Incentivizing Peace


Incentivizing Peace
How International Organizations CanHelp Prevent CivilWars
inMember Countries

JAROSL AV TIR AND JOHANNE S KARRETH


Oxford University Press is a department of the University of Oxford. It furthers the University’s objective of excellence in research,
scholarship, and education by publishing worldwide. Oxford is a registered trade mark of Oxford University Press in the UK and in
certain other countries.
Published in the United States of America by Oxford University Press
198 Madison Avenue, New York, NY 10016, United States of America.
© Oxford University Press 2018
All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any
means, without the prior permission in writing of Oxford University Press, or as expressly permitted by law, by license, or under terms
agreed with the appropriate reproduction rights organization. Inquiries concerning reproduction outside the scope of the above should be
sent to the Rights Department, Oxford University Press, at the address above.
You must not circulate this work in any other form and you must impose this same condition on any acquirer.
Library of Congress Cataloging-in-Publication Data
Names: Tir, Jaroslav, 1972- author. | Karreth, Johannes, author.
Title: Incentivizing peace : how international organizations can help prevent civil wars
in member countries / Jaroslav Tir and Johannes Karreth.
Description: New York, NY : Oxford University Press, [2018] | Includes bibliographical references.
Identifiers: LCCN 2017026625 (print) | LCCN 2017052236 (ebook) | ISBN 9780190699543 (epub) |
ISBN 9780190699536 (updf) | ISBN 9780190699512 (hardcover) | ISBN 9780190699529 (pbk.)
Subjects: LCSH: International agencies–Membership. |
Peace-building–International cooperation. |

Conflict management–International cooperation. | Civil war–Prevention.
Classification: LCC JZ4850 (ebook) |
LCC JZ4850.T57 2018 (print) | DDC 303.6/4–dc23
LC record available at />

CONTENTS

List of Figures
List of Tables
Acknowledgments

1. Introduction
1.1.
1.2.
1.3.
1.4.

THE PROBLEMS OF CIVIL WAR AND CIVIL WAR MANAGEMENT
A NOVEL PERSPECTIVE ON CIVIL WAR MANAGEMENT
CONTRIBUTIONS AND IMPLICATIONS
OUTLINE OF THE BOOK

2. Managing Civil Wars from the Perspective of Their Development
2.1.
2.2.
2.3.
2.4.
2.5.
2.6.


INTRODUCTION
CAUSES, CONSEQUENCES, AND MANAGEMENT OF CIVIL WARS
CIVIL WAR DEVELOPMENT AND ESCALATION
WHAT DOES IT TAKE TO PREVENT LOW-LEVEL ARMED CONFLICT FROM ESCALATING TO CIVIL WAR?
THE (IN)ADEQUACY OF COMMON THIRD-PARTY CONFLICT MANAGING POLICIES IN PREVENTING
CONFLICT ESCALATION
OUR THEORETICAL FRAMEWORK FOR ESCALATION PREVENTION

3. The Interplay Between Civil War Development and Highly Structured Intergovernmental
Organizations
3.1.
3.2.
3.3.
3.4.
3.5.

THE DEFINITION, EVOLUTION, AND DISTRIBUTION OF HIGHLY STRUCTURED IGOS
THE IMPORTANCE OF IGOS’ INDEPENDENCE FROM AND LEVERAGE OVER MEMBER STATES
THE ROLE OF HIGHLY STRUCTURED IGOS IN PREVENTING CIVIL WAR
HOW HIGHLY STRUCTURED IGOS EXERCISE INFLUENCE
THE PRIMARY HYPOTHESIS

4. The Empirical Record of Highly Structured Intergovernmental Organizations and Armed Conflict
Escalation
4.1.
4.2.
4.3.
4.4.
4.5.
4.6.

4.7.

A SYSTEMATIC EXAMINATION OF CONFLICT ESCALATION
THE UNIVERSE OF CASES
ARMED CONFLICT ESCALATION
MEASURING THE INFLUENCE OF HIGHLY STRUCTURED IGOS
OTHER POTENTIAL DETERMINANTS OF ESCALATION
HIGHLY STRUCTURED IGOS REDUCE THE RISK OF ARMED CONFLICT ESCALATION
CONCLUSION

5. The Logic of Institutional Influence: Conceptual and Methodological Implications
5.1.
5.2.
5.3.
5.4.
5.5.

COSTS AND BENEFITS
PEACEFUL SETTLEMENTS
MEDIATION AND INTERVENTION
POTENTIAL MEMBERSHIP SCREENING BY HIGHLY STRUCTURED IGOS
OTHER POTENTIAL INFLUENCES ON CONFLICT ESCALATION


5.6.
5.7.
5.8.

THE ADDED VALUE OF HIGHLY STRUCTURED IGOS INEXPLAINING ESCALATION
CONCLUSION

APPENDIX

6. Case Evidence: Conflict Trajectories in Indonesia, Ivory Coast, and Syria
6.1.
6.2.
6.3.
6.4.
6.5.

CASE SELECTION
INDONESIA AND EAST TIMOR
IVORY COAST
SYRIA
CONCLUSION

7. Conclusion
7.1.
7.2.
7.3.
7.4.
7.5.
7.6.

BRIEF SUMMARY OF FINDINGS
LESSON 1: THE INTERNATIONAL ENVIRONMENT CAN CONTRIBUTE TO CIVILWAR PREVENTION
LESSON 2: ECONOMIC INFLUENCE IS KEY TO CIVILWAR PREVENTION
LESSON 3: COORDINATION INCREASES THE IMPACT
LESSON 4: FURTHER INVESTIGATE THE IMPACT OF HIGHLY STRUCTURED IGOS ON REBELS
LESSON 5: PEACE PAYS OFF (FOR REBELS, TOO)


Data Appendix
List of Software Packages Used
Notes
Bibliography
Index


LIST OF FIGURES

3.1
3.2
3.3
3.4
3.5
4.1
4.2
4.3
4.4
4.5
4.6
5.1
5.2
5.3
5.4
5.5
5.6
5.7
5.8
5.9
5.10

5.11
5.12
5.13
5.14
5.15

Count of highly structured IGOs in the international system over time
Comparison of the membership size of all highly structured IGOs
Membership size of highly structured IGOs over time
States’ participation in highly structured IGOs over time
The geography of highly structured IGOs
Map of all low-level armed conflicts, 1946–2000
Map of escalated conflicts, 1946–2000
Highly structured IGOs and estimated escalation risk
Changes in escalation risk
Political institutions and escalation risk
Changes in the risks of conflict onset and escalation
HSIGOs commanding substantial resources and changes in escalation risk
HSIGOs commanding substantial resources and conflict escalation
Changes in the probability of low-level armed conflict settlements
Highly structured IGOs and low-level armed conflict settlement
Distribution of mediation and intervention efforts
The role of highly structured IGOs intervention, and mediation
Mediation, intervention, highly structured IGOs, and conflict escalation
Low-level armed conflict escalation patterns over time
Highly structured IGOs and escalation risk, using instrumental variable estimates
HSIGOs, conflict escalation, and natural resources
HSIGOs, rebel strength, and escalation risk
HSIGOs, space, time, and changes in escalation risk
HSIGOs, conflict diffusion, and changes in escalation risk

Variable inclusion in BMA
Posterior probabilities of coefficients in BMA


LIST OF TABLES

2.1
3.1
3.2
4.1
4.2
4.3
4.4
5.1
5.2
5.3
5.4
5.5
5.6
5.7
5.8
5.9
5.10
5.11
6.1
6.2
6.3

An Assessment of Common Third-Party Conflict Management Policies
List of Highly Structured IGOs

An Assessment of Third Parties’ Roles in Pre–Civil War Bargaining
Low-Level Armed Conflicts That Escalated to Civil Wars, 1946–2000
Descriptive Statistics for All Variables in Model 2 in Table 4.3
Probit Estimates of the Escalation of Low-Level Armed Conflict to Civil War
Heckman Probit Estimates of Escalation, Accounting for Selection of Low-Level Armed
Conflict
Descriptive Statistics for All Analyses
The Influence of Highly Structured IGOs Commanding Substantial Resources
Highly Structured IGOs and Conflict Settlement
Mediation, Intervention, Highly Structured IGOs, and Escalation
Instrumental Variable Estimates
Natural Resources and Escalation
Natural Resources and a Possible Conditional Impact of Highly Structured IGOs
Rebel Strength and Escalation
Spatial and Temporal Trends of Escalation
Conflicts in Proximity and Escalation
Territorial Conflicts and Escalation
Summary of Key Expectations and Anticipated Evidence
Summary of Evidence Pertaining to the Three Cases
Expectations and Observed Evidence


ACKNOWLEDGMENTS

This book has grown out of a long-term collaborative project between the authors. Along the way,
several colleagues and friends have generously provided feedback and offered helpful thoughts on
our work. For their comments on the manuscript along its different stages, we thank Kyle Beardsley,
Ken Bickers, Margit Bussmann, Courtenay Conrad, Paul Diehl, Bryan Early, Page Fortna, Michael
Greig, Håvard Hegre, Ann Karreth, Carmela Lutmar, Helen Milner, Sara Mitchell, Glenn Palmer,
Burcu Savun, Carolyn Tir, and Geoffrey Wallace. John Vasquez, Patrick James, Paul Diehl, and

Matthew Ingram provided advice and guidance in developing and placing the book manuscript.
At Oxford University Press, we are grateful to David McBride for his editorial guidance. The
feedback from two anonymous reviewers greatly improved the manuscript. We also thank Emily
Mackenzie, Ed Robinson, Claire Sibley, and Kathleen Weaver for their assistance during the
production of this book and Susan McClung for copy-editing our manuscript.
In facilitating this research, we have benefited from institutional support from the Departments of
Political Science at the University of Colorado Boulder and the University at Albany, the Department
of Politics and International Relations at Ursinus College, and the Kroc Institute for International
Peace Studies at the University of Notre Dame. For research assistance, we thank Steven Beard, Cori
Cichowicz, Timothy Passmore, Elisa Elvove, and Charmaine Willis. We also owe a debt of gratitude
to the scholars and researchers who collected and made publicly available the data we use throughout
this book.
The authors contributed equally and interactively to this book. This book is one of several outputs
of a joint project, and the order of their names follows a principle of rotation.


Incentivizing Peace


1

Introduction

1.1 The Problems of Civil War and Civil War Management
Civil wars are one of the most pressing problems facing the world today. Beyond causing millions of
casualties and immeasurable human suffering in the post–World War II era, 1 such wars have
devastating social and economic consequences. Public health institutions and outcomes in countries
experiencing civil war systematically suffer. 2 The mass migration of people fleeing violence and
destruction can destabilize neighboring countries and regions, creating new challenges.3 The
destruction of physical and human capital during civil wars creates serious and lasting damage to

economic prosperity and international trade.4 Experiences of violence lead to substantial increases in
political and social intolerance of ethnic, religious, and other out-groups.5
These and other consequences of civil wars are in part responsible for setting the stage for future
civil wars. Nearly 50 percent of civil wars return to violence within five years of the initial cessation
of hostilities.6 The danger of recurrence is so great that researchers have characterized many
domestic conflicts as enduring internal rivalries.7
Given the far-ranging and well-known consequences of civil wars, academics and policymakers
have devoted much attention to finding solutions for managing these conflicts and their consequences.
This attention has led to some positive outcomes. Among a variety of approaches to civil war
management by third parties, the most common ones include mediation, peacekeeping, and militarized
interventions. These third-party responses have frequently helped. They have contributed to the
transition of civil war–plagued countries toward normalcy and to the prevention of civil war
recurrence. Some of the oft-heralded success stories of third-party conflict management include
Cambodia, Sierra Leone, Angola, the Eastern Slavonia region of Croatia, Kosovo, and the Dayton
Agreement in Bosnia. More broadly, and focusing more on international wars, Joshua Goldstein
argues that the world has become more peaceful in recent times and gives at least some credit for this
development to concerted international conflict management efforts.8
Yet, despite some successes, the sad fact remains that civil wars continue to occur, and recur, to
this day with notable regularity. 9 Indeed, critics note that the successful track record of conflict
management efforts is marred by a notable undersupply of third-party involvement or by outright
failures. Some high-profile cases where third parties failed to manage and halt conflicts in the last 25
years include pre-Dayton Bosnia, Rwanda, Somalia, Sudan, and Syria. In these and other cases,
conflict management failed either because third parties took far too long to engage in robust, decisive
conflict management efforts or because they did not supply any meaningful conflict management at all.
Multilateral peacekeeping missions in particular are deployed to only about one-third of civil
wars.10 And in cases where international peacekeepers were involved, their engagement often leads
to speculation about whether the presence of peacekeepers actually helped local populations, and
whether they made substantial contributions to stabilization and peace.11 This mixed success record



of peacekeeping missions has also created a robust debate on how and why third parties choose when
to get involved in conflict management and when to stay out. One criticism maintains that
peacekeeping missions’ main agenda relates more to pursuing goals relevant to their sending
organization or states. These include accusations of selecting civil wars that are easier to resolve in
order to make organizations such as the United Nations look successful and helpful.12 Another
critique notes that peacekeepers are often sent to help allied or economically important civil war
states, while ignoring others.13 Therefore, even if peacekeeping were a panacea to political violence,
such selective deployment of peacekeeping efforts is one of the reasons why civil wars persist today.
Questions of bias favoring the interests of sending states as opposed to civil-war states are even
more prominent in the case of militarized interventions. In these cases, intervening states generally
aim to protect or advance their economic or strategic interests and to make sure that their favored side
wins the war. 14 And an unfortunate consequence of military interventions is that they can intensify
political violence and prolong armed conflicts,15 further amplifying the negative long-term
consequences of civil wars noted previously.
International mediators are closest to being truly neutral actors who have the interests of the civilwar state and its population at the forefront of their agenda. Yet while mediation attempts receive
much press for helping negotiate short-term ceasefires, it is not all that clear that they are particularly
effective in actually ending civil wars.16 Additionally, mediators do not appear to go to all the
conflict zones or do so particularly quickly. For example, in the context of sub-Saharan Africa,
Michael Greig reports that the chances that a conflict is mediated in its first three years are less than 1
in 100.17
In short, despite many conflict management successes, scholars’ and practitioners’ understanding of
how to manage civil wars can still be substantially improved. With a better understanding of how to
use existing international structures effectively to mitigate civil wars, countries facing high risks of
political violence could improve their odds for domestic peace and stability. Addressing this issue
and highlighting a potential solution are the subjects of this book. We preview this approach and our
argument next.

1.2 A Novel Perspective on Civil War Management
As the previous discussion suggests, the international community has had only limited success in
managing civil wars. For that reason, we argue for the need to fundamentally rethink how third parties

can approach civil wars. In this book, we propose that international, third-party conflict management
efforts will be particularly effective when they focus on civil war prevention, rather than on dealing
with already ongoing civil wars. Civil wars do not occur overnight, but take time to develop.
Acknowledging this fact allows us to investigate whether and how emerging conflicts can be stopped
before they escalate to full-scale civil wars. Unfortunately, scholarly attention has thus far mostly
focused on how typical third-party policy responses (peacekeeping, mediation, and intervention)
affect later phases of a conflict. Existing studies often ask whether mediation can help reach
ceasefires, or whether peacekeepers prolong the peace after civil war. These outcomes are all
relevant once hostilities have already escalated to civil war. The timing of the respective efforts
illustrates this issue. Mediators, intervention forces, and peacekeepers are typically deployed only
after a conflict has already become a full-blown civil war. Across the post–World War II period that
we investigate, and according to the data that we use in this book, the median timing for mediation is


48 months into the conflict. The typical intervention is deployed 23 months into the conflict.
Peacekeepers are almost never sent before a conflict becomes an all-out civil war; just about all
peacekeeping operations in the history of the United Nations did not start until at least one—and most
often several—years after the initial fighting began.18 In contrast, the threshold of a conflict’s
escalation to a full-scale civil war is usually reached within 12 months after hostilities began. These
temporal patterns suggest that a reorientation toward preventing armed conflict escalation is both
necessary and potentially fruitful. Nevertheless, third-party involvement in nascent, pre–civil war
conflicts has received only scant scholarly attention.
To be sure, much work by scholars and practitioners has focused on the role of third parties in
preventing conflict by improving structural conditions. Many of these efforts focus on development
assistance and institution-building.19 But while such long-term efforts may help mitigate the general
conditions that give rise to political violence, such efforts and the actors associated with them have
little ability to help defuse concrete and time-sensitive armed hostilities between governments and
opposition movements.20
We therefore focus on the early stages of domestic armed conflicts in this discussion. In a typical
scenario, civil wars start as relatively small clashes between the government and a domestic

opposition: that is, minorities, political movements, or other insurgent groups, all of whom can
potentially turn into organized rebel groups. After initial eruptions of violence, governments and
opposition groups typically face the important choice of mobilizing further, which risks conflict
escalation to civil war, or of trying to return to peace by accommodating each other’s demands
without significant further use of force. Taking a cue from the bargaining theory of war, 21 we see both
the rebels and the government as actors motivated by prospective gains, but also sensitive to the costs
that they may have to endure. Thus, both actors follow cost-benefit calculations to determine whether
continued fighting or a peaceful settlement would be more beneficial. Importantly for our perspective,
further large-scale organized violence can also be avoided if these cost-benefit calculations can be
changed to favor a peaceful resolution.
We argue that certain international third parties can help change these calculations early in the
conflict. The answer, however, is unfortunately less straightforward than merely urging that
mediation, peacekeeping, or intervention efforts take place sooner. As we explain in the next chapter,
these approaches are a poor fit for civil war prevention. They fall short on one or more of the
following dimensions. First, deploying them quickly is often not feasible. Second, their politicalstrategic nature introduces undesirable uncertainty into the calculations of both opposition and
government overwhether these activities are indeed forthcoming. And third, they are temporary fixes
that are not all that well suited to addressing the particularly challenging long-term credible
commitment problem that stands in the way of ending conflicts.22 This commitment problem in
conflict bargaining occurs because, lacking a powerful, long-term guarantor of peace, each side fears
that the other side will not hold up its end of a potential peace deal. Instead, each party will use
negotiation phases or ceasefires to better prepare itself for a future offensive and strike when the
conditions are favorable.23 Due to this problem and both actors’ anticipation of it, peace deals are
often difficult to achieve at the early stage of low-level armed conflicts. Conflict parties may fail to
sign them, or they may not proceed with their implementation. That is, if signed, agreements often
unravel before peace stabilizes.
In this book, we argue that there is a significant opportunity in taking a closer look at the role of
international third-party actors that have much potential to be quite effective at dealing with these


challenges in civil war prevention. We go beyond the “default” third-party response types of

international mediation, peacekeeping, and (military) intervention to focus on the role of a particular
subset of intergovernmental organizations ( IGOs) that share some key structural features and a natural
self-interest in conflict prevention. These IGOs include multilateral financial organizations such as
the World Bank, various regional development banks, and the International Monetary Fund (IMF).
While these organizations are well known, their role in managing the developmental trajectory of
domestic armed conflicts is not. Yet it is precisely these organizations and their influence — often in
concert with other regional organizations or third-party states — that has pushed both government and
rebels in low-level armed conflicts toward settling their differences before their country slid into
civil war. This influence has materialized in countries as different as Indonesia and Ivory Coast,
which we explore in more depth in this book.
More generally, we argue that a certain type of IGOs can be particularly effective in steering
emerging, low-level armed conflicts away from becoming full-scale civil wars. These IGOs can be a
useful tool in preventing civil wars in the first place. We identify these organizations as highly
structured IGOs. They are a subset of the broader range of hundreds of IGOs that have come into
existence since World War II. While the primary task of most IGOs is to coordinate policies between
member states, IGOs also have a self-interest in keeping member countries free from civil war in
order to achieve the IGOs’ missions. For instance, promoting economic development and
international trade requires, among other conditions, that member countries maintain internal peace
and stability. And a subset of IGOs, namely, highly structured IGOs, possess potent policy tools and
tangible resources that can be used toward successful civil war prevention. Key features of these
organizations are that they are designed in ways that provide them with notable formal-legal leverage
over member countries and that they often command valuable tangible resources that can be offered
to, or be withheld from, member countries. Member states value these resources highly. Access to
them is the reason why these states joined the organization and incurred the sovereignty costs of
submitting themselves to the institutional structures in the first place.
The central thesis of this book is that international institutional structures are a key factor in
preventing civil war. The extent to which a country experiencing a low-level domestic armed conflict
is embedded in highly structured IGOs can shape whether the emerging armed conflict on its territory
will escalate to full-scale civil war. We briefly preview this argument here and then expand upon it
throughout the book. Highly structured IGOs share the ability to act against the preferences or

objections of a member government, command over valuable resources, and an interest in member
states’ stability. These three factors enable highly structured IGOs to shape the incentives of
governments and potential rebels in the bargaining process before conflict escalation to civil war
takes place. For instance, development banks and trade organizations alike have frequently promised
assistance with postconflict reconstruction to countries at the verge of, or in the midst of, civil wars
—but always under the condition of stopping the violence first. And because domestic instability
unduly affects the mandate of highly structured IGOs, governments and their opponents can count on
the involvement of these IGOs. Otherwise, these institutions lose resources that they have previously
invested, risk having their missions compromised, and suffer reputational costs as well.
This is not necessarily the case with other types of third parties. Their involvement is often more
uncertain and subject to political decision making. United Nations peacekeeping, unilateral or
multilateral interventions, and even mediation are all subject to political and strategic incentives of
their own, as well as to the approval of multiple governments, or at least political actors within the
third-party itself. Moreover, third-party involvement, especially in the form of other states or alliance


organizations, is often partisan.24 The costs of continued fighting may thus raise for one side (e.g., the
rebels), but decrease for another (e.g., the government). This gap in costs makes peace more elusive.
In contrast, any involvement of highly structured IGOs is biased toward peace because this is the most
desirable outcome for these organizations.
Highly structured IGOs can also address the credible commitment problem. The threat of costs
imposed by these IGOs not only alters the rebels’ and government’s incentives toward peace in the
present time, but also constrains them from resuming the fight in the future. The interest of these
institutions in maintaining the peace over the long run, as well as their continued ability to provide
tangible benefits and impose material punishments, help curtail potential future misbehavior by both
the rebels and the government. By reducing uncertainty about future behavior, highly structured IGOs
thus help mitigate commitment problems. In this vein, they increase the odds of prolonged peace. This
argument mostly builds on the ability of highly structured IGOs to put pressure on member
governments. But, as we explain in Chapter 3, highly structured IGOs also provide a similar, yet
indirect, incentive structure for the rebel side to pursue and commit to peace as well.

Our argument implies that conflict management effects of highly structured IGOs will be greatest
when a state is simultaneously a member of multiple such IGOs. This helps reinforce the conflict
management function of each IGO. It also makes penalties more multifaceted and severe and provides
more rewards for desirable behavior, such as honoring peace commitments. It also increases the
chances that one highly structured IGO will take a leadership role in helping resolve the conflict; and
it creates an opportunity for other IGOs to multiply the incentives that push both sides toward peace.
The previous paragraphs constitute only a preview of the argument that we fully develop in
Chapter 3. In Chapters 4, 5, and 6, we carefully check the historical record for evidence on this
argument. Across all low-level armed conflicts all over the world since 1945, we find that countries
that are members of larger numbers of highly structured IGOs experienced substantially lower rates of
conflict escalation. Follow-up analyses show that the conflict-mitigating influence of these IGOs also
results in a better record of conflict settlement, and that highly structured IGOs exercise their
influence by shaping the costs and benefits of conflict escalation. In case studies of Indonesia in the
late 1990s and Ivory Coast in 2010–2011, we show in more detail the direct and indirect role that
highly structured IGOs took in these conflicts. Conversely, we also investigate bargaining in the
absence of a strong influence of highly structured IGOs by considering the conflict between the Syrian
government and opposition in 2011–2012. That case suggests that where highly structured IGOs are
less present, an emerging conflict is more likely to escalate to a full-scale civil war.

1.3 Contributions and Implications
This book makes multiple contributions to academic research. First, by focusing on more subtle ways
in which IGOs can effectively shape domestic conflicts, we go beyond the traditional IGO conflict
management roles of restraining interstate conflicts.25 In doing this, we also demonstrate that domestic
conflict management is not an exclusive domain of organizations such as the United Nations. Instead,
many IGOs have the potential to mitigate domestic conflicts, even though this is not their core
mission. These IGOs engage in shaping belligerents’ choices in domestic armed conflicts even before
conflicts escalate to civil war. Often, they do so effectively and prevent the escalation of tensions to
civil war. This suggests that future research on civil war prevention should incorporate international
institutions in both theoretical work and empirical investigations.
Second, our findings show that there are positive externalities to certain institutional designs. The



benefits of international institutional design of the type that we identify in highly structured IGOs
extend well beyond the original core institutional mandates, such as trade promotion, to the politics of
domestic conflict. An important stream of the institutionalist literature has focused on the “rational
design” of international institutions.26 This research suggests that states design institutions to make
credible commitments and to reap the benefits of cooperation. By demonstrating that IGO structure
can also affect domestic political developments, our findings show that high levels of international
institutionalization can have benefits both beyond the international level of analysis and the core
policy issue of the respective IGO. Although few of the IGOs that we classify as highly structured
were established to prevent the escalation of domestic armed conflict as their core mission, states’
involvement in these IGOs is more likely to result in more effective domestic conflict management.
That is, spillover effects and positive externalities from memberships in highly structured IGOs for
domestic conflict management are an important, non-trivial addition to well-established arguments on
desirable international effects of institutional design.
Third, the book demonstrates the importance of international institutional environments for the
trajectory of domestic conflict processes. In doing so, it contributes to the emerging discussion on the
transnational dimensions of domestic armed conflict, advocated by Kristian Gleditsch, Jeffrey
Checkel, and others.27 In this context, scholars have investigated the role of external actors such as
states; ethnic, religious, or political diasporas;28 international mediation attempts;29 or changes in the
international system.30 Focusing on states’ participation in networks of highly structured IGOs, our
study extends the concept of transnational aspects of domestic conflict, pointing to an understudied
aspect of more indirect transnational conflict management. In combination with our findings, the
proliferation of institutionalized cooperation between states31 and ever-increasing levels of IGOs’
decision-making independence from member states32 imply that effects of centralized IGO structures
will likely become an even more influential factor in shaping the trajectories of domestic armed
conflicts in the future.
Finally, by highlighting the concept of domestic conflict escalation, we help bring together the lowlevel armed conflict and civil war strands of the domestic conflict literature. And while our results
imply that the determinants of lower-level political violence differ somewhat from those of full-scale
civil war—especially with regard to the role of a state’s institutional embeddedness at the

international level—this is an important integrative step that can help spur new directions in this
well-known research area. While the extant literatures on low-level domestic political violence and
civil wars are both rich, less effort has been made to integrate the two. Treating the two phenomena
as part of one trajectory provides the opportunity to approach conflict management efforts from the
vantage point of conflict escalation. This opens new possibilities for investigating potentially
effective ways of preventing full-scale civil war.
Among the practical implications of our work, one important consideration is that we shed light on
an important third-party international actor whose contributions to civil war prevention have not
received due attention. As we demonstrate, highly structured IGOs have the potential to be
particularly effective in shaping the trajectory of low-level armed conflict. While domestic conflict
management is typically not the mission of highly structured IGOs, these IGOs have the potential to
mitigate conflicts. Arguably, they are more effective at this task than traditional conflict-managing
IGOs, such as the United Nations. By focusing on highly structured IGOs and their roles, we point
policymakers interested in civil war management toward an already existing and potentially highly
effective tool. As civil wars continue to inflict their many human, societal, economic, psychological,


and material costs, having a better handle on civil war management is of the utmost priority. Highly
structured IGOs can help here, as we find that their involvement can stop internal armed conflicts
before they turn into civil wars.
Also noteworthy is the consideration that civil war prevention through highly structured IGOs is
relatively straightforward. These institutions do not have to be persuaded that preventing a civil war
in a member country is something that should concern them. Decision-makers and staff in these
institutions already know this, as domestic instability unduly affects their missions. Thismakes them
all the more likely to take an interest in emerging conflicts and seek to use their influence to prevent
their escalation. In other words, these institutions’ awareness of and involvement in conflicts has the
potential to be automatic, precisely because domestic peace and stability in member countries is part
of the self-interest of these IGOs. Their goals of peace and stability mean that the institutions’ interest
in emerging conflicts is neutral, not favoring one side or the other. Highly structured IGOs instead
prioritize the return to peace and stability. Furthermore, highly structured IGOs can effectively

perform the long-termrole of a guarantor of peace that is critical to overcoming the credible
commitment problem. Performing this role is again in these institutions’ self-interest, meaning that the
organizations and their staff have sufficient motivation to engage in it. These points all compare
favorably to conflict management efforts by peace-keepers, mediators, alliances, or third-party states
contemplating militarized intervention. To greater or lesser extents, such efforts tend to be subject to
political decision making, partisan preferences favoring government or rebel sides, delays, and lack
of long-terminterest. These problems lead to an increased likelihood of escalation and prolonged
civil wars, with all their deleterious consequences. Highly structured IGOs can and do circumvent
these issues by acting fast, by being neutral, by their ability to impose penalties and offer benefits, and
by their interest in the long-term stability of member countries.
Finally, our approach of treating low-level armed conflicts and civil wars as part of one trajectory
is not only of academic interest, but is also important for international policy. External actors have
little influence on spontaneous and rapid occurrences of political protest turning violent. But there is
an opportunity for international institutions to affect the further development of such episodes of
political violence. Such opportunities come specifically from enhancing the conflict-ending
commitments of participating actors and signaling negative external ramifications of prolonged
violence. As we demonstrate, highly structured IGOs are rather effective at preventing low-level
armed conflict escalation to full-scale civil wars. In this vein, our research also encourages emerging
efforts of highly structured IGOs to coordinate their activities on armed conflict. Some highly
structured IGOs have already established steady communication channels and share knowledge on
this issue. Our findings about the effectiveness of highly structured IGOs in conflict prevention should
further encourage these efforts.

1.4 Outline of the Book
In Chapter 2, we continue with the themes raised here. We further develop the argument that civil
wars have been a particularly perplexing and important issue facing countries around the globe and
the international community. This sets the stage for arguing that novel ways of examining civil wars
and their management are needed. Specifically, we advocate in favor of a developmental view of
civil war and highlight the problem of conflict escalation. The issue of how to avoid conflict
escalation is key to preventing full-scale civil war. Chapter 2 also surveys the existing literature on

civil wars, suggesting that both structural approaches and, especially, explanations focusing on


bargaining failure are central here. Then, applying our insights to the bargaining failure approach, we
identify key problems that would need to be overcome in order for third parties to prevent full-scale
civil war. These challenges include (a) swiftness of response, (b) the will and ability to impose
tangible costs on (and offer benefits to) governments and rebels, and (c) long-term involvement. Our
analysis shows that the typical third-party civil war management approaches (mediation,
peacekeeping, and intervention) fail to adequately address one or more of these issues, making them
inadequate tools for the prevention of low-level armed conflict escalation to civil war. This
motivates our argument in favor of focusing on a different type of third-party that could arguably play
a particularly constructive role in civil war prevention. Highly structured IGOs are these third
parties.
In the first part of Chapter 3, we define highly structured IGOs and list all organizations that match
the definition. Of particular importance are institutional features that give these institutions leverage
over member states. We also describe temporal and spatial patterns in the evolution of highly
structured IGOs and states’ participation in them. The second part of the chapter presents the core of
our theoretical argument, focusing on the interplay between civil war development and highly
structured IGOs. Here, we argue that highly structured IGOs have an inherent, vested self-interest in
the domestic peace and stability of member states. Additionally, due to their institutional structure and
command over substantial tangible resources, highly structured IGOs can act quickly to alter the costbenefit calculations of both the government and rebel sides in their domestic bargaining interaction.
The stakes of highly structured IGOs in member countries’ internal peace and stability are enduring.
This means that these IGOs can continue to deter rebels and governments from steps toward conflict
escalation long after an initial ceasefire. Such enduring involvement is critical to overcoming the
credible commitment problem. This satisfies all three of the conditions for successful conflict
management and civil war prevention that we identify in Chapter 2. Chapter 3 ends with a description
of exactly how highly structured IGOs engage in sanctioning behavior and with the exposition of our
main hypothesis.
The main purpose of Chapter 4 is to put the hypothesis to a systematic empirical test. To this end,
we present a quantitative research design for examining the role of highly structured IGOs in a large

number of low-level domestic armed conflicts. We describe our empirical domain, unit of analysis,
variables, and note our statistical methodology. In keeping with standard practice, we define a lowlevel armed conflict as the occurrence of politically motivated violence within a country resulting in
at least 25 battle deaths within one year. 33 Our main outcome variable, escalation to civil war,
delineates whether a low-level armed conflict became a full-scale civil war and surpassed the
threshold of 1,000 casualties, a long-standing cut-off in the literature on civil wars.34 Since World
War II, roughly one-third of more than 260 separate low-level armed conflicts have escalated to civil
war. The analysis provides strong evidence in favor of the hypothesis that countries that belong to a
larger number of highly structured IGOs face a significantly lower likelihood that an emerging lowlevel armed conflict on their territories will escalate to a full-scale civil war. The impact is
substantial: the likelihood of escalation for typical cases at the higher end of participation in highly
structured IGOs is cut to less than half compared to cases at the lower end.
Chapter 5 further probes the validity, robustness, and additional implications of the main finding
presented in Chapter 4. We present a series of analyses to that end. For instance, we investigate
whether the correlation of IGO participation and lower escalation risk is spurious and find that this is
highly unlikely to be the case. We also show that low-level armed conflicts in countries that
participate in larger numbers of highly structured IGOs indeed end in the more desirable outcome, as


our theory suggests. They are more likely to end through peace settlements than by one side defeating
the other militarily before escalation to civil war took place. Examining a key mechanism behind our
theory, we also focus in particular on those highly structured IGOs that have the greatest ability to
impose costs and offer benefits to member governments and rebels. This helps us isolate not just the
organizational structure, but also the ability and will to impose costs, as the factors that drive our
main finding of a decreased escalation likelihood. In a final set of different analyses, we expose our
initial findings to more scrutiny by comparing them to other common correlates of armed conflict and
civil war.
The goal of Chapter 6 is to provide more detailed and nuanced qualitative case evidence
underpinning our argument and main empirical finding. We demonstrate that highly structured IGOs
indeed undertake the particular conflict-preventing activities identified by our theory, and that rebels
and governments respond to these activities. Specifically, we discuss evidence that highly structured
IGOs threaten and sanction member states that are at risk of conflict escalation. Furthermore, highly

structured IGOs offer benefits conditional on successful resolution of low-level armed conflicts.
Finally, we show that member-state governments are aware of IGOs’ concerns and respond to their
pressure—as do the rebels. We use evidence from three different domestic armed conflicts and
follow their developmental trajectories. We select conflicts with different escalation outcomes and
different levels of participation in highly structured IGOs to illustrate the link between these two
variables.
The first case is Indonesia’s response to East Timorese demands for autonomy in the late 1990s,
with Indonesia being comparatively well integrated into highly structured IGOs. The case illustrates
the constructive influence of highly structured IGOs in managing the East Timor Crisis, during which
a very real possibility of a civil war was averted. The second case is that of Ivory Coast in late 2010
and early 2011. With this case, we show that even in instances in which there is a more traditional
military response (i.e., a French intervention) in an emerging conflict, highly structured IGOs still
play an important conflict-managing role. As our narrative demonstrates, highly structured IGOs were
instrumental in placing sanctions on a regime that was trying to hang on to power despite losing
elections, as well as providing incentives to the rebels to settle the conflict. While French forces
actively helped remove the old regime from power, highly structured IGOs were arguably key to
preventing further violence and getting the opposing sides to settle. Through this influence, they
prevented future conflict escalation and set the stage for subsequent peace. This is no small feat, as
research on military interventions demonstrates that such interventions often prolong domestic
conflicts and thus actually contribute to civil wars.35 That such an outcome did not take place in Ivory
Coast is arguably a result of the escalation-preventing involvement of highly structured IGOs.
The early stages of the conflict in Syria in 2011 constitute our final case. Despite the undesirable
eventual outcome of conflict escalation and its dramatic consequences, this case is consistent with our
theorizing. It demonstrates that in a countrywith a shallow portfolio of participation in highly
structured IGOs, the international community had few ways of curbing the emerging violence. Due to
few memberships, the Syrian government faced few constraints or potential benefits for avoiding
escalation. Thus, it had few incentives not to continue to fight the rebels and saw little point in settling
with and accommodating the opposition. Without highly structured IGOs’ counterweight to curtail the
government in its desire to fight the rebels, the rebels themselves saw little reason to stop their armed
resistance. In addition, the case illustrates the difficulty of obtaining the necessary agreement among

conventional external powers, either states or the UN Security Council, to take decisive action. The
result is a particularly brutal and deadly civil war.


The final chapter of the book, Chapter 7, offers concluding thoughts. We summarize our argument
and findings, as well as discuss five key lessons learned from our analyses. These lessons include
suggestions for future research on political violence, conflict management, and international
organizations. In the same context, we present additional implications for policy-makers concerned
with managing armed conflicts and point to opportunities for making use of our findings to further
capitalize on the role of highly structured IGOs in preventing conflict escalation.


2

Managing Civil Wars from the Perspective of Their Development

2.1 Introduction
In this chapter, we lay the foundation for our theoretical argument about the role of international
institutions in civil war prevention. We do this by first examining civil wars, noting their
consequences, and providing an overview of the state of scholarly knowledge regarding their causes.
We divide the scholarly literature on the causes into two broad categories: the structural perspective
and the bargaining approach, focusing on interactions between governments and rebels. Along the
way, we note what the findings from each perspective imply for policymakers interested in conflict
management. The discussion shows that while existing theoretical approaches have provided much
valuable insight and there have been many successful policy efforts to manage civil wars, these wars
nevertheless continue to occur with disturbing frequency and consequences. We argue that, therefore,
an ontological shift in how we think about civil wars and their management is necessary. Management
efforts have largely focused on how to end already ongoing civil wars and how to prevent their
recurrence after (interim) periods of peace. Many of these efforts occur rather late, after civil wars
have already caused severe suffering and damage. Instead, we argue that it is necessary to think about

ways of managing domestic armed conflicts, with an eye toward civil war prevention.
In the second part of the chapter, therefore, we examine civil wars from the perspective of their
development. Thinking about civil wars in developmental terms allows us to consider the issue of
escalation of low-level armed conflict into full-scale civil war. Importantly, we can then ask what
can be done to prevent this escalation, thereby avoiding full-scale civil war. Relying on insights from
the bargaining approach, we analyze the dynamics in the bargaining between a government and an
opposition group or potential rebels. The analysis suggests that changing the government’s and rebels’
cost-benefit calculi regarding the costs of escalation early in their emerging armed conflict can have
the positive effects of making a peaceful settlement more desirable in both the short and long run.
Considering both short and long time horizons helps ensure that the peaceful settlement not only is
reached, but also remains attractive even after the fighting has stopped. The latter issue helps guard
against escalation through conflict recurrence at a later point.
Third parties in the international environment constitute one important set of actors that can
potentially affect the cost-benefit calculations of rebels and governments engaged in an armed
struggle, and thereby provide potent antiescalatory incentives. Yet, as we discuss in the third part of
the chapter, forms of third-party involvement that have been receiving much scholarly and practitioner
attention—such as mediation efforts, peacekeeping, and diplomatic, economic, and military
interventions—suffer from some inherent shortcomings that often make them inadequate for the task of
preventing the escalation of low-level government-rebel armed conflict into full-fledged civil war.
We provide the logic underlying this conclusion. It generally rests on three observations. First, these
conventional types of policy efforts are simply not geared to address the conflict before escalation to
civil war has already occurred. Second, third parties’ self-interests create time-inconsistency
problems that undermine their efforts to stop armed hostilities. That is, third parties involved in


mediation, intervention, and peacekeeping efforts tend to sit out the emerging conflicts on the
sidelines all too often, or for too long. They hence cannot be reliably counted on to become involved
in emerging armed conflicts at all, or, if they do so, to act quickly enough to prevent escalation to
civil war. Third, these efforts’ long-term effects are questionable. Most mediation, intervention, and
peacekeeping missions are involved only in the short to medium term, eventually leaving the conflict

zone.
Instead, a novel understanding of the available options for preventing escalation is needed. The
core of this understanding is that conflict prevention benefits a good deal from a third-party actor that
has a more automatic, consistent, and long-term self-interest in preventing civil wars. We argue that
such an actor can be found in the form of a subset of intergovernmental organizations (IGOs) to which
the state at risk of civil war already belongs. While this topic is introduced in this chapter, its full
exploration follows in Chapter 3.

2.2 Causes, Consequences, and Management of Civil Wars
Conservative estimates of civil wars’ cost in human life in the post–World War II period range
between 5 and 7 million deaths.1 And in addition to a continuously mounting toll of casualties, civil
wars have devastating social and economic consequences. For example, public health institutions and
procedures of post–civil war states are often severely disrupted,2 and subsequently, populations of
such states face higher risks of suffering or dying from various infectious diseases.3 Displaced
persons fleeing armed conflicts can destabilize politically not only neighboring countries, but also
whole regions.4 Furthermore, civil wars inhibit economic growth within affected countries5 and their
neighbors6 by harming political stability, 7 individual households, and the labor force;8 these
ramifications of civil war are typically stronger than those suffered after interstate war. Finally,
recent studies report that civil wars trigger high levels of political and social intolerance.9
These consequences of civil wars are in part responsible for setting the stage for future civil wars,
making civil war recurrence a major problem for countries that have experienced civil wars in their
past. That is, the economic, social, and political consequences of civil wars are also the core factors
predicting civil war onset, putting states into the so-called conflict trap from which an escape is
difficult. The result is that nearly half of all civil war countries return to violence within five years of
the cessation of hostilities, making civil war recurrence one of the greatest problems on the security
map of the world.10 Accordingly, preventing civil war recurrence has taken on increased salience for
policymakers and academics. Yet, the fact that civil wars continue to occur, and recur, to this day
with grim regularity11 suggests that there is much room for improvement in how the international
community deals with civil wars—but not for lack of trying. Given civil wars’ dire consequences, it
is not altogether surprising that scholars and policymakers have expended much effort over the past

decades in order to both understand causes of civil wars and develop policies for managing them. To
these topics we now turn.
2.2.1 Civil Wars from the Structural Perspective
After over twenty years of intense research, scholars have identified a set of factors typically
associated with civil wars that tap into both the motive and opportunity to rebel. For instance, lowl
evels of economic development create grievances that motivate rebellion. Poverty also renders the


opportunity cost for rebelling low, as well as generally depriving the state of resources needed to
successfully deter or quickly stamp out potential rebellions.12 Inaccessible terrain makes it easier for
rebels to find sanctuaries and harder for the state to destroy rebel forces,13 whereas larger
populations offer greater opportunities for rebel recruitment.14 Meanwhile, unstable or
underdeveloped political institutions, or institutional setups that systematically exclude ethnic
minorities, may not be able to channel and resolve grievances,15 so they tend to build up; mounting
frustrations can turn into reasons to rebel.
Additionally, easily extractable but highly valued natural resources (e.g., oil, diamonds, and
timber) may provide both a temptation to rebel out of greed or may function as a means for funding a
rebellion that may be seeking to address a “legitimate” grievance (e.g., insufficient political rights).16
Finally, though often investigated, the factor of ethnic diversity is only occasionally clearly linked
with civil war. 17 Ostensibly, ethnic divisions provide an opportunity for governments to decide
whom to provide and whom to deny political benefits or scarce economic resources. And for
potential rebels, ethnic divisions help identify and recruit compatriots, mobilize populations, and
highlight grievances. Even though scholars have had a hard time demonstrating a clear, systematic
link between raw ethnic diversity scores and civil war onset, more recent research shows that what
matters is whether ethnic minority populations are systematically excluded from power and wealth
sharing; if they are, this provides civil war–motivating grievances.18
A smaller subset of civil war research has started to tie in the issue of how transnational factors
may be relevant to civil war onset as well. For example, civil wars may spill over into neighboring
countries via refugee flows. Additionally, rebels seeking sanctuaries across borders can carry
violence into neighboring countries, especially if they are pursued by their governments.19 Rebellions

in one country may encourage rebellions in others through the diffusion of norms, identities, or
motivations.20 In the recent Arab Spring, for instance, demands for political reforms quickly spread
across states in North Africa and the Middle East.
Foreign interventions, another transnational factor, are typically thought of as occurring in the midst
of civil wars, presumably often triggered by concerns over which side is likely to prevail. But some
scholars point out that the expectation of a future partisan intervention also could encourage or
discourage the opposition in its decision-making overwhether to rebel, depending on whether it
anticipates a third-party to intervene and, if so, on whose behalf.21 Returning to the Arab Spring,
Western involvement in Libya may have created what Alan Kuperman has termed a moral hazard of
intervention,22 leading political opposition in other states in the region to falsely believe that foreign
help would be forthcoming if they pressed their regimes for reforms.
A case such as the Bosnian civil war from 1992 to 1995 illustrates many of these dynamics well.
The fighting resulted in an estimated 200,000 casualties and 2 million displaced persons. These fled
mostly to the neighboring countries of Croatia and Serbia, severely taxing their resources and
somewhat destabilizing them as well. Additional refugees fled to countries such as Germany,
Sweden, Switzerland, Canada, and the United States. This illustrates the broad reach of the war’s
consequences. The fighting in Bosnia also severely damaged the country’s infrastructure while nearly
completely wiping out the country’s economy, as well as its public health and educational structures.
Many of the related aftereffects are quite pronounced to this day.
Prewar Bosnia lines up relatively well with structural accounts of civil wars. It belonged to the
lower end of middle-income countries and had rather young and untested political institutions. The
design of these new institutions implied a change to the prior setup, which had favored the Serbian


population. The new design would allow two ethnic groups to ally and exclude the third. This
occurred when the Muslim and Croatian sides pushed for a referendum on Bosnia’s independence
from Yugoslavia, which the Serbian population did not want. Once violence started, the heavily
mountainous terrain made it relatively easy and safe for the Serbian rebel forces to lay long-term
sieges on valley cities such as Sarajevo and Bihać, persistently shelling the trapped populations for
over three years. Reflecting transnational dynamics, the secession of Bosnia was part of a broader

wave of ethnic secessions in the region occurring in the aftermath of the end of the Cold War.
Furthermore, one could speculate that the secessions were in part encouraged by the erroneous
expectation of Western military help with shedding socialist political systems. And on the surface at
least, the war seems to mesh well with the idea that ethnic diversity is a contributing cause, as
members of Serbian, Bosniak Muslim, and Croatian ethnic groups have all fought one another. But a
deeper analysis also reveals important instances in which ethnic and religious dividing lines were
less clear cut. For example, Bosniak Muslims fought alongside Christian Croatians against Christian
Serbs; and Bosniak Muslims from the westernmost reaches of Bosnia allied themselves with
Christian Serbs against the Muslim-dominated government in Sarajevo. The latter example means that
there was intra-Bosniak Muslim fighting—a phenomenon that occurred on a smaller scale on the
Croatian side as well.
A result of the war is that Bosnia seems to be a prime candidate for the conflict trap and civil war
recurrence, with a significantly damaged economy, fragile and problematic institutions, poor public
health and educational systems, a tense and unstable political situation, and other problems. That this
outcome has not (yet) occurred may be due to over two decades’ worth of massive international
assistance. Yet, many observers have questioned whether the war in Bosnia will resume after the
eventual departure of international personnel and suspension of aid.23 This suggests that one of the
keys to preventing future fighting may lie precisely in the long-term—ideally permanent—
involvement of international actors. We return to the feasibility of this scenario later in this chapter.
Although the structurally oriented stream of research has increased our understanding of civil war
tremendously, converting the related causes into potential solutions to actual civil wars is
unfortunately not all that straightforward. Policy implications stemming from conclusions based on
structural arguments, for example, would include eradicating poverty and developing stable, mature,
and ethnically inclusive democratic institutions. Although such goals are certainly noble, they are not
that easy to achieve. The transformation of poorer countries with unstable political institutions into
wealthy, stable, and inclusive democracies is challenging and likely to take generations. 24 Decades of
efforts and resources provided by the World Bank and regional development banks in countries such
as Haiti and Uganda attest to the difficulty of successfully bringing about economic and political
development. Furthermore, altering other factors from the list—for example, preempting expectations
of future foreign military assistance, reducing large populations, decreasing ethnic diversity, and

flattening mountainous terrain—may not be desirable or even possible. In short, many of the common
structural factors linked with civil war can generally be understood as either immutable, especially in
the short to medium run, or undesirable to change.
2.2.2 Civil Wars from the Perspective of Bargaining Theory
A more recent stream of research, in contrast, takes a more dynamic view of why and how civil wars
develop. In its simple, abstract form, bargaining theory25 suggests that wars are a result of an
interaction between two actors that disagree over the distribution of some valuable resource. Both


actors are rational, in the sense that they are motivated by prospective gains, but they are also
sensitive to the costs that they may have to pay to achieve their goals. Thus, throughout their
interactions, both actors follow cost-benefit calculations to determine whether fighting or a peaceful
settlement would be more beneficial. Wars occur when the benefits of prospective gains from fighting
exceed the anticipated benefits of peace.
But fighting can also be avoided, or stopped, if these cost-benefit calculations change. This can be
due to, for example, mounting costs, changing relative power balance between the sides, or better
information about each other’s strengths and weaknesses. Apeace deal that divides a contested
resource according to the sides’ relative strengths can then presumably be struck and (further) fighting
avoided. This bargain, however, is subject to the expectation that both sides will honor the peace
deal not only now, but also in the future. The issue of whether each side can be trusted to be truly
committed to peace—rather than simply waiting for a future opportunity to revise the new status quo
to be more in its favor—is particularly tricky. This credible commitment problem helps explain why
potential peace deals fall through—that is, why fighting sometimes occurs or resumes even though the
actors can agree in principle on how to divide contested resources.
Civil wars and the bargaining framework
We follow this brief sketch of the bargaining framework by applying it to civil wars. 26 Rebellions
leading to civil wars take place because the rebels desire to alter the status quo, which favors the
government, in terms of the distribution of political and/ or economic benefits. The rebels thus seek to
negotiate a new bargain that would provide them with a more favorable distribution of political,
economic, or some other resources. If the disputants (the government and rebels) know each other’s

strengths, they can theoretically negotiate a revised status quo that reflects their relative power and
hence avoid the costs of (prolonged) fighting.
Initial stages of the fighting have the side effect of revealing information about the relative strengths
of the rebel and government sides. For example, the rebels’ ability to recruit supporters, access
resources, develop sanctuaries, or secure financing is revealed as the fighting proceeds. Similarly,
the rebels learn more about the government’s will to fight, the skill of its soldiers, popular support,
and tactics. So while information problems that would reveal the true relative power of each side are
not necessarily trivial, as soon as the fighting begins, the two sides start to develop an understanding
of each other’s strengths and weaknesses. This information can then be used to allow the two sides to
strike a bargain that would redistribute the contested resources in proportion to their relative
strengths. Yet overcoming the information problem alone does not necessarily mean that the fighting
will be stopped all that easily.27
Although both the government and rebels may feel ready and willing to strike a bargain to end the
fighting—and even agree over the settlement of grievances and demands—this process is seriously
hampered by what is known as the commitment problem. In particular, the government often cannot
credibly commit to a fighting-ending agreement. The specific sticking point is that such an agreement
typically calls for the rebels to disarm. The rebels have to be more or less disarmed to fulfill one of
the basic definitions and functions of a state: to have the sole monopoly over the legitimate use of
coercive force on its territory. Astate cannot have two different militaries, so the rebels have to lay
down their arms; in some cases, the now former rebels are then integrated into the state military
structure. Yet, once the peace deal is signed and the rebels have disarmed, the government is placed
in a powerful position, as the sole armed actor. Rebels likely may fear that such a scenario may