Competition is commonplace among militant groups. Although political scientists have begun recognizing its importance, they lag behind other fields in the general study of competition. This is critical due to the strategic depth that competition brings. How one group behaves affects another group, and vice versa. Moreover, target governments and international organizations can manipulate the environment in which the groups must then interact. This chapter argues that building models to examine these issues is a useful strategy, but that the literature on political violence has not yet explored the implications. We then set the stage for the results we develop throughout the book.

Target governments can reduce grievances among disaffected populations who might otherwise pledge support to a group. Incorporating this into the workhorse model, we show an unexpected relationship between the total number of groups and total violence observed. When few groups exist, the target state has little incentive to reduce grievances. Due to the lack of competition, the government calculates that paying that price in violence is worth offering fewer concessions. In contrast, when many groups exist, the competition instills great fear in the target state. As a result, it may calculate that entirely abandoning the objectionable policy is the best solution. Without any supporters to recruit, the groups then drop their violence outputs. Thus, violence may decrease in the number of competing groups because violence deters the government. We characterize the circumstances under which the deterrent effect dominates the competition effect and provide broader tips for the empirical literature on outbidding.

One way a target government can try to mitigate outbidding violence is to increase enforcement efforts to intercept contributions and arrest volunteers to militant groups. We expand the workhorse outbidding model to account for this decision. States with greater enforcement capacity indeed benefit, partially from directly stopping contributions and partially from deterring supporters from making contributions in the first place. The decreased prize therefore also tempers outbidding violence. As a result, competition is contingent on enforcement capacity, with the effect of another group growing larger as that capacity declines. Statistical analysis finds broad empirical support for our mechanism: competitive violence is most pronounced when governments incur higher marginal costs of enforcement. These results increase our confidence that competition drives violence more broadly, as competing explanations do not predict this conditional effect.

This chapter reflects on the generalizable lessons that our theoretical and empirical results generate. Two central ideas emerge. First, strategic interaction is a central component of political violence. Failure to account for it risks generating invalid theoretical mechanisms and ineffective policy recommendations. Second, there is no silver bullet for terrorism. Some policies may be more effective on average than others. But even some seemingly sensible solutions can backfire under the wrong circumstances. As such, policymakers wishing to influence political violence outcomes must have a strong understanding of the causal process that guides the violence before making interventions. We also unite various subthemes that have reoccurred throughout the book, such as the role international institutions play in affecting terrorism patterns.

This chapter develops the workhorse model we explore throughout the book. We begin by substantively motivating many aspects of competitive violence: the marketplace has limited resources, violence is costly but increases a group’s share of those resources, opposing violence decreases one’s own share, and others. These components lead us to conclude that a "contest" model is ideal to study the implications of competition. Doing so allows us to recover a central implication from existing theories of outbidding: that more groups imply more total violence output. However, our model concludes that outbidding is a collective effect rather than an individual one. Even as total violence increases in group numbers, the per-group rate of violence drops. These central results are robust to a variety of alternative assumptions.

We directly assess the empirical evidence of the propositions derived in the previous chapter. First, we conduct a large-n analysis of terrorist violence in every country between 1970 and 2015. We then examine whether there is a relationship between the number of active militant groups in a state and the aggregate amount of violence. We find evidence to support the basic outbidding hypothesis: more militant groups are significantly associated with more violence at the state level. We subsequently analyze the effect of increasing numbers of groups on per group violence. In accordance with our model expectations, we find that while increasing competition appears to lead to more violence overall, per group violence declines on average. Finally, to more fully explore the causal mechanisms at work in this process, we examine in detail the multi-dimensional insurgency in Northeast India since 2009. We find that as aggregate violence in the country and the region increased, groups curtailed their own use of violence due to concerns over diminishing returns and increased costs.

This chapter endogenizes a would-be militant group’s decision to enter the marketplace for violence. We show that an existing group may overproduce violence to corner the market and make its potential rivals calculate that recuperating their costs will be impossible. As a result, violence may be greater when we observe one group than when we observe many. We then investigate four ways in which a target government might mitigate the violence: offensive measures that undermine the lead group’s marginal cost of violence, defensive measures that absorb a portion of all violence, deterrent measures that increase the cost of group formation, and concessions to the group’s audience to reduce grievances. Of these, only specific types of defensive measures are guaranteed to decrease violence. In contrast, increasing the burden of entry and decreasing grievances can counterintuitively increase violence.

Chapter seven takes a closer look at the rebel group themselves, their make-up and propensity to project outward force and includes a focus on the individual participant in violent groups. Here we find that personal determinants and group behaviour shed a different light on escalation, serving an inner logic or social bonds rather than strategic concerns of the group. The cases used to illustrate the mechanisms are the Khalistan Movement in India, the German Rote Armee Fraktion and the Red Brigades in Italy. In the Indian case external pressure had clear repercussions for the internal dynamics of the Sikh movement. The German Rote Armee Fraktion showcases a personal extremity shift after the repression by the German state. The Italian case is notable for the role of individual ways out of violence to explain a de-escalation of the struggle. The chapter points to escalatory mechanisms focusing on extremity shifts within armed groups.