If a purely private market in catastrophe insurance is to be viable, it must solve the fundamental problem of the mismatch between the size of annual premiums and the size of the expected loss. This can be done in one of two ways. The premiums can be adjusted to the losses, or the losses can be adjusted to the premiums. We discuss each case in turn.
Accumulating Premiums to Meet Losses
The science of catastrophe prediction is not well developed, in the sense that the size, frequency, and location of recent major hurricanes and earthquakes have surprised most expects. Meteorologists, for example, are now debating whether the increased frequency of hurricanes in the 1990s compared to earlier decades is an exception or a return to a more normal frequency. Similarly, the last two California earthquakes occurred on fault lines that were not even known to seismologists prior to the events.
For the sake of discussion, however, suppose we assume that it is certain that a $30 billion earthquake will hit California in the next 30 years, and suppose we consider an insurance company with 10% of the market, i.e. a company facing a $3 billion loss. What premium strategy ought this company to pursue? Because of the absence of stability in loss experience, any premium policy based on averaging recent losses clearly will not work. A more viable strategy would be to divide the $3 billion into equal annual premiums of $100 million. (For simplicity we ignore time value of money issues). If these premiums can be unambiguously set aside in a fraud, perhaps being placed in a trust subsidiary, it will eventually provide coverage against the $3 billion loss.
It is clear that this arrangement provides no help if the $3 billion loss should occur in the first year of operation. If there is no access to capital markets, there is simply no way that an insurance company can accumulate premiums fast enough to guarantee that it will have sufficient surplus to meet a catastrophic loss.
Reinsurance
The contract of reinsurance is the traditional means by which an insurance company reduces the size of its potential losses. Although catastrophe reinsurance capacity has recently grown sharply, international reinsurance catastrophe capacity is still very limited. Furthermore, an individual primary insurance company is unlikely to obtain more than perhaps $500 million in catastrophe reinsurance. These amounts appear small compared to the possibility of an aggregate $100 billion earthquake loss or a $10 billion loss for a large company.
In the contract of reinsurance, total annual premiums P and total losses L are split up and held by the direct insurer and the reinsurer according to some sharing rule. The repackaging of P and L can be shown to be in general beneficial to both the primary insurer and the reinsurer, but it cannot avoid the arithmetic fact that when the contracts of insurance and reinsurance are considered together, the ration of losses to premiums, L/P, in any one year is what it was before the reinsurance and is, in the case of catastrophes, potentially high.
Act of God Bonds
Recognizing that committed capital is one of the keys to viability, a number of investment banks have developed bond instruments which provide capital to an insurance company in advance of a catastrophe. Since these instruments will only be exercised in the event of a catastrophe they are known as Act of God bonds (Catastrophe Bonds).
Catastrophe bonds (also known as cat bonds) are risk-linked securities that transfer a specified set of risks from a sponsor to investors. They are often structured as floating rate bonds whose principal is lost if specified trigger conditions are met. If triggered the principal is paid to the sponsor. The triggers are linked to major natural catastrophes. Catastrophe Bonds are typically used by insurers as an alternative to traditional catastrophe reinsurance.
For example, if an insurer has built up a portfolio of risks by insuring properties in Florida, then it might wish to pass some of this risk on so that it can remain solvent after a large hurricane. It could simply purchase traditional catastrophe reinsurance, which would pass the risk on to reinsurers. Or it could sponsor a cat bond, which would pass the risk on to investors. In consultation with an investment bank, it would create a special purpose entity that would issue the cat bond. Investors would buy the bond, which might pay them a coupon of LIBOR plus a spread, generally (but not always) between 3 and 20%. If no hurricane hit Florida, then the investors would make a healthy return on their investment. But if a hurricane were to hit Florida and trigger the cat bond, then the principal initially paid by the investors would be forgiven, and instead used by the sponsor to pay its claims to policyholders.