Department of Defense (DoD) force planners use integrated security constructs and multiservice force deployment scenarios to project the numbers and types of demands likely to be placed on U.S. forces in future years. The documents define possible “states of the world,” each of which includes steady-state activities and small- and larger-scale contingency scenarios that would require surges in U.S. military forces in a given region. Although these projections do make use of data, models, and simulations, they also rely heavily on two key assumptions—one used to estimate the frequency of future contingencies and the other used to estimate the...

The argument that military interventions exhibit temporal dependence suggests that these interventions are not independent events but are instead related in systematic and predictable ways over time. More specifically, temporal dependence predicts that the likelihood of an intervention in one year is a function of interventions in previous years. If this relationship is positive, the likelihood of a military deployment increases with the number of interventions in past periods, resulting in clusters of interventions over time. These clusters are not chance events but rather reflect a systematic relationship between interventions in one period and in the next.¹ The interventions in...

Tests for temporal dependence between military deployments look for relationships between deployments over time. If temporal dependence exists, the likelihood of an intervention in any one period should depend on the number of interventions in previous periods, even once underlying instability, conflict, and relevant political and economic factors are controlled. In this chapter, I test for temporal dependence using a set of 66 U.S. Army “interventions” over the period 1949 to 2010 and a carefully chosen set of control variables.

Although it can be defined in many ways, for the purpose of this report, I defineinterventionsto include company-size...

The results in the previous chapter suggest that, at times, U.S. military interventions exhibit temporal dependence that increases the chances they will occur in dependent clusters rather than following the serially independent distribution assumed in force planning processes. This result appears somewhat sensitive to the characteristics and political dynamics of the governing geopolitical regime and, as a result, may not always be a relevant concern for force planners and decisionmakers. However, where it does occur, clustering may have significant effects on force requirements that are worth considering in some additional detail.

To understand the implications of temporal dependence on the...

This report provides initial evidence that temporal dependence does affect the timing of military interventions and contributes to deployments, although this result is sensitive to underlying geopolitical and prevailing strategic considerations and appears more likely under certain international regimes than others. When temporally dependent clustering occurs, each military intervention heightens the risk of another intervention in the near future. Importantly, even a small amount of temporal dependence between interventions could have a sizeable effect on the number of deployments and the demands placed on military personnel if this heightened risk aggregates over several periods or leads to multiple overlapping interventions....