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

DIGGING FOR INFORMATION WORTH ITS WEIGHT IN GOLD

Reprints

A West Coast-based company decided not to take chances with Mother Nature when the National Weather Service started predicting a devastating El Nino this year.

A member of the risk management staff was assigned the task of gathering information on the company's weather-related exposures.

He then downloaded historical weather information and forecasts from the Boulder, Colo.-based National Oceanic and Atmospheric Administration via the Internet at www.noaa.gov.

Added to this mix were historical claims data on weather-related losses, the values of which were updated to current dollars.

The purpose of the analysis, still ongoing, is to enable the company to better prepare for El Nino-related losses.

It's also a perfect example of "data mining," a trendy term for the kind of research that risk managers and risk analysts do almost every day.

But the growth of the Internet and personal computing technology is making that research process faster and less arduous.

For example, it takes only a few clicks with a mouse for a risk analyst to drill down through Occidental Petroleum Corp.'s casualty exposure information.

The 2-year-old Corporate Casualty Insurance Survey System warehouses annual exposure data from the Los Angeles-based petroleum company's operating units worldwide.

"In the past, we had to collect the data manually," said Randal K. Bilodeau, senior risk management associate in Occidental's casualty department. "Now we use a diskette."

Information from payroll, accounting, human resources, fleet administration, operating facilities and foreign locations is assembled on a customized survey diskette sent annually to each of the company's operating groups for completion.

"It's a very user-friendly system," Mr. Bilodeau said. "The data input screens are in Windows. All you have to do is point and click."

Among the exposures listed are vehicles, joint-venture information, offshore platforms, pipe-lines, oil wells, refineries, oil storage tanks and vessels, tanker trucks and construction projects.

The information from the completed diskettes is then downloaded into a Microsoft Access-based database system for consolidation, review and analysis.

The final survey reports are sent to Occidental's foreign and domestic insurance brokers to assist in preparing underwriting submissions during the annual renewal process.

Mr. Bilodeau also performs analyses when exposure values change more than 10% from one year to the next.

The original software for this data warehousing and data mining project was written in 1987 on a mainframe-based "Clipper" system. Clipper systems allow users to build customized databases. The program was converted for use in the Windows platform two years ago, said Mr. Bilodeau.

Occidental Casualty Claims Manager Mark Oliver is using computer modeling to help the company resolve litigation with its liability insurers over coverage for toxic tort claims and other long-tail liabilities.

The model quantifies all potential insurance proceeds -- by insurer and by policy -- to determine whether the insurer is worth pursuing and to establish a starting point for negotiations.

It also calculates various coverage scenarios using any of four trigger theories: continuous, exposure, manifestation and injury-in-fact trigger.

The use of the model has resulted in insurance buyouts totaling $235 million and settlement agreements involving multiple insurers, Mr. Oliver said.

Chad Karls, an associate actuary with Milliman & Robertson in Milwaukee, recently mined data on behalf of an insurer that was considering entering the employment practices liability market.

"We used external data to come up with premium rates," Mr. Karls said.

Among the sources that Mr. Karls "mined" was the Equal Employment Opportunity Commission database, which contained the number of EEOC charges filed in each state.

"This helped determine the frequency (of EEOC claims) on a countrywide basis and by state," he pointed out.

Mr. Karls also used EEOC descriptions of resolved cases to calculate claims severity.

To determine the underlying exposure base, Mr. Karls used information from the U.S. Statistical Abstract compiled by the U.S. Census Bureau, which provides the number of full-time workers in each state.

Also included in the EPL report was information on defense costs derived from interviews with attorneys who handled such claims on behalf of insurers.