Were the men who set off bombs in the London
Underground domestic or foreign terrorists?
For British purposes, it may not matter. In a
similar situation, however, American insurers would be forced to
ask: How does one classify citizens who carry out acts inspired
and encouraged by participants in an international conflict?
As most readers know, the U.S. federal
terrorism reinsurance program makes a distinction between “certified”
acts of terrorism carried out by foreign interests (covered under
the program) and “non-certified” acts of domestic terrorism
(not covered under the program).
That distinction, unique among all the
publicly sponsored terrorism insurance arrangements in
industrialized countries, demonstrates the distinct logic behind
the Terrorism Risk Insurance Act (TRIA) of 2002.
From its inception, TRIA embodied a view that
the U.S. government would limit its involvement in terrorism
insurance to providing a backstop for losses arising from war-like
attacks emanating from overseas.
If the Treasury Dept. had its way, application
of the program would be restricted even further.
In a letter accompanying the Treasury Dept.’s
June 30 report assessing the progress of TRIA, Treasury Secretary
John Snow stated that the Bush Administration would accept an
extension of TRIA beyond its scheduled “sunset” only under
certain conditions.
For TRIA to be extended, Snow wrote, the size
of an event triggering coverage would need to be increased from $5
million to $500 million in losses, the amount of deductibles and
co-payments would have to be increased, and certain lines of
insurance would have to be excluded from the program.
Under the original act, the federal government
committed to provide reinsurance liquidity--not a subsidy--for a
portion of losses arising from an event that caused $5 million or
more in commercial insurance losses, provided the event was
certified as an act of foreign terrorism by specified federal
officials.
In testimony before a U.S. House committee
following the release of the report, Secretary Snow suggested that
the $500 million threshold could be an annual aggregate, a
critical shift in tone and substance from the adminstration’s
initial recommendation.
Whether an “event threshold” applies per
event or per year could determine whether the federal government
strictly limits its role to covering large events, or whether it
provides support through a sustained campaign of smaller attacks,
the most common form of terrorist threat.
The debate over whether to extend the federal
terrorism reinsurance backstop, in what form, and for how long,
was just beginning as this publication went to press.
So, at this crossroads, going on four years
since Sept. 11th, what have we learned that will help us determine
the appropriate private and public roles in insuring terrorism
losses?
After Sept. 11, Americans often heard how they
were no longer protected by the oceans that divided them from the
rest of the world. That’s not exactly true.
Since 2001, terrorists have continued to
strike--in Bali, Madrid, London, and elsewhere--but no terrorist
attacks of note have occurred on American soil.
The campaign against the American presence in
Iraq demonstrates that Islamic militants yearn to strike at
Americans. However, they are apparently unable to get
operatives into the United States or mobilize
those already here for a sustained campaign of attacks, as they
can in Europe.
Given that, U.S. Treasury officials believed,
at least initially, that it was reasonable to limit federal
terrorism reinsurance to large, infrequent events.
Also, since there have been no notable acts of
domestic terrorism since the Oklahoma City bombing, the Treasury
Dept. has not offered to extend federal reinsurance coverage to
attacks by domestic terrorists.
Even if the threat to our own shores is
contained, however, insurer groups argue that a permanent federal
role in terrorism reinsurance is warranted because the threat is
affected by federal policy. Depending on one’s point of view,
federal policy can be either too permissive or too provocative in
dealing with the sources of terrorism.
“Since the public sector controls virtually
all of the ways and means of influencing, deterring, and
destroying terrorists, how is it consistent to turn to the private
sector and demand protection from the financial consequences of
its own actions?,” wrote Robert Hartwig, senior vice president
and chief economist of the Insurance Information Institute, in a
September 2004 report.
The RAND Corporation would add that domestic
terrorism merits consideration in a federal terrorism reinsurance
program.
In a 2005 report, RAND researchers say that
anti-globalization sentiment brings together radical anarchists,
environmentalists, and animal rights extremists, and even attracts
heavily-armed right wing militia groups, one of which was found in
2002 to have a storehouse of weapons, including 800 grams of pure
cyanide.
In light of the threat from domestic
extremists, RAND reports that “a significant gap exists in the
current system for managing and mitigating the financial risk of
terrorist attacks in the United States.”
Before September 2001, many underwriters were
unaware that businesses in a single skyscraper could have
different street addresses, even different ZIP Codes, and that a
carrier could unwittingly accumulate concentrated property
exposures.
No more. “Accumulation management” has
become a standard practice in property underwriting, made possible
by the use of geocoding and mapping software that allows carriers
to plot precisely the locations of their insured properties.
The ability to identify accumulations has been
an essential component of terrorism risk modeling, which the
Treasury Dept. cites as a “positive” development for the
creation of a private market for terrorism insurance.
“Terrorism risk modeling . . . has made
great strides in modeling the costs resulting from various types
of attacks [and] have made the magnitude of the risk more
understandable,” wrote researchers Glenn Hubbard and Bruce Deal
in a 2004 report for several insurance trade groups entitled “The
Economic Effects of Federal Participation in Terrorism Risk.”
Yet most observers, including the Treasury
Dept., acknowledge the shortcomings of terrorism modeling compared
to models developed to predict the frequency and severity of
losses due to natural catastrophes.
“Terrorism risk modeling allows potential
losses resulting from various attack scenarios to be quantified,
but falls short of increasing the predictability of future
attacks,” reads a recent report on terrorism risk insurance by
the Organization for Economic Cooperation and Development (OECD),
an inter-governmental association of industrialized countries.
“Response rates to the reinsurer survey were
too low to warrant the presentation of any quantitative results.”
So reads a frustrating statement in the Treasury report.
In any decision regarding the scope and
duration of the federal role in terrorism reinsurance, there is no
more important question than, “What will the reinsurers do?”
European-based reinsurers absorbed more than
half the losses from Sept. 11, preventing a tragedy from becoming
a crippling event for the U.S. economy. It was their determination
to exclude terrorism coverage after Sept. 11--as much as the event
itself--that forced U.S. carriers to seek exclusions of their own
and federal help to address the peril.
Will reinsurers come back if the federal
government retrenches on terrorism coverage?
Since Sept. 11, the role of foreign reinsurers,
primarily based in Bermuda and Europe, has grown to where they or
their American subsidiaries account for 80% of the U.S.
reinsurance market, yet most of the major trans-Atlantic
reinsurers have seen their surplus and ratings decline since 2001.
Given that, it is of little help to read
statements urging domestic carriers to “use” private
reinsurance to address terrorism risk. One can only buy
reinsurance if someone is willing to sell it in sufficient volume
at an acceptable price.
A similar admonition applies for catastrophe
bonds.
After years of being promoted as a means to
tap the capital markets, catastrophe bonds outstanding at the end
of 2004 amounted to $4.4 billion worldwide, less than 3% of the
$164 billion in reinsurance revenue for that year.
Only two catastrophe bonds have been written
to address potential losses from terrorism, and terrorism is only
one of many causes of loss addressed by those contracts.
There are several reasons why catastrophe
bonds are a tough sell to investors, but terrorism bonds have an
additional drawback in that terrorist attacks often trigger
declines in capital markets. Thus, an investor can lose principal
on a terrorism catastrophe bond at a time his other investments
lose value.
Under TRIA, insurers are required to offer
certified terrorism coverage with commercial policies, but
insureds are not required to accept it. Time will tell whether
that is a recipe for adverse selection.
For the time being, both the Treasury Dept.
and private sources agree that more firms have purchased coverage
since TRIA took effect.
Based on survey data, the Treasury Dept.
estimates that, in 2004, about 55% of commercial policyholders had
coverage for certified and/or non-certified terrorism losses,
almost a doubling of the percentage of insureds that had terrorism
coverage in 2002, before TRIA was enacted.
Marsh Inc. reports similar findings in its
annual Marketwatch studies of terrorism insurance. Marsh finds
that “take-up” rates (the percentages of firms that purchase
terrorism coverage) vary by size, region, and industry.
While more than half of firms with $500
million or more in insured values purchased certified terrorism
coverage in 2004, only a third of those with values under $100
million did so.
Also, while Boston led major cities with a 69%
take-up rate--followed by Washington, D.C., Chicago, Dallas and
New York-- Los Angeles and San Francisco had rates of only 39% and
37%, respectively, even though rates for coverage in those cities
was lower than in other major cities.
Within specific industrial sectors, Marsh
finds that financial institutions, real estate firms, and
health-care facilities had the highest overall take-up rates, each
exceeding 60%.
Perhaps the ultimate test of policyholder
support for federal terrorism reinsurance is the vigor of their
support for an extension of TRIA in some form.
“Frankly, it’s not us who are going to get
[an extension] passed,” said Joel Wood, senior vice president of
the Council of Insurance Agents and Brokers, in a recent
teleconference. “It’ll be the policyholder commitment that
gets it done.”
|
What have
we learned...
