Strategic assessment in war: A bounded rationality model of how organizations evaluate policy effectiveness.

Abstract

Despite volumes of military history, surprisingly few scholars attempt to model decision makers' actual wartime strategic choices. I develop and test a bounded rationality model which predicts how decision makers evaluate the success of their implemented policies. My model captures the forces influencing policy makers' preferences by analyzing the rate of change of key indicators. Well suited for predicting organization's strategic assessments in war, the model applies generally to other, non-war situations. The model precisely predicts British naval decision making in both World Wars, US ground strategy assessments in Vietnam and the timing of the Carter Administration's hostage rescue attempt. I argue that decision makers form preferences in response to sudden changes in the behavior of their policies. Among the sources decision makers use to measure policy performance are quantitative indicators, such as ships sunk per month, or the amount of cocaine captured per raid. Decision makers choose the indicators which they think will best measure a policy's ability to pursue their organizational interests. I call the indicators which decision makers monitor to asses policy performance "dominant indicators" which together form an organization's "dominant indicator set.". By analyzing the rate of change of the dominant indicators (represented by their first and second derivatives) the model captures the primary forces structuring the decision maker's preferences. I hypothesize that record movement of the dominant indicators in a positive direction results in decision makers' satisfaction with their policy. Conversely, record movement in a negative direction leads to dissatisfaction. I also predict that if the indicators within the decision maker's dominant indicator set move in record opposite directions, the organizations will become conflicted and be unable to reach a consensus assessment. This bounded rationality model provides a powerful, general way to predict organizational leader's policy assessments and includes a number advantages over common organizational theory and security studies approaches. This approach also suggests new ways to think about other war and organizational issues, such as war-termination and the principal-agent problem.