Interstate Resource Conflicts: A Network-Centric Resource Access Security Perspective

Abstract

This paper introduces the theoretical concept of perceived resource access security in order to assess the link between natural resource scarcity and interstate conflict. It develops the hypothesis that resource access through international trade is an important determinant for conflict behaviour of states arising from resource concerns, implying a shift in focus away from endowments and towards competition for access. The main analysis is conducted in the form of a unit fixed effects logistic regression covering country-dyads of the period 1962–2010 with military interstate dispute (MID) initiation as dependent variable. The empirical findings consistently demonstrate that a diversified position in resource supply networks mitigates the conflict propensity of states characterized by outside dependence for access to natural resources, and thereby help to determine conditions under which trade reduces conflict.

Keywords

international trade interstate conflict networks resource access security resource scarcity resource realism

Introduction

The widespread opinion about the role of natural resources in interstate conflicts is clear: natural resources are an obvious and significant conflict-increasing factor and states’ (violent) foreign policies are guided by resource security concerns. In short – states fight over natural resources. However, in the scientific literature, the existence of a resource-conflict link is much less clear (O’Lear 2005; Koubi et al. 2014; Meierding, 2016).

Only recently, a number of focused studies has taken on the task of systematically investigating the multitude of posited resource conflict links, often yielding conflicting results: Reuveny and Barbieri (2014) establish the significance and directionality of effects for a range of natural resources on interstate conflict; Colgan (2010) and de Soysa, Gartzke, and Lee (2011) find support for increased aggressive foreign policy behaviour of (revolutionary) petro-states, but reject the resource war hypothesis in which states fight over oil; by contrast, Strüver and Wegenast (2011; 2018) and Caselli, Morelli, and Rohner (2015) find general and robust support for the resource war hypothesis, especially so when resources are located close to state borders, presumably decreasing the cost of conquest. Finally, Meierding (2016) questions the prevalence of interstate conflicts over resources in general and posits that most suggested resource conflicts are not primarily driven by resource concerns, but rather by conditions of desperation. Regardless of the perspective, it is widely accepted that every state has the need for natural resources, ultimately to ensure survival (Baccini, Lenzi and Thurner 2013; Kaldor, Karl, and Said (2007).

Research has implicitly focused on the effect of large foreign resource concentrations, rather than on domestic conditions of resource scarcity. The small number of studies that focuses on the conflict-enhancing effect of resource scarcity is limited to environmental variables (e.g. Stalley 2003), or remains a tentative test of evidence (e.g. Reuveny and Barbieri 2014). In addition to the missing important distinction between mechanism sets (Figure 1), the conceptualization of resource scarcity conditions vis-à-vis interstate conflict is largely inadequate, as the possibility for states to acquire resources through trade has mostly been neglected.

Figure 1.

Organisation of posited conflict mechanisms linking natural resources and interstate conflict.

In light of its evident importance, yet little understood effect on state behaviour, this paper sets out to investigate the effect of natural resource scarcity conditions on the likelihood of military interstate dispute (MID) initiation, and in that context identifies international trade as an important determinant of conflict behaviour of states arising from resource concerns. In contrast to existing studies, it considers an explicit distinction between two main underlying conditions across mechanism sets and adopts a trade dependence perspective to introduce a concept of resource access security, where security is assessed as the extent and shape of a state’s resource import network. In other words, this paper has two aims: first, it conceptualizes resource scarcity as perceived resource access security and highlights the role of trade relationships on the network level in shaping this security. Second, it examines the empirical relationship between resource scarcity and the initiation of militarized disputes against other states.

Resource Scarcity in Terms of Resource Access

Rediscovering the Importance of Trade for a Scarcity Model

Most studies concerned with the effect of natural resource(s) (scarcity) on conflict behaviour employ metrics capturing an aspect of (domestic) resource endowments (see Section A1.3.1 - Sections A1-A7 refer to parts in the online appendix) rather than access in general. While the internal dimension (endowments within state) is of evident importance for acquiring natural resources, a more holistic perspective that also incorporates the external dimension (Figure 2) and provides a framework for assessing it is necessary to better explain state behaviour with regard to resource concerns.

Figure 2.

Modes of resource access.

Surprisingly, almost all studies concerned with theorizing and testing the effect of resource scarcity conditions on interstate conflict neglect one fundamental means to acquire natural resources nowadays: international trade. In fact, this is very much the reason for which resource conflict-related theories, such as the lateral pressure theory by Choucri and North (1989) have been criticized. For instance, Rosecrance (1986) observes that the availability of markets reduces the propensity of states to fight over resources; Harris and Samaraweera (1984) argue that lateral pressure theory fails to recognize trade as an important way for states to access resources and that “the first line of defence against resource conflicts is an informal institution: The market.” (Sources as cited in Meierding 2010, 22).

In fact, it is reasonable to assume that for a decision-maker trade is, once in existence, the most important factor in assessing a state’s access to resources and therefore the presence of resource scarcity. Compared to the access to resources through domestic endowments, which tends to be stable, access through trade is more volatile (less reliable), increasing the perception of insecurity. In other words, assuming a state derives natural resources through domestic endowments and international trade, the latter access point is the weak link in terms of security and needs to be assessed in order to evaluate the overall security level regarding resource access. As a result, it becomes important to evaluate the reliability of a state’s trade access in terms of (default) risks. In fact, it can be conjectured that under certain conditions the degree of trade access security becomes a strong predictor for interstate conflicts arising from resource security concerns. Assuming that a state cannot access sufficient resources through domestic endowments, the absence of trade would only yield two outcomes: a violent action to acquire the needed resources or no access (which can lead to state failure or non-survival).¹

Examples where a country was isolated in terms of resource trade (or a combination of high outside dependence, negative trade balance, unreliable trade partners or enemies as trade partners) and therefore chose to take violent action to acquire the needed resources can be found throughout history. One example is Japan during the beginning of the Second World War: a highly populated state with almost no natural resource endowments on its own, it had to rely on outside sources to satisfy its resource requirements. In the period leading to World War II international trade was disrupted through the Great Depression, which resulted, amongst other reasons, in increased aggressions against China in order to gain control over needed resources. In the face of Japan’s aggressive and expansionist stance towards China, the United States considered how to respond. However, faced with a public opinion opposed to Americans fighting in Asia, the US merely responded with economic sanctions, limited military assistance for China and the refusal to recognize conquered territory. As a result, Japan found itself isolated from international markets unable to satisfy its resource requirements, leading it to further conquests of territories controlled by France, the Netherlands, and Great Britain in the Southeast Asia and the South Pacific region (Yergin 2011; Choucri 2013; Klare 2016).

Similar to Japan, World War II Germany was also a state highly dependent on external sources for access to natural resources and increasingly isolated from international trade as its aggressions progressed. Inspired by the theory of Lebensraum and having failed to become self-sufficient through resource substitutes, part of Hitler’s violent conquest throughout Europe had a direct resource focus on the oil fields of Romania and a further drive towards the oil-rich Caucasus region and the vast territories of Russia (Meierding 2010; Cashman 2014; Kelanic 2016; Klare 2016).

More recent examples of conflicts where natural resources reportedly played a role are the Falkland War (e.g. Burnett 2016) and the Persian Gulf War in 1991 (e.g. O’Connor 1992).² Even more recently, tensions between China and other states in the South China Sea are prevailing. Both the United Kingdom and China are highly dependent on outside sources for resource access, and in 1991 the shale revolution in the US had also not yet come to pass. Finally, changes to the environment due to climate change could decrease global resource availability in general.

In fact, the perceived threat of possible isolation is still relevant today and it appears that states assign high importance to the mitigation of this threat through a diversified position in terms of resource imports. For instance, the United States, historically dependent on the conflict-prone Persian Gulf area as a major oil supplier, has taken steps to diversify its position: under President Clinton, the Baku-Tbisli-Ceyhan pipeline was constructed in order to gain access to the energy resources of the Caspian region independently from Russia, and the George W. Bush administration has increased its focus on resource-rich regions of Africa, providing training, aid and US arms. Similarly, Henderson (2016) shows that the European Union has been able to diversify its access to gas vis-à-vis Russia through increasing competition amongst suppliers. Despite its established ties in Africa, also China seeks new sources of access as its outside dependence is increasing and domestic endowments are insufficient given future resource requirements (Klare 2016). It emerges that a central position in the global resource supply network for a resource importer mitigates the power of large resource suppliers that may use their large endowments as a tool of coercion vis-à-vis the importer (Hughes and Gholz 2016). Even to a larger extent than the major powers, which can compete over the direct control of resources, the majority of resource importers are intent on creating redundant and continuous networks of supply (Nicolas 2009, as quoted in Baccini, Lenzi, and Thurner 2013). This drive towards greater diversification of access can be observed even in the presence of large domestic resource endowments and significantly increased production. As Klare (2016) notes, even in the face of the latest wave of hydrocarbon abundance, access to resources is still regarded as a matter of national security and states appear to constantly pursue an increase in the degree of access security.³

While it is not central to the main argument, the importance attributed to reliable trade access also extends to the securitization of international trade routes. One of the main examples is the Suez Crisis where Israeli, British and French military forces responded to the nationalization of the Suez Canal by Egypt in 1956. Further instances can also be found in more recent history, as especially the US has taken on the role of protecting free flow of resources throughout the globe (Chapman 2008). For instance, after Kuwaiti oil tankers had been attacked by Iran, Reagan authorized reflagging the ships with US flags since an interruption of oil flow was considered a security threat (Palmer 1992). In fact, the US has declared on numerous occasions that it is prepared to undertake military operations in the event of an Iranian attempt to block the Strait of Hormuz (Talmadge 2008) (sources as cited in Klare 2016). Other countries are also concerned with the free flow of resources; one example is the joint effort of EU countries to curb pirate activity on the north-eastern coast of Africa (mission EU NAVFOR). More recently, the UK and Iran found themselves in a standoff with each side having seized an oil tanker from the other and Iran threatening to close the Strait of Hormuz. Regardless of the affected country in question, the free flow of resources is considered a paramount security concern, since a disruption of supplies is likely to have a significant negative impact on the stability of global economies and therefore also security.

A Multilevel Resource Access Framework

The previous section discusses the importance of natural resources for states and observes that trade appears to play a central role in accessing those resources. Further, it notes that states seem to engage in diversification strategies with the aim to increase the reliability of their access through trade. In order to enable an adequate empirical assessment of these dynamics, this section will develop a systematic conceptual framework of resource access including the different ways states can acquire natural resources to satisfy their resource needs (Figure 2). A concept and measure of resource scarcity should capture the degree to which a given state cannot access the desired amount of resources, i.e. the availability of resources relative to the required resources is limited and therefore perceived as scarce.

Internal access mode

Internal resource access refers to the acquisition of endowments within state borders. Due to the direct control and therefore high degree of supply security, this mode is preferred by most states (Meierding 2010). While this preference should be true vis-à-vis the forced access mode, it may not necessarily be so over the trade access mode. As will be shown later, especially a highly secured (hedged) trade access position may almost be considered as secure as internal access, but with possibly lower costs. If this holds true, a slight increase in risk may be traded off against reductions in costs.⁴ Further, there are special circumstances where strategic considerations demand avoiding the internal access mode in order to avoid domestic depletion (Klare 2016). Nevertheless, if available, domestic endowments should constitute an important source for accessing resources.

Forced access mode

The external access category consists of two sub-categories: trade access and forced access. It refers to all access modes in which the desired resource is located outside the state’s borders.

Forced access refers to all access modes in which a state gains access to resources that it did not have access to before through either the threat to use force (access through intimidation) or the use of force (access through intimidation or conquest). Even though the resource becomes situated within state borders if the access is enforced through conquest, it should not be considered as internal access mode, because the resource was outside of state borders before the violent acquisition. Similar to the internal access mode, the forced access mode allows for direct control, yet is characterized by a significantly higher degree of costs and risks in modern times.

Trade access mode

The last method for a state to access natural resources is through the trade access mode.⁵ This mode refers to the acquisition of resources through any kind of trade, mostly through international markets. At the heart of globalization lies the highly efficient provision of goods and services. Simple comparative trade theory dictates that the efficiency of resource acquisitions (and therefore the cost aspect of the access mode) should be maximized in the trade access mode, since the globally most efficient producers should be the ones to provide the resource through trade (with the caveat that the allocation of natural resources across states is fixed). As a result, the trade access mode can offer a desirable cost-utility trade-off, even when compared to the internal access mode. However, the question of risk remains, which, for a concept of resource access security, should be more important than considerations about cost. Compared to the internal access mode, the risk associated with trade access is higher on average. However, there are certain strategies that a state can employ to reduce the risk, for instance, it could increase the number of trading partners, focus on reliable partners, consider the interconnectedness amongst partners, create multiplex dependencies, etc. Costs and risks are further reduced by the existence of international trade organisations, which act to facilitate free and reliable trade flows by providing an institutionalized framework for the allocation of limited resources, through increasing assurances regarding continued resource flows, and through the provision of independent conflict resolution mechanisms such as the dispute settlement process of the World Trade Organization. Consequently, the possibility that even a resource-producing state is obtaining the same resources through trade becomes reasonable, as trade access appears to offer a desirable cost structure at an acceptable level of control.⁶

Choosing a Modus

This section provides a systematization of the cost, risk and utility components and the access requirements of the resource access modes. Under the assumptions that (1) states are unitary actors; (2) the aim of states is survival; (3) and one means for survival is access to resources;⁷ a given state is bound to choose (at least) one mode as not doing so results in state failure (non-survival). When choosing an access mode, states consider utility, costs and risks, which constantly shift given changing (structural) conditions.⁸ In doing so, states are risk averse (in terms of survival), cost averse (in terms of resource acquisition) and utility maximizing (in terms of resource access) and they balance these factors against each other with an on average diminishing preference for low risk over low cost.⁹

As partly addressed above, the three access modes are characterized by varying risk – cost – utility factors, which shape whether and to what extent other conditional factors (such as the world market price for a respective resource) determine the net utility of each access mode and therefore the way a state accesses required resources.

The important cost factor for the internal access mode is extraction costs and potential environmental degradation costs, while the main utility is direct control and also often cost efficiency, since the resource is situated within the state’s own territory and is therefore not dependent on others for access (i.e. no premium, with the exception when specialized extraction technology needs to be provided by another state). The risk factor is the attraction of potential aggressors with regard to the territory where the respective resource is located.

For the trade access mode the main risk factors are market isolation (e.g. in form of trade sanctions) and dependence on world prices, while costs are incurred in the form of a reduction in other tradeable resources. Considering basic trade theory, the utility of this mode is high cost efficiency and a reasonable degree of control, which is further improved by the presence of international institutions.

Finally, the forced access mode can be characterized by a number of cost and risk factors. Before a target territory is occupied there is the risk of failure of conquest (e.g. when facing unexpected military capability of the opponent) and after a territory has been occupied there is the risk of recapture and risk of retribution. In addition to the risks, the factors costs of conquest, costs of occupation, and international reputation costs pose a negative burden on this access mode. The main utility factors of this mode are the direct control and the potential strategic impact. Importantly, this mode may be available when the two others are not.

When considering the type of risk – cost – utility factors of the different access modes – and assuming no access requirements - it can be concluded that, on average, a state will first choose the internal access mode, then the trade access mode and finally the forced access mode.¹⁰

However, there are certain conditions when not all access modes are available to a given state. In fact, each access mode has certain requirements that need to be satisfied, otherwise the state cannot acquire resources through that mode. Specifically, the access requirements for the internal access mode are the prevalence of domestic resource endowments, the extractability of resources, and the availability of extraction technology. For the trade access mode the requirements are market access, ownership of tradable resources (money can be a proxy for the set of availably tradable resources), and the availability of required resources in the market. Finally, the forced access mode requires access to high military capability (possibly through allies) relative to the military capability of the target (coalition) and the ability to apply it where necessary.¹¹

Yet, even when assuming the absence of entry requirements, the preference ranking of the access modes is not static, but is contingent on a set of additional factors. For instance, under low world prices, it is possible that the net utility of the trade access mode is higher than that of the internal one, arising from a shift in the pay-off structure of the modes. Numerous other considerations impacting the utility, risk, and cost perception for each mode exist, e.g. domestic resources should be saved or their extraction is costly; the aim is to increase power or to decrease the power of another; etc.

Those shifts are also true in a broader historical context. In the not very distant past, access to resources was ensured through empires. For instance, during the colonial period western states were faced with favourable terms for choosing the forced access mode: the discovery of America presented new and resource rich territory, for which the cost of conquest and occupation appeared low as the indigenous defenders were highly inferior in terms of military capability. (1) The absence of reliable international markets and (2) the highly favourable trade-off between low risks and costs relative to (3) high potential gains in form of great and lootable resource concentrations; (4) the competition vis-à-vis other empires; (5) and the diminishing low international reputation costs (possibly even negative) resulted in the forced access mode being the modus operandi with regard to securing access to needed and desired natural resources.¹²

However, the military costs of upholding an empire are significant, especially so when the size of the territory under control increases. With the rise of globalization and reliable international markets in the 20^th century, the trade access mode has gained in importance, as its costs and risks have decreased significantly. Furthermore, the establishment of international political institutions and new international norms, e.g. with regard to human rights, increase the cost aspect of the forced access mode. As a result, direct control over resources is traded off against a more cost-effective type of access. For instance, the empirically supported strategic oil hypothesis of De Soysa, Gartzke, and Lie (2011) states that powerful consumers prefer to allow local control as long as oil is not too expensive. However, if prices rise too high, importing states shift their preference from commerce to conquest.¹³ An indication of the importance of direct access to vital resources can also be found in the significant time difference between the decolonization of oil-rich colonies and all other types of colonies (Colgan 2014).

It is important to acknowledge that under certain conditions direct control (forced access mode) is favoured (especially when access is considered a security issue).

The fact that states may choose the forced access mode has also been predicted by a number of models. For example, Acemoglu et al. (2012) develop a dynamic theory of resource wars with a focus on the interaction between scarcity of resources and incentives for war. Their result shows that in the case of inelastic demand for a resource, war incentives increase over time and conflict becomes inevitable. In some cases, this spiral can be broken by regulation of prices and quantities by the resource-rich state. Further, Maxwell and Reuveny (2000) establish that per capita resource scarcity in fact raises conflict, and, if resources are destroyed through conflict, the system may become destabilized and break down.

Overall, it becomes likely that states use the internal and trade access mode before the forced access mode, since today’s cost, risk and utility trade-off is in favour of these. For states with domestic resource endowments the dominant access strategy may be a balance between the internal and trade access, while for other states trade should constitute the main basis for their access to natural resources. Nevertheless, there appear to be instances in which the forced access mode is still preferred.

Furthermore, it becomes evident that control over access to resources (rather than direct control) is of importance in the context of survival and can therefore be considered a source of power. Normally, the risk of losing access to natural resources within the trade access mode (excluding a lack of tradable resources) arises from competitive, political, and ideologically driven power-dynamics between states; conflict arising from resource concerns needs to be considered in the context of competition for access. Even if we assume that a resource producing state would not willingly give up access to the lucrative global market, it could be forced by other powerful states to cease providing access to the resource to third states (e.g. see Stalley 2003). Importantly, this means that it is not necessarily the producer of resources where the power is concentrated.¹⁴

The Resource Access Security Concept and the Central Role of Trade Networks

The foregoing discussion suggests that trade should be a key factor in determining how vulnerable (volatile) a state’s access to required resources is. Building on this, the following section develops a resource access security perspective with a focus on the network level. It allows for an adequate assessment of the effects of resource scarcity vis-à-vis conflict, as it is a highly suitable and forward-looking concept for understanding and assessing existing and future levels of access security.

Examples of the relevance of the network level of analysis with regard to international conflict are numerous. For instance, Dorussen and Ward (2008, 2010) show that both direct and indirect general trade ties, as well as indirect links through IGO membership tend to reduce international conflict. Yet, these relationships are not always clear cut, e.g. Hafner-Burton & Montgomery (2006) establish that many states with the same degree in IGO networks in fact increase the likelihood of conflict, and Maoz et al. (2007) show a similar effect when imbalances of positive and negative ties (‘the friend [enemy] of my friend is my friend [enemy]’) occur. Regarding trade, Haim (2016) demonstrates that the effect of alliance networks on bilateral trade is much larger than that of dyadic alliances, and Chyzh (2016) suggests that indirect trade channels enable states to participate in trade without the need to adhere to human rights. As Gartzke and Westerwinter (2016) suggest for economic relationships and Dorussen, Gartzke, and Westerwinter (2016a) in general, extradyadic ties are very relevant in determining and understanding factors driving conflict. The JPR special issue Dorussen, Gartzke, and Westerwinter (2016b) offers additional examples on networks and conflict.

The degree of perceived resource access security (Figure 3 for a concept overview and Figure A6.1 for a detailed overview) is broadly defined as a function of (1) the degree of existing (current) resource scarcity, referring to a situation in which the availability of resources relative to the required resources is limited and therefore perceived as scarce; and (2) the level of anticipated future access, referring to the expectation of the future level of access to resources. In other words, it could be said that the concept of resource access security captures the overall degree of vulnerability of a state with regard to conditions of resource scarcity vis-à-vis interstate conflict. Furthermore, it should be noted that resource access security is a theoretically necessary but empirically unobserved link in the causal chain (and realistically unobservable in a large N study). Since it concerns policy-makers’ assessment of future developments, it can also be seen as an indicator of how vulnerable access is perceived to be in relation to changes in the international context.

Figure 3.

Concept overview - The network centric resource access security concept.

Determining the Anticipated Resource Access Through Trade

Aside from more structural considerations, such as the prevalence of international organisations in the system, there are two main sets of factors that have a direct influence on the anticipated resource access through trade (which can be understood as trade access security): (1) The reliability of individual trade partners (dyadic level) and (2) the position within the global resource trade network (more precisely the direct and indirect connectivity and therefore reachability as captured by the network level) – due to the great scope of this subject matter this paper focuses almost exclusively on the latter, only touching upon the assessment of the former.

While this study limits itself to capturing the reliability of individual trade partners through the size of trade flows attributed to them, a number of additional factors can be considered on a dyadic level between importer and exporter. These include present and past relationships, such as the degree of past diplomatic exchange, military presence in the foreign state, the amount of FDI between states, similarity of the UNGA voting behaviour, exchange of arms, the level of power asymmetry, etc. - or similarities between states, e.g. in terms of regime type, cultural affinity, foreign policy, rule of law, colonial past, etc.. Ideally, these factors lead to dependencies of the exporting state on the importing state resulting in an interdependent relationship.

Factors of more structural nature, such as the existence of international trade organisations, play an important role in shaping the trade access mode in general (e.g. such organisations act to reduce transaction costs and uncertainties with regard to international trade), yet also affect the reliability of individual trading partners directly. For instance, if a trading partner is a member in the same international trade organization it should be regarded as more reliable, because institutionalized structures are in place to uphold the bilateral trade flows (also through the provision of conflict resolution mechanisms).

These considerations mostly refer to the dyadic relation between importer and exporter, yet the degree of access security can also be assessed through the network level. While most of the factors discussed above still play a role (mostly when considered in composite), new ones arise, namely those describing the supply network of a state and its position therein. For instance, the overall number of trade partners is an important direct factor for evaluating the degree of anticipated access, as additional trade partners can compensate for the default of individual ones (to a certain extent). In fact, considering the network level could be seen as more important: an individual reliable partner is less secure in terms of access than a diversified position with regard to a large number of independent and individually less reliable partners. One reason for this is the higher exposure in terms of impact magnitude regarding resource trade flows to outside influences for an individual state than it is for a set of partly independent states. In this regard, the anticipated access through trade could also be dependent on the degree of independence amongst the suppliers of a given state (e.g., the ability of one state to supply resources to another might depend on a third state), or the dependence of a supply network on the importing state across other dimensions than resource trade (e.g. arms trade), increasing the opportunity costs of breaking up a tie (Maoz 2009).

These network factors add to the overall degree of diversification, as they should be critical in determining the expected volatility of access through trade and therefore overall access to resources. Considering relevant factors limited to individual states should play a comparatively subordinate role, as this fails to account for the network structure of a state’s trade access. However, these (and other network measures, such as degree of interdependence) should also be relevant on the network level, and future research needs to account for them in order to arrive at a more comprehensive framework of access security.

To capture these complex sets of relationships empirically, this work uses the toolbox of Social Network Analysis. Networks consist of nodes and edges, the former one represents the individual units of observation and the latter one the relational information between those units. For the case at hand, the nodes are the states within the international system and the edges are the trade flows with regard to natural resources. With this information a global resource trade network is constructed. The use of Social Network Analysis enables to quantify the degree of access security to resources through trade for a state based on information about not only the individual reliability and number of the respective state’s trading partners, but also the state’s entire supply network.

The dynamics arising from the network level then capture the anticipated level of resource access (Figure 3). The position of a state in specific networks, alongside its actual resource endowments, determines to a large degree whether it experiences a security risk due to a lack of access to resources. In other words, it is posited that a state’s overall degree of diversification is negatively related to the default risk of the trade access mode for that state, and therefore to its level of access security, and thus its conflict propensity.

Importantly, the degree of diversification contains information about the future level of access volatility: ceteris paribus, the future level of resource access through trade is robust to outside shocks if the level of diversification is high. Even in the face of acute (momentary) scarcity or deficiency (insufficient resources arrive in the state) a central (and diversified) position in the supply network for the required resource will overrule the actual resource scarcity, because future resource supplies are secured through a reliable network, based on the position therein. On the other hand, an undiversified position within (trade) networks may increase the volatility of the degree of future access, resulting in a resource scarcity threat for the state. This means the current degree of diversification (captured through the network measures) can be regarded as a proxy for the anticipated level of resource access (an expectation of a future state), which is in turn the decisive factor in determining the overall level of access security (Figure 3). In other words, the degree of diversification in the trade access mode approximates a future state of resource access and thereby determines the degree to which a current scarcity impacts the level of access security (and therefore conflict propensity) of states.

As a result, the introduced concept can be considered as forward oriented, an important aspect for a concept of scarcity,¹⁵ and hence resource access security.¹⁶ While it is normal practice to employ objective measures to explain state behaviour, states don’t necessarily act upon the information used to describe a current condition, but rather based upon the expectation of a condition in the future, which should be reflected in the employed metric.¹⁷

Overall, this section suggests that the network perspective is particularly helpful in generating more precise hypotheses on state behaviour with regard to the securitization of resource access.

Evaluation of Direct Relationships

Considering the state level, the most direct measure to capture the degree of diversification becomes the number of direct trade partners a state has,¹⁸ which is captured by what Social Network Analysis calls the degree centrality for the respective state (Freeman 1978).

Higher levels of degree centrality imply a higher level of resource access security, because the state has a wider spectrum of trading partners (it has a high degree of diversification in terms of trading partners). Losing one out of 10 trading partners does not have the same negative impact as losing one out of two, on average. Furthermore, assuming that a given state is insufficiently endowed with natural resources and characterized by a low number of actual trade ties (it has a low degree of diversification), the state may perceive that it is isolated and that it cannot easily establish new trade relationships.¹⁹ As a consequence, the possibility of an interstate conflict should become more likely (assuming this is within the capability scope of the affected state). For example, for the interaction between resources, sanctions and interstate conflict, Hasan and Lahiri 2015, 91) find that “[…] whether an anticipated future sanction on one country will reduce war intensity depends on the level of resource stock; and finally the effect of a permanent sanction on both countries is uncertain and war intensity will fall only if the resource stocks of the countries are sufficiently high.” This illustrates that under conditions of (imminent) resource scarcity induced by low domestic resource stocks and sanctions (imposed isolation) can have a positive impact on the likelihood of interstate conflict.²⁰

In addition to testing and comparing the effects of the number of trading partners versus the total amount of trade flows (Figure 4), a measure that here will be called balance-sensitive degree centrality also allows for the evaluation of the effect of supply networks where trade flows are equally distributed across trading partners versus networks with an unequal distribution (Figure 5). Assuming a random default by one of the trading partners, the effect is always the same in a balanced network, but can vary significantly in an unbalanced one. Consequently, also in terms of predictability, a balanced network should have a larger positive effect on the access security of a state, since the potential maximum negative impact of one supplier defaulting is larger in an unbalanced network (the lower effect of a minor supplier default does not fully compensate for the higher effect of a major supplier default).

Figure 4.

Number of trading partners vs. (total) value of trade flows.

Figure 5.

Balance versus unbalanced distribution of trade flows across trading partners. Note: The thickness of the arrow signifies the amount of resources imported from the respective state.

Following hypotheses 1 and 2, respectively, arise:

The number of trade partners of State A is negatively related to the probability of State A initiating a military interstate dispute.

Higher concentration of bilateral trade flows for State A is positively related to the probability of State A initiating a military interstate dispute.

Evaluation of Indirect Relationships

A major contribution of the network perspective is that, in addition to the direct dyadic relationships, also indirect relationships can be assessed. In this regard, recalling the example from the preceding section, the assessment of the degree of diversification should not be limited to the first-degree import network of a state (number of direct trading partners), as its position relative to the entire network possibly yields additional relevant information. Even if all direct trading partners are responsible for the same amount of trade flows, they may vary in terms of embeddedness in the overall trade network themselves. For instance, a high degree of ties for the direct trade partners (some states both sell and buy the same natural resource) can indicate a higher degree of robustness vis-à-vis trade defaults in a similar manner the direct import ties indicate for the initial state. Considering the right node in Figure 6, the upper direct trading partner could then be perceived as more reliable with 4 direct ties than the lower one with only 2, ceteris paribus. For the analysis at hand, this implies that the degree to which direct trading partners are in turn centrally embedded in the resource network can be taken into consideration when assessing the resource access security of a state.

Figure 6.

Example of differences in overall reachability.

Furthermore, a direct trading partner can be in a central position with regard to the entire network, and therefore be perceived as more valuable in terms of access to other nodes in the network compared to other partners accounting for a similar amount of direct trade flows, yet being less embedded. If such partners are well connected with regard to the overall network the state in question will be able to reach other network members through fewer degrees of separation. In other words, such states could serve as gates to a greater number of resource exporters, increasing the scope of resource access beyond direct ties and therefore overall access security from an importer perspective. Notably, in Figure 6 all nodes are connected and both states have the same degree centrality (number of direct trading partners), yet the left node is characterized by a lower level of overall reachability.

Incorporating information about trade ties beyond the first degree adds an additional layer to the evaluation of the overall resource access security of a state.

The network metric eigenvector centrality is a centrality measure where a node’s centrality is its summed connections to others weighted by their centralities (Bonacich 1987). For the context at hand, it evaluates the reliability of a state’s supply network based on the number of direct ties of the state’s trading partners.²¹

Following hypothesis 3 arises:

The number of trade partners of nodes in State A’s direct resource supply network is negatively related to the probability of State A initiating an interstate conflict.

The network measure closeness centrality captures the sum of shortest paths to reach any node in the network (Bavelas 1950). This can serve as proxy for the overall access to international resource markets; a high score implies the shortest access to resources relative to the entire market (on average lower order relations have to be used to reach any resource exporter in the market).

Following hypothesis 4 arises:

The degree of reachability across all nodes in a resource supply network by State A is negatively related to the probability of State A initiating an interstate military dispute.

Empirical Analysis

Research Design

The main analysis is conducted in the form of a unit fixed effects logistic model employing a dyad year unit of analysis with standard errors clustered on the dyad-level and covering politically relevant country-dyads of the period 1962–2010 with Military Interstate Dispute (MID) initiation from the COW dataset as dependent variable.²²

A crucial part for testing the presented conflict mechanism is the delimitation from other mechanisms (as for instance discussed by Humphreys 2005 and demonstrated in Section A4.3). The presented hypothesis specifically refers to the resource conflict mechanism in which a state initiates an interstate conflict in order to acquire access to the required resource, rather than predicting a change in conflict propensity in general. Therefore, the dyadic research design seems most appropriate, as it allows to consider a specific situation (namely resource scarcity in State A and resource presence in State B of a given dyad), therefore excluding other conflict mechanisms that may be triggered and ensuring that the effect is not driven by the confounding effect of those. For instance, when faced with civil unrest induced by domestic resource scarcity, an interstate conflict could be initiated for diversionary purposes, strengthening the government in place. Taking natural resources in the target state under consideration ensures that a conflict could in fact have been initiated due to interests related to acquiring access to natural resources. In other words, in the tested mechanism, resource scarcity causes conflict when there is a prospect of mitigating it.

The research design is complemented by the application of the quadratic assignment procedure (QAP) (Section A4.4), which serves as an important robustness test, as it takes into account network interdependencies (Krackardt 1988).²³ An in-depth discussion on the research design, such as the coding of on-going conflicts and joiners or different ways to operationalize the dependent variable, can be found in Section A1.

The variables of interest, outside trade dependence and the respective centrality measures (both standardised) are based on the resource oil, as captured by the SITC categories 3300 – 3399, and measured in current USD value. The complete batch of dyadic trade data is retrieved from the Observatorium of Economic Complexity (OEC), which comprises revised trade data from the Center for International Data from Feenstra et al. (2005) for the years 1962–2000 and from the UN COMTRADE database for the period 2001–2014, cleaned by the BACI team (Gaulier and Zignago 2010). All variables of interest are based on this data. For greatest coverage, GDP data in current USD is taken from Gleditsch (2002). The exact operationalization can be found in Figure A11 and Sections A1.3.2 & A1.3.3.

The set of standard control variables is based on Gleditsch, Salehyan, and Schultz (2008), including the temporal control from Beck, Katz, and Tucker (1998) (see Section A1.4 for a detailed account of covariates). Section A2 comprises a table with summary descriptive statistics for all variables and Section A3 a list of the data sources.

A number of findings suggest simultaneity between conflict and natural resources; for instance resources could be destroyed as a result of the conflict, which again has an effect on conflict (e.g. Maxwell and Reuveny 2000). In order to address potential endogeneity and ensure that MIDs do not affect the independent variables, a 1-year lag is employed. A further discussion on endogeneity and binary outcome variation can be found in Section A1.6.

The following standard model S can be specified

(S) C o n f l i c t_{ij,t} = α + β 1 V I_{ki,t} - 1 + X_{ij,t}^{'} - 1 + υ_{ij,t}

Where

C o n f l i c t_{ij,t}

is a dummy taking the value of 1 when state i initiated a conflict with state j at time t and otherwise 0;

V I_{ki,t} - 1

is the variable of interest for resource k and state i at time t-1,

X_{ij,t}^{'} - 1

is the vector of the control variables between the states i and j for time t - 1, and

υ_{ij,t}

is the error term.

Results

Table 1 presents the results of the standard model S for all centrality measures for the resource oil with the standard COW variable for conflict initiation (State A) as dependent variable, also displaying all included control variables. Table 2 shows results for the in-degree centrality measure taking into account the presence or absence of resources in the target state respectively. Table 3 and Section A4.4 present results of the QAP. All models consistently yield significant results in favour of the presented mechanism. Notably, for the fixed-effects model S, all components of the interaction between outside oil dependence and the most direct centrality measure, in-degree centrality, are significant with effect directions entirely consistent with expectations. As predicted by the theory, states with high outside dependence are more prone to initiate an interstate conflict, if their position in the resource trade network can be characterized as not diversified, and vice versa. This result is supported by the corresponding marginal effect plots (Figure 7 and additional plots in Section A5.2),²⁴ and remains robust (or improves) when considering various fixed effects²⁵ (Table A4.2.1.2), different sets of control variables (Table A4.2.1.3), the monadic level of analysis for the standard model²⁶ (including a number of robustness tests yielding entirely consistent results, Section A4.2.2), or different ways to operationalize the dependent variable, various coding decisions regarding ongoing conflict years and joiners, in addition to further robustness tests such as the inclusion of all dyads (Table A4.2.1.5).²⁷

Table 1.

Standard Model S - Centrality Measures (All Covariates Displayed).

	(1)	(2)	(3)	(4)	(5)	(6)
	no_FE	FE	no_FE	FE	no_FE	FE
Outside oil dependence	0.179	0.439**	−0.211	0.0321	1.618*	2.522**
Outside oil dependence	(0.154)	(0.188)	(0.147)	(0.0940)	(0.913)	(1.132)
Degree centrality	0.429	−1.655**
Degree centrality	(0.688)	(0.751)
Outside oil dependence#Degr ∼ y	−4.978***	−3.891***
Outside oil dependence#Degr ∼ y	(1.843)	(1.495)
Eigenvector centrality			0.577	−1.425**
Eigenvector centrality			(0.682)	(0.693)
Outside oil dependence#Eige ∼ y			−2.800	−0.724
Outside oil dependence#Eige ∼ y			(2.847)	(1.533)
Closeness centrality					−1.437**	−1.869***
Closeness centrality					(0.558)	(0.654)
Outside oil dependence#Clos ∼ y					−3.536**	−4.684**
Outside oil dependence#Clos ∼ y					(1.780)	(2.076)
Net oil imports	4.08 × 10⁻¹²**	2.57 × 10⁻¹²*	4.52 × 10⁻¹²***	1.26 × 10⁻¹²	5.43 × 10⁻¹²***	1.78 × 10⁻¹²
Net oil imports	(1.68 × 10⁻¹²)	(1.36 × 10⁻¹²)	(1.46 × 10⁻¹²)	(1.36 × 10⁻¹²)	(1.41 × 10⁻¹²)	(1.34 × 10⁻¹²)
Territorial contiguity	−0.0767	2.849***	0.0386	−2.856***	0.0886	−2.854***
Territorial contiguity	(0.868)	(0.961)	(0.869)	(0.957)	(0.881)	(0.959)
Distance, log	−0.164	−0.541***	−0.148	−0.544***	−0.138	−0.543***
Distance, log	(0.112)	(0.119)	(0.112)	(0.119)	(0.113)	(0.119)
Colonial contiguity	0.675***	0.559***	0.689***	0.562***	0.706***	0.559***
Colonial contiguity	(0.156)	(0.193)	(0.155)	(0.193)	(0.155)	(0.191)
Democracy both	−0.902***	−0.830***	−0.904***	−0.838***	−0.896***	−0.845***
Democracy both	(0.203)	(0.168)	(0.207)	(0.169)	(0.207)	(0.167)
Capability ratio, log	−0.207***	−0.233***	−0.204***	−0.236***	−0.204***	−0.236***
Capability ratio, log	(0.0367)	(0.0344)	(0.0364)	(0.0342)	(0.0367)	(0.0340)
Alliance	0.000937	−0.0570	−0.0143	−0.0442	−0.0333	−0.0376
Alliance	(0.0930)	(0.104)	(0.0905)	(0.105)	(0.0906)	(0.105)
Trade dependence	−6564.9	3571.2	−7422.1	3448.1	−7420.9	2914.4
Trade dependence	(4969.4)	(2718.2)	(5337.0)	(2714.2)	(5256.5)	(3120.0)
Shared IGO memberships	0.00594	0.0142***	0.00725**	0.0123***	0.00881**	0.0123***
Shared IGO memberships	(0.00372)	(0.00371)	(0.00350)	(0.00366)	(0.00348)	(0.00359)
Civil conflict one	0.543***	0.244***	0.535***	0.241**	0.534***	0.241**
Civil conflict one	(0.0883)	(0.0934)	(0.0890)	(0.0938)	(0.0888)	(0.0935)
Civil conflict both	0.882***	0.397**	0.877***	0.388**	0.880***	0.389**
Civil conflict both	(0.134)	(0.167)	(0.134)	(0.168)	(0.135)	(0.167)
Peace years	−0.382***	−0.264***	−0.376***	−0.263***	−0.378***	−0.265***
Peace years	(0.0253)	(0.0237)	(0.0250)	(0.0237)	(0.0252)	(0.0238)
Spline 1	−0.00244***	−0.00149***	−0.00230***	−0.00147***	−0.00234***	−0.00146***
Spline 1	(0.000323)	(0.000317)	(0.000318)	(0.000318)	(0.000318)	(0.000317)
Spline 2	0.00126***	0.000663**	0.00114***	0.000647**	0.00118***	0.000611**
Spline 2	(0.000258)	(0.000259)	(0.000256)	(0.000259)	(0.000254)	(0.000257)
Spline 3	−0.000116	0.0000266	−0.0000685	0.0000310	−0.0000780	0.0000617
Spline 3	(0.000102)	(0.000103)	(0.000103)	(0.000104)	(0.000100)	(0.000102)
Constant	−1.579*	4.104***	−1.716*	3.917***	−1.032	4.967***
Constant	(0.893)	(1.093)	(0.893)	(1.080)	(0.931)	(1.172)
Observations	93,804	75,868	93,804	75,868	93,804	75,868

Standard errors in parentheses | All independent variables (except peace years and splines) are lagged one time period.

*p < 0.1, **p < 0.05, ***p < 0.01.

Table 2.

Standard Model S – Results for Subset of Dyads Defined by the Presence or Absence of Oil in the Target State.

	Oil Presence in Target States					Oil Absence in Target State
	(1)	(2)	(3)	(4)	(5)	(6)	(7)
Outside oil dependence	0.970**	0.761***	0.313	0.379*	1.229***	0.187	0.257
Outside oil dependence	(0.461)	(0.290)	(0.199)	(0.194)	(0.291)	(0.300)	(0.316)
Degree centrality	−1.927	−1.819	−1.599*	−1.371	−3.403***	0.730	0.912
Degree centrality	(1.706)	(1.396)	(0.904)	(0.875)	(1.156)	(1.679)	(1.720)
Outside oil dependence#D ∼ y	−6.014*	−5.527**	−3.024*	−3.229**	−10.82***	0.856	1.041
Outside oil dependence#D ∼ y	(3.356)	(2.470)	(1.610)	(1.468)	(2.865)	(3.175)	(3.299)
Net oil imports	−1.67 × 10⁻¹²	2.61 × 10⁻¹²	2.86 × 10⁻¹²*	3.34 × 10⁻¹²**	5.14 × 10⁻¹²***	−7.61 × 10⁻¹²	−1.87 × 10⁻¹¹*
Net oil imports	(1.40 × 10⁻¹¹)	(1.11 × 10⁻¹¹)	(1.56 × 10⁻¹²)	(1.58 × 10⁻¹²)	(1.86 × 10⁻¹²)	(9.00 × 10⁻¹²)	(1.03 × 10⁻¹¹)
Constant	5.231***	5.331***	3.696***	3.540***	0.975	3.329*	5.162**
Constant	(1.889)	(1.561)	(1.154)	(1.125)	(1.565)	(1.857)	(2.600)
	15,273	23,859	48,607	49,892	33,239	24,539	15,864

Standard errors in parentheses.

*p < 0.1, **p < 0.05, ***p < 0.01.

(1) oil reserves (2) oil extraction (3) oil production (4) oil deposits (5) oil rents (6) no oil reserves (7) no oil extraction. Sources for corresponding data can be found in Section A3. Not all models converge.

Table 3.

MR-QAP - p-Values for Individual Years with a 5-Year Gap, as well as the Average Across All Years (See Section for All Years A4.4).

MR-QAP - Standard Model S (5 years gaps)
Year	Degree Centrality
1965	0.118
1970	0.098
1975	0.213
1980	0.064
1985	0.095
1990	0.091
1995	0.031
2000	0.072
2005	0.079
2010	0.019
AVG	0.09

Figure 7.

Marginal effect plots. Note: The curve refers to the frequency distribution. The plots are calculated with the interflex package for Stata.

The interaction containing closeness centrality is also significant for the standard model (Table 1), as well as most other robustness tests (Table A4.2.1.5). Eigenvector centrality (Table 1) remains insignificant for the standard model, yet, when considering other ways to capture the initiation of a conflict (i.e. revisionist state), it also becomes statistically significant with effect directions consistent with expectations (Table A4.2.1.5).

Given the apparent importance of the interaction with the degree centrality measure (and its individual components) and the insignificance of resource (oil) imports when considered in isolation (Table A4.2.1.1 for the nested model), the degree centrality measure for oil covering a shift in relative importance starting from the number of trade partners and moving towards the amount of total imports should confirm these results (Figure A6.1 or Section A1.3.3 for the exact operationalization). As expected, it can be observed that the significance level is at the 1% level when considering the number of trade partners and then fades until becoming insignificant when moving towards total imports (Table A4.1.1). This further supports the proposed theory and underlines the previous finding that the conflict decreasing effect is highest when the respective state is diversified in terms of the number of trading partners. Furthermore, in models with the balance-sensitive degree centrality measure, the significance level is highest when trade flows are distributed equally across trading partners rather than concentrated on a few partners, which limits the maximum level of negative impact when faced with the default of an important trading partner (Table A4.1.2).

Importantly, results are consistent for the subset of violent conflicts or only interstate wars (highest conflict intensity) (Table A4.2.1.4). Notably, the significance level and coefficient are larger for violent conflicts (as opposed to non-violent) and considerably larger for wars (as opposed to non-wars).

Finally, the effect size, for instance for the degree centrality measure, is considerable.²⁸ For the monadic model without unit fixed effects and only the standardized degree centrality considered in isolation, while keeping all other covariates at their mean, the average likelihood of conflict decreases by ca. 2.5% with an increase of the degree centrality variable from the 25th to the 75th percentile, by about 5.5% with an increase from the 5th to 95th percentile and by ca. 8% when considering the maximum difference. To put this in perspective, comparing the conflict propensity between a major and minor power in this model yields a difference of ca. 4%. The results appear to be largely robust given the broadly similar odds ratios for the same model with and without unit fixed effects and across the two levels of analysis.

Conclusion

This paper introduces the theoretical concept of perceived resource access security for states with the purpose of establishing a causal chain between natural resources and conflict behaviour of states, and thus allow for an adequate assessment thereof. It posits that the degree of access security to resources through trade is a crucial dimension for investigating the resource-conflict link. Indeed, considerable empirical support has been found that the degree of diversification within international resource trade networks plays an important role in determining the effect of resource scarcity vis-à-vis interstate conflict. Specifically a central position in a resource supply network decreases a state’s conflict propensity arising from resource concerns.

Overall, the empirically supported perspective presented in this work suggests a shift in focus from ‘how much’ to ‘who has control’ in the resource-conflict debate. In other words, the struggle to obtain natural resources should not be framed in terms of absolute quantity (‘peak oil discussion’), but much rather in terms of competition for resource access. Importantly, this means that even in the face of general resource abundance, situations of individual resource scarcity are very possible and even likely, implying that controlling access rather than the endowments directly becomes a source of power. Finally, the findings have meaningful implications for the trade-conflict debate as they provide new insights concerning the conditions under which trade reduces or increases conflict, arising from considerations about interdependencies.

Understanding and acknowledging the central role of trade access to natural resources vis-à-vis conflict may become increasingly relevant in the future. In times of climate change and potentially weakened international institutions, the global availability of access to important natural resources could be threatened. Even more so, this is true for future potential financial crises or trade wars that have the potential to significantly disrupt global trade flows. Given the theory at hand, a collapse of global markets could shift access security considerations of states and have a significant impact on the conflict propensity in the international system over a longer period of time.

Considering these points, it appears important to rekindle research efforts that address the fundamental interplay between resources, survival, and power,²⁹ in what could be considered the main subject of a resource realism in its own right.

Supplemental Material

Supplemental Material - Interstate Resource Conflicts: A Network-Centric Resource Access Security Perspective

Supplemental Material for Interstate Resource Conflicts: A Network-Centric Resource Access Security Perspective by Luka Bareis in Journal of Conflict Resolution

Supplemental Material

Supplemental Material - Interstate Resource Conflicts: A Network-Centric Resource Access Security Perspective

Supplemental Material for Interstate Resource Conflicts: A Network-Centric Resource Access Security Perspective by Luka Bareis in Journal of Conflict Resolution

Footnotes

Acknowledgements

I’m very grateful to the editor and to the JCR reviewers for their insightful comments. Thanks also to Indra de Soysa for his helpful suggestions and comments. Special thanks to David Davison for advising me on the particularities of the English language. Last but not least, I want to express my great gratitude to Mathias Koenig-Archibugi for his unwavering support of my research in this field of study.

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

Luka Bareis

Supplemental Material

Supplemental material for this article is available online.

Notes

References

Acemoglu

Daron

Golosov

Mikhail

Tsyvinski

Aleh

Yared

Pierre

. 2012. “A Dynamic Theory of Resource Wars.” The Quarterly Journal of Economics 127 (1): 283-331. doi: 10.1093/qje/qjr048

Baccini

Leonardo

Lenzi

Veronica

Thurner

Paul W.

. 2013. “Global Energy Governance: Trade, Infrastructure, and the Diffusion of International Organizations.” International Interactions 39 (2): 192-216. doi: 10.1080/03050629.2013.768512

Bavelas

Alex

. 1950. “Communication Patterns in Task-Oriented Groups.” The Journal of the Acoustical Society of America 22 (6): 725-30. doi: 10.1121/1.1906679

Beck

Nathaniel

Katz

Jonathan N.

Tucker

Richard

. 1998. “Taking Time Seriously: Time-Series-Cross-Section Analysis with a Binary Dependent Variable.” American Journal of Political Science 42 (4): 1260-1288.

Bennett

D. Scott

Stam

Allan C.

. 2000. “EUGene: A Conceptual Manual.” International Interactions 26 (2): 179-204.

Bonacich

Phillip

. 1987. “Power and Centrality: A Family of Measures.” American Journal of Sociology 92 (5): 1170-1182.

Brown

Gardner M.

Field

Barry

. 1979. “The Adequacy of Measures for Signalling the Scarcity of Natural Resources.” Scarcity and Growth Reconsidered 13.

Burnett

M. Troy

. 2016. Natural Resource Conflicts: From Blood Diamonds to Rainforest Destruction [2 Volumes]: From Blood Diamonds to Rainforest Destruction. ABC-CLIO.

Caselli

Francesco

Morelli

Massimo

Rohner

Dominic

. 2015. “The Geography of Interstate Resource Wars.” The Quarterly Journal of Economics 130 (1): 267-315.

10.

Cashman

Greg

. 2014. What Causes War? An Introduction to Theories of International Conflict. Lanham, MD: Rowman & Littlefield.

11.

Chapman

Duane

. 2008. “Gulf Oil and International Security: Can the World’s Only Superpower Keep the Oil Flowing?” Energy Security and Global Politics. 17 July 2008. doi: 10.4324/9780203892831-10

12.

Choucri

North

R. C.

. 1989. “Lateral Pressure in International Relations: Concept and Theory.”. Handbook of War Studies, edited by Midlarsky

Manus I.

. Ann Arbor, MI: University of Michigan Press.

13.

Choucri

Nazli

. 2013. Challenge of Japan Before World War II, Vol. 6. Routledge.

14.

Chyzh

Oga

. 2016. “Dangerous liaisons: An endogenous model of international trade and human rights.” Journal of Peace Research 53 (3): 409-423.

15.

Colgan

Jeff D.

2010. “Oil and Revolutionary Governments: Fuel for International Conflict.” International Organization 64 (4): 661-694.

16.

Colgan

Jeff D.

2014. The Political Economy of the End of Empire. International Political Economy Society Annual Conference.

17.

Dorussen

Han

Ward

Hugh

. 2008. “Intergovernmental Organizations and the Kantian Peace. A Network Perspective.” Journal of Conflict Resolution 52 (2): 189-212.

18.

Dorussen

Han

Ward

Hugh

. 2010. “Trade networks and the Kantian peace.” Journal of Peace Research 47 (1): 29-42.

19.

Dorussen

Han

Gartzke

Erik

Oliver

Westerwinter

. 2016a. “Networked international politics: Complex interdependence and the diffusion of conflict and peace.” Journal of Peace Research 53 (3): 283-291.

20.

Dorussen

Han

Gartzke

Erik

Oliver

Westerwinter

, eds. 2016b. “Special Issue on Networked International Politics.” Journal of Peace Research 53(3).

21.

Feenstra

Robert C

Lipsey

Robert E

Deng

Haiyan

Alyson C

Hengyong

. 2005. “World Trade Flows: 1962-2000.” Working Paper 11040. National Bureau of Economic Research. doi: 10.3386/w11040

22.

Freeman

Linton C.

1978. “Centrality in Social Networks Conceptual Clarification.” Social Networks 1 (3): 215-239.

23.

Gartzke

Erik

Rohner

Dominic

. 2011. “The Political Economy of Imperialism, Decolonization and Development.” British Journal of Political Science 41 (3): 525-556.

24.

Gartzke

Erik

Westerwinter

Oliver

. 2016. “The complex structure of commercial peace contrasting trade interdependence, asymmetry, and multipolarity.” Journal of Peace Research 53 (3): 325-343.

25.

Gaulier

Guillaume

Zignago

Soledad

. 2010. BACI: International Trade Database at the Product-Level (the 1994-2007 Version). SSRN Scholarly Paper ID 1994500. Rochester, NY: Social Science Research Network. https://papers.ssrn.com/abstract=1994500

26.

Gleditsch

Kristian Skrede

. 2002. “Expanded Trade and GDP Data.” Journal of Conflict Resolution 46 (5): 712-724.

27.

Gleditsch

Kristian Skrede

Salehyan

Idean

Schultz

Kenneth

. 2008. “Fighting at Home, Fighting Abroad: How Civil Wars Lead to International Disputes.” Journal of Conflict Resolution 52 (4): 479-506.

28.

Hafner-Burton

Emilie M.

Montgomery

Alexander H.

. 2006. “Power Positions: International Organizations, Social Networks, and Conflict.” Journal of Conflict Resolution 50 (1): 3-27.

29.

Haim

Dotan A.

2016. “Alliance networks and trade: The effect of indirect political alliances on bilateral trade flows.” Journal of Peace Research 53 (3): 472-490.

30.

Harris

John R.

Samaraweera

Vijaya

. 1984. “Economic Dimensions of Conflict.” Multidisciplinary Perspectives on Population and Conflict: 123-156.

31.

Hasan

Md Didarul

Lahiri

Sajal

. 2015. “A Two-Period Model of Natural Resources and Inter-Country Conflicts: Effects of Trade Sanctions.” International Journal of Development and Conflict 5 (2): 76-100.

32.

Henderson

James

. 2016. “Does Russia Have a Potent Gas Weapon?” The Palgrave Handbook of the International Political Economy of Energy, 461-486. Springer.

33.

Hendrix

Cullen S.

2017. “Oil Prices and Interstate Conflict.” Conflict Management and Peace Science 34 (6): 575-596.

34.

Hughes

Llewelyn

Gholz

Eugene

. 2016. “Energy, Coercive Diplomacy, and Sanctions.” The Palgrave Handbook of the International Political Economy of Energy, 487-504. Springer.

35.

Humphreys

Macartan

. 2005. “Natural Resources, Conflict, and Conflict Resolution: Uncovering the Mechanisms.” Journal of Conflict Resolution 49 (4): 508-537.

36.

Kaldor

Mary

Lynn Karl

Terry

Said

Yahia

. 2007. Oil Wars. Pluto Press.

37.

Kelanic

Rosemary

. 2016. “The Petroleum Paradox: Oil, Coercive Vulnerability, and Great Power Behaviour.” Security Studies 25: 181-213.

38.

Klare

Michael T

. 2016. “No Blood for Oil? Hydrocarbon Abundance and International Security.” The Palgrave Handbook of the International Political Economy of Energy, 419-439. Springer.

39.

Koubi

Vally

Gabriele Spilker Böhmelt

Tobias

Bernauer

Thomas

. 2014. “Do Natural Resources Matter for Interstate and Intrastate Armed Conflict?” Journal of Peace Research 51 (2): 227-243.

40.

Krackardt

David.

1988. “Predicting with networks: Nonparametric multiple regression analysis of dyadic data.” Social Networks 10 (4): 359-381.

41.

Le Billon

Philippe

Cervantes

Alejandro

. 2009. “Oil Prices, Scarcity, and Geographies of War.” Annals of the Association of American Geographers 99 (5): 836-844.

42.

Maoz

Zeev

Terris

Lesley G.

Kuperman

Ranan D.

Talmud

Ilan

. 2007. “What is the enemy of my enemy? Causes and consequences of imbalanced international relations, 1816–2001.” Journal of Politics 69: 100-15.

43.

Maoz

Zeev

. 2009. “The Effects of Strategic and Economic Interdependence on International Conflict across Levels of Analysis.” Journal of Political Science 53 (1): 223-240.

44.

Maxwell

John W.

Reuveny

Rafael

. 2000. “Resource Scarcity and Conflict in Developing Countries.” Journal of Peace Research 37 (3): 301-322.

45.

Meierding

Emily L.

2010. “No Blood for Oil? The Dynamics of Interstate Petroleum Disputes.” The University of Chicago. ProQuest Dissertations and Theses.

46.

Meierding

Emily L.

2016. “Dismantling the Oil Wars Myth.” Security Studies 25 (2): 258-288.

47.

Nicolas

Francoise

. 2009. ‘ASEAN Energy Cooperation: An Increasingly Daunting Challenge’. Paper from Institut Francais Des Relations Internationles (IFRI), Available at IFRI: www.Ifri.Org/Downloads/Fnicolas.Pdf

48.

O’Connor

James

. 1992. “Murder on the Orient Express: The Political Economy of the Gulf War.” Social Justice 19 (1 (47)): 62-75.

49.

O’Lear

Shannon

. 2005. “Resource Concerns for Territorial Conflict.” GeoJournal 64 (4): 297.

50.

Palmer

Michael A

. 1992. Guardians of the Gulf. New York: Free Press.

51.

Palmer

Glenn

d’Orazio

Vito

Kenwick

Michael

Lane

Matthew

. 2015. “The MID4 Dataset, 2002–2010: Procedures, Coding Rules and Description.” Conflict Management and Peace Science 32 (2): 222-242.

52.

Piccone

Theodore JE

Trinkunas

Harold A.

. 2014. The Cuba-Venezuela Alliance: The Beginning of the End? Latin American Initiative, Foreign Policy at Brookings.

53.

Reuveny

Rafael

Maxwell

John W.

Davis

Jefferson

. 2011. “On Conflict over Natural Resources.” Ecological Economics 70 (4): 698-712.

54.

Reuveny

Rafael

Barbieri

Katherine

. 2014. “On the Effect of Natural Resources on Interstate War.” Progress in Physical Geography 38 (6): 786-806.

55.

Rosecrance

Richard N.

1986. The Rise of the Trading State: Commerce and Conquest in the Modern World. New York: Basic Books.

56.

Soysa

Indra de

Gartzke

Erik

Grete Lie

Tove

. 2011. “Oil, Blood, and Strategy: How Petroleum Influences Interstate Disputes.” Typescript. The Norwegian University of Science and Technology and the University of California, San Diego.

57.

Stalley

Phillip

. 2003. “Environmental Scarcity and International Conflict.” Conflict Management and Peace Science 20 (2): 33-58.

58.

Strüver

Georg

Wegenast

Tim

. 2011. “Ex Oleo Bellare? The Impact of Oil on the Outbreak of Militarized Interstate Disputes”.

59.

Strüver

Georg

Wegenast

Tim

. 2018. “The Hard Power of Natural Resources: Oil and the Outbreak of Militarized Interstate Disputes.” Foreign Policy Analysis 14 (1): 86-106.

60.

Talmadge

Caitlin

. 2008. “Closing Time: Assessing the Iranian Threat to the Strait of Hormuz.” International Security 33 (1): 82-117.

61.

Tomz

Michael

Wittenberg

Jason

King

Gary

. 2001. CLARIFY: Software for Interpreting and Presenting Statistical Results. Version 2.0.

62.

Yergin

Daniel

. 2011. The Prize: The Epic Quest for Oil, Money & Power. Simon and Schuster.

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.

93.83 MB

0.00 MB

2.86 MB