Difficult decisions: Coercion in asymmetric security cooperation

Abstract

How do minor states protect themselves against coercion from their major power allies? Asymmetric security cooperation is often reduced to tradeoffs between security and autonomy, but coercion is another factor that minor states fear while cooperating. Coercing partner states can take the form of major powers threatening or undertaking regime change, and minor states must weigh the benefits of cooperation against this risk. To this end, I suggest that minor states with anti-major power political oppositions cooperate more extensively with their major power partners than minor states with pro-major power oppositions. A pro-major power opposition provides opportunities for major powers to threaten or replace the incumbent regime; therefore, such minor states limit their cooperation out of fear of coercion. I employ original data on opposition characteristics to present evidence from security relations between the United States and 65 minor states during 1950–1991 to support the hypothesis.

Keywords

coercion cooperation major power minor state security

Research in international relations has often focused on asymmetric security cooperation between major powers and minor states (Baum and Potter, 2015; Beckley, 2015; Keohane, 1971; Kim 2011; Krasner, 1999; Lake 1996; Mattes, 2012; Morrow, 1991; Waltz, 1979). Such cooperation is described as a tradeoff between autonomy and security, i.e. acquiring much-needed security while balancing the loss of autonomy. However, another important factor for the minor state ally is the risk of facing coercion from its major power partner. Coercion is the ability to change behavior through threatening or inflicting punishment. In security cooperation, coercion can take the form of threatening or undertaking regime change, and major powers often employ such coercion towards their minor state allies. For instance, in 1963, the United States backed factions of the South Vietnamese military in overthrowing a pro-US regime under Diem and installing a junta more amenable to US interests in the Southeast Asia region (Cima and Library of Congress Federal Research Division, 1989). In this case, the major power resorted to coercing—and successfully replacing—its ally in hopes of obtaining better policy concessions from opposing factions.

But why would coercion be experienced by a minor state's incumbent regime allied with a major power? Coercion may be expected in hostile inter-state relations, but why does it materialize in security cooperation? Major powers define their national security agendas broadly, which often include the domestic politics of other states (Lake, 2013). Owing to the frequency of interaction, the potential for disagreement between divergent foreign policy agendas, as well as the presence of important major power resources in the minor state—troops, military-to-military contacts, etc.—coercion can become a prominent risk in asymmetric security cooperation. In this case, major powers can not only withhold security benefits from minor state allies, but they can also credibly threaten and replace incumbent regimes in other states. For this reason, managing the risk of coercion while cooperating with major powers is an important consideration for minor states, as security cooperation with the major power is costly to give up. Nonetheless, such coercion in cooperation remains theoretically under-developed. Keeping this in mind, I posit that coercion is a crucial factor in determining the extent of cooperation in asymmetric security cooperation, and that minor states protect against coercion by strategically limiting the scope of their security cooperation with major powers.

To this end, I argue that the basic foreign policy stance of the political opposition in the minor state towards the major power (pro/anti) serves as an important constraint. Minor states with pro-major power political oppositions fear coercion in the security relation, as such opposition provides viable opportunities for the major power to threaten and replace the incumbent regime. In other words, the presence of a pro-major power political opposition provides an outside option to the major power in the security relation; hence, lower cooperation between major powers and minor states is expected. On the other hand, in the presence of an anti-major power political opposition in the minor state, the major power is unlikely to obtain a more favorable domestic policy outcome since the viable alternative to the incumbent regime, i.e. the political opposition, is not favorable towards the major power. In such a situation, coercing the incumbent regime can lead to a worse and undesirable outcome for the major power. Anti-major power oppositions therefore reduce the possibility of coercion from the perspective of the minor state, and allow regimes to cooperate more extensively with the major power. Original coding on opposition foreign policy stances is combined with data on US troop placements to test this argument in the case of security relations between the United States (US) and 65 minor states in the Cold War (1950–1991) period.

Existing approaches to security cooperation

The acquisition of security—taken to mean the ability of a state to have control over its territory and the freedom to choose its own form of rule (Lake, 1996)—is fundamental for the survival of the incumbent regime in a state. Within this framework, states can decide to cooperate with one another and acquire security. A primary assumption in this regard is that states can achieve security through the aggregation of security capabilities of other, more powerful states they choose to partner with (Morrow, 1991). Such asymmetric cooperation between major powers and minor states is the focus of this research.¹

However, security cooperation between states of different sizes and capabilities is marked by a bargain, such that the more powerful states provide security—something that the minor state may need but cannot produce on its own—in exchange for the minor states’ allegiance or control over some of their policies, in the words of Krasner (1999), the more sovereignty is essentially traded for security. This “security–autonomy” tradeoff was popularized by Morrow (1991), who posited that in asymmetric alliances the minor state can obtain security support from the major power in return for losing its autonomy. Through such asymmetric security cooperation, therefore, major powers can be expected to benefit by reducing the number of potentially adversarial states, whereas minor states can benefit from acquiring security for their domestic territories with the help of the major power's resources.

Such cooperation is necessarily hierarchic, as the major power seeks authority over the minor state (Lake, 2008). This hierarchical characteristic of asymmetric security relations is salient as it allows the major power to constrain the minor state through conditions of economic and security dependence. These hierarchical relations have been studied previously in extant literature, but an important constraint of this literature is that it focuses on developing theoretical arguments only from the major power's perspective. Most research on security cooperation (Biglaiser and DeRouen, 2009; Butt 2013; Gowa and Mansfield, 1993; Ikenberry and Kupchan, 1990; Kim, 1989, 2011; Lake 1996; McDonald, 2015; Mattes, 2012; Mousseau, 1997) delineates incentives for cooperation between asymmetrically powerful states only from the perspective of the major power. For example, research undertaken by Lake (1996) and McDonald (2015) attributes international security arrangements resulting in concrete cooperative or hostile outcomes to the decisions taken by major powers. Such decisions often involve minor states, but they are not given adequate agency in the theoretical construct. Additionally, Lake (1996), theorizing about the structure of security cooperation, never attributes any role to minor states, assuming instead that minor states will simply follow the direction set by the major power. Similarly, Mousseau (1997) fails to incorporate the varied incentives for minor states to cooperate with major powers under democracy. Furthermore, in the work of McDonald (2015) there is no room for domestic explanations related to minor states since all agency pertaining to the lack of conflict only comes from major powers.

Additionally, the literature on alliance formation (Altfeld, 1984; Leeds et al., 2002; Morrow, 1991) also suffers from a narrow focus on major powers. Beyond alliance formation, theoretical considerations about alliance design and reliability also demonstrate a bias to portray only the major power's preferences. For instance, Mattes (2012) discusses how power asymmetry can impact alliance design—and even contends that minor states can seldom induce favorable changes—but stops short of developing a theoretical framework for coercion in alliances. Moreover, Kim (2011) analyzes American alliances with South Korea, Japan, and Spain, but only considers tinkering that was done to the alliance by the major power. Beckley (2015) also recognizes that major powers have immense “freedom of action” to prevent getting themselves entangled abroad through security relations, as they can freely exert their influence on minor state allies.²

Contributions of this research: Coercion considerations

The tendency in the security cooperation literature to ignore minor states in theoretical constructs is a significant hurdle towards a better understanding of security relations, since neglecting the minor state as an important player leads to several theoretical problems. Asymmetry in security relations is often ignored, and research seeking to highlight this gap, such as Harrison (2005), Jackson (2015), Keohane (1971), Park (1975), and Womack (2016), remains rare. The main issue here is that minor states are neither replicas of major powers nor obedient partners in security relations, and in ignoring these critical aspects, existing approaches to security cooperation fail to incorporate important inter-state dynamics. Security cooperation with a major power does not bear identical costs and benefits for minor states, as minor states significantly differ in their motives and capacities for security cooperation from major powers. Furthermore, minor state foreign policy is characterized by constraints different from the ones experienced by major powers (Krause and David Singer, 2001) and they have also been shown to be more sensitive towards major international developments in comparison with major powers (Hey, 2003).

Consequently, existing theoretical frameworks fail to capture a crucial element of asymmetric security cooperation, i.e. coercion by the major power towards the minor state, that is built into the very fabric of asymmetric security relations. Coercion is the ability to change behavior through threatening or inflicting punishment, and it can be reasonably argued that asymmetric security cooperation provides ample incentives to the major power for coercing its minor state security partners. Some examples of coercion in security relations can be the covert empowerment of the political opposition, withholding security benefits, or even pushing for regime change in the minor state. This latter manifestation of coercion—the threat of regime change—is a salient factor for incumbents in minor states, as it directly relates to regime survival concerns; hence, this article theorizes about this particular form of coercion.

In this context, I suggest that asymmetric security cooperation cannot be completely explained by the “security–autonomy” tradeoff, as balancing finite resources with decisions regarding policy independence (Morgan and Palmer, 2003; Morrow, 1991; Powell, 1999) is inherently more difficult for minor states that cooperate with a major power. Since a major power can define its national interests broadly to include the domestic politics of the minor state (Lake, 2013) as well as enjoying access to resources that can be used to coerce the minor state for asymmetrically favorable policies—military-to-military contacts, troops, etc.—a minor state that chooses to cooperate with a major power must choose between its desire to continue its independent function and acquiring security from the major power, all within the shadow of coercion.

It is important to note that this article focuses only on the coercion experienced by minor states from their major power allies. Even though coercion towards allies remains theoretically underdeveloped, several historical cases of coercion by one major power, the United States, to undermine and replace the incumbent regimes in security allies such as Japan (1952–1968), Iran (1952–1953), South Vietnam (1963, 1967), and Thailand (1965) provide ample historical context to this inquiry.³ Major powers care about the domestic policies of their minor state security partners since they establish their foreign policy goals to often include domestic policy in other minor states; therefore, obtaining more favorable policy outcomes in other minor states is a desirable goal for any major power. In pursuit of this goal, major powers can choose to coerce even their allies when favorable conditions are present. Such coercion frequently involves attempts from the major power to induce regime change in the weaker security partner.

Coercion and regime change

Even though coercion in asymmetric security relations can materialize in both overt and covert ways, regime change is often the key objective whenever coercion is at play. As the cases of many Middle-Eastern, Southeast Asian, and Latin-American countries demonstrate,⁴ the common strand between different coercive actions is that all such actions are aimed at inducing regime change within a minor state, as the prospect of more favorable outcomes for the major power can only be achieved under new, more favorable leadership. Regime change is the preferred mode of coercive action by the major power since it results in overcoming important barriers to cooperation (O’Rourke, 2013: 64–68). First, it increases the similarity in shared interests between the two security partners. The reason coercion materializes is precisely due to the lack of such commonalities in the security relation, and as such, then, regime change solves this problem.

Second, coercion also provides the major power an opportunity to reduce the amount of uncertainty within a security relation. As distinct and independent entities, minor states guard their preferences for future decisions, which can result in the major power being uncertain of the minor state's intentions. Coercing and replacing the regime corrects this imbalance in favor of the major power, thereby making this option a unique one. Third, regime change through coercive means also provides more leverage to the major power. If the major power succeeds in its attempts, it can grow bolder in its demands; additionally, the newly-installed regime is in a worse position as well, as it knows how much influence the major power can wield in case of disagreement.⁵

Historical examples demonstrate this mechanism at play. In the case of South Vietnamese leader Ngo Dinh Diem, the United States staunchly supported his rule in the country after French colonization ended in the wake of the Second World War, despite him lacking widespread domestic support. American support lasted until the winter of 1963, when in the presence of a more lucrative domestic alternative, i.e. a military junta more amenable to American interests, the major power pushed for regime change (Jones, 2003). The South Vietnamese case is instructive and supports the theoretical argument presented here, as the regime's repressive actions towards Buddhists in the summer of 1963, while a concern for the major power, did not trigger regime change. In fact, regime change only became a viable option six months later, once reliable promises from factions within the South Vietnamese military guaranteeing more control to the US emerged. A few years earlier in 1961, the United States had also helped depose Rafael Trujillo in the Dominican Republic, after initially helping him prop up his rule since 1930. Unlike the South Vietnamese case, however, a favorable alternative did not come to power in the Dominican Republic as Trujillo's son subsequently assumed the reins of power (Diederich, 2000). In both cases, the major power resorted to coercion towards allies, which highlights the point that minor states need to strike a balance between their desire for security and the risk of coercion by the major power, a choice distinct from the security–autonomy tradeoff.

Therefore, once the possibility of coercion in security relations is accounted for, it is not difficult to see that minor states interacting with a major power need to account for the risk of coercion by the major power. Hence, any minor state interacting with a major power faces a dilemma: it can cooperate and gain security, but it also incurs the possibility of being coerced by the major power, even when the two states may be security allies and the minor state's leadership may be friendly towards the major power. Major powers coerce the incumbent regimes in minor states to acquire favorable policy outcomes, and if such outcomes remain difficult to achieve under the incumbent leadership, major powers can take actions to undermine and replace the leadership with a more favorable alternative, i.e. the political opposition.

For this reason, fear of coercion is the cause of fundamental differences in the preferences of the two states that engage in a security relation, and it makes the list of potential choices, costs, and benefits for each state asymmetrical. However, security cooperation with major powers is costly to give up once established, and in this scenario the possibility of coercion—and therefore the extent of cooperation—in a security relation between a major power and minor state are inextricably linked to the domestic situation prevalent in the minor state. In particular, viable alternatives to the incumbent regime in the form of political oppositions can substantially alter the nature of asymmetric cooperation as well as impacting the potential for coercion. We can therefore expect incumbent regimes in minor states to act strategically and weigh the risks and benefits of allying with major powers. To this end, I hypothesize that owing to the fear of coercion, minor states with pro-major power political oppositions cooperate less extensively with their major power partners than minor states with anti-major power oppositions.

Political oppositions and security cooperation

Oppositions in a minor state can comprise actors or factions that compete for the control of government. As far as the role of oppositions towards security relations is concerned, two relevant factors make opposition characteristics crucial for the incumbents. First, it can be reasonably argued that incumbent regimes in minor states are inherently concerned with regime survival (Bueno de Mesquita et al., 2003); subsequently, ensuring the continuation of their regime forms a primary concern for incumbent regimes in minor states (Morgan and Palmer, 2003; Powell, 1999). Consequently, countering the opposition a regime faces and limiting the opposition's chances to come to power is a central concern for a minor state's regime. To this end, enacting policies that would help the opposition gain political resources and public support plays counter to the goals of the incumbents, and these strategic concerns have been shown to be salient towards foreign policy decisions (Fordham, 2019). These concerns are evident in democratic setups but they are present even in authoritarian regimes, as opposition groups or internal factions within the regime can also lead to similar outcomes.

Second, oppositions in a minor state can be seen as viable alternatives to the incumbents by the major power, especially if the opposition views the major power in a favorable light. This possibility is anathema to incumbents trying to hold on to power, since in the presence of viable domestic alternatives major powers could readily coerce them and install more amicable factions in their place. When oppositions in minor states do not share the major power's policy outlook, coercing the incumbents with the aim of regime change is much harder for the major power.

Therefore, following Putnam (1988), it is reasonable to expect that the incumbent regime must take into account the domestic opposition it faces while negotiating international agreements—such as security cooperation—as doing so is required for considerations of regime survival. In this context, if the opposition in the minor state is more pro-major power than the incumbent regime, then making concessions to the major power through security cooperation can benefit the opposition and harm the incumbent regime's chances to limit the possibility of coercion. The major power can use its position as leverage to get a better outcome by empowering the pro-major power opposition instead of the incumbent regime through overt or covert means. Since the major power stands to get a better deal—asymmetrical benefits in return for side-payments, for example—by replacing the incumbent regime, the presence of a pro-major power opposition can spell disaster for the incumbent regime's survival. Hence, the risk of coercion becomes very real for minor states that extend concessions to major powers in the presence of pro-major power political oppositions, as the major power has an outside option in the security relation.⁶ Therefore, incumbent regimes in minor states that face a pro-major power political opposition should show restraint when extending security concessions to the major power.

On the other hand, if the opposition in the minor state is anti-major power, then extending security concessions is a viable option for the incumbent regime. When the political opposition to the incumbent regime in a minor state is characterized by an anti-major power positions in its foreign policy, making security concessions to the major power protects the regime from coercion since the major power cannot hope to get a better deal from the regime's alternative. Even if the incumbent regime in the minor state is dependent on the major power's security cooperation, the minor state's regime does not face significant threats to its survival since the political opposition cannot offer a better deal to the major power. In this way, the incumbent regime in the minor state can reduce the possibility of coercion in asymmetric security cooperation; this results in more security concessions from the minor state to the major power.⁷ A related theoretical expectation is given below.

Hypothesis: Minor states with anti-major power political oppositions allow for higher levels of security concessions to their major power security partners than minor states with pro-major power political oppositions.

Before introducing the research design chosen to test the above hypothesis, a few important points bear mention. First, it must be noted that major powers are not necessarily limited in coercing only their allies, as the presence of a security relation is not a prerequisite for coercion. As a matter of fact, major powers can and do coerce non-allies as well (O’Rourke, 2018) but this article focuses on coercion directed at allies for the following reasons. First, coercion in security cooperation is theoretically interesting because it is unexpected, especially given the extant literature that poses symmetric consequences for both states. Theories of cooperation seldom mention coercion between allies, and almost always choose to ignore it while building models of cooperation. Second, security cooperation engenders frequent contact and therefore it is more likely for policy differences—and opportunities for coercion—to occur between allies than non-allies due to limited contact between the latter. Lastly, it is also easier for a major power to coerce an ally since it has important resources on the ground, in the form of actual troops or military-to-military contacts, which can be called upon to exert influence in the minor state's domestic politics or to engineer outright coups. For these reasons, this article focuses only on coercion within security relations.

Second, the difficulty here is that the pattern of behavior hypothesized here takes place in the background, and is shielded from the public. Therefore, it is not something on which we would expect the relevant actors to comment extensively, and only the consequences of the decision—such as lower or higher cooperation—can be observed. Such behavior is not anomalous, however, as many social and political phenomena exhibit similar patterns. For instance, selection processes that shape who participates in a decision-making process (for instance, a legislative body) can end up determining the outcome, even when such processes remain difficult to observe. However, if these processes are stringent enough—so much so that only a certain type of individual runs for office—then they can produce a group of actors with homogenous views on key issues, even if the members of the groups are entirely unaware of them. The only observable outcome in this case would be the homogenous views, but the presence of the underlying processes cannot be ignored. Similarly, the pattern of higher cooperation in the presence of anti-major power oppositions also points to a process that is difficult to observe, but the observed outcome leads to a reasonable expectation about the underlying fear of coercion in the security relation.

Data and methods

The empirical test assesses the above-stated hypothesis in the case of the United States and its security relations during the Cold War period (1950–1991). Owing to the presence of a strong rival—Union of Soviet Socialist Republics (USSR)—in this duration, United States expanded its security relations broadly across the globe to include many minor states. This historical setting is appropriate for testing the hypothesis, as first, extending the argument empirically to American security relations provides broad coverage in terms of minor states. More importantly, however, the pattern of interaction between the United States and other minor states is similar to the mechanisms highlighted in the theoretical discussion: within these asymmetric security relations, coercion can be reasonably expected as minor states wary of the major power's reach find ways to limit their cooperation with the stronger partner. For these reasons, the case of the United States and its security partners in the Cold War makes for an appropriate test of the hypothesis.⁸

The empirical model used is OLS regression with robust standard errors, the unit of analysis for the empirical test is country-year, and the time period of interest is from 1950 to 1991. Additionally, the sample for the empirical test includes all independent minor states recognized by the United Nations (UN) that had severe security issues which would make them desirous of seeking a security relation with the United States. By limiting the sample to only states that could enter into a security relation with the US, we can be reasonably confident that observing the absence of a security relation in this limited sample is only due to no actual security relation materializing between the US and the respective minor state. Therefore, the sample is limited to the universe of all minor states for which their respective security threats were severe enough to warrant considerations of leaning towards the US and relying on its military personnel despite the fear of coercion. Since the time period under investigation is the Cold War era, the states that satisfy the above criteria, i.e. states that could potentially cooperate with the United States, are all minor states that shared a territorial border with the Soviet bloc countries or other Soviet-aligned states from 1950 to 1991. The sample therefore contains states that were under threat both territorially as well as strategically from the former Soviet Union's sphere of influence. Before choosing to enter security relations with the US, such minor states effectively weighed their options and balanced their need for security against the fear of coercion. Therefore, the sample contains all countries during the time period for which the theoretical choice presented in the article—balancing security against coercion—was operational.⁹

Measuring cooperation: US troop stationing

For providing empirical evidence to support the argument, this article uses US troop deployment data in other minor states; this security concession is therefore the dependent variable for the empirical test. US troop stationing data are available through the work done by Kane (2004) for the Heritage Foundation. Troop deployment data are based on original US Department of Defense reports, and the natural log of actual troop numbers is used in the test to remove the effect of outliers and to transform the data into a normal distribution.

Using troop stationing as a proxy for measuring security cooperation is a reasonable option due to several factors. First, stationing its own troops on the minor state's territory is a powerful tool for major powers to expand their influence abroad by providing security assistance to other minor states in need. In other words, such cooperation allows the major power to establish a security dependency over the minor state (Lake, 2008). More importantly, however, the presence of American troops makes the possibility of coercion by the major power quite pronounced. Such cooperation affords the major power the luxury of developing military-to-military contacts in the minor state. As military institutions are highly regularized and follow predictable chains of command, the major power gains a certain foothold in a domestic situation which may otherwise be unpredictable. If the loyalties of the incumbents change unfavorably, having the minor state's military on its side is a major advantage for the major power. If the US decides to seek regime change, it can already count on the support of a powerful domestic ally, i.e. the military.¹⁰ In this case, the theory presented above suggests that minor states will seek to limit US troops on their territories out of fear of coercion. Moreover, while the deployment of a major power's troops is sometimes observed in conflict situations (Little and Leblang, 2004), troops are more often stationed in minor states for cooperation purposes. As Biglaiser and DeRouen (2007) discuss, national security interests frequently involve more than interstate disputes; therefore, the deployment of troops on foreign soil can be considered an essential element of foreign policy.¹¹

Explanatory and control variables

The main independent variable is the foreign policy stance of the political opposition in the minor state (pro-/anti-US). This variable is constructed through original coding from various sources, such as the Manifesto Project data (Volkens et al., 2018), the World Bank's DPI dataset (Scartascini et al., 2017), as well as historical sources such as the Library of Congress’ Country Studies (various citations, see References) and other archival texts. Regarding the pro- or anti-US stance of the opposition, the following coding guidelines are followed. Political oppositions with leftist, socialist, or Communist ideological leanings are categorized as anti-US, whereas rightist, conservative, and republican parties are categorized as pro-US. Given the Cold War context, such a classification makes sense: in the case of Communist political parties/factions, the anti-US stance of the regime and/or opposition is straightforward due to the affinity towards the Soviet Union as a mentor and role model for such parties. However, even in case of leftist and socialist parties, such a classification is reasonable since such parties typically had ideologies that could be treated as similar to the ones taken on by the Communist parties. Furthermore, rightist and conservative parties during the same time period can be seen to have similar ideological stances as those of the United States in the Cold War era, specifically owing to their refusal to adopt Communist policies.¹² Moreover, centrist parties—as reported in the DPI and Manifesto datasets and other coding resources—are also included in the empirical test but no theoretical expectations are attached to these opposition parties.

If relevant opposition characteristics are not available through the DPI or Manifesto data, original coding is done using the Library of Congress’ Country Studies and other historical texts. Through the use of archival sources, the opposition in a minor state is identified and coded as pro- or anti-US based on the following: its ideology (leftist, socialist, or Communist ideological leanings are categorized as anti-US, whereas rightist, conservative, and republican parties are categorized as pro-US), or a documented history of opposing (anti-US) or supporting (pro-US) American foreign policy domestically. For inclusion in the empirical test, a factor variable equaling 0 for centrist, 1 for pro-US, and 2 for anti-US oppositions is constructed, and relevant beta coefficients are reported in the analysis. A robustness check with a binary indicator equaling 1 for anti-US oppositions and 0 for pro-US or centrist oppositions is also included.¹³

The empirical test also makes use of various control variables, which, in addition to controlling for the effect of exogenous factors, are also used to check for sensitivity of results based on model specification.¹⁴ First, a binary variable coding for the foreign policy stance of the regime in the minor state—indicating whether the regime is anti-US—is used in the regression. This variable is collected through original coding, following the coding criteria used to code for opposition characteristics. This variable equals 1 when the regime in the minor state is leftist, socialist, or Communist ideologically, and equals 0 when rightist, conservative, republican, and centrist regimes are present.

Second, in order to account for the regime's stance towards the US, the ideological affinity of the minor state towards the US is also used in the regression. This variable is constructed as a similarity index between the minor state and the US through comparison of UN votes (Bailey et al., 2015) and it ranges from 0 (low affinity) to 1 (high affinity). This variable is expected to have a positive beta coefficient, as higher ideological affinity with the US could make cooperation more likely.

Third, the natural log of per capita GDP of the minor state is added to the empirical test in order to control for the varying resource levels of minor states. This variable comes from the Penn World Tables (Feenstra et al., 2015) and is expected to have a negative beta coefficient, as countries endowed with higher levels of wealth may not need foreign troops for maintaining their security. Additionally, the variable is also appropriate in this context since minor states such as Germany, Japan, and Norway went from requiring a substantial number of US troops on their territories in the initial time period to requiring relatively lower levels of US troops, owing to significant changes in their national wealth.

Additionally, the level of internal conflict that a minor state faces in a given year is also added to the test. This is the case because some political oppositions can be more aggressive than others—as in the case of civil wars in Cambodia, Laos, and Vietnam in the 1960s and 1970s—and need to be accounted for as such. Additionally, higher levels of threat are experienced by incumbent regimes in minor states that are in a state of civil war, and military deployments by the major power might be required to address the severity of the security situation. Therefore, a dummy variable equaling 1 for every year that a minor state is embroiled in civil war is added to the regression. This variable comes from the UCDP/PRIO Armed Conflict dataset (Gleditsch et al., 2002) and is expected to have a positive beta coefficient in the regression.

Moreover, a binary variable coding for whether a minor state is the rival of a state in the Soviet bloc is used. This variable is taken from the strategic rivalry data compiled by Colaresi et al. (2008) and is expected to have a positive beta coefficient, since a minor state's antagonistic attitude towards another state in the Soviet bloc would impact its likelihood of having US military personnel in higher numbers as compared to other minor states. This variable also helps to control for the effect of territorial contiguity towards the level of threat that a minor state faces, as strategic rivalries account for this factor by definition.

Furthermore, a variable for the economic importance of the US to the minor state is also included in the test, since the extant literature points to economic links being salient when it comes to the likelihood of establishing security exchanges or defense pacts between major powers and minor states (Fordham, 2010; Long, 2003). This variable is expected to have a positive beta coefficient, and it is constructed from the Gleditsch (2002) data on trade by adding the minor state's imports to and exports from the US (in real dollars) together and then dividing them by the minor state's GDP (in real dollars) for a given year. The resulting number provides an estimate of the relative economic importance of the US to the minor state, with the expectation that minor states perceiving the US’s economic importance to their domestic economy as more valuable would opt to station higher levels of US military personnel on their territory in comparison to minor states that do not perceive the US as a vital economic partner.

In addition to the above, the Polity Score of each minor state for every year in the data is added to the analysis by using the PolityIV data (Marshall et al., 2016); this variable ranges from −10 (autocracy) to 10 (democracy). Since democracies routinely allow for the existence of political oppositions and political participation, the fear of coercion and regime turnover is much less pertinent as compared to autocratic regimes for such minor states. This is the case because transparent information transmission mechanisms in democracies, combined with opportunities for expression, provide the public with avenues for outcry in case of coercive regime change, thereby increasing the costs of coercion for the major power. Hence, minor state leaders, cognizant of this structural characteristic, can allow for higher numbers of the major power's troops to be stationed on their territories; therefore, this variable is expected to have a positive beta coefficient.

Further, the model uses the natural log of the National Material Capabilities variable from the Singer et al. (1972) data. This variable is a proxy for a minor state's total military capabilities, with higher numbers denoting militarily stronger states.¹⁵ This variable is expected to have a negative beta coefficient, implying that more militarily capable minor states will have a lower demand for security from the US. Lastly, a binary variable indicating a formal defense pact with the United States is included, and is expected to have a positive beta coefficient. Table 1 provides a list of the variables used and their descriptive summary statistics.

Table 1.

Summary statistics.

Variables	Expected direction	N	Mean	St Dev.	Min	Max
Year		1492	1976	10.05	1950	1991
US troops (ln)		1492	3.937	2.821	0	12.45
Opposition type (centrist = 0, pro-US = 1, anti-US = 2)		1492	1.748	0.464	0	2
Regime stance (centrist/pro-US = 0, anti-US = 1)	−	1492	0.176	0.381	0	1
Ideological affinity w/ US	+	0.327	0.194	0	1
GDP/capita (ln)	−	1492	8.405	1.161	5.946	12.43
Civil conflict	+	1492	0.187	0.390	0	1
Soviet bloc rival	+	1492	0.272	0.445	0	1
US trade importance	+	1492	0.127	0.453	0	5.740
PolityIV score	+	1492	−0.839	7.777	−10	10
National material capabilities (ln)	−	1492	−6.437	1.471	−10.07	−2.835
Formal US ally	+	1492	0.131	0.337	0	1

Results and discussion

Table 2 contains the results of the empirical analysis; the dependent variable is the natural log of US troops stationed and the main independent variable is a factor variable that codes for the foreign policy stance of the opposition in the minor state (pro/anti/centrist) towards the United States. Models 2a and 3a have the same sample size, but Model 3a uses a binary indicator as an alternative independent variable that equals 1 when the opposition in the minor state is anti-US, and 0 when the opposition is either pro-US or centrist. Similarly, Models 2b and 3b share the same, smaller sample size; Model 2b excludes some states from the sample for which less reliable data is available and Model 3b uses this reduced sample with the alternative independent variable (from Model 3a). For Models 2b and 3b, all minor states for which coding regarding the stance of the opposition was done using secondary historical sources are excluded from the sample.¹⁶

Table 2.

Predicted troop levels given opposition stance (OLS).

Variables	Model 2a	Model 2b	Model 3a	Model 3b
Pro-US opposition	0.533**	0.542**
	(0.251)	(0.233)
Anti-US opposition	1.502***	1.555***
	(0.238)	(0.222)
Merged opposition (binary; anti-US = 1)			1.060***	1.105***
			(0.141)	(0.142)
Regime stance (anti-US = 1)	−0.129	−0.182	−0.043	−0.095
	(0.162)	(0.160)	(0.154)	(0.151)
Ideological affinity w/ US	2.052***	2.271***	2.073***	2.297***
	(0.242)	(0.252)	(0.241)	(0.251)
GDP/capita	0.033	0.049	0.032	0.046
	(0.046)	(0.058)	(0.046)	(0.058)
Civil conflict	−0.006	−0.140	0.009	−0.123
	(0.126)	(0.130)	(0.125)	(0.129)
Soviet Bloc rival	0.333***	0.377***	0.326***	0.366***
	(0.106)	(0.117)	(0.106)	(0.117)
US trade importance	0.928***	1.024***	0.931***	1.028***
	(0.100)	(0.111)	(0.100)	(0.110)
PolityIV score	0.036***	0.022***	0.036***	0.022***
	(0.007)	(0.007)	(0.007)	(0.007)
National material capabilities	0.915***	0.799***	0.910***	0.793***
	(0.040)	(0.044)	(0.039)	(0.044)
Formal US ally	3.029***	2.980***	3.042***	2.995***
	(0.154)	(0.157)	(0.153)	(0.156)
Constant	7.064***	6.259***	7.470***	6.681***
	(0.558)	(0.645)	(0.550)	(0.644)
Observations	1492	1237	1492	1237
R²	0.674	0.645	0.673	0.644

*** p 0.01, ** p 0.05, * p 0.1. Robust standard errors in parentheses.

The results are all noteworthy, and in particular we can see that the beta coefficient for anti-US opposition is positive and significant for all models, which was expected from the theory. This result implies that when the incumbent regime in a minor state faces an anti-US opposition, the level of cooperation extended to the major power is higher, based on the possibility that the incumbents fear less coercion in the security relation as the major power does not have viable alternatives to the regime. The base category for the opposition stance variable is centrist parties, for which no theoretical expectations are stated. The results stay the same even when alternative specification of the independent variable (Model 2b) is used.

To provide a more intuitive understanding of the substantive effects, Figure 1 plots the predicted levels of US troops stationed in a minor state (ln) over the values of the opposition stance variable. Panels (a) and (b) in Figure 1 demonstrate the predicted levels of US troops stationed in a minor state for centrist, pro-US, and anti-US opposition parties from Models 2a and 2b. As can be seen, the prediction for anti-US opposition parties is positive and significant, leading to the conclusion that such minor states fear less coercion in the security relation and are able to allow for higher security concessions to the major power (in the form of troops stationed).

Figure 1.

Predicted values of US troops stationed given opposition stances. (a) Model 2a, (b) Model 2b, (c) Model 3a, (d) Model 3b.

Models 3a and 3b—with an alternative independent variable—also provide support for the hypothesis, even though the sample for these models is different. Even with a reduced sample and alternative specification of the main independent variable, the beta coefficients for anti-US opposition are positive and significant, as expected. A more intuitive understanding of these results could be found in panels (c) and (d) of Figure 1, which plot the predicted values of US troops stationed over all values of the opposition stance variable from Models 3a and 3b. The panels clearly demonstrate evidence supporting the hypothesis, as the prediction for anti-US oppositions is higher, positive, and significant, whereas the predictions for pro-US parties are lower.

Time-dependent variation

It is possible that the number of US troops stationed in a minor state within a given year is dependent on time, i.e. troops present in period t − 1 dictate the number of troops in period t. This could especially be the case as stationing troops is a time-consuming endeavor, one that is likely to not vary greatly year on year. For this reason, assessing the empirical analysis with an eye on the time component for the US troops variable is important. One approach in this case would be to lag the dependent variable and include it in the regression as a covariate. However, this is likely to produce biased results as the lagged variable can soak up the variance to be explained by other variables. More importantly, however, including a lagged variable is a specific approach suggested in cases where the time-series shows signs of near-integration and not otherwise.¹⁷

A better approach instead is to treat the US troops variable as a time-series, and exclude its time-dependent component from the analysis. In other words, we can separate the variance that can be explained by time from the variance explained by other covariates in the regression. This approach, proposed by Kwiatkowski et al. (1992), among others, starts with a test to see whether the time-series has a unit root.¹⁸ To detect the presence of a unit root, the Augmented Dickey-Fuller (ADF) test (Hamilton, 1994) modified as the Fisher-type test for panel data (Choi, 2001) is used. The tests—performed in STATA—indicate that the time series of US troops stationed in minor states is stationary, i.e. it does not have a unit root. The absence of a unit root also allays any fears about the order of integration of the time-series with its t − 1 counterpart.

However, the presence of a trend in the time series is nonetheless detected. A trend is long-term change in the mean level of the time-series; in particular, the absence of a unit root in the series from the Fisher-type test indicates that the trend in the present time series is deterministic, i.e. linear, and we need to control for such variation. In order to do so, a regression is run with the time-series on the trend itself, i.e. a regression with the time index variable and a constant. In other words, this approach subtracts the trend from the time-series, thereby allowing us to use the remainder as the dependent variable. Put simply, a time index variable (number of years in this case) is regressed on the US troops variable, and the residual from this simple regression is used as the new dependent variable afterwards. This process eliminates any time-dependent characteristics emanating from the trend in a stationary time-series, and produces unbiased regression coefficients as the dependent variable is detrended, i.e. all variation attributed to a linear trend in the series is eliminated.

Model 4a in Table 3 makes use of this process. The US troops variable is regressed on the year variable, and the residual from this regression (US Troops_detrended) is used as the dependent variable. As the table shows, when the time-series is corrected for its trend, the evidence still provides support for the main hypothesis, as the beta coefficient for anti-US oppositions is significant and corresponds to higher values than the pro-US opposition variable. Lastly, Model 4b incorporates an alternative way to think about the time-dependency in the dependent variable, i.e. the inclusion of year fixed effects in the regression. The inclusion of fixed effects controls for unobserved heterogeneity when such heterogeneity is presumed constant over time. This treatment is accomplished by including dummy variables for all time periods in the regression on the right-hand side, and the beta coefficient estimates are subsequently calculated. As Model 3b demonstrates, the inclusion of fixed effects also does not change the support for the hypothesis. Figure 2 plots the results from Table 3.¹⁹

Figure 2.

Predicted values of US troops stationed for all values of opposition stance (time-based models). (a) Model 4a, (b) Model 4b, (c) Model 4a, (d) Model 4b.

Table 3.

Predicted troop levels given opposition stance (time variation).

Variables	Model 4a	Model 4b	Model 4c	Model 4d
Pro-US opposition	0.470*	0.178
	(0.269)	(0.265)
Anti-US opposition	1.504***	1.378***
	(0.256)	(0.249)
Merged opposition (binary; anti-US = 1)			1.116***	1.231***
			(0.144)	(0.144)
Regime stance	−0.073	0.221	0.003	0.251
	(0.165)	(0.165)	(0.157)	(0.154)
Ideological affinity w/ US	2.030***	4.286***	2.048***	4.302***
	(0.247)	(0.443)	(0.246)	(0.440)
GDP/capita	0.048	0.016	0.047	0.015
	(0.046)	(0.044)	(0.046)	(0.044)
Civil conflict	−0.013	0.007	−0.000	0.012
	(0.126)	(0.123)	(0.125)	(0.122)
Soviet Bloc rival	0.325***	0.428***	0.318***	0.426***
	(0.107)	(0.105)	(0.106)	(0.105)
US trade importance	0.953***	0.895***	0.956***	0.895***
	(0.105)	(0.106)	(0.105)	(0.106)
PolityIV score	0.034***	0.016**	0.034***	0.016**
	(0.007)	(0.007)	(0.007)	(0.007)
National material capabilities	0.900***	0.878***	0.895***	0.876***
	(0.041)	(0.041)	(0.040)	(0.040)
Formal US ally	3.027***	2.990***	3.039***	2.994***
	(0.157)	(0.157)	(0.156)	(0.156)
Constant	2.930***	2.032	3.287***	2.161
	(0.565)	(1.327)	(0.552)	(1.320)
Observations	1492	1492	1492	1492
Fixed effects		YES		YES
R²	0.666	0.686	0.666	0.686

*** p 0.01, ** p 0.05, * p 0.1. Robust standard errors in parentheses.

Conclusion

In conclusion, this article has presented a theoretical framework positing that coercion plays an important and distinct role in minor state decision-making with regards to asymmetric security relations. Through the examination of one particular concession, namely, allowing the stationing of a major power's troops on the minor state's soil during 1950–1991 in 65 minor states, this article has provided empirical evidence to support its central claim: the possibility of coercion looms large in the minds of minor state leaders when extending security concessions to major powers. In this context, it can be seen that, first, the decision-making processes that lead minor states to enter security relations with major powers are distinct from the ones undertaken by the leaders of major powers. Second, coercion plays a critical role in this regard: this article contends that regimes in minor states decide on their policy choices in conjunction with the character of political oppositions they face at home, as doing so is inherently tied to the potential for coercion in asymmetric security relations. Not only do the theory and empirical evidence presented here point to a need for understanding foreign policies of states through the lens of their internal politics, but they also contribute towards research on asymmetric security relations through incorporating a role for coercion and minor state agency in the larger academic discussion as well.

Furthermore, it must be noted that while the analysis pertaining to the stationing of US troops in minor states is an easily-observed proxy for the underlying theoretical processes, several other phenomena may also pose the same risks for minor states. Additionally, the theory presented here can be enhanced by introducing strategic interactions between the incumbent regime and the opposition in the minor state, in addition to the minor state and the major power.²⁰ In the same vein, further refinements to the theory can elaborate on possibilities where major powers can seek to side with more amenable oppositions even in the case of pro-major power regimes, as well as incorporating scenarios where incumbent regimes in minor states can also leverage the presence of an anti-US opposition to their advantage. Moreover, future projects within the theoretical framework presented here can also theorize about any changes that may take place within the character of the political opposition when a major power intervenes in the domestic politics of a minor state. Lastly, as this article has theorized about the fear of coercion, investigating when coercion actually takes place by the major power is also a useful extension of the present argument. At the current stage, this article posits about possible pathways only from the perspective of the incumbent regime in the minor state regarding cooperation under the shadow of coercion, but it is important to note that other perspectives, theoretical models, and empirical tests can also be developed from this line of reasoning. In conjunction with the theory and empirical evidence presented above, it is evident that the above strand of research serves as an encouraging avenue for further research.

Supplemental Material

sj-pdf-1-acp-10.1177_20578911221076222 - Supplemental material for Difficult decisions: Coercion in asymmetric security cooperation

Supplemental material, sj-pdf-1-acp-10.1177_20578911221076222 for Difficult decisions: Coercion in asymmetric security cooperation by Syed Rashid Munir in Asian Journal of Comparative Politics

Footnotes

Acknowlegments

The author is grateful to Benjamin O. Fordham for providing invaluable advice on earlier versions of this manuscript. The author would also like to thank Hayley Munir, Ekrem Karakoc, David Clark, and other members of the Political Science department at Binghamton University (SUNY) for their feedback.

Declaration of conflicting interests

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

Funding

The author(s) received no financial support for the research, authorship and/or publication of this article.

ORCID iD

Syed Rashid Munir

Supplemental material

Supplemental material for this article is available online.

Notes

References

Altfeld

(1984) The decision to ally: A theory and test. Western Political Quarterly 37(4): 523–544.

Bailey

Strezhnev

Voeten

(2015) Estimating dynamic state preferences from United Nations voting data. Journal of Conflict Resolution 61(2): 430–456.

Baum

Potter

PBK

(2015) War and Democratic Constraint. Princeton, NJ: Princeton University Press.

Beckley

(2015) The myth of entangling alliances: Reassessing the security risks of U.S. defense pacts. International Organization 39(4): 7–48.

Biglaiser

DeRouen

Jr. (2007) Following the flag: Troop deployment and U.S. foreign direct investment. International Studies Quarterly 51(4): 835–854.

Biglaiser

DeRouen

Jr. (2009) The interdependence of U.S. troop deployments and trade in the developing world. Foreign Policy Analysis 5: 247–263.

Bueno de Mesquita

Smith

Siverson

, et al. (2003) The Logic of Political Survival. Cambridge, MA: MIT Press.

Butt

(2013) Anarchy and hierarchy in international relations: Examining South America’s war-prone decade, 1932–41. International Organization 67: 575–607.

Choi

(2001) Unit root tests for panel data. Journal of International Money and Finance 20: 249–272.

10.

Cima

and Library of Congress Federal Research Division (1989) Vietnam: A Country Study. Washington, DC: Federal Research Division, Library of Congress. Available at: https://www.loc.gov/item/88600482/ (accessed 20 May 2020).

11.

Colaresi

Rasler

Thompson

(2008) Strategic Rivalries in World Politics: Position, Space and Conflict Escalation. Cambridge, MA: Cambridge University Press.

12.

Cooley

Nexon

(2013) “The empire will compensate you”: The structural dynamics of the U.S. overseas basing network. Perspectives on Politics 11(4): 1034–1050.

13.

Diederich

(2000) Trujillo: The Death of the Dictator. Princeton, NJ: Markus Wiener Publishers.

14.

Fearon

(1995) Rationalist explanations for war. International Organization 49: 379–414.

15.

Feenstra

Inklaar

Timmer

(2015) The next generation of the Penn World Table. American Economic Review 105(10): 3150–3182.

16.

Fordham

(2010) Trade and asymmetric alliances. Journal of Peace Research 47(6): 685–696.

17.

Fordham

(2019) The domestic politics of world power: Explaining debates over the United States battleship fleet, 1890–91. International Organization 73: 435–468.

18.

Gavin

(2015) Strategies of inhibition: U.S. Grand strategy, the nuclear revolution, and nonproliferation. International Security 40: 9–46.

19.

Gerzhoy

(2015) Alliance coercion and nuclear restraint: How the United States thwarted West Germany’s nuclear ambitions. International Security 39: 91–129.

20.

Gleditsch

(2002) Expanded trade and GDP data. Journal of Conflict Resolution 46: 712–724.

21.

Gleditsch

Eriksson

Sollenberg

, et al. (2002) Armed conflict 1946–2001: A new dataset. Journal of Peace Research 39(5): 615–637.

22.

Gowa

Mansfield

(1993) Power politics and international trade. American Political Science Review 87(2): 408–420.

23.

Grow

(2008) U.S. Presidents and Latin American Interventions: Pursuing Regime Change in the Cold War. Lawrence, KS: University of Kansas Press.

24.

Hamilton

(1994) Time Series Analysis. Princeton, NJ: Princeton University Press.

25.

Harrison

(2005) Driving the Soviets up the Wall: Soviet-East German Relations, 1953–1961. Princeton, NJ: Princeton University Press.

26.

Hey

JAK

(ed.) (2003) Small States in World Politics: Explaining Foreign Policy Behavior. Boulder, CO: Lynne Rienner.

27.

Ikenberry

Kupchan

(1990) Socialization and hegemonic power. International Organization 44(3): 283–315.

28.

Jackson

(2015) The showdown that wasn’t: U.S.-Israeli relations and American domestic politics, 1973–75. International Security 39(4): 130–169.

29.

Jones

(2003) Death of a Generation: How the Assassinations of Diem and JFK Prolonged the Vietnam War. New York, NY: Oxford University Press.

30.

Kane

(2004) Global U.S. Deployments, 1950–2013. Washington, DC: The Heritage Foundation. https://www.heritage.org/defense/report/global-us-troop-deployment-1950-2005 (accessed 19 May 2021).

31.

Keohane

(1971) The big influence of small allies. Foreign Policy 2: 161–182.

32.

Kim

(2011) Why alliances entangle but seldom entrap states. Security Studies 20(1): 350–377.

33.

Kim

(1989) Power, alliance, and major wars, 1816–1975. Journal of Conflict Resolution 33(2): 255–273.

34.

Krasner

(1999) Sovereignty: Organized Hypocrisy. Princeton, NJ: Princeton University Press.

35.

Krause

David Singer

(2001) Minor powers, alliances, and armed conflict: Some preliminary patterns. In: Reiter

Gartner

(eds) Small States and Alliances. Heidelberg: Physica, 15–23.

36.

Kwiatkowski

Phillips

PCB

Schmidt

, et al. (1992) Testing the null hypothesis of stationarity against the alternative of a unit root. Journal of Econometrics 54: 159–178.

37.

Lake

(1996) Anarchy, hierarchy, and the variety of international relations. International Organization 50(1): 1–33.

38.

Lake

(2008) The new American empire? International Studies Perspectives 9: 281–289.

39.

Lake

(2013) Legitimating power: The domestic politics of U.S. international hierarchy. International Security 38(2): 74–111.

40.

Leeds

Ritter

McLaughlin Mitchell

, et al. (2002) Alliance treaty obligations and provisions, 1815–1944. International Interactions 28(3): 237–260.

41.

Little

Leblang

(2004) Military securities: financial flows and the deployment of U.S. troops. In: Presented at the Annual Meeting of the American Political Science Association.

42.

Long

(2003) Defense pacts and international trade. Journal of Peace Research 40(5): 537–552.

43.

Marshall

Gurr

Jaggers

(2016) Polity IV Project: Political Regime Characteristics and Transition, 1800–2015. College Park, MD: Center for Systemic Peace. Available at: http://www.systemicpeace.org/inscrdata.html (accessed 10 Jan 2021).

44.

Mattes

(2012) Reputation, symmetry, and alliance design. International Organization 66(4): 679–707.

45.

McDonald

(2015) Great powers, hierarchy, and endogenous regimes: Rethinking the domestic causes of peace. International Organization 69(2): 557–588.

46.

Morgan

Palmer

(2003) To protect and to serve: Alliances and foreign policy portfolios. Journal of Conflict Resolution 47(2): 180–203.

47.

Morrow

(1991) Alliances and asymmetry: An alternative to the capability aggregation model of alliances. American Journal of Political Science 35(4): 904–933.

48.

Mousseau

(1997) Democracy and militarized interstate collaboration. Journal of Peace Research 34(1): 73–87.

49.

O’Rourke

(2013) Secrecy and security: U.S.-orchestrated regime change during the Cold War . PhD thesis, University of Chicago, USA.

50.

O’Rourke

(2018) Covert Regime Change: America’s Secret Cold War. Ithaca, NY: Cornell University Press.

51.

Park

(1975) The influence of small states upon the superpowers: United States-South Korean relations as a case study, 1950–53. World Politics 28(1): 97–117.

52.

Powell

(1999) In the Shadow of Power: States and Strategies in International Politics. Princeton, NJ: Princeton University Press.

53.

Putnam

(1988) Diplomacy and domestic politics: The logic of two-level games. International Organization 42: 427–460.

54.

Scartascini

Cruz

Keefer

(2017) The database of political institutions 2017. Available at: https://publications.iadb.org/handle/11319/8806 (accessed 15 Sep 2020).

55.

Singer

Bremer

Stuckey

(1972) Capability distribution, uncertainty, and major power war, 1820–1965. In: Russett

(ed) Peace, War, and Numbers. Beverly Hills, CA: Sage, 19–48.

56.

Volkens

Lehmann

Matthieß

, et al. (2018) The manifesto data collection. Manifesto project (MRG/ CMP/MARPOR). Version 2018a. Wissenschaftszentrum Berlin für Sozialforschung. https://doi.org/10.25522/manifesto.mpds.2018a.

57.

Waltz

(1979) Theory of International Politics. Reading, MA: Addison-Webley.

58.

Womack

(2016) Asymmetry and International Relationships. Cambridge, MA: Cambridge University Press.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.15 MB