Insurers Need To Rethink Catastrophe Reserves

It suggested that the reinsurance/insurance industry would react to this potential catastrophic threat by improving its modeling, changing the current per-occurrence model of catastrophe coverage, and reviewing levels of risk retention.

Mr. Smith said the reserving question was highlighted by the number of large companies whose recent fourth-quarter financial reports included millions of dollars in additional loss numbers for hurricane events that exceeded their first estimates. "We're going to see more as the [financial] result season goes on," he said.

A risk retention review, the paper said, could lead to companies reducing the amount of risk they take on. The firm also expects the use of per-event deductibles by companies will declinereplaced by seasonal deductibles. In Florida, it was noted, insurers estimate this change will increase rates 3 percent.

The report said that the 2004 hurricane season was unusual in one crucial wayseriality.

Mr. Smith, found that, as of today, none of the currently available catastrophe models take into account the cumulative effect of serial storms, such as increased cost of raw materials, lack of availability of claim adjusters, and increased susceptibility due to damage from prior storms.

"Many insurers will have to critically re-evaluate their appetite for risk retention, improve their models, and review how deductibles should be handled moving forward, particularly if the market moves to a per-season deductible structure," Mr. Smith said.

"Based on published scientific studies, insurers and reinsurers can expect heightened levels of hurricanes to continue for another 10-to-40 years," he added. "I strongly believe that the modeling of hurricanes, especially those in clusters, will have to improve."

According to the ReAdvisory findings, until last year's hurricane season, the likelihood of a $20 billion loss was mainly measured as the likelihood of a single large event causing the damage. The 2004 hurricane season, the company said, has shown that the risk of losses accumulating from multiple storms must also be considered.

ReAdvisory said its findings are based upon known and measurable facts, but there are two "wild cards" that need further consideration and could bring the future into focus.

One area of concern that has not been factored in the predicted return cycle of the hurricane "cluster" year is global warming, ReAdvisory said. "Weather patterns continue to change and we cannot yet determine how global warming will sway current hurricane forecasts," Mr. Smith said.

ReAdvisory noted the importance of being mindful of where the hurricanes land. Are there any physical or meteorological phenomena that affect hurricane paths?

The firm noted that if Hurricane Andrew hit Miami only 10 miles north, potential damage was estimated at three-to-four times what actually occurred. "If any of the four major storms of 2004 had hit a major city, the damage could have been astronomical. More work is needed in this area," Mr. Smith said.

Mr. Smith said that among other topics the paper raises is the issue of how the industry is going to look at catastrophe models. Some of the catastrophe modeling companies, he said, will make fundamental changes. "It's going to change the industry," he predicted.

The "2004 Hurricane Season In Review" white paper is available online at http://www.carvill.com.

Reproduced from National Underwriter Edition, February 25, 2005. Copyright 2005 by The National Underwriter Company in the serial publication. All rights reserved.Copyright in this article as an independent work may be held by the author.