Property and Casualty (C) Committee
Public Hearing
The Regulation of Catastrophe Modelers
September 28, 2007

The Property and Casualty Insurance Committee held a public hearing during the NAIC 2007 Fall National Meeting. The purpose of the hearing was to gather information on the appropriate regulatory framework for monitoring the activities of catastrophe risk modelers and to explore whether laws and regulations governing rating or advisory organizations are broad enough to be applied to catastrophe modeling vendors. In addition, the committee explored the use of near-term catastrophe models for pricing property insurance.

Following opening remarks by Florida Insurance Commissioner Kevin McCarty, the Committee heard from J. Robert Hunter of the Consumer Federation of America. Mr. Hunter provided insight into consumer’s perspectives. He said that consumers feel that insurers have not honored promises made when the use of catastrophe models was introduced. He was critical of the use of what is known as “near-term” models―models that use a five-year projection that emphasizes the use of the adverse experience from 2004 and 2005. These models have resulted in rapidly escalating rates for policyholders and record profits for insurers, he maintained. He urged greater regulatory oversight of the modelers.

Mr. Hunter was followed by Howard Eagelfeld representing the Florida Commission on Hurricane Loss Projection Methodology. Mr. Eagelfeld explained how the commission analyzes catastrophe models in Florida.

A panel of modelers responded to questions regarding their regulation and the use of near-term models. Mitch Sattler (Risk Management Solutions), Thomas Larsen (EQECAT, Inc.) and David Lalonde (AIR Worldwide) said that there is already significant regulation of catastrophe modelers. They did not think that the regulatory framework for rating organizations fit with oversight of catastrophe modelers. Some suggested that another layer of regulation would be unwelcome, however, if a central framework could be developed that replaced the existing oversight; it might be preferable if consistently applied. Each maintained that an unbiased model with consistent results was the goal. Catastrophe models use historical information and apply scientific and engineering principles to the information to predict future losses. The modelers suggested that near-term models provided another option for insurers to consider when assessing future catastrophe losses.

A panel of representatives of insurer trade associations defended insurers’ use of near-term catastrophe models and agreed with modelers that the framework for regulating advisory organizations was inappropriate for modelers. David Kodoma (Property Casualty Insurers of America), Robert Detlefsen (National Association of Mutual Insurance Companies) and Eric Goldberg (American Insurance Association) encouraged regulators to continue to allow insurers to select whether a near-term or other model best fits their needs. They observed that financial rating agencies have encouraged insurers to move toward near-term models for financial ratings.

Birny Birnbaum (Center for Economic Justice) presented his perspectives on what was said by the modelers and the insurer trade organization representatives. He encouraged regulators to investigate whether there was collusion among insurers or modelers, to regulate the modelers as advisory organizations, to develop a public model for benchmarking purposes, to provide guidance to modelers for key modeling components that are public policy, rather than scientific, decision points and to get insurers and modelers involved in loss mitigation efforts.