THE insurance industry faces claims of at least $40 billion tied to the World Trade Center attacks, and it remains exposed to significant risk from future incidents. Retail insurance companies are considering exclusions for terrorist damage, leaving property owners unable to buy the coverage they need.
The problem lies in the reinsurance industry, the wholesale market where insurers lay off large risks to pools of investors willing to absorb them. Warren E. Buffett's reinsurance company, General Re, has done well in the last few years offering property reinsurance because there have not been major hurricanes or earthquakes. But its luck ran out last month: General Re has said it is liable for over $2 billion of the total industry losses in the Sept. 11 attacks.
Earthquakes and hurricanes, unpleasant as they are, present manageable risks: insurers know roughly how often they occur in various places, what their likely magnitudes are and how much property damage might be expected.
Terrorist risk is much harder to quantify, leading to the current paralysis in the reinsurance market. The insurance industry has been lobbying in Washington to make the federal government the insurer of last resort against terrorism, which means that taxpayers will end up bearing the financial risk of future attacks.
Britain adopted a government-backed reinsurance market several years ago and, so far, it has worked reasonably well. But there is another way to supplement traditional reinsurance markets that has been attracting increasing attention: catastrophe bonds. These bonds, generally sold to large institutions, have typically been tied to natural disasters, like earthquakes or hurricanes, but they could, in principle, be used to provide financial backing for terrorism insurance.
Here is how they work. A financial intermediary, like a reinsurance company or an investment bank, issues a bond tied to a particular insurable event, like a Los Angeles earthquake. If there is no earthquake, investors are paid a generous interest rate. But if the earthquake occurs and the claims exceed an amount specified in the bond, investors sacrifice their principal and interest.
How generous does the interest rate have to be? That is left up to the market. If the bond sales offer cheaper reinsurance coverage than do traditional private placements used in the reinsurance market, these bonds, known as cat bonds, are a preferred method of finance.
Cat bonds are a form of ''contingent security,'' a concept first formulated by Kenneth J. Arrow of Stanford University, winner of the 1972 Nobel in economic science.
Back in 1952, Professor Arrow came up with the idea of a security that would pay a fixed amount of money depending on whether or not some event occurred. He showed that portfolios of such contingent securities could be used to allocate virtually any kind of risk in an efficient manner.
The analysis by Professor Arrow was long thought to be of only theoretical interest. But it turned out that all sorts of options and other derivatives could be best understood using contingent securities. Now Wall Street rocket scientists draw on this 50-year-old work when creating exotic new derivatives.
There are a number of special cases of contingent securities. Consider a security tied to my house burning down. This would certainly affect my net worth, but would have negligible impact on the owners of the other 70 million housing units in the United States.
This kind of personal risk can easily be shared among many people, each paying only a small amount if the event occurs, and homeowners insurance is to ease this kind of risk sharing.
This works fine for small independent risks, but there are other sorts of events that impose large costs on many people at the same time, like earthquakes or hurricanes. Insurance companies are able to deal with such events using reinsurance markets, where the risk is transferred to large investors.
There are other financial institutions that also allow for risk transfer for this sort of widespread risk. The orange juice futures market in Chicago is essentially a market in a contingent security tied to whether it freezes in Orlando, Fla.
A futures market allows market participants who bear a significant risk (like farmers worried that their crop might be destroyed) to transfer some of that risk to others, who are, of course, paid to absorb it. The futures market allows shifting of risk rather than a simple sharing of risk.
Risk sharing and risk shifting are familiar terms, but there is another phenomenon worth noting, something I like to call ''risk shafting.'' This is when some risk -- often a particularly large risk -- is transferred to a third party who is forced to bear it involuntarily.
In many cases, these third parties are taxpayers, who are not even aware of their potential liability.
Hence the debate that is going on in Washington now: should the insurance industry be able to transfer the risk from future terrorist attacks to taxpayers? Or would it be better to use existing reinsurance markets or new financial instruments like catastrophe bonds to shift the risk to those willing and able to bear it if appropriately compensated?
Cat bonds have some attractive features. They can spread risks widely and can be subdivided indefinitely, allowing each investor to bear only a small part of the risk. The money backing up the insurance is paid in advance, so there is no default risk to the insured.
As the market for cat bonds matures, secondary markets may develop, particularly if the bonds become standardized and bundled into portfolios. If so, the market price of cat bonds will reflect the market perceptions of the likelihood of the event associated with the bond, just as price changes on the orange juice futures market reflect the likelihood of a freeze in Florida.
The market for cat bonds is still immature, and is only a fraction of the size of traditional reinsurance. But cat bonds and other sorts of contingent securities may well end up being part of the long-run solution to the problems of the reinsurance industry.