Is there still a place in the industry for finite reinsurance despite its tarnished reputation? Or is it just a bad word that "gives people the shivers"? asks Ronald Gift Mullins.

The use of so-called finite reinsurance has received considerable attention because of its misuse by some high-profile insurers, the National Association of Insurance Commissioners (NAIC) diplomatically admitted a few years ago.

Initially, the appeal of finite reinsurance was that although there had to be some transfer of risk for the transaction to appear as an asset on the insurer's balance sheet, the appearance of risk transfer was what was desired even if very little risk transfer had actually taken place. This was the observation of a former CEO of a reinsurance company, who asked not to be identified. "As long as regulators and auditors didn't look too closely," he explained, "nothing much was said about actual transfer of risk. It was only when Eliot Spitzer (then New York attorney general) was made aware of the accounting irregularities of finite risk reinsurance by the actions the Securities and Exchange Commission (SEC) and rode it hard to bolster his run for the New York governorship that it became a major media deal. Consequently, rigid rules have been mandated to forestall this slight of accounting from happening again ... or at least anytime soon."

Finite risk reinsurance transfers only a finite or limited amount of risk to the reinsurer. Risk is reduced through accounting or financial methods, along with the actual transfer of limited economic risk. By transferring less risk to the reinsurer, the insurer receives coverage on its potential claims at a lower cost than traditional reinsurance, and losses are spread over several years. There are a number of different types of finite risk reinsurance that are all commonly used by insurers wanting to get out of a particular line of business, corporations that have been self-insuring their exposures and seek to transfer their claims operations, or to enable a captive insurance company to withdraw from writing third party business.

But most often used, and where trouble often arises, is through manipulating loss reserves. Straight discounting of an insurer's loss reserves is not allowed by US accounting rules, but with one type of finite reinsurance, a portion of loss reserves can be transferred for a fee via reinsurance, thus freeing up additional surplus and improving the insurer's balance sheet. Or the reverse can occur; loss reserves can be bolstered by finite reinsurance contracts, although transfer of risk has to be included in the transaction.

Obscure side deals

From about 1990 until recently, the acceptable amount of risk transfer in a finite reinsurance transaction had been commonly agreed to be at least 10% of the total policy value. But due to the highly complex structure of these risk instruments, there were abuses where not enough or no risk at all was transferred. Consequently, the transaction should have been treated as a loan on the company's balance sheet. And through side agreements, premiums were ceded back to the insurer. These side deals, not shared with regulators, accounting firms or shareholders, lacked transparency but allowed the insurer to increase surplus or reserves, and if needed, smooth out a bad earnings quarter.

Unlike traditional reinsurance contracts, where reinsurers can be on the hook for substantial losses stemming from a catastrophe such as 9/11, finite risk contracts require primary insurers to ultimately pay for a majority of ceded losses themselves, but gain by spreading the payments over a longer period of time. One of the most serious abuses of finite reinsurance occurred in 2001 in Australia with the sudden collapse of the HIH Insurance Group, which cost policyholders, taxpayers and creditors more than $3.8bn. Following an investigation by the Australian government, finite reinsurance was blamed in part for the insurer's failure. Through the use of massive amounts of finite reinsurance, executives of FAI Insurance, which had been acquired by HIH, managed to hide onerous claims which allowed the company to show positive earnings for three years. In fact, if it had not benefited from finite reinsurance, FAI would have gone into default and would not been purchased by HIH.

Alarmed by the HIH failure and the investigation into Enron's catastrophic bankruptcy in late 2001, which entailed the concealed use of off-balance sheet transactions and offshore captives, the SEC, the US Department of Justice and Eliot Spitzer began requesting information concerning finite reinsurance transactions. American International Group (AIG), Berkshire Hathaway, ACE, Platinum Underwriters, MBIA, Swiss Re, Zurich Financial Services and St Paul Travelers, among others, were all targeted.

Noise brings attention

Marc Oberholtzer, chairperson of the American Academy of Actuaries Committee on property and liability financial reporting, said that some accounting irregularities regarding finite reinsurance were initially highlighted by the rating agencies Fitch and Standard & Poor's. "They made noises about certain finite re contracts not having enough risk being transferred to qualify as reinsurance, so more attention came about following their statements."

In addition to the rating agencies, the accounting profession had raised concerns about the lack of risk transfer in finite reinsurance contracts and had urged the regulators of the insurance and reinsurance industry to develop strong practices and rules governing this type of reinsurance.

Finite risk reinsurance is not bad practice in itself. According to Joseph Sieverling, senior vice president and director of financial services, Reinsurance Association of America (RAA), "It is not a way to cook the books." There was not much finite reinsurance used from 1996 up to 2000, he continued, as reinsurance was cheap due to excess capacity, lots of competition and naive capital. There was real under-pricing from 1997 to 2000, and "if cheap reinsurance is available, why bother with finite reinsurance. Then with the losses from 9/11 and a tightening of the market following, interest in finite reinsurance increased."

Sieverling says it is difficult to identify how much finite risk reinsurance is sold, but estimates that it represented about 2% to 3% of the world reinsurance premiums in 2004 or 2005. He acknowledged, "It is tough to say if problems with finite reinsurance have hit rock bottom. Between the Justice Department and the SEC there have been about 20 investigations. Of them, only a few have been finalised." He observed that a number of companies had restated earnings in 2005 and 2006, but the amounts involving "finite reinsurance were not material adjustments as to overall capital and surplus of a specific company. I think there is a feeling that the worst is behind us."

NAIC taskforce

In early 2005, the NAIC formed a taskforce to access the risk transfer requirement and other important issues related to finite risk reinsurance. Accordingly, the group promulgated a set of guidance rules to govern reporting of finite risk reinsurance. These rules were implemented too late for the 2007 reporting year, but will become effective 31 December 2007. But stipulations in the new rules are closely related to requirements in effect now.

The rules require an insurer to report to state insurance regulators any agreement that has the effect of altering policyholders' surplus by more than 3%, or representing more than 3% of premium or losses. The new disclosure is also designed to identify any reinsurance contract that has been accounted for differently under statutory accounting principles compared to general financial statement purposes.

A standard attestation form must be signed by the insurer's CEO and CFO acknowledging reinsurance contracts that the company has taken "credit" for on its financial statements. The provisions of the attestation include:

- That there are no separate agreements between the insurer and the reinsurer that could serve to modify the actual or potential losses under the contract; and

- That the insurer complies with all requirements of the NAIC's statement of statutory accounting principle No 62, "Property and Casualty Reinsurance".

"The proposed enhanced disclosure requirements, in addition to an attestation by company management of entities that engage in these transactions, should clarify the overall impact of finite reinsurance on the industry," said Joe Fritsch, director of insurance accounting policy for the New York Insurance Department and chair of the study group, in a statement.

Hot cakes

With the new detailed rules for finite reinsurance contracts in place, the outlook remains cautious for increased production of this type of reinsurance. Enda McDonnell, president and CEO of Access Reinsurance, said the Bermuda-based brokerage had done a couple of multi-year contracts in the last 12 months. "Even though the finite re product has changed dramatically," he said, "I wouldn't call what we put together finite risk reinsurance. The word 'finite' gives people the shivers. There is much more risk transfer now in finite reinsurance, but finite re is not being bought by the number of companies who used to buy it in its heyday in the late 1990s and early 2000s."

"I would guess there is more comfort now within the industry for finite reinsurance," RAA's Sieverling observed. "There is a bit more being done now than last year, because a lot of regulatory and enforcement factors have been ironed out. People know what the rules are."

The creativity of utilising finite reinsurance to remove the potential liability of adverse loss development in an insurer's financial statement on the surface can be circumspect. According to Andrew Barile, a reinsurance consultant, "The terms and conditions of the finite reinsurance agreement must be scrutinised, along with the correspondence file. It is only after full disclosure that intelligent opinions can be made with respect to custom and practice in the buying of finite reinsurance agreements."

Eventually, the former reinsurance company CEO believes, finite reinsurance contracts will come back. "But however popular they become," he offered, "they'll never sell like hot cakes."

- Ronald Gift Mullins is an insurance journalist based in New York City.