Time, Money, and a Possible New "Maginot Line"?

June 2007

For the first few months of this year, many insurance industry observers believed that the 2-year extension of the Terrorism Risk Insurance Act (TRIA) would be reauthorized long before its scheduled year-end expiration. It seemed like a sure thing.

by James Macdonald1
Navigant Consulting

Trade journals reported that a new bill was possible by the end of April. In February and March, House and Senate hearings focused not on "whether" but rather on "how" the program should be extended. But despite the momentum, nothing happened.

The popular thinking now is that there will be an extension of the federal terrorism insurance backstop. However, much like the late December 2005 enactment of Extension Act, we may not know the details until the waning hours of 2007.

Part 1 of this article considered the changing nature of the terrorism threat. In this, the second installment, I explain the reasons why some form of an extension appears to be certain. I then consider several important issues related to either timing or money, and some creative options that need to be considered as potential building blocks to a constructive new approach.

What Explains the Optimism?

Understanding the reasons for the current optimism provides insight into the specific reforms that may be in the offing. There are at least four reasons for the confidence expressed by many stakeholders. The three most recognized are the following.

  1. The surprising November elections putting key Democrats in favor of the program in charge of important Congressional Committees (notably Dodd, Frank, and Kanjorski);
  2. The general belief that the threat of another large-scale terrorism attack inside the United States is at least as real today as it was in late 2002 when the terrorism insurance program was created; and
  3. The consensus that the federal backstop program has been highly successful in achieving almost all of its original goals (i.e., assuring an affordable and available market for terrorism insurance, giving insurers time to rebuild their capital bases in the wake of September 11, and also providing time for insurers and reinsurers to develop new approaches to measure and price this risk.) (See Marsh MarketWatch™ 2006.) Simply stated: "Why mess with success?"

On a more specific but less recognized level, the foundation for common ground between all parties may have been cemented by important shared conclusions in two separate September 2006 reports from the Presidential Working Group (PWG) and the Government Accountability Office (GAO). Both reports noted that a terrorist attack involving a nuclear, biological, chemical, or radiological (NBCR) device would be largely uninsured today. One reason for this deficiency is that the TRIA law subordinates the "make available" terrorism insurance requirement to the normal policy terms and conditions that otherwise apply. As a result, property or liability insurance losses from a weapon of mass destruction could largely not be covered (assuming the standard nuclear, contamination, pollution, war, or some other policy limitation is deemed to be applicable). Another reason is that surveys show many buyers are not buying the insurance or investing in loss mitigation because they do not believe they are at risk.

The reports also noted:

  • The inherent limitations of new catastrophe models at estimating possible losses from "worst-case" attack scenarios (especially considering that most information about specific terrorism threats is classified, PWG page 28, GAO page 8).
  • The unique challenge that terrorism risk presents to the nation's critically important workers compensation insurance line, given the fact that no state gives insurers the ability to limit or exclude terrorism (PWG page 8, GAO pages 2 & 15).
  • The limited capacity from the private market reinsurance market, estimated by the Reinsurance Association of America at not more than $8 billion in 2006 for a conventional weapons attack, and another $900 million to $1.6 billion for an NBCR attack (PWG page 26 and GAO page 8).

Importantly, both reports predict little to no probable future interest in the traditional reinsurance or alternative capital markets in insuring NBCR (with the GAO saying the development of private sector NBCR capacity is "highly unlikely in the foreseeable future" and the PWG report concluding that "there may be little potential for future market development").

By essentially concluding that, despite the many successes of the program, "the sum of all fears" remains largely uninsured, the PWG and GAO have defined what could be specific minimum necessary public policy goals for this year's legislative reform. A full discussion of all the issues being debated exceeds the limits of this article. But let's consider two important obstacles which, in short, focus on time and money.

Time: Does TRIA "Time Out" the Development of New Capacity?

The initial TRIA legislation was explicitly defined as a "temporary" measure to address economic disruptions in the wake of the September 11 attacks. This is a favorite refrain of the program's most fervent opponents. Once again, their strong desire is to keep any extension to a minimum. Their basic rationale for another short extension is that, because there is no premium charged for the federal portion of the program, the program "crowds out" the development of new terrorism insurance capacity from traditional or alternative capital market sources. Legislators supporting TRIA have stated that they favor an extension of at least 8 to 10 years, and one has argued for an unlimited extension subject to periodic reviews.

If we take a closer look, TRIA's critics are missing a key point with their focus on the "crowd out" argument. What it really does is "time out" the creation of new approaches.

For example, from early 2003 through the first quarter of 2004, 15 large and small insurers cosponsored an inquiry into the feasibility of forming a voluntary terrorism reinsurance pool for workers compensation. I was personally involved with the effort as an employee of one of the larger insurance companies. The exhaustive collaboration did not proceed to implementation, but it did produce an informative and frequently cited report detailing the key findings and hurdles to forming a voluntary pool. (The Towers Perrin report is still available for download.)

In my opinion, the uncertain duration of the initial federal program was one of the major impediments to the possible creation of the pool. If, for example, we had known that the federal TRIA backstop would be in place for another 5 years or more, there is a real possibility that this pool may have been formed by at least some of the participants.

Alternative capital market solutions also take considerable time and effort to launch. It can take well more than a year to go from concept, to approaching investors, to the creation of new risk-taking legal entities, to the final implementation of operations.

Because two installments of the federal terrorism insurance program have been certain only for an initial 3-year period, and then a 2-year increment, there simply has not been enough time or certainty to test the possibility of developing creative, new capacity solutions. Any extension of less than 5 years will almost certainly continue to render hypothetical the feasibility of new pooling arrangements or capital market solutions for terrorism risk.

Money: Is TRIA Really a Benefit to Federal Taxpayers?

Critics of the terrorism insurance program argue passionately that TRIA is an unjustified "subsidy" of an industry that is enjoying historical profits. They accurately point out that commercial insurers have more than recovered from losses on September 11 (with domestic property and casualty insurer capital and surplus approaching $500 billion at year end 2006). To limit the post-loss funding risk they believe the program presents to federal taxpayers, these critics argue that the required insurer retentions should be progressively increased.

What no one seems to have considered is the possibility that TRIA has actually been a benefit to federal taxpayers. The PWG report came close to recognizing this contrarian point when it estimates that after-tax TRIA terrorism premiums paid to insurers "increased policyholder surplus by a total of approximately $1.7 billion during 2002 through 2004" (page 33). Although the report projects that additional capacity for terrorism risk could grow as this contribution to after-tax surplus develops, there is no attempt to quantify the incrementally larger investment income growth that would be possible if, like many European terrorism insurance programs, American insurers were allowed to set aside pre-event and pre-tax terrorism loss reserves.

It is also remarkable that the PWG report does not recognize the significant real-world benefit to federal revenues from the federal taxes that insurers have been paying on these premiums. Citing a number of sources, the PWG authors estimate the total collected terrorism insurance premium to be between $3 billion and $8 billion through year-end 2005 (page 33). If we add another $2 billion to $4 billion for 2006 and 2007, and assume the federal corporate tax rate of 35 percent, federal tax coffers will be the cumulative beneficiaries of between $2 billion and $4 billion in tax revenues by the end of this year.

As Senator Everett Dirksen might have said,2 this is "real money." It is at least as worthy of public policy-making consideration as the theoretical macroeconomic costs and benefits of the program argued by both sides of the ongoing debate.

A second "money" issue very much in dispute is whether a premium should be charged for the federal backstop protection. Insurers make a strong case that the "free reinsurance" is best understood as a "trade-off" in return for the federal requirement that they provide the terrorism insurance (despite the fact that many believe that this risk, like war risk, is inherently uninsurable at least for a major attack). If private capital is to remain in the market, and the TRIA "make available" requirement is continued, they argue cogently that this "trade-off" is essential and fair.

Opponents of the current approach offer some compelling rebuttal arguments for the introduction of some federal premium charge. If the pricing is set at a level above a conservative actuarial estimate of the needed price, it would set a new incentive for the development of new capacity. Charging a premium for the federal protection would also enable a federally managed fund to accumulate, gradually limiting pos-loss taxpayer risk.

Establishing federal premium charges could also enable some creative solutions to major issues that are now simply not feasible. For example, if the federal program set different prices for optional retentions or triggers, small insurance companies could be allowed to "buy down" the minimum coverage trigger from the current $100 million to some lower amount more reflective of their capital bases. When we consider that many hundreds of domestic commercial insurers have well less than a total of $500 million in capital supporting all their obligations, a breakthrough of this nature could be critical to preserving the competitive balance in the commercial marketplace.

Conclusion: A New "Maginot Line" Would (Once Again) Be a Bad Idea

In recent weeks, there has been some speculation that the extension of TRIA may only cover workers compensation (for conventional attacks and NBCR) and only NBCR losses to the other commercial lines still included in the law. Although it is easy to understand why this thinking may be popular, it would be a major error. If we agree that one of the main goals of private sector terrorism insurance is to optimize economic resiliency after an attack, then limiting the triggers to that responsiveness to only nuclear, biological, chemical, or radiological attacks seems ill advised. Daunting as an NBCR attack would be, the simple fact is that terrorism experts believe that the probability of a major attack using conventional weapons is much greater than an attack using a NBCR agent or device.

For the insurance program to be effective, it needs to be as flexible as the threat itself.

In many ways, this idea is reminiscent of the "Maginot Line" built by the French after World War I to block the anticipated route of a future generation of German or Italian forces. If the French learned anything as they watched Nazi tanks sweep around their defenses in World War II, it was that static, fixed defenses against highly mobile enemies are a bad idea. In a similar vein, a federal program limited to NBCR could simply increase the likelihood that the next major attack—like the attacks on September 11—will not involve these agents or devices.

Part 3 of this series considers the two additional issues reportedly presenting major hurdles to policymakers: coverage and insurer retentions.


1The opinions in this article are solely the personal views of the author and do not express or imply the position of Navigant Consulting, Inc.

2The late Senator Everett Dirksen is generally considered the originator of the now popular phrase: "A billion here, a billion there, and pretty soon you're talking real money." See: http://www.dirksencenter.org/print_emd_billionhere.htm.


Opinions expressed in Expert Commentary articles are those of the author and are not necessarily held by the author’s employer or IRMI. This article does 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.