The North Korean Nuclear Dilemma: Does China Have Leverage?

Abstract

North Korea has repeatedly defied the international community with regards to its nuclear weapons programme. Many look to China for leverage to change North Korea’s behaviour. This study reviews the development of the China–North Korean relationship and conducts a statistical analysis on the impact of China’s influence on North Korea. Our analysis finds China’s leverage on North Korea to be nuanced. We maintain that North Korea has been wary of China’s influence. Complete isolation or pressure from China under certain conditions will render North Korea more resistant to China’s influence. We also suggest that the key to the North Korean issue is still in the hands of the United States and the entire international community through the use of an engagement strategy.

Keywords

Nuclear weapons sanctions North Korea China United States

Introduction

North Korea has repeatedly defied the international community with regards to its nuclear weapons programme despite diplomatic pressure and international economic sanctions. Given that North Korea is believed to have an alliance-like relationship with China and that it relies on the latter for food and energy supplies, many in the international community believe that China has leverage on North Korea to change its behaviour and that North Korea’s defiance to the international pressure is attributable to China’s protection. Therefore, Beijing has been called on to utilise this leverage and persuade North Korea to stop its nuclear weapons programme.

In response to the request of the international community, China has reduced its trade with North Korea. According to the International Trade Centre (2017), a joint agency of the World Trade Organization and the United Nations (UN), when North Korea conducted its first nuclear test in 2006, China cut back its cereal exports to North Korea from US$49 million to US$16 million, a 66 per cent decline (Figure 1). After North Korea’s second nuclear test in 2009, China began pressuring North Korea by cutting fuel exports from US$586 million to US$327 million, a 44 per cent drop. North Korea’s third nuclear test in 2013 triggered an 81 per cent drop in mineral fuels imports and an 89 per cent drop in cereal imports from China. In 2017, China banned exports of petroleum to and imports of textile and coal from North Korea and ordered all North Korean businesses in China to be closed (Clover, Harris, & Lockett, 2017; Doubek, 2017).

Despite rising pressure from China during this period, little change was seen in North Korea’s behaviour. Table 1 shows that North Korea’s nuclear and missile tests have become increasingly frequent since 2006. China has leverage on North Korea only if the actual withholding of something yields a change or particular response from the latter. Yet even with the drop in commodity exports to North Korea, North Korea has tested missiles, such as the intercontinental ballistic missile, Hwasong-14, in July 2017. Other nuclear-capable missiles were launched earlier in 2017, including the Pukguksong-2 missile. These missile tests were followed by China announcing that it would enact UN Security Council sanctions by suspending imports of coal from North Korea. Even after this announcement and use of ‘leverage’, North Korea proceeded to test more missiles and is suspected of killing Kim Jong-Nam, Kim Jong-Un’s more pro-Chinese half-brother. Scenarios such as these lead us to question why China’s influence fails to change North Korea’s behaviour.

Figure 1.

Source: Data collected from International Trade Centre’s Trade Map (2017) database.

Table 1.

North Korea Nuclear and Missile Tests

Year	1998	1999	2000	2001	2002	2003	2004	2005	2006	2007
No. of tests	1	0	0	0	0	3	1	2	10	7
Year	2008	2009	2010	2011	2012	2013	2014	2015	2016	2017
No. of tests	6	16	0	0	6	9	18	18	23	24

Source: Data collected from the Centre for Strategic and International Studies (2017).

In this study, we seek to systematically test whether particular actions by China and other stakeholders of the North Korean nuclear issue over the past decade have influenced North Korean foreign policy behaviour. We argue that believing North Korea would respond positively to China’s influence because of its reliance on China is based on unfounded assumptions. The contribution of this article is threefold. First, it contributes to the foreign policy literature by emphasising how coercion through the use of sanctions is not always successful when dealing with isolated states like North Korea. Second, our empirical evidence suggests that relying on China’s economic leverage on North Korea is a misguided approach to the situation. Our findings indicate a nuanced pattern of North Korean response to changes in its imports and exports. Although a cut in North Korea’s exports to China may contain its aggression, a significant cut in its imports from China is futile. Moreover, an increase in North Korea’s import reliance on China could motivate it to escalate the tension by becoming less cooperative and/or adopting a more aggressive foreign policy action with its nuclear weapons programme. Third, our results make a significant contribution to the literature by examining other forms of foreign policy outside the implementation of sanctions and highlight how diplomatic meetings with North Korea has also yet to impact the country’s nuclear programme. This article proceeds with an analysis of Democratic People’s Republic of Korea (DPRK) foreign policy strategy, its relationship with China over time, followed by an explanation of our methods employed and the results of the models.

Leverage, Failed Coercion and North Korea’s Foreign Policy Strategy

The concept of leverage in this study is derived from Levitsky and Way’s studies on how the international community impacts regime change through leverage and linkage (2006, 2010). Specifically, we define leverage as a ‘government’s vulnerability to external pressure’ (Levitsky & Way, 2006, p. 382). Leverage is ‘rooted primarily in the size and strength of countries’ states and economies’ (Levitsky & Way, 2006, p. 382); therefore, we focus on North Korea’s economic vulnerability and pressure from China. While weak states with economies that rely on foreign aid are more vulnerable to outside pressure, this can be mitigated by a country having a substantial military. Although North Korea relies on China for a significant portion of its trade and economic aid, its military capabilities by way of its nuclear weapons programme causes the country to be able to withstand any punitive action levied against it (as outside countries are less likely to use such punitive actions to begin with), thus creating a low leverage situation.

In order for one country to have leverage on another, pressure must be applied. This pressure comes in various forms, including ‘positive conditionality, punitive sanctions, diplomatic persuasion, and military force’ (Levitsky & Way, 2006, p. 382). We refer to economic pressure by way of sanctions as being coercive in nature. The effectiveness of coercive foreign policy is limited (Elliott, 1998; Pape, 1997); in fact, we posit that such coercive measures will cause North Korea to be more resistant and likely to cease cooperating with the international community or even conduct defiant actions such as nuclear tests and withdrawal from negotiations. One reason coercive policies are ineffective is that they require collective action from multiple states, such as economic sanctions not always being faithfully employed by all states after being announced. Many have claimed that the reason Pyongyang has not changed its foreign policy behaviour is because China has simply not maximised its position or leverage. For instance, despite UN Security Council sanctions and China’s cuts on food and energy supplies to North Korea, the total China–DPRK trade volume rose between 2006 and 2012. Even when the trade volume declined and China announced that it would suspend coal imports from North Korea, this did not equate to the actual total suspension of coal imports as resources were still traded on the basis that the North Korean’s ‘livelihood’ would be at stake if such imports were to cease. In addition to general coercion, Bader (2015) finds that only export dependence on China helps the survival of an autocratic regime like North Korea. Thus, pressuring China to decrease total trade may not be the best course of action when seeking a change in behaviour from North Korea on its nuclear programme.

The target state of coercion sometimes can circumvent international isolation (Drezner, 2003). For instance, the North Korean regime has used its diplomats to create front companies in order to evade sanctions (Blancke, 2016; Einbinder, 2017). However, these factors only mitigate the impact of coercion. Therefore, the logic of coercion still holds, that is, increasing pressure will eventually change the payoff structure of North Korea’s policy options and make cooperation with the international community relatively more profitable or less costly than defiance.

Even if there is still room for China to further press North Korea and implement more sanctions, it is not guaranteed that North Korea would become more compliant with the international community’s demands as coercive policy could be ineffective due to political reasons. Foreign policymakers are influenced by their strategic thinking and perception about international and domestic environments (Jervis, 1976; Rose, 1998). In the case of North Korea, where leaders enjoy personalised power and ultimate control over the government and society (Oh & Kassig, 2000, p. 96; Park, 2002), its leaders’ strategic concern plays a much more important role in foreign policymaking than social or bureaucratic factors (Hermann & Hermann, 1989; Holsti, 1976).

A leader seeks to maximise the chance of regime survival in an anarchic international system as well as his or her own survival in domestic politics. At the international level, a small power such as North Korea is particularly concerned about being abused by great powers (Waltz, 1979). A small state’s interactions with the major powers surrounding it is conditioned by the small state’s perception of a threat (Jervis, 1976; Kang 2003; Walt, 1987). The perception of a state on another can be influenced by the power structures in the international and domestic political systems (Moravcsik, 1997). A state perceives its allies as less threatening than its adversaries and is more receptive to the former’s requests than that of the latter even if doing so would incur short-term costs because it expects to gain more in the long run. Conversely, the state would be reluctant to cooperate with its adversary because it expects greater long-term costs in the relationship. North Korea views the United States and its allies in Northeast Asia as major threats because they have the capability and expressed intention to threaten its regime survival, while Russia and China would be considered less threatening because of ideological similarities.

When dealing with an external threat, a state can bandwagon with the threatening state, ‘pass the buck’ to other states or balance against the threat. Although small states are most likely to bandwagon (Powell, 1999; Schroeder, 1994; Schweller, 1994; Walt, 1985), this is not a good option for North Korea because the United States is viewed by North Korea as an imperialist power that is still at war with North Korea and has illegally occupied the southern part of the country for more than half a century. In other words, the North Korean regime’s legitimacy is built on defending the country from imperial invasion and reunification with the south. Therefore, bandwagoning with the United States would jeopardise regime legitimacy. Furthermore, bandwagoning with the United States would cause others to view North Korea as having easily conceded to US demands (Drezner, 1998, 2003; Drury, 2000), which would put North Korea at a disadvantaged position in future negotiations with the United States.

Buck-passing is not feasible for North Korea either because this strategy requires multiple status quo powers in the region that are able and willing to balance against the United States on behalf of North Korea. During the Cold War, North Korea was used by major powers as a buffer zone between the communist and capitalist worlds. Today, no state in Northeast Asia seeks open, direct conflict with the United States. Thus, North Korea has no one to pass the buck.

North Korea’s only feasible strategy is balancing. Balancing can be pursued internally by improving its capabilities or externally by forming alliances. Internally, North Korea began its nuclear programme in the 1950s with assistance from the Soviet Union to balance against the US nuclear threat in South Korea (Kristensen & Norris, 2017). Nonetheless, its nuclear weaponisation process during the Cold War was slow due to the constraint of the Soviet Union’s involvement. The collapse of the Soviet Union left North Korea unprotected and motivated Kim Jong-Il to adopt the ‘military first’ policy and accelerate the nuclear programme, which resulted in a crisis between North Korea and the United States. Even though the two countries reached the Framework Agreement in 1994, the deal collapsed when the United States failed to provide two light-water nuclear reactors to North Korea as the agreement required, and North Korea resumed its nuclear programme. In the 2000s, the US diplomatic strategy leading up to and during the Iraq War along with the deaths of Saddam Hussein in Iraq and Muammar Gaddafi in Libya convinced North Korean leaders that only nuclear weapons could provide their regime with adequate security (KCNA, 2017a).

Before North Korea acquires the sufficient capabilities needed to deter foreign invasion, it can form alliances to balance against threats. For instance, after the Korean War, North Korea quickly signed mutual military assistance treaties with the Soviet Union and China. When the Sino–Soviet relationship developed strains in the 1960s, Kim Il-Sung naturally chose the more powerful Soviet Union for balancing purposes. In the meantime, China began to normalise its relationship with the United States, and later with Japan and South Korea. These shifts in Chinese foreign policy convinced Kim Il-Sung that China was unreliable. However, the collapse of the Soviet Union made China the only option which North Korea could use to balance against the United States and help sustain its domestic economy. China became so important to North Korea after the Cold War that the latter overlooked China’s betrayal of recognising South Korea. Studies of North Korean attitudes towards China show that the North Korean media has maintained a positive view of China even when China’s foreign policy hurt North Korean interests (Zhan, 2016; Zhang & Zinoviev, 2017).

Figure 2 illustrates that Chinese and North Korean leaders have exchanged frequent visits since the late 1990s. These visits helped North Korea quickly rebuild its relationship with China and facilitated a rapid recovery in trade (Figure 3). Interactions between the top leaders from each country declined in 2002 when China went through a leadership transition and the United States labelled North Korea an ‘axis of evil’. The frequency of visits rebounded in 2004 and 2005 and then dropped again due to the worsening international environment surrounding North Korea’s nuclear test in 2006. However, the number of diplomatic meetings and communications rose sharply in 2010 when Kim Jong-Il frequently visited China to seek economic assistance. The unexpected death of Kim Jong-Il in December 2011 and the subsequent leadership transition to Kim Jong-Un resulted in a significant drop in leadership interactions and diplomacy between North Korea and China.

Figure 2.

Source: Data collected from NK News Leadership Tracker database (2017).

Figure 3.

Source: Data collected from the Trade Map by International Trade Centre (2017).

In addition to external threats, totalitarian leaders and dictators also face domestic challenges. They must ensure their own political survival within the regime before concerning themselves with regime survival at the international level. At the domestic level, a state tends to fear its neighbours the most because their close geographic proximity makes foreign invasion or intervention in domestic affairs more likely (Gause, 2007). Moreover, leaders’ perception of threats is also influenced by their knowledge of international events, especially those that affected their country and lives when they were young (Jervis, 1976). Invasions by the Mongolians, Japanese, Chinese and Russians likely contributed to North Korea’s distrust of outside influence. For instance, North Korea’s official Juche ideology professes independence of foreign influence in policymaking:

Juche means adopting the attitude of a master towards the revolution and construction of one’s country. It means maintaining an independent and creative standpoint in finding solutions to the problems which arise in the revolution and construction. It implies solving those problems mainly by one’s own efforts and in conformity with the actual conditions of one’s own POLITICS country. The realization of independence in politics, self-sufficiency in the economy and self-reliance in national defence is a principle the Government maintains consistently.

The Korean people value the independence of the country and nation and, under the pressure of imperialists and dominationists, have thoroughly implemented the principle of independence, self-reliance and self-defence, defending the country’s sovereignty and dignity firmly. (DPRK, 2011).

Due to China’s proximity, capability and long history of interference in Korean politics, Korean leaders have reason to be wary of China’s influence in the country. For example, among the founders of the Korean Workers’ Party (KWP) were a group of Korean generals who were part of the Chinese communist military during the Second World War and the Chinese civil war. They returned to Korea after the wars to help rebuild the country. Because of their experience in China, they had a very good relationship with the Chinese leaders and were referred to as the ‘Yan’an faction’ in North Korea (Lankov, 2013). Since Kim Il-Sung was backed by the Soviet Union, he was concerned about the growing influence of the Yan’an faction in the KWP and the possibility of China using the Yan’an faction to interfere in Korean politics and undermining his authority in the party. Consequently, when Kim Il-Sung planned to invade South Korea in 1950, the Yan’an faction was excluded from the decision-making process. Additionally, while Kim Il-Sung travelled to Moscow in April 1950 to discuss a war plan, China was not involved at all until September 1950 when Kim Il-Sung needed Chinese assistance (Shen, 2003; Suh, 2004).

China’s support during the Korean War helped North Korea survive; it also strengthened the influence of China and the Yan’an faction in Korean politics. Under these circumstances, Kim Il-Sung announced the Juche ideology in 1955. One year later, Kim requested all foreign militaries to withdraw from Korea (Chung & Choi, 2013). Since the United States was not expected to grant this request and the Soviet military units were stationed in China, China was the only other foreign country with a military presence in North Korea. In other words, Kim Il-Sung’s request was implicitly for China to leave Korea. Kim’s dedication to cut off China’s influence motivated the Yan’an faction in a vain attempt to remove Kim from leadership in August 1956. Believing that China was behind the coup, Kim completely purged the Yan’an faction from Korean politics (Chung & Choi, 2013).¹

The pro-Soviet faction also participated in the coup but was not purged as harshly as the Yan’an faction.

The pro-China political force in North Korea was then eliminated until the 1990s when North Korea needed China to take over the Soviet Union’s role as a balancer.

The China–DPRK Treaty of Friendship, Cooperation and Mutual Assistance of 1961 has been viewed by many as proof of a China–DPRK alliance. However, the treaty focusses more on both countries refraining from interfering with the other’s domestic affairs than on mutual aid. In fact, the Chinese government once claimed that the treaty did not commit Chinese military to defend North Korea (Kim, 2015, p. 188). In comparison, the alliance among NATO members and the treaties between the United States and Japan and the United States and South Korea specifically require ‘the use of force’ and that all parties will ‘act to meet the common danger’.

During the remainder of the Cold War, North Korea relied on the Soviet Union for food and energy. The collapse of the Soviet Union resulted in a sharp drop in the supplies of these resources in North Korea. The supply vacuum created in the early 1990s was not immediately filled in by China because of the establishment of a diplomatic relationship between China and South Korea in 1992. The severe drought in North Korea in the mid-1990s exacerbated the economic and living conditions in North Korea. Consequently, North Korea began asking China for assistance and became reliant on China for food and energy.

Economic dependence can become a source of power (Hirschman, 1980; Kirshner & Rawi, 1999; Knorr, 1975; Levy, 2003). According to trade data from the International Trade Centre (2017), China–DPRK trade, on average, accounts for more than 80 per cent of North Korea’s total trade. The trade reliance on China puts North Korea in a disadvantaged position in the bilateral relationship. In addition to trade dependence, China’s successful economic reform has attracted a great deal of interest among North Korean elites. Consequently, China’s economic strength revived the pro-China faction in North Korea and drew the attention of the North Korean leadership.

One example of the pro-China faction capturing the attention of the DPRK leadership is Jang Song-Thaek’s downfall in 2013. Jang was Kim Jong-Il’s brother-in-law and designated mentor to Kim Jong-Un. Because of his connection to the Kim family, Jang held key positions in the KWP since the 1970s and was later viewed as the second most powerful person in North Korea. Jang was also an advocate for Chinese-style economic reform. Having been in charge of economic cooperation with China, Jang not only developed good relationships with Chinese leaders, but also opened dozens of Chinese-style special economic zones in North Korea. He was considered to be the person that ‘China counted on to move the economy in North Korea’ (Perlez & Choe, 2013).

Jang and his followers posed a significant challenge to Kim Jong-Un’s authority. Consequently, one year after Kim Jong-Un inherited command of the regime, he arrested and executed Jang and purged his followers. Jang’s indictment accused him of organising a faction against Kim Jong-Un and selling resources cheaply to China. To mitigate China’s influence, many North Korean businessmen were recalled from China and high-level leadership interactions stopped (Figure 2) (Yonhap News, 2013).

To further secure his authority, Kim Jong-Un assassinated Kim Jong-Nam, Kim Jong-Il’s eldest son who had been under China’s protection. Even though Kim Jong-Nam expressed his disinterest in politics in North Korea (Chow & Park, 2017), his identity as Kim Jong-Il’s eldest son could have given him legitimacy should he have sought control of the regime in North Korea, and his close relationship with China would have made him a powerful competitor with domestic and foreign support. Kim Jong-Un’s concern resulted in another attempted assassination of Kim Jong-Nam’s son in China, months after Kim Jong-Nam’s death (Kong, 2017a).

The reason why Kim Jong-Un was much more concerned about China’s influence than Kim Jong-Il had been during his time in power is that they enjoyed different levels of legitimacy in North Korea. Kim Jong-Il started governing North Korea with Kim Il-Sung in the 1970s and gained legitimacy over many years. In comparison, Kim Jong-Un did not have any significant legitimacy when he was appointed to be Kim Jong-Il’s successor. The fact that Kim Jong-Un was young, had little governing experience and spent most of his life outside of North Korea resulted in his low recognition in Korean society.²

This is according to the author’s interview with a North Korean tour guide during his visit to North Korea in 2016.

This made him vulnerable to the senior leaders in the KWP and military. Therefore, Kim Jong-Un was more worried about his own political survival in North Korea than the regime’s survival at the international level. China became a major threat to Kim Jong-Un because it controls North Korea’s trade, has supporters among North Korean elites and could potentially replace Kim Jong-Un with someone that has higher levels of legitimacy. Therefore, in addition to killing Jang Song-Thaek and Kim Jong-Nam, Kim Jong-Un also quickly replaced most of the old generation political and military elites with his own followers and shifted power from the military back to the KWP.

In addition to rooting out competitors in and outside the country, Kim Jong-Un seeks to gain legitimacy from economic development. Estimates suggest that the North Korean economy under Kim Jong-Un’s rule has been growing and reached a 3.9 per cent growth rate in 2016 despite international sanctions (Kim & Chung, 2017). More recently, during the 7th KWP Congress, Kim Jong-Un stressed improvement in self-reliance on food and energy and diversifying its trade partners (Kim, 2016). Zhang and Zinoviev examined the North Korean media attitude towards China and found that while China plays an increasingly important role in the North Korean economy, the Kim Jong-Un administration has carefully downplayed China’s importance to the North Korean economy but skilfully maintained a high level of nominal praise of China for its help during the Korean War (Zhang & Zinoviev, 2018).

China plays two roles in North Korea’s foreign policy strategy. At the international level, China is the key to North Korea’s balancing strategy against the United States, at least before North Korea develops serious deterrent capabilities. Thus, North Korea needs to maintain a good relationship with China. At the domestic level, China’s economic and political influence is a threat to North Korea. To North Korean leaders, their own political survival in the regime is more important than regime survival. North Korea’s need for China depends on how China helps it balance against the United States. If China fails to serve in the balancing role, then North Korea does not need China. For instance, a 2017 Korean Central News Agency (KCNA) commentary criticised China for ‘dancing to the tunes of the US’ to ‘check its nuclear program’ (Kong, 2017b) and warned China to ‘ponder over grave consequences to be entailed by its reckless act of chopping down the pillar of the DPRK–China relations’ (KCNA, 2017b). China’s pressure on North Korea to change its behaviour would prove its ‘unreliability’ to Kim Jong-Un and motivate him to focus on an internal balancing strategy and accelerate the country’s nuclear programme. Although China’s economic assistance contributes to North Korean economic recovery, as the history of the ‘arduous march’ in the mid-1990s shows, a strong economy is helpful but not necessary for the North Korean regime to survive. Rather, China’s economic assistance is considered by the North Korean regime as a service fee that China owes North Korea for ‘protecting [the] peace and security of China’ from the Western influence and a potential US invasion (KCNA, 2017b).

Based on our theoretical and historical analyses, we argue that North Korea has maintained a nominally good relationship with China but is wary of China’s influence. Thus, we contend that China does not have economic leverage on North Korea. Although North Korea may be viewed as vulnerable as it relies on China for various resources and aid, there is no solid linkage between the two countries. Eliciting a change in a foreign government requires a linkage or ‘density of ties and cross-border flows’ (Levitsky & Way, 2006, p. 383). In fact, we maintain that because of North Korea’s past interactions with China, the regime lacks trust in China and will respond accordingly. Even if China applies pressure in the form of coercive foreign policies such as economic sanctions, North Korea will not respond in a more cooperative manner. We hypothesise that the more China attempts to maintain leverage on North Korea, the more likely North Korea will take actions that resist China’s influence. For instance, we maintain that the more economically vulnerable North Korea is in its reliance on China and the more pressure China puts on North Korea by reducing its trade with the country, the more likely North Korea is to stop cooperating or even shift from maintaining the status quo to actively resist China’s influence by taking actions that decrease the region’s stability. If we find support for this hypothesis, it would prove imperative that the United States and the rest of the international community work to normalise their relationships with North Korea without relying on sanctions and/or pressure from China.

Methods

To test our hypothesis, we collected data on North Korea’s foreign policy behaviour and factors that influence it. The dependent variable is North Korea’s action in response to international demands in a month. The variable has three categories: cooperation (0), status quo or no observed action (1) and resistance (2). ‘Cooperation’ includes any event that could improve regional stability or de-escalate the crisis, such as reaching diplomatic agreements. Since China seeks regional stability and de-escalation of the crisis, ‘resistance’ refers to any North Korea-initiated event that worsens regional instability or escalates the crisis, such as nuclear or missile testing, withdrawing from international negotiations or treaties, making threats against other countries and so on. If no action is detected, then the observation is categorised as ‘status quo’.

The event information is collected from the Chronology of United States-North Korea Nuclear and Missile Diplomacy which is compiled by the Arms Control Association. Our data set covers the January 2005 to December 2017 period, and observations are structured on a monthly basis to capture short-term changes. We chose monthly data because long time intervals such as a yearly change cannot capture North Korea’s frequent change in foreign policy. Likewise, too short of a time interval such as a daily change creates problems when gathering data for other variables and inflates the data set with too many invariable observations. There are a total of 156 observations (or months), 17 of which are cases of cooperation and 45 are cases of resistance.

The independent variable operationalises China’s economic leverage on North Korea.³

We also considered China’s security leverage on North Korea. Security vulnerability is a state’s relative capability. We used the Composite Index of National Capability from the Correlates of War data set to calculate North Korea’s relative national capability compared to China’s capability. The results showed that North Korea’s capability from 2005 to 2012 had declined from 7.4 per cent of China’s capability to 6 per cent. Despite China’s capability advantage, there has been little security pressure. For instance, China has maintained a mutual security treaty with North Korea and has not sought to take advantage of North Korea’s vulnerability due to concerns over border security and North Korea’s nuclear retaliation. In other words, China has little security leverage over North Korea. Additionally, there are no data available for monthly changes in national capability as they vary little on a monthly basis. Therefore, the security factor is not included in the following analysis.

Since the concept of leverage indicates that a government is exposed to outside pressure, we consider North Korea’s belief that its economy and security are vulnerable in our operationalisation of the variable. Regarding economic vulnerability, North Korea’s economy relies on foreign resources, mainly from China. Its economic vulnerability is measured with the proportion of its imports and exports that is contributed by China. North Korea imports essential resources to maintain its basic living standards. Its exports to China help it earn foreign currencies to purchase foreign products or resources. North Korea is more vulnerable the more it relies on China for exports or imports. Previous research such as Bader’s analysis of China’s impact on the longevity of other autocratic regimes found that export dependence on China helps authoritarian regimes survive (Bader, 2015). While Bader’s large-N study examines China’s trade relationship with 155 countries and its impact on their regime survival, our model tests whether and to what extent North Korea responds to its import and export dependence on China.

Pressure is another key element of leverage. China is capable of pressuring North Korea by reducing its trade with North Korea due to the latter’s asymmetrical dependence on the former. This pressure is measured by utilising two variables, the growth rates of the value of China’s exports to and imports from North Korea. A decline in the trade volume between the two countries would imply that China is pressuring North Korea. Nonetheless, it is reasonable to suspect international trade to follow seasonal patterns, and the data on China-North Korea trade do show seasonality. However, it does not necessarily mean that the relationship between economic leverage and foreign policy reaction is spurious. To test the impact of seasonality, we include a factor variable of a month. The logic is that if seasonality plays a role, we should expect North Korea’s foreign policy behaviour to reflect seasonal patterns too. If North Korea’s behaviour is not affected by seasonality, then it is reasonable to claim that the impact of trade on foreign policy derives from economic pressure.

Leverage requires both vulnerability and pressure to be effective; therefore, it is measured with an interaction term of vulnerability and pressure. We collected monthly trade data for these variables. The data are from the Trade Map Database of the International Trade Centre and are available from January 2006 to December 2017.

Ideally, our analysis would include North Korea’s trade with all of its trade partners, especially the six-party talks members. However, due to the lack of such data between North Korea and all the six-party talks countries, we focus only on China–North Korea trade. Excluding the economic influence of other countries in the model does not fundamentally distort our analysis because they account for only a small fraction of North Korea’s trade. For instance, South Korea offered to help North Korea via its Sunshine Policy in 1998, under which restrictions were relaxed on the South Korean private sector to invest in North Korea. But, according to the Korean International Trade Association, North Korea has not imported anything from South Korea since 1998 (KITA, 2017). North Korea’s exports to South Korea peaked in 2006 at 0.026 per cent of North Korea’s total exports.⁴

Percentages presented are calculated by the authors based on the data collected from the KITA website.

Even this small amount of exports to South Korea came to a halt after North Korea’s nuclear tests in 2006.

Regarding an alternative explanation, some in the international community have claimed that China has not faithfully enforced international sanctions. Therefore, it is possible that North Korean foreign policy is more responsive to the signals of international sanctions than the application of sanctions. For instance, North Korean leaders almost always verbally ‘fight back’ or show resistance in some way after a state or an international organisation announces sanctions on North Korea. To test this alternative explanation, we extended our analysis of Chinese influence to international influence and created a dummy variable of economic sanctions to include in the model. When a unilateral or multilateral economic sanction against North Korea, including those issued by the United States, is announced in a month, that month is coded as 1, otherwise 0. Sanctions information and dates are collected from the Chronology of US–North Korea Nuclear and Missile Diplomacy.

In addition to economic leverage, external political pressure can shape North Korea’s foreign policy behaviour. To parcel out the effect of economic leverage on North Korean foreign policy behaviour versus diplomatic pressures via meetings, we test the effect of China’s economic pressure on North Korea as well as the latter’s vulnerability while controlling for the political pressure North Korea faces in bilateral or multilateral meetings with countries that participate in the six-party talks. We distinguish between bilateral and multilateral meetings in the model because to whom North Korea talks and how they talk matters in terms of the level of political pressure North Korea may encounter. On the one hand, research shows that multilateral negotiations may be less fruitful than bilateral negotiations because the former requires agreement from more players (Oye, 1985). On the other hand, multilateral negotiations could offer more reliable and sustainable agreements and have more resources to offer or take away from North Korea, which translates into political leverage.

We utilise dummy variables for each bilateral meeting and multilateral meeting between North Korea and another country that participated in the six-party talks. When a talk takes place in a month, the observation for that month is coded as 1, otherwise 0. The bilateral diplomacy variables are labelled ‘Meet with China’, ‘Meet with US’, ‘Meet with Russia’, and ‘Meet with South Korea’ while the multilateral diplomacy variable is labelled ‘Multilateral Talks’. The information about these meetings are collected from the NK Leader Tracker Database prepared by NK News (2017). The data include 31 bilateral meetings between North Korean and Chinese leaders, 12 meetings with US officials, 7 meetings with their South Korean counterparts and 5 meetings with Russian officials. No bilateral interaction occurred between North Korean and Japanese leaders.

The period covered by this study includes two North Korean regimes. It is possible that some fundamental differences between Kim Jong-Il’s and Kim Jong-Un’s administrations could be the sources of North Korean foreign policy behaviour changes. Additionally, as Kim Jong-Un started as a young, new leader surrounded by a group of senior political elites, it took time for him to consolidate power, build up legitimacy and gain confidence in ruling. Thus, we divide the time frame of the data set into three periods. The first period is the Kim Jong-Il era (coded as 0), from the beginning of the available data to December 2011 when Kim Jong-Il died; the second period is from Kim Jong-Il’s death to Jang Sung-Taek’s execution in 2013 (coded 1). The execution of Jang Sung-Taek symbolises the consolidation of Kim Jong-Un’s power and the end of the pro-China faction in North Korea. The third period is the post-Jang era (coded 2). If different regimes or stages of the Kim Jong-Un regime affect North Korea’s foreign policy differently, then this variable would capture the impact.

Table 2

. Correlation Among Variables

Variables*	1	2	3	4	5	6	7	8	9	10	11	12	13	14	15	16
1	1.000
2	0.002	1.000
3	0.286	0.102	1.000
4	0.027	0.973	0.126	1.000
5	-0.031	0.607	0.025	0.599	1.000
6	0.184	0.004	0.754	0.021	0.042	1.000
7	-0.044	0.617	0.046	0.638	0.964	0.069	1.000
8	-0.094	-0.087	-0.127	-0.105	-0.039	-0.115	-0.066	1.000
9	-0.166	0.006	-0.138	-0.022	0.081	-0.022	0.058	0.203	1.000
10	-0.048	0.099	0.073	0.092	0.237	0.002	0.220	0.189	0.094	1.000
11	-0.114	-0.096	-0.129	-0.084	0.005	-0.147	0.002	0.066	0.213	-0.043	1.000
12	0.138	0.058	0.059	0.076	0.066	0.049	0.091	-0.066	-0.077	-0.053	-0.059	1.000
13	-0.117	-0.045	-0.165	-0.041	0.061	-0.140	0.039	0.233	0.117	-0.035	0.174	-0.047	1.000
14	0.204	-0.054	0.660	-0.042	-0.084	0.749	-0.069	-0.253	-0.191	-0.156	-0.171	0.002	-0.139	1.000
15	0.244	-0.013	0.233	-0.016	0.099	0.237	0.117	-0.094	-0.117	-0.115	-0.053	0.189	-0.043	0.204	1.000
16	-0.042	0.010	0.116	-0.009	-0.071	0.061	-0.071	0.001	0.002	0.030	0.046	-0.084	0.037	-0.012	-0.151	1.000

Source: The authors.

Note: *Due to space limitations, variable names are replaced with numbers.

Variable Number	Variable Names	Variable Number	Variable Names
1	Behaviour	9	Meet United States
2	Growth of imports from China	10	Meet Russia
3	% of North Korean imports from China	11	Meet South Korea
4	Growth of imports × % of imports from China	12	Sanction
5	Growth of exports to China	13	Six-party talks takes place
6	% of North Korean exports to China	14	Stage of Kim Jong-Un
7	Growth of exports × % of exports to China	15	Previous behaviour
8	Meet China	16	Month

Lastly, we take into account the effect of time on North Korea’s behaviour. Foreign policies do not always change drastically overnight. A state’s foreign policy at one point in time is also not totally independent of the previous policy. North Korean foreign policy behaviour during a month could be significantly influenced by its previous policy behaviour. Therefore, in addition to examining the impact of the aforementioned variables on North Korean behaviour in the same month, behaviour in the previous month is also taken into consideration. In addition to policy dependency, we create a second model, Model 2, that examines whether North Korean foreign policy behaviour in a month is driven by not only the previous policy but also the independent variables and control variable measurements in the previous month. This model can tell us whether North Korea responds quickly to recent events or takes time to plan before acting.

We lag the dependent variable by one month which is a widely adopted approach in time-series studies to examine whether time plays any role in the model. We do not choose to lag the variables by more than one month because the effect of time declines with each additional month. If we adopt a two-month lag, for example, and examine the extent to which the policy behaviour in the month t is affected by the behaviour in the previous two months, t-1 and t-2, then it would be methodologically difficult to distinguish between the influence of the policy in t-1 and that in t-2 on t (t-1 is also affected by t-2). Our choice of a one-month lag does not mean that North Korean foreign policy is only affected by the policies of one month ago. Rather, we use a one-month lag to test whether time plays a role in North Korean foreign policy changes.

Table 2 reports the correlation among all the variables. The correlations do not display major issues of collinearity. The growth rate of North Korean exports to China and the growth rate of imports from China have a correlation of 0.61, and China’s share of North Korean exports and its share of North Korean imports have a correlation of 0.75. The export and import measurements represent two aspects of North Korean trade and are therefore expected to have a positive correlation.

Models and Results

Two times-series ordered logit models are employed. Model 1 examines the impact of explanatory variables in one month on North Korea’s behaviour in the same month. Model 2 examines the impact of the explanatory variables in one month on North Korea’s behaviour in the next month. We chose time-series ordered logit models because the dependent variable is ordinal and the models analyse a time-series data set. The ordered logit model estimates the impact of the explanatory variables on the probability of a change between the three ordered categories of the dependent variable. If there is support for our hypothesis, then we should find that North Korea becomes more likely to resist China’s economic leverage by defying the international community and escalating the crisis.

One caveat of the ordered logit model is that the model is built on the parallel regression assumption,⁵

It is also called proportional odds assumption.

that is, the degree of impact of each explanatory variable on all categories of the dependent variable is the same. For this study, it means that our models assume that each variable has the same degree of impact on the probability of North Korea changing from cooperative policies to maintaining the status quo as that of shifting from the status quo to resistance. This assumption is not always valid for ordered logit models. If our models violate this assumption, then our results could be unreliable. To test whether our models violate the parallel regression assumption, we fit the data with a generalised ordered logit model (Williams, 2006).⁶

Generalized ordered logit model is less restrictive than ordinal logit model in that the former allows independent variables to have different degrees of impact on the dependent variable at each level and at the same time still takes into consideration the hierarchy of the dependent variable. We used the ‘gologit2’ command in STATA developed by Williams (2006).

The results of the generalised model suggest that the model does not violate the parallel regression assumption, which means that we may utilise ordered logit models to test the hypothesis, and the results would be reliable.

The results of the regression analysis are reported in Table 3. The estimated coefficients presented in the table are in the odds ratio format.⁷

Mathematically, odds ratio equals Pr(y≤k)/ Pr(y>k).

An odds ratio that is greater than one indicates that the independent variable has a positive effect on the dependent variable whereas an odds ratio smaller than one indicates a negative effect.

The results of Model 1 indicate that, first, an increase in North Korea’s import reliance, or vulnerability, results in a statistically and quantitatively significant increase in the probability of resistance from North Korea. However, the growth of imports from China has little impact. In other words, a worsening of North Korea’s import vulnerability, but not absolute import volume, would make it more likely to resist. Second, unlike its import vulnerability, North Korea’s export reliance on China has no statistically significant impact on its behaviour, but its exports growth does, which partially supports Bader’s findings that trading with China helps authoritarian regimes survive.

Table 3

. Impact of China’s Leverage on North Korean Behaviour

	Model 1		Model 2
Independent Variables	Odds Ratios	SE	Odds Ratios	SE
Growth of imports from China	0.04	−0.09	0.36	−0.67
% of imports from China	51.08**	−98.11	0.82	−1.51
Growth of imports × % of imports	94.81*	−256.61	2.53	−6.49
Growth of Exports to China	226.72*	−683.6	0.02	−0.06
% of Exports to China	0.37	−0.62	1.4	−2.34
Growth of Exports × % of Exports	0.00**	0	1884.47	−8893.94
Meet China	0.93	−0.45	2.13	−1.04
Meet United States	0.49	−0.4	0.78	−0.59
Meet Russia	0.8	−0.88	0.59	−0.62
Meet South Korea	0.51	−0.49	0.32	−0.29
Sanction	2.06	−1.5	2.51	−1.93
Six-Party Talks	0.47	−0.51	0.42	−0.47
Stage of Kim Jong-Un	1.00	−0.37	1.63	−0.6
Month	0.99	−0.06	1.02	−0.06
Previous behaviour	2.02**	−0.69	1.98**	−0.66
Cut1
Constant	1.51	−1.62	0.45	−0.47
Cut2
Constant	61.86***	−71.98	15.79**	−17.15
sigma2_u
Constant	1.00	0.00	1.00	0.00
Chi-square	18.86		17.71
p > Chi-square	0.09		0.12
N	133		133

Exponentiated coefficients.

Source: The authors.

Notes: * p<0.1 ** p < 0.05 *** p < 0.001; SE: Standard Error.

One explanation for these two interesting phenomena is that North Korea only imports for the sake of necessity rather than economic efficiency. Thus, relying on China only for strategically important resources would give China tremendous influence, which the North Korean leader would like to resist. This departure from cooperative behaviour could be seen before 2018 as North Korea became more resistant to the international community’s demands while at the same time its import reliance on China was on the rise. After examining the trade data closely, we find that the rise in North Korea’s import reliance on China has been exacerbated by a decline in the trade volume with other trade partners due to international sanctions. In other words, the sanctions forced North Korea to rely on China for imports, which caused the former to be more resistant to the latter’s demands. Exports to China can provide extra sources of economic growth; however, the regime does not rely on exports to China to politically survive. Therefore, its export reliance on China has little impact on its behaviour. It is worth noting that although the trade data show seasonal patterns, the statistically insignificant ‘month’ variable in Model 1 suggests that seasonality does not affect North Korea’s foreign policy.

Third, interestingly, the two interaction terms for economic leverage have different impacts on North Korean behaviour. China’s import leverage (interaction between import growth and import reliance) has a positive impact on the probability of North Korea’s resistance, whereas its export leverage has a negative impact. Additionally, the increase in the interaction terms reflects a rise in vulnerability and a decrease in pressure. These intriguing findings will be analysed with marginal effects analysis in the next section.

Next, neither bilateral meetings nor multilateral negotiations have an impact on North Korean policies. This result can be understood as a consequence of North Korean Juche ideology, or problems may exist with the measurement of political pressure. Negotiations or meetings are not the only ways for states to apply political pressure on each other. Additionally, our data only include publicly reported meetings and negotiations and exclude all the secret negotiations, which are arguably more important than the former because they not only set up the foundations for negotiation outcomes but also determine whether public meetings happen in the first place. Model 1 also shows that the policy in one month follows the policy direction set by the previous period, which implies continuity and some degree of certainty in North Korean policymaking. Lastly, although Kim Jong-Un and Kim Jong-Il have many differences and Kim Jong-Un’s power status grew significantly in North Korea, our results indicate there is no statistically significant difference among the three periods in terms of foreign policy reaction to external pressure. Overall, Model 1 has a p value of 0.09, which indicates the model is moderately well fit.

Model 2 examines whether international economic and political leverage in one month affect North Korean foreign policy behaviour in the future. In this model, only the time variable is statistically significant. This suggests that although North Korean leaders tend to follow their previous policy, they are more responsive to the immediate development in the external environment, as Model 1 indicates. However, Model 2 has a p value of 0.12, which indicates that the model as a whole does not have a good fit for the data and thus the results of Model 2 are not very reliable.

Table 4.

Average Marginal Effects Analyses on the Probability of Resistance and Cooperation

	Resistance		Cooperation
Independent Variables	Coefficients	SE	Coefficient	SE
Growth of imports from China	−0.52	0.32	0.27	0.17
% of imports from China	0.66**	0.32	−0.35**	0.18
Growth of imports × % of imports	0.76*	0.45	−0.40*	0.24
Growth of exports to China	0.90*	0.49	−0.47*	0.28
% of exports to China	−0.16	0.28	0.09	0.15
Growth of exports × % of Exports	−1.57**	0.77	0.83*	0.44
Meet China	−0.01	0.08	0.01	0.04
Meet United States	−0.11	0.14	0.06	0.07
Meet Russia	−0.04	0.18	0.02	0.10
Meet South Korea	−0.11	0.16	0.06	0.08
Sanction	0.12	0.12	−0.06	0.07
Six-party talks	−0.13	0.18	0.07	0.10
Stage of Kim Jong-Un	0.00	0.06	−0.00	0.03
Month	−0.00	0.01	0.00	0.00
Previous behaviour	0.12**	0.06	−0.06**	0.03
N	133		133

Source: The authors.

Note: * p < 0.1 ** p < 0.05 *** p < 0.001; SE: Standard error.

Marginal Effects Analysis

Although our statistical analysis shows some support for our hypothesis, single coefficients in a non-linear model, such as the ordered logit model, are not capable of capturing non-linear change. Thus, a marginal effects analysis is applied to give us a more refined picture of the impact of China’s economic leverage on North Korea’s behaviour. Table 4 presents the results of the average marginal effect for the probability of North Korea resisting and cooperating and they confirm our overall findings in the regression analysis. For instance, a one-unit increase in the import vulnerability would increase the probability of North Korea resisting by 66 per cent and lower the probability of its cooperation by 35 per cent. A one-unit increase in China’s export growth rate would increase the probability of North Korea resisting by 90 per cent and decrease that of cooperation by 47 per cent. These results also indicate a strong tendency for North Korean resistance. However, the overall marginal effects analysis still is subjected to the issues of using a single value to measure the impact of a variable on the dependent variable over a range of values. More importantly, the impact of interaction variables is still unclear. The reason it is difficult to find the impact of economic leverage is that it is a product of two continuous variables, while marginal effects requires holding one variable constant and allowing the other variable to change, which is difficult to do in our model.

To examine the impact of the interaction variables, we conducted average marginal effects analysis on one interacting variable (or one factor in the multiplication) while controlling for the other variable at specific values. Table 5 shows the marginal effect of import growth rate from China on the probability of North Korea resisting, while controlling the levels of import vulnerability from 0 to 100 per cent with 10 per cent intervals. The result is also illustrated in Figure 4. Although import growth rate is not statistically significant in the regression model or in the overall average marginal effects analysis, its impact is statistically significant when North Korea relies on China for more than 80 per cent of its imports. Further, the more North Korea depends on China for imports, the more likely a decrease in imports from China renders North Korea resistant. The results also suggest that allowing North Korea to increase imports from China may lower the prospect of a resistant behaviour. This could be because imports such as food and electricity from China alleviate domestic social pressure on the North Korean government. Consequently, North Korea would have one less reason to escalate the nuclear crisis as goods from China help alleviate domestic political tensions.

Table 5.

Average Marginal Effects of Growth of North Korean Imports from China on the Probability of Resistance at Different Levels of Reliance

Import reliance	0	0.1	0.2	0.3	0.4	0.5
Marginal effect (dy/dx)	−0.06	−0.09	−0.12	−0.17	−0.23	−0.31
Standard error	(0.09)	(0.12)	(0.14)	(0.17)	(0.20)	(0.23)
Import reliance	0.6	0.7	0.8	0.9	1
Marginal effect (dy/dx)	−0.40	−0.50	−0.59*	−0.66*	−0.70*
Standard error	(0.27)	(0.31)	(0.36)	(0.39)	(0.42)

Source: The authors.

Note: * p < 0.1.

Figure 4.

Source: The authors.

Relatedly, Table 6 and Figure 5 show the marginal effect of import vulnerability on the probability of North Korean resistance while controlling for the import growth rates. It should be noted that the minimum import growth rate in the data set is −0.8. The marginal effect at the growth rates lower than −0.8 is estimated based on regression results. The result displays an interesting non-linear pattern of change in probability. When North Korea’s imports from China are cut by more than 60 per cent, import reliance on China has little impact. This finding is logical as a significant cut in trade volume often implies that economic ties have been severed due to a damaged diplomatic relationship.⁸

Cutting import volume does not necessarily lower North Korea’s import reliance on China, which is affected by North Korea’s total import volume.

Therefore, North Korea would have no reason to listen to China’s request. However, when the decrease in import volume is moderate (between 0% and −5%), the action signals that China is using the import cut as a punishment to influence North Korean behaviour rather than severing ties. Under this circumstance, North Korean leaders would resist China’s pressure in order to protect their control over decision-making. However, as import growth rises, the impact of import vulnerability declines and has little effect on North Korean behaviour. One explanation is that North Korea’s imports from China consist mostly of necessities such as food and energy. These resources contribute to people’s basic survival and contribute little to state capability.

Table 6.

Average Marginal Effects of Import Reliance on China on the Probability of Resistance at Different Levels of Imports Growth Rate

Import growth rate	−1	−0.9	−0.8	−0.7	−0.6	−0.5
Marginal effect (dy/dx)	0.47	0.53	0.59	0.64	0.68*	0.71**
Standard error	(0.46)	(0.44)	(0.42)	(0.39)	(0.37)	(0.36)
Import growth rate	−0.4	−0.3	−0.2	−0.1	0	0.1
Marginal effect (dy/dx)	0.72**	0.71**	0.68**	0.64**	0.59**	0.53*
Standard error	(0.35)	(0.34)	(0.33)	(0.31)	(0.30)	(0.28)
Import growth rate	0.2	0.3	0.4	0.5	0.6	0.7
Marginal effect (dy/dx)	0.46*	0.40	0.34	0.28	0.23	0.19
Standard error	(0.27)	(0.26)	(0.25)	(0.24)	(0.22)	(0.20)
Import growth rate	0.8	0.9	1
Marginal effect (dy/dx)	0.16	0.13	0.11
Standard error	(0.18)	(0.16)	(0.14)

Note: * p < 0.1 ** p < 0.05.

Source: The authors.

Figure 5.

Source: The authors.

Our results indicate a different pattern for export leverage. For instance, Table 7 and Figure 6 show that at nearly all levels of export reliance, an increase in the growth of North Korea’s exports to China would make it more likely to resist. In other words, increasing revenue from exports could allow North Korea to acquire resources to focus more on developing state capabilities rather than feeding its population. The results agree with the aforementioned results of Model 1 in Table 4, that is, a decrease in the export growth would undermine or slow down the growth of North Korea’s capabilities and consequently make North Korea less capable of resistance. Marginal effects analysis results in Table 8 also reconfirm that North Korea’s export reliance on China has little impact on its behaviour regardless of changes in its exports to China.

Table 7.

Average Marginal Effects of Growth of Exports to China on the Probability of Resistance at Different Levels of Reliance

Export reliance	0	0.1	0.2	0.3	0.4	0.5	0.6
Marginal effect (dy/dx)	1.06*	1.04*	1.02*	1.00*	0.98*	0.95*	0.92*
Standard error	(0.57)	(0.57)	(0.56)	(0.55)	(0.53)	(0.52)	(0.50)
Export vulnerability	0.7	0.8	0.9	1
Marginal effect (dy/dx)	0.89*	0.86*	0.83*	0.79
Standard error	0.49	0.48	0.48	0.49

Source: The authors.

Note: * p < 0.1.

Table 8

. Average Marginal Effects of Exports Reliance on China on the Probability of Resistance at Different Levels of Export Growth Rates

Export growth	−1	−0.9	−0.8	−0.7	−0.6	−0.5
Marginal effect (dy/dx)	0.00	0.00	−0.01	−0.01	−0.02	−0.03
Standard error	(0.01)	(0.01)	(0.02)	(0.03)	(0.04)	(0.06)
Export growth	−0.4	−0.3	−0.2	−0.1	0	0.1
Marginal effect (dy/dx)	−0.05	−0.07	−0.09	−0.12	−0.15	−0.17
Standard error	(0.09)	(0.12)	(0.16)	(0.21)	(0.25)	(0.28)
Export growth	0.2	0.3	0.4	0.5	0.6	0.7
Marginal effect (dy/dx)	−0.18	−0.17	−0.16	−0.14	−0.11	−0.09
Standard error	(0.30)	(0.29)	(0.27)	(0.24)	(0.21)	(0.17)
Export growth	0.8	0.9	1
Marginal effect (dy/dx)	−0.07	−0.05	−0.04
Standard error	(0.14)	(0.11)	(0.09)

Source: The authors.

Note: * p < 0.1.

Figure 6.

Source: The authors.

Conclusion

This study examines why increasingly heavier economic pressure from China has failed to contain North Korean behaviour. We argue that North Korea’s foreign policy strategy involves China in two different ways. At the international level, North Korea relies on China to balance against external threats like the United States and the regime needs to maintain a good relationship with China. At the domestic level, North Korean leaders view China as a foreign power that seeks to intervene in Korean affairs and as a threat to their authority. Acquiring nuclear weapons not only helps the North Korean leaders mitigate external security threats but also allows them to focus more on dealing with international and domestic political threats. Looking back at the development of North Korea’s domestic political changes, as North Korea’s nuclear and missile technology matures, Kim Jong-Un moved to eliminate the pro-China faction and shifted political power and the control over nuclear weapons from the military to the party or himself.

Our historical and statistical analyses of the North Korea–China relationship show that the Sino–Korean relationship is not as close as many believe, and China may not have leverage on North Korea. How can the international community prevent North Korea from resisting? We find that North Korea responds to import and export leverage in different ways. Rising import reliance renders North Korean leaders vulnerable to China’s influence. Generally, the more North Korea relies on China for imports, the more resistant North Korea becomes because its leader seeks to stave off China’s influence in his own domestic politics before he protects the regime’s security from external threats. This is even more so the case when the external threat is not a major concern because of North Korea’s possession of nuclear weapons. Moreover, when China moderately cuts its exports to North Korea and the latter is heavily dependent on the former, not only does North Korea’s concern over China’s influence turn into a reality but it also triggers strong repercussions from North Korea. Our analysis suggests that North Korea is less likely to resist either when it receives more imports from China while other countries do not export to the former or when the country becomes less reliant on Chinese imports. The latter can be achieved by allowing the international community to fill in the gap when China cuts exports to North Korea. Unlike its response to imports, North Korea is unresponsive to its level of export reliance on China but is more likely to resist when its exports to China grow. Hence, cuts on North Korean exports to China could effectively contain its behaviour.

Our analysis indicates that strategies that seek to rely on total isolation to change North Korean behaviour are unsubstantiated and a carefully designed engagement strategy from the international community, particularly the United States but not China, would produce a more rewarding result. This is supported by empirical evidence. The strategies adopted by the international community have gone through three phases. The first strategy adopted prior to 2016 by the United States and its allies was referred to as ‘strategic patience’. Acknowledging that resources and influences are not infinite, under this strategy, the United States would not respond to North Korea nor make compromises with the regime until it showed credible steps to denuclearise. While the United States sought to be ‘patient’ with North Korea, it placed a great degree of importance on China’s actions with regards to North Korea (Office of the Press Secretary, 2014).

The strategic patience phase ended when the current US President Donald Trump and his administration declared that ‘the era of strategic patience is over’ (Diamond, 2017; Kennedy, 2017), but also called upon China to put more economic pressure on North Korea. In addition to the requests, the United States adopted a more coercive approach towards China and North Korea (Schectman & Brunnstrom, 2017). It was during these two phases when North Korea became increasingly resistant.

Some argue that if sanctions on North Korea are kept long enough, the regime will eventually collapse on its own under the rising domestic pressure. However, history shows that sanctions have not worked effectively in totalitarian regimes and they are unlikely to have much impact on regime stability in North Korea either. After living under sanctions for decades, North Korean people have become ‘used to the life under sanctions’ and developed ways to adapt to it.⁹

This was determined via the author’s interview with a North Korean guide during a trip to North Korea in 2016.

Moreover, the majority of the population in North Korea face tremendous difficulties and risks when planning and participating in collective mass movements against the government, whereas those who have the power to change the regime choose to leave the country instead of staying to fight for changes.

While relying on China to exert heavy pressure on North Korea did not elicit a change in the regime, it was not until the United States, followed by South Korea, changed its policy from coercion to engagement when de-escalation was provoked. For instance, not only have there been no nuclear tests or missile tests by the regime, but Kim Jong-Un also met with President Trump at a summit and announced that the country would dismantle its nuclear test site and pursue denuclearisation.

Our analysis suggests that the current North Korean leader has reasons to view China as a potential threat to his rule; therefore, it is not surprising that Kim Jong-Un would welcome an external power that is willing to offset the rising influence of China. Although North Korea officially considers the United States its enemy and the deep distrust between the two countries has prevented the United States from engaging with North Korea, the engagement strategy had a record of success with North Korea who for years has been seeking direct bilateral talks and normalisation of its relationship with the United States (Choe, 2009). For instance, the Clinton Administration utilised an engagement strategy with North Korea in 1994 when attempting to convince the country to dismantle its nuclear weapons programme. Similarly, the fourth round of the six-party talks in 2005 reached a joint statement on agreed steps towards denuclearisation on the Korean peninsula after all parties made compromises. The Agreement Framework failed in part because the US domestic political struggle prevented President Clinton from fulfilling his US commitments. Further, the joint statement of the six-party talks was never implemented when the United States requested new amendments to the statements and sanctioned North Korean trading companies and their partners, such as Banco Delta Asia of Macau.

To conclude, coercion has been ineffective and a disservice to all stakeholders. The United States might consider continuing the current course of engagement, without disruptions from domestic political struggles. Normalising the relationship with North Korea does not come at a high cost for the United States.¹⁰

Many scholars disagree with this statement as the normalization of the US relationship with North Korea may cause the relative power of the United States to decline since it is negotiating with a much smaller country.

Instead, a relationship would motivate North Korea to be less aggressive towards the United States and in turn cause the United States to believe it is less threatened by North Korean nuclear weapons. This solution will also release China from the dilemma in which it has to choose between maintaining a US-China relationship and border security. The Trump Administration’s strategy on North Korea has been shifting back and forth between coercion and engagement. While the latest strategy is that of engagement, it is unclear if the administration will indeed proceed to engage with North Korea, but our findings lend support for the notion to do so.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.

Funding

The authors received no financial support for the research, authorship and/or publication of this article.

Data Availability Statement

The data that support the findings of this study are openly available.

References

Bader

(2015). China, autocratic patron? An empirical investigation of China as a factor in autocratic survival. International Studies Quarterly , 59, 23–33.

Blancke

(2016 November 15). Sanctions busting: How North Korea smuggles illicit goods. NK Pro .

Centre for Strategic and International Studies. (2017, January 1). North Korean missile launches. Missile threat. Retrieved from https://missilethreat.csis.org/north-korea-missile-launches-1984-present/

Choe

S-H.

(2009, November 1). North Korea presses US to agree to bilateral talks. The New York Times . Retrieved from http://www.nytimes.com

Chow

Park

J. M.

(2017, February 14). North Korea suspected behind murder of leader’s half-brother. Reuters . Retrieved from https://uk.reuters.com/article/uk-northkorea-malaysia-kim/north-korea-suspected-behind-murder-of-leaders-half-brother-u-s-sources-idUKKBN15T1EA

Chronicle of US-North Korean Nuclear and Missile Diplomacy. (2017). Arms control association . Retrieved from https://www.armscontrol.org/factsheets/dprkchron

Chung

J. H.

Choi

(2013). Uncertain allies or uncomfortable neighbors? Making sense of China–North Korea relations, 1949–2010. The Pacific Review , 26(3), 243–264.

Clover

Harris

Lockett

(2017, September 28). North Korean companies in China ordered to close. Financial Times . Retrieved from https://www.ft.com/content/be406ed8-a4b3-11e7-9e4f-7f5e6a7c98a2?mhq5j=e5

Democratic People’s Republic of Korea (DPRK). (2011). Juche ideology. Retrieved from http://www.korea-dpr.com/juche_ideology.html

10.

Diamond

(2017, June 30). Trump: US patience with North Korean regime is over. CNN . Retrieved from http://www.cnn.com/2017/06/30/politics/trump-moon-jae-in-rose-garden/index.html

11.

Doubek

(2017, Setember 23). China moves to limit fuel, textile trade with North Korea. NPR . Retrieved from http://www.npr.org/sections/thetwo-way/2017/09/23/553087705/china-moves-to-limit-fuel-textile-trade-with-north-korea

12.

Drezner

D. W.

(1998). Conflict expectations and the paradox of economic coercion. International Studies Quarterly , 42(4), 709–731.

13.

Drezner

D. W.

(2003). The hidden hand of economic coercion. International Organization , 57(3), 643–660.

14.

Drury

A. C.

(2000). U.S. presidents and the use of economic sanctions. Presidential Studies Quarterly , 30, 623–642.

15.

Einbinder

(2017, October 3). How North Korea uses front companies to help evade sanctions. PBS . Retrieved from https://www.pbs.org/wgbh/frontline/article/how-north-korea-uses-front-companies-to-help-evade-sanctions/

16.

Elliott

K. A.

(1998). The sanctions glass: Half full or completely empty. International Security , 23, 50–65.

17.

Gause

F. G.

(2007). Threats and threat perceptions in the Persian Gulf region. Middle East Policy , 14(2), 119–125.

18.

Hermann

(1989). Who makes foreign policy decisions and how? International Studies Quarterly , 33(4), 361–388.

19.

Hirschman

(1980). National power and the structure of foreign trade . Berkeley, CA: University of California Press.

20.

Holsti

(1976). Foreign policy viewed cognitively. In Axelrod

(Ed.), Structure of decision . Princeton, NJ: Princeton University Press.

21.

International Trade Centre. (2017). Trade Map. Retrieved from https://www.trademap.org/Index.aspx

22.

Jervis

(1976). Perception and misperception in international politics . Princeton, NJ: Princeton University Press.

23.

Kang

(2003). Getting asia wrong: The need for new analytical frameworks. International Security , 27(4), 57–85.

24.

KCNA. (2017a, October 29). Rodong Sinmun calls for stubborn struggle against U.S. Retrieved from https://kcnawatch.co/newstream/1509260484-528114070/rodong-sinmun-calls-for-stubborn-struggle-against-u-s/

25.

KCNA. (2017b, May 3). Commentary on DPRK–China relations. Retrieved from http://www.kcna.kp/kcna.user.article.retrieveNewsViewInfoList.kcmsf#this

26.

Kennedy

(2017, April 17). Pence tells North Korea: The era of strategic patience is over. National Public Radio . Retrieved from http://www.npr.org/sections/thetwo-way/2017/04/17/524316419/pence-tells-north-korea-the-era-of-strategic-patience-is-over

27.

Kim

Chung

(2017, July 20). North Korea 2016 economic growth at 17-year high despite sanctions—South Korea. Reuters . Retrieved from https://www.reuters.com/article/uk-northkorea-economy-gdp/north-korea-2016-economic-growth-at-17-year-high-despite-sanctions-south-korea-idUKKBN1A6083

28.

Kim

J-U.

(2016). Kim Jong Un’s Speeches at the 7th Workers’ Party Congress . Retrieved from https://www.ncnk.org/resources/publications/KJU_Speeches_7th_Congress.pdf

29.

Kim

S. S.

(2015). Sino-North Korea relations in the post-cold war world. In Kihl

Y. W.

Kim

H. N.

(Eds.), North Korea: The politics of regime survival . New York, NY: Routledge.

30.

Kirshner

Rawi

(1999). Strategy, economic relations, and the definition of national interests. Security Studies , 9, 119–56.

31.

Knorr

(1975). The power of nations: The political economy of international relations . New York, NY: Basic Books.

32.

Kong

(2017a, October 30). China breaks up plot to kill Kim Jong Un’s nephew, report says. Bloomberg Politics . Retrieved from https://www.bloomberg.com/news/articles/2017-10-30/china-breaks-up-plot-to-kill-kim-jong-un-s-nephew-report-says

33.

Kong

(2017b, February 24). North Korea says China ‘dancing to U.S. tune’ in rare spat. Bloomberg Politics . Retrieved from https://www.bloomberg.com/politics/articles/2017-02-24/north-korea-says-china-dancing-to-u-s-tune-in-rare-criticism.

34.

Korean International Trade Association (KITA). (2017, July 13). K-Statistics. Retrieved from http://www.kita.org/kStat/overview_BalanceOfTrade.do

35.

Kristensen

Norris

(2017). A history of US nuclear weapons in South Korea. Bulletin of the Atomic Scientists , 73(6), 349–357.

36.

Lankov

(2013). The real North Korea: Life and politics in the failed stalinist utopia . New York, NY: Oxford University Press.

37.

Levitsky

Way

L. A.

(2006). Linkage verses leverage: Rethinking the international dimension of regime change. Comparative Politics , 38(4), 379–400.

38.

Levitsky

Way

L. A.

(2010). Competitive authoritarianism: Hybrid regimes after the cold war . New York, NY: Cambridge University Press.

39.

Levy

(2003). Balances and balancing: Concepts, propositions, and research design. In Vasquez

J. A.

Elman

(Eds.), Realism and the balance of power: A new debate . Upper Saddle River, NJ: Prentice Hall.

40.

Moravcsik

(1997). Taking preferences seriously: A liberal theory of international politics. International Organization , 51(4), 513–553.

41.

NK News. (2017, January 1). NK leadership tracker. Retrieved from www.nknews.org

42.

Office of the Press Secretary. (2014, April 25). Press conference with President Obama and President Park of the Republic of Korea. The White House . Retrieved from https://www.whitehouse.gov/the-press-office/2014/04/25/press-conference-president-obama-and-president-park-republic-korea

43.

K. C.

Kassig

R. D.

(2000). North Korea through the looking glass . Washington, DC: Brookings Institute Press.

44.

Oye

(1985). Explaining cooperation under anarchy: Hypotheses and strategies. World Politics , 38(1), 1–24.

45.

Pape

(1997). Why economic sanctions do not work. International Security , 22(2), 90–1236.

46.

Park

(2002). North Korea: The politics of unconventional wisdom . Boulder, CO: Lynne Rienner Publishers, Inc.

47.

Perlez

Choe

S.-H.

(2013, December 10). Public Ouster in North Korea Unsettles China. New York Times . Retrieved from http://nytimes.com

48.

Powell

(1999). In the shadow of power: States and strategies in international politics . Princeton, NJ: Princeton University Press. Rose

(1998). Neoclassical realism and theories of foreign policy. World Politics , 51(5), 144–172.

49.

Schectman

Brunnstrom

(2017, November 21). US sanctions 13 Chinese and North Korean organizations. Reuters . Retrieved from https://www.reuters.com/article/us-northkorea-missiles-usa/u-s-sanctions-13-chinese-and-north-korean-organizations-idUSKBN1DL2G6.

50.

Schroeder

(1994). Historical reality vs. neo-realist theory. International Security , 19(1), 108–148.

51.

Schweller

R. L.

(1994). Bandwagoning for profit: Bringing the revisionist state back in. International Security , 19(1), 72–107.

52.

Shen

(2003). Weihu dongbeiya anquan de dangwu zhi ji’ [The urgent task in safe- guarding northeast asian security]. Shijie Jingji Yu Zhengzhi [World Economy and Politics] , 9, 55–60.

53.

Walt

(1985). Alliance formation and the balance of world power. International Security , 9(4), 3–43.

54.

Walt

. (1987). The origins of alliances . Ithaca, NY: Cornell University Press.

55.

Waltz

(1979). Theory of international relations . Reading, MA: Addison-Wesley.

56.

Williams

(2006). Generalized ordered logit/ partial proportional odds models for ordinal dependent variables. The Stata Journal , 6(1), 58–82.

57.

Yonhap News. (2013, December 14). N.K. businessmen in China summoned back after Jang’s execution. Retrieved from http://english.yonhapnews.co.kr/northkorea/2013/12/14/82/0401000000AEN20131214001400315F.html.

58.

Zhan

(2016). Analysis of changes in North Korea’s cognition of China. The Korean Journal of Defense Analysis , 28(2), 199–221.

59.

Zhang

Zinoviev

(2018). How North Korea views China: Quantitative analysis of Korean Central News Agency Reports. Korean Journal of Defense Analysis , 30(3), 377–396.