Aggressive or Peaceful Rise? An Empirical Assessment of China’s Militarized Conflict, 1979–2010

2.1 The Kantian peace

According to the Kantian peace theory, countries that are joint democracies, trade with each other, and share common memberships in international organizations (IOs) are less likely to engage in war or military conflict (Russett and Oneal 2001). Maoz and Russett (1993) discuss a normative model and a structural model to support the argument that wars are less likely to occur between democracies. The political norm of “regulated political competition through peaceful means” (p. 625) and the structural requirement of “the mobilization of both general public opinion and of a variety of institutions that make up the system of government” (p. 626) both lower the probability of war between democracies. Other scholars suggest alternative explanations for the democratic peace theory (e.g. Bueno de Mesquita et al. 1999). States that trade with each other are less likely to engage in conflict. The game-theoretical analysis shows that the opportunity cost argument and the costly signaling proposition both predict an inverse trade conflict relationship (Gartzke, Li, and Boehmer 2001; Polachek and Xiang 2010). Opportunity costs arise because trading states tend to forgo the benefits from trade when a war erupts. Because of the opportunity costs, trade can also serve as a costly signal to reduce the probability of war in an incomplete information game. In addition, ample empirical evidence demonstrates this inverse relationship in the data. (e.g. Oneal and Russett 1999; Polachek 1980; Xiang, Xu, and Keteku 2007). In an influential study, Russett, Oneal, and Davis (1998) argue that international organizations can promote peace through six functions: “coercing norm breakers; mediating among conflicting parties; reducing uncertainty by conveying information; problem-solving…; socialization and shaping norms; and generating narratives of mutual identification” (pp. 444–445). Other studies supply additional arguments and empirical evidence to support the pacifying roles of international organizations (e.g. Caruso 2006; Oneal, Russett, and Berbaum 2003; Pevehouse and Russett 2006).

Because China has significantly increased its international trade and participation in IOs since the economic reform at the end of 1978, the Kantian peace theory suggests that China is rising peacefully.^[9] For illustration, we compare China’s figures on international trade and participation in IOs for the years 1979 and 2010, which are the starting and ending years of our data analysis. Based on WTO trade statistics,^[10] China’s total trade volume was 29.2 billion USD in 1979, compared to 2.97 trillion USD in 2010.^[11] This is an increase of 10,071%. In terms of percentage of world trade, China accounted for less than 1% of the world’s trade in 1979, whereas in 2010 China accounted for almost 10% of world trade. The same is true for China’s participation in IOs. China was a member of 27 IOs in 1979. This number almost tripled in 2005, as China became a member of 75 international organizations.^[12] Given the pacifying effects of trade and IOs, all these figures suggest that China should become less likely to be involved in conflict over this 30-year time period.

Numerous scholars who write on China have utilized the Kantian peace theory (i.e. trade and IOs) to argue that China rises peacefully. For example, Broomfield (2003), Tammen and Kugler (2006), Brandt, Rawski, and Zhu (2007), Goldsmith (2007), and Weede (2010) all argue that since China has substantially benefited from growing trade with other countries, China is rising peacefully. The bilateral relationship between China and Japan clearly exemplifies this argument (e.g. Beeson and Li 2012). Despite the “traumatic history” between these two countries, the deep and growing economic interdependence^[13] makes China “adopt a more conciliatory attitude toward Japan in pursuit of long-term gains” (Beeson and Li 2012, 48).^[14] Similarly, an increased participation in IOs has led China to be more cooperative with other states. China has become more willing to embrace the norms and to preserve the status quo of the existing IOs (e.g. Beeson 2009; Chin and Thakur 2010; Hempson-Jones 2005; Johnston 2003, 2008; Reilly 2012; Zhang and Tang 2005). As a result, China’s increased involvement in IOs implies it rises peacefully.

2.2 Soft power

When a country possesses soft power, it is able to achieve many of its objectives in international relations without the threat to use or the actual use of force (Nye 1990, 2004, 2008). Soft power is defined as “the ability to affect others to obtain the outcomes one wants through attraction rather than coercion or payment,” and a country may use soft power to “obtain the outcomes it wants in world politics because other countries want to follow it, admiring its values, emulating its example, and/or aspiring to its level of prosperity and openness” (Nye 2008, 94). Therefore, assuming its objectives are fixed, a country that has more soft power will be less likely to employ military force.

Many scholars argue that China has employed soft power in its long history and China has increased its use of soft power in recent years.^[15] Based on these arguments, it is expected that China is unlikely to increase its use of military force and as a result China rises peacefully. Ding (2008) and Li and Worm (2011) argue that soft power is not a novelty to China and the essence of soft power has been part of China for 2000 years. Ding (2008) states that “such ideas as attraction and agenda-setting have always embedded in ancient Chinese philosophy and culture” (p. 195). In recent years, the Chinese government has invested substantially in several main areas to advance its soft power. First, the Chinese government has attempted to promote educational and cultural exchanges. To establish Confucius Institutes that teach Chinese language and culture around the world (e.g. Paradise 2009) and to attract international students to study in China are clear examples of China’s attempt to advance its soft power. In addition, the Chinese government has expanded its news programs throughout the world. For example, the China Central Television (CCTV) broadcasts news in the world’s major languages – English, French, Spanish, Arabic, and Russian – in an effort to present a friendly China (Zhu 2010). Third, China has increased its involvement in multilateral organizations. China has significantly increased its UN peace keeping missions and joint military exchanges with the Shanghai Cooperation Organization, which Nye (2008) argues can generate soft power.

2.3 Chinese history as prelude

Some China scholars argue that because the history of Chinese international relations demonstrates significantly more peace than conflict, China will continue in that trajectory and rise peacefully (e.g. Ding 2011; Hsiung 2009; Li and Worm 2011; Shih and Huang 2013). Hsiung (2009) and Ding (2011), among other authors, use the example of Zheng He’s peaceful explorations to argue that China is more peaceful than other great powers in history. Although China was significantly more advanced than any other country at the time of his expeditions (1405–1433), Zheng He traded with countries in Africa and Asia instead of conquering the lands or people. This contrasts with other great powers (e.g. the UK, Germany, and Japan) of aggressive risings.^[16] As a result, Li and Worm (2011) conclude that “Chinese history does not support that kind of prediction that China will use non-peaceful means to rise” (p. 71).

Kang (2007) and Shih and Huang (2013) provide an explanation why the history of Chinese international relations is mainly characterized by peace. Shih and Huang (2013) state: “In the first place, there is no tradition in Chinese political thought, modern as well as pre-modern, that recognizes or even lays the intellectual foundations for recognizing potential duties of the state beyond its borders” (p. 351). Similarly, Kang (2007) proposes that it is “a shared understanding among East Asian states that although China will most likely reemerge as the regional core, its aims are limited” (p. 198).

2.4 Realism

Realist scholars tend to be pessimistic and argue that a rising China poses a threat to international security. Arguably “the oldest and most dominant approach to international relations theory” (Rathbun 2010, 3), realism assumes the primary goal of all states in an anarchic international system is to survive and military power is of fundamental importance to this goal. The balance of power theory proposes that an equilibrium of power enhances national security. It argues that “the equilibrium must aim at preventing any element from gaining ascendancy over the others” (Morgenthau 1985, 189) and that “states fight wars to prevent others from achieving an imbalance of power in their favor” (Waltz 1979, 204). Therefore, based on the balance of power theory, when a state acquires inordinate power war will become more likely. Offensive realism contends that “great powers are primed for offense” (Mearsheimer 2001, 3). For great powers, “[t]he overriding goal of each state is to maximize its share of world power,” and when opportunity arises, “they will use force to alter the balance of power” (Mearsheimer 2001, 2). Although it uses a different argument than the balance of power theory, offensive realism reaches the same conclusion that a rapidly rising great power will lead to an increased probability of war.

The realists’ arguments suggest that China rises aggressively. In a number of publications, Mearsheimer puts forth the influential argument that China’s rise is likely to be unpeaceful (Mearsheimer 2001, 2006, 2009, 2010). He argues that China will strive to establish hegemony in Asia, and in response the US and China’s neighboring countries (e.g. India, Japan, Russia) will seek to thwart China from achieving this goal (Mearsheimer 2006). Furthermore, a rising China is possible to pursue a foreign policy similar to that of “Imperial Germany, Imperial Japan, and Nazi Germany” (Mearsheimer 2009, 252). Therefore, it is likely “[t]he result would be an intense security competition between China and its rivals, with the ever-present danger of great-power war hanging over them” (Mearsheimer 2001, 4).^[17]

In addition, other scholars argue that China is likely to rise non-peacefully. For instance, Panda’s (2013) discussion of the rivalry between China and India throughout Asia over the issues such as the South China Sea and pipelines in Central Asia suggests that China’s rise is aggressive as predicted by realism. Shambaugh (2011) argues that since 2009, China has become increasingly a self-interested power maximizer. This contrasts with China before the global financial crisis and the Beijing Olympic Games, which Shambaugh (2005) characterizes as having a benign regional security posture. Because China has significantly elevated its status after 2008, this contrast supports realists’ argument that China’s rise is likely to be non-peaceful.

2.5 Power transition theory

The power transition theory makes a prediction of the relationship between the existing dominant state, the US, and the rising potential challenger, China. Contrary to the balance of power theory, the power transition theory argues that “wars are most likely when there is an approaching balance of power between the dominant nation and a major challenger” (Organski 1958, 376). This probability of war is higher when the challenger is dissatisfied with the existing international order and the challenger rises rapidly. To support his theory, Organski (1958) states that “the major wars of recent history have all been wars involving the dominant nation and its allies against a challenger who has recently risen in power” (p. 376).^[18]

A direct implication from the power transition theory is that a rapid rising great power China is likely to challenge and wage war against the US, and this implication has inspired many studies to explain and predict the future Chinese-US relations. Some scholars argue that while it is possible China will surpass the US to become the dominant power, it is uncertain whether or not China will challenge the US (e.g. Efird, Kugler, and Genna 2003; Lemke 2003; Lemke and Tammen 2003; Rapkin and Thompson 2003). For example, Efird, Kugler, and Genna (2003) employ a simulation to demonstrate that “levels of joint satisfaction” determine the forecast outcome: “a reconciliation of preferences between the U.S. and China” prevents war whereas dissatisfaction with each other raises the possibility of a major war (p. 308). Organski (1958) uses the example of “the transfer of power from England to the United States” to argue that it is also possible China overtakes the US without a war (p. 372).

Other scholars contend that possibly China may never replace the US to become the dominant power (e.g. Beeson 2009; Chan 2004). Chan (2004) argues that “there is scant evidence pointing to an ongoing or impending power transition between China and the United States and suggest[s] instead that concerns about this possibility are founded less on material conditions and more on ideational construction” (p. 106). In addition, scholars propose that because China has significantly benefited from participating in the existing international economic system, the probability that China challenges the US is reduced (e.g. Beeson 2009; Chan 2004).

3.1 Research design

This study conducts a dyadic analysis of China’s use of militarized conflict to empirically test whether China rose peacefully from 1979 to 2010.^[19] It uses all state dyads that include China as one member, and the dyads are non-directional. To measure our dependent variable, we rely on the widely employed Militarized Interstate Disputes (MIDs) data. The dependent variable is coded 1 if a threat to use force, a display of force, or an actual use of force is observed between China and another state in a year, and 0 otherwise. This study begins its analysis from 1979 because the Chinese economic reform started in December 1978. It is widely observed that China’s economic achievements today are a direct result of the Chinese economic reform, and therefore, an appropriate assessment of China’s rise should examine the time period since the Chinese economic reform. The analysis ends at the year 2010 because MIDs data ends in that year.

The debate on China’s rise points to China’s significantly increased economic power and military power. We measure them using China’s GDP and Military Spending. Furthermore, CINC score is used to measure China’s overall capability. The central task of this study is to empirically assess whether or not a larger value in China’s GDP, military spending, or CINC score led to more China involved conflict. If this relationship holds, it suggests that China rose aggressively. On the other hand, if this relationship does not exist in the data, it demonstrates that China rose peacefully during this time period.

In addition, this study follows the existing conflict literature to include the following independent variables: GDP of Other State, Military Spending of Other State, CINC of Other State, GDP/Military Spending/CINC Ratio, Trade, GDP Per Capita of China and Other State, Alliance, Contiguity, Distance, International Organizations and Regime Dissimilarity. Finally, this study includes three Peace Year variables to control for the temporal dependence presented in the binary dependent variable. To reduce the potential endogeneity problem, all independent variables are lagged by 1 year.

GDP data is measured in constant 2005 USD, which is downloaded from the World Development Indicators.^[20] We use two measures of military spending: one from the National Material Capabilities data (Singer, Bremer, and Stuckey 1972),^[21] and the other from the SIPRI Military Expenditure Database. The former data is available till the year 2007 and the latter is available since 1989. As previously mentioned, CINC scores measure each country’s share of the interstate system’s total military, industrial, and demographic resources (Bremer 1980). This variable is also taken from the National Material Capabilities data, and it covers the years up to 2007. In addition, we create GDP ratio, military spending ratio, and CINC ratio to evaluate the hypothesis that the more unbalanced the powers, the less likely they are to engage in conflict.^[22] Trade is defined as dyadic tradetotal trade, China×dyadic tradetotal trade, other state, a measure introduced by Barbieri (1996) and employed by many trade conflict studies. The trade data is from the Bilateral Trade dataset (Barbieri and Keshk 2012; Baribieri, Keshk, and Pollins 2009). It is expected that China is less likely to have conflict with a state that has more bilateral trade with China. GDP per capita is measured by real GDP per capita taken from the Penn World Table 7.1.^[23] This variable is used to address the argument that high development reduces the probability of conflict. Alliance is a dummy variable that equals 1 if China and the other state within a dyad have a mutual defense treaty, a neutrality pact, or an entente; otherwise, it is coded 0. This data is from the Formal Alliances dataset (Gibler 2009). Allied states are expected to engage in less conflict. Contiguity is also a dummy variable, which is equal to 1 when China and the other state share a land border or are contiguous across up to 400 miles by water, and 0 otherwise. We obtain this data from Stinnett et al. (2002). It is often argued that contiguous states are more likely to engage in conflict. Distance is calculated as the distance between Beijing, China and the other state’s capital. It is expected that when China and the other state have a greater distance they are less likely to have conflict. We download this variable from the Expected Utility Generation and Data Management Program (Bennett and Stam 2000). The variable international organizations is calculated as a simple count of the number of international organizations joined by both China and the other state in a dyad, with its data from Pevehouse, Nordstrom, and Warnke (2004). Regime Dissimilarity is computed using the larger Polity2 score minus the smaller Polity2 score within a dyad. We collect the Polity scores from the Polity IV Project. It is expected that a larger regime dissimilarity increases the probability of conflict. Finally, following the suggestion in Carter and Signorino (2010), this study creates three Peace Year variables to control for the temporal dependence in the binary dependent variable.

3.2 Empirical results

In this section, we present our empirical results to show whether or not China has increased its use of force from 1979 to 2010. Since our three measures of China’s rise – GDP, military spending, and CINC – are highly correlated with each other, three regressions are estimated to avoid the multicollinearity problem: one using GDP, one using military spending, and one using CINC. A logit regression is employed to estimate the binary dependent variable.

The main regression results are presented in Table 2. Since military spending is measured by both COW data and SIPRI data, it occupies two columns. For each regression, we list the time period of the analysis. As clearly demonstrated from all models in Table 2, GDP of China, military spending of China, and CINC score of China all fail to show a statistically significant effect on China’s conflict. These findings shed light on the debate on China’s rise. If China rose aggressively from 1979 to 2010, the coefficients of China’s GDP, military spending, and CINC score are expected to be positive and statistically significant. On the other hand, if China rose peacefully during this time period, their coefficients are expected to be statistically insignificant. The findings in Table 2 lend consistent and strong support to the argument that China rose peacefully from 1979 to 2010.

The results of the other independent variables in Table 2 are to a large extent consistent with the existing theoretical arguments. GDP, military spending, or CINC score of the other state increases the probability of conflict with China. This finding does not contradict the result that GDP, military spending, or CINC score of China has no effect on conflict. The positive effect of GDP, military spending, or CINC score of the other state is largely driven by the cross-state variations rather than the time variations. However, GDP ratio, military spending ratio, or CINC ratio has no effect on the dependent variable. One interesting finding from Table 2 is that except for the regression using GDP, international trade fails to show a pacifying effect on conflict. This finding of trade is a little puzzling since the existing trade conflict research overwhelmingly argues a pacifying effect of trade on conflict.^[24] Nevertheless, this study suggests that trade exerts only a weak effect on militarized conflict between China and its trade partners, a surprising yet interesting finding that has not been predicted by the existing research on China’s rise.^[25] The explanations of the remaining variables are straightforward. Development level in general and alliance do not affect the probability of conflict.^[26] More conflicts occur between China and its neighboring states, and geographical distance decreases the probability of conflict.

Due to the significance of the Kantian peace theory, we add international organizations and regime dissimilarity to the existing regressions and rerun the analysis.^[27] These new regressions are important robustness check, and their sample of analysis is from 1979 to 2006.^[28] The results are presented in Table 3. Once again, the findings from Table 3 unanimously suggest that China’s increased GDP, military spending, or CINC score did not lead to more militarized conflict with other states. All the coefficients of China’s GDP, military spending, and CINC are statistically insignificant. These results echo the conclusion of Table 2 that China rose peacefully from 1979 to 2010.

The findings of the other independent variables from Table 3 largely reinforce the findings in Table 2.^[29] One difference comes from GDP ratio: it shows that more balanced powers tend to engage in more conflict.^[30] Regime dissimilarity supplies weak evidence to buttress the argument that a larger regime difference between China and another state is likely to increase the probability of militarized conflict. In addition, participating in international organizations has no effect on China’s militarized conflict. In short, Table 3 provides little support of the Kantian peace argument during China’s rise from 1979 to 2006. Put differently, the findings in Table 3 revise the conventional wisdom that participating in trade and international organizations reduces China’s involvement in militarized conflict. Nonetheless, Table 3 provides additional and consistent evidence to suggest China rose peacefully during this time period.

Finally, to address the influential argument that China has turned into an aggressive rise since 2008, we conduct an analysis to compare China’s pre- and post-2008 conflict behavior. In the existing research on China’s rise, many studies argue that the year 2008 is a turning point in the trajectory of China’s rise (e.g. Buzan and Cox 2013; Hughes 2011; Shambaugh 2011; Yahuda 2013; Zhao 2013). Both the success of the Beijing Olympic Games in 2008 and China’s continued high economic growth in spite of the global financial crisis of 2007–2008 have significantly elevated China’s international status. Therefore, many scholars argue that China has since then turned into an aggressive rise.

The implication of this theoretical argument for our statistical analysis is that we need to treat the year 2008 as a structural break point and to test if China rises aggressively since 2008. We follow the common approach to use a dummy variable and an interaction term to empirically test this hypothesis. This dummy variable is named Post-2008 Dummy, taking the value of 1 for the years after 2008, and 0 otherwise. Two interaction terms are created, which are GDP of China×Post-2008 Dummy and Military Spending of China×Post-2008 Dummy. If China has turned into an aggressive rise since 2008 as predicted in the existing research, the dummy variable and/or the interaction term will show statistically significant coefficients. In addition to these new created variables, our regressions include the same independent variables as in Table 2. Because the COW military spending data and the CINC data end in the year 2007, they are not available for this empirical test.

The regression results are shown in Table 4. Contrary to the existing argument that China has turned into an aggressive rise since 2008, our regression analysis shows that China rose peacefully during this post-2008 era. Across both models, the coefficients for GDP of China and military spending of China remain statistically insignificant. Furthermore, our post-2008 dummy and both interaction terms are statistically insignificant. All these findings demonstrate that China rose peacefully during both the pre-2008 and the post-2008 time periods. Scholars have pointed to China’s more assertive stance on sporadic issues such as the South China Sea and the Diaoyu Islands to conclude that China has become more aggressive in recent years. Our large-N empirical analysis, however, demonstrates that their conclusion is biased by observational selection. When we take into account all conflicts, an increase in China’s GDP or military spending did not cause more conflict between China and other states during the post-2008 era.

Although our post-2008 sample (i.e. 2009–2010) is relatively short due to data availability, we argue this sample nevertheless constitutes a good test of the hypothesis. Since the effects of the Beijing Olympic Games and the global financial crisis tend to decline over time, we will be more likely to observe China’s aggressiveness during the immediate years after these events. If, on the other hand, our analysis of 2009–2010 suggests no evidence of China’s aggressive rise, it is very likely that we will reach the same conclusion based on a longer post-2008 time period. Overall, our empirical analyses in Tables 2–4 suggest the same conclusion that China rose peacefully from 1979 to 2010.