The Boston Globe brings us this, erm, “joyful” news:
Risk Management Solutions now says its earlier gloom and doom about the cost of potential hurricane damage in coastal areas was far too rosy. The California company is predicting that hurricanes will occur with much greater frequency and intensity over the next five years, and is telling insurers they need to increase their annual loss estimates by 25 percent to 30 percent in New England and the mid-Atlantic states, and 40 percent across the Gulf Coast, Florida, and Southeast.[...]
Most hurricane models use historical data to forecast hurricane frequency and intensity, but Risk Management says that approach is no longer sufficient.
”There’s enough data and science to support the view that hurricane activity will be different from what it has been on a historical basis,” said Paul VanderMarck, executive vice president at Risk Management.
Kerry A. Emanuel, a professor of meteorology at the Massachusetts Institute of Technology who served on a four-member panel of specialists consulted by Risk Management, said there are two schools of thought about Atlantic hurricane activity.
One school holds that hurricane activity rises and falls in multidecade cycles, with the Atlantic currently in an active cycle that is expected to last another 10 to 20 years. The other school of thought holds that climate warming is changing hurricane dynamics, and the ups and downs of the past are giving way to a more stable trend line that will increase long-term risks for insurers.
Emanuel says he supports the more pessimistic view. ”If it is correct,” he said in an e-mail, ”there is then no reason to expect a decadal time-scale downturn in Atlantic hurricane activity.”
AIR Worldwide of Boston, the other major hurricane modeling company, takes a more traditional approach, relying on long-term historical data to predict hurricane losses, and sees no need to alter its approach.
The Fair Plan’s rate request for this year illustrates just how different the Risk Management and AIR models are. In seeking a 25 percent rate increase on the Cape, where its average premium is $1,306, the Fair Plan averaged the outputs of the two models.
”I certainly don’t know which one is right, but it doesn’t make sense to take just one and bet everything that it’s accurate,” Golembeski said.
But Attorney General Thomas F. Reilly did just that. In his Fair Plan rate filing, Reilly discarded the Risk Management model, which he said would yield a Cape rate increase of nearly 60 percent, and adjusted the AIR model on his own to yield a 1.2 percent increase.
And so forth. Those of us who work with property insurance have been having an inkling that more changes were coming, and RMS isn’t shy about shocking the industry with big changes in their models.
What I can’t glean from the article is if the “big changes” represent RMS’s short-range forecast, representing the increased risk faced by insurers from the current warm cycle in Atlantic waters, or if they’ve bumped everything up for the long term.
The article mentions that there are two schools of thought on the cause of the recent bad years of Atlantic hurricanes — multi-decadal cycles vs. global warming. There’s actually a third school, to which I belong — that the bulk of the recent uptick is the result of long-term cycles, but that there also very likely is a global warming component that is, as of yet, somewhat minor, but has the potential to be come much more significant, and noticeable. As far as the 2004 and 2005 Atlantic seasons go… well, there’s significant variance around the overall trend curve. And we’ve seen significant variance.
Over at WeatherUnderground, Jeff Masters has been working on the question of “are cat 4 and cat 5 hurricanes becoming more common”. It’s worth checking out if you have an interest.