AIR U.S. Hurricane Model Update Adds Business Interruption

Boston - AIR Worldwide Corp., (AIR) released Version 9.0 of its U.S. Hurricane Model, which includes enhanced methodology for estimating business interruption (BI) losses, which accounts for both building and business characteristics when estimating total BI downtime and includes indirect losses from sources other than physical damage to the insured building. The release also includes enhancements to the model’s demand surge function, the vulnerability of residential contents and pool enclosures, and incorporates research by AIR meteorologists and climate scientists into the link between elevated sea-surface temperatures (SSTs) and U.S. landfall activity. “In the aftermath of the 2004 and 2005 hurricane seasons, AIR undertook an intensive research and development effort to improve the way catastrophe models estimate business interruption losses,” according to Dr. Jayanta Guin, senior vice president of research and modeling at AIR Worldwide.  The software also estimates indirect business interruption losses—those stemming from sources other than physical damage to the insured building—such as utility service interruption, actions taken by civil authorities, dependent building damage, and extended period coverage. The new model has been validated using detailed claims data from 2004 and 2005 hurricane seasons. Boston-based AIR is also providing an update to Version 9.0 of the U.S. Hurricane Model’s near-term catalog, which incorporates an additional 12 months—and several man-years—of research by AIR meteorologists and climate scientists into the link between elevated SSTs and U.S. landfall activity.  Rather than relying on highly uncertain point forecasts of sea-surface temperatures, the Version 9.0 near-term catalog is instead conditioned on the assumption that currently elevated SSTs are likely to remain elevated for the next five years. As a result, the inclusion of one additional hurricane season will not significantly change estimates of near term risk. However, uncertainty in near-term estimates of landfall frequency remain significant, so AIR is once again releasing the near-term catalog as a supplement to, rather than a replacement for the standard catalog, which is based on more than 100 years of historical data. Analysis of claims data from recent hurricane seasons has revealed that contents vulnerability for single-family homes has decreased significantly in recent years. The updated model accounts for this trend. The model has also been enhanced to explicitly account for the impact of hurricane winds on pool enclosures, structures that—according to AIR’s latest research—show a higher vulnerability to wind damage than previously estimated. Finally, the impact of demand surge—the increase in material, services, and labor costs due to increased demand following a catastrophic event—was fine-tuned and validated using high resolution construction cost time series data for the 2004 and 2005 hurricanes taken from XactAnalysis, a reporting tool created by ISO subsidiary Xactware. Research at AIR shows higher levels of demand surge for time element, including BI. A state-of-the-art model alone is not sufficient for generating accurate BI loss estimates, notes the company. AIR’s analyses also revealed significant issues with the accuracy and completeness of insurers’ business interruption exposure data. For many companies, large numbers of locations have very low BI exposure values. In most cases, business interruption limits, which reflect part-year business income exposure, have been set equal to annual BI exposure. AIR also found evidence that companies are using general “rules of thumb” to determine the BI limit, rather than the use of BI worksheets for each location in multi-location policies. Finally, the number of locations that may sustai damage in a catastrophe is often underestimated. Together, these issues regarding the quality of BI exposure data lead to significant underestimation of modeled BI losses by many companies.  “Modeled loss estimates are only as accurate as the exposure data input into the catastrophe model,” continued Dr. Guin. “Insurers must continue to put an emphasis on improving the quality and completeness of their BI exposure data to improve the accuracy of the catastrophe risk information used by company management.” Source: AIR Worldwide Corp.