The Sleeping Giants: Notable Absences in 2013 U.S. Cat Season

"[Last year] was one of the lightest hurricane seasons in terms of catastrophe losses, but storm frequency was comparable to previous years," explains Joe Louwagie, assistant vice president of Property Claim Services (PCS), a division of Verisk Analytics.

Nevertheless, a succession of unfortunate weather patterns in late November left many ambling to find shelter for the holidays and for the foreseeable future. As with any insured loss, there were lessons to be gleaned and motivation for insurers to do a better job of handling claims while mitigating future risk.                  

By reflecting on this past catastrophe season and drawing from the data derived, insurance adjusters and risk managers especially can analyze trends and patterns, identifying potential future complications and weather events. Historically, catastrophe forecasting has been a precarious endeavor at best. At worst, it has been a disaster of its own making.

The field of weather forecasting has been riddled with skepticism since its inception. One could even think back to mythology where the town codger—loon or prophet?—would foreshadow impending disaster, only to be met with derision.

A Conspicuous Absence

Luckily the tools at the industry's disposal have progressed eons since, and forecasters have become respected members of the claims community. Still, the revered scientists at Colorado State University (CSU) have some explaining to do. By all accounts and most measures, their projections were a colossal failure. Phil Klotzbach and William Gray had predicted an "above average" 2013 Atlantic hurricane season, which would later be characterized precisely by what was missing—namely, a storm of Sandy proportions.

Klotzbach and Gray had predicted that 18 tropical storms and nine hurricanes—four reaching Category 3 or higher—would form in the Atlantic. Hurricane Humberto took the initiative on Sept. 11, but it fell mere hours short of setting a record. Then in early October, Tropical Storm Karen was poised to hit the Gulf Coast and had the potential to develop into a hurricane while approaching the United States. Yet, in the hours leading up to landfall, Karen had a change of heart. Atmospheric conditions shifted, dismantling her resolve.

Consequently, Karen's impact was described as "minimal" when moving through the coastal area. Ultimately, 2013 would witness only 13 named storms, with just two reaching "hurricane" status, neither of which were defined as "major." To their credit, the veteran CSU forecasters did issue a mea culpa of sorts.

"It was one of the largest busts for our research team in the 30 years we've been issuing this report," said Klotzbach, adding "These [predictions] should be judged on their overall track record of success, not on a single or a few unsuccessful seasonal forecasts."

Lukewarm Waters

The previous year, there were 19 tropical storms, 10 of which were upgraded to hurricanes. Of those 10, two were classified as major.
Tepid hurricane activity is interesting on many levels and carries with it some notable implications for the p&c industry in terms of forecasting and insurer solvency. Take, for instance, the findings presented recently in Moody's annual report. You'll find a summary at the end of this article. When evaluating exposures, hurricanes top the list in terms of the capability to rock insurers at their very core. So it would seem the industry dodged the proverbial bullet. Right?

Even so, Moody's contends that other severe storms—including tornadoes and hailstorms—have been a source of significant volatility in recent years. According to PCS, between 1993 and 2012, tornadoes comprised about 36 percent of all U.S. insured catastrophe losses. Although such storms do not generally threaten insurer capital to the extent that their drier counterparts do (hurricanes and earthquakes), they nevertheless can cause significant earnings volatility.

Moreover, from a risk modeling standpoint, severe storms are problematic. Several factors make them notoriously difficult to accurately model. In comparison to hurricanes and earthquakes, real numbers are more difficult to observe, with historic data more elusive, PCS adds. Insurers and their risk management cohorts are responding with creative strategies, however. For starters, many companies have increased annual catastrophe budgets to reflect the elevated high-frequency, low-severity events in recent years.

The Global Picture

To redirect our focus to the overall 2013 picture, despite the lows for hurricane, wildfire, and tornado losses, other events garnered considerable public attention. Surprisingly, abnormal sinkhole activity took center stage, as three separate sinkhole catastrophes transpired in Florida. In addition to sparking rampant media coverage, those events served to raise public awareness of the often overlooked peril.

Aside from sinkholes, thunderstorms were a major factor in the year's overall catastrophe analysis.  

"Thunderstorms dominated catastrophe activity, causing nearly 90 percent of insured catastrophe losses in the United States," Louwagie explains. "Wildfire emerged as one of the most discussed perils of the year, given the recent devastation in Colorado Springs. It's an emerging risk to follow proactively in the year ahead."

In 2012, there were 11 weather and climate disasters, with losses exceeding $1 billion each in the United States. Those disasters claimed 377 lives, by CoreLogic's estimation. These natural events caused a combined total of more than $110 billion in damages, making 2012 the second costliest year on record behind 2005.

A report issued by Swiss Re in December 2013 confirms the U.S. experienced a moderate year when compared to cat losses on the global scale. In 2013, insured losses from disasters were down significantly from about $81 billion in 2012.

Economic losses from cats in 2013 amounted to about $130 billion, compared with $196 billion in 2012. Swiss Re adds that the loss of human life last year was greater—25,000 died as a result from natural and man-made disasters compared to 14,000 in 2012.

The highest loss of life occurred from Typhoon Haiyan, which killed at least 7,000 people and packed some of the strongest winds ever recorded, bringing heavy rain and storm surges.

Swiss Re notes the June flooding in central and Eastern Europe caused $4 billion in insured losses, becoming the second most expensive fresh water flood on record, just behind the 2011 Thailand flood, which generated more than $16 billion in insured claims. PCS also designated five catastrophe events in Canada in 2013, the two most significant being the floods in Calgary and Toronto.

On the next page, we take a closer look at the findings presented by Moody's in its annual report.

Catastrophes Present Key Risks for P&C Insurers

In a report released on Dec. 13, Moody's Investors Service asserts that, in spite of an uncharacteristically quiet 2013 hurricane season, catastrophes will remain a top credit risk and key area of risk management focus for p&c insurers.

"The U.S. p&c industry's current strong capital levels enable it to withstand volatility from natural and man-made catastrophes even up to severe levels such as an event with a 1-in-250 probability, but the risk nevertheless remains significant for the industry for 2014 and beyond," the company explains in "Catastrophe Exposure Remains Key Risk For US P&C Insurance Firms."

Paul Bauer, Moody's analyst and author of the report, presents the rating agency's key credit observations from discussions with company management teams and Moody's 2013 surveys of p&c catastrophe exposure. The analysis puts baseline catastrophe losses on average to be equivalent to between 4 and 6 percent of earned premium in a given year.

Variance in Exposure

Bauer notes that severe storms pose particular risk management challenges to p&c insurers, even though severe storms do not threaten capital to extent of hurricanes and earthquakes. Nevertheless, severe storms can cause significant earnings volatility and have traditionally been difficult to model. He also points out that catastrophe exposure "varies widely by company" but that based on surveys, even severe catastrophes—or a series of them—can be absorbed by a moderate portion of equity capital rather than threatening solvency. 

"In the most-exposed quartile of its rated companies, average loss estimates on a 1-in-250-year basis are 30 percent of policyholders' surplus net-of-reinsurance," he explains. "Conversely, about a quarter of rated p&c companies could withstand a 1-in-250-year loss with an equity impact of less than 10 percent net of reinsurance."

Companies continue to enhance their risk management processes, by better testing model sensitivity, enhancing data validation, and adjusting modeled results based on portfolio experience," Moody's adds.

Among Low Probability but High Severity Risks, Hurricanes are Predominant

The chart below shows the relative size of various individually modeled net losses at a 1/250 return period. As illustrated, the largest potential losses at the low probability but high severity level (the "tail") come from hurricanes, followed by California earthquake.

Note that though Florida hurricane risk is fairly high in the chart below, it is still lower than it otherwise would be because the data does not include the public entity Citizens Property Insurance Corporation run by the State of Florida, which is expected to absorb a significant share of losses in the event of a Florida hurricane. Citizens' gross hurricane exposure is estimated at about $31 billion for a 1-in-250 event, and significantly exceeds the 1-in-250 PMLs of all of our rated primary P&C companies combined.

Many of the large national property writers rated by Moody's have been actively working to reduce their exposure to Florida over time, leaving a greater share of the market not only with Citizens but also with small, unrated regional insurers.  

Source: Moody's