Property Catastrophe Risk and Capital Markets Risk Transfer
August 2008
Will we ever be able to transfer nonproperty,
noncatastrophic risk into the capital markets?
by Donald
J. Riggin
Albert Risk
Management Consultants
Before we dive into the topic, it's important that we define a few terms.
Nonproperty risk is of course casualty risk. It also may be characterized
as frequency risk, which is the opposite
of catastrophic risk. Frequency risk is best exemplified by workers compensation.
Given enough historical loss data, actuaries can predict, with varying degrees
of confidence, future loss activity. Insurance premiums are based on these predictions,
which underlay a large portion of the insurance industry. Predictability begets
premiums, which begets (sometimes) a little underwriting profit and (always)
investment income. Property insurance is predicated on individual risk characteristics,
(construction, occupancy, protection, and exposures—COPE), and usually a catastrophe
loss model or two.
This brings us to property catastrophe risk and why it lends itself to capital
markets risk transfer. Cat risk insurers and reinsurers do not rely on COPE
characteristics to set rates; they use the underlying pricing and the aforementioned
damage models to establish a probability of loss for a specific geographic region
and/or specific perils. These data are used to set reinsurance rates as well
as insurance-linked security (cat bonds, etc.) pricing.
The primary reason why cat bonds work is that the data used to set pricing
is relatively transparent. I say relatively
transparent because the average investor certainly does not understand the complexities
of the damage models used to determine loss probabilities. They do, however,
understand that an A rating from Standard and Poor's ensures that at some level,
the creditworthiness of the bond has been vetted. Cat bonds usually exist in
multiple tranches (French for "layer"). The tranches are similar to layers of
reinsurance. The lowest tranche, the one closest to a loss, often does not earn
an investment-grade rating. The excess tranches, however, are usually rated
investment grade.
Loss Triggers
There are three commonly used loss triggers in insurance-linked securities:
indemnity, index, and parametric. The indemnity trigger is identical to that
which is found in every insurance and reinsurance contract—if a loss is covered
under the terms of the contract it will be paid. An index is a single value
that represents an aggregation of individual values. Those individual values,
however, must be highly correlated, meaning that their values must all be susceptible
to the same unpredictable forces. The best example of an index is the Dow—30
blue-chip stocks combined into a single value that fluctuates daily based on
a variety of forces too numerous to mention here.
A cat bond's index is an aggregation of defined property losses within a
specific period of time. A parametric trigger uses the magnitude of the event,
such as a 7.5 magnitude earthquake (on the Richter scale).
Exchange-Traded Catastrophe Futures
Over the last couple of years, three new exchange-traded products for catastrophe
risk have hit the scene. Identical to cat bonds, these futures contracts are
designed to provide funding for catastrophic loss due to natural causes except
earthquake. Unlike cat bonds, which are technically structured financial products,
these products are true derivatives in that they are "contracts for difference."
This means that they are traded anonymously and the traders can make or lose
money based only on the price movements of the contracts and need not own the
underlying asset, or in this case, the index. These products, however, only
cover property damage losses above a $25 million loss threshold, as determined
by their particular index.
Summary
So will we ever see frequency loss exposures traded in the capital markets?
Two major obstacles are the lack of information transparency and the lack of
information itself. Markets only work when investors are sure that the rules
for determining settlement are clear and applicable to every participant, and
there is a guarantor for each trade—the exchange platform. Moreover, catastrophe
indexes can only exist because of organizations like Property Claims Services®
(PCS®) that survey every natural disaster and assign a potential
damage value. We have no way to document every general liability loss that occurs.
Another barrier to frequency losses being traded in the capital markets is
the very nature of the loss exposures. While all insurable risks, frequency
and catastrophe, create loss events, only catastrophe risk produces an event
large enough to warrant an index based on a single, identifiable, event. Perhaps
the National Council on Compensation Insurance (NCCI) could produce an index
based on its knowledge of workers compensation losses, but where would such
an index begin and end? Which industry classes would be included? These questions
may not be insurmountable, however.
The biggest lack-of-information hurdle is the
cause of loss. We know what happens in a
natural catastrophe—a hurricane flattens real property. The problem with casualty
losses is that the causes of loss are manifold. A tradable insurance-linked
security needs a definitive peril such as windstorm. What peril would we assign
to workers compensation bodily injury? Okay, then how would we deal with the
multiyear loss development patterns for casualty losses in a tradable security?
Property damage losses can be established fairly quickly, allowing for index
settlement, but workers compensation loss values are moving targets.
We certainly may conclude that frequency-driven loss exposures may be too
variegated and complex to be boiled down into a tidy futures contract. We have
no access to reliable loss data, multiple causes of loss that magnify the loss
possibilities, and multiyear loss payout periods. While I think that while these
barriers could be surmountable some day, for the time being, insurance and reinsurance
underwriter's jobs are not in danger.
Opinions expressed in Expert Commentary articles are those of the author and are
not necessarily held by the author's employer or IRMI. Expert Commentary articles
and other IRMI Online content do not purport to provide legal, accounting, or other
professional advice or opinion. If such advice is needed, consult with your attorney,
accountant, or other qualified adviser.