. For quantitative analysis, I use 538 military alliances formed between 1815 and 2003. These agreements include defense, offense, neutrality, or consultation obligation. I exclude from my analysis nonaggression pacts without other alliance obligations because many of them are agreements between enemies rather than allies (for example, North and South Korea; the United States and the Soviet Union).50 Although there are important differences among defense, offense, neutrality, and consultation obligations, these military commitments all facilitate the success of the allies in the event of military conf lict. I demonstrate the causal mechanisms of my arguments primarily through case studies of America’s alliances with Japan (1951–), South Korea (1953–), and Spain (1963–). These three alliances are not only substantively important but also theoretically interesting, individually as well as collectively. The U.S.–Japan alliance consists of two of the largest national economies in the world and yet is based on extremely asymmetric military responsibilities. The U.S.–South Korea alliance is one of the most militarized alliances in the contemporary world, but its origin and development are hardly captured by the capability aggregation model of alliances (Kim 2011). In addition to providing a comparison with America’s East Asian alliances, the U.S.–Spain case shows that an alliance can really be a marriage of “convenience” and not even one of necessity: Neither the United States nor Spain really needed the alliance to fight external threats but benefited from their loose cooperation, which gradually developed into an institutionalized alliance. Collectively, the three alliances have significant variations in such factors as geography, regime types, and external threats. In addition to the quantitative analysis of the ATOP dataset, the Spanish case shows that the pressures and issues discussed in this book are not simply about the murk and mess of Asian international relations. There are also diplomatic records of the negotiations between the United States and each of the three allies, with which we can directly examine the variables of interest rather than infer them from the circumstances. For the primary sources, I use Foreign Relations of the United States and have collected documents at the National Archives at College Park. For

26 introduction

the secondary sources, I primarily use works in English but also some in Japanese and Korean. I have conducted interviews in Japan (2006, 2008, 2012) and South Korea (2008, 2012) for this project. Outline Building on the concept of the alliance market, the following chapters develop a general theory of military alliances and apply it to U.S. alliances in East Asia. Focusing on the characteristics of the alliance market, Chapter Two presents the broad theoretical map and explains important concepts in the market theory of alliances. The chapter justifies the market approach of this project and explains how my theory is related to the existing literature. Chapter Three models the effects of systemic variables on the relationships between allies, and Chapter Four presents empirical support for my argument. My systemic model focuses on the demand and supply of military commitments and how polarity affects the bargaining power of great powers. The existing literature is heavily biased toward the analysis of demand for security and does not pay sufficient attention to the logic on the supply side. In addition to its focus on the supply side, my model is novel in its deductive reasoning about the effects of polarity. The model analyzes how the number of great powers affects the fear of abandonment and entrapment/entanglement. This fear shapes bargaining positions of allies and affects the patterns of formation, duration, and termination of alliances in different systems. Deductive reasoning is particularly important for our analysis of polarity because shifts of polarity are infrequent and are also correlated with other variables. As others have pointed out, we need to distinguish the effects of unipolarity from other characteristics of the contemporary world (Ikenberry, Mastanduno, and Wohlforth 2009, 25–26; Jervis 2009, 200–206). Similarly, inductive reasoning based on U.S.–Soviet relations is not an adequate basis for analyzing possible Sino–American bipolarity in the future. Chapter Five explains the effects of three domestic variables on a state’s intra-alliance bargaining power, and Chapter Six supports the argument with case studies. As the existing theories of international relations argue, the presence of effective domestic opposition to cooperation increases a state’s bargaining power in an alliance. The chapters advance our understanding of the domestic sources of bargaining power by further explaining disordinal interaction effects of a leader’s attitude toward an alliance and his or her vulnerability. When a leader is likely to stay in power for a long time, the other state either has to accept concessions to the tougher demands of an



introduction 27

anticooperation leader or does not need to prop up a procooperation leader. In contrast, when a leader’s rule is shaky, the other state’s likely response is either to undermine an anticooperation leader by withholding concessions or to support a procooperation leader by making concessions. Thus, the effects of a leader’s attitude and vulnerability vary, depending on the value of each variable; vulnerable procooperation leaders and nonvulnerable anticooperation leaders have bargaining advantages in intra-alliance bargaining. In addition, these two chapters address the difference between bargaining power based on strength and bargaining power based on weakness. Chapter Seven concludes the study by discussing this book’s implications for international relations theory and policy of the United States, its East Asian allies, and China. In particular, the chapter emphasizes that the rise of China affects both the demand-side and supply-side factors in the alliance market.

2

A Market Theory of Alliances

It is impossible to speak of international relations without referring to alliances; the two often merge in all but name. For the same ­reasons, it has always been difficult to say much that is peculiar to alliances on the plane of general analysis. George Liska (1962, 3) We are going to do something terrible to you—we are going to deprive you of an enemy. Georgi Arbatov, the Soviet Union’s top foreign policy expert, addressing Americans before the end of the Cold War (quoted in Talbott 1988)

Arbatov’s statement captures an important element of the alliance market because an enemy creates demand for military commitment. As I show in Chapters Three and Four, however, the end of the Cold War hurt the allies of the United States rather than the remaining superpower. The end of the Cold War reduced the levels of military threats in many parts of the world, but it also removed competition from the supply side of the alliance market. Under unipolarity, U.S. allies needed to prove their worth to the United States, not the other way around. To understand this, we need to see military alliances from a new angle. I conceptualize military alliances as contracts in which states agree to a continuous exchange of goods and services, at least one of which is a military commitment. Although members of alliance agreements typically have some form of military commitments in formality, the real contributions of many alliance members lie outside the provision of military force. The goals of alliances vary across cases and time, but states participate in military alliances to obtain efficiency gains, regardless of their specific goals. With this key assumption, a market theory of alliances should explain intra-alliance bargaining as well as the formation, duration, and termination of alliances. There are many possible ways to build such a theory, but a market theory of alliances should be based on the characteristics of the alliance market.



a market theory of alliances 29

First, there is no “currency” in the alliance market, and alliance partners exchange goods without a medium of exchange. The most commonly exchanged good in the alliance market is military security. The exchanges, however, should not be limited to that single good because the market participants have other needs and also because not everyone can provide the good. We should therefore analyze the role of nonmilitary goods in the alliance market. Second, the alliance market has a limited number of participants and is far from perfect competition. Especially important is the limited number of actors who can supply effective military protection. Thus, I focus on the effects of the number of great powers (that is, polarity) at the systemic level. Third, payments in the alliance market are not simultaneous, and there is no world government that enforces alliance agreements. Because the market participants face the risk of opportunism and inadvertent failures, we should analyze how states design alliance agreements to cope with the risk.1 Finally, the variety of potential stakeholders in the alliance market is higher than that in corporations in commercial markets. Individuals pursue their self-interest in commercial enterprises, too, but they are more likely to agree on what is good for their companies (for example, maximizing monetary profits). Because national interests are more multifaceted than the interests of commercial enterprises, the interests of substate groups and individuals in the alliance market are more likely to be diverse, diverging from their representative. Thus, we should analyze how domestic politics affect transactions between allies. This chapter provides a guide to the general ideas of my theory and illustrates where in the literature my theory is placed. Some works from major paradigms of international relations theory (realism, institutionalism, liberalism, constructivism) have explained the continuation of the Cold War alliances, but none of them explains the increase of alliance formation and the inactivity of the United States after the Cold War.2 My theory of the alliance market is in the tradition of institutionalism in its focus on efficiency gains, but variables from neorealism (that is, systemic polarity) and neoclassical realism or liberalism (that is, domestic political variables) also play important roles in it.3 In the following passages, I first explain why the market approach is necessary for a sound and comprehensive understanding of alliances. Then I lay out the basic concepts for my theory of alliance market, with a brief explanation of my arguments at the systemic and domestic levels.

30

a market theory of alliances

Why a Market Approach? Alliance Games There is no condition that deserves permanently the name either of friendship or hostility. Both friends and foes arise from considerations of interest and gain. Friendship becomes changed into enmity in the course of time. A foe also becomes a friend. Self-interest is very powerful. From the Mahabharata, an epic of ancient India (Ganguli 2009, 410)

As in the majority of alliance literature, I assume that states, and their leaders, pursue self-interest, but I find in the literature a bias toward associating self-interest with hostility rather than with cooperation. For instance, George Liska (1962, 12) writes that “alliances are against, and only derivatively for, someone or something.” This kind of thinking is the reason behind Georgi Arbatov’s comment quoted at the beginning of this chapter. In major theories of international relations, states form alliances to balance against power (Waltz 1979) or against threats (Walt 1987). Even Glenn Snyder, who analyzes both interactions between allies and interactions between adversaries, defines alliances as “formal associations of states for the use (or nonuse) of military force, in specified circumstances, against states outside their own membership” (1997, 4). As a result, alliance literature tends to emphasize conf licting rather than cooperative elements of international relations, even though alliances are fundamentally cooperative institutions. Alliances and friendship are different concepts—the latter is less formal and usually involves normative connotations rather than mere self-interest.4 The two concepts, however, share certain properties as well—actors cooperate better with each other within the clubs of alliances or friendships, and the insiders value good relationships among them. Students of international politics have been careful not to confuse alliances with friendships because they are advised to focus on the self-interest of international actors. The existing literature on alliances has been trapped in the other end of errors—alliances have been analyzed predominantly in terms of strategic interactions between adversaries. Granted, self-interest is more conspicuous in the context of enmity than in friendship, but one still should not uncritically link alliances with enmity. Your enemy’s enemy is not necessarily your ally (for example, Saddam Hussein and Osama Bin Laden, enemies of the United States, were not allies), and your ally’s enemy is not necessarily your enemy (for example, Greece and Turkey, both NATO allies of the United



a market theory of alliances 31

States, were long-time foes). Although military dependence caused by hostility is a major driving force of alliances, the explanations it offers are often inadequate. This book is aimed to correct the literature’s bias toward adversary games as opposed to alliance games.5 The independent dynamics of alliance games deserve more attention from international relations research because alliance politics is not exclusively, and perhaps not even primarily, driven by considerations of adversary games. Ironically, the alliance literature has focused more on interactions between adversaries (that is, how alliances deter or fight enemies) than on those between allies (that is, how allies help each other) because of the widespread notion that alliances are formed in response to external threats. For instance, although the term alliance often appears in discussions of balancing, the concept of balancing is predominantly discussed in the logic of adversary games. Granted, states sometimes balance against the power or threat of an adversary, but external (and to a lesser extent, internal) balancing is still a function of relationships between potential and actual allies.6 In a situation where external threat for a state is scarce, the state’s balancing policy should be more inf luenced by alliance games than by adversary games. A state may have an interest in strengthening its security tie with another state because of their political or economic tie, and not because they share a common external threat. In such cases, contrary to Liska (1962, 12), alliances are for, and only derivatively against, someone or something. During the Cold War, Japan’s policy against communist states was a derivative of its relationship with the United States. In the future, even without external threats such as North Korea and China, Japan and South Korea will still benefit from having a formal tie with the United States, given the deep interdependence across the Pacific. Relationships between military allies cannot be reduced to cooperation against external threats. For instance, Miyahara (2014) argues that Japan sustained the Anglo–Japanese alliance (1902–1923) for over two decades because of political and economic benefits from the alliance, even though the threat of Russia and the credibility of British military support both declined over time. The revitalization of the U.S.–Philippines alliance since 1999 also illustrates this point. From a traditional perspective, threats from China appear to have caused this reinvigoration. This assessment is at least partially accurate, and the Chinese threat has since become even more important. We should not, however, discount the other benefits the Philippines gains from its alliance with the United States—benefits that

32

a market theory of alliances

had significantly declined after the Philippines refused to renew the military bases agreement in 1991. From a perspective with more focus on the benefits, the Philippine government seems to have “portrayed China as an immediate threat in order to induce Filipinos to support revitalizing the alliance . . . Congressman Gregorio Andolana even accused the Ramos government of casting the Chinese as a bogeyman to lure the Americans back into a revitalized alliance, and to justify signing a new access agreement with the U.S.” (Park 2011, 275).7 In sum, to understand what alliances are for (and not just what they are against), we need an analytical framework of alliance games independent of adversary games. Reconceptualization Without the premise of external threat motivating every member of alliances, we need an alternative rationale for military alliances, at least for some members. This is exactly the problem many policy makers faced after the Cold War. For Americans, reasons to keep their alliances are not as clear in a unipolar era as in the Cold War era (Menon 2007). Many doubted the continuation of the Cold War alliances (Mearsheimer 1990; Waltz 1993), but the Western alliances have been largely intact. Meanwhile, despite predictions made by some realists (Layne 1993; Waltz 1993), no serious alliance has been formed against the American unipole (Ikenberry 2002). Furthermore, there had been anomalies to the threat-based alliance theory well before the advent of the unipolar system. Asymmetric alliances were common in multipolar and bipolar systems too, and states used to purchase military services from their allies by forming subsidy treaties, in which one state loaned its troops to another. Many states sought something other than military protection in these alliances. Conceptualization of military alliances as a solution only to adversary games cannot explain these new and old anomalies. To understand functions of military alliances other than capability aggregation, we need to focus on what allies exchange with each other. Schroeder (1976) explains how alliances were used for management and control of alliance members in addition to capability aggregation against third parties. In alliances of management or control, some or all of the members were expected to restrain their policy rather than to provide military force. Morrow (1987, 1991, 1993, 2000) points out that smaller partners in asymmetric alliances cannot contribute much to capability aggregation, and thus larger partners must have obtained something other than security benefits from their alliances. Morrow (1991) develops a model of asymmetric alliances, where allies exchange security and autonomy. As examples



a market theory of alliances 33

of increased autonomy due to an alliance formation, Morrow (1991, 921) raises concessions in an ally’s internal or external policies and military bases. This study adopts the exchange framework and pays special attention to the distinction between military commitment and other kinds of goods in the alliance market.8 We can find a rationale for military alliances in a classic explanation of international trade. David Ricardo ([1817] 1821) illustrated the principle of comparative advantage with the wine-for-cloth trade between Portugal and England. Irrespective of the absolute productivity, two countries can benefit from specializing in the production of goods with which they have a comparative advantage and then trading those goods. What is not in economics textbooks is that the series of the Anglo–Portuguese alliances—the world’s oldest alliance still in force—have explicit clauses on commercial arrangements, including those for wine and cloth.9 Scholars have compared states of the international system to states of the market (for example, Boulding 1958, 330; Waltz 1979), but interactions between allies should resemble the market even better because conf licting and violent interactions between adversaries are often not captured by the market analogy. The market is a place for mutual cooperation based on self-interests, and commercial interactions are more similar to interactions between allies than to those between adversaries.10 From the problems already discussed, it naturally follows that we need to take account of exchanges of heterogeneous goods among allies. Heterogeneous exchanges are possible even within military realms (for example, basing rights and protection), but the inclusion of nonmilitary goods vastly expands the scope and benefit of alliance politics. Although most of the literature has neglected nonmilitary aspects of military alliances, there is a notable exception—a group of researchers who studied the effects of alliances on trade (for example, Gowa and Mansfield 1993; Gowa 1994; Mansfield and Bronson 1997; Long 2003; Gowa and Mansfield 2004). In addition to the causal effects these works have shown, which are mostly one sided (from alliances to trade), the effects of international trade on military alliances deserve more attention.11 Moreover, it may be useful to conceptually integrate the analyses of alliances and economic cooperation. As discussed in Chapter One, recent scholarship shows that many alliance agreements include clauses on economic cooperation (Long and Leeds 2006), and many trade agreements include military commitments (Powers 2004). Alliances and monetary cooperation affect each other as well: Gavin (2003, 217), for example, argues

34

a market theory of alliances

that the most important factors that explain U.S. monetary policy during the 1960s were “questions of security and alliance politics in Europe.” The market theory of alliances provides a ground for theoretical integration of the studies on alliances and economic cooperation. Military Protection Is Not the Only Good in the Market We cannot explain the rationale for asymmetric alliances, let alone understand the variety of exchanges that take place between allies, without taking into account nonmilitary goods in the alliance market. A state sometimes needs military assistance from an ally weaker than itself, but when the power disparity between the allies becomes large, the ally’s military force becomes less relevant. Let us draw a line at the capability ratio of 10 because approximately half (44.2 percent) of all dyad-years in the interstate system between 1816 and 2003 have a tenfold capability disparity or greater12 and because the ratio of 10 seems to be sufficiently large to make the weaker ally’s force trivial.13 The capability ratio was about 6 for the U.S.–Japan alliance in 1960, 41 for the U.S.–South Korea alliance in 1953, and 12 in the U.S.–Spain alliance in 1976. Somewhat surprisingly, 38.8 percent of allied dyad-years between 1816 and 2003 have a capability ratio of 10 or more (27.5 percent for 20 times or more, and 17.2 percent for 50 times or more). In fact, the mean capability ratio of allied dyad-years is 278 whereas it is 209 for nonallied dyad-years. That is, on average, one of two allied states is 278 times stronger than the other. When we drop observations after 1945, the mean ratios drastically change (42 for allied dyads and 48.1 for nonallied dyads) because there are far more small states in the post–World War II era. However, 33.7 percent of allied dyads have a capability ratio over 10 even during the multipolar period (1816–1945), when the concern for the balance of power was supposed to be strong and the capability aggregation model should work well.14 This suggests that many states have formed and maintained alliances with partners too weak for the purpose of capability aggregation, and it is not a new phenomenon. Of course, small states sometimes provide valuable military assets such as basing rights, but only about 16 percent of the allied dyad-years with a capability ratio of 10 or more (1816–2003) have such arrangements for troop stationing (“BASE” = 2 or 3 in the ATOP dataset, version 3.0).15 Thus, it is hard to explain away so many asymmetric alliances with the military values of small states’ contributions. Asymmetric alliances are not only numerous but can also be very inf luential. The U.S.–Japan alliance, arguably the most important and clearest



a market theory of alliances 35

asymmetric alliance in our time, shows that an alliance of great importance and stability can have an extreme asymmetry of military responsibilities. The alliance between two of the world’s largest national economies is a linchpin of security orders in East Asia and has endured numerous changes in the international environment since 1951. Yet Japan’s military contribution to the alliance is limited. Japan does not have an obligation to support the United States militarily. In fact, it has been prohibited from supporting the United States because the Japanese constitution has until very recently been interpreted as forbidding the exercise of the right of collective self-defense.16 Because of the pacifist political culture of postwar Japan, Japanese leaders did not publicly use the term alliance until Prime Minister Zenko Suzuki’s 1981 joint communiqué with U.S. President Ronald Reagan; even then, Suzuki later denied military implications of the alliance, and his minister and vice-minister of foreign affairs resigned as a result of the political turmoil (Fujimoto 1999). The alliance has been nonetheless beneficial to both parties. As Kenneth Pyle points out, Japan desired the alliance “to bring about an early end to the occupation while still being able to draw on US aid . . . thereby not only guarantee Japan’s security but provide the country with the economic and technological aid necessary for recovery” (Pyle 2010, 6). By relying on U.S. military protection, Japanese leaders were able to focus on economic development and avoid the political costs of rearmament, which would have alienated neighboring countries and the pacifist population within Japan. In terms of security provision, it is also important that the United States no longer poses a military threat to Japan. This issue is seldom publicly discussed, but Japan does not have to worry about the strongest military power on the planet (and the only successful conqueror of Japan) because of the alliance. Although Japan has benefited from American protection and economic support, the United States has also received various benefits. Politically, the United States established some control over Japan through the alliance, keeping the former enemy closely in the Western camp.17 The Japanese designed their foreign policy toward such important neighbors as China and the Soviet Union in line with U.S. policy. Militarily, bases in Japan have allowed the United States to project its military power in the Western Pacific, and the Japanese industrial capacity also made these bases valuable. As the Japanese economy developed, its economic contribution to the alliance became important as well. The level of contribution eventually became the highest among U.S. allies. For instance, the Japanese

36

a market theory of alliances

g­ overnment covered between 75 and 79 percent of total local stationing costs of U.S. troops in the period between 1995 and 2002 (Calder 2007, 192–193).18 Japan’s foreign aid, although also serving its economic national interest, was directed toward countries strategically important for the United States (Boyer 1993; Yoshitsugu 2011). When the U.S. current account deficit and fiscal deficit grew in the 1980s, it was Japan that financed the American economy. Although economic interdependence between the two countries is strong in itself, Japan’s military dependence on the United States reinforces American inf luence on Japanese economic policy, and the United States is always able to count on Japan’s support for its global economic policy (Dwyer 2000, 117, 120–121). In sum, states often form and maintain an alliance with another state that is unable to provide military force for allied defense. An imbalance of military capabilities does not mean that there cannot be a valuable exchange between allies. Neglecting nonmilitary goods in alliance politics, therefore, is problematic, biasing our view.

Allies Exchange Goods I argue that military alliances are a mechanism in which countries exchange goods, at least one of which is military commitment.19 When the goods exchanged are limited to the provision of military force, the exchange model approximates the capability aggregation model. The demand–­supply curves for military commitment, however, should look different for the capability aggregation model. In the model, the actors with purchasing power are limited to states that can provide meaningful military commitment. A market in which participants exchange only food should work differently from a market where participants can purchase food with wine, cloth, and other goods. Traditional theories of alliances were so focused on the fear of military defeat (starvation) that they assumed states exchanged only military force (food) in the alliance market. Military goods include the provision of troops, military bases, and nuclear deterrence. Nonmilitary goods include economic concessions, public support of the ally’s policy, and coordination of foreign policy (for example, refraining from approaching the allies’ enemies). Aside from the difference in goods and services exchanged, military alliances are quite similar to trade contracts. No wonder there are overlaps between alliance treaties and trade agreements. Kinds of goods exchanged vary: States sometimes exchange similar goods (for example, both allies providing troops),



a market theory of alliances 37

but they also exchange goods from different fields (such as military protection in exchange for economic or political concessions) or different goods even within the same field (such as protection in exchange for the provision of military bases). Subsidy treaties England signed with continental powers are a typical trade (alliance) of military and nonmilitary goods (Schweizer 1989). Although England was powerful itself, it was often cheaper to subsidize continental powers such as Russia and Prussia to fight for English interests. Economic incentives can be used both to attract military allies and to sustain them (Grieco and Ikenberry 2003, 184–185). In his analysis of the relationships between international finance and alliance politics in the period between 1880 and 1914, Jacob Viner (1929, 447) comes to the following conclusion: In almost every important instance, the successful negotiation of an entente or alliance was aided by loan negotiations, and in several cases the financial inducements appear to have been a decisive factor in the success of the negotiations. On the other hand, refusals to permit loans to be negotiated were on several occasions an important factor in destroying the chances of establishing a favorable alliance or of securing the termination of a hostile one.

Poast (2012) shows that offers of trade linkage substantially increased the probability of alliance formation in Europe between 1860 and 1945. Fordham (2008) argues that exports have indirect effects on American intervention in foreign conf licts by shaping alliance commitments in the long run. As noted in Chapter One, Portugal bought its alliance with England in 1661 (Belcher 1975), and French loans to Russia helped the formation of the Franco–Russian alliance of 1892 (Viner 1929, 410–411). During the interwar era, while Nazi Germany cultivated the economic dependence of Bulgaria, Hungary, and Romania on its economy (Hirschman [1945] 1980), France offered commercial concessions to Belgium (Long and Leeds 2006), and Great Britain to the United States (Skalnes 1998). Davis (2009) demonstrated that the formation and management of the Anglo–Japanese alliance (1902–1923) was also, at least partially, based on the exchange of military and economic goods. Economic incentives play important roles even after a war breaks out. During the Thirty Years’ War, French financial support to Sweden, Dutch Protestants, and German princes ensured that they continued to fight against the Habsburg (Adams and Parker 1997, 63; Parker 1997, 111). In the Second World War, the United States provided military and economic assistance to its allies through the Lend-Lease agreement.

38

a market theory of alliances

Exchanges of heterogeneous goods are prevalent even in more recent periods. Since the late 1960s, the United States has sometimes used its alliance with Japan to obtain trade concessions (Stone 1999). The security– economy linkage is not limited to trade issues, and “the U.S.–Japan security relationship will continue to constrain . . . the degree of assertiveness with which Japan can confront the United States over any financial issue” (Dwyer 2000, 117). Even in the Atlantic alliance, at “one juncture the United States hinted that the continuation of the American nuclear guarantee of Europe would depend upon economic concessions from the Europeans” (Mandelbaum 1981, 168). Similarly, to obtain West Germany’s monetary cooperation, the United States repeatedly threatened troop withdrawals from the key ally (Gavin 2003). Alliance members do not publicize exchanges of heterogeneous goods because such exchanges tend to create a sense of inequity and emotional backlash. Measuring the values of different goods is difficult, and exchange of military protection with other goods also creates an impression of exploitation. For instance, Cha and Katz noted that the passage of the U.S.–ROK free trade agreement in Seoul could be “complicated by criticism among opposition parties that South Korea bowed to U.S. pressure in exchange for U.S. security assurances during the December 2010 round of negotiations, which coincided with the aftermath of North Korea’s attack on Yeongpyeong Island at a time when South Korea was feeling particularly vulnerable” (Cha and Katz 2011, 62). The existing literature is mostly concerned with exchanges of military force, but a narrow focus on capability aggregation leads us to ignore a broad range of functions that military alliances play in international politics. Although combining military capabilities for defense and deterrence is an important part of exchanges between allies, alliances also provide their participants with efficiency gains based on comparative advantage, economies of scale, and informational and institutional effects of diverse exchanges.20 Comparative Advantage Irrespective of the absolute productivity, two states can benefit from specializing in the production of goods with which they have a comparative advantage and then trading those goods. Even when one country has an absolute advantage in the production of every good, it still gains from trade because it is more efficient to concentrate on the state’s strongest areas and to exchange the goods with different goods produced by other states.



a market theory of alliances 39

The same principle applies to the alliance market. Although the United States has an absolute advantage in many areas, it still gains from the division of labor with its allies. Because the United States has a comparative as well as an absolute advantage in military capabilities, it supplies military protection to its allies and in return receives various concessions from them. The United States has also provided its allies with economic concessions (such as economic aid or trade concessions), but some of the allies have come to have a comparative—though not absolute—advantage in economic fields as their economies have developed. In the postwar era, states such as Japan and Germany came to possess disproportionately high comparative advantage in the supply of economic goods because of political constraints on their military postures. Most developed countries share the responsibility of managing international trade and monetary regimes with the United States. In fields such as developmental assistance, many have contributed more than the superpower, relative to the size of their economies. Boyer (1993) shows that what has been seen as free riding in Western alliances is actually only one aspect of the division of labor among allies: The United States disproportionately contributes to the alliances militarily, but others have played important roles in foreign aid and monetary cooperation.21 We can find a division of labor within military deployment as well, because the principle of comparative advantage works at various levels.22 In general, the United States contributes more in air and naval power, whereas its allies provide ground troops, which are less capital and technology intensive. When it is more efficient for logistical reasons, the allies contribute in air and naval force as well; for example, Japan Maritime SelfDefense Force developed an extensive cooperation with the U.S. navy in the Western Pacific during the Cold War. The Czech Republic has specialized in chemical warfare since its time in the Warsaw Pact, the Netherlands has focused on sealift and land support by its navy, and Denmark has sacrificed its blue-water navy in favor of brown-water capacities (Mowle and Sacko 2007, 602). The division of labor also exists between different stages of military operations: The United States fights a war, and the allies work in the postwar reconstruction and peacekeeping phase. Although these divisions of labor are efficient in terms of material costs, they can also become a source of acrimony between allies because asymmetric responsibilities create a sense of inequity: “In many circumstances, allies complain that Washington demands an excessive degree of leadership for the planning and conduct of operations while allies are relegated to second-class roles, collectively

40

a market theory of alliances

l­eading to the colloquial expression that ‘the United States does the cooking and we wash the dishes’” (Tertrais 2004, 140). Economies of Scale Economies of scale are obtained when the cost of production per unit decreases as the scale of production increases. When there is a high fixed cost, or when there is learning by doing, cost per unit decreases as output increases. Efficiency gains do not have to be based on comparative advantage because a division of labor and specialization can be beneficial even to actors with identical factor endowments when there are economies of scale. Each state benefits from specializing in an area rather than diversifying its resources across fields and failing to take advantage of economies of scale. Naturally, diseconomies of scale (efficiency loss) are also possible in some alliances or fields of allied cooperation (Holsti, Hopmann, and Sullivan 1973, 22). Perhaps the clearest example of economies of scale is to be found in arms production.23 Despite many countering effects, arms industries tend to concentrate in a limited number of states. Having arms industries in one’s home country is good for obvious military-strategic reasons, and it can also be good for the domestic economy because the industries create jobs and spillover effects from research and development. Yet most countries import arms, especially from their allies. If there were no economies of scale, many would have had more motivation to develop their own arms industry. During the Cold War, the United States and the Soviet Union dominated the arms industry not just because of power politics but also due to the economy of scale. Signaling and Enhancement of Commitment through Exchanges Benefits of exchanges are not limited to those based on comparative advantage and economies of scale. When states exchange goods, they send information to each other and third parties.24 Within alliances, exchanges clarify expectations about cooperation, thereby facilitating integration of allied forces and reducing the fear of opportunistic behavior. The more exchanges allies have with each other, the easier it is to coordinate their actions and to detect deviations from the agreed common policy. Exchanges signal the allies’ close cooperation to those outside the alliance and enhance the deterrence of enemies. The more exchanges allies have, the clearer their common interest is to their enemies. For example, the U.S.–East Asian alliances



a market theory of alliances 41

run smoother and look more credible than the Sino–North Korean alliance, which lacks diverse exchanges of goods, despite North Korea’s economic dependence on China. Exchanges not only signal but also reinforce existing commitments. As a division of labor develops, allies become more vulnerable without the continuation of the alliance. For instance, many American allies depend on the United States for air, naval, and nuclear force, as well as for military intelligence. The allies have developed these capabilities less than they would without the alliances; they would be more vulnerable if American protection was suddenly withdrawn than they would be if they were not allied with the United States in the first place. The United States and its allies also exchanged economic goods with one another for security externalities (Gowa 1994), and the institutionalization of the exchanges also strengthened the alliance commitments. A division of labor creates mutual dependence, and it makes allies more committed to the continuation of the alliance. This commitment mechanism supplements the contractual aspect of alliances, which in an anarchic world is mostly based on reputation costs and, perhaps, moral obligation.25 Thus, the commitment mechanism reinforces the deterrence effect of exchanges. NATO’s successful history provides us with an excellent example of these informational and institutional effects.26 Without active trans-Atlantic interactions in political, economic, social, and military fields, NATO as a military alliance would probably have been less effective and credible.27 Similarly, U.S. alliances in East Asia have established the expectation of cooperation in diverse fields, and strong interdependence among the members has reinforced the military alliances. Exchanges of goods between allies is more than simple giving and taking. Similar to Skalnes’s (2000) discussion of foreign economic policy as strategic instruments, giving goods to one’s ally increases the ally’s loyalty, signals intentions to the third parties, and strengthens the ally. Besides the obvious benefit, receiving goods from an ally also signals intentions to others and strengthens the capability of the alliance. These benefits suggest the importance of exchanges between allies on top of the efficiency gains from comparative advantage and economies of scale. As efficiency gains do not create an extreme division of labor in international trade, nor do they make every state allied with every other state. There are transaction costs in alliance formation, and finding a reliable partner is not easy. Some goods are less tradable than others because they lose value over distance or because dependence on foreign states for the goods makes the state vulnerable. States are particularly sensitive about how they

42

a market theory of alliances

trade military goods. Moreover, alliance commitments are highly rivalrous (subtractable), and forming one alliance sometimes means giving up another alliance option. Because most states have limited capacity of force projection, the number of potential allies who can supply military force to any given state is fairly limited in the alliance market; consequently, the number of great powers has significant effects on the supply of military protection in the market.

Contracts and Risks Alliances as Contracts Military alliances are contracts where countries commit themselves to each other to exchange goods according to their agreements. Unlike alignments, which are based on coincidence of interest, alliances require commitment. Allies exchange various goods, but they seldom give and take these goods at the same time. Consequently, states always face risks of opportunistic behavior or inadvertent failures by their allies. An ally may not deliver what it promised (that is, abandonment), or an ally may cause a disaster by taking advantage of alliance commitments (that is, entrapment/­entanglement). The risk of opportunistic behavior threatens the stable continuation of exchanges and may even prevent the formation of an alliance in the first place. Moreover, even without opportunism, allies may inadvertently fail to deliver promised goods or impose unexpected costs on one another. Only by crafting a good contract can allies safely reap the benefits of exchanges. The nature of the alliance market creates a special need for contracts. Unlike for most of our daily commercial interactions, military allies cannot change partners without serious difficulty. The number of buyers and sellers in the alliance market is limited, and a state’s alliance options are constrained by the existing network of commitments. When an alliance is dissolved, states lose what they have invested in the alliance and efficiency gains obtained through integration. Many investments that states make to support transactions in a military alliance (for example, adjustment of diplomatic policy and military strategy) are specific to that particular relation and cannot be used for other purposes—that is, asset specificity is high in the alliance market.28 Because betrayal by one’s ally is costly, management of opportunistic behavior is a central part of alliance politics. Oliver Williamson, a leading scholar of transaction cost economics, has written on the use of hostages as a mechanism for commitment—which we know was common among mili-



a market theory of alliances 43

tary allies. The following example he refers to also reveals the importance of commitment to military alliances: A still earlier example of an effort to craft a credible commitment was recently unearthed in Mesopotamia. Tablets dated around 1750 B.C. show that curses were used to deter the breach of treaties. One of these reads as follows: When you ask us for troops, we will not withhold our best forces, we will not answer you with evasions, we shall brandish our maces and strike down your enemy . . . As wasted seeds do not sprout, may my seed never rise, may someone else marry my wife under my very eyes, and may someone else rule my country. (Williamson 1996, 91)

Because states are fully aware of fateful consequences of compliance and noncompliance, they carefully design alliance contracts. Previously, the reliability of military alliances had been considered to be as low as 27 percent (Sabrosky 1980) or 23 percent (Siverson and King 1980) because these earlier works observed only whether states joined their allies in time of war. With the ATOP dataset, which codes specific obligations of alliance treaties, Leeds, Long, and Mitchell (2000) find that alliances are indeed reliable 74.5 percent of the time; although the previous datasets coded inaction of allies as a violation of alliance agreements, many cases were outside the scope of alliance obligations.29 Because the contents of alliance agreements have substantial effects on how allies behave, we should expect states to design military alliances with great care. Thus, the analysis of alliances can be linked to the literature on the designs of international institutions (for example, Lake 1999; Koremenos, Lipson, and Snidal 2001). Treating alliances as contracts may seem problematic because of the anarchy of international relations, but my approach is consistent with the development of economics literature on contracting and transaction cost. Alliance treaties are “weak” or “soft” even when compared with other international agreements, and if a state “refuses to come to the aid of another under the terms of an alliance, nothing can force it to” (Baxter 1980, 3). Contracts, however, can be made self-enforcing, and economics literature has demonstrated various ways in which contracts can be designed to mitigate the problems such as asymmetry of information, opportunism, and uncertainty of the future (for example, Williamson 1983, 1985; Hart 1995; Williamson and Masten 1999; Baker, Gibbons and Murphy 2002). Therefore, some international relations scholars have drawn on transaction cost economics (for example, Lake 1999; Weber 2000) and incomplete contracting theory (Cooley and Spruyt 2009) to explain various forms of governance in international relations. Military alliances may provide a fertile ground for theoretical and empirical research for economists because some

44

a market theory of alliances

of the major issues in alliance politics are central to the economic theories cited above (for example, asset specificity and military bases, and incomplete contracting and renegotiation of alliance agreements). There are two major risks associated with alliance contracts, abandonment and entanglement (the current literature mislabels it as entrapment). Whereas abandonment is a result of opportunistic behavior, entanglement is not necessarily so. Abandonment and Intra-Alliance Bargaining Power Abandonment is defection by an ally, and “It may take a variety of specific forms: the ally may realign with the opponent; he may merely de-align, abrogating the alliance contract; he may fail to make good on his explicit commitments; or he may fail to provide support in contingencies where support is expected” (Snyder 1984, 466).30 From the perspective of the state being abandoned, it is a problem on the benefit side of its alliance. In addition to various case studies on how the fear of abandonment affects international relations (for example, Mandelbaum 1981; Snyder 1984, 1997; Cha 1999), recent works have shown what conditions affect the risk of abandonment by explaining when states tend to violate their alliance obligations (Leeds 2003a; Leeds and Savun 2007; Leeds, Mattes, and Vogel 2009; Kim and Sciubba 2015). As in other human activities such as business and social life, the fear of being abandoned ref lects dependence between military allies. Dependence between states is the central variable that explains intra-alliance bargaining as well as formation, duration, and termination of alliances.31 Therefore, in Chapter Three, I analyze how systemic polarity affects the fear of abandonment. Because systemic polarity not only ref lects relative capabilities but also affects the availability of alternative allies, it tells us more about the respective bargaining power of the allies than their capabilities alone can. Entrapment/Entanglement and the Dilemma According to Glenn Snyder, entrapment “means being dragged into a conf lict over an ally’s interests that one does not share, or shares only partially” (Snyder 1984, 467). There is usually a trade-off between measures to reduce the risk of abandonment and entrapment. When a state tries to avoid abandonment by tightening its alliance, it faces an increased risk of entrapment. Conversely, when a state tries to avoid entrapment by distancing itself from its ally, it faces an increased risk of abandonment. This trade-off is called the secondary alliance security dilemma (Snyder 1984).32



a market theory of alliances 45

The dilemma has long been considered to be important to policy makers (Snyder 1984, 1997), but recent research offers evidence that the trade-off affects the public opinion as well. Hata and Iida (2015) conducted a survey experiment in late July 2014, shortly after the cabinet of Prime Minister Shinzo Abe decided to reinterpret Article 9 of the Japanese constitution to exercise the right of collective self-defense. With this reinterpretation, the Japan Self-Defense Force can now aid allies such as the United States even when Japan itself is not attacked. In their experiment involving 1,788 Japanese respondents, Hata and Iida found that those who were given explanations about the risk of abandonment were significantly less likely to disapprove the cabinet decision than those who were not. Given that the cabinet decision is a legal pathway for Japan to be entangled in another state’s military conf lict, the respondents’ attitude was likely to have been affected by the trade-off between the risk of abandonment and the risk of entrapment/ entanglement. Although the term entrapment is popularly used, the label is sometimes misleading and is better replaced by entanglement. Researchers have pointed out that alliances are associated with the spread of war (for example, Siverson and King 1979; Christensen and Snyder 1990), but a state’s participation in its ally’s war does not automatically constitute entrapment. The state may have involved itself in the war regardless of the alliance, and even when the alliance entangled the state in the war, the state may be a willing participant rather than a victim of “entrapment.” As can be seen in the case of Japan in the post–Cold War era, a state could positively desire entanglement for an improved relationship with its ally.33

Systemic and Domestic Aspects of the Alliance Market Systemic Polarity To analyze the dynamics of alliance games at the systemic level, I focus on systemic polarity as Kenneth Waltz did for analyzing adversary games (Waltz 1979).34 Like Waltz, I define polarity by the number of great powers because the number has significant effects on transactions in the alliance market. In international politics, only great powers have sufficient military capabilities to affect the systemwide supply of security and, consequently, the price of military protection—the most important good in the alliance market. My primary focus is on the effects of systemic polarity on the supply side of the alliance market. Although shifts in polarity also affect demand for security because some states face great powers as enemies, the

46

a market theory of alliances

effect is less consistent and less predictable. Although demand for security is, for most states, largely determined by local situations, the availability of an alternative security supplier is more directly affected by systemic changes. Because we need to analyze more idiosyncratic factors on the demand side, we get more leverage by focusing on the supply side for the purpose of systemic analysis.35 Before the end of the Second World War, there were multiple great powers, and the system was multipolar. During the Cold War, the United States and the Soviet Union were the two great powers, and the system was bipolar. Since shortly before the collapse of the Soviet Union, the United States has been the sole great power, and the system has been unipolar (Mastanduno 1997; Wohlforth 1999, 2002; Thalakada 2012; Monteiro 2014). Chapter Three develops a model that explains the effects of systemic polarity on allies’ concerns for abandonment and entrapment/entanglement. My approach, of course, is not the only way to analyze the systemic dynamics of the alliance market, but it is a reasonable first cut. My goal is not to simply translate what most analysts already have in their mind (about demand for security) to the language of economics. Unlike demand for security, supply of security has not received attention in alliance theory. The alliance literature is underdeveloped also in terms of a theory that focuses on why and how systemic polarity affects alliances, even though the effects of polarity on international relations have been studied widely (for example, Kaplan 1957; Deutsch and Singer 1964; Waltz 1979; Wohlforth 1999; Buzan 2004; Ikenberry, Mastanduno, and Wohlforth 2009).36 Domestic Politics and Intra-Alliance Bargaining Under the surface of international transactions in the alliance market, domestic actors pursue their own interests, and their behavior is constrained by political conditions of their country.37 As “the classical Realists were already aware, the effects of domestic factors are not limited to the process of interest formation, but affect strategy and bargaining outcomes as well” (Moravcsik 1993, 14–15). There are many ways to formulate a domestic theory of alliance politics, but I focus on the determinants of intra-­a lliance bargaining power and when allies are more willing to make concessions (that is, supply of various goods). The theory should help us explain who gets what for how much and the conditions that govern the exchanges between allies. Chapter Five theoretically explains and Chapter Six empirically illustrates how domestic politics of allies systematically affect intraalliance bargaining outcomes.



a market theory of alliances 47

I argue that the presence or absence of effective domestic opposition to alliance cooperation, a leader’s vulnerability, and the leader’s attitude toward an alliance are the three particularly important factors. The presence of effective domestic opposition to alliance cooperation increases a state’s intra-alliance bargaining power because domestic opposition increases the cost of concessions to an ally. The effect of domestic opposition has been widely studied in international relations literature (for example, Putnam 1988; Evans, Jacobson, and Putnam 1993; Fearon 1994; Schultz 2001; Howell and Pevehouse 2007), but Chapters Five and Six extend the domestic theory of international relations by introducing the effects of a leader’s attitude and vulnerability. Two simple observations tell us much about the effects of the two variables: (1) A proalliance/procooperation leader is more willing to concede to an ally than an anticooperation leader is; (2) the ally prefers to have a pro­ alliance leader as its partner. A proalliance leader is more willing to concede to the ally, but the ally also has an interest in conceding to a proalliance leader if the concession keeps the more cooperative and less demanding partner in power. An anticooperation leader is less willing to concede to the ally and is a tougher negotiator, but the ally has an interest in undermining the uncooperative leader by withholding concessions. Therefore, a leader’s attitude has two opposite effects, depending on his or her vulnerability. When a leader of a state is vulnerable, the country’s ally has incentives to make (withhold) concessions to a procooperation (anticooperation) leader because the ally benefits from propping up (undermining) a procooperation (anticooperation) leader. When a leader’s hold on power is expected to last, however, the ally does not need to help a procooperation leader and has to accept tougher demands of an anticooperation leader. Table 2.1 summarizes the effects of the three variables on a state’s intraalliance bargaining power. When a leader becomes too strongly opposed to an alliance, or when domestic opposition to an alliance becomes too strong, the alliance is not feasible. My cases, the U.S. alliances with Japan, South Korea, and Spain did not experience such extreme situations, but Chapters Five and Six illustrate the variations in intra-alliance bargaining power as a function of the changes in these domestic explanatory variables. The actual alliance bargaining process, of course, is affected by both international and domestic factors. For instance, the strengthening of Japan’s military commitment to the United States in Shinzo Abe’s second term as prime minister (2012–) is attributable not only to the rising threat to Japan

48

a market theory of alliances table 2.1 Domestic politics and intra-alliance bargaining power.

When a leader is not vulnerable

Procooperation leader

Anticooperation leader

Strong opposition to alliance Weak opposition to alliance

Moderate bargaining power Weak bargaining power

Strong bargaining power Moderate bargaining power



Procooperation leader

Anticooperation leader

Strong opposition to alliance Weak opposition to alliance

Strong bargaining power Moderate bargaining power

Moderate bargaining power Weak bargaining power

When a leader is vulnerable

from China but also to Abe’s domestic power base, which is much stronger than in his first term. With respect to South Korea, many are rightly interested in Sino–South Korean relations and strategic alignments around the Korean Peninsula, partly because the U.S.–South Korea alliance has been stable in the last several years. Domestic politics, however, can haunt the U.S.–South Korea relations in the coming years. Park Geun-hye’s domestic political vulnerability can increase as a result of unforeseen events like the sinking of the ferry Sewol. Furthermore, the U.S. presidential election in 2016 can lead to disagreement between the allies over policy toward North Korea. For analytical purposes, I separate the systemic and domestic levels. In the following chapters, I develop my theory of the alliance market in more detail and test my arguments with the ATOP dataset and case studies. Chapters Three and Four deal with the fact that there are a limited number of important security suppliers in the alliance market, and they explain the effects of systemic polarity on alliance politics. In Chapters Five and Six, I extend the market theory by taking into account the divergent interests of actors within national borders.

3

A Systemic Theory of the Alliance Market

The vice to which great powers easily succumb in a multipolar world is inattention; in a bipolar world, overreaction; in a unipolar world, overextension. Kenneth Waltz (2000, 13)

In the statement above, Waltz succinctly describes the structural incentives for great powers in different international systems. Although his proposition sounds intuitive and somewhat reasonable, it makes sense only in terms of structural incentives in adversary games.1 In the context of alliance politics, I argue, the vice to which great powers easily succumb in a multipolar world is overreaction; in a bipolar world, overextension; in a unipolar world, inattention. Allies of the United States, including Japan and South Korea, have, for example, been working hard to deal with the inattention of the United States after the end of the Cold War. In 1999, then British Prime Minister Tony Blair pointed out this problem of unipolarity for the allies of the United States: At the end of this [20th] century the US has emerged as by far the strongest state . . . If anything Americans are too ready to see no need to get involved in affairs of the rest of the world. America’s allies are always both relieved and gratified by its continuing readiness to shoulder burdens and responsibilities that come with its sole superpower status. We understand that this is something that we have no right to take for granted, and must match with our own efforts.2

Why is a unipolar power prone to inattention in the alliance market? How is it different from multipolarity and bipolarity? When we think about the past and future of the international system, it is useful to articulate the characteristics of each system, especially in terms of relationships between allies. Analyzing the distribution of capabilities at the regional level, Aaron Friedberg (1993) argued that a multipolar Asia was “ripe for rivalry,” and Robert Ross (1999) predicted a stable regional bipolarity between the United States and China. It is important that Waltz, Friedberg, and Ross all share a focus on relationships between adversaries rather than allies. By developing a model that explains alliance dynamics of different systems, I aim to shed new light on international relations.

50

a systemic theory of the alliance market

In this chapter, I develop a systemic theory of military alliances that explains both the persistence of numerous Cold War alliances and the high rate of alliance formation in the post–Cold War era that has taken place without the involvement of or balancing against the United States.3 This chapter also explains why China’s alliance policy has so far been dormant and why it would be more active under Sino–U.S. bipolarity. These global trends and post–Cold War developments in East Asian international relations share the same causes—the military primacy of the United States and the increased bargaining power of the United States under unipolarity. It may seem almost tautological to argue that the United States enjoys advantages under the current American unipolarity, but this contradicts both the balance of power theory and the public goods model of alliances. Despite the predominance of the U.S. military capability, other states have not balanced against the sole superpower, and they have been paying more to the United States for American military protection. The main topics of this chapter are as follows. Systemic polarity (that is, the number of great powers; see Waltz 1979; Snyder 1997) ref lects the number of significant security suppliers in a system. Therefore, systemic polarity, in combination with demand for security, affects the bargaining positions of importers and exporters of security. This chapter also brief ly explores the role of nonmilitary goods exchanged among allies, as the supply–­demand analysis of security is incomplete without them. Because the concept of “poles” is central to the supply analysis of security, I explain why China is not (yet) a pole. China’s lack of power projection capacity and the current distribution of military capabilities have kept the middle kingdom’s alliance policy rather inactive. My analysis of the supply side of the alliance market complements the vast literature on the rise of China and the threats it creates (for example, Fravel 2010; Mearsheimer 2010; Friedberg 2011; Johnston 2013; Ross 2013; Scobell and Harold 2013; Acharya 2014) that helps us understand the demand side of the alliance market in Asia. The chapter further explains how polarity affects the fear of abandonment and entrapment/entanglement and, by extension, intra-alliance bargaining. I conclude this chapter with hypotheses derived from the systemic model and test the hypotheses in Chapter Four.

Alliance Market at the Systemic Level In my theory, a military commitment is one common good across all exchanges between allies, but it is not the only good in the alliance market.4



a systemic theory of the alliance market 51

Changes in systemic polarity affect the price and quantity of security produced relative to nonsecurity goods, and my theory is based on the assumption that allies exchange heterogeneous as well as homogeneous goods with one another. Like international trade, alliances help states obtain efficiency gains from exchanges of goods based on comparative advantage and economies of scale (Boyer 1993; Hartley 2006). When demand for security decreases, all other things being equal, the relative price of security decreases (that is, less bargaining power for a security exporter and more bargaining power for a security importer). Similarly, when supply of security decreases, all other things being equal, the price of security increases. In other words, decreased demand for military protection improves the terms of an alliance for an ally that imports military protection, but decreased supply of military protection from other states increases a security-exporting state’s bargaining power. The existing theories of alliances reviewed in the previous chapters, especially those based on the capability aggregation model, focus exclusively on the demand curve because their central explanatory variable is military threats. According to these theories, the price of security and the bargaining power of the United States should have decreased after the end of the Cold War because there was a leftward shift of the demand curve (see Figure 3.1). The supply curve, however, also shifted toward the left, and I argue that its shift was more

Relative price of security

Higher price (my theory)

S’

S

D Lower price, lower quantity (old theories)

D’

Quantity of security produced relative to nonsecurity goods

Figure 3.1. The shifts of demand and supply curves.

52

a systemic theory of the alliance market

s­ignificant than that of the demand curve. The price for security provided by the United States, therefore, is higher after the end of the Cold War. In addition to the price of security, we should also pay attention to the quantity of security produced relative to nonsecurity goods. When demand or supply of security decreases, the quantity of security produced declines relative to non-security goods. As will be shown in Chapter Four, the quantity of security offered in the alliance market has declined in the post– Cold War alliance market even though numerous new alliances have been formed in the period. A higher price for security tempts its importers to produce more security on their own, but it is difficult for nongreat powers to be self-sufficient, let alone to become security exporters. Moreover, security exporters can adjust their policy to prevent the emergence of competitors. For instance, as prescribed in the 1992 Defense Planning Guidance, the United States maintained its alliances in Western Europe and East Asia after the Cold War to discourage the emergence of a rival.5 For an American-led unipolar structure of the international system, the maintenance of the alliances in the two regions was particularly important, as they included countries such as Germany and Japan, which could potentially transform into self-­sufficient security consumers or security exporters. Some predicted a multipolar Europe (Mearsheimer 1990) and a multipolar Asia (Friedberg 1993) after the Cold War, but the U.S. alliance policy discouraged such shifts. This chapter explains how the structure of the alliance market inf luences the transactions between its participants, thereby complementing the intuitive demand and supply analysis of security. Systemic polarity inf luences how goods are exchanged among states, just as the number of sellers affects commercial transactions.6 I focus on military polarity because military commitment is the most prominent good in the alliance market and also because international relations research has paid special attention to the distribution of military capabilities. Let me, however, look at goods exchanged with military commitment in the alliance market before I start the systemic analysis of alliance politics. Nonmilitary Goods in the Alliance Market The importance of nonmilitary goods in alliance politics motivates and justifies the market approach. We cannot explain the rationale for asymmetric alliances or understand the variety of exchanges that take place between allies without taking into account nonmilitary goods in the alliance mar-



a systemic theory of the alliance market 53

ket. As noted in Chapter Two, in 38.8 percent of allied dyad-years between 1816 and 2003, one of the alliance partners is 10 times or more stronger than the other (27.5 percent over 20, and 17.2 percent over 50)—a disparity that makes the capability aggregation model suspect. Assuming that militarily weak states have less security surplus to export, weaker allies do not have much to contribute in terms of military force and, rather, drain force from stronger allies. Because the cost of military commitment to a weaker ally can be substantial, stronger states should be reluctant to form an alliance with weaker states for military contributions. One could broadly interpret the capability aggregation model and argue that stronger states form asymmetric alliances for military bases, but, as explained in Chapter Two, most asymmetric alliances do not have basing arrangements. Moreover, if a stronger state is willing to accept a heterogeneous good (that is, military bases) as payment for military protection, there is no apparent reason why they would refuse to accept nonmilitary goods. Nonmilitary goods in alliance politics include economic and political concessions, the latter covering both foreign and domestic policy. Ignoring nonmilitary goods is problematic because, in many cases, bargaining for or within an alliance includes more than exchanges of military goods alone. Because wealth and the military capability of a state are significantly correlated, a militarily stronger state tends to have comparative advantage in economic fields as well. This leaves political and asymmetric military concessions (for example, provision of military bases) as more likely options for weaker states.7 During the Cold War, most U.S. allies received military protection and economic support from the United States while they supplied military bases and political concessions to the superpower. Sometimes, it was the smaller ally that provided military goods (albeit in the forms of military bases) in pursuit of economic and political goals. Franco’s Spain after the Second World War, for instance, pursued an alliance with the United States to break out of political isolation and to obtain economic assistance (Whitaker 1961; Viñas 2003; Chislett 2005b; Calvo-Gonzalez 2006). Similarly, Japan’s decision to ally with the United States was significantly affected by its desire for economic aid and access to the international market, and the negotiation for the alliance was inextricably linked with one for the end of U.S. occupation (Sakamoto 2000). When the provision of military force has political significance, as in the cases of the American-led coalitions in Vietnam (Blackburn 1994; Colman and Widen 2009) and Iraq (Newnham 2008), the militarily

54

a systemic theory of the alliance market

stronger state sometimes exchanges economic goods for the military force of weaker allies.8 Typically, a military threat causes alliance formation by creating demand for military protection, but not all alliance partners need to share the military threat.9 A state in need of military protection can buy an alliance by providing military, economic, or political concessions to another state. To return to the case of the Anglo–Portuguese alliance, when Charles II returned to England at the time of the Restoration in 1660, the king was expected to ally with Spain but instead entered into an alliance with the rebellious, then-Spanish province, Portugal. Charles II’s move was interpreted, by both his contemporaries and many historians, as joining Louis XIV of France, who was supporting Portugal against Spain. A close look at the formation of the alliance, however, reveals that Charles’s decision was based on the financial and commercial benefits offered by the Portuguese rather than on considerations of England’s relationship with Spain and France (Belcher 1975). The Portuguese literally bought their alliance with England, and Charles II’s behavior shows the importance of nonmilitary (and a leader’s personal) motivations for alliance formation. Poast’s (2012) quantitative analysis of trade linkage and alliance formation in Europe between 1860 and 1945 also suggests that nonmilitary goods play important roles in alliance negotiations. In a similar vein, the continuation of an alliance does not necessarily require the continuation of the threat that originally motivated the formation of the alliance, because the allies can adjust the terms of the alliance by altering what they exchange. Leeds and Savun’s (2007) findings support this view: namely, that linkages to other issues reduce the risk of alliance termination by violation. In their estimate, alliance agreements that include provisions for nonmilitary cooperation are 61 percent less likely to be abrogated opportunistically. America’s East Asian alliances were originally motivated by the threat of communism, but none of the alliances faced a serious risk of dissolution after the Cold War. The stability of the alliances is partially attributable to the nonmilitary effects of the alliances. As ref lected in Article 2 of the 1960 U.S.–Japan security treaty, America’s alliances helped these states strengthen “their free institutions . . . and eliminate conf lict in their international economic policies and . . . encourage economic collaboration between them.”10 The South Korean government’s enthusiastic pursuit of a Free Trade Agreement (FTA) with the United States is a good recent example of the linkage between exchanges of military and nonmilitary goods.11 Despite his



a systemic theory of the alliance market 55

leftist support base, former South Korean President Roh Moo-hyun began the FTA negotiation with the United States in 2006, partly because of the security implications of the agreement.12 A former official of the U.S. government notes that South Korean leaders were willing to make concessions because “failure would look like a setback to the political and security relationship. Once we got into negotiations with the ROK, look at how many times we reneged even after we signed a deal. . . . We asked for changes in labor and environment clauses, in auto clauses and the Koreans took it all” (Brooks, Ikenberry, and Wohlforth 2012, 44). According to a Korean foreign policy expert I interviewed in Seoul in 2008, the U.S. side also had noneconomic motivations: The American National Security Council (NSC) was involved in the negotiations between the U.S. Trade Representative (USTR) and South Korea’s Ministry of Foreign Affairs and Trade. The U.S. NSC’s involvement itself is not surprising, but the NSC at the time perceived South Korea to be too close to China, and it urged the South Korean presidential office to intervene as well.13 Poles of the International System Now, let us turn to the providers of military goods in the alliance market. In my alliance theory, poles refer to military great powers. During the Cold War, Japan became economically stronger than the Soviet Union but never constituted a pole. Saudi Arabia is a pole in the oil market but is not one in the international system or in the alliance market.14 There are many ways to periodize polarity (for example, Modelski 1974; Rapkin, Thompson, and Christopherson 1979), but I follow Snyder (1997, 18): The international system was multipolar until the end of the Second World War, a bipolar era follows through 1989, and a unipolar era thereafter. This focus on polarity based on military capabilities might seem contradictory to my emphasis on the exchanges between allies. My approach, however, is consistent because the provision of a military commitment by at least one party is required in my conceptualization of military alliances. Polarity affects the supply patterns of military commitments, the most important good in the alliance market. Among the variety of goods allies exchange, military commitment is the most appropriate good on which to build my theory, and the simplification does not negate the utility of the exchange model. The next section discusses what is required of a pole in the alliance market with the example of China.

56

a systemic theory of the alliance market

China as a Nonpolar Power and Its Alliance Policy China’s national capabilities have grown remarkably in the last few decades (Friedberg 2011; White 2012), but it is not a pole in alliance politics (yet). Unlike the Soviet Union, which was perceived to have conventional military capabilities in Europe superior to those of the United States (for example, see Brown 2012, 135), China is not able to match the United States militarily even in East Asia. As Michael Beckley (2011, 58) points out, economic size “does not necessarily make China a contender for superpower status. After all, China was the largest economy in the world throughout most of its ‘century of humiliation,’ when it was ripped apart by Western powers and Japan.” Because of its relatively lower per-capita income, China is short in “disposable surplus” wealth, which is important for military capabilities (Knorr 1956, 231). China also lags far behind the United States in nuclear weapons, despite the modernization of the Chinese nuclear arsenal (Christensen 2012). In his recent book, China Goes Global: The Partial Power (2013), David Shambaugh argues that China is not a true superpower yet, despite its growing power and global ambition. Militarily, “China is not able to pro­ ject power outside of its Asian neighborhood (other than ballistic missiles, space program, and cyber warfare capacities), and even within Asia its military power projection capacities remain limited” (Shambaugh 2013, 9). In its Annual Report to Congress, the U.S. government recently stated that the People’s Liberation Army Navy’s “goal over the coming decades” was to “become a stronger regional force that is able to project power across the globe for high-intensity operations over a period of several months, similar to the United Kingdom’s deployment to the South Atlantic to retake the Falkland Islands in the early 1980s.”15 Similarly, the People’s Liberation Army Air Force does not have global reach (Shambaugh 2013, 288). It is not clear when, if ever at all, China will be able to project military force over long distance in a way that matches the U.S. military. China’s relative lack of power projection capabilities (Blair 2008) presents a serious problem for it to become a security exporter. China is a formidable military power in its own territory, but what matters from the perspective of potential security importers is the military values the exporter can deliver in a distant location. One major reason for China’s inactivity in alliance politics is the simple lack of capabilities to provide military protection, especially over long distances. Whereas the United States has fought wars



a systemic theory of the alliance market 57

across the globe, China has fought only in neighboring areas such as Korea, Vietnam, and the borders with India and the Soviet Union. To date, alliances with China have not been attractive to many states.16 China lacks power projection capabilities and possesses “negative soft power” (Van Ness 2002, 143). Moreover, an alliance with China is likely to antagonize the United States, which is a far more attractive alliance partner. The Soviet Union was not as attractive an ally as the United States, but the choice in the postwar alliance market was limited for many Eastern European states because of the Soviet military occupation and the geography (compare members of the Warsaw Pact and Yugoslavia). Because China has not militarily surpassed the United States, countries that are threatened by China are likely to prefer buying security from the United States rather than from China itself. China’s existing alliance arrangements seem to ref lect the limits of its power. The Sino–North Korean alliance, signed in 1961, is the only alliance agreement in which China is obligated to provide active military assistance. To North Korea, China had previously proven its military value by playing a crucial role in the Korean War (1950–1953), albeit with great costs to itself. China also has a bilateral consultation pact with Russia and a multilateral consultation pact (Shanghai Cooperation Organization) with Russia, Kazakhstan, Kyrgyzstan, Tajikistan, and Uzbekistan. Among the members of the Shanghai Cooperation Organization, only Uzbekistan does not share borders with China. Thus, ref lecting its lack of power projection capabilities, China’s alliance policy has been geographically limited and underdeveloped. In its official discourse, in fact, the Chinese government has refrained from the language of military alliances altogether, advocating “independent and self-reliant foreign policy of peace” since 1982 (Zhang 2012, 129). Nevertheless, there are some signs of change in China’s alliance policy. According to Feng Zhang, “Alliance thinking is making a comeback in China’s intellectual and policy communities” (Zhang 2012, 130). According to this proalliance thinking, Alliances with neighbours would contribute to building strategic trust with them, thus dampening their suspicions and consolidating strategic cooperation, while at the same time reducing the attraction of the United States as a security provider. At a minimum, building alliances with neighbours might save China from losing any more of an already very small number of friends in its periphery, a worry recently deepened by changes in Myanmar. (Zhang 2012, 135)

58

a systemic theory of the alliance market

Zhang further notes that the new thinking on alliances is attributable to the growth of Chinese power (Zhang 2012, 138). If China’s military capabilities continue to grow and reach the level of a polar power, the alliance market will be transformed to a bipolar (or multipolar with other great powers) structure. Incentives in a bipolar alliance market will be quite different from those in the current unipolar market.17 Sino–American Competition in the Alliance Market Although the alliance market is still unipolar, American policy seems to be increasingly geared toward the prevention of (and preparation against) China’s emergence as the second pole. Aaron Friedberg, in his book A Contest for Supremacy: China, America, and the Struggle for Mastery in Asia, explains the rationale for increasing US military commitment in Asia: Without active cooperation from its regional partners, Washington cannot hope in the long run to balance against a rising China. On the other hand, without strong tokens of its continuing commitment and resolve, America’s friends may grow fearful of abandonment, perhaps eventually losing heart and succumbing to the temptations of appeasement. (Friedberg 2011, 275)

In the context of rising Chinese power, “the so-called hub-and-spokes arrangement that took shape during the Cold War remains indispensable,” and Asian democracies (for example, Japan, South Korea, India, Indonesia) “ultimately form the hard core, or rather the sturdy outer rim, of the American position in Asia” (Friedberg 2011, 281). Open conf licts between the United States and China will be detrimental to the interests of most states in the world, but if there emerges a Sino–American bipolarity, competition between the two superpowers may give certain advantages to their potential clients. In fact, the Sino– American competition has already begun yielding benefits to other states. For instance, the U.S. accession in 2009 to ASEAN’s Treaty of Amity and Cooperation is considered to be an effort to catch up with China’s accession in 2003 (Womack 2014, 279). Sheldon Simon (2010, 55) points out that the United States restored its relationship with Kopassus, the Indonesian special forces unit accused of human rights violations, after the unit’s commander suggested the possibility of developing ties with China.18 The competition, though most intense in Asia, is not limited to that region. In February 2014, the Chinese Minister of Defense Chang Wanquan visited Djibouti, where the only U.S. military base in Africa is located, thereby causing speculation that China might establish a base there in the near future. The United States responded by sending National Security



a systemic theory of the alliance market 59

Advisor Susan Rice to Djibouti in the following month. The Obama administration was initially critical of the Djiboutian President Ismail Omar Guelleh’s undemocratic policy, but Obama welcomed Guelleh in the White House in May 2014.19 At the meeting, the base agreement was extended, with the leasing cost jumping from US$38 million per year to US$63 million per year. The episode is reminiscent of the U.S.–Soviet bipolar competition for clients and military bases. Within the existing network of U.S. alliances, American policy makers consider South Korea to be increasingly susceptible to China’s inf luence (McDevitt 2004). When Chung (2005) interviewed 56 American policy elite in 2002 and 2003, 86 percent of the interviewees expected South Koreans to choose China, rather than the United States, as the most inf luential country in ten years. Yan Xuetong, an inf luential Chinese scholar, has been arguing for a China–South Korea alliance since 2013, suggesting that South Korea can keep its alliance with the United States and stay neutral in the Sino–American competition (Yan 2013; Chen 2014; Han 2014).20 In this context, for example, the South Korean government’s choice of the Korea Air Missile Defense System rather than the U.S. missile defense system can be considered to be in deference to China.21 American policy makers were also irritated by the stance taken by the South Korean government on the Japanese Prime Minister Shinzo Abe’s new policy regarding collective selfdefense: Seoul seemed to be aligning with Beijing rather than with Tokyo, another ally of the United States. Not surprisingly, the U.S. government and foreign policy experts have begun expressing concerns about the increasing inf luence of China over South Korea. Reportedly, for example, the United States (unsuccessfully) pressured the South Korean government not to participate in the Asia Infrastructure Investment Bank (AIIB), which China is preparing to set up as a counterbalance against the (U.S.- and Japan-led) Asian Development Bank.22 Although “U.S. government officials have been adamant that they have no concerns about ‘losing’ Korea to China,” Americans are also uneasy about the growing security cooperation between China and South Korea.23 Before the Conference on Interaction and Confidence-Building Measures in Asia (CICA) in Shanghai in May 2014, China, albeit unsuccessfully, tried to make South Korea sign a joint statement that calls for the end of alliances in Asia. Michael Green implicitly criticizes Seoul’s approach toward Beijing when he writes that it “is not a good thing that Beijing thought Korea could be convinced to sign on to a statement in Shanghai that challenged the legitimacy of U.S. alliances in the region.”24 Pointing out that South Korea

60

a systemic theory of the alliance market

is “infatuated” with China, Victor Cha warns that South Korea “without the U.S. alliance” will be “treated by China like a small province.”25 These warnings from the United States should not be lightly dismissed, as South Korea faces a real risk of alienating the United States, but such warnings also suggest that the United States cannot take South Korea for granted in its competition with China. Increased threat from China will lead to a higher demand for U.S. military protection, but the competition in the supply side might offset the effect and even increase the bargaining power of security-importers. Let us now turn to a more general discussion of the effects of polarity. Basic Intuition on the Effects of Polarity As brief ly explained in Chapters 1 and 2, systemic polarity inf luences how goods are exchanged among states. In a multipolar system, a state can satisfy its demand for military protection through many alliance options. Great powers under multipolarity also have many potential buyers of their military commitments, including their fellow great powers. In a bipolar system, geography permitting, a nongreat power facing a threat could purchase protection from either or both of the two great powers. Two superpowers under bipolarity compete for client states, and the competition increases the bargaining power of smaller states. During the Cold War, as historian John Lewis Gaddis points out, There were limits to how much either Moscow or Washington could order smaller powers around, because they could always defect to the other side, or at least threaten to do so. The very compulsiveness with which the Soviet Union and the United States sought to bring such states within their orbits wound up giving those states the means of escape. (Gaddis 2006, 128)

Moreover, even leaders who had clear ideological positions could still threaten their bloc leaders by encouraging “fears that their regimes might fall if their respective superpower sponsors did not support them” (Gaddis 2006, 129). Although few U.S. allies would have considered switching camps during the Cold War, the United States had to keep the price of its commitment low when the Soviet Union was around. Now that there exists only one great power, a nongreat power has less bargaining power vis-à-vis the most likely supplier of military force, the single superpower. One of the most contentious issues in recent international relations literature, the lack of balancing against the United States in the current unipolar system, is not particularly puzzling or surprising from the perspective of this book (cf. Ikenberry 2002). The shift from bipolarity to unipolarity gave



a systemic theory of the alliance market 61

the United States advantages in the alliance market by increasing its relative capability and reducing the attractiveness of alternative allies; the lack of balancing against the United States ref lects the sole superpower’s position in the new structure of this market. As noted earlier, these advantages also explain the increasing use of coalition building in U.S. foreign policy. Because the United States, the employer, has the upper hand in the alliance market, it no longer needs to recruit formal allies and can now hire parttime partners. As the United States came to almost monopolize the security market after the end of the Cold War, it raised the price and reduced the supply of its formal alliance commitment. Thus we see neither counter­ alliances against the United States nor new alliances formed by the United States in the unipolar system. To understand the supply-side logic of the alliance market better, let us now turn to a more detailed analysis of systemic polarity. Polarity affects alliance politics because it is related to the number of potential security suppliers and to the relative capabilities of actual and potential allies. Number of Security-Suppliers In the alliance market, we can effectively consider the number of poles to be the number of security suppliers. The number of states who can supply security to any given state is fairly small because most states have limited capacity of force projection and most states do not have sufficient military force to make a difference. When a state is facing only a weak enemy, its neighboring states may be able to supply sufficient security, but military protection by a great power is more effective and is often the only effective protection if the enemy is allied with a great power. In any case, the larger the number of poles, the larger the number of potential security suppliers for any given state. As the number of poles increases, importers of security have more options to choose from, and the bargaining power of great powers vis-àvis nongreat power allies decreases. In unipolarity, nongreat powers have nowhere else to turn but to the unipole for effective military protection, and the bargaining power of the unipole is very strong. The bargaining power of two superpowers in a bipolar system, however, is not stronger than that of great powers in a multipolar system. This is because great powers in a multipolar system export security to fellow great powers whereas two superpowers in a bipolar system are not likely to export security to each other (see Figure 3.2).

62

a systemic theory of the alliance market Unipolarity

Bipolarity

Multipolarity

Figure 3.2.  Polarity and exports of military protection.

In a bipolar system, the most significant external threat to a superpower is likely to come from the other superpower. Although a weaker state might form a bandwagoning alliance with such a threat (Weitsman 2003, 30), importing military protection from the source of threat does not make sense for a superpower, as it does not need to capitulate to the other superpower. Collusion between bipolar powers is theoretically possible, but they still do not need to import active military protection because they need only a passive form of security from each other, that is, a promise of nonaggression. In contrast, great powers in a multipolar system actively export and import security among them. Consequently, nongreat powers as importers of security are less important to great powers in a multipolar system than in a bipolar system. Therefore, I argue that the bargaining power of bipolar powers vis-à-vis nongreat power allies is weaker than that of multipolar powers. This argument becomes even stronger when we assume that two superpowers in a bipolar system cannot collude. In the next section, I embed the analysis of polarity in a broader discussion of the effects of capabilities on alliance politics. Albeit loosely, systemic polarity ref lects the relative capabilities of actual and potential allies, especially those of great powers (for example, a unipolar power is stronger than either of the bipolar powers). Systemic polarity concisely describes systemic distribution of power, which is hard to express in simple ways, unlike dyadic relative capabilities.26 Relative Capabilities and Fear of Abandonment and Entrapment (Entanglement) The following analysis is based on two assumptions about bargaining power of alliance partners and explains the effects of systemic polarity on the fear of abandonment and entrapment.27 First, other things being equal, the more



a systemic theory of the alliance market 63

powerful a state is, the more attractive it is as an alliance partner because powerful states have more capacity to supply goods. Second, other things being equal, a state has more bargaining power vis-à-vis its ally when it has more attractive—not just more numerous but also larger—alternative partners outside the alliance. The previous section focused only on the number of potential security suppliers and did not explain how the relative capabilities of allies affect their relationships with each other and other states. Systemic polarity summarizes not only the number of security suppliers but also the distribution of capabilities. In addition to the reasoning given in the previous section, polarity affects the bargaining power of a state because relative capabilities matter, both within alliances and with respect to the attractiveness of alternative allies. States obtain efficiency gains from participating in the alliance market, but they face the risk of abandonment and entrapment in doing so. Because alliance politics are largely made of states’ reactions to these risks, I analyze how the number of great powers affects the fear of abandonment and entrapment. Abandonment is defection by an ally, and “the ally may realign with the opponent; he may merely de-align, abrogating the alliance contract; he may fail to make good on his explicit commitments; or he may fail to provide support in contingencies where support is expected” (Snyder 1984, 466). On the other hand, entrapment “means being dragged into a conf lict over an ally’s interests that one does not share, or shares only partially” (ibid., 467). Thus, from the perspective of the state being abandoned or entrapped, these problems represent malfunctioning of the alliance in the benefit and cost sides, respectively. As in other human activities, such as business and social life, the fear of being abandoned is a function of mutual dependence between military allies. Therefore, the fear of abandonment is directly related to intra-­ alliance bargaining power; namely, an ally with less fear of abandonment has more bargaining power. I also analyze the fear of entrapment (entanglement) because states balance the fear of abandonment and entrapment, and there is usually a trade-off between measures to address the two problems (Snyder 1984, 1997). States adjust the risks of abandonment and entrapment through the designs of alliances (for example, commitment mechanisms to reduce the risk of abandonment, conditional activation of alliance obligations to reduce the risk of entrapment), but my analysis here is still relevant in determining the original risk levels. When we think of what determines the attractiveness of a state as an ally and the purchasing power of a state in the alliance market, its capability and

64

a systemic theory of the alliance market

interest are the primary factors, but the latter is very difficult to generalize and depends more on what happens inside states. Therefore, I use the distribution of capabilities to explain the properties of the alliance market.28 To be sure, this model does not address all variables relevant to the fear of abandonment and entrapment, but it provides a workable framework on which we can base our conjecture about the differences of the fear across states and across systems.29 The systemic model provides only a partial answer to how alliances work, but it is nevertheless useful. As in commercial markets, who buys what from whom for how much is determined by many factors, such as the nature of goods and locations of stores, but the number and sizes of buyers and sellers are important. The following analysis assumes all other things being equal. Crudely speaking, fear of abandonment can be defined as follows: fear of abandonment = (probability of abandonment) × (cost of abandonment) – (probability of getting an alternative ally) × (likely benefit of an alternative alliance) – (self-reliance) The following tables show positive or negative relationships between the key factors. For the analysis, the relevant factors are the shares of the total system power that is occupied by the state and its bargaining partner, the ally (or allies in a multilateral alliance). The denominator of power shares is the combined total power of all states in the system rather than the combined power of states in a particular alliance because alliance bargaining is conducted with attention toward states outside the alliance as well. I suspect that the most important intra-alliance bargaining is likely to be held bilaterally even within multilateral alliances, but in some instances, the relevant bargaining partner may be two or more other members of the alliance.30 The first component consists of the probability of being abandoned and the cost of abandonment from the perspective of the state A (see Table 3.1). The larger A’s power share is, the less likely it is to be abandoned: (a) because it has more resources for satisfying its ally’s (or allies’) changing needs and (b) because A plays more important roles in the ally’s foreign policy. I argue that this relationship should hold despite a selection bias in alliance formation. The effect of power share should be somewhat attenuated in intra-alliance bargaining because, in observed alliances, members have agreed on the terms of the alliances after considering differences of their capabilities. In other words, if a state is weak and still has persuaded another



a systemic theory of the alliance market 65 table 3.1 Probability and cost of abandonment.

The larger the power share of the state A The larger the power share of the current ally (allies)

Probability of A being abandoned

Cost for A of being abandoned



(–)

+

+

state to form an alliance, the weak state is likely to be contributing something that compensates for its lack of capability. Nevertheless, I expect a stronger state to be less likely to be abandoned. A stronger state is better able to adapt to new situations than a weaker state. When the costs and benefits of an alliance change, a more resourceful state has room to adjust its contributions, whereas a less resourceful state may simply be unable to make new concessions that satisfy its ally.31 Moreover, although the bargaining process for alliance formation may create a rough symmetry of benefits to member states (sometimes through asymmetric types of contributions), such bargaining does not guarantee similar degrees of dependence among allies. A strong state tends to play larger roles in its ally’s foreign policy portfolio and thus is more difficult to abandon. For similar reasons, the stronger A’s current ally is, the more likely A is to be abandoned by the ally: (a) A strong ally’s changing needs require more resources to satisfy, and (b) A’s contribution is relatively less important to the foreign policy of a strong state. Stronger states are more likely to have larger strategic interests with larger variance in strategic needs, and they are also more difficult to satisfy with side payments. For instance, when the United States decided to change its policy toward the People’s Republic of China, there was not much Taiwan could do to prevent its abandonment by the United States. Neither was South Africa able to stop the United Kingdom from terminating their defense arrangement, the Simonstown Agreement, in 1975. Meanwhile, I expect the cost of being abandoned to be smaller when A is stronger, although I see this issue as a problem of self-reliance as well. State A suffers less from abandonment when it is strong, because the ally plays a smaller role in A’s foreign policy portfolio. U.S. abandonment of South Korea (for example, withdrawal of troops from South Korea), for instance, would have been much more damaging to South Korea’s national interests before than it currently would because South Korea’s military dependence on the superpower has been decreasing. The cost for A of being

66

a systemic theory of the alliance market table 3.2 Probability of getting an alternative ally and its likely benefit.



A’s probability of getting an alternative ally

The larger the power share of the state A + The larger the power share of the current ally (allies) N/A

Likely benefit of an alternative alliance

Polarity matters for great powers’ alternatives

a­ bandoned and the ally’s power share should be positively related, because A is more dependent on the ally when the ally is stronger. Being abandoned by the United States is more consequential than being abandoned by a small state such as Malta or even by a major power such as France or the United Kingdom. The second component consists of the probability of getting an alternative ally and the likely benefit of the alternative alliance (see Table 3.2). The probability of getting an alternative ally is higher for stronger states because their larger resources help attract a potential ally. Historically, stronger states have formed larger number of alliances than weaker states have: Approximately 29 percent of alliance memberships in the ATOP dataset (version 3.0, 1815–2003) are those of only nine major powers as defined by the Correlates of War project (version 2011). The effect of the ally’s power share on A’s probability of getting an alternative ally is indeterminate. It is also difficult to say what effects the power shares of A and its ally have on the likely benefit of an alternative alliance when they are both nongreat powers. When A is a great power, however, polarity affects how much of the power shares in the system is potentially available as alternative allies. As a unipolar power or one of multipolar powers, A has all the other states minus the current ally as potential alternative allies. As a bipolar power, in contrast, a large portion of the system’s power shares (that is, that of the other bipolar power) is not available to A, assuming that A cannot collude with the other bipolar power. The third component consists of self-reliance, that is, how much a state satisfies its needs with its own supply (see Table 3.3). In general, the larger a state’s power share is, the more the state can rely on itself for whatever goods it needs (such as military force, natural resources, and markets for its products). For an analytical purpose, it is useful to distinguish the vulnerability caused by the alliance, which is related to the cost of abandonment, from



a systemic theory of the alliance market 67 table 3.3 Self-reliance. Self-reliance

The larger the power share of the state A

+

table 3.4 Probability and cost of entrapment.

The larger the power share of the state A The larger the power share of the current ally (allies)

Probability of entrapment

Cost of entrapment



N/A

+

N/A

the vulnerability inherent in a state’s inability to supply everything it needs (that is, the problem of self-reliance). It is important to note that abandonment creates costs that a state would not suffer if it never entered the alliance (recall the discussion about the division of labor and vulnerability in Chapter Two). As trade ties create dependence on trade partners, alliances also make states more vulnerable without them. The larger the exchange partner is, the more costly the severance of their ties (that is, abandonment) is. Because the measures to reduce the fear of abandonment tend to increase the fear of entrapment (and vice versa), it is important to analyze the fear of entrapment as well (see Table 3.4). Fear of entrapment can be defined as follows: fear of entrapment = (probability of entrapment) × (cost of entrapment) Other things being equal, power shares of the state and its ally should not affect the cost of entrapment because the cost is primarily determined by the nature of the event with which a state is entangled and by how divergent the interests of the allies are. Information about the allies alone can explain some variations in the cost of abandonment, whereas the cost of entrapment largely depends on actors or events outside the alliance.32 It is very costly, for example, for an ally of the United States to be abandoned by the United States even in peacetime, without consideration of any specific adversary. In contrast, the cost of entrapment by the United States will vary significantly, depending on the event or the adversary (for example, sending troops to Iraq for political support versus fighting against China over Taiwan). Timing and circumstances matter for both a­bandonment

68

a systemic theory of the alliance market

and entrapment, but “By whom?” is the most important question for the cost of abandonment, whereas “Into what?” matters most for the cost of entrapment. With respect to the probability of entrapment, I argue that the larger A’s power share is, and the smaller the ally’s power share is, the less likely A is to become entrapped. First, stronger states have more capacity to refuse entrapment and coerce weaker states into entrapment. Second, stronger states have better control of the overall situations outside the intra-alliance relationship. Third, weaker states tend to be more dependent on their allies, and the dependence makes entrapment more likely. In short, stronger states should be more successful in entrapping their allies than weaker states are; hence, there is a larger likelihood of entrapment for weaker states. Many analyses focus on stronger states’ fear of entrapment rather than that of weaker states (for example, Cha 1999, 43; Morrow 2000, 79; Cha 2010), but this does not mean that stronger states have more fear of entrapment than weaker states do. As Glenn Snyder (1984, 1997) shows, states balance the fears of abandonment and entrapment. The absolute levels of the fears and the relative balance between them should be distinguished. Stronger states have more room for worrying about entrapment because they worry less about abandonment than weaker states do. This, of course, does not mean that the absolute risk of entrapment (as opposed to the risk relative to that of abandonment) is higher for stronger states than for weaker states. In addition, we should note that the literature on entrapment has mostly focused on military affairs. Stronger states worry about military entrapment because they supply military protection to their weaker partners. Weaker states, however, have more reasons to worry about entrapment in general than stronger states do. Although weaker states may lack military resources to be exploited, they are more likely to be dragged into an undesirable situation by their allies.33 When stronger states are involved in their allies’ military conf licts, the stronger states are more likely to be able to pursue their own interests.34 Within individual alliances, states adjust the risk of abandonment and entrapment by various institutional mechanisms (for example, stationing troops in allies’ territories or integrated command), but my systemic argument explains the sources of these concerns before the adjustment takes place. Even if institutional forms of alliances explain a great deal of the variation in the fear of abandonment and entrapment, we still need to identify the effects of systemic and domestic factors in order to understand why certain arrangements were made in an alliance.



a systemic theory of the alliance market 69

Intersystem Comparison I now derive some conjectures about systems with different numbers of great powers by considering the relative capabilities of states. To summarize the analysis of the previous section with respect to capabilities: (a) The stronger A is and the weaker its ally is, the less A fears abandonment; (b) the stronger A is and the weaker its ally is, the less A fears entrapment. Generally, a unipolar power should be stronger than a bipolar power, and a bipolar power should be stronger than a multipolar power. Then, a great power A should have the least fear of abandonment and entrapment under unipolarity, and the fear should be larger when there are more great powers (that is, poles).35 From the perspective of the great power A’s ally, in turn, the fear of abandonment and entrapment should be smaller when there are more poles in the system. These considerations suggest that the intraalliance bargaining power of a great power decreases as the number of poles increases. This result is, in part, contradictory to a widely held view in the literature, namely, that smaller powers have less fear of abandonment (Snyder 1984, 483–485) and more bargaining power with their patrons in a bipolar system than in a multipolar system (Keohane 1971; Risse-Kappen 1995; Harrison 2003).36 The gap between the model and the standard argument comes from a neorealist assumption that two superpowers in a bipolar system do not ally with each other.37 Snyder justifies the assumption as follows: “The two superpowers will always be rivals, never allies, so long as they are superpowers. The sufficient reason for this is that there is no other state in the system powerful enough to threaten either one and thus to provide an incentive to ally” (1997, 19).38 When we assume that not all great powers can be in the same camp, great powers in a bipolar system are put at a special disadvantage because they have to forsake a large portion of the power share left in the system for an alliance. A bipolar power cannot ally with the other bipolar power, whereas that is irrelevant for a unipolar power, and great powers in multipolar systems can ally with other great powers except for one. A loss of a nongreat power ally is a relatively minor problem for a great power. With the neorealist assumption, however, the minor loss is more important in a bipolar system because the loss is not readily recovered by an alternative alliance, and the loss is most likely a gain for the other great power. Conceivably, this is the reason for the standard argument. When a great power has more fear of abandonment—in the case of bipolarity, because a great power has

70

a systemic theory of the alliance market

l­imited alliance options—it would tighten its alliances, and its weaker partners would have less fear of abandonment and more bargaining power; hence, the small powers have more inf luence in a bipolar system. Changes in systemic polarity affect the fear of abandonment and entrapment by altering the relative capabilities of the allies and the availability of an alternative ally. The most important points are that, other things being equal, (a) a great power has weaker intra-alliance bargaining power vis-à-vis its lesser allies when there are more poles in the system, but (b) two great powers in a bipolar system are at a special disadvantage because they do not ally with each other. These points, in combination with the analysis in the previous sections, explain basic characteristics of different systems. For instance, a great power in a multipolar system has to be sensitive about other states’ alliance policies. A state has more options for alliances in a multipolar system, but so do its allies, including the great power allies, who can make a big difference. In contrast, alliance politics in a bipolar system tend to be less f luid because every state has fewer options for security suppliers and especially because the two superpowers have a stable alliance policy—the bipolar powers more readily extend their commitment and try to retain the existing alliances because they do not have good alternatives. Because neither of the two great powers would replace their allies with the other giant in the system, other states have less need to worry about abandonment and have more leverage over their patrons. In a unipolar system, the single superpower has huge advantages in alliance games because it has the least fear of abandonment and entrapment. When a state has little fear of abandonment, it can demand more from its allies. When a state has little fear of entrapment, it can pursue alliance policies that are risky vis-à-vis third parties (for example, NATO expansion to the east). The sole superpower’s bargaining power is strong, and it becomes reluctant to extend or retain alliance commitments while other states work hard to be allied with the unipole. Because the sole superpower has few structural constraints, its alliance policy can become haphazard. U.S. foreign policy in the unipolar era, therefore, is characterized not only by weak restraints in adversary games but also in alliance games.39 Thus, in the context of alliance politics, the vice to which a great power easily succumbs in a multipolar world is overreaction; in a bipolar world, overextension; in a unipolar world, inattention.40 Because the model analyzes the source of intra-alliance bargaining power in terms of the fear of abandonment, it has implications for the termination,



a systemic theory of the alliance market 71

duration, and formation of alliances. In the following passages, I list the hypotheses to be tested in Chapter Four. The first hypothesis concerns the modes of alliance termination in different systems: Hypothesis 3.1: Other things being equal, alliances in a multipolar system are more likely to be abrogated than are those in a bipolar or unipolar system. There are important differences in how alliances end: I focus here on the opportunistic termination of alliance agreements. Alliances are sometimes abrogated (that is, prematurely terminated in violation of their terms), but alliances can end also because they fulfilled their goals (and are not renewed), because one or more members lost independence or because they are replaced with a new agreement with different obligations (Leeds and Savun 2007). I have explained that great powers have incentives to overreact in alliance politics of a multipolar system, but such incentives primarily relate to abrogation and not to the other forms of termination. Under bipolarity, states can have minor intra-alliance disputes without completely upsetting their entire alliance arrangements. The two bipolar powers have incentives to maintain their respective alliances, and this knowledge has a stabilizing effect on other states as well. For example, the United States partially abandoned the United Kingdom and France in the Suez Crisis of 1956, but this did not seriously cast doubts on American commitment to the Western alliance. Charles de Gaulle’s partial abandonment of NATO did not arouse the fear of radical changes in France’s alliance policy either.41 Because intra-alliance disputes under bipolarity can be addressed with more confidence that the alliance agreements will continue, they are less likely to prompt abrogation. In contrast, the model suggests that terminations of alliances would be abrupt and radical—that is, alliances are less likely to muddle through with minor adjustments and partial abandonment—in a multipolar system. Alliances under multipolarity are more likely to experience opportunistic termination and even drastic realignment (for example, joining an enemy camp). For instance, under multipolarity, Greece, Serbia and Bulgaria were allies who fought against the Ottoman Empire in the First Balkan War (1912–1913), but in the Second Balkan War (1913), Bulgaria fought against its former allies Greece and Serbia. Austria and Prussia ended their alliance in 1866 with the Austro–Prussian War, although they afterwards formed an alliance that lasted until the First World War.

72

a systemic theory of the alliance market

As Snyder points out, states in a multipolar system “can readily defect and realign” because “the system itself does not foreclose alternatives” (Snyder 1997, 19); “the structural availability of realignment options,” in turn, “maximizes fears of allies’ defection” (Snyder 1997, 346). In the f luid alliance market of a multipolar system, states can change their course of foreign policy more radically, and they have incentives to preempt other states’ betrayal. Because states tend to be sensitive about each other’s alliance policy in a multipolar system, their alliance policy can become hyperreactive and cause a decisive realignment. In extreme cases, such realignments led to war between allies.42 Finally, systemic dynamics under unipolarity should largely depend on the choice the unipolar power makes, but I expect them to be less hyperreactive than under multipolarity. In the case of the actual unipolarity after the Cold War, the United States chose to retain most of its alliances and to reassure its allies about continuity, albeit with more demands on its lesser allies. Moving to another aspect of alliance politics, Hypothesis 3.2: Other things being equal, alliance memberships are more durable in a bipolar system than in a multipolar system. My model suggests that alliances, especially great powers’ alliances with nongreat powers, are more durable in a bipolar system than in a multipolar system because great powers have more incentives to retain their allies under bipolarity. Numerous Cold War alliances are still active today, and even unsuccessful alliances such as the South East Asia Treaty Organization (SEATO, 1954–1977) and the Central Treaty Organization (CENTO, 1955–1979) lingered for decades under bipolarity, despite their dismal performances. Termination and duration of the unipolar power’s alliances are hard to predict because the unipole can make arbitrary decisions. Moreover, two opposite forces are working in a unipolar system. Whereas the unipole is reluctant to continue its commitment (for the same price), its allies are willing to pay more to retain the sole superpower’s commitment. In a transition to unipolarity, therefore, either continuation or termination of the unipole’s alliances is consistent with my theory—what my theory explains is who becomes willing to pay more. Thus, I focus on the difference between multipolarity and bipolarity in Hypothesis 3.2. The same prediction can be derived from the conventional wisdom that alliances are f luid under multipolarity and rigid under bipolarity (Waltz 1979; Duncan and Siverson 1982; Snyder 1997), but the reasoning is impor-



a systemic theory of the alliance market 73

tant, especially in relation to other aspects of alliance politics. Hypothesis 3.2 is based on the deductive argument developed earlier, and alliance memberships are more durable under bipolarity because the two great powers’ bargaining position is weak. Although Waltz (1979, 169–171) observes the rigidity of alliances under bipolarity, he argues that two great powers in a bipolar system do not worry about lesser allies’ policy because they do not matter in the central balance of power. For capability aggregation or the balance of power, the two bipolar powers should not care much about maintaining alliances because the military values of lesser allies are not large and the two superpowers can rely on their own military power. In contrast, the allegiance of nongreat powers is important in my t­ heory, in which great powers need customers for the military protection they export. As I repeatedly argue in this book, military force is not the only good exchanged between allies. From the exchange perspective, it is wasteful not to export military commitments, and the lesser allies are the only available customers for the two great powers in a bipolar system. As noted earlier, lesser allies are less important to great powers of a multipolar system, because there are great power security importers. The importance of lesser allies under bipolarity imposes asymmetric constraints on the two great powers in a bipolar system. When Snyder (1997, 196) states that “the allegiance of allies is hardly in question” under bipolarity, he assumes that both great powers and nongreat powers are constrained by the systemic forces. In my model, bipolarity’s systemic constraint is only on the two great powers, who have to compete for the lesser allies’ loyalty because they lack good alternatives. Lesser allies during the Cold War indeed had clear alliance policy, but this clarity was not due to the number of great powers.43 Western states were loyal to the United States because they preferred U.S. military protection over the Soviet alternative. Eastern European states were “loyal” to the Soviet Union because that was the only option available to them; defection by an Eastern European state was likely to invite a Soviet intervention, as seen in Hungary in 1956 and in Czechoslovakia in 1968. In other words, during the Cold War, the quality of the product and the location of the suppliers shaped the incentives for nongreat powers, whereas the number of suppliers inf luenced the two great powers’ policy.44 Turning to the issue of alliance formation, Hypothesis 3.3: Other things being equal, a great power is most selective in offering alliance commitments in a unipolar system and least selective in a bipolar system.

74

a systemic theory of the alliance market

The model predicts that the sole superpower in a unipolar system is likely to be selective in forming new alliances because it does not compete with other suppliers. In contrast, superpowers in a bipolar system more readily extend their commitments to lesser powers because of their limited alliance options. In a multipolar system, great powers compete as suppliers, but the competition for lesser allies is limited because a large part of alliance transactions under multipolarity is between great powers. In other words, the supply curve for a great power’s alliance commitment is furthest toward the right under bipolarity and furthest toward the left under unipolarity. With respect to bargaining power of a great power in the alliance market, we can derive the following general hypothesis from the systemic model: General Hypothesis: Other things being equal, the bargaining power of a great power in the alliance market is the strongest in a unipolar system, weaker in a multipolar system, and weakest in a bipolar system. The supply of military commitment (that is, the quantity of commitment offered for a given price) from a great power, therefore, should be largest under bipolarity and smallest under unipolarity. Of course, the assumption of other things being equal does not hold in reality. For instance, the advent of unipolarity coincided with the demise of the Soviet Union and a significant decline in demand for security. This is a problem when we focus on the quantity of security produced because we cannot clearly separate out the effect of reduced supply from reduced demand (did the production reduce because states are less willing to sell or less willing to buy?). However, evidence of a higher price for security would support my argument because reduced demand would have led to a lower price unless supply declined even more. I argue that the United States became more unilateralist after the Cold War, not just because it became the unipolar power in “adversary games” but also because it turned into the unipolar power in “alliance games” (Snyder 1997). Given the relatively peaceful international environment in the post–Cold War era, what really affected American foreign policy was the attractiveness of the United States in the alliance market rather than its power to pressure adversaries. Unipolarity in adversary games (that is, being potentially the biggest bully) may trigger balancing from other states, but unipolarity in alliance games creates a different dynamic, as will be explained in the next chapter. Table 3.5 summarizes the preceding analysis. From the general hypothesis on the bargaining power and systemic polarity, I derive the following hypotheses, assuming that other factors



a systemic theory of the alliance market 75 table 3.5 Implications of the model for alliance politics.

Multipolarity Bipolarity Unipolarity

Mode of termination Abrupt/radical Gradual Indeterminate Duration Short Long Indeterminate Formation Great powers The two great powers The single great are active are not selective power is selective Great powers’ bargaining power vis-à-vis nongreat Moderate Weak Strong powers

(most important, demand for military commitment) do not change too drastically: Hypothesis 3.4: Other things being equal, when the international system becomes unipolar, the single great power reduces its commitment to existing and potential allies. Whereas Hypothesis 3.3 is about forming alliance ties, Hypotheses 3.4 to 3.6 focus on the adjustment of goods exchanged between allies. Because the single great power has a huge advantage in the unipolar alliance market, it can raise the price or reduce the quantity of its export, military commitments. The quantity of military commitment provided by a great power should decline with a shift to unipolarity, and the price of the alliance commitment should increase unless the demand declines more than the supply. The observable implications of this hypothesis would be reduced troop levels committed to the defense of the allies, reduced military and economic aid to the allies, and a lower level of alliance commitment promised to new allies. Meanwhile, other states try to retain or buy an alliance with the unipole, even for an increased price, and they also actively seek new allies, who now have a relatively cheap price. Thus, Hypothesis 3.5: Other things being equal, when the international system becomes unipolar, nongreat powers increase their commitment to existing and potential allies. Assuming that the demand for the unipole’s commitment does not decline more than the supply, the price other states pay to the unipole should increase. We should observe increased concessions to the unipolar power in military, economic, and political affairs. We should also find active exchanges of these goods among nongreat powers (that is, active alliance formation) who cannot afford to buy the unipole’s commitment.

76

a systemic theory of the alliance market

The increased bargaining power of the unipole should also affect arrangements of its alliances, in the direction that favors the sole superpower. Hypothesis 3.6: Other things being equal, when the international system becomes unipolar, the single great power removes restraining arrangements with its allies. Increase in unilateral policies by the unipole in general would be evidence for this hypothesis, but my test for this hypothesis will focus on one particularly important arrangement between military allies, that of military bases. Bases on allies’ territories are specific assets—assets that have lower values outside the specific relationship—for basing countries, although they may not be so for hosting countries.45 When an alliance dissolves, the investment made for the bases is wasted and cannot be used for alternative purposes. This creates an incentive to continue the alliance, and this incentive is different from trip-wire effects and commitments, which are about reliability in wartime rather than continuation of an alliance in general. The unipole can forgo such restraining arrangements because it has a strong bargaining advantage. This chapter developed a systemic theory of alliance bargaining with a focus on the incentives for great powers. In Chapter Four, I will test the hypotheses discussed in the preceding pages.

4

Alliance Politics and Polarity

This chapter tests the hypotheses presented at the end of Chapter Three. As set in Chapter One, my samples all include a defense, offense, neutrality, or consultation obligation and exclude pure nonaggression pacts. I will later explain the variations among agreements with different obligations, but these agreements all facilitate the success of the allies in the event of military conf lict. The final section of this chapter discusses the theoretical implications of the systemic theory and offers some predictions about U.S. alliances in the long and short terms.

Modes of Alliance Termination I argue that the terminations of alliances tend to take more radical forms in a multipolar system, where states have the maximum f lexibility in choosing their allies. Hypothesis 3.1: Other things being equal, alliances in a multipolar system are more likely to be abrogated than are those in a bipolar or unipolar system. My test for this hypothesis builds on Leeds and Savun (2007), who analyze factors that inf luence the likelihood of abrogation while explicitly acknowledging the other possible modes of alliance termination (fulfillment, exogenous loss of independence, and renegotiation). The sample for this hypothesis consists of alliance-years between 1816 and 2001 (but limited to bilateral alliances formed between 1816 and 1989).1 I follow their use of competing risks duration analysis and treat modes of termination other than abrogation as censored. The dependent variable is abrogation of an alliance agreement, which happens when “(1) a major provision is violated and the governments do not indicate their intention to continue to recognize the alliance in spite of the violation or (2) one or more of the allied governments specifically declares that it will no longer recognize or be bound by the alliance agreement despite the fact that the alliance has not reached its scheduled termination date” (Leeds and Savun 2007, 1124). My central explanatory variable

78

alliance politics and polarity

for this hypothesis is MULTIPOLAR, which is coded 1 when the year of observation is between 1816 and 1945 and coded 0 if the year of observation is between 1946 and 2001. To examine the effect of systemic polarity, it is necessary to control for the effects of other variables. The following control variables are all from Leeds and Savun (2007), and I refer readers to their work for more details. CHANGE IN INTERNATIONAL POWER is a dummy variable that is coded 1 when either of the alliance members has experienced a more than 10 percent change in the Composite Index of National Capabilities (CINC from the Correlates of War; Singer 1988) since the formation of the alliance. CHANGE IN POLITICAL INSTITUTIONS is a dummy variable that is coded 1 when either of the alliance members has experienced at least a two-point change in their democracy scale and autocracy scale of Polity IV dataset (Marshall and Jaggers 2002) since the formation of the alliance. DECREASE IN EXTERNAL THREAT is a dummy variable that is coded 1 when the external threat faced by either member has decreased more than 30 percent since the formation of the alliance. The external threat is based on the capabilities and foreign policy orientation of states that are in the “politically relevant international environment” (Maoz 1996) and do not share an alliance with the state in question. NEW OUTSIDE ALLIANCE is a dummy variable that is coded 1 when either of the alliance members forms a new alliance with a state outside the alliance. DEMOCRATIC MEMBERS is coded 0 if neither of the alliance members is democratic (here meaning a score of 6 or higher on the composite democracy scale of Polity IV data; Marshall and Jaggers 2002); 1 if only one of the members is democratic and 2 if both members are democratic. ASYMMETRIC POWER is a dummy variable that is coded 1 for alliances between a major power and a minor power and coded 0 for alliances that are between two major powers or two minor powers according to the Correlates of War project’s classification of major powers (Small and Singer 1982). NONMILITARY COOPERATION is a dummy variable that is coded 1 when the alliance agreement has a clause on nonmilitary cooperation. RATIFICATION is a dummy variable that is coded 1 when the alliance was formed by a treaty requiring ratification. Finally, PEACETIME COORDINATION is a trichotomous variable that is coded 0 for low, 1 for moderate, and 2 for high levels of institutionalization regarding peacetime military coordination (Leeds and Anac 2005).



alliance politics and polarity 79 table 4.1 Cox partial likelihood estimates of alliance abrogation. Variables

MULTIPOLAR CHANGE IN INTERNATIONAL POWER CHANGE IN POLITICAL INSTITUTIONS DECREASE IN EXTERNAL THREAT NEW OUTSIDE ALLIANCE DEMOCRATIC MEMBERS ASYMMETRIC POWER NONMILITARY COOPERATION RATIFICATION PEACETIME COORDINATION N of observations (alliances) Log likelihood

Coefficients (robust standard errors)

Substantive effects (minimum to maximum)

0.676* +97% (0.252) 0.927* (0.303) 0.543* (0.268) 0.368 (0.242) 0.805* (0.204) –0.555* (0.224) –0.779* (0.240) –0.551* (0.270) –0.172 (0.276) 0.733* (0.143) 3955 (293) –454

+153% +72% No significant effect +124% –67% –54% –42% No significant effect +333%

* p < 0.05

Table 4.1 presents the result of the Cox proportional hazard model that treats modes of termination other than abrogation as competing risks. The “substantive effect” is shown as the percentage change in the hazard rate when the value of the independent variable changes from the minimum to the maximum (0 to 1 or 0 to 2). The Efron method is used for handling ties. Robust standard errors are calculated “assuming potential nonindependence among cases associated with the same alliance” (Leeds and Savun 2007, 1129; Leeds, Mattes, and Vogel 2009, 473). Following BoxSteffensmeier and Jones’s recommendation (2004, 137), Grambsch and Therneau’s global proportional hazards test statistic and Harrell’s rho were calculated, and I found that the proportionality assumption was not violated. The results for the control variables are very similar to the findings of Leeds and Savun (2007), but, most important, the effect of multipolarity on the likelihood of abrogation is statistically and substantively significant. Controlling for all the variables discussed in the preceding paragraphs, an alliance under multipolarity is almost twice as likely to end in violation

80

alliance politics and polarity

of its terms as an alliance in other systems. Thus, systemic polarity affects how alliances end. In the next section, I test the hypothesis on the effects of polarity on the duration of alliance memberships.

Duration of Alliance Memberships I argue that alliances, especially those involving great powers, are more durable in a bipolar system than in a multipolar system because the two bipolar powers are more willing to retain their allies than multipolar powers, and the bipolar powers’ alliance policy has an assuring effect on other states’ policy as well. As shown in the previous section, alliance members under multipolarity have a higher likelihood of facing drastic changes in their allies’ policy, and the risk has destabilizing effects. My theory is agnostic about the duration of alliances under unipolarity because it depends on the policy of the unipole, which can be haphazard.2 Other states’ alliance policies will be significantly affected by what the unipole decides to do (imagine, for example, a counterfactual world in which the United States had abandoned its alliances after the end of the Cold War). Moreover, on the empirical side, too many alliances formed in the unipolar era are still in effect to make a comparison with other systems.3 In the following analysis, I focus on individual memberships rather than alliances as a whole because there are many instances of a member exiting an alliance without bringing the entire alliance to an end and because an exit of a member is important even when the rest of the alliance continues. Hypothesis 3.2: Other things being equal, alliance memberships are more durable in a bipolar system than in a multipolar system. According to the Alliance Treaty Obligations and Provisions (ATOP) dataset (version 3.0), Russian or the Soviet Union’s alliance memberships that began in the multipolar period lasted on average about 90 months whereas its alliance memberships beginning during the Cold War have the average duration of 247 months. U.S. alliance memberships that began in the multipolar period on average lasted only 39 months whereas those beginning in the bipolar era lasted for 359 months. Such comparison, however, is only partially informative for my current purpose; for example, duration reported is up to December 2003 in the ATOP dataset, and many Cold War alliances are still active (and thus right censored). Taking into account the right censoring, Figure 4.1 shows the Kaplan-Meier survival curves for all alliance members and for the United States and Russia/the Soviet Union from 1815 to 1989. The figure clearly shows that alliance memberships are



alliance politics and polarity 81 1.00 0.75

All alliance memberships

0.50 0.25 0.00 0

10,000

20,000

30,000

Time in days 1.00 Only memberships of United States and Russia/USSR

0.75 0.50 0.25 0.00 0

10,000

20,000

30,000

Time in days Multipolar

Bipolar

Figure 4.1.  Survival of alliance memberships (1815–1989).

more likely to endure under bipolarity than under multipolarity. It is, however, important to control for the effects of other variables that affect the duration of alliance memberships. I used Cox models to analyze the alliance membership duration because many alliance memberships that began during the Cold War are still active (that is, their durations are right censored) and also because there is no strong theoretical reason to expect a particular form of the duration dependency (Box-Steffensmeier and Jones 2004). My dependent variable for the models is the conditional probability that an alliance membership ends in a given year. The duration is measured by days to minimize the number of observations with tied survival times, and the fifteenth day of the month was used for some observations where the date of termination is missing in the ATOP dataset (version 3.0). For multiphase alliances, the ATOP’s member-level data have separate entries for different phases of memberships

82

alliance politics and polarity

(for example, NATO’s first phase lasts only from 1949 to 1951), but I analyze the entire duration of the memberships. The time scope is between 1816 and 1989 because I am comparing multipolarity with bipolarity.4 My central explanatory variable is BIPOLARITY, a binary variable that is coded 1 if the year of observation is between 1946 and 1989 and coded 0 if between 1816 and 1945.5 The literature on alliance duration and reliability has identified some important explanatory variables, and I include them as control variables here (Bennett 1997; Reed 1997; Leeds and Savun 2007; Leeds, Mattes, and Vogel 2009).6 Alliances formed in wartime are likely to have limited duration, and the ATOP dataset has a dummy variable WARTIME, which is coded 1 “if any member of the alliance was a participant in a war . . . at the time the alliance was initially formed” (Leeds 2005). Because the literature has found military capabilities to be an important factor in alliance politics, I included the members’ capabilities (Composite Index of National Capability from the Correlates of War version 4.0; Singer, Bremer, and Stuckey 1972) in the year of observation and their proportional increase from the first year of membership.7 To account for the distribution of capabilities within the alliances, I also used CAPRATIO-LOG, the natural log of the ratio of the strongest member’s capability to that of the weakest member. Democratic states are expected to have more durable alliance commitments (Gaubatz 1996; Bennett 1997; Reed 1997), and I use DEMOCRACY, the revised combined polity score of the Polity IV project (Marshall and Jaggers 2002; version 2012) to control for the effect. Because a change in the regime type is likely to lead to a change in the foreign policy interest of the state, changes in the polity score, in either direction, may prompt alliance membership termination. Thus, I control for the absolute value of the difference between DEMOCRACY in the first year of membership and DEMOCRACY in the year of observation. To account for each alliance’s collective level of democracy, I also checked the effects of the average, the highest, and the lowest polity score of each alliance’s members (DEMOCRACY-MEAN, DEMOCRACY-HIGH, and DEMOCRACYLOW) in the year of observation. Institutional features of alliances also affect the duration. Ratification requirement is expected to increase the duration by giving political significance to the agreement. Ratified agreements have support from both the executive and legislative branches of the member government. The variable RATIFICATION is the dummy variable ESTMODE from the ATOP dataset, which is coded 1 “if the written document establishing the alliance



alliance politics and polarity 83

is a formal treaty requiring domestic ratification” (Leeds 2005). Alliance agreements that include provisions for nonmilitary cooperation either foster or ref lect common interests among the members, and they are expected to be associated with a longer duration. NONMILITARY-COOPERATION is the dummy variable NOMICOOP from the ATOP dataset, which is coded 1 “if the alliance agreement includes provisions for cooperation on non-military issues, for instance, economic cooperation, protection of minorities, scientific or cultural exchange, environmental protection, etc.” (Leeds 2005). Both bilateral and multilateral alliances face the systemic pressure explained in Chapter Three, but I control for bilateral alliance memberships in case they may have a different dynamic from those of multilateral memberships.8 The ATOP dataset reports when an alliance agreement has a specified duration, either by time period or by duration of a specific event (SPECLGTH = 1 or 2), and this should impose a limit on the duration of an alliance membership. I create a dummy variable SPECIFIED-LENGTH that is coded 1 when the agreement has a specified duration. Finally, for variables that violate the assumption of nonproportional hazards, I add an interaction term between the covariate and a function of time (for example, natural log of duration days = ln[t]) to account for the nonproportional hazards (Box-Steffensmeier and Zorn 2001).9 The Efron method is used for handling ties. Robust standard errors are calculated “assuming potential non-independence among cases associated with the same alliance” (Leeds and Savun 2007, 1129; Leeds, Mattes, and Vogel 2009, 473). The results are presented in Table 4.2. The interpretation of Table 4.2 is not straightforward because I have to include interaction terms to account for nonproportional hazards. Substantive effects of the variables can be calculated with a formula provided by Box-Steffensmeier and Jones (2004, 60), but we need to consider the interactions. The coefficient for BIPOLAR is positive, suggesting that the risk of termination is actually higher under bipolarity on the first day of membership, but the effect becomes insignificant as early as the second day. Meanwhile, the coefficient for its interaction term (BIPOLAR × natural log of duration) is negative and, after about 371 days (that is, exponential of [the coefficient for BIPOLAR/the coefficient for BIPOLAR × natural log of duration]), memberships under bipolarity begin to have lower hazard rates than those under multipolarity. Only about 6.2 percent of the observations for the Cox model in Table 4.2 have 371 or fewer days of duration, and bipolarity prolongs alliance memberships for the vast majority.

84

alliance politics and polarity table 4.2 Bipolarity and the risk of alliance membership termination.1

Coefficients (robust standard errors)

BIPOLAR at the time of observation 2.72** (1.39) BIPOLAR × natural log of duration –0.46*** (0.17) WARTIME entry 2.92** (1.32) WARTIME × Natural log of duration –0.26 (0.18) PROPORTIONAL INCREASE in CAPABILITY –0.08 (0.09) DEMOCRACY –0.14** (0.07) DEMOCRACY × natural log of duration 0.016* (0.009) DEMOCRACY-CHANGE 0.05*** (0.01) RATIFICATION –0.65*** (0.18) NONMILITARY-COOPERATION 0.53 (1.22) NONMILITARY-COOPERATION × natural log of duration –0.13 (0.16) BILATERAL –0.10 (0.21) BILATERAL × duration 0.00003 (0.00004) SPECIFIED-LENGTH –1.53 (1.04) SPECIFIED-LENGTH × natural log of duration 0.23* (0.14) N of observations 13538 (memberships) (991) (alliances) (356) *** p < 0.01, ** p < 0.05, * p < 0.1 1ATOP ID 1467, formed by Bolivia and Ecuador in 1911, is by far the oldest alliance active as of the end of 2003, but the results are very similar even when I drop that alliance from the analysis.

Figure 4.2 shows the effects of bipolarity on hazard rate over time.10 After 1,062 days, the upper 95 percent confidence bound goes below 0, meaning that the membership-prolonging effect of bipolarity becomes statistically significant. For over three-quarters of the observations (with duration of 1,675 days or more), the risk of membership termination is more than 50 percent lower under bipolarity than under multipolarity. The risk is 65 percent lower under bipolarity for memberships that lasted for 10 years, and 74 percent lower at 20 years. Thus, even after controlling for the other



alliance politics and polarity 85 150%

100%

50%

0%

–50%

–100% 300

5,000

10,000

15,000 20,000 Analysis time in days

25,000

30,000 33,861

Hazard rate relative to multipolarity

Figure 4.2.  Effects of bipolarity on hazard rate after 300 days. The dotted lines indicate the 95 percent confidence interval.

variables, alliance memberships are more durable under bipolarity than under multipolarity once they survive the first few years.11

Unipolarity and Reduced Supply of Military Protection With respect to alliance formation, I argue that the unipole should be reluctant to form new alliances or, more generally, to extend alliance commitments, while the two superpowers in a bipolar system actively seek alliance ties: Hypothesis 3.3: Other things being equal, a great power is most selective in offering alliance commitments in a unipolar system and least selective in a bipolar system. The following analysis does not actually control for other factors and is only a partial test, but the trend of U.S. alliance formation seems to support the hypothesis. The United States has been less enthusiastic about offering alliance commitments than during the bipolar era. Because the United States (employer) has large advantages in the alliance market, it can now pay less to its allies (employees) and court support of temporary coalition partners (part-time workers) without giving them formal commitments.12

86

alliance politics and polarity

10% 8% 6% 4% 2% 0% 1910

1920

10%

1930

1940

1950 1960 Year

1970

1980

1990

2000

1990

2000

Limited to ties with defense of offense obligation

8% 6% 4% 2% 0% 1910

1920

1930

1940

1950 1960 Year

1970

1980

Figure 4.3. The proportion of new U.S. alliance ties in the system.

The United States has formed 20 alliances since the end of the Second World War, but it formed no new alliances after 1981, despite its ever more extensive military presence across the world and its newly formed security arrangements with numerous countries. This is not due to the lack of demand from other states: In 2015, for instance, the White House rejected a proposal from Gulf nations to forge a common defense treaty with the United States (Hudson 2015). Since 1990, the United States has offered alliance commitments only when new members joined the existing alliances it formed during the Cold War—that is, when the newly unified Germany replaced West Germany in NATO, and when NATO or OAS expanded. Even when we count the modest expansion of these existing alliances, which does not seem to be motivated by American security concerns, the American share of new alliance ties has significantly declined from that of the Cold War period. Figure 4.3 shows the proportion of new U.S. alliance ties to the new ties formed across the system each year—measured in directed dyads and not counting multiple agreements between the same pair of states—between 1908 and 2003. The upper part of the figure includes neutrality pacts and consultation pacts, and the lower part is limited to ties with a defense or



alliance politics and polarity 87

offense obligation.13 Ref lecting its isolationist tradition and more modest international profile, the United States did not form many alliance ties (none with defense or offense obligation) until 1942, within the time scope of the ATOP dataset (version 3.0).14 The bar charts indicate that the United States formed new alliance ties much more frequently (more years and higher proportions) during the Cold War than in the post–Cold War period. Furthermore, the post–Cold War U.S. alliance ties were limited to the expansion of existing alliances, and new members of NATO were required to satisfy certain conditions (Kydd 2001). The United States accounted for 1.9 percent (48/2,548) of new allied directed dyad-years between 1815 and 1945, 2.2 percent (45/2,010) between 1946 and 1989, and 1.0 percent (13/1,270) between 1990 and 2003. The trend is similar when I count multiple alliance agreements between the same pair of states. Between 1815 and 1945, the United States accounted for 2.1 percent (69/3,252) of alliance commitments extended across the system.15 Between 1946 and 1989, the United States extended 104 alliance commitments, accounting for 2.8 percent of the system total (104/3,736). Between 1990 and 2003, in contrast, the United States offered 13 alliance commitments (NATO and OAS), and it accounts for only 0.7 percent of 1,850 alliance commitments made during the period.16 One reason the United States did not form many new alliance ties in the unipolar era may be the large number of its existing allies. In Figure 4.4, therefore, I show the proportion of U.S. alliance ties to the system total, with not just the first years of directed dyads but also including the subsequent years. Although the United States maintained most of its Cold War alliances, the U.S. presence in the alliance market proportionately declined because it did not actively seek new allies. Of course, a state’s share of new alliance ties depends on various factors, and the foregoing analysis is only a first cut. In addition to the willingness of the state to offer commitments, the pattern of alliance formation should be affected by other states’ willingness as well. For instance, the Soviet Union retained great power status while transitioning from multipolarity to bipolarity, but it actually had a lower share of alliance ties under bipolarity than under multipolarity. During the Cold War, many states were reluctant to form ties with the Soviet Union because such ties were incompatible with American alliance networks and also because these states had their own reservations about communism.17

88

alliance politics and polarity

10%

8%

6%

4%

2%

0% 1910

1920

1930

1940

All alliance ties

1950 1960 Year

1970

1980

1990

2000

Defense/offense obligation

Figure 4.4. The proportion of U.S. alliance ties.

Reduced Supply and Higher Price of Military Protection I argue that the bargaining power of a great power vis-à-vis lesser powers is the strongest in a unipolar system and the weakest in a bipolar system.18 The systemic transition from bipolarity to unipolarity, therefore, should have profound effects on alliance politics of the unipole and the other participants of the alliance market. Although the contractual nature of alliance agreements moderates incentives for behavioral change, alliance politics under unipolarity seems to be substantially different from that of the bipolar era. Closely related to Hypothesis 3.3, the unipole has incentives to reduce the supply of security and to raise its price. The following discussion focuses on supply of military protection, but the United States has reduced its supply of other goods as well. In the U.S.– Philippines alliance, for example, Washington not only downgraded the security partnership with the island nation but substantially cut down economic and military aid to Manila (Fisher 1999; Park 2011). The proportion of foreign aid in U.S. government spending had been steadily declining throughout the Cold War, but it seems no coincidence that the lowest point came in the mid-1990s (Tarnoff and Nowels 2004, 14–15).



alliance politics and polarity 89

Hypothesis 3.4: Other things being equal, when the international system becomes unipolar, the single great power reduces its commitment to existing and potential allies. Some might object to the hypothesis by pointing out that only a couple of U.S. alliances dissolved after the Cold War. The U.S. alliance with Israel terminated when the agreement’s target, the Soviet Union, collapsed, and the U.S. alliance with Panama ended with the return of the Panama Canal in 1999.19 The limited number of alliance terminations does not negate the hypothesis, however. A major rationale for forming an alliance, after all, is to make it difficult to sever relationships when circumstances change. Only five out of the 20 alliances formed by the United States during the Cold War had a specific time period for their effects, and it is not surprising that the United States has not terminated its Cold War alliances unilaterally. Instead of terminating alliances, the United States has become more demanding toward old and new allies. For example, for new allies, the United States has raised the price of entry into its alliances; new NATO members must meet such political requirements as settlement of international and ethnic disputes, and democratic control of the military—criteria Greece and Turkey did not meet long after joining the organization in 1952. Neither was Portugal, one of the original NATO members, democratic until the mid-1970s. Despite the stricter conditions for entry, new members of NATO from East and Central Europe “feel stronger loyalty toward the United States than do their Western counterparts” because they are “grateful to Washington for pushing NATO expansion to the East” (Pehe 2003, 33). Furthermore, these alliances are not as costly or risky for the United States as they were when the Soviet Union posed a real threat to Western security, and American allies are working hard to keep the superpower satisfied. The United States simply reduced the cost of its alliances without dissolving them. U.S. military expenditure has increased in the unipolar period, especially under President George W. Bush, but the increased expenditure hardly means an increased commitment to its allies, unlike during the Cold War. Figure 4.5 shows that the percentage of American troops stationed on foreign soil significantly declined after the Cold War. Although the percentage surged in 2003 as a result of the Iraq War, the troops in Iraq do not represent American commitment to U.S. allies, unlike troops elsewhere. The number of U.S. troops in Europe was reduced by two-thirds after the Cold War, although U.S. forces in East Asia did not experience such a

source: Kane (2006).



5.0

10.0

15.0

20.0

25.0

30.0

35.0

Figure 4.5.  Percentage of American troops stationed in foreign countries (1950–2005).

1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005



alliance politics and polarity 91

drastic downsizing. Not coincidentally, the host nation support of Japan and South Korea to U.S. forces is substantially above that of NATO countries (Wolf and Zanini 1998; Calder 2006; Yoda 2006). One could argue that U.S. forces are kept in East Asia because the threat level did not diminish there as much as in Europe, but that is at best a partial explanation. In terms of burden sharing, for example, the Japanese government covered between 75 and 79 percent of total local U.S. stationing costs in the period between 1995 and 2002 (Calder 2007, 192–193). South Korea has rivaled Germany in its host nation support even though both its national economy and the number of American troops stationed there have been smaller than those of Germany. Even then, the number of U.S. troops in Korea declined as a result of the global transformation of U.S. military deployment and the Iraq War. Incidentally, a continued troop presence does not necessarily mean a continued level of commitment or concessions on the part of the United States. Sometimes, the United States stations its troops on a reluctant host country. For instance, although the Japanese public are generally satisfied with the U.S.–Japan alliance, they are not eager to have American troops on their soil. In fact, the Japanese government even agreed to pay more than US$6 billion to transfer 8,000 U.S. Marines from Okinawa to Guam.20 Skeptics might also point out the fact that 12 Central and Eastern European states have joined NATO after the end of the Cold War. In addition to the stricter standards for entering the alliance, I offer two different explanations for the expansion. The first explanation, which is more directly related to my argument, is that the expansion of NATO was meant to prevent the rise of Europe as the second pole in international security affairs. Successive U.S. administrations in the post–Cold War period have adopted a NATO-first policy to discourage independent European security cooperation ( Jones 2007, 235–238). This is consistent with my systemic argument because the unipole should try to prevent the emergence of a competitor in the alliance market. The United States wants Europe to be both dependent and independent because Americans want Europeans to remain steady customers for its alliance goods. Europeans need to be in a sufficiently good condition to make payments to the United States, but they should not be self-sufficient, let alone competitive with the unipole.21 Meanwhile, the old members of NATO had to agree to NATO expansion to avoid the expected decline of NATO after the Cold War, that is, to avoid abandonment by the United States. Whereas the United States as a unipolar power has an incentive to deter the emergence of new poles, the United States as one of the bipolar ­powers

92

alliance politics and polarity

did not discourage Europe’s entry to the alliance market as a supplier: “By 1946, some officials in the State Department began urging a policy that encouraged European unity, building Europe into a third force alongside the United States and Soviet Union. The idea was to foster a multipolar postwar system in which Europe would be a relatively independent and unified geopolitical power center, with Germany integrated within it” (Ikenberry 2011, 203). Considering the disadvantages of bipolarity for great powers (see Chapter Three), such a shift toward multipolarity could have increased U.S. bargaining power in the alliance market. The second explanation for the enlargement of NATO is that it was a result of several historical contingencies rather than military necessities; the United States spontaneously extended its alliance ties to the new NATO members even though it could have withheld such commitments. The single great power does not “have to” withhold its commitment in a unipolar system. The United States can use its advantage in whatever way it wishes, especially when opposition from old allies count less than before. The NATO expansion does not look so strange when the strategic luxury is combined with such factors as the idea of democratic peace, political incentives to cultivate voters of East European extraction, lobbying by the American arms industry, desire for continued primacy, and Article 10 of the North Atlantic Treaty, which keeps the organization open to other European states.22 Because NATO enlargement was something the United States wanted for particular reasons, and not out of military necessity, the case does not hurt my argument that the United States enjoys the advantages of the unipolar system.

More Alliances, More Exchanges, but Less Military Protection Because the supply of security is limited under unipolarity, and also because demands for military protection decreased in many parts of the world after the Cold War, we should observe reduced production of security in the alliance market. It is important to note that my theory does not necessarily predict a decrease of alliances in the face of reduced supply or reduced demand for security; states can continue to benefit from exchanges with their allies by altering what they exchange.23 What we should pay more attention to is the content of allies’ exchanges because reduced supply or reduced demand for security means only reduced production of security and not reduced exchanges. Many states have actively formed alliances since the end of the Cold War, but the types of alliances formed ref lect the decline in the supply of security



alliance politics and polarity 93

from major powers, especially the United States, as well as the decline in the demand for security. Leeds and Mattes (2007) point out that post–Cold War alliances involve fewer major powers and are connected to broader cooperative projects than those before them.24 The percentage of new alliances that include at least one major power declined from 49.4 percent (1945–1989) to 37.4 percent (1990–2003), even though the number of major powers (not great powers) in the COW dataset increased from five to seven in 1991 (Leeds and Mattes 2007, 187). Although 68.8 percent of alliance agreements formed between 1946 and 1989 include nonmilitary cooperation, 95 percent do so between 1990 and 2003. Furthermore, only 13.5 percent (22/163) of the alliances formed since 1990 require active military assistance (that is, defense or offense), whereas the number is 76.8 percent (129/168) for alliances formed between 1946 and 1989. Ref lecting its role as a supplier of military protection, the United States has a defense obligation in 17 of the 20 alliances it formed during the Cold War. In contrast, the United States has not created even a consultation pact since the end of the Cold War. Table 4.3 shows the geographic distribution of new alliance memberships, and Europe is by far the most active stage of alliance formation after the Cold War (72.3 percent of the world total). Europe has been traditionally an active stage of alliance formation (57.9 percent for 1815–1989 period), but Europe’s increased share is counterintuitive at first glance because it is a relatively peaceful region and the main beneficiary of the end of the Cold War. When we look at the new alliance memberships with defense or offense obligations, however, Europe’s share has actually declined from 58.1 percent to 40.3 percent in the post–Cold War era. In contrast to other table 4.3 New alliance memberships.

1815–1989 1990–2003

Memberships in general

Memberships with defense or offense Memberships obligations in general

Memberships with defense or offense obligations

Europe Middle East Africa Asia Americas Total

482 (58.1%) 90 (10.8%) 70 (8.4%) 39 (4.7%) 149 (18.0%) 830 (100%)

48 (40.3%) 10 (8.4%) 31 (26.1%) 20 (16.8%) 10 (8.4%) 119 (100%)

686 (57.9%) 162 (13.7%) 90 (7.6%) 77 (6.5%) 170 (14.4%) 1185 (100%)

334 (72.3%) 22 (4.8%) 46 (10.0%) 49 (10.6%) 11 (2.4%) 462 (100%)

note: One new membership = one state joining one alliance, and the number includes new members joining existing alliances.

94

alliance politics and polarity

regions, the vast majority of new alliances in Europe do not require defense or offense obligations. In the post–Cold War era, the shares of militarized alliances for Asia and Africa have significantly increased. In sum, as the exchange model suggests, the decline in the demand and supply of security in many parts of the world changed not only who forms alliances but also the goods exchanged within such alliances. It is difficult to tell how much of the change is attributable to the advent of unipolarity and changes in supply and not to the changes in military threats and demand for security. We do, however, know that supply of security from the United States has decreased and that other states have not been filling the gap despite the surge in the number of alliances formed after the Cold War. Only 13.5 percent (22/163) of the new alliances and 25.8 percent (119/462) of the new memberships that began between 1990 and 2003 require defense or offense obligations. This significant decrease in the proportion of alliance agreements with defense or offense obligations might tempt readers to think that predictions of the capability aggregation model and realist theories of alliances were right after all because militarized alliances did decline after the Cold War (Mearsheimer 1990; Layne 1993; Waltz 1993). Ignoring less militarized alliances, however, is not justifiable even when we focus only on military security. Alliance agreements without defense or offense obligation still have important military implications. For instance, consultation pacts such as the Baghdad Pact (the Central Treaty Organization) and the Soviet– Vietnamese treaty of 1978 were considered military alliances. Moreover, alliances without defense or offense obligations can evolve, and their military value can increase over time as in the cases of the Triple Entente before World War I and the Anti-Comintern Pact among fascist states before World War II. Waltz (1993) predicted that NATO would decline and disappear, and he later claimed that “the expectation” had “been borne out” in “a basic sense” because the alliance’s function changed (Waltz 2000, 19), but I disagree with his assessment. Adjustments in alliance exchanges are an integral part of alliance management, and an alliance does not die just because the partners change the goods exchanged. Waltz correctly pointed out the decline in the demand for security, but he did not seriously consider the possibility of alliances’ survival by adaptation. For analyzing the future of alliance politics, we need a theory that explains alliances’ survival and evolution, as the market theory of alliances does.



alliance politics and polarity 95

Reactions of the Other States If the end of the Cold War meant only reduced demand for security, the price of security should have decreased. As discussed in the following section, the price of security in the current unipolar system seems to be higher than during bipolarity, and this suggests that the decline in the supply of military protection was more significant than the decline in the demand. Hypothesis 3.5: Other things being equal, when the international system becomes unipolar, nongreat powers increase their commitment to existing and potential allies. As discussed in Chapter One, many alliances have endured beyond the end of the Cold War, and states in the unipolar era have actively formed alliances. Given that the end of the Cold War reduced demand for security in many regions of the world, one might wonder why nongreat powers are eager to form and maintain alliances. Nongreat powers formed new alliances or held on to the existing alliances, even for a higher price in the case of America’s alliances, because the shift from bipolarity to unipolarity meant a reduced supply of military protection from the remaining superpower. Because demand for security also declined in many regions, and also because nongreat powers cannot fill the role of security suppliers well, post–Cold War alliances are not as military oriented as their predecessors, as already noted. The attenuated forms of military alliances may appear to be a bad sign for the future of alliances, but the market theory of alliances suggests that states can obtain efficiency gains through alliances even without clear military threats. Post–Cold War alliances indeed have adjusted themselves to the new environment; although 29.1 percent of 174 alliances formed between 1945 and 1989 mention specific sources of threat, only 1.3 percent of 163 alliances formed between 1990 and 2003 do so (Leeds and Mattes 2007). Because nongreat powers are not good at producing military protection, they have tried to compensate for their military weakness by increasing alliance commitments to one another and, as will be explained soon, by more closely following the lead of the sole superpower.25 Although ties with the United States are the most important, many states have also tried to strengthen their other alliance ties. For instance, Posen (2006) argues that the European Union has been developing its own military capabilities not because it is threatened by the United States but because it needs to guard against the risk of abandonment by the United States and also wishes to have political inf luence. Many alliances not involving the United States

96

alliance politics and polarity table 4.4 Countries with the largest number of alliance agreements in 1989 and 2003.

Ranking

Number of alliance agreements in 1989 Ranking

  1. Russia 23   1. Russia   2. United States 12   2. France   3. France 10   3. Romania   4. Bulgaria 10   4. Bulgaria   4. Hungary 10   5. Hungary   6. East Germany   9   6. Poland  7. Czechoslovakia  8  7. Ukraine  8. Romania  7  8. Slovakia  8. Poland  7  9. Spain 10. Mozambique   5   9. Uzbekistan 10. Libya   5 ... 10. South Yemen   5 15. United States

Number of alliance agreements in 2003

36 31 24 20 19 18 16 13 12 12 ... 10

have formed because the sole superpower is reluctant to extend its commitment and has raised the price of its commitment. Table 4.4 lists countries that had the largest number of alliance agreements in 1989 and 2003. The United States fell from number 2 to number 15 in terms of the number of existing alliance agreements.26 Between 1990 and 2003, 96 countries joined at least one alliance agreement, but the United States did not join any over the same period. The void caused by the demise of the Eastern Bloc was largely filled by its former members, but data suggest that these states were not satisfied with the minor-power alliances. States from the former Eastern Bloc formed numerous alliances among themselves in the early 1990s, but they did so largely because they could not afford to buy the U.S. alliance commitments at the time. In about 1994, these states began to have the prospect of joining NATO, and they gradually stopped pursuing the cheaper but ineffective alternatives to U.S. military commitment. This trend is partially attributable to the presence of the large number of alliances created in the early 1990s, but these countries’ behaviors differ before and after they signed the membership agreement with NATO. Between 1994 and 2003, ten former communist states that later joined NATO also signed 35 other alliance memberships, and all but one membership—Romania’s consultation pact with Russia in July 2003—were signed before each state signed NATO’s accession protocols.27 Thus, their behavior illuminates the difficulty of gaining security from minor powers and the attractiveness of U.S. military protection. This leads to my next point.



alliance politics and polarity 97

Under unipolarity, U.S. allies seem willing to pay an increased price for their alliances with the United States. The trend is counterintuitive from the perspective of traditional alliance theories because military threats have declined in many regions of the world. If anything, decreased demand for military protection should lead to a decline in military alliances and less commitment from the allies importing security from the United States. This, presumably, was the reasoning behind pessimistic predictions about U.S. alliances in the wake of the Cold War (for example, Waltz 1993, 75–76); however, the shift to unipolarity explains the failure of these predictions.28 Considering the marginalized status of military alliances in U.S. foreign policy, the predictions were partially correct. However, America’s allies (and for that matter, most states in the international system) seem desperate to retain or obtain their alliances with the United States, and the bargaining position of the United States looks better than ever. For countries allied to the unipole, serious efforts are required to avoid abandonment or entrapment by the much stronger partner. Europeans, for instance, opened NATO to new members and reinforced the alliance while simultaneously developing the European Security and Defense Policy (ESDP).29 The price of security is determined not only by demand for it—which is created by military threat—but also by its supply. Because the United States no longer has to compete against the Soviet Union in the alliance market, other states have to live with the increased price of America’s commitments. It may seem bizarre to treat the collapse of the Soviet Union as a cause of reduced supply of security for U.S. allies, but U.S. allies lost the potential for courting a rival of the United States, and U.S. military forces abroad were indeed reduced. Even when one likes shopping at Store A, having Store B in the market helps keep Store A from raising its prices. Under bipolarity, for example, the initially reluctant Washington was compelled to provide protection to Kuwaiti tankers during the Iran–Iraq War because the Kuwaiti government approached Moscow for protection (Caron 1990, 155–156; Hansen 2000, 125). For many states, the end of the Cold War shifted the situation from cell I to cell IV, rather than cell I to cell II, in Table 4.5. To illustrate the effects of the shift in polarity, I describe in the following sections how much the structural shift has pushed American allies in favor of the superpower. The shift is most visible in regions where military threats remain serious, not in absolute terms but relative to the bipolar era. For instance, because of its relative strategic safety during the Cold War,

98

alliance politics and polarity table 4.5 The end of the Cold War, polarity, and threat.



Threat high

Threat low

Bipolarity I II (two superpowers willing to supply security) Cold War Unipolarity IV III (the sole superpower reluctant to supply security) Post–Cold War

and also because of the rise of China and North Korea’s provocative actions after the Cold War, Japan’s needs for military protection did not decline as significantly as in European or South Korean cases.30 Japan Japan has increased its contribution to the U.S.–Japan alliance after the Cold War. One could argue that it is a result of greater needs for Japanese contribution arising from developments such as the Gulf War, the Afghanistan War, the Iraq War, and the threat of North Korea and China. That argument, however, severely undervalues the level of military threats and needs for Japanese contribution during the Cold War, and it does not explain why Japan has been willing to contribute more despite the relatively peaceful international environment of the post–Cold War era. Significantly, the change in Japanese policy began well before North Korea’s nuclear tests and the heightening of tension between Japan and China. Although Japan has indeed been more seriously seeking U.S. military protection in the last several years, an explanation based solely on the demand side of the alliance market is partial at best. Japan is still an important ally for the United States and has a capacity to provide valuable alliance goods, but its importance in the alliance market has declined with the shift to unipolarity. In a report for a White House meeting held in May 2003, Japan was ranked the highest among U.S. allies in terms of factors such as economic power, geopolitical importance, science and technology, and population (Hisae 2005, 12). Commenting on the report, Secretary of Defense Donald Rumsfeld stated that, in Asia, Japan was the only country that could become a global partner for the United States. In the unipolar alliance market, however, the United States does not need to match any competing suppliers of military protection. The collapse of the Soviet Union cast doubt on the Japanese notion that the island nation was indispensable to the United States. For most of the



alliance politics and polarity 99

Cold War period, Japan had little fear of abandonment even though its military contribution to the U.S.–Japan alliance was never satisfactory to the United States. The so-called Yoshida Doctrine, under which Japan focused on economic development and relied on the United States for military security, was “a tactically brilliant way to take advantage of the bipolar era,” but “it was immediately outmoded when the bipolar conf lict ended” because the “United States was no longer willing to provide automatic guarantees of Japanese security and demanded greater reciprocity in the alliance” (Pyle 2010, 6–7). The shift in bargaining power is ref lected in Japan’s economic contribution. The so-called omoiyari-yosan (sympathy budget or compassion budget), a subset of host nation support (HNS) and sometimes confused with the entire HNS, was originally introduced in 1978 to compensate for the increasing financial burden for the United States resulting from the appreciating yen.31 Shin Kanemaru, the head of the Defense Agency at the time, used the term omoiyari because he could not justify the budget in terms of the legal obligations under the Status of Forces Agreement. Omoiyari-yosan was introduced to alleviate the financial difficulty of U.S. forces in Japan, but the amount vastly increased in the 1990s, ironically a period of economic boom for the United States. The initial budget was a modest 6.2 billion yen in 1978, but items covered by the budget gradually increased, and the support reached its peak of 275.6 billion yen in 1999.32 Japan’s economic contribution is not limited to host nation support or bilateral cooperation. Japan’s financial contribution to the Persian Gulf War (1990–1991) was substantial (US$13.5 billion), although it was criticized as “too little, too late.” For the reconstruction of Iraq after the Iraq War, by May 2009, Japan had provided or pledged US$1.5 billion in grants, US$3.5 billion in concessionary loans, and US$6 billion in debt relief, making the country the second largest contributor.33 Japan has also contributed over US$4 billion to Afghanistan since 2001.34 In the less military context of financial international contribution, Japan was the world’s largest contributor of Official Development Assistance (ODA) from 1991 through 2000. In addition to the goods it has traditionally supplied, Japan has increased its military commitment to the United States after the Cold War (Nishihara and Tsuchiyama 1998; Sotooka, Honda, and Miura 2001; Akiyama 2002; Okamoto 2002). After it was criticized for “checkbook diplomacy” in the Gulf War, Japan dispatched mine sweepers to the Persian Gulf in 1991 (the Japan Self-Defense Forces’ first mission abroad) and its first UN Peace Keeping Operations personnel to Cambodia in 1992.35 The first North

100

alliance politics and polarity

Korean nuclear crisis of 1993 and 1994 revealed that Japan was not ready to support U.S. forces should anything happen on the Korean peninsula. Therefore, following the 1996 U.S.–Japan Joint Declaration on Security and the 1997 revision of the Defense Guidelines with the United States, the Japanese government painstakingly worked for reinforcement of the U.S.– Japan alliance through legislation, such as one government act that made it possible for the JSDF to provide logistic support to U.S. forces in the event of contingencies in the areas surrounding Japan. In February 2005, Japan departed from its previous stance by declaring that Japan and the United States had a “common strategic objective” in encouraging “the peaceful resolution of issues concerning the Taiwan Strait” (Tamamoto 2005, 57).36 As a U.S.–Japan Security Consultative Committee document notes, Japan and the United States have adapted “the alliance to the changing regional and global security environment” (emphasis added).37 The change is significant, given that the 1960 U.S.–Japan Security Treaty has a so-called Far East clause (Article 6) that limits the purpose of the American military presence in Japan. In the wake of the September 11 terrorist attacks in 2001, the Japanese Diet passed the Antiterrorism Special Measures Law, which permitted logistical support for U.S.-led military campaigns against terrorist groups. In addition to the Air Self-Defense Force and the Maritime Self-Defense Force, the Japanese government even sent ground troops to Iraq for the country’s reconstruction.38 The scope of the alliance continues to expand. On April 27, 2015, the U.S. and Japanese governments revealed new “Guidelines for U.S.–Japan Defense Cooperation.”39 The new guidelines eliminated geographic restrictions, and the alliance will now “respond to situations that will have an important inf luence on Japan’s peace and security.” Japan’s own defense buildup has also been a measure to reduce the risk of abandonment, with its attendant risk: “As much, if not all, of Japan’s force modernization is taking place within the constraints of the U.S. alliance— for instance, the joint development of ballistic missile defence systems—this may significantly exacerbate Japan’s security dependency on the U.S. and thus limit Tokyo’s future ability to hedge against entrapment within the alliance” (Goh 2011, 20). In sum, Japan has been paying more for the ride on the American bandwagon, even though the military threats the alliance faces have been not nearly as significant as those it faced during the Cold War. This trend is now further reinforced by the demand-side factor, the increasing threat of China. China’s military expenditure has been rapidly increasing, and it is 3.2 times that of Japan in 2014 (SIPRI 2015, in constant



alliance politics and polarity 101

2011 U.S. dollars). Moreover, tensions between China and Japan have been rising for the last several years because of incidents over their territorial disputes in the East China Sea. Japan’s detention of a Chinese ship captain in September 2010 and the Japanese government’s purchase of three of the disputed Senkaku/Diaoyu Islands severely strained the bilateral relationship (Hagström 2012; Johnston 2013; Moore 2014; Takeuchi 2014). Although this book focuses on the neglected, supply side of the alliance market, it is important to acknowledge the recent increase in demands for US military protection in Asia. Europe European states have been also making efforts to court the unipole, although there is a considerable variation among them. In his examination of small European states’ military policies after the Cold War, Rickli (2008) notes that these states, including those that were neutral during the Cold War, shifted their defense policy toward cooperation with the United States, and they did so by adopting niche strategies, that is, by drawing on their comparative advantage. As Press-Barnathan (2006, 285) points out, such a division of labor for NATO members was “directly linked to an increase in effective burden sharing with the hegemon,” which “will make the alliance more worthwhile for the hegemon” under unipolarity. Moreover, in terms of foreign assistance, non-U.S. NATO members’ total contribution has been much larger than that of the United States (2.7 to 3.9 times larger in the period between 1995 and 2002).40 Despite the transatlantic disagreement over the invasion of Iraq and their approaches toward counterterrorism in general, NATO members have been generally cooperative to U.S. foreign policy, as shown in their contribution to the International Security Assistance Force in Afghanistan. In 2003, a year in which the trans-Atlantic relation was in a great turmoil because of the controversy over the U.S. invasion of Iraq, non-U.S. NATO members contributed a 5.6 times larger number of multinational peace operations personnel than the United States, with France and Germany each almost matching the U.S. contribution of 4,569 personnel.41 The Iraq War was exceptionally controversial, and opposition to it should not be interpreted as a decline of the allies’ commitment. In fact, it is surprising how many governments supported the Iraq War, given the strong opposition worldwide.42 Spanish Prime Minister Jose Maria Aznar, for instance, “thought of the Iraqi crisis as a unique opportunity to improve Spain’s relationship with the U.S.” because he shared “Bush’s

102

alliance politics and polarity

unipolar vision of the world” (Iglesias-Cavicchioli 2007, 3–4). Similarly, Tony Blair was criticized for “‘tumbling over himself ’ to get close to Bush” (Dyson 2007, 650). Many Central and Eastern European states supported the Iraq War, despite the substantive cost in their relationship with opponents of the war, possibly jeopardizing their prospect for joining the European Union (Fawn 2006). Fundamentally, European states all have a strong interest in good relations with the United States. Even the opponents of the war, such as France and Germany, had amended their relationships with the United States well before President Barack Obama took office with an emphasis on multilateralism. These European efforts to court the United States are more attributable to changes in the supply side of the alliance market, but the demand-side factors remain important as well. As a result of Russia’s intervention in Ukraine in 2014, European states now have even stronger incentives to seek better relationships with the United States. In particular, Central and Eastern European states are seeking more U.S. military support to counter the Russian threat. President Obama in June 2014 asked Congress for an additional $1 billion to boost U.S. military presence in Europe, but these allies are still anxious. Poland, for instance, announced its intention to increase its military budget and also suggested that the United States could open a major military base on its territory.43 In sum, European states will continue to court the United States in the alliance market for the foreseeable future. South Korea South Korea has been working hard to increase its inf luence over U.S. policy toward North Korea and to retain its alliance with the United States, that is, to avoid both entrapment/entanglement and abandonment by the United States (Choi and Park 2008). On one hand, nonstructural factors, that is, factors not related to the number of poles, led South Korea to fear entanglement in the post–Cold War era. On the other hand, the structural shift to unipolarity increased South Korea’s fear of abandonment by the United States. In addition, the rise of China has presented new challenges (and opportunities) for South Korea’s relationships with the United States. South Korea has the fear of entrapment by the United States vis-à-vis North Korea and China. After witnessing the difficulty of the German reunification, the South Korean government no longer wants a sudden collapse of the North Korean regime. North Korea’s nuclear weapons are



alliance politics and polarity 103

threatening to the South, but Seoul has been vulnerable to North Korea’s conventional force for a long time, and its priorities are different from those of the United States or Japan. South Korea now faces a much weaker enemy in the North, whose collapse could create serious national problems. Therefore, it is only natural that the United States and South Korea have frictions over their policy toward North Korea. South Korea also has an interest in avoiding entanglement in a rivalry between the United States and China. Seoul’s political relationships with Beijing have significantly improved since the early 1990s and, more recently, South Korea’s economic dependence on the rising Chinese market created an incentive to avoid antagonizing Beijing. China’s share in South Korea’s total export in 2013 was 26.1 percent, more than double the share for the second largest export market, the United States (11.1 percent).44 These factors, however, did not lead Seoul to take distance from Washington because South Koreans are also worried about abandonment by the United States. Despite North Korea’s decline and South Korea’s improved relationships with Russia and China, Seoul has been working hard to strengthen the U.S.–ROK alliance. Unlike Japan, South Korea’s strategic value to the United States has been traditionally rather limited, and South Korea obtained its alliance with the United States mostly because of the Cold War and the logic of alliance politics under bipolarity.45 Geopolitically, the “Korean Peninsula is not a major factor in the balance of power or in U.S. protection of shipping lanes,” and “the increased threat to Japan from U.S. military withdrawal from the peninsula and greater cooperation between Beijing and Seoul would be marginal” (Ross 1999, 112). South Korea’s strategic value to the United States may increase when China becomes another pole in the international system, but that is not yet materialized, as explained in Chapter Three. The South Korean government under unipolarity, therefore, has worked hard to please its ally by bearing more of the cost of the US troops there and by giving more f lexibility to U.S. Forces Korea. As Jae-Jeok Park points out, “South Korea closely cooperated with the U.S. in the latter’s efforts to transform its overseas military from specific threat-based to capabilities-­ based, even though domestic criticism was strong and the effort was contrary to South Korea’s short-term security interests” (Park 2011, 269). Even the leftist Roh Moo-hyun administration sent troops to Iraq, despite the public’s opposition and the inclination of the president. As an editorial of a conservative newspaper pointed out, the Iraq War “was a good war for Seoul’s relationship with the US” (Lee 2006, 492).46 Lee Myung-bak,

104

alliance politics and polarity

who became president in February 2008, was more conservative and pro-­ American than the previous two presidents and worked even harder to restore the alliance with the United States. This effort continues under the current Park Geun-hye administration, which is also considered to be proalliance, albeit with simultaneously good relations with Beijing. In the coming years, South Korea’s alliance policy will be significantly affected by South Koreans’ perception of China. Sino–South Korean relations have been generally good since the end of the Cold War, and their bilateral relations have improved even more under President Park. The Chinese President Xi Jinping received Park’s state visit to Beijing in June 2013 and had his own state visit to Seoul in July 2014—the first time that a Chinese top leader visited Seoul without having first visited Pyongyang. At the same time, however, South Koreans are increasingly wary of the challenges that the rise of China imposes on the U.S.–South Korea alliance (Kim 2014b). South Koreans might take significantly more than the current distance from the United States if they worry about entanglement in a Sino–American conf lict, but Seoul’s incentives to retain its alliance with Washington would become even stronger if China becomes threatening toward South Korea. Rivals of the United States Finally, countries that should be balancing against the United States (for example, Russia and China) have been taking action “that amounts to bandwagoning” (Wohlforth 1999, 37). Aside from the rhetoric, no country is seriously challenging the primacy of the United States through either internal or external balancing.47 China’s rapid military buildup is a natural consequence of its economic growth. It has spent only around 2 percent (from 1.6 to 2.2 percent) of its gross domestic product (GDP) on its military for the last two decades—a figure well below that of the United States (SIPRI 2015). Wohlforth (1999) argues that other states do not balance against the United States because it is too costly to do so. From the perspective of this study, other states side with the United States because the single superpower is the most attractive partner in the alliance market.48 Some argue that other states are tacitly balancing—soft balancing—the United States (Walt 2002; Paul 2004, 2005; Pape 2005), while others dispute the existence of soft balancing (Brooks and Wohlforth 2005; Lieber and Alexander 2005). The key to the debate, so far, has been whether other major powers view the United States as a potential adversary. My theory instead focuses on the value of the United States as an actual or potential



alliance politics and polarity 105

ally and explains the rationale for bandwagoning. A proponent of the softbalancing argument maintains that soft-balancing behavior occurs when “the dominant state is a major source of public goods in both the economic and security areas that cannot simply be replaced” (Paul 2005, 59). Because the United States is in a position to determine the f low of many valuable private goods as well, other states have incentives to please the unipole. Even the leaders of the states that are the targets of American enmity, such as those of North Korea and Iran, are interested in improved relations with the United States and are pursuing negotiations with the nemesis, to the extent that their domestic politics allow (for example, Oberdorfer 2001; Leverett 2006). Of course, other states do not accept every single policy of the United States, but we do not need to use the term balancing to explain why they try to “delay, frustrate, and undermine aggressive unilateral U.S. military policies” (Pape 2005, 10; emphasis added). Given the unpopularity of American unilateralism, it would have been more natural if the United States faced more resistance from other states. In the case of the Iraq War, behavior of the most participants of the coalition of the willing could be called “soft bandwagoning,” because most states simply paid lip service and did not seriously join the United States. Moreover, even if soft balancing is happening, other states are not distancing themselves from the United States; they are trying to constrain the United States within the existing alliances or with closer bilateral relationships. When “second-tier states . . . engage the hegemon and develop institutional links with it to ward off possible retaliatory actions” in their pursuit of soft balancing, their behavior conforms to my theory (Paul 2005, 59).

Decline of Restraining Arrangements under Unipolarity In addition to being reluctant to form new alliances and to support its old allies, the United States in the post–Cold War era became increasingly unilateralist, although costly wars in Afghanistan and Iraq as well as the 2007–2008 economic crisis probably had a moderating effect on the attitude. Hypothesis 3.6: Other things being equal, when the international system becomes unipolar, the single great power removes restraining arrangements with its allies. The United States has no enemy to fear in adversary games and no competitor to match in alliance games. Consequently, U.S. military policy under unipolarity has relied on ad hoc coalitions not bound by the concerns of its

106

alliance politics and polarity

traditional allies (Campbell 2004; Tertrais 2004; Menon 2007). The George W. Bush administration rejected the Comprehensive Nuclear Test Ban Treaty (CTBT) and the Kyoto Protocol, and it waged a preventive war against Iraq despite the worldwide protests and opposition of its allies. As Michael Mastanduno (2009, 143) points out, unipolarity also “provided strong incentives for the United States to pursue its economic agenda far more aggressively than it had in the past.” As military–strategic calculation no longer dictated the United States to court the support of its allies against the Soviet Union, President Bill Clinton’s diplomacy aggressively focused on economic issues, for example vis-à-vis Japan. Without the economic recovery of the United States in the 1990s and the rise of the Chinese economy, the economic frictions between the superpower and its allies might have been a much more serious problem. The subsequent economic boom in the United States reinforced the effects of military unipolarity. With the U.S. economy going well and the U.S. military strategy not requiring allies’ support, little incentive was left for the United States to behave itself in the alliance market. This trend has been moderated in recent years. The global economic crisis of 2007–2008 and the financial burdens of the wars in Afghanistan and Iraq should have increased the U.S. needs for economic concessions from its allies. Economic difficulties are as much a source for alliance cooperation as for economic friction, to the extent that the United States and its allies can successfully exchange what they have for what they want. Mutual dependence creates incentives for mutually restraining arrangements. One of the most important arrangements between allies is deployment of military force, and the alliance dynamics of the unipolar era have been affecting the redeployment of U.S. forces across the world. To be sure, this transformation was driven by technological innovations and new strategic needs to fight the War on Terror, but we should pay attention to its political aspect as well. The network of U.S. bases did not experience such a radical shift during the Cold War, despite many technological innovations and shifts in diplomatic climate.49 This was so because, as Kurt Campbell, a senior official in the Clinton and Obama administrations, pointed out, “Forward bases and presence are extraordinarily sensitive diplomatically.”50 A base agreement does more than simply enhance military capabilities: In addition to deterrence and commitment effects, it helps in the management of an alliance by giving the allies better control of each other. American bases made it easier for the United States to control the hosting countries, but they also restrained the United States by creating a vested interest.



alliance politics and polarity 107

Except for some “hubs” in countries such as Britain and Japan, large American bases in the territories of traditional allies of the United States were replaced with smaller facilities in locations such as Southern Europe, the Middle East, and Central Asia (Campbell and Ward 2003; Calder 2007).51 The United States will have less difficulty in giving up these small facilities than in relinquishing large military bases. As Kent Calder (2007, 231) notes, “Diversifying deployments to multiple redundant locations” reduces “the political risk of dependence on any one.” Similar to the American reliance on coalitions of the willing rather than on formal allied cooperation, the network of American military bases is becoming more f luid and less restraining for the United States. Although the troop redeployment became conspicuous during former Defense Secretary Donald Rumsfeld’s tenure, the trend had begun earlier with the end of the Cold War. The number of troops in Germany, by far the largest host of U.S. troops during the Cold War, has been constantly declining (from over 200,000 during the Cold War to 51,000 in 2011), and other old NATO allies except Italy have also experienced troop reductions (Kane 2006).52 There has been a similar trend in East Asia as well. In April 1990, the U.S. Congress introduced the first East Asian Strategic Initiative (EASI-I) and set down the gradual reduction of U.S. forces stationed in South Korea. Even after the announcement of the EASI-I and the first phase of the reduction, arguments for reduction did not cease, with certain inf luential advocates calling for a total withdrawal (Murata 1998a, 266). With suspicions over North Korea’s nuclear development deepening, the scheduled troop reduction was slowed down and eventually suspended in November 1991, but the structural transformation of the alliance continues. In 2004, the U.S. government announced a one-third reduction of the 37,500 troops in South Korea, although it later suspended the reduction at the force level of 28,500 in April 2008.53 On top of the troop reduction, U.S. forces in Korea have been redeployed from areas near the demilitarized zone to the southern region of the peninsula, an action that has reduced American commitments as well as the troops’ vulnerability to a North Korean attack. The same can be said about the proposed return of wartime operational command to South Korea. Ostensibly, the return symbolizes South Korea’s increased inf luence in the alliance, but it effectively means reduced American military commitment because the return means that the two countries’ militaries will fight under separate commands in wartime. Peacetime control of South Korean troops had transferred back to the country in 1994. The Pentagon originally proposed that the transfer be

108

alliance politics and polarity

made in 2009, but the two governments have been postponing the transfer at the request of the South Korean government.54 Although the changes discussed in the preceding pages represent partial abandonment by the United States, the removal of restraining arrangements also increases the risk of entanglement or entrapment by the United States. During the Cold War, the purpose of the U.S. military presence in Korea was simply deterring and defending against North Korea. With South Korea’s increased capability and the rise of China, the United States demanded and South Korea accepted the “strategic f lexibility” of the U.S. Forces Korea (USFK).55 Because American bases in Korea are close to Beijing, and because the USFK are no longer tied to the peninsula, the realignment of the USFK is worrisome to the Chinese and South Koreans. Ironically, but not surprisingly, much of the cost of the realignment is paid for by the South Korean government.56 Quantitatively, the American military presence in Japan seems to be intact for the moment, but the U.S. bases there are no less affected by the global transformation of the U.S. military. The Japanese side was optimistic about the realignment of the U.S. Forces Japan (USFJ), but they were surprised by the contents of the U.S. proposal when the two allies began specific negotiations in November 2003 (Hisae 2005, 60). For instance, the U.S. side proposed to relocate U.S. Army I Corps from Fort Lewis in Washington to Camp Zama in Kanagawa and the U.S. Fifth Air Force from Yokota Air Base to Guam, although these relocation plans were significantly modified over the course of the negotiations.57 In Japan, the issue of the USFJ is discussed in terms of reducing Okinawa’s burden and maintaining deterrence, but the United States is pursuing its policy for reasons of its own. The number of American troops in Japan has not changed as much as elsewhere, but their mission has become increasingly f lexible and less directly relevant to the defense of Japan. The Japanese government nevertheless is paying for the relocations of American troops and the construction of new facilities. For instance, Japan agreed to pay up to US$6.09 billion for the U.S. Marines’ move from Okinawa to Guam, covering about 60 percent of the estimated total cost of US$10.3 billion.58 This means financing by Japan of facilities and infrastructure for U.S. forces stationed in U.S. territories. Regarding redeployments within Japan, the decision to relocate U.S. jets from Atsugi to Iwakuni was not really an American concession, but Japan is expected to pay for it.59 In April 2006, Richard Lawless, then deputy undersecretary of Defense for Asia and



alliance politics and polarity 109

Pacific Affairs, estimated that the overall cost of the realignment for Japan would be about US$26 billion.60 In short, the United States is shedding its presence abroad much more casually than during the bipolar era, and it is increasingly free of the restraints that accompany its large bases abroad. Even in places such as Japan, where host nation support is generous, the U.S. deployment pattern is primarily accommodating the needs of the U.S. military, not those of the allies. The rise of China’s threat (see, for example, Mearsheimer 2010; Friedberg 2011; Ross 2013; Scobell and Harold 2013) will increase demands in Asia for U.S. military presence, but basing policy will naturally be determined by U.S. interest. Whether the United States sees China as the second pole (or tries to prevent China from becoming one) in the international system, along with how much host nation support the host states offer, will affect the U.S. willingness to accommodate the allies’ requests (and the supply side of the alliance market).

International System and the Future of Alliance Politics Chapters Three and Four have explained the importance of both military and nonmilitary goods in alliance politics and changes that are caused by shifts in systemic polarity. Although military alliances can have various goals, the common function of all military alliances is to be a contract for exchanging goods and services between states. There is a variety of goods exchanged between allies, and it is important to understand the demand and supply of those goods. The alliance market is far from perfectly competitive, and polarity (that is, the number of great powers) affects the patterns of transactions there. Some of the alliance literature has linked alliances and trade, but the deeper conceptual connection has yet to be realized, and the theoretical integration will cultivate a new ground for a general theory of alliances. As Glenn Snyder (1984, 1997) demonstrates, alliance politics is played at the crossroad of alliance games and adversary games, but the majority of the alliance literature to date has been based on the logic of adversary games. By paying attention to the exchanges between allies, we will have a better understanding of alliance games independent of adversary games. In Chapter Four, I have examined the effects of systemic polarity, especially those of unipolarity, to demonstrate the real-world implications of the alliance market theory at the systemic level. In the current unipolar world, the United States has advantages not only in adversary games but also in

110

alliance politics and polarity

alliance games by being an ever more attractive alliance partner. As frictions over the Iraq War showed, however, inattention to allies can create serious problems, and even the single superpower can suffer the consequences. All other states in the unipolar system are at a disadvantage in the alliance market, but they should not be too pessimistic. Nonmilitary goods are important in alliance politics, and the importers of military protection will be better off as military threats decline. If a systemwide decline of military threats takes place, the protection by the United States will not be particularly important except as an insurance against ambiguous security risks or as a protection against the unipole itself. Chapters Three and Four deal with the long-term systemic shifts in alliance politics, but the major themes are directly relevant to the U.S. alliance policy in the short term, too. My theory predicts that the United States will start forming new alliances when potential allies come to possess goods that are high in demand (for example, cooperation in the war on terror or global economic crisis) and short in supply, or when a serious competitive supplier of military protection emerges.61 The Obama administration’s rebalancing toward Asia, for example, can be considered to be a monopolist’s attempt to prevent a competitive supplier’s (that is, China) entry to the alliance market.62 If the international system returns to bipolarity, the United States and the other superpower (for example, China) will have incentives to extend and retain alliance commitments to lesser powers, and the bargaining power of minor states will increase, with other things being equal. Ross (1999) argued that the East Asian regional order is likely to remain stable because of the structure and geography of the system, based on the regional system being bipolar between China’s land power and U.S. maritime power. Ross’s argument mostly focuses on relationships between rivals, but my systemic model suggests that alliance politics under bipolarity will also be more stable. In terms of geography, South Korea, being on mainland Asia, is disadvantaged in maintaining its alliance with the United States (Ross 1999, 112). Under bipolarity, however, the United States will have stronger incentives to keep the alliance. My findings on alliance termination and alliance duration suggest that alliances under multipolarity tend to be more f luid and hyperreactive. Therefore, with other things being equal, multipolarity is probably not conducive to peace in East Asia. As Aaron Friedberg pointed out, “An Asia in which alignments were more f luid, more complex, and less certain might be more likely to see crises escalate into wars” (Friedberg 1993, 31). Under multipolarity, the bargaining power of Japan and South Korea relative to



alliance politics and polarity 111

the United States will be stronger than under unipolarity but weaker than under bipolarity. Multipolarity, especially one through Japan’s militarization, might be precipitated by the dissolution of America’s alliances in East Asia, but the reverse causation is also possible; that is, America’s alliances in East Asia will be more likely to dissolve if the international system becomes multipolar. The existing U.S. alliance relationships have been good in recent years because the unilateralist tendency of the U.S. foreign policy has been weakened under President Obama. The United States is still the sole superally but has needed the nonmilitary support of its allies to overcome the economic crisis. On one hand, with the unipolar distribution of military capabilities, the systemic incentives Presidents Clinton and George W. Bush faced still remain. On the other hand, however, the bargaining positions of America’s allies improved as the global financial crisis raised the demand for nonmilitary goods in the alliance market.63 Thus, dependence between the allies is more mutual than before, and the room for exchange is larger. The Obama administration has had better relationships with traditional U.S. allies not only because of learning from the previous administrations but also because of the systemwide increase in demand for economic concessions. In the next chapter, I extend my market theory of alliances to the domestic level by explaining how domestic politics affect military allies’ willingness to supply concessions in the alliance market. For systemic analysis, it is analytically more useful to see states as unitary actors, but “national” interests expressed in the alliance market in fact have important domestic elements. Those groups and individuals who inf luence alliance policy sometimes have sets of interest significantly different from those assumed by the systemic model to be the interests of the actors in the alliance market.

5

A Domestic Theory of Intra-Alliance Bargaining

Paradoxically, as Japan grows economically stronger her bargaining power in Washington is in some sense weakened, since she can less easily argue that without American assistance she faces economic collapse. Miriam Farley (1955, 142) It is an irony that President Roh, initially skeptical of the raison d’etre for the alliance ended his presidency after rendering a great contribution to the strengthening of the alliance. He accommodated difficult demands from the United States, such as troop reduction, redeployment and the concept of strategic flexibility for U.S. forces in Korea, dispatch of Korean troops to Iraq. He also initiated the arrangement for the transfer of the wartime operational control, a welcome development for the United States. Tong Kim1 As soon as Franco disappeared from the scene, on his way to God and the judgment of History, all the Governments of the new Spanish democracy were bent on achieving a rebalancing of the relationship with the leading Western power. Angel Viñas (2003, 24)

This chapter presents a domestic theory of intra-alliance bargaining. In addition to international factors, allied states’ specific interests in an issue and the domestic politics of the allies affect intra-alliance bargaining. Because a state’s interest varies issue by issue, we should look for a systematic pattern in the domestic politics of the allies that affects the allies’ willingness to concede in bargaining. I argue that the presence of effective domestic opposition to an alliance, a leader’s attitude toward the alliance, and the leader’s vulnerability, are the three particularly important factors.2 Effects of domestic opposition to cooperation have been widely studied—among others, in the two-level game literature (for example, Putnam 1988; Evans, Jacobson and Putnam 1993)—but this chapter increases the explanatory power of domestic variables by adding the role of leaders to the



a domestic theory of intra-alliance bargaining 113

picture. Beyond the personal affinity between leaders of alliance partners, certain characteristics of the leaders affect bargaining between allies. For instance, some would argue that a proalliance or procooperation leader is better positioned to obtain concessions from an alliance partner because the ally is more willing to concede to a cooperative partner. On the other hand, others would argue that a more neutralist or anticooperation leader is a better negotiator because he or she can more credibly take a tough stance in negotiations. Both mechanisms are at work in interstate negotiations, but I argue that the vulnerability of the leader determines which effect prevails. When a leader is likely to stay in power for a long time, the other state either has to accept concessions to the tougher demands of an anticooperation leader or does not need to prop up a proalliance leader. In contrast, when a leader’s rule is precarious, the other state either benefits from undermining an anticooperation leader by withholding concessions or needs to support a procooperation leader by making concessions. In the following passages, I present a domestic theory of intra-alliance bargaining by explaining how a leader’s attitude and vulnerability, as well as domestic opposition to an alliance, affect the bargaining power of a state vis-à-vis its ally. In Chapter Six, I conduct a “structured, focused comparison” of the U.S. alliances with Japan, South Korea, and Spain with respect to the three independent variables and intra-alliance bargaining (George 1979; George and Bennett 2004). The final section of Chapter Six discusses the implications of my findings as well as future research agendas.

International and Domestic Push An outcome of bargaining between allies, like that of any other international negotiations, is determined by both international and domestic factors. A mutually beneficial agreement can be reached when two or more states’ interests overlap; the ranges of outcomes acceptable to the negotiating parties become larger or smaller according to both the international environment and domestic politics (Putnam 1988). For instance, increased external military threats would raise the value of alliance cooperation and increase the range of acceptable deals for the threatened state. Alternatively, when one of the negotiating parties is approached by a third party for a new alliance, the bargaining position of the approached state improves as the state can substitute the existing alliance with a new alliance. When we look at the interests of subnational actors, a leader with a neutralist inclination or a leader who faces domestic­

114

a domestic theory of intra-alliance bargaining

opposition to an alliance has a smaller range of acceptable outcomes in intra-alliance negotiations. Moreover, when the alliance partner has certain interests in the domestic political situation of the state (for example, preserving an anticommunist regime), the range of acceptable deals for the alliance partner changes as domestic politics of the state shift. Capabilities or resources of states have effects on both international and domestic factors, and the effects are sometimes not straightforward to understand. When we look at the international equilibrium of the alliance market, a state’s increased capability should lead to an improved bargaining position for the state. When we look at the interaction between international bargaining and domestic politics, however, the effects of increased resources are not so clear. As Thomas Schelling (1960, 22) pointed out, qualities such as “more financial resources, more physical strength, more military potency . . . are by no means universal advantages in bargaining situations; they often have a contrary value.” Because lack of resources creates domestic constraints against concessions to an ally, it makes the state’s refusal to give in to the ally’s demands more credible and understandable. In some cases, therefore, a shift that appears to be favorable to a state at the international level can actually lead to a less favorable outcome to the state if the shift erodes the domestic sources of bargaining power at the negotiating table. Before we proceed, two points are worth mentioning. First, actors’ expectations for future interactions play an important part in the following analysis. I assume that the actors do not overly discount the future. This study focuses on alliances, but the argument should be applicable to many situations where actors bargain in the context of long-term mutual dependence. Second, the variables discussed are important only to the extent that they are visible to one’s ally, and what really matter are the perceptions of these variables rather than the reality. For instance, American policy makers were seriously concerned about the risk of Japan becoming neutral in the early years of the Cold War. That perception played a decisive role in the revision of the U.S.–Japan security treaty, although in hindsight Japan was unlikely to distance itself from the United States.

Domestic Opposition to Alliance Cooperation Domestic opposition to cooperation increases a state’s bargaining power vis-à-vis its ally because “the power of a negotiator often rests on a manifest inability to make concessions and meet demands” (Schelling 1960, 19). The



a domestic theory of intra-alliance bargaining 115

opposition raises the cost of alliance cooperation for a leader, thus limiting his or her room for concessions in intra-alliance bargaining.3 Domestic opposition makes concessions politically costly to both procooperation and anticooperation leaders, and even a powerful and secure leader needs to spend political capital to implement an unpopular policy.4 Regardless of the leader’s preference and vulnerability, cooperation is costlier when there is effective domestic opposition to cooperation.5 Opposition to alliance cooperation can come from various sources, such as the general public, political parties, and mass media. There is always some opposition in almost any case, but I focus on effective opposition visible to the ally—opposition that imposes clear costs on leaders and that can harm the future of the alliance unless appeased with some concessions. For opposition to translate into bargaining power, it must be visible to the negotiating partner. In fact, as long as the partner perceives a significant opposition to cooperation, the opposition does not have to be actually strong; for instance, Japanese conservatives benefited from the Americans’ exaggerated perception of Japanese domestic opposition to the U.S.–Japan alliance (Kataoka and Myers 1989, 20). The opposition from the Japanese leftist groups was real but not particularly strong, and the general public accepted the alliance. Because effective opposition to alliance cooperation is difficult under nondemocratic regimes, democratization often brings about a change in this variable. For instance, Ferdinand Marcos’s dictatorship had a large impact on the base negotiations between the United States and the Philippines: “The exigencies of martial law . . . reduced Marcos’s bargaining leverage over Washington,” and the “more open Philippine polity” after Marcos “has both complicated the negotiations and diminished U.S. bargaining leverage” (Stromseth 1989, 186).6 Domestic opposition can impose costs on nondemocratic leaders as well, but the costs are generally smaller. Opposition to an alliance or policy, however, does not always mean opposition to a government or vice versa. Opposition to an alliance is not necessarily stronger or more effective in a liberal political environment. People oppose an alliance or a concession to an ally for various reasons. Leftists may oppose an alliance with an “imperialist” power for ideological reasons or for fear of their country’s militarization. In contrast, rightists may oppose an alliance with a foreign state because they worry that the alliance weakens the sovereignty of the state and deprives it of independence and dignity. The mainstream population can also oppose cooperation for a variety of practical reasons such as the risk of military entanglement and

116

a domestic theory of intra-alliance bargaining

seemingly unfair military or economic burdens. When domestic opposition groups’ preferences are heterogeneous, the negotiation process becomes more complicated, and a concession from the other party does not please all the opposition groups (Putnam 1988, 443–444). For instance, the sale of a new weapons system to a state may please the country’s right-wing groups but repulse left-wing groups who oppose the militarization of the state. For my current purposes, I examine opposition in the aggregate, taking into account both opposition and support to a policy from various groups. Even when different groups in State A are opposing alliance cooperation for different reasons, State B can gauge the overall cost imposed on State A’s leader. Negotiating parties often discuss domestic obstacles to cooperation and form some—albeit not necessarily accurate—opinions about the level of domestic opposition to cooperation in the other state. I should also note that opposition groups to an alliance do not have to be veto players, that is, “individual or collective actors whose agreement is necessary for a change of the status quo” (Tsebelis 2002, 19). Opposition makes cooperation in an alliance politically costly and difficult but seldom impossible.7

Attitude of the Leader Leaders’ attitudes in an alliance vary from proalliance (procooperation) to neutralist (anticooperation). A proalliance leader is a leader whose utility for cooperation in an alliance is high, and a neutralist leader has low utility for intra-alliance cooperation. I use these terms because they have been popularly used to describe leaders in my cases, but the labels simply mean procooperation and anticooperation with respect to alliance policy. For the current purpose, the readers should not attach any historical meaning to the term neutralist; the concept is related but separate from neutralism during the Cold War.8 Leaders’ attitudes vary by issue area (for example, a leader who is procooperation in basing rights may be anticooperation in economic issues), but most leaders have (or are seen to have) a relatively consistent attitude about the overall utility of the alliance; hence, labels such as pro­ alliance and pro-American. The ally perceives a leader to be procooperation or anticooperation relative to alternative leaders in the state. Therefore, the larger the gap between the current leader’s attitude and that of the likely alternative, the starker the effects of this variable are. The unit of analysis for this variable is the top leader of a state. Individual leaders’ relationships with their alliance partners can widely vary, but some leaders’ political fates are closely tied to their alliance partners. Proalliance leaders of smaller states are often criticized as a puppet of the stronger alli-



a domestic theory of intra-alliance bargaining 117

ance partners. The difference of attitudes is sometimes expressed along the party line. During the Cold War, leftist parties in many countries questioned the utility of their countries’ alliances with the United States. Even in the post–Cold War era, party differences are important, as seen in the People’s Party of Spain’s support for the Iraq War and the Spanish Socialist Workers’ Party (PSOE)’s decision to withdraw the Spanish troops from Iraq. Personal differences can be large even within a political party. Japanese Prime Minister Nobusuke Kishi (1957–1960), for example, was considered to be much more proalliance than his two predecessors, Ichiro Hatoyama (1954–1956) and Tanzan Ishibashi (1956–1957), though all three belonged to the Liberal Democratic Party. South Korean Presidents Kim Dae-jung (1998–2003) and Roh Moo-hyun (2003–2008) had a similar base of political support, but Roh was considered to be neutralist whereas Kim was regarded as a clear advocate of the alliance with the United States. It is difficult to examine an alliance’s real utility to a leader, but my task is a more manageable one—to examine the perception of the ally. I will therefore examine the American perceptions of the leaders of Japan, South Korea, and Spain for this variable. Like leaders who face effective opposition against alliance cooperation, neutralist leaders genuinely have a higher cost for cooperation with an ally, and yet this does not automatically lead to a bargaining advantage for neutralist leaders. When allies fail to cooperate, the consequence (for example, stalemate, the deterioration of alliance relationships, or the dissolution of the alliance) looks less gloomy to neutralist leaders than to proalliance leaders. If alliance politics is a one-shot game and its players are fixed, the low cost of noncooperation simply raises the bargaining power of a neutralist leader. Allied states, however, interact with each other many times, and their leaders are replaced from time to time. When the long-term interests are considered, the ally has an incentive to resist tough demands from a neutralist leader. By withholding concessions and refraining from cooperation, the ally may lose some benefits from the alliance in the short term, but if it helps topple the neutralist leader, then the ally can recover those lost benefits in its future interactions with more proalliance leaders. The ally can sometimes inf luence the partner’s future policy by rewarding proalliance leaders and punishing neutralist leaders of the target state. Over the base relocation issues in Okinawa, for example, the United States did not make concessions to Prime Minister Yukio Hatoyama (2009–2010), who seemed neutralist. Subsequently, Hatoyama was replaced by Naoto Kan. With the lessons of Hatoyama’s fiasco over the Futenma

118

a domestic theory of intra-alliance bargaining

relocation plan, Kan quickly adopted a policy that conformed more to the U.S. government’s position, even though he had been previously considered to be relatively neutralist among the leaders of the Democratic Party of Japan (DPJ). DPJ’s manifesto in 2009 pledged to reexamine the role of U.S. military bases in Japan, but the clause was removed from the manifesto for the summer 2010 election for the House of Councillors.9 Thus, a neutralist (that is, anticooperation) leader is a tough but not necessarily good negotiator because the ally has incentives to avoid helping the inconvenient negotiation partner.10 In extreme cases, states actively undermine and violently replace a neutralist leader of their ally. The Chilean President Salvador Allende was killed in the 1973 coup d’état supported by the United States after economic sanctions and various measures were placed against him.11 The Afghan President Hafizullah Amin, whose loyalty to the Soviet Union came into question, was killed by Russian KGB agents in 1979, and the Soviet Union fought a war to support the proalliance Marxist People’s Democratic Party of Afghanistan.12 By the same token, states are more willing to help out proalliance leaders of their allies, even though proalliance leaders require fewer concessions to be satisfied. In sum, the attitude of a leader alone does not tell us much about the bargaining leverage of the leader.

Vulnerability of the Leader I define the vulnerability of a leader as the expected duration of the leader’s power and the sensitivity of the leader to concessions from the allied state.13 This variable can alternatively be called the “replaceability” of a leader. What matters to an alliance partner is how long a leader continues to control the country’s policy, given what the ally can do to inf luence the duration. As noted earlier, the leader’s perceived vulnerability rather than actual vulnerability matters; however, public information, such as results of opinion polls and electoral outcomes, helps to estimate the perceived vulnerability because the ally keeps modifying its perception through observation. Although the political system of a country significantly affects the vulnerability of a leader, this variable is not solely determined by the regime type. Democratic leaders often have institutional limits to their time in power, but there is a substantial variation among democracies. Presidents in democracies have fixed terms; for instance, South Korean presidents after democratization are in power for only one five-year term. Prime ministers in parliamentary democracies have various expected durations, depending



a domestic theory of intra-alliance bargaining 119

on the prospect of elections. Still, in other cases like Japan, prime ministers were replaced largely according to the intraparty dynamics of the dominant party.14 For the leaders of nondemocratic states, the variation is even larger. Some dictators such as Francisco Franco (1939–1975) and Kim Il-sung (1948–1994) stay in power for decades, but others disappear in a matter of weeks or days. For this variable, therefore, it is desirable to examine the perception of the alliance partner about the expected duration of a leader’s power rather than simply relying on the regime type. A leader’s expected duration of power is important, independent of his or her susceptibility to concessions by an ally. When the duration of power is expected to be short, an ally can refrain from cooperation with a neutralist leader without losing too much benefit from the alliance. In contrast, when a neutralist leader is expected to stay in power for a long time, an ally is more strongly compelled to deal with said leader because a stalemate is more costly to the ally than to the leader. When a proalliance leader’s time in office is expected to be long, the ally can wait for concessions by the more forthcoming leader, who is eager to achieve cooperation. In addition to considerations of a leader’s expected length of term, an alliance partner’s move in a negotiation is inf luenced by how much its concessions to the target state will affect the leader’s power. This argument, unlike the previous one, hinges on the idea that concessions will help keep leaders in power. For most issues, concessions by the ally should positively affect the leader’s tenure. Even when the domestic public opinion is critical of an alliance, concessions such as more foreign aid and less risk of military entanglement should lead to the increased popularity of the leader. It should be noted that vulnerability is partially defined by sensitivity to concessions but not by the amount of concessions—if the latter was the case, we would have a problem of endogeneity (that is, amount of concessions defines vulnerability, and vulnerability affects amount of concessions). In general, the ally should prefer to make concessions to a proalliance leader rather than to a neutralist leader. Such concessions can help a more convenient partner to stay in power and can also encourage conformity to the ally’s preference. However, a proalliance leader with a very limited expected duration of power may not be worth supporting because the ally is not likely to recover the investment and also risks appearing to be an unhelpful supporter. The ally needs a proalliance leader in office so that the leader can implement desirable policy. A deal made with a proalliance leader could bind the future governments to a certain extent, but there are many ways in which the future governments can avoid the implementation

120

a domestic theory of intra-alliance bargaining table 5.1 Interaction of the leader’s attitude and vulnerability.



Proalliance leader

Neutralist leader

Vulnerable Nonvulnerable

Bargaining advantage Bargaining disadvantage

Bargaining disadvantage Bargaining advantage

of the agreement. For instance, the 1996 U.S.–Japan agreement on the relocation of the Futenma base has yet to be implemented. The issue of compliance and implementation may give further advantages to vulnerable proalliance leaders, as opposed to nonvulnerable proalliance leaders.15 Some agreements take too much time for a leader to implement, and some policies can be easily reversed. When a procooperation leader adopts a policy and is then replaced by an anticooperation leader, the deal might not be fulfilled and the adopted policy might be reversed by the less cooperative successor. Therefore, the negotiation partner may offer more generous concessions to a vulnerable procooperation leader not only to help the cooperative leader but also to make cooperation more acceptable to the successors. The vulnerability of a leader, considered together with his or her attitude, has disordinal (crossover) interaction effects on the bargaining power of a state (see Table 5.1). When a leader is not vulnerable, being proalliance does not help the leader because the ally knows that the proalliance leader is more cooperative and does not need to be propped up. A nonvulnerable neutralist leader has a strong bargaining position because he or she is genuinely reluctant to cooperate with the ally, and yet the ally still has to keep dealing with the tough negotiator. In contrast, when a leader is vulnerable, a proalliance attitude gives him or her an advantage, because the ally has to maintain its support to avoid his or her replacement by a neutralist leader.16 In addition to losing a convenient partner, the ally will suffer reputational damage if a loyal friend in the target state falls from power.17 A vulnerable neutralist leader can be demanding, but the ally no longer has an incentive to concede to him or her because the ally can simply wait for a proalliance leader to emerge. The following hypotheses can be derived from the foregoing analysis: Hypothesis 5.1: Other things being equal, a procooperation attitude creates an intraalliance bargaining advantage for a vulnerable leader and an intra-alliance bargaining disadvantage for a nonvulnerable leader.



a domestic theory of intra-alliance bargaining 121

Hypothesis 5.2: Other things being equal, an anticooperation attitude creates an intra-alliance bargaining advantage for a nonvulnerable leader and an intra-alliance bargaining disadvantage for a vulnerable leader. Although the attitude and vulnerability of leaders have interaction effects, I assume that opposition to alliance cooperation has additive effects on a state’s bargaining power; that is, the effect of domestic opposition does not depend on the value of the other variables. Regardless of the leader’s own preference, I expect domestic opposition to increase the political cost of making concessions. Hypothesis 5.3: Other things being equal, a leader has an intra-alliance bargaining advantage when he or she faces strong domestic opposition to alliance cooperation. Table 2.1 in Chapter 2 summarizes the effects of the three explanatory variables on intra-alliance bargaining power. The analytical framework can be applied to both sides of a negotiation, but I hold one state’s domestic politics constant because analyzing the domestic politics of two sides at the same time significantly complicates the analysis. When a leader becomes too strongly opposed to an alliance, or when domestic opposition to an alliance becomes too strong, the continuation of the alliance becomes infeasible (for example, the U.S.–Iran relations after the 1979 Islamic revolution). My cases, the U.S. alliances with Japan, South Korea, and Spain did not experience such extreme situations, and the next chapter illustrates the variations in intra-alliance bargaining power rather than formation and dissolution of alliances.

Profit versus Inf luence My main focus in the market theory of alliances is how much a state is able to take relative to what it gives in an alliance, but Chapter Six also brief ly discusses shifts in the inf luence of allies. Inf luence of a state is determined by how much it is perceived to be able to give. Having more inf luence helps a state extract more benefit from its ally, but there is a trade-off between pursuits of inf luence and profit. Pursuing inf luence requires moderation in the pursuit of profit because those who extract too much profit at the expense of their ally lose inf luence in their alliances. Related to the earlier discussion in the section titled “International and Domestic Push,” changes in state resources have opposite effects on the state’s international inf luence and bargaining power based on domestic constraints.

122

a domestic theory of intra-alliance bargaining

For instance, when asked about shifts in intra-alliance bargaining power, Japanese experts on the U.S.–Japan alliance often mention two contradicting trends.18 On one hand, Japan’s political, military, and economic recovery from World War II led to a significant increase of its inf luence in the alliance. On the other hand, however, these same factors made it increasingly difficult for Japan to play “the strength of weakness” strategy (Schelling 1960, 22, 37, 52) in the alliance. Japanese people became too rich to demand economic concessions from the United States; in fact, they were not particularly troubled by their government’s economic concessions to the superpower. Because the Japanese leadership was, more or less, always proalliance, Japan’s bargaining power also decreased as the vulnerability of their leaders declined. As the rule of pro-U.S. conservative political groups became stable, Japan had to share an increasing burden of the alliance, and the profitability of the alliance declined accordingly for Japan. In Chapter Six, I test the hypotheses presented in this chapter and also discuss the overall trends in the U.S. alliances with Japan, South Korea, and Spain.

6

Case Studies of Domestic Politics and Alliances

In this chapter, I conduct a “structured, focused comparison” (George 1979; George and Bennett 2004) of the U.S. alliances with Japan, South Korea, and Spain to test the three hypotheses presented in Chapter Five.1 The final section of Chapter Six discusses the implications of my findings as well as future research agendas.

Case Selection There are significant variations in the three domestic explanatory variables across and within the three American alliances. The effectiveness of antialliance groups varied, as did the vulnerability of the leaders in these states. With regard to the attitudes of the leaders, none of the countries was ever ruled by an absolute neutralist or antialliance leader, but what matters is the leaders’ attitude relative to their alternatives, and there is a large variation in the perception of the American government to affect the bargaining power of the allies. These variations are partially attributable to the political institutions of the states; for instance, Japan achieved democratization as soon as the U.S.-led occupation ended whereas South Korea and Spain took decades to do so. These variables, however, were also affected by the terms of the alliances and by specific events, and they cannot be operationalized solely by the regime types. In the following case studies, I focus on the periods when the three domestic variables changed in the most significant ways. I conduct case studies on the domestic politics of the U.S. allies but not on those of the United States. Although U.S. domestic politics may also be of interest, analyzing the domestic politics of two sides at the same time significantly complicates the analysis. I chose to focus on the U.S. allies for the following reasons. First, although one might think that the domestic politics of stronger states should matter more, what I analyze here is the strength of weakness as per Schelling (Schelling 1960, 22, 37, 52). The weaker state can more credibly threaten noncooperation because it has less capacity to meet the demands of a negotiation partner. As the more powerful state, the United States also knew that its policy has a large impact on the domestic

124

case studies of domestic politics and alliances

politics of its allies, and Washington had more incentives to adjust its concessions. Second, during the lifetime of the alliances, domestic politics in the United States has been more stable than those of its allies in terms of the three variables discussed—largely because the alliances were more salient and contentious for the Japanese, Korean, and Spanish governments.

U.S.–Japan Alliance Japanese Domestic Opposition to the U.S.–Japan Alliance The U.S.–Japan alliance was born in 1951 amid the heightening tensions of the Cold War and the war on the Korean peninsula, but the attitude of the Japanese public toward the alliance was lukewarm at best. Unlike its neighbor South Korea, Japan was not directly facing clear military threats from the communist powers, and the formation of the alliance was a continuation of the status quo from the occupation era. Consequently, the formation of the alliance seemed to be a passive acceptance of the security ties with the United States rather than Japan’s own policy. Although many acknowledged the desirability of close ties with the United States for political, military, and economic reasons, others worried about getting entangled in America’s conf licts in Asia or getting deprived of autonomy under another form of occupation, with American troops stationed in Japan. Opposition to the alliance came mainly from the left wing of the political spectrum, although there was some opposition to the alliance from the ultranationalist groups as well. Most important, the leftist groups in postwar Japan consisted of pacifists who wanted to dissociate themselves from all things military. They were afraid that the military ties with the United States could push Japan back into the road of militarism both internationally and domestically. In addition, the leftists were ideologically critical of the United States and sympathetic to the Soviet Union and the People’s Republic of China.2 The U.S. government recognized the inf luence of the leftist groups on Japan’s security policy. A telegram from the embassy in Japan to the Department of State points out that “there is an instinctive yearning on part of most Japanese for world where they would not have to side with either American or Soviet giant but could sit it out on sidelines,” and the “latent neutralism is fed on anti-militarist sentiments, pacifism, fuzzy-mindedness, nuclear neuroses and Marxist bent of Japanese intellectuals and educators” (Foreign Relations of the United States [FRUS], 1958–1960, vol. 18, 380).3 According to a National Security Council Report, the Socialists “are domi-



case studies of domestic politics and alliances 125 40% 35% 30% 25% 20% 15% 10% 5%

19

5 19 2 5 19 3 5 19 5 5 19 8 6 19 0 6 19 3 6 19 7 6 19 9 7 19 2 7 19 6 7 19 9 8 19 0 8 19 3 8 19 6 9 19 0 9 19 3 9 20 6 0 20 0 0 20 3 0 20 5 0 20 9 1 20 2 14

0%

Figure 6.1.  Leftist parties’ share of seats in the Japanese House of Representatives (election results).

nated by extreme left-wing elements advocating a Communist-oriented neutralism,” and they are “presenting difficult problems for the United States, since these extremists are strongly anti-United States” (FRUS, 1958–1960, vol.18, 338).4 In addition, the Communist Party, “though numerically weak, exercises a significant inf luence over Japanese opinion particularly through its penetration of mass organizations, labor, education and the information media” (ibid). The opposition groups that advocated the abrogation of the U.S.–Japan alliance never became the mainstream in Japanese politics, but it was not clear at all in the early postwar years that they would fall into the status of a permanent opposition. In addition, even without taking power, “the inf luence of the Socialists on conservative policy” was considerable because conservatives were “impelled to say and do things considered popular” (FRUS, 1955–1957, vol. 23, part 1, 206).5 Figure 6.1 shows the leftist parties’ share of seats in Japan’s House of Representatives, which is the lower house of the Diet and the more powerful chamber of the bicameral legislature.6 When we look at the entire time span, the figure shows a trend of decline for the leftist parties, but the power of the leftist parties seemed to be rapidly growing in the late 1950s. The U.S. government was so concerned about the rising power of the leftist parties that it provided funds to the Liberal Democratic Party and conservative elements within the Socialist Party

126

case studies of domestic politics and alliances

through the Central Intelligence Agency (Schaller 1997, 135–136, 159, 165). According to historian Michael Schaller (1995), as much as US$10 million annually may have been spent for this operation between 1958 and 1960.7 Opposition to the alliance from the general public followed the trend in the Diet, in that it peaked around 1960 and gradually became marginal in the public discourse. In the early years of the Cold War, the Japanese were seriously concerned about getting entangled in America’s war in Asia. The memory of World War II was still vivid, and there was a hot war in the neighboring Korean peninsula. The Soviet Union tested its first nuclear weapon in 1949, and the Soviet launching of Sputnik in 1957 increased the sense of vulnerability among many Japanese, who feared that American bases could become the target of the Soviet nuclear weapons (for example, Kataoka and Myers 1989, 18; and Nishihara and Tsuchiyama 1998, 183). The first (1954–1955) and the second (1958) Taiwan Strait crises also increased the tension in East Asia and the Japanese fear of entanglement. The Japan Self-Defense Forces ( JSDF) were not sent abroad until after the end of the Cold War (minesweepers to the Persian Gulf in 1991 and UN Peacekeeping Operation in Cambodia in 1992), but, without the wisdom of hindsight, the Japanese public was concerned that the newborn JSDF might become militarily active.8 Another source of Japanese dissatisfaction in the 1950s was the terms of the 1951 security treaty with the United States. As I explain later, the terms of the treaty were unfavorable to Japan, and even proalliance conservative elites resented the inequity of the treaty. Base-related crimes and nuisances also troubled the general public, and their anger was further exacerbated by the special status of the American military personnel.9 For instance, in 1957, the Japanese public was outraged when a Japanese housewife picking up spent rif le cartridges was shot and killed by an American soldier, William Girard. After jurisdictional disputes between the Japanese authorities and the U.S. Army, Girard was eventually turned over to the Japanese court but was handed only a three-year suspended sentence and went back to the United States after the trial. In the field of economy, Japan in the 1950s was still in the recovery period, and its need for economic concessions from the United States was credible. A State Department document predicted that “prolonged economic distress would weaken moderate political forces, encourage extremist parties of both right and left, and probably lead eventually to an ultranationalist resurgence”; it also indicated a concern that in “the event of an imminent threat of general war, Japan might attempt to assume a neu-



case studies of domestic politics and alliances 127

tral position in an effort to avoid nuclear destruction” (FRUS, 1955–1957, vol. 23, part 1, 133).10 Many Japanese also complained about the restrictions imposed by the United States on their trade with China, which was a major market for prewar Japan. The Japanese gradually realized that the Chinese economy was not as important to them as before because of communist China’s economic policy and the different international economic environment. The mental adaptation, however, took some time, and the economic opportunity cost of the U.S.–Japan alliance seemed high in the early postwar years. Therefore, the United States opened its market to Japan for strategic reasons; Americans were so successful in making Japan dependent on the American market that trade friction became a major issue for the two countries by the end of the 1960s (Schaller 1997).11 Japanese top leaders exploited the domestic opposition to the alliance from the leftist groups and the general public to their advantage. Opposition to the alliance was used as bargaining leverage even before the U.S.-led occupation ended in 1952; Prime Minister Shigeru Yoshida secretly encouraged anti-American movements to obtain concessions from the United States during the peace treaty and alliance treaty negotiations (Igarashi 1986, 250–253). Yoshida in 1954 again refused American demands for faster rearmament of Japan by referring to Article 9 of the Japanese constitution and the threat of left-wing backlash (Schaller 1997, 279). As former Prime Minister Noboru Takeshita (1988–1989) explained, “The Liberal Democrats had used the possibility of criticisms by the Socialists to def lect demands by the United States to speed rearmament or play a more active role internationally” (Schaller 1997, 120). The leftist groups “provided a screen for the conservatives” and “Washington, hostage to fear of communism, concluded from the riots of 1960 that Japan had to be left alone with its domestic priorities” (Kataoka and Myers 1989, 20). Domestic opposition to the U.S.–Japan alliance became most visible around 1960, when the alliance treaty was substantially revised for the first and last time. In 1957, those satisfied with the security treaty were less than one-third of those who were dissatisfied.12 American policy makers were acutely aware of the situation. Secretary of State John Foster Dulles noted that the American posture in Japan and Okinawa could not continue safely: “If we try merely to sit on our treaty rights,” Dulles remarked, “we shall end by being blown out by popular sentiment, spearheaded by a Japanese government of hostile and neutralist, if not pro-Communist, sentiment.”13 The intra-alliance bargaining power of Japan increased because of the presence of effective opposition to the U.S.–Japan alliance. Despite

128

case studies of domestic politics and alliances

American concessions in the 1960 treaty revision, the renewal of the alliance caused the largest political turmoil in postwar Japan. Millions of Japanese engaged in protest against the revision of the treaty, and some of the radical groups took violent measures. When White House Press Secretary James Hagerty arrived in Japan on June 11, 1960, to prepare for President Eisenhower’s visit, his car was surrounded by the protesters, and a Marine helicopter had to retrieve him and Ambassador MacArthur after the Japanese police created room for it to land. On June 15, in a confrontation between the police and the student groups, a female student from the University of Tokyo was killed. Prime Minister Kishi used gangs and rightwing groups against the protesters and even attempted, unsuccessfully, the mobilization of the Self-Defense Forces, but he was not able to control the situation. Kishi had hoped to celebrate the ratification of the treaty with Eisenhower, but the American president’s trip to Japan was canceled in the last minute for security reasons.14 Understandably, both the U.S. and Japanese governments became very cautious in their management of the alliance in the 1960s, although they gradually realized that Japanese domestic opposition to the alliance was not a serious threat as long as the two countries did not radically challenge the status quo of their security relationships (Buckley 1992; Schaller 1997). The Vietnam War aroused the fear of entrapment in Japan, but Nixon’s disengagement policy alleviated the fear of entrapment and instead created a fear of abandonment among Japanese political leaders (Yoshida 2007, 2009; Izumikawa 2010, 145–146). By 1970, when the 1960 treaty was about to expire, the alliance had become much less controversial; in a poll conducted in February 1970, only 19 percent of Japanese respondents opposed the automatic extension of the security treaty, and 39 percent supported the extension.15 Over time, the U.S.–Japan alliance became something like “air” to the Japanese public (Sunohara 2007, 245).16 Whereas only 20 percent wanted to continue the U.S.–Japan alliance in the late 1960s (Schaller 1997, 194–195), 67.6 percent of respondents preferred to continue (and only 6 percent to discontinue) the alliance in a poll in 1978.17 Support for the U.S.–Japan alliance and the bilateral relationship in general has been consistently high since then. In polls by Yomiuri Shimbun in 2009 and 2010, 75 percent and 76 percent of the respondents answered that the U.S.–Japan alliance was conducive to the peace of the Asia Pacific region, whereas only 14 percent and 13 percent answered that the alliance was not conducive to peace.18 On goodwill toward the United States, Japan’s Cabinet Office poll in October



case studies of domestic politics and alliances 129

2012 recorded the highest score (84.5 percent having a friendly feeling) since the poll began in 1975.19 The Japanese public can be critical of specific issues, but they are generally favorable to alliance cooperation with the United States. Without some special event that galvanizes anti-­A mericanism in Japan (for example, base-related incidents), the Japanese public’s opposition to the alliance will continue to be negligible, and a neutralist policy is unlikely to be supported. Attitude of Japanese Leaders The most visible changes in this variable occurred in 1957, 1993, and 2009, when Nobusuke Kishi, Morihiro Hosokawa, and Yukio Hatoyama, respectively, became prime minister. Since the end of the U.S.-led occupation, the Japanese government has never been led by serious neutralists. In 1994, when Tomiichi Murayama, the head of the Socialist Party, became prime minister in the coalition government with the LDP and New Party Sakigake, even the Socialists abandoned their neutralist policy. There is nevertheless a variation in the degrees of Japanese leaders’ proalliance attitudes, and it had an impact on the bargaining power of Japan. Despite his wartime record, Kishi in the 1950s stood out as a proalliance leader even among conservative political elites. Kishi, “America’s favorite war criminal,” was a member of the Tojo cabinet and cosigned the declaration of war against the United States in 1941.20 After the Japanese surrender, he spent three years in prison as a class A war crime suspect. However, as the priority in the U.S. government’s policy toward Japan shifted in accordance with the rising tension of the Cold War, Kishi gained the trust of American policy makers as a strong anticommunist conservative politician. Kishi was one of the leaders who unified the conservative factions and created the LDP. He became the American choice for the prime minister as early as in the end of 1954, when another pro-American leader, Shigeru Yoshida, was losing political support. Despite American support, however, two other conservative leaders occupied the office of premier before Kishi took power in 1957.21 Ichiro Hatoyama, who succeeded Yoshida as prime minister in 1954, was “not regarded as ‘pro-American’” (FRUS, 1955–1957, vol. 23, part 1, 26).22 According to an assessment by the U.S. government, Hatoyama was “emotional, naïve in international affairs” and loved “public acclaim” (ibid., 25). The Hatoyama cabinet was expected to have “little inclination to increase political ties with the free world, particularly in efforts to combat Communism in Asia” and “Hatoyama’s advocacy of increased defense

130

case studies of domestic politics and alliances

efforts” would be only “for the purpose of hastening United States troop withdrawals” (ibid., 26). In the American view, the Hatoyama government “showed an increasingly independent and uncooperative attitude evident in a number of ways” including “expansion of trade with Communist China, expansion of social welfare measures, reduction of taxes and defense expenditures” (ibid.). Hatoyama was considered to have “consistently ignored US interest in handling almost all pending US–Japanese problems” (ibid., 32). In short, he was an inconvenient leader for the United States, who would “trade on Japan’s ‘indispensability’ to the United States to extract maximum concessions for the minimum cooperation” (ibid., 26). American policy makers hoped Ichiro Hatoyama would be succeeded by Kishi, but to their surprise Tanzan Ishibashi, another leader of the LDP, became prime minister in December 1956. The U.S. government “had put their money on Kishi” and hoped that Ishibashi would “not last too long” because they considered Ishibashi anti-American and a “headstrong . . . rabble rouser” (Schaller 1997, 124). Ishibashi had advocated Japan’s diplomatic normalization with communist states and was particularly critical of American restrictions on Japan’s trade with Communist China. Ishibashi resigned as the prime minister in two months due to a brain infarction, but he visited China in September 1959 despite then Prime Minister Kishi’s objection. Kishi’s proalliance attitude was contrasted with these relatively neutralist attitudes of his predecessors. Kishi was obviously proalliance compared with the leaders of opposition parties, who at the time still seemed to have some chance of taking power, and he was also proalliance relative to other leaders in the conservative groups. A telegram from the American Embassy in Tokyo to the State Department illustrates the American view of Kishi’s attitude toward the alliance: Kishi has taken calculated domestic political risk in staking future [of] his govt on central thesis of cooperation with free nations, notably US and Asian countries. He has come out openly against Communism and neutralism and in favor of free world cooperation. Moreover he has explained and defended this policy before [the] public as even Yoshida never did. (While Yoshida, Hatoyama, and Ishibashi govts all considered relations with US as the basic factor in Japan’s international policies, Japanese people as [a] whole wanted greater independence from America. This explains to [a] considerable extent popularity [of] Hatoyama’s position for restoring relations with Soviets and also [the] enthusiasm for Ishibashi who first pushed policy of expanding trade with Communist China.)23

Kishi was “by far [the] best leader in sight in terms of US objectives,” and if “he loses out, his successor [was] likely [to] be weaker or less cooperative,



case studies of domestic politics and alliances 131

or both” (FRUS, 1955–1957, vol. 23, part 1, 520). In the same telegram from Tokyo, Ambassador Douglas MacArthur II argued that it was “strongly in US interest to do whatever” appropriate to help Kishi and they “should be thinking about things” they “might do which would strengthen Kishi just as” they “have done with [West German Chancellor] Adenauer in last two elections” (ibid.). Another important Japanese leadership change is Morihiro Hosokawa’s premiership (1993–1994), which came out of the end of the 38 years of LDP rule. Although Hosokawa is one of the Japanese conservative elites, the circumstances of his ascendance to power cast a doubt in the American minds on his attitude toward the alliance. Because the LDP consistently supported the U.S.–Japan alliance, Hosokawa’s position was considered inevitably more neutralist, relative to his predecessors. Although the Hosokawa government declared that it would not abandon the foreign and economic policies of the LDP, and, although Green and Samuels (1994) are right in pointing out the continuity rather than changes from the LDP, there was uncertainty about the direction of the non-LDP government. American policy makers’ attention was mostly focused on trade talks rather than on security issues in the early 1990s, but American experts on Japan and the Department of Defense were concerned about the future of the Japanese government’s security policy. Cronin and Green (1994) warned that the U.S.–Japan alliance was drifting, and as evidence they pointed to the so-called Higuchi report written by the advisory panel created by Hosokawa. The U.S. side considered the draft of the Higuchi report to be inclined too much toward multilateral security cooperation at the expense of the U.S.–Japan alliance. After warnings from the U.S. government and the Japanese Ministry of Foreign Affairs, the final version of the Higuchi report emphasized close cooperation between the United States and Japan (Sotooka, Honda and Miura 2001, 491–498; Akiyama 2002, 44–56).24 Japan did not change the course of its security policy under Hosokawa, but Hosokawa in fact wanted to reconsider Japanese security policy in accordance with the end of the Cold War. He wanted either a visible reduction in defense expenditures or some other defense policy that symbolized the end of the Cold War (Tanaka 1997, 325; Akiyama 2002, 31–34). Four years after his premiership ended, he advocated the reduction of host nation support and the withdrawal of U.S. troops from Japan in the journal Foreign Affairs, although he did not argue that the alliance should be dissolved (Hosokawa 1998). Questioning the utility of American military presence in Japan is a taboo for mainstream Japanese political leaders.25 The

132

case studies of domestic politics and alliances

United States, therefore, would not have been unreasonable in viewing him as a neutralist leader, especially given that alternative Japanese leaders were more proalliance. In 2009, the Democractic Party of Japan (DPJ) won a landslide victory, and Yukio Hatoyama, the grandson of Ichiro Hatoyama, became the prime minister. David Arase points out that the power transition in Japan was “unsettling to the Pentagon, which is accustomed to deferential behavior from LDP governments,” which “owed the U.S. a great deal for helping” the LDP “lay the basis for its perennial rule”; in contrast, the “DPJ is friendly, but its leaders owe no special debt of gratitude to the U.S.” (Arase 2010, 51). With a PhD from Stanford University, Yukio Hatoyama was rather pro-American at a personal level, but two factors placed him as a neutralist leader in the eyes of American policy makers. First, because most LDP leaders have been considered pro-American, Hatoyama’s attitude inevitably looked relatively neutralist.26 On the one hand, Hatoyama appeared to be relatively anticooperation even though he wanted to look pro-American in the eyes of American policy makers. Hatoyama sought to establish a good relationship with President Obama, but his attempt ended in failure. Hatoyama’s “Trust me” statement to Obama and the subsequent failure to solve the Futenma relocation issue suggest that the Japanese prime minister’s one-way admiration made the problem worse. On the other hand, Hatoyama intentionally took a more neutralist policy to appeal to the Japanese public. As a leader of the DPJ, Hatoyama needed to distance himself from the traditional foreign policy stance of the LDP. As a departure from LDP policy, he stressed the need to have a more “equal relationship” with the United States, and he emphasized the importance of regional cooperation in Asia. His criticism of Japan’s dependence on the United States and his call for the East Asian Community did not help his image among Americans. Second, there were reasons for American policy makers to be critical of Hatoyama well before the DPJ victory. In an article published in 1996, for example, he advocated the reduction of the U.S. military presence in Japan.27 As a leader of the largest opposition party, Hatoyama had also criticized the Japanese government’s dependence on the United States. Hatoyama’s supposedly “anti-American” essay in the Japanese conservative magazine Voice was translated into English and caught up by American media outlets in this context.28 In American and Japanese mass media, Japanese conservatives and American government officials raised concerns about the foreign policy orientation of the new government. These media accounts emphasized



case studies of domestic politics and alliances 133

Hatoyama’s criticism of American-led globalization and his call for a closer relationship with Asia. Their views of Hatoyama were overblown but not baseless, given his past behavior.29 This book focuses on the top leaders of allied states, but attitudes of other domestic actors are important too. Although we need to avoid the trap of conspiracy theories, it is undeniable that the United States has significant inf luence over the attitudes of Japanese politicians, journalists, bureaucrats, and others. Those who share the American views of the alliance are rewarded with information and access to the U.S. government. The United States can also pressure the Japanese government to remove inconvenient actors; according to Yuriko Koike, who in 2007 became the minister of Defense in the first Abe cabinet, Morihiro Hosokawa replaced his Chief Cabinet Secretary Masayoshi Takemura in April 1994 because of American pressure.30 Because many Japanese politicians seek American support, they sometimes tie the hands of Japanese bureaucrats who engage in negotiations with the United States. Of course, bureaucrats themselves have incentives to cultivate good relationships with the United States. The “village of the US–Japan security treaty,” in this sense, is a product of skillful American diplomacy.31 Vulnerability of Japanese Leaders Throughout the history of the alliance, no Japanese leader has been invulnerable, although the LDP as a ruling party seemed nonvulnerable for a long time, holding office nonstop between 1955 and 1993. Since Japan regained national sovereignty in 1952, and until the current government of Shinzo Abe (2012–), there have been 49 cabinets and 29 prime ministers. American occupation transformed Japan to a democracy, and, although the LDP was dominant in much of the postwar period in Japan, intraparty competition was sufficiently active to make every leader vulnerable to political challenge. Factional fighting has been particularly important with regard to Japanese conservatives. For instance, a memorandum of the U.S. Department of State characterized Ichiro Hatoyama’s position as weak, suffering “from [internal] dissensions accentuated by the expected brevity of his political life” (FRUS, 1955–1957, vol. 23, part 1, 25).32 American policy makers were particularly sensitive to the vulnerability of Japanese leaders in the 1950s because the opposition parties still seemed to present a neutralist alternative to the proalliance policy of the conservatives. Economic concessions from the United States were important to the power of Japanese leaders in the early postwar era, when the Japanese economy

134

case studies of domestic politics and alliances

was still weak. Ties with the United States were important in noneconomic fields as well, and the support of the United States was important political capital. Even in recent years, Kenji Yamaoka and Katsuya Okada, members of the then-governing Democratic Party of Japan, have privately remarked that the party’s ability to remain in power is significantly affected by its relationship with the United States.33 Because the United States benefits most when its ally has a nonvulnerable proalliance leader, Nobusuke Kishi appeared to be an attractive partner for the U.S. government.34 The U.S. government saw in Kishi a strong leader who could unite the conservative groups and implement unpopular policy for the United States. Before Kishi’s visit to Washington in June 1957, Secretary of State John Foster Dulles reported to President Dwight Eisenhower that Kishi “gives every indication of being the strongest Government leader to emerge in postwar Japan,” and that “the time has come to take the initiative in proposing a readjustment of our relations with Japan” (FRUS, 1955–1957, vol. 23, part 1, 346–348).35 Kishi’s position, however, gradually weakened. Kishi was a competent bureaucrat and power broker, but he never had as much appeal to the general public as Ichiro Hatoyama did. Because of his association with Hideki Tojo and the Pacific War, many in the leftist groups and the press regarded Kishi as a reactionary who was trying to push Japan back into the road of militarism. In fact, so many Japanese opposed the 1960 treaty, despite changes favorable to Japan because it was advocated by Kishi. American policy makers knew Kishi’s lack of popularity but saw it as a necessary price for a responsible leader: Kishi’s position, according to Ambassador MacArthur, had no “popular appeal and therefore leaves him open to attack from those elements, not only Socialist, neutralists, and Communists, but also some conservative circles who desire looser association with US and closer ties with Asia including Communist China” (FRUS, 1955–1957, vol. 23, part 1, 519).36 Because of intraparty competition, many LDP politicians raised questions about the strategic issues of the new treaty. Although, or rather because, the revision of the security treaty was a major goal of the party, they tried to undermine the negotiation to prevent Kishi from consolidating his power within the party (Hara 1991; Schaller 1997; Sarantakes 1999). Because Kishi could not lose the support of his own party in the face of the leftist opposition to the treaty revision, his power within the LDP was fairly limited during the negotiation process of the new treaty.



case studies of domestic politics and alliances 135

The leftist opposition at the time still had a strong appeal to the Japanese public, and Kishi was a controversial political figure who needed tacit but firm support of the United States. In the process of ratification, some American senators objected to the unilateral concessions to Japan in the new treaty, but the volatile situation in Japan made it difficult for the United States to do anything that could further undermine Kishi’s position. Secretary of State Christian Herter’s following explanation to skeptics in the Senate illustrates the effect of a proalliance leader’s vulnerability: If we are not in a position to exchange ratifications fairly shortly, the pressures building up in Japan for a dissolution of the Diet to regularize the present abnormal political situation may force Prime Minister Kishi to resign and dissolve the Diet before the exchange takes place. In this event, the position of those in Japan who favor close ties with the United States, whom we believe to be at the present time a substantial majority of the Japanese people, would be considerably weakened during the election campaign and thereafter. If the Japanese conservatives are able to enter the election campaign without [sic] the treaty an accomplished fact, this should create conservative unity behind a policy of close cooperation with the United States and help maintain the dominance in Japan of conservative, pro-Western elements whose position has been shaken by recent events. On the other hand, if the United States does not act promptly this could be interpreted in Japan as a lack of confidence in the future of the U.S.–Japanese partnership. It could also lead to pressures in Japan for reviewing the treaty with the goal of altering the provisions of the treaty in a manner which will satisfy certain Communist-inspired demands.37

As a side note, it is interesting to note that some American leaders thought of using domestic opposition as a bargaining leverage in this episode. Albert Gore Sr. made the following remark to the Secretary of State: “I raised the question as to timing of ratification because I thought it might possibly be helpful to have a little show of critical questioning, if not possible opposition. If that would be helpful to play this little game out a little bit, I will be glad to cooperate. If it would be helpful for a show of unanimity I shall not raise a question further” (Sarantakes 1999).38 Because Kishi’s tactics in the Diet fueled the anger of the public, American policy makers became critical of Kishi in his last days in office, but the U.S. government clearly made many concessions to save Kishi, as will be discussed later. After Kishi was forced to resign, Hayato Ikeda, another proalliance leader, took power. Shortly before the transition, Ambassador MacArthur commented in a telegram that Ikeda was by far the best successor to Kishi from the viewpoint of U.S. interest because he believed staunchly in Japanese–American partnership and was “militantly anti-Communist” (FRUS, 1958–1960, vol. 18, 377).39

136

case studies of domestic politics and alliances

When the LDP fell from power more than three decades later, Japan was no longer vulnerable to socialist inf luence or neutralist sentiment, but its leader Morihiro Hosokawa’s position was shaky at best. Hosokawa’s vulnerability was not attributable to the public’s opposition to him or to the alliance but to the unstable political balance that put him in the office of prime minister. The Hosokawa government was a coalition of eight political parties whose only common goal was to defeat the LDP. Although the LDP lost the majority of the House of Representatives, it still had the plurality. The largest group in the coalition was the Socialist Party, which at the time had not abandoned its traditional security policy. Coordination within the coalition was inevitably difficult, and Hosokawa’s power was not expected to last long. Hosokawa resigned in nine months because of a financial scandal, but the short life of the Hosokawa government was somewhat inevitable given the discord within the coalition; Tsutomu Hata, who succeeded Hosokawa, was in the office for only 64 days, and then the LDP returned to power by forming a coalition with the Socialist Party and New Party Sakigake. With the historic power transition to the DPJ in 2009, Yukio Hatoyama became prime minister with high approval ratings (more than 70 percent), but he was vulnerable much like the LDP prime ministers after Junichiro Koizumi. Many voted for the DPJ in the 2009 general election not necessarily because they supported the party but because they wanted a change from the government of the LDP and Komeito (Clean Government Party). With respect to intraparty dynamics, Hatoyama’s group was neither large nor inf luential in the DPJ, although Hatoyama was a prominent politician. Hatoyama did not appear to be exercising leadership, and many considered Ichiro Ozawa to have the real power in the DPJ. According to Asahi Shimbun’s analysis of Wikileaks documents related to the issue, both DPJ politicians and Japanese bureaucrats suggested that Hatoyama’s soft personality created confusion and ambiguity over his policy stance. On top of a financial scandal over his political funds, Hatoyama’s poor handling of the Futenma relocation plan led to a rapid decline of his approval ratings. On January 4, 2010, the political risk consulting company Eurasia Group listed Japan as one of the world’s top ten risks for the year. Eurasia Group’s report noted that it was “quite possible that Hatoyama won’t last the year.”40 In February, The Economist also discussed the vulnerability of the Hatoyama government and predicted that Hatoyama would be gone before the summer was out.41 Yukio Hatoyama indeed did not survive the first half of the year and resigned on June 8, 2010.



case studies of domestic politics and alliances 137

Effects of Japanese Domestic Politics Let us examine the long-term trend before looking at the effects of the more short-term changes described in the preceding pages. In general, the profitability of the alliance has steadily declined for Japan. Because of increased state resources, it became more and more difficult for Japan to refuse demands from the United States. Lack of state resources does not necessarily make the public more critical of alliances per se, but it creates stronger opposition against concessions to allies. In the aftermath of the 1960 crisis, pressuring the proalliance government in Japan for concessions was not a serious option for the U.S. government. Combined with the logic of bipolarity during the Cold War, the historical lesson explains why Paul Giarra commented that “the culture of US–Japan alliance” used to be “Do not forcibly push or criticize Japan” (Sunohara 2007, 222). For military and strategic reasons, President Kennedy took a generous economic policy toward Japan to increase the country’s dependence on the United States. Although the Japanese economy was rapidly growing, President Johnson continued the policy because he needed the support of the Asian ally for the Vietnam War.42 There was a significant change in the intra-alliance bargaining power during the Nixon administration, and it was not just the consequence of the new international alignment precipitated by Nixon’s China policy. In the late 1960s, the Japanese public was critical of the alliance; in a poll, only 20 percent wanted to continue the alliance, which could be terminated after 1970 (Schaller 1997, 194–195). By the time Nixon took power, however, Japan’s bargaining power based on domestic politics had decreased considerably. The memory of the 1960 crisis faded away, and the support for the alliance increased after Nixon began withdrawing troops from Vietnam. On the American side, the burden of the Vietnam War caused a lack of state resources and created opposition against concessions to allies. It was time for American allies to start paying more if they wanted to keep good relations with the United States. The United States began using security policy as leverage to gain economic concessions from the allies (Schaller 1997; Stone 1999). As Shingo Yoshida (2007, 113) points out, Japan’s security policy changed significantly in about 1970 as a result of Japan’s economic great-power status and the relative decline of the United States. These two factors are international variables, but they decreased the profitability of the alliance to Japan by weakening Japanese domestic opposition to concessions and

138

case studies of domestic politics and alliances

by increasing American domestic opposition to concessions. To be sure, there was a causal process at the international level as well. Because of the Sino–American rapprochement and the apparent decline in the supply of American military protection, Japan feared abandonment by the United States. To cope with this problem, Japan increased self-help efforts in its national defense policy and military cooperation with the United States. In 1971, Japan accepted the cancellation of the return of the U.S. naval base in Yokosuka, and the base subsequently became the homeport for the aircraft carrier task force of the U.S. Seventh f leet (Yoshida 2007, 108–109). Japan has constantly been one of the largest foreign contributors to the U.S. military in the world (Calder 2007). Since 1978—not coincidentally after President Carter’s attempt to withdraw U.S. troops from South Korea—the Japanese government has been paying additional host nation support beyond what is required in the Status of Forces Agreement. Japan still benefits from the alliance, but its military and economic contribution, relative to the benefit, is much higher than it used to be; without effective opposition against concessions, the Japanese government can and does concede more to the United States. In addition to the long-term trend, the profitability of the alliance for Japan varied in accordance with the attitude and the vulnerability of the leaders and opposition to the alliance. Since its inception in 1951, the U.S.– Japan security treaty has had only one significant revision, and it is no coincidence that the change took place during Kishi’s premiership.43 Because the original 1951 treaty was negotiated under the U.S.-led occupation, the bargaining power of Japan at the time was limited, and many provisions of the treaty were unequal. The United States was granted the exclusive basing rights in Japan without being obligated to defend Japan. There was no institution for consultation between the two governments where Japan could voice its opinion. The United States was even allowed to intervene in domestic disturbances of Japan. Because the treaty had no specified lifetime or termination clause, the inequality that symbolized the loss of Japan’s sovereignty was to perpetuate as long as the United States liked. Prime Minister Shigeru Yoshida (1946–1947, 1948–1954), who negotiated the alliance treaty and the peace treaty, thought that the loss of sovereignty was a price Japan had to pay for its dependence on the United States, but the revision of the treaty became a major goal for Japanese political leaders. Prime Minister Ichiro Hatoyama sent his foreign minister Mamoru Shigemitsu to Washington in 1955 to propose the revision of the treaty, but the American side f latly rejected the request.



case studies of domestic politics and alliances 139

The U.S. government became willing to negotiate the revision only with the rising opposition to the alliance in Japan and the premiership of Kishi, a more proalliance leader who needed American support. Compared with the 1951 treaty, the United States made many concessions and received few in return in the 1960 treaty and the supplementary agreements. Among the major changes, the United States was now obligated to defend Japan; a consultation clause was added, and the U.S. freedom of using the military bases and introducing nuclear weapons to Japanese soil was limited; the domestic disturbance clause was deleted; the Administrative Agreement was revised to the Status of Forces Agreement (SOFA), which followed the more equal standard of the North Atlantic Treaty Organization; the duration of the treaty was set to be 10 years, after which either side could terminate the treaty with one year’s notice. For the United States, the only improvement from the previous treaty was that a sovereign Japanese government, no longer under occupation, voluntarily agreed to support the U.S. bases in Japan. As one senator complained, the treaty was “a one-way street” where the United States gave much and gained “practically nothing” that they did not already have.44 In contrast to Kishi’s case, the vulnerability of the Hosokawa government was not something the United States had to worry about; Americans had little incentive to help Hosokawa out given that he seemed more neutralist than other Japanese leaders. In Hosokawa’s case, the low bargaining power predicted by my theory is not so visible, and the effect of his premiership seems to be better seen in the changes after he left the office. This, I argue, is because of the noticeably weak bargaining position of Japan already present when Hosokawa succeeded the LDP government. Japanese prime ministers changed quite frequently, but the LDP as a proalliance group became nonvulnerable, as no serious opposition or alternative to the status quo was present in Japan for a long time. Therefore, the bargaining power of Japan did not change much, as the leadership transferred from the nonvulnerable proalliance group to the vulnerable and relatively neutralist Hosokawa government. No significant change took place in the alliance’s defense policy, although economic friction between the allies exacerbated. The Clinton administration took an aggressive stance toward Japan in their trade talks. Deputy Treasury Secretary Roger Altman declared that the United States would “wait until hell freezes over” for the Japanese to accept U.S. demands (Bhagwati 1994, 7). The U.S.–Japan relations in defense and economic fields did not show visible improvement until Prime Minister Ryutaro Hashimoto (1996–1998),

140

case studies of domestic politics and alliances

with whom the LDP regained the office of prime minister. The frictions over trade finally subsided, and the 1996 U.S.–Japan Joint Declaration on Security marked the end of the drift of the alliance. Admittedly, the improved U.S.–Japan relations were also attributable to international factors such as the changes in the structure of the international economy, the rising threat of China, and lessons from the first North Korean nuclear crisis, but domestic variables also played an important role. Hashimoto was clearly a proalliance leader, the LDP was more vulnerable than during the Cold War, and domestic opposition to the alliance had increased over the problems of American bases in Okinawa. In September 1995, the Japanese public’s long dormant anti-Americanism was galvanized by an incident in Okinawa, where a 12-year-old girl was raped by three U.S. Marines. The United States initially did not give custody of the men to the Japanese authority, and this led to criticism of the Status of Forces Agreement (SOFA) and of the presence of U.S. military bases in Okinawa. The incident forced the restructuring of American bases in Okinawa, which is still ongoing. The change from the LDP government (proalliance, nonvulnerable, little opposition) to the Hosokawa government (neutralist, vulnerable, little opposition) did not have much impact on the alliance because the Japanese side in both cases had weak bargaining power for different reasons. Hosokawa’s premiership, however, had important effects on Japan’s bargaining power in later periods, as it made the LDP look more proalliance and more vulnerable. Developments during the Hashimoto government also suggest the importance of domestic opposition to the bargaining power of a state. The LDP as a group became more vulnerable than during the Cold War, but one could argue that there was a relatively nonvulnerable proalliance leader. With 1980 days in office, Junichiro Koizumi is the third-longestserving prime minister (2001–2006) in postwar Japan and was a relatively nonvulnerable leader. The Koizumi cabinet recorded the highest approval rating in Japanese political history, and Koizumi enjoyed much popular support. At the same time, he was widely known as a pro-American and proalliance leader who established a close personal tie with President George W. Bush. As a relatively nonvulnerable proalliance leader, Koizumi accommodated many American requests. After the September 11 attacks, the two houses of the Japanese National Diet in total took only 10 hours to pass the Antiterrorism Special Measures Law. In November 2001, the Japanese government sent military ships to the Arabian Sea to support U.S. military operations in Afghanistan. Koizumi also sent ground troops to Iraq, and



case studies of domestic politics and alliances 141

Japan’s financial contribution to the reconstruction of Iraq is second only to that of the United States. In December 2003, the Koizumi cabinet also decided the introduction of the missile defense program in Japan. A senior U.S. official commented that it took only a year for Koizumi’s government to complete what the Bush administration had hoped to achieve in four years vis-à-vis Japan (Hisae 2005, 42). In contrast, the subsequent prime ministers from the LDP—Shinzo Abe’s first term, Yasuo Fukuda, and Taro Aso—stayed in power around only a year and did not have enough power to carry out significant proalliance policy for the United States. The U.S. government had high hopes for Abe, who had been known for his hawkish ideology and interest in expanding Japan’s military role, but his contribution to the United States was rather limited in his first term. The post-Koizumi LDP leaders made efforts to improve ties with Asia, but the United States was not as critical toward these LDP leaders as it was toward Yukio Hatoyama subsequently. Being relatively neutralist and vulnerable, Yukio Hatoyama had a difficult time with the United States. Michael Auslin warned that “few in the Obama administration believe relations will get better anytime soon, and certainly not while Mr. Hatoyama remains in office.”45 Andrew Oros (2010, 33) was correct in pointing out that Japanese voters “have criticized the DPJ for undermining U.S.–Japan relations,” but they would have had the opposite impression if Hatoyama obtained concessions from the United States in the Futenma relocation plan. Hatoyama lost “domestic support . . . because he appeared unrealistic, indecisive, and ineffectual in international negotiations” (Rosenbluth 2011, 44). George Packard (2010, 102) noted that it “was foolish for the Pentagon to try to bully Hatoyama just one month after he came to power into carrying out an agreement that the previous Japanese government had made with the Bush administration.” According to my theory, it was quite a rational policy to take against a vulnerable neutralist leader. Because of his problems with the Futenma relocation plan and a financial scandal, Hatoyama resigned on June 8, 2010. The U.S. strategy to withhold concessions from the neutralist leader worked even before Hatoyama’s resignation. Facing American pressure, Hatoyama stepped back from his earlier statement that the Futenma base should “at least” move out of the Okinawa prefecture. As Michael Green observed, the Hatoyama government modified “its rhetoric and policies on the alliance,” the “foreign and defense ministers quickly got behind the existing plans for realignment,” and the “renewed realism” was “ref lected in the pro-alliance make-up of

142

case studies of domestic politics and alliances table 6.1 Domestic variables and bargaining power of Japan.



Opposition Attitude Vulnerability Bargaining power

Ichiro Hatoyama/ Ishibashi Kishi

Hosokawa (non-LDP)

Moderate Strong Weak Weak Neutralist Proalliance Proalliance Neutralist High High Low High Weak Strong Weak Weak

Hashimoto Koizumi

Opposition Attitude Vulnerability Bargaining power

LDP after Kishi

Yukio Hatoyama (DPJ)

Kan (DPJ)

Moderate Weak Weak Weak Proalliance Proalliance Neutralist Moderate Moderate Low High High Moderate Weak Weak Weak

the government’s new defense advisory panel” (Green 2010a, 17–18). As the reason for giving up the relocation of the Futenma base out of Okinawa, Hatoyama in May 2010 emphasized the importance of the U.S. Marines for deterrence, but the policy reversal was a result of American pressure. In fact, in February 2011, he revealed that the deterrence argument was just an excuse, and he did not actually believe that the Marines in Okinawa were directly contributing to deterrence (for Japan’s national defense).46 Witnessing these developments, Naoto Kan, Hatoyama’s successor, emphasized the importance of the U.S.–Japan alliance and was accommodating toward the United States on the Futenma relocation plan. Unlike all his predecessors after the socialist Tomiichi Murayama, Naoto Kan never belonged to the LDP in his career, and his career in leftist civil movements makes him appear to be less pro-American than most politicians in Japan. As a potential successor to Hatoyama, Kan made conscious efforts to avoid the appearance of anti-Americanism. After assuming the premiership, Kan “set a positive tone for the US–Japan relationship by repeatedly referring to the US–Japan alliance as the axis or cornerstone of Japanese diplomacy” (Green 2010b). Table 6.1 summarizes the foregoing analysis of the U.S.– Japan alliance.

U.S.–South Korea Alliance South Korean Domestic Opposition to the US–ROK Alliance For a long time, there was no effective opposition to the alliance in South Korea. South Korea has been a pro-American and anti-Communist state



case studies of domestic politics and alliances 143

because of the memory of the Korean War and its dependence on American military protection. Moreover, any antialliance or neutralist element in the society was severely suppressed by the successive authoritarian regimes, under which anti-Americanism was denounced as being “an act of subversion” (Sook-Jong Lee 2005, 94).47 We can consider the South Korean political system to have been nondemocratic from the establishment of the republic in 1948 to 1987.48 Although elections were regularly held under the authoritarian governments, political freedom was limited, and dictators— both civilian and military—arbitrarily revised the constitution for political convenience. The two turning points for South Korean domestic opposition to the alliance are democratization in the late 1980s and the anti-Americanism of the 2000s. In the late 1980s, anti-Americanism became active along with the liberalization of the political process in South Korea, and South Korean domestic opposition gained political significance for the first time in the history of the alliance.49 In the early 1980s, resentment of the United States increased among the South Korean public, most importantly because they identified the United States with the military junta of their country. The assassination of President Park Chung-hee in 1979 raised the hope of democratization after 18 years of Park’s dictatorship, but military officers led by Chun Doo-hwan quickly took power through a coup d’état within the military. Protests against the military takeover were severely suppressed, and hundreds were killed in the city of Gwanju. Although the U.S. government did not support Chun’s actions and merely faced the fait accompli, the South Korean public held it responsible for the coup d’état and the Gwangju massacre. At the time of the coup, the U.S. government had the operational control of the two countries’ militaries in Korea, and the U.S. failure to stop Chun was considered to be an intentional act. Furthermore, President Ronald Reagan took a friendly approach toward Chun, giving him a warm White House reception and legitimating his rule.50 Criticism of the United States became radical among student groups, who even set fire to the U.S. Cultural Center in Gwangju in December 1980 and to the U.S. Cultural Center in Busan in 1982.51 As the democratization movement gathered broader and broader support among the public, anti-Americanism also became less of a taboo. After the transition to a democratically elected government in 1987, the liberalized political environment further opened up the possibility for opposition against the U.S.–South Korea alliance.

144

case studies of domestic politics and alliances

Another turning point for opposition to the alliance came in the early 2000s. The inter-Korean summit meeting between President Kim Dae-jung and North Korean leader Kim Jong-il in 2000 transformed South Korean public’s perception of North Korea, which had been gradually changing since the 1990s. The threat of North Korea in the minds of South Koreans def lated to the lowest level in history. Meanwhile, the U.S. policy toward North Korea became hard line as President George W. Bush departed from the Clinton administration’s engagement policy. South Korea and the United States had different priorities—the former pursuing national reconciliation and the latter nuclear nonproliferation—and the gap of their threat perception led to serious friction within the alliance. The already strained relationship was further exacerbated in 2002 by an accident in which two 14-year-old Korean girls were killed by a U.S. Army’s armored vehicle. As the two U.S. solders responsible for the accident were acquitted by the American military court, large anti-American demonstrations were organized across South Korea, and the opposition to the U.S. military’s presence in Korea gathered unprecedentedly wide support among the general public. In addition to these political developments, economic and demographic changes also strengthened the support base for anti-Americanism—a trend that continues today. Because of the rapid economic growth of South Korea, the United States has become less generous toward the ally in economic matters, and American pressures to open the South Korean market have strained the bilateral relationship.52 Meanwhile, the increased economic power led to higher self-confidence on the part of South Korean people, and their national pride fueled anti-Americanism. South Korea’s younger generations have a strong national pride, and they do not have the memory of the traumatic Korean War (1950–1953). Members of the so-called 386 generation, who led the democratization movement in the 1980s and have now acquired important positions in South Korean society, were born in the 1960s. The worldview of the younger generation is significantly different from that of the older generation, who have admired and relied on the United States for military and economic support since the time of the Korean War. It is only natural that domestic opposition to the U.S.–South Korea alliance increased with the generational change. Aside from the long-term trends, there are short-term f luctuations, of course. According to polls by Pew Research Center, the percentage of South Koreans having favorable view of the United States has significantly increased from the Bush era (52 percent in 2002 and 46 percent in 2003) to the Obama era (78 percent in 2009 and 79 percent in 2010).53 President



case studies of domestic politics and alliances 145

Obama is more popular than President Bush in many parts of the world, but anti-Americanism in South Korea was particularly strong in the early 2000s because of the gap between the two countries’ policies toward North Korea as well as the tragic accident that killed the two schoolgirls. Reopening of the Korean market to U.S. beef in 2008 was controversial because of the public’s fear of bovine spongiform encephalopathy (BSE, commonly known as mad cow disease), but the large protests were primarily against the South Korean government, rather than against the United States. Attitude of South Korean Leaders The most visible change in this variable came with the election of President Roh Moo-hyun (2003–2008). Roh Moo-hyun symbolized the changes in South Korean society in recent decades, and his policy ref lected the growing desire of the South Korean public for autonomy from the United States. Let us, however, first review the other leaders’ attitudes. Because South Korea has been heavily dependent on the United States to deter North Korea, South Korean leaders are even more proalliance than Japanese leaders. There has been virtually no South Korean political leader who opposed the alliance since its inception.54 There is, however, a meaningful variation even among the proalliance leaders in absolute terms because what matters in my theory is the attitude of a leader relative to others. Presidents Rhee Syng-man (1948–1960) and Park Chung-hee (1963–1979, in power from 1961, President Park Geun-hye’s father), were not neutralist in any traditional sense, but neither were they proalliance (procooperation) in my framework.55 Despite South Korea’s dependence on the United States, both dictators took a firm attitude toward the United States and were rather “neutralist” (anticooperation) relative to alternative leaders, given that South Korean leaders were all proalliance in the absolute sense.56 Rhee’s obstinacy, especially in the process of the armistice talks for the Korean War, tempted American policy makers to remove him forcefully (“Operation Everready”), although the plan was never implemented.57 Park Chung-hee, who took power by a military coup about a year after Rhee was ousted by protesters, was not a cooperative partner for the United States either. The discord between Park and U.S. President Jimmy Carter is well known because of the latter’s emphasis on human rights and his attempt to withdraw U.S. troops from Korea. In fact, Park never really had a good relationship with any U.S. administration. Park and Americans distrusted one another so much that U.S. intelligence in the 1960s installed listening devices in Park’s presidential office, and Park took countermeasures against

146

case studies of domestic politics and alliances

them (Oberdorfer 2001, 33). In reaction to the Nixon Doctrine, which in Korea led to the withdrawal of the 20,000 U.S. troops of the Seventh Division, Park began a secret nuclear weapons program and gave it up only after strong pressures from the United States (ibid., 68–74). Park also infuriated Washington by his attempt to bribe U.S. policy makers (ibid., 50–51). Compared with Rhee and Park, the successive South Korean leaders after them (Chun Doo-hwan, Roh Tae-woo, Kim Yong-Sam, and Kim Dae-jung) seem more proalliance and cooperative. President Kim Daejung (1998–2003) is the first South Korean top leader whose political basis was not in conservative and proalliance groups. Kim Dae-jung had support from the traditional opposition groups with leftist inclinations, who later became the support base for his successor, Roh Moo-hyun. Kim, however, was an experienced and pragmatic politician who also had a strong personal tie with the United States, which had repeatedly saved his life and harbored him during his years as a political dissident. Kim was therefore a proAmerican and proalliance leader, although his policy toward North Korea was significantly different from that of the U.S. government after President George W. Bush was sworn in.58 Roh Moo-hyun’s presidency (2003–2008) was a result of the generational change in South Korean society, and his electoral victory in December 2002 is often attributed to his pro–North Korean and anti-American attitude. With the support from the younger generation and the rising anti-­ Americanism after the accident in 2002 that killed two schoolgirls, Roh defeated Lee Hoi-chang, the proalliance and conservative candidate. In addition to continuing Kim Dae-jung’s conciliatory policy toward North Korea, Roh Moo-hyun pursued many policies that seemed dangerously neutralist in the eyes of proalliance groups. Roh’s request for renegotiation of the Status of Forces Agreement (SOFA) was not so controversial, but many criticized his policy to increase South Korean autonomy in the alliance (Bechtol 2005, 2006).59 Roh expressed his desire that the wartime operational control be returned to South Korea, a move that the South Korean military establishment feared would decrease the deterrence and war-fighting capability of the alliance. In a speech on March 22, 2005, Roh also presented a controversial strategic view in which South Korea plays the role of a “balancer” in Northeast Asia (Pastreich 2005; Kim 2005). In the new regional order he envisioned, Roh said, Korea would “calculate and cooperate if need be, and move forward with its proper authority and responsibility.”60 Given the history of the U.S.–South Korea alliance, his statement sounded shockingly neutralist, although it is unlikely that Roh



case studies of domestic politics and alliances 147

meant the traditional meaning of the term balancer. As will be discussed later, Roh’s foreign policy initiatives were mostly unsuccessful. Although Roh held the second inter-Korean summit meeting with Kim Jong-il in 2007, the meeting did not have as much impact as Kim Dae-jung’s meeting with the North Korean leader. In the presidential election in December 2007, Lee Myung-bak, a conservative and pro-American candidate, put an end to the decade of rule by leftist presidents. Lee was widely considered to be proalliance and took a tougher stance against North Korea. Lee received 48.6 percent of the vote, whereas the liberal candidate Chung Dong-young received only 26.3 percent, and another conservative candidate, Lee Hoi-chang, received 15.1 percent. Although the turnout was 62.9 percent and a record low, it is significant that Lee won the election with the largest margin since direct ­elections of South Korean presidents began in 1987, even with the presence of another conservative candidate.61 Vulnerability of South Korean Leaders Presidents of democratic South Korea still have strong power, but South Korean top leaders wielded enormous power before democratization in the late 1980s. Successive dictators (Rhee Syngman 1948–1960; Park Chung-hee, 1961–1979; Chun Doo-hwan, 1980–1987) eliminated political oppositions, and they all looked nonvulnerable until shortly before their rule ended.62 Therefore, the vulnerability of South Korean leaders increased as a result of democratization. On one hand, a president’s office is guaranteed for the term of five years. On the other hand, however, a president in democratic South Korea cannot be reelected and has only one term. Unlike in the authoritarian period, a democratic president’s power is constrained by the National Assembly, the unicameral legislature of the country. A president can quickly lose political power even within the five-year term, and American support is an important source of political capital. Although the most visible change in the vulnerability of South Korean leaders was brought about by the democratization of the country, and although there is a common tendency for democratic Korean presidents to lose power as their terms approach the end, Roh Moo-hyun’s presidency was unique in terms of its vulnerability as well. Roh inherited from Kim Dae-jung a National Assembly in which the president’s party was in the minority. Kim Dae-jung was a skilled politician and managed the opposition well, especially while his “Sunshine Policy” toward North Korea seemed to be working well and accepted by the United States and Japan.

148

case studies of domestic politics and alliances

Roh, on the other hand, was not so skilled in his dealing with the conservative opposition, and he was “notorious for . . . using frank but inappropriate words.”63 Roh also faced tougher scrutiny on his foreign policy as the euphoria of the 2000 inter-Korean summit subsided and the domestic and foreign audiences began to question the merits of the conciliatory policy toward North Korea. Roh’s relationship with the conservative opposition quickly deteriorated. In March 2004, the National Assembly voted to impeach Roh for minor electoral violations. This move backfired, and Roh’s low popularity temporarily rose. In the election in April 2004, Roh’s ruling party gained the majority of the National Assembly, and the Constitutional Court dismissed the impeachment in May 2004. Support for Roh, however, quickly decreased after the reinstatement, and his ruling party also suffered losses in the subsequent elections. Roh’s popularity became so low that he attempted to form a grand coalition with the conservative opposition, but this plan also failed. Roh’s approval ratings also suffered from the public’s dissatisfaction with his performance in economic policy, and his approval ratings dropped to 11 percent.64 Overall, filled with political controversies, Roh’s presidency was fragile, and his vulnerability was high. The political polarization that hurt Roh also limited the power of President Lee Myung-bak (2008–2013).65 As a conservative politician with an emphasis on pragmatic economic policy, Lee was expected to restore the strained relationships with the United States. However, large antigovernment demonstrations ensued because of the perception that he failed to negotiate hard over the safety issues of beef imports from the United States concerning BSE. Lee’s popularity was limited also because he did not meet the expectation of the public about his economic performance as he had campaigned on the perception that he as a “CEO president” would lift the South Korean economy. Lee’s approval ratings had ups and downs, but the numbers were relatively stable, compared with those for Roh Moo-hyun.66 Effects of South Korean Domestic Politics As in Japan’s case, the long-term trend is for South Korea to gain more inf luence and less profit in the alliance. With respect to inf luence, South Korea has gradually gained more equal position in the alliance. After the 1953 armistice, the UN Command, headed by an American general, had the operational control of the two countries’ militaries. In 1978, the Combined Forces Command (CFC) was created, and South Korean officers



case studies of domestic politics and alliances 149

began to participate in the command. South Korea regained the peacetime operational control of its military in 1994, and the process of returning the wartime operational control has been ongoing, albeit with delays. It should be noted, however, that the increased inf luence does not mean more profits for South Korea because South Korea incurred a larger burden as its responsibility broadened. The profitability of the alliance has been declining for South Korea in the long term. Working as a surrogate for domestic opposition, a lack of state resources gave South Korea a credible reason to demand concessions from the United States, and South Korea’s economic growth gradually reduced this kind of bargaining leverage. The South Korean economy has been much weaker than the Japanese economy, and South Korea’s per capita income surpassed that of North Korea only in the 1970s. The per capita income level of South Korea was roughly equal to that of Ghana in 1957 and lower than that of the Philippines in 1960 (Werlin 1991). Because South Korea was at the front line of the Cold War and the country genuinely needed aid from the United States, South Korea was able to gain considerable concessions from the United States: It received US$7 billion of military aid between 1945 and 1979 and US$5.5 billion of economic aid between 1945 and 1973 (Leonard 2006, 102). Economic and military aid from the United States played crucial roles in the country’s reconstruction after the Korean War and its defense against North Korea, but the amount of aid steadily declined as the South Korean economy grew. South Korea experienced a rapid economic growth based on an export-oriented development strategy, joined the Organization for Economic Cooperation and Development (OECD) in 1996 and was in 2014 the thirteenth largest economy in the world (according to the International Monetary Fund). As early as in the mid-1960s, the United States began to phase out its aid program to South Korea, although loans and other forms of economic concessions (for example, opening the U.S. market to South Korean products) continued.67 Starting in the 1970s, when the American commitment to East Asia seemed to be declining, South Korea began to make economic contributions to the alliance. South Korea first provided host nation support in kind for U.S. forces in Korea, such as the 1974 storage and maintenance of war stockpiles program (Calder 2007, 150). It also began official host nation support in 1979, although the amount of monetary contribution was not much until the 1990s.68 By the late 1990s, South Korea became one of the top providers of host nation support, rivaling Germany (Calder 2007, 195).

150

case studies of domestic politics and alliances

The profitability of the alliance has been declining for South Korea in terms of U.S. military contribution as well. The number of American troops in South Korea can be considered to be a concession from the United States, because U.S. forces in Korea have been almost exclusively for defense and deterrence against North Korea. Unlike in the Japanese or Spanish case, South Koreans wanted the troops to stay in the country. Every South Korean top leader except Roh Moo-hyun opposed the reduction of American troop presence, but the number of American soldiers stationed in Korea has been steadily declining as South Korea’s military capability improved and as the U.S. strategic priority shifted (Oberdorfer 2001; Levkowitz 2008). At the end of the Korean War in 1953, there were more than 320,000 American soldiers in Korea, but by April 2008, the number had been reduced to 28,500.69 Figure 6.2 shows that the troop level has a downward trend with some f luctuations. In addition to the number of troops, the United States also reduced its commitment by relocating its troops away from the demilitarized zone in the early 1970s and the early 2000s. Now, aside from the long-term trend, what impact did South Korean domestic politics have on the U.S.–South Korea alliance? Since the 1953 Mutual Defense Treaty has never been revised, we need to look at other arrangements between the allies. In addition to the shifts in economic and military contributions of the allies previously discussed, the revisions of the Status of Forces Agreement (SOFA) ref lect the changes of the two states’ bargaining power. The SOFA in South Korea was signed in 1966 after years of bitter negotiations. From the South Korean point of view, the agreement granted too many privileges to the U.S. military, including their use of land and facilities without constraints and America’s exclusive jurisdiction over its soldiers in South Korea. Despite South Korean public’s dissatisfaction, the agreement was not revised until after the democratization, ref lecting the absence of effective domestic opposition to the alliance in the authoritarian period. The SOFA was revised for the first time in 1991, and another revision took place in 2001. In these revisions, South Korea obtained concessions from the United States concerning issues such as the criminal and civil jurisdiction over American soldiers and the prevention of pollution in facilities granted to the United States. The increased domestic opposition to the alliance created bargaining leverage for the South Korean government, but President Roh Moo-hyun was not able to capitalize on the advantage. His failure to do so was ironic, given that Roh represented the neutralist sentiment of the South Korean public. As the accident that killed two girls in 2002 infuriated the South

Figure 6.2.  U.S. troops stationed in Korea (1955–2005).

55 57 59 61 63 65 67 69 71 73 75 77 79 55 83 85 87 89 91 93 95 97 99 01 03 05 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 20 20 20

source: Kane (2006).



10,000

20,000

30,000

40,000

50,000

60,000

70,000

80,000

152

case studies of domestic politics and alliances

Korean public at the implementation of the SOFA, Roh requested its further revision, but the United States was not responsive. In fact, despite Roh’s inclination, his presidency was characterized by South Korea’s concessions to the United States. The discord between Roh and President Bush was widely publicized,70 but its effect was fairly asymmetric in that the United States gained much and conceded little during the Roh administration. A former White House official of the Bush administration “hailed President Roh as having contributed more to the alliance than pro-American former president Chun Doo-hwan or Roh Tae-woo.”71 Roh sent South Korean troops to Iraq despite severe criticism from his constituents; in fact, those 3,600 Korean soldiers made the third largest contribution to the coalition force after the United States and the United Kingdom. Given the already strained relationship between the Bush administration and his, Roh had to make the decision, which he later said was a historical error. If South Korea did not send troops, Roh said in an interview, “Americans could have felt betrayed,” and “Korea’s commitment to the United States is just a reality Korea should accept.”72 Roh accommodated many other American demands, including troop reduction, redeployment, and the concept of strategic f lexibility for U.S. forces in Korea. Despite opposition from his own support base, Roh signed a free trade agreement with the United States, arguing that it would strengthen security ties as well as economic cooperation between the allies.73 The transfer of the wartime operational control of the South Korean military began under Roh’s presidency, and it was actually a welcome development for the United States. Unlike in the early years of the alliance, the United States no longer worries about South Korea’s adventurism against North Korea, and the transfer of wartime operational control would help reduce the burden of the United States. South Korea, on the other hand, will bear a larger cost in the alliance and will have a weakened American commitment to its defense. In sum, the neutralist president, despite his anti-American image, conceded much to the United States and gained little in return. When Lee Myung-bak became president in 2008, his most important foreign policy agenda was the restoration of the alliance. His proalliance policy, however, did not resonate well with the South Korean public. His soft stance on beef imports from the United States angered the South Korean public, who felt that the president prioritized his relationship with the United States over their safety. Given that Lee’s power was moderately limited and vulnerable, my theory predicts that Lee’s bargaining power



case studies of domestic politics and alliances 153 table 6.2 Domestic variables and bargaining power of South Korea. South Korea Rhee/Park Chun

Opposition Attitude Vulnerability Bargaining power

Democratic presidents

Roh Moo-hyun

Lee Myung-bak

Weak Moderate Moderate Strong Moderate Neutralist Proalliance Proalliance Neutralist Proalliance Low Low Moderate High Moderate Moderate Moderate Moderate Weak Moderate

vis-à-vis the United States would be stronger than that of Roh Moo-hyun, unless Lee’s power became extremely weak and did not warrant the political investment of the United States. Under Lee’s presidency, the U.S.–ROK relationships were exceptionally good. Lee enjoyed close personal relationships not only with Bush—somewhat expected given their political backgrounds—but also with Obama. At the G20 summit in Toronto in 2010, Obama stated that the U.S.–ROK alliance was the “linchpin” of the security for the two countries and the Pacific even though the term had been traditionally reserved for the U.S.– Japan alliance (Cha and Katz 2011, 55).74 Due to increased tension on the Korean peninsula, the allies had very close coordination over their North Korea policy under the Obama administration. In December 2009, one high-level official described the two countries’ North Korea policies as being “not just on the same page, but on the same paragraph” (quoted in Manyin et al. 2013, 9). Lee succeeded in postponing the transfer of wartime operational control, thereby alleviating the perception of abandonment by the United States. These developments were certainly affected by the allies’ shared perception of North Korean threats, but Lee’s proalliance credentials also helped. Americans knew that Lee, as a proalliance leader, could easily be “perceived as caving to U.S. demands,” for example, on the issue of beef imports (Cha and Katz 2011, 62). When Lee was perceived to be vulnerable, therefore, the U.S. policy makers probably had incentives to help him out. Table 6.2 summarizes the major points of the U.S.–South Korean case.

U.S.–Spain Alliance Spanish Domestic Opposition to the U.S.–Spain Alliance The U.S.–Spanish alliance was one of the most important policies of the Spanish dictator Francisco Franco (in power between 1939 and 1975), and opposition to the alliance was not viable as long as Franco was alive.75 The

154

case studies of domestic politics and alliances

Alliance Treaty Obligations and Provisions (ATOP) dataset codes the U.S.– Spain alliance from 1963, but the bilateral security relations began in 1953 with the so-called Pact of Madrid, which was essentially an exchange of economic and military assistance from the United States and military bases for American use. The base agreement gave Franco political legitimacy at home and abroad, as well as economic and military assistance. Although Franco’s Spain remained neutral during World War II, the postwar world’s public opinion was hostile to the dictator, whose regime was established with assistance from the Axis Powers. Therefore, in the 1953 agreement, Franco’s “greatest single gain consisted in the mere fact that the agreement had been signed” (Whitaker 1961, 43). As an analyst put it in 1953, Spain had gone “from United Nations outcast to United States partner.”76 Two major groups opposed the formation of the bilateral security relationships between Spain and the United States. On the right end of the political spectrum was the conservative Catholic group, who “protested against bartering the ‘Catholic conscience’ of Spain for ‘heretical dollars.’” Franco appeased this group by completing the negotiation of a concordat with the Vatican in August 1953, a month before the signing of the Pact of Madrid (Whitaker 1961, 41). The other group, leftist relative to the Franco government, had a much wider support base. This group opposed the alliance mostly because the American tie legitimized and strengthened Franco’s rule. The leftists were suppressed under Franco’s regime and did not present a serious challenge to the alliance. The U.S. government did not support the political opposition for the fear of jeopardizing their use of military bases in Spain, and anti-American sentiments grew among the Spanish public.77 After Franco’s death in November 1975, the liberalization of Spain’s political environment made it possible for the public to openly challenge the alliance, although advocates of democratization carefully proceeded with the transition to democracy.78 At the mass level, the opposition to the alliance came in the form of demands for more equal standing in the alliance, reduced American military presence, and opposition to joining the North Atlantic Treaty Organization.79 From the very beginning of the alliance, Franco had unsuccessfully attempted to join NATO (and later demanded treatment on equal terms with America’s NATO allies);80 ironically, now that Franco, the largest obstacle in the way of Spain’s entry, was gone, the Spanish public was not eager to join the multilateral alliance led by the United States. Although the Spanish public desired to join the European Economic Community (EEC) and appreciated the benefits of the



case studies of domestic politics and alliances 155

U.S.–Spain bilateral alliance, they were skeptical about the further integration of Spain into the Western defense structure.81 At the elite level, democratization enabled Spanish civilian elites to take a more critical stance toward the United States. During Franco’s rule, the Spanish military repeatedly obstructed Spanish diplomats’ efforts to improve the terms of the alliance. Spanish negotiators often found that Spain’s bargaining chips had been thrown away by its military officers, who had special interests in the alliance. Because the Spanish military “were anxious not to loosen the contact with their only source of materiel and technical expertise,” American negotiators often bypassed Spanish negotiators and made deals with powerful figures of the Spanish military such as Admiral Carrero Blanco and Captain General Muñoz Grandes (Viñas 2003, 12).82 The end of the military dictatorship, therefore, activated intraregime criticism of the alliance and improved the bargaining position of Spanish negotiators. Attitude of Spanish Leaders Franco was a quintessential proalliance leader in that he had a disproportionately high utility for cooperation in the alliance compared with other Spaniards. Although Spain had a neutralist tradition in its foreign policy, Franco was clearly anticommunist. More important, the alliance served his personal interest by strengthening his regime’s political, economic, and military foundation. Moreover, once the United States chose to ally with Franco, alternative leaders of Spain became antialliance by implication. As Angel Viñas (2003, 4) points out, the U.S.–Spain alliance deterred not just external enemies but also Franco’s internal enemies, and Americans knew this effect of the alliance.83 When neutralist sentiments arose within the regime, Franco and his conservative clique removed neutralist elites for the interest of the alliance. Spanish Foreign Minister Fernando Maria Castiella (1957–1969), a successful diplomat, was dismissed from the office after American negotiators and Spanish conservatives began to worry about his anti-American and neutralist tendency (Pollack 1987, 49; Viñas 2003, 10–11). Because the proalliance attitude of Franco’s clique was convenient for American strategic planners, the stance of the U.S. government toward the opposition became distant despite the otherwise prodemocratic inclination of American foreign policy.84 Even in December 1975, one month after Franco died, Secretary of State Henry Kissinger was not enthusiastic

156

case studies of domestic politics and alliances

about the prospect of Spanish democratization: In his conversation with Spanish Foreign Minister Jose Maria de Areilza, Kissinger said that “the ability to compromise of Spanish individuals is not . . . what made Spain great,” and he did not expect pluralism to come tomorrow in Spain (FRUS, 1969–1976, vol. E-15, part 2, 694).85 Stating that one of the worst mistakes Americans ever made was to encourage the opening to the left in Italy, Kissinger assured the Spanish Foreign Minister that the U.S. government would not pressure Spain to democratize. Considering the historical tie between the Spanish dictatorship and the United States, it was natural for the Spanish opposition to be critical of the United States. After the transition years, they finally took power in 1982. As Prime Minister Felipe Gonzalez Marquez (1982–1996) stated in a speech during his visit to the United States in 1985, “Those defeated in the civil war and the democratic opposition in general . . . have viewed [the alliance as] . . . American support for the dictatorship and a blow for the hopes of a rapid democratic restoration in Spain.”86 Although the Spanish Socialist Workers’ Party (PSOE) did not abrogate the alliance, they demanded a more equal standing in the alliance and less American military presence in Spain. The socialists also opposed Spain’s membership in NATO and pledged to put the issue on a national referendum—though they later changed their stance and supported the continued membership in NATO for the sake of Spain’s integration into the European Community (Stanton 1993). Consequently, the range of acceptable deals for Spain visibly shrank under the socialist government. During Felipe Gonzalez’s visit in 1985, Secretary of State George Schultz threatened that the United States would leave Spain if the troops were not welcome there. Gonzalez replied that that was not what he wanted, but, unlike his predecessors, he was ready to start discussing the exit if the United States wished to leave (Viñas 2003, 19). In sum, the attitude of Spanish leaders changed from extreme proalliance to somewhat neutralist as the former leftist opposition took power. Spanish leaders became proalliance when the conservative People’s Party (PP) took power in 1996, but the leadership shifted to a more neutralist line again when PSOE won the election in 2004, and Spain withdrew its troops from Iraq. Party differences still matter in Spanish policy toward the United States, but the U.S.–Spain alliance has become politically less controversial. PP won the election in November 2011, and its leader Mariano Rajoy is considered to be pro-American, but Spanish politics have been preoccupied with economic issues.



case studies of domestic politics and alliances 157

Vulnerability of Spanish Leaders Franco was dependent on the United States for many things, but he was still a relatively nonvulnerable leader. Although Franco’s Spain had a pariah status after World War II, the Allied Powers did not have sufficient incentives to invade it because Spain had stayed neutral during the war. As the Cold War intensified, and as the United States broke the isolation of Spain by signing the Pact of Madrid, Franco’s power base became secure. Until the very end of his life, Franco’s hold on power was strong. The conservative leaders who succeeded Franco were clearly more vulnerable. Transition to democracy gave real power to the opposition groups, and there was deep uncertainty about the future of the regime. Factional fighting had been going on in Franco’s final years within the conservative camp; after his death, there was no consensus among the ruling elites on the transition to democracy. As Franco himself arranged, the Spanish monarchy was restored, and King Juan Carlos became the head of state. The king, however, did not attempt to maintain the authoritarian regime because he understood that his legitimacy could be derived only from helping the country’s transition to democracy.87 Moderate reformists eventually prevailed over reactionaries within the regime, but the power balance was precarious throughout the transition period, with a failed coup d’état on February 23, 1981, and another attempt shortly before the general election of October 1982. In the legislature, the Union of the Democratic Center (UCD), which represented the moderate conservatives, never obtained a majority and was forced to work with leftist and other rightist parties. In contrast, Felipe Gonzalez, the socialist prime minister, was a relatively nonvulnerable leader within the democratic political structure of new Spain. The PSOE, illegal between 1939 and 1977, won a landslide victory in the 1982 general election and occupied 202 of 350 seats in the lower house of the Spanish legislature. Felipe Gonzalez pushed the PSOE in a more centrist direction and widened the party’s support base. Consequently, he was nonvulnerable within the party as well. He led the government from 1982 to 1996 and remains the longest-serving prime minister in democratic Spain. Although the vote for the PSOE was the highest in 1982 (48.1 percent) and declined in the following elections (44.1 percent in 1986, 39.6 percent in 1989, 38.8 percent in 1993, 37.6 percent in 1996), the party still kept the majority of the seats until 1993.88 Jose Maria Aznar replaced Felipe Gozalez as prime minister in 1996, as the conservative People’s Party (PP) formed a minority government with

158

case studies of domestic politics and alliances

support from regional parties. Aznar was clearly more pro-American than Gonzalez, but he initially “worked within the framework of the bipartisan consensus with the Socialist Party” (Iglesias-Cavicchioli 2007, 4). His position was strengthened in the 2000 election, when the PP obtained an absolute majority (183 seats) of the lower house. In his second term, Aznar adopted a very strong pro-American foreign policy, supporting the Iraq War despite the overwhelming domestic opposition (90 percent in 2003) against the invasion (Chislett 2005a).89 Aznar’s support of the Bush administration was unpopular, and the voters were further angered by his attempt to blame the Basque terrorists for the 2004 Madrid bombings by Islamic terrorists. In the general election held three days after the bombings, the PSOE won 164 seats against PP’s 148, and its leader Jose Luis Rodriguez Zapatero became prime minister. Zapatero won again in 2008 (169 seats against 154 of PP) and retained the office of prime minister until 2011. Overall, no leaders in democratic Spain seem to be as nonvulnerable as Franco or Felipe Gonzalez was. Effects of Spanish Domestic Politics Again, the long-term trend is increased inf luence and reduced profitability for Spain, especially in economic terms, although democratization increased Spain’s bargaining power and profitability in security affairs. Democratization activated domestic opposition to the alliance, but economic growth weakened the most credible source of opposition to concessions, lack of resources. At the time of the 1953 base agreement, Spain was simply a provider of military bases, and the United States did not acknowledge any other formal relations between them. In the process of the successive renewals of the base agreement, Spain gradually acquired the status of a U.S. ally (ATOP codes the alliance from 1963). The bilateral relationship was institutionalized and widened to cover issues other than the provision of military bases. As Spain’s inf luence in the alliance increased, the consultation mechanism of the alliance was strengthened, and the United States took Spanish demands more seriously. As a result of the growth of the Spanish economy, the Spanish negotiators became more critical of their country’s security arrangement but less insistent on demanding economic benefits from the alliance. Spain received from the United States US$500 million of military aid in grant form between 1954 and 1961 and US$1.238 billion in the form of loans (US$727 million) and grants (US$511 million) between 1962 and 1982.90 Since 1983, the U.S. military aid has been in the form of sales under con-



case studies of domestic politics and alliances 159

cessional credit terms, and even this military credit has faded away as the Spanish military became more self-sufficient. In the 1986–1988 round of negotiations over the bilateral defense cooperation, the Spanish negotiators requested the reduction of U.S. military presence and accepted the termination of American military and economic aid that had accompanied the previous base agreements. To explain the critical changes in the alliance, let us now turn to the effects of the three domestic variables. As Viñas (2003) points out, Spain’s intra-alliance bargaining power during the authoritarian period was weak because of Franco’s dependence on the United States, but the explanation can be misleading when applied to other cases. A leader’s dependence on an ally can also increase his or her bargaining power, as in the case of Nobusuke Kishi. Spain’s bargaining power was weak in the Franco era because there was no effective opposition against the alliance and because Franco was a nonvulnerable pro­alliance leader. Effective opposition to the alliance did not come from either the public or Franco himself, and the United States did not have to support the proalliance dictator for his survival. Consequently, there was little need for the United States to make concessions to the ally. Although the United States did provide military aid to Spain during the Franco era, the amount was far below what the Spanish negotiators asked for. Franco gave away many important concessions in security issues, and even in the issue of financial compensation for the bases, he caved in when the United States took a firm stance. Franco’s death immediately led to an improved bargaining position for Spain. Spanish leaders were still proalliance, but they were far more vulnerable than the late dictator, and American negotiators also saw the shadow of opposition against the alliance. The United States, therefore, had an incentive to make concessions to Spain in order to consolidate the proalliance group’s hold on power in the transition period. This move was also consistent with a more idealistic motivation of many Americans to give support to Spain’s democratization. U.S. policy makers’ connections with the Spanish opposition groups were weak, and they accurately perceived the opposition groups to be more neutralist than the conservative elites (Chislett 2005a, 2005b). Thus, it made sense for the United States to make the bilateral relationship more formalized and strong in the anticipation of future shocks to the alliance. In 1976, the United States gave treaty status to its ties with Spain for the first time in the bilateral security relationships and provided Spain with the highest level of security guarantee in their history.91 Because the United

160

case studies of domestic politics and alliances

States in the past negotiations always referred to Franco as the obstacle to Senate approval, one could alternatively argue that this development was a natural result of Spain’s democratization. The treaty status of the 1976 agreement, however, was not a foregone conclusion even a month after Franco’s death. In his conversation with Secretary of State Henry Kissinger on December 16, 1975, Spanish Foreign Minister Jose Maria Areilza requested that the agreement be given a status of treaty. Kissinger replied that he was sympathetic but was concerned that it might make a precedent for other base agreements. It seems probable that the Spanish foreign minister’s following reasoning was at least partially shared by American policy makers: If they “could show a Treaty, the connection with the Atlantic, and money, it would help counter all things the Communist Party is saying” (FRUS, 1969–1976, vol. E-15, part 2, 694).92 It is difficult to see a clear winner in the 1982 base agreement. Although there were improvements from the negotiations in the authoritarian period, and although some “institutional arrangements for the U.S. forces in Spain found a preliminary solution based on the models applied within” NATO, Viñas (2003, 18) argues that the terms of the agreement were largely in favor of the United States. Most important, the United States was able to keep its military presence in Spain intact. The Spanish government did not have a bargaining advantage in this round of negotiations. Although its leader Leopoldo Calvo-Sotelo was proalliance, the UCD government was too vulnerable and expected to end soon. An ally has an incentive to support a vulnerable proalliance leader but not one who is likely to lose power soon regardless of the ally’s support. The conservative camp was divided before the 1982 election, and the PSOE led all the public opinion polls.93 The 1982 agreement was signed in July, only three months before the general election in which the PSOE won an overwhelming majority. Perhaps American negotiators did not want to waste their political capital in negotiations with the moribund UCD government. Moreover, the United States by that point had less need for the UCD government to stay in power. The Spanish legislature had approved Spain’s membership in NATO—which the U.S. government had long supported— in December 1981, and Spain had been an official member of NATO since May 1982. Because the leftist oppositions within the parliament and the majority of the Spanish public were against Spain’s membership in NATO, the United States needed to save its political capital for future negotiations.94 By the time of the negotiations between 1986 and 1988, “the climate in which the bilateral relationship was developing had undergone a funda-



case studies of domestic politics and alliances 161

mental change,” and “the Spanish negotiating position was . . . very strong” (Viñas 2003, 20). The socialist opposition had taken power in 1982, shortly after the UCD government and the United States signed the base agreement. Spanish public opinion was still critical of Spain’s NATO membership and American military presence. A poll by El País in May 1985 showed that a substantial majority still opposed Spain’s continued membership in NATO, and 64 percent disagreed with the view that “the United States was a true and loyal friend of Spain” (Pollack 1987, 153). President Ronald Reagan’s visit to Spain in the same month met huge demonstrations against him. Felipe Gonzalez was able to confront the United States from a position of strength because he had a comfortable majority in the legislature and the support of public opinion. Although the PSOE government had changed their policy and supported Spain’s membership in NATO in the 1986 referendum, the socialist government pushed the United States hard in the negotiations for the 1988 agreement: “Ultimately, Spain’s manifest determination to invoke its sovereign rights—which included a threat to end the base agreement entirely if satisfactory terms could not be reached—forced the U.S. side to give up far more than it had hoped” (Basora 2009, 93). Spain did not join NATO’s integrated military command; nuclear weapons continued to be banned from the Spanish soil; Spanish control on the American use of the bases was strengthened; a gradual reduction of the American military presence was agreed—reducing the U.S. military presence by 40 percent (Viñas 2003, 18–21; Chislett 2005b, 31–32). As a result of the increased Spanish bargaining power, noted a Spanish scholar, “The enshrined imbalances, dependence, and trends toward a lack of proper supervision of U.S. activities in Spain were transformed into a well-balanced compact of duties, rights and responsibilities strictly respecting the full sovereignty of both parties” (Viñas 2003, 21). According to a British journalist, the 1988 agreement “laid the foundations of a more balanced relationship and no longer made Spain a kind of vassal of the United States” (Chislett 2005b, 32). The provisions and implementation of the base agreements continued to experience minor changes, but they were not as significant as those already discussed. The subsequent events also conform to my theory. The proalliance Aznar government (1996–2004) made concessions to the United States in its search for a special relationship. In addition to reinforcing Spain’s integration into the NATO and strengthening the U.S.–Spain bilateral relationship (as opposed to the socialists’ advocacy of Europeanization), the conservative government offered the enlargement of the Rota Naval Station and the

162

case studies of domestic politics and alliances table 6.3 Domestic variables and bargaining power of Spain.

Spain Franco

Opposition Attitude Vulnerability Bargaining power

Post-Franco conservatives (1976) UCD (1982)

Felipe Gonzalez (1988)

Weak Moderate Moderate Moderate Proalliance Proalliance ProAlliance Neutralist Low High Extremely high Low Weak Strong Moderate Strong

use of Moron Air base during the review of the 1988 agreement in 2002 (Cantalapiedra 2009, 5–6). The position of strength given to Aznar in the 2000 election made it possible for the conservative government to make these concessions and to support the unpopular Iraq War. Table 6.3 summarizes the main points of the U.S.–Spanish case. Overall, findings from the case studies of this chapter all support or at least do not contradict Hypotheses 5.1, 5.2, and 5.3. Vulnerable procooperation leaders and nonvulnerable anticooperation leaders had bargaining advantages in their relations with the United States, and domestic opposition to alliance cooperation increased the leaders’ bargaining power as well. In some cases, the causal mechanisms predicted in my theory were clearly traceable in diplomatic records. All this being said, however, the effects of domestic politics are not always strong, and it is difficult to tell how much domestic politics matter relative to international factors. The dependent variable of this study, the negotiation partner’s willingness to make concessions, is difficult to measure without going into the specifics of the cases, but generalization of the argument would require further studies with a larger number of observations.

Implications of the Findings In Chapters Five and Six, I have explained the effects of three domestic variables on intra-alliance bargaining with the case studies of America’s alliances with Japan, South Korea, and Spain. As previous literature noted, the presence of effective domestic opposition to cooperation increases a state’s bargaining power vis-à-vis its negotiation partner (Schelling 1960; Putnam 1988; Evans, Jacobson, and Putnam 1993). Democratization had distinct effects on South Korean and Spanish bargaining power by activating opposition to their alliances with the United States. However, domestic opposition, much less regime types alone, does not provide us with satisfactory explanations for the variations of intra-alliance bargaining.



case studies of domestic politics and alliances 163

My theory increased the explanatory power of domestic politics by additionally taking account of a leader’s attitude and vulnerability. By emphasizing the interaction effect of the two variables, I explained why a leader matters in intra-alliance bargaining and why looking at one of the variables alone can be misleading. A leader’s attitude alone cannot predict the reactions of the ally. Kishi’s proalliance attitude served him well, but that of the subsequent LDP leaders or Franco reduced their bargaining power. The neutralist attitude of Felipe Gonzalez brought rewards to Spain, but that of Roh Moo-hyun backfired. Likewise, the vulnerability of a leader does not have a uniform effect. Although some leaders such as Roh Moo-hyun, Ichiro Hatoyama, and Morihiro Hosokawa suffered from their vulnerability, others such as Kishi and post-Franco Spanish conservatives benefited from their shaky positions. When a leader seems vulnerable, a proalliance attitude increases (and a neutralist attitude reduces) his or her bargaining power because the ally has an incentive to support the proalliance leader (and remove the neutralist leader) by manipulating its concessions. In contrast, when a leader seems nonvulnerable, the ally is less generous toward a proalliance leader and accepts a tough stance of a neutralist leader. When a leader is extremely vulnerable, the ally would not support even a proalliance leader, but this is consistent with my theory because the ally does not gain from making concessions to a leader whom it will not face in future interactions.95 This chapter also showed that a state tends to receive less benefit from its alliance as its state resources increase, even though its inf luence in the alliance tends to grow with its capability. My findings about the trade-off between inf luence and profitability caution us against assuming a simple relationship between the capability and the bargaining power of a state. When we look at the long-term trends from the American perspective, the alliances have become more profitable to the United States, whereas the superpower has lost the enormous inf luence it used to have in the early decades of the alliances. To avoid complicating the argument, this chapter did not analyze American domestic politics, but the American side of the stories would be pertinent to the study of certain periods when economic or military hardships made the American public more critical toward U.S. commitments abroad (for example, the Nixon era). Policy implications of my argument are clear; certain combinations of the domestic variables improve the bargaining position of a state in an alliance. Policy makers as well as those who select them should carefully consider the effects of domestic opposition to an alliance, the attitude of a

164

case studies of domestic politics and alliances

leader, and his or her vulnerability. It should be noted, however, that an aggressive pursuit of profit hurts the foundations of an alliance, which cannot be sustained without mutual benefits. Even if one could manipulate the domestic variables without losing control of the subnational actors, one still risks the erosion of its alliance by pursuing high profitability. The range of acceptable deals for the ally f luctuates as international variables and the ally’s domestic politics change. For long-term national interest, therefore, it would be prudent to keep the profitability of an alliance not too high and have sufficient inf luence in the management of the alliance. Theoretically, my analysis in this chapter should be applicable to other international bargaining situations as well. We can look at domestic opposition and leaders’ attitudes toward a different kind of international cooperation. My theory, however, will work well only when it is applied to relations that are expected to continue for a sufficiently long period—for instance, trade partnerships between economically interdependent states, relationships between donors and recipients of foreign aid, or relationships between oil companies and host governments. As noted earlier, in other contexts such as arms control negotiations and crisis bargaining between adversaries, the applicability would be limited because the actors have strong uncertainty about the continuation of the bargaining in future.96 Within alliance literature, my theory brings a new angle to research on the effects of domestic politics on alliance management (for example, Kupchan 1988; Gaubatz 1996; Bennett 1997; Reed 1997; Snyder 1997; Narizny 2003a, 2003b; Gartzke and Gleditsch 2004). For instance, my ­theory provides alternative explanations to the finding of Leeds, Mattes, and Vogel (2009) that a state is more likely to end its alliance in violation of the terms when a leader with a different supporting coalition comes to power, but the effect is mitigated in democratic regimes. Leeds and her coauthors argue that democratic leaders have stronger constraints against changing the course of foreign policy or are more careful in making commitments, or both. Although these explanations are plausible, we could also explain the finding with the framework presented in this chapter. When the societal coalition of the leadership changes in nondemocracies, domestic opposition to an alliance suppressed under the previous leader is likely to surface and can drastically shrink the range of acceptable deals for the country. Because domestic opposition to an alliance is tolerated in democracies and even ref lected in alliance policy, such a deterioration of alliance relationships is less likely in political transitions within democratic regimes. Moreover,



case studies of domestic politics and alliances 165

because the public is not likely to elect a leader whose alliance policy is radically different from that of the society at large, democracies are less likely to have extreme shifts in alliance policy than nondemocracies. Whereas Leeds, Mattes, and Vogel (2009) emphasize the careful screening of alliance commitments by democratic leaders, democratic publics also screen their leaders according to the latter’s attitude toward alliances; for instance, a leader who proposes to dissolve the alliance with the United States has little chance of coming to power in Japan or South Korea. Further exploration of the effects of the three independent variables is warranted within and outside alliance literature. Finally, at least two future research agendas are left for the domestic theory of intra-alliance bargaining. First, I assumed that domestic politics of the United States did not have a strong impact on the intra-alliance bargaining of the three alliances. For my cases, simultaneously accounting for domestic politics of two states will hurt more than it will help the analysis (the number of cells in Table 2.1 will increase from 8 to 64), but the benefit of a more complicated model might exceed the cost for other cases. Second, I assume that actors do not discount the future benefits from an alliance too much—a reasonable assumption for my cases. In other cases, however, the assumption may not hold for one or both of the negotiating parties. What matters to a leader may be the benefit he or she receives from the ally during his or her time in power rather than what the leader’s state would gain in the future. Then, the strategic calculation about making concessions would be much more complicated as the future benefits get discounted by the leader and as the ally interacts with the leader with its own discount rate. In sum, there are a great many ways in which domestic politics can affect intra-alliance bargaining.

7

Conclusion

For the second largest power to balance against the top power, it needs strategic allies more urgently than economic profits. Yan Xuetong (2014, 159) Doubts about our continuing commitment, combined with economic inducements and diplomatic pressures emanating from Beijing, could compel some of our longtime friends to reappraise their own national security policies, including their alignments with us. Aaron Friedberg (2011, 6)

The findings of this book cast new light on the above statements of the leading Chinese and American international relations scholars. Yan Xuetong predicts the emergence of a Sino–American bipolarity, and this book has explained why two superpowers in a bipolar system actively seek military allies. Aaron Friedberg’s warning should be understood in the context of China’s possible future role as a supplier in the alliance market and how that affects the position of the United States in international politics. The two statements are noteworthy from the perspective of the alliance market also because the strategists acknowledge the role of nonmilitary goods in the alliance market. In this chapter, I summarize the arguments and findings of this book and discuss their theoretical and policy implications.

Market Theory of Alliances By developing a market theory of military alliances, this book has explained how the transactions between military allies have evolved, with a special focus on America’s relationships with Japan, South Korea, and Spain. As explained in the first chapter, I have been motivated to develop the theory of the alliance market by the inability of the existing theories to explain anomalous trends of alliances in the post–Cold War era. By conceptualizing military alliances as contracts of exchange, my market theory explains why many Cold War alliances persist, why many alliances were formed after the Cold War despite reduced military threats in many regions of the



conclusion 167

world, and why the United States is neither a member nor a target of these new alliances. With a deductive model, I have argued that the vice of alliance politics to which great powers easily succumb in a multipolar world is overreaction; in a bipolar world, it is overextension; in a unipolar world, it is inattention. I have then shifted the level of analysis to domestic politics, where I have shown the effects of domestic variables on intra-alliance bargaining. I found, as in the previous literature, that domestic opposition to alliance cooperation generally improves a state’s bargaining position vis-à-vis its ally. Domestic opposition to the state’s leader (and the leader’s vulnerability), however, was found to have different effects on the state’s bargaining power, depending on whether the leader is seen as proalliance. Namely, a vulnerable proalliance leader tends to benefit from the ally’s concessions whereas a vulnerable anticooperation leader is disadvantaged. Similarly, a nonvulnerable proalliance leader tends to be not successful in obtaining concessions from the ally, whereas a nonvulnerable anticooperation leader is likely to have a bargaining advantage. I have analyzed different aspects of the alliance market separately, but they are all connected by the common theme of this book—that alliances are contracts of exchange with which actors pursue their self-interests. As previous works indicate (Boulding 1958, 330; Waltz 1979), the system-level analysis of international relations has a natural affinity with the market approach, but other aspects also have an integral part in the alliance market. Contractual aspects of alliance agreements are important because the nature of the alliance market makes undelivered promises very costly, and the effects of domestic politics have to be considered because subnational actors bring their own agendas when participating in intra-alliance bargaining. Factors arising from the different aspects of alliance politics affect one another. Systemic factors (such as distribution of capabilities or supply and demand for military protection) affect conditionality of alliance agreements; they also have a significant impact on the domestic politics of alliance members. Contents of alliance contracts affect the benefits and risks of alliance agreements, thereby inf luencing how leaders and the public see the alliances and reinforcing or mitigating the effects of systemic variables. Domestic political variables affect intra-alliance bargaining and the contents of alliance contracts, and domestic politics also have some effect on systemic factors because changes in domestic politics in the long run contribute to systemic shifts of capabilities and threats.

168 conclusion

Theoretical Implications The market theory of military alliances is a good theoretical tool for integrating different fields of international relations. A general theory of alliances should not be confined to the field of international security, since international relations and alliances “often merge in all but name” (Liska 1962, 3). Scholars have studied the effects of alliances on trade (for example, Gowa and Mansfield 1993; Gowa 1994; Mansfield and Bronson 1997; Long 2003; Gowa and Mansfield 2004) and of trade on alliances (for example, Papayoanou 1999; Powers 2004; Long and Leeds 2006; Davis 2009; Fordham 2010), but this study advances the field by developing a new conceptual definition for alliances, thereby opening a possibility of theoretical integration between studies of alliances and economic cooperation. The market approach also contributes a new perspective to alliance literature, which is currently focused too much on the demand for security (that is, threats that states face: Liska 1962; Walt 1987).1 As I explain in Chapter Three, we need to pay more attention to the supply of security and the demand for nonsecurity goods in alliance politics. With appropriate modifications, economic theories can be applied to issues of international security. Charles Lipson once wrote that there were “few equivalents in the security field to the comprehensive, rule-guided arrangements in trade and money,” but he made an exception for alliances (Lipson 1984, 12). The field of international security tends to focus on conf lict rather than on cooperation, but, unlike many other subjects in the field, alliances are primarily about cooperation and are amenable to a market approach, which is about cooperation based on self-interest. With the market approach, my theory at the systemic level casts a new light on the issue of polarity, which has attracted attention because of uncertainty about the systemic property of the unipolar system (Wohlforth 1999; Hansen 2000; Ikenberry, Mastanduno, and Wohlforth 2009; Monteiro 2014) and also because of the power shift between the United States and China. Our analysis of unipolarity is inevitably inf luenced by our experience in the “current” and “American” unipolarity ( Jervis 2009), just as our analysis of multipolarity was colored by the European history of balance of power and as bipolarity was studied in the context of the U.S.–Soviet Cold War. By treating polarity as the number of security suppliers, however, I have developed a deductive model. Deductive analysis is especially fitting for unipolarity, which is historically novel. Moreover, our analysis of bipolarity and multipolarity should



conclusion 169

not be exclusively based on past experience. For instance, the systemic model discussed in Chapter Three made it clear that our understanding of bipolarity has been crucially affected by the neorealist assumption that two superpowers never collude with each other in a bipolar system. If in the future we have a bipolar system with two superpowers that are willing to collude with each other, we can modify the assumption and still use the systemic model to analyze it. As Kenneth Waltz’s works (1964, 1979) were inf luenced by the actual bipolarity, our analyses of unipolarity are (and should be) affected by our experience of the unipolar world, but we should still make efforts to develop deductive theoretical models. My theory is not about just any market. It is about a market of repeated exchanges, and contracts play an integral role in this market. From the realist view that emphasizes the anarchy of international relations (for example, Waltz 1979; Mearsheimer 1994), contracts may appear to be antithetical to international relations, especially in the field of international security. The field of security studies is structured around the question of war, which could suddenly end cooperation between states. In the analysis of adversary games, therefore, we often assume that it is difficult to withstand betrayal and to recover the loss from it in the future. Alliances, however, are about continuous cooperation. In fact, benefits of alliances can be reaped only with multiple interactions between allies, even in cases of temporary wartime alliances. There are temporal gaps between allies’ payments, and allies cooperate in the environment of asymmetric information. To address these problems as well as to cope with the uncertainty of the future, states form alliance contracts. These contracts have substantial effects despite the absence of government and third-party enforcement in the international system. Kim (2011), for instance, demonstrates that states carefully craft alliance agreements to avoid the negative consequences of alliance obligations. Many alliance agreements also have strategic incompleteness (Morrow 2000, 81). For some time now, economists have worked on self-enforcing contracts (for example, Telser 1980; Baker, Gibbons, and Murphy 2002) and incomplete contracts (for example, Grossman and Hart 1986; Hart and Moore 1988; Hart 1995), and military alliances may provide them with fertile ground for developing contract theory. At the domestic level, I have used the market theory of alliances to present a new perspective on interstate negotiations. The role of domestic opposition has been studied extensively in the literature (for example, Putnam 1988; Evans, Jacobson, and Putnam 1993; Fearon 1994; Schultz

170 conclusion

2001; Howell and Pevehouse 2007), but I have differentiated opposition to policy from opposition against a leader. By bringing in the leadership variables (attitude and vulnerability), my theory has expanded the explanatory power of domestic politics for international bargaining. It is also more relevant to policy analysis than theories that focus on regime types of bargaining parties. As my analysis reveals in Chapters Five and Six, democracy is an important permissive cause for domestic opposition to a government and its policy, but we cannot rely solely on regime types as the explanatory variable for our analysis of domestic politics. There is much variation of domestic politics within the same regime type, and regime types do not change frequently.

Policy Implications The market theory of alliances predicts that military alliances will continue to play an important role in international relations as long as interstate relations and demand and supply of military commitments exist. Unless one of the conditions disappears, states will continue to obtain efficiency gains from alliances by adjusting the contents of exchanges and contracts in accordance with situations they face. Even at the height of the unipolar system, when the George W. Bush administration used ad hoc coalitions in lieu of formal alliances, alliance politics were still active. As Campbell (2004) argued, we should expect formal alliances to remain relevant. Moreover, alliance politics are likely to be an important subject in the coming decades, for the reasons I discuss in the following pages. The distribution of military capabilities in the international system is likely to remain unipolar for the foreseeable future, but many perceive a decline in U.S. power, especially relative to the rising power of China. This perception of power shift affects both the demand side and the supply side of the alliance market. Moreover, the rise of China and the relative decline of the United States alter the costs and benefits of the alliances for the United States. Although we should not uncritically assume that U.S. military power will decline relative to that of China (Beckley 2011), it is imperative to consider such a scenario when discussing the future of alliance politics in East Asia and across the world. On the demand side of the alliance market, the rise of Chinese military threat increases the demands for military protection. China’s military budget now dwarfs those of its neighbors, and many consider China’s policy on territorial disputes to have become more assertive in recent years



conclusion 171

(Mearsheimer 2010; Swaine 2010; Friedberg 2011; Scobell and Harold 2013; but also see Johnston 2013). Russia’s intervention in Ukraine has increased demands for U.S. military protection in Europe as well. These demandside factors suggest that the United States as the most important supplier of military protection will have more bargaining power in the coming years, if other things remain equal. Other things, however, are unlikely to remain the same: The supply side of the alliance market will significantly change. On the one hand, cuts in the U.S. military budget will restrict the U.S. capacity and willingness to supply military protection to other states, which will add upward pressures on the price of U.S. military protection. Domestic politics of the United States can push the superpower toward a less internationalist foreign policy, and many security studies scholars in the United States have advocated retrenchment in security affairs (for example, Gholz, Press, and Sapolsky 1997; Walt 2006; Posen 2007).2 On the other hand, the United States will face the need to prepare against the emergence of China as a second pole, and Sino–American competition in the alliance market is likely to increase the supply of protection. In addition to these changes in the demand and supply of military commitment, the rise of China will have direct impact on the costs and benefits to the United States of certain alliances. As tensions rise between China and its neighbors, the United States will have a higher risk of entanglement. American policy makers have been particularly concerned about the possibility of conf lict between China and Japan, as the two East Asian states have escalated disputes over the Senkaku/ Diaoyu Islands.3 Perhaps ref lecting such concerns, support in U.S. public opinion for the U.S.–Japan alliance sharply declined after Japan’s nationalization of three disputed islands in September 2012. Since 1996, annual surveys for Japan’s Ministry of Foreign Affairs have asked the American public and American opinion leaders whether they supported the continuation of “the current Japan–U.S. Security Treaty.”4 The poll conducted in July and August 2013 recorded the lowest support for the alliance in the surveys’ history, with a sharp decline from the previous year (67 percent of the public and 77 percent of the opinion leaders supported the continuation as opposed to 89 percent and 93 percent in February/March 2012). The concerns for entanglement are not limited to military conf licts: Economic and political cooperation with China is also important to the United States. Unlike the Soviet Union during the Cold War, China today is a major economic partner of the United States. This will affect the U.S.

172 conclusion

attitude toward its military allies as well. China is not only one of the largest trading partners of the United States but also one of the largest foreign holders of U.S. government debt. The United States also benefits from political cooperation with China on a number of issues, ranging from nuclear nonproliferation to climate change and to food safety. As in the alliance market at large, nonmilitary goods matter in the context of Sino– American relations. China’s military rise is threatening not only to its neighbors but also to the United States itself. When we focus on the threat China poses to the United States, China’s neighbors will be more important to the extent that these states can offer militarily valuable goods to the United States (for example, military bases and cooperation over intelligence). Moreover, the United States needs to consider the military value these states can offer to China as well. The U.S. alliance policy needs to serve a “preemptive” function (Fedder 1968, 67) to stop military cooperation between China and its neighbors beyond a certain level. Thus, the more the United States is threatened by China relative to the other states’ perception of Chinese threat, the more generous the United States will be toward these other states. In this regard, the United States probably has more incentive to court South Korea than to court Japan because Japan has a strong incentive to balance against China whereas South Korea is more wary of getting entangled in a conf lict between China and the United States. As can be seen in the preceding discussion, the systemic analysis presented in Chapter Three is useful for planning for a postunipolar era or for regional security at the subsystemic level.5 For instance, Ross (1999) has predicted the emergence of Sino–American bipolarity in East Asia, and the ­theory presented in Chapter Three would predict an increase in the bargaining power of U.S. regional allies vis-à-vis the United States in such a scenario. An analysis based solely on the threat of China, which neglects supply-side factors, will lead to a significantly different (and misleading) prediction. Increased military capabilities of China will not only create demands for American military protection but will also raise the systemic supply of military protection by making China a potential security-exporter. As China’s projectable military power increases, other states will be more likely to seek China’s security assurance (including through an alliance with China). Thus, rebalancing of the United States toward Asia should be understood in the context of both China as a potential competitive security supplier and China as a security threat.6



conclusion 173

The general theoretical framework of the market theory (which emphasizes interactions between allies more than interactions between adversaries) and the empirical findings discussed in Chapter Four also have policy implications. Military allies are fighting against each other less and less, and alliance contracts also have restraining effects on allies’ military policy toward third parties (Schroeder 1976; Gelpi 1999; Weitsman 2003; Pressman 2008; Johnson and Leeds 2011; Kim 2011; Fang, Johnson, and Leeds 2014). Although changes in the contents of alliance agreements reflect rather than cause changes in international relations, contents of alliance agreements are not epiphenomenal because renegotiating an alliance agreement is costly (Gourevitch 1999) and reputation matters to states (Miller 2003, 2011; Gibler 2008). The evolution of alliance agreements from 1815 to 2003 shows that the nature of alliance agreements is becoming less offensive and less militarized (Chapters Three and Four; Leeds and Mattes 2007); we should continue this trend by refining military alliances as a tool for international security management. As a unipolar power, the United States has a particularly important role in this endeavor because, if it chooses to do so, it can use its bargaining advantage in the alliance market for the development of a more peaceful alliance system. For policy makers, day-to-day management of alliances is most affected by the domestic politics of the allies. Chapters Five and Six explained that the domestic political situations of American allies had important effects on intra-alliance bargaining despite the huge capability gap between the United States and its allies. It is unlikely that regime types will change in Japan, South Korea, or Spain in the foreseeable future. However, baserelated issues (such as crimes, accidents, environmental pollution) or controversial foreign policy (for example, the Vietnam War, the Iraq War) can quickly arouse domestic opposition to the alliances. Such a shift in public opinion can lead in turn to the emergence of neutralist or anticooperation leaders. In the short term, the political vulnerability of Shinzo Abe and Park Geun-hye, in addition to the 2016 U.S. presidential election, may be the most important variable to watch for observers of the U.S. alliances with Japan and South Korea. Provided that the United States wishes to keep its alliances, it will be better off making preemptive efforts to remove sources of friction that turn the allies’ populace against these alliances. Calder (2007), for instance, demonstrates that friction over American military bases is limited when bases are located in less densely populated areas and when financial compensation

174 conclusion

and mediating institutions are available for the local population affected by the bases. Aside from all the other concerns, this point suggests that the United States actually benefits from relocations of bases such as those in Futenma, Okinawa, and Yongsan, Seoul.7 American allies, of course, also need to pay attention to the domestic politics of the United States, especially under unipolarity, where the single superpower has fewer strategic constraints ( Jervis 2009; Walt 2009). After all, the military rationale for retaining alliances has become weak for the United States. Sino–American competition may create bargaining advantages for the existing and potential allies of the United States, but exploiting the advantages also entails the risk of alienating the United States. The allies of the United States have to convince the American public that the goods they supply are valuable, though different in nature from those provided by the United States. The research presented in this book aims to contribute to understanding of alliance politics and to beneficial exchanges among alliance partners.

Reference Matter

Notes

Chapter 1 1.  Public-goods models of alliances introduced by Olson and Zeckhauser (1966) are notable exceptions, but it is problematic to assume that military protection is a public good. Military protection is largely a rival (subtractable) and excludable good. 2.  McLeary 2015. 3.  Interviews in 2006, 2008, and 2012. I keep the anonymity of my interviewees except for cases in which they gave me explicit approval to quote them. 4.  “Koizumi Unequivocal on Iraq War,” Japan Times, March 19, 2003; available at http://search.japantimes.co.jp/cgi-bin/nn20030319a2.html. Also see Midford (2008, 15–18) and Izumikawa (2010, 150–154). 5.  On May 21, 2015, U.S. President Barack Obama announced that Tunisia would become a major non-NATO ally, even though the status does not include any security commitment. See Peter Bakermay, “Obama Upgrades Tunisia’s Status as a U.S. Ally,” New York Times, May 21, 2015. 6.  China formed nine nonaggression pacts between 1992 and 2003, but this study does not regard pure nonaggression pacts as military alliances. 7.  Though less institutionalized than American alliances formed after World War II, such hegemonic alliances were also formed by ancient Rome and nineteenth-century Britain. 8.  The number of alliance “ties” and the number of alliance “memberships” within the existing U.S. alliances slightly increased, and my quantitative analysis takes into account that increase. An alliance can be signed without ratification, and 13 out of 27 alliance memberships of the United States in the ATOP dataset (version 3.0) are through executive agreements. 9.  See, for instance, Gulick ([1955] 1967), especially 33–34, 58–62; Haas (1953); and Wight (1966). 10.  For the debate on American primacy and “soft” versus “hard” balancing, see Brown et al. (2009). A lively debate has developed to explain the absence of external

178

Notes to Chapter 1

balancing against the United States (see Ikenberry 2002; Walt 2002; Paul 2004; Brooks and Wohlforth 2005; Lieber and Alexander 2005; and Pape 2005). 11.  Neither the balance of power nor the balance of threat explains this phenomenon well (Waltz 1979 and Walt 1987). Realists predicted the demise of NATO after the Cold War (for example, Mearsheimer 1990 and Waltz 1993). Institutionalists (for example, McCalla 1996 and Haftendorn, Keohane, and Wallander 1999) offer an explanation for alliance persistence but do not predict the increase of alliance formation. On the issue of alliance persistence, also see Duffield (1994); McCalla (1996); Walt (1997); Wallander (2000); and Thalakada (2012). 12.  These numbers exclude pure nonaggression pacts. 13.  The number of active alliances has been increasing. For historical trends in alliance politics, see Leeds and Mattes (2007). Figures 1.1, 1.2, and 1.3 exclude 72 alliance agreements that were simply replaced by new agreements including the members of the replaced alliances, but the pictures are very similar when I treat the renewals as terminations and formations. 14.  Although capability aggregation for defense and deterrence has been the dominant framework for analyzing military alliances (for example, Liska 1962; Walt 1987; and Snyder 1997), there have been works that explain other aspects of military alliances. Scholars have noted that states form alliances not only for increasing security against external threats but also for enhancing internal security (Barnett and Levy 1991 and David 1991), binding their ally (Schroeder 1976 and Weitsman 2003), revising the status quo (Schweller 1994), or managing ambiguous security risks (Haftendorn, Keohane, and Wallander 1999). More concretely, “Alliances have been formed to recognize the independence of states, to clear passage or territory for trade, to put down insurgencies on both land and sea, to create regional organizations, and even to exchange or secure territories among alliance members” (Gibler 2009, 553). Capability aggregation is an effective means for some of these ends but an inadequate one for others. 15.  Albania and Croatia signed the accession protocols in 2008 and joined NATO in 2009, a fact not yet ref lected in the ATOP dataset (version 3.0). Though much less conspicuous, the Organization of American States (OAS) also expanded when Belize and Guyana joined it in January 1991. 16.  For critiques of NATO expansion, see Brown (1995) and Mandelbaum (1996). For a supportive view, see Ball (1998). 17.  In addition, there is a moral hazard problem, in that once these states join the alliance, they have less incentive to live up to the standard of the alliance (Wallander 2002, 4). 18.  For critiques of the capability aggregation model, see Schroeder (1976) and Morrow (1991). As Goertz, Gibler, and Powers (2008) point out, the traditional theories of alliances are not necessarily wrong but rather “radically incomplete.” 19.  Patrick E. Tyler, “U.S. Strategy Plan Calls for Insuring No Rivals Develop,” New York Times, March 8, 1992. 20.  Such exchanges take place not just in military alliances. See Lake (1999, 2001, 2007, 2009) for exchanges in hierarchical relationships. See Palmer and Morgan (2006) for a two-good model of foreign policy. 21.  As these efficiency gains do not create an extreme division of labor in international trade, neither do they make every state allied with every other state. (Non) tradability of goods and services as well as transaction costs make some exchanges unworthy. Moreover, alliance commitments are highly rivalrous (subtractable). Forming



Notes to Chapter 1

179

one alliance sometimes means giving up another alliance option. Increasing opportunity costs also hinder states from fully specializing in one good. 22.  For a portfolio diversification model of alliances, which also focuses on efficiency gains, see Conybeare (1992, 1994). 23.  In the context of hegemonic stability theory, Russett had pointed out that “many of the gains from hegemony have been less collective goods than private ones, accruing primarily to the hegemon and thus helping maintain its hegemony” (Russett 1985, 208). For critiques of the public goods models of alliances, see, for example, Sandler and Cauley (1975); Boyer (1993); and Goldstein (1995). 24.  “A subsidy treaty was a written agreement between two (occasionally three) states concerning troop-hire terms, and usually included political cooperation” (Wilson 1995, 85). 25.  See Meernik (2008) for a supply and demand theory of U.S. military policy, although supply in his theory does not include support from other states. 26.  The Sino–American rapprochement “signalled a relative reduction in the strategic importance of Japan for the United States, along with a corresponding readiness to view Japan as much as an economic competitor as a strategic ally” (Ito 2003, 2). 27.  Increased demands for U.S. military protection against China can be seen in other Asian states such as the Philippines and Vietnam as well. For an analysis of global public opinion on the threat of China and the United States as an ally, see Pew Research Center, July 2014, “Global Opposition to U.S. Surveillance and Drones, but Limited Harm to America’s Image.” From the Chinese perspective, the reinforcement of the U.S.–Japan alliance and other U.S.-led security cooperation have been threatening (Christensen 1999). Military threats are affected by alliances as well, although the potential endogeneity is outside the scope of this study. Intra-alliance dynamics can affect outside actors and cause security dilemma. For classical works on alliances and security dilemma, see Snyder (1984, 1997). 28.  Russia and North Korea reached a new alliance agreement in 2000, but, unlike their defense pact during the Cold War, the agreement contains only the obligation of consultation. 29.  On the effects of the collapse of the Soviet Union on North Korea’s trade, see Eberstadt (2011), especially 67. 30. On the advantages North Korea and other socialist states gained from the competition between the Soviet Union and China, see, for example, Zagoria (1962, 394–399) and Christensen (2011). 31.  This does not mean that nongreat powers cannot form alliances among themselves, but such minor power alliances are militarily less consequential. 32.  For a thorough review of the literature on polarity, see Buzan (2004). Buzan differentiates superpowers and great powers. 33.  Chapter Three develops a model that explains the effects of systemic polarity on concerns for abandonment and entrapment. Simply put, abandonment is a betrayal by one’s ally, and entrapment means being dragged into a conf lict of one’s ally. 34.  Palmer and Morgan (2006) similarly develop a two-good theory of foreign policy, where states pursue efficient mixtures of maintenance and change. 35.  Although most of the literature has neglected the nonmilitary aspects of military alliances, there are some notable exceptions. For instance, a group of researchers have studied the effects of alliances on trade (for example, Gowa and Mansfield 1993; Gowa 1994; Mansfield and Bronson 1997; Long 2003; and Gowa and Mansfield 2004). More

180

Notes to Chapter 1

recently, there have been works that explore the effects of trade on alliances (for example, Papayoanou 1999; Long and Leeds 2006; Davis 2009; Fordham 2010; and Poast 2012). My task, however, is not just to account for the two-way causation between alliances and trade; the problem is the total conceptual separation of alliances and trade. Aside from types of goods exchanged, military alliances and trade partnerships have quite similar mechanisms. In fact, recent scholarship shows that many alliance agreements include clauses on economic cooperation (Long and Leeds 2006), and many trade agreements include military commitments (Powers 2004). A market theory of alliances can integrate theoretical insights from studies on alliances and economic cooperation. 36.  Capability is derived from the Composite Index of National Capability (CINC) score (Singer, Bremer, and Stuckey 1972) of the Correlates of War project, using ­EUGene (version 3.204). Alliance data are from the ATOP dataset (version 3.0) and exclude nonaggression pacts. 37.  Benjamin Fordham found that “the relationship between trade and asymmetric alliance formation was stronger when major powers faced greater international threats” (Fordham 2010, 694). His result is consistent with an exchange framework if we assume the mutual reinforcement between military and economic cooperation. 38.  Shinzo Abe, “Toward an Alliance of Hope.” April 29, 2015, before the U.S. Congress. 39.  As Fedder (1968, 67) points out, some alliances serve a “preemptive” function (that is, to stop another state’s power being added to a potential adversary), and the political loyalty of countries like Japan and West Germany was militarily important as well. 40.  Most agree that the world was multipolar until 1945, bipolar until roughly 1989, and then unipolar (see, for example, Snyder 1997, 18). For operationalizations of polarity, see Modelski (1974) and Rapkin, Thompson, and Christopherson (1979). Haass (2008) argues that we are entering the age of nonpolarity, and Schweller (2010) argues that polarity has become less important in international relations. 41.  This study draws on the work of Glenn Snyder (1984, 1997), who developed theories on the relationships between alliance games and adversary games. Although Snyder’s works are innovative, there are three shortcomings in his analyses of military alliances. Namely, he excludes cases from nonmultipolar systems (that is, bipolar and unipolar systems), cases of asymmetric alliances (in terms of military capability), and cases that show exchanges of nonmilitary goods among allies. The exclusion of bipolar and unipolar cases is problematic not only because it limits his theory’s applicability but also because it reduces the opportunity to demonstrate the wide spectrum of alliance dynamics. Systemic polarity affects alliance dynamics through the availability of alternative allies and the distribution of capability. Snyder, however, only brief ly discusses how alliance structure rigidity decreases abandonment fears under bipolarity, and he does not distinguish the effects of bipolarity on great powers from its effects on nongreat powers, as explained in Chapter Three. The exclusion of asymmetric alliances is problematic not only because asymmetric alliances are a common and important form of alliances but also because it biases our view of military alliances in general. Snyder’s cases are all alliances formed between great powers that are militarily dependent on each other. Consequently, his cases overrepresent the capability aggregation aspects of alliances. The third shortcoming, which is related to the second point, is that ignoring nonmilitary goods is problematic because not all states can supply military goods.



Notes to Chapters 1 and 2

181

42.  For instance, when the United States was reluctant to provide protection during the Iran–Iraq War, the Kuwaiti government sought Soviet protection for its tankers. The United States then agreed to protect Kuwaiti tankers by registering the ships so that they f ly the f lag of the United States and come under American protection (Hansen 2000, 125; and Caron 1990, 155–156). Nonpolar powers can gain nonsecurity benefits as well: The leftist military government that took power in Peru in 1968 “made it clear to the Nixon Administration that it would not hesitate to turn to the Soviets for aid if the American cut it off ” (Maurer 2011, 1). 43.  Asahi Shimbun, June 30, 2011, Morning Edition, 27. 44. See Asahi Shimbun online at www.asahi.com/special/wikileaks/. 45.  DPJ’s call for “seiji shudo” (politicians’ leadership) worried American and Japanese bureaucrats who have been the managers of the alliance. Tatsumi (2010) warns that the politicization of the alliance management can put the long-term health of the alliance at risk. 46.  With respect to the effects of domestic politics on intra-alliance bargaining, existing theories of international relations suggest that domestic opposition to an alliance increases a state’s bargaining power vis-à-vis its ally. When a leader’s hands are tied by domestic opposition, he or she can credibly take a tough stance in international negotiations (Schelling 1960; Putnam 1988; and Evans, Jacobson, and Putnam 1993). Because domestic opposition is weak in nondemocratic regimes, regime types have a significant impact on intra-alliance bargaining (Dong Sun Lee 2007). Historical accounts of military alliances suggest that top leaders also have significant inf luence on alliance relationships, but theoretical literature on alliances is strangely silent about their role. See, for example, Kataoka and Myers (1989), Buckley (1992), and Schaller (1997) on the U.S.–Japan alliance; Hong (1999) and Oberdorfer (2001) on the U.S.–South Korea alliance; and Rubottom and Murphy (1984), Pollack (1987), and Viñas (2003) on the U.S.– Spain alliance. In Chapters Five and Six, in addition to the role of domestic opposition to an alliance, I examine the effects of a leader’s attitude toward an alliance and his or her vulnerability on intra-alliance bargaining. 47. For a macrohistorical study on war and state development, see Tilly (1990). Gibler and Wolford (2006) explain the effects of military alliances on democratization of alliance members, and Gibler (2007) argues that removal of territorial issues between neighbors causes both peace and democracy. 48.  Neoclassical realists have advanced our knowledge on the interaction of domestic politics and international security (Rose 1998; Schweller 2003). 49.  The applicability of alliance theory might be even broader; Shenkar and Arikan (2009) argue that researchers of nation-state alliances and interfirm alliances can benefit from interdisciplinary enrichment. Conybeare and Kim (2010) illustrate the similarities between corporate alliances and military alliances. 50.  On the differences of nonaggression pacts from other alliances, see Mattes and Vonnahme (2010) and Leeds and Savun (2007, 1125). Chapter 2 1.  On the contractual aspects of alliances, see Kim (2011). 2. For a concise overview of the paradigms, see Jack Snyder (2004). For examples of works that explain the continuation of Cold War alliances, see Walt (1997) for realism; McCalla (1996) and Haftendorn, Keohane, and Wallander (1999) for

182

Notes to Chapter 2

i­nstitutionalism; and Suh (2007) for the mixture of realism, institutionalism, liberalism, and constructivism. 3.  For reviews of neoclassical realism, see Rose (1998) and Schweller (2003). 4.  Two psychologists have proposed an “alliance hypothesis for human friendship,” where they argue that human friendship is like alliance politics (DeScioli and Kurzban 2009). What they mean by alliance politics, however, is creating support groups for potential conf licts, and they explicitly distinguish allies from exchange partners. Interestingly, Plato in Lysis writes that if “the enemy were to go away, [the friend] is no longer, it seems, a friend to us” (quoted in Ludwig 2010, 134). 5.  For our purpose, games here simply mean interactions between adversaries or allies. For three modes of these games, see Snyder (1997, 33–34). Miller (1995) explains the effects of multipolarity and bipolarity on adversary games. This book does not address an important aspect of alliance politics—inf luence attempts targeted at enemy alliances. On wedge strategy, see Crawford (2008, 2011) and Izumikawa (2013). Izumikawa (2015) analyzes binding strategies to maintain the cohesion of an alliance. 6.  The literature on balancing is large. See Waltz (1979) and Walt (1987) for the most popular theories of balancing and Schweller (2006) for when balancing is likely to be absent. 7.  Also see Storey (1999) and Buszynski (2002). 8.  Morrow defines security as a state’s “ability to maintain the current resolution of the issues that it wishes to preserve” and autonomy as “the degree to which” a state “pursues desired changes in the status quo” (1991, 908–909). Accordingly, he does not differentiate military and nonmilitary goods, and the same good can be categorized as security or autonomy depending on how we see the status quo. For a similar approach using maintenance and change, see Palmer and Morgan (2006). 9.  For the relationship between the Anglo–Portuguese alliance and the two countries’ commercial relationship, see Belcher (1975) and Shaw (1998). 10.  The market analogy may not be straightforwardly applicable, however. For example, see Glenn Snyder’s following conjecture: To use another image, alliance making in a multipolar system is not like perfect competition in an economic system, where many buyers and sellers are present but the market sets the price, nor is it like bilateral monopoly, where the price is set by bargaining between a single buyer and single seller, analogous to a bipolar system. Structurally, it is like oligopoly, with a few sellers (or buyers) collaborating to set the price, but behaviorally, it tends toward duopoly, that is, the few are often only two. (Snyder 1997, 148)

As will be seen in Chapter Three, I do not consider alliance politics under bipolarity to be bilateral monopoly. 11.  More recently, there have been works that explore the effects of trade on alliances (for example, Papayoanou 1999; Long and Leeds 2006; Davis 2009; Fordham 2010; and Poast 2012). 12.  The unit of analysis is dyad-years. Capability is derived from the Composite Index of National Capability (CINC) score of the Correlates of War project, using the EUGene, version 3.204 (available at www.eugenesoftware.org/). Alliance data are from the Alliance Treaty Obligations and Provisions (ATOP) dataset, version 3.0 (available at http://atop.rice.edu/) and exclude nonaggression pacts. 13.  It might be better to have different thresholds for different periods because technologies affect the “loss of strength gradient” (Boulding 1962).



Notes to Chapter 2

183

14. Benjamin Fordham’s research offers an explanation for the big difference in capability ratios. He argues that great powers form asymmetric alliances with minor power trading partners to protect their trade relationships but notes that it may be a post–World War II phenomenon. He suggests that colonial relationships served similar functions as asymmetric alliances before the war (Fordham 2010, 692). 15.  Even when I limit the analysis to dyad-years with defense commitment, only about 17 percent of the highly asymmetric allied dyad-years have basing arrangements. Because “BASE” is an alliance-level variable in the ATOP dataset, all dyads in a multilateral alliance are coded to have basing arrangements if the alliance has any basing arrangements. 16.  On July 1, 2014, the cabinet of Prime Minister Shinzo Abe decided to reinterpret Article 9 of the Japanese constitution to exercise the right of collective self-defense. 17.  On the role of the U.S.–Japan alliance in Japan’s national identity and Japan’s connection to the West, see Wirth (2015). 18.  In 1995, Japan’s contribution to U.S. forces in Japan amounted to US$4.1 billion, compared with the US$1.7 billion of South Korea and the US$1.6 billion of NATO allies combined (Wolf and Zanini 1998, 11). Despite fiscal problems, Japan’s contribution remains to be the most generous among U.S. allies (US$4.4 billion in 2002; Yoda 2006). Also see Calder (2007). 19.  Cooley and Spruyt (2009) discuss the exchange and transfer of sovereignty between states. 20.  Capability aggregation is not necessarily for efficiency gains. Even if aggregation is less efficient (that is, output/input ratio is lower than individual states’ military efforts), if defeating an enemy requires a certain amount of output, aggregation is necessary and, to that extent, effective. My focus on efficiency gains, of course, does not negate the importance of such effectiveness, which still requires contracts of exchange. 21.  See Norrlof (2010) for the economic benefits the United States receives from being a security supplier. 22.  Choi (2003; 2004) argues that democratic alliances are better at capturing the benefits of comparative advantages than are nondemocratic alliances. 23. See Hartley (2006) for an analysis of defense industrial policy in military alliances. 24.  Bearce, Flanagan, and Floros (2006) argue that the internal information effects of alliances reduce military conf lict among member-states. 25.  In their survey experiment with a sample of 950 U.S. citizens, Tomz and Weeks (2015) found that a sense of moral duty was the most important mechanism that explains public support for U.S. military intervention for an ally under attack. They also found that the respondents were more willing to honor alliance commitments when the costs of intervention were low and when the country under attack was a democracy. 26. Sandler and Hartley (1999), however, argue that role specialization within NATO is limited. 27.  McCalla (1996, 461) sees NATO as “part of a broad multilevel and multi-issue relationship among” member states. Gavin (2002, 2003, 2004) and Zimmermann (2001) show the importance of monetary cooperation for the alliance. Risse-Kappen (1995) explains the importance of societal actors in the transatlantic alliance. Ikenberry (2011, 183–190) discusses the long-term military, political, and economic commitments built in the Western alliance.

184

Notes to Chapters 2 and 3

28.  Alliance is analogous to relational contracting, whereas alignment is analogous to normal trade, where coincidence of interest is sufficient and commitment is not required. According to Williamson (1985), relational contracting occurs when frequency of transaction is recurrent (as opposed to occasional) and when investment characteristics are mixed or idiosyncratic (as opposed to nonspecific). 29.  Alastair Smith (1995, 1996) points out selection effects concerning alliance reliability (that is, a state with an unreliable ally is more likely to be targeted than one with a reliable ally). The interdependence of alliance reliability and occurrence of war creates a sample bias. Leeds argues that targeting effects would be less important once we identify the factors that inf luence challenger estimates of alliance reliability (2003a, 818). See Leeds (2003b) for roles that specific contents of alliance treaties play in initiation of militarized interstate disputes. 30.  The concepts of abandonment and entrapment were first developed by Mandelbaum (1981) and popularized by Snyder (1984, 1997). 31.  On the effects of asymmetric interdependence on international relations, see Hirschman ([1945] 1980) and Keohane and Nye (1977). 32. The primary alliance security dilemma occurs during the process of alliance formation, with the formation of one alliance pushing the adversary to form a counteralliance. 33.  Entrapment should mean only undesirable entanglement in which the entangling state adopts a risky or offensive policy not specified in the alliance agreement (Kim 2011). Examining U.S. militarized interstate disputes from 1948 to 2010, Beckley (2015) finds that even entanglement is a relatively uncommon reason for U.S. military intervention abroad. 34.  For examples of works that operationalize polarity, see Modelski (1974) and Rapkin, Thompson, and Christopherson (1979). Schweller (1998) argues that tripolarity deserves a distinctive analytical category. 35.  Moreover, the effect of the demand side, which depends on the threats that a state faces, has been extensively studied in the previous literature, such as Walt (1987) and Snyder (1997). 36. Snyder (1997) extensively discusses cases from multipolarity but only brief ly deals with bipolarity, and his theoretical focus is more on the interplay between alliance games and adversary games. Walt (2009) analyzes alliances in a unipolar world. 37.  On political leaders’ motivations for alliance and alignment policy, see Barnett and Levy (1991); David (1991); Levy and Barnett (1992); and Kim and Sciubba (2015). On the effects of regime types on the choice of allies, see Siverson and Emmons (1991); Siverson and Starr (1994); Simon and Gartzke (1996); Weart (1998); Lai and Reiter (2000); and Leeds et al. (2002). There is also literature on the effect of regime types on the durability (for example, Gaubatz 1996; Reed 1997; and Bennett 1997) and reliability (for example, Leeds 2003a; Gartzke and Gleditsch 2004; and Leeds and Savun 2007) of alliances. Chapter 3 1.  See Snyder (1997) for different levels (preparedness, diplomacy, and action) of adversary games and alliance games. 2.  Tony Blair. “Doctrine of the International Community.” Speech at the Economic Club of Chicago, April 22, 1999.



Notes to Chapter 3

185

3.  My focus is explicitly on formal military alliances, though I argue that they have nonmilitary aspects as well. For case selection, I follow the operational definition of the Alliance Treaty Obligations and Provisions (ATOP) dataset: Alliances are “written agreements, signed by official representatives of at least two independent states, that include promises to aid a partner in the event of military conf lict, to remain neutral in the event of conf lict . . . or to consult/cooperate in the event of international crises that create a potential for military conf lict” (Leeds et al. 2002, 238). 4.  See Chapters One and Two. 5. The Pentagon’s February 18 draft of the Defense Planning Guidance for the Fiscal Years 1994–1999. Excerpts are available on New York Times, March 8, 1992; www .nytimes.com/1992/03/08/world/excerpts-from-pentagon-s-plan-prevent-the-reemergence-of-a-new-rival.html?pagewanted=all. 6.  When buyers are inf luential, changes in the number of buyers can also be im­ portant. As noted in Chapter One, China in a sense became a customer of U.S. military protection in the early 1970s. China as a new buyer of U.S. protection meant a reduced bargaining power for Japan vis-à-vis the United States. Toward China, Nixon and Kissinger “repudiated the ideologically-driven U.S. policy . . . in favor of pursuing a quasi-alliance,” but toward Japan, “they emphasized traditional anti-Communist ideology in order to draw Japanese concessions over textiles and the reentry of nuclear weapons into the Okinawa islands, in exchange for their nominal reversion to Japan” (Ito 2003, 3). 7.  The two-goods model in Figures 1.4 and 3.1 is obviously a simplification, and states exchange a variety of goods. Given the military might of the United States, it could also be argued that many industrialized allies of the United States came to possess comparative (though not absolute) advantage in economic fields. Some countries such as Japan and Germany in the postwar era came to possess a disproportionately high comparative advantage in the supply of economic goods because of political constraints on their military postures. For instance, West Germany supported the U.S. dollar during the 1960s, “helping in effect to finance the US war effort in Vietnam, and it was one of the largest contributors to South Vietnam economically” (Colman and Widen 2009, 491). On alliance politics of smaller states, see Reiter and Gaertner (2001). 8.  See Tago (2008) for the relationship between U.S. aid policy toward participants and nonparticipants in U.S.-led coalitions. 9. See, for example, Schroeder (1976); Altfeld (1984); Morrow (1987, 1991, 1993, 2000); Barnett and Levy (1991); David (1991); Levy and Barnett (1992); and Weitsman (2003). 10.  See the text of the treaty at www.mofa.go.jp/region/n-america/us/q&a/ref/1 .html. 11.  “Officials across administrations of different parties stress that the desire of Korea and Australia to tighten their security relationships with the United States was a core reason why Washington was able to enter into free trade agreements with them and to do so on terms favorable to U.S. economic interests.” (Brooks, Ikenberry, and Wohlforth 2012, 43–44). 12.  Resolution of trade frictions, of course, was an important goal. On the security implications of the FTA, see Heo (2008) and Sohn and Koo (2011). On the role of the U.S. Congress in the FTA, see Seo (2015). Im (2015) attributes Roh’s FTA policy to the appreciation of the South Korean currency (won).

186

Notes to Chapter 3

13.  Although Japan and Australia are not military allies, they offer another recent example of such an exchange, albeit in an implicit manner. Former Australian Ambassador to Japan (2004–2011) Murray McLean assesses that Australia’s decision to provide security for the Japanese Self-Defense Forces ( JSDF) contingent in Samawah, Iraq was a factor in making it possible for then Japanese Prime Minister Junichiro Koizumi to manage and overcome opposition from within the Liberal Democratic Party’s agricultural lobby against economic cooperation with Australia. Koizumi agreed to conducting a feasibility study into a possible Free Trade Agreement/Economic Partnership Agreement between Japan and Australia when then Australian Prime Minister John Howard visited Japan in April 2005, shortly after Australia had announced its decision to take on the responsibility of providing the protection for the JSDF contingent in southern Iraq (my personal communication with Murray McLean). 14.  The United States supplies military protection to both Japan and Saudi Arabia, whereas the latter two supply different goods to the United States. Saudi Arabia is not a formal ally of the United States, according to the ATOP dataset. 15.  Office of the Secretary of Defense. 2013. “Military and Security Developments Involving the People’s Republic of China 2013,” p. 38. 16.  According to a survey of 11 Asian countries in 2014, respondents in Malaysia and Pakistan see China as the “greatest ally” whereas those in Bangladesh, India, Indonesia, Japan, Philippines, South Korea, Thailand, and Vietnam chose the United States. The Chinese respondents chose Russia as the number one partner. Pew Research Center, July 2014, “Global Opposition to U.S. Surveillance and Drones, but Limited Harm to America’s Image.” 17.  Yan Xuetong (2014) argues that China should shift from the strategy of keeping a low profile to the strategy of striving for achievement. With a Sino–American bipolarity in his mind, Yan contends that Chinese foreign policy should now focus on making friends and allies (by providing military protection and economic benefits to other states) rather than on making money. 18.  Ian Storey (2009) argues that reforms within the Indonesian military and Indonesia’s critical role in counterterrorism were more important than the China card. 19.  Nobuyoshi Sakajiri. “Afurika beigun kyoten, Chugoku no te [On U.S. military base in Africa, China’s hand].” Asahi Shimbun, June 24, 2014. 20.  Wang Yiwei, another inf luential Chinese scholar, argues that China and South Korea should sign a friendship and partnership agreement that approximates an alliance treaty. See You Sang-chul, “China Gets Chummy.” Joongang Daily, October 16, 2014; available at http://koreajoongangdaily.joins.com/news/article/article.aspx?aid=2996088. 21.  South Korea, of course, may decide to deploy the American missile defense system in the future. See, for example, Banyan 2015, “Shield and Spear: A Row over Missile Defence Tests South Korea’s Friendships with America and China.” May 2, The Economist. 22. “Japan Reluctant to Join China-Led New Asian Investment Bank,” July 6, 2014 (Mainichi Japan); available at http://mainichi.jp/english/english/newsselect/ news/20140706p2g00m0bu007000c.html. 23. Michael Green, “An Optimistic Relationship,” Korea Joongang Daily, July 11, 2014; available at http://koreajoongangdaily.joins.com/news/article/article.aspx?aid= 2991842. 24. Michael Green, “Korea in the Middle,” Korea Joongang Daily, June 11, 2014; available at http://koreajoongangdaily.joins.com/news/article/article.aspx?aid=2990401.



Notes to Chapter 3

187

25.  Victor Cha, “Five theories of unification,” Korea Joongang Daily, July 22, 2014; available at http://koreajoongangdaily.joins.com/news/article/article.aspx?aid=2992317. 26.  Another simple way to describe the systemic distribution of capabilities is to focus on the concentration of capabilities. On capability concentration and polarity, see Wayman (1984) and Mansfield (1993). 27.  I use the term entrapment here because it is a widely used term, but it is better labeled as entanglement (Chapter Two). 28.  Capabilities in the model ref lect resources of potential and actual allies, who exchange a wide variety of goods to obtain efficiency gains. 29.  Other variables include, but are not limited to, external threats, how much allies care about their reputation, the allies’ past behavior, tightness of an alliance, geography, and the terms of an alliance. For effects of reputation on alliances, see Miller (2003, 2011) and Gibler (2008). 30.  For instance, in the case of the United States abandoning New Zealand by suspending its obligation to defend the island country in 1986, what mattered was the U.S.–New Zealand bilateral relation (over nuclear ship visits) within the trilateral Australia, New Zealand, United States Security Treaty (ANZUS) (Catalinac 2010). In the case of France’s withdrawal from NATO during Charles de Gaulle’s presidency, France probably had multiple meaningful bargaining partners within the multilateral alliance. 31.  Even a very strong state can be abandoned when a weaker state’s interests fundamentally change, as in the case of Iran abandoning its alliance with the United States after the Islamic Revolution. Such occurrence is rare, however, despite numerous upheavals in the domestic politics of U.S. allies. 32.  As noted earlier, the model assumes other things to be held constant, but in reality the probability and cost of abandonment are also affected by the adversary. The ally’s decision to intervene is likely to be affected by the strength of A’s adversary, and the consequence of abandonment will also be more serious to A when the adversary is strong. 33.  See Lee (2006) and Saunders (2007) on how Asian allies of the United States involved themselves in the Iraq War. 34.  This may partially explain Beckley’s (2015) finding that entanglement has been a relatively uncommon cause for U.S. military intervention abroad. 35.  These points are consistent with Stephen Walt’s following argument: “For great powers, in short, the abandonment/entrapment dilemma will be most intense under multipolarity, somewhat diminished under bipolarity, and least worrisome under unipolarity” (Walt 2009, 98–99). As explained in the following discussion, however, my model, with the neorealist assumption, predicts that great powers fear abandonment by lesser powers more in bipolarity than in multipolarity. 36.  Another popular argument is that alliances are f luid in a multipolar system and rigid in a bipolar system (Waltz 1979; Duncan and Siverson 1982). The following discussion is consistent with this view in that great powers in a bipolar system have incentives to tighten their alliances. 37.  This does not have to be true, but it was historically true for the United States and the Soviet Union. The assumption would have been more problematic, for example, if we had a U.S.–UK bipolarity. 38.  Snyder, in another article, also writes: “The two superpowers have no common enemy strong enough to motivate them to ally, and their allies either have no incentive

188

Notes to Chapters 3 and 4

to align with the opposite superpower, or if they do have an incentive, they will be prevented from acting upon it by their own patron” (Snyder 1990, 118). In my opinion, the statement about lesser allies depends on other factors such as the geography and political systems of these states. 39.  This certainly depends on the situational contexts. For example, certain U.S. allies may prove to be important suppliers of goods that the United States needs for fighting the War on Terror (Byman 2006). Uzbekistan’s high strategic value to the United States after the September 11 attacks is interesting in this regard. 40.  My theoretical argument is most directly relevant to great powers’ alliance policies, but systemic characteristics described in the following pages are also likely to affect alliances that do not involve great powers. 41.  Alliances are abrogated under bipolarity as well, but the process may be more gradual. For instance, although the U.S. policy toward China and Taiwan clearly shifted in the early 1970s, the United States did not abandon Taiwan until 1979 and even then retained informal relations with the former ally through the Taiwan Relations Act. 42.  Intra-alliance war, in fact, is not uncommon (Bueno de Mesquita 1981, 81–83), and certain alliance arrangements are more conducive to intra-alliance conf licts than others (Krebs 1999; Long, Nordstrom, and Baek 2007). Some states use alliances “to tether threats,” and “tethering alliances, formed between states seeking to manage their conf lict of interest, are not usually terribly cohesive” (Weitsman 2003, 12). Bearce, Flanagan, and Floros (2006) argue that military allies are less likely to fight one another because alliances provide information about the members’ military capabilities and reduce the information problems that lead to war. If, as I argue, there are systematic differences between systems with different numbers of poles, polarity might affect the degrees of both the incomplete information problem and the commitment problem, two major causes of war (Fearon 1995). 43.  Snyder considers lesser allies to be constrained by bipolarity because he assumes that bipolarity leads to bipolarization: “The confrontation between superpowers” in a bipolar system “tends to homogenize the interests of the allies clustered around each pole” (Snyder 1997, 349). 44.  Where security importers lacked purchasing power, as in Africa, the two superpowers exported fewer formal alliance commitments. The Soviet Union allied with Somalia, Angola, Mozambique, Ethiopia, and Congo, but the alliances are all consultation pacts. The United States, perhaps because its alliance commitment was more popular, was more selective and allied only with Liberia (a consultation pact since 1959), a country founded by former U.S. slaves. 45.  On asset specificity, see, for example, Williamson (1985), Lake (1999), and Weber (2000). Chapter 4 1.  “We exclude alliances formed after 1989 because an extremely small percentage of these alliances terminated prior to 2001, when our data set ends. Because multilateral alliances may have different termination dates for different members, we have decided to begin by evaluating our arguments only with respect to bilateral alliances” (Leeds and Savun 2007, 1125).



Notes to Chapter 4

189

2.  On the durability of a unipolar system itself, see Wohlforth (1999) and Monteiro (2014). 3.  At the alliance level, over 95 percent (155/163) of alliances that began between 1990 and 2003 were active as of the end of 2003. Even at the membership level, over 86 percent (398/462) of alliance memberships that began between 1990 and 2003 were active as of the end of 2003. 4.  Data on national capabilities are available only from 1816. 5.  I count alliance memberships that ended in 1945 as observed under multipolarity because much of their duration was under multipolarity and most of them ended due to the end of World War II. 6.  Duration, however, is not a good proxy for reliability because alliance terminations and violations of alliance obligations are two separate phenomena. 7.  When the capability is missing for the first year of the membership, I used the second year for the calculation. 8.  Cha (2010), for example, argues that the United States preferred bilateral alliances in East Asia because Americans thought that tight bilateralism restrained allies better than a multilateral alliance. 9.  Following Box-Steffensmeier and Jones’s recommendation (2004, 137), Grambsch and Therneau’s global proportional hazards test statistic and Harrell’s rho were calculated. Keele (2010) cautions against confusing specification errors with a violation of proportional hazards assumption. 10.  On substantive interpretation of nonproportional hazards in Cox models, see Licht (2011). I exclude from the figure observations with duration under 300 days because the confidence interval takes extreme values in the early days of memberships. 11.  My focus here is on the effects of polarity, but the effects of other variables are also interesting. The membership-shortening effect of WARTIME is very strong, and the negative coefficient of the interaction term is too small to change the direction of the effect over time. The life-shortening effect of specified alliance duration naturally increases over time because such a clause terminates an alliance membership only after certain periods of time. 12.  On the changing roles of alliances, see Tertrais (2004). On “the end of alliances,” see Menon (2007). 13.  I use directed dyad-years data because State A and State B can have different obligations to each other. Here, I count only one tie in a directed dyad, even when they have multiple alliance agreements between them. 14.  The U.S.–France alliance of 1778 is not included in the ATOP dataset (version 3.0), which starts in 1815. 15.  I used the ATOP’s directed dyad-year dataset (version 3.0), counting only the first allied dyad-year for an alliance. I counted multiple alliance ties between a pair of states when they are allied through more than one alliance. 16.  In addition to extending commitments to ten former Eastern Bloc states in 1997 and 2003, the United States made commitments to Germany in 1990 (replacing West Germany) and to Belize and Guyana in 1991 (new members of OAS). 17.  This increased the bargaining power of the Soviet allies. See, for example, Reisinger (1992) and Harrison (2003). 18.  Though my systemic analysis focuses on the demand for and supply of military commitment, other factors in reality are not held constant. The early years of the

190

Notes to Chapter 4

­ nipolar period coincided with the economic boom in the United States. The effects of u the shift to unipolarity discussed in the following pages were reinforced by the relative lack of needs for nonmilitary goods on the part of the United States. The global economic crisis of 2007–2008 changed the picture by increasing the demands for various economic goods in the alliance market. 19.  The U.S.–Israel alliance began with the signing of the Memorandum of Understanding in November 1981. It was soon suspended until 1983, but their cooperation continued (Peretz 2002, 278). The formal alliance ceased to exist in December 1991 because the agreement was specifically aimed against the Soviet Union. 20.  “Record of Decision for Guam and CNMI Military Relocation.” Department of the Navy and Department of the Army, September 2010; available at www.guambuildupeis .us/documents/record_of_decision/Guam_Record_Of_Decision_FINAL.pdf. 21.  On the ambivalent attitude of the United States toward European military cooperation, see Larres (2002, 588–591). Gibbs (2009) argues that Western interventions in the former Yugoslavia in the 1990s were significantly motivated by the U.S. desire to retain dominance over its Western allies. 22.  Anthony Lake, Clinton’s national security advisor, is a well-known proponent of democratic peace. See Eyal (1997), Goldgeier (1998, 1999), and works in Rauchhaus (2001) for explanations of the decisions that led to the NATO enlargement. 23.  When demand for military commitment disappears or when no country can supply military commitment, alliances will cease to exist. 24.  Not all major powers are great powers and constitute poles (Correlates of War Project, available at www.correlatesofwar.org/). 25.  Another way to cope with the reduced supply of military protection is to increase the efficiency and self-reliance of arms production. The two objectives are hard to pursue at the same time but are not necessarily incompatible. For instance, intraEuropean codevelopment and coproduction of weapons, as opposed to Europe’s collaboration with the United States or others, increased from 42 percent in the 1970s and 43 percent in the 1980s to 57 percent in the 1990s ( Jones 2007, 159). See Caverley and Kapstein (2012) on more recent trends in the international arms trade. 26.  In terms of the number of allies, the United States continues to have the highest number of allies in the world because it leads large multilateral alliances. In fact, the number of U.S. allies increased from 52 in 1989 to 63 in 2003. As shown in Figure 4.4, however, the proportion of the U.S. allies in the system declined in the post–Cold War era. Second highest in this regard is Canada, with 46 allies in 1989 and 59 allies in 2003. Although the members of the Organization of American States (OAS) dominated the top ranks in 1989, we see more European states in 2003 (France, the third highest, with 48 allies, and Romania, fourth, with 36 allies). 27.  Three states signed the accession protocols on December 16, 1997, and seven states on March 26, 2003. 28.  My theory cannot be tested in terms of disintegration or tightening of alliances because both are consistent with it. The important questions for us are which state is distancing itself and which is making efforts to tighten alliances. 29.  For European responses to unipolarity through NATO and ESDP, see PressBarnathan (2006) and Posen (2006), respectively. 30.  Subjectively, in fact, the Japanese public seem to feel more threatened in the post–Cold War era (Izumikawa 2010, 149–154).



Notes to Chapter 4

191

31. It was also a time when Japan was taking measures to alleviate the risk of abandonment. 32.  See the Ministry of Defense, available at www.mod.go.jp/j/approach/zaibeigun/ us_keihi/suii_table_53-60.html. On Japan’s host nation support, see Yoda (2006) and Calder (2007). 33.  See the Ministry of Foreign Affairs, available at www.mofa.go.jp/mofaj/area/ iraq/shien.html. 34.  Japan’s Ministry of Foreign Affairs, available at www.mofa.go.jp/mofaj/area/ afghanistan/the_kabul_conference1007/shien.html. 35.  The discourse, like many others, is not politically neutral, given that this episode was extremely useful for expanding Japanese military contribution. In 2000, Brent Scowcroft, the U.S. National Security Advisor under George H. W. Bush, commented that the financial contribution of Japan was more useful than military contribution at the time of the Gulf War (Sotooka, Honda, and Miura 2001, 417–418). 36.  Joint Statement of the U.S.–Japan Security Consultative Committee, February 19, 2005; available at www.mofa.go.jp/region/n-america/us/security/scc/joint0502 .html. 37.  “U.S.–Japan Alliance: Transformation and Realignment for the Future,” October 29, 2005; available at www.mofa.go.jp/region/n-america/us/security/scc/doc0510 .html. 38.  This is a significant action given postwar Japan’s pacifism (Berger 1998 and Katzenstein 2008). 39.  “The Guidelines for Japan–U.S. Defense Cooperation,” April 27, 2015; available at www.mod.go.jp/e/d_act/anpo/pdf/shishin_20150427e.pdf. 40.  Office of the Secretary of Defense, Department of Defense, Allied Contributions to the Common Defense. 2004 Statistical Compendium. 41. Ibid. 42.  According to the White House, 49 countries were publicly committed to the “coalition of the willing” in March 2003; available at www.whitehouse.gov/news/ releases/2003/03/20030327-10.html. Newnham (2008) shows how the United States used nonmilitary goods (economic linkage) to form the coalition. Thompson (2009) explains U.S. policy toward the UN Security Council and Iraq by focusing on strategic information transmission. 43.  Adam Taylor. “Why Poland Wants a U.S. Military Base.” Washington Post, June 3, 2014. 44.  South Korea’s export dependency on China was 24.5 percent in 2012, 24.1 percent in 2011, and 25.1 percent in 2010. Korea Customs Service; available at www.customs .go.kr/kcshome/. 45. On the origins of the U.S.–ROK alliance, see Ahn (1998); Murata (1998b); Tucker (1998); Hong (1999); and Stueck (2012). 46.  The Vietnam War had a very different meaning to South Korea because North Vietnam was an Asian and communist state. Still, South Korea sent Vietnam the largest number of troops among U.S. allies mostly because of Seoul’s ties with Washington. In addition to economic rewards, South Korea also received political rewards from fighting in Vietnam: “According to the White House, South Korean participation gave them a ‘special place in our feelings,’ and the right to be ‘genuinely consulted [about strategic decisions].’ This led to a series of high-level exchanges between Washington

192

Notes to Chapter 4

and Seoul, including trips by President Johnson, Vice President Humphrey and ranking officials” (Ehrhardt 2004, 673). 47.  On internal and external balancing, see Waltz (1979). 48.  In fact, some Chinese scholars have argued that an alliance with the United States would be more beneficial to China than an alliance with Russia is (Zhang 2012, 140). 49.  Of course, bases came and went during the Cold War, too. Given America’s willingness to extend its base network almost wherever it could, we find the strongest explanatory variables of base politics in domestic politics of host nations (for example, regime change). On the global network of U.S. military bases, see Calder (2007). 50.  Vernon Loeb. 2003. “New Bases Ref lect Shift in Military.” Washington Post, June 9, A01. 51.  Also see Vernon Loeb. 2003. “New Bases Ref lect Shift in Military.” Washington Post, June 9, A01. 52.  Incidentally, Italy is the largest financial contributor among NATO allies to U.S. troops stationed in-country per soldier (Wolf and Zanini 1998, 13). The percentage of U.S. forces in Europe based in Italy tripled from around 5 percent to more than 15 percent between 1991 and 2013. See David Vine, 2013, “The Pentagon’s Italian Spending Spree,” Tomgram; posted on October 3 and printed on October 9. Available at www .tomdispatch.com/blog/175755/. 53. On October 1, 2013, U.S. Defense Secretary Chuck Hagel stated that the troop level would not decline from 28,500 (available at http://america.aljazeera.com/ articles/2013/10/1/us-to-maintain-symbolicmilitarypresenceinkorea.html). 54.  Ashley Rowland. “Year in Review: US, South Korea Agree to Again Delay Handover of Wartime Operational Control to Seoul.” Stars and Stripes, December 22, 2014. 55.  A South Korean newspaper reported that the U.S. Deputy Assistant Secretary of Defense Richard Lawless threatened the South Korean Ambassador to the United States with the withdrawal of the USFK if Seoul did not accept the strategic f lexibility; Hankyoreh, June 10, 2005, p. 3. 56. Wook-sik Cheong. 2006. “ROK–U.S. Alliance: More Harm Than Good: A South Korean Progressive’s Perspective about the Alliance.” Ohmynews International (Seoul), April 4, 2006; available at http://english.ohmynews.com/articleview/article_ view.asp?at_code=321054. The realignment has been repeatedly delayed and the details of the cost-sharing are not clear. “Korea to Pay for Yongsan Relocation.” Korea Joongang Daily, August 30, 2010; available at http://joongangdaily.joins.com/article/view .asp?aid=2925308. 57.  On the implications of these moves, see Hisae (2005) and Sunohara (2007). 58.  On the relocation, see Kan and Niksch (2010). The move of Marines from Okinawa to Guam is linked to the relocation of the Futenma base, the implementation of which has been repeatedly delayed. Because Guam is more distant from mainland Asia, the original plan was to move command units rather than combat units (Sunohara 2007, 196–199). The composition of the units to relocate has changed since then and is not clear. For a more recent discussion of the base problems, see Heginbotham, Ratner, and Samuels (2011). The relocation plan and the budget continue to change. 59.  Because the surrounding area of the Atsugi base is densely populated and bright at night, Iwakuni is a better place for night landing practice (Hisae 2005, 183).



Notes to Chapters 4 and 5

193

60.  “$26 billion bill coming Japan’s way: Pentagon.” Japan Times, April 27, 2006; available at http://search.japantimes.co.jp/cgi-bin/nn20060427a2.html. 61.  As in NATO’s enlargement, domestic politics can also push the United States to form new alliances. 62.  Rothkopf (2015) argues that allies and adversaries alike are “strategically rebalancing” away from the United States and toward China in the Middle East. 63.  Ref lecting on the positive effects of the U.S.–Japan alliance on the two countries’ economic relationship during the Cold War, Michael Finnegan made the following remark about the future of the alliance: “How the perception of a less-than-useful security alliance would affect thinking in other areas of the relationship is unpredictable but risky: Japan’s decreasing value as a security partner could in turn reduce its value as a partner in other spheres (e.g., economics)” (Finnegan 2010, 13). Japan’s value as an economic partner, however, can increase its value as a security partner, and U.S. bases in Japan continue to be valuable. Chapter 5 1. Tong Kim. 2008. “Irony of Roh Moo-hyun.” The Korea Times. February 24; available at www.koreatimes.co.kr/www/news/opinon/2008/02/137_19498.html. 2.  The following argument is based on Kim (2014a). 3.  For examples of works that discuss the effects of domestic constraints on security affairs, see Fearon (1994); Schultz (2001); and Howell and Pevehouse (2007). 4.  The vulnerability of leaders is not entirely independent of domestic opinion on cooperation; for instance, being proalliance in the face of domestic opposition increases a leader’s vulnerability. It is, however, analytically more useful to separate the two concepts. 5.  The use of domestic opposition for bargaining could backfire, however. Opposition to alliance cooperation increased in the Philippines after its democratization, and the Philippines demanded more concessions from the United States, but this led to the closure of American bases and reduced aid to the Philippines because the United States was not ready to concede (Park 2011). 6.  Yuko Kawato pointed out to me that opposition in nondemocratic regimes sometimes plays an important role. According to her, in fact, Marcos used domestic opposition to obtain concessions from the United States. 7.  Tsebelis’s theory explains policy stability or instability but not the direction of change. In contrast, domestic opposition explains the direction of policy change but not policy stability or instability. Opposition can cause policy stability by denying concessions to an ally, but it can also lead to policy change by pushing an ally to concede more. 8.  On neutralism in the context of the Cold War, see Morris (1960) and Anabtawi (1965). 9.  See the DPJ’s manifesto in 2009 at www.dpj.or.jp/english/manifesto/manifesto 2009.pdf. 10.  See McGillivray and Smith (2008) for a theory of leader-specific punishment. 11.  Some of the U.S. documents have been declassified (see Kornbluh 1998). 12.  See Andrew and Mitrokhin (2005). In the ATOP dataset, the alliance terminated when the Soviet Union installed a new leader in place of Amin.

194

Notes to Chapters 5 and 6

13.  For an empirical work on the effects of domestic political vulnerability on deterrence, see Leeds and Davis (1997). They analyze vulnerability attributable to the environment a leader faces; that is, economic decline and electoral cycles. Their findings are more closely related to my argument about the effects of domestic opposition on international bargaining. They find, for example, that “other states are less likely to make demands on” a state “when its economy is performing poorly” (Leeds and Davis 1997, 831). 14.  Depending on the scope of the research, it may be sometimes useful to treat a group or a series of leaders as one actor. For instance, the successive LDP leaders may be treated as a single nonvulnerable proalliance leader to the extent that their rule and alliance policy was stable and homogeneous, although personal differences among them were important in some cases. 15.  I thank an anonymous reviewer for pointing out this issue. On the effects of leader replacement on international bargaining, see, for example, Schultz (2005); Wolford (2007); McGillivray and Smith (2008); and Park and Hirose (2013). 16.  Licht (2010) demonstrates that leaders who face a higher risk of losing office are significantly more likely to receive foreign aid. Her finding is consistent with my argument, given that leaders who receive such aid are likely to be those who are friendly toward the donor state. 17.  As noted earlier, if it has to make concessions, the ally prefers to make concessions to a proalliance leader to encourage conformity to the ally’s position. Such reputation is important outside the bilateral relationship, too. 18.  These experts include scholars, and current and former high-ranking officials of the Ministry of Foreign Affairs, Defense Agency (now, the Ministry of Defense), and the Japan Self-Defense Forces. Chapter 6 1.  Part of the analysis is based on Kim (2014a). 2.  For the same reason, right-wing groups, who were otherwise unhappy with Japan’s subordination to the United States, were willing to support the alliance because it served as a barrier against communism. 3.  “Telegram from the Embassy in Japan to the Department of State,” Tokyo, June 24, 1960. 4.  NSC 6008/1, “United States Policy toward Japan,” Washington, DC, June 11, 1960. 5.  “Memorandum from the Acting Director of the Office of Northeast Asian Affairs (Hemmendinger) to the Assistant Secretary of State for Far Eastern Affairs (Robertson),” Washington, DC, August 22, 1956. 6.  I count the seats of Rightist Socialists, Leftist Socialists, the Socialist Party, Communist Party, Labor and Farmer Party, and Social Democratic Party. On the electoral results, see, for example, Ishikawa and Yamaguchi (2010). 7.  “The investment paid off handsomely. In the May 1958 election to the Diet’s lower house, the LDP retained nearly all its seats while the frustrated Socialists fell to bickering, culminating in a party split at the end of 1959. Meanwhile one of the Americans involved in the operation remarked cynically that Japanese politicians proved they were like those everywhere else—anyone was welcome to play in their game as long as they put up the money” (Schaller 1995).



Notes to Chapter 6

195

8.  Unofficially, Japanese minesweepers had been sent to the Korean War in 1950, four years before the JSDF was formed. 9. Calder (2007) explains how “compensation politics” developed to solve base-­ related problems in Japan. 10.  “Memorandum from the Secretary of State’s Special Assistant for Intelligence (Armstrong) to the Acting Secretary of State,” Washington, DC, October 10, 1955. 11.  “President John F. Kennedy’s advisers envisioned a future in which Japan doubled or tripled its exports to the U.S., making Japan so dependent on American consumers that it could never contemplate neutrality . . . this trade strategy contributed to an economic transformation barely imaginable to leaders on either side of the Pacific” (Schaller 1995). 12.  See “Kokusai mondai ni kansuru chosa (Polls on International Affairs),” M095, M106, and M107; available at at https://ssjda.iss.u-tokyo.ac.jp/. 13.  Dulles’s statement is quoted in “Memorandum of Howe to Robertson,” January 24, 1958, Lot File 61 D 68, RG 59, Stack 250, Row 49, Compartment 19, Shelf 6, E-1341, Subject Files Relating to Japan, 1957–1959, Box 12, National Archives at College Park, MD. 14.  For American reactions to the 1960 crisis, see Sarantakes (1999). 15.  See “Kokusai mondai ni kansuru chosa (Polls on International Affairs),” M095, M106, and M107; available at https://ssjda.iss.u-tokyo.ac.jp/. 16.  Nye (1995) argues that the U.S. military presence in East Asia provides “oxygen” to the region. 17. Cabinet Office, Government of Japan; available at www8.cao.go.jp/survey/ index-gai.html. 18.  See polls by Yomiuri Shimbun at www.yomiuri.co.jp/feature/fe6100/koumoku/ 20091211.htm and www.yomiuri.co.jp/feature/fe6100/koumoku/20101222.htm. 19. Cabinet Office, Government of Japan. Available at http://survey.gov-online .go.jp/h24/h24-gaiko/2-1.html. 20.  For Kishi’s rise as a pro-American leader in postwar Japan, see Schaller (1995). Some point out that Kishi was technically not a war criminal because he was never indicted. 21.  The CIA supported another conservative politician, Taketora Ogata, but Ogata died in January 1956. In 2009, declassified CIA documents revealed close cooperation between the intelligence agency and Ogata. See Mainichi Shimbun, July 26, 2009, 1–2. 22.  “Memorandum from the Acting Assistant Secretary of State for Far Eastern Affairs (Sebald) to the Under Secretary of State (Hoover),” Washington, DC, March 4, 1955. 23.  FRUS, 1955–1957, vol. 23, part 1, 519. “Telegram from the Embassy in Japan to the Department of State,” Tokyo, October 18, 1957. 24.  While the report was being written, Japan’s prime minister changed from Hosokawa to Tsutomu Hata to Tomiichi Murayama (1994–1996). 25.  For instance, in February 2009, Ichiro Ozawa, then the head of the Democratic Party of Japan, was severely criticized for his remark that the U.S. 7th f leet would be sufficient for the American military presence in the Far East. 26. Allies’ perception matters more than reality, but perception will eventually catch up with the reality. Yukio Hatoyama and other leaders of the DPJ are fundamentally proalliance, perhaps slightly neutralist relative to the LDP leaders. There is less un­certainty about a non-LDP Japanese government than there was at the time of

196

Notes to Chapter 6

the H ­ osokawa government. The members of the DPJ, like those of the LDP, vary significantly in their views on foreign policy, and the party does not have a consensus strong enough to alter the status quo of the Japanese foreign policy radically. The Social Democratic Party’s initial presence in the coalition did not push the DPJ’s policy significantly to the left, because the DPJ had 308 of 480 seats in the House of Representatives. Unlike in the early years of the Cold War, there is not a stark contrast between the opposing political groups of Japan about alliance policy, and the effects of government turnover are likely to be limited. 27.  Yukio Hatoyama. “Watashi no seiken kousou [My Plan for the Future Government].” Bungei Shunju, November 1996. 28.  Hatoyama’s original essay in Voice was translated, edited, and published in the Christian Science Monitor and the New York Times. See Yukio Hatoyama. “Watashi no seiji tetsugaku (My Political Philosophy).” Voice, September 2009, available at www .csmonitor.com/Commentary/Opinion/2009/0819/p09s07-coop.html and www .nytimes.com/2009/08/27/opinion/27iht-edhatoyama.html?pagewanted=all. 29.  See, for example, Mark Landler and Martin Fackler. 2009. “U.S. Is Seeing Policy Thorns in Japan Shift.” New York Times, September 1; available at www.nytimes .com/2009/09/02/world/asia/02diplo.html?_r=0; and Isabel Reinolds and Chisa Fujioka. 2009. “U.S. Open to Talks on Japan Alliance Issues: Envoy.” Reuters, September 18, 2009; available at www.reuters.com/article/idUSTRE58H2XL20090918; and postelection articles published by Japanese major newspapers such as Asahi, Yomiuri, Mainichi, and Sankei. 30.  Yuriko Koike. “Hosokawa shushou taijin no hikigane ha ‘kitachousen yuuji’ datta [North Korea Crisis Triggered the Resignation of Prime Minister Hosokawa],” July 2002, Seiron. 31.  Asahi Shimbun, June 30, 2011, morning edition, 27. There are various divisions, even within groups that are generally proalliance. Over the realignment of U.S. Forces Japan, for example, Hisae (2005) and Sunohara (2007) illustrate the divisions between Japan’s Ministry of Foreign Affairs and Ministry of Defense as well as the frictions between the U.S. Department of State and Department of Defense. Interservice rivalry within the U.S. military also plays an important role in alliance politics. In addition to Hisae (2005) and Sunohara (2007), see, for example, Packard (2010, 101). 32.  “Memorandum from the Acting Assistant Secretary of State for Far Eastern Affairs (Sebald) to the Under Secretary of State (Hoover),” Washington, DC, March 4, 1955. 33. See Asahi Shimbun’s analysis of U.S. diplomatic cables at www.asahi.com/ politics/update/0506/TKY201105060408_01.html and www.asahi.com/politics/ update/0511/TKY201105110530_01.html. 34.  “From this perspective, the following remark by Michael Green is interesting: Previous examples of alliance redefinition have all involved strong leaders, such as Nakasone, Hashimoto, or Koizumi, who were able to articulate a new vision for the security relationship while making all the political deals necessary to reconsolidate bases and expand security cooperation” (Green 2010a, 18). It is not surprising that Japan’s contribution to the U.S.–Japan alliance has significantly expanded in Shinzo Abe’s second term as prime minister (2012–), as Abe has a much stronger power base than in his first term. 35.  “Memorandum from the Secretary of State to the President,” Washington, DC, June 12, 1957.



Notes to Chapter 6

197

36.  “Telegram from the Embassy in Japan to the Department of State,” Tokyo, October 18, 1957. 37. Quoted in Sarantakes (1999). Herter to Johnson, June 20, 1960, Foreign Relations–­Japan Folder, Box 770, Senate File, LBJ Archive, LBJL. 38.  “Treaty of Mutual Cooperation and Security with Japan,” June 7, 1960, EightySixth Congress, Second Session, 1960 (Washington, DC, 1960), 28; “Treaty of Mutual Cooperation and Security with Japan,” June 7, 1960, Executive Sessions of the Senate Foreign Relations Committee (Historical Series) vol. 12, Eighty-Sixth Congress, Second Session, 1960 (Washington, DC, 1982), 410. 39.  “Telegram from the Embassy in Japan to the Department of State,” Tokyo, June 21, 1960. 40.  “Eurasia Group’s Top Risks of 2010,” 5; available at http://eurasiagroup.net/ item-files/1001-04%20Top%20Risks%20announcement.pdf. 41.  “How Long for Hatoyama?” February 15, 2010; available at www.economist .com/blogs/banyan/2010/02/how_long_ japans_hatoyama. 42.  See Havens (1987) for Japan’s roles in the Vietnam War. 43.  There were discussions of further revisions. Plans to strengthen the military aspects of the alliance have been incrementally adopted without formal revisions. There were opposite kinds of plans, too. For instance, the Japanese Ministry of Foreign Affairs has at some points suggested reducing the geographic scope of the so-called Far East clause, limiting the stationing of U.S. forces to emergencies and making the alliance “political” rather than military. The U.S. side, of course, rejected these ideas (Yoshida 2009, 28–29). 44.  Quoted in Sarantakes (1999). Los Angeles Times, June 23, 1960. 45.  Michael Auslin. “Japan Dissing.” Wall Street Journal Asia, April 22, 2010; available at www.aei.org/article/101953. 46.  “Hatoyama shi yokushiryoku hatsugen ‘houben’ mitomeru (Mr. Hatoyama Admits That the ‘Deterrence’ Statement Was an Excuse.” Asahi Shimbun, February 14, 2011; available at www.asahi.com/politics/update/0214/TKY201102140100.html. 47.  Presidents Rhee Syng-man and Park Chung-hee sometimes encouraged antiAmerican demonstrations when these served their interests, but the protests were issue specific and were never meant to hurt South Korea’s security relationship with the United States (Hong 1999; Oberdorfer 2001). 48.  For the modern history of South Korea and its democratization, see, for example, Diamond and Shin (1999); Oberdorfer (2001); and Kil and Moon (2001). For the effects of democratization on the alliance, see Dong Sun Lee (2007). 49.  For an analysis of anti-Americanism in Korea from various perspectives, see ­David Steinberg’s edited volume (2005). 50.  U.S. pressure, however, saved the life of Kim Dae-jung, the opposition leader who later became president. The U.S. government saved Kim’s life during Park Chunhee’s regime too. 51.  “Thereafter, further acts of violence perpetrated by Korean students against U.S. interests included another seizure of the US Cultural Center in Seoul (February 1988), a home-made bomb thrown into the US Embassy in Seoul by seven students (May 1988), a firebomb thrown into a KorAm Bank branch office (May 26, 1988), another firebomb incident at a US military residential area in Hannam-dong (November 1988), and occupation of the US ambassador’s living room by six students (October 6, 1989). The media

198

Notes to Chapter 6

provided detailed coverage of these incidents, which created quite a shock among Korean society overall” (Kang-ro Lee 2004). 52.  According to a report of the U.S. Congressional Research Service, however, the “intensity of the disputes has diminished considerably since the late 1980s and early 1990s, in part because South Korea has enacted a set of sweeping market-oriented reforms as a quid pro quo for receiving a US$58 billion package from the International Monetary Fund (IMF) following the near collapse of the South Korean economy in 1997” (Manyin 2006, 2). 53.  Global Indicators Database, Pew Research Center; available at http://pewglobal .org/database/?indicator=1&country=116&response=Favorable. 54. In Japan, the leftist opposition at times had approximately one-third of the Diet seats, and their neutralist alternative presented a serious threat to the U.S. policy makers. 55.  After Rhee was ousted as the result of protests and riots in April 1960, there was a brief period of parliamentary democracy before Park took power by coup d’état in May 1961. The U.S. government considered Prime Minister Chang Myon “to be a good man who wants to work with the Americans” (FRUS, 1958–1960, vol. 18, 691); “Memorandum of Conference With President Eisenhower,” Washington, DC, September 14, 1960. 56.  Rhee and Park, however, were able to stand firm against the United States exactly because there was no serious alternative to them. 57.  For Rhee’s bargaining style with the United States, see Suhrke (1973) and Hong (1999). 58.  Kim Dae-jung also had personal connections in Japan from the time of his political exile. The relationships between South Korea and Japan, which are often acrimonious, were good during Kim’s presidency. 59.  Most scholarly critiques of Roh’s policy, especially those in English, are from the perspective of proalliance groups. For a critique from a leftist perspective, see Cheong (2005). 60.  “Roh Hints at New East Asian Order,” Chosun Ilbo, March 22, 2005; available at http://english.chosun.com/w21data/html/news/200503/200503220024.html. 61. Bruce Klingner, “Conservative Landslide Marks New Era in South Korea.” December 20, 2007; available at www.heritage.org/Research/Reports/2007/12/ Conservative-Landslide-Marks-New-Era-in-South-Korea. 62.  As far as their personal security is concerned, however, they were not really in a safe position. There have been assassination attempts, both by North Korean agents and South Koreans (Oberdorfer 2001). 63.  Tong Kim. 2008. “Irony of Roh Moo-hyun.” The Korea Times. February 24; available at www.koreatimes.co.kr/www/news/opinon/2008/02/137_19498.html. 64. Norimitsu Onishi. “South Korea’s President Sags in Opinion Polls,” New York Times, November 27, 2006; available at www.nytimes.com/2006/11/27/world/ asia/27korea.html. 65. In May 2009, Roh Moo-hyun committed suicide after bribery allegations against him. His death led to a strong criticism of Lee Myung-bak because many South Koreans believed “that the prosecutors, conservative media” and the Lee administration “were complicit in the tragedy” (Moon 2010, 58). 66. Andy Jackson, “Lee Myung-bak’s Poll Numbers: Steady in a Time of Uncertainty.” December 30, 2010; available at http://asiancorrespondent.com/45042/



Notes to Chapter 6

199

lee-myung-baks-poll-numbers-steady-in-a-time-of-uncertainty/; and Andy Jackson, “Lee Myung-bak Ratings Tumble amid Inf lation Fears, Scandals.” February 26, 2011; available at http://asiancorrespondent.com/48975/lee-myung-bak-faces-ratings-tumbleamid-inf lation-fears-corruption-scandals/. 67.  Country Studies/Area Handbook Program, Federal Research Division of the Library of Congress; retrieved on May 25, 2015, from www.country-studies.com/southkorea/foreign-economic-relations.html. 68.  See Pagliano (1991); Easley (2006); and Reports on Allied Contributions to the Common Defense published by the U.S. Department of Defense. 69.  The troop level was still at 28,500 as of October 2013; see http://america .aljazeera.com/articles/2013/10/1/us-to-maintain-symbolicmilitarypresenceinkorea .html. 70.  As the Economist put it, the two leaders were at best “awkward bedfellows”; see www.economist.com/world/asia/displaystory.cfm?story_id=7887978. Also see Bechtol (2005) and Stueck (2012). 71.  Tong Kim. 2008. “Irony of Roh Moo-hyun.” The Korea Times. February 24; available at www.koreatimes.co.kr/www/news/opinon/2008/02/137_19498.html. 72. Won-sup Yoon. 2007. “Sending Troops to Iraq Was Historical Error: President.” Korea Times, November 11, 2007; available at http://gopkorea.blogs.com/ south_korean_politics/usf ksofa/. 73.  See Heo (2008) for security implications of the free trade agreement. 74.  See “Remarks by President Obama and President Lee Myung-bak of the Republic of Korea after Bilateral Meeting,” Toronto, June 26, 2010, the White House; available at www.whitehouse.gov/the-press-office/remarks-president-obama-andpresident-lee-myung-bak-republic-korea-after-bilateral-. 75.  For the U.S.–Spain bilateral relationships, see Cortada (1980); Rubottom and Murphy (1984); and Botero (2001), in addition to the works cited later in this section. 76.  Quoted in Chislett (2005b, 17). The 1953 base agreement is not coded as a military alliance in the ATOP dataset, and its extension in 1963, albeit with a rather limited military commitment, is the official beginning of the alliance. A Spanish foreign minister remarked that Spain was willing to collaborate in security affairs with the United States “the same, lesser, or even more . . . depending on the degree of security and protection that the United States would be in a position to offer,” but the relationship until 1976 was asymmetrical in that the Spanish legislature approved the base agreements as treaties while the United States treated them as executive agreements. See April 1970 folder, box 11, Department of State, Entry 5600, Bureau of European Affairs, 1963–1976, RG 59, Stack 150, Row 73, National Archives at College Park, MD, “NOTE” dated April 12, 1970, p. 4. 77.  See Cooley (2005, 2008) for similar cases of American “base politics.” For antiAmericanism in Spain, see Chislett (2005a). 78. On Spain’s democratization, see, for example, Carr and Fusi (1979); Preston (1986); Gunther, Montero, and Botella (2004); and Tusell (2007). 79.  Stanton (1993) examines the politicization of the NATO membership issue and how the indifferent Spanish public turned against it. 80.  Publicly, Spanish officials did not express desire to join NATO, but Franco actually desired an invitation. See, for example, Foreign Relations of the United States, 1958–1960, vol.7, part 2, Western Europe, 714–715. “Intelligence Report,” August 7, 1958.

200

Notes to Chapters 6 and 7

81.  On Spain’s entry into NATO, see Gil and Tulchin (1988). 82.  For example, see the following analysis in “Memorandum from the President’s Assistant for National Security Affairs (Kissinger) to President Nixon.” Washington, DC, March 17, 1969: “There have been definite indications that the Spanish have grown progressively nervous since the first of this year that the US might decide to reduce its presence drastically in Spain, or even pack up and go home. The original hard Spanish negotiating posture was basically associated with Foreign Minister Cas­ tiella; the progressive softening ref lects the apprehension of the conservative and powerful Spanish military, who do not want to risk losing their link to the United States” (FRUS, 1969–1976, vol. 41, 846). 83.  Spain had minor external threats to its colonies in North Africa, but Americans clearly refused to support Spain in this matter. 84.  When an ally was heavily dependent on the United States for security and had little to offer in return, the United States could take a critical stance toward the ally’s domestic policy, as with South Korea. The United States did not pressure Franco concerning his domestic policy because Spain was not so dependent on the United States and Spanish bases were valuable strategic assets for the United States. 85.  “Memorandum of Conversation.” Paris, December 16, 1975. 86.  Quoted in Viñas (2003, 19). 87.  On the role of King Juan Carlos in the transition, see Bernecker (1998). 88. Percent out of valid votes. Spain’s interior ministry; see www.infoelectoral .mir.es/. 89.  Although the domestic perspective of this chapter explains the timing of Aznar’s policy change, the systemic theory of Chapter Three illuminates the reasoning behind Aznar’s pro-U.S. attitude: “Aznar shares Bush’s unipolar vision of the world,” and he “decided to give an absolute priority to the bilateral relationship with the U.S.” (Iglesias-­Cavicchioli 2007, 3). Aznar’s Spain was entangled in the Iraq War, but “Aznar . . . thought of the Iraqi crisis as a unique opportunity to improve Spain’s relationship with the U.S. to the highest possible level” (ibid., 4). 90.  See the Country Studies/Area Handbook Program, Federal Research Division of the Library of Congress, available at www.country-data.com/cgi-bin/query/r-13104 .html. 91. The United States also accepted the Spanish request that U.S. submarines equipped with Poseidon missiles be withdrawn from the naval base in Rota, Spain. 92.  “Memorandum of Conversation.” Paris, December 16, 1975. 93.  Associated Press. “Spain’s Assembly Dissolved; New Vote to Be Held Oct. 28.” New York Times, August 28, 1982, Section 1; Page 3, Column 5. 94.  On the Spanish public opinion at the time, see Mujal-León (1983, 105). 95.  Rewarding loyalty is important as a signal, but it is also bad public relations to support a leader and still fail to save him or her. 96.  My argument, however, may be applicable to adversarial relations as well. Clare (2014) argues that dovish leaders have an advantage in negotiations with adversaries and supports his argument with statistical analysis of territorial disputes. Chapter 7 1.  According to this traditional view, alliances “are against, and only derivatively for, someone or something” (Liska 1962, 12).



Notes to Chapter 7

201

2.  See Brooks, Ikenberry, and Wohlforth (2012) for a case against retrenchment. 3.  Commenting on negotiations for revision of the U.S.–Japan Guidelines for Defense Cooperation, a U.S. official stated that their role is not just to contain the military rise of China but also to prevent escalation by Japan; “Taichuu senryaku, nichibei de ondosa [China-Policy, the US and Japan Different Degrees of Enthusiasm]” Asahi Shimbun, October 28, 2014. By tightening the U.S.-Japan alliance, the new guidelines of 2015 increased the risk of entanglement for both the United States and Japan. Rinehart (2015) points out that the “expansion of alliance cooperation to explicitly include ‘gray zone’ contingencies appears likely to make it nearly impossible for the United States to avoid entanglement in a Japan-China conf lict surrounding the Senkaku/Diaoyu Islands. Conversely, the Abe Administration’s decision to enable collective self-defense will facilitate Japan’s involvement in more U.S. conf licts, in more significant ways.” 4.  Japan’s Ministry of Foreign Affairs; available at www.mofa.go.jp/mofaj/area/usa/ yoron.html. 5.  Because there are many regional “poles” at the subsystemic level, we could perhaps apply the logic of the systemic model to other regional systems such as some parts of Africa. The application, however, will not be useful when outside inf luence is strong. 6.  On the rebalancing, see, for example, Ling (2013); Ratner (2013); and Tow and Stuart (2015). 7.  U.S. Defense Secretary Donald Rumsfeld visited Okinawa in November 2003. He and the Department of Defense realized that the continued use of the Futenma base was dangerous for the future of the alliance. A Marine Corps helicopter crashed into the campus of Okinawa International University near the Futenma base on August 13, 2004 (Hisae 2005, 83–86). Between the return of Okinawa to Japan in May 1972 and the end of 2011, Okinawa Prefecture had 522 accidents related to U.S. military aircrafts, including 43 crashes; see www.pref.okinawa.jp/site/chijiko/kichitai/documents/h24toukei-5 .pdf.

References

Abe, Shinzo. 2015. “Toward an Alliance of Hope.” Address to a Joint Meeting of the U.S. Congress. April 29. Available at http://japan.kantei.go.jp/97_abe/statement/201504/ uscongress.html. Acharya, Amitav. 2014. “Power Shift or Paradigm Shift? China’s Rise and Asia’s Emerging Security Order.” International Studies Quarterly 58 (1): 158–173. Adams, Simon, and Geoffrey Parker. 1997. “Europe and the Palatine War.” In The Thirty Years’ War, second edition, edited by Geoffrey Parker, 55–63. London: Routledge. Ahn, Byung-joon. 1998. The Origins and Evolution of the Korean–American Alliance: A Korean Perspective. Stanford, CA: Asia Pacific Research Center, Stanford University. Akiyama, Masahiro. 2002. Nichibei no senryakutaiwa ga hajimatta—anpo saiteigi no butaiura (Japan–U.S. Strategic Dialogue Has Begun: The Backstage of the Revision of the Security Treaty). Tokyo: Aki Shobou. Altfeld, Michael. 1984. “The Decision to Ally: A Theory and Test.” Western Political Quarterly 37 (4): 523–544. Anabtawi, Samir N. 1965. “Neutralists and Neutralism.” Journal of Politics 27 (2): 351–361. Andrew, Christopher, and Vasili Mitrokhin. 2005. The World Was Going Our Way: The KGB and the Battle for the Third World. New York: Basic Books. Arase, David. 2010. “Japan in 2009: A Historic Election Year.” Asian Survey 50 (1): 40–53. Baker, George, Robert Gibbons, and Kevin Murphy. 2002. “Relational Contracts and the Theory of the Firm.” Quarterly Journal of Economics 117 (1): 39–84. Ball, Christopher. 1998. “Nattering NATO Negativism? Reasons Why Expansion May Be a Good Thing.” Review of International Studies 24 (1): 43–67. Barnett, Michael, and Jack Levy. 1991. “Domestic Sources of Alliances and Alignments: The Case of Egypt, 1962–73.” International Organization 45 (3): 369–395. Basora, Adrian. 2009. “U.S.–Spain Relations from the Perspective of 2009.” CIDOB International Year Book 2009, Keys to Facilitate the Monitoring of the Spanish Foreign Policy and the International Relations in 2008. Barcelona: Barcelona Centre for International Affairs.

204 references Baxter, Richard R. 1980. “International Law in ‘Her Infinite Variety.’” International and Comparative Law Quarterly 29 (4): 549–566. Bearce, David, Kristen Flanagan, and Katharine Floros. 2006. “Alliances, Internal Information, and Military Conflict among Member-States.” International Organization 60 (3): 595–625. Bechtol, Bruce E. 2005. “The ROK–US Alliance during the Bush and Roh Administrations: Differing Perspectives and Their Implications for a Changing Strategic Environment.” International Journal of Korean Studies 9 (2): 87–115. ———. 2006. “Force Restructuring in the ROK–US Military Alliance: Challenges and Implications.” International Journal of Korean Studies 10 (2): 19–41. Beckley, Michael. 2011. “China’s Century? Why America’s Edge Will Endure.” International Security 36 (3): 41–78. Beckley, Michael. 2015. “The Myth of Entangling Alliances: Reassessing the Security Risks of U.S. Defense Pacts.” International Security 39 (4): 7–48. Belcher, Gerald. 1975. “Spain and the Anglo-Portuguese Alliance of 1661: A Reassessment of Charles II’s Foreign Policy at the Restoration.” Journal of British Studies 15 (1): 67–88. Bennett, D. Scott. 1997. “Testing Alternative Models of Alliance Duration, 1816–1984.” American Journal of Political Science 41 (3): 846–878. Bennett, D. Scott, and Allan Stam. 2000. “EUGene: A Conceptual Manual.” International Interactions 26 (2): 179–204. Berger, Thomas. 1998. Cultures of Antimilitarism: National Security in Germany and Japan. Baltimore: Johns Hopkins University Press. Bernecker, Walther L. 1998. “Monarchy and Democracy: The Political Role of King Juan Carlos in the Spanish Transicion” Journal of Contemporary History 33 (1): 65–84. Bhagwati, Jagdish. 1994. “The U.S.–Japan Rift: Samurais No More.” Foreign Affairs 73 (3): 7–12. Blackburn, Robert M. 1994. Mercenaries and Lyndon Johnson’s “More Flags”: The Hiring of Korean, Filipino, and Thai Soldiers in the Vietnam War. Jefferson, NC: McFarland. Blair, Dennis. 2008. “Military Power Projection in Asia.” In Strategic Asia 2008–09: Challenges and Choices, edited by Ashley J. Tellis, Mercy Kuo, and Andrew Marble, 391–420. Seattle, WA: National Bureau of Asian Research. Botero, Rodrigo. 2001. Ambivalent Embrace: America’s Troubled Relations with Spain from the Revolutionary War to the Cold War. West Port, CT: Greenwood Press. Boulding, Kenneth. 1958. “Theoretical Systems and Political Realities: A Review of Morton Kaplan, System and Process in International Politics.” Journal of Conflict Resolution 2 (4): 329–334. ———. 1962. Conflict and Defense: A General Theory. New York: Harper. Box-Steffensmeier, Janet, and Bradford Jones. 2004. Event History Modeling: A Guide for Social Scientists. New York: Cambridge University Press. Box-Steffensmeier, Janet, and Christopher Zorn. 2001. “Duration Models and Proportional Hazards in Political Science.” American Journal of Political Science 45 (4): 972–988. Boyer, Mark. 1993. International Cooperation and Public Goods: Opportunities for the Western Alliance. Baltimore: Johns Hopkins University Press. Brooks, Stephen, G. John Ikenberry, and William C. Wohlforth. 2012. “Don’t Come Home, America: The Case against Retrenchment.” International Security 37 (3): 7–51.

references 205 Brooks, Stephen G., and William C. Wohlforth. 2005. “Hard Times for Soft Balancing.” International Security 30 (1): 72–108. Brown, Harrold, with Joyce Winslow. 2012. Star Spangled Security: Applying Lessons Learned over Six Decades Safeguarding America. Washington, DC: Brookings Institution Press. Brown, Michael. 1995. “The Flawed Logic of Expansion.” Survival 37 (1): 34–52. Brown, Michael, Owen R. Cote Jr., Sean M. Lynn-Jones, and Steven Miller, eds. 2009. Primacy and Its Discontents: American Power and International Stability. Cambridge, MA: MIT Press. Buckley, Roger. 1992. US–Japan Alliance Diplomacy, 1945–1990. New York: Cambridge University Press. Bueno de Mesquita, Bruce. 1981. The War Trap. New Haven, CT: Yale University Press. Buszynski, Leszek. 2002. “Realism, Institutionalism, and Philippine Security.” Asian Survey 42 (3): 483–501. Buzan, Barry. 2004. The United States and the Great Powers: World Politics in the TwentyFirst Century. Cambridge, UK: Polity Press. Byman, Daniel L. 2006. “Remaking Alliances for the War on Terrorism.” Journal of Strategic Studies 29 (5): 767–811. Calder, Kent E. 2006. “Beneath the Eagle’s Wings? The Political Economy of Northeast Asian Burden-Sharing in Comparative Perspective.” Asian Security 2 (3): 148–173. ———. 2007. Embattled Garrisons: Comparative Base Politics and American Globalism. Princeton, NJ: Princeton University Press. Calvo-Gonzalez, Oscar. 2006. “Neither a Carrot nor a Stick: American Foreign Aid and Economic Policymaking in Spain during the 1950s.”Diplomatic History 30 (3): 409–438. Campbell, Kurt. 2004. “The End of Alliances? Not So Fast.” Washington Quarterly 27 (2): 151–163. Campbell, Kurt M., and Celeste Johnson Ward. 2003. “New Battle Stations?” Foreign Affairs 82 (5): 95–103. Cantalapiedra, David García. 2009. “Spain, Burden-Sharing, and NATO Deterrence Policy.” Strategic Insights 8 (4) (September). Monterey, CA: Center for Contemporary Conflict. Caron, David D. 1990. “Choice and Duty in Foreign Affairs: The Reflagging of the Kuwaiti Tankers.” In The Iran Iraq War: Lessons for Strategy, Diplomacy & Law, edited by Christopher Joyner, 153–172. Westport, CT: Greenwood Press. Carr, Raymond, and Juan Pablo Fusi. 1979. Spain: Dictatorship to Democracy. London: George Allen and Unwin. Catalinac, Amy. 2010. “Why New Zealand Took Itself out of ANZUS: Observing ‘Opposition for Autonomy’ in Asymmetric Alliances.” Foreign Policy Analysis 6 (4): 317–338. Caverley, Jonathan, and Ethan Kapstein. 2012. “Arms Away: How Washington Squandered Its Monopoly on Weapons Sales.” Foreign Affairs 91 (5): 125–132. Cha, Victor. 1999. Alignment Despite Antagonism: The United States–Korea–Japan Security Triangle. Stanford, CA: Stanford University Press. ———. 2010. “Powerplay: Origins of the U.S. Alliance System in Asia.” International Security 34 (3): 158–196. Cha, Victor, and Katrin Katz. 2011. “South Korea in 2010: Navigating New Heights in the Alliance.” Asian Survey 51 (1): 54–63.

206 references Chen, Dingding. 2014. “Is a China–South Korea Alliance Possible?” The Diplomat. July 8. Cheong, Wook-sik. 2005. Dongmeng ui Deot (The Trap of the Alliance). Seoul: Samin Press. Chislett, William. 2005a. “Anti-Americanism in Spain: The Weight of History.” Working Paper 47/2005, Real Instituto Elcano, Madrid. Available at www.realinstitutoelcano .org/wps/portal/rielcano_eng/Content?WCM_GLOBAL_CONTEXT=/Elcano_ in/Zonas_in/USA-Transatlantic+Dialogue/DT47-2005. ———. 2005b. Spain and the United States: The Quest for Mutual Rediscovery. Madrid: Real Instituto Elcano. Choi, Ajin. 2003. “The Power of Democratic Cooperation.” International Security 28 (1): 142–53. ———. 2004. “Democratic Synergy and Victory in War, 1816–1992.” International Studies Quarterly 48 (3): 663–682. Choi, Kang, and Joon-Sung Park. 2008 “South Korea: Fears of Abandonment and Entrapment.” In The Long Shadow: Nuclear Weapons and Security in 21st Century Asia, edited by Muthiah Alagappa, 374–403. Stanford, CA: Stanford University Press Choo, Jaewoo. 2008. “Mirroring North Korea’s Growing Economic Dependence on China: Political Ramifications.” Asian Survey 48 (2): 343–372. Christensen, Thomas. 1999. “China, the U.S.–Japan Alliance, and the Security Dilemma in East Asia.” International Security 23 (4): 49–80. ———. 2011. Worse Than a Monolith: Alliance Politics and Problems of Coercive Diplomacy in Asia. Princeton, NJ: Princeton University Press. ———. 2012. “The Meaning of the Nuclear Evolution: China’s Strategic Modernization and US–China Security Relations.” Journal of Strategic Studies 35 (4): 447–487. Christensen, Thomas, and Jack Snyder. 1990. “Chain Gangs and Passed Bucks: Predicting Alliance Patterns in Multipolarity.” International Organization 44 (2): 137–168. Chung, Jae Ho. 2005. America’s Views of China–South Korea Relations: Public Opinions and Elite Perceptions. Korean Journal of Defense Analysis 17 (1): 213–234. Clare, Joe. 2014. “Hawks, Doves, and International Cooperation.” Journal of Conflict Resolution 58 (7): 1311–1337. Clinton, Hillary. 2011. “America’s Pacific Century.” Foreign Policy, November, 56–63. Colman, Jonathan, and J. J. Widen. 2009. “The Johnson Administration and the Recruitment of Allies in Vietnam, 1964–1968.” History 94 (316): 483–504. Conybeare, John. 1992. “A Portfolio Diversification Model of Alliances: The Triple Alliance and Triple Entente, 1879–1914.” Journal of Conflict Resolution 36 (1): 53–85. ———. 1994. “The Portfolio Benefits of Free Riding in Military Alliances.” International Studies Quarterly 38 (3): 405–419. Conybeare, John, and Kim Dong-Hun. 2010. “Democracy, Institutionalization, and Corporate Alliances.” Journal of Conflict Resolution 54 (5): 715–744. Cooley, Alexander. 2005. “Base Politics.” Foreign Affairs, 84 (6): 79–92. ———. 2008. Base Politics: Democratic Change and the U.S. Military Overseas. Ithaca, NY: Cornell University Press. Cooley, Alexander, and Hendrik Spruyt. 2009. Contracting States: Sovereign Transfers in International Relations. Princeton, NJ: Princeton University Press. Cortada, James, ed. 1980. Spain in the Twentieth-Century World: Essays on Spanish Diplomacy, 1898–1978. Westport, CT: Greenwood Press. Crawford, Timothy W. 2008. “Wedge Strategy, Balancing, and the Deviant Case of Spain, 1940–41.” Security Studies 17 (1): 1–38.

references 207 ———. 2011. “Preventing Enemy Coalitions: How Wedge Strategies Shape Power Politics.” International Security 35 (4): 155–189. Cronin, Patrick. 1996. “The U.S.–Japan Alliance Redefined.” Strategic Forum, 75 (May). Cronin, Patrick, and Michael Green. 1994. “Redefining the U.S.-Japan Alliance: Tokyo’s National Defense Program.” McNair Paper 31. Washington, DC: Institute for National Strategic Studies. David, Steven R. 1991. “Explaining Third World Alignment.” World Politics 43 (2): 233–256. Davis, Christina. 2009. “Linkage Diplomacy: Economic and Security Bargaining in the Anglo–Japanese Alliance, 1902–23.” International Security 33 (3): 143–179. Dempsey, Judy. 2012. “The United States and China: The Return of a Bipolar World.” Published electronically, November 12. Available at http://carnegieeurope.eu/ strategiceurope/?fa=49969. DeScioli, Peter, and Robert Kurzban. 2009. “The Alliance Hypothesis for Human Friendship.” PLoS ONE, 4 (6): e5802. doi:10.1371/journal.pone.0005802. Deutsch, Karl W., and J. David Singer. 1964. “Multipolar Power Systems and International Stability.” World Politics 16 (3): 390–406. Diamond, Larry Jay, and Doh Chull Shin. 1999. Institutional Reform and Democratic Consolidation in Korea. Stanford, CA: Hoover Institution Press, Stanford University. Dingman, Roger. 1979. “Theories of, and Approaches to, Alliance Politics.” In Diplomacy: New Approaches in History, Theory, and Policy, edited by Paul Gordon Lauren, 245–266. New York: Free Press. Duffield, John. 1994. “NATO’s Functions after the Cold War.” Political Science Quarterly 109 (5): 763–787. Duncan, George, and Randolph Siverson. 1982. “Flexibility of Alliance Partner Choice in a Multipolar System: Models and Tests.” International Studies Quarterly 26 (4): 511–538. Dwyer, Jennifer Holt. 2000. “U.S.–Japan Financial-Market Relations in an Era of Global Finance.” In New Perspectives on U.S.–Japan Relations, edited by Gerald Curtis, 82–128. Tokyo: Japan Center for International Exchange. Dyson, Stephen Benedict. 2007. “Alliances, Domestic Politics, and Leader Psychology: Why Did Britain Stay out of Vietnam and Go into Iraq?” Political Psychology 28 (6): 647–666. Easley, Leif-Eric. 2006. “Forward-Deployed and Host Nation Interaction: U.S.–ROK Cooperation under External Threat and Internal Frictions.” Korean Journal of Defense Analysis 18 (4): 123–149. Eberstadt, Nicholas. 2011. The North Korean Economy: Between Crisis and Catastrophe. New Brunswick, NJ, and London: Transaction Publishers. Ehrhardt, George. 2004. “The Evolution of US–ROK Security Consultation.” Pacific Affairs 77 (4): 665–682. Evans, Peter, Harold Jacobson, and Robert Putnam, eds. 1993. Double-Edged Diplomacy: International Bargaining and Domestic Politics. Berkeley: University of California Press. Eyal, Jonathan. 1997. “NATO’s Enlargement: Anatomy of a Decision.” International Affairs 73 (4): 695–719. Fang, Songying, Jesse Johnson, and Brett Ashley Leeds. 2014. “To Concede or to Resist? The Restraining Effect of Military Alliances.” International Organization 68 (4): 775–809.

208 references Farley, Miriam. 1955. “Mr. Shigemitsu in Washington.” Far Eastern Survey 24 (9): 140–143. Fawn, Rick. 2006. “Alliance Behaviour, the Absentee Liberator and the Influence of Soft Power: Post-Communist State Positions over the Iraq War in 2003.” Cambridge Review of International Affairs 19 (3): 465–480. Fearon, James. 1994. “Domestic Political Audiences and the Escalation of International Disputes.” American Political Science Review 88 (3): 577–592. ———. 1995. “Rationalist Explanations for War.” International Organization 49 (3): 379–414. Fedder, Edwin H. 1968. “The Concept of Alliance.” International Studies Quarterly 12 (1): 65–86. Finnegan, Michael. 2010. “Return to Basics: Recalibrating the U.S.–Japan Security Alliance.” Asia Policy 10: 10–15. Fisher, Richard. 1999. “Rebuilding the U.S.–Philippine Alliance.” In Heritage Foundation Backgrounder, No.1255: 1–12. Washington, DC: Heritage Foundation. Fordham, Benjamin. 2008. “Power or Plenty? Economic Interests, Security Concerns, and American Intervention.” International Studies Quarterly 52 (4): 737–758. ———. 2010. “Trade and Asymmetric Alliances.” Journal of Peace Research 48 (2): 685–696. Foreign Relations of the United States [FRUS], 1955–1957, vol. 23, part 1, Japan. Washington, DC: Government Printing Office. Foreign Relations of the United States [FRUS], 1958–1960, vol. 7, part 2, Western Europe. Washington, DC: Government Printing Office. Foreign Relations of the United States [FRUS], 1958–1960, vol. 18, Japan; Korea. Washington, DC: Government Printing Office. Foreign Relations of the United States [FRUS], 1969–1976, vol.41, Western Europe; NATO, 1969–1972. Washington, DC: Government Printing Office. Foreign Relations of the United States [FRUS], 1969–1976, vol. E-15, part 2. Documents on Western Europe, 1973–1976. Washington, DC: Government Printing Office. Fravel, M. Taylor. 2010. “International Relations Theory and China’s Rise: Assessing China’s Potential for Territorial Expansion.” International Studies Review 12 (4): 505–532. Friedberg, Aaron. 1993. “Ripe for Rivalry: Prospects for Peace in a Multipolar Asia.” International Security 18 (3): 5–33. ———. 2011. A Contest for Supremacy: China, America, and the Struggle for Mastery in Asia. New York: W. W. Norton. Fujimoto, Kazumi. 1999. “Nichibei shunou kaidan to seiji katei: 1981nen 5gatsu no Suzuki–­Reagan kaidan wo megutte [Japan–US Summit Meetings and Political Process: Over the May 1981 Suzuki–Reagan meeting].” Senshu daigaku hougaku kenkyusho kiyou. 24: 3–50. Gaddis, John Lewis. 2006. The Cold War: A New History, reprint edition. East Rutherford, NJ: Penguin Group USA. Ganguli, Kisari Mohan, trans. 2009. The Mahabharata of Krishna-dwaipayana Vyasa, vol. 12 Charleston, SC: BiblioBazaar, LLC. Gartzke, Erik, and Kristian Skrede Gleditsch. 2004. “Why Democracies May Actually Be Less Reliable Allies.” American Journal of Political Science 48 (4): 775–795. Gaubatz, Kurt Taylor. 1996. “Democratic States and Commitment in International Relations.” International Organization 50 (1): 109–139.



references 209

Gavin, Francis J. 2002. “The Gold Battles within the Cold War: American Monetary Policy and the Defense of Europe, 1960–1963.” Diplomatic History 26 (1): 61–94. ———. 2003. “Ideas, Power, and the Politics of U.S. International Monetary Policy during the 1960s.” In Monetary Orders: Ambiguous Economics, Ubiquitous Politics, edited by Jonathan Kirshner, 195–217. Ithaca, NY: Cornell University Press. ———. 2004. Gold, Dollars, and Power: The Politics of International Monetary Relations, 1958–1971. Chapel Hill: University of North Carolina Press. Gelpi, Christopher F. 1999. “Alliances as Instruments of Intra-Allied Control.” In Imperfect Unions: Security Institutions over Time and Space, edited by Helga Haftendorn, Robert Keohane and Celeste Wallander. Oxford, UK: Oxford University Press. George, Alexander. 1979. “Case Studies and Theory Development: The Method of Structured, Focused Comparison.” In Diplomacy: New Approaches in History, Theory, and Policy, edited by Paul Gordon Lauren, 43–68. New York: Free Press. George, Alexander, and Andrew Bennett. 2004. Case Studies and Theory Development in the Social Sciences. Cambridge, MA: MIT Press. Gholz, Eugene, Daryl G. Press, and Harvey M. Sapolsky. 1997. “Come Home, America: The Strategy of Restraint in the Face of Temptation.” International Security 21 (4): 5–48. Gibbs, David. 2009. First Do No Harm: Humanitarian Intervention and the Destruction of Yugoslavia. Nashville, TN: Vanderbilt University Press. Gibler, Douglas. 1999. “An Extension of the Correlates of War Formal Alliance Data Set, 1648–1815.” International Interactions 25 (1): 1–28. ———. 2007. “Bordering on Peace: Democracy, Territorial Issues, and Conflict.” International Studies Quarterly 51 (3): 509–532. ———. 2008. “The Costs of Reneging: Reputation and Alliance Formation.” Journal of Conflict Resolution 52 (3): 426–454. ———. 2009. International Military Alliances, 1648–2008. Washington, DC: CQ Press. Gibler, Douglas, and Scott Wolford. 2006. “Alliances, then Democracy: An Examination of the Relationship between Regime Type and Alliance Formation.” Journal of Conflict Resolution 50 (1): 129–153. Gil, Federico, and Joseph Tulchin, eds. 1988. Spain’s Entry into NATO: Conflicting Political and Strategic Perspectives. Boulder, CO, and London: Lynne Rienner Publishers. Gilpin, Robert. 1981. War and Change in World Politics. Cambridge, UK: Cambridge University Press. Goertz, Gary, Douglas Gibler, and Kathy Powers. 2008. “Reconceptualizing Alliances: How Treaty Nesting Can Fundamentally Change the Meaning of a Concept.” Working Paper. Available at http://bama.ua.edu/~dmgibler/pubs/Goertz.Gibler. Powers.012008.pdf. Goh, Evelyn. 2011. “Japan, China, and the Great Power Bargain.” EAI Fellows Program Working Paper Series No. 32. Seoul: East Asia Institute. Goldgeier, James. 1998. “NATO Expansion: The Anatomy of a Decision.” Washington Quarterly 21 (1): 85–102. ———. 1999. Not Whether but When: The U.S. Decision to Enlarge NATO. Washington, DC: Brookings Institution Press. Goldstein, Avery. 1995. “Discounting the Free Ride: Alliances and Security in the Postwar World.” International Organization 49 (1): 39–71. Gourevitch, Peter. 1999. “The Governance Problem in International Relations.” In

210 references Strategic Choice and International Relations, edited by David Lake and Robert Powell, 137–164. Princeton, NJ: Princeton University Press. Gowa, Joanne. 1989. “Bipolarity, Multipolarity, and Free Trade.” American Political Science Review 83 (4): 1245–1256. ———. 1994. Allies, Adversaries, and International Trade. Princeton, NJ: Princeton University Press. Gowa, Joanne, and Edward D. Mansfield. 1993. “Power Politics and International Trade.” American Political Science Review 87 (2): 408–420. ———. 2004. “Alliances, Imperfect Markets, and Major-Power Trade.” International Organization 58 (4): 775–805. Green, Michael. 2010a. “Redefining and Reaffirming the U.S.–Japan Alliance.” Asia Policy 10: 16–20. ———. 2010b. “US–Japan Relations: New Realism.” Comparative Connections, July 2010. Available at http://csis.org/files/publication/1002qus_japan.pdf. Green, Michael, and Richard Samuels. 1994. “Recalculating Autonomy: Japan’s Choices in the New World Order.” In Analysis, 5–22. Seattle, WA: National Bureau of Asian Research. Grieco, Joseph, and John Ikenberry. 2003. State Power and World Markets: The International Political Economy. New York: Norton. Grossman, Sanford, and Oliver Hart. 1986. “The Costs and Benefits of Ownership: A Theory of Vertical and Lateral Integration.” Journal of Political Economy 94 (4): 691–719. Gulick, Edward Vose. (1955) 1967. Europe’s Classical Balance of Power. New York: Norton. Gunther, Richard, José Ramón Montero, and Joan Botella. 2004. Democracy in Modern Spain. New Haven, CT: Yale University Press. Haas, Ernst. 1953. “The Balance of Power: Prescription, Concept, or Propaganda?” World Politics 5 (4): 442–477. Haass, Richard. 2008. “The Age of Nonpolarity.” Foreign Affairs 87 (3): 44–56. Haftendorn, Helga, Robert Keohane, and Celeste Wallander, eds. 1999. Imperfect Unions: Security Institutions over Time and Space. New York: Oxford University Press. Hagström, Linus. 2012. “‘Power Shift’ in East Asia? A Critical Reappraisal of Narratives on the Diaoyu/Senkaku Islands Incident in 2010.” The Chinese Journal of International Politics 5 (3): 267–297. Han, Sukhee. 2014. “China’s Charm Offensive to Korea: A New Approach to Extend the Strategic Buffer.” The Asan Forum 2 (4). Published electronically June 13 at www .theasanforum.org/chinas-charm-offensive-to-korea-a-new-approach-to-extendthe-strategic-buffer/. Hansen, Birthe. 2000. “The Unipolar World Order and Its Dynamics.” In The New World Order, edited by Birthe Hansen and Bertel Heurlin, 112–133. London: Macmillan. Hara, Yoshihisa. 1991. Nichi-bei kankei no kozu: Ampo kaitei o kensho suru (The Design of the United States–Japan Relationship: An Examination of the Security Treaty Revision of 1960). Tokyo: Nihon Hoso Shuppan Kyokai. Harrison, Hope M. 2003. Driving the Soviets up the Wall: Soviet–East German Relations, 1953–1961. Princeton, NJ: Princeton University Press. Hart, Oliver. 1995. Firms, Contracts and Financial Structure. New York: Oxford University Press. Hart, Oliver, and John Moore. 1988. “Incomplete Contracts and Renegotiation.” Econometrica, 56 (4): 755–785.

references 211 Hartley, Keith. 2006. “Defense Industrial Policy in a Military Alliance.” Journal of Peace Research 43 (4): 473–489. Hata Masaki and Takeshi Iida. 2015. “Survey Experiment on Japanese Attitudes toward the Cabinet Decision on Collective Self-Defense.” Unpublished manuscript, version April 2. Havens, Thomas. 1987. Fire across the Sea: The Vietnam War and Japan l965–l975. Princeton, NJ: Princeton University Press. Heginbotham, Eric, Ely Ratner, and Richard Samuels. 2011. “Tokyo’s Transformation.” Foreign Affairs 90 (5): 138–148. Heo, Uk. 2008. “The US–ROK Alliance: Security Implications of the South Korea–US Free Trade Agreement.” Pacific Focus 33 (3): 365–382. Hirschman, Albert O. (1945) 1980. National Power and the Structure of Foreign Trade. Reprint. Berkeley: University of California Press. Hisae, Masahiko. 2005. Beigun Saihen: Nichi-Bei “Himitsu Kosho” de Naniga Attaka (Reorganization of U.S. Forces: What Happened in the U.S–Japan “Secret Negotiations”?). ­Tokyo: Kodansha. Holsti, Ole, Terrence Hopmann, and John Sullivan. 1973. Unity and Disintegration in International Alliances. New York: John Wiley & Sons. Hong, Yong-Pyo. 1999. State Security and Regime Security: President Syngman Rhee and the Insecurity Dilemma in South Korea, 1953–60. New York: Palgrave Macmillan. Hosokawa, Morihiro. 1998. “Are U.S. Troops in Japan Needed? Reforming the Alliance.” Foreign Affairs 77 (4): 2–5. Howell, William, and Jon Pevehouse. 2007. While Dangers Gather: Congressional Checks on Presidential War Powers. Princeton, NJ: Princeton University Press. Hudson, John. 2015. “White House Rejected Defense Treaty Proposal Ahead of Gulf Summit.” Foreign Policy, May 11. Available at http://foreignpolicy.com/2015/05/11/ white-house-rejected-defense-treaty-proposal-ahead-of-gulf-summit/. Hwang, Jaeho. 2014. “The ROK’s China Policy under Park Geun-hye: A New Model of ROK–PRC Relations.” CEAP Visiting Fellow Working Paper. Washington, DC: The Brookings Institution. Igarashi, Takeshi. 1986. Tainichi kowa to reisen—sengo nichi-bei kankei no keisei (The Japanese Peace Treaty and the Cold War—The Formation of Postwar U.S.–Japan Relations). Tokyo: Tokyo University Press. Iglesias-Cavicchioli, Manuel. 2007. “A Period of Turbulent Change: Spanish–US Relations since 2002.” The Whitehead Journal of Diplomacy and International Relations 8 (2): 1–17. Ikenberry, G. John, ed. 2002. America Unrivaled: The Future of the Balance of Power. Ithaca, NY: Cornell University Press. ———. 2011. Liberal Leviathan: The Origins, Crisis, and Transformation of the American World Order. Princeton, NJ: Princeton University Press. Ikenberry, G. John, Michael Mastanduno, and William C. Wohlforth. 2009. “Introduction: Unipolarity, State Behavior, and Systemic Consequences.” World Politics 61 (1): 1–27. Im, Koonsam. 2015. “Korean Exchange Rate and FTAs under the Roh Moo-Hyun Administration.” International Relations of the Asia-Pacific 15 (2): 367–396. Ishikawa, Masumi, and Jiro Yamaguchi. 2010. Sengo Seijishi (Postwar Political History), third edition. Tokyo: Iwanami Shoten.

212 references Ito, Go. 2003. Alliance in Anxiety: Détente and the Sino–American-Japanese Triangle. New York: Routledge. Izumikawa, Yasuhiro. 2010. “Explaining Japanese Antimilitarism: Normative and Realist Constraints on Japan’s Security Policy.” International Security 35 (2): 123–160. ———. 2013. “To Coerce or Reward? Theorizing Wedge Strategies in Alliance Politics.” Security Studies 22 (3): 498–531. ———. 2015. “Binding Strategy and Alliance Cohesion: Explaining the Soviet–JapaneseU.S. Diplomatic Tug of War in the Mid-1950s.” Unpublished manuscript. Jervis, Robert. 2009. “Unipolarity: A Structural Perspective.” World Politics 61 (1): 188–213. Johnson, Jesse C., and Brett Ashley Leeds. 2011. “Defense Pacts: A Prescription for Peace?” Foreign Policy Analysis 7 (1): 45–65. Johnston, Alastair Iain. 2013. “How New and Assertive Is China’s New Assertiveness?” International Security 37 (4): 7–48. Jones, Seth. 2007. The Rise of European Security Cooperation. New York: Cambridge University Press. Kan, Shirley, and Larry Niksch. 2010. “Guam: U.S. Defense Deployments.” CRS Report for Congress. Washington, DC: Congressional Research Service, 7-5700, RS22570. Kane, Tim. 2006. “Global U.S. Troop Deployment, 1950–2005.” Washington, DC: The Heritage Foundation, Center for Data Analysis. Available at www.heritage.org/ research/reports/2006/05/global-us-troop-deployment-1950-2005. Kaplan, Morton. 1957. System and Process in International Relations. New York: Wiley. Kataoka, Tetsuya, and Ramon Myers. 1989. Defending an Economic Superpower: Reassessing the U.S.–Japan Security Alliance. Boulder, CO: Westview Press. Katzenstein, Peter. 2008. Rethinking Japanese Security: Internal and External Dimensions. London and New York: Routledge. Keele, Luke. 2010. “Proportionally Difficult: Testing for Nonproportional Hazards in Cox Models.” Political Analysis 18 (2): 189–205. Keohane, Robert. 1971. “The Big Influence of Small Allies.” Foreign Policy, 2, 161–182. ———. 1984. After Hegemony: Cooperation and Discord in the World Political Economy. Princeton, NJ: Princeton University Press. Keohane, Robert, and Joseph S. Nye. 1977. Power and Interdependence: World Politics in Transition. Boston, MA: Little, Brown. Kil, Soong Hoom, and Chung-In Moon. 2001. Understanding Korean Politics: An Introduction. Albany: State University of New York Press. Kim, Choong Nam. 2005. “The Roh Moo Hyun Government’s Policy toward North Korea.” International Journal of Korean Studies 9 (2): 1–34. Kim, Tongfi. 2011. “Why Alliances Entangle but Seldom Entrap States.” Security Studies 20 (3): 350–377. ———. 2014a. “The Role of Leaders in Intra-Alliance Bargaining.” Asian Security 10 (1): 47–69. ———. 2014b. “South Korea’s Middle-Power Response to the Rise of China.” In Middle Powers and the Rise of China, edited by Bruce Gilley and Andrew O’Neil, 84–103. Washington, DC: Georgetown University Press. Kim, Tongfi, and Jennifer Dabbs Sciubba. 2015. “The Effect of Age Structure on the Abrogation of Military Alliances.” International Interactions 41 (2): 279–308. Kindleberger, Charles. 1973. The World in Depression, 1929–39. Berkeley: University of California Press.

references 213 Knorr, Klaus. 1956. The War Potential of Nations. Princeton, NJ: Princeton University Press. Koremenos, Barbara, Charles Lipson, and Duncan Snidal. 2001. “The Rational Design of International Institutions.” International Organization 55 (4): 761–799. Kornbluh, Peter. 1998. National Security Archive Electronic Briefing Book No. 8, “Chile and the United States: Declassified Documents Relating to the Military Coup, September 11, 1973,” on National Security Archive Web site at www2.gwu.edu/~nsarchiv/ NSAEBB/NSAEBB8/nsaebb8i.htm. Krebs, Ronald R. 1999. “Perverse Institutionalism: NATO and the Greco–Turkish Conflict.” International Organization 53 (2): 343–377. Kupchan, Charles. 1988. “NATO and the Persian Gulf: Examining Intra-Alliance Behavior.” International Organization 42 (2): 317–346. Kydd, Andrew. 2001. “Trust Building, Trust Breaking: The Dilemma of NATO Enlargement.” International Organization 55 (4): 801–828. Lai, Brian, and Dan Reiter. 2000. “Democracy, Political Similarity, and International Alliances, 1816–1992.” Journal of Conflict Resolution 44 (2): 203–227. Lake, David. 1999. Entangling Relations: American Foreign Policy in Its Century. Princeton, NJ: Princeton University Press. ———. 2001. “Beyond Anarchy: The Importance of Security Institutions.” International Security 26 (1): 129–160. ———. 2007. “Escape from the State of Nature: Authority and Hierarchy in World Politics.” International Security 32 (1): 47–79. ———. 2009. Hierarchy in International Relations. Ithaca, NY: Cornell University Press. Larres, Klaus. 2002. “North Atlantic Treaty Organization.” In Encyclopedia of American Foreign Relations: Studies of the Principal Movements and Ideas, revised and expanded second edition, vol. 2, edited by Alexander DeConde, Richard Dean Burns, and Fredrik Logevall, 573–593. New York: Scribner’s Sons. Layne, Christopher. 1993. “The Unipolar Illusion: Why New Great Powers Will Rise.” International Security 17 (4): 5–51. Lee, Dong Sun. 2007. “Democratization and the US–South Korean Alliance.” Journal of East Asian Studies 7 (3): 469–499. Lee, Gerald Geunwook. 2006. “South Korea’s Faustian Attitude: The Republic of Korea’s Decision to Send Troops to Iraq Revisited.” Cambridge Review of International Affairs 19 (3): 481–493. Lee, Kang-ro. 2004. “Critical Analysis of Anti-Americanism in Korea.” Korean Journal of International Relations 44 (4): 74–98. Lee, Sook-Jong. 2005. “Allying with the United States: Changing South Korean Attitudes.” Korean Journal of Defense Analysis 17 (1): 81–104. Leeds, Brett Ashley. 2003a. “Alliance Reliability in Times of War: Explaining State Decisions to Violate Treaties.” International Organization 57 (4): 801–827. ———. 2003b. “Do Alliances Deter Aggression? The Influence of Military Alliances on the Initiation of Militarized Interstate Disputes.” American Journal of Political Science 47 (3): 427–439. Leeds, Brett Ashley. 2005. “ATOP Codebook version 3.0.” Available at http://atop.rice .edu/download/ATOPcdbk.pdf. Leeds, Brett Ashley, and Sezi Anac. 2005. “Alliance Institutionalization and Alliance Performance.” International Interactions 31 (3): 183–202.

214 references Leeds, Brett Ashley, and David Davis. 1997. “Domestic Political Vulnerability and International Disputes.” Journal of Conflict Resolution 41 (6): 814–834. Leeds, Brett Ashley, Andrew Long, and Sara McLaughlin Mitchell. 2000. “Reevaluating Alliance Reliability: Specific Threats, Specific Promises.” Journal of Conflict Resolution 44 (5): 686–699. Leeds, Brett Ashley, and Michaela Mattes. 2007. “Alliance Politics during the Cold War: Aberration, New World Order, or Continuation of History?” Conflict Management and Peace Science 24 (3): 183–199. Leeds, Brett Ashley, Michaela Mattes, and Jeremy S. Vogel. 2009. “Interests, Institutions, and the Reliability of International Commitments.” American Journal of Political Science 53 (2): 461–476. Leeds, Brett Ashley, Jeffrey M. Ritter, Sara McLaughlin Mitchell, and Andrew G. Long. 2002. “Alliance Treaty Obligations and Provisions, 1815–1944.” International Interactions 28 (3): 237–260. Leeds, Brett Ashley, and Burcu Savun. 2007. “Terminating Alliances: Why Do States Abrogate Agreements?” Journal of Politics 69 (4): 1118–1132. Leonard, Thomas M. 2006. Encyclopedia of the Developing World. New York: Taylor & Francis. Leverett, Flynt. 2006. “Dealing with Tehran: Assessing U.S. Diplomatic Options Toward Iran.” New York: The Century Foundation. Available at http://tcf.org/publications/ pdfs/pb595/leverett_diplomatic.pdf. Levkowitz, Alon. 2008. “The Seventh Withdrawal: Has the US Forces’ Journey Back Home from Korea Begun?” International Relations of the Asia-Pacific 8 (2): 131–148. Levy, Jack, and Michael Barnett. 1992. “Alliance Formation, Domestic Political Economy and Third World Security.” Jerusalem Journal of International Relations. 14 (4): 19–40. Licht, Amanda. 2010. “Coming into Money: The Impact of Foreign Aid on Leader Survival.” Journal of Conflict Resolution 54 (1): 58–87. ———. 2011. “Change Comes with Time: Substantive Interpretation of Nonproportional Hazards in Event History Analysis.” Political Analysis 19 (2): 227–243. Lieber, Keir, and Gerard Alexander. 2005. “Waiting for Balancing: Why the World Is Not Pushing Back.” International Security 30 (1): 109–139. Ling, Wei. 2013. “Rebalancing or De-Balancing: U.S. Pivot and East Asian Order.” American Foreign Policy Interests 35 (3): 148–154. Lipson, Charles. 1984. “International Cooperation in Economic and Security Affairs.” World Politics 37 (1): 1–23. Liska, George. 1962. Nations in Alliance: The Limits of Interdependence. Baltimore, MD: Johns Hopkins University Press. Long, Andrew. 2003. “Defense Pacts and International Trade.” Journal of Peace Research 40 (5): 537–552. Long, Andrew, and Brett Ashley Leeds. 2006. “Trading for Security: Military Alliances and Economic Agreements.” Journal of Peace Research 43 (4): 433–451. Long, Andrew, Timothy Nordstrom, and Kyeonghi Baek. 2007. “Allying for Peace: Treaty Obligations and Conflict between Allies.” Journal of Politics 69 (4): 1103–1117. Ludwig, Paul W. 2010. “Without Foundations: Plato’s Lysis and Postmodern Friendship.” American Political Science Review 104 (1): 134–150. Mandelbaum, Michael. 1981. The Nuclear Revolution: International Politics before and after Hiroshima. Cambridge, UK, and New York: Cambridge University Press.

references 215 ———. 1996. The Dawn of Peace in Europe. New York: Twentieth Century Fund Press. Mansfield, Edward. 1993. “Concentration, Polarity, and the Distribution of Power.” International Studies Quarterly 37 (1): 105–128. Mansfield, Edward, and Rachel Bronson. 1997. “Alliances, Preferential Trading Arrangements, and International Trade.” American Political Science Review 91 (1): 94–107. Manyin, Mark. 2006. “South Korea–U.S. Economic Relations: Cooperation, Friction, and Prospects for a Free Trade Agreement (FTA).” Congressional Research Service Report for Congress RL30566. Manyin, Mark, Mary Beth Nikitin, Emma Chanlett-Avery, Ian E. Rinehart, and William Cooper. 2013. “U.S.–South Korea Relations” Congressional Research Service Report for Congress, R41481. Maoz, Zeev. 1996. Domestic Sources of Global Change. Ann Arbor: University of Michigan Press. Marshall, Monty, and Keith Jaggers. 2002. Polity IV Project: Dataset Users’ Manual. College Park: University of Maryland. Available at www.systemicpeace.org/polity/ polity4.htm. Mastanduno, Michael. 1997. “Preserving the Unipolar Moment: Realist Theories and U.S. Grand Strategy after the Cold War.” International Security 21 (4): 49–88. ———. 2009. “System Maker and Privilege Taker: U.S. Power and the International Political Economy.” World Politics 61 (1): 121–54. Mattes, Michaela, and Greg Vonnahme. 2010. “Contracting for Peace: Do Nonaggression Pacts Reduce Conflict?” The Journal of Politics 72 (4): 925–938. Maurer, Noel. 2011. “Much Ado about Nothing: Expropriation and Compensation in Peru and Venezuela, 1968–75.” Harvard Business School Working Paper No. 11-097, March. McCalla, Robert. 1996. “NATO’s Persistence after the Cold War.” International Organization 50 (3): 445–475. McDevitt, Michael. 2004. “U.S. Korea Security Alliance in Transition: A U.S. Perspective.” International Journal of Korean Studies 8 (1): 1–22. McGillivray, Fiona, and Alastair Smith. 2008. Punishing the Prince: A Theory of Interstate Relations, Political Institutions, and Leader Change. Princeton, NJ: Princeton University Press. McLeary. Paul. 2015. “Japan and U.S. Double Down on Defense Ties.” April 27, Foreign Policy. Mearsheimer, John. 1990. “Back to the Future: Instability in Europe after the Cold War.” International Security 15 (4): 5–56. ———. 1994. “The False Promise of International Institutions.” International Security 19 (3): 5–49. ———. 2010. “The Gathering Storm: China’s Challenge to US Power in Asia.” The Chinese Journal of International Politics 3 (4): 381–396. Meernik, James. 2008. “A Supply and Demand Theory of US Military Policy.” Conflict Management and Peaces Science 25 (1): 33–48. Menon, Rajan. 2007. The End of Alliances. New York: Oxford University Press. Midford, Paul. 2008. “Japanese Mass Opinion toward the War on Terrorism.” In Japanese Public Opinion and the War on Terrorism, edited by Robert D. Eldridge and Paul Midford, 11–42. New York: Palgrave Macmillan. Miller, Benjamin. 1995. When Opponents Cooperate: Great Power Conflict and Collaboration in World Politics. Ann Arbor: University of Michigan Press.

216 references Miller, Gregory. 2003. “Hypotheses on Reputation: Alliance Choices and the Shadow of the Past.” Security Studies 12 (3): 40–78. ———. 2011. The Shadow of the Past: Reputation and Military Alliances before the First World War. Ithaca, NY: Cornell University Press. Miyahara,Yasufumi. 2014. “Nippon wa naze nichiei doumei wo jizokusasetanoka: kokusai seijigaku no kasetsu wo enyoushite [The Sustainment of the Anglo-Japanese Alliance by Japan: An International Political Science Approach]. Military History Studies Annual, 17: 1–27. Modelski, George.1974. World Power Concentration. Morristown, NJ: General Learning Press. Monteiro, Nuno. 2014. Theory of Unipolar Politics. Cambridge, UK: Cambridge University Press. Moon, Chung-in. 2010. “South Korea in 2009: From Setbacks to Reversal.” Asian Survey 50 (1): 56–64. Moore, Gregory J. 2014. “‘In Your Face’: Domestic Politics, Nationalism, and ‘Face’ in the Sino–Japanese Islands Dispute.” Asian Perspective 38 (2): 219–240. Moravcsik, Andrew. 1993. “Introduction: Integrating International and Domestic Theories of International Bargaining.” In Double-Edged Diplomacy: International Bargaining and Domestic Politics, edited by Peter Evans, Harold Jacobson, and Robert Putnam, 3–42. Berkeley: University of California Press. Morris, Ivan. 1960. “Japanese Foreign Policy and Neutralism.” International Affairs 36 (1): 7–20. Morrow, James. 1987. “On the Theoretical Basis of a Measure of National Risk Attitudes.” International Studies Quarterly 31 (4): 423–438. ———. 1991. “Alliances and Asymmetry: An Alternative to the Capability Aggregation Model of Alliances.” American Journal of Political Science 35 (4): 904–933. ———. 1993. “Arms Versus Allies: Trade-Offs in the Search for Security.” International Organization 47 (2): 207–233. ———. 2000. “Alliances: Why Write Them Down?” Annual Review of Political Science 3, 63–83. Mowle, Thomas, and David Sacko. 2007. “Global NATO: Bandwagoning in a Unipolar World.” Contemporary Security Policy 28 (3): 597–618. Mueller, John. 2004. The Remnants of War. Ithaca, NY: Cornell University Press. Mujal-León, Eusebio. 1983. “Rei(g)ning in Spain.” Foreign Policy, 51, 101–117. Murata, Koji. 1998a. Daitouryou no Zasetsu Carter Seiken no Zaikan Beigun Tettai Seisaku (President Carter’s Troop Withdrawal Policy from South Korea). Tokyo: Yuhikaku. ———. 1998b. The Origins and Evolution of the Korean–American Alliance: A Japanese Perspective. Stanford, CA: Asia Pacific Research Center, Stanford University. Narizny, Kevin. 2003a. “Both Guns and Butter, or Neither: Class Interests in the Political Economy of Rearmament.” American Political Science Review 97 (2): 203–220. ———. 2003b. “The Political Economy of Alignment: Britain’s Commitment to Europe, 1905–1939.” International Security 27 (4): 184–219. Newnham, Randall. 2008. “‘Coalition of the Bribed and Bullied?’ U.S. Economic Linkage and the Iraq War Coalition.” International Studies Perspectives 9 (2): 183–200. Nish, Ian. 1972. Alliance in Decline. London: Athlone Press. Nishihara, Masashi, and Jitsuo Tsuchiyama, eds. 1998. Nichibei Doumei Q & A 100 (The Japan–U.S. Alliance Q & A 100). Tokyo: Aki Shobou.

references 217 Norrlof, Carla. 2010. America’s Global Advantage: US Hegemony and International Cooperation. Cambridge, UK: Cambridge University Press. Nye, Joseph. 1990. Bound to Lead: The Changing Nature of American Power. New York: Basic Books. ———. 1995. “The Case for Deep Engagement.” Foreign Affairs 74 (4): 90–102. Oberdorfer, Don. 2001. The Two Koreas: A Contemporary History. Revised and updated edition. New York: Basic Books. Office of the Secretary of Defense. 2013. “Military and Security Developments Involving the People’s Republic of China 2013.” Annual Report to Congress. Available at www.defense.gov/Portals/1/Documents/pubs/2013_China_Report_FINAL.pdf. Okamoto, Yukio. 2002. “Japan and the United States: The Essential Alliance.” Washington Quarterly 25 (2): 59–72. Olson, Mancur, and Richard Zeckhauser. 1966. “An Economic Theory of Alliances.” Review of Economics and Statistics 48: 266–279. Oros, Andrew. 2010. “The 50th Anniversary: Time for a ‘Renewal of Vows.’” Asia Policy 10: 28–34. Packard, George. 2010. “The United States–Japan Security Treaty at 50: Still a Grand Bargain?” Foreign Affairs 89 (2): 92–103. Pagliano, Gary J. 1991. “Allied Burdensharing in Transition: Status and Implications for the United States.” Washington, DC: Congressional Research Service. Palmer, Glenn, and T. Clifton Morgan. 2006. A Theory of Foreign Policy. Princeton, NJ: Princeton University Press. Papayoanou, Paul. 1999. Power Ties: Economic Interdependence, Balancing, and War. Ann Arbor: University of Michigan Press. Pape, Robert. 2005. “Soft Balancing against the United States.” International Security 30 (1): 7–45. Park, Jae-Jeok. 2011. “A Comparative Case Study of the U.S.–Philippines Alliance in the 1990s and the U.S.–South Korea Alliance between 1998 and 2008.” Asian Survey 51 (2): 268–289. Park, Jong Hee, and Kentaro Hirose. 2013. “Domestic Politics, Reputational Sanctions, and International Compliance.” International Theory 5 (2): 300–320. Parker, Geoffrey. 1997. “1630–1632: The Intervention of Sweden.” In The Thirty Years’ War, second edition, edited by Geoffrey Parker, 108–118. London: Routledge. Pastreich, Emanuel. 2005. “The Balancer: Roh Moo-hyun’s Vision of Korean Politics and the Future of Northeast Asia.” Japan Focus. Retrieved on March 26, 2009, from www.japanfocus.org/-Emanuel-Pastreich/2041. Paul, Thazha Varkey. 2004. “The Enduring Axioms of Balance of Power Theory.” In Balance of Power: Theory and Practice in the 21st Century, edited by T.V. Paul, James J. Wirtz, and Michel Fortmann, 1–28. Stanford, CA: Stanford University Press. ———. 2005. “Soft Balancing in the Age of U.S. Primacy.” International Security 30 (1): 46–71. Pehe, Jiri. 2003. “A View from Eastern Europe.” Foreign Policy, July/August, 137, 33. Peretz, Don. 2002. “US Policy toward the Middle East.” In The Middle East Enters the Twenty-First Century, edited by Robert Freedman, 267–287. Gainesville: University Press of Florida. Pew Research Center. 2014. “Global Opposition to U.S. Surveillance and Drones, but Limited Harm to America’s Image.” July 14. Available at www.pewglobal.org/ files/2014/07/2014-07-14-Balance-of-Power.pdf.

218 references Poast, Paul. 2012. “Does Issue Linkage Work? Evidence from European Alliance Negotiations, 1860 to 1945.” International Organization 66 (2): 277–310. Pollack, Benny. 1987. The Paradox of Spanish Foreign Policy: Spain’s International Relations from Franco to Democracy. New York: St. Martin’s Press. Posen, Barry R. 2006. “European Union Security and Defense Policy: Response to Unipolarity?” Security Studies 15 (2): 149–186. ———. 2007. “The Case for Restraint.” American Interest 3 (1): 7–17. Powers, Kathy. 2004. “Regional Trade Agreements as Military Alliances.” International Interactions 30 (4): 373–395. Press-Barnathan, Galia. 2006. “Managing the Hegemon: NATO under Unipolarity.” Security Studies 15 (2): 271–309. Pressman, Jeremy. 2008. Warring Friends: Alliance Restraint in International Politics. Ithaca, NY: Cornell University Press. Preston, Paul. 1986. The Triumph of Democracy in Spain. New York: Methuen. Putnam, Robert. 1988. “Diplomacy and Domestic Politics: The Logic of Two-Level Games.” International Organization 42 (3): 427–460. Pyle, Kenneth. 2010. “Troubled Alliance.” Asia Policy 10: 3–9. Rapkin, David, William Thompson, and Jon Christopherson. 1979. “Bipolarity and Bipolarization in the Cold War Era: Conceptualization, Measurement, and Validation.” Journal of Conflict Resolution 23 (2): 261–295. Ratner, Ely. 2013. “Rebalancing to Asia with an Insecure China.” The Washington Quarterly 36 (2): 21–38. Rauchhaus, Robert, ed. 2001. Explaining NATO Enlargement. Portland, OR, and London: Frank Cass Publishers. Reed, William. 1997. “Alliance Duration and Democracy: An Extension and CrossValidation of ‘Democratic States and Commitment in International Relations.’” American Journal of Political Science 41 (3): 1072–1078. Reisinger, William. 1992. Energy and the Soviet Bloc: Alliance Politics after Stalin. Ithaca, NY: Cornell University Press. Reiter, Erich, and Heinz Gaertner, eds. 2001. Small States and Alliances. Heidelberg and New York: Physica-Verlag. Revere, Evans. 2012. “Park Geun-hye’s Electoral Victory: A Sigh of Relief from Washington?” Brookings Institution. December 19. Available at www.brookings.edu/ blogs/up-front/posts/2012/12/19-south-korea-president-revere. Ricardo, David. (1817) 1821. On the Principles of Political Economy and Taxation, third edition. London: John Murray. Rickli, Jean-Marc. 2008. “European Small States’ Military Policies after the Cold War: From Territorial to Niche Strategies.” Cambridge Review of International Affairs 21 (3): 307–325. Rinehart, Ian E. 2015. “New U.S.–Japan Defense Guidelines Deepen Alliance Cooperation.” In CRS Insight, April 28. Washington, DC: Congressional Research Service, IN10265. Risse-Kappen, Thomas. 1995. Cooperation among Democracies: The European Influence on U.S. Foreign Policy. Princeton, NJ: Princeton University Press. Rose, Gideon. 1998. “Neoclassical Realism and Theories of Foreign Policy.” World Politics 51 (1): 144–172. Rosenbluth, Frances. 2011. “Japan in 2010.” Asian Survey 51 (1): 41–53.

references 219 Ross, Robert. 1999. “The Geography of the Peace: East Asia in the Twenty-First Century.” International Security 23 (4): 81–118. ———. 2013. “US Grand Strategy, the Rise of China, and US National Security Strategy for East Asia.” Strategic Studies Quarterly 7 (2): 20–40. Rothkopf, David. 2015. “The Middle East’s Pivot to Asia.” Foreign Policy, April 24. Rubottom, Richard, and Carter Murphy. 1984. Spain and the United States: Since World War II. New York: Praeger. Russett, Bruce. 1985. “The Mysterious Case of Vanishing Hegemony; or, Is Mark Twain Really Dead?” International Organization 39 (2): 207–231. Sabrosky, Alan Ned. 1980. “Interstate Alliances: Their Reliability and the Expansion of War.” In The Correlates of War II: Testing Some Realpolitik Models, edited by J. David Singer, 161–198. New York: Free Press. Sakamoto, Kazuya. 2000. Nichibei domei no kizuna: Anpo–joyaku to sogosei no mosaku (The Japan–U.S. alliance nexus: The Security Treaty and the Search for Mutuality). Tokyo: Yuhikaku. Sandler, Todd, and Jon Cauley. 1975. “On the Economic Theory of Alliances.” Journal of Conflict Resolution 19 (2): 330–348. Sandler, Todd, and Keith Hartley, 1999. The Political Economy of NATO: Past, Present, and into the 21st Century. Cambridge, UK, and New York: Cambridge University Press. Sarantakes, Nicholas Evan. 1999. “Alliance in Doubt: American Reaction to the 1960 US–Japanese Security Treaty Crisis.” American Diplomacy, 4 (4). Available at www .unc.edu/depts/diplomat/AD_Issues/amdipl_13/sarantakes1.html. Saunders, Phillip. 2007. “The United States and East Asia after Iraq.” Survival 49 (1): 141–152. Schaller, Michael. July 1995. “America’s Favorite War Criminal: Kishi Nobusuke and the Transformation of U.S.–Japan Relations.” Working Paper 11. Oakland, CA: Japan Policy Research Institute. Available at www.jpri.org/publications/workingpapers/ wp11.html. ———. 1997. Altered States: The United States and Japan since the Occupation. New York: Oxford University Press. Schelling, Thomas. 1960. The Strategy of Conflict. Cambridge, MA: Harvard University Press. Schroeder, Paul. 1976. “Alliances, 1815–1945: Weapons of Power and Tools of Management.” In Historical Problems of National Security, edited by Klaus Knorr, 247–286. Lawrence: University of Kansas Press. Schultz, Kenneth. 2001. Democracy and Coercive Diplomacy. Cambridge, UK: Cambridge University Press. ———. 2005. “The Politics of Risking Peace: Do Hawks or Doves Deliver the Olive Branch?” International Organization 59 (1): 1–38. Schweizer, Karl. 1989. England, Prussia, and the Seven Years War: Studies in Alliance Policies and Diplomacy. Lewiston, NY: E. Mellen Press. Schweller, Randall. 1994. “Bandwagoning for Profit: Bringing the Revisionist State Back In.” International Security 19 (1): 72–107. ———. 1998. Deadly Imbalances: Tripolarity and Hitler’s Strategy of World Conquest. New York: Columbia University Press. ———. 2003. “The Progressiveness of Neoclassical Realism.” In Progress in International Relations Theory, edited by Colin Elman and Miriam Fendius Elman, 311–347. Cambridge, MA: MIT Press.

220 references ———. 2006. Unanswered Threats: Political Constraints on the Balance of Power. Princeton, NJ: Princeton University Press. ———. 2010. “Entropy and the Trajectory of World Politics: Why Polarity Has Become Less Meaningful.” Cambridge Review of International Affairs 23 (1): 145–163. Scobell, Andrew, and Scott W. Harold. 2013. “An ‘Assertive’ China? Insights from Interviews.” Asian Security 9 (2): 111–131. Seo, Jungkun. 2015. “Security Ties or Electoral Connections? The US Congress and the Korea–US Free Trade Agreement, 2007–2011.” International Relations of the Asia-Pacific 15 (2): 217–243. Shambaugh, David. 2013. China Goes Global: The Partial Power. New York: Oxford University Press. Shaw, L. M. E. 1998. The Anglo–Portuguese Alliance and the English Merchants in Portugal, 1654–1810. Aldershot, UK: Ashgate. Shenkar, Oded, and Ilgaz Arikan. 2009. “Business as International Politics: Drawing Insights from Nation-State to Inter-Firm Alliances.” Business and Politics 11 (4): 1–31. Simon, Michael, and Erik Gartzke. 1996. “Political System Similarity and the Choice of Allies: Do Democracies Flock Together, or Do Opposites Attract?” Journal of Conflict Resolution 40 (4): 617–635. Simon, Sheldon. 2010. “US–Southeast Asia Relations: Growing Enmeshment in Regional Affairs.” Comparative Connections, October: 49–61. Singer, J. David. 1988. “Reconstructing the Correlates of War Dataset on Material Capabilities of States, 1816–1985.” International Interactions 14 (2): 115–132. Singer, J. David, Stuart Bremer, and John Stuckey. 1972. “Capability Distribution, Uncertainty, and Major Power War, 1820–1965.” In Peace, War, and Numbers, edited by Bruce Russett, 19–48. Beverly Hills, CA: Sage. SIPRI. 2015. Stockholm International Peace Research Institute Military Expenditure Database. Available at www.sipri.org. Siverson, Randolph, and Juliann Emmons. 1991. “Birds of a Feather: Democratic Political Systems and Alliance Choices in the Twentieth Century.” Journal of Conflict Resolution 35 (2): 285–306. Siverson, Randolph, and Joel King. 1979. “Alliances and the Expansion of War.” In To Augur Well: Early Warning Indicators in World Politics, edited by J. David Singer and Michael D. Wallace, 37–49. Beverly Hills, CA: Sage Publications. ———. 1980. “Attributes of National Alliance Membership and War Participation, 1815–1965.” American Journal of Political Science 24 (1): 1–15. Siverson, Randolph, and Harvey Starr. 1994. “Regime Change and the Restructuring of Alliances.” American Journal of Political Science 38 (2): 145–161. Skalnes, Lars. 1998. “Grand Strategy and Foreign Economic Policy: British Grand Strategy in the 1930s.” World Politics 50 (4): 582–616. ———. 2000. Politics, Markets, and Grand Strategy: Foreign Economic Policies as Strategic Instruments. Ann Arbor: University of Michigan Press. Small, Melvin, and J. David Singer. 1982. Resort to Arms: International and Civil War, 1816–1980. Beverly Hills, CA: Sage. Smith, Alastair. 1995. “Alliance Formation and War.” International Studies Quarterly 39 (4): 405–425. ———. 1996. “To Intervene or Not to Intervene: A Biased Decision.” Journal of Conflict Resolution 40 (1): 16–40.

references 221 Snyder, Glenn H. 1984. “The Security Dilemma in Alliance Politics.” World Politics 36 (4): 461–495. ———. 1990. “Alliance Theory: A Neorealist First Cut.” Journal of International Affairs 44 (1): 103–124. ———. 1997. Alliance Politics. Ithaca, NY: Cornell University Press. Snyder, Jack. 2004. “One World, Rival Theories.” Foreign Policy, 145: 53–62. Sohn,Yul, and Min Gyo Koo. 2011. “Securitizing Trade: The Case of the Korea–US Free Trade Agreement.” International Relations of the Asia-Pacific 11 (3): 433–460. Sotooka, Hidetoshi, Masaru Honda, and Toshiaki Miura. 2001. Nichibei Domei Hanseiki: Anpo to Mitsuyaku (Half Century of the Japan–U.S. Alliance: Security Treaty and Secret Promises). Tokyo: Asahi Shimbunsha. Stanton, Sheldon. 1993. “Foreign Policy as Theater: Understanding Spain and NATO.” PhD dissertation, Ohio State University. Steinberg, David, ed. 2005. Korean Attitudes toward the United States: Changing Dynamics. London: M. E. Sharpe. Stone, Laura 1999. “Whither Trade and Security? A Historical Perspective.” In The U.S.–Japan Alliance: Past, Present, and Future, edited by Patrick Cronin and Michael Green, 247–267. New York: Council on Foreign Relations. Storey, Ian. 1999. “Creeping Assertiveness: China, the Philippines and the South China Sea Dispute.” Contemporary Southeast Asia 21 (1): 95–118. ———. 2009. “China and Indonesia: Military-Security Ties Fail to Gain Momentum.” China Brief 9 (4): 6–9. Stromseth, Jonathan. 1989. “Unequal Allies: Negotiations over US Bases in the Philippines.” Journal of International Affairs 43 (1): 161–188. Stueck, William. 2012. “Power and Culture: The Origins and Durability of the Korean– American Alliance.” In Pacts and Alliances in History, edited by Melissa Yeager and Charles Carter, 11–36. London: I. B. Tauris. Suh, Jae-Jung. 2007. Power, Interest, and Identity in Military Alliances. New York: Palgrave Macmillan. Suhrke, Astri. 1973. “Gratuity or Tyranny: The Korean Alliances.” World Politics 25 (4): 508–532. Sunohara,Tsuyoshi. 2007. Domei Henbo: Nichi-Bei Ittaika no Hikari to Kage (Alliance Metamorphosed: The Bright and Dark Sides of the Increasing Interoperability of the U.S.– Japan Alliance). Tokyo: Nihon Keizai Shinbunsha. Swaine, Michael. 2010. “Perceptions of an Assertive China.” China Leadership Monitor, 32: 1–19. Tago, Atsushi. 2008. “Is There an Aid-for-Participation Deal?: US Economic and Military Aid Policy to Coalition Forces (Non)Participants.” International Relations of the Asia-Pacific 8 (3): 379–398. Takeuchi, Hiroki. 2014. “Sino–Japanese Relations: Power, Interdependence, and Domestic Politics.” International Relations of the Asia-Pacific 14 (1): 7–32. Talbott, Strobe. 1988. “East–West No More Mr. Tough Guy?” Time, May 23. Tamamoto, Masaru. 2005. “How Japan Imagines China and Sees Itself.” World Policy Journal 22 (4): 55–62. Tanaka, Akihiko. 1997. Anzen hosho: sengo gojunen no mosaku (National Security: Fifty Years of Postwar Struggle). Tokyo: Yomiuri shimbunsha. Tarnoff, Curt, and Larry Nowels. 2004. “Foreign Aid: An Introductory Overview of U.S. Programs and Policy.” Congressional Research Service (98-916, updated April 15).

222 references Tatsumi, Yuki. 2010. “A Short-Term Challenge to the U.S.–Japan Alliance Putting Its Long-Term Health at Risk.” Asia Policy 10: 35–41. Telser, L. G. 1980. “A Theory of Self-Enforcing Agreements.” Journal of Business 53 (1): 27–44. Tertrais, Bruno. 2004. “The Changing Nature of Military Alliances.” Washington Quarterly 27 (2): 135–150. Thalakada, Nigel. 2012. Unipolarity and the Evolution of America’s Cold War Alliances. London: Palgrave Macmillan. Thompson, Alexander. 2009. Channels of Power: The UN Security Council and U.S. Statecraft in Iraq. Ithaca, NY: Cornell University Press. Tilly, Charles. 1990. Coercion, Capital, and European States, AD 990–1992. Revised edition, Malden, MA: Wiley-Blackwell. Tomz, Michael, and Jessica Weeks. 2015. “Military Alliances and Public Support for War.” Working paper. Stanford University. Available at http://web.stanford.edu/~tomz/ working/TomzWeeks-Alliances-2015-02-14.pdf. Tow, William T., and Douglas Stuart. 2015. The New US Strategy Towards Asia: Adapting to the American Pivot. London: Routledge. Tsebelis, George. 2002. Veto Players: How Political Institutions Work. Princeton, NJ: Prince­ton University Press. Tucker, Nancy Bernkopf. 1998. The Origins and Evolution of the Korean-American Alliance: An American Perspective. Stanford, CA: Asia Pacific Research Center, Stanford University. Tusell, Javier. 2007. Spain: From Dictatorship to Democracy, 1939 to the Present. Malden, MA: Blackwell. Van Ness, Peter. 2002. “Hegemony, Not Anarchy: Why China and Japan Are Not Balancing US Unipolar Power.” International Relations of the Asia-Pacific 2 (1): 131–150. Viñas, Angel. 2003. “Negotiating the U.S.–Spanish Agreements, 1953–1988: A Spanish Perspective.” Jean Monnet/Robert Schuman Paper Series, 3 (7). Available at www.miami .edu/eucenter/vinasfinal.pdf. Viner, Jacob. 1929. “International Finance and Balance of Power Diplomacy, 1880–1914.” Southwestern Political and Social Science Quarterly 9 (4): 407–451. Wallander, Celeste. 2000. “Institutional Assets and Adaptability: NATO after the Cold War.” International Organization 54 (4): 705–735. ———. 2002. “NATO’s Price.” Foreign Affairs 81 (6): 2–8. Walt, Stephen M. 1987. The Origins of Alliances. Ithaca, NY: Cornell University Press. ———. 1997. “Why Alliances Endure or Collapse.” Survival 39 (1): 156–179. ———. 2002. “Keeping the World ‘Off Balance.’” In America Unrivaled: The Future of the Balance of Power, edited by G. John Ikenberry, 121–154. Ithaca, NY: Cornell University Press. ———. 2006. Taming American Power: The Global Response to U.S. Primacy. New York: W. W. Norton. ———. 2009. “Alliances in a Unipolar World.” World Politics 61 (1): 86–120. Waltz, Kenneth. 1959. Man, the State, and War: A Theoretical Analysis. New York: Columbia University Press. ———. 1964. “The Stability of a Bipolar World.” Daedalus 93 (3): 881–909. ———. 1979. Theory of International Politics. New York: McGraw-Hill. ———. 1993. “The Emerging Structure of International Politics.” International Security 18 (2): 44–79.

references 223 ———. 2000. “Structural Realism after the Cold War.” International Security 25 (1): 5–41. Wayman, Frank Whelon. 1984. “Bipolarity and War: The Role of Capability Concentration and Alliance Patterns among Major Powers, 1816–1965.” Journal of Peace Research 21 (1): 61–78. Weart, Spencer. 1998. Never at War: Why Democracies Will Not Fight One Another. New Haven, CT: Yale University Press. Weber, Katja. 2000. Hierarchy Amidst Anarchy: Transaction Costs and Institutional Choice. Albany: State University of New York Press. Weitsman, Patricia. 2003. Dangerous Alliances: Proponents of Peace, Weapons of War. Stanford, CA: Stanford University Press. Werlin, Herbert. 1991. “Bottlenecks to Developments: Studies from the World Bank’s Economic Development Institute.” Public Administration and Development 11 (3): 245–255. Whitaker, Preston. 1961. Spain and Defense of the West: Ally and Liability. New York: Harper. White, Hugh. 2012. The China Choice: Why America Should Share Power. Collingwood, Australia: Black Incorporated. Wight, Martin. 1966. “The Balance of Power.” In Diplomatic Investigations: Essays in the Theory of International Politics, edited by Herbert Butterfield and Martin Wight, 149–175. London: Allen & Unwin. Williamson, Oliver. 1983. “Credible Commitments: Using Hostages to Support Exchange.” American Economic Review 73: 519–540. ———. 1985. The Economic Institutions of Capitalism: Firms, Markets, Relational Contracting. New York: Free Press. ———. 1996. The Mechanisms of Governance. New York: Oxford University Press. Williamson, Oliver, and Scott Masten eds. 1999. The Economics of Transaction Costs. . Cheltenham, UK, and Northampton, MA: Edward Elgar Publishing. Wilson, Peter H. 1995. War, State and Society in Württemberg, 1677–1793. Cambridge, UK, and New York: Cambridge University Press. Wirth, Christian. 2015. “Securing Japan in the ‘West’: The US–Japan Alliance and Identity Politics in the Asian Century.” Geopolitics 20 (2): 287–307. Wohlforth, William. 1999. “The Stability of a Unipolar World.” International Security 24 (1): 5–41. ———. 2002. “U.S. Strategy in a Unipolar World.” In America Unrivaled: The Future of the Balance of Power, edited by John Ikenberry, 98–118. Ithaca, NY: Cornell University Press. Wolf, Charles, and Michele Zanini. 1998. Benefits and Burdens: The Politically Dominated Economics of U.S. Alliances with Japan and Korea. Stanford, CA: Asia Pacific Research Center, Stanford University. Wolford, Scott. 2007. “The Turnover Trap: New Leaders, Reputation, and International Conflict.” American Journal of Political Science 51 (4): 772–788. Womack, Brantly. 2014. “China’s Future in a Multinodal World Order.” Pacific Affairs 87 (2): 265–284. Yan, Xuetong. 2011. “From a Unipolar to a Bipolar Superpower System: The Future of the Global Power Dynamic.” Global Times, December 30. ———. 2013. Lishi de guanxing [Inertia of History]. Beijing: Zhongxin chubanshe. ———. 2014. “From Keeping a Low Profile to Striving for Achievement.” Chinese Journal of International Politics 7 (2): 153–184.

224 references Yoda, Tatsuro. 2006. “Japan’s Host Nation Support Program for the U.S.–Japan Security Alliance.” Asian Survey 46 (6): 937–961. Yoshida, Shingo. 2007. “Beikoku no Gunji Purezensu Shukusho wo Meguru Nihon no Anzenhosho Seisaku” (Japan’s Security Policy Concerning the Reduction in the U.S. Military Presence). Kokusai Anzenhosho 35 (2): 97–118. ———. 2009. “Detanto no Seiritsu to Nichibei Domei” (The Formation of Détente and the U.S.–Japan Alliance). Kokusai Anzenhosho 36 (9): 19–39. Yoshitsugu, Kosuke. 2011. Nichibei domei ha ikani tsukuraretaka: “Anpo taisei” no tenkanten 1951–1964 (How the Japan–US Alliance Was Constructed: The Turning Point of the “Security Treaty Regime” 1951–196)]. Tokyo: Kodansha. Zagoria, Donald. 1962. The Sino–Soviet Conflict, 1956–1961. Princeton, NJ: Princeton University Press. Zhang, Feng. 2012. “China’s New Thinking on Alliances.” Survival 54 (5): 129–148. Zimmermann, Hubert. 2001. Money and Security: Troops, Monetary Policy, and West Germany’s Relations with the United States and Britain, 1950–1971. Cambridge, UK: Cambridge University Press.

Index

abandonment, 17, 58, 108, 184n30, 187nn30–32; defined, 42, 63, 179n33; Japanese fear of, 128, 138, 191n31; measures to reduce risk of, 44–45, 95, 97, 100, 102, 180n41, 191n31; South Korean fear of, 4, 103; systemic polarity and fear of, 24, 26, 44, 46, 50, 51, 62–67, 68, 69–71, 97, 102, 179n33, 187n35 Abe, Shinzo, 47–48, 133, 173; attitudes regarding collective self-defense, 45, 59, 183n16, 201n3; attitudes regarding U.S.-Japan alliance, 2, 20, 45, 141, 196n34, 201n3; on TPP, 20 abrogation of alliances, 44, 54, 71–72, 77–80, 164–65, 188n41 Acharya, Amitav, 50 Adams, Simon, 37 Adenauer, Konrad, 131 Afghanistan: Soviet war in, 118, 193n12; U.S. war in, 98, 99, 101, 105, 106, 140 Ahn, Byung-joon, 191n45 Akiyama, Masahiro, 2, 99, 131 Albania, 178n15 Alexander, Gerard, 104 alignments vs. alliances, 42 Allende, Salvador, 118 alliances: alliance games vs. adversary games, 30–32, 45, 49, 70, 74, 105, 109–10, 169, 173, 180n41, 182n5, 184nn36,2; Alliance Treaty Obligations and Provisions (ATOP) dataset, 5, 6, 9, 14,

25, 34, 43, 48, 80, 81, 82, 83, 84, 87, 154, 158, 177n8, 178n15, 180n36, 182n12, 183n15, 185n3, 186n14, 189nn14,15, 193n12, 199n76; as asymmetric, 19–20, 25, 32–36, 52–53, 78, 79, 180n41, 183nn14,15, 184n31, 199n76; capability aggregation model of, 14, 19, 25, 32, 34, 36, 38, 51, 53, 73, 94, 178nn14,18, 180n41, 183n20; vs. coalitions, 22, 61, 105–6, 107, 170; competition in alliance market, 13, 18, 29, 109, 110, 182n10; consultation obligations, 6, 25, 57, 77, 86–87, 93, 96, 139, 179n28, 188n44; contents of alliance agreements, 23, 24, 29, 33–34, 43, 54, 82–83, 100, 126–28, 138–39, 160–61, 167, 169, 173, 180n35, 184n29, 189n11; as contracts, 5, 14, 23, 24, 28, 36–37, 41, 42–44, 109, 166–67, 169, 173, 183n20, 184n28; defense obligations, 6, 25, 57, 77, 86–87, 88, 93–94, 139, 183n15; defined, 5, 30, 185n3; duration of, 18, 28, 70–71, 72–73, 75, 80–85, 110–11, 184n37, 189n10, 189nn5,6,11; efficiency gains from, 14, 15, 28, 29, 38, 40, 41, 42, 63, 95, 170, 178n21, 179n22, 183n20; exchanges in, 14–21, 23, 24, 28, 29, 32–42, 46, 50, 52–55, 60, 73, 75, 92–94, 106, 109, 166–67, 169, 174, 178n21, 179n34, 180nn37,41, 183n19; formation of, 7, 10–11, 12–13, 18, 28, 29, 50, 53, 71, 73–74, 75, 85–88, 89, 92–94, 95–98,

226 index alliances (continued) 121, 166–67, 177n8, 178nn11,13,14, 184n32, 189n3; vs. friendships, 30; vs. international trade, 15, 33, 36–37, 41–42, 51, 67, 178n21, 180n37; intra-alliance bargaining, 1–2, 14–15, 18–19, 23, 24, 26–27, 28, 44, 46–48, 50, 52, 63, 64, 65, 69, 112–22, 127–28; intra-alliance war, 71, 72, 188n42; military aspects, 5, 14, 15–20, 25, 28, 29, 30–33, 34, 36–42, 45, 51, 52–54, 68, 73, 88, 93–94, 95–98, 99–100, 106, 109, 110, 166, 172, 180n35, 182n8, 186n14, 190n23; negotiation of, 23–24, 25, 29, 37, 38, 54, 112–22, 123–24, 127–28, 134–35, 150, 152, 155, 158–63, 165, 169–70, 185n6, 194nn13,15, 198n57, 200n82; neutrality obligations, 25, 77, 86–87; nonmilitary aspects, 5, 14, 15, 19–21, 25, 29, 32–42, 50–51, 52–55, 75, 78, 79, 83, 84, 88, 93, 99, 109, 110, 111, 168, 172, 179n35, 180n41, 182n8, 185nn3,7, 186n14, 188n39, 191n35; number formed, 7, 10–11, 12–13, 51, 95–96; offense obligations, 25, 77, 86–87, 88, 93–94; portfolio diversification model of, 179n22; profit vs. inf luence in, 121–22, 137, 138, 148–50, 158, 163, 164; public-goods models of, 15, 19, 50, 177n1, 179n23; relationship to international trade, 20, 37, 38, 39, 54–55, 109, 127, 130, 139, 144, 145, 148, 149, 152, 164, 168, 179n35, 182n11, 183n14, 185nn12,12, 199n73; relationship to military threats, 1, 2, 3–4, 7, 15, 16, 17–18, 20, 24, 28, 30–32, 51, 54, 60, 78, 79, 95, 97, 98, 100–101, 110, 113, 179n27; reliability of, 43, 184nn29,37; and reputation, 187n29, 194n17; termination of, 7, 8–9, 18, 21, 28, 54, 70–72, 75, 76, 77–80, 82, 89, 110–11, 121, 164–65, 178n13, 189n6 Altfeld, Michael, 7, 19, 185n9 Altman, Roger, 139 Amin, Hafizullah, 118, 193n12 Anabtawi, Samir N., 193n8 Anac, Sezi, 78 Andolana, Gregorio, 32 Andrew, Christopher, 193n12 Angola, 188n44 Anti-Comintern Pact, 94 Arase, David, 132 Arbatov, George, 28, 30

Areilza, Jose Maria de, 156, 160 Arikan, Ilgaz, 181n49 Aso, Taro, 141 Asahi Shimbun, 22, 136, 196n31, 201n3 ASEAN Treaty of Amity and Cooperation, 58 Asia Infrastructure Investment Bank (AIIB), 59 Asian Development Bank, 59 asset specificity, 42, 44, 188n45 Auslin, Michael, 141 Australia, New Zealand, United States Security Treaty (ANZUS), 187n30 Australia: relations with Japan, 186n13; relations with United States, 185n11 Austria, 71 Austro-Prussian War, 71 Aznar, Jose Maria, 101–2, 157–58, 161–62, 200n89 Baek, Kyeonghi, 188n42 Baghdad Pact, 94 Baker, George, 43, 169 balance of power theory, 34, 50, 178n11 balance of threat, 178n11 balancing, 6–7, 31, 50, 58, 60–61, 74, 146–47, 166, 172, 182n6, 192n47; soft vs. hard, 6, 104–5, 177n10 Balkan Wars, 71 Ball, Christopher, 178n16 bandwagoning, 103–4, 105 Bangladesh, 186n16 bargaining power, 16–17, 51, 63; relationship to domestic politics, 1, 3, 15, 23, 24, 26–27, 29, 46–48, 111, 112–22, 123, 127–28, 129, 137–42, 148–53, 158–65, 167, 169–70, 173, 181n46; relationship to systemic polarity, 1, 3, 14–15, 18–19, 21–22, 24, 26, 28, 29, 44, 50, 61–62, 69–76, 99, 172 Barnett, Michael, 1, 178n14, 184n37, 185n9 Basora, Adrian, 161 Baxter, Richard R., 43 Bearce, David, 183n24, 188n42 Bechtol, Bruce E., 146 Beckley, Michael, 56, 170, 184n33, 187n34 Belcher, Gerald, 20, 37, 54, 182n9 Belgium: relations with France, 20, 37 Belize, 178n15, 189n17 Bennett, Andrew, 113 Bennett, D. Scott, 13, 82, 164, 184n37 Berger, Thomas, 191n38



index 227

Bernecker, Walther L., 200n87 Bhagwati, Jagdish, 139 Bin Laden, Osama, 30 bipolarity, 59, 137, 180nn40,41, 187n37, 188n43; duration of alliances in multipolarity vs. (Hypothesis 3.2), 72–73, 80–85; future Sino-American bipolarity, 4–5, 14, 26, 50, 58; vs. multipolarity, 21, 60, 61–62, 69, 70, 71, 72–73, 75, 80–85, 110–11, 168–69, 182n10, 187n35; and overextension, 49, 70, 167; selectivity in alliance commitments in unipolarity vs. (Hypothesis 3.3), 73–74, 75, 85–88; superpower collusion in, 62, 66, 169; vs. unipolarity, 4–5, 6, 19, 21–22, 28, 32, 46, 49, 55, 58, 60–62, 69, 72, 73–74, 75, 85–88, 95–99, 103, 108, 110, 166–67, 168, 172–73. See also Cold War Blackburn, Robert M., 53 Blair, Dennis, 56 Blair, Tony, 49, 102 Bolivia, 84 Botella, Joan, 199n78 Botero, Rodrigo, 199n75 Boulding, Kenneth, 33, 167, 182n13 Box-Steffensmeier, Janet, 79, 81, 83, 189n9 Boyer, Mark, 15, 19, 36, 39, 51, 179n23 Bremer, Stuart, 82, 180n36 Bronson, Rachel, 33, 168, 179n35 Brooks, Stephen, 55, 104, 185n11, 201n2 Brown, Harrold, 56 Brown, Michael, 177n, 178n16 Buckley, Roger, 128, 181n46 Bueno de Mesquita, Bruce, 188n42 Bulgaria, 20, 37, 71, 96 Bush, George H. W., 14, 191n35 Bush, George W., 89, 101–102, 106, 111, 140–41, 144–45, 146, 152, 153, 158, 173, 200n89 Buszynski, Leszek, 182n7 Buzan, Barry, 46, 179n32 Byman, Daniel L., 188n39 Calder, Kent E., 107, 149, 183n18, 192n49; on Japan’s host nation support, 36, 91, 138, 173–74, 191n32, 195n9 Calvo-Gonzalez, Oscar, 53 Calvo-Sotelo, Leopoldo, 160 Cambodia, 99, 126 Campbell, Kurt, 106, 107, 170 Cantalapiedra, David García, 162 Caron, David D., 97, 181n42

Carr, Raymond, 199n78 Carrero Blanco, Luis, 155 Carter, Jimmy, 138, 145 cases and methods, 25–26 Castiella, Fernando Maria, 155, 200n82 Catalinac, Amy, 187n30 Cauley, Jon, 179n23 Caverley, Jonathan, 190n25 Central Treaty Organization (CENTO), 72, 94 Cha, Victor, 38, 44, 59, 68, 153, 189n8 Chang Myon, 198n55 Chang Wanquan, 58 Charles II, 54 Chen, Dingding, 59 Cheonan, 3 Cheong, Wook-sik, 198n59 Chile, 118 China: alliance policy, 5–6, 14, 50, 56–58, 166, 186n17; economic conditions, 56, 103, 104, 106, 127, 171–72, 191n44; military capabilities, 50, 56–58, 172; military expenditure, 16; nonaggression pacts involving, 177n6; as nonpolar, 50, 56–58; nuclear weapons, 56; People’s Liberation Army, 56; policy regarding territorial disputes, 101, 170–71, 201n3; power projection by, 50, 56–57, 172; relations with Japan, 2, 3, 16, 31, 35, 47–48, 98, 100–101, 127, 130, 134, 140, 171, 179n27, 201n3; relations with North Korea, 6, 17, 41, 57; relations with Philippines, 31–32, 179n27; relations with Russia, 6, 57, 192n48; relations with South Korea, 4, 17, 31, 48, 55, 59–60, 102, 103, 104, 172, 186n20, 191n44; relations with Soviet Union, 179n30; relations with United States, 1–2, 5–6, 14, 16, 26, 27, 49, 50, 58–60, 65, 103, 104, 108, 109, 110, 137, 138, 166, 168, 170, 171–73, 174, 179n26, 185n6, 187n41, 192n48, 201n3 Chislett, William, 53, 158, 159, 161, 199n76 Choi, Ajin, 183n22 Choi, Kang, 102 Choo, Jaewoo, 17 Christensen, Thomas, 45, 56, 179nn27,30, 180n40 Christopherson, Jon, 55, 184n34 Chun Doo-hwan, 143, 146, 147, 152, 153 Chung, Jae Ho, 59 Chung Dong-young, 147

228 index Clare, Joe, 200n96 Clinton, Bill, 106, 111, 139, 190n22 Clinton, Hillary, 1 coalitions, 22, 61, 105–6, 107, 170, 191n42 Cold War: end of, 2, 3, 4, 5–7, 14–15, 17, 18–19, 28, 29, 32, 49, 50, 51, 52, 54, 72, 74, 85–88, 91, 92, 92–93, 94, 95–101, 104, 107, 131, 166–67, 178n11, 190n26; Japan after, 98–101, 131, 190n30; Japan during, 1, 3, 31, 39, 55, 98–99, 100, 114, 124–26, 193n63, 196n26; persistence of alliances made during, 50, 52, 72, 80, 81, 86, 87, 89, 94, 97, 166–67, 178n11, 181n2; South Korea during, 1, 4, 103, 149; Soviet Union during, 14, 16, 17, 21, 26, 35, 40, 46, 73, 80–81, 87, 89, 168, 171, 179n28, 188n44; United States after, 2, 3, 6, 7, 17, 18–19, 28, 29, 32, 49, 51–52, 54, 61, 72, 74, 85–92, 93, 95–97, 99–101, 106, 107, 190n26; United States during, 1, 3, 4, 21, 40, 46, 53, 57, 58, 60, 73, 80–81, 85–88, 93, 103, 106, 107, 108, 114, 117, 124, 125, 137, 149, 157, 168, 171, 180n40, 192n49, 193n63. See also bipolarity Colman, Jonathan, 53, 185n7 comparative advantage, 33, 38–40, 41, 51, 53, 101, 183n22 Composite Index of National Capability (CINC), 19, 78, 82, 180n36, 182n12 Comprehensive Test Ban Treaty (CTBT), 106 concessions, 1, 4, 22–23, 46, 55, 65, 112, 124, 149, 150, 152, 158, 161–62, 165; economic concessions, 14, 20, 36, 37–38, 39, 53, 54, 55, 75, 99, 106, 111, 122, 126–27, 133–34, 137–38, 149, 158–59; political concessions, 14, 17, 20, 21, 37, 53, 54, 75; relationship to ally’s perception of state’s leader, 5, 23, 26–27, 47, 113, 117–18, 119, 120, 130, 135, 141, 159, 163, 167, 194n17; relationship to opposition to alliance cooperation, 5, 23, 26, 47, 114–15, 116, 121, 127–28, 137, 138, 139, 193n7 Conference on Interaction and ConfidenceBuilding Measures in Asia (CICA), 59 Congo, 188n44 constructivism, 29, 182n2 Conybeare, John, 179n22, 181n49 Cooley, Alexander, 43, 183n19 cooperation, 30, 33–34, 40, 168, 169, 172 Correlates of War project, 19, 78, 82, 93, 180n36, 182n12, 190n24

Cortada, James, 199n75 Cox proportional hazard model, 79, 81, 83, 189n10 Crawford, Timothy W., 182n5 Croatia, 178n15 Cronin, Patrick, 2, 131 Czechoslovakia, 73, 96 Czech Republic, 39 David, Steven R., 1, 178n14, 184n37, 185n9 Davis, Christina (2009), 20, 37, 168, 180n35, 182n11 Davis, David, 194n13 Defense Planning Guidance, 14, 52 De Gaulle, Charles, 71, 187n36 democracy, 78, 79, 82, 84, 89, 118–19, 170, 181n47, 183nn22,25, 190n22, 193n6; in Japan, 123, 133; in Philippines, 115, 193n5; in South Korea, 118, 123, 143, 147, 150, 153, 162; in Spain, 123, 155–56, 157, 158, 159–60, 162, 199n78 Dempsey, Judy, 5 Denmark, 39 DeScioli, Peter, 182n4 Deutsch, Karl W., 46 Diamond, Larry Jay, 197n48 Dingman, Roger, 24 division of labor, 15, 39–40, 41 Djibouti, 58–59 domestic politics, 192n49, 193n3; in Japan, 22, 35, 45, 47, 115, 117, 119, 122, 123, 124–42, 165, 173, 194nn14,2,6,7, 195nn20,21,24–26, 196n34, 198n54; opposition to alliance cooperation, 5, 23, 26–27, 47–48, 112, 113, 114–16, 121, 123, 124–29, 133, 134–35, 137–38, 139, 140, 142–45, 153–55, 156, 158, 159, 160–61, 162, 163–64, 167, 169–70, 173–74, 181n46, 193nn4–7, 194n13; regime type, 23, 24, 25, 118–19, 123, 142–43, 147, 150, 153–56, 157, 158, 159–60, 162, 164–65, 170, 173, 181nn46,47, 184n37, 193nn5,6; relationship between intra-alliance bargaining advantage and domestic opposition (Hypothesis 5.3), 121, 162; relationship between intraalliance bargaining advantage and leaders’ anticooperation attitude and nonvulnerability (Hypothesis 5.2), 121, 162; relationship between intraalliance bargaining advantage and leaders’ procooperation attitude and



index 229

vulnerability (Hypothesis 5.1), 120, 162; relationship to state bargaining power, 1, 3, 15, 23, 24, 26–27, 29, 46–48, 111, 112–22, 123, 127–28, 129, 137–42, 148–53, 158–65, 167, 169–70, 173, 181n46; in South Korea, 3, 4, 22, 38, 48, 117, 118, 123, 124, 142–53, 162, 173, 197nn47–51, 198nn55,65, 200n84; in Spain, 117, 123, 124, 153–62, 173, 200n84; state leaders’ attitude toward alliance, 12–21, 23, 24, 26–27, 47–48, 112, 113–14, 116–18, 119, 120–21, 123, 129–33, 137–42, 153–56, 162, 163–64, 165, 167, 170, 181n46, 193n4, 194nn16,17, 198n59, 200n95; state leaders’ vulnerability, 23, 26–27, 47–48, 112, 113, 118–21, 122, 123, 133–42, 147–48, 157–58, 160, 162, 163, 164, 167, 170, 173, 181n46, 193n4, 194nn14,15, 200n95; in United States, 24, 123, 124, 163, 165, 171, 173, 174, 193n61. See also concessions Duffield, John, 178n11 Dulles, John Foster, 127, 134 Duncan, George, 72 Dwyer, Jennifer Holt, 36, 38 Dyson, Stephen Benedict, 102 Easley, Leif-Eric, 199n68 East Asian Strategic Initiative (EASI-I), 107 East China Sea, 101 Eastern Bloc, 18, 96, 189n16 Eberstadt, Nicholas, 179n29 economic crisis of 2007–2008, 105, 106, 190n18 economies of scale, 38, 40, 41, 51 Ecuador, 84 Efron method, 79, 83 Ehrhardt, George, 192n46 Eisenhower, Dwight, 128, 134 Emmons, Juliann, 184n37 entrapment/entanglement, 69, 108, 115, 171, 184n30, 187nn27,34,35, 201n3; defined, 42, 63, 179n33, 184n33; Japanese fear of, 124, 126, 128; measures to reduce risk of, 44–45, 97, 100, 102; South Korean fear of, 4, 103, 104, 172; systemic polarity and fear of, 24, 26, 46, 50, 51, 62–64, 67–68, 70, 97, 179n33 Ethiopia, 188n44 EUGene, 13, 180n36, 182n12 European Economic Community (EEC), 154, 156

European Security and Defense Policy (ESDP), 97, 190n29 European Union, 95, 102 Evans, Peter, 47, 112, 162, 169, 181n46 expectations for future interactions, 114, 119, 165 Eyal, Jonathan, 190n22 Falkland Islands, 56 Fang, Songying, 173 Farley, Miriam, 112 Fawn, Rick, 102 Fearon, James, 47, 169, 188n42, 193n3 Fedder, Edwin H., 172, 180n39 Finnegan, Michael, 20, 193n63 Fisher, Richard, 88 Flanagan, Kristen, 183n24, 188n42 Floros, Katharine, 183n24, 188n42 Fordham, Benjamin, 37, 168, 180nn35,37, 182n11 Foreign Relations of the United States, 25 France, 66, 101, 190n26; Louis XIV, 54; and NATO, 71, 187n30; number of alliance agreements, 96; relations with Belgium, 20, 37; relations with Russia, 20, 37; relations with Spain, 54; relations with United States, 71, 102, 189n14; during Thirty Years’ War, 37 Franco, Francisco, 24, 53, 119, 158, 199n80; and alliance with U.S., 153–55, 157, 159, 160, 162, 163, 200n84; death of, 112, 154–56, 157, 159, 160 Fravel, M. Taylor, 50 Friedberg, Aaron, 49, 50, 52, 56, 109, 110, 166, 171; A Contest for Supremacy, 58 Fujimoto, Kazumi, 35 Fukuda, Yasuo, 141 Fusi, Juan Pablo, 199n78 Futenma relocation plan, 117–18, 120, 132, 136, 141–42, 174, 192n58, 201n7 Gaddis, John Lewis, 60 Gaertner, Heinz, 185n7 Gartzke, Erik, 164, 184n37 Gaubatz, Kurt Taylor, 82, 164, 184n37 Gavin, Francis J., 33–34, 38, 183n27 Gelpi, Christopher F., 173 George, Alexander, 113 Germany, 92, 101, 107; East Germany, 6, 96; economic conditions, 39, 185n7; host nation support (HNS), 91, 149; and NATO, 86; Nazi Germany, 20, 37;

230 index Germany (continued) relations with United States, 39, 52, 91, 102, 131, 185n7, 189n16; reunification, 102; West Germany, 38, 86, 180n39, 185n7, 189n16 Gholz, Eugene, 171 Giarra, Paul, 137 Gibbons, Robert, 43, 169 Gibbs, David, 190n21 Gibler, Douglas, 9, 173, 178nn14,18, 181n47, 187n29 Gil, Federico, 199n81 Gilpin, Robert, 6 Girard, William, 126 Gleditsch, Kristian Skrede, 164, 184n37 Goertz, Gary, 178n18 Goh, Evelyn, 100 Goldgeier, James, 190n22 Goldstein, Avery, 179n23 Gonzalez Marquez, Felipe, 156, 157, 158, 161, 162, 163 Gore, Albert, Sr., 135 Gourevitch, Peter, 173 Gowa, Joanne, 21, 33, 41, 168, 179n35 Grambsch and Therneau’s proportional hazards test statistic, 79, 189n9 Great Britain, 56, 107, 152; relations with Japan, 20, 23–24, 31, 37; relations with Portugal, 20, 33, 37, 54, 182n9; relations with Prussia, 20, 37; relations with Russia, 20, 37; relations with South Africa, 65; relations with Spain, 54; relations with United States, 37, 71 great powers vs. major powers, 190n24 Greece, 30–31, 71, 89 Green, Michael, 59, 131, 141, 142, 196n34 Grieco, Joseph, 37 Grossman, Sanford, 169 Guam, 91, 192n58 Guelleh, Ismail Omar, 59 Guidelines for U.S.–Japan Defense Cooperation, 2, 100, 201n3 Gulick, Edward Vose, 177n9 Gunther, Richard, 199n78 Guyana, 178n15, 189n16 Haas, Ernst, 177n9 Haas, Richard, 180n40 Haftendorn, Helga, 178nn11,14, 181n2 Hagel, Chuck, 192n53 Hagerty, James, 128 Hagström, Linus, 101

Han, Sukhee, 59 Hansen, Birthe, 97, 168, 181n42 Hara, Yoshihisa, 134 Harold, Scott W., 50, 109, 171 Harrell’s rho, 79, 189n9 Harrison, Hope M., 69, 189n17 Hart, Oliver, 43, 169 Hartley, Keith, 15, 51, 183nn23,26 Hashimoto, Ryutaro, 139–40, 142, 196n34 Hata, Masaki, 45 Hata, Tsutomu, 136, 195n24 Hatoyama, Ichiro, 117, 132, 133, 134, 138, 142, 163 Hatoyama, Yukio, 117, 129–30, 132–33, 136, 141–42, 195n26 Havens, Thomas, 197n42 hegemonic stability theory, 6, 177n7, 179n23 Heginbotham, Eric, 16, 192n59 Heo, Uk, 20, 185n12, 199n73 Herter, Christian, 135 Higuchi report, 131 Hirose, Kentaro, 194n15 Hirschman, Albert O., 20, 37, 184n31 Hisae, Masahiko, 98, 192nn57,59, 196n31, 201n7 Holsti, Ole, 40 Honda, Masaru, 2, 99, 131, 191n35 Hong, Yong-Pyo, 181n46, 191n45, 197n47, 198n57 Hopmann, Terrence, 40 Hosokawa, Morihiro, 129, 131, 133, 136, 139, 142, 163, 195n24 Howard, John, 186n13 Howell, William, 47, 169, 193n3 Hudson, John, 86 Humphrey, Hubert, 192n46 Hungary, 20, 37, 73, 96 Hussein, Saddam, 30 Hwang, Jaeho, 4 Igarashi, Takeshi, 127 Iglesias-Cavicchioli, Manuel, 102, 158, 200n89 Iida, Takeshi, 45 Ikeda, Hayato, 135 Ikenberry, G. John, 26, 46, 55, 92, 168, 183n27, 185n11, 201n2; on lack of balancing against U.S., 21, 32, 60 Im, Koonsam, 185n12 incomplete contracting theory, 43, 44 India, 58, 186n16 Indonesia, 58, 186nn16,18

index 231 institutionalism, 29, 178n11, 181n2 International Monetary Fund (IMF), 198n52 International Security Assistance Force (ISAF), 101 international trade, 171–72, 190n25, 191n44, 195n11; vs. alliances, 15, 33, 36–37, 41–42, 51, 67, 178n21, 180n37; efficiency gains in, 41, 51, 178n21; relationship to alliances, 20, 37, 38, 39, 54–55, 109, 127, 130, 139, 144, 145, 148, 149, 152, 164, 168, 179n35, 182n11, 183n14, 185nn12,12, 199n73 Iran: Islamic Revolution of 1979, 121, 187n31; relations with United States, 105, 121, 187n31 Iran-Iraq War, 97, 181n42 Iraq War, 89, 91, 98, 105, 106, 110, 173, 187n33, 191n42; Japan during, 2–3, 99, 100, 140–41, 186n13; South Korea during, 103, 112; Spain during, 101–2, 117, 156, 158, 162, 200n89 Ishibashi, Tanzan, 117, 130, 142 Ishikawa, Masumi, 194n6 Israel: relations with United States, 5, 89, 190n19 Italy, 107, 156, 192n52 Ito, Go, 179n26, 185n6 Izumikawa, Yasuhiro, 128, 182n5, 190n30 Jacobson, Harold, 47, 112, 162, 181n46 Jaggers, Keith, 78, 82 Japan: Antiterrorism Special Measures Law, 100, 140; after Cold War, 98–101, 131, 190n30; during Cold War, 1, 3, 31, 39, 55, 98–99, 100, 114, 124–26, 193n63, 196n26; Communist Party, 125, 194n6; constitution, 2, 35, 45, 127, 183n16; Democratic Party of Japan (DPJ), 22, 118, 132, 134, 136–37, 141–42, 181n45, 195nn25,26; Diet/ House of Representatives, 125, 194n6, 196n26; domestic politics, 22, 35, 45, 47, 115, 117, 119, 122, 123, 124–42, 165, 173, 194nn14,2,6,7, 195nn20,21,24–26, 196n34, 198n54; economic conditions, 35–36, 55, 112, 126–27, 130, 133–34, 137, 149, 185n7, 193n63, 195n11; host nation support (HNS), 99, 131, 138, 183n18, 191n32; during Iraq War, 2–3, 99, 100, 140–41, 186n13; Komeito (Clean Government Party), 136; leaders’ attitudes regarding alliance cooperation,

129–33, 137–42; leaders’ vulnerability, 133–42, 173; Liberal Democratic Party (LDP), 117, 125–26, 127, 129, 130, 131, 132, 133, 134, 136, 139, 140–41, 142, 163, 194nn14,7, 195n26; Maritime Defense Force, 39; military expenditure, 16; Ministry of Defense, 194n18, 196n31; Ministry of Foreign Affairs, 171, 194n18, 196n31, 197n43; New Party Sakigake, 129, 136; omoiyari-yosan, 99; opposition to alliance cooperation in, 124–29, 130, 137–38, 139, 140, 198n54; during Persian Gulf War, 2, 99, 126, 191n35; relations with Australia, 186n13; relations with China, 2, 3, 16, 31, 35, 47–48, 98, 100–101, 127, 130, 134, 140, 171, 179n27, 201n3; relations with Great Britain, 20, 23–24, 31, 37; relations with North Korea, 1–2, 3, 16, 31, 98, 99–100; relations with Russia, 31; relations with South Korea, 59, 198n58; relations with Soviet Union, 35; relations with United States, 1, 2–3, 14, 15, 16, 20, 21, 25, 31, 34–36, 38, 39, 45, 47, 49, 52, 53, 54, 58, 91, 98–101, 103, 106, 108–9, 110–11, 112, 113, 114, 117, 120, 121, 122, 123–42, 150, 153, 159, 162, 163, 165, 166, 172, 173, 174, 179n26, 180n39, 181nn45,46, 183n17, 185n6, 186n14, 191nn31,35,38, 193n63, 194nn18,2,7, 195nn14,20,21,25, 196nn31,34, 197n43, 198n54, 201nn3,7; Self-Defense Forces ( JSDF), 45, 99, 100, 126, 140–41, 186n13, 194n18, 195n8; Social Democratic Party, 196n26; Socialist Party, 124–26, 127, 129, 136, 194nn6,7; Status of Forces Agreement (SOFA), 139, 140 Jervis, Robert, 26, 168, 174 Johnson, Jesse C., 173 Johnson, Lyndon, 137, 192n46 Johnston, Alastair Iain, 50, 101, 171 Jones, Seth, 79, 81, 83, 91, 189n9, 190n25 Juan Carlos, King, 157, 200n87 Kan, Hideki, 22 Kan, Naoto, 117–18, 142 Kan, Shirley, 192n58 Kane, Tim, 107 Kanemaru, Shin, 99 Kaplan, Morton, 46 Kaplan-Meier survival curves, 80–81 Kapstein, Ethan, 190n25

232 index Kataoka, Tetsuya, 115, 126, 127, 181n46 Katz, Katrin, 38, 153 Katzenstein, Peter, 191n38 Kawato, Yuko, 193n6 Kazakhstan, 57 Keele, Luke, 189n9 Kennedy, John, 137, 195n11 Keohane, Robert, 6, 69, 178nn11,14, 181n2, 184n31 Kil, Soong Hoom, 197n48 Kim, Choong Nam, 146 Kim, Dong-Hun, 181n49 Kim, Tong, 112 Kim Dae-jung, 3, 117, 144, 146, 147, 197n50, 198n58 Kim Il-sung, 119 Kim Jong-il, 144, 147 Kim Yong-Sam, 146 Kindleberger, Charles, 6 King, Joel, 43, 45 Kishi, Nobusuke, 128, 138, 159; as proalliance leader, 117, 129, 130–31, 134–35, 139, 142, 163; during WWII, 129, 134, 195n20 Kissinger, Henry, 155–56, 160, 185n6, 200n82 Knorr, Klaus, 56 Koike, Yuriko, 133 Koizumi, Junichiro, 3, 136, 140–41, 142, 186n13, 196n34 Koo, Min Gyo, 20, 185n12 Korean War, 57, 126, 143, 144, 145, 152, 195n8 Koremenos, Barbara, 43 Kornbluh, Peter, 193n11 Krebs, Ronald R., 188n42 Kupchan, Charles, 164 Kurzban, Robert, 182n4 Kuwait: relations with United States, 97, 181n42 Kydd, Andrew, 87 Kyoto Protocol, 106 Kyrgyzstan, 57 Lai, Brian, 184n37 Lake, Anthony, 190n22 Lake, David, 43, 178n20, 188n45 Larres, Klaus, 190n21 Lawless, Richard, 108–9, 192n55 Layne, Christopher, 32, 94 Lee, Dong Sun, 181n46, 197n48 Lee, Gerald Geunwook, 103, 187n33

Lee, Kang-ro, 197n51 Lee, Sook-Jong, 143 Lee Hoi-chang, 146, 147 Lee Myung-bak, 3, 103–4, 147, 148, 152–53, 198n65 Leeds, Brett Ashley, 20, 71, 78, 82, 83, 93, 95, 173, 178n13, 181n50, 184n37, 188n1, 194n13; on abrogation of alliances, 44, 54, 77, 79, 164–65; on definition of alliances, 5, 185n3; on domestic politics and alliances, 23; on reliability of alliances, 43, 184n29; on trade and alliances, 33, 37, 168, 180n35, 182n11 Lend-Lease agreement, 37 Leonard, Thomas M., 149 Leverett, Flynt, 105 Levkowitz, Alon, 150 Levy, Jack, 1, 178n14, 184n37, 185n9 liberalism, 29, 182n2 Libya, 96 Licht, Amanda, 189n10, 194n16 Lieber, Keir, 104 Ling, Wei, 201n6 Lipson, Charles, 43, 168 Liska, George, 1, 28, 30, 31, 168, 178n14, 200n1 Long, Andrew, 20, 43, 168, 179n35, 188n42; on trade and alliances, 33, 37, 180n35, 182n11 MacArthur, Douglas, II, 131, 134, 135 Malaysia, 186n16 Mandelbaum, Michael, 38, 44, 178n16, 184n30 Mansfield, Edward, 33, 168, 179n35, 187n26 Manyin, Mark, 198n52 Maoz, Zeev, 78 Marcos, Ferdinand, 115, 193n6 Marshall, Monty, 78, 82 Mastanduno, Michael, 26, 46, 106, 168 Masten, Scott, 43 Mattes, Michaela, 23, 44, 79, 82, 83, 93, 95, 164–65, 173, 178n13, 181n50 Maurer, Noel, 181n42 McCalla, Robert, 178n11, 181n2, 183n27 McDevitt, Michael, 59 McGillivray, Fiona, 193n10, 194n15 McLean, Murray, 186n13 McLeary, Paul, 177n2 Mearsheimer, John, 50, 94, 109, 169, 171; on end of Cold War, 5, 32, 52, 178n11 Meernik, James, 179n25

index 233 Menon, Rajan, 32, 106, 189n12 Middle East, 193n62 Midford, Paul, 177n4 military bases, 14, 20, 33, 36, 37, 44, 76; role in asymmetric alliances, 34, 53, 183n15. See also U.S. military bases Miller, Benjamin, 182n5 Miller, Gregory, 173, 187n29 Mitchell, Sara McLaughlin, 43 Mitrokhin, Vasili, 193n12 Miura, Toshiaki, 2, 99, 131, 191n35 Miyahara, Yasufumi, 31 Modelski, George, 55, 180n40, 184n34 monetary regimes, 39 Monteiro, Nuno, 46, 168, 189n2 Montero, José Ramón, 199n78 Moon, Chung-In, 197n48, 198n65 Moore, Gregory J., 101 Moore, John, 169 moral hazard, 178n17 Moravcsik, Andrew, 46 Morgan, T. Clifton, 15, 178n20, 179n34, 182n8 Morris, Ivan, 193n8 Morrow, James, 19, 32–33, 68, 169, 178n18, 182n8, 185n9 Mowle, Thomas, 39 Mozambique, 96, 188n44 Mueller, John, 7 Mujal-León, Eusebio, 200n94 multipolarity, 4–5, 32, 34, 46, 52, 55, 58, 180nn40,41, 184n36, 189n5; abrogation of alliances in bipolarity and unipolarity vs. (Hypothesis 3.1), 71–72, 77–80; vs. bipolarity, 21, 60, 61–62, 69, 70, 71, 72–73, 75, 80–85, 110–11, 168–69, 182n10, 187n35; duration of alliances in bipolarity vs. (Hypothesis 3.2), 72–73, 80–85; and overreaction, 49, 70, 71–72, 167; vs. unipolarity, 69, 71–72, 74, 75, 77–80, 110–11, 187n35 Muñoz Grandes, Agustín, 155 Murata, Koji, 107, 191n45 Murayama, Tomiichi, 129, 142, 195n24 Murphy, Carter, 181n46, 199n75 Murphy, Kevin, 43, 169 Myanmar, 57 Myers, Ramon, 115, 126, 127, 181n46 Nakasone, Yasuhiro, 196n34 Narizny, Kevin, 164 national interests, 3, 29, 46, 167, 168

NATO, 30–31, 41, 82, 94, 139, 178n11, 183nn18,26,27, 190n29; expansion of, 6, 7, 70, 86, 87, 89, 91, 96, 97, 178nn15,16,17, 189n16, 190nn22,27; and France, 71, 187n30; and Russia, 7; and Spain, 154, 156, 160, 161, 199nn79–81; and United States, 6, 7, 14, 30–31, 86, 87, 89, 91–92, 96, 101, 107, 189n16, 192n52, 193n61 Netherlands, 39 Newnham, Randall, 53, 191n42 New Zealand: relations with United States, 187n30 Niksch, Larry, 192n58 Nish, Ian, 23–24 Nishihara, Masashi, 2, 99, 126 Nixon, Richard, 128, 137, 163, 181n42, 185n6, 200n82; Nixon Doctrine, 16, 146 nonaggression pacts, 25, 77, 177n6, 178n12, 180n36, 181n50 Nordstrom, Timothy, 188n42 Norrlof, Carla, 183n21 North Korea: economic conditions, 17, 149; nuclear weapons, 16, 17, 98, 102–3, 107, 140; relations with China, 6, 17, 41, 57; relations with Japan, 1–2, 3, 16, 31, 98, 99–100; relations with Russia, 179n28; relations with South Korea, 1–2, 3–4, 17, 25, 31, 38, 48, 102–3, 144, 145, 146, 147–48, 153; relations with Soviet Union, 179nn28–30 North Yemen, 6 Nowels, Larry, 88 Nye, Joseph, 15, 184n31, 195n16 Oberdorfer, Don, 17, 21, 105, 146, 150, 181n46, 197nn47,48, 198n62 Official Development Assistance (ODA), 99 Ogata, Taketora, 195n21 Okada, Katsuya, 134 Okamoto, Yukio, 2, 99 Okinawa, 91, 108, 117, 140, 141–42, 174, 185n6, 192n58, 201n7 Olson, Mancur, 15, 177n1 opportunistic behavior, 40, 42–43, 44, 54, 71–72 Organization for Economic Cooperation and Development (OECD), 149 Organization of American States (OAS), 86, 87, 178n15, 189n16, 190n26

234 index Oros, Andrew, 141 Ottoman Empire, 71 Ozawa, Ichiro, 136, 195n25 Packard, George, 141, 196n31 Pact of Madrid, 154, 157 Pagliano, Gary J., 199n68 Pakistan, 186n16 Palmer, Glenn, 15, 178n20, 179n34, 182n8 Panama Canal, 89 Papayoanou, Paul, 168, 180n35, 182n11 Pape, Robert, 104, 105 Park, Jae-Jeok (2011), 32, 88, 103, 193n5 Park, Jong Hee, 194n15 Park, Joon-Sung, 102 Park Chung-hee, 143, 145–46, 147, 153, 197nn47,50, 198nn55,56 Park Geun-hye, 3, 4, 48, 104, 145, 173 Parker, Geoffrey, 37 Pastreich, Emanuel, 146 Paul, Thazha Varkey, 104, 105 Pehe, Jiri, 89 Peretz, Don, 190n19 Persian Gulf War, 2, 98, 99, 126, 191n35 Peru, 181n42 Pevehouse, Jon, 47, 169, 193n3 Pew Research Center, 144, 179n27, 186n16 Philippines, 186n16; domestic politics, 115, 193nn5,6; relations with China, 31–32, 179n27; relations with United States, 31–32, 88, 115, 179n27, 193n5 Plato’s Lysis, 182n4 Poast, Paul, 37, 54, 180n35, 182n11 Poland, 96, 102 policy implications, 170–74 Polity IV project, 78, 82 Pollack, Benny, 155, 161, 181n46 Portugal: and NATO, 89; relations with Great Britain, 20, 33, 37, 54, 182n9 Posen, Barry R., 95, 171, 190n29 Powers, Kathy, 33, 168, 178n18, 180n35 Press, Daryl G., 171 Press-Barnathan, Galia, 101, 190n29 Pressman, Jeremy, 173 Preston, Paul, 199n78 Prussia, 20, 37, 71 Putnam, Robert, 47, 112, 113, 116, 162, 169, 181n46 Pyle, Kenneth, 21, 35, 99 Rajoy, Mariano, 156 Ramos, Fidel, 32

Rapkin, David, 55, 180n40, 184n34 Ratner, Ely, 16, 192n59, 201n6 Rauchhaus, Robert, 190n22 Reagan, Ronald, 35, 143, 161 realism, 29, 32, 69, 94, 169, 178n11, 181nn48,2, 187n35 Reed, William, 82, 164, 184n37 Reisinger, William, 189n17 Reiter, Dan, 184n37 Reiter, Erich, 185n7 relative capabilities: capability ratios, 34, 82, 182nn12,13, 183n14; and systemic polarity, 62, 63–76, 82, 187nn26,28 Revere, Evans, 3 Rhee Syng-man, 145, 147, 153, 197n47, 198nn55,56 Ricardo, David: on principle of comparative advantage, 33 Rice, Susan, 59 Rickli, Jean-Marc, 101 Rinehart, Ian E., 201n3 Risse-Kappen, Thomas, 69, 183n27 Roh Moo-hyun, 22, 55, 146–48, 185n12, 198n65, 199n70; attitudes regarding U.S.-South Korea alliance, 3, 4, 112, 117, 145, 146–47, 150, 152, 153, 163, 198n59; policy regarding Iraq War, 103; relationship with George W. Bush, 152, 199n70 Roh Tae-woo, 146, 152 Romania, 20, 37, 96, 190n26 Rose, Gideon, 181n48, 182n3 Rosenbluth, Frances, 141 Ross, Robert, 5, 49, 50, 103, 109, 110, 172 Rothkopf, David, 193n62 Rubottom, Richard, 181n46, 199n75 Rumsfeld, Donald, 98, 201n7 Russett, Bruce, 179n23 Russia, 80–81, 186n16; and NATO, 7; number of alliance agreements, 96; relations with China, 6, 57, 192n48; relations with Great Britain, 20, 37; relations with Japan, 31; relations with North Korea, 179n28; relations with South Korea, 17, 103; relations with United States, 104; and Ukraine, 102, 171. See also Soviet Union Sabrosky, Alan Ned, 43 Sacko, David, 39 Sakamoto, Kazuya, 53 Samuels, Richard, 16, 131, 192n58

index 235 Sandler, Todd, 179n23, 183n26 Sapolsky, Harvey M., 171 Sarantakes, Nicholas Evan, 134, 195n14 Saudi Arabia: relations with United States, 186n14 Saunders, Phillip, 187n33 Savun, Burcu, 23, 71, 78, 82, 83, 181n50, 184n37, 188n1; on abrogation of alliances, 44, 54, 77, 79 Schaller, Michael, 181n46, 195n20; on U.S. Japan relations, 126, 127, 128, 130, 134, 137, 194n7, 195n11 Schelling, Thomas, 114, 162, 181n46; on strength of weakness strategy, 122, 123 Schroeder, Paul, 32, 173, 178nn14,18, 185n9 Schultz, George, 156 Schultz, Kenneth, 47, 169, 193n3, 194n15 Schweizer, Karl, 20, 36 Schweller, Randall, 178n14, 180n40, 181n48, 182nn3,6, 184n35 Sciubba, Jennifer Dabbs, 44, 184n37 Scobell, Andrew, 50, 109, 171 Scowcroft, Brent, 191n35 secondary alliance security dilemma, 44–45 security: vs. autonomy, 182n8; price of, 15–16, 18, 51–52, 74, 88, 95–97; quantity of, 52, 74, 88, 92–94; supply and demand regarding, 1–2, 3, 15–16, 17–19, 18–19, 24, 26, 28, 45–46, 50–52, 60, 74, 75, 88–94, 95, 95–98, 100–101, 102, 167, 168, 170–71, 172, 184n35, 189n18 self-reliance, 64, 65, 66, 67 Senkaku/Diaoyu Islands, 101, 171, 201n3 Seo, Jungkun, 185n12 September 11th attacks, 100, 140, 188n39 Serbia, 71 Sewol sinking, 48 Shambaugh, David: China Goes Global, 56 Shanghai Cooperation Organization, 6, 57 Shaw, L. M. E., 182n9 Shenkar, Oded, 181n49 Shigemitsu, Mamoru, 138 Shin, Doh Chull, 197n48 Simon, Michael, 184n37 Simon, Sheldon, 58 Singer, J. David, 46, 77, 78, 82, 180n36 SIPRI, 16, 100, 104 Siverson, Randolph, 43, 45, 72, 184n37 Skalnes, Lars, 37, 41 Slovakia, 96 Small, Melvin, 78 Smith, Alastair, 184n29, 193n10, 194n15

Snidal, Duncan, 43 Snyder, Glenn H., 1, 50, 164, 178n14, 179n27, 184n35; on abandonment, 24, 63, 68, 69, 180n41; on alliance games and adversary games, 74, 109, 180n41, 182n5, 184n2, 184n30; on bipolarity, 55, 73, 180n40, 180n41, 187n38, 188n43; definition of alliances, 30; on entrapment, 24, 63, 68; on multipolarity, 55, 72, 180n40, 182n10, 184n36; on secondary alliance security dilemma, 44–45; on systemic polarity and exchange, 18; on unipolarity, 55, 74, 180n40 Snyder, Jack, 181n2 Sohn, Yul, 20, 185n12 Somalia, 188n44 Sotooka, Hidetoshi, 1, 99, 131, 191n35 South Africa, 65 South East Asia Treaty Organization (SEATO), 72 South Korea: during Cold War, 1, 4, 103, 149; Combined Forces Command (CFC), 148–49; domestic politics, 4, 22, 38, 48, 117, 118, 123, 124, 142–53, 162, 173, 197nn47–51, 198nn55,65, 200n84; economic conditions, 17, 103, 144, 145, 148, 149, 152, 153, 191n44, 198n52; Free Trade Agreement (FTA), 20, 54–55; Gwangju massacre, 143; host nation support (HNS), 149, 183n18; during Iraq War, 103, 112; leaders’ attitude regarding alliance cooperation, 145–47, 153, 165, 198n59; leaders’ vulnerability, 147–48, 153, 173, 198n62; Ministry of Foreign Affairs and Trade, 55; missile defense, 59, 186n21; National Assembly, 147, 148; opposition to alliance cooperation in, 142–45, 150, 152, 153, 162, 197n51; relations with China, 4, 17, 31, 48, 55, 59–60, 102, 103, 104, 172, 186n20, 191n44; relations with Japan, 59, 198n58; relations with North Korea, 1–2, 3–4, 17, 25, 31, 38, 48, 102–3, 144, 145, 146, 147–48, 153; relations with Russia, 17, 103; relations with United States, 1, 3–4, 14, 17, 18, 20, 21, 25, 31, 34, 38, 47, 48, 49, 54–55, 58, 59, 65, 91, 102–4, 107–8, 110–11, 112, 113, 117, 121, 122, 123–24, 138, 142–53, 162, 163, 165, 166, 172, 173, 174, 181n46, 185nn11,12, 186n16, 191nn45,46, 192nn53,55,56, 197nn47,50,

236 index South Korea (continued) 198nn56,57, 200n84; Status of Forces Agreement (SOFA), 146, 150, 152; 386 generation, 144; during Vietnam War, 191n46 South Yemen, 6, 96 Soviet Union, 25, 28, 56, 59, 124, 181n42, 187n37, 189n17, 190n19; during Cold War, 14, 16, 17, 21, 26, 35, 40, 46, 73, 80–81, 87, 89, 168, 171, 179n28, 188n44; collapse of, 6, 46, 74, 97, 98, 179n29; nuclear weapons, 126; relations with China, 16, 57, 179n30; relations with Japan, 35, 126; relations with North Korea, 179n28; relations with Vietnam, 94. See also Cold War; Russia Spain: Communist Party, 160; domestic politics, 117, 123, 124, 153–62, 173, 200n84; economic conditions, 156, 158–59; during Iraq War, 101–2, 117, 156, 158, 162, 200n89; leaders’ attitudes regarding alliance cooperation, 153–56, 162; leaders’ vulnerability, 157–58, 160, 162; Moron Air base, 162; and NATO, 154, 156, 160, 161, 199nn79–81; number of alliance agreements, 96; opposition to alliance cooperation in, 153–55, 156, 158, 159, 160–61, 162; People’s Party (PP), 117, 156, 157–58; relations with France, 54; relations with Great Britain, 54; relations with United States, 14, 24, 25, 34, 47, 53, 101–2, 112, 113, 117, 121, 122, 123–24, 150, 153–62, 163, 166, 173, 181n46, 199n76, 200nn82–84,89,91; Rota Naval Station, 161–62, 200n91; Socialist Workers’ Party (PSOE), 117, 156, 157, 158, 160–61; Union of the Democratic Center (UCD), 157, 160, 161, 162 Spruyt, Hendrik, 43, 183n19 Stackpole, Henry, 21 Stam, Allan, 13 Stanton, Sheldon, 199n79 Starr, Harvey, 184n37 Steinberg, David, 197n49 Stone, Laura, 38, 137 Storey, Ian, 182n7, 186n18 Stromseth, Jonathan, 115 Stuart, Douglas, 201n6 Stuckey, John, 82, 180n36 Stueck, William, 191n45 subsidy treaties, 15, 32, 36, 37, 179n24

Suez Crisis of 1956, 71 Suh, Jae-Jung, 182n2 Suhrke, Astri, 198n57 Sullivan, John, 40 Sunohara, Tsuyoshi, 128, 137, 192nn57,58, 196n31 Suzuki, Zenko, 35 Swaine, Michael, 171 systemic polarity, 69–76, 179n32, 180nn40,41, 184n34, 188n42, 201n5; poles of international system, 50, 55–58, 61–62, 69; relationship to duration of alliances, 80–85; relationship to fear of abandonment, 24, 26, 44, 46, 50, 51, 62–67, 68, 69–71, 97, 102, 179n33, 187n35; relationship to fear of entrapment/entanglement, 24, 26, 46, 50, 51, 62–64, 67–68, 70, 97, 179n33; relationship to state bargaining power, 1, 3, 14–15, 18–19, 21–22, 24, 26, 28, 29, 44, 50, 61–62, 69–76, 99, 172; relationship to termination of alliances, 77–80; and relative capabilities, 62, 63–76, 82, 187nn26,28. See also bipolarity; multipolarity; unipolarity Tago, Atsushi, 185n8 Taiwan: relations with United States, 24, 65, 188n41 Taiwan Relations Act, 188n41 Taiwan Strait crises, 126 Tajikistan, 57 Takemura, Masayoshi, 133 Takeshita, Noboru, 127 Takeuchi, Hiroki, 101 Tamamoto, Masaru, 100 Tanaka, Akihiko, 131 Tarnoff, Curt, 88 Tatsumi, Yuki, 181n45 Telser, L. G., 169 Tertrais, Bruno, 39, 106, 189n12 Thailand, 186n16 Thalakada, Nigel, 46, 178n11 Thirty Years’ War, 37 Thompson, Alexander, 55, 191n42 Thompson, William, 180n40, 184n34 Tilly, Charles, 181n47 Tojo, Hideki, 129, 134 Tomz, Michael, 183n25 Tow, William T., 201n6 transaction cost economics, 43

index 237 Trans-Pacific Partnership (TPP), 20 Triple Entente, 94 Tsebelis, George, 116, 193n7 Tsuchiyama, Jitsuo, 2, 99, 126 Tucker, Nancy Bernkopf, 191n45 Tulchin, Joseph, 199n81 Tunisia, 177n5 Turkey, 30–31, 89 Tusell, Javier, 199n78 Ukraine, 102, 171 unipolarity, 180nn40,41, 184n36, 189n2, 190n29; and inattention, 49, 70, 167; increase in commitment of nongreat powers to alliances (Hypothesis 3.5), 75–76, 95–98; vs. multipolarity, 69, 71–72, 74, 75, 77–80, 110–11, 187n35; reduction in alliance commitments in (Hypothesis 3.4), 75, 88–92; removal of restraining arrangements in (Hypothesis 3.6), 76, 105–9; selectivity in alliance commitments in bipolarity vs. (Hypothesis 3.3), 73–74, 75, 85–88; and United States, 4, 21–22, 24, 28, 32, 46, 49, 50, 52, 60–61, 70, 74, 85–92, 94, 95–110, 168, 169, 170–71, 173, 174, 189n18, 200n89; vs. bipolarity, 4–5, 6, 19, 21–22, 28, 32, 46, 49, 55, 58, 60–62, 69, 72, 73–74, 75, 85–88, 95–99, 103, 108, 110, 166–67, 168, 172–73 United States: alliance policy, 7, 14, 37, 38, 52, 58–60, 61, 70, 72, 85–92, 93, 96, 99, 105–6, 107, 110, 123–24, 137, 172, 177n7, 188n44, 189n8; arms industry in, 40; George W. Bush administration, 89, 101–2, 106, 111, 140, 141, 144–45, 146, 152, 153, 158, 170, 199n70, 200n89; CIA, 126, 195n21; coalition building by, 21; after Cold War, 2, 3, 6, 7, 17, 18–19, 28, 29, 32, 49, 51–52, 54, 61, 72, 74, 85–92, 93, 95–97, 99–101, 106, 107, 190n26; during Cold War, 1, 3, 4, 21, 40, 46, 53, 57, 58, 60, 73, 80–81, 85–88, 93, 103, 106, 107, 108, 114, 117, 124, 125, 137, 149, 157, 168, 171, 180n40, 192n49, 193n63; domestic politics, 24, 123, 124, 163, 165, 171, 173, 174, 193n61; economic conditions, 36, 99, 105, 106, 111, 163, 171–72, 183n21, 189n18, 193n63; Lend-Lease agreement, 37; military capabilities, 56–57, 171; National

Security Council (NSC), 55, 124–25; and NATO, 6, 7, 14, 30–31, 86, 87, 89, 91–92, 96, 101, 107, 189n16, 192n52, 193n61; Nixon administration, 16, 128, 137, 146, 155–56, 160, 163, 181n42, 185n6, 200n82; nuclear weapons of, 56, 139, 185n6, 187n30, 200n91; number of alliance agreements, 95–96; Obama administration, 102, 110, 111, 132, 141, 144–45, 153, 177n5; rebalancing toward Asia, 110, 172, 193n62, 201n6; relations with Australia, 185n11; relations with Chile, 118; relations with China, 1–2, 5–6, 14, 16, 26, 27, 49, 50, 58–60, 65, 103, 104, 108, 109, 110, 137, 138, 166, 168, 170, 171–73, 174, 179n26, 185n6, 187n41, 192n48, 201n3; relations with Djibouti, 58–59; relations with France, 71, 102, 189n14; relations with Germany, 39, 52, 91, 102, 131, 185n7, 189n16; relations with Great Britain, 37, 71; relations with India, 58; relations with Indonesia, 58, 186n18; relations with Iran, 105, 121, 187n31; relations with Israel, 5, 89, 190n19; relations with Japan, 1, 2–3, 14, 15, 16, 20, 21, 25, 31, 34–36, 38, 39, 45, 47, 49, 52, 53, 54, 58, 91, 98–101, 103, 106, 108–9, 110–11, 112, 113, 114, 117, 120, 121, 122, 123–42, 150, 153, 159, 162, 163, 165, 166, 172, 173, 174, 179n26, 180n39, 181nn45,46, 183n17, 185n6, 186n14, 191nn31,35,38, 193n63, 194nn18,2,7, 195nn14,20,21,25, 196nn31,34, 197n43, 198n54, 201nn3,7; relations with Kuwait, 97, 181n42; relations with Liberia, 188n44; relations with New Zealand, 187n30; relations with North Korea, 1, 48, 103, 105, 108, 144, 145, 146, 150, 153; relations with Peru, 181n42; relations with Philippines, 31–32, 88, 115, 179n27, 193n5; relations with Poland, 102; relations with Russia, 104; relations with Saudi Arabia, 186n14; relations with South Korea, 1, 3–4, 14, 17, 18, 20, 21, 25, 31, 34, 38, 47, 48, 49, 54–55, 58, 59, 65, 91, 102–4, 107–8, 110–11, 112, 113, 117, 121, 122, 123–24, 138, 142–53, 162, 163, 165, 166, 172, 173, 174, 181n46, 185nn11,12, 186n16, 191nn45,46, 192nn53,55,56, 197nn47,50, 198nn56,57, 200n84;

238 index United States (continued) relations with Spain, 14, 24, 25, 34, 47, 53, 101–2, 112, 113, 117, 121, 122, 123–24, 150, 153–62, 163, 166, 173, 181n46, 199n76, 200nn82–84,89,91; relations with Taiwan, 24, 65, 188n41; relations with West Germany, 38; relation with Uzbekistan, 188n39; and unipolarity, 4, 21–22, 24, 28, 32, 46, 49, 50, 52, 60–61, 70, 74, 85–92, 94, 95–110, 168, 169, 170–71, 173, 174, 189n18, 200n89. See also Afghanistan; Iraq War; U.S. military bases; Vietnam War U.S.–Japan Joint Declaration on Security, 100, 140 U.S.–Japan Security Treaty, 114, 126, 127–28, 134–35, 138–39, 171; Far East clause (Article 6), 100 U.S. military bases, 89–91, 106–9, 172, 192n49; in Djibouti, 58–59; in Germany, 91, 149; in Italy, 192n52; in Japan, 15, 35, 91, 107, 108–9, 117–18, 120, 126, 129, 131–32, 136, 138, 139, 140, 141–42, 173–74, 174, 183n18, 185n6, 192nn58,59, 195n9, 196n31, 197n43, 201n7; in Philippines, 32, 115, 193n5; in South Korea, 91, 107–8, 144, 149–50, 151, 173–74, 192n53, 199n69; in Spain, 154, 156, 158, 159, 160, 161–62, 199n76, 200nn84,91 U.S.–South Korea Mutual Defense Treaty, 150 U.S. Trade Representative (USTR), 55 Uzbekistan, 57, 96, 188n39 Van Ness, Peter, 57 Vietnam, 94, 179n27, 186n16 Vietnam War, 53–54, 128, 137, 173, 185n7, 191n46 Viñas, Angel, 53, 112, 155, 156, 159, 160, 161, 181n46 Vine, David, 192n52 Viner, Jacob, 20, 37 Vogel, Jeremy S., 23, 44, 79, 82, 83, 164–65 Voice, 132 Vonnahme, Greg, 181n50 Wallander, Celeste, 178nn11,14,17, 181n2 Walt, Stephen, 1, 23, 46, 168, 171, 174, 178nn11,14, 181n2, 184n35; on

abandonment/entrapment dilemma, 187n35; on balancing, 104, 182n6; on bipolarity, 187n35; on formation of alliances, 30; on multipolarity, 187n35; on unipolarity, 184n36, 187n35 Waltz, Kenneth, 33, 167, 192n47; on adversary games, 45; on balancing, 182n6; on bipolarity, 72, 73, 169; on end of Cold war, 5, 32, 94, 97, 178n11; on formation of alliances, 30; on systemic polarity, 18, 49, 50 Wang Yiwei, 186n20 Ward, Celeste Johnson, 107 War on Terror, 2, 106, 110, 188n39 Warsaw Pact, 39, 57, 73 Wayman, Frank Whelon, 187n26 Weart, Spencer, 184n37 Weber, Katja, 188n45 wedge strategy, 182n5 Weeks, Jessica, 183n25 Weitsman, Patricia, 62, 173, 178n14, 185n9, 188n42 Werlin, Herbert, 149 Whitaker, Preston, 53, 154 White, Hugh, 56 Widen, J. J., 53, 185n7 Wight, Martin, 177n9 WikiLeaks, 22, 136 Williamson, Oliver, 42–43, 184n28, 188n45 Wilson, Peter H., 179n24 Wirth, Christian, 183n17 Wohlforth, William, 21, 26, 46, 55, 104, 168, 185n11, 189n2, 201n2 Wolf, Charles, 91, 183n18, 192n52 Wolford, Scott, 181n47, 194n15 Womack, Brantly, 58 World War I, 71, 94 World War II, 37, 46, 126, 134, 154, 157, 189n5 Xi Jinping, 104 Yamaguchi, Jiro, 194n6 Yamaoka, Kenji, 134 Yan, Xuetong, 5, 59, 166, 186n17 Yeonpyeong Island, 3, 38 Yoda, Tatsuro, 91, 183n18, 191n32 Yokosuka military base, 138 Yomiuri Shimbun, 128 Yongsan, Seoul, 174

index 239 Yoshida, Shigeru, 127, 129, 130, 138; Yoshida Doctrine, 99 Yoshida, Shingo, 128, 137, 197n43 Yoshitsugu, Kosuke, 36 Yugoslavia, 57 Yugoslavia, former, 190n21

Zagoria, Donald, 179n30 Zanini, Michele, 91, 183n18, 192n52 Zapatero, Jose Luis Rodriguez, 158 Zeckhauser, Richard, 15, 177n1 Zhang, Feng, 14, 16, 57–58, 192n48 Zimmermann, Hubert, 183n27 Zorn, Christopher, 83

Studies in Asian Security Amitav Acharya, Chief Editor, American University David Leheny, Chief Editor, Princeton University Protests Against U.S. Military Base Policy in Asia: Persuasion and Its Limits By Yuko Kawato 2015 How India Became Territorial: Foreign Policy, Diaspora, Geopolitics By Itty Abraham 2014 Wronged by Empire: Post-Imperial Ideology and Foreign Policy in India and China By Manjari Chatterjee Miller 2013 Looking for Balance: China, the United States, and Power Balancing in East Asia By Steve Chan 2012

sponsored by the east-west center, 2004–2011 Muthiah Alagappa, Founding Series Editor Rethinking Japanese Public Opinion and Security: From Pacifism to Realism? By Paul Midford 2010 The Making of Northeast Asia By Kent Calder and Min Ye 2010 Islam and Nation: Separatist Rebellion in Aceh, Indonesia By Edward Aspinall 2009 Political Conflict and Economic Interdependence across the Taiwan Strait and Beyond By Scott L. Kastner 2009

(Re)Negotiating East and Southeast Asia: Region, Regionalism, and the Association of Southeast Asian Nations By Alice D. Ba 2009 Normalizing Japan: Politics, Identity, and the Evolution of Security Practice By Andrew L. Oros 2008 Reluctant Restraint:The Evolution of China’s Nonproliferation Policies and Practices, 1980–2004 By Evan S. Medeiros 2007 Why Taiwan? Geostrategic Rationales for China’s Territorial Integrity By Alan M. Wachman 2007 Beyond Compliance: China, International Organizations, and Global Security By Ann Kent 2007 Dangerous Deterrent: Nuclear Weapons Proliferation and Conflict in South Asia By S. Paul Kapur 2007 Minimum Deterrence and India’s Nuclear Security By Rajesh M. Basrur 2006 Rising to the Challenge: China’s Grand Strategy and International Security By Avery Goldstein 2005 Unifying China, Integrating with the World: Securing Chinese Sovereignty in the Reform Era By Allen Carlson 2005 Rethinking Security in East Asia: Identity, Power, and Efficiency Edited by J. J. Suh, Peter J. Katzenstein, and Allen Carlson 2004