Help

BI’s Article search uses Boolean search capabilities. If you are not familiar with these principles, here are some quick tips.

To search specifically for more than one word, put the search term in quotation marks. For example, “workers compensation”. This will limit your search to that combination of words.

To search for a combination of terms, use quotations and the & symbol. For example, “hurricane” & “loss”.

Login Register Subscribe

Insurer losses from rare November tornados expected to top $1 billion

Reprints
Insurer losses from rare November tornados expected to top $1 billion

The Nov. 17 tornado outbreak across the Midwest was a rare event in a year experts describe as unusually mild for U.S. tornado frequency, highlighting the unpredictable nature of tornadoes and the escalating severity of the insured losses.

Last month's catastrophic weather produced a preliminary count of 81 tornadoes, including two of powerful EF-4 strength in Illinois, among them the storm that devastated Washington, Ill. They were the first November EF-4 tornadoes on record in Illinois.

The twisters resulted in eight deaths and estimated total insured losses of more than $1 billion. The spate of tornadoes is expected to be the first-ever billion-dollar November storm loss in the country and one of possibly eight billion-dollar-plus weather-related U.S. catastrophes in 2013 — a year without a big Atlantic hurricane.

The increasing number of tornadoes with billion-dollar losses — driven in large part by increasing property values, according to some experts — is one more factor drawing insurers' and reinsurers' focus to noncoastal weather events and providing urgency for greater tornado resilience in both residential and commercial structures.

“We're starting to see more and more larger and more severe storms in middle America that are causing larger impacts on middle America,” said Andy Castaldi, head of catastrophe perils for the Americas at Swiss Re Ltd. in Armonk, N.Y. “Years ago, you wouldn't expect to see a tornado or a hailstorm hit the billion-dollar mark in losses. But losses of that size are becoming more frequent in any given year.

“That's a trend that we have to slow down, and that goes to things like better building codes,” Mr. Castaldi said.

%%BREAK%%

However, better building codes might not be able to protect a building from a strong tornado's 200 mph winds, he said. And property owners always face a cost/benefit question, with some concluding the costs of increasing buildings' tornado resilience isn't merited given the odds of the property being struck by such a storm.

“Part of it might be a misconception by people that if my building is built to the building code, then it's safe. The building code is sort of a minimum standard,” said Tanya Brown, a research engineer at the Insurance Institute for Business and Home Safety's Research Center in Richburg, S.C.

In addition, tornadoes are “such a spatially rare event. Your risk of getting hit by a tornado in any given year is pretty low,” Ms. Brown said.

“I think we need to distinguish between low-probability events and zero-probability events,” said James Waller, research meteorologist at Guy Carpenter & Co. L.L.C. in Philadelphia.

“EF-2s or weaker constitute 95% of all tornadoes. That's a range where structural resilience measures can really make a difference,” Mr. Waller said. “The first thing to go is usually the roof, and if you can keep the roof on for just a bit longer it can make a big difference in terms of the safety and the resilience of the home. The same thing goes for commercial structures as well.”

Regarding the unusual Nov. 17 tornado outbreak in the Midwest, Bill Bunting, forecast operations chief at the National Weather Service Storm Prediction Center in Norman, Okla., said, “They do happen, but to see that many tornadoes that far north certainly is a very rare occurrence in November.”

%%BREAK%%

Mr. Bunting said that among the conditions that led to the Nov. 17 tornadoes were a very strong jet stream and intensifying low pressure system at the surface.

“That increased the degree of wind shear, and it brought very warm, moist air into the Ohio Valley, where you normally don't see it in November,” he said.

While the storm system was unusual, drawing any connection between the tornado outbreak and climate change is difficult, if not impossible, in the absence of more data and history, experts say.

“The challenge is that November tornado events are so rare,” Mr. Bunting said. “It's hard to infer a trend.”

“Events like this may have happened in the past or could have happened in the past, and it could be the normal variation of outcomes in the current climate,” Mr. Castaldi said. “However, this is something that we could expect to see as a result of climate change, and 20 years from now we might look back and say this was the start of something.”

Mr. Waller said U.S. tornado outbreaks tend to demonstrate “crazy variability from year to year. There's no evidence to detect any statistically relevant trends when you look back over time.

%%BREAK%%

“We really don't have as robust a data set as we need to make a conclusion one way or another at this time,” he said. “You can't really point to one event and make an argument for climate change. You have to look at the aggregation of those events over time.”

Though climate change might not produce more or more severe tornadoes, it could produce more outside of spring and summer, said Matthew Nielsen, director of model project management at Newark, Calif.-based Risk Management Solutions Inc.

“It certainly has been speculated that these types of events, happening at weird times — off of the peak — that could happen more frequently in the future,” Mr. Nielsen said.

While a link between tornadoes and climate change might not be clear, the increased frequency of large tornado losses is getting insurers' and reinsurers' attention, though their response remains to be seen, Mr. Nielsen said. “This is no longer just background noise as they focus on earthquakes and hurricanes,” he said.

“What I've seen a lot lately from a lot of people in the insurance industry that I work with is people trying to model these events,” he said, lessening their reliance on historic claims information in assessing tornado risk.

But modeling tornadoes is challenging, Mr. Waller said. “It's an ongoing effort in the industry,” he said. “Tornadoes are really, really tricky to model because they happen on such a small scale. Hurricanes are big.

“Tornadoes are smaller, they're more difficult to model. The random nature is more difficult. They really are low-probability events,” Mr. Waller said. “That's the challenge that the modelers face. They've made advances in that regard, and there are more advances to be made, I'm sure.”