A.M. Best Methodology: Property/Casualty Supplemental Rating Questionnaire.

OLDWICK, N.J. -- In an effort to provide more time to all parties associated with the completion of A.M. Best Co.'s Property/Casualty Supplemental Rating Questionnaire (SRQ), A.M. Best has placed the updated natural catastrophe-related questions on its website. The changes to the existing questions, as well as new questions related to the use of non-traditional catastrophe mitigation techniques, warranted the need to notify interested parties earlier than the January 2007 release date.

The natural catastrophe modeling questions are designed to gain further insight into a company's catastrophe risk management. With these questions, A.M. Best continues to put a high degree of emphasis on an organization's overall catastrophe management capabilities and its impact on financial strength. Going forward, A.M. Best will more closely scrutinize the entire probable maximum loss curve to account for the range of possibilities a company may face. Specifically, A.M. Best will be requesting Tail Value at Risk (TVAR) or Tail Conditional Expectation (TCE) for the various return periods. In the near-term, A.M. Best will continue to utilize the existing single occurrence probable maximum loss methodology with assessment of the loss curves used for sensitivity analysis. Through the SRQ, A.M. Best will conduct additional industry analysis to determine if a revised approach to assessing catastrophe risk is appropriate.

Other key revisions to the natural catastrophe portion of the Property/Casualty SRQ include providing additional details regarding the various data quality metrics. As noted with events of 2005, data quality or lack thereof can have a significant impact on the credibility of modeled output. In cases where A.M. Best believes data quality and/or all material sources of catastrophe risk are not incorporated into the model, modeled output will be adjusted. Accordingly, A.M. Best encourages companies to be forthcoming regarding the catastrophe model report, detailing such factors as construction type, location and key model assumptions.

While they are only tools and not the sole risk management mechanisms available to companies, catastrophe models, in A.M. Best's view, are valuable tools in an organization's overall risk management approach. As a point of clarification regarding some recent industry commentary, A.M. Best does not require multiple models or a specific modeler. Instead, A.M. Best's assessment of catastrophe risk is based on a comprehensive review of both modeled and actual results. For those companies that do utilize multiple catastrophe models, A.M. Best's baseline approach is to take a straight average. In cases where a company and the natural peril modeler can effectively explain key assumptions and differences, a weighted average may be used in the rating analysis.

A.M. Best only expects companies to provide responses to those questions if they are actively using a catastrophe model in the risk management of their catastrophe exposures. Risks that are considered immaterial or models that are only run occasionally for validation of actively run models do not require SRQ responses.

The new catastrophe-related questions can be accessed at www.ambest.com/ratings/methodology/srq-catastrophe.pdf.

A.M. Best Co., established in 1899, is the world's oldest and most authoritative insurance rating and information source. For more information, visit A.M. Best's Web site at www.ambest.com.