TRIA: It's Time to Examine New Terrorism Approaches

Insurer Concerns

Insurers reported a range of concerns as respects impediments to the supply of terrorism insurance. The main issues reported by survey respondents included: the availability and cost of traditional reinsurance; the large net loss potential because of the current TRIA deductible and coinsurance requirements; the challenge of modeling probable loss scenarios (with frequency being the main hurdle); and the views of rating agencies regarding what constitutes acceptable concentrations of risk and adequate capital levels.

To address the concerns of insurers and insurance buyers, the GAO report offers comments on five options for policymakers to consider. Here is a summary of each:

1. Reset Insurer Deductibles following a Major Terrorist Attack: Under this legislative effort (proposed in 2007 by New York Senator Charles E. Schumer), a terrorist attack resulting in insured losses of $1 billion or more would result reduce a commercial insurer's deductible from 20 to only 5 percent of the prior year's direct earned premium. This after-the-fact approach appears to introduce too much uncertainty to an insurer's capital management challenge. As such, it is unlikely to get much support.

   There is surprisingly no mention of the changes that the 2005 House bill that was approved by a huge majority. This bill would have introduced with retentions subject to a "risk based" formula, e.g., workers compensation would require a lower retention than commercial auto since it exposes insurers to theoretically unlimited losses. The bill also would have reduced insurer retentions for nuclear, biological, chemical, or radiological (NBCR) losses to a fraction of the 20 percent normally required. It is not clear why the GAO decided to omit any comment on the seemingly more plausible options in the House bill.

2. Permit Insurers to Establish Pre-Loss Tax-Deductible Reserves: This option is often also mentioned as a solution to the supply of insurance for all forms of extreme events, including earthquakes and hurricanes. The basic idea is to take the same approach afforded to the private sector in many European nations by changing the US tax code to allow insurers to set aside of pretax loss reserves for future extreme events (be they man-made or natural).

   Proponents argue that this would reduce the cost of capital of insuring catastrophic losses and thereby improve affordable capacity in the commercial market. As the pre-event loss reserves accumulate, the risk to an insurer's capital adequacy would decrease. Critics counter that insurers already enjoy enough tax advantages. Because this decrease tax revenues, it is not likely to receive serious consideration in the near term, particularly in Congress returns to the "pay as you go" approach. The GAO also notes significant challenges in determining the size of reserves that should be permitted.

3. Form a Pool of Insurers To Underwrite Terrorism Risk: The GAO cites an innovative Aon proposal in 2006 for the creation of a new terrorism reinsurance pool funded by the private sector and designed to be able to finance two $40 billion losses. This pool would replace the federal government as the first tier of protection above required insurer retentions. In the event that a loss depleted the pool's funds, the overage would be funded either by some combination of bonds or a second layer of protection from the Treasury Department.

   The GAO mentions two examples of operating pools that provide coverage for extreme events: Florida's mandatory Hurricane Catastrophe Fund (FHCF) and the United Kingdom's voluntary Pool Reinsurance Company, Ltd. (Pool Re), as examples of pools that have been created to insure extreme events.

   The report cites numerous obstacles to forming a pool including the difficulty in determining the optimal size for a pool given the uncertainty inherent to estimates of the frequency and severity of terrorism attacks. In addition, the GAO mentions a key conclusion of a major 2004 report into the feasibility of creating a workers compensation pool, i.e., that a pool in itself does not expand the total amount of capacity that is available for the private sector to insure terrorism risk. The GAO fails to note, however, that this study assumed that pool participation would be in a voluntary basis and never considered the prospects of a mandatory approach.

4. Facilitate Issuance of Onshore Catastrophe Bonds: To create new risk-bearing capacity, the GAO considers the feasibility of encouraging capital market solutions through new forms of debt obligations. Like the loss reserve option, this would require the changes to the US tax code, and it would likely not be revenue neutral. One can also reasonably assume that rating agencies would be reluctant to give the investment grade ratings that many investors require, particularly in the aftermath of the subprime crisis. The GAO concludes that, even with changes in the tax code, a viable market for catastrophe bonds is unlikely to develop because of the lack of reliable data to form the basis for estimates of terrorism losses. Much like the first two options, this approach is unlikely to gain legislative traction at a time when reducing the deficit is an imperative concern.
5. Revise State Regulatory Requirements: Some states require insurers to pay for fire losses following a terrorist attack, even if the policy excludes loss from terrorism. These so-called Special Fire Policy (SFP) states arguably limit the capacity that private insurers are willing to commit. Several insurers also reported that certain state regulators limit capacity by preventing insurers from charging what they consider to be actuarially fair, risk-based prices. The GAO correctly notes that commercial pricing is not strictly regulated in high-risk areas of concern. Although the authors concede that the "fire following" issue may limit available capacity, they provide no final position, concluding instead that other factors, such as potential losses on high-value buildings, are of greater import.

Report Omissions

In general, the report is most notable for the options it ignores rather than the limited number of alternatives it explores. There are no clear recommendations or even preferred next steps for policy makers to consider. The slant of the authors appears to be almost entirely negative, with an almost singular focus on what won't work. For example, in the discussion of pools, we see no mention what is probably the best example of a successful program similar to Terrorism Risk Insurance Act (TRIA): the American Nuclear Insurers Pool. This successful public and private sector program was launched in 1957 in response to the Price Anderson Amendment to the Atomic Energy Act. Similar to TRIA, the goal was to assure the availability of affordable liability insurance for nuclear incidents in order to encourage the development of nuclear energy facilities. One would think this would serve as a better point of comparison than the hurricane fund or a private sector funded feasibility study that failed to produce any substantive results.

The most surprising (and disappointing) omission from the GAO report is its failure to explicitly address what many believe is TRIA's greatest failing, i.e., the lack of adequate or affordable insurance for terrorist attacks involving NBCR agents or devices. The 2006 PWG report (see PWG Report on the Future of Terrorism Insurance) and broker surveys of the property insurance market have consistently concluded that, despite the fact that Treasury guidance has been clear that TRIA would cover these losses (if the primary insurance applied), almost all commercial policies exclude these perils. The limited coverage that is available is generally considered far too expensive, as reported in recent reports to the PWG from Aon and CIAT. The sole exception is workers compensation, due to the fact that no state allows insurers to exclude or place sublimits on losses from terrorist attacks.

At minimum, one would have expected the GAO to consider some of the bold changes proposed in the revision to TRIA approved by the House in December 2005 (HR 4314, 109^th Congress, 1^st Session), or by the strong opposition by many conservatives to what they call the "subsidy" TRIA needlessly affords to highly profitable private insurers.

Other Proposals

The notable proposals that merit future consideration as we approach the final years of TRIPRA include:

• Eliminating the "subsidy" aspect of the program by requiring an actuarially sound premium for the federal reinsurance;
• Including group life insurance (especially given the significant losses incurred on 9/11 and continuing exposure to terrorist attacks); and
• Encouraging more capacity for NBCR losses by lowering the insurer deductible to less than 10 percent.

Conclusion

Since it was issued in 2008, the GAO report's report has, for very solid reasons, already become a distant memory. In all probability, it will be replaced by one or more improved studies more typical of the normal GAO work product. If the current intensely competitive commercial insurance market turns, or if we sustain another major attack, some fresh thinking may be needed to maintain a stable market. In this event, Aon's thoughtful 2006 pool proposal will likely receive renewed attention as the basis for a possible long-term solution.