Counting claims

Alan Wickman ( awickman@doi.state.ne.us )
Sun, 28 Feb 1999 20:44:39 -0600

I am the chair of an NAIC working group that is looking at (among many =
other things) the fashion in which insurers count claims, with personal =
auto being the primary focus. Not all insurers count claims the same =
way, and we believe that this can create problems when data for many =
insurers is combined, as it is by statistical agents for reports to the =
states.

Although there are variations, most insurers can be said to count claims =
on either an "accident" or "claimant" basis. (These are my labels, and =
are not "official" in any way.) Suppose that your insured is at fault =
in an accident that injures two people. Both injured people bring =
actions against your insured. If you set up one BI liability reserve =
for the accident, then you count claims on a "accident" basis. If you =
set up two reserves, then you do things on a "claimant" basis.

There are a lot of technical data management nuances to this area, =
because insurers that keep track of claimants need to pull this =
information back together for various purposes. Well done, a claimant =
reporting basis provides more flexibility to users, because it can tell =
you everything that an "accident" basis can, plus more. But that is not =
the question, nor do I want to know which way that you count claims. (A =
survey that we took indicated that almost all ISO-reporting insurers =
count on a "claimant" basis, while non-ISO insurers split about 50-50 =
between the two methods.)

The question that I have is this: Is anyone aware of any studies, formal =
or informal, published or unpublished, that provide any quantification =
of the average claim frequency differences that occur purely because of =
these two different ways of counting claims? The answer would appear to =
be different for BI and PD.

Thank you.

Alan Wickman
awickman@doi.state.ne.us
Nebraska Department of Insurance
941 "O" Street, Suite 400
Lincoln, NE 68512

The NAIC group is the "Statistical Handbook Working Group."

Visit the CAS Web Site at http://www.casact.org
===============================================
To subscribe or unsubscribe from CASNET:
Send an e-mail to caslists@lists.casact.org
Type in the body join casnet to subscribe
or leave casnet to unsubscribe.