Combating Weather-Related Claims Fraud

By John Deegan and Stephen Massa

The variability of weather and insurance losses

The severe winter weather in the United States last winter may have caused more than $1 billion in insured losses, putting 2011 on a path toward equaling, or even surpassing, the $2.6 billion in winter-related insurance claims generated nationwide in 2010, according to the Insurance Information Institute.

Insurers are hard-pressed to anticipate the effect of weather events and accurately predict losses. Advances in meteorological models and tools over the past several years have proven extremely useful in predicting severe weather, but year-to-year variability in meteorological conditions is still difficult to gauge. Weather-related losses can increase or decrease substantially and be influenced by other factors, such as the location of severe weather events and the presence of fraudulent claims.

Trends in weather-related claim reporting and the incidence of fraud

In recent years, the debate over climate change and possible profound changes in weather patterns and their impact have led to heightened awareness within the property/casualty insurance industry that weather may play a more prominent role in fraud activity than once considered. Even with no change in the frequency of severe weather events, the current weakness in the economy — which in itself typically leads to increases in fraudulent claims — makes greater vigilance and more proactive analysis of weather-related losses essential.

The National Insurance Crime Bureau (NICB) provided evidence of a trend toward greater insurer vigilance of claims related to severe weather through its analysis of ISO ClaimSearch® insurer-reported hail claims data for the period January 1, 2006, through March 31, 2010. It also analyzed and contrasted questionable claims reported by insurers to the NICB's questionable-claims system during the same period.

In June 2010, the NICB ForeCASTTM Report on this analysis noted the number of hail loss claims grew from 256,000 in 2006 to 413,178 in 2009, a 61 percent increase. As the number of claims trended higher, the ratio of hail loss claims referred to the NICB as questionable increased from 1.2 questionable claims per 1,000 claims in 2006 to 1.7 questionable claims per 1,000 claims in 2009. The increase in the ratio is evidence of an overall rise in potentially fraudulent hail loss claims activity. (See Figures 1 and 3.)

Further NICB analysis of questionable hail claims filed in the first half of 2010 showed that seven of the top ten states with the highest ratio of questionable claims to claims for hail losses are not in the central section of the country, where most hail losses are generally concentrated. This suggests that fraudulent hail claims can and do occur in any part of the country.

Figure 1
Analysis of Questionable Claims

  2006 2007
Total hail loss claims 256,000 413,178
Questionable Claims 301 711
Ratio (per 1,000 claims) 1.2 1.7
Source: National Insurance Crime Bureau

The table above illustrates the questionable-claims-to-claims ratio per 1,000 claims — or the number of suspected fraud claims reported per 1,000 total claims for the peril.

Challenges of claims analysis

Typically, adjusters face conflicting objectives when trying to settle weather-related claims — namely, how to balance the delivery of responsive customer service in a timely fashion with the need to protect their business against fraud risk. Unfortunately, as insurers rush to settle these claims, proper due diligence may not be rigorous enough. Fraudsters are only too quick to prey on the situation.

Turning to weather data resources for help

Insurers are increasingly joining forces with forensic meteorologists to manage weather-related insurance fraud more efficiently. There is now a greater emphasis on identifying and exploiting a full range of weather data assets and related services to help analyze claims quickly. Government agencies such as the National Oceanic and Atmospheric Administration (NOAA) / National Weather Service (NWS) and the National Severe Storms Laboratory (NSSL) offer valuable weather information at minimal or no cost. But the data is not easy for claims departments to use, and the "freshness" of the data can sometimes be an issue. A number of private organizations also specialize in weather forensic information.

What information insurers need

Proactive analysis of weather-related claims starts with the reporting of loss claims to ISO ClaimSearch, the industrywide all-claims database, as a preliminary step in analyzing the loss patterns of a risk. Full reporting of loss is also key to improving the quality of the data available to the industry for claims evaluation and special investigations.

Adjusters achieve additional value when they document the date, time, and occurrence of specific severe weather events near the location of the loss. The ease of access to accurate and reliable information as part of the claims-handling workflow means claims personnel can deliver timely and responsive customer service.

Figure 2
Hail Location

Average communte time in CA after spacial smoothing

Example of geomapping of severe weather reporting points in relation to location of loss. This includes the proximity of the closest National Weather Service observation station and local weather spotters. In this case, valuable on-site observations of storm conditions such as hail size can help adjusters

How Verisk Insurance Solutions is helping insurers

By joining forces, ISO and Atmospheric and Environmental Research (AER), both members of the Verisk Insurance Solutions group, offer a set of reports that can help mitigate claims fraud and facilitate processing of meritorious claims. They include hail, lightning, severe wind, and tornado Weather Forensic Reports delivered through the ISO ClaimSearch service and its integrated third-party information portal, Decision Net®. Standard features of the reports include a severe weather event summary, geomapping of weather in relation to the location of interest (see Figure 2), and a summary of general weather observations, such as temperature and wind speed, which can prove valuable in the absence of definitive severe weather information.

Figure 3
Ratio of Hail Loss Questionable Claims to Claims (2006 to 2009)


This graph depicts the ratio of suspicious hail claims referred to the NICB to the total number of industrywide hail claims reported to ISO ClaimSearch for the period 2006 through 2009 inclusive.

The 12 percent average annual increase in this measure from 1.2 to 1.7 questionable claims per 1,000 reported to ISO ClaimSearch suggests a growing incidence of fraud in hail damage claims.

Average communte time in CA after spacial smoothing

AER is an authority in the field of weather, climate, and environmental data analytics. Its areas of expertise span climate change, defense, hydrology, meteorology, oceanography, space, remote sensing, and atmospheric and environmental science. AER also offers expert verification and assessment of weather-related events through its weather consultation services.

John Deegan is product development manager for ISO ClaimSearch Decision Net at ISO. Stephen Massa is director of data services at Atmospheric and Environmental Research.