The third quarter may have been easy, with Bermuda companies reporting an average annualized return of 20 percent, but Bermuda executives still needed to answer a lot of tough questions during recent analyst conference calls.
However, several answered the most obvious one voluntarily:
What happened to the hurricanes?
William Riker, president of RenaissanceRe Holdings Ltd., explained that the seasonal forecast models that fueled predictions of an active storm season "are statistical in nature. They are the result of trolling through data to look for correlations between climatic factors and hurricane activity," he said, identifying sea-surface temperature as one factor widely considered.
"The key thing to recognize is that…no one has figured out the true physical interactions between these statistical indicators and hurricane formation. The statistical correlations appear to exist, but the physics of the situation are still uncertain," he said.
"The science of seasonal forecasting is still in its infancy, and its effectiveness in managing risk on an annual basis is still suspect," he concluded.
At Endurance Specialty, Chief Underwriting Officer David Cash said that in addition to sea-surface temperature, scientific forecasters sometimes indicate that the presence or absence of a mid-Atlantic high-pressure zone can be another indicator of land-falling hurricanes. But this year, he said, the factor he's heard discussed most by scientists involves theories about dust clouds and the formation of hurricanes on the African coast.
"I'm not crisply enough informed about it that I can describe it to you, [but while] everyone in the industry is breathing a sigh of relief, no one is taking that as a sign that in the future we may benefit," said Mr. Cash–adding he understood that within the forecasting world, a year such as 2005 registers "like a triple-witching hour."
A combination of the right circumstances in Africa, the right sea-surface temperature, the right positioning of the mid-Atlantic oscillator–and maybe a fourth factor, la Nina conditions–"can be tremendously powerful for the industry in terms of profits," he said, while "getting them all wrong," as in 2005, "can be exactly the converse."
Mr. Riker presented figures comparing storm activity this year to historical averages, noting that while 2006 comes in below averages for the most recent 10 years, based on a longer-term view, this year would have been considered a normal season.
o The number of named storms in 2006 was nine, compared to the long-term historical average of 10.
o The number of 2005 hurricanes was five, compared to a long-term average of six. For the most recent 10 years, the average is higher–at about eight per year.
o The number of intense hurricanes was two, compared to a long-term average of two, but a 10-year average of four.
"As a reinsurer, we are most concerned about big storms hitting populated areas," Mr. Riker said. "Even within a heightened period, hurricane damage is still an uncertain distribution." While 2006 will likely be a year "when our concerns are not realized," he added, "we continue to believe we are in a heightened period of hurricane activity and we're managing our risk accordingly."
© 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.