Finite Insurance: What is the real problem?

A number of insurance industry practices have recently been reviewed and criticized by New York state Attorney General Eliot Spitzer, as well as various U.S. state insurance regulators. After initial attention was focused on the payment of contingent profit commissions to brokers, regulators then examined the use of finite reinsurance. From the perspective of Spitzer, it appeared finite reinsurance was nothing more than an attempt by insurance companies to manipulate and smooth their earnings using complicated reinsurance transactions for which there is not adequate disclosure.

WHAT IS FINITE REINSURANCE?

One of the basic difficulties in determining whether there is a real problem with finite reinsurance is that there is no clear difference between it and normal reinsurance. Finite reinsurance was introduced in the ’80s as a way to lower the cost of reinsurance. It is a legitimate product when the accounting for it reflects the economic reality. However, some types of finite reinsurance are really loans used in order to improve the appearance of the financial condition of an insurance company. When properly used, finite reinsurance can protect against volatility, while allowing the ceding company to retain greater risk. As a result, it is a more cost-effective type of risk transfer. In other cases, there may be no economic justification for finite reinsurance.

The following indicators are signs that a reinsurance arrangement may be considered to be finite reinsurance:

* the agreement is retroactive;

* there is a difference between statutory and GAAP accounting for the transaction;

* it is used for financial engineering;

* it results in a delayed recognition of losses or smoothes them over a period of time;

* there is a side letter;

* there is a greater than 50% chance that the agreement will be commuted;

* there is material use of deficit accounts or experience refund accounts;

* the agreement is in effect for more than one year.

It is necessary to look at the purpose of the agreement and the amount of risk transfer. It is also necessary to look at the effect on the financial statements of an insurance company. The most common use of finite reinsurance in the past has been to provide for discounting for loss reserves when they have not been permitted on a statutory basis.

One of the difficulties in determining whether there is a real problem with finite reinsurance is the fact that all types of insurance and reinsurance are also intended to smooth their earnings and reduce volatility. As a result, it is difficult to determine where to draw the line between practices considered to be proper and improper. It is interesting to note that caps, stop losses, loss corridors and aggregates also have finite elements to them and are commonly used in normal reinsurance agreements.

The issue of finite reinsurance was reviewed a number of years ago by various regulatory and accounting groups in the U.S.; as a result, FASB 113 was developed to provide guidelines with respect to the minimum amount of risk transfer required for finite reinsurance. The traditional 10/10 rule for the evaluation of finite reinsurance transactions was developed. It required that there be a 10% probability of a 10% loss for there to be a sufficient amount of risk transfer for an agreement to be considered a reinsurance arrangement rather than a loan.

SPITZER’S REVIEW

Spitzer and various state insurance regulators have argued that insurance companies facing trouble will use finite reinsurance as a way to conceal their problems. Finite reinsurance may be viewed as simply being an accounting issue for which new disclosure rules may be necessary. It can also be viewed as an arrangement that can be used to misrepresent to regulators and rating agencies the true financial condition of a company. Manipulation of financial statements and the concealing of losses results in issues similar to what occurred in Enron. In extreme cases, a finite reinsurance arrangement could even be considered to be fraud.

A second type of finite reinsurance transaction – the use of a side letter — has attracted the attention of Spitzer and various insurance regulators. In particular, they question the use of a traditional reinsurance arrangement that has been modified by a side letter that may not be disclosed. The use of a side letter may be viewed as a deliberate attempt to conceal the true economic nature of the arrangement. The effect of a side letter is to override or amend certain provisions of a reinsurance agreement. For example, a side letter might provide for an automatic commutation after a certain period of time.

As a result, it will be necessary to determine whether finite reinsurance is a legitimate product with some minor abuses, or an inherently bad product. The failure of HIH in Australia has been used as an example of a company that used finite reinsurance to conceal its real problems and as a result delayed insurance regulators taking action.

One of the interesting issues in the current review of finite reinsurance in the U.S. is whether Spitzer and other state attorneys general, the SEC or the NAIC should have primary jurisdiction to deal with the matter. The International Association of Insurance Supervisors – in which the Office of the Superintendent of Financial Institutions (“OSFI”) participates – has also issued a discussion paper on finite reinsurance. The appropriate person to be taking action will likely depend on whether the issues regarding finite reinsurance are viewed as being accounting fraud or insurance regulatory in nature.

REGULATORY INITIATIVES

Insurance regulators have generally responded in one of two basic ways to the finite reinsurance issue. One has been to promote enhanced disclosure of finite reinsurance arrangements in statutory reporting by insurance companies. It will now be necessary for insurance companies to disclose in their statutory filings arrangements that have finite characteristics.

The second approach, which has resulted in more controversy, is a requirement for CEO/CFO attestation that all reinsurance agreements for which a company has taken credit do not have any side agreements; moreover, if there is any agreement for which the risk transfer “is not considered self-evident,” there is documentation provided concerning risk transfer analysis and the economic intent of the agreement.

The result of these proposals is that a detailed risk transfer analysis and testing will be required for all reinsurance agreements with finite aspects to them.

OSFI has requested that insurance companies in Canada report to it on their positions regarding the use of finite reinsurance. OSFI has not yet issued a guideline indicating its position on finite reinsurance. However, it is likely OSFI will follow a similar approach to that taken by U.S. insurance regulators.

THE FUTURE OF FINITE REINSURANCE

It is likely that significantly less finite reinsurance will be used in the current environment, since companies will not want their reinsurance arrangements to be investigated. In particular, the requirement for CEO/CFO attestations in some jurisdictions could significantly reduce the willingness of companies to use finite reinsurance due to the possibility of individual liability.

There is no question that the insurance industry has recognized the importance of additional disclosure with respect to finite reinsurance transactions. However, it is unlikely that finite reinsurance will disappear.

The abuses Spitzer has identified, such as side letters, are really a failure to comply with current accounting standards and insurance regulatory requirements. It is unlikely there has been a wide spread abuse of finite reinsurance by the insurance industry.

Finite reinsurance will continue to be used in the future in appropriate situations. Provided that there is adequate amount of risk transfer and proper disclosure of the economic effect of the transaction on the companies involved, there will always be a place for finite reinsurance as a legitimate product to be used by the insurance industry.