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

Predictive models problematic in underwriting directors, officers cover

Reprints
Predictive models problematic in underwriting directors, officers cover

ORLANDO, Fla. — Challenges surrounding data quality and availability need to be surmounted if predictive models are to play a larger role in directors and officers insurance, speakers said during the Professional Liability Underwriting Society’s 26th international conference.

“If knowledge is power, these tools represent the future of D&O underwriting,” Sean M. Fitzpatrick, New York-based chief operating officer of Dutch-based specialty insurer ANV Group, said Tuesday during a panel discussion.

Nonetheless, Randy Hein, vice president of Warren, N.J.-based Chubb Specialty Insurance, said modelers face challenges in adapting predictive models for D&O insurance. One challenge is that the historical data needed to populate the model is limited as the underlying business of each company being underwritten is unique and constantly changing.

“D&O underwriting is much more dynamic than property underwriting because public companies are dynamic,” he said.

Aziz Lookman, senior director of analytics for New York-based American International Group Inc., agreed that the quality of data feeding D&O models will vary and predictive models used in other lines of insurance may not work as well in D&O lines.

“For somebody planning to use an off-the-shelf model, do your due diligence to make sure that the model is all that it is cracked up to be,” Mr. Lookman said during the Orlando, Fla., conference.

Michael Klausner, professor of law at Stanford University, said one advantage that D&O modelers have is that public companies are forced to release a great deal of information. Along with a team of students and graduate assistants, Mr. Klausner has compiled a database of information from publicly held companies that he has used to model the likelihood that D&O lawsuits will succeed.

Business Insurance's digital coverage of the 2013 Professional Liability Underwriting Society's international conference is sponsored by Ace. To view all the Digital Daily news and related content in its ideal form, use a nonmobile browser to visit www.businessinsurance.com/PLUS2013.