As the property-casualty industry anxiously awaits the opening of the hurricane season in three days, many are keeping their fingers crossed–with some scientific backing–that it will not be a repeat of last year.
Last week, the National Oceanic & Atmospheric Administration officially predicted an active Atlantic hurricane season with 13 to 16 named storms. But less than half–four to six–will turn into major hurricanes.
NOAA noted that last year's hurricane season, a record breaker in terms of insurance losses at $56.8 billion, produced 28 named storms, 15 of which were hurricanes with seven classified as major.
Further indications that this season will not pack the record-breaking wallop of 2005 came from other quarters. Florida State University Professor Jim Elsner, for one, told investors on a Bear Stearns conference call that last year included additional abnormalities, such as the storms tracking over a warm current in the Gulf of Mexico, which will in all likelihood fail to affect this year's activity.
Mr. Elsner forecast activity levels 50 percent above normal for the upcoming season and said he expects continued elevation for the next five-to-eight years.
Higher Atlantic sea surface temperatures, although not as high as last year, and a negative North Atlantic Oscillation combine to create the potential of storms increasing in frequency and severity, he said.
Commenting on the potential impact of Professor Elsner's predictions, Bear Stearns analyst David Small said if 2006 shows continued high frequency but lower severity, then reinsurers should perform better than primary insurers, due to generally higher property-catastrophe attachment points.
In addition, five more years of intense storms could lead both to higher property prices and losses and more share price volatility, he said.
Looking back at the 2005 storm season, Mr. Small said it should not have been the surprise it was to the insurance industry, drawing his conclusion based on information gleaned from a series of conference calls, including Mr. Elsner's.
The information, he said, underscored that scientists have been discussing frequency and severity increases for quite some time now. And also, modelers have run scenarios that gave fairly accurate insight to potential losses from large storms.
During the most recent Bear Stearns call, Frank Fischer, primary insurance marketplace leader for the Boston-based catastrophe modeler AIR Worldwide, attempted to debunk the widespread notion that the models missed the mark in 2005. "AIR's review of the data and benchmarking of their estimates against underwriters reported losses indicates that their models were reasonably accurate," Mr. Fischer said.
Many of the issues that arose were due to incomplete or inaccurate exposure data such as missing locations, incorrect replacement value and incorrect property coding used by underwriters in the modeling process.
The hurricane season runs June 1 through Nov. 30.
© Arc, All Rights Reserved. Request academic re-use from www.copyright.com. All other uses, submit a request to TMSalesOperations@arc-network.com. For more information visit Asset & Logo Licensing.