It sounds to me like these principles, highly questionable in Life, have
been extended to P/C. Your MedMal example, while extreme, is an excellent
case in point. How about WC and the effect of pension cases? Auto BI? This
whole concept sounds to me like something very ripe for CAS to address.
Brad Gile
-----Original Message-----
From: Ralph.S.Blanchard@travelers.com <Ralph.S.Blanchard@travelers.com>
To: casnet@lists.casact.org <casnet@lists.casact.org>
Date: Thursday, October 14, 1999 7:49 AM
Subject: Re: Michael Lamb's premium deficiency reserve (PDR) discussion
>
>
>I would like to take Michael Lamb up on his offer to write about PDR
reserves,
>but have to first read the recent PDR-relevant papers presented at the last
>CLRS. Would the best response be a brand new paper, or discussions of
those
>papers?
>
>As to whether or not to allow discounting, a prior contributor to this
>discussion raised an interesting issue about whether the existance of a
non-zero
>PDR implies that rates are deficient. Most states require that investment
>income be taken into account in ratemaking. For a long tail line, to NOT
do so
>would result in a rate that the state would call "excessive". Therefore,
states
>supposedly would not allow rates (for long-tail lines) that do not reflect
>investment income. If the same state then does not allow discounting to
the
>loss payment date in the PDR calculation, it could force a monoline
long-tail
>line writer to hold a premium deficiency reserve. In other words, to
prevent a
>rate from being excessive, it must charge rates that the state then labels
>"deficient". Does anyone else see a contradiction here besides me?
>
>One incentive for disallowing discounting for Statutory PDRs is that
inadequate
>rates can be a leading indicator of a potential insolvency. If companies
with
>inadequate rates have to set up (or at least evaluate) premium deficiencies
>imbedded in the UPR, could it lead to timely corrective action.
>
>My experience in the calculation of PDRs is that it is not a trivial
exercise.
>It requires detailed accounting info (i.e. deeper than the IEE and annual
>statement), investigation into unusual expense values on a calendar
year/line of
>business level of detail, knowledge of the company's
projected/planned/budgetted
>future loss ratio, knowledge of line of business shifts, knowledge of how
>various direct and reinsurance programs affect written premium, adjustments
to
>the held UPR for future audits and reinsurance adjustments (e.g. possible
>reinstatement premiums), knowledge of how the various contingent commission
>plans work (WP incentive programs?, or responsive to calendar year loss
ratio?),
>etc. etc..
>
>Given infinite (and unbiased) resources at zero cost, a full detailed
>calculation of a PDR probably would reduce the chance of at least a few
future
>insolvencies. But resources are not infinite, they do cost, and the
resulting
>estimates are sometimes biased. Therefore, in order to possibly reduce
>insolvencies, does it make sense to require all companies to reallocate
>resources to do a full PDR calculation? Are the companies with inadequate
rates
>for which these rates raise an insolvency risk the type of company to
recognize
>the problem in advance, or are they more likely not to see (or acknowledge)
the
>problem until its almost too late? Are companies that would recognize
>inadequate rates in a PDR on a timely basis the type more likely to correct
the
>problem before it becomes an insolvency issue, hence the existance of a PDR
>would not be a factor?
>
>Given the above, I would recommend that the PDR calculation be consistent
with
>premium rate requirements, that if there is any concern with anticipation
of
>future investment income, that this be addressed in the RBC calculation (it
may
>already be), and that the PDR calculation be designed to be a simple
screen,
>calculated as simply and inexpensively as possible. If regulators feel
that a
>certain company faces extra risks not covered by this simple screen, that
they
>require additional reporting from that company, and not burden an entire
>industry.
>
>P.S. As to future catastrophes, the PDR covers "anticipated" losses, not
actual
>losses. The PDR calculation at 6/30 would cover the expected level of cats
for
>the next month, from business currently on the books, not the actual cats.
If a
>storm hits between 6/30 and the filing of 6/30 financials, those storm
losses
>are a subsequent events issue, not a PDR issue.
>
>
>
>
>
>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.
>
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.