Earthquake papers
Foreign Earthquake Insurance Programs
Richard Roth Sr.
October 1999
ICLR Research Paper Series B - No. 8
Abstract
Strictly speaking, the earthquake risk is not insurable by the private insurance industry. It is not insurable because it lacks three of the requirements for insurability, which are italicized below. Although there are differing opinions on the conditions of insurability, one that has generally been accepted over the years is as follows:
There should be a large number of homogeneous, independent exposures to permit the operation of the theory of probability.
1. The insured event must be accidental. The timing or the severity of the loss should be out of the control of the insured.
2. The peril must produce a loss definite in time and amount. The insurer must be able to verify the loss promptly, and objectively measure its amount.
3. The loss must cause an economic hardship on the insured.
4. The risk must not be subject to a catastrophic loss where a large number of exposure units can be damaged or destroyed in a single event.
5. The chance of loss must be calculable. The insured group of risks must not be exposed to an incalculable catastrophe hazard.
6. The premium must be reasonable in relation to the potential financial loss so that there will be sufficient demand for policies to cover the up-front costs of development and marketing.
If the above is true, one may ask why does the private insurance industry offer insurance against earthquake damage? One explanation is that the coverage was initiated in the last years of the 19th century or the early years of the 20th century when scientists believed that the planet earth was shrinking as it cooled, and that the contraction produced occasional earthquakes. It was not until the 1960s when the tectonic plate theory was accepted, that it was realized that what we have is a relatively thin layer of the earth_s crust consisting of different plates constantly moving on a molten interior. As these plates impinge on each other, eventually pent up energy is released resulting in earthquakes of different magnitudes proportional to the amount of energy released. We now know that the question is not whether an earthquake will occur, but when.
Although the earth scientists are increasing their knowledge at a rapid pace, the fact is that the probability of the occurrence of an extremely large earthquake is not known. The same is true for estimating the amount of resulting damage.
The private insurance industry today is currently at risk for the earthquake coverage written regardless of the size of the earthquakes. Further, insurance companies in Canada and the United States are subject to an even greater exposure because a loss due to fire following an earthquake is considered a fire coverage, and is covered under the fire policy. Thus, even though earthquake coverage may be included in only a percentage of their fire insurance policies, 100 percent ofthese policies are exposed if the damage is the result of fire following an earthquake. The earthquake peril without a federal, or central government partnership is not insurable in the long run by the private insurance industry. Only a central government has the resources able to weather the financial losses caused by an extremely large earthquake which occurs once in five hundred years, or once in a thousand years, or longer. It should be noted that even if a given size earthquake has a probability of occurring only once in a hundred years, that doesn_t necessarily mean that its occurrence will be delayed to one hundred years from now. It could happen at any time!
The government has a lot at stake in the event of a mega-earthquake, and will not be able to avoid its responsibility to the citizenship. It is much to its advantage if it seeks a partnership with the private insurance industry by offering a very high level excess of loss reinsurance program. A reinsurance program of this nature would have several benefits for the central government in the event of a mega-disaster. First, it would keep the private insurance industry viable preventing an almost certain economic disaster. Second, it would give it an opportunity to insist on earthquake loss-reduction measures as a condition of reinsurance. The only long-term solution to reducing natural disaster losses is the careful and constant attention to appropriate building construction and land use.
It seems, therefore, that the most desirable program would be one in which there is a partnership between the private insurance industry and the government. Private insurance companies, then, would be able to offer earthquake insurance in a similar manner to all of their other lines of insurance. This would include policy-servicing, setting rates, and claims handling. It should also allow companies to establish catastrophe reserves for their part of the exposure. Above the given earthquake magnitude the government reinsurance program would come into play. It would reimburse the participating private insurance companies for losses above a given threshold. The new computer models would be helpful in establishing the proper thresholds.
The programs that seem to operate best are those requiring mandatory participation by the public. Studies indicate that the average insured is not inspired to pay money for insurance that has a low probability of occurring, even if the consequences are very substantial.