Abstract

Since 1990, China has used coercion in its maritime territorial disputes in the South China Sea, despite adverse implications for its image. China is curiously selective in its timing, targets, and tools of coercion: China rarely employs military coercion, and it does not coerce all countries that pose similar threats. An examination of newly available primary documents and hundreds of hours of interviews with Chinese officials to trace the decisionmaking processes behind China's use and nonuse of coercion reveals a new theory of when, why, and how China employs coercion against other states, especially in the South China Sea. Contrary to conventional wisdom, the findings show that China is a cautious bully that does not use coercion frequently. In addition, when China becomes stronger, it tends to use military coercion less often, choosing instead nonmilitary tools. Moreover, concerns with its reputation for resolve and with economic cost are critical elements of Chinese decisionmaking regarding the costs and benefits of coercing its neighbors. China often coerces one target to deter others—“killing the chicken to scare the monkey.” These findings have important implications for how scholars understand states' coercive strategies and the future of Chinese behavior in the region and beyond.

Introduction

Claiming all land features in the South China Sea, China has maritime territorial disputes with Brunei, Malaysia, the Philippines, and Vietnam, and maritime jurisdictional disputes with Indonesia. To manage these disputes and shape the behavior of its neighbors and rivals, China has utilized the full spectrum of coercive tools, including diplomatic sanctions, economic sanctions, and military coercion. For example, in 1995, China seized the Mischief Reef, also claimed by the Philippines. In 2012, it banned Philippine banana exports to China after the Philippines arrested Chinese fishermen in the disputed Scarborough Shoal. In 2014, it used maritime law enforcement vessels to ram Vietnamese vessels in response to Vietnam's opposition to Chinese oil rigs operating in the disputed Paracel Islands. When, why, and how does China use coercion against other states over disputes in the South China Sea? By “coercion,” I mean the use or threat of negative actions such as economic sanctions and military means to force the target state to change its behavior. In the South China Sea disputes, China engaged in military coercion in the 1990s, refrained from coercion from 2000 to 2006, and resumed coercion after 2007. Since 2007, however, China has largely used nonmilitarized coercive tools, which is striking, considering the quantitative and qualitative advances in its military capabilities in recent years. When China decides to pursue coercion, why does it opt for military coercion in some cases and nonmilitary coercion in others?

Identifying and explaining broad patterns regarding when and how China coerces other states in the South China Sea has both theoretical and policy relevance. First, although the literature on coercion is vast, the focus has been on evaluating its effectiveness.1 There is, therefore, ample room to theorize when states decide to use coercion and what influences their choice of coercive tools, especially for rising powers such as China.2 Moreover, the literature examines the effectiveness of individual coercive tools, yet the question of when and why states choose one coercive measure over another has been understudied.3 Second, explaining Chinese coercive behavior has policy implications for managing China's rise and maintaining peace in the Asia Pacific. It sheds light on how contemporary rising powers try to translate their power into influence and on their choice of policy instruments.4 It also illuminates how decisionmakers in Beijing craft security policies for a state likely to become one of the most significant great powers in the twenty-first century. Moreover, experts on Chinese foreign policy have hotly debated whether China is becoming more assertive. To date, this debate has lacked systematic coding of how to measure assertiveness.5 Third, the literature on rising powers has not examined the question of how they use coercion, focusing instead on theories of war and peace.6 China's increasing power offers an intuitive explanation for its use of coercion—that is, when states are more powerful, they become more coercive—but the evidence suggests otherwise. China used military coercion in the 1990s, when it was weaker than in other periods, but chose not to use military coercion when it grew stronger. Fourth, this article examines disputes in the South China Sea. Those disputes are potential threats to regional security, because they increase the risk of armed conflict between China and the United States while endangering U.S. national interests, including freedom of navigation and the credibility of U.S. treaty commitments to allies in the region.

In this article, I develop a theory of coercion that emphasizes the expected costs and benefits to the state in choosing to coerce or not coerce and in choosing one coercive tool over another in response to national security threats. This “cost-balancing theory” helps explain, first, when and why China coerces one target to deter other potential challengers. Second, it suggests that China is more likely to use coercion when the need to establish a reputation for resolve is high and the economic cost is low. “Economic cost” here refers to the extent to which China needs the target state for markets or supply. A reputation for resolve is the resolve a state demonstrates for defending its national security interests. Third, my theory posits that China prefers to use nonmilitarized coercive tools when the geopolitical backlash cost is high. The term “geopolitical backlash” as used in this article refers to concerns of the coercing state that the target state might balance against it.

My study yields three key findings. First, contrary to the conventional wisdom, China is a cautious bully. Second, China employs coercion only infrequently. Third, when it becomes stronger, it uses military coercion less often, instead resorting to mostly nonmilitarized tools. Therefore, decisions about when to pursue coercion and which tools to use cannot be explained by focusing on material capabilities. My theory highlights the centrality of the need to establish resolve and concerns about economic cost. China coerces one target to deter others.

In the next section, I describe the full spectrum of coercive tools available to states. In the second section, I develop my cost-balancing theory of coercive behavior and describe my research design and measurement of the variables. In the third section, I conduct a congruence test to explain Chinese coercive patterns in the South China Sea from 1990 to 2015, introducing original data on maritime disputes and Chinese coercive behavior. The fourth section presents a case study of the 2012 Scarborough Shoal incident, drawing on exclusive interviews with Chinese officials and internal Chinese documents to illustrate the causal mechanisms at work in my theory. I conclude with a discussion of some of the implications of my study for researchers and policymakers concerned with the role of coercion in international relations and rising powers, as well as with U.S.-China relations in the South China Sea.

Full Spectrum of Coercive Measures

The classic definition of coercion comes from Thomas Schelling, who uses the term “compellence” when describing a strategy designed to make an adversary act in a particular way; the strategy usually involves the use of punishment until the enemy acts in the desired manner.7 Robert Art and Patrick Cronin further specify that in coercive diplomacy, the coercer compels the adversary either to start doing something it is not doing or to stop doing something it is doing.8 Strictly speaking, the concept of interest here is compellence, but the term “coercive diplomacy” has become the convention.9 Therefore, I use the term “coercion,” not “compellence,” but broaden the scope of compellence to include military and nonmilitary coercive tools.

Following the literature, I define “coercion” as the threat or use of negative actions by a state to demand a change in the behavior of another state. I consider both physical actions and threats, yet maintain that all else being equal, physical actions should signal resolve more credibly than threats of action. Coercion has two goals: to make the target either stop action it is taking or to take new action. An attempt to coerce an adversary should make clear the kind of action the coercer wants the target to take. Aggression or brute force, whose end goal is to take a piece of land rather than make the target state do something, is not coercion.10 In this article, coercion falls along a spectrum. At one end is inaction: the decision by a state not to take physical action, even when it has the ability to do so. Instead, it may resort to rhetorical protest or remain silent, both of which constitute inaction. Inaction is forbearance. At the other end of the spectrum is military coercion.

Diplomatic sanctions constitute the coercer's deliberate interruption of its relations with the target state. Tara Maller codes the following as diplomatic sanctions: the short, temporary recall of the ambassador to the target state, a downgrade in diplomatic status, and closure of the embassy—the first being the least severe and the last being the severest.11 The complete break of bilateral relations, however, affects the ability to gather intelligence and the ease of communication.12 As such, states may choose to maintain some level of relations—for example, closing consulates, canceling important meetings, or terminating senior-level communications.

“Economic sanctions” refer to instructions by the government to certain actors to withdraw from trade or financial relations so as to force the target to change a foreign policy the coercer dislikes.13 Trade sanctions include embargos, increases in tariffs, withdrawal of most-favored-nation status, quotas, blacklisting, denial of licenses, preclusive buying, and other discriminatory actions. Financial sanctions include the freezing of assets, aid suspension, expropriation, unfavorable taxation, and the imposition of controls on the import or export of capital.14 Being strictly nonmilitary, diplomatic and economic sanctions can be used to send signals to the target state while minimizing the risk of escalation.

Another form of coercion—so-called gray-zone coercion—straddles nonmilitary and military coercion and has attracted growing attention in recent years.15 According to Michael Mazarr, states involved in gray-zone conflict employ “civilian instruments to achieve objectives sometimes reserved for military capabilities.”16 Like others, however, Mazarr's conceptualization is too expansive and includes military force.17 Some military actions might be non-kinetic, but they are still militarized. Emphasizing the civilian aspect, I define gray-zone coercion as physical violence by government agencies to force the target state to change its behavior. Similar to military coercion, gray-zone coercion can cause tangible damage to the target. Covert actions conducted by the Central Intelligence Agency during the Cold War are examples of gray-zone coercion. Also, gray-zone coercion is analytically distinct from military coercion, because it is imposed by civilians and the instruments involve much smaller capabilities than those available to the military. Being nonmilitarized, gray-zone coercion is useful for escalation control, because it allows states to claim plausible deniability: states can deny that they are using military force, thus reducing the likelihood of military escalation triggered by defense treaty commitments. Finally, if the coercer can prevail with gray-zone coercion, its incentives to use military force are reduced.

Military coercion represents the most escalatory level of coercion. Chas Freeman divides military coercion into two categories: (1) the nonviolent use of military power and (2) the use of force.18 Following Freeman, I define “military coercion” as consisting of the display, threat, and use of force short of war. Nonviolent military actions include shows of force, such as temporary deployments, military exercises, and naval visits.19 Such displays of force could emphasize the possibility of escalated and intensified confrontation.20 Acts of military coercion are “physical and so menacing that the threat of hostile intent is implicit in their use.”21 They also risk escalation into militarized conflicts.

The Cost-Balancing Theory

In this section, I describe the cost-balancing theory and how it explains when and why states use coercion. The core benefit of coercion is that it demonstrates the coercing state's reputation for resolve: other states view it as credible. The costs can be economic—the loss of markets or supply—or geopolitical—balancing behavior by the target or other states against the coercer (i.e., geopolitical backlash).

ISSUE IMPORTANCE

When devising national security policies, states weigh the importance of the issues at hand. Taylor Fravel notes that states are more likely to escalate to the use of force when the conflict involves territory they value highly.22 Vesna Danilovic emphasizes the relevance of stakes—the importance of the issue—in states' use of deterrence.23 As the crisis bargaining literature suggests, an actor will be resolved about certain issues when the stakes are high.24

States' logic for choosing coercion is similar. Threats to national security are, by definition, high-stakes issues. Not every national security issue, however, is weighted equally. As such, states use coercion for issues they consider to have high importance and not for issues they consider of less importance.25 When the state considers the issue to be of the highest importance, it may resort to coercion. Thus, when the need to establish a reputation for resolve and economic cost are both high, states use coercion only for issues of the highest importance.

Nevertheless, there are still temporal and cross-national variations as to when and against whom a state uses coercion, even for the same issue. That is, issue importance does not dictate when states decide to use coercion: the importance of the issue varies across issues but remains constant for the same issue. Even for the same issue, the state chooses to coerce in certain periods and target certain countries, but not others. This is when the specific benefits and costs of coercion become critical.

BENEFITS OF COERCION: THE NEED TO ESTABLISH A REPUTATION FOR RESOLVE

The intended benefit of coercion is external; that is, other states view the coercer as having resolve. States take coercive measures to achieve specific goals, yet I argue that they do so not just to influence the target. States fear that if they do not use coercion, other states might not consider them credible, instead viewing them as weak, which might lead to failure in deterring future aggression. In Joshua Kertzer's words, resolve is “a state of firmness or steadfastness of purpose.”26 Reputation, according to Jonathan Mercer, involves a “judgment of someone's character (or disposition) that is then used to predict or explain future behavior”; a reputation is formed when an observer uses “dispositional or character-based attributions” and “past behavior to explain or predict another's behavior.”27

This focus on past behavior is found in the work of Schelling, who argues that to be convincing, commitments should be backed by precedents.28 In addition, actions are more credible and less ambiguous than rhetoric.29 Robert Jervis similarly notes that issues of little intrinsic value can become indices of resolve.30 In this sense, both Schelling and Jervis suggest that states use coercion to signal their commitment to defend their national security.31

This logic of establishing a reputation for resolve is in line with recent scholarship. Nicholas Miller argues that economic sanctions imposed by the United States on some of its allies that were pursuing nuclear proliferation deterred other potential proliferants.32 In explaining why states tend to be sincere when engaging in diplomacy rather than resort to bluffing, Anne Sartori argues that a state that has a reputation for bluffing will experience reduced credibility when it issues future deterrent threats.33 In an experiment, Dustin Tingley and Barbara Walter find that the majority of their participants “invest more heavily in reputation building if they believe a game will be repeated many times.”34 Todd Sechser notes that compliance with a coercive threat entails reputation costs for the target: it raises the possibility that the coercer will make additional demands in the future, thus leading to compellence failure.35 Although Sechser focuses on explaining when compellent threats are ineffective, one can also apply the logic of reputation costs to the coercer. Indeed, Walter finds that governments might fight civil wars against secessionist groups to look tough and discourage other rebel groups from making demands.36 As such, empirical evidence suggests that the logic of establishing a reputation for resolve manifests itself in international relations.

In sum, a state's need to establish a reputation for resolve goes beyond a particular incident. It has implications for other issues and for the state's reputation vis-à-vis other states more generally. States use coercion to deter other states from engaging in undesired activity. However, just because states perceive the need to establish resolve does not mean that they will automatically gain resolve when they engage in coercive behavior.37 My article focuses on explaining states' coercion decisions, not on whether states gain resolve based on these decisions.38

ECONOMIC COST OF COERCION

Potential coercers may worry about incurring domestic economic costs resulting from economic dependence on the target state or the loss of markets. Coercion may generate economic costs for both the coercer and the target, affecting their bilateral trade or capital flows. Albert Hirschman argues that commerce can be an alternative to war only when it is “extremely difficult” for the target to replace the coercer as a market and a source of supply with other countries; that is, it has no “exit options.”39 If the coercer has exit options but its target does not, then it has leverage. Building on this argument, Robert Keohane and Joseph Nye use vulnerability dependence to indicate the “costliness of making effective adjustments to a changed environment.”40 In other words, how costly is it for the coercer to find exit options? A state is less likely to take coercive measures if it is dependent on the target for markets or supply or if it is concerned about losing important markets, even if it does not depend on the target.

Like war, coercion can be economically costly.41 As Scott Kastner notes, political conflicts short of war have an impact on economic relations, the Cold War being one example.42 State leaders may place restrictions on foreign economic ties in the face of political conflict, and conflicts could make it riskier for firms to operate in affected countries.43 Coercion is a form of political conflict. There is no theoretical expectation that the target state will respond to political conflicts “in kind”—that is, diplomatic measures for diplomatic coercion and military measures for military coercion. For example, diplomatic sanctions might stall bilateral economic relations, especially if both sides rely on bilateral meetings of senior government officials to negotiate significant purchases (e.g., aircraft) or investment deals. Military coercion may reduce the incentive for companies in the target state to invest in the coercing state. Therefore, if a state needs the target state for markets or critical supply, it will consider the potential economic cost of using coercion.

GEOPOLITICAL BACKLASH COST OF COERCION

Threats and negative actions can be self-defeating if they elicit counteraction from the other side, thereby setting in motion a costly cycle.44 One such cost could be a geopolitical backlash, in which the target state seeks to balance the coercer by creating or aggregating military power through internal mobilization or the formation of alliances.45 Stephen Walt argues that states tend to balance against threats rather than bandwagon and that larger states balance more often than smaller ones.46 Therefore, if a state is aware that acts of coercion may be interpreted as threats, it will be concerned about geopolitical backlash—the target might side with other states to balance the coercer.

The reason why geopolitical backlash cost influences the state's choice of coercive tools is that greater geopolitical pressure could potentially trigger a military alliance. What I mean by economic cost influencing coercion decisions and geopolitical backlash cost affecting coercive tools is that, theoretically, economic cost should be the most critical factor for whether a state uses coercion, if at all, whereas geopolitical cost is most relevant for what coercive tools states choose.

SYNTHESIS AND PREDICTIONS—A COST-BALANCING THEORY

In an ideal world without economic or geopolitical constraints, a state could take coercive action to increase its reputation for resolve any time it faces a challenge. In reality, however, states face economic and geopolitical constraints that force them to balance the costs of coercion against the need to establish resolve. This calculation accords with Kertzer's argument that “risk aversion increases sensitivity to both the costs of fighting and the costs of backing down.”47 Sometimes, states have to take the middle path and make cost-balancing calculations.

THE DECISION TO PURSUE COERCION. States will refrain from using coercion when the need to establish a reputation for resolve is low. For issues of similar importance, states engage in coercion when that need is high and when the economic cost is low. In circumstances when both the need to establish resolve and the economic cost are high, states will engage in coercive behavior only when the issue is of the highest importance. This article focuses only on the South China Sea, where issue importance is held constant.48

MILITARY VERSUS NONMILITARY COERCION. In theory, military coercion is more escalatory than are other forms of coercion. I hypothesize that states will be cost conscious and optimizing; that is, they will maximize the utility of coercion while minimizing the cost. States tend to prefer to use nonmilitarized tools of coercion, especially when the geopolitical backlash cost is high.

WEIGHING THE COSTS AND BENEFITS OF COERCION

The cost-balancing theory identifies a state's need to establish a reputation for resolve and the potential costs and benefits that may be involved. First, although changing the behavior of the target state is a benefit of coercion, as the signaling and reputation literature indicates, the expected benefit of coercion centers on the reputation for resolve; an example is U.S. concern about the credibility of its resolve during the Cold War.49

Second, although states may benefit from using coercion to increase their domestic legitimacy, their concern for legitimacy is not an independent factor influencing when and why they pursue coercion. Rather, the need to establish resolve precedes concerns about domestic legitimacy: it is sometimes through foreign media exposure (which first increases the coercer's need to establish resolve) that the domestic public begins to be informed about issues threatening its state's national security.

Third, economic costs influence when states decide to engage in coercion, because economic indicators are crucial in determining whether leaders will remain in office. The logic of this relationship between economic costs and leadership longevity applies to authoritarian and democratic states alike.50 Calculations of geopolitical costs influence decisions to escalate to military coercion, because high geopolitical costs could push the target state to call on its allies to provide military assistance, which could lead to a military confrontation with the coercer. Moderate use of economic sanctions, in contrast, is unlikely to trigger defense treaty obligations.

MEASURING THE VARIABLES

In my study, all variables are binary. I do not treat the costs and benefits as low, medium, or high, because decisionmakers are not mathematicians. With many decisions to make on a daily basis, they simplify their decision-making process.51

THE NEED TO ESTABLISH A REPUTATION FOR RESOLVE. In addition to speech evidence in which officials stressed the need to show resolve and expressed concerns about appearing weak, I use two objective indicators for purposes of cross-checking: (1) the number of incidents (i.e., challenges from other states that threaten the coercer's national security) and (2) the visibility and salience of the incidents in question. It is important to note that I am measuring the level of the state's need to establish resolve, not the level of resolve that the state already has.52 Thus, the focus is on the coercer, not on how resolved other states view the coercer.

When the visibility and salience of the target state's action are high, the coercing state might fear that potential challengers will observe this action and that if the state does not use coercion, other states may take similar actions in the future (or the target may continue its action or escalate), in the belief that the state will not be willing to use coercion. I measure visibility with the level of media coverage—that is, whether the issue threatening the coercer's national security receives lots of coverage, especially in highly influential media outlets such as Agence France-Presse, the Associated Press, and Reuters.

From the perspective of the coercer, the visibility of national security issues through media reports is one indicator it uses to measure the need to establish resolve, because coercion is not only about the challenger. The challenger might have excellent intelligence regarding the issue at stake, but other states might not accord it the same level of attention. For example, not all states use their intelligence services to track when a particular government will receive the Dalai Lama, or foreign fishermen fishing in waters claimed by China, if English-language news sources do not report them. The greater the media visibility, the more likely it is that other states or potential challengers might be watching the coercer's response. All else being equal, the lack of a response in the face of high-visibility incidents might make other states view the coercer as less resolved than if the incidents have lower visibility. If the coercer does not respond despite the incidents' high level of visibility, other states might think that it will similarly refrain from using coercion in the future. If the incident is not highly visible, other states might think that the coercer and the challenger have a private arrangement. In measuring status-altering events, Jonathan Renshon similarly notes that such events should be highly visible and salient, because “leaders and their advisors face severe constraints on their time and attention” and therefore “cannot pay attention to everything that happens in the world.”53 Theoretical and empirical studies in international relations, sociology, criminology, and economics likewise show that an increase in the visibility of rule-breaking behavior may strengthen the propensity of individuals to break social norms, laws, or regulations;54 that is, increased publicity of a particular behavior may lead others to follow suit. These studies suggest that visibility and salience have external validity and are not ad hoc measures.

As for the number of incidents, when there is more than one challenger threatening a state's national security or when one challenger engages in the same action multiple times—especially during a concentrated period and when the perpetrators are smaller states—the state uses coercion to avoid being seen as weak and unwilling to defend its interests. Other states may be watching the state's reaction, so if it does not take action to halt repeated transgressions, other states may take this as a green light to undertake similar transgressions in the future. As such, the higher the visibility of the issue and the greater the number of perpetrators, the more pressure there is on the state to establish a reputation for resolve. This is not to say that reputation concerns disappear when visibility of the issue is low and when there are fewer challengers. States do not have unlimited resources to respond to every challenge and therefore have to rank order when the need to establish resolve is high.

ECONOMIC COST. Economic cost is measured in terms of the economic relations between the coercer and the target, as well as the economic relations between the coercer and states in the region where the target is located. When the economic cost is high, one should first notice that objective economic relations indicate an asymmetry favorable to the target state. Indicators of bilateral economic relations include trade dependency and levels of foreign direct investment. Second, one should also observe government policy analysts and officials talking about such asymmetry, including how the state needs the target state for markets or supply. Government analysts and officials should note the existence of alternative markets and supply when they decide to apply coercive measures.

GEOPOLITICAL BACKLASH COST. I measure the geopolitical backlash cost as the capability of the target state to balance against the coercer. This capability includes both immediate military retaliation (from allies or neighbors of the target state) and long-term balancing, which is the target's forming or strengthening of alliances with its neighbors or great powers, especially the United States.

I use two kinds of indicators. The first kind consists of official threat assessments and threat assessments of government policy analysts affiliated with the coercing state. When the geopolitical backlash cost is high, one should first observe government officials and analysts making threat assessments, including analysis of the potential target state's bilateral relations with other states. If they perceive competition between the target and other states and are confident that the target will be unable to balance against the state, the state will use military coercion. Official assessments of other states' past and current policies, past crisis behavior, and statements—prior to the decision of whether to coerce—are therefore crucial. In cases where states do not use military coercion, one should see statements by government officials and scholars about their concerns of a geopolitical backlash from the target state, such as immediate military retaliation based on existing alliances or the formation of a long-term alliance. I also use U.S. national security documents, including the National Security Strategy, for cross-checking purposes.55

RESEARCH DESIGN AND SOURCES

I first use congruence tests to demonstrate that temporal variation in Chinese coercion in the South China Sea is in line with the cost-balancing theory. I then process trace the 2012 Scarborough Shoal incident to indicate that the causal mechanisms of the theory hold in this case. Below, I list the materials used and cross-check them against one another.

PRIMARY WRITTEN MATERIALS. I used three kinds of sources, categorized by their level of authority (i.e., whether they are official sources). The most authoritative evidence is official government documents, including the annual book (Zhongguo Waijiao) from China's Ministry of Foreign Affairs (MFA), the biannual defense white paper from China's Ministry of Defense, China's State Council's annual government work report, and annual maritime development reports published by China's State Oceanic Administration (SOA). I also used official chronologies of Chinese leaders and statements from the MFA, the People's Daily, and the SOA. Finally, I used data from China's Customs and Ministry of Commerce.

I also used semi-official documents and reports written by government think tanks, as well as articles written by zhongsheng in the People's Daily. An apparent homophone for “the voice of China,” zhongsheng is written by the editorial staff of the People's Daily International Department.56 I used the following semi-official reports, some of which are for internal use only: the annual Yellow Book of International Politics, published by the Chinese Academy of Social Sciences (CASS); the annual Strategic and Security Review, published by the China Institute of Contemporary International Relations (CICIR); internal reports by the National Institute of South China Sea Studies (NISCSS); the annual Bluebook of International Situation and China's Foreign Affairs, published by the China Institute of International Studies (CIIS); and the annual Strategic Assessment from the Chinese Academy of Military Science (AMS).57 I also used memoirs of Chinese leaders.

Finally, I used scholarly writings, which are not authoritative and provide the least strong evidence.

INTERVIEW DATA. I first interviewed former Chinese and foreign officials, who provide the strongest evidence in the interview category. I also interviewed government policy analysts with access to internal government information. Finally, I interviewed a variety of Chinese and foreign scholars. Interviews took place in Beijing, Guangzhou, Haikou, Nanjing, Shanghai, Wuhan, Xiamen, and Washington, D.C. By diversifying the geographical locations and kinds of interviewees, I reduce organizational, geographical, and occupational biases.

SECONDARY SOURCES. When constructing the dataset, I used both Chinese and foreign accounts of particular incidents, to avoid bias. I also used secondary sources to triangulate the measurements of the costs and benefits in my theory.

China's Use of Coercion in the South China Sea over Time

As mentioned, China has maritime territorial disputes with Brunei, Malaysia, the Philippines, and Vietnam in the South China Sea. Figure 1 shows opportunities for and instances of China's use of coercion in these disputes from 1990 to 2015. The vertical axis indicates the number of incidents or cases of Chinese coercion. The dark gray bars denote the total number of incidents—actions taken by other South China Sea disputants to which China could react by pursuing or not pursuing coercion. These incidents, which are not cases of coercion, fall into two categories: (1) other disputants' control of land features in the South China Sea and (2) energy exploration in disputed waters.58 Specifically, incidents regarding control over land features include other claimants seizing and building infrastructure on land features—for example, Vietnam's seizure of a land feature in the Spratly Islands in 1991. Incidents regarding resource exploration include oil and gas exploration activities and the signing of production-sharing contracts with foreign companies—for example, the Philippines signing such contracts with foreign oil companies. By reactive, I do not mean that China is the victim in maritime disputes. Of course, China is not always reactive and has used proactive coercion, including land reclamation in the South China Sea, especially in 2015.59

Figure 1.

Opportunities for and Instances of Chinese Coercion regarding Maritime Disputes in the South China Sea, 1990–2015

Figure 1.

Opportunities for and Instances of Chinese Coercion regarding Maritime Disputes in the South China Sea, 1990–2015

As the light gray bars in figure 1 show, China's use of coercion from 1990 to 2015 exhibits both temporal variation and variation in its use of coercive instruments. China pursued coercion against South China Sea disputants in the mid-1990s, taking a more dramatic, sometimes militarized, form from 1994 to 1996. In the early 2000s, however, China refrained from using coercion. Starting in 2007, China greatly increased its use of coercion, especially gray-zone coercion.60 Yet, unlike the early 1990s, these were all cases of nonmilitary coercion, including diplomatic and economic sanctions and gray-zone coercion.61 Gray-zone coercion included the use of civilian law-enforcement ships to ram the vessels of other South China disputants as well as the blocking of foreign ships from conducting oil and gas exploration (e.g., throwing dried tree branches in their way to interrupt seismic surveys).62 Since the 1990s, China has not used brute force in any of its territorial disputes in the South China Sea. If the cost-balancing theory is correct, China should use coercion when its need to establish a reputation for resolve is high and the economic cost is low. It should choose nonmilitarized coercive tools when the geopolitical backlash cost is high.

THE NEED TO ESTABLISH A REPUTATION FOR RESOLVE

China's need to establish a reputation for resolve was high in the 1990s, declined between 2000 and 2006, and rose after 2007. Figure 2 shows the number of challenges to Chinese sovereign claims from 1990 to 2015.63 As the figure demonstrates, the mid-1990s witnessed a surge in claimants' challenges to China in the South China Sea, though these declined dramatically from 2000 to 2006. None of the challenges in this latter period were particularly concerning to China. The claimants seized land features in the 1990s but focused more on building infrastructure on land features they had already taken. The slight bump in 2003 had more to do with officials of other claimants visiting land features they had taken in the 1990s.64 The post-2007 period witnessed a resurgence of actions. These trends are corroborated by the amount of exposure they received in the People's Daily and international media.

Figure 2.

Number of Incidents (Challenges to Chinese Sovereign Claims) in the South China Sea, 1990–2015

Figure 2.

Number of Incidents (Challenges to Chinese Sovereign Claims) in the South China Sea, 1990–2015

Figure 3 presents the results of my Factiva search of Agence France-Presse, the Associated Press, and Reuters reports that mention either “South China Sea” or “Spratlys” disputes.65 These three are the most influential English-language news agencies in the world. Reporting by them would increase the salience of the South China Sea issue and the pressure on China to establish resolve. I read every report to exclude those with topics that have no relevance to this study; the topics include typhoons, positive developments in the South China Sea, and use of Chinese coercion. In so doing, I am not capturing the dependent variable itself. In line with the findings in figure 2, international media exposure was high in the 1990s, contracted from 2000 to 2006, and picked up again starting in 2009.66

Figure 3.

Associated Press, Agence France-Presse, and Reuters Reports of “South China Sea” or “Spratly” Disputes, 1990–2015

Figure 3.

Associated Press, Agence France-Presse, and Reuters Reports of “South China Sea” or “Spratly” Disputes, 1990–2015

China's Ministry of Foreign Affairs was keenly aware of the concentrated activities of South China Sea claimants in the 1990s (especially in the early to mid-1990s) and was quick not only to respond to them, but to take steps to prevent their recurrence.67 Internal CASS publications in 1993 and 1994 also documented such behavior by South China Sea claimants, reflecting concerns about the growing trend of “internationalization”—that is, the increasing salience of and international attention paid to these disputes.68 Internal CASS reports in the early 1990s asserted that other claimants had begun to “carve up” the Spratly Islands, because China had not taken measures to assert its sovereign rights since 1988.69 SOA's internal publication in March 1992 reasoned that only by showing more resolve would China be able to make great powers stop investing in Vietnam for oil exploration in China's waters.70 Thus, the need to establish resolve was high.

From 2000 to 2006, official and semi-official government threat assessments noted the reduced pressure on China to establish resolve. For example, China's official defense white papers indicated in 2000 and 2002 that the situation in the South China Sea was “basically stable”; in 2004 the South China Sea was not even mentioned.71 The China Institute for Maritime Affairs, a government institute under the SOA, indicated in its 2004 and 2005 reports that the situation in the South China Sea was relaxed.72 Similarly, the internal 2003 and 2004 reports from the NISCSS described the general situation in the South China Sea as “overall stable.”73 Interviews with current government officials and government policy analysts were in line with the above assessments.74

In the post-2007 period, increasing actions by other claimants in the South China Sea heightened Chinese concerns about growing international attention. Starting in 2008, internal NISCSS assessments reported that the situation in the South China Sea had become complicated and that disputes were becoming “salient.”75 An internal NISCSS report published in 2008 suggested that China strengthen its regular patrolling of the Spratlys and “selectively disrupt and stop” other claimants' actions.76 China's 2010 defense white paper stated that pressure on China to defend its maritime rights had increased.77 Semi-official documents shared this assessment. One internal CASS report indicated in 2011 that China's maritime security environment had worsened in 2010 and that China would face “regularized” pressure in the maritime realm, observations echoed in CICIR reports.78 Furthermore, the publicity and salience of the South China Sea issue added to China's need to establish resolve. For example, the 2008 NISCSS report expressed particular concern about Vietnam and the Philippines because of their attempts to publicize the South China Sea issue.79 As such, the deputy chief of staff of China's People's Liberation Army (PLA) stated in early 2010 that “we are against actions of drastically publicizing the South China Sea issue.”80

Interviews with officials and government analysts in various parts of China also confirm the logic of using coercion to establish a reputation for resolve and avoid being seen as weak.81 Government policy analysts and scholars stated that China used coercion to “kill the chicken to scare the monkey” (shaji jinghou), warning all claimants against taking action in the future.82 Chinese coercion thus aimed at deterring any future encroachment of China's sovereign rights in the South China Sea.83 As an official from the maritime surveillance team of the SOA indicated, China needed to show its resolve that it would not lose any island or maritime area.84 In short, China's need to establish resolve was high in the 1990s, low from 2000 to 2006, and high after 2007.

CHINA'S ECONOMIC COST

The economic cost for China to pursue coercion was low in the 1990s, rose briefly from 2000 to 2006, and fell again in the post-2007 period. Turning first to objective indicators, China's exports to the Association of Southeast Asian Nations (ASEAN) paled in comparison to its exports to Japan, the European Union (EU), and the United States, especially in the 1990s (see figure 4).85 Even though Chinese exports to ASEAN grew continuously as a share of total Chinese exports in the late 2000s, they were still far below the level of Chinese exports to the EU and the United States, each constituting an average of 15 percent of Chinese exports. Since the mid-2000s, however, Sino-ASEAN trade has increasingly become an important component of ASEAN's overall trade relations (see figure 5).86

Figure 4.

Exports to ASEAN, Japan, the European Union, and the United States as a Share of China's Exports, 1992–2014

Figure 4.

Exports to ASEAN, Japan, the European Union, and the United States as a Share of China's Exports, 1992–2014

Figure 5.

ASEAN Exports to and Imports from China as a Share of ASEAN Exports and Imports, 1997–2014

Figure 5.

ASEAN Exports to and Imports from China as a Share of ASEAN Exports and Imports, 1997–2014

In line with objective indicators, Chinese government policy analysts indicated that China sought to attract investment from Japan and the United States in the 1990s.87 Of course, China would have liked to expand its economic ties with Southeast Asian countries, but at the time, this was not a priority.

Interestingly, the objective data do not show the nuances: there was a brief period from 2000 to 2006 when the economic cost for China to coerce ASEAN countries was high. From the early 2000s, China began to increase its economic cooperation with ASEAN (e.g., creating the ASEAN-China free trade zone [FTZ]). According to Zhang Yunling, a senior government policy analyst involved in the FTZ negotiations, China initiated the talks for economic reasons.88 China, in the 1990s, was focused on gaining membership to the World Trade Organization (WTO), which it did in 2001.

One of China's economic strategies following its accession to the WTO was to find ways to increase regional economic cooperation.89 ASEAN was an ideal starting place given its concerns about the implications of China's membership in the WTO for competition of market share and foreign direct investment.90 Zhang Yunling indicated that China was aware of these concerns and wanted to lower them.91 For example, in November 2000, Premier Zhu Rongji suggested that China and ASEAN begin discussions involving free trade.92 Also, China considered ASEAN more likely than more advanced trading blocs to negotiate an FTZ.93 In other words, China had no exit options regarding an FTZ, whereas ASEAN did with, for example, Japan, the United States, and the EU.94 To improve Sino-ASEAN economic relations, China refrained from taking coercive action, as noted in interviews and internal SOA reports in 2002.95

Over time, China's economic cost associated with the FTZ gradually began to decline. Increasingly, China came to believe that ASEAN depends more on China than vice versa. For example, the 2009 NISCSS report noted that as a result of the global financial crisis, ASEAN countries would need China's markets for a long period.96 Because the Chinese economy was in better shape compared to advanced economies, China believed that it could stand firm on the issue of coercion.97 Also, after 2007 the Chinese government began the transition from an export-oriented to a consumption-oriented economy, reducing the importance of the China-ASEAN FTZ.98 Further, by April 2009, China had completed negotiations with ASEAN regarding all aspects of the FTZ.99

In sum, China's economic cost of using coercion was low in the 1990s, high from roughly 2000 to 2006, and low in the post-2007 period.

CHINA'S GEOPOLITICAL BACKLASH COST

The geopolitical backlash cost to China of pursuing coercion was low in the 1990s but rose in the post-2000 period. I turn first to official Chinese and U.S. documents, including the MFA's annual China's Foreign Affairs and the U.S. National Security Strategy (see table 1). Whether and how China's MFA used the word “multipolarity” in China's Foreign Affairs is an important indicator of the geopolitical pressure that China felt coming from the United States. In the Chinese political context, multipolarity means greater flexibility for China in the international system and less geopolitical pressure from the United States, the hegemon. The more optimistic China was in its description of multipolarity, the less unipolar China's perception of the international balance of power became and the less pressure China felt from the United States. MFA assessments appeared confident about the progress of multipolarity in the 1990s (see table 1). Beginning in the early 2000s, however, the number of times they mentioned multipolarity decreased.100

Table 1.

China's Official Assessments of Geopolitical Costs

MFA Annual China's Foreign Affairs Assessments of the International SituationMFA Annual China's Foreign Affairs Assessments on the United StatesU.S. National Security Strategy
1990s (low) Assessments mainly used wording such as “rapid,” “quick,” and “unstoppable” to describe what it perceived to be the progress of multipolarity. Highlighted U.S. withdrawal from the Subic Bay; reduction of troops in Asia; Europe as U.S. priority European stability is vital to U.S. security. East Asia became of growing significance for U.S. security and prosperity 
Post-2000 period (high) Decreased number of mentions of multipolarity; wording such as “in obstacle” Increased U.S. efforts in Asia, especially since the terrorist attacks of September 11, 2001; Philippines and Thailand as “major non-NATO allies” 2001 was the first time East Asia was named a vital U.S. interest. 
MFA Annual China's Foreign Affairs Assessments of the International SituationMFA Annual China's Foreign Affairs Assessments on the United StatesU.S. National Security Strategy
1990s (low) Assessments mainly used wording such as “rapid,” “quick,” and “unstoppable” to describe what it perceived to be the progress of multipolarity. Highlighted U.S. withdrawal from the Subic Bay; reduction of troops in Asia; Europe as U.S. priority European stability is vital to U.S. security. East Asia became of growing significance for U.S. security and prosperity 
Post-2000 period (high) Decreased number of mentions of multipolarity; wording such as “in obstacle” Increased U.S. efforts in Asia, especially since the terrorist attacks of September 11, 2001; Philippines and Thailand as “major non-NATO allies” 2001 was the first time East Asia was named a vital U.S. interest. 

MFA stands for Ministry of Foreign Affairs. China's Foreign Affairs is published annually by the MFA.

Second, despite the conventional wisdom that China was concerned about a geopolitical backlash as a result of the 1989 Tiananmen incident and the end of the Cold War, the geopolitical backlash cost regarding the Spratly disputes was low in the 1990s. MFA assessments maintained that the United States and Russia had decreased their presence in Southeast Asia. The 1993 issue of China's Foreign Affairs Overview noted that the United States had withdrawn its forces from the Subic Bay Naval Base in the Philippines.101 The 1997 issue of China's Foreign Affairs claimed that Europe was the priority of U.S. global strategy, a view the MFA held until 2000.102 Official Chinese national defense white papers made similar threat assessments.103 This position was corroborated by the U.S. National Security Strategy, which treated Europe as the vital interest until 2000.104

Unlike the 1990s, China's concerns about a geopolitical backlash have grown serious since the 2000s. Official Chinese threat assessments in the post-2000 period expressed worry about the United States returning to Southeast Asia. The 2001 issue of China's Foreign Affairs stressed that the United States had reinstated joint military exercises with the Philippines and that its secretary of defense had visited Vietnam for the first time since the Vietnam War.105 The 2002 issue of China's Foreign Affairs stated that after the terrorist attacks of September 11, 2001, the United States had sought greater counterterrorism cooperation with ASEAN countries in response to rampant terrorist activity in Southeast Asia.106 An internally circulated document on great power issues, classified as “secret,” from the Central International Liaison Department of the Chinese Communist Party declared in 2004 that the United States had begun to establish counterterrorism battlegrounds in Southeast Asia.107 Finally, every issue of China's Foreign Affairs from 2007 to 2014 cited U.S. efforts to strengthen relations with ASEAN. China's national defense white papers made similar observations.108

Shifts in geopolitical costs were also evident in internal reports and interviews with government policy analysts.109 Several interviewees explicitly indicated that Chinese military coercion in the South China Sea during the mid-1990s was related to the U.S. withdrawal from Subic Bay, which had created a “geopolitical power vacuum” that China was eager to fill.110 CICIR noted, however, that after 2000 the United States sought to develop alliance or quasi-alliance relations with ASEAN countries.111 CASS, the AMS, and the CIIS issued similar assessments.112 An internal CASS report indicated in 2011 that the United States viewed ASEAN's role in the Asia Pacific as critical.113

In sum, the geopolitical backlash cost of coercion for China was low in the 1990s and high in the post-2000 period.

TEMPORAL VARIATION IN CHINESE COERCION AND CHOICE OF COERCIVE TOOLS

Table 2 offers a summary of changes in China's need to establish resolve, the associated economic and geopolitical backlash costs, and patterns of Chinese coercion. It demonstrates that variations in these variables are in line with the cost-balancing theory.

Table 2.

Cost Balancing and China's Use/Nonuse of Coercion

Costs
Need to Establish a Reputation for ResolveEconomic CostGeopolitical Backlash CostCoercion
1990–99 high low low yes (some militarized) 
2000–06 low high high no 
2007–present high low high yes (no militarized coercion) 
Costs
Need to Establish a Reputation for ResolveEconomic CostGeopolitical Backlash CostCoercion
1990–99 high low low yes (some militarized) 
2000–06 low high high no 
2007–present high low high yes (no militarized coercion) 

The China-Philippines Scarborough Shoal Incident of 2012

The Scarborough Shoal (Huangyandao) is located in the Macclesfield Bank in the South China Sea. On April 10, 2012, a Philippine naval ship tried to arrest Chinese fishermen for fishing illegally around the disputed shoal.114 In previous years, China had used diplomatic channels to secure the release of such fishermen.115 April 10 marked the first time in the post-2000s that China used multiple coercive tools to take the shoal.

Offensive realists would predict that China would adopt militarized coercive measures in the Scarborough case. Jervis's deterrence model, however, would argue that, in light of the decision by the United States to “rebalance” to Asia in 2011, China should be deterred from taking such measures. As my theory predicts, China used coercion in this case because the need to establish a reputation for resolve was high and the economic cost was low. Given concerns about the geopolitical backlash cost, it restricted itself to using nonmilitary coercion.

THE MAGNITUDE AND GOALS OF COERCION

China engaged in three forms of nonmilitarized coercion against the Philippines. First, using gray-zone coercion, the head of the South China Sea section of the SOA immediately ordered two maritime surveillance ships to rescue the Chinese fishermen on April 10.116 A fishery administration ship arrived at the Scarborough Shoal on April 11.117 On April 17, the Philippines urged China to bring the dispute to the International Tribunal on the Law of the Sea, but China refused.118 On May 2, China dispatched four more maritime surveillance ships.119 By May 9, China had blocked Filipino fishermen from entering the shoal and forced them to leave.120 Afterward, China maintained regular patrols around the shoal.121 The Philippines eventually withdrew, although China did not.122 Second, China imposed economic sanctions beginning in early May 2012, quarantining Philippine fruits. Beginning on May 11, China ultimately prevented 1,500 containers of bananas from the Philippines from entering Chinese ports, citing “pest infestation.”123 Philippine media estimated that the ban, which lasted for about a month, led to the loss of 1 billion Philippine pesos (about $23 million).124 Third, China imposed diplomatic sanctions on the Philippines. According to government policy analysts, China terminated all senior-level (ministerial-level and above) bilateral visits. From 2013 to 2015, no formal meetings were held between the foreign ministers of the two states.125

Chinese behavior in the Scarborough Shoal case constitutes coercion because it involved the following factors: state action; clearly identified targets; the threat/use of coercion to inflict pain; and, most importantly, clear goals. China's direct goal was to stop the Philippines from controlling the shoal: the Chinese MFA repeatedly demanded that Philippine vessels withdraw.126 Further, the Chinese MFA called on the Philippines to return to bilateral talks and respect Chinese sovereignty claims.127 The broader goal was to stop other states from viewing China as weak and engaging in actions that threatened Chinese interests in the South China Sea.128

WHY CHINA USED COERCION

China's need to establish a reputation for resolve was high in the Scarborough Shoal case. Chinese policy analysts believed that the Philippines had been trying to increase the international salience and exposure of South China Sea disputes, especially through media reports and its government officials' call for using the UN and ASEAN to resolve the disputes.129 Prior to the 2012 incident, the Philippines had also increased the frequency of its small challenges to the Chinese in the South China Sea. In May 2011, the Philippine navy removed three markers that China had placed on reefs and banks in the Spratlys.130 In June, it announced plans to award offshore gas and oil drilling rights to foreign companies in the Spratlys. China claimed that two of the three blocks lay within its nine-dash line.131 In July, China announced plans to build a loading ramp and upgrade a runway on Thitu Island.132 Additionally the number of Philippine media reports on the South China Sea increased sharply in 2011, more than doubling in number from 2008.133 This media exposure increased pressure on China to establish resolve. In addition, despite Beijing's rejection of Manila's request for UN arbitration, the president of the Philippines told Reuters in September 2011 that his government was seeking other options,134 including a push by the Philippines for a joint statement on the South China Sea during the ASEAN leaders' meeting in November 2011.135 In particular, the Philippines publicized the arrest of the Chinese fishermen, prior to China's decision to take coercive action. The Philippine navy and foreign ministry released photos of the arrested Chinese fishermen, with an armed Filipino naval officer standing behind them.136 Reuters reported on these photos before China responded.137

None of the Philippine actions above was enough to tilt the balance of power in the South China Sea. Still, the Chinese government was unhappy. As early as August 2, 2011, zhongsheng had noted that a Philippines infrastructure project on Flat Island would soon be completed.138Zhongsheng continued that China's principle of “shelving disputes for joint development” did not mean that China would let the Philippines take this as an opportunity to encroach upon China's territory and that if the Philippines made a serious strategic miscalculation, it would “pay the price.”139 Similarly, another semi-official government source—a regional security assessment of CASS published in January 2012—noted the above-mentioned Philippine actions in 2011.140

On February 28, 2012, a Chinese MFA spokesperson warned that the Philippines should not “take actions that further complicate and expand the South China Sea disputes.”141 The following day, in response to the Philippines bidding on energy contracts in exclusive economic zone blocks claimed by China, zhongsheng blamed the Philippines for “instigating trouble” in the South China Sea,142 stating that the Philippines would be wrong to view China's efforts to push for cooperation among South China Sea claimants as “a sign of weakness.”143Zhongsheng further emphasized that “China was resolute in defending its sovereignty and would take necessary measures.”144 A comprehensive search of the People's Daily for the words “weakness” or “weak and bulliable” from 1990 to 2017 indicates that China used this wording only infrequently to describe its foreign affairs.145 In fact, this zhongsheng statement was the first time China ever used such wording vis-à-vis the Philippines.

During the standoff, China's deputy foreign minister, Fu Ying, summoned Philippine diplomats on May 7 to tell them that, in the past month, the Philippines had failed to realize its grave mistake and instead had made matters worse: he urged the Philippines to withdraw its ships.146 Fu emphasized that the Philippines should avoid miscalculation and that China was prepared to take action.147 Fu's statement demonstrates that China did not want the Philippines to think that Beijing lacked resolve in this situation. On May 8, the People's Daily underscored China's position: “The Philippines thought that China wanted to avoid trouble … Yet the Philippines did not see things clearly—China would not give in to issues of sovereignty, the Philippines should not view China's friendliness as weak and susceptible to bullying … China would not mind creating a ‘Scarborough model’ to stop the opponent and to deter any transgression.”148 The same statement appeared on the front page of the overseas edition of the People's Daily, intended for an international audience. On May 15, Dai Bingguo, a state councilor and one of the highest-ranking figures in Chinese foreign policy, reaffirmed that being modest did not mean that China would stand being bullied by other countries, “especially small countries like the Philippines.”149 Chinese officials' statements before and during the Scarborough incident showed consistency and were not post hoc justifications.

Interviews with government policy analysts, former government officials, and scholars confirm China's need to establish resolve. One former senior SOA official who was involved in the Scarborough incident stated bluntly that China took measures in 2012 because the Philippines “had done too much in the past.”150 Another former official agreed that China was pressured to establish resolve to defend its rights in this incident.151 One former diplomat explained that China thought that if it withdrew, the Philippines would believe that China would compromise yet again.152 Other government policy analysts noted that if China did not take coercive measures, it would signal a green light to the Philippines and Vietnam, thereby encouraging more states to encroach on China's sovereignty.153 A senior government policy analyst stressed that China needed to “achieve a deterrent effect on surrounding countries,” termed explicitly by another scholar as “establishing resolve” (li wei).154 One former government analyst even noted that during the Scarborough Shoal incident, China was also thinking about Japan, as their dispute over the Senkaku Islands had begun to heat up around roughly the same time as the Scarborough incident, a point corroborated by a Japanese diplomat.155

The economic cost of coercing the Philippines was low in this case. In 2010, China was the Philippines' third largest trading partner.156 By 2011, China had become its third largest export destination.157 China is the second largest export destination for the member companies of the Pilipino Banana Growers and Exporters Association (PBGEA), constituting about 25 percent of PBGEA's annual exports. China is also the largest export market for non-PBGEA member companies (i.e., independent growers and cooperatives).158 In contrast, the Philippines was China's sixth largest trading partner in bilateral trade with ASEAN countries.159 This asymmetry gave China leverage during the dispute.

Speech evidence concurs with objective measures of economic costs, which were low for China in this case. Chinese government officials and policy analysts had noted China's economic importance to the Philippines long before the Scarborough incident.160 Bai Ming, an official in China's Ministry of Commerce, stated that Chinese-Philippine trade was asymmetrical, with bilateral trade constituting 30 percent of total Philippine trade but only 0.89 percent for China.161 Bai emphasized that China “could impose economic sanctions and isolate the Philippines,” while strengthening economic relations with other ASEAN countries.162 Former government officials also stated that using coercion would hurt the Philippines much more than it would China, given the size of the Chinese economy and the Philippines's greater reliance on China.163

The geopolitical backlash cost for China was high in this case, which limited China's choice of coercive tools. Concerned about a potential backlash, especially immediate escalation, China chose nonmilitarized coercive tools.164 Chinese government policy analysts believed that it was fine to use coercive measures, but that militarization would escalate the disputes and push ASEAN countries closer to the United States.165

Indeed, the United States was the most critical factor in restraining China's choice of coercive tools. In internal conferences and internal publications, Chinese government policy analysts, fearing U.S. military containment, stressed that China needed to avoid direct confrontation with the United States in the South China Sea.166 One former official in the PLA Navy was particularly concerned that if China used military coercion, the U.S. Navy might become directly involved; he admitted that the United States was still “no. 1.”167 In short, China believed that military means were too costly to use in South China Sea disputes and peace remained the priority168

Semi-official Chinese assessments made before China used coercion in the Scarborough incident indicated U.S. unwillingness to use force to intervene in territorial disputes in the South China Sea.169 Government policy analysts and scholars emphasized that the United States would not start a “backlash” against China, especially when the Philippines had lost legitimacy by sending in naval vessels.170 In an internal conference, one government policy analyst noted that on June 23, 2011, when U.S. Secretary of State Hillary Clinton met with Philippine Foreign Minister Albert del Rosario, “Clinton avoided promising to unconditionally support the Philippines in South China Sea disputes.”171 Despite del Rosario's demand, Clinton did not explicitly state that the U.S.-Philippine Mutual Defense Treaty was applicable to South China Sea issues.172 The analyst concluded that the United States did not want direct conflict with China.173 Scholars and government policy analysts indicated that China's rationale in the Scarborough incident was that as long as Chinese action remained controlled and nonmilitarized, the United States would not get involved.174

Chinese analysts were probably right in this assessment. On April 22, 2012, U.S. Lt. Gen. Duane Thiessen took a Filipino reporter's question about the applicability of the U.S.-Philippines defense treaty to the Scarborough Shoal. The general answered ambiguously that the treaty “guarantees that we get involved in each other's defense and that is self-explanatory.”175 He did not elaborate on what kind of assistance the United States would provide, stating that “there is no tie between Scarborough Shoal and U.S. movement in the Pacific.”176 Similarly, when the U.S. secretary of defense and secretary of state met with their Philippine counterparts on April 30, they did not clarify whether the treaty covered the Philippines' offshore claims, nor did they promise direct U.S. intervention.177

Alternative Explanations for China's Use of Coercion

There are several alternative hypotheses regarding when and why a state decides to use coercion against other states. First, as the sanctions literature suggests, powerful domestic lobbies can pressure their governments to take such action. Official documents suggest, however, that Chinese coercion in the South China Sea is both regularized and centralized. Detailed and modularized plans describe how crews on maritime surveillance and fishery administration ships should behave when dealing with foreign counterparts. In Guangdong Province, when foreign fishing vessels engage in illegal fishing in Chinese EEZs or when foreign administrative ships attempt to harass Chinese fishermen, fishery administration ships are instructed to report the incident to the command center of the fishery administration.178 Measures such as expelling foreign ships have to be approved by sub-bureaus of the SOA.179

Additionally, interviews with former Chinese officials indicate that the central government decides when to take coercive measures.180 According to one scholar, every South China Sea incident involving China and another country is reported to the central government.181 Citing internal seminars with officials from the SOA, the Coast Guard, and the Maritime Surveillance Agency, several government analysts indicated that Chinese administrative patrol ships strictly adhere to instructions and follow orders from the center.182 A former PLA Navy colonel who once participated in patrols in the South China Sea stated that there are institutionalized plans about how ships' crews should act when encountering foreign vessels.183 Local governments similarly do not have much leeway about what to do in these situations.184

The second alternative explanation for when and why a state decides to use coercion against other states involves domestic politics. According to this argument, faced with elite power struggles, domestic social issues, and popular nationalistic sentiments, leaders will pursue coercion against other countries to increase their domestic legitimacy. Since the 1990s, however, Chinese leaders have constantly had to deal with elite power struggles and myriad social issues. Yet, during this period, there was noticeable variation in when and how China has used coercion. In the 1990s, for example, China experienced intense elite power struggles during leadership transitions and periods of high inflation.185 In the 2000s, when inflation began to slow, the number of social protests—some of which turned violent—increased significantly186 Of course, the cost-balancing theory can be falsified and does not claim to explain all cases of Chinese coercion. Nevertheless, as Chinese government policy analysts pointed out, legitimacy concerns do not drive coercion decisions.187

In addition, China experienced critical leadership transitions in 1997, 2002, and 2012. None of these transitions, however, accords with the temporal variation in China's use of coercion. Among Jiang Zemin, Hu Jintao, and Xi Jinping, Hu is said to have been the weakest leader and Xi the most assertive.188 If individual leaders are critical, then instances of Xi leading coercive efforts against other countries should have been more numerous than for Jiang or Hu. Yet, China used coercion (sometimes militarized coercion) seven times in the 1990s during Jiang's rule. China again began taking coercive measures in 2007, during Hu's term. Despite his supposed weakness, Hu pursued coercion more than Jiang—ten times in all. In an internal speech during the Central Foreign Affairs Conference in August 2006, Hu stated that “China needed to be more proactive in foreign affairs,” which undermines the notion that Xi championed greater proactive action.189 Thus far, Xi has used coercion six times, none with a militarized component. One of Xi's former political secretaries revealed that Xi's viewpoints are closely in line with those of the center.190 Interviews with Chinese government analysts also confirm that individual leadership does not dictate coercion decisions.191 In particular, decisions to use coercion during the Scarborough Shoal incident were made collectively by the Politburo Standing Committee, China's highest decisionmaking body.192

Regarding popular nationalist sentiment, as Jessica Weiss points out, the Chinese government uses nationalism instrumentally, as it did in the 2012 Scarborough Shoal incident.193 A search for the word Huangyandao (Scarborough Shoal) on the highly nationalistic Tiexue website reveals that heated discussions started only after the Chinese official media released news about what was happening on April 12 (two days after the incident began). From January 1, 1990, to April 11, 2012, there were only twenty-five pages on the Tiexue website that included the word huangyandao; between April 12 and June 1, 2012, after People's Daily released the news, the number increased to seventy-six pages—a clear indication that the government controls how nationalistic the Chinese public should be regarding maritime disputes. Finally, empirical studies have shown that nationalism has only a moderate impact on China's foreign policy.194

A third alternative explanation for China's use of coercion focuses on the power variable, be it relative power, overall power, or the balance of power. The “relative power” argument follows a preventative logic. According to Fravel, states are more likely to use force in territorial disputes when their claims of strength are declining, which is partly a function of their power projection capability. Yet, in the post-1990s, China's projection capability consistently dwarfed that of other South China Sea claimants. If the relative power argument is correct, China should have relied on coercion less often when its relative power position had improved—a pattern that is not supported by the empirical evidence.195

As for the “overall power” argument, offensive realists predict that as its power grows, China will become more aggressive, possibly using of force. Again, the trend does not support this argument. China used military coercion when it was weak in the 1990s, refrained from taking coercive measures in the early 2000s, and resorted to nonmilitarized coercion beginning in 2007. In fact, the ratio of instances of Chinese coercion to number of incidents (actions taken by other states that challenge Chinese claims in the South China Sea) has remained low. Even the highest ratio of Chinese coercion to incidents has remained under 40 percent since 1990. China's use of coercion does not demonstrate a linear increase: China has not become more militarily aggressive over time. One could argue that as overall Chinese capability grows, China will rely on gray-zone tools.196 China did, however, use militarized coercion in its border disputes with India in 2017, despite having gray-zone coercive tools at its disposal.

Finally, the “balance of power” argument suggests that China uses coercion to balance against the more powerful state (i.e., the United States). There is scant empirical evidence, however, to indicate that China is balancing the United States in the South China Sea. If it were, then it would puzzling why China uses coercion in some cases, but not others.

Conclusion

This article has presented a theory—the “cost-balancing theory”—to explain when, why, and how China uses coercion in disputes in the South China Sea. I argue that the need to establish a reputation for resolve while considering the economic and geopolitical costs associated with coercive action are central to China's calculus. When the need to establish a reputation for resolve exceeds economic cost, China uses coercion. When the likelihood of a geopolitical backlash is high, it prefers to use nonmilitarized coercion. China believes that having capabilities but not demonstrating the willingness to use them may lead to deterrence failure. In a sense, China uses coercion for purposes of deterrence, blurring the line between the two.197

These findings contribute to the coercion, signaling, and credibility literature in several ways. The article demonstrates that China's decisions to use coercion extend beyond trying to change the behavior of target states. Signaling resolve to deter other states is central to China's rationale for using coercion. For rising powers such as China, coercion can be a cost-effective way to stop a target state from taking undesired actions while deterring other states from taking similar actions in the future.

I also show that China weighs its need to establish resolve against the economic and geopolitical costs of coercion. Asymmetric economic interdependence provides China coercive leverage, but at the same time, military coercion may lead to conflicts and instability that could adversely affect China's economic growth. As such, rising powers such as China pursue coercion, but at the nonmilitarized level, when the geopolitical cost is high. China has always been a risk-averse bully and is less belligerent than previous rising powers: the United States in the late nineteenth and early twentieth centuries, Germany under Otto von Bismarck, Wilhelmine Germany, and interwar Japan tended to use force against other powers.198 But given today's globalized production and supply chains, contemporary rising powers may face a dilemma that these earlier powers did not—showing resolve while minimizing the economic and geopolitical costs.199 This research thus complements a growing literature that links international security and political economy in calling for scholarly work that compares the coercive behavior of historical and contemporary rising powers.200

There is a rich literature on audience costs as a form of costly signals.201 Yet, my research reinforces Schelling's notion that states need to show physical evidence of resolve. China mostly engages in coercive action, as opposed to making coercive threats; the rationale is that physical actions increase China's reputation for resolve, especially if other states are watching and if the purpose of Chinese coercion is deterrence. This article therefore builds on the argument that military action sends strong signals because they are physical, and it expands that argument to suggest that nonmilitary physical signals can also be costly signals.202 Relatedly, the manner in which China pursues coercion is significant. Unlike the United States, when China threatens or imposes economic sanctions, it rarely makes a public announcement.203 One explanation for this is that the lack of publicity helps China eschew WTO rules; in a way, China can plausibly deny that it has explicitly imposed economic sanctions.

Empirically, in contrast to historical rising powers, China is a cautious coercer; it does not coerce frequently; and it relies on military coercion less often the stronger it becomes, instead employing unconventional tools such as gray-zone coercion.204 China's use of coercion in maritime disputes thus challenges the notion that China suddenly became assertive in the wake of U.S. decline following the 2009 global financial crisis.

Moreover, this study shows that China uses the United States' statements and past actions in assessing U.S. alliance commitments in the Asia Pacific. Whether and how the United States gets involved in South China Sea disputes significantly affects China's decisions regarding the use of coercion. China's use of military coercion in the 1990s against the Philippines and Vietnam after the U.S. withdrawal from the Subic Bay provides a useful example.

Finally, the cost-balancing theory is generalizable to other issues and other states. For example, China's use of coercion in response to the reception of the Dalai Lama by foreign leaders, the awarding of the Nobel Peace Prize to the human rights activist Liu Xiaobo, and the deployment of Terminal High Altitude Area Defense in South Korea all confirm the cost-balancing logic: China balances between the need to establish resolve and the economic cost of doing so.205

Acknowledgments

For helpful comments on earlier versions of this article, the author thanks Marsin Alshamary, Lena Andrews, John-Michael Arnold, Ryan Baker, Andrew Bell, Mark Bell, Michal Ben-Josef Hirsch, Meredith Blank, Austin Carson, Chong Chen, Binn Cho, Kate Cronin-Furman, Fiona Cunningham, Benjamin Denison, Michael Desch, Marina Duque, Alexander Evans, Taylor Fravel, Mathias Frendem, Mayumi Fukushima, Payam Ghalehdar, Andrea Gilli, Charles Glaser, Kelly Greenhill, Victoria Hui, Jamie Gruffydd-Jones, Daniel Jacobs, Michael Joseph, Tyler Jost, Joshua Kertzer, Nadiya Kostyuk, Marika Landau-Wells, Dan Lindley, Philip Martin, Timothy McDonnell, Andrew Miller, Aidan Milliff, Daisuke Minami, Kacie Miura, Sara Moller, Vipin Narang, Dani Nedal, Cullen Nutt, Rachel Esplin Odell, Reid Pauly, Brad Potter, Sara Plana, Sebastian Rosato, Erik Sand, Anne Sartori, Nina Silove, Jennifer Spindel, Rachel Tecott, Richard Samuels, and Peter White. The author also thanks participants in the Yale Brady-Johnson International Security Studies Research Workshop, the Boston International Security Graduate Conference, the Harvard International Security Conference, as well as the anonymous reviewers. The research for this article was supported by a fellowship from the Smith Richardson Foundation.

Notes

1. 

Thomas C. Schelling, Arms and Influence (New Haven, Conn.: Yale University Press, 1966); Robert A. Pape, Bombing to Win: Air Power and Coercion in War (Ithaca, N.Y.: Cornell University Press, 1996); Branislav L. Slantchev, Military Threats: The Costs of Coercion and the Price of Peace (Cambridge: Cambridge University Press, 2011); and Wendy Pearlman and Boaz Atzili, Triadic Coercion: Israel's Targeting of States that Host Nonstate Actors (New York: Columbia University Press, 2018). For work on economic sanctions effectiveness, see Taehee Whang, Elena V. McLean, and Douglas W. Kuberski, “Coercion, Information, and the Success of Sanction Threats,” American Journal of Political Science, Vol. 57, No. 1 (January 2013), pp. 65–81, doi.org/10.1111/j.1540-5907.2012.00629.x; Navin A. Bapat and Bo Ram Kwon, “When Are Sanctions Effective? A Bargaining and Enforcement Framework,” International Organization, Vol. 69, No. 1 (Winter 2015), pp. 131–162, doi.org/10.1017/S0020818314000290; Kenneth A. Rodman, Sanctions beyond Borders: Multinational Corporations and U.S. Economic Statecraft? (Lanham, Md.: Rowman & Littlefield, 2001); and Meghan L. O'Sullivan, Shrewd Sanctions: Statecraft and State Sponsors of Terrorism (Washington, D.C.: Brookings Institution Press, 2003). For work on the effectiveness of new coercive tools, see Kelly M. Greenhill and Peter Krause, eds., Coercion: The Power to Hurt in International Politics (Oxford: Oxford University Press, 2018); and Erica D. Borghard and Shawn W. Lonergan, “The Logic of Coercion in Cyberspace,” Security Studies, Vol. 26, No. 3 (2017), pp. 452–481, doi.org/10.1080/09636412.2017.1306396.

2. 

The literature does discuss costs to the coercer, including audience costs. Yet, such costs arise after a decision to use coercion has been made. They matter for whether one backs down from a threat. The literature does not analyze the conditions leading states to use coercion or why the coercer targets some states but not others. Some scholars do analyze sanctions decisions, but underspecify the costs and benefits for the coercer and focus on Western democracies. See Daniel W. Drezner, The Sanctions Paradox: Economic Statecraft and International Relations (Cambridge: Cambridge University Press, 1999); David A. Baldwin, Economic Statecraft (Princeton, N.J.: Princeton University Press, 1985); A. Cooper Drury, “Sanctions as Coercive Diplomacy: The U.S. President's Decision to Initiate Economic Sanctions,” Political Research Quarterly, Vol. 54, No. 3 (September 2001), pp. 485–508, doi.org/10.2307/449267; Kenneth A. Schultz, Democracy and Coercive Diplomacy (Cambridge: Cambridge University Press, 2001); and Jonathan Markowitz, “Prices or Power Politics? When and Why States Coercively Compete over Resources,” in Greenhill and Krause, Coercion, pp. 271–290.

3. 

The literature on foreign policy substitutability does not explain when states choose certain coercive tools over others. See T. Clifton Morgan and Glenn Palmer, “A Model of Foreign Policy Substitutability: Selecting the Right Tools for the Job(s),” Journal of Conflict Resolution, Vol. 44, No. 1 (February 2000), pp. 11–32, doi.org/10.1177/0022002700044001002.

4. 

For a historical example, see Fareed Zakaria, From Wealth to Power: The Unusual Origins of America's World Role (Princeton, N.J.: Princeton University Press, 1999).

5. 

Alastair Iain Johnston, “How New and Assertive Is China's New Assertiveness?” International Security, Vol. 37, No. 4 (Spring 2013), pp. 7–48, doi.org/10.1162/ISEC_a_00115. Current theories of Chinese foreign policy do not analyze gray-zone coercion. M. Taylor Fravel, for example, focuses on diplomacy and the use of force. See Fravel, Strong Borders, Secure Nation: Cooperation and Conflict in China's Territorial Disputes (Princeton, N.J.: Princeton University Press, 2008).

6. 

For example, John J. Mearsheimer, “Can China Rise Peacefully?” National Interest, October 25, 2014, http://nationalinterest.org/commentary/can-china-rise-peacefully-10204; and Robert Gilpin, War and Change in World Politics (New York: Cambridge University Press, 1981).

7. 

Schelling, Arms and Influence, p. 70.

8. 

Robert J. Art and Patrick M. Cronin, eds., The United States and Coercive Diplomacy (Washington, D.C.: United States Institute of Peace, 2003), p. 8.

9. 

Alexander L. George coined the term “coercive diplomacy.” See George, David K. Hall, and William E. Simons, The Limits of Coercive Diplomacy: Laos, Cuba, Vietnam (Boston: Little, Brown, 1971); and Alexander L. George and William E. Simons, The Limits of Coercive Diplomacy, 2nd ed. (Boulder, Colo.: Westview, 1994).

10. 

Schelling, Arms and Influence.

11. 

Tara J. Maller, “Diplomacy Derailed: The Consequences of U.S. Diplomatic Disengagement,” Ph.D. dissertation, Massachusetts Institute of Technology, 2011, p. 92.

12. 

Ibid., pp. 95–97.

13. 

Baldwin, Economic Statecraft, p. 32.

14. 

Ibid., p. 41. Threatening to withdraw positive inducements such as foreign aid is a form of coercion, but providing positive inducements without the threat of withdrawal is not.

15. 

Frank G. Hoffman, “The Contemporary Spectrum of Conflict: Protracted, Gray Zone, Ambiguous, and Hybrid Modes of War,” in 2016 Index of U.S. Military Strength (Washington, D.C.: Heritage Foundation, 2016), pp. 25–36, https://s3.amazonaws.com/ims-2016/PDF/2016_Index_of_US_Military_Strength_ESSAYS_HOFFMAN.pdf.

16. 

Michael J. Mazarr, Master the Gray Zone: Understanding a Changing Era of Conflict (Carlisle Barracks, Pa.: United States Army War College Press, 2015), pp. 58, 62.

17. 

See also Michael Green et al., “Countering Coercion in Maritime Asia: The Theory and Practice of Gray Zone Deterrence” (Washington, D.C.: Center for Strategic and International Studies, May 2017), https://www.csis.org/analysis/countering-coercion-maritime-asia.

18. 

There is a rich literature on the use of force. See, for example, Pape, Bombing to Win.

19. 

Chas W. Freeman Jr., Arts of Power: Statecraft and Diplomacy (Washington, D.C.: United States Institute of Peace Press, 1997), p. 53.

20. 

Ibid., p. 54.

21. 

Slantchev, Military Threats, p. 3.

22. 

Fravel, Strong Borders, Secure Nation, pp. 14–16. For earlier studies centering on issue importance, see Paul F. Diehl, “What Are They Fighting For? The Importance of Issues in International Conflict Research,” Journal of Peace Research, Vol. 29, No. 3 (August 1992), pp. 333–344, doi.org/10.1177/0022343392029003008; and Hemda Ben-Yehuda, “Territoriality and War in International Crises: Theory and Findings, 1918–2001,” International Studies Review, Vol. 6, No. 4 (December 2004), pp. 85–105, doi.org/10.1111/j.1521-9488.2004.00451.x.

23. 

Vesna Danilovic, When the Stakes Are High: Deterrence and Conflict among Major Powers (Ann Arbor: University of Michigan Press, 2002), p. 5.

24. 

Joshua D. Kertzer, “Resolve, Time, and Risk,” International Organization, Vol. 71, Supplement S1 (April 2017), p. S113, doi.org/10.1017/S0020818316000424.

25. 

This is similar to the logic of deterrence. See Glenn H. Snyder, Deterrence and Defense: Toward a Theory of National Security (Princeton, N.J.: Princeton University Press, 1961), p. 13.

26. 

Joshua D. Kertzer, Resolve in International Politics (Princeton, N.J.: Princeton University Press, 2016), p. 3.

27. 

Jonathan Mercer, Reputation and International Politics (Ithaca, N.Y.: Cornell University Press, 1996), p. 6. For an overview of reputation, see Paul K. Huth, “Reputations and Deterrence: A Theoretical and Empirical Assessment,” Security Studies, Vol. 7, No. 1 (Autumn 1997), pp. 72–99, doi.org/10.1080/09636419708429334.

28. 

Thomas C. Schelling, The Strategy of Conflict (Cambridge, Mass.: Harvard University Press, 1980), p. 34.

29. 

Schelling, Arms and Influence, p. 150.

30. 

Robert Jervis, The Logic of Images in International Relations (New York: Columbia University Press, 1989), pp. 78–79.

31. 

For discussion on images and signals, see ibid.

32. 

Nicholas L. Miller, “The Secret Success of Nonproliferation Sanctions,” International Organization, Vol. 68, No. 4 (Fall 2014), pp. 913–944, doi.org/10.1017/S0020818314000216.

33. 

Anne E. Sartori, Deterrence by Diplomacy (Princeton, N.J.: Princeton University Press, 2005), p. 14. Other forms of maintaining credibility include repaying expensive loans. See David Leblang, “To Devalue or Defend? The Political Economy of Exchange Rate Policy,” International Studies Quarterly, Vol. 47, No. 4 (December 2003), pp. 533–560, https://www.jstor.org/stable/3693635; and Michael Tomz, Reputation and International Cooperation: Sovereign Debt across Three Centuries (Princeton, N.J.: Princeton University Press, 2007).

34. 

Dustin H. Tingley and Barbara F. Walter, “The Effect of Repeated Play on Reputation Building: An Experimental Approach,” International Organization, Vol. 65, No. 2 (Spring 2011), p. 344, doi.org/10.1017/S0020818311000026.

35. 

Todd S. Sechser, “Goliath's Curse: Coercive Threats and Asymmetric Power,” International Organization, Vol. 64, No. 4 (October 2010), pp. 627–660, doi.org/10.1017/S0020818310000214; and Todd S. Sechser, “Reputations and Signaling in Coercive Bargaining,” Journal of Conflict Resolution, Vol. 62, No. 2 (February 2018), pp. 318–345, doi.org/10.1177/0022002716652687.

36. 

Barbara F. Walter, Reputation and Civil War: Why Separatist Conflicts Are So Violent (Cambridge: Cambridge University Press, 2009).

37. 

Kertzer's Resolve in International Politics contains excellent research on who is more likely to display resolve. Other scholars examine whether or when states gain resolve. See Daryl Press, Calculating Credibility: How Leaders Assess Military Threats (Ithaca, N.Y.: Cornell University Press, 2007); Roseanne W. McManus, Statements of Resolve: Achieving Coercive Credibility in International Conflict (Cambridge: Cambridge University Press, 2017); and Timothy M. Peterson, “Sending a Message: The Reputation Effect of U.S. Sanction Threat Behavior,” International Studies Quarterly, Vol. 57, No. 4 (December 2013), pp. 672–682, doi.org/10.1111/isqu.12017.

38. 

A large-n study shows that “weaker states are more likely to initiate territorial disputes if stronger states have yielded in previous disputes,” proving the coercer's concern about resolve. See Chong Chen, “Territorial Dispute Initiation by Weaker States,” Chinese Journal of International Politics, Vol. 11, No. 3 (Autumn 2018), pp. 339–372, doi.org/10.1093/cjip/poy009.

39. 

Albert O. Hirschman, National Power and the Structure of Foreign Trade (Berkeley: University of California Press, 1945), p. 17.

40. 

Robert O. Keohane and Joseph S. Nye, Power and Interdependence: World Politics in Transition (Boston: Little, Brown, 1977), p. 13.

41. 

For discussion of economic and opportunity costs of war, see Kertzer, Resolve in International Politics, p. 17.

42. 

Scott L. Kastner, Political Conflict and Economic Interdependence across the Taiwan Strait and Beyond (Palo Alto, Calif.: Stanford University Press, 2009), pp. 11–12. Japan's “comprehensive security policy” also emphasizes economic costs. See Tsuneo Akaha, “Japan's Comprehensive Security Policy: A New East Asian Environment,” Asian Survey, Vol. 31, No. 4 (April 1991), pp. 324–340, doi.org/10.2307/2645387.

43. 

Kastner, Political Conflict and Economic Interdependence across the Taiwan Strait and Beyond, pp. 7, 14.

44. 

Jervis, Perception and Misperception in International Politics, pp. 58–60.

45. 

Randall L. Schweller, Unanswered Threats: Political Constraints on the Balance of Power (Princeton, N.J.: Princeton University Press, 2006), p. 9.

46. 

Stephen M. Walt, The Origins of Alliances (Ithaca, N.Y.: Cornell University Press, 1983).

47. 

Kertzer, “Resolve, Time, and Risk,” p. S120.

48. 

Elsewhere, I analyze China's hierarchy of issue importance and examine issues that Beijing considers the most important—for example, Taiwan, where China pursued coercion despite equally high economic cost and a need to establish resolve. Ketian Zhang, “Calculating Bully: Explaining Chinese Coercion,” Ph.D. dissertation, Massachusetts Institute of Technology, 2018.

49. 

Ryan Brutger and Joshua D. Kertzer, “A Dispositional Theory of Reputation Costs,” International Organization, Vol. 72, No. 3 (Summer 2018), p. 693, doi.org/10.1017/S0020818318000188.

50. 

Andrew J. Nathan, “China's Changing of the Guard: Authoritarian Resilience,” Journal of Democracy, Vol. 14, No. 1 (January 2003), pp. 6–17, doi.org/10.1353/jod.2003.0019; and Gary W. Cox and Barry R. Weingast, “Executive Constraint, Political Stability, and Economic Growth,” Comparative Political Studies, Vol. 51, No. 3 (March 2017), pp. 279–303, doi.org/10.1177/0010414017710254.

51. 

Charles R. Schwenk, “Cognitive Simplification Processes in Strategic Decision-Making,” Strategic Management Journal, Vol. 5, No. 2 (April/June 1984), pp. 111–128, doi.org/10.1002/smj.4250050203.

52. 

Kertzer measures resolve in Resolve in International Politics.

53. 

Jonathan Renshon, Fighting for Status: Hierarchy and Conflict in World Politics (Princeton, N.J.: Princeton University Press, 2017), p. 24.

54. 

Allison Carnegie and Austin Carson, “The Spotlight's Harsh Glare: Rethinking Publicity and International Order,” International Organization, Vol. 72, No. 3 (Summer 2018), p. 628, doi.org/10.1017/S0020818318000176.

55. 

Specific balancing measures include defense technology cooperation, establishment of new bases, and force rotation. See Adam P. Liff, “Whither the Balancers? The Case for a Methodological Reset,” Security Studies, Vol. 25, No. 3 (2016), p. 436, doi.org/10.1080/09636412.2016.1195624.

56. 

Michael D. Swaine has a useful discussion of Chinese language sources. See Swaine, “Chinese Leadership and Elite Responses to the U.S. Pacific Pivot,” China Leadership Monitor, Vol. 38 (Summer 2012), p. 1, http://carnegieendowment.org/files/Swaine_CLM_38_Final_Draft_pdf.pdf.

57. 

CASS is under China's State Council; CICIR is under the Ministry of State Security; NISCSS is under the MFA and the State Council; CIIS is under the MFA; and AMS is under the Chinese People's Liberation Army. They report to their respective government branches, and some of the analysts are former government officials.

58. 

I code these two categories because an internal document by China's SOA stated in 2002 that sovereign, jurisdictional, and administrative rights are the core of maritime rights and that resource exploration in one's exclusive economic zones (EEZ) and continental shelves is an exclusive right that is “quasi-sovereign,” indicating its significance. Internal materials edited by SOA's China Institute for Maritime Affairs (CIMA), Zhuanshu jingji qu he dalujia (EEZs and the continental shelf) (Beijing: Oceanic Press, 2002), pp. 395, 398. For data and sources, see the online appendix at doi.org/10.7910/DVN/7HTIQ1. For any incident or case of Chinese coercion to become part of the database, I use both Chinese-language materials and English-language media reports to avoid any bias. I include all four kinds of Chinese coercion from 1990 to 2015. This database adds to Todd S. Sechser's database on militarized compellent threats. See Sechser, “Militarized Compellent Threats, 1918–2001,” Conflict Management and Peace Science, Vol. 28, No. 4 (September 2011), pp. 377–401, doi.org/10.1177/0738894211413066.

59. 

Zhang, “Calculating Bully.”

60. 

China used proactive coercion in 2015, but did not use coercion when responding to energy-related or land feature–related incidents.

61. 

Despite reform in March 2018, which put the Chinese Coast Guard under the Chinese Armed Police, the Chinese Coast Guard is not assigned a number of PLA troops. “Reforms Regarding Party and Government Organizations,” Xinhua News, March 21, 2018, http://www.xinhuanet.com/politics/2018-03/21/c_1122570517_6.htm. See also Ministry of Defense, “The Chinese Armed Police Force Is Centralized and Unified by the Party Central Committee and the Central Military Commission,” press release (Beijing: Ministry of Defense, December 28, 2017), http://news.163.com/17/1228/19/D6P2D0UE00018AOQ.html; author interview, KZ #123, Beijing, China, December 26, 2018; and author interview, KZ #125, Beijing, China, January 8, 2019.

62. 

Author interview with a former Chinese diplomat appointed to Southeast Asia, KZ #91, Beijing, China, June 7, 2016. In China's gray-zone coercion against Vietnam in 2014, Chinese naval ships were looming far back. Yet, this does not make Chinese behavior militarized coercion. First, Chinese naval ships did not directly confront Vietnamese vessels, which was different from the 1990s, when China used naval ships to expel foreign vessels. Second, Chinese gray-zone coercion in the post-2007 period involved mostly low-level cases.

63. 

See the online appendix. As mentioned, these incidents include other claimants' seizure of land features, fortification of previously occupied land features, and oil and gas contracts and exploration activities with foreign companies. Separating incidents regarding land features and oil exploration yields similar trends as those in figure 2.

64. 

See the online appendix.

65. 

I used the following search in Factiva: “(South China Sea or Spratly) not (typhoon or storm or piracy or pirate or rescue or refugee or EP-3 or code of conduct or declaration of conduct or crash or CNOOC or HD-981 or 981 or Ocean Oil 981 or Hai Yang Shi You or reclamation or oil rig or flight 370 or MH370 or artificial island),” yielding 11,621 articles. This search excludes positive developments between China and other claimants (e.g., signing of the Declaration on the Conduct of the Parties in the South China Sea); Chinese coercion (e.g., land reclamation); and irrelevant reports (e.g., typhoons, the MH370 plane crash, Taiwan Strait crisis, and piracy). I read all 11,621 articles to exclude additional articles that were about Chinese coercion or positive developments, or were simply irrelevant to the topic. I made sure that the data contain only reports of actions by other claimants and articles that mention South China Sea disputes, not Chinese coercion or positive developments.

66. 

The Factiva search shows that the number of reports peaked in 2016 then declined from 2017 to 2019 to the 2013–14 level. Incidents of Chinese land reclamation peaked in 2015 then slowed after 2017, coinciding with the decrease of reports.

67. 

MFA, Zhongguo waijiao gailan, 1991 (China's foreign affairs overview, 1991) (Beijing: World Knowledge Press, 1991), p. 49.

68. 

Sun Xiaoying, “Buzhan erzheng de heping zhanlue yu heping jiejue nansha zhengduan” (Using the peaceful strategy to resolve the Spratly disputes), in Asia-Pacific Office of CASS, ed., Nansha wenti yanjiu ziliao (Materials regarding the South China Sea issue) (Beijing: Asia-Pacific Office of CASS, 1993), p. 278.

69. 

Ibid., p. 280; and Lu Jianren, “Nansha zhengduan ji duice” (Countermeasures for the Spratly disputes), in ibid., p. 307.

70. 

CIMA of the SOA, Nanhai zhudao xueshu taolunhui lunwen xuanbian (Papers of the seminar on islands in the South China Sea), internal publication (Beijing: CIMA of the SOA, March 1992), p. 63.

71. 

White Paper on China's National Defense in 2000, http://www.mod.gov.cn/regulatory/2011-01/07/content_4617805.htm; White Paper on China's National Defense in 2002, http://www.mod.gov.cn/regulatory/2011-01/06/content_4617806.htm; and White Paper on China's National Defense in 2004, http://www.mod.gov.cn/regulatory/2011-01/06/content_4617807.htm, all published by China's Ministry of Defense.

72. 

Jia Yu et al., “Zhongguo zhoubian haiyang xingshi zongshu” (China's maritime situation in 2004–05), in Gao Zhiguo and Zhang Haiwen, eds., Haiyang guoce wenji (Studies on national maritime policies) (Beijing: Oceanic Press, 2007), pp. 207, 242.

73. 

NISCSS, 2003nian nanhai diqu xingshi pinggu baogao (The 2003 report regarding the situation in the South China Sea), printed by NISCSS in July 2004 for internal use, p. 5; and NISCSS, 2004nian nanhai diqu xingshi pinggu baogao (The 2004 report regarding the situation in the South China Sea), printed by NISCSS in 2005 for internal use, p. 4. These books are available in the library of the NISCSS in Haikou.

74. 

Author interview, KZ #114, Beijing, China, December 29, 2016; and author interview, KZ #64, Beijing, China, April 27, 2016.

75. 

NISCSS, 2007nian nanhai diqu xingshi pinggu baogao (The 2007 report regarding the situation in the South China Sea), printed for internal use (Haikou: NISCSS, 2008), p. 4; and NISCSS, 2008nian nanhai diqu xingshi pinggu baogao (The 2008 report regarding the situation in the South China Sea), printed for internal use (Haikou: NISCSS, 2009), pp. 3–4.

76. 

NISCSS, 2007nian nanhai diqu xingshi pinggu baogao, pp. 15, 41.

77. 

China's Ministry of Defense, White Paper on China's National Defense in 2010, http://www.mod.gov.cn/regulatory/2011-03/31/content_4617810.htm.

78. 

Zhang Jie and Zhong Feiteng, “2010nian zhongguo zhoubian anquan xingshi yu zhongguo duice” (The regional security environment in 2010 and China's countermeasures), in Zhang Jie and Yang Danzhi, eds., Zhongguo zhoubian anquan xingshi pinggu (Assessment of China's regional security environment) (Hong Kong: Hong Kong Social Sciences Press, 2011), p. 7. This book was circulated internally in China; CICIR, Guoji zhanlue yu anquan xingshi pinggu (Strategic and security review, 2011/2012) (Beijing: Shishi Press, 2012), pp. 114–115.

79. 

See NISCSS, 2008nian nanhai diqu xingshi pinggu baogao, p. 11.

80. 

Wang Guopei, “Jiefangjun fuzongzhang: fandui nanhaiwenti guojihua, fandui waibushili jieru” (The Deputy Chief of Staff of the PLA: against the internationalization of South China Sea issues and intervention from external actors), Dongfang zaobao (Eastern Morning Daily), April 28, 2010.

81. 

Author interview, KZ #4, Beijing, China, September 15, 2015; author interview, KZ #5, Beijing, China, September 16, 2015; author interview, KZ #11, Beijing, China, October 14, 2015; author interview, KZ #12, Beijing, China, October 21, 2015; author interview, KZ #16, Guangzhou, China, November 30, 2015; author interview, KZ #17, Guangzhou, China, December 1, 2015; author interview, KZ #18, Guangzhou, China, December 3, 2015; author interview, KZ #19, Guangzhou, China, December 4, 2015; author interview, KZ #30, Haikou, China, January 6, 2016; author interview, KZ #34, Haikou, China, January 8, 2016; author interview, KZ #53, Atlanta, United States, March 17, 2016; and author interview, KZ #69, Shanghai, China, May 5, 2016.

82. 

Author interview, KZ #8, Beijing, China, October 6, 2015; and author interview, KZ #11.

83. 

Author interview, KZ #26, Nanjing, China, December 30, 2015; and Zhang Jie, “Huanngyandao moshi yu zhongguo haiyang weiquan zhengce dezhuanxiang” (The Scarborough model and shifts in China's maritime rights protection), Dongnanya yanjiu (Southeast Asia Studies), No. 4 (2013), pp. 25–31, http://niis.cass.cn/webpic/web/niis/upload/2013/12/d20131203161227727.pdf.

84. 

Wang Yong, “Weihu haiyang quanyi shiyixiang changqi de zhanlue renwu” (Maritime rights protection is one long-term mission), in Wu Shicun and Zhu Huayou, eds., Jujiao nanhai—diyuan zhengzhi, ziyuan, hangdao (Focusing on the South China Sea—geopolitics, resources, and SLOCs) (Beijing: China Economic Publishing House, 2009), p. 160.

85. 

Data for import and export totals come from editions of the China Statistical Yearbook, available at the China Data Online database, https://www.china-data-online.com/; data regarding ASEAN are from China Customs Data, available in CEIC Data's China Premium Database, https://www.ceicdata.com/en/products/china-economic-databasec; for data regarding ASEAN from 1992 to 1996, see the official versions of the annual Almanac of China's Foreign Economic Relations and Trade from 1993 to 1998, compiled by the Chinese Ministry of Foreign Economics and Trade; and, for data regarding the United States, the EU, and Japan, see the China's Commerce yearbooks, the yearbooks of China's Foreign Economic Relations and Trade, and the MFA's China's Foreign Affairs.

86. 

CEIC Data, China Customs Data. Data regarding ASEAN's export and import totals come from World Trade Organization, International Trade Statistics, 2008 (Geneva: World Trade Organization, 2008), https://www.wto.org/english/res_e/statis_e/its2008_e/its2008_e.pdf.

87. 

Author interview, KZ #40, Beijing, China, January 22, 2016; author interview, KZ #39, Beijing, China, January 22, 2016; and author interview, KZ #42, Beijing, China, January 25, 2016.

88. 

Zhang Yunling, Zai lixiang yu xianshi zhijian: wodui dongya hezuo de yanjiu, canyu, he sikao (Between ideals and reality: My analysis, participation, and thoughts regarding East Asian cooperation) (Beijing: China Social Science Press, 2015), p. 12. Premier Zhu Rongji instructed the MFA, the Ministry of Foreign Economics and Trade, and CASS to establish an expert group to evaluate the ASEAN-China FTZ. Zhang was a member.

89. 

Ibid., p. 97.

90. 

He Xiaoqin, “Zhongguo dongmeng zimaoqu de mubiao, jincheng, yu chengbenshouyi fenxi” (Goals, process, and benefit analysis of the Sino-ASEAN FTZ), Shijie jingji yanjiu (World Economic Research), No. 6, 2003, pp. 70–74.

91. 

Zhang, Zai lixiang yu xianshi zhijian, p. 113.

92. 

Zhang Zhen and Peng Yun, “Shixi goujian zhongguo-dongmeng ziyou maoyiqu zhongde dongmeng yinsu” (ASEAN factors regarding the establishment of the Sino-ASEAN FTZ), Dongnanya zongheng (Around Southeast Asia), No. 10, 2002, pp. 7–11.

93. 

Zhang, Zai lixiang yu xianshi zhijian, p. 97.

94. 

Ministry of Foreign Affairs of Japan, “The Japan-ASEAN Plan of Action” (Tokyo: Ministry of Foreign Affairs of Japan, 2003), http://www.mofa.go.jp/region/asia-paci/asean/year2003/summit/action.pdf; and Tong Fuquan, “Zhongguo-dongmeng ziyou maoyiqu gouxiang yu nanti” (The construction and difficulty of the ASEAN-China FTZ), Guoji maoyi (International Trade), No. 2 (2002), p. 25.

95. 

Author interview, KZ #59, Wuhan, China, April 18, 2016; author interview, KZ #64; and CIMA, Zhuanshu jingji qu he dalujia, p. 144.

96. 

NISCSS, 2008nian nanhai diqu xingshi pinggu baogao, p. 51.

97. 

Author interview, KZ #19; and author interview, KZ #35, Beijing, China, January 18, 2016.

98. 

China's State Council, “China's Government Work Report, 2008,” Beijing, March 5, 2008, http://www.gov.cn/premier/2009-03/16/content_1260198.htm; Hu Jintao, Hu Jintao wenxuan disan juan (Hu Jintao's selected works), Vol. 3 (Beijing: People's Publishing House, 2016), p. 335; and China's State Council, “China's Government Work Report, 2011,” Beijing, March 5, 2011, http://www.gov.cn/2011lh/content_1825233.htm.

99. 

Zhang Yunling, ed., Zhongguo duiwai guanxi: huigu yu sikao (China's foreign relations: Retrospection and thoughts) (Beijing: Social Sciences Academic Press, 2009), p. 222.

100. 

Some analysts suggest that China's national defense white papers contain more nuance. The white papers, however, are not the best indicator of China's view on multipolarity. First, unlike the MFA's annual China's Foreign Affairs, which began publication in 1987, the biannual defense white papers began only in 2000. What they would have stated in the years when they were not published is unknown. My conjecture is that it is in line with the MFA, as seen in assessments in 2010 and 2012. I do not mean that the MFA has only negative assessments of multipolarity after 2000. From 2000 to 2015, there were eleven years—a majority—in which the MFA either had a negative assessment of multipolarity or did not mention it at all. Second, to measure geopolitical cost, I also use official U.S. documents and Chinese official and government think-tank assessments— some of which are internal—for cross-checking purposes. Third, the trend of U.S. military visits to Southeast and East Asia also corroborates the high geopolitical cost since the 2000s. For data on the visits, see Vito D'Orazio “U.S. Military Visits: 1990–2010” (Vito D'Orazio's personal website, n.d.), https://www.vitodorazio.com/data.html.

101. 

MFA, Zhongguo waijiao gailan, 1993 (China's foreign affairs overview, 1993) (Beijing: World Knowledge Press, 1993), p. 27.

102. 

MFA, Zhongguo waijiao, 1997 (China's foreign affairs, 1997) (Beijing: World Knowledge Press, 1997), p. 549; MFA, Zhongguo waijiao, 1998 (China's foreign affairs, 1998) (Beijing: World Knowledge Press, 1998), p. 540; MFA, Zhongguo waijiao, 1999 (China's foreign affairs, 1999) (Beijing: World Knowledge Press, 1999), p. 432; and MFA, Zhongguo waijiao, 2000 (China's foreign affairs, 2000) (Beijing: World Knowledge Press, 2000), p. 471.

103. 

China's Ministry of Defense, White Paper on China's National Defense in 2000; and China's Ministry of Defense, “China's National Defense, 1998” (Beijing: State Council Information Office, July 1998), http://www.gov.cn/zwgk/2005-05/26/content_1107.htm.

104. 

White House, “A National Security Strategy for a Global Age,” December 2000, National Security Strategy Archive, Dover, New Hampshire, http://nssarchive.us/NSSR/2001.pdf.

105. 

MFA, Zhongguo waijiao, 2001 (China's foreign affairs, 2001) (Beijing: World Knowledge Press, 2001), pp. 449–454.

106. 

MFA, Zhongguo waijiao, 2002 (China's foreign affairs, 2002) (Beijing: World Knowledge Press, 2002).

107. 

Qi Ju (Seventh Bureau), Daguo wenti yanjiu zhuanti baogao huibian (Compiled reports on the studies of great power issues), internal publication (Beijing: CCP Central International Liaison Department, April 2004), p. 297.

108. 

See White Paper on China's National Defense in 2008, http://www.mod.gov.cn/regulatory/2011-01/06/content_4617809.htm; White Paper on China's National Defense in 2010; White Paper on China's National Defense in 2013, http://www.mod.gov.cn/regulatory/2013-04/16/content_4617811_2.htm; and White Paper on China's National Defense in 2015, http://www.mod.gov.cn/regulatory/2015-05/26/content_4617812.htm, all published by China's Ministry of Defense.

109. 

Sun, “Buzhan erzheng de heping zhanlue yu heping jiejue nansha zhengduan,” p. 280.

110. 

Author interview, KZ #16; author interview, KZ #25, Nanjing, China, December 30, 2015; and author interviews, KZ #17 and KZ #19.

111. 

CICIR, Guoji zhanlue yu anquan xingshi pinggu (Strategic and security review, 2001/02) (Beijing: Shishi Press, 2002), p. 27. Nina Silove's article and my research on U.S.-ASEAN presidential meetings also note this trend. See Silove, “The Pivot before the Pivot: U.S. Strategy to Preserve the Power Balance in Asia,” International Security, Vol. 40, No. 4 (Spring 2016), pp. 45–88, doi.org/10.1162/ISEC_a_00238; and Ketian Zhang, “A View from the United States,” Asan Forum, Vol. 7, No. 3 (May–June 2019), http://www.theasanforum.org/a-view-from-the-united-states-6/.

112. 

CASS, 2003nian guoji xingshi huangpishu (2003 yellow book of international politics) (Beijing: Social Sciences Academic Press, 2003), pp. 102–104; AMS, 20003nian ban zhanlue pinggu (2003 strategic assessment) (Beijing: Academy of Military Sciences Press, 2004), pp. 12–22, 168; and CIIS, 2005/2006nian guoji xingshi he zhongguo waijiao lanpishu (2005/2006 bluebook of international situation and China's foreign affairs) (Beijing: Contemporary World Press, 2006), pp. 16, 28.

113. 

Zhang and Zhong, “2010nian zhongguo zhoubian anquan xingshi yu zhongguo duice,” pp. 1, 4.

114. 

Floyd Whaley, “Philippines and China in a Standoff at Sea,” New York Times, April 12, 2012, https://archive.nytimes.com/query.nytimes.com/gst/fullpage-9C05E1D91431F931A25757C0A9649D8B63.html; author interview, KZ #34; author interview, KZ #84, Guangzhou, China, May 21, 2016; and author interview, KZ #91.

115. 

Author interview, KZ #106, Washington, D.C., September 22, 2016.

116. 

Author interview, KZ #92, Xiamen, China, June 13, 2016; Chinese embassy in Manila, “Huangyandao shiwen” (Ten questions regarding the Scarborough Shoal), June 15, 2012, https://www.fmprc.gov.cn/ce/ceph/chn/zgxw/t941671.htm; and Whaley, “Philippines and China in a Standoff at Sea.”

117. 

“Philippines, China ‘Set Aside’ Protests to Ease Tensions over Sea Dispute,” BBC Monitoring Asia Pacific, April 14, 2012, supplied by BBC Worldwide Monitoring.

118. 

Jaime Marie Elona, “Philippines Urges China to Bring Shoal Row to International Court,” Inquirer, April 17, 2012, http://globalnation.inquirer.net/33579/philippines-urges-china-to-bring-Shoal-row-to-international-court.

119. 

Carnegie Endowment, “Nanzhongguohai dashiki” (A chronology of South China Sea events) (Washington, D.C.: Carnegie Endowment, September 4, 2012), http://carnegieendowment.org/files/SCS_Timeline_since_2006.pdf.

120. 

Michael Green et al., “Countering Coercion in Maritime Asia.”

121. 

CIMA, 2014 zhongguo haiyang fazhan bagao (2014 report of China's maritime development) (Beijing: Oceanic Press, 2014), p. 347.

122. 

See “U.S. Strategists Face Dilemma over Beijing Claim in South China Sea,” Financial Times, July 9, 2014; and Rodel Rodis, “Did Trillanes Commit Treason in the Loss of Scarborough Shoal?” Inquirer, May 25, 2016, http://globalnation.inquirer.net/139658/139658.

123. 

“Feilvbin xiangjiaoshang: 3yue yilai yizai zhongguo sunshi yue 10yi bisuo” (Philippine banana sellers have lost about 1 billion pesos in China since March), Qianjiang Wanbao (Qianjiang Evening News), May 14, 2012, http://news.ifeng.com/mainland/special/nanhaizhengduan/content-3/detail_2012_05/14/14502214_0.shtml?_from_ralated.

124. 

“Feixiangjiao sunshi yida 1.5yi renminbi, nongye guanyuan fanghua qiuqing” (The Philippine banana loss has reached 0.15 billion RMB, agricultural officials visited China for forgiveness), Renmin Wang (People's Net), May 17, 2012, http://news.ifeng.com/mainland/special/nanhaizhengduan/content-3/detail_2012_05/17/14608569_0.shtml?_from_ralated. People's Net is the internet complement of People's Daily.

125. 

Author interview, KZ #9, Beijing, China, October 9, 2015; and author interview, KZ #17. See also Ge Hongliang, “Nanhai diqu anquan xingshi” (Security situations in the South China Sea region), in Ju Hailong, ed., Nanhai diqu xingshi baogao 2013–2014 (Report on the South China Sea) (Beijing: Shishi Press, 2015), p. 7.

126. 

MFA, press conference, April 12, 2012, http://www.mfa.gov.cn/ce/ceee/chn/ztlm/zgwjbfyrth/t922354.htm; MFA, press conference, April 27, 2012, http://www.mfa.gov.cn/web/fyrbt_673021/jzhsl_673025/t927033.shtml; MFA, press conference, April 28, 2012, http://www.fmprc.gov.cn/web/fyrbt_673021/dhdw_673027/t927488.shtml; MFA, press conference, April 30, 2012, http://www.fmprc.gov.cn/web/fyrbt_673021/dhdw_673027/t927973.shtml; MFA, press conference, May 11, 2012, http://www.mfa.gov.cn/ce/ceee/chn/ztlm/zgwjbfyrth/t930962.htm; and MFA, press conference, May 30, 2012, http://www.fmprc.gov.cn/ce/ceae/chn/wjbfyrth/t936494.htm.

127. 

MFA, press conference, April 27, 2012, http://www.mfa.gov.cn/web/fyrbt_673021/jzhsl_673025/t927033.shtml; MFA, press conference, April 30, 2012, http://www.fmprc.gov.cn/web/fyrbt_673021/dhdw_673027/t927973.shtml; MFA, press conference, May 11, 2012, http://www.mfa.gov.cn/ce/ceee/chn/ztlm/zgwjbfyrth/t930962.htm; and MFA, press conference, May 30, 2012, http://www.fmprc.gov.cn/ce/ceae/chn/wjbfyrth/t936494.htm.

128. 

Author interview, KZ #54, Haikou, China, April 8, 2016; author interview, KZ #57, Haikou, China, April 14, 2016; and author interview, KZ #69.

129. 

Zhang Liangfu, Jujiao zhongguo haijiang (Focusing on China's maritime territory) (Beijing: Oceanic Press, 2013), p. 109.

130. 

Barbara Mae Dacanay, “Navy and Coast Guard Remove Chinese Markers,” Gulf News (United Arab Emirates), June 16, 2011.

131. 

Lindsay Murdoch, “Islands Off Agenda but Still Split ASEAN,” The Age (Melbourne, Australia), April 4, 2012; and Agence France-Presse, “Philippines to Ramp Up Oil Exploration,” April 26, 2011, https://m.vietnambreakingnews.com/2011/04/philippines-to-ramp-up-oil-exploration.

132. 

Agence France-Presse, “Filipinos Fly Flag in South China Sea,” Times of Oman, July 23, 2011.

133. 

Data on the Philippines come from Factiva searches of the Philippine Daily Inquirer, Manila Times, and Philippine Star. The wording of the searches includes “South China Sea,” “West Philippine Sea,” and “Spratlys.”

134. 

“Interview, Philippines: S. China Sea Code Doesn't Solve Sovereignty,” Reuters, September 16, 2011.

135. 

Manuel Mogato and Paul Eckert, “Clinton in Manila amid ASEAN Row over South China Sea,” Reuters, November 15, 2011, https://www.reuters.com/article/us-asean-usa-southchinasea-idUSTRE7AE0RW20111115.

136. 

For example, “China-Philippines Dispute: Timeline,” Wall Street Journal, n.d., https://graphics.wsj.com/embeddable-carousel/?slug=Hague-china; and Jim Gomez, “Philippine Warship in Standoff with China Vessels,” Star Advertiser, April 10, 2012, http://www.staradvertiser.com/2012/04/10/breaking-news/philippine-warship-in-standoff-with-china-vessels.

137. 

Manuel Mogato, “Manila Summons China's Envoy over South China Sea Standoff,” Reuters, April 10, 2012, https://www.reuters.com/article/us-philippines-china-sea/manila-summons-chinas-envoy-over-south-china-sea-standoff-idUSBRE83A02G20120411.

138. 

Zhongsheng, “Yanzhong de zhanlue wupan” (A serious strategic miscalculation), People's Daily, August 2, 2011, http://military.people.com.cn/GB/15305363.html.

139. 

Ibid.

140. 

Zhang Jie and Zhong Feiteng, eds., 2012nian Zhongguo zhoubian anquan xingshi pinggu (The 2012 assessment of China's regional security environment) (Beijing: Social Sciences Academic Press, January 2012), pp. 97–98.

141. 

MFA, “Press Conference with Foreign Ministry Spokesperson Hong Lei,” February 28, 2012, http://www.fmprc.gov.cn/web/fyrbt_673021/jzhsl_673025/t909155.shtml; for similar Chinese accusations, see MFA, “Press Conference with Foreign Ministry Spokesperson Hong Lei,” March 29, 2012, http://www.fmprc.gov.cn/web/fyrbt_673021/jzhsl_673025/t918673.shtml; and MFA, “Press Conference with Foreign Ministry Spokesperson Hong Lei,” March 22, 2012, http://www.fmprc.gov.cn/web/fyrbt_673021/jzhsl_673025/t916303.shtml.

142. 

Zhongsheng, “Feilvbn zai naihai wenti shang xuxuehui ziwo yueshu” (The Philippines should learn about self-restraint in the South China Sea issue), People's Daily, February 29, 2012, http://world.people.com.cn/GB/17248392.html.

143. 

Ibid.

144. 

Ibid.

145. 

See the results from the People's Daily database, http://data.people.com.cn/rmrb/20190513/1?code=2.

146. 

“Zhongguo jiu huangyandao zhize feilvbin kuoda shitai” (China accused the Philippines of escalating tension in the Scarborough Shoal), BBC News (Chinese), May 8, 2012, https://www.bbc.com/zhongwen/simp/chinese_news/2012/05/120508_china_philippines.

147. 

Ibid.

148. 

Qin Hong, “Miandui feilvbin, women you zugou shouduan” (Faced with the Philippines, we have adequate means), Overseas Edition of People's Daily, May 8, 2012, http://cpc.people.com.cn/pinglun/GB/17831131.html.

149. 

“Dai Bingguo: Fei deng xiaoguo buneng qifu daguo” (Dai Bingo said that small states like the Philippines must not bully great powers), Nanyang Sin-Chew Lianhe Zaobao, May 16, 2012, https://www.zaobao.com.sg/wencui/politic/story20120516-178565.

150. 

Author interview, KZ #88, Guangzhou, China, May 25, 2016.

151. 

Author interview, KZ #54.

152. 

Author interview, KZ #112.

153. 

Author interview, KZ #58, Haikou, China, April 14, 2016; and author interview, KZ #20, Beijing, China, December 9, 2015.

154. 

Author interviews, KZ #57 and KZ #84, respectively. See also author interviews, KZ #92 and KZ #64; and author interview, KZ #6, Beijing, China, September 28, 2015.

155. 

Author interview, KZ #10, Beijing, China, October 13, 2015; and author interview, KZ #122, Palo Alto, California, October 6, 2018.

156. 

Economic and Commercial Counselor's Office of the Chinese Embassy in the Philippines, “Zhongfei jingmao guanxi gaikuang” (Overview of Sino-Philippine relations), January 26, 2010, http://ph.mofcom.gov.cn/article/zxhz/hzjj/201001/20100106762477.shtml.

157. 

See data from the World Bank Group, “Philippines Exports, Imports, and Trade Balance, by Country and Region,” n.d., https://wits.worldbank.org/CountryProfile/en/Country/PHL/Year/2011/TradeFlow/EXPIMP/Partner/all.

158. 

Slides from Pilipino Banana Growers and Exporters Association, “Philippine Banana Exports to China: Dealing with Sanitary and Phytosanitary Concerns” (Davao City, Philippines: Pilipino Banana Growers and Exporters Association, June 11, 2015), http://appfi.ph/images/2015/presentations/6_Pres_PBGEA_Philippine_Banana_Exports_to_China.pdf.

159. 

Economic and Commercial Counselor's Office of the Chinese Embassy in the Philippines, “Zhongfei jingmao guanxi gaikuang.”

160. 

He Shengda, “Dongmeng duihua guanxi de xianzhuang yu weilai” (The reality and future of Sino-ASEAN relations), in Zhang Yunling, ed., Zhongguo yu zhoubian guojia: goujian xinxing huoban guanxi (China and neighboring countries: Constructing new partnerships) (Beijing: Social Science Academic Press, 2008), p. 92.

161. 

Chinese Ministry of Commerce, “Fei suowei jingjizhicai zitaokuchi” (The Philippines would be asking for trouble if it sought economic sanctions), May 14, 2012, http://chinawto.mofcom.gov.cn/article/e/s/201205/20120508122972.shtml.

162. 

Ibid.

163. 

Author interviews, KZ #54, KZ #57, and KZ #69.

164. 

Author interview, KZ #12.

165. 

Xue Li and Hu Bo, “Shendu jiexi zhongguo de haiyang qiangguo zhilu” (A deep analysis of China's path to becoming a maritime power), Fenghuang Bobao (Ifeng News), August 4, 2015, http://uscnpm.com/model_item.html?action=view&table=article&id=5133; author interview, KZ #85, Guangzhou, China, May 23, 2016; and author interviews, KZ #92 and KZ #106.

166. 

Liu Zaorong, “Meiguo nanhai zhengce xin dongxiang” (New moves in U.S. South China Sea policy), presented at a seminar by the Collaborative Innovation Center of South China Sea Studies, Beijing, August 15–16, 2015, p. 232. This is a print copy of the conference materials, internally circulated and available at the China Institute of Boundary and Ocean Studies of Wuhan University. See also Wu Shicun, Zonglun nansha zhengduan (On the South China Sea dispute) (Haikou: Hainan Press, 2005), internal circulation, p. 189.

167. 

Author interview, KZ #84.

168. 

Author interview, KZ #4.

169. 

Ju Hailong, “Aobama zhengfu nanhai zhengce yanjiu” (Analysis of the Obama administration's South China Sea policies), Dangdai yatai (Journal of Contemporary Asia-Pacific Studies), No. 3, 2011, p. 111; and author interview, KZ #85.

170. 

Author interviews, KZ #57 and KZ #64; and author interview, KZ #90, Guangzhou, China, May 25, 2016.

171. 

Liu, “Meiguo nanhai zhengce xin dongxiang,” p. 229.

172. 

Ibid.; author interview, KZ #103, Philadelphia, Pennsylvania, September 2, 2016; and author interview, KZ #106.

173. 

See author interviews, KZ #103 and KZ #106.

174. 

Author interview, KZ #53; author interview, Beijing, China, June 30, 2014; and author interview, KZ #106.

175. 

Agence France-Presse, “U.S. Commander Reaffirms Philippines Defense Treaty,” Rappler, April 22, 2012, http://www.rappler.com/nation/4205-us-commander-reaffirms-philippines-defense-treaty.

176. 

Ibid.

177. 

Hillary Clinton et al., “Remarks with Secretary of Defense Leon Panetta, Philippines Foreign Secretary Albert del Rosario, and Philippines Defense Secretary Voltaire Gazmin after Their Meeting” (Washington, D.C.: U.S. Department of State, April 30, 2012); and U.S. Department of State, “Joint Statement of the United States-Philippines Ministerial Dialogue,” press release (Washington, D.C.: U.S. Department of State, April 30, 2012), quoted in Michael Green et al., “Countering Coercion in Maritime Asia,” p. 110.

178. 

Li Zhujiang and Zhu Jianzhen, eds., Haiyang yu yuye yingji guanli (Ocean and fishery emergency administration) (Beijing: Oceanic Press, 2007), p. 314.

179. 

China Ministry of Agriculture, Zhuanshu jingji qu yuzheng xunhang gongzuo guifan, shixing (Regulations regarding the fishery administration patrol in EEZs) (Beijing: China Ministry of Agriculture, June 2007). See the appendix for a copy of this document.

180. 

Author interview, KZ #17; author interview, KZ #29, Haikou, January 5, 2016; and author interview, KZ #35.

181. 

Author interview, KZ #7, Beijing, September 29, 2015.

182. 

Author interview, KZ #24, Nanjing, December 29, 2015; and author interviews, KZ #30 and KZ #34.

183. 

Author interview, KZ #26.

184. 

Author interview, KZ #32, Haikou, China, January 7, 2016.

185. 

Author interview, KZ #27, Nanjing, China, December 31, 2015; author interviews, KZ #16 and KZ #32; author interview, KZ #113, Beijing, China, December 29, 2016; Hu Angang, Zhongguo jiti lingdao tizhi tanjiu (A study of China's collective leadership institution) (Hong Kong: Chunghwa, 2014), p. 35; Jiang Liu et al., eds., Shehui xingshifenxi yu yuce (1994–1995 report on analysis of social conditions) (Beijing: Social Sciences Press, 1995), pp. 8, 31, 48, 195; and Ru Xin et al., eds., Shehui xingshi fenxi yu yuce (1998 report on analysis of social conditions) (Beijing: Social Sciences Press, 1998), p. 162.

186. 

Andrew Wedeman, “Enemies of the State: Mass Incidents and Subversion in China,” paper presented at the annual meeting of the American Political Science Association, Toronto, Canada, September 3–6, 2009, https://ssrn.com/abstract=1451828.

187. 

Author interviews, KZ #6, KZ #7, KZ #8, KZ #11, and KZ #32.

188. 

Author interview, KZ #69; Wen Siyong and Ren Zhichu, Talingdao zhongguo: Hu Jintao xinzhuan (He leads China: A new biography of Hu Jintao) (New York: Mirror, 2010); and Gao Xiao, Tajiang lingdao zhongguo: Xi Jinping zhuan (He will lead China: Biography of Xi Jinping) (New York: Mirror, 2010).

189. 

Hu Jintao, Hu Jintao wenxuan dierjuan (Hu Jintao's selected works), Vol. 2 (Beijing: People's Publishing House, 2016), p. 518. Hu's speech during this conference was previously not made public.

190. 

Liang Jian, Xi Jinping xinzhuan (New biography of Xi Jinping) (New York: Mirror Books, 2012), p. 498.

191. 

Author interview, KZ #103.

192. 

Author interview, KZ #57.

193. 

Jessica Chen Weiss, Powerful Patriots: Nationalist Protest in China's Foreign Relations (Oxford: Oxford University Press, 2014).

194. 

See Duan Xiaolin, “Unanswered Questions: Why We May Be Wrong about Chinese Nationalism and Its Foreign Policy Implications,” Journal of Contemporary China, Vol. 26, No. 108 (2017), pp. 886–900, doi.org/10.1080/10670564.2017.1337312; and Kai Quek and Alastair Iain Johnston, “Can China Back Down? Crisis De-escalation in the Shadow of Popular Opposition,” International Security, Vol. 42, No. 3 (Winter 2017/18), pp. 7–36, doi.org/10.1162/ISEC_a_00303.

195. 

My study focuses on coercion, not territorial disputes per se. The alternative explanations, therefore, should concern coercion. For recent works on territorial disputes, see David B. Carter, “The Strategy of Territorial Conflict,” American Journal of Political Science, Vol. 54, No. 4 (October 2010), pp. 969–987, doi.org/10.1111/j.1540-5907.2010.00471.x; and Hein E. Goemans and Kenneth A. Schultz, “The Politics of Territorial Claims: A Geospatial Approach Applied to Africa,” International Organization, Vol. 71, No. 1 (Winter 2017), pp. 31–64, https://doi.org/10.1017/S0020818316000254.

196. 

Andrew S. Erickson and Ryan D. Martinson, eds., China's Maritime Gray Zone Operations (Annapolis: Naval Institute Press, 2019).

197. 

The distinction between deterrence and compellence seems sharper than it really is. Glenn Snyder, for example, states that deterrence is the power to dissuade, not the power to compel. Snyder, Deterrence and Defense, p. 9.

198. 

Ian Nish, Japanese Foreign Policy in the Interwar Period (Westport, Conn.: Praeger, 2002); Imanuel Geiss, Germany Foreign Policy, 1871–1914 (London: Routledge and Kegan Paul, 1976); Prosser Gifford and W.M. Roger Louis, eds., Britain and Germany in Africa: Imperial Rivalry and Colonial Rule (New Haven, Conn.: Yale University Press, 1967); Dana Gardner Munro, Intervention and Dollar Diplomacy in the Caribbean, 1900–1921 (Princeton, N.J.: Princeton University Press, 1964).

199. 

For contested status signaling, see Xiaoyu Pu, Rebranding China: Contested Status Signaling in the Changing Global Order (Palo Alto, Calif.: Stanford University Press, 2019).

200. 

Stephen G. Brooks argues that globalized production chains reduce conflicts among great powers. Brooks, Producing Security: Multinational Corporations, Globalization, and the Changing Calculus of Conflict (Princeton, N.J.: Princeton University Press, 2005). See also Erik Gartzke and Oliver Westerwinter, “The Complex Structure of Commercial Peace Contrasting Trade Interdependence, Asymmetry, and Multipolarity,” Journal of Peace Research, Vol. 53, No. 3 (May 2016), pp. 325–343, doi.org/10.1177/0022343316637895.

201. 

James D. Fearon, “Domestic Political Audiences and the Escalation of International Disputes,” American Political Science Review, Vol. 88, No. 3 (September 1994), pp. 577–592, doi.org/10.2307/2944796; and Jessica L. Weeks, “Autocratic Audience Costs: Regime Type and Signaling Resolve,” International Organization, Vol. 62, No. 1 (January 2008), pp. 35–64, doi.org/10.1017/S0020818308080028; and Bruce Bueno de Mesquita and Alastair Smith, “Domestic Explanations of International Relations,” Annual Review of Political Science, Vol. 15 (June 2012), pp. 161–181, doi.org/10.1146/annurev-polisci-070209-174835. For scholarship challenging audience costs, see Marc Trachtenberg, “Audience Costs: An Historical Analysis,” Security Studies, Vol. 21, No. 1 (2012), pp. 3–42, doi.org/10.1080/09636412.2012.650590. See also “Do Audience Costs Exist? A Symposium,” special issue, Security Studies, Vol. 21, No. 3 (2012).

202. 

For scholarship on non-audience cost signals, see Austin Carson and Keren Yarhi-Milo, “Covert Communication: The Intelligibility and Credibility of Signaling in Secret,” Security Studies, Vol. 26, No. 1 (2017), pp. 124–156, doi.org/10.1080/09636412.2017.1243921.

203. 

See James Reilly, “China's Unilateral Sanctions,” Washington Quarterly, Vol. 35, No. 4 (Fall 2012), pp. 123–133, doi.org/10.1080/0163660X.2012.726428.

204. 

This is in line with Todd Sechser's dataset on militarized compellent demands. In the dataset, China engaged in militarized compellence four times in the post–1949 period, all during the Cold War, which was fewer than India (six times), the United States (seventeen times), the Soviet Union/Russia (seven times), or interwar Japan when it was a rising power (twenty-eight times). Sechser, “Militarized Compellent Threats, 1918–2001.”

205. 

Elsewhere, I examine case studies in which China did not coerce the target state given its low need for resolve. Zhang, “Calculating Bully.” Moreover, China's wartime negotiation strategy demonstrated Beijing's concern for showing resolve. Ahsan Butt's research on factors leading to the 2003 U.S. war in Iraq highlights the importance that U.S. decisionmakers gave to establishing resolve. Instead of a classic security dilemma, there may be a “resolve dilemma”: both the coercer and the target perceive a need to show resolve, which could make negotiations difficult. See Oriana Skylar Mastro, The Costs of Conversation: Obstacles to Peace Talks in Wartime (Ithaca, N.Y.: Cornell University Press, 2019); and Ahsan I. Butt, “Why Did the United States Invade Iraq in 2003?” Security Studies, Vol. 28, No. 2 (2019), pp. 250–285, doi.org/10.1080/09636412.2019.1551567.