Arms transfers and international relations theory: Situating military aircraft sales in the broader IR context

Abstract

Quantitative research on arms transfers has not adequately accounted for broader theories of international relations. We review the specialized literature on arms transfers and examine how arms transfers fit with the broad international relations theories. We derive and test seven hypotheses based on realist, liberal and constructivist theories using a dataset of all non-US/Russian aircraft transfers between 1990 and 2010. We find limited support for realist hypotheses. We find little support for hypotheses derived from the Democratic Peace literature, but some support for liberal trade arguments. We also find some support for constructivist arguments based on shared identity and prestige measures.

Keywords

Arms theory arms transfers geography International Relations Theory realism

Do transfers of coercive capabilities between states, especially military aircraft, conform to broader theoretical explanations of international relations (IR), or are these transfers somehow unique? The study of war is central to many of our theories about both the causes and consequences of world politics, but the willing transfer of capabilities to wage war has been steadily understudied by IR research. The acquisition of arms by states to increase their military power is taken for granted in major theories of IR. Yet the theoretical mechanisms for states to acquire military equipment from outside states have not been adequately studied by security scholars. Internal and external balancing are long-recognized mechanisms for states to protect themselves against threats from other states. However, most states lack the capacity to manufacture advanced weapons systems on their own. This works against the theoretical construct that the international system is “self-help” in any meaningful sense, since the bulk of states need to seek alliances or arms from abroad in order to maintain their security.

Recent scholarship has begun to address this problem. Vucetic and Tago (2015) examined how the transfer of fighter aircraft fits into different theories of IR. They tested fighter aircraft purchases using conceptions of power and hierarchy by Lake (2009), and found that purchasing is closely related to measures of military and economic dependence on the USA. Spindel (2018: 3) argued that arms transfers matter because they serve as a credible signal about relationships between states—even when those transfers do not meaningfully affect the balance of power. Caverley and Kapstein (2012) argued the strategic wisdom of the USA catering more to the export market in order to expand and cement influence with different states.

This scholarship represents an important step toward linking arms sales to other IR problems. We seek to contribute to the understanding of arms IR by testing hypotheses generated by major theories of IR using aircraft transfers. We focus on aircraft since they are primarily used for interstate conflict and give the possessor power-projection capabilities. Aircraft have long lifespans of up to 30 years with proper support and maintenance (Johnson, 2019), so there is an opportunity for importers to export their older models when they upgrade their fleet.

Our tests allow us either to begin to integrate this research with mainstream arguments of IR—a difficult prospect in the past—or to begin eliminating potential explanations as we seek to find more general theories of arms transfers. We further limit these sales to states that are not the USA and the USSR/Russia. Examination of such mid-tier states comes with some trade-offs. The primary cost from our sample choice is that the USA and Russia account for the majority of the global arms trade historically. However, mid-tier producers comprise a sizable portion of the trade. These suppliers are understudied compared with the superpowers, despite comprising a sizeable portion of the market. They have complicated motivations and constraints in the arms trade that deserve their own study.

We focus on two gaps in the literature as identified in some of the most recent comprehensive work on arms transfers. The first is the general lack of theorizing about arms transfers and their effect on IR (Spindel, 2018: 15; see also Erickson, 2015: 31). The second is that specialized arms research often uses large-n research design to test hypotheses, something that the literature on arms transfers that deals with broader IR theories has consistently failed to do well (Spindel, 2018: 31). We fill in these two gaps by clearly identifying the expectations of each major IR paradigm separately and by conducting quantitative tests that parse out how well the theoretical expectations perform empirically. Quantitative methods are necessary since we examine the 24 mid-tier state aircraft producers.

We first address the literature on arms transfers and identify areas where our understanding of arms transfers could be informed by theories of IR. Next, we theorize about how arms-exporting states may take security, economics and identity into consideration when exporting offensive weapons. We test the hypotheses generated from this theory using ordinary least squares (OLS) and logistic regression models with variables to account for the different theoretical explanations. We find that trade, rather than democracy, is the best “liberal” predictor of arms transfers. We observe some minor support for realist expectations in aircraft transfers—states take into account region and power differentials before transferring aircraft. We also find mixed support for our constructivist hypotheses for arms transfers.

What we know about arms sales

Arms scholars have thought about the capabilities of arms producers in terms of states’ ability to use transfers to affect change in recipient states (Catrina, 1998; Kinsella, 1998; SIPRI, 1975). Recently, security scholars have used case studies of arms transfers to describe different relationships between major power states and their clients or protégés. Yarhi-Milo et al. (2016) frame their discussion of arms transfers by the USA to Israel and Taiwan under the logic of the “Patron’s Dilemma” outlined by Snyder (1997). This dilemma is unique to a particular type of powerful state in the international system and is not generalizable to most state relations.

While important, work that focuses only on the exceptionally powerful states in the system has limited explanatory power for a large portion of arms transfers. Such work also largely ignores the growing corpus of work that has systematically examined both the economic and security factors considered by exporting and importing states (e.g. Brauer, 2003; Hartley and Martin, 2003; Krause, 2004; Levine et al., 2003). The literature on arms exports has its own theoretical shortcoming: a focus on state-level economic factors in exporting states to explain arms transfers. Work which ignores the international context and the realities of power and coercion of both the arms transferred and the actors involved in those transfers can provide only partial explanations of arms transfers (Akerman and Seim, 2014; Comola, 2012; Qingmin, 2006).

Arms sales are treated as an economic problem by many studies in the empirical arms literature (e.g. Akerman and Seim, 2014; Comola, 2012). These studies, grounded in the economics and International Political Economy literature, focus on the domestic factors that incentivize states to export arms. However, the strict regulation of arms sales by most exporting states, and the involvement of the state in their sale even in states that are non-restrictive and non-discriminating, like Russia, mean that arms are treated differently than other economic goods (Willardson, 2013). Additionally, realist theories that discuss state security typically attribute state power to internal factors or to alliances. Arms are not included in the elements of power in the most well-known measure of state power (Singer et al., 1972), even though military composition is a key element of immediate threat to states (Gray, 1993).

An early study of arms transfers classified supplier states into three types in relation to their exports to Third World countries that were incapable of supporting weapons development and production on their own. The three types of suppliers were hegemonic, using arms to influence the behavior of recipient states; industrial, using arms to maintain domestic production in the exporting state; and restrictive, supplying weapons, but careful not to become involved in local or international conflict (SIPRI, 1975: 21–26).

A slightly different classification looked at arms production capabilities from an economic and technical perspective. This was necessary as major weapon systems—aircraft, air defense systems, armored vehicles, artillery, missiles, naval weapons, and ships based on classifications by the Stockholm International Peace Research Institute (SIPRI)—require varying levels of technical sophistication that impact R&D costs and support/maintenance costs. These variations occur across and within categories (e.g. aircraft vs. armored vehicles and trainer vs. fighter aircraft) (Johnson, 2017). A linked issue is the interaction between R&D costs and lifetime of advanced weapon systems—producers are continually creating newer models of major weapon systems despite some systems, such as fighter aircraft, having a lifespan of up to 30 years when properly maintained (Johnson, 2019).

Under these conditions, only a few states were able to produce advanced weapons and sustain domestic production without aggressive export policies; some states could produce sophisticated weapons, but only if they also aggressively exported weapons to support domestic industry; and some states were simply incapable of producing sophisticated weapons (Krause, 1992; Smith et al., 1985). Arms transfer scholars using these classifications have concluded that states export weapons for power, influence and/or economic reasons (Akerman and Seim, 2014; Blanton, 2005; Comola, 2012; Erickson, 2015; Keller and Nolan, 1997; Platte and Leuffen, 2016; Stohl and Grillot, 2009).

Only the USSR and the USA fell into the top category because those states were capable of producing advanced weapons systems across the entire military spectrum. They had the economic and military capacity to absorb the costs associated with large amounts of military spending—especially R&D. More recently, Willardson (2013) showed that neither the USA nor Russia are fully capable of ignoring the need for exports to maintain domestic production. For the USA (and Russia to a lesser degree), economics is not the driving factor behind all arms sales decisions, but simply one of many domestic and foreign policy pressures.

It is the middle-tier exporter states that find themselves in an interesting position. Production capability hinges on a few distinct competencies and two stand out. First is the technical capability of producing extremely sophisticated military equipment. Second is the economic ability to turn technical ability into production capability without damaging the overall economy. The second dominates the foreign policy of arms sales—the imperative to increase the market for weapons systems in order to offset the expense of R&D and production (Krause, 1992: 141). States need to disseminate coercive technology in order to keep the capability of producing it for themselves (Kapstein, 1992; Kinsella, 2002; Pearson, 1994; Sanjian, 1991). This pressure can potentially reduce the role of strategy and politics in arms-export decisions (Smith et al., 1985).

Paradoxically, most economic benefit arguments for producing and exporting weapons are specious. Arguments about the strategic importance of arms sales may hold true, but economic arguments have fallen short. The balance of payments has been a primary justification for exports (Eikenberry, 1995; Keller and Nolan, 1997; Krause, 1992). When the UK began actively increasing exports in 1965, balance of payments was the primary consideration (Ministry of Defence, 1965). Later research showed a one-third reduction in British exports would only decrease the balance of payments deficit by 0.5% (Hartley, 2000). The Ministers of the RAF and Royal Navy also noted that commitments to arms exports would have the “effect on British industry of tying up capital and skilled labour in armaments” (Ministry of Defence, 1965: 201).

The arms literature covers the supply side (Blanton, 1999, 2005; Erickson, 2011, 2015; De Soysa and Midford, 2012), demand side (Kinsella, 2002), and the effect of transfers (Johnson and Willardson, 2018; Kinsella, 1994; Kinsella and Tilemma, 1995; Mintz and Heo, 2014). There are few attempts to apply IR theory to large-n empirical studies of arms transfers with the exception of some very recent studies (Vucetic and Tago, 2015). This lack of a theoretical link between IR and quantitative studies of arms is surprising given the growing trend of quantitative studies in the discipline. One explanation is an over-focus on the superpowers when applications of IR theory are made. With two states, qualitative case studies and historical research can make a valuable contribution, but with around two dozen mid-tier suppliers, quantitative tests are necessary.

Arms transfers are by definition an exchange of power, and power is the driving force in large swaths of IR theory. Structural (Waltz, 1979) and power transition (Kadera, 2001; Organski and Kugler, 1989) explanations of IR focus on explanations of long-term state power. These theories describe the behavior of the most powerful states that shape the international system. Gilpin (1988: 612) noted that states consciously decide to go to war, while seeking to avoid system-altering war. This difference suggests that for most states the primary security concern is about the short-term capabilities of other states.

Arms sales epitomize immediate capability shifts as finished weapon systems are transferred.¹ One manifestation of arms transfers is arms races between states (Buzan and Herring, 1998). They represent immediate capability balancing. States that find themselves in competition monitor and match their own arms imports in order to not be caught off balance, leading to a continuing cycle of uncertainty about the balance of immediate capabilities. A fundamental difficulty in tying arms sales to balancing behavior is that most balancing literature focuses on long-term capabilities.

Air power

Air power has been a key component of national military power since its effectiveness and use were demonstrated decisively in World War II. The extent to which air power can stand alone in terms of military strategy is debated by military and security experts. Air power holds an important place in studies of international security (Pape, 1990, 1996) and military history (Black, 2016), and in the public mind. The debate about military air power is not primarily about the extent of its uses.

The importance of air power in the public mind (see Vick, 2015), and more importantly, in civilian and military leaders’ minds (Byman and Waxman, 2000), leads to an important puzzle regarding international arms transfers. Why would states transfer a weapon that can be used coercively against it? Why put powerful weapons in the hands of friends, if those friends could one day be enemies? Why put them in the hands of those states that are not currently friends or allies? This puzzle is one that has not been adequately addressed in the literature on the arms market. The puzzle deepens when weapons are undeniably offensive.

Aircraft are offensive weapons (Garthoff, 1980). Aircraft provide greater power projection capabilities than land weapon systems. There are limitations in power projection beyond the short-term without backup from land and sea forces. Troops placed by aircraft can be supported by aircraft. As aircraft are offensive in nature, they provide a good test of realist assumptions about balancing against both power and threat. Neorealist theory would assume that states should not allow the exchange of powerful weapons at all. A more nuanced reading, based on the ideas of either hegemony, or theories of offense/defense, may allow for the exchange of major weapons systems when they are clearly defensive in nature. Deterrence theory supports the idea that great power states transfer weapons to strengthen ties between the defender and its protégé (Huth and Russett, 1984: 523).

Aircraft transfers may also grant the more powerful state leverage (Sislin, 1994) and economic advantages. Aircraft transfers can make alliance partners more capable, and therefore increase the strength of the alliance. Such arguments may account for why aircraft transfers from mid-tier states may be expected. They do not account for the biggest issue, however, which is the change in short-term balance of capabilities even if these capabilities are naturally decaying over time.

A hegemonic power or exporter may not worry about transfers of offensive weapons that do not significantly challenge its military dominance, and which may help allies. Hegemonic extended deterrence is a possible explanation for arms transfers to its allies. Security hierarchies are another explanation for the transfer of offensive arms (see Lake, 2009; Vucetic and Tago, 2015). Mid-tier states² are capable of producing advanced weapons, but are not capable of supporting the development and production of those weapons solely for the domestic market. Such states are usually not in a position to be the defender as defined in the extended deterrence literature. Some of the mid-tier state arms producers (France, UK, Germany, China) are great powers that should behave in ways consistent with realist theories.

Theoretical explanations of arms transfers

Arms races are the most common exploration of arms transfers in mainstream IR research. The security dilemma can lead to a long-term cycle of defense spending and acquisition for the states involved. The acquisition of new weapons systems through production or purchase leads to a stronger sense of insecurity in rivals, triggering additional arms production or purchases. Arms races have been explored in relationship to conflict (Diehl and Crescenzi, 1998; Sample, 2018; Wallace, 1979), deterrence (Gibler et al., 2005), costs (Kydd, 2000), and interstate relations and foreign policy (Klare, 1993). Some scholars have questioned the usefulness of the concept of arms races (Gray, 1996).

Arms races serve as a bridge between the realist concepts of power and security in IR and arms, which are tools used to provide that security. Gray (1993) noted that weapons do not make war; states and their leaders do. They do so based on their strategy, interests, and assessment of the intentions and capabilities of other states. Even if weapons themselves are not the proximal cause of war, it is difficult to imagine civilian and military leaders ignoring the arsenal of their potential adversaries.

Previous work has outlined an economic logic for arms transfers. By examining the logic of realism, liberalism, and constructivism and testing whether that logic applies in the face of strong evidence from previous studies that economics is the primary driver, we have a strong test for determining which theory/argument drives arms transfers. Realism allows us to examine power and security motives, while liberalism’s identification of factors leading to pacific relationships may also lead to the provision of arms. Constructivism allows for examination of the role that ideas, images, and prestige may play in import decision making.

Realist explanations

A key logic of realism is that states exist in an anarchical environment in which they must make decisions and survive (Bull, 1979; Waltz, 1979; Wendt, 1999). This environment incentivizes states to focus first on their own survival. The power of the state is measured against other states, because other states pose the primary external threat to survival. Power in structural realism is measured in the long-term capabilities of states to generate war-making materials and populations. The short-term balance of power between states, however, must take into account the present military power of both states (Diehl and Crescenzi, 1998: 113). Military power comes both from arms and alliances (Chatterjee, 1975; Colaresi and Thompson, 2005; Morrow, 1993; Yarhi-Milo et al., 2016); however, traditional capability measures do not examine alliances. The effectiveness of alliances is also conditioned on the text of the treaty (Leeds et al., 2002) and so it is important to take treaty language seriously when using treaties to measure potential balancing behavior via defense pacts.

States are sovereign only if they can reasonably dictate what happens within their own borders and protect themselves from violent incursions and the dictates of other states (Rotberg, 2004). Sovereignty, security, and power are interwoven and are often vague and slippery terms (Biersteker, 2002) when discussing the actual actions and interactions of states. We treat states as sovereign actors who seek security through the accumulation of power through internal and external balancing (Morrow, 1993).

The purchase of arms helps a state internally balance by increasing the capability of the state rather than externally balancing by committing to, or shifting alliance ties to other states (Lieber and Alexander, 2005: 119). Acquisition of weapons is a signal to potential partners that the state is a capable ally. The capability of allies has been used as a measure of alliance reliability (Holmes, 1988: 173). States increasing military capabilities signal both their own capabilities and their potential to be a capable ally to other states in the international system. Spindel (2018) argues that arms transfers are a signal about a state’s status, and so alliances can be strengthened and their status more deeply understood through an examination of arms transfers.

Balancing against other states’ power through alliances is the primary mechanism in the neorealist conception of IR. Other realists disagree about the mechanics of Waltz’s (1979) balancing and order, but agree with the basic tenets of realism. Schweller (1997: 928) characterizes realism as a theory, or philosophy, that IR are governed by group competition under conditions of scarcity. In his conception, competition is not about the importance of absolute advantages, but about the relative advantage that states hold in discrete contests of power.³ The military balance (Holmes, 1988) is a measure of the present ability of those states to compete in a discrete contest for power.

The provision of arms from one state to another represents an immediate enhancement of the recipient state’s coercive capacity. We do not expect that any state would provide another state with such coercive capabilities, preferring instead to balance with alliances. This naïve assumption deals with the issue of internal balancing by building weapons and externally by forming alliances.

Alliances are useful because they can be changed when the underlying or immediate power capabilities of states change. Coercive capabilities cannot be changed quickly unless they are destroyed. War is the primary method of destroying other states’ capabilities. Logically, the transfer of weapons should be avoided by status quo states because such a provision could lead to the need for future conflict if the intentions of recipient states change. We have two caveats to this explanation. The first is recipient dependency on exporters (Johnson, 2019). Combat aircraft require continued support from the exporter to remain air-worthy, giving the exporter some leverage after transfer. However, recipients may be willing to shoulder the costs involved with finding a new supplier (Rounds, 2019). The second caveat is that arms become relatively less capable over time, so any single arms transfer can be considered threatening only temporarily.

There is some evidence that states may prefer a strategy arming allies. The US “Nixon Doctrine” explicitly pledged to provide economic and military aid to states while insisting those states provide their own manpower to defend themselves (see Thies, 2013: 280).⁴ This is also the logic of the patron–client dilemma (Yarhi-Milo et al., 2016). Such logic may not apply to exporting states without overwhelming advantages in power over their import partners. The logic is much clearer for the two Cold War superpowers. We do allow for the potential that alliances may increase the probability of arms transfers. Even so, we expect that such transfers would favor defensive over offensive weapons. Since alliances are a key component in both realism and liberalism, we condition our hypotheses by stating that any effect should be increased with the presence of a defense pact.

The patron–client relationship is one where balancing via arms transfers is seen as better for the exporting state. In this case there is a clear power imbalance between supplier and recipient. In both examples by Yarhi-Milo et al. (2016), the recipients were also located far away from the supplying state. Our naïve assumption that states will not transfer weapons is tempered by the logic of perceived threat. The greater the power imbalance and the greater the distance, the less the perceived threat, and the greater likelihood that arms will be transferred.

The most-used measure of state capabilities focuses on the long-term power potential of the state via an index created from six key factors (Singer and Small, 1966). The balance of power is inherently important for military decision making (Rosenberg, 1990: 290) and the Composite Index of National Capabilities (CINC) measure (Singer and Small, 1966) includes factors representing the long-term military balance in contrast to the immediate military balance. Over the short term, the military capability of potential adversaries—states with whom the exporter does not have pacific relationships—is more important for decision-making. The present shift in actual military balance is the more salient measure for understanding the actions of states than is the potential for accumulating such capabilities (Lobell, 2018). Balance of power considerations damage the logic of selling arms to potential adversaries.

H1: Mid-tier states will be more likely to supply aircraft only to states that have lower material capabilities.

This hypothesis relies on the logic that states may operate under rational time horizons (Toft, 2006). If a recipient state is significantly less capable than the supplying state—and this is a condition that should be somewhat static—a supplier state might take the calculated risk of providing that state with arms. While such action potentially weakens strong realist assumptions, it is a reasonable exception for states to carve out—especially for states located in a relatively peaceful international environment facing strong domestic incentives to export arms.

H2: Mid-tier states will be more likely to supply aircraft to states with similar or higher overall capabilities if those states are outside of the supplier state’s geographic region.

This assumes that states act rationally in terms of balancing threats. Most threats come from states that are nearby. This is true of great power states as well as mid-tier arms producers. The logic of this hypothesis comes from in an opportunity/willingness (Most and Starr, 1989) framework, or in work on the geographic determinants of conflict (Vasquez, 1995).

We expect that Hypothesis 1 may be conditioned by geographical proximity. States will be more likely to transfer out of the region and to transfer to states that are less powerful. We also believe that states will be less likely to transfer weapons to more powerful states outside their region if those weapons could potentially be used against them. More powerful states have greater power projection capabilities that may negate the effects of proximity that we account for with the region measure. To examine Hypothesis 2, we use an interaction term of capabilities and geographic region.

Liberal explanations

Liberalism shares the same underlying assumptions of an anarchic international system and state as the primary actor in IR as realism. Liberal arguments, however, have a comparatively larger number of quantitative tests providing support for their assertions. Democracies rarely go to war against each other owing to the structural and normative features of these regimes. Maoz and Russett (1993) found stronger support that normative factors work because democratic states look like each other and trust the openness and intentions of other similar states. Structural features of democracies prevent escalation through democratic states’ domestic veto players and public discussion. Regardless of which set of arguments is more valid, both explanations are directly applicable to arms transfers.

Democratic peace arguments revolve around trust and openness. States with similar norms and values where information is available are less likely to escalate disagreements to levels where military force results in the loss of life. This applies to arms transfers directly: if democracies do not engage in high-level militarized conflict with one another, then they can be trusted with weapons providing coercive capability. Providing coercive capabilities creates opportunities to an importer that were not present before, which is different than just not fighting. Structural arguments suggest that transferring arms to other democracies is comparatively safe since both states have the relevant understanding of the other’s capability and resolve, leaving greater room for negotiation and non-violent outcomes.

H3: Mid-tier states will be more likely to supply aircraft to states if they share a democratic regime type.

The other primary relationship in the liberal peace argument is trade. Hirschman (1980: 17) provides a clear rationale for democracies to supply one another with coercive capabilities:

states may trade weapons with each other in order to protect valuable shared trade and economic linkages.

Liberal peace arguments apply where higher trade creates more losses if conflict occurs; thus, trade partners can be trusted more with weapons. Additionally, states that have increased trade will be likely to trade across the full range of products, including arms. However, trade and economic incentives are the most used explanation in the arms literature. Arms transfers typically occur with trade offsets (Brauer and Dunne, 2004).

H4: Mid-tier states will be more likely to supply aircraft to states as overall trade increases.

The final relationship from the liberal peace arguments is joint intergovernmental organization (IGO) membership. States that share a greater number of IGO memberships have greater overlapping interests and contact, which results in lower levels of international conflict with each other.

H5: Mid-tier states will be more likely to supply aircraft to states as joint IGO overlap increases.

Constructivist explanations

Constructivist theory somewhat overlaps with liberal theory, but with different underlying assumptions. Constructivism assumes that states are the main unit of analysis, the system is intersubjective (as opposed to materialist in rationalist theories), and state identity and interests are in part determined by the structure of the system (Zehfuss, 2002). Boiled down, constructivism asserts that ideas, culture, and identity matter (Jepperson et al., 1996); however, differentiating between liberalism and constructivism empirically with quantitative tests is difficult (e.g. Maoz, 2006).

Vucetic and Tago (2015: 102) tie arms sales to constructivist thought by noting that “arms are manifestations of socialization, recognition, identity relations, and status-seeking.”Vucetic and Tago (2015) focus on hierarchy in IR as their independent variable. Their variable is derived from Lake’s work, and focuses on security and economic reliance on the USA to establish hierarchy.

To unpack constructivism and arms imports we link how importer culture, identity, and ideas are impacted by the possession of arms and their relationships with other states. Colonial relationships are linked with constructivism. Many mid-tier major weapon exporters were colonial powers and had a cultural influence in their possessions. The successor society was socialized by the former possessor and they share similar values. One way this is done is by preserving military links between the two states, which can be formalized with arms transfers. For example, comparison of order and delivery dates in SIPRI’s arms trade register with state formation dates from the Polity IV Project shows that many newly independent states ordered major weapon systems from their former colonial power in the year(s) before independence occurred. Another way is to provide officer training to former colonies. For example, the UK’s arms relationship with Jordan was directly linked to providing officer training and the UK saw a threat in Jordanian officers beginning to be trained in the USA (Foreign & Commonwealth Office, 1967: 57). Thus, a shared military culture is linked by colonial relationships in multiple ways.

H6: Mid-tier states will be more likely to supply aircraft to states if they are a former colonial possession.

How does a former colonial relationship relate to arms beyond a shared military culture linked with security concerns? First, we consider how identity—specifically national identity—links with arms imports. The importance of national identity explains why states import weapons they do not need for security (Wezeman, 2010). Suchman and Eyre (1992) compare the symbolic linkage of weapon possession with a flag—they both are a symbol of national sovereignty. Thus, a so-called “modern” military gives a state an identity by providing an image of power regardless of any actual security threat. Morgenthau (1962) noted that fighter-jet imports by less-developed states provide no real military benefit. Instead, import can provide prestige, a belief echoed by a wide body of research (Kaldor, 1982; Kinsella and Chima, 2001; Suchman and Eyre, 1992; Wendt and Barnett, 1993). Spindel (2018: 3) argues that arms sales are credible signals of political relationships.

Which states need greater prestige in terms of aircraft imports? Prestige is given by others (Galtung, 1964) and has recently been linked with alliance centrality (e.g. Maoz, 2011; Renshon, 2016). States with lower alliance centrality have less prestige and, in turn, wish to become more important internationally. Aircraft imports could be a signal that a state wishes to increases its status and that states with lower capability alliance networks have a greater motivation to do so. As aircraft are of less military use to these states, there is little reason to deny the transfer.

H7: Mid-tier states will be more likely to supply aircraft to states with lower capability alliance networks.

Data and tests

Our data cover 1990–2010. We focus on the post-Cold War period since the arms market completed its shift from a seller’s market into a buyer’s market. Additionally, the previous bipolar nature of the system theoretically limited transfer options for states, but in the post-Cold War environment the limit in customer base was largely removed. The unit of analysis is the supplier–importer dyad.

All producing states of planes and helicopters are included in the sample except the superpowers—the USA and Russia.⁵ Much of the literature that bridges the arms–IR gap deals with the USA and USSR. We have reason to believe that the motivations and logic for arms exports are different for these states. We also use this fact to leverage our other explanations. If the mid-tier states are exporting primarily for economic purposes, we would expect the logics of liberalism and trade to be dominant in their patterns of arms sales.

The dependent variables are coded using data from the SIPRI arms trade register and Jane’s Defence publications via Johnson (2017) and SIPRI trend-indicator valued (TIV) data broken down by model of aircraft transferred. The Johnson data has arms exports broken down by the weapon’s type, age, condition (new/used), version, etc., and SIPRI’s TIV model allows for application of Johnson’s categories using TIV. The TIV is a measure of the military value of a transfer, not the amount paid for the transfer. We leverage this data by examining the transfer of different types of aircraft and their capabilities to test the hypotheses.

There are two dependent variables, one binary and one continuous, for the transfer of planes and the transfer of helicopters. A value of “1” is given if a transfer occurred and the logged value of the TIV is the continuous measure.

The independent variables come from a variety of sources. For Hypotheses 1 and 2, we use data from the Correlates of War National Material Capabilities (NMC) data version 5.0 (Singer et al., 1972; Singer, 1987). The National Material Capabilities is an index created from total population, urban population, iron and steel production, energy consumption, military personnel, and military expenditures. The Capability Ratio is calculated by taking the log of exporter capability divided by importer capability; thus, positive values indicate that the exporter is stronger than the importer and negative values indicate that the importer is stronger than the exporter. While the composites of the index may seem better suited on their own, the dataset authors caution against using the data in this manner.

The location of the importer is the other part of the conditional Hypothesis 2. We create a dummy variable indicating if the exporter and importer are in Different Regions based on the Correlates of War country codes. A “1” indicates that the exporter and importer are in different regions. Owing to our interest in focusing on realist assumptions in our tests, we create interaction terms between Capability Ratio and Different Region. The interaction terms allow us to identify how our variables of interest have a conditional effect on transfer decision-making based on multiple factors (Franzese and Hays, 2007).

Regime type data comes from the Polity IV project polity levels (Marshall et al., 2010). Specifically, the Polity2 measure ranging from −10 to 10 is used. Higher values indicate that the state has more democratic characteristics. Joint Democracy is coded as “1” if both the exporter and importer have a Polity score greater or equal to 7.

Trade data comes from Gleditsch (2002) measured in millions of dollars. Trade is the logged value of yearly trade between the exporter and importer. The IGO data come from the Correlates of War data by Pevehouse et al. (2020). Joint IGO Count is the logged value of overlapping IGO membership.

Defense pacts are included since alliance participation is a key to balancing. Defense pact is coded using the Alliance Treaty Obligations and Provision (ATOP) data (Leeds et al., 2002). If a defense pact is present it is coded as “1”.

Former colony status is included as a proxy for constructivism via shared culture. It is coded as a “1” if the importer gained independence from the exporter post-World War II. These data are from Hensel and Mitchell (2007).

Ally capability is created from the CINC and ATOP data. It is the additive value of the capabilities of the importer’s alliance network. Based on alliance network capability, states are then reverse ranked so importers with more powerful alliances have a higher value. Owing to the distribution we log this variable. It ranges from 0 to 4.96 where a value of 0 indicates that the importer has no alliances.

States that produce aircraft are included as a binary control where a value of “1” indicates that the state produces at least one type of plane or helicopter in that year. Producer states are important as their need is arguably lower; thus, the threat of transfer is also lower. This data comes from Johnson (2017).

Tests

We conduct two sets of tests: whether a transfer occurs vs. the size of the transfer. We conduct logistic regressions on the binary dependent variable that measures the presence or absence of a transfer. Previous research shows that the decision to transfer is affected more by the traditional IR independent variables than the size of transfer (e.g. Blanton, 2005; Johnson and Willardson, 2018). We conduct OLS regressions on the continuous variable that measures the size of transfers. In both sets of tests, standard errors are clustered on the exporter–importer dyad.

Results

We present results in two tables and two figures. Table 1 shows the results of the models for Hypothesis 1 with the non-interaction models and Table 2 shows the results for the interaction models for Hypothesis 2. Hypotheses 3–7 are tested in both sets of tests. The two figures show marginal effects of capability ratio when varying the presence of the liberal and constructivist independent variables and interaction effects of capability ratio and different region. The former allows us to evaluate the realist vs. non-realist arguments. The latter allows us to evaluate the strength of the realist results when accounting for location. We first present a cursory discussion of the models in terms of transfers vs. size of transfer and planes vs. helicopters before examining substantive differences.

Table 1.

Binary logit and OLS models on determinants of aircraft transfers.

	Plane	Plane	Helicopter	Helicopter
	Logit	OLS	Logit	OLS
Capability Ratio (Log)	−0.106*	−0.002**	0.131**	0.001*
	(0.04)	(0.00)	(0.04)	(0.00)
Joint Democracy	−0.181	−0.005	0.090	−0.000
	(0.14)	(0.00)	(0.16)	(0.00)
Trade (Log)	0.265**	0.006**	0.308**	0.003**
	(0.02)	(0.00)	(0.04)	(0.00)
Joint IGO Count (Log)	−0.013	−0.003	0.187	0.001
	(0.14)	(0.00)	(0.21)	(0.00)
Former Colony	1.044**	0.026	0.613	0.001
	(0.31)	(0.02)	(0.33)	(0.00)
Ally CINC Rank	−0.070	−0.002*	−0.096*	−0.001
	(0.04)	(0.00)	(0.04)	(0.00)
Defense Pact	0.707**	0.022**	0.523	0.014*
	(0.20)	(0.01)	(0.28)	(0.01)
Importer Producer	−0.779**	−0.014*	−0.135	0.000
	(0.20)	(0.01)	(0.26)	(0.00)
Constant	−5.895**	0.015	−8.103**	−0.009
	(0.48)	(0.02)	(0.79)	(0.01)
Observations	71,694	71,694	71,694	71,694
R ²		0.005		0.003

*p < 0.05; **p < 0.01.

Standard errors clustered by directed dyad in parentheses.

Table 2.

Binary logit and OLS models on determinants of aircraft transfer–interaction models.

	Plane	Plane	Helicopter	Helicopter
	Logit	OLS	Logit	OLS
Capability Ratio (Log)	0.064	0.000	0.294**	0.003**
	(0.07)	(0.00)	(0.08)	(0.00)
Different Region	0.566**	0.013*	1.049**	0.011**
	(0.20)	(0.01)	(0.32)	(0.00)
Capability Ratio × Different Region	−0.207**	−0.003	−0.177*	−0.002*
	(0.08)	(0.00)	(0.09)	(0.00)
Joint Democracy	−0.177	−0.005	0.129	0.000
	(0.15)	(0.00)	(0.16)	(0.00)
Trade (Log)	0.269**	0.006**	0.329**	0.003**
	(0.02)	(0.00)	(0.04)	(0.00)
Joint IGO Count (Log)	0.068	−0.001	0.449	0.003
	(0.14)	(0.00)	(0.26)	(0.00)
Former Colony	1.054**	0.025	0.496	0.001
	(0.31)	(0.02)	(0.33)	(0.00)
Ally CINC Rank	−0.069	−0.002*	−0.123**	−0.001*
	(0.04)	(0.00)	(0.04)	(0.00)
Defense Pact	0.822**	0.026**	0.858**	0.018**
	(0.21)	(0.01)	(0.30)	(0.01)
Importer Producer	−0.737**	−0.013*	−0.191	0.001
	(0.20)	(0.01)	(0.27)	(0.00)
Constant	−6.699**	−0.003	−10.022**	−0.024**
	(0.54)	(0.02)	(1.08)	(0.01)
Observations	71,694	71,694	71,694	71,694
R ²		0.005		0.004

*p < 0.05; **p < 0.01.

Standard errors clustered by directed dyad in parentheses.

The tables present the results for plane transfers in the first two columns and helicopters in the second two columns. The results in Table 1 provide support for Hypothesis 1 with the helicopter models while the opposite occurs in the plane models. As the exporter’s capabilities increase compared with those of the importer, the likelihood of the transfer of planes decreases as does the size of a transfer. With helicopters, transfers are more likely to occur as the exporter becomes comparatively stronger.

In Table 2, the realist variables and their interaction term show that there is support for Hypothesis 2 in both helicopter models. The coefficients for capability ratio and different region are positive and significant in those models, indicating that, on their own, mid-tier exporters transfer to increasingly less powerful states and states outside of their region. The interaction term coefficients, however, are negative, which indicates that when the importer is in a different region there is a reduction in the likelihood and size of transfer as the capability ratio grows in the exporter’s favor. The importance of capability is reduced when the importer is in a different region, supporting Hypothesis 2 for helicopters. For planes, capability ratio is insignificant while the different region coefficients are positive, which indicates that capability is substantively less important than region when considering exports—providing minor support for, or at least not negating, Hypothesis 2. In the plane logit models, the interaction term is also negative and significant. Interaction terms need to be visualized for a full interpretation of the results.

Hypotheses 3–5 on the impact of joint democracy, trade, and joint IGO overlap, respectively, on arms transfers are consistent across all of the models. While joint democracy and joint IGO overlap have been shown to have a negative impact on war and conflict in a plethora of studies, they are consistently statistically insignificant across all of the models in both tables. Trade, however, has positive significant coefficients across all of the models, supporting Hypothesis 4. The lack of significance for joint democracy and joint IGO overlap weakens any assertion that arms trade happens owing to liberal hypotheses.

The constructive hypotheses do not fare much better. We find mild support for Hypothesis 6. In both tables’ logit models only the transfer of planes to former colonial possessions is positive and statistically significant. Hypothesis 7 has greater support. In two of four models in Table 1, and three of four models in Table 2, as the importer’s alliance network capability rank increases, transfers decrease. In other words, as the capability of the alliance network increases, the probability of one of the suppliers providing aircraft decreases.

Defense pact is positive and significant in seven of the eight models; the non-significant coefficient has a p-value of 0.06. Exporters are more willing to export aircraft, and in larger numbers, to states they have a defense pact with. This consistent finding is not surprising—if you are willing to fight with someone you should be willing to provide weapons. However, assigning defense pacts to a single paradigm is difficult as they are related to realism via balancing, to democracy and liberalism via shared norms and values, and linked via shared identify to constructivism.

The last control variable measures when the importer is a producer of aircraft. In all of the plane models, the coefficient is negative and significant whereas they are all insignificant in the helicopter models. Aircraft producers are less likely to import planes than non-producers, presumably because they do not need to import as many planes if they have some production capability. There are fewer helicopter producers than plane producers.

While the tables show us the impact of the individual coefficients, we are also interested in the substantive impact while controlling for the effect of other variables. We do this visually with interaction and marginal effects, respectively. While we calculated the effect for all of the independent variables, we only present the findings that had significant effects in Table 2 when varying the relevant independent variable and capability ratio while holding the other variables in the model at their mean. We visually compare the impact of capability ratio vs. other relationships. Figure 1 shows all of the plane models and Figure 2 the helicopter models with 95% confidence intervals.

Figure 1.

Marginal effect of independent variables on plane transfers with 95% confidence intervals.

Figure 2.

Marginal effect of independent variables on helicopter transfers with 95% confidence intervals.

Across the 12 graphs in the two figures, we present the interaction terms, trade, and defense pact as all the other calculated substantive effects had overlaps in the confidence intervals across the entire range of capability ratio in both tests. Thus, there is no substantive support for the constructivist hypotheses despite some statistical significance in the models, which means we are now determining if the transfer of aircraft occurs under realist beliefs or in a way that cannot be accounted for by the three primary theories of IR.

Figure 1 presents the two models for planes. Figure 2 presents the two models for helicopters. In both, the first column is the interaction effects of capability ratio and different region, the second column is the marginal effects of trade accounting for capability ratio, and the third column is the marginal effects of defense pact when accounting for capability ratio.

In the first column of Figure 1, we see what is expected—no significance in the bottom graph owing to the lack of statistical significance of capability ratio in Table 2. The top graph has an almost imperceptible split in confidence intervals at the −1 value in the top graph, which indicates that realism via Hypothesis 2 does not do a great deal in motivating transfers of planes. This is shown by the overlapping confidence intervals that negate the importance of region when accounting for capability and the mostly flat slope of the solid lines when different region equals zero.

Trade—in the middle column—is a significant predictor of transfer in both models. The values used are that of no trade, mean level of trade, and trade 2 standard deviations above the mean. In both graphs there is separation of the confidence intervals across wide ranges of capability ratio on the x-axis, which indicates that greater levels of trade lead to transfer of planes and as exporter capability increases vs. importer, the impact of trade is tempered a little. Exporters are more willing to transfer planes to more powerful importers who are also large trade partners.

The impact of defense pacts is present in both graphs where separation of the confidence intervals occurs when the importer and exporter have near similar capabilities through the exporter having moderately greater capabilities than the importer. In both graphs a defense pact indicates a greater willingness to engage in plane transfers.

The graphs in Figure 2 are similar except that the slope of the effect lines are opposite with a positive slope as capability ratio increases. There is no effect in the logit model graphs in the top row for region or defense pact, which means that the substantive impact of region and defense pact are diminished when accounting for the results of the full model. However, high levels of trade are a significant factor in the middle column with a positive relationship with helicopter transfers. Unlike planes, as helicopter exporters’ capabilities increase in relation to the importer, they are more likely to transfer helicopters to high-level trade partners and in larger amounts, as shown in the bottom middle graph. The visualization of the interaction effect, in the bottom row, shows that different region matters for helicopter transfers when the importer has greater capabilities than the exporter—the willingness to transfer helicopters to states in a different region with more capabilities supports Hypothesis 2. Lastly, defense pact matters in a similar way to Figure 1.

Discussion and conclusions

The combined implications of these tests are that aircraft transfers do not clearly follow the logic of any of the major paradigms. The greater willingness for exporters to transfer helicopters to less capable states provides some support for the realist arguments we made. Joint democracy and joint IGO membership are consistently insignificant, discounting liberalism. Likewise, our constructivist hypotheses are not supported. The former colony variable is largely insignificant and substantively insignificant when other variables are accounted for. Alliance capability rank as a proxy for prestige seeking is negative regardless of statistical significance. While there is a clear relationship between trade and arms transfers, and trade is a key part of liberalism, we believe the trade results are better explained by the economics of the arms trade, particularly the use of countertrade in arms deals as opposed to trade creating pacific relationships. Lastly alliances play a role in all three paradigms; thus, the consistent positive and significant coefficients of defense pact inform us that alliances matter. Mid-tier exporters support alliance partners. Importers with stronger alliance networks import fewer aircraft.

The transfer of coercive capabilities between states is complicated. Our strongest statistical result is that arms sales follow trade. Trade is a key component of liberal theories of IR, and our results make clear that even goods that have the potential to cause harm are traded. Trade may also be an indicator of other types of cooperative relationships and shared values: strong results for trade may point towards the potential of constructivist arguments of arms sales.

The results hint at cautious application of realist principles by arms exporting states that are not at the top of the security hierarchy (Lake, 2009). There is some support for the idea that mid-tier states prefer to export out of their region, but very little evidence that power differential matters. We asserted that in a realist anarchical world all other states are potential future adversaries. Helicopter-exporting states seem to assess the current state of relations by transferring larger orders out of their own region when the power dynamic is not in their favor. This is a significant finding about the prudent behavior of states embedded within a broader security hierarchy. This is true even though these transfers seem to be driven first and foremost by economic and trade concerns.

Recent literature on arms has highlighted the role that arms play in balancing power (Yarhi-Milo et al., 2016). Such theorizing is important to understanding the role of arms transfers at all levels of power in the international system. We focused on states that do not necessarily share the same capabilities and vulnerabilities, but found that caution has a limited role in arms sales. Our models also suggest that less powerful states may export planes to more powerful states that are part of the same defense pact.

Given the arguments in the specialized arms transfer literature, the lack of support for most liberal factors and minor support for realist and constructivist factors is somewhat surprising. This is, to our knowledge, the first study which has systemically attempted to assess these dynamics based on large-n empirical analysis of the transfer of arms between states. We have focused on non-superpower states as superpower arms transfers have been the area studied more robustly in the past.

Our tests of the three paradigms were limited in terms of engaging many specific propositions from those theories. We expect that there are connections between offense–defense theories and arms transfers, for example. We suspect that this may be one of the factors that accounts for the differences between fixed-wing and rotary aircraft transfers in our analysis.

This may be particularly true of defensive weapons systems that may be transferred to maintain the status quo, or to protect states from other states that may have revisionist tendencies. By arming neighboring states with defensive capabilities, a state may attempt to stabilize its neighborhood (Krause, 1992), although the opposite may happen (Eikenberry, 1995; Stanley and Pearton, 1972). A neighborhood full of defensively hardened states is not an attractive target for an individually powerful country—especially when those states are allied with one another. Such defensive arming may also serve to discourage other states’ leaders within the neighborhood from contemplating revisionist behavior. Future arms transfer research should examine this further.

This paper offers potential topics for researchers who are interested in looking more closely at decision-making processes within states. Factors that may be difficult to discern in large-n observational studies may be discoverable in other research designs.

States buy and sell arms, but they do so for specific economic, identity, and security concerns, and they take into account many factors when doing so. While researchers have systematically studied the transfer of arms between states for more than 40 years, the mechanisms of this decision-making are still in need of deep study, especially work that ties the specialized arms transfer literature to theoretical and empirical work on IR.

Supplemental Material

sj-pdf-1-cmp-10.1177_0738894221992034 – Supplemental material for Arms transfers and international relations theory: Situating military aircraft sales in the broader IR context

Supplemental material, sj-pdf-1-cmp-10.1177_0738894221992034 for Arms transfers and international relations theory: Situating military aircraft sales in the broader IR context by Spencer L Willardson and Richard AI Johnson in Conflict Management and Peace Science

Footnotes

Funding

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

ORCID iD

Spencer L Willardson

Supplemental material

All data, replication materials, and instructions regarding analytical materials upon which published claims rely are available online through the SAGE CMPS website:

Notes

References

Akerman

Seim

(2014) The global arms trade network 1950–2007. Journal of Comparative Economics 42(3): 535–551.

Biersteker

(2002) State, sovereignty and territory. In Carlsnaes

Risse

Simmons

(eds) Handbook of International Relations. London: Sage.

Black

(2016) Air Power: A Global History. Lanham, MD: Rowman & Littlefield Publishers.

Blanton

(1999) Instruments of security or tools of repression? Arms imports and human rights conditions in developing countries. Journal of Peace Research 36(2): 233–244.

Blanton

(2005) Foreign policy in transition? Human rights, democracy, and US arms exports. International Studies Quarterly 49(4): 647–668.

Brauer

(2003) Potential and actual arms production. In Levine

Smith

(eds) The Arms Trade, Security and Conflict. Abingdon: Routledge.

Brauer

Dunne

(2004) Arms Trade and Economic Development: Theory, Policy, and Cases in Arms Trade Offsets. New York: Routledge.

Bull

(1979) The state’s positive role in world affairs. Daedalus 108: 111–123.

Buzan

Herring

(1998) The Arms Dynamic in World Politics. Boulder, CO: Lynne Rienner.

10.

Byman

Waxman

(2000) Kosovo and the great air power debate. International Security 24: 5–38.

11.

Catrina

(1988) Arms Transfers and Dependence. New York: Taylor & Francis.

12.

Caverley

Kapstein

(2012) Arms away. Foreign Affairs 91: 125–132.

13.

Chatterjee

(1975) Arms, Alliances and Stability: The Development of the Structure of International Politics. London: Macmillan.

14.

Colaresi

Thompson

(2005) Alliances, arms buildups and recurrent conflict: Testing a steps-to-war model. The Journal of Politics 67: 345–364.

15.

Comola

(2012) Democracies, politics, and arms supply. Review of International Economics 20(1): 150–163.

16.

De Soysa

Midford

(2012) Enter the dragon! An empirical analysis of Chinese versus US arms transfers to autocrats and violators of human rights, 1989–20061. International Studies Quarterly 56(4): 843–856.

17.

Diehl

Crescenzi

MJC

(1998) Reconfiguring the arms race–war debate. Journal of Peace Research 35: 111–118.

18.

Eikenberry

(1995) Explaining and Influencing Chinese Arms Transfers. Coillingdale, PA: DIANE Publishing.

19.

Erickson

(2011) Market imperative meets normative power: Human rights and European arms transfer policy. European Journal of International Relations 19(2): 209–234.

20.

Erickson

(2015) Dangerous Trade: Arms Exports, Human Rights, and International Reputation. New York: Columbia University Press.

21.

Foreign & Commonwealth Office. (1967) Prospects for Arms Sales: Policy. FCO 46/188. Accessed as National Archives, January 2017.

22.

Franzese

Hays

(2007) Spatial econometric models of cross-sectional interdependence in political science panel and time-series-cross-section data. Political Analysis 15: 140–164.

23.

Galtung

(1964) A structural theory of aggression. Journal of Peace Research 1(2): 95–119.

24.

Garthoff

(1980) American reaction to Soviet aircraft in Cuba, 1962 and 1978. Political Science Quarterly 95: 427–439.

25.

Gibler

Rider

Hutchison

(2005) Taking arms against a sea of troubles: Conventional arms races during periods of rivalry. Journal of Peace Research 42: 131–147.

26.

Gilpin

(1988) The theory of hegemonic War. Journal of Interdisciplinary History 18: 591–613.

27.

Gleditsch

(2002) Expanded trade and GDP data. Journal of Conflict Resolution 46 (5): 176–214.

28.

Gray

(1993) Weapons Don’t Make War: Policy, Strategy, and Military Technology. University Press of Kansas.

29.

Gray

(1996) Arms races and other pathetic fallacies: A case for deconstruction. Review of International Studies 22: 323–335.

30.

Hartley

(2000) The benefits and costs of the UK arms trade. Defence and Peace Economics 11: 445–459.

31.

Hartley

Martin

(2003) The economics of UK arms exports. In Levine

Smith

(eds) The Arms Trade, Security and Conflict. Abingdon: Routledge.

32.

Hensel

Mitchell

(2007) The Issue Correlates of War (ICOW) Project Supplementary Data Set: Colonial History Data Set. Available from: https://doi.org/10.7910/DVN/5EMETG, Harvard Dataverse, V2.

33.

Hirschman

(1980) National Power and the Structure of Foreign Trade. Berkeley, CA: University of California Press.

34.

Holmes

(1988) Measuring the conventional balance in Europe. International Security 12: 166–173.

35.

Huth

Russett

(1984) What makes deterrence work? Cases from 1900 to 1980. World Politics 36: 496–526.

36.

Jepperson

Wendt

Katzenstein

(1996) Norms, identity, and culture in national security. In Katzenstein

(ed.) The Culture of National Security: Norms and Identity in World Politics. New York: Columbia University Press.

37.

Johnson

RAI

(2017) The role and capabilities of major weapon systems transferred between 1950 and 2010: Empirical examinations of an arms transfer data set. Defence and Peace Economics 28: 272–297.

38.

Johnson

RAI

(2019) Decision-making in the arms of a dependent relationship: Explaining shifts in importer acquisition patterns of major weapon systems, 1955–2007. Defence and Peace Economics. DOI: 10.1080/10242694.2019.1618651

39.

Johnson

RAI

Willardson

(2018) Human rights and democratic arms transfers: Rhetoric versus reality with different types of major weapon systems. International Studies Quarterly 62(2): 453–464.

40.

Kadera

(2001) The Power–Conflict Story: A Dynamic Model of Interstate Rivalry. Ann Arbor, MI: University of Michigan Press.

41.

Kaldor

(1982) The Baroque Arsenal. New York: Hill and Wang.

42.

Kapstein

(1992) The Political Economy of National Security: A Global Perspective. New York: McGraw-Hill.

43.

Keller

Nolan

(1997) The arms trade: Business as usual? Foreign Policy 109: 113–125.

44.

Kinsella

(1994) Conflict in context: Arms transfers and third world rivalries during the Cold War. American Journal of Political Science 38(3): 557–581.

45.

Kinsella

(1998) Arms transfer dependence and foreign policy conflict. Journal of Peace Research 35(1): 7–23.

46.

Kinsella

(2002) Rivalry, reaction, and weapons proliferation: A time-series analysis of global arms transfers. International Studies Quarterly 46(2): 209–230.

47.

Kinsella

Chima

(2001) Symbols of statehood: Military industrialization and public discourse in India. Review of International Studies 27(3): 353–373.

48.

Kinsella

Tillema

(1995) Arms and aggression in the Middle East: Overt military interventions, 1948–1991. Journal of Conflict Resolution 39(2): 306.

49.

Klare

(1993) The next great arms race. Foreign Affairs 72: 136–152.

50.

Krause

(1991) Military statecraft: Power and influence in Soviet and American arms transfer relationships. International Studies Quarterly 35(3): 313–336.

51.

Krause

(1992) Arms and the State: Patterns of Military Production and Trade. Cambridge: Cambridge University Press.

52.

Krause

(2004) Hazardous weapons? Effects of arms transfers and defense pacts on militarized disputes, 1950–1995. International Interactions 30(4): 349.

53.

Kydd

(2000) Arms races and arms control: Modeling the hawk perspective. American Journal of Political Science 44: 228–244.

54.

Lake

(2009) Hierarchy in International Relations. Cornell University Press.

55.

Leeds

Ritter

Mitchell

Long

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

56.

Levine

Mouzukis

Smith

(2003) Arms export controls and emerging domestic producers. In Levine

Smith

(eds) The Arms Trade, Security and Conflict. Abingdon: Routledge.

57.

Lieber

Alexander

(2005) Waiting for balancing: Why the world is not pushing back. International Security 30: 109–139.

58.

Maoz

(2006) Network polarization, network interdependence, and international conflict, 1816–2002. Journal of Peace Research 43: 391–411.

59.

Maoz

(2011) Networks of Nations: The Evolution, Structure, and Impact of International Networks, 1816–2001. New York: Cambridge University Press.

60.

Maoz

Russett

(1993) Normative and structural causes of democratic peace, 1946–1986. The American Political Science Review 87: 624–638.

61.

Marshall

Gurr

Jaggers

(2010) Polity IV Project.

62.

Ministry of Defence (1965) Arms sales organisation. DEFE 23/1. Accessed at National Archives 14 January 2017.

63.

Mintz

Heo

(2014) Triads in international relations: The effect of superpower aid, trade, and arms transfers on conflict in the Middle East. Peace Economics, Peace Science and Public Policy 20(3): 441–459.

64.

Morgenthau

(1962) Politics Among Nations: The Struggle for Power and Peace. New York: Knopf.

65.

Morrow

(1993) Arms versus allies: Trade-offs in the search for security. International Organization 47: 207–233.

66.

Most

Starr

(1989) Inquiry, Logic, and International Politics. University of South Carolina Press.

67.

Organski

AFK

Kugler

(1981) The War Ledger. University of Chicago Press.

68.

Pape

(1990) Coercive air power in the Vietnam War. International Security 15: 103–146.

69.

Pape

(1996) Bombing to Win: Air Power and Coercion in War, 1st edn. Ithaca, NY: Cornell University Press.

70.

Pearson

(1994) The Global Spread of Arms: Political Economy of International Security. Oxford: Westview Press.

71.

Pevehouse

JCW

Nordstrom

McManus

Jamison

(2020) Tracking organizations in the world: The Correlates of War IGO Version 3.0 Datasets. Journal of Peace Research 57(3): 492–503.

72.

Platte

Leuffen

(2016) German arms exports: Between normative aspirations and political reality. German Politics 25(4): 561–580.

73.

Qingmin

(2006) The bureaucratic politics of US arms sales to Taiwan. The Chinese Journal of International Politics 1: 231–265.

74.

Renshon

(2016) Status deficits and war. International Organization 70(3): 513–550.

75.

Rosenberg

(1990) What’s the matter with realism? Review of International Studies 16: 285–303.

76.

Rotberg

(2004) When States Fail: Causes and Consequences. Princeton University Press.

77.

Rounds

(2019) Sourcing air supremacy: Determinants of change in the international fighter jet network. PhD thesis, Georgetown University. Available at: https://search.proquest.com/docview/2211495649/abstract/6BE4DB64FB494086PQ/1. (accessed 7 August 2019).

78.

Sample

(2018) Anticipating war? War preparations and the steps-to-war thesis. British Journal of Political Science 48: 489–511.

79.

Sanjian

(1991) Great power arms transfers: Modeling the decision-making processes of hegemonic, industrial, and restrictive exporters. International Studies Quarterly 35: 173–193.

80.

Schweller

(1997) New realist research on alliances: Refining, not refuting, Waltz’s balancing proposition. The American Political Science Review 91: 927–930.

81.

Singer

(1987) Reconstructing the Correlates of War Dataset on Material Capabilities of States, 1816–1985. International Interactions, 14: 115–132.

82.

Singer

Small

(1966) The Composition and status ordering of the international system: 1815–1940. World Politics 18: 236–282.

83.

Singer

Bremer

Stuckey

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

(ed.) Peace, War, and Numbers. London: Sage.

84.

SIPRI (1975) The Arms Trade with the Third World: Revised and Abridged Version. New York: Holmes and Meier.

85.

Sislin

(1994) Arms as influence: The determinants of successful influence. Journal of Conflict Resolution 38: 665–689.

86.

Smith

Humm

Fontanel

(1985) The economics of exporting arms. Journal of Peace Research 22: 239–247.

87.

Snyder

(1997) Alliance Politics. Ithaca, NY: Cornell University Press.

88.

Spindel

(2018) Beyond military power: The symbolic politics of conventional weapons transfers. Dissertation, University of Minnesota. Available at: https://conservancy.umn.edu/handle/11299/199011. (accessed 10 July 2019).

89.

Stanley

Pearton

(1972) The International Trade in Arms. New York: Praeger.

90.

Stohl

Grillot

(2009) The International Arms Trade. Cambridge: Polity Press.

91.

Suchman

Eyre

(1992) Military procurement as rational myth: Notes on the social construction of weapons proliferation. Sociological Forum 7(1): 137–161.

92.

Thies

(2013) The roles of bipolarity: A role theoretic understanding of the effects of ideas and material factors on the Cold War. International Studies Perspectives 14: 269–288.

93.

Toft

(2006) Issue indivisibility and time horizons as rationalist explanations for war. Security Studies 15: 34–69.

94.

van Evera

(1998) Offense, defense, and the causes of war. International Security 22: 5–43.

95.

Vasquez

(1995) Why do neighbors fight? Proximity, interaction, or territoriality. Journal of Peace Research 32(3): 277–293.

96.

Vick

(2015) Proclaiming Airpower: Air Force Narratives and Public Opinion from 1917 to 2014. Santa Monica, CA: Rand Corporation. Available at: https://www.rand.org/pubs/research_reports/RR1044.html#download (accessed 9 February 2018).

97.

Vucetic

Tago

(2015) Why buy American? The international politics of fighter jet transfers. Canadian Journal of Political Science 48: 101–124.

98.

Wallace

(1979) Arms races and escalation. Journal of Conflict Resolution 23: 3–16.

99.

Waltz

(1979) Theory of International Politics, 1st edn. McGraw-Hill.

100.

Wendt

(1999) Social Theory of International Politics. New York: Cambridge University Press.

101.

Wendt

Barnett

(1993) Dependent state formation and third world militarization. Review of International Studies 19(4): 321–347.

102.

Wezeman

(2010) The global arms trade after the Cold War. In The Global Arms Trade. A Handbook. New York: Routledge, pp. 193–207.

103.

Willardson

(2013) Under the influence of arms: The foreign policy causes and consequences of arms transfers. PhD Dissertation, University of Iowa. Available at: http://ir.uiowa.edu/cgi/viewcontent.cgi?article=4789&context=etd.

104.

Yarhi-Milo

Lanoszka

Cooper

(2016) To arm or to ally?: The patron’s dilemma and the strategic logic of arms transfers and alliances. International Security 41: 90–139.

105.

Zehfuss

(2002) Constructivism in International Relations: The Politics of Reality. Cambridge: 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

0.09 MB