I nterest in peacekeeping has blossomed since the end of the cold war. However, academics have only recently begun to study third-party interventions in conflict. 1 We review the flourishing new literature on third-party intervention and point to areas of research in which economic theory may be useful to enhance scholars' and laymen's understanding. Our review highlights three aspects of the literature on thirdparty intervention. First, what are the goals of third parties who intervene and do

more »

... hey achieve those goals? Second, we review academic work concerning United Nations interventions. And third, paying attention to the recent extension of theory that models conflict as destructive, we suggest that this theory might be usefully grafted onto the theory of third-party intervention. Goals and success of interventions Regan discusses the issue of third-party objectives in civil conflict. He states that interventions can reduce or prolong conflict, depending on whether intervention is in behalf of the government or the opposition. He hypothesizes that early intervention in support of the government will shorten the conflict, while early intervention on behalf of the opposition will lengthen duration. He tests these hypotheses using a hazard model applied to monthly data on 150 conflicts drawn from 1944-1999. Contrary to his hypothesis, the empirical results suggest that early intervention in favor of the government does not statistically affect the duration of conflict. However, the estimates show that intervention on behalf of the opposition does tend to lengthen the duration of conflict, although the substantive impact is small. When Regan takes account of the type of intervention-military or economic-he finds that third-party involvement greatly increases the expected duration of conflict. Even neutral interventions-those made on both sides of a conflict-are found to lengthen conflict duration. He writes: "Clearly, the notion of what constitutes an effective intervention strategy is an important element in understanding how civil conflicts are managed. When focusing exclusively on military and economic forms of outside interventions, conflicts appear to be not only ineffectively managed, but the interventions themselves also appear to prolong the conflict ... The results further suggest that policy makers need to think more critically about the role of military or economic interventions if their objective is to manage the violent aspects of a conflict." 2 Collier and Sambanis cite Regan as the gold standard in third-party intervention research. But they view Regan's work as lacking crucial normative content, because third parties likely have strong views on "the perceived justice of the rebellion." This critique might extend to Siquira, who assumes that third parties only value peace. 3 Collier and Sambanis also note that even if the third party seeks swift conflict resolution, it may not be clear which combatant is strongest; that researchers lack data on the cost of both conflict and intervention; and that Regan omits important variables (although Collier and Sambanis do not specify which variables). If, as Regan finds, third parties do actually prolong conflict, then either they wish to prolong conflict or their actions work against their intentions. However, other factors which are not currently accounted for, such as the intensity of an ethnic conflict, may lead to both third-party involvement and longer expected conflict duration (econometrically speaking, this is an endogeneity problem). Elbadawi and Sambanis, for example, effectively deal with endogeneity issues. They employ economic theory to ascertain how intervention affects conflict, basing their empirical work on the theoretical economic approach of Intriligator and Brito. Elbadawi and Sambanis predict that external assistance to a rebel party will increase duration of conflict, ceteris paribus, later theoretically confirmed by Chang, Potter, and Sanders, and by Chang and Sanders. This is because external intervention will serve to raise rebel capabilities, allowing them to continue their resistance. 4 But Elbadawi and Sambanis note that this causal relation between intervention and duration is difficult to show empirically. Does intervention on behalf of rebels cause a longer lasting conflict, or do longer lasting conflicts lead to external intervention? To resolve this issue Elbadawi and Sambanis employ the econometric technique of instrumental variables. First, they estimate a random effects probit model to predict interventions. They find that several factors lead to external intervention. Civil wars with high casualties invite intervention; ethnic conflicts invite less intervention; and intervention is more likely when war was present in the previous period. Next, they use these results in an ordered probit model to estimate the duration of conflict. As predicted by theory, the coefficient of expected intervention is positive; hence, they conclude that external intervention on behalf of rebels will increase the duration of conflict. Chang, Potter, and Sanders, and Chang and Sanders do not assume that interveners Our review highlights three aspects of the literature on third-party intervention. First, what are the goals of third parties who intervene and do they achieve those goals? Second, we review academic work concerning United Nations interventions. And third, paying attention to the recent extension of theory that models conflict as destructive, we suggest that this theory might be usefully grafted onto the theory of third-party intervention.