Insuring Against Terror
The troubling post-911 insurance market.

By Scott E. Harrington, professor of insurance and finance at the University of South Carolina, and Tom Miller, director of health-policy studies at the Cato Institute.
November 5, 2001, 8:45 a.m.

The insurance, banking, construction, and real-estate industries are making the rounds on Capitol Hill to demand that the federal government return their balance sheets to what they looked like on September 10th.

The total cost of insurance claims from the September attacks is at least $35 billion. A good portion of that amount will fall on foreign reinsurers. The insurance industry remains reasonably well capitalized despite those unprecedented losses. Several entities have announced entry into the market for terrorist coverage.

Nevertheless, funding potentially large losses through private insurance markets presents significant problems in quantifying the risk and spreading the risk over time. It requires large amounts of capital. Even higher premiums are needed to cover the costs of holding that capital because they must include, at least for domestic insurers, corporate taxes on insurers’ investment income.

Reinsurers are reportedly refusing to offer coverage for terrorist claims when most contracts are due for renewal on January 1, or they will be willing to do so only for much higher rates. If primary insurers cannot obtain sufficient reinsurance, or insurance price controls in some states limit their ability to pass higher reinsurance prices on to their policyholders, insurers will likely seek to exclude terrorist losses from coverage. But many regulators won’t permit terrorism exclusions for certain lines of insurance. Property/casualty coverage for many commercial property owners and businesses could evaporate if insurers seek to limit their exposures by declining to quote on renewal coverage.

Federal intervention offers a different way for insurers to limit their risk exposures and shift costs. The Bush administration has proposed direct federal reimbursement of a portion of terrorist claims for three years. In 2002, the federal government would pay 80% of the first $20 billion in insured losses from terrorism and 90% of the next $80 billion. The taxpayers’ share would decline somewhat, and kick in at higher loss thresholds, in 2003 and 2004. The Bush proposal includes no risk premium for those taxpayer guarantees.

The Senate seems poised to skip the plan’s first year and make insurers bear $10 billion of loss before the Treasury writes any checks. A new proposal by House Banking Committee Republicans sets even lower requirements for federal assistance, but it chooses to deliver taxpayer assistance in the form of “loans” to insurers.

This initial round of federal reinsurance proposals would create adverse effects on risk assessment and incentives to reduce the risk of loss and settle claims efficiently. Government-controlled insurance invariably results in subsidized rates that are, at best, only crudely related to the risk of loss. Its incentives for in-claim settlements are relatively weak. In the two main federal insurance programs, crop and flood insurance, the government insures a disproportionate number of high-risk entities at inadequate rates, thus requiring large taxpayer subsidies. Rather than lose money and disappear, federal insurance programs tend to lose money and expand, crowding out viable private-sector coverage. Risky activity and the amount of losses increase as parties adapt risk management to the terms of subsidized coverage.

Subsidized federal reinsurance could make citizens more vulnerable to harm by discouraging rational responses to post-September 11 risks. If insurance against terrorist attacks is made available at substantially lower costs due to federal subsidies, will businesses be more or less likely to disperse their operations, relocate away from high-risk urban centers, and invest in risk reduction?

The risk of loss from terrorist attacks is not currently significant enough to make terrorism broadly uninsurable. Private insurers are handling tens of billions of dollars in claims from the World Trade Center attacks. What is uninsurable, by either private or public parties, is the harm caused by unquantifiable fears of the unknown.

We are told that the mere possibility of massive uninsured and (presumably) uncompensated losses will shut down our economy because lenders will no longer finance new construction and business operations will close. A more plausible scenario, barring the unlikely event of another major attack in the near term, is that bankers, builders, businesses, and at least some insurers will still want to make money rather than shut their doors to new and renewed business. They will adjust by repricing and reconfiguring the shared costs of terrorism risks. If some insurers throw their hands up and say, “When the going gets tough, the tough may get going, but we’re outta here,” the remaining parties, and perhaps many new ones, will bear the risk.

Look for accelerated entry in offshore reinsurance markets and a burgeoning market in catastrophe bonds and insurance derivatives, which offer larger returns to investors willing to handle greater risks. In high-risk regional markets for office space and business construction, borrowing costs and down payments will be higher. Fewer buildings will be constructed. New office space will be designed differently, and business operations will become less concentrated.

Bankers, builders, insurers, and owners of commercial office space — particularly in higher risk locations — will be less wealthy than they expected to be less than two months ago. With or without insurers providing a layer of risk-sharing protection, our markets won’t be failing. They simply will be bringing us news that we don’t like. We won’t manage this risk effectively through mechanisms that try to tune out, rather than take in, market signals.

Private interests never lobby for public insurance in order to ensure market prices. Low prices for politically brokered insurance displace viable private coverage and expand the public program over time. The government then has strong incentives to intervene more directly in the remaining “private” layers of the primary insurance market. With taxpayers at risk, greater federal regulation of the pricing, sale, and coverage of insurance becomes a necessary quid pro quo. Even if private insurers later seek to “take back” a greater market-priced share of the insurable risk market, they will have forfeited much of their political credibility. How much can business customers (and voters) rely on insurers to manage difficult risks in the future through private market pricing — when those same insurers appeared so eager to cover their bets with taxpayer funding?

The insurance industry’s current central message is that taxpayers should assume risks that insurers are unwilling to assume. Yet if private experts at risk assessment (insurers) won't put up their own company’s money to offer insurance coverage, why should the public?

Several alternatives would make more sense.

1. Allow insurers and reinsurers to do their job by accumulating some amount of capital (loss reserves) on a tax-deferred basis. Reducing the tax “penalty” on insurers’ capital would expand private sector capacity to insure potentially large losses from terrorism. Coverage would become cheaper and more readily available.

2. Distinguish between low-level liability and truly catastrophic risk exposures. Liability caps and exclusions at the upper end in private insurance contracts should be allowed and enforced. Any new federal reinsurance should only enter at the thresholds that truly threaten systemic breakdown of private insurance protection. The commercial insurance industry already is benefiting from an improved pricing climate and looking to expand in tax-friendly jurisdictions like Bermuda. It’s quite premature to suggest that levels for federal relief are anywhere near as low as those suggested by the Bush administration.

3. Consider authorizing a temporary system of ex post assessments, patterned loosely after existing state-guaranty funds, as another buffer against direct federal involvement. In this manner, one could spread a portion of insured losses from terrorist attacks broadly among insurers, their policyholders, and, beyond an ultimate threshold, the federal government. If annual losses exceeded a substantial industry-wide amount, all property/casualty insurers could be assessed a percentage of their premiums to finance a proportion of losses above the threshold. Annual assessments against insurers could be capped to limit insurers’ risk, with insurers allowed to borrow any shortfall of assessments from the Treasury, to be paid back over time from future assessments. If losses exceeded a second, much higher threshold, it might make sense to have the federal government reimburse a proportion of the losses above that amount. Other insurers would work hard to make sure that any claims are settled at reasonable cost and that the program be phased out as soon as any “crisis” passed.

Levels of catastrophic risks that truly are "uninsurable" won’t be managed efficiently with hastily constructed public/private "partnerships" that masquerade as insurance and corrupt private markets. To handle those most unlikely events, we need clearer ex ante guidelines for the ex post, compassionate relief that would be forthcoming from taxpayers to all eligible injured parties (not just those with commercial insurance coverage). We also should press ahead to remove tax and regulatory disincentives that impede the growth of private sector risk bearing capacity.

The federal government’s role in the fight against terror is far more crucial than subsidizing private insurance markets. By upgrading national security, Washington can make a real difference. Success in permanently reducing the risk of terrorism and lowering its costs would reduce the volatility and uncertainty we see now in insurance markets. We need more security, not subsidies, to handle the new world in which we awoke on September 11. Change the risk. Don’t hide its cost.