I am involved with the NAIC's Statistical Task Force (STF). The STF is
addressing changes to the model P&C rating laws to deal with disclosure
versus confidentiality of statistical data and reports that come into a
state's possession. Although changes to an NAIC model law lack any
official impact until one or more states pass the changes, the NAIC's work
with regulatory reengineering appear to make it likely that a significant
number may consider changes to their P&C rating laws within the next few
years. As such, changes made at this time might have a better chance of
adoption by the states than would otherwise be the case.
The language in the current model laws is subject to several
interpretations, but many read them to say that data submitted pursuant
to required statistical reporting are open to public viewing. In the
past, this really hasn't made much difference, because most states have
only received very high level summarizations of data -- nothing that would
get down to the level of individual insurers or claimants (or generally be
of much commercial value). Although states could have requested more
detail, or raw data for that matter, this was rarely done on account of the
difficulties that would occur with using such volumes of information. Over
the past 5 or 10 years, however, the proliferation of inexpensive yet
powerful PC's has effectively removed this inhibiting factor and it has
become more relevant to address whether highly detailed information in
state possession should be released.
So that you will know where I am coming from, I favor an alternative that
would:
o Protect the confidentiality of individual insureds and claimants,
although worded in such a fashion so as to allow WC experience rating to
continue. (The current law provides no such protection.)
o Protect the confidentiality of individual insurer loss data. Loss data
would only be provided in statistical reports which aggregate the
experience of many insurers, making it relatively unlikely that data
would be identifiable to individual insurers.
o Would allow disclosure of premiums and exposures for individual
insurers. For instance, individual insurer's writings by zip code
(when the state requires that to be reported) could be disclosed.
o Protect the confidentiality of data collected by statistical agents
that goes beyond the detail required by regulators. For instance, ISO
collects detail that goes beyond regulatory requirements for most states
and lines.
To date, every official response (that is, all letters) from insurers have
opposed any sort of disclosure of information that identifies individual
insurers, where the proposal that I have just outlined would allow premiums
and exposures to be identified for individual insurers. I really can't say
with any confidence where the votes on the STF are lined up at this point,
except to say that I have reason to believe that some are to the right of
my proposal (that is, they'd agree with the industry's comments to date),
some are to the left (that is, they'd favor release of individual insurer
loss information as long as it didn't identify individual claimants or
insureds), and some are pretty close to my position.
The argument that is made against the amount of disclosure that I propose
(although less than that now in effect) is that it would reveal valuable
insurer trade secrets. I tend to pooh-pooh that argument, as I believe
that an insurer wanting to find out who is in various markets (that being
what premium or exposure information would provide) can find that out in
other ways. I question whether the real reason for much of the opposition
is a fear of consumer groups getting information that will cause insurers'
underwriting practices to be called into question.
I would really like to get an idea how individual actuaries see this. In
this regard, although anything posted to CASNET is "public", I will neither
mention anyone's name nor intrepret his/her views as being those of his/her
employer. Rather, if a good discussion comes out of this, I may paraphrase
it (omitting any mention of names) during the STF's discussion of the
matter. Of course, I have no control over what anyone else may do with
postings to CASNET, but I think that this sort of use is polite and
anything more than that would often not be polite.
BTW, if you want to make a comment that will actually be distributed to STF
members, I suggest that you do so by putting words on paper and sending
them to me via fax or snail mail. If you want to do so, you may want to
contact me first to get a full set of the background materials.
Thank you.
Alan Wickman, Casualty Actuary Phone: (402) 471-4646
Nebraska Department of Insurance Fax: (402) 471-4610
941 "O" Street, Suite 400 e-mail: ins1008@vmhost.cdp.state.ne.us
Lincoln, NE 68508