Pricing Excess-of-loss Reinsurance Contracts Against Catastrophic Loss

This paper develops a pricing methodology and pricing estimates for the proposed Federal excess-of- loss (XOL) catastrophe reinsurance contracts. The contracts, proposed by the Clinton Administration, would provide per-occurrence excess-of-loss reinsurance coverage to private insurers and reinsurers, where both the coverage layer and the fixed payout of the contract are based on insurance industry losses, not company losses. In financial terms, the Federal government would be selling earthquake and hurricane catastrophe call options to the insurance industry to cover catastrophic losses in a loss layer above that currently available in the private reinsurance market. The contracts would be sold annually at auction, with a reservation price designed to avoid a government subsidy and ensure that the program would be self supporting in expected value. If a loss were to occur that resulted in payouts in excess of the premiums collected under the policies, the Federal government would use its ability to borrow at the risk-free rate to fund the losses. During periods when the accumulated premiums paid into the program exceed the losses paid, the buyers of the contracts implicitly would be lending money to the Treasury, reducing the costs of government debt. The expected interest on these "loans" offsets the expected financing (borrowing) costs of the program as long as the contracts are priced appropriately. By accessing the Federal government's superior ability to diversify risk inter-temporally, the contracts could be sold at a rate lower than would be required in conventional reinsurance markets, which would potentially require a high cost of capital due to the possibility that a major catastrophe could bankrupt some reinsurers. By pricing the contacts at least to break even, the program would provide for eventual private-market "crowding out" through catastrophe derivatives and other innovative catastrophic risk financing mechanisms. We develop prices for the contracts using two samples of catastrophe losses: (1) historical catastrophic loss experience over the period 1949-1994 as reported by Property Claim Services; and (2) simulated catastrophe losses based on an engineering simulation analysis conducted by Risk Management Solutions. We used maximum likelihood estimation techniques to fit frequency and severity probability distributions to the catastrophic loss data, and then used the distributions to estimate expected losses under the contracts. The reservation price would be determined by adding an administrative expense charge and a risk premium to the expected losses for the specified layer of coverage. We estimate the expected loss component of the government's reservation price for proposed XOL contracts covering the entire U.S., California, Florida, and the Southeast. We used a loss layer of $25-50 billion for illustrative purposes.