(The other three participating groups were the American Insurance Association in Washington; the Alliance of American Insurers in Downers Grove, Ill.; and the National Association of Mutual Insurance Companies in Indianapolis.)

Mr. Cleasby said that the results culled by the four insurance company groups were presented to regulators during an interim meeting in Chicago, Aug. 22-23. As a result, he said, regulators agreed to make the data time period prospective–from Jan. 1 to June 30, 2003, rather than from July 1, 2001 to June 30, 2002.

The decision is important, according to Mr. Cleasby, because “the stretch on the [information technology] staff would be very difficult for companies.”

Indeed, the survey results pointed to this difficulty. It asked respondents if companies that are part of a group use the same data systems. Thirty of 60 companies responded that they were part of the same data system.

However, Mr. Cleasby noted that the companies that said they are not part of the same system have slightly or conceivably very different data systems, complicating data call efforts.

Several questions addressed the issue of whether companies retrieved information electronically, manually or through both means.

For example, a question on available claims information found that 68.4 percent would retrieve information electronically, but 31.6 percent would use a combination of both electronic and manual retrieval.

Although a decision was made by regulators not to include claims data in the data call, the industry survey included information on claims.

For instance, when asked what kind of claims information is available, 57 of 63 companies said that information is available on the date a claim is made or opened; 38 on the date of a claims settlement; 54 on the date a draft or check was issued; and 55 on the date a claim file was closed.

When asked about sorting information by lines of coverage, 27 could sort auto claims data by bodily injury, 26 by property damage, 25 by collision, 25 by comprehensive, 26 by UM/UIM, and 24 by PIP.

A total of 38 of 63 companies had information readily available by number of auto cancellations during the first 59 days after a policy's effective date; 37 had the number of cancellations 60 days or more after the effective date; 40 had information on the number of auto non-renewals; 52 on the number of autos insured; and 54 on the number of auto policies in force.

The same question was asked about homeowners and renters insurance. Thirty of 63 companies had information readily available by number of cancellations during the first 59 days after the effective date; 28 had the number of cancellations 60 days or more after the effective date; 35 had information on the number of non-renewals; 43 on the number of homes insured; and 50 on the number of policies in force.

Mr. Cleasby said the results are instructive because they point out the need to make sure that the benefits of having the data outweigh the costs of retrieving it. Those costs, he said, are just for the p-c industry, and do not take into account the expenses that state insurance departments would face in analyzing the data.

More discussion is needed on the value of that information and whether it will be used to target examinations, according to Laura Kersey, a counsel with the AIA. Removing the retrospective requirement will reduce costs, she added.

Ms. Kersey said it is a good thing that states will be collecting data rather than having it collected in a central place that could open up access to what companies consider confidential information. However, she expressed concern that if data is gathered, that states actually make use of it.

Jim Connolly is a senior editor with NU's Life & Health/Financial Services edition.

Reproduced from National Underwriter Property & Casualty/Risk & Benefits Management Edition, September 2, 2002. Copyright 2002 by The National Underwriter Company in the serial publication. All rights reserved.Copyright in this article as an independent work may be held by the author.