Arational Decision Stages: The Breakdown of Rationality in Strategic Planning and Implementation

EXECUTIVE SUMMARY The rational decision making stages often collapse during an imminent natural disaster, such as Hurricane Katrina. Each stage is shaped by independent moderating forces, such as the of the decision maker expertise, resource constraints and imperfect information, as well as arational mediators, such as favored decision premises, political agendas and personal opportunism. The power of independent moderators tends to match the influence of the arational mediators. In highly uncertain situations these arational "black box" dynamics, tend to suspend rational decision processes. Keywords: Crisis decision making, crisis management, arational decision making, decision stages, Hurricane Katrina NORMATIVE DECISION MAKING REVISITED Decision making is the science of choosing the best alternative. The problem with classical normative decision models lies in intractable constraints, particularly a lack of time, staff and resources for search, evaluation and selection of alternatives (Abelson & Levi, 1985; Bazerman, 2005). Normative theorists argue decision making during natural disasters should be more rational due to fewer constraints - immense government resources, foreseeable problems, and highly motivated decision-makers fearing for their jobs (McCarthy, 2003). Common steps in the decision making processes in crisis situations include: preparation, needs analysis, search for alternatives, evaluating alternatives, authorization and selection, communication & implementation, explanation and justification, and finally, retrospective evaluation (Dennis, 1995; Eisenhardt & Bourgeois, 1989; Turner, 1978). Descriptive theorists warn that high uncertainty can either constrain rational processes, or even undermine them to the point they become irrelevance (Bazerman, 2005; Heracleous, 1994, Quarantelli, 1988; Rosenthal, 1998). Combining moderating constraints with likely arational mediating factors yields Figure 1: 1. Preparation consists of contingency planning followed by simulations and stockpiling, both of resources and redundant systems (Quarantelli, 1988; Turner, 1978). Effective response to a widespread disaster requires an interdependent, interorganizational effort between the Department of Homeland security (DHS), the Federal Emergency Management Agency (FEMA), the National Guard, the Department of Defense, and a host of state and local agencies. Senate and Congressional committees investigating the Hurricane Katrina disaster both concluded that this catastrophe was both predicted and predictable (Gelbspan, 2005; Marek, 2006b). In New Orleans there were three areas of preparation, all requiring significant resources: (1) the levy system designed to prevent flooding, (2) FEMA equipment, supplies and resources stockpiled for relief efforts, and (3) disaster planning and simulation exercises to coordinate relief efforts following the disaster. Unfortunately political patronage undermined preparation efforts by shifting resource allocation elsewhere. The levee system was never designed to withstand anything more than a weak hurricane and Washington repeatedly refused to allocate the resources necessary for upgrades (Brinkley, 2006; Gelbspan, 2005). Similarly, FEMA's pitiful performance is not surprising, given its chronic under-funding and under-staffing as resources were shifted away towards homeland security issues. FEMA never received funding to be prepared for anything more than minor disasters (Marek, 2006b; Singer, 2006). 2. Needs analysis ensures the most important problems are being addressed. Usually expert panels identify critical issues, verify feedback and information, and maximize objectivity and quality (Bazerman, 2005; Rosenthal, 1998). Needs analysis should result in a decision process cycle which continues until either reaching the objective, or discounting the need for the objective (Ablesen & Levi, 1985; Harrison, 1999). Obviously in a disaster scenario, there is neither the time nor the information for an optimal needs analysis. …