The US Congress is due to weigh up the success or failure of the US Terrorism Risk Insurance Act with a view to possibly extending it Joseph Lebens and Bruce Hockman discuss the implications.

Sometime between now and next summer, the US Congress will consider whether to extend the Terrorism Risk Insurance Act (TRIA), passed in 2002, whose provisions for supporting terrorism insurance with what is effectively federal government reinsurance come to an end on 31 December 2005. Legislation to extend the law has already been introduced, with the support of many members of the re/insurance industry. The legislation has been opposed by 'consumer' advocacy organisations, public policy activists, and even some members of the industry.

The debate over the extension of TRIA raises three major questions for re/insurers, both as they become involved in this public policy issue and as they consider how to manage their businesses under the cloud of terrorism:

- whether TRIA has been a success or failure;

- whether TRIA represents 'good' public policy; and

- what the implications are for insurers and their customers if the Act is or is not extended.

Success or failure?

Judging whether TRIA has been a success or failure depends to a large extent on what one believes were the goals and intentions of the Act.

Its provisions are certainly clear enough: commercial lines insurers were required to offer commercial customers coverage for acts of terrorism during the life of the Act. In return for this mandated coverage, the federal government pledged to cover 90% of specifically designated terrorism-related insurance losses up to $100bn a year, above a 'deductible' for individual companies that increased from 7% of premium in the first year of the Act to 15% during the last.

If the primary goal of TRIA was to remove the obstacle to future economic development in areas at risk of terrorism attacks caused by the lack of insurance protection, then the Act has worked. It enabled coverage to be made available for commercial risks while insurers, reinsurers, and their ultimate clients dealt with an event that no-one thought would happen.

The Act brought stability to segments of the industry and the economy when stability was most needed.

From another standpoint however, the success of TRIA may be more problematic.

In the view of some observers, one hope was that, while re/insurers had a promise from the government in the short term for assistance with covering potentially catastrophic losses from a subsequent act or acts of terrorism, the industry itself would be able to develop a solution, or solutions to assessing and managing the risk of loss from terrorism, and therefore be willing and able to offer coverage without a federal backstop. That has yet to happen in any large-scale way.

There are a number of reasons why private insurers believe they are not, three years after 9/11, in any better a position to offer terrorism coverage on their own without federal assistance. First, insurers cannot assess the risk of future acts of terrorism in the same way that they can assess the risks of other kinds of disasters. Acts of terrorism are human-made, depending upon the minds of terrorists, the agenda of their organisations, and other unknowable and unpredictable factors. Therefore, many observers believe terrorism is not an insurable risk.

Second, the losses resulting from acts of terrorism spread across nearly all lines of business. Terrorism wipes out the usual assumption of disaggregation of effects of a single event. If the next terrorist act is nuclear, chemical, or biological, the costs to insurers could double or triple the costs of 9/11 and, in fact, exceed the total current capacity of the industry.

Third, how long it might take for re/insurers to re-capitalise after such an event or events is also unknown and unpredictable. It depends on the collateral activities, such as the actions of the stock market, that would or could follow from another terrorist assault, which are equally unpredictable.

Currently, nearly all private re/insurers believe they do not have the means to offer terrorism coverage. Some insurance industry leaders have thought that it might be possible to create industry 'pools' for the lines of business perceived to be most impacted (such as workers compensation or healthcare), so that catastrophic losses could be shared among participating insurers.

A feasibility study conducted by the Tillinghast business of Towers Perrin at the request of leading workers compensation insurers indicates that designing an equitable means of sharing risks and costs would be extraordinarily complex. Further, insurers would need time to build a pool of sufficient financial size to protect against catastrophic losses that modellers believe are possible if not probable. Whether there would be enough time is another unknowable and unpredictable factor.

Good public policy?

Much of the public debate about TRIA will hinge on the Act's wisdom.

Some of the arguments for TRIA as good public policy are essentially the same as when the Act was originally passed. If the state and/or federal government has an interest in mandating terrorism insurance coverage, then it has an equal interest and responsibility in seeing that such a mandate is effectively carried out.

If TRIA goes away, it is not at all clear that the mandate to the industry will go away. That mandate will certainly come from state governments through their insurance departments, especially in the highest risk states.

All the credible evidence suggests that private insurers and their reinsurers could not carry out that mandate without putting the commercial insurance industry itself at risk.

It has been suggested that a precedent for this kind of federal involvement in re/insurance already exists. Federally provided flood damage insurance provides useful analogies to TRIA-provided terrorism reinsurance. Both floods and terrorist acts are extremely difficult to predict and can create widespread damage affecting multiple parties and, potentially, multiple kinds of risks. Insurers decided some time ago that because major floods presented an 'uninsurable risk' they could not be the primary provider of coverage for flood damage. In 1968 the federal government created the National Flood Insurance Program to make flood coverage available for specified liabilities in specified locations. This programme has been accepted as sound public policy for nearly 40 years.

Some opponents of extending TRIA argue, however, that federal flood assistance should not be an accurate analogy to terrorism insurance; rather, federal disaster relief is. In the event of a natural disaster, the federal government provides disaster aid to victims that supplement other kinds of assistance, including insurance.

This sort of federal aid is not an insurance programme, nor does it involve the government in the private insurance market. The private market operates to serve the interests of its policy holders and shareholders. The federal government serves the interests of its citizens. Those who argue that TRIA is bad public policy suggest that acts of terrorism should be treated as any other large-scale disaster risk. This public policy position would clearly not satisfy business and business lenders in high risk areas.


If TRIA is extended under current proposals, we can anticipate something like the current situation continuing, at least for a time. That said, insurers would be required to retain greater amounts of loss before federal coverage would be available. The cost of such additional risk would have to be passed on to policyholders, with such resultant costs further impacting buying decisions.

The extension could provide further opportunity for the re/insurance industry to more fully explore alternatives, such as the workers compensation pooling mechanism, but there is no guarantee that the nature of the risk would allow for any ultimate response without some level of state and or federal involvement.

If TRIA goes away, either now or after a one-time extension, but state level mandates to offer terrorism insurance don't, then we have the situation of insurers being commanded to offer a product that puts their business at risk, without the availability of reinsurance protection to sufficiently safeguard their capital position. Conceivably, insurers could decide to simply withdraw from markets where they are faced with that choice. Insurers who did not or could not make that choice would be depending on the goodwill of the federal government to provide them whatever assistance they needed if an event did take place.

If TRIA and the mandates go away, a 'normal' private market will operate.

Re/insurers will assess the risks, exclude terrorism coverage in some places, offer it at a prudent price in others, and reallocate their capital in a way that makes the most sense to the totality of their customer, policyholder and shareholder base. In all likelihood, the number of re/insurers willing to operate in this market would be very limited, with market established prices tied to geographic location, and coverage aggregations carefully guarded.

What to do now

Insurers need to take steps now to better manage the major risk related to terrorism - employee or policyholder aggregation. In many jurisdictions, that risk may be just as great, or even greater, for catastrophic loss that results from natural hazards. Understanding the potential for such a loss falls under the definition of critical underwriting.

For those insurers who have not yet begun to collect data regarding employee per insured location, it is time to do so. Mapping, such as that used for years in property catastrophe underwriting, can be a useful tool, but accurate data is critical. Modelling firms continue to improve their capabilities in providing yet further assistance in evaluating potential loss scenarios, once the data is secured. Even if TRIA is extended, the likely increase in company retentions and co-participations require this level of underwriting discipline in protecting an insurer's balance sheet.

- Joseph Lebens is a consultant with Towers Perrin in Hartford, specialising in capital management projects and outstanding liability analyses. Bruce Hockman is a principal of Towers Perrin and senior vice president and workers compensation practice leader for the firm's reinsurance business.