Q: It sounds like there has been some resistance in Congress to including group life coverage in any extension of TRIA. Why is it important that group life be included?

A: Since the attacks of Sept. 11, the catastrophe reinsurance that companies use to protect themselves against large-scale losses from terrorism and other events has become much scarcer and more restrictive and expensive. However, group life companies at this point have not passed on any dramatic cost increases to their customers, and for the most part have not made their offerings significantly more restrictive or less available. The lack of impact on consumers has led members of Congress to think that there has been no disruption in the market, but in fact what has happened is that the very efficient scheme of risk spreading that was in place through reinsurance prior to the 9/11 attacks is gone. There is a much, much greater risk to specific companies now if another attack occurs. So I think it is short-sighted not to include group life in some sort of federal backstop.

Q: Why are group life policies especially susceptible to terrorism risk?

A: There are three reasons why that’s the case. First of all, the risk in group life insurance is heavily concentrated. When an insurance company writes a group insurance policy, it is insuring a bunch of employees who work for the same employer. Many times those employees work in the same building or in sites very close to each other, which means that there is the potential for a single catastrophic event to result in a large number of deaths from that one given population. With individual life insurance, the risk is more spread out; the individual policyholders live all over the country and a single company would not face such a great concentration of risk.

Group life has another complication when it comes to terrorism. With life insurance, the entire amount insured has to be paid out immediately, and that puts tremendous financial pressure on the insurer in the event of large losses. Compare that to something like disability insurance, where the payments only begin after three months or six months and are spread out over a long period of time.

Finally, with group life insurance, the premium that is collected is quite low relative to the company’s total risk, since the mortality risk is extremely low. Compare this to the medical market, where companies collect vast sums of premium and much of the premium is paid out in claims. There is a lot more cash flow coming in and out of the medical market in a given year, which can help if there’s a need to absorb a larger payout. This is not true of group life.

Q: Is there reduced reinsurance capacity for group life since the Sept. 11 attacks?

A: Yes, there is. There are a couple of different forms of reinsurance that group life carriers buy. One involves excess reinsurance on a given life; it helps carriers deal with a particularly large face amount on a single insured person (e.g., for any claim over a half-million dollars, the reinsurer pays the excess amount). That reinsurance is still available; there really hasn’t been a material change in that market. The second form of reinsurance, which has changed considerably, is called “catastrophe insurance.” Catastrophe reinsurance reimburses an insurance company for a large level of losses that are the result of a single event. Now, that form of reinsurance traditionally has been very widely used by group life carriers; in fact, I think nearly every direct writer had some form of catastrophe reinsurance prior to the Sept. 11 attacks. Since then, catastrophe reinsurance has become less widely available and much more expensive and restrictive, with deductibles growing from perhaps a few million dollars to up to $50 or $100 million. The average rates for catastrophe reinsurance have gone up at least five to 10 times, and many catastrophe reinsurance policies do not even cover terrorism losses. Many companies feel like that coverage no longer provides meaningful protection to them, and as a result have chosen not to buy this coverage or to buy a reduced amount or an amount that does not cover terrorism losses. That’s a profound change in the group life market.

Q: Why can’t we just use catastrophe bonds instead of TRIA as terrorism reinsurance? Why can’t we use risk pooling? Why do we need to do this federally?

A: That’s an interesting question. You mentioned two options: catastrophe bonds and risk pooling. Let me talk about those separately.

Catastrophe bonds involve transferring the risk of losses from a catastrophe to investors who are outside the insurance market in the form of a bond that’s issued by the insurer or reinsurer. Catastrophe bonds have worked pretty well for some coverages—such as property damage due to natural disasters because there is an abundance of data about, say, hurricanes that can be used to model the risk. With terrorism losses, though, the losses are much less frequent and much more volatile, and I don’t think anyone has a good understanding of the potential frequency or severity of those losses. So a catastrophe bond for terrorism losses is not really a very attractive investment. If the insurance industry itself has trouble managing that risk, imagine how difficult it is for someone outside of the industry to feel confident in that kind of investment.

Risk pooling can, in fact, be very effective in spreading the risk of loss among insurance companies, and there has been at least one instance of using a risk pool to backstop terrorism losses. I mentioned how group life is very susceptible to a concentration of risk, meaning that a single event can have a disproportionate effect on one company relative to others. Risk pooling can help spread that out. But there are some limits to risk pooling. The capacity of a given risk pool depends on how many companies participate and how large those companies are. There is only a certain-sized loss that can be absorbed through a risk pool, even with efficient spreading, before the capacity of its members is simply used up. So I think the primary issue with risk pooling is the total capacity that’s available. It’s useful for dealing with small and medium-sized events, but not the largest events. So even if you use risk pooling to help manage terrorism risk, you would probably still need some form of high-level backstop to ensure full protection.

Q: What other options, beyond risk pooling, are available to group life insurers in the absence of TRIA protection? Is self-insurance an option?

A: Self-insurance is kind of a default option; that’s a nice way of saying no reinsurance! The issue there is one of capacity. It’s easy to self-insure, but if the loss is big enough, you simply can’t cover it. Most companies, when they start to take a close look at the potential risks on group life, are uncomfortable with a full self-insurance program; only the very largest companies would be comfortable with self-insurance. Having said that, there are some companies that are self-insuring today because they haven’t found another solution that they think is workable, and they are unwilling to pay the higher cost for catastrophe reinsurance.

Many companies have simply tried to restructure their catastrophe reinsurance in such a way that they feel that they at least have some protection from the losses. For instance, many have decided that they will be willing to accept a very high deductible: $50 million, $100 million, $200 million. They might not be happy with having to pay that much in the event of a terrorist attack, but, based on their capital and financial strength, they could do it if they really had to. And then they’ve tried to make sure that once they’re above that deductible they would have a meaningful level of coverage without exclusions. Most of the companies that I’ve talked to are not completely happy with that solution, in terms of cost and in terms of level of protection, but at least they feel they have a little bit of protection.

Q: How do you respond to claims that including group life in TRIA would prevent a private market solution to the terrorist risk problem? Is there a private market solution?

A: I think that’s a simplistic viewpoint. The point that is generally missed when making that argument is that TRIA would have a very high attachment point, so that only the most expensive losses would be covered. If group life is included in the TRIA extension, the attachment point needs to remain high, so that TRIA only comes into play in the face of truly catastrophic losses, the type of losses that would threaten company solvency. There is significant room for private market solutions below that level. In fact, I think it’s likely that a federal program with a high attachment point would revitalize the private catastrophe reinsurance market because some companies would feel that they could go out and buy a meaningful layer of coverage below the attachment point. Because the backstop would be available for the really high-end losses, they would feel like the mid-level coverage is worth the investment.

Q: Given that, and despite the frequent claim that TRIA was never intended as a permanent solution, it sounds like you’re suggesting that a federal backstop with a high attachment rate is in fact part of the long-term permanent solution.

A: TRIA in its original form was not intended as a permanent solution. It was intended to be a stopgap or a temporary solution, but a couple of things have been discovered. First of all, the ongoing impact of the attacks on the property and casualty market was more severe than people expected. Second, the remedy provided by TRIA proved to be more effective than people thought. I think TRIA in its initial form had a few drawbacks, so in this regard a permanent solution wouldn’t necessarily have to look the same as the original version of TRIA.

I think it’s important to address perceptions that TRIA is some sort of federal handout or subsidy to the insurance industry. That certainly wasn’t the intention and I don’t believe that to be the case. TRIA is a form of federal reinsurance for the highest level of losses, where government payments are recovered in arrears via policyholder surcharges. It is a loss-spreading mechanism, not a handout. In considering a permanent solution, I believe that something more akin to a traditional reinsurance scheme, with clearly identified premiums paid by the participating insurers, would probably be attractive to companies if it provided meaningful protection as part of a permanent solution.

Q: If there were to be an attack that tapped reinsurance capacity, how would that affect the price of life insurance at the consumer level?

A: It’s hard to say, but I think there could potentially be a significant impact on both the cost and the availability of group life insurance to consumers. We didn’t see much impact on price and availability for consumers as a result of the Sept. 11 attacks, but that was because there was such an efficient network of catastrophe reinsurance in place. The industry spread the losses from that attack remarkably well. So companies didn’t feel the immediate need to pass on costs to the consumers; and I think they hesitated to do that because they don’t want to exit the market, they don’t want to have a major business disruption. And now they are crossing their fingers. Catastrophe reinsurance is to a large extent unavailable or restrictive and the group writers are hoping for the best, but if there was another attack, it would be much more financially painful for the insurance companies. If there is another attack, we may see a number of companies decide to exit that business.

Q: Are there other insurance products currently outside the scope of TRIA that could potentially require some sort of federal reinsurance if there was a major terrorist attack? Are there other risks that TRIA should cover that are not currently on the table?

A: In the world of life and health insurance, group life faces the highest risk from terrorist attacks, for the reasons I’ve already discussed: the combination of the risk concentration, the immediate payments, and the low premium rates.

Still, it is worth considering the effect that an attack might have on other lines of insurance. Disability insurance has a lot of overlap with workers’ compensation, and workers’ comp is covered by TRIA, so to the extent that terrorism losses occurred at the workplace, workers’ compensation would pick up much of the costs. And benefits from disability are paid out gradually over time, so there is not the immediate cash impact on the insurance companies, even though the total cost could be quite high. This allows for more time in the aftermath of an attack to develop a financial solution.

With medical coverage, there would probably be more emphasis placed on providing immediate healthcare rather than worrying at the outset how the insurance claims are going to work. Also, similar to group disability, to the extent that losses occurred at the workplace, workers’ compensation would cover most of the cost.

There is potential for significant individual life insurance losses from terrorist attacks and we saw significant life losses from the Sept. 11 attacks, but there isn’t the same risk concentration for individual life. Losses would generally be spread more evenly across the insurance industry. These losses might be difficult to absorb, but they probably would not pose as great a threat to financial strength and solvency. So while I think all of these lines can be affected by terrorism losses, group life insurance has the greatest need for a federal backstop.

DAN SKWIRE, is a principal and consulting actuary in the Portland, Maine office. He is part of an industry-wide effort to support the inclusion of group life in the extension of the Terrorism Risk Insurance Act (TRIA), and has written several white papers on the topic.