The property insurance industry in most countries is competitive, often intensely competitive, and the actuarial models that insurance companies use to estimate losses – and therefore set premiums – are treated as proprietary “black boxes.” While understandable from a business perspective, the proprietary nature of these industry loss models, particularly for low frequency but high impact or catastrophic events, places regulatory agencies, public officials, the media, and the public in general at a serious disadvantage in trying to assess the accuracy or “fairness” of loss estimates and the translation of those into premiums, especially to residential homeowners.
As a type of balance or check on the proprietary models, Professor Shahid Hamid of the College of Business Administration and colleagues from multiple disciplines in the International Hurricane Research Center at Florida International University (FIU) have developed a public loss model to assess hurricane risk and project annual expected insured losses for specific properties or by coverage, construction type, zip code, county, and region in Florida. The model also estimates the probable maximum losses that should be re-insured.
The current model, whose development has been supported by nine public and private sector organizations and is certified by the Florida Commission on Hurricane Loss Projection Methodology, consists of three major components – wind, vulnerability/susceptibility to damage, and actuarial insured loss. The model may also be used to quantify the economic benefits of hurricane mitigation. Efforts are currently underway at FIU to enhance the model with a storm surge and flooding component.
More detailed information concerning this effort can be found at http://www.cis.fiu.edu/hurricaneloss/
