CASNET has had a bit of a discussion over the past few days regarding
the changing definition of ALAE versus ULAE. The annual statement blank
has one definition; statistical agents may have a different definition
and reinsurance treaties may treat things yet another way. There are
discussions going on now at the NAIC to try to have the statistical
definitions of ALAE / ULAE get a little closer to the annual statement
definitions, but there are practical cost-based barriers that will
probably keep that from being mandated completely.
Yet another area where there is a decided lack of uniformity is in the
counting of claims. To my knowledge (please correct me if I am wrong),
there is no direction in the annual statement for how claims are to be
counted. Specifically, if an auto accident injures two 3rd parties,
some insurers will count this as one claim and others will count it as
two claims. I will refer to the first method as accident-based and the
other as claimant-based.
For statistical reporting, ISO wants claimant-based reports. For the
example just given, there would be 2 reserves ultimately replaced by 2
paid amounts. NAII and NISS will accept claim counting on either basis,
but they request that insurers remain consistent with whatever method
that they select to use.
As a regulatory actuary, I cannot think of any compelling reason to
force claim count practices to be the same for every insurer for
Schedule P purposes. Average claim sizes between insurers will vary for
many reasons other than claim counting practices. Mandating the same
counting practices would not make them comparable between each other,
and that would be the only reason that I could think of to mandate a
change. (However, one might consider the insertion of an interrogatory
to determine if an insurer had changed its claim counting practices.)
Statistical reporting is different. With statistical reporting, I would
like to be able to compare claim frequencies and average severities over
time and between different groupings of insureds, but it mucks things up
if different insurers count things in different ways, and they have
differing market penetrations for the groups (territories, zips, age
classes, etc.) in which I am interested. As such, there is a legitimate
reason for the regulatory community to be interested in this and it is
being discussed at meetings of the NAIC's Statistical Handbook Working
Group.
A poll of the major private passenger auto writers revealed that reports
to ISO were mostly claimant-based (as expected), but revealed about a
50-50 split between accident-based and claimant-based reporting for
insurers reporting to NAII and NISS. (In most states, the bulk of the
private passenger auto market reports to other than ISO.) The dilemma
faced by the NAIC working group is that a mandated change from accident-
based reporting to claimant-based reporting would be a fundamental and
therefore expensive change for a large number of insurers. It wouldn't
be as simple as just adding another data element (e.g., whether the
insured is a homeowner or not).
Some alternate ideas have been put forward but have had very little
discussion to date. These ideas would leave claimant-based reporting
alone, as it already provides all the information needed. Rather, one
or two data elements would be added for insurers that utilize accident-
based reporting. The elements that might be added would be the number
of claimants for each "claim" and the dollar amount of the "claim" that
is in excess of financial responsibility limits. In this fashion, claim
counting, average severities, average limited severities, and losses
above and below financial responsibility limits could be studied
without distortion. It would fall short of claimant-based reporting
for working with size-of-loss distributions, although presumably there
would be enough claimant-based experience available that one could
still do plenty of that type of work.
Has anyone with an insurer that counts and tracks claims on an accident
basis attempted to capture information for the purposes just described?
The intended advantage to do things the way that I have described is to
get most of the benefits that would be realized in going to claimant-
based reporting with a much smaller cost. (There would no impact
whatsoever on insurers that already report on a claimant basis.) Has
anyone considered an approach like this? Do you have any thoughts
worth sharing, either privately or back on CASNET? Thank you.
Alan Wickman, ACAS
Administrator, Actuarial Division
Nebraska Department of Insurance Fax: (402) 471-6559
941 "O" Street, Suite 400 Phone: (402) 471-4646
Lincoln, NE 68508 Internet: ins1008@vmhost.cdp.state.ne.us
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.