Explaining Peace Duration through Foreign Aid and Civil War Outcome

Abstract: 

This study explores the relationship between foreign aid and civil war outcome and analyzes the impact these factors have on peace duration in a post-conflict environment. Previous literature has focused on the factors of war duration and civil war outcome (Brandt et al. 2008; Mason et al. 2011), while other scholars have observed the effect that the amount or timing of foreign aid has on political stability (Breuning and Ishiyama 2007). Previous research has not, however, examined the relationship between civil conflict outcome and sources of foreign aid to observe how the interaction of these factors may affect peace duration in a post-conflict environment. The theory holds that bilateral aid provides biased support when giving aid to recipient countries by supporting one group over another, which may decrease peace duration depending on the civil war outcome. Interaction variables are created to measure the relationship between civil conflict outcome and foreign aid in a post-conflict environment. A logit regression model is used to test my theory regarding the relationship between the interaction of foreign aid and civil war outcome with peace duration. Ultimately, I found that the relationship between the civil war outcomes of government victory and truce and a high bilateral aid ratio has a significant effect on peace duration. Specifically, a high proportion of bilateral aid in a truce civil war outcome has a greater probability of durable peace than a government victory in a post-conflict environment.

Table of Contents: 

    Introduction

    Since 1956, Sudan has been mired in a period of violent civil conflict. In 2005, the contending groups signed the Nairobi Comprehensive Peace Agreement, which called for a referendum on the status of the southern aspect of Sudan in six years, but also allowed the opposing groups to remain armed, which incited conflict in border regions. Recently, on July 9, 2011, after this lingering civil war, the south successfully seceded from the north, creating a new state. The conflict stemmed from ethnic and economic differences between the north and south. Non-Arab (including animist and Christian) groups located in the south held longstanding grievances towards the Sudanese government, which was dominated by a minority group of Arab and Muslim people in the north. The grievances of the groups in the south included political and economic exclusion despite the mass reserves of crude oil in the south. Furthermore, despite the natural resource richness in the region, the south remained underdeveloped and had limited infrastructure in comparison to the northern region of Sudan. The case study of Sudan demonstrates how a durable peace may be difficult to achieve in a post-conflict environment while dual sovereignty continues between the government and rebel groups. Dual sovereignty arises when the government no longer solely controls the primary means of coercion within the state and a challenge develops for the legitimate sovereignty over the state between the government and groups outside of the government (Tilly 1978). Some main factors that help groups acquire dual sovereignty in the state include military strength and financial sustainability, so weak governance may create government instability and economic frailty. Post-conflict environment countries are in need of international assistance to help sustain the economy and keep state legitimacy; therefore, foreign aid is given to developing countries. Foreign aid assistance is given from a variety of donor incentives; some include international goals, while others include goals that may only benefit the donors. Multilateral aid is given by international organizations which aim to help recipient countries that are most in need of aid. Bilateral aid donors incorporate policies in exchange for aid, which may support a particular party or political actor and may not be beneficial for the recipient country. As a result, aid donors who do not focus on implementing policies to increase political and economic stability may harm the recipient state and incite tension among belligerents. Therefore, ongoing civil conflict suggests the question: How does the interaction between different sources of aid and the outcome of the civil conflict affect durable peace?

    Developing countries that have weak governments and slow economic growth are more likely to experience difficulties preserving peace in a post-conflict environment. Collier et al. (2003) argue that the key root cause of conflict is the failure of economic development. Poverty is the one commonality of all countries that experience civil war, and civil war itself has drastic economic consequences in the post-conflict environment, thus developing a conflict trap. Fearon (2004) argues that the number of civil wars does not represent an increasing number of new civil wars but rather a number of unresolved wars. Therefore, unresolved wars provide a path to the conflict trap in developing countries (Fearon 2004: Collier et al. 2003).

    The concepts of civil war outcome and foreign aid have been analyzed separately by previous scholarship, but none of the previous work has tested the interaction of the factors. Previous studies disagree on the impact of foreign aid in post-conflict settings as some scholars argue that foreign aid helps countries adopt practices to further develop their economy (Collier and Hoeffler 2002). However, others argue that foreign aid does not promote economic growth and peace, but rather incites civil conflict (Grossman 1992). On the other hand, it has been argued that the impact of civil conflict outcome (which may be rebel victory, government victory, or negotiated settlements), may have a different effect on peace duration (Brandt et al. 2008). Thus far, studies have not analyzed the impact that different sources of aid may have on post-conflict environments based on the type of civil war outcome.

    The objective of this research is to explain how bilateral aid may affect dual sovereignty in accordance with the civil war outcome, whether it is government victory, rebel victory, truce, or negotiated settlement. This article will help evaluate whether bilateral aid has an effect on dual sovereignty and if previous research overlooked a significant factor that may affect peace duration. This paper will first review existing literature on foreign aid and civil conflict outcome and the rationale scholars use to argue that these factors may affect peace duration in a post-conflict setting. The second section of the article describes my theory regarding the relationship that foreign aid and the civil war outcome have with dual sovereignty and their effect on the duration of peace in a post-conflict environment. The third section discusses the methodology used to test my theory and the operationalization of my dependent and independent variables. The fourth section of the study analyzes the results of the test and states the implications in relation to the theory. Finally, the fifth section of the article discusses my conclusions from this research and future consideration for research regarding foreign aid and civil war outcome that are not addressed in this paper.

    Literature Review

    Previous studies argue that the duration of civil war, rebel incentives, and rebel costs are a few factors that may affect the duration of peace in a post-conflict environment (Mason et al. 2011). Tilly (1978) argues that dual sovereignty (which involves a legitimate group that has power over a state and an opposition group challenging such power) may lead to conflict due to competing means of obtaining power of the state. He claims that the government has the choice of either satisfying the opposition by allowing them to become part of the polity or maintaining the status quo with the opposition remaining outside of the polity, in which case the opposition may decide to resort to political violence (Tilly 1978). Unresolved wars, therefore, may increase the opportunity for dual sovereignty because it decreases the government’s legitimacy of power. On a similar note, Mason et al. (2011) and Brandt et al. (2008) argue that peace duration after civil war depends on whether the civil war ended in a rebel victory, a government victory, or a negotiated settlement. The outcome of a civil war is an important factor in peace duration because it may lead to dual sovereignty if one group did not gain complete control over the state.

    Scholars have analyzed previous civil war outcomes and their impact on peace duration in a post-conflict environment. Kreutz (2010) argues that if a previous conflict is fought with rebels aiming for total control over government, the risk of conflict recurrence increases. On the other hand, if total government control is not the aim of the rebel group, then government victories or the deployment of peacekeepers should reduce the likelihood that conflict will recur (Kreutz 2010). In addition, other studies argue that rebel victories produce a more stable peace than government victories, but only if the new regime controlled by former rebels survives the first few years after the conflict (Brandt et al. 2008; Quinn et al. 2007).

    Tilly (1978) and Walter (2004) argue that a rebel group may try to gain recognition by persuading a sizeable percentage of the population that it holds sovereignty, which creates a dual sovereignty between the government and the rebel group. Walter (2004) argues that people will not join a rebel group if the opportunity costs are high. Opportunity costs may increase if the population does not see the rebel group as having sovereignty, or if the population does not have a particular objective (such as increasing income or gaining rights) for joining the rebel group (Walter 2004). Civil wars typically last longer if there is an unequal distribution of income and a low average income within a state because the opportunity costs for rebellion are low and the government has a low state capacity (Collier et al. 2003). On a similar note, Collier et al. (2003) argue that the key root cause of conflict is the failure of economic development. One possible contribution to economic development in a post-conflict setting is foreign aid. Quinn et al. (2007) argue that the international community has policy tools at its disposal to structure intervention strategies that will reduce the probability of civil war recurrence.

    Bueno de Mesquita (2010) describes foreign aid as the contribution of money, goods, or services to other governments or to “people” or “communities” in foreign countries. Foreign aid can be given as large subsidized loans, for free or for a price below the market price (Bueno de Mesquita 2010). He also argues that increasing aid may help alleviate poverty and therefore help to prevent civil conflict from recurring, but Grossman (1992) argues that aid increases the chance of civil conflict due to the resource opportunity in foreign aid. Similarly, Breuning and Ishiyama (2007) argue that neither the amount nor the timing of foreign aid is related to political stability in a post-conflict environment. They suggest that donors should encourage political stability in post-conflict societies by advocating political institutions to promote the inclusiveness of political groups.

    Theory

    Peace duration in a post-conflict environment is essential in developing countries because peace may enhance economic growth and government legitimacy. Civil conflict damages infrastructure, and diminishes the income and self-confidence of the population, so the duration of peace helps regain what was lost. Recurring civil wars have been proven to be more common than new civil wars in recent research, which implies that post-conflict environments are significantly important to analyze in peace duration research (Brandt et al. 2005). Recurring wars may be caused by actors who are politically interested in gaining some, if not all, political sovereignty from the developing state. Tilly (1978) argues that opposition groups begin civil wars to challenge the government’s power. The rebel group’s ultimate goal is to gain control over the state (Tilly 1978). Thus, dual sovereignty may cause political and economic instability because only one group of belligerents may completely accomplish such a goal.

    In a post-conflict environment, the government strives to maintain legitimacy in order to avoid reemergence of dual sovereignty. Post-conflict environments are economically devastated and many accept outside assistance to sustain peace and to help maintain political and economic stability. Given that developing countries generally endure slow economic growth and many receive foreign aid, outside actors often take advantage of the financial need of developing states to make agreements that may benefit the donor. As a result, different sources of aid may affect the recipient country in terms of the duration of peace in a post-conflict environment.

    Bilateral aid and multilateral aid differ in many regards, particularly policies, type, and amount, as well as incentives of donors. Understanding differences in the sources of foreign aid will help determine if the results of previous research overlooked a significant gap by including bilateral and multilateral aid as a single source. Many recipient states receive aid from numerous bilateral and multilateral donors which vary in policies and conditions. The requirements from donors may vary according to the amount of aid, and donor-recipient relationship with the country. Bilateral aid differs from multilateral aid in that bilateral aid has a stronger donor-recipient relationship between the parties. Bilateral donors typically receive preferential treatment within the recipient country and choose to give aid based on factors such as whether the recipient is a former colony of the donor, they share a common language, or have transitioned to a democracy, to name just a few factors. As a consequence, this may provoke competition between the former belligerents in the recipient state particularly if the bilateral donor supports one group over another. Typically, the donor gives aid to the recipient’s government and not to a rebel group. According to Bueno de Mesquita (2010), foreign aid (bilateral) enriches petty dictators and secures them in office. Thus, bilateral aid may increase the tension between belligerents, which may lead to the recurrence of dual sovereignty.

    On the other hand, multilateral aid such as International Monetary Fund loan agreements, contain formal performance criteria including economic policy targets. Although disagreements arise before a consensus is reached, such agreements offer many advantages in policies toward more efficient economic growth and political stability in recipient countries. Multilateral aid donors distribute to all the recipient countries which belong to a particular international organization. Consequently, multilateral aid is more likely to have policies and economic distribution that may benefit a larger percent of the population in the recipient state in comparison to bilateral aid.

    Bilateral aid may have a varied effect in a post-conflict setting depending on the civil war outcome. The civil war outcome has a significant effect because it influences the condition of dual sovereignty within the post-civil war environment. A government is typically weak in a post-conflict environment because it is trying to rebuild its legitimacy. Thus, a rebel victory, government victory, or negotiated settlement may affect the duration of peace in a post-conflict environment (Mason et al. 2011). The outcome of a civil war may interact with the effects of bilateral aid to affect its impact on peace duration.

    In a government victory, the inclination for dual sovereignty to reemerge is not as direct as the other outcomes. On one hand, a victory by the government is established, but the supporters of rebels as well as the rebel group itself remain in the country after the civil war. As a result, peace duration may be difficult to predict. Mason et al. (2011) argue that government victories remain stable only because it takes time for rebel groups to regroup and rearm. To follow this concept, in a government victory, the ratio of bilateral aid should be proportionate to multilateral aid to keep rebels from acquiring an incentive for dual sovereignty in a post-conflict environment. Bilateral aid donors implement policies based on self-interest in comparison with multilateral donors who implement policies to benefit the recipient country.

    Hypothesis 1: High ratio of bilateral aid over multilateral aid decreases peace duration in a government victory in post-conflict environments.

    The findings of Quinn et al. (2007) reveal that rebel victories are less likely to break down into renewed conflict than government victories. Unlike a government victory in which the rebel group remains in the state, during a rebel victory, the former members of the government are exiled from the country and the rebels solely control the new government. In a rebel victory the government loses its sovereignty because the rebels become the new government. Consequently, dual sovereignty is less likely to recur in a rebel victory because the former government no longer exists within the country (Mason et al. 2011). Thus, bilateral aid donors are less likely to have an impact on rebel victories because there is a less likelihood for dual sovereignty to recur. On that note, after a rebel victory, donors should maintain long term bilateral aid agreements during the first few years of a post-conflict environment until the recipient country attains political stability. Therefore, a high ratio of bilateral aid in a government victory will likely have a positive effect on durable peace in a post-conflict environment.

    Hypothesis 2: High ratio of bilateral aid over multilateral aid increases peace duration in a rebel victory in post-conflict environments.

    Brandt et al. (2008) argue that, over time, negotiated settlements are not more fragile in comparison to decisive victories. On the other hand, Mason et al. (2011) argue that negotiated settlements often produce a more fragile peace than government victories. Negotiated settlements represent post-conflict environments where the former belligerents reached some form of “consensus” and are therefore less likely to cause dual sovereignty to recur than government victories. Although both belligerents remain in the post-conflict environment, there may be a chance for durable peace because the belligerents reached a negotiation, and each side gained something from the negotiated settlement. Alternatively, skepticism may arise between the government and opposing group to follow the agreement, which may incite one group to break the contract. As a result, bilateral donors may negatively impact durable peace if donors give a high ratio of aid to a post-conflict environment with a negotiated settlement.

    Hypothesis 3: High ratio of bilateral aid over multilateral aid decreases peace duration in an outcome of negotiated settlement in post-conflict environments.

    Unlike negotiated settlements where the belligerents reached a negotiated settlement, a ceasefire only calls for belligerents to put their arms down. Ceasefire may be the outcome of different issues that came up during the conflict including costs. According to Mason et al. (2011), if the rebels do not see a victory soon and are not able to cover the costs for a prolonged war, they are most likely to agree to a ceasefire. During this type of civil war outcome, both the government and rebels continue to operate within the state and remain armed and mobilized. Therefore, dual sovereignty is likely to recur because the rebels do not gain anything from the conflict and if they remain in the state, they are most likely to challenge the government again. On this note, a high ratio of bilateral aid may cause tension between belligerents and trigger dual sovereignty to recur. Since bilateral aid donors are more likely to support one group over another and the aid goes directly to the governments, the rebels (who are in need of additional funds) observe the government receiving aid from a third party and may challenge the state’s power.

    Hypothesis 4: High ratio of bilateral aid over multilateral aid decreases peace duration in an outcome of ceasefire in post-conflict environments.

    The policies in aid agreements are very important because they help implement the aid to its best advantage for the recipient state. Bilateral donors typically give aid based on their own political gains and may not consider other actors within the country that may need foreign aid. Additionally, they may not consider the effects of the distribution of bilateral aid in a post-conflict environment. This action may cause conflict to recur. Corruption is very common in certain forms of foreign aid, which emphasizes the importance of having good policies to avoid corruption from the donor or the recipient. Balancing the ratio of multilateral and bilateral aid in post-conflict environments may assist in prolonging peace duration because they will avoid supporting one group over the other, and subsequently recreating the condition of dual sovereignty.

    Research Design

    I use the Mason et al. (2011) dataset on peace duration to test my hypotheses concerning civil war outcome and foreign aid. According to Mason et al. (2011), the dataset contains 96 peace attempts, but 48 of the peace spells ended in new civil wars or peace failure. The dataset also includes information on the civil war outcome particularly if the outcome of the civil war was a government victory, rebel victory, negotiated settlement, or truce. The peace duration dataset documents peace duration as the number of years after a civil war has ended, and includes all civil wars that started between 1945 and 2002 (Mason et al. 2011). The unit of analysis for this article is peace spells.

    Previous research has focused on the circumstances in which a nation’s political and economic institutions encourage dual sovereignty (Mason et al. 2011). This article focuses on how outside factors including sources of foreign aid may destabilize durable peace depending on the fragile stability left from the previous conflict, which may incidentally encourage dual sovereignty. I am interested in analyzing the impact of the foreign aid ratio and the civil war outcome, along with other explanatory variables on peace duration. Given that durable peace is difficult to achieve in post-conflict environments, this paper advances the proposal that international actors may affect peace duration in developing countries.

    As noted earlier, the conflict in Sudan was ongoing even after a peace treaty attempt in 2005, and as a result, made Sudan highly dependent on foreign aid. Therefore, peace attempts between the belligerents did not eliminate the condition of dual sovereignty, and foreign aid sources may have influenced this by implementing policies that may have excluded some of the former belligerents. Some of these policies made by donors may include keeping an actor or government party in office in the recipient state, which keeps other parties from participating.

    It is also important to take into consideration that some variables have time varying effects and may not have the same number of cases regarding civil war outcomes, which may lead to biased results. A logistic regression model is used to test the relationship between durable peace and the interaction between foreign aid ratio and the civil war outcome. A logit regression model was the best fit for this study because my dependent variable is dichotomous. The results using a binomial logistic model are provided in Table 1 (pg. 25). In binomial logit models, we can observe whether the coefficients are statistically significant or not. The actual coefficients produced by the logit regression, however, are limited in their interpretation. When using a binomial logit model, one cannot interpret the coefficients directly in terms of a change in the dependent variable, y, for a unit change in the independent variable, x (Long 1997). Therefore, I ran four logit models with odds ratios that can be directly interpreted to determine the effect of my independent variables on my dependent variable, peace duration.

    Dependent Variable

    The dependent variable for this study is peace duration, which is measured on an ordinal scale. Using the Mason et al. dataset (2011), a new dichotomous variable was created to simply measure whether the civil conflict ended in durable peace. Peace is measured by the number of years without civil armed conflict. The dataset also measures peace duration through the concept of dual sovereignty which was previously discussed in Tilly (1978). Peace duration represents the time belligerents engage in peace. A state with durable peace is defined as having 5 or more years of peace. On that note, a state that does not have durable peace is defined to have 5 years or less of peace.

    Peace duration measured by year is not descriptive enough to measure the time it took the rebels to rearm, and recruit to challenge the government, subsequently resuming the condition of dual sovereignty. Apart from political actors acting as obstacles for dual sovereignty, other obstacles may have prevented the rebels from engaging in conflict, including costs. Although peace duration does not describe what specific incentive led to the recurrence of civil conflict, it is safe to generalize that the rebels achieved dual sovereignty as a fracture of peace duration.

    This article focuses on peace spells for the unit of analysis. A post-conflict year unit of analysis is preferred over dyads because only governments receive foreign aid, and dyads vary within state borders. Thus, post-conflict year analysis works best for this research.

    Independent Variables

    Several independent variables are defined to test my hypothesis. First and foremost, the sources of foreign aid are gathered from the OECD dataset (www.oecd.org) which contains data from 1960 to 2009. The dataset contains all donor and recipient amounts from all foreign aid transactions. This research, however, specifically focuses on the foreign aid that is received from each recipient state. Data are separated by the total amount of foreign aid that is received from each recipient state, and the total amount of multilateral foreign aid. Therefore, total aid was subtracted from multilateral aid in order to document the amount of bilateral aid that was received by each recipient country. To measure the hypothesis that bilateral aid has a negative impact on peace duration, the variable ratio was created to measure the proportion of bilateral aid over multilateral aid. Foreign aid is measured in U.S. dollars and is a continuous variable. A high a ratio is the division of bilateral aid over multilateral aid. Therefore, a high ratio demonstrates a higher proportion of bilateral aid over multilateral aid.

    Second, I include a variable representing the civil war outcome from the dataset (Mason et al. 2011), which signifies a government victory, rebel victory, negotiated settlement, or truce.  Given that previous research has already observed the relationship between civil war outcome and peace duration, I constructed four interaction variables that describe the relationship between ratio and civil war outcome to test its impact on peace duration. The first interaction variable, interact_reb, includes the relationship between foreign aid ratio and rebel victory. Rebel victory means that the rebel wins and the government is overthrown.  Thus, dual sovereignty no longer takes place because the government is replaced by the opposing group. The second interaction variable,interact_gov, includes the relationship between foreign aid ratio and government victory. Government victory means that the government continues to fight and rebel quits, or the government wins and the rebel is defeated. The third interaction variable, interact_settle, consists of the relationship between foreign aid ratio and negotiated settlements. A settlement means that both the government and the rebel choose to quit at the same time. The fourth interaction variable, interact_truce incorporates the relationship between foreign aid ratio and truce, which means that both the government and the rebel decide to agree to a ceasefire without a formal agreement. The interaction variable between foreign aid ratio and civil war outcome suggests that there is a relationship between ratio and civil war outcome, thus these variables are added to determine if a relationship exists between the interaction variables and peace duration (Marshall and Jaggers 2000).

    Control Variables

    Several control variables are added that may influence peace duration, including democracy, GDP per capita, GDP growth, and total aid given to a recipient country. Polity 2 was included as a measurement of democracy and its effect on peace duration, which was gathered from the Polity IV Political Regime Characteristics and Transitions Dataset, where the scale of democracy is measured from -10 (least democratic) to 10 (most democratic).

    GDP per capita and GDP growth of the country measured in U.S. dollars are included as control variables in order to analyze the income necessity that may support previous research regarding opportunity costs (Walter 2004).  GDP per capita and GDP growth are already included in the Mason et al. (2011) dataset. Opportunity costs may influence enough people to join the rebel groups in order to gain dual sovereignty (Walter 2004). Thus, GDP per capita and GDP growth are measured as continuous variables.

    The total amount of aid is also included as a control variable. The total amount of aid includes bilateral and multilateral aid, which is included as a continuous variable measured in U.S. dollars.  The total amount of aid data is gathered from the OECD database (www.oecd.org).

    Analysis

    Table 1 presents the results from the logit regression tests for the hypotheses. I altered the controls for each model to verify the strength of the relationship between the interaction variables and peace duration. Model 1 contains all the interaction variables along with GDP per capita, GDP growth, total aid, and Polity2. Model 2 results do not have Polity2 as a control variable. Model 3 results do not have total aid as a control variable, and model 4 does not have GDP per capita as a control variable. The results for Models 1, 2, and 4 show that aid ratio along with government victory and truce as civil war outcomes have significant results on peace duration.

    In Table 1, Models 1, 2, 3, and 4 provides the outcome of hypothesis 1, which concerns the interaction variable for interact_gov and its impact on peace duration in post-conflict environments. The logit regression model shows a significant relationship between interact_gov and peaceduration in Models 1, 2, and 4, therefore the null hypothesis was rejected because it demonstrated a small p-value of .05 or less in Models 1, 2, and 4. In addition, the odds ratio model indicates that the odds of peace duration increase by 1.18 as the ratio increases in a government victory outcome. A larger proportion of bilateral aid over multilateral aid given to the recipient country in a post-conflict environment decreases the opportunity for dual sovereignty. Bilateral aid is given to the government, many times to a particular party to keep the party in office, thus increasing sovereignty for the government and decreasing the opportunity for dual sovereignty.

    Mason et al. (2011), state that in a government victory, there is a high probability of durable peace in relation to the first few years of peace. Specifically, Mason et al. (2011) argue that there is a high probability of peace frailty in a post-conflict setting after four years in a government victory than in a rebel victory. In addition, Collier et al. (2002; 2003) argue that a large amount of aid provided in the first four years would produce more political instability. Thus, these two variables were not previously combined to analyze a relationship with durable peace. Although Breuning and Ishiyama (2007) argue that neither the amount nor the timing of aid leads to political stability, they aggregated their variable for foreign aid and did not analyze the sources of foreign aid separately. Consequently, given that government victories in a post-conflict environment provide a high probability of peace duration for the first few years, the results show that a high ratio of bilateral aid also adds to this probability.

    Models 1, 2, 3, and 4 on Table 1 provide the outcome of hypothesis 3, which demonstrates the relationship between the interaction variable of interact truce and peace duration in a post-civil war environment. The logit results demonstrate a statistical significance between interact truce and peace duration. The odds of peace duration increase by 4 as the ratio increases with a truce outcome in a post-conflict environment. Therefore, if the bilateral aid ratio increases after a negotiated truce, there is a high probability for peace duration. The results emphasize the finding produced in Mason et al. (2011), which claimed that if rebels do not see a victory soon and are not able to cover the costs for a prolonged war, they are most likely to cease fire. After a negotiated truce, both government and rebel forces put their weapons down, so the opportunity for dual sovereignty is high. As previously stated, bilateral aid will give financial assistance to the government thus decreasing dual sovereignty as the government gains more economic power. Based on the results, one explanation is that the government receives bilateral aid, which may cause a higher probability for dual sovereignty to be eliminated for a few years because the ratio of bilateral aid increases.

    Moreover, the introduction of a high ratio of bilateral aid after a government victory and negotiated truce may be effective for peace duration considering the action that the government decides to make afterwards. In order to prevent dual sovereignty from recurring, the government may decide to change to a democratic regime.

    Models 1, 2, 3, and 4 demonstrate that there is no statistical significance between the relationship of ratio and negotiated settlements, or rebel victory to impact peace duration in a post-conflict environment. In a negotiated settlement, the government and the rebel reach a consensus which may involve sharing or splitting some sources in order to refrain from violence which involves costs. Mason et al. (2011) argue that negotiated settlements produce a more fragile peace than government victories, while Brandt et al. (2008) argue that negotiated settlements are not entirely more fragile than decisive victories. Reaffirming both arguments, negotiated settlements differ extensively from contract to contract. Thus, costs may not be a large factor in this situation because in a negotiated settlement outcome, none of the belligerents see victory in the near future and instead aim for settlement to acquire some political gain between belligerents.

    A high ratio of bilateral aid in a rebel victory did not indicate to be significant to peace duration in any of the models. Rebel victories are less likely to break down into renewed conflict than government victories (Quinn et al. 2007), and for this reason, a high ratio of bilateral aid in a rebel victory outcome does not have a statistical significance towards peace duration. After a rebel victory, dual sovereignty is eliminated, which means that there is less dependency on bilateral donors because there is no competition for sovereignty.

    GDP per capita, total aid, GDP growth, and polity2 had a different effect towards peace duration. Models 1, 2, and 3 indicate that GDP per capita is statistically significant towards peace duration. On the other hand, polity2, GDP growth, and total aid did not have a statistical significance towards peace duration.

    Conclusion

    This study aimed to determine whether an interaction between the sources of foreign aid and outcome of a civil war have an effect on the duration of peace. Although domestic affairs have a large impact on peace duration, international actors may also play a role on the condition of dual sovereignty in the recipient countries that experience ongoing conflict. Previous studies have argued that foreign aid does not have an impact in the recipient country in regard to political stability and economic development. Many scholars have included peacekeepers as international actors that may influence peace duration, yet previous studies have not included how different sources of foreign aid may have an influence in peace duration. Thus, this study attempted to explain how foreign aid may influence dual sovereignty depending on the civil war outcome. Given that bilateral aid is more likely to give aid that is biased on personal preferences, and not necessarily to those who need it the most, bilateral aid may shift dual sovereignty. Thus, interaction variables were developed to indicate the relationship between the ratio of bilateral aid over multilateral aid and civil war outcome. A logit regression model was constructed to explain the relationship between the interaction variable of ratio and civil war outcome and the impact these factors have towards peace duration.

    Ultimately, it was found that a high ratio of bilateral aid introduced in the outcomes of government victory and negotiated truce may be effective in prolonging peace duration. This study emphasizes the findings of previous studies which argue that a method to prolong peace duration includes the elimination of dual sovereignty (Tilly 1978). Dual sovereignty may lead to conflict due to competing means of obtaining power of the state. The results from this research demonstrate support for this theory. A high ratio of bilateral aid given to a recipient country that experienced a negotiated truce or government victory may help decrease the opportunity for dual sovereignty because the government receives the aid, which increases government sovereignty. Thus, if a government is receiving a high ratio of bilateral aid in a post-conflict environment, some of that aid may also serve as military aid, which strengthens the military capacity of the state and may use it against the rebels if they decide to attempt to resume conflict.

    The actions a government pursues after regaining sovereignty may also impact peace duration. In order to prevent dual sovereignty from reoccurring, the government may decide to transition toward democracy in order to signal to the former rebels that they are attempting to co-opt the former rebel. Quinn et al. (2007) argue that a negotiated settlement outcome involves the construction of power-sharing institutions that offer belligerents the opportunity to compete peacefully for political office. Democracy requires the redistribution of political and economic power between governments (Quinn et al. 2007). This concept, therefore, supports the results for having no significant effect for having a high ratio of bilateral aid and negotiated settlement outcome impacting durable peace because there is a compromise for redistribution. Thus, bilateral aid may be received by the former belligerents involved in the negotiated settlement. Unlike a negotiated settlement, a truce outcome involves belligerents who compromised to put down their weapons, but no formal permanent agreement takes place. Thus, a high bilateral aid ratio decreases dual sovereignty because the government gains sovereignty and may use such aid to its advantage.

    Quinn et al. (2007) also argue that military victories, which include both government and rebel victories, are less likely to pursue a democratic regime in a post-war environment, because there is no major gain. Rebel victories have a durable peace because the condition of dual sovereignty no longer exists as there are no longer groups challenging state sovereignty. Thus, the results demonstrate that a high ratio of bilateral aid will not have a large impact on peace duration after a rebel victory.

    Although there is significant difference for peace duration in a government victory, a high ratio of bilateral aid may only impact peace duration for the first few years. If the rebels continue to live in the same state after conflict, they may rearm and regain dual sovereignty at some point in time. For this reason, a high proportion of bilateral aid in a truce civil war outcome has a greater probability of durable peace than in a government victory in a post-conflict environment.

    The data gathered for bilateral aid was calculated based on the information provided from the OECD dataset. The bilateral aid data included loans which may have impacted results; therefore further research should attempt to include bilateral data without loans. Furthermore, this research examines interactive factors including sources of foreign aid and civil conflict outcome which have not been analyzed together to measure peace duration. Previous studies have included peacekeepers as an international factor that may influence peace duration in a post-conflict environment, thus this research provides additional international factors that impact peace duration.

    Future research should attempt to include non-OECD countries. Non-OECD countries such as nations in Asia tend to give industrial forms of foreign aid. Therefore, non-OECD donors may have a different impact on peace duration than OECD donors considering the different types of foreign aid.

    As previously stated, a high ratio of bilateral aid introduced in the civil war outcomes of government victory and truce civil war outcome may be effective for peace duration considering the action that the government decides to take. Dual sovereignty suggests that, in order to challenge the government, the rebels have to cover the costs of rebellion to compete with the recruitment and weaponry from the government. Thus, if a government receives a high ratio of bilateral aid in a post-conflict environment, some of the aid may also serve as military aid which decreases the probability for rebels to rapidly rearm and recruit. In order to further analyze this observation, military aid data from each recipient country would be needed. Nevertheless this study was able to analyze additional factors that contribute to search for peace duration in post-conflict environments.

    References

    • Brandt, P., Mason, D., Gurses, M., McLeod, P., Petrovsky, N., & Radin, D. 2005. “Never-Lasting Peace: Explaining the Duration of Civil Wars.” Conference Papers-International Studies Association, 1, 1-28. Retrieved July 25, 2011, from the Academic Search Complete database.
    • Brandt, P., Mason, D., Gurses, M., Petrovsky, N., & Radin, D. 2008. “When and How the Fighting Stops: Explaining the Duration and Outcome of Civil Wars.” Defense & Peace Economics, 19(6), 415-434. Retrieved July 19, 2011, from the Academic Search Complete database.
    • Breuning, M., & Ishiyama, J. 2007. “Foreign Aid, Democracy and Political Stability in Post-conflict Societies.” Alternatives: Turkish Journal of International Relations, 6(1/2), 82-91.
    • Collier, P., & Hoeffler, A. 2002. “Aid, Policy and Peace: Reducing the Risks of Civil Conflict.” Defense and Peace Economics, 13(6), 435-451.
    • Collier, P., Elliott, V., Hegre, H., Hoeffler, A., Reynal-Querol, M., & Sambanis, N. 2003. Breaking the Conflict Trap. Washington, DC: A World Bank Policy Research Report.
    • De Mesquita, B. B. 2010. Principles of International Politics (Fourth Edition ed., pp. 246-273). Washington, D.C.: Bruce Bueno de Mesquita.
    • Esman, M. J., & Herring, R. J. 2001. Carrots, Sticks, and Ethnic Conflict: Rethinking Development assistance. Ann Arbor: University of Michigan Press.
    • Fearon, J. 2004. “Why do some civil wars last so much longer than others?” Journal of Peace Research, 41(3), 275-301. Retrieved July 25, 2011, from the JSTOR database.
    • Grossman, H. 1992. “Foreign Aid and Insurrection.” Defense Economic, 3, 275-288.
    • Joshi, M. 2010. “Post-Civil War Democratization: Promotion of Democracy in Post Civil War States, 1946–2005.” Democratization 17:5, 826-855.
    • Kang, A. S., & Meernik, J. 2004. “Determinants of Post-Conflict Economic Assistance.” Journal of Peace Research 41(2), 149-166.
    • Kelegama, S. 2005. “Transforming conflict with an economic dividend: The Sri Lankan experience.” Round Table 94(381), 429-442.
    • Kreutz, J. 2010. “How and when armed conflicts end: Introducing the UCDP Conflict Termination dataset.” Journal of Peace Research 47(2), 243-250. Retrieved July 25, 2011, from the Academic Search Complete database.
    • Long, J. 1997. “Regression Models for Categorical and Limited Dependent Variables. Advance Quantitative Techniques in the Social Sciences.” Number 7. Sage Publications.
    • Mason, D., Gurses, M., Brandt, P., & Quinn, J. 2011. “When Civil Wars Recur: Conditions for Durable Peace after Civil Wars.” International Studies Perspectives 12(2), 171-189.
    • Marshall, Monty G. and Keith Jaggers. (2000). “Polity IV Project: Political Regime Characteristics and Transitions, 1800-2000.”
    • Nielsen, R., Findley, M., Candland, T., & Nielson, D. 2011. “Foreign Aid Shocks as a Cause of Violent Armed Conflict.” American Journal of Political Science 55(2), 219-232.
    • Quinn, M., Mason, D., & Gurses, M. 2007. “Sustaining the Peace: Determinants of Civil War Recurrence.” International Interactions 33(2), 167-193. Retrieved July 25, 2011, from the Academic Search Complete database.
    • Tilly, C. 1978. From mobilization to revolution. New York: McGraw-Hill.
    • Walter, B. 2004. “Does Conflict Beget Conflict? Explaining Recurring Civil War.” Journal of Peace Research 41(3), 371-388.

    Table 1: Models 1-4, Odds Ratio Estimates, Whether Ratio and Civil War Outcome Leads to Peace Duration

    Variables

    Model 1

    Model 2

    Model 3

    Model 4

    Ratio and Truce

    4.41 (.01)***

    4.37 (.01)***

    4.19 (.02)**

    5.39(.004)***

    Ratio and
    negotiated
    settlement

    .86 (.17)

    .85 (.12)

    .84 (.11)

    0.91 (.32)

    Ratio and rebel
    victory

    .76 (.13)

    .74 (.11)

    .75 (.13)

    0.8 (.15)

    Ratio and
    government
    victory

    1.15 (.05)**

    1.16 (.04)**

    1.12 (.07)*

    1.25 (.005)***

    Polity2

    .98 (.31)

     

    .98 (.30)

    0.99 (.95)
    GDP growth

    1 (.82)

    .99 (.96)

    1 (.80)

    1(.84)

    GDP per capita 1(.000)***

    1 ( .000)***

    1(.000)***

     
    Total aid

    .99 (.30)

    .99 (.31)

      0.99(.24)
    Dependent
    Variable: Peace Duration
    *p< .10
    ** P<.05
    *** p<.01
    Number of observations: 632
    Log likelihood:
    Pseudo R2:
    0.0882