Oil Wealth,Post-conflict Elections,and Postwar Peace Failure

Abstract

New research has emerged that suggests there is a troubling relationship between elections and civil wars; primarily, elections increase the risk of civil war recurrence. I investigate this relationship further by examining the economic factors associated with the connection between postwar elections and peace failure. Specifically, how does the presence of oil wealth impact the risk posed by postwar elections. Drawing on previous findings in the democratization literature, I suggest the immobility of oil wealth dramatically increases the stakes associated with postwar elections. As postwar elites use irregular electioneering to consolidate their control of oil revenue, it increases the incentives for postwar opposition to use violence as a means to achieve their objectives. Using post-civil war data from 1945 to 2005, I demonstrate that postwar elections that occur in oil-rich economies dramatically decrease the durability of postwar peace. Once controlling for petro elections, though, I demonstrate that subsequent postwar elections actually increase the durability of postwar peace.

Keywords

civil wars democratic institutions democratization resource extraction

Within the peace science scholarship, researchers over the last decade have begun to examine the puzzle of why so many countries are unable to sustain peace within their borders. For instance, after twenty years of fighting between the Angolan government and the National Union for the Total Independence of Angola (UNITA) rebels, a settlement to the conflict was finally agreed upon. Three years later, the country descended back into civil war (Kreutz 2010). Fearon and Laitin (2003) demonstrate that, out of the many factors that influence whether or not a country will become embroiled in a civil war, one of the most important is whether that country had previously fought a civil war. The puzzle of how countries can break the conflict trap (Collier et al. 2003) has spawned numerous studies. Recently, though, new research has begun to ask what role elections play in preventing or spurring peace failure (Flores and Nooruddine 2012; Brancati and Snyder 2013); adding to more literature within the discipline that has examined the nexus of elections and violence (Wilkinson 2004; Dunning 2011; Cedarman et al. 2013). The findings suggest that postwar elections tend to ignite tensions within a fragile society and lead to a renewal of civil war (Flores and Nooruddine 2012; Brancati and Snyder 2013). I advance the current scholarship on the connection between elections and postwar peace failure by examining how the presence of oil impacts the salience of postwar elections.¹ Specifically drawing on Boix’s (2003) findings on asset specificity and democratization, this article seeks to incorporate economic factors into the discussion of elections and postwar violence. I examine whether oil wealth plays a similar role in the post-conflict environment, as it does within the democratization process. As Boix (2003) demonstrates, asset wealth that is immobile at the point of democratization increases the chance that the elites within society will take all means necessary, including potentially resorting to violence, so as to protect their wealth from future redistributive policies. Therefore, elites may attempt to instigate violence that leads to renewed civil war rather than allow their assets to be used for redistributive policies. This study, then, seeks to provide more insights as to why elections following civil wars are so detrimental to the postwar peace.

I argue that oil wealth increases the salience of the first postwar elections, as these elections pose a risk to those in power. Given that a loss at the polls would mean that elites within the government will no longer be able to extract personal revenue from oil wealth, they are more likely to engage in illicit electioneering so as to hang onto power. This process may be particularly dangerous in the postwar environment. As noted by Brancati and Snyder (2013), the first postwar elections usually pit former militants against one another in the electoral contest. The use of violence, bribery, and intimidation pushes past civil war protagonists to restart the civil war so as to renegotiate the status quo.

This article proceeds in three parts. First, I will explore the findings of previous literature on the determinants of peace failure. This section will highlight the recent emergence of the literature on elections and post-conflict peace failure as well as highlight how my research will expand on previous findings. Second, I will provide a novel theory of post-conflict peace failure that emphasizes the role that oil plays in magnifying the riskiness of post-conflict elections. As oil wealth may be the only wealth within the country, control of future revenue becomes especially important for postwar elites. As the first elections threaten access to future oil wealth, elites will use illicit electioneering as a means of preserving access to future revenues. The use of illicit electioneering will in turn push opponents to engage in violence as a means to renegotiate the status quo. Finally, I will provide an empirical analysis of the novel propositions provided in the previous section. As the results demonstrate, postwar elections that occur in oil-rich countries have a large and statistically significant impact on the risk of peace failure.

Determinants of Peace Failure

The current scholarship on peace duration tends to emphasize two broad themes in preventing civil war recurrence: factors associated with the previous civil war and the post-conflict arrangements. As to the former, previous research suggests that greater battle deaths (Walter 2004; Quinn, Mason, and Gurses 2007; Mason et al. 2011; Gurses, Rost, and McLeod 2008) tend to have a positive relationship to peace failure, with bloodier wars often giving rise to future conflicts. On the other hand, other works suggest that greater conflict duration tends to reduce the risk that future combatants will resume hostilities (Doyle and Sambanis 2006; Fortna 2004); the shorter the duration of the previous conflict, the less opportunity combatants have to learn about one another. While still other work suggests that identity-based conflicts, such as those that relied on ethnic-based recruiting or ideological grounds, tend to produce more unstable peace, increasing the risk that a new war will break out (Binningsbo and Rustad 2012; Mason et al. 2011). Along similar lines, the presence of foreign peacekeeping forces has a consistent positive effect on the promotion of peace following civil wars (Fortna 2003, 2004; Gurses, Rost, and McLeod 2008; Mason et al. 2011). This finding generally underscores one of the few options the international community has in promoting peace, by raising the costs of reneging (Mattes and Savun 2009) on peace agreements. Though, this determinant of post-conflict peace is hindered by the uncertainty that follows the exit of third party guarantors.

Beyond the characteristics of the pervious war, previous research suggests that the type of institution following a civil war also matters, as democracies tend to benefit from more sustained peace following civil wars than autocracies (Gurses, Rost, and McLeod 2008; Quinn, Mason, and Gurses 2007). These findings, though, are tempered by the fact that democracies that are within the “transitioning” stage (or anocracies) tend to have a greater risk of experiencing renewed civil wars (Mason et al. 2011). This is a common phenomenon, though, as transitioning nations also suffer from negative practices such as greater human rights abuses than do either democracies or autocracies (Fein 1995; Hegre et al. 2001; Davenport and Armstrong 2004). While institutions are commonly termed a sticky influence in politics, as they are difficult to change, it is not uncommon for civil wars that end in negotiated settlements that often include specific mandates calling for institutional reform. For example, power-sharing arrangements are a common tool utilized in the establishment of negotiated settlements. Such power-sharing arrangements tend to promote greater peace spells (Hartzell 1999; Hartzell, Hoddie, and Rothchild 2001; Hartzell and Hoddie 2003; Mattes and Savun 2009). The findings, though, have been somewhat mixed. For instance, power-sharing arrangements that combine economic and military power sharing (such as including members of different ethnic groups in the army or revenue sharing arrangements) tend to produce more sustained peace (Hartzell and Hoddie 2003). But Derouen, Lea, and Wallensteen (2009) argue that, on the contrary, costly power-sharing arrangements push governments to renege on the settlement; in other words, increasing the costs will push actors to seek to use violence to renegotiate the terms of the agreement. Additionally, revenue-sharing programs have been promoted as a way to reduce the grievance elements (such as inequality within society) that push combatants to take up arms (Walter 2004; Binningsbo and Rustad 2012). Intuitively this makes sense, though unfortunately there appears to be little evidence of its efficacy (Binningsbo and Rustad 2012). Given this body of research, it appears that power-sharing arrangements that focus solely on dividing control of the military appear to have the greatest impact on reducing the risk of a renewed civil war (Derouen, Lea, and Wallensteen 2009).

Other research has questioned the general use of such power-sharing arrangements. Earlier work on power dividing institutions, such as a “checks and balances” institutional arrangement, are much more likely to promote peace than power-sharing institutions (Roeder and Rothchild 2005), but this finding has not been confirmed through iterated testing. Rather, the relationship between institutions and peace duration may be much more subtle. For instance, Joshi and Mason (2011) find that peace duration often depends on whether the settlement expanded the size of the governing coalition, with smaller coalitions leading to more unrepresented interests who choose to take up arms. This finding fits well with Roessler (2011), who argues that many leaders are willing to shrink the size of the governing coalition, thereby increasing the risk of civil war, as long as it decreases the risk of the government suffering a military coup.

The current scholarship on postwar peace duration has begun to further unpack the impact of institutions (specifically democratic institutions) in generating more durable peace. The relationship between elections and post-conflict peace duration is a recent addition to the study of the post-conflict environment. Generally, findings suggest that early elections appear to increase the risk of renewed conflict emerging (Brancati and Snyder 2013; Flores and Nooruddin 2012; Cederman, Gleditsch, and Hug 2013). The post-conflict government is left with many challenges, such as a lack of general security, infrastructure damage, and low capital (Addison et al. 2005; Collier 2009; Flores and Nooruddin 2012). Given that the state engages in fostering democracy rather than addressing many of the fundamental issues with the country, the citizenry becomes aggravated with the process (Autesserre 2009). Elections appear to antagonize preexisting tensions remaining from the previous civil war (Brancati and Snyder 2013; Flores and Nooruddin 2012). The empirical research suggests that the first election appears to be the most contentious, while the second and third decrease the risk of future peace failure (Flores and Nooruddin 2012; Cederman, Gleditsch, and Hug 2013). Given that factors such as election timing (Brancati and Snyder 2013) enhance the risk of some elections and therefore play a critical role in generating stability in the postwar environment, it should be intuitive that the context in which postwar elections take place will also impact the durability of peace in the postwar environment, as more contentious environments should increase the risk of postwar elections. I argue that the first postwar elections that occur in oil-rich countries play a similar role in enhancing the risk of peace failure, as these initial elections generate a risk to elite access to future oil revenues.

Theory

I argue that the initial postwar elections in oil-rich countries generate considerably more risk of peace failure than postwar elections that occur within countries that do not have oil. This is because oil wealth increases the salience of postwar elections as opposed to postwar elections that occur in countries without substantial oil wealth, as access to oil wealth creates an opportunity for elites within the postwar government to increase their own fortune as well as the fortune of their supporters. As noted by Flores and Nooruddin (2012), the first postwar election is typically the most hazardous to peace duration. Following the end of the conflict, the first postwar elections set the stage for leaders to lose access to that wealth which in turn incentivizes them to use coercive means to hold onto power. Leaders may use outright violence to hold onto power thereby suppressing opponents, or they may attempt to use the coercive apparatus of the state to effectively steal the election through voter intimidation which pushes opponents to seek redress through open rebellion (Dunning 2011; Brancati and Snyder 2013; Cederman, Gleditsch, and Hug 2013). Additionally, I argue that postwar elections that operate in the absence of oil wealth tend to be more stable as they are less salient. Given that mobile wealth has either fled the country during the fighting or that elites may simply withdraw their wealth if they lose the election, politicians should be more amenable to peacefully handing over power to their opponents. This is not to say that postwar elections are not contentious affairs. Rather, drawing on conclusions of Boix (2003), I argue that the immobility of oil wealth generates more salience associated with the initial postwar elections which push elites to use coercive means to insure their grasp on power, which in turn spurs a resumption of fighting.

This argument rests on a few propositions. First, I make the claim that elites within the postwar government can increase their own wealth, and the wealth of their supporters, through stealing the revenue generated from oil. Second, I make the claim that elected officials will use violence, vote buying, or other forms of irregular electioneering as a means of preserving their access to oil revenue in the future. Third, I claim that access to oil revenue creates much more salient elections, as prospects for future oil revenue increases the potential cost of losing an election. Finally, I assume that postwar elections that occur in the absence of oil wealth are less salient and, over time, increase the durability of postwar peace as they provide a nonviolent mechanism for political entrepreneurs to make claims upon the state.

Oil Wealth and the Prospects of Oil Wealth

Elites within the postwar government are incentivized to use revenue generated from oil wealth to supplement their own wealth following the civil war. As demonstrated from recent experiences in Nigeria as well as the Republic of Congo (or Congo-Brazzaville) where government officials repeatedly steal large amounts of oil wealth, this is not an unfounded expectation (Nossiter 2014). Vast amounts of oil revenue are often siphoned off by ruling officials as a means to supplement their own incomes. Revenue may also be used to reward individuals within the elites’ winning coalitions (Bueno de Mesquita et al. 2003), or in other words, elites within the government may use oil wealth to ensure loyalty from their supporters. Specifically, oil wealth provides an attractive option for elites in government as collecting it generally does not require the consent of the country’s population as opposed to taxes. Additionally, it forces companies who do the actual drilling to rely on the state for security and infrastructure as a means to safely transport oil out of the postwar state and onto the international market. This reliance ensures that those within the state receive a steady paycheck from foreign petroleum companies.

Additionally, a postwar government may have to rely heavily on oil wealth following an end to fighting, as petroleum reserves may constitute the only source of revenue for the state. Following a civil war, there is a dearth of wealth within the country. During a civil war, those assets that are mobile are shifted abroad as capital (both human and otherwise) are transferred to less risky areas (Addison et al. 2005; Bigombe et al. 2000; Collier 2009). This leaves the state with little resources necessary to assuage public concerns or payoff potential rivals. If a country then relied on mobile assets preceding the civil war, then I assume that those assets were moved abroad during the conflict. The logic behind this claim is intuitive, as business leaders are concerned with reducing the risk toward their assets (Jensen 2003; Büthe and Milner 2008). The onset of a civil war creates clear political risks for investors within the country (both domestic and foreign), as rebels may target those assets to weaken the state or the state may expropriate businesses to offset the costs of fighting an insurgency. Therefore, if business leaders are able to shift their resources abroad during the conflict, they will do so. With few domestic assets left within the country following a civil war, there is little that the postwar government can generally tax. Given that there are few mobile assets that the postwar government can tax, oil wealth becomes a very attractive source of wealth for elites within the postwar government. For instance, following the end of Angola’s twenty-seven-year-old civil war in 2002, the government still relies heavily on oil wealth as a means of revenue, with government officials taking a large share of the profits to line their own pockets (The Economist, 2014). Given Angola’s postwar culture of corruption, the rentier economy forces many business leaders and foreign investors to “partner” with government officials or military leaders. Building off of this notion, I make the claim that elites then have a vested interest in consolidating their control of oil wealth in the postwar environment.

Oil and the Democratic Process

Boix’s (2003) well-known work on the impact of asset specificity on the process of democratization underscores why oil wealth should make postwar elections so contentious. As demonstrated by his work, Boix suggests that elites, when faced with opponents who will likely institute redistributive policies, would rather suffer open violence than sacrifice their wealth in the future. Therefore, if wealth is immobile, it increases the chance that elites will resort to violence as a means of protecting their access to it within the democratization process. Acemoglu and Robinson (2001) come to a similar conclusion, demonstrating that the democratization process is most at risk in the least developed societies as elites generally fear overly redistributive policies, and will help to overthrow democratic governments to maintain their access to wealth. I adapt this conclusion to the postwar setting by suggesting that elections that occur after civil wars end place those in the postwar government in a difficult position. As elections may be part of the peace settlement (as was the case in Mozambique) or part of a new rebel government’s mandate to reform the system, elites within the state may feel compelled to follow through with an open election.

It should be noted, though, that the literature on oil wealth and democratization is somewhat mixed. While Boix (2003) posits that immobile wealth such as oil should hinder the transition toward democracy, other scholars have proposed a more nuanced narrative as to oil’s impact on the democratization process. Some scholars (Morrison 2007; Dunning 2008) have suggested that oil provides a degree of stability, preventing major transitions either toward or away from democratization. On the other hand, work by Gurses (2011) suggests that oil wealth should ameliorate divides between elites and the poor. Oil wealth can be used to improve the infrastructure of society and supplement state revenue so as to increase public goods while reducing the need for redistributive policies. In light of this evidence, one might consider that oil wealth should improve the prospects for postwar democratization. Given the debate on the role of oil in the democratization process, this article does not seek to address the process of democratization. Rather, I borrow lessons from the democratization literature to highlight the role oil plays in destabilizing one aspect of the democratization process: the first postwar election. While in the long run oil wealth could improve the prospects for postwar democratization, elites within the government and militants within civil society must first get past the initial postwar election. If the country descends back into war, then the prospects for future reconciliation and cooperation through the democratic process seem rather bleak.

While elections can be contentious affairs in general, the postwar setting raises the stakes even further. As noted by Suhrke and Berdal (2012, 3), “While in the long run associated with nonviolent conflict resolution, in times when rules of the new order are being defined in the aftermath of war the stakes are high and the democratization process has historically been punctured by violence.” As many in government may have been the perpetrators of violence during the civil war, once removed from office during elections those former leaders may be exposed to rivals who wish to use reprisal violence for past transgressions. Joshi (2013) generates a similar conclusion in his findings on postwar democratization. Joshi suggests that postwar leaders fear possible repression from past opponents, which is why they usually seek to reduce the degree of democratization in the postwar environment. Therefore, elections following an end to the civil war may pose a serious threat to those in government following a cessation of open hostilities. The risk of losing access to oil as a result of the first postwar election therefore spurs actors to engage in risky behavior increasing the likelihood that the civil war will resume once more.

Petro Elections and Peace Failure

I argue that the presence of oil in postwar countries increases the precarious nature of postwar elections. While early elections increase the risk of peace failure in general, I posit that the presence of oil in those countries increases the salience of those postwar elections. Previous research has demonstrated that the elections that directly follow an end to the war often carry a substantial risk to postwar peace (Brancati and Snyder 2013; Flores and Nooruddin 2012). As mentioned earlier, these elections generate potential contention between former military opponents. Elites, fearing that they will be removed from office, engage in brutality and intimidation of opponents, spurring their opposition to return to fighting. Additionally, the international community often pushes for elections rather than directing its resources to solve the issues that led to fighting in the first place (Autesserre 2009). I build on these findings by suggesting that the proverbial stakes are raised when elites in postwar government may face the loss of future oil revenue once out of office. Additionally, political entrepreneurs within society have a greater incentive to place themselves within the government as they and their supporters expect to benefit from future oil revenue in the future. As suggested by Boix (2003) and Acemoglu and Robinson (2001), when elites within society fear that their wealth may be removed from them as a result of the democratic process, they may resort to unconventional means (such as political violence or coups) so as to preserve their wealth in the future. As oil wealth may be the only revenue generating asset in the postwar environment, elites are likely to use violence to consolidate their control of the revenue stream in the face of an open election. As suggested by Tilly (2003), governments usually are the beneficiaries of exclusion and inequality, where the wealth developed in a state is often used to buttress the rule of those in charge. Tilly posits that the government will use the coercive apparatus of the state to solidify its control. Furthermore, Brancati and Snyder (2013) posit that in the immediate postwar environment, armed groups from the previous conflict are typically the most powerful political competitors in the elections. As noted by the authors, “When postconflict elections are held soon after civil wars end, former combatants generally possess greater material resources, more extensive organizational networks, and stronger ties to society than newly formed, proreform groupings” (Brancati and Snyder 2013, 6). These militant groups (regardless of whether they are associated with the incumbent government or rebels) are then in a special position to challenge the results militarily. I argue that the presence of oil wealth creates the incentive to do so, as control of the government is equated to de facto control of future petroleum revenues. The presence of oil, then, substantially increases the salience of the first postwar election, which incentivizes political entrepreneurs and elites within the government to use violence as a means of consolidating their control of the state.

To reiterate, it is the initial or first postwar election that increases the risk of peace failure as opposed to all elections following the end of the civil war. Subsequent postwar elections should carry less risk for a number of reasons. First, subsequent postwar elections should allow the state to rebuild much of its lost capacity. The environment that immediately follows the end of the civil war is characterized by a dearth of resources. As much of the wealth within the country has been looted or sent abroad, the postwar regime lacks the ability to appropriately police the population. Additionally, with damaged infrastructure and large portions of the population displaced, it is difficult for the military and the police to monitor dissidents, thereby giving opposition forces the opportunity to generate a military challenge in response to illicit electioneering. As the state is able to ramp up oil production in the future, it should be able to generate the ability to properly monitor the population.

Subsequent postwar elections should also be less contentious in that elites within the government may use previous electoral competition as a way to identify opponents within the population. Once elites consolidate their control of the state following the first election, they may identify past challengers and co-opt them through the use of oil rents. As noted by much of the literature on oil wealth and democracy, oil wealth has stabilizing properties (Morrison 2007, 2009; Gurses 2011; Smith 2004). Elites are able to use nontax revenue such as oil wealth so as to ameliorate citizen grievances and payoff opponents. By using the first postwar election as a learning experience, elites are able to identify the most dangerous challengers and pay them off before they can challenge them in future elections.

It should be noted that I do not expect oil wealth by itself to be the causal factor in instigating a postwar peace failure. Scholars who study the effects of natural resources on civil war onset have long noted that the presence of oil can instigate violent intrastate conflict (Ross 2004; Fearon 2005; Lujula 2010). Additionally, scholars of democratization have found that oil wealth can have a stabilizing effect on regimes, as they are able to use oil wealth to co-opt political opponents and ease dissent within the population through financial benefits (Ross 2001; Smith 2004; Gandhi and Przeworski 2006; Morrison 2007; Gurses 2011). I argue instead that the initial postwar elections in oil-rich countries following a civil war generate considerable risk to political elites in office, as an unfavorable result in the votes could risk them losing access to future oil revenue. Przeworski (1986) famously suggested that democracy is the process of institutionalizing uncertainty. I argue that given the uncertainty in election results, elites within the government should be willing to use coercive means to maintain their grasp on power and that the use of coercion spurs a new civil war. Given that the first election following civil wars is the most hazardous, as armed groups are the most organized and the future of elite control is still at risk (Brancati and Snyder 2013), the presence of oil further aggravates the electoral process.

Hypothesis 1: The first postwar elections in oil-rich countries increase the risk of a peace failure as compared to postwar elections in non-oil-rich countries.

A comparison of Angola and Liberia’s different experiences should underscore this point. As noted by Chaudrey (2012), the 2005 elections in Liberia were notable in that none of the rebel factions competed as parties. While some of the military leaders were elected, the vast majority of political parties and politicians were from civil society rather than being associated with military factions. Liberia’s experience contrasts remarkably with Brancati and Snyder’s (2013) accurate description of postwar elections, as they emphasize that the strongest parties are ones affiliated with armed groups. I maintain that Liberia’s divergent experience may be attributed to the fact that the nation was not a major oil exporter at the time of the Accra Peace Accords. Liberia’s relatively peaceful transition should be compared with the Angolan civil war or Iraq’s recent experience with the democratic process. Following the signing of the Bicesse Accords, both the Angolan government and the UNITA rebels agreed to postwar elections as part of the negotiated settlement. Unlike the Liberian experience, UNITA supporters and election monitors claimed that the 1992 election was fraught with widespread irregularities, triggering a resumption of violence throughout the country ( National Society for Human Rights, Namibia 2000). I argue that these divergent experiences with postwar elections can be attributed to the vast oil wealth within Angola as compared to Liberia. Faced with the risk of losing access to future oil revenues, the dos Santos government engaged in irregular electioneering as a means to ensure its grasp on power, thereby spurring opponents to resume fighting as a means of renegotiating the status quo. Along similar lines, the post-conflict stability generated from the co-optation of Sunni militants and the surge of American Troops, the Iraqi government began a fierce campaign to consolidate its control through targeting former Sunni allies. Actions such as charging the Sunni Vice President with organizing militias, as well as rounding up Sunnis for possible ties to militants, marginalized the Sunni population. The resurgence in violence has been attributed to this brutal strategy by the Maliki government (Ghazi and Arango 2014).

While other commodities may influence the willingness of leaders to use coercion to stay in office, oil should be of particular concern in the postwar environment. First, the scholarship on the connection between the resource curse and civil war has noted oil’s importance as a factor that spurs civil war onset, while other commodities merely impact the duration of civil wars (Ross 2004; Fearon 2005; Lujula 2010). Second, as noted earlier, as oil is not easily extracted or moved, the government should maintain an unfair advantage when it comes to local or foreign companies drilling and transporting the wealth abroad. This allows elites within the government to extract rents from those involved in the process of gathering and selling oil. Given that the postwar economy is typically defined by the dearth of economic activity, oil wealth should be of particular concern for postwar elites. Therefore, maintaining access to future oil wealth should be of particular concern to those within the postwar government. On the other hand, the absence of oil wealth should lower the stakes for postwar elections, as the profit opportunity for being in government is much lower. As there may be few options to extract revenue from the postwar economy (Collier 2009; Mason et al. 2011), elites should be more amenable to abdicating authority.²

Alternative Rival Hypothesis

One could easily posit an alternative rival hypothesis, that oil wealth should make the first postwar election more stable, thereby reducing the risk of peace failure. If elites within the postwar government have sufficient nontax revenue on hand, then they should be able to co-opt armed militant groups and reduce the risk of fighting. Much of the literature on democracy and oil wealth would support this conclusion. As noted by Morrison (2009), nontax revenue such as oil wealth has stabilizing properties. Given that the government does not need to tax either the rich (who would be prone to engage in a coup d’état) or the poor for wealth, it is able to support state goals without burdening the population. Additionally, Morrison suggests that governments who have access to nontax revenue are more likely to use that wealth to support social spending, generating more stability. Oil wealth can, in effect, ease tensions between the rich and the poor within countries, reducing the demand for a redistribution of wealth (Gurses 2011). As noted by Gurses (2011, 166), “an abundance of natural resources can lubricate interactions between actors, ameliorating social conflict and thereby stabilizing the regime.” Finally, Dunning (2005, 2008) proposes a much more nuanced approach to oil wealth and stability. Fearful of opposition elements in society, weak governments may prevent the diversification of the national economy, so that they maintain a monopoly on the national wealth, preventing potential opponents from accruing the resources necessary to challenge them.

If anything, this literature suggests that regimes that have access to oil wealth should be able to co-opt opponents and promote greater stability (Smith 2004). Within the postwar context, elites may be savvy enough to use oil wealth (or the promise of future oil wealth) to payoff opponents and generate greater regime stability. By offering some form of revenue sharing agreement, elites may co-opt opposition figures and reduce the desire for groups to control the state so as to benefit from oil wealth. As noted by Fjelde (2009), savvy leaders in oil-rich countries engage in co-optation of political opponents so as to prevent future armed resistance. If this process continues into the postwar environment, there should be sufficient reason to believe that elites should negotiate an equitable share of revenue to prevent future armed opposition.

Alternative Rival Hypothesis: The first postwar election in oil-rich countries should reduce the risk of peace failure.

Research Design

To evaluate the claim that elections that take place in immobile asset economies increase the risk of a new civil war, I examine all peace spells from 1946 to 2005. Peace spells are defined as the duration of peace that occurs following the end of a civil war (Mason et al. 2011). Peace spells end when a new civil war emerges within the state. Specifically, this phenomenon is referred to as peace failure (Walter 2004; Fortna 2004). The study of peace failure within the postwar environment is in essence the study of the durability of peace that occurs after civil wars end; given this point, researchers often study what impacts the durability of peace after civil wars end. My population of peace spells is drawn from the Joshi and Mason (2011) data set. This data set is an expansion of Sambanis’s (2004) data set where a civil war is coded as occurring if at least 500 individuals died in the first year. As noted by Sambanis (2004), his data set provides some nuanced techniques for sidestepping conceptual problems in coding what is and is not a civil war.³

My specific unit of analysis will be the country peace spell year. This differs from Brancati and Snyder’s analysis of post-conflict elections, as their unit of analysis is simply the peace spell. My decision to use the peace spell year is motivated by the fact that I will be using maximum likelihood (MLE) estimation, and relying on simply all peace spells will give me fewer than 100 observations. As MLE estimation relies on asymptotic properties for consistent results, the use of time series allows for more reliable estimates. In addition, variables such as Gross Domestic Product (GDP) and Population vary with time, which would be ignored. Consistent with other studies of peace duration (Mason et al. 2011; Joshi and Mason 2011; Walter 2004; Fortna 2004; Gurses, Rost, and McLeod 2008), my dependent variable peace failure will be operationalized as a binary indicator; with a one representing the start of a new civil war and a zero representing continued peace. As the analysis is of peace duration, countries that experience a new civil war will be dropped from the analysis and reenter when the civil war ends.⁴

Independent Variables

My election data are drawn from the Brancati and Snyder (2013) data set on post-conflict elections and peace failure.⁵ I measure the effects of the first national election that occurs after the civil war ends using a binary variable, with a one representing the year in which the election occurred and a zero otherwise. As noted by Flores and Nooruddine (2012), the first postwar election is typically the most hazardous to peace duration. Additionally, I include a binary variable for all subsequent elections (“Any Elections”) with a one representing the year in which an election occurs and a zero otherwise. Data for this second election variable are drawn from both Brancati and Snyder (2013) as well as Beck et al.’s (2001) data. Furthermore, I use Brancati and Snyder’s (2013) election timing data, to account for when the election occurs following the end of the civil war; specifically the time since the civil war ends that the election occurs. My primary variable (“first petro elections”) measures the first national elections that occur in oil-rich countries. To operationalize my primary independent variable (“first petro elections”), I generate an indicator that measures the impact of the first national election within an oil-rich country. As a dichotomous variable, the year in which a first national election occurs in an oil-rich country is coded as a one, with all other years coded as a zero. My variable for oil-rich countries (“Oil Dependence”) is drawn from Doyle and Sambanis’s (2006) data set, where a country is coded as being oil rich if at least 33 percent of its exports are from petroleum.⁶ To ensure that the presence of oil is not driving my results, I include the Oil Dependence control in all models tested. Finally, all election indicators will be lagged so as to further clarify the causal relationship. Cederman, Gleditsch, and Hug (2013) note that civil war violence that emerges as a result of elections typically takes place after electoral misconduct. Violence that occurs prior to the election tends to be low level, instrumental violence as a means for political leaders to insure electoral victory (Wilkinson 2004; Bratton 2008).⁷

Control Variables

As the effects of postwar elections may vary depending on the regime type of the country, I include a control variable for the presence of democracy within the country (a binary variable with a one representing a Polity II score of six or higher). In addition, I include a binary control variable to account for a country’s previous experience with democracy.^8,9 To account for the effects of the previous civil war on the likelihood of peace failure, I control for the outcomes of the previous conflict (Mason et al. 2011); whether or not the previous conflict was fought along ethnic lines (Joshi and Mason 2011); whether or not the previous conflict was meant to overthrow the government or as a secessionist conflict (Buhuag 2006); as well as the duration of the past conflict and the logged number of battle deaths from the previous conflict (Quinn, Mason, and Gurses 2007; Gurses 2009). Previous research has shown that previous conflicts fought over ethnic differences tend to lead to very tenuous post-conflict environments (Sambanis 2004). Furthermore, previous work has demonstrated that civil wars can be thought of as an information sharing process (Walter 2009). The longer the conflict lasts, the more information combatants have about one another, contributing to more durable peace. Conflict outcomes also shape the likelihood of a peace failure, as negotiated settlements tend to be much more fragile than military victories (Mason et al. 2011).

To account for the impact of power-sharing arrangements on the durability of peace, I include variables to account for the presence of political and military power-sharing arrangements that exist between the rebel group(s) and the government. The data on power-sharing arrangements are drawn from Hartzell and Hoddie (2003) as well as Joshi and Mason (2011). Following Joshi and Mason’s (2011) coding procedure, both military and political power-sharing arrangements are measured using a binary variable (with a one representing the presence of the particular arrangement and a zero if no arrangement is in place). Additionally, the variables are only coded if the power-sharing arrangement was actually implemented as opposed to simply agreed to.^10,11

Finally, I control for the presence of peacekeepers (Fortna 2004; Gurses, Rost, and McLeod 2008) as well as the lagged log of a country’s GDP. As shown by Walter (2004), a country’s economic development plays a role in whether dissidents seek to challenge the state in a new civil war. Additionally, past research has consistently demonstrated that the presence of peacekeepers reduces the likelihood of peace failure.¹²

All models are estimated using a Weibull Hazard model, which calculates the risk or “hazard” of peace failure occurring. Assuming that peace becomes more durable over time, the Weibull model would appear to be a more appropriate tool to analyze the impact of petro elections on peace failure. To identify which model is best for examining the impact of the first petro elections on peace failure, I run a distributional model selection test with Akaike information criterion and Bayesian information criterion. The Weibull Hazard model and the log logistic model were shown to be the best fit for the data. Therefore, I elected to use the Weibull model as well as run a robustness check with a log logistic model. I also present the results using a Cox Proportional Hazard model in the Appendix. The results appear to be largely unaffected by model specification.

Findings

I present the results in hazard ratios as opposed to coefficients, so as to allow for simple interpretations of the magnitude of each of the variables. All values that drop below “1” represent a decrease in the risk of peace failure, while all values that are over 1 represent increases in the risk of peace failure. Model 1 includes all control variables as well as the First National Election_t−1 within postwar countries. Model 2 includes both First National Election_t−1 and Any Election_t−1 as well as all of the control variables. Model 3 includes both First National Election_t−1 and First Petro Election_t−1 variables in addition to all of the control variables. Finally, model 4 includes all election variables as well as the control variables (Table 1).

Table 1.

The Impact of Oil Wealth on First National Elections and Peace Failure.

	Model 1	Model 2	Model 3	Model 4
Constant	0.001 (0.003)**	0.001 (0.03)**	0.001 (0.004)*	0.001 (0.004)*
War duration	0.99 (0.004)	0.99 (0.004)	0.99 (0.004)	0.99 (0.004)
Log of battle deaths	1.17 (0.11)*	1.16 (0.11)**	1.16 (0.11)	1.16 (0.11)*
Ethnic conflict	17.2 (14.5)***	17.0 (14.7)***	17.4 (14.8)***	17.1 (14.9)***
Oil dependence	2.97 (1.73)*	3.03 (1.80)**	2.63 (1.64)	2.69 (1.72)
Government incompatibility	0.31 (0.16)**	0.31 (0.16)**	0.33 (0.18)**	0.34 (0.18)**
First Petro Election _t−1			12.4 (12.8)**	12.3 (12.9)**
First National Election _t−1	0.88 (0.52)	0.73 (0.51)	0.48 (0.44)	0.40 (0.38)
Any Election _t−1		1.24 (0.60)		1.20 (0.57)
Election timing (national)	0.94 (0.04)	0.93 (0.11)	0.95 (0.04)	0.93 (0.10)
Election timing (any)		1.02 (0.12)		1.02 (0.11)
Democracy	0.98 (0.005)***	0.98 (0.005)***	0.98 (0.005)***	0.98 (0.005)***
Past democratic experience	2.05 (0.93)	2.06 (0.92)	2.12 (0.98)	2.12 (0.98)
Log of GDP_t−1	1.06 (0.31)	1.06 (0.32)	1.07 (0.32)	1.06 (0.33)
Log of population	0.97 (0.31)	0.98 (0.33)	0.94 (0.31)	0.95 (0.33)
Peacekeeping operation	1.84e-07 (1.84e-07)***	1.40e-07 (1.41e-07)***	1.92e-07 (1.94e-07)***	1.90e-07 (1.94e-07)***
Negotiated settlement	0.69 (0.41)	0.70 (0.40)	0.70 (0.43)	0.70 (0.41)
Government victory	0.30 (0.15)**	0.30 (0.16)**	0.31 (0.17)**	0.31 (0.17)**
Government power-sharing arrangement	1.43 (0.78)	1.40 (0.77)	1.43 (0.78)	1.40 (0.77)
Military power-sharing arrangement	0.35 (0.19)*	0.36 (0.19)**	0.36 (0.20)*	0.37 (0.20)*
P	1.25	1.26	1.26	1.26
N	1,024	1,024	1,024	1,024
Wald χ ²	696.02***	733.81***	1,128.75***	1,131.44***

Note: Hazard ratios with robust standard errors in parentheses presented above. Two-tailed test. Results clustered around country codes. GDP = gross domestic product.

*p < .10.

**p < .05.

***p < .01.

The first and second models suggest that the election variables, both the first national election and all subsequent elections, have no statistical impact on the risk of peace failure. Model 3 suggests strong statistical support for Hypothesis 1, as the occurrence of the first petro election appears to have a positive relationship with peace failure, rejecting the null hypothesis with over 95 percent confidence. In addition to its statistical significance, the first petro election has a large marginal effect on peace failure, increasing the risk of peace failure by 1,140 percent from the baseline hazard rate. The effects of the first petro election appear to be independent of the effects of oil within the country. This would suggest that a reliance on oil may not be the decisive factor in spurring peace failure, but the possible risk that elites may lose access to it. The relationship between first petro elections and peace failure is presented graphically in Figure 1. The value of first petro elections is changed from zero to one, with all other variables held constant at their means.

Figure 1.

The impact of oil wealth on first national elections and peace failure.

The impact of the first petro elections also appears to be independent of a country’s past experience with democracy as well as the presence of working democratic institutions. While democracy appears to reduce the risk of peace failure (p > .01), it does not appear to remove the impact of postwar petro elections. Furthermore, the impact of democracy appears to be rather weak, as more democratic institutions only reduce the risk of peace failure by 2 percent. This would suggest that there must be more to the story than simply democratic institutions.

Finally, both ethnic-based conflicts and the presence of peacekeeping forces appear to be highly significant with large marginal effects which fit with much of the previous literature on peace duration. Furthermore, military power-sharing arrangements, incompatibility with the government, government victories, and the log of the previous war’s battle deaths were significant. Again, these findings fit with much of the previous literature on the determinants of peace failure. While the connection between government victories and greater peace duration has been challenged, findings by Kreutz (2010) and Mason et al. (2011) tend to underscore that government victories in civil wars generally stabilize the postwar peace process. Interestingly, negotiated settlements as well as population and the country’s wealth fail to meet statistical significance. Robustness checks with infant mortality rates as opposed to the log of a country’s GDP demonstrate that economic factors fail to change the significance or marginal effects of the election data.

Qualitative Evidence: The Insurgency in Aceh

As demonstrated by the data, the presence of oil greatly impacts the risk of the first postwar election. Unfortunately, statistical analysis is limited in its ability to underscore how the causal mechanisms connect to the phenomenon or phenomena under investigation. Researchers must often incorporate qualitative examples to demonstrate their point. To further illustrate the impact that oil has on the first postwar elections, I have included a brief discussion of the second Aceh rebellion (Aspinall 2009). Unlike many of the islands in Indonesia, Aceh is rich in natural resources. The island has considerable deposits of natural gas and oil. These oil wells became (and still are) a major source of income for the Indonesian government, with the military receiving large portions of the resource revenue to fund everything from officer salaries to military equipment (Walter 2009). By the 1980s, over 30 percent of Indonesia’s oil exports were being extracted from Aceh (Aspinall 2009). As noted by Aspinall (2009) and Walter (2009), this dependence on Aceh’s oil and gas revenue has shaped much of the state’s policy in regard to allowing greater autonomy for the island. Oil wealth has reduced the incentive to negotiate over the years and has pressed the military to further consolidate its control over the territory and its people. In Michael Ross’s (2003) study of the civil war in Aceh, he notes that the presence of oil increased the military’s incentive to use coercive means to maintain control of the island.

Not only was the army able to fund itself through the profits of the exported oil, but the government was able to monopolize the production process. Mobil Oil, the company that did much of the drilling on the island, was forced to rely heavily on the Indonesian government for production and for transportation of the product (Coll 2013). While the government was able to profit heavily from the oil production, much of the rural indigenous population of Aceh was unable to benefit from much of the revenue generated by the drilling. On the other hand, the rural population suffered the most from costs of oil extraction: specifically, the confiscation of their lands for drilling purposes as well as the severe pollution that resulted from the extraction process (Ross 2003).

The Gerakan Aceh Merdeka (GAM), or “Free Aceh Movement,” began in earnest in 1976 as nationalist forces within the Island of Aceh began advocating for independence from Indonesia (Ross 2003). Motivated in part by grievances against state abuses as well as a greater demand for self-determination, the GAM moved from citizen demands for autonomy into a separatist insurgency (Aspinall 2009). This initial rebellion was quickly repressed, and the Indonesian state had effectively consolidated its control of the island (Ross 2003). Following the defeat of the Acehnese insurgents, the government sought to further consolidate its control of the island. As noted by Walter (2009), the military had grown to rely heavily on the oil produced on the island. In order to eliminate any threat to that control, the government effectively rigged the electoral process, so that pro-independence parties in Aceh would be unable to sufficiently gain access to the state. While Indonesia was effectively under Suharto’s rule, there were still competitive elections within the country. Following the end of the first Aceh insurgency, opposition parties won handily in elections in Aceh, specifically Partai Persatuan Pembangunan (PPP; Aspinall 2009). Despite these victories, the government instituted a barrier between electoral outcomes and gaining access to executive office. As the military was able to fill specific seats within the local government, opposition parties were unable to gain majorities in the legislature. This came at a time when the central government failed to institute many of the promises it had made while campaigning on the island (Ross 2003).

The absence of opposition voices within the legislature created a political vacuum that the GAM was able to fill. Given the government’s desire to maintain control of the oil production in Aceh, they reformed the institutions and participated in illicit electioneering so as to reduce the ability for opposition forces to challenge their control. Additionally, Ross (2003) notes that the use of human rights abuses by the state and the knowledge among Acehnese that the government relied heavily on oil and gas revenues generated great distrust of the central government. Following the civil war, GAM was one of the few organized groups operating in Aceh. With most political parties unable to voice opposition to government practices through electoral institutions, the GAM was able to capitalize on the postwar political grievances and generate more support for their cause. Aspinall (2009) notes that GAM generated cross-cultural ties in this period so as to reinvigorate the insurgency and restart the civil war by 1982. Notably, while the initial insurgency may not have started over the natural resources (Aspinall 2007), the oil and gas wealth became a prominent issue among the insurgents during the second uprising. With the central government effectively stealing the election in Aceh, local antipathy was stoked by the perception that Acehnese had little control over the oil and gas wealth drilled from their territory. If anything, Aceh’s civil war recurrence underscores the role oil plays in impacting the postwar electoral process. The presence of oil heavily influenced the government’s policy toward Aceh, especially with regard to their role in rigging the electoral process. Oil also played a role in shaping the population’s perception of the government’s behavior. Given that the Acehnese were unable to make their way into local or national offices, their ability to profit from oil wealth was greatly diminished.

Conclusion

In the results presented above, I build on the findings of Brancati and Snyder (2013) as well as Flores and Nooruddin (2012) to suggest that oil plays a prominent role in postwar elections. Specifically, the postwar elections that occur in countries that are heavily dependent on oil exports are much more contentious than elections that occur in the absence of oil-rich countries. Drawing on previous work by Boix (2003) and Acemoglu and Robinson (2001), I find that the initial postwar elections that occur in oil-rich countries present a risk to postwar elites. A loss at the polls would eliminate government elite access to oil revenues for personal gain as well as for support from those within their winning coalition. Given the risk that government officials will lose access to future oil revenue, elites will be more likely to engage in illicit electioneering or open violence as a means to hang onto power, thereby spurring their opponents to use violence. On the other hand, elections that occur in the absence of oil revenues will be less likely to restart the civil war as the opportunity for rentierism is far less, as many of the assets within the country were likely removed during the civil war. Given the reduction in potential profit from being in office, elites may be more amenable to cooperating with the democratic process.

This research also begins to unpack the role democracy plays in promoting postwar peace. Elections, though, are only on aspect of democracy. It would be informative to investigate some of the substantive aspects of the democratic process such as citizen participation, access to civil liberties and rights (which are often curtailed during wars), and the nature of the country’s party system. Finally and most importantly, the results demonstrate that the context in which postwar elections occur also has a significant impact on whether elections contribute to peace failure. While this article specifically looks at the role oil plays in making postwar elections detrimental to peace duration, there are other contentious issues within postwar society. For instance, the role of ethnic parties may make peace duration far less feasible following wars fought along ethnic lines.

One other avenue that should be investigated is whether other immobile assets play a role in making elections more or less contentious following the end of the civil war. Again, the results presented above can only be applied to the role oil plays in enhancing the risk of the initial postwar election. But the theory presented here borrows heavily from Boix’s (2003) conclusion about the role immobile assets play in disrupting the democratization process. While oil may be representative of immobile assets in general, there are many different forms of asset specificity which may impact the salience of postwar elections.¹³

Finally, scholars of peace science should be interested in what ways the international community can ameliorate the impact of oil on postwar elections. Given that oil dramatically increases the salience of postwar elections, we should be interested in which institutions or forms of peace agreements that can reduce the risk of peace failure. As the data suggest, subsequent elections can actually reduce the risk of peace failure within postwar countries. Given that this one aspect of democracy can assist with future peace in war torn countries, it would be incumbent on scholars to identify policies that make elections less risky within these countries. This may, of course, lead researchers to conclude that elections are not feasible for many postwar countries. For instance, while postwar elections may be useful in promoting durable peace in the Solomon Islands, they may not be a proper tool for promoting peace within South Sudan or Iraq. Rather, perhaps other aspects of democracy should be fostered so as to promote a more durable peace within the future.

Footnotes

Appendix

Correlation tests	Oil dependence	First Petro Election _t−1	First National Election _t−1
Oil dependence	1.0
First Petro Election _t−1	0.1174	1.0
First National Election _t−1	0.0171	0.2625	1.0
Cox Proportional Hazard models		Cox Proportional Hazard models
War duration		0.99 (0.003)
Log of battle deaths		1.17 (0.12)*
Ethnic conflict		12.7 (8.31)***
Oil dependence		2.12 (1.08)
Government incompatibility		0.44 (0.20)*
First Petro Election _t−1		14.7 (18.5)**
First National Election _t−1		0.75 (0.50)
Any Election _t−1		0.40 (0.25)
Election timing (national)		0.95 (0.08)
Election timing (any)		1.01 (0.08)
Democracy		0.98 (0.006)**
Past democratic experience		2.06 (0.91)
Log of GDP_t−1		1.10 (0.24)
Log of population		0.92 (0.20)
Peacekeeping operation		Dropped
Negotiated settlement		0.76 (0.36)
Government victory		0.34 (0.19)**
Government power-sharing arrangement		1.00 (0.47)
Military power-sharing arrangement		0.52 (0.30)
N		1,024
Wald χ²		210.25***

Note: Hazard ratios with robust standard errors in parentheses presented above. Two-tailed test. Results clustered around country codes. GDP = gross domestic product.

*p < .10.

**p < .05.

***p < .01.

Declaration of Conflicting Interests

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

Funding

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

Supplemental Material

The supplemental materials are available at .

Notes

References

Acemoglu

Daron

Robinson

James

. 2001. “A Theory of Political Transitions.” American Economic Review 91:938–63.

Addison

Tony

Geda

Alemayehy

Billon

Philippe Le

Murshed

S. Mansoob

. 2005. “Reconstructing and Reforming the Financial System in Conflict and ‘Post-conflict’ Economies.” Journal of Development Studies 41 (4): 703–18.

Aspinall

Edward

. 2007. “The Construction of Grievance: Natural Resources and Identity in a Separatist Conflict.” Journal of Conflict Resolution 51 (6): 950–72.

Aspinall

Edward

. 2009. Islam and Nation: Separatist Rebellion in Aceh, Indonesia. Palo Alto, CA: Stanford University Press.

Autesserre

Séverine

. 2009. “Hobbes and the Congo: Frames, Local Violence, and International Intervention.” International Organization 63 (2): 249–80.

Beck

Thorsten

Clarke

George

Groff

Alberto

Keefer

Philip

Walsh

Patrick

. 2001. “New Tools in Comparative Political Economy: The Database of Political Institutions.” World Bank Economic Review 15 (1): 165–76.

Bigombe

Betty

Collier

Paul

Sambanis

Nicolas

. 2000. “Policies for Building Post-conflict Peace.” Journal of African Economies 9 (3): 323–48.

Binningsbo

Helga Malmin

Rustad

Siri Aas

. 2012. “Sharing the Wealth: A Pathway to Peace or a Trail to Nowhere?” Conflict Management and Peace Science 29 (5): 547–66.

Boix

Carles

. 2003. Democracy and Redistribution. Cambridge, UK: Cambridge University Press.

10.

Brancati

Dawn

Snyder

Jack L.

. 2013. “Time to Kill: The Impact of Election Timing on Postconflict Stability.” Journal of Conflict Resolution 57 (5): 822–53.

11.

Bratton

Michale

. 2008. “Vote Buying and Violence in Nigerian Election Campaigns.” Electoral Studies 27: 621–32.

12.

Bueno de Mesquita

Bruce

Smith

Alastair

Siverson

Randolph M.

Morrow

James D.

. 2003. The Logic of Political Survival. Cambridge, MA: The MIT Press.

13.

Buhaug

Halvard

. 2006. “Relative Capability and Rebel Objective in Civil War.” Journal of Peace Research 43 (6): 691–708.

14.

Büthe

Tim

Milner

Helen V.

. 2008. “The Politics of Foreign Direct Investment into Developing Countries: Increasing FDI through Trade Agreements.” American Journal of Political Science 52 (4) (October): 741–62.

15.

Cederman

Lars-Erik

Gleditsch

Kristian Skrede

Hug

Simon

. 2013. “Elections and Ethnic Civil War.” Comparative Political Studies 46 (3): 387–417.

16.

Chaudrey

Torunn Wimpelmann

. 2012. “Political Economies of Violence in Post-war Liberia.” In The Peace in Between: Post-war Violence and Peacebuilding, edited by Suhrke

Astri

Berdal

Mats

, 248–66. New York: Routledge.

17.

Coll

Steve

. 2013. Private Empire: ExxonMobil and American Power. New York: Penguin Books.

18.

Collier

Paul

. 2009. “Post-conflict Recovery: How Should Strategies be Distinctive?” Journal of African Economies 18:99–131.

19.

Collier

Paul

Elliot

V. L.

Hegre

Havard

Hoeffler

Anke

Reynal-Querol

Marta

Sambanis

Nicholas

. 2003. Breaking the Conflict Trap: Civil War and Development Policy. Oxford, UK: Oxford University Press.

20.

Davenport

Christian

Armstrong

David

. 2004. “Democracy and the Violation of Human Rights: A Statistical Analysis from 1976-1996.” American Journal of Political Science 48 (3): 538–54.

21.

Derouen

Karl

Lea

Jenna

Wallensteen

Peter

. 2009. “The Duration of Civil War Peace Agreements.” Conflict Management and Peace Science 26 (4): 367–87.

22.

Doyle

Michael W.

Sambanis

Nicholas

. 2006. Making War and Building Peace: United Nations Peace Operations. Princeton, NJ: Princeton University Press.

23.

Dunning

Thad

. 2005. “Resource Dependence, Economic Performance, and Political Stability.” Journal of Conflict Resolution 49 (4): 451–82.

24.

Dunning

Thad

. 2008. Crude Democracy: Natural Resource Wealth and Political Regimes. New York: Cambridge University Press.

25.

Dunning

Thad

. 2011. “Fighting and Voting: Violent Conflict and Electoral Politics.” Journal of Conflict Resolution 55 (3): 327–39.

26.

Fearon

James D.

2005. “Primary Commodity Exports and Civil War.” Journal of Conflict Resolution 49 (4): 483–507.

27.

Fearon

James D.

Laitin

David D.

. 2003. “Ethnicity, Insurgency, and Civil War.” American Political Science Review 97 (1): 75–90.

28.

Fein

Helen

. 1995. “More Murder in the Middle: Life-integrity Violations and Democracy in the World, 1987.” Human Rights Quarterly 17 (1): 170–91.

29.

Fjelde

Hanne

. 2009. “Buying Peace? Oil Wealth, Corruption, and Civil War, 1985-99.” Journal of Peace Research 46 (2): 199–218.

30.

Flores

Thomas Edward

Nooruddin

Irfan

. 2012. “The Effect of Elections on Post Conflict Peace and Reconstruction.” The Journal of Politics 74 (2): 558–70.

31.

Fortna

Virginia Page

. 2003. “Inside and Out: Peacekeeping and the Duration of Peace after Civil and Interstate Wars.” International Studies Review 5 (4): 97–114.

32.

Fortna

Virginia Page

. 2004. “Does Peacekeeping Keep Peace? International Intervention and the Duration of Peace after Civil War.” International Studies Quarterly 48 (2): 269–92.

33.

Gandhi

Jennifer

Przeworski

Adam

. 2006. “Cooperation, Cooptation, and Rebellion under Dictatorships.” Economics and Politics 18 (1): 1–26.

34.

Ghazi

Yasir

Arango

Tim

. 2014. “Qaeda-linked Militants in Iraq Secure Nearly Full Control of Falluja.” New York Times. Accessed January 20, 2014. http://www.nytimes.com/2014/01/05/world/middleeast/shelling-in-iraqi-city-held-by-qaeda-linked-militants-kills-at-least-8.html.

35.

Gurses

Mehmet

. 2009. “State-sponsored Development, Oil and Democratization.” Democratization 16 (3): 508–29.

36.

Gurses

Mehmet

. 2011. “Elites, Oil, and Democratization: A Survival Analysis.” Social Science Quarterly 92 (1): 164–84.

37.

Gurses

Mehmet

Rost

Nicolas

McLeod

Patrick

. 2008. “Mediating Civil War Settlements and the Duration of Peace.” International Interactions 34 (2): 129–55.

38.

Hartzell

Caroline A.

1999. “Explaining the Stability of Negotiated Settlements to Intrastate Wars.” Journal of Conflict Resolution 43 (3): 2–22.

39.

Hartzell

Caroline A.

Hoddie

Mathew

. 2003. “Institutionalizing Peace: Power Sharing and Post-civil War Conflict Management.” American Journal of Political Science 47 (2): 318–22.

40.

Hartzell

Caroline A.

Hoddie

Matthew

Rothchild

Donald

. 2001. “Stabilizing the Peace after Civil War: An Investigation of Some Key Variables.” International Organization 55 (1): 183–208.

41.

Hegre

Havard

Ellingsen

Tanja

Gates

Scott

Gleditsch

Nils Petter

. 2001. “Toward a Democratic Civil Peace? Democracy, Political Change, and Civil War, 1816-1992.” American Political Science Review 95 (1): 33–48.

42.

Jensen

Nathan

. 2003. “Democratic Governance and Multinational Corporations: Political Regimes and Inflows of Foreign Direct Investment.” International Organization 57:587–616.

43.

Joshi

Madhav

. 2013. “Inclusive Institutions and Stability of Transition toward Democracy in Post-civil War States.” Democratization 20 (4): 743–70.

44.

Joshi

Madav

Mason

T. David

. 2011. “Civil War Settlements, Size of Governing Coalition, and Durability of Peace in Post-civil War States.” International Interaction 37 (4): 388–413.

45.

Kreutz

Joakim

. 2010. “How and When Armed Conflicts End: Introducing the UCDP Conflict Termination Dataset.” Journal of Peace Research 47 (2): 243–50.

46.

Lujala

Paivi

. 2010. “The Spoils of Nature: Armed Civil Conflict and Rebel Access to Natural Resources.” Journal of Peace Research 47 (1): 15–28.

47.

Mason

T. David

Gurses

Mehmet

Brandt

Patrick T.

Quinn

Jason Michael

. 2011. “When Civil Wars Recur: Conditions for Durable Peace after Civil Wars.” International Studies Perspectives 12:171–89.

48.

Mattes

Michaela

Savun

Burcu

. 2009. “Fostering Peace after Civil War: Commitment Problems and Agreement Design.” International Studies Quarterly 53:737–59.

49.

Morrison

Kevin M.

2007. “Natural Resources, Aid, and Democratization: A Best-case Scenario.” Public Choice 131:365–86.

50.

Morrison

Kevin M.

2009. “Oil, Nontax Revenue, and the Redistributional Foundations of Regime Stability.” International Organization 63:107–38.

51.

National Society for Human Rights. 2000. “Ending the Angolan Conflict: Our Time Has Come to be Heard.” Accessed August 1, 2014. http://www.paulbogdanor.com/left/angola/ending.pdf.

52.

Nossiter

Adam

. 2014. “Governor of Nigeria’s Central Bank Is Fired After Warning of Missing Oil Revenue.” The New York Times, February 20, 2014. Accessed May 5, 2014. http://www.nytimes.com/2014/02/21/world/africa/governor-of-nigerias-central-bank-is-fired-after-warning-of-missing-oil-revenue.html?_r=0.

53.

Przeworski

Adam

. 1986. “Some Problems in the Study of the Transition to Democracy.” In Transitions from Authoritarian Rule: Comparative Perspectives, Vol. III, edited by O’Donnell

Guillermo

Schmitter

Philippe C.

Whitehead

Laurence

, 47–63. Baltimore, MD: The Johns Hopkins University Press.

54.

Quinn

Jason Michael

Mason

T. David

Gurses

Mehmet

. 2007. “Sustaining the Peace: Determinants of Civil War Recurrence.” International Interactions 33 (2): 167–93. Accessed May 1, 2013. http://www.psci.unt.edu/∼mg003/SustainingthePeace-II.pdf.

55.

Roeder

Philip G.

Rothchild

Donald

, eds. 2005. Sustainable Peace: Power and Democracy after Civil Wars. London, UK: Cornell University Press.

56.

Roessler

Philip

. 2011. “The Enemy Within: Personal Rule, Coups and Civil War in Africa.” World Politics 63 (2): 300–46.

57.

Ross

L. Michael

. 2001. “Does Oil Hinder Democracy?” World Politics 53 (April): 325–61.

58.

Ross

L. Michael

. 2003. “Resources and Rebellion in Aceh, Indonesia.” In Understanding Civil War: Europe, Central Asia and Other Regions, edited by Collier

Paul

Sambanis

Nicholas

, 35–58. Washington, DC: The World Bank.

59.

Ross

Michael L.

2004. “What Do We Know about Natural Resources?” Journal of Peace Research 41 (3): 337–56.

60.

Sambanis

Nicholas

. 2004. “What Is a Civil War? Conceptual and Empirical Complexities of an Operational Definition.” Journal of Conflict Resolution 48 (6): 814–58.

61.

Smith

Benjamin

. 2004. “Oil Wealth and Regime Survival in the Developing World, 1960-1999.” American Journal of Political Science 48 (2): 232–46.

62.

Suhrke

Astri

Berdal

Mats

, eds. 2012. The Peace in Between: Post-war Violence and Peacebuilding. New York: Routledge.

63.

The Economist . 2014. “Till Much Too Oily: Angola Badly Needs to Diversify its One Dimensional Economy.” April 12, 2014. Accessed May 5, 2014. http://www.economist.com/news/middle-east-and-africa/21600693-angola-badly-needs-diversify-its-one-dimensional-economy-still-much-too-oily.

64.

Tilly

Charles

. 2003. The Politics of Collective Violence. Cambridge: Cambridge University Press.

65.

Walter

Barbara F.

2004. “Does Conflict Beget Conflict? Explaining Recurring Civil War.” Journal of Peace Research 41 (3): 371–88.

66.

Walter

Barbara F.

2009. Reputation and Civil War: Why Separatist Conflicts Are So Violent. New York: Cambridge University Press.

67.

Wilkinson

Steven

. 2004. Votes and Violence: Electoral Competition and Ethnic Riots in India. New York: 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.

0.65 MB

0.00 MB