Show Summary Details

Page of

Printed from Oxford Research Encyclopedias, Politics. Under the terms of the licence agreement, an individual user may print out a single article for personal use (for details see Privacy Policy and Legal Notice).

Subscriber: Google Scholar Indexing; date: 27 February 2024

Effectiveness of Peacekeeping Operationsunlocked

Effectiveness of Peacekeeping Operationsunlocked

  • Jessica Di SalvatoreJessica Di SalvatoreDepartment of Politics and International Relations, University of Oxford
  •  and Andrea RuggeriAndrea RuggeriDepartment of Politics and International Relations, University of Oxford


Peacekeeping has been one of the main conflict management tools used by the international community to restore or safeguard peace and security. Since 1948, the United Nations has established 70 peace operations and has substantially evolved, adopting approaches to peace that extend beyond purely military concerns. Indeed, the promises of peacekeeping as effective instrument of conflict reduction may, to some extent, explain the evolution toward multidimensional missions and the unprecedented number of peacekeepers deployed in the last decade. As consequence, the growing importance of peacekeeping effectiveness has sparked a new wave of research that empirically investigates whether and under which conditions UN peacekeeping works.

Peacekeepers are mostly deployed in conflict or postconflict environments where violence is either ongoing or lingering. Thus, violence remains a priority for peace missions. Consequently, peacekeeping is deemed successful or effective according to whether it curbs conflict in several dimensions. Effective missions are those responsible for decreasing the intensity of battle violence, protecting civilians, and containing conflict diffusion and recurrence in the postwar phase. Each mission, however, is deployed in different contexts and operates under variable conditions that affect the operation’s capacity to influence conflict. Concerning mission features, peacekeeping success is more likely when large contingents are deployed under robust mandates. Mission type, size, and composition signal credible commitment from the international community and empower peacekeepers to halt violence while guaranteeing the implementation of peace agreements. These nuanced understandings of peacekeeping stem from the availability of new data on both conflict and peace operations at the national and subnational levels of analysis. Moreover, the empirical study of the effectiveness of peace operations has recently been flanked by simulation-based forecasting, field experiments, and surveys investigating local-level outcomes of peace missions.

Unsurprisingly, the focus on violence and conflict outcomes as indicators of success is debatable. First, in dealing with violence, peacekeeping operations produce spillover effects that are largely neglected, such as refugee flows and terrorist violence. Second, given the wide range of functions performed by UN peacekeepers, including electoral assistance, economic reconstruction, and state building, it is reasonable to include these aspects when defining effectiveness. Third, and relatedly, no assessment of short- versus long-term implications of peacekeeping for political, social, and economic development in the host country has been forthcoming. While reducing infant mortality, inequality, and crime are not necessarily tasks for peacekeepers, it is vital to study whether and how UN missions may have shaped the quality of peace in host countries.


  • Contentious Politics and Political Violence
  • World Politics


Since its establishment in 1945, the United Nations has authorized more than 70 peace missions. In addition, regional organizations, such as the African Union and the European Union, have conducted their own peace operations—an additional 65 missions between 1946 and 2016 (see Williams, 2017). As Figure 1 shows, the number of countries contributing to UN peace operations, merely 45 after the end of the Cold War, has reached almost 120 in the contemporary period (Figure 1, bins, left-hand Y-axis).1 In the same temporal span, the number of “Blue Helmets,” or peacekeeping troops, increased from 10,000 to almost 100,000 globally (Figure 1, thick line, right-hand Y-axis).

Figure 1. Number of peacekeepers and number of countries contributing (Kathman, 2013). Figure created by authors.

The United Nations’ increasingly active role after the end of the Cold War has drawn further attention to peacekeeping. The critical questions for both academic and policy-making communities are: Does peacekeeping work? And are peacekeeping operations effective?

This article does not aim to provide complete answers to these questions, seeking instead to review and contextualize the wealth of quantitative research on this very topic. Importantly, before appropriate answers are found, lingering challenges must be addressed. First, what is meant by effectiveness? Several conceptualizations and operationalizations have been suggested as measures of peacekeeping’s utility, such as compliance with ceasefires (Fortna, 2008), the combination of peaceful spells and stable polities (Doyle & Sambanis, 2000), protecting civilians from violence (Bove & Ruggeri, 2015; Hultman, Kathman, & Shannon, 2013), and reducing levels of violence between belligerents (Hultman, Kathman, & Shannon, 2014).

Second, how should an operations’ effectiveness be studied, and which mission features matter? Some scholars have focused on the mere presence of peace operations or have extended their analysis to different mission mandates. More recently, quantitative research has investigated whether the size of a mission, its composition of troops or observers, and the heterogeneity of nationalities can affect its effectiveness.

Third, does it matter how the effectiveness of peace operations is studied? The research design selected to evaluate and study the effectiveness of peace operations is, itself, a salient issue. For example, are peace operations deployed to the easy or hard cases? How have specific research designs accounted for this nonrandom assignment of the treatment and omitted viable bias? Moreover, further debates on the empirical study of peace operations have focused on temporal and spatial analytical units: should researchers focus only on country–year measures to understand the effectiveness of peace operations or use more granular temporal and geographical units? Missions have been deployed in very large countries such as Angola, Democratic Republic of Congo, or Sudan, but the physical deployment occurs in more specific locations. Should the capacity of peace operations to stop violence at the subnational level also be studied?

The remaining sections of this article discuss the aforementioned issues and questions and are organized as follows. First, the conceptualizations and operationalizations used in research on the effectiveness of peace operations are reviewed. Second, the theoretical framework usually used to study peace operations effectiveness in quantitative research is summarized. Moreover, advantages and disadvantages of the range of research designs used in the empirical literature is discussed. Third, the main findings using tables capturing results of different dependent and explanatory variables are presented, including trends of peace operations in terms of size of deployments, contributing countries, and evolution of mandates. The article concludes with a section that highlights what remains absent in empirical studies of peacekeeping effectiveness.

Quantitative work on peace operations is available,2 and qualitative scholars have provided substantial and critical contributions to understanding the effectiveness of peace operations (see Autesserre, 2010, 2014; Bellamy, Williams, & Griffin, 2010; Howard, 2008). However, the study of peacekeeping must handle an emergent imbalance, given that the United Nations has a longer history of peace operations than other regional organizations and, consequently, dominates the quantitative literature. Furthermore, this article focuses on peace operations in civil wars, mostly due to the structural changes in conflict dynamics and in the deployment of peace operations the post-Cold War era, and the summary, while inclusive, is not exhaustive.

What Do We Mean by the Effectiveness of Peace Operations?

Defining peace, theoretically and empirically, has been a perennial issue for scholars of international relations and peace studies. The concept of “negative peace,” or the absence of war, has been challenged by differing perspectives on “positive peace” (Galtung, 1969), including concepts such as “participatory peace” (Doyle & Sambanis, 2000) to “bottom-up peace” (Autesserre, 2010). The operationalization of conflict has also been defined according to different thresholds (Gleditsch et al., 2002; Stinnett et al., 2002), specifying different targets of violence (Eck & Hultman, 2007), and referring to temporal and within-country variation (Sundberg & Melander, 2013). Hence, the definition of the effectiveness of peace operations is vigorously debated in the peacekeeping literature, and it has only become more relevant as the field has adopted systematic empirical assessments of mission performance.

Since its inception in the early 1990s, the so-called “second wave” of peacekeeping studies struggled to define and measure success (Fortna & Howard, 2008). In 1988, Paul Diehl published what may be one of the first studies on peacekeeping effectiveness that moves beyond the descriptive approach of the first wave of peacekeeping studies (Fortna & Howard, 2008). The core and main criterion for peacekeeping success, according to Diehl, depends on the capacity of the mission to limit armed conflict and prevent its renewal. A secondary criterion concerns the mission’s ability to foster peaceful settlement (Diehl, 1988). Based on these criteria, Diehl evaluates six UN missions and concludes that peacekeeping is more likely to succeed when it is consent based, when it deploys neutral lightly-armed personnel, and when the geographical conditions allows separation of combatants and detection of violations. In general, he also observes that peacekeeping is mostly successful in its core objective (i.e., limiting conflict) but contributes less to peaceful settlements. A similar set of criteria has been proposed by Brown (1993) who explores whether a specific mission fulfilled the actual mandate, reduced conflict, and contributed to peaceful resolution. The last two criteria correspond to Diehl’s concept of success, while the first criterion circumscribes the expectation of success to the mandated aims of the mission. In criticizing Diehl’s approach, Johansen (1994) argues that applying the aforementioned criteria to each mission is controversial because Diehl focuses on the renewal of violence or conflict settlements—conditions over which peacekeepers may have had, at best, only a marginal impact. By considering outcomes that are indirect results of peacekeeping activities, Diehl’s evaluation can be too optimistic or too pessimistic, depending on the mission under scrutiny. In anticipation of continuing concerns regarding the impact of the peacekeeping mission(s), Johansen proposes an alternative framework that evaluates the effects of the mission on the quality of life of local population. He places emphasis on “what-if” counterfactuals, namely the likely conditions had a peace mission not been deployed.3 In a sense, both the evaluation of counterfactual scenarios (e.g., the expected outcomes without a peacekeeping mission) and accounting for which tasks peacekeepers are mandated to perform, echoes recent appeals for evaluating success against a set of baseline expectations (Diehl & Druckman, 2010; Han Dorussen, 2014). Ultimately, the basic idea of an operational standard for success may be misguided. Diehl & Druckman warn against using a single standard for success because “conclusions drawn on only one set of standards will lead policymakers to adopt certain policies without being aware of the full consequences of those policies” (2010, p. 5). To avoid unintended consequences of peacekeeping operations, what success entails in a broader sense must be specified.

Over the years, the debate on the concept of peacekeeping effectiveness has centered on one crucial dimension: peace survival. However, other dimensions shape this debate, such as the divide between systematic and interpretative approaches. This section, however, focuses on how scholars advocating empirical evaluations of peacekeeping have contributed to the concept of peacekeeping effectiveness and, consequently, how its operationalization in statistical analysis is informed by large-N data.

The divide between short- and long-term effects of peace missions has drawn significant attention to how missions reduce different dimensions of conflict and strengthen postconflict peace (see Sambanis, 2008). Hence, notwithstanding the much-discussed criteria proposed by Diehl, empirical evaluations of peacekeeping have embraced a parsimonious conceptualization of success. Short-term analyses of peacekeeping success tend to focus on whether missions contain and limit conflict violence. Researchers have measured the conflict-reducing effect of peace missions in different ways. While the operationalization of success as a dependent variable varies across studies, all alternatives share a focus on the immediate (yearly or monthly) conflict-reducing capacity of peacekeeping. The most common measure of success in early empirical studies of peacekeeping effectiveness was peace duration. Fortna (2003, 2004a, 2004b) operationalized effectiveness as peace spells terminating when new violence occurred within the same conflict dyad after a ceasefire. Similarly, Hultman et al. (2015) use postconflict stability to evaluate mission success. However, their analysis differs from Fortna’s not only in how peacekeeping is measured (i.e., size of deployment rather than mere mission presence) but also in how peace duration is measured. As in Hartzell et al. (2001), peace duration is the count of months of peace in the postconflict period.

This strand of literature assumes that peace is an established condition and that a peacekeeper’s job is to maintain it. Notably, however, the United Nations started sending troops to ongoing conflicts soon after the end of the Cold War. Thus, missions should be evaluated not only against their ability to keep peace but also according to their capacity for making peace. Relatedly, success was then evaluated in terms of war duration. Successful peace operations are expected to reduce the length of conflict and facilitate peaceful settlement (or at least ceasefires). For example, Beardsley (2012) looks at stalemates and victories after crises and investigates whether peacekeeping (and other types of military involvements) lengthens the time parties take to compromise or emerge victorious compared to diplomatic actions. DeRouen, Jr. (2003) further disaggregates UN diplomatic actions into mediation, dispatch of observer groups, calls for action, and emergency deployments. All these actions are deemed possible determinants of success, measured in terms of termination of a crisis. Notably, using war duration as the standard of success tends to conflate peacekeeping with peacemaking—both studies referred to here do not clearly distinguish peacekeeping from other types of intervention. Nonetheless, prolonging peace is not the same as shortening wars. Yet, these have both become important standards for success, and contemporary peacekeeping is often expected to produce both outcomes. The propensity for peacekeeping to either prolong peace or shorten war is likely determined by different configurations of factors, such as breadth of the mission’s mandate, military composition, and duration and timing of deployment. Gilligan and Sergenti (2008) present one of the few studies that evaluates peacekeeping’s capacity to shorten war and sustain peace. Ruggeri et al. (2016a) propose a measure of localized success by exploring the impact of UN peace mission presence and size on the duration of conflict and its onset within subnational 50-km × 50-km geographical areas within countries. Because the authors condition conflict onset on peace in the previous year, they are evaluating peacekeeping relative to its capacity to shorten conflict and lengthen peace.

Whether peacekeepers are good at ending war is crucial because ceasefire or other temporary settlements are important conditions for reaching peace agreements. Greig and Diehl (2005) point out that peacekeeping and peacemaking are interrelated conflict-resolution tools. Here, successful peacekeeping should not just shorten wars but also enable and strengthen peace processes (this is the second criterion of success as defined by Diehl, 1988). Not only do Greig and Diehl find that peacekeeping is negatively associated with likelihood of negotiations or mediation, but peacekeeping also decreases the chances of settlement when such processes are undertaken. Thus, while peacekeeping may halt or limit hostilities, it has strong negative externalities on peacemaking and, thus, on the long-term resolution of the conflict. The authors refer to this as the peacekeeping–peacemaking dilemma.

Another and perhaps most obvious outcome for short-term success is decreasing conflict intensity. In one of the first analyses of peacekeeping effectiveness, Heldt (2001) evaluates peacekeeping success based on presence of civil war. Using a global sample of all civil wars from 1960 to 1999, he explores whether mission or conflict features have different marginal impacts on war. However, the research design used in this analysis, in line with many others, does not tackle the problem of nonrandom assignments for peacekeeping missions.

Most recent work has advanced the research design to empirically test peacekeeping success. In particular, a distinction is usually made between battle-related deaths and one-sided violence (or violence against civilians). Hultman et al. (2014) find that larger deployment of UN peacekeeping troops (but not UN police and observers) reduces the intensity of violence on the battlefield in ongoing civil wars on a monthly basis. The authors replicate a similar analysis focusing on one-sided violence, and they highlight somewhat different conclusions suggesting that battlefield violence and civilian killings results from distinct dynamics (Hultman et al., 2013). Kathman and Wood (2014) also examine each mission’s composition and its impact on civilian victimization in the aftermath of conflict. Bove and Ruggeri (2015) propose an additional factor associated with protection of civilians beside size and personnel type, namely the national heterogeneity of UN contingents. However, civilian victimization can occur on different scales and can even reach genocidal levels. Peace missions have been deployed in scenarios with high risk of genocide; thus, they have also been evaluated in the literature based on their capacity to quell or forestall genocides and politicides (Kathman & Wood, 2011).

Geographic containment is another criterion used to judge whether peace missions work. Beardsley (2011) illustrates how the presence of peacekeepers reduces the risk of conflict onset in neighboring states. This containment effect also exists within countries. Indeed Beardsley and Gleditsch (2015) argue that peacekeeping constrains diffusion of violence during civil wars. In their analysis, using a subnational and geographically disaggregated research design, they assess the extent to which peacekeepers manage to contain and shrink conflict areas.

All criteria for success presented so far share a limited time perspective focusing on the yearly or monthly effect of peacekeeping on a set of outcomes. Other scholars have adopted a longer temporal span, which allows the evaluation of broader peacebuilding objectives. For example, Diehl et al. (1996) discusses the risk of renewed interstate conflict within 10 years; Beardsley (2013) also posits that UN missions reduce peace fragility in the long term. Diplomatic engagement or sanctions, he finds, are only temporary stopgaps for violence, which usually re-erupts within 5 to 10 years from conflict termination. Among others, the most notable definition of long-term success is provided by Doyle and Sambanis (2000). They propose a measure of success that covers 2, 5, and 10 years of the postwar phase. Two versions of successful peacebuilding are proposed. The lenient version is concerned with negative peace and requires (a) no renewal of war, (b) no low-level violence and (c) uncontested sovereignty. The strict version adds another requirement to these: democratization (or political openness). This definition of success (and oftentimes its operationalization) has been used in numerous studies: Bigombe et al. (2000), Fortna (2003, 2004a), Gilligan and Stedman (2003), Sambanis and Schulhofer-Wohl (2008), Gizelis (2009), and Kim (2015).

The increasing relevance of peacebuilding and long-term effects of peace operations has produced a growing number of studies focusing on outcomes that are not strictly related to security and conflict. Some of these outcomes are more limited in scope and thus are insufficient for defining success. Others are intertwined with the concept of positive peace. These outcomes include whether specific peacekeeping tasks are associated with more cooperative responses from the local population (Gizelis, 2011; Ruggeri et al., 2012; Dorussen & Gizelis, 2013); the number of violent attacks against peacekeepers (Salverda, 2013; Fjelde et al., 2016); reported cases of sexual exploitation and abuses (Beber et al., 2017; S. Karim & Beardsley, 2016; Nordås & Rustad, 2013); local perceptions of peacekeeping effectiveness (Karim, 2016); satisfaction with the mission performances (Kelmendi & Radin, 2016); democratic transition (Joshi, 2013); democratization (Steinert & Grimm, 2015); as well as economic development (Caruso et al., 2016) and improvements in quality of life (Kim, 2015).

Two main shifts are evident in “what is” the effectiveness of peace operations, both conceptual and analytical. First, an important evolution of the political phenomena is under scrutiny. Or, to put it more technically, a shift in the dependent variable has occurred, from a focus on peace survival to the effects of peacekeeping on violence dynamics. Clearly, this shift is neither complete nor necessarily desirable, but recent empirical studies, on average, exhibit this tendency. This shift has also occurred in consonance with an analytical move toward a comparatively disaggregated study of peacekeeping as the unit of analysis has switched from country-year or conflict spell to models featuring the higher temporal granularity of conflict dynamics (usually monthly variation) while endorsing more precise spatial variation within countries.

The Effectiveness of Peace Operations: Theories and Research Designs

This section briefly summarizes the main theoretical tenants of quantitative empirical study of peacekeeping effectiveness. Most of these theories (e.g., Doyle & Sambanis, 2000; Virginia Page Fortna, 2008; Hultman, Kathman, & Shannon, 2013, 2014) are clearly rooted in the rationalist explanations of conflict and, therefore, suggest that peace operations can modify parties’ incentives by increasing the costs of fighting, mitigating commitment problems, and facilitating information flows among belligerents. Second, this section highlights the data and research designs that have been used to evaluate the effectiveness of peace operations.

Peace Effectiveness and Theoretical Mechanisms

The theoretical arguments put forward to explain the effectiveness of a peace mission hinge on the idea that peacekeeping increases the cost of fighting. Doyle and Sambanis (2000) propose a simple model in which peacebuilding outcomes are a function of three factors: (a) the level of hostility prevalent at the start of the peace process, (b) local capacities for reconstruction and development, and (c) international capacities for peacebuilding mostly thorough UN peace operations. Peacekeeping effectiveness seems to work mostly through three mechanisms: deterrence, commitment, and information (Bove & Ruggeri, 2015; Ruggeri, Dorussen, & Gizelis, 2016b).

First, peacekeeping deterrence helps prevent conflicts from spilling over into noncombatant areas, thus reducing and limiting violence. Second, because ceasefires provide opportunities for government and rebel authorities to increase their bargaining power, the local presence of peacekeepers matters because it commits leaders to follow previously agreed rules, including combatants’ interdiction from civilian areas. Third, information flows can be crucial, as government and rebel leaders often lack information about their relative strength. By providing such information, peacekeepers can assist the peace process. Furthermore, peacekeepers obtain vital information through their frequent interactions with locals, which allow more proactive protection of civilians (Bove & Ruggeri, 2015).

Further elaborating the informational argument, monitoring is a commonly proposed mechanism for effectiveness. Peacekeepers observe and report on parties’ behavior and conflict processes. Thus, the peacekeepers’ presence may spur cooperation by both reducing uncertainty and containing defection, particularly in areas where UN personnel are stationed. Collecting information through monitoring has also indirect effects on the cost–benefit of violence: it reduces chances of surprise attacks and increases the state’s cost of heavily repressing rebels.

Observing and monitoring, per se, are not sufficient conditions for deterring parties from using violence. Indeed, monitoring activities are more effective when troops are deployed instead of unarmed observers. Signaling commitment and willingness to punish violations by deploying large military contingent is key for the success of peace missions. Ruggeri et al. (2016a) highlight a difference between commitment and deterrence. They argue that commitment relies on ensuring that parties do not subvert and violate agreements, while deterrence is used to avoid local actors to spoil peace when national principals loosely control them. As consequence, military presence is even more relevant for deterrence than for commitment, as the latter can ultimately be political at its core. Relatedly, when comparing mission cross-nationally, Fortna (2008) notices that Chapter VI and Chapter VII missions (the latter with robust enforcement mandates) do not have a distinguishable effect on halting conflict. She identifies the mechanisms through which peacekeeping reduces violence as political and economic, and are not, therefore, a function of military capability. But such translations should be undertaken carefully when translating dynamics at the national level to political actors with local dynamics.

Political and economic leverages (or commitment) may be insufficient for deterring local actors from engaging in violence, especially when elites lack control over their constituents. The mechanism of commitment works at the institutional or macro level and is independent of military capacity; for the role of military deployment, it is more salient for deterrence at the local level. Peacekeepers can also play more active roles to signal commitment, instead of relying on passively monitoring compliance, including by enforcing peace agreements or ceasefires and by reassuring parties. Both enforcement and reassurance require more engagement and proactive stances; therefore, they are more reliant on the mission’s military capacity. Enforcement missions are not consent based, which implies that at least one party must be coerced not to use violence. Exerting this compelling effect requires a clear, visible, and credible threat to use force, more than deterrence does. Similarly, reassurance is usually associated with security guarantees offered by the United Nations, whose credibility is predicated on its military presence. In other words, both enforcement and reassurance mechanisms rely on higher degrees of coercion than the others.

Data and Research Designs

While these theoretical mechanisms imply reasonable expectations on the effectiveness and impact of peace missions, their empirical testing presents several challenges. First, data collection on peacekeeping has lagged behind an increasing interest in disaggregating the analytical unit, in time and space. The first quantitative wave (Diehl, 1988; Doyle & Sambanis, 2000; Fortna, 2008) focused mostly on country-years or conflict spells. Only with the second wave has the empirical study of peacekeeping started moving to temporal (monthly) and spatial (subnational) disaggregation (Costalli, 2014; Di Salvatore, 2017; Hultman, Kathman, & Shannon, 2013, 2014; Ruggeri, Gizelis, & Dorussen, 2012; Ruggeri, Dorussen, & Gizelis, 2016a, 2016b). Moreover, new data on different dimensions of peacekeeping are now available, such as personnel contribution to missions (Kathman, 2013), gender composition (Karim & Beardsley, 2016), mission leadership (Bove, Ruggeri, & Zwetsloot, 2017), and georeferenced event data (Dorussen & Ruggeri, 2017).4

A second challenge is causal identification. Peacekeeping is not random, which creates problems when estimating the treatment effect. The evolution of empirical approaches has substantially increased our confidence in the evidence supporting peacekeeping effectiveness. Early empirical approaches used cross-sectional country-level data in which outcomes were compared without correcting for the influence of nonrandom assignment of peace operations to conflicts. In the best case, peacekeeper presence was measured as a dummy, which hid many substantial differences across peace missions. Using a dummy to measure peacekeeping in host countries and comparing them to other civil wars without peacekeeping captures any significant heterogeneity among these groups but hides all variations across missions. Again, the problem of nonrandom assignment has usually been overlooked or has not been addressed.

Gilligan and Sergenti (2008) were the first to propose matching for pre-process data and to allow comparison between most similar cases. Matching allows researchers to balance the treatment and control group based on a set of observable factors to generate meaningful counterfactuals. Once the sample is matched, observations are weighted so that the two groups are comparable. Thus, the difference in conflict-related violence between the matched treated factual (civil wars with peacekeeping) and the untreated counterfactual (civil wars without peacekeeping) is an estimate of the average effect of peacekeeping on violence. Matching has become a popular tool for addressing issues stemming from nonrandom assignment of peace operations and is now widely used in statistical models (Hultman, Kathman, & Shannon 2013; Costalli, 2014; Ruggeri, Dorussen, & Gizelis, 2016b). One problem with such an approach, however, is that nonobservable factors could determine whether civil wars receive peacekeepers. For instance, some countries could be more likely to host Blue Helmets because of political motivations that may not be known or easily controlled for in statistical models. To overcome this issue, other approaches to address selection bias have been proposed, including instrumental variables approaches.5 Vivalt (2014) and Carnegie and Mikulaschek (2016) propose the rotation of seats in the Security Council as possible source of exogenous variation. Vivalt argues that countries are less likely to receive peace operations if they are sitting temporarily in the Security Council when conflict is ongoing. Carnegie and Mikulaschek exploit the idea that when rotation provides seats to more African countries, they use it to get peace operations to nearby states to reduce the negative externalities of conflict. Rotation is based on election, though, so it is not random. Hence, they add as a further instrument the African presidency in the Security Council, which is based on the alphabetical order of council members’ names. Ijaz (2014) proposes supply of peacekeeping as an instrument, arguing that domestic shocks in troop provision reduces the odds of enforcement missions being deployed. As she acknowledges, however, variation in supply explains the lower likelihood of a specific type of mission only. Overall, the instrumental variable approach is an exception rather than the rule in studies on the effectiveness of peace operations, in part because of the challenges in finding strong and accurate instruments.

Recent work has shown promising new research designs and methods in the empirical study of the effectiveness of peace operations. The use of surveys to explore how peace operations affect attitudes toward peace attempts and other conflict dynamics has been encouraging (Mironova & Whitt, 2015; Kelmendi & Radin, 2016). In addition, field experiments in the study of peace operations have been used to increase the internal validity of empirical tests of theoretical mechanisms (Mvukiyehe & Samii, 2010; Beber et al., 2017). Finally, a recent wave of research is providing advanced and sophisticated methods based on empirical estimates to simulate and forecast the effectiveness of peace operations (Hegre, Hultman, & Nygard, 2011).

Empirical Findings and Trends

Over time, the literature has introduced new and more nuanced measurements of peacekeeping as explanatory variables for conflict. These range from mere presence of peace operations (Diehl, 1988) and mandate types (Doyle & Sambanis, 2000; Fortna, 2008), to size and composition of peacekeeping personnel (Hultman et al., 2013, 2014; Bove & Ruggeri, 2015) and, more recently, local presence of peacekeepers (Costalli, 2014; Ruggeri, Dorussen, & Gizelis, 2016b).

Table 1 summarizes the main findings of empirical studies of peacekeeping effectiveness, which includes several outcomes of interest related to conflict and peace processes that are now commonly used as parameters of effectiveness. One of the broadest distinctions is made with regard to mandate type: Variation in success overlaps with variations in mandate type. Doyle and Sambanis (2000) find that the mandate is key for success, and after distinguishing between observer, traditional, multidimensional, and enforcement peace missions, they posit that each has a different impact on peacebuilding and ending violence. Traditional missions, which encompass military deployment for interposition between belligerents, have no impact on these two outcomes, whereas multidimensional missions have a beneficial effect on both participatory peace and tend to shorten conflicts.

Table 1. Peace Operation Effectiveness and Peace/Conflict Outcomes



Peacekeeper Presence

Troops Size



Peace duration



Conflict duration


Civilian killings



✓in long-term, x in short-term

Battle-related violence



Peace settlements


Geographic containment



Conflict duration, local

Peace duration, local





Notes: ✓: beneficial effect; x: nonbeneficial effect; NS: nonsignificant effect.

Enforcement is also effective in reducing violence, but it does not contribute significantly to peacebuilding. Fortna (2003) uses the same categories of peace missions to assess their performances on the duration of peace. Unsurprisingly, traditional missions have the strongest deterrence capacity and reduce the chances of conflict recurrence. Deterrence capacity diminishes observer and multidimensional missions, respectively. Counterintuitively, enforcement missions are not associated with more durable peace. These findings contrast with the conclusion of Hultman et al. (2015) that more UN troops (usually deployed in enforcement missions) produce more stable postconflict environments. Peacekeeper observers and police, however, seem to have no relevant impact on postconflict peace. Other studies have also shown that most successful missions have robust mandates (Hultman, 2010; Kreps, 2010; Beardsley & Gleditsch, 2015). Hultman (2010) categorizes mandates based on robustness and protection of civilians tasks (POC), and her analysis indicates that only POC reduced violence against civilians. Other studies have only incidentally controlled for mandate robustness. For example, Beardsley and Gleditsch (2015) show that robustness reinforces the containment effect of missions. In fact, mandates are often ignored in the empirical analysis, relegated instead to serve as control variables. This lack of attention to mandates has resulted in a striking heterogeneity of mandate classifications. Furthermore, while these mandates are often coded consistently within a piece of research, mandate coding is often inconsistent among different works. Thus, mandates do affect conflict dynamics, but comparisons across studies may fail to categorize them consistently.

Other operationalizations of peacekeeping have produced more consistent and comparable results. First, peace missions (especially UN missions) produce more durable peace (Fortna, 2004a, 2003, 2004b; Gilligan & Sergenti, 2008), produce comparatively geographically contained conflicts, shorter episodes of local violence (Ruggeri, Dorussen, & Gizelis, 2016b), improve the odds of peacebuilding success (Doyle & Sambanis, 2000), and lower the odds of genocide over the long term (Kathman & Wood, 2011). Importantly, some findings at the national level do not automatically translate at the local level. As mentioned, Gilligan and Sergenti’s findings (2008) at the national level resonate with Diehl’s; they find peacekeepers to be better at keeping peace than making it. A more disaggregated approach, however, tells a different story at the local level. At this scale, Ruggeri and colleagues (2016b) find that peacekeepers shorten conflict episodes, but the empirical results on the capacity to deter local conflict onset is inconclusive. With regard to mass killings, the effect of interventions appears conditional on whether the mission directly challenges and engages the perpetrators. While this would suggest that only some missions reduce large-scale civilian killings, Melander (2009) finds that, after controlling for unobservable factors leading to deployment, peacekeeping does have a clear preventive effect on mass killings. Kathman and Wood (2011) also find that third-party intervention can trigger more intense genocides and politicides in the months following the deployment. Over time, however, violence significantly decreases if the intervention is perceived as impartial. More recent studies measuring the size of deployment and the type of deployed personnel have further supported most of these results: large deployments of peacekeeping troops have consistently beneficial impacts on all conflict outcomes examined in the literature (Hultman, Kathman, & Shannon, 2013, 2014, 2015; Beardsley & Gleditsch, 2015; Ruggeri, Dorussen, & Gizelis, 2016b). One notable exception is peacebuilding, which does not seem to be affected by the actual number of deployed personnel (Doyle & Sambanis, 2000).

Few studies include peacekeeping police and observer components, and they present mixed results. Hultman et al. (2014) find that larger deployment of UN peacekeeping troops (but not UN police and observers) reduces the intensity of violence on the battlefield in ongoing civil wars on a monthly basis. With regard to one-sided violence, more UN troops and police also effectively decrease violence against civilians, although observers produce the opposite effect (Hultman et al., 2013). Interestingly, UN troops can still effectively protect civilians in the aftermath of conflict (Kathman & Wood, 2014). Hence, on the one hand, peacekeeping police reduce the number of civilian killings but do not contribute to longer peace spells or reduce the number of battle-related deaths (Hultman, Kathman, & Shannon, 2014). Even more worrisome, police seem to increase violence rather than contain it (Beardsley & Gleditsch, 2015). Furthermore, observers have no statistical impact on peace duration, battle-related violence, and geographic containment. Ultimately, the only significant effect found in the existing literature is that deployment of observers is associated with greater intensity of one-sided violence (Hultman, Kathman, & Shannon, 2013; Beardsley & Gleditsch, 2015). Unarmed observers do not pose a threat to perpetrators of violence; thus, cannot exert a significant deterrent effect against mass killings.

Different dimensions of peacekeeping affect conflict dynamics, and each of these operationalizations captures something different but nonexhaustive about peacekeeping. In other words, if peacekeeping is a latent variable, then it makes sense to reframe Table 1 according to conflict outcome instead of by operationalization. For example, regardless of how peacekeeping is measured, substantial evidence indicates that it fosters durability of peace. Similarly, consistent results on the benefits of peace operations are reported for civilian and battle deaths and geographic containment of violence. Some studies also agree that some missions promote peacebuilding, but this also depends on mandate type. At the local level, peacekeeper capacity shows encouraging results for halting violence but exerts a weaker effect on the duration of local peace. Finally, peacekeeper capacity to increase the likelihood of peace settlements is conditional on the type of settlement, such as compromise, victory, mediated, or negotiated agreement (Beardsley, 2012; Greig et al., 2005).

Table 2 summarizes the main finding on the relationship between peace missions and non-conflict-related outcomes. These include cooperation with and attacks against mission personnel, incidence of sexual exploitation and abuses (SEA), economic and social development, democratization, and human rights. This shift from ending violence to improving the quality of the peace has been central in qualitative work on peacekeeping but has only recently emerged in the empirical literature on peacekeeping. Indeed, in most cases the findings summarized in Table 2 refer to only one study, and conclusions must be drawn carefully.

Dorussen and Gizelis (2013) examine which activities carried out by peacekeepers are more effective at getting support from both governments and rebel groups. They find that activities related to improvements in human rights or otherwise entail active involvement of the United Nations do not elicit cooperative behavior from either side. However, activities that strengthen state capacity are usually supported by both government and rebels, with the latter possibly being motivated by future rent seeking. In another analysis, Gizelis (2011) shows that cooperation with peacekeepers is more likely if the prewar status of women in the host country is high. She measures the status of women using educational enrollment rates, however, which does not allow her to establish the extent to which international actors include women in peacebuilding. Another condition that spurs cooperation with UN personnel is the asymmetry of capabilities between rebels and governments. In the presence of weak rebels, cooperation is more likely because peacekeepers play a central role in reducing commitment problems and, to some degree, reducing the imbalance with the government (Ruggeri, Gizelis, & Dorussen, 2012). Research on physical attacks against peacekeepers produces consistent mirroring results. First, rebels are more likely to target Blue Helmets if they lose ground on the battlefield and when they perceive that peacekeepers are taking the government’s side. By doing so, rebels aim at gaining or retaining bargaining power and preserving internal cohesion (Fjelde, Hultman, & Lindberg Bromley, 2016). In addition, stronger rebel groups are more likely to launch attacks against peacekeepers (Salverda, 2013), consistently with the finding suggesting that cooperation is associated with a group’s weakness.

Table 2. Peace Operation Effectiveness and Nonconflict Outcomes


Does peacekeeping have a beneficial effect?

Cooperation with peacekeeping

x if PK addresses human rights issues

✓ if peacekeeping addresses government capacity and has supportive role

Attacks against peacekeeping

x if peacekeeping is progovernment and rebels lose militarily

Sexual exploitations and abuses

✓if more women participate in peacekeeping

Economic development

Mixed (local vs. national)

Social development (quality of life)

✓on health; NS on literacy rate

Political development (democracy)

Human rights


Notes: ✓: beneficial effect; x: nonbeneficial effect; NS: nonsignificant effect.

Scholarly attention toward conflict-related sexual abuses has also increased, and several studies have focused on the effectiveness of peacekeeping at reducing sexual exploitation and abuses (SEA). Karim and Beardsley (2016) argue that peacekeeping missions that receive increased military contributions from countries with strong gender equality values reduce SEA. The presence of female peacekeepers also reduces SEA, although this evidence is less robust, perhaps due to the fact that the involvement of women can reduce SEA depends on the actual function they perform within missions.6 More worrisome is the survey finding of Beber et al. (2017) that deployment of UN troops in Monrovia, Liberia, is highly associated with episodes of transactional sex.

The study of the economic impact of peacekeeping has produced mixed results. Hoeffler et al. (2010) systematically explore the effect of peacekeeping on national growth during the first 3 years of peace and find that UN presence increases growth of approximately 2.4%. However, when peacekeeping is measured in terms of personnel size and expenditure, the effect disappears. In their study on economic recovery in South Sudan, Caruso et al. (2016) point out that the presence of Blue Helmets and the “security spillover” leads to an increased cereal production in counties where they are deployed. Another interesting finding on the disaggregated or local impacts of peacekeeping suggests that the arrival of UN staff also affects the local economy and may result in the emergence of distorted “peacekeeping economies” (Jennings & Nikolić-Ristanović, 2009). While the extent of these peacekeeping economies is empirically underdefined, Mvukiyehe and Samii (2010) found that peacekeeping stimulates the local economy in several ways, including increasing the employment opportunities leading to higher incomes and inflation. Carnahan et al.’s (2006) report on the economic impact of peacekeeping provides more information on how deployment may alter economic indicators locally and nationally, but the report does not provide definite conclusions regarding the causal effect of UN missions.

In addition to economic development, previous research offers some indications that peacekeeping missions may influence quality of life in the postconflict phase. Kim (2015) argues that multilateral interventions result in governments being more responsive to citizens’ needs, while unilateral military interventions are more self-interested and often focus on protecting the government. Hence, UN intervention improves health conditions and resources for reconstruction, although it does not influence literacy rates.

Regarding political development, peacekeeping seems to promote democratization. The two main studies on this issue are Joshi (2013) and Steinert and Grimm (2015), and these authors use different measures of democratization—changes in the Polity IV score and Freedom House Index, respectively. They both find that missions do promote democratic processes, which can be measured as improvements on the scale of their selected index. While these results are encouraging, both studies suffer the same limitation; they merely identify a democratization process independently of whether peacekeepers contributed to this. Peacekeeping may be incidental to the outcome, and this leaves open the possibility that democratic transition led by external parties creates more instability than autonomous processes. In other words, the presence of peacekeepers may or may not translate to UN-led democratization.

Finally, peacekeeping can also improve human rights in host countries but, as in other cases, mandates and activities in specific policy areas make the difference. Murdie and Davies (2010) point out that peacekeeping does not unconditionally help human rights. Mediation attempts and humanitarian missions, they argue, have a positive effect on physical integrity rights. While mediation improves both current and future human rights conditions, humanitarian missions have a longer-term effect and can be detrimental to physical security in the short run. Peksen (2012) explores this finding further and concludes that pro-government and neutral interventions may increase human rights violations by lowering incentives to reduce repression. Hostile interventions do not generate major improvements in types of violations; in fact, they result in more political imprisonments in the affected country. Although these findings depict a grim picture, it is important to note that Peksen (2012) examines both multilateral and unilateral military interventions. Whether UN peace missions produce similarly detrimental effects requires further investigation.

What Are We Missing in Evaluating Peacekeeping Effectiveness?

Consensus on the “gold standard” for assessing peacekeeping effectiveness is still missing, and the necessity of such an ultimate standard is unclear. If effectiveness is a function of a mission’s mandate, then there are as many standards as there are types of mandates. On the other hand, advocating for a gold standard demands that judgment on missions must be confined to a more parsimonious set of outcomes than those reviewed here. However, the criterion or set of criteria selected should not lead to a belief that other outcomes are irrelevant for understanding peacekeeping as a practice. For example, restricting evaluation of effectiveness to the peacekeepers’ capacity to maintain peace and reduce military fights should not imply that protection of civilians is less relevant. Peacekeeping operations are complex and affect the evolution of conflict within countries at various levels (e.g., local population, rebels, and political elites) and scales (i.e., local, national, and transnational). Although the literature has proposed a variety of concepts of effectiveness, aspects of the dynamic relationship between peace operations and conflict have received less attention and may have the same influence. Some are related to the conflict itself, while others concern the long-term transformation of fragile countries into peaceful and stable societies (see Sambanis, 2008, for an exception).

Among conflict-related aspects that have been sidelined within the literature, conflict trajectory is certainly a relevant and surprising absentee. Indeed, timing of third-party interventions is crucial in the conflict management literature because each conflict phase requires targeted approaches and allows for limited objectives. In the literature on conflict management, well-designed and successful mediation efforts take place when conflicts are ripe for intervention (Zartman & Berman, 1982), but theories of ripeness do not help predict when ripe moments will occur. It is probably more useful to think about conflict phases and which interventions are more likely to succeed in each scenario. The disaggregation of conflict into phases can be summarized as follows: initial disputes of parties over an issue; at least one party threatens to use military force; actual use of military force; violence ceases but the dispute is not settled and force is still used as threat; military resolution is no longer an option but no settlement has been reached yet; finally, termination (Bloomfield & Leiss, 1969). Naturally, external interventions will set different objectives and criteria for effectiveness according to the phase of conflict. During the Cold War, UN peacekeeping missions were predominantly deployed in the last phase of conflict, when some form of settlement had been achieved and hostilities had ceased. In the early 1990s, the Security Council authorized some missions with a conflict prevention mandate to avoid escalation to military confrontation (e.g., in the Balkans). Most contemporary UN peace missions operate amid violence. Thus, whether peacekeepers are better at keeping rather than making peace may depend on the conflict phase they operate in. While UN peacekeeping, generally, has adapted its capacity to work in different scenarios with some success, most conflicts do not follow a linear trajectory. Civil wars can go through escalation and de-escalation repeatedly, putting the flexibility and adaptability of peacekeeping missions to the test. Often, peace operations may be deployed in the settlement phase—when violence is absent—only to witness an unexpected re-eruption of the conflict. Ideally, missions that successfully deter violence and promote a shift from hostility to posthostility phase should eventually adapt to the new scenario to avoid this risk. The United Nations Mission in South Sudan (UNMISS) stands as proof. The Security Council authorized the dispatch of UN peacekeepers in South Sudan in summer 2011. The resolution highlighted the need to consolidate peace and outlined a mandate that prioritized postconflict peacebuilding and institution building (United Nations Security Council, 2011). Despite the awareness that violence was still present, the mission was designed under the assumption that South Sudan was in a conflict termination phase. The spiraling of violence in the following years and the outbreak of civil war in December 2013, made clear that UNMISS, as designed in 2011, was neither appropriate for conflict prevention nor for conflict resolution. As the head of the mission later declared, everyone was taken by surprise and the mission’s capacity was overwhelmed by the crisis (Nzambanita, 2014). This forced the Security Council to modify the original mandate and increase the commitment of military personnel to UNMISS to realign the mission’s capacity to the evolving situation. The mission continues to struggle to stabilize the country. In sum, peacekeeping missions vary in the timing of deployment; thus, the criteria to evaluate their success should be selected to match. In the very least, empirical studies on peacekeeping effectiveness should account and control for these differences. Most studies control for the intensity of violence, signature of ceasefires, and other settlements, but these variables fail to capture the combination of factors that define conflict phases. Consequently, it is difficult to determine which missions are more effective at different stages of the conflict or which are simply better equipped in responding to quick shifts across those phases. An explicit modeling of conflict dynamics in the empirical analysis of peacekeeping is still missing. Dixon (1996) acknowledged that international crises are fluid and that conflict phases can be accompanied by readjustment in strategies of interventions. Although his analysis includes several types of interventions and incidentally refers to military ones, his suggestion of using escalatory transitions to judge effectiveness is insightful. Furthermore, responsiveness to the evolution of conflict may be an additional important criterion to assess mission performances, especially in conflicts that are more susceptible to escalation. Even more crucially, it is possible—in principle—to identify an ideal entry point for peacekeepers along the conflict trajectory that reduces the likelihood of escalation and facilitates peacekeeping tasks.

In addition to conflict dynamics, mission dynamics are also overlooked in the existing literature. First, whether mission phases can be discussed in the way that conflicts are discussed is an open question. Systematic identification of stages in peacekeeping operations would provide interesting insights into which main phases the most successful missions go through. Second, the impact of mission duration on conflict has not yet been the subject of thorough investigation. The length of the peace operation is often included in a model’s specification to reduce temporal dependencies, but the state-of-the-art peacekeeping literature does not explore the ideal duration of successful missions. At the same time, experience shows the duration of UN peacekeeping missions is increasing. On average, missions that started in the 1990s lasted 26 months compared to 50 months for missions authorized since 2000.7 This trend is a result of broader mandates that are increasingly concerned with rebuilding states after conflict, which demands long-term commitments.

As highlighted in our third section, widely used indicators of effectiveness seem to be dominantly security focused. With conflict being the main concern for peace operations, improvements in terms of levels and intensity of violence will continue to be of primary importance. However, this focus on security misses an important aspect of the evolution of peace missions. Traditional peacekeeping operations were militarily demanding, but fewer peacekeeping mandates of the traditional type were seen. Most recent missions have multidimensional mandates (Figure 2) that “include elements of longer-term postconflict economic, social, and political development or ‘peacebuilding’” (Fortna & Howard, 2008). Figure 2 illustrates these changes by showing which tasks UN peacekeepers were mandated within each authorized mission from 1948 to 2015. The end of the Cold War resulted in a clear expansion of responsibilities for peace missions differently from the pre-1990s period, when missions were uniquely focused on military activities, ceasefire monitoring, and occasional assistance to governments.

Figure 2. Changing mandates of UN peace operations (Di Salvatore, 2017). Figure created by authors.

Researchers are just beginning to focus on the long-term impacts of peacekeeping. Instead, evaluations of peacekeeping are framed in relation to absence of violence (negative peace) rather than building peaceful societies (positive peace). Conditions for economic, political, and social development are intertwined with this concept of positive peace and the importance of reducing the risks for future conflict. Regarding economic development, a few studies examine the contribution of peace missions to national and local economies. For example, Hoeffler et al. (2010) identify a positive but temporary (3-year) positive effect of peacekeeping mission on economic growth, but this is unexpectedly independent of mission size. The only published subnational study that captures the economic effect of peacekeeping is from Caruso et al. (2016), who show that stronger mission presence increases cereal production in South Sudan. While the results of this study are limited to the South Sudan, it is important to recall that complementary research highlights the distorting effect of peacekeeping at the local level through the establishment of so-called peacekeeping economies (Jennings & Nikolić-Ristanović, 2009). While more rigorous empirical work on economic success of peacekeeping is ongoing, even less is known about its effect on other developmental outcomes such as infant mortality rates and human development.

Peacekeeping may reduce conflict and stop wars without creating conditions that reduce risk of recurrence. Political and economic exclusion of some groups, such as ethnic groups, have been shown to generate horizontal inequalities and thus increase the odds of violent mobilization (Cederman, Gleditsch, & Buhaug, 2013). Establishing peace through negotiated settlements without removing those enabling conditions for civil war cannot reasonably produce durable peace. UN peacekeepers are mandated to assist governments in building new institutions and reforming existing ones. Indeed, steering democratic transition is one criterion of successful peacebuilding, according to the widely cited definition of Doyle and Sambanis (2000), but democratic transition does not equate to more equal societies. It remains unknown whether the changes in the polity score are due to improvements on procedural democracy or on indices of inclusiveness. Relatedly, persistence of inequalities is an important obstacle for potential returnees after conflicts have ceased. Refugees and internally displaced people often refrain from returning to their homes even if violence stops. It is possible that the presence of peacekeepers can reduce uncertainty and favor normalization, but this is only likely to occur under certain circumstances. For example, whether the UN contributed to war termination, negotiated a political agreement with power-sharing arrangements, or effectively protected civilians are reassuring pull-factors for refugees and internally displaced people. Future research should engage with peacekeeping’s effects on these and other indicators. In this sense, the contribution of peacekeeping to the quality of peace in postconflict countries remains understudied in the quantitative literature. Notably, the peacekeeping literature is focused on negative peace (or the absence of violence) but this focus is, in turn, limited to conflict-related violence. Social violence (i.e., riots and violent protests), criminal violence, and terrorism are not accounted for in identifying peacekeeping’s potential violence-reduction impact. These forms of violence may be less political and only tangentially related to the main conflict; however, they shape perceptions of safety among the local population. On balance, the empirical work on peace operations effectiveness has contributed to a robust foundation of cumulative knowledge on how and under what conditions negative peace lasts, civilians are protected, and belligerents stop fighting. However, further empirically grounded work is needed to understand whether the effectiveness of peace operations can be extended to major challenges such as encouraging refugees and internally displaced people to return home, providing stable polities, strengthening economies, and aiding in processes of social development.


  • Autesserre, S. (2010). The trouble with the Congo: Local violence and the failure of international peacebuilding (Vol. 115). Cambridge, U.K.: Cambridge University Press.
  • Autesserre, S. (2014). Peaceland: Conflict resolution and the everyday politics of international intervention. Cambridge, U.K.: Cambridge University Press.
  • Angrist, J. D., & Pischke, J.-S. (2008). Mostly Harmless Econometrics: An Empiricist’s Companion. Princeton, NJ: Princeton University Press.
  • Beardsley, K. (2011). Peacekeeping and the contagion of armed conflict. Journal of Politics, 73(4), 1051–1064.
  • Beardsley, K. (2012). UN intervention and the duration of international crises. Journal of Peace Research, 49(2), 335–349.
  • Beardsley, K. (2013). The UN at the peacemaking-peacebuilding nexus. Conflict Management & Peace Science, 30(4), 369–386.
  • Beardsley, K., & Gleditsch, K. S. (2015). Peacekeeping as conflict containment. International Studies Review, 17(1), 67–89.
  • Beber, B., Gilligan, M. J., Guardado, J., & Karim, S. (2017). Peacekeeping, international norms, and transactional sex in Monrovia, Liberia. International Organization, 71(1), 1–30.
  • Bellamy, A. J., Williams, P. D., & Griffin, S. (2010). Understanding peacekeeping. Cambridge, U.K.; Malden, MA: Polity Press.
  • Bigombe, B., Collier, P., & Sambanis, N. (2000). Policies for building post-conflict peace. Journal of African Economies, 9(3), 323–348.
  • Bloomfield, L. P., & Leiss, A. C. (1969). Controlling small wars. New York: Knopf.
  • Bosco, L. (2017). UN Peacekeeping data: The known. International peacekeeping (Special Data Review Forum).
  • Bove, V., & Elia, L. (2011). Supplying peace: Participation in and troop contribution to peacekeeping missions. Journal of Peace Research, 48(6), 699–714.
  • Bove, V., & Ruggeri, A. (2015). Kinds of blue. Diversity in U.N. peacekeeping missions and civilian protection. British Journal of Political Science, 46(3), 681–700.
  • Bove, V., Ruggeri, A., & Zwetsloot, R. (2017). What do we know about UN peacekeeping leadership? International Peacekeeping (Special Data Review Forum).
  • Brown, M. A. (1993). United Nations peacekeeping: Historical overview and current issues. Washington, DC: Congressional Research Service.
  • Carnahan, M., Durch, W., & Gilmore, S. (2006). United Nations Economic Impact of Peacekeeping—Final Report.
  • Caruso, R., Khadka, P., Petrarca, I., & Ricciuti, R. (2016). The economic impact of peacekeeping. Evidence from South Sudan. Defence & Peace Economics, 1, 1–21.
  • Cederman, L.-E., Gleditsch, K. S., & Buhaug, H. (2013). Inequality, grievances and civil war. New York: Cambridge University Press.
  • Clayton, G., Kathman, J., Beardsley, K., Gizelis, T.-I., Olsson, L., Bove, V., et al. (2017). The known knowns and known unknowns of peacekeeping data. International Peacekeeping, 24(1), 1–62.
  • Costalli, S. (2014). Does peacekeeping work? A disaggregated analysis of deployment and violence reduction in the Bosnian war. British Journal of Political Science, 44(2), 357–380.
  • Derouen, K., Jr. (2003). The role of the UN in international crisis termination, 1945–1994. Defence & Peace Economics, 14(4), 251–260.
  • Diehl, P. F. (1988). Peacekeeping operations and the quest for peace. Political Science Quarterly, 103(3), 485–507.
  • Diehl, P. F., & Druckman, D. (2010). Evaluating peace operations. Boulder, CO: Lynne Reinner.
  • Diehl, P. F., Reifschneider, J., & Hensel, P. R. (1996). United Nations intervention and recurring conflict. International Organization, 50(4), 683–700.
  • Di Salvatore, J. (2017). Peacekeepers against ethnic and criminal violence. PhD diss., University of Amsterdam.
  • Dixon, W. J. (2016 [1996]). Third-party techniques for preventing conflict escalation and promoting peaceful. International Organization, 50(4), 653–681.
  • Dorussen, H. (2014). Peacekeeping works, or does it? Peace Economics, Peace Science & Public Policy, 20(4).
  • Dorussen, H., & Ruggeri, A. (2017). Peacekeeping event data: Determining the place and space of peacekeeping. International Peacekeeping (Special Data Review Forum).
  • Dorussen, H., & Gizelis, T.-I. (2013). Into the lion’s den: Local responses to UN peacekeeping. Journal of Peace Research, 50(6), 691–706.
  • Doyle, M., & Sambanis, N. (2000). International peacebuilding: A theroretical and quantitative analysis. American Political Science Review, 94(4), 779–801.
  • Eck, K., & Hultman, L. (2007). One-sided violence against civilians in war: Insights from new fatality data. Journal of Peace Research, 44(2), 233–246.
  • Fjelde, H., Hultman, L., & Bromley, S. L. (2016). Offsetting losses: Bargaining power and rebel attacks on peacekeepers. International Studies Quarterly, 60(4), 611–623.
  • Fortna, V. P. (2003). Inside and out: Peacekeeping and the duration of peace after civil and interstate wars. International Studies Review, 5(4), 97–114.
  • Fortna, V. P. (2004a). Does peacekeeping keep peace? International intervention and the duration of peace after civil war. International Studies Quarterly, 48, 269–292.
  • Fortna, V. P. (2004b). Interstate peacekeeping: Causal mechanisms and empirical effects. World Politics, 56(4), 481–519.
  • Fortna, V. P. (2008). Does peacekeeping work? Shaping belligerants’ choice after civil war. Princeton, NJ: Princeton University Press.
  • Fortna, V. P., & Howard, L. M. (2008). Pitfalls and prospects in the peacekeeping literature. Annual Review of Political Science, 11(1), 283–301.
  • Gaibulloev, K., George, J., Sandler, T., & Shimizu, H. (2015). Personnel contributions to UN and non-UN peacekeeping missions: A public goods approach. Journal of Peace Research, 52(6), 727.
  • Galtung, J. (1969). Violence, peace, and peace research. Journal of Peace Research, 6(3), 167–191.
  • Gilligan, M., & Sergenti, E. (2008). Do UN interventions cause peace? Using matching to improve causal inference. Quarterly Journal of Political Science, 3(2), 89–122.
  • Gilligan, M., & Stedman, S. (2003). Where do peacekeepers go? International Studies Review, 5(4), 37–54.
  • Gizelis, T. I. (2009). Gender empowerment and United Nations peacebuilduing. Journal of Peace Research, 46(4), 505–523.
  • Gizelis, T. I. (2011). A country of their own: Women and peacebuilding. Conflict Management & Peace Science, 28(5), 522–542.
  • Gleditsch, N. P., Wallensteen, P., Eriksson, M., Sollenberg, M., & Strand, H. (2002). Armed conflict 1946–2001: A new dataset. Journal of Peace Research, 39(5), 615–637.
  • Greig, J. M., & Diehl, P. F. (2005). The peacekeeping-peacemaking dilemma. International Studies Quarterly, 49(4), 621–646.
  • Hartzell, C., Hoddie, M., & Rothchild, D. (2001). Stabilizing the peace after civil war: An investigation of some key variables. International Organization, 55(1), 183–208.
  • Hegre, H., Hultman, L., & Nygard, H. M. (2011). Simulating the effect of peacekeeping operations 2010–2035. In J. Salerno, S. J. Yang, D. Nau, & S.-K. Chai. (Eds.), Social computing, behavioral-cultural modeling and prediction (pp. 325–332). Berlin: Springer.
  • Heldt, B. (2001). Conditions for successful intrastate peacekeeping missions. In A. Woodcock & D. Davis (Eds.), The Cornwallis Group VI: Analysis for assessment, evaluation and crisis management (pp. 109–136). Clementsport, NS: Canadian Peacekeeping Press.
  • Hoeffler, A., Ijaz, S. S., von Billerbeck, S., & The World Bank Group. (2010). Post-conflict recovery and peacebuilding. Washington, DC: World Bank.
  • Howard, L. M. (2008). UN peacekeeping in civil wars. Cambridge: Cambridge University Press.
  • Hultman, L. (2010). Keeping peace or spurring violence? Unintended effects of peace operations on violence against civilians. Civil Wars, 12(1–2), 29–46.
  • Hultman, L., Kathman, J., & Shannon, M. (2013). United Nations peacekeeping and civilian protection in civil war. American Journal of Political Science, 57(4), 875–891.
  • Hultman, L., Kathman, J., & Shannon, M. (2014). Beyond keeping peace: United Nations effectiveness in the midst of fighting. American Political Science Review, 108(9), 737–753.
  • Hultman, L., Kathman, J. D., & Shannon, M. (2015). United Nations peacekeeping dynamics and the duration of post-civil conflict peace. Conflict Management & Peace Science, 33(3), 231–249.
  • Hultman, L., Kathman, J. D., & Shannon, M. (2016). United Nations peacekeeping dynamics and the duration of post-civil conflict peace. Conflict Management and Peace Science.
  • Ijaz, S. S. (2014). The selection problem in peacekeeping missions: Developing an instrumental variable. Political Analysis, 20(1), 1–24.
  • Jennings, K. M., & Nikolić-Ristanović, V. (2009). UN peacekeeping economies and local sex industries: Connections and implications. IDEAS Working Paper Series from RePEc.
  • Johansen, R. C. (1994). U.N. peacekeeping: How should we measure success? Mershon International Studies, 38(2), 307–310.
  • Joshi, M. (2013). United Nations peacekeeping, democratic process, and the durability of peace after civil wars. International Studies Perspectives, 14(3), 362–382.
  • Karim, S., & Beardsley, K. (2016). Explaining sexual exploitation and abuse in peacekeeping missions: The role of female peacekeepers and gender equality in contributing countries. Journal of Peace Research, 53(1), 100–115.
  • Karim, S. (2016). Reevaluating peacekeeping effectiveness: Does gender neutrality inhibit progress? International Interactions, 1–26.
  • Kathman, J. D. (2013). United Nations peacekeeping personnel commitments, 1990–2011. Conflict Management & Peace Science, 30(5), 532–549.
  • Kathman, J. D., & Wood, R. M. (2011). Managing threat, cost, and incentive to kill: The short- and long-term effects of intervention in mass killings. Journal of Conflict Resolution, 55(5), 735–760.
  • Kathman, J. D., & Wood, R. M. (2014). Stopping the killing during the “peace”: Peacekeeping and the severity of postconflict civilian victimization. Foreign Policy Analysis, 12(2), 149–169.
  • Kelmendi, P., & Radin, A. (2016). UNsatisfied? Public support for postconflict international missions. Journal of Conflict Resolution, 1–29.
  • Kim, S. K. (2015). Third-party intervention in civil wars and the prospects for postwar development. Journal of Conflict Resolution, 2, 1–28.
  • King, G., & Zeng, L. (2007). When can history be our guide? The pitfalls of counterfactual inference. International Studies Quarterly, 51(1), 183–210.
  • Krain, M. (2005). International intervention and the severity of genocides and politicides. International Studies Quarterly, 49, 363–387.
  • Kreps, S. (2010). Why does peacekeeping succeed or fail? Peacekeeping in the Democratic Republic of Congo and Sierra Leone. Modern War & the Utility of Force: Challenges, Methods, & Strategy, 2(2), 90–118.
  • Melander, E. (2009). Selected to go where murderers lurk? The preventive effect of peacekeeping on mass killings of civilians. Conflict Management & Peace Science, 26(4), 389–406.
  • Murdie, A., & Davis, D. R. (2010). Problematic potential: The human rights consequences of peacekeeping interventions in civil wars. Human Rights Quarterly, 32(1), 49–72.
  • Mvukiyehe, E. N., & Samii, C. (2010). The subtle micro-effects of peacekeeping: Evidence from Liberia. APSA 2010 Annual Meeting paper.
  • Nordås, R., & Rustad, S. C. A. (2013). Sexual exploitation and abuse by peacekeepers: Understanding variation. International Interactions, 39(4), 511–534.
  • Nzambanita, P. (2014). South Sudan crisis brings questions for UNMISS | IPI Global Observatory. IPI Global Observatory.
  • Peksen, D. (2012). Does foreign military intervention help human rights? Political Research Quarterly, 65(3), 558–571.
  • Ruggeri, A., Dorussen, H., & Gizelis, T.-I. (2016a). On the frontline every day? Subnational deployment of United Nations peacekeepers. British Journal of Political Science, 6, 1–21.
  • Ruggeri, A., Dorussen, H., & Gizelis, T.-I. (2016b). Winning the peace locally: UN peacekeeping and local conflict. International Organization, 71(1), 163–185.
  • Ruggeri, A., Gizelis, T.-I., & Dorussen, H. (2012). Managing mistrust: An analysis of cooperation with UN peacekeeping in Africa. Journal of Conflict Resolution, 57(3), 387–409.
  • Salverda, N. (2013). Blue helmets as targets: A quantitative analysis of rebel violence against peacekeepers, 1989–2003. Journal of Peace Research, 50(6), 707–720.
  • Sambanis, N. (2008). Short- and long-term effects of United Nations peace operations. World Bank Economic Review, 22(1), 9–32.
  • Sambanis, N., & Doyle, M. W. (2007). No easy choices: Estimating the effects of United Nations peacekeeping (response to King and Zeng). International Studies Quarterly, 51(1), 217–226.
  • Sambanis, N., & Schulhofer-Wohl, J. (2008). Evaluating multilateral interventions in civil wars. In D. Bourantonis, K. Ifantis, & P. Tsakonas (Eds.), Multilateralism and security institutions in an era of globalization (pp. 252–287). London; New York; Routledge.
  • Sandler, T. (2017). International peacekeeping operations. Journal of Conflict Resolution.
  • Steinert, J. I., & Grimm, S. (2015). Too good to be true? United Nations peacebuilding and the democratization of war-torn states. Conflict Management & Peace Science, 32(5), 513–535.
  • Stinnett, D. M., Tir, J., Schafer, P., Diehl, P. F., & Gochman, C. (2002). The Correlates of War (COW) project direct contiguity data, version 3.0. Conflict Management & Peace Science, 19(2), 59–67.
  • Sundberg, R., & Melander, E. (2013). Introducing the UCDP georeferenced event dataset. Journal of Peace Research, 50(4), 523–532.
  • United Nations Security Council. (2011). S/RES/1996 (2011).
  • Vivalt, E. (2014). Peacekeepers help, governments hinder. Working paper, New York University.
  • Ward, H., & Dorussen, H. (2016). Standing alongside YOUR Friends: Network centrality and providing troops to UN peacekeeping operations. Journal of Peace Research, 53(3), 392–408.
  • Williams, P. D. (2017). Global and regional peacekeepers: Trends, opportunities, risks and a way ahead. Global Policy, 8(1), 124–129.
  • Zartman, W. I., & Berman, M. R. (1982). The practical negotiation. New Haven, CT: Yale University Press.


  • 1. The literature on “supply” of peacekeeping is not reviewed here. Works by Bove and Elia (2011), Gaibulloev et al. (2015) and Ward and Dorussen (2016) are great starting points for understanding who provides peacekeepers and why.

  • 2. For a recent work reviewing quantitative work on peacekeeping, see also Sandler (2017).

  • 3. See King and Zeng (2007) and Sambanis and Doyle (2007) on counterfactuals and study the effects of peace operation.

  • 4. See Clayton et al. (2017) in the special issue, G. Clayton (Ed.), The known knowns and known unknowns of peacekeeping data in international peacekeeping, for a survey of peacekeeping data the piece by Bosco (2017).

  • 5. See Angrist and Pishcke (2008) for technical details.

  • 6. See Karim (2016) on barriers to entry for female peacekeepers.

  • 7. This average includes only mission that started and ended in each decade, so missions that were deployed in the 1990s and continued in the 2000s are obviously excluded.