Finite Matters

What a difference a decade makes. In the mid-to-late 1990s, reinsurance companies and brokers could not talk enough about finite risk or financial reinsurance products. Several leading reinsurance companies and brokers set up finite reinsurance units, in various countries around the world, and actively marketed “non-traditional” products. Publications from the world’s top reinsurers touted these new approaches to risk transfer as the future of reinsurance.

Typical of these was a 2000 publication from Munich Reinsurance Co. about finite reinsurance, which put the estimated worldwide finite risk reinsurance premium at the time at US$8-$10 billion, equal to 6%-7% of total global reinsurance premiums. It confidently predicted that finite reinsurance premiums would double by 2005 (some estimates put the current level of finite reinsurance premiums at $50 billion). “Finite risk reinsurance has developed into an effective tool to allow reinsurance companies to retain more risk, but, at the same time to smooth results over time by combining risk financing and risk transferring techniques,” Munich Re notes.

Another publication from Swiss Reinsurance Company in 1997 states that “the prospects of finite risk reinsurance are very promising” and “finite risk reinsurance is gaining increasing significance as a component of insurance companies’ risk management”. And, the Internet website of one reinsurance broker holds that “as the leading finite reinsurance intermediary, Guy Carpenter has placed the largest reinsurance programs in the industry, with limits from $10 million to $2 billion”.

Fast-forward to today and reinsurance companies and brokers are tight-lipped about the current and future status of finite reinsurance (representatives of several leading reinsurers and brokers in Canada either declined to participate in this article or did not return phone calls). This, however, is hardly surprising given the ongoing investigation of certain finite reinsurance deals in the U.S.

U.S. INVESTIGATIONS

At the center of ongoing investigations in the U.S. is a finite reinsurance arrangement between AIG and General Re that dates back to 2000/2001. The heart of the matter is whether premiums transferred from General Re to AIG, and repayable later, involved a true risk transfer or were simply a loan. A standard rule of thumb for legitimate risk transfer is the “10/10 rule,” meaning that the insurer should face, at a minimum, a 10% chance of losing 10% of the policy amount for the contract to be considered insurance. If there were no significant degree of risk transfer, then it would be a loan.

AIG accounted for the $500 million in premiums it received as a loan, or more specifically as additional reserves. At the end of March this year AIG’s directors concluded that the General Re transaction was not properly classified as insurance and said the company would reduce its reserves by $250 million and increase liabilities by $245 million. The investigation into this deal prompted the recent retirement of AIG chairman/CEO Maurice Greenberg, who in mid-April, also invoked his “Fifth Amendment right” against self-incrimination rather than discuss the issue with regulators (in November 2004, AIG also agreed to pay $126 million to the SEC and U.S. Department of Justice to settle investigations relating to finite insurance transactions for PNC Financial Services).

The fallout has extended well beyond one company and one deal. A wide-ranging investigation by Spitzer’s office and the SEC has seen subpoenas issued to several companies, including ACE, AXA Re, Chubb, MBIA, Platinum Underwriters, St, Paul, Swiss Re, Travelers and Zurich Financial Services, amongst others. Several companies are the targets of ongoing investigations into finite reinsurance deals.

OUTWARD RIPPLES

Regulatory attention to finite reinsurance has also extended to the U.K. In March this year, the country’s Financial Services Authority (FSA) started a formal inquiry into finite reinsurance, building on an earlier investigation conducted by the watchdog body in 2002.

At present, there are no pending investigations in Canada. “We have not found evidence…of these types of alleged abuses of finite reinsurance that have occurred internationally,” confirms Rod Giles, a spokesman for the Office of the Superintendent of Financial Institutions (OSFI). “At OSFI, we continue to monitor the reputational risk management systems and practices of financial institutions in relation to finite reinsurance developments.”

“In Canada, on the finite side, I think the problem you have is scale,” says Andr Fredette, senior vice president and general manager of CCR Canada. “A lot of these treaties are fairly complex and you need a bunch of lawyers to look at it on both sides, which is expensive. Traditional reinsurance was so cheap that, by and large, it seems to have met the needs of the market. People have tended to use finite risk products on unusual classes, where you couldn’t find a traditional market.”

CANADIAN CONCERNS

Yet, global reinsurance activity and trends do filter into Canada. “I think the finite deals have been around for a long time, and I certainly have been offered the opportunity to participate in them or I have heard of them,” says David Wilmot, chief agent in Canada for the Toa Reinsurance Co. “The suggestion is that, at one point, they represented 25% of the deals that went down globally, and I believe there is some truth to that.”

A simple definition of finite reinsurance is elusive, but there are some characteristics that distinguish it from traditional reinsurance. Finite reinsurance is intended to limit companies’ exposure to unknown liabilities or to soften the impact of losses over a longer period of time. It is a complex transaction, and at least part of the allure of the product is how it can be potentially used for tax, accounting and earnings benefits that bolster a company’s financial position.

“When I heard it described to me by one expert, I said that it sounds unnecessarily complex,” notes Wilmot. “But the person responded that it was ‘necessarily complex’ for the exact reason that it would be unfathomable to many people, including regulators. That is when I knew I didn’t want to have anything to do with it.”

One of the key distinctions of finite risk reinsurance is that it “represents a combination of risk transfer and risk financing which emphasizes the time value of money,” according to a Swiss Re report. In other words, there is an understanding that money to be paid in the future is worth less than the same amount of money payable now. The study lists several characteristics that apply to finite reinsurance products, including the limited assumption of risk by the reinsurer, a multi-year contract term, explicit inclusion of investment income in the contract and sharing of the results with the insured/cedent.

Finite arrangements can be either prospective, “protecting against the contingencies associated with future loss events,” according to a Munich Re document, or “retrospective covers [that] insulate insurance companies from the negative impact of an adverse loss development and/or from the accelerated payout of losses already incurred.”

HISTORICAL DEVELOPMENT

Finite risk reinsurance is not new. These financial transactions first appeared in the 1970s and by the mid-1980s were widely used, often with very limited risk transfer. In some cases, signs of abuse emerged more than 20 years ago. Writing in a 1997 Sigma report, Swiss Re stated that “many of the prejudices against finite risk reinsurance go back to the early 1980s, when some few cedents in the USA misused certain products to conceal the precarious situation of their companies. Such misuse has been a thing of the past for a long time now and is no longer imaginable within the context of the finite solutions offered today.”

The question of what constitutes an authentic risk transfer was addressed in the ea
rly 1990s, when the U.S. Financial Accounting Standards Board (FASB) established specific guidelines for risk transfer and accounting practices for loss payments over time.

Finite reinsurance continued to gain strength into the late 1990s and early 2000s.

Sources maintain that there are important distinctions between the abuses of finite risk, or financial, reinsurance and legitimate forms of reinsurance risk transfer. Wilmot says the loose interchange of financial and finite risk reinsurance is “careless, in my opinion”. He adds, “the term financial reinsurance is widely used but often misunderstood as a blanket term for reinsurance transactions more likely to be described as accounting strategies rather than reinsurance risk transfer. At its worst, one may describe financial reinsurance as ‘non-concurrent accounting’. That means taking it off your balance-sheet at this point and putting it on somebody else’s, or doing it in such a way that the liabilities or profitability do not appear on either balance-sheet.”

In the world of reinsurance, there is nothing intrinsically wrong with the words “finite,” which reinsurers use to limit their exposure, or “financial,” which, in the form of quota-share treaties, represents the truest form of reinsurance and “lending” of capital. Many argue that finite risk reinsurance is a legitimate product that has been tarred with the brush of illegal actions by a few companies. The sticking point is the level of real risk transfer that exists in a reinsurance agreement.

“The danger is that we throw the baby out with the bathwater,” says Philip Cook, CEO of Omega Insurance Holdings Inc. “The fact is that insurers buy reinsurance to improve their balance-sheets. Now, buying it for the purposes of falsifying the balance-sheet is blatantly illegal and should be prosecuted. You have to be very careful on how you define these issues and look what was behind the strengthening of balance-sheets.”

“Finite reinsurance transactions are appropriate as long as they are used for a true transfer of risks,” notes Giles. “The concern by OSFI and other regulators is that some contracts could involve no or little amount of risk transfer between the ceding companies and reinsurers, and the primary motive of entering into reinsurance is to smooth earnings or inflate capital.”

“I think OSFI has made it clear what they consider to be outside the expected norms, and they have in fact acted on occasion over the years,” says Wilmot. “I can even think of situations 10 years ago, in which they were stepping in on particular contracts and saying ‘no, we don’t feel there is a sufficient risk transfer involved’.”

QUESTIONABLE USE

Companies have got into trouble when they have used finite risk reinsurance transactions less as an insurance policy and more as a loan designed to either lower reported losses or inflate reported surpluses – what Fitch rating agency termed “financial engineering” in a report issued last November. These kinds of transactions “may look normal, but you would have to dig underneath and say to yourself, ‘this doesn’t smell right,'” observes Wilmot.

In addition, the issue of undisclosed “side agreements” has come to light – agreements that could materially alter the terms of the contract. Regulators in Tennessee and Virginia have filed civil lawsuits against General Re alleging “non-contractual understandings” and two unreported side agreements in transactions involved in the bankruptcy of professional liability insurer Reciprocal of America. “To me, this is practically criminal to have something like that,” says Wilmot. “It means there is the contract for the regulators to see, but also this real [other] agreement on the side.”

The New York State Insurance Department is now requiring CEOs of insurance companies to sign-off on finite reinsurance transactions, In particular, CEOs are required to disclose, under penalty of perjury, that there are “no separate written or oral agreements” and that “for each reinsurance contract, the reporting entity has an underwriting file documenting the economic intent of the transaction and the risk transfer analysis evidencing the proper accounting treatment”. The U.S.’s National Association of Insurance Commissioners (NAIC) is also looking at proposals to require more disclosure of finite reinsurance products.

In Canada, while no investigation is pending, regulators are looking into clarifying rules around risk transfer. “OSFI has written to the Canadian Institute of Actuaries requesting the actuarial profession to provide guidelines…on the minimum degree of risk transfer required for recognition of reinsurance agreement in the policy valuation,” according to Giles. “OSFI will continue to review and update guidelines as needed.”

Despite attempts to more closely monitor finite risk reinsurance, the damage may already be done to the product. Fredette says this will have some ramifications for the long-term viability of reinsurer units specializing in finite risks. “Given the investigations on finite, I think it will be difficult for those units that do this class of business to develop it,” he notes. “The knee-jerk reaction on the company side when it comes to a finite treaty is ‘no, I don’t want to be investigated by Spitzer, there is too much heat on this product’.”

And, Wilmot notes that a decline in finite reinsurance products will mean greater transparency in financial reporting. “This greater transparency could play a role in is a shorter market cycle,” he observes. “It is just one factor that may give us a much shorter soft market than some expect.”

Wilmot acknowledges the confusion around finite risk reinsurance products and how these will be singled out, defined and regulated in the future. “There is a certain correlation,” he says, “with the sometimes used definition of pornography: I don’t know how to define it, but I know it when I see it.”