Are U.S. Military Interventions Contagious over Time?

Current DoD force planning processes assume that U.S. military interventions are serially independent over time. This report challenges this assumption, arguing that interventions occur in temporally dependent clusters in which the likelihood of an intervention depends on interventions in the recent past. The author used data on 66 U.S. Army contingency and peacekeeping deployments of at least company size between 1949 and 2010 and found evidence of temporal dependence between military interventions even when controlling for political, economic, and other security factors. However, the results also suggested that clustering is affected by the nature of the geopolitical regime and is stronger at certain points than others, for example, after the Cold War as compared to during the Cold War. The results suggested that as few as two military interventions above average is often enough to trigger interventions in subsequent years. Because current planning processes address only the direct force demands of a given deployment and ignore the heightened risk for additional demands created by temporal dependence, these processes may project force requirements that understate the demands placed on military deployments during a period of clustered interventions. This analysis suggests that DoD should consider modifying the integrated security constructs to incorporate serial correlation of interventions, making assumptions about the nature of the current or future geopolitical regime explicit, and assessing whether the existing set of force planning frameworks reflects the spectrum of potential future geopolitical regimes.