NU Online News Service, Oct. 28, 9:44 a.m. EST

The extraordinary path of Hurricane Irene through various inland regions of the Northeastern United States partly validates aspects of the guidance provided by the new Atlantic Hurricane Model from Risk Management Solutions.

RMS' Version 11 upgrade to its catastrophe model posited a greater likelihood of hurricane exposures beyond U.S. coastlines. The company recalibrated its model to account for the higher risk-potential of a hurricane affecting a greater swath of geography—not just coastal properties, but inland regions.

AIR's upgrade to version 12 made similar strides in advancing our understanding of hurricane risks. For instance, AIR updated wind damage functions based on the latest findings from their analysis of detailed claims data from recent hurricane seasons. This resulted in increased inland loss estimates and the addition of new states.

While the industry should applaud the increasing sophistication of the information provided by catastrophe modelers, the models are not a replacement for human ingenuity in the underwriting process.

Cat models only provide estimates of hurricane (or other catastrophe) losses. The process is conceptually simple: generate a sample event, follow it across the country, and simulate the losses based on both storm characteristics and policy conditions. Then, aggregate the results over thousands of years. However, there will always be a need for flexibility in assessing individual property risks and accumulated exposures in a particular region.

Although some climatologists cite a “new normal” in hurricane activity—a greater propensity for windstorms to cause damage in geographic areas formerly deemed “non-critical”—placing too much emphasis on the catastrophe models fails to account for the plethora of other risk factors at play.

Yet, many insurance carriers seem to be doing just that—letting technology via the models make the underwriting decisions for them. Why?

Part of this is due to the pressure from the rating agencies and reinsurers, not to mention elected officials and regulators. Standard insurance carriers, to a good extent, have their hands tied. Their cost of goods sold has gone up, insofar as the capital they must have to address the higher apparent risk of their property catastrophe exposures.

Why higher apparent risk? The risk of loss hasn't changed. The assessment of that risk has. This is not a problem for the excess and surplus lines marketplace, which enjoys the benefit of remarkable flexibility in its underwriting process and approach.

Given the current upheaval in the property catastrophe market, with prices firming, underwriting terms and conditions tightening, and retail brokers shopping accounts left in the lurch, the E&S market has the rare opportunity to take the lead in developing a flexible underwriting solution to these challenges.

The market's greater freedoms are needed in this time of volatility, and there are significant business opportunities for wholesale brokers and E&S carriers that step up and assume this vital role.

A Different Storm

RMS Version 11 was the consequence of information gleaned primarily but not exclusively from Hurricane Ike—the third costliest hurricane to make landfall in the United States.

RMS took the hurricane's atypical activity into account in its underlying calibration data and modeling methodologies, releasing Version 11 this past February. Compared to Version 10, the differences are significant.

Whether new data increases or decreases risk estimates is beyond the modeler's control—that's the job of insurance underwriters. As the organization attests, “We believe companies should have access to all new and relevant information known to and reviewed by us so they can adjust underwriting decisions and capital requirements appropriately. The alternative would lead to compromised decisions and surprises when catastrophes strike, and has associated liability implications.”

Certainly, the insurance industry looks to AIR, RMS, and other modelers to provide greater details assisting the underwriting process. But, in this case the new models have brought about significant changes in underwriting that pose serious challenges for buyers of property catastrophe insurance, many of them in regions of the country heretofore considered to pose non-critical risk.

The new models have also made an impact on regulators, rating agencies and reinsurers, compelling the standard insurance market to review its risk aggregations and pricing. The recent changes have resulted in extraordinary volatility in the homeowners and business premises insurance markets, spurred on by both capital and regulatory concerns. We are at the beginning of a sea change in the underwriting treatment of these risks, one with enormous social and economic implications.

This puts the spotlight on the underwriter, and the level of skill and experience that he or she brings to bear on the risk. Insurers that employ experienced property underwriters with the skill and expertise to evaluate the plethora of weather-related risks and non-disaster exposures confronting building structures of different type, age and condition will likely see the models as a single part of the underwriting process.

What does this mean for brokers and their clients?

This difference in property underwriting has implications for the broker-client experience. A wholesale broker or managing general agency that partner with a carrier who has expert property underwriters will find that customers are more satisfied with the experience. This assists greater customer retention rates. Similarly, the ability of a carrier to tailor the policy to the clients' needs will result in a more satisfied customer.

Some underwriters place too much emphasis on a single underwriting criterion, such as a sophisticated (and perhaps well-intentioned) catastrophe model. Rather than assess the wealth of historical data drawn from each carrier's respective books of business, a single piece of advice has prevailed.

Instead of evaluating a particular property catastrophe risk insofar as the many primary and secondary modifiers that more explicitly tell its story, a single catastrophe model is being relied on for the tale. The human approach to underwriting seems to be disappearing.

Financial Flexibility

Quality data to drive underwriting decisions can be accumulated and assessed.

This does not mean that RMS Version 11 and other stringent models from AIR Worldwide and Eqecat should be discounted. The more information the better the underwriting process, which leads to improved decision-making. And that's the point—people make decisions, not machines.

People drive ingenuity, not institutions. People are the entrepreneurial engine of an insurance company's growth, not machines, not technology.

RMS and other modelers take into account such risk modifiers as a premises' construction quality, the type of roof framing, the age and anchoring of roofs, and the wind resistance of windows, among many other considerations. This is valuable insight, but until human eyes discern and digest these risk factors they remain pieces of information.

Underwriters must evaluate property catastrophe risk on a location-by-location basis, digging into the detailed construction data. Catastrophe models tell the general story, but there are deviations to the standard. The models are telling, but they are just one tool, not the entire toolbox.

Certainly, disciplined underwriting is required, given the “new normal” of climatic events. Regions formerly considered “non-critical” no longer escape the brunt of today's hurricanes. It's a new world.

At the same time, these geographic areas are simply not at the risk level of truly critical regions like the Gulf Coast and the Florida shoreline. Intellectually, these exposures do appear to be greater, and the path of storms obviously is atypical—a tornado hitting Massachusetts is an eye-opener for sure.

The market must price for these enhanced exposures, and when regulators do not permit such clear-thinking actions, the E&S industry must step forward.

In the volatile market environment, the flexibility to appropriately price the perceived risk exposure is vital. The industry can take the lead in forging a “new normal” with underwriting expertise, refinement and diligence.

Michael J. Carr, is senior vice president, excess and surplus lines/property, at Liberty International Underwriters

Special Reports /

Catastrophe Models Are a Tool, Not the Toolbox

Related Stories

Recommended For You