Quantifying Insurance Terrorism Risk

The World Trade Center disaster was a stark reminder to the insurance industry of the potential dire consequences of accumulating high concentrations of insured value and underestimating a hazard to which they are exposed. By imposing strict coverage limits, or stopping to offer terrorism cover for large commercial policies, initial steps can be taken to address the accumulation problem. Subsequently, exploration of the impact of some hypothetical future terrorist scenarios can guide the control of risk accumulations. But estimating Probable Maximum Loss, as well as the pricing of terrorism risk, require the hazard issue also to be addressed. This will never be resolved as well as hurricane or earthquake hazard, but some insight into its ranking as a peril is urgently needed. With the insurance industry struggling with the uncertainty of how to deal with terrorism risks, hopes will be placed on a reduction of the terrorist threat, now that there is a concerted global will to combat terrorism. Terrorism hazard is considered here in the wake of this international governmental resolution, and the destruction of the al-Qaeda training camps in Afghanistan. The frequency and severity of attacks depend crucially on organizational structure. To minimize detection by counter-terrorist forces, advantage may be taken by terrorists of alternative forms of network architecture, adopted by drugs syndicates, pirates, and other criminals, within which sporadic pulsing swarm attacks might be effectively launched. The constraints of this network architecture, with sustained pressure from counterterrorist forces, will influence the relative likelihood of different scenarios favored by alQaeda. A terrorism cost function, involving planning time, technical difficulty, and consumption of resources, may be defined to quantify relative scenario likelihood, and thereby allow a loss severity curve to be derived. As part of the task of normalizing this curve to strike frequency, a logical event-tree is outlined for computing the probability that a planned attack is successful. Any probabilistic framework for quantifying terrorism risk, however logically designed, will ultimately have to involve a measure of expert judgement. Extensive expert consultation exercises have already been commissioned by the Pentagon, and should prove as insightful to the insurance industry as for government agencies.