Roughly five years after finite reinsurance was tarnished by high-profile scandals, coverage by that name is no longer being written, though similar products continue to exist.

Finite or financial reinsurance is a form of financial leverage that often was used to smooth losses over a number of years, although it has fallen out of favor in recent years chiefly because of increased regulatory scrutiny.

The key difference in today's products, which often carry the label “structured,” is they incorporate substantial amounts of risk, observers say. A lack of genuine risk transfer in certain finite deals is what proved to be particularly treacherous for many in the industry in recent years (see box, page 19).

Finite reinsurance in some ways took off with the 1988 founding of Bermuda-based Centre Reinsurance Holdings Ltd., which specialized in the business and introduced more complex financial transactions.

But while finite reinsurance proved useful in managing loss volatility, there developed growing concerns it was being structured less like insurance and more like loans, becoming in effect a financial engineering tool that was abusively used to hide insurers' true financial situation. 

In 1993, the Financial Accounting Standards Board stepped in and said there must be a reasonable chance of a significant loss for a transaction to be accounted for as insurance. Absent the risk transfer, such contracts were supposed to be booked as deposits.

But ambiguity remained as to how much risk must be transferred. Although it was never officially codified by any accounting, insurance or regulatory authority, the market developed the “10/10” rule, which determined that a transaction could be considered insurance if there was at least a 10% chance of a 10% loss.

Then came investigations by then-New York Attorney General Eliot Spitzer and others that, among other developments, eventually led to the criminal convictions of several former executives at New York-based American International Group Inc. and Stamford, Conn.-based General Reinsurance Corp. in connection with finite risk deals (BI, Sept. 20).

Meanwhile, in 2005, the National Assn. of Insurance Commissioners approved finite reinsurance disclosure requirements for property/casualty insurers. It also said that an insurer's chief executive officer and chief financial officer must attest that risk transfer occurred and there were no side agreements that eliminated risk.

The rule has been effective, according to Joseph Fritsch, director of insurance accounting policy for the New York State Insurance Department and chairman of the NAIC's property/casualty reinsurance study group. Since introducing the NAIC disclosure requirements, “we haven't seen any major cases of abuses,” said Mr. Fritsch.

There also is little of what once was called finite reinsurance, observers say.

“Today, there's very little, if any, being written,” said Robert P. Hartwig, president of the New York-based Insurance Information Institute. The investigations and convictions in connection with finite reinsurance have “obviously been an enormous deterrent to participating in this market which, at its height, maybe accounted for 5% or 6% of the reinsurance market,” said Mr. Hartwig.

Paul Walther, CEO and principal consultant with Lake Mary, Fla.-based Reinsurance Directions Inc., said he has not heard much of finite reinsurance activity recently.

It “just seems the traditional ways of doing things have been restored to popularity,” he said. “The bloom went off the finite rose” when AIG and Gen Re came under the microscope, he said. “Because of that, I think people got scared” of such deals.

“Everyone seemed relatively comfortable 10 years ago that they knew how much risk transfer” was required, said Stephen K. Bolland, president of reinsurance intermediary Gill & Roeser Inc. But Mr. Spitzer “just basically changed the goal posts as to how much risk has to be transferred, unfortunately,” Mr. Bolland said.

“It's very difficult to say there is sufficient risk transfer when you don't know what that risk transfer is supposed to be anymore,” he said. “I don't know if too many actuaries and accountants would be comfortable signing off at the moment, because it's unclear where that line is as to how much risk transfer has to take place.”

There continue to be products, however, that incorporate at least some of the elements of finite reinsurance, although all have substantial amounts of risk transfer. Mr. Bolland said that “technically, traditional reinsurance can be adjusted,” so as not to replace finite reinsurance, “but get close to it.”

“One of the interesting things about finite reinsurance is it was done for a huge amount of different reasons. It wasn't just a finite product that was sitting there on the shelf and somebody said, "Oh, I want a bit of that,'” Mr. Bolland said.

“You could use it for a whole host of different reasons, so a whole host of different things has effectively taken its place,” said Mr. Bolland. For instance, much of finite reinsurance was done on aggregate basis, with limited risk transfer. Now, insurers continue to buy aggregate excess reinsurance, but there is more risk transfer and it is more costly.

“Effectively, people are paying more, but they're getting more protection,” Mr. Bolland said.

“The deals are being done, but they're not being called finite anymore. They're being called "structured'” products, said Donald J. Riggin, senior consultant with Spring Consulting Group L.L.C. in Boston. However, these deals no longer follow the 10/10 rule of thumb, said Mr. Riggin. “It's got to be 20/20 or 30/30. There really has to be a significant loss transfer.”

Edward Hochberg, Philadelphia-based principal with Towers Perrin's reinsurance brokerage business, said, “In the old days, it was much more about achieving an accounting result,” such as discounting loss reserves. “Today, the primary objective is risk transfer,” he said.

And those who are involved in these transactions “are being very careful to document pretty extensively the legitimate risk transfer,” said Howard Mills, New York-based director and chief adviser for the industry group of consultants at Deloitte & Touche USA L.L.P.

Mr. Hochberg estimates that, where finite risk products once accounted for about 5% to 7% of the market, structured products now account for 50% to 60% of that share. “It's still a pretty vibrant market,” he said.

“We've definitely seen a pickup in activity over the last couple of years. The initial runaway has started to move in the other direction,” said Mr. Hochberg.

Painting all the finite products with the same brush was an overreaction, “and I think we're starting to come out of it,” Mr. Hochberg said.

However, Donald A. Paterson, CEO of Los Angeles-based Paterson2, which designs customized risk-transfer products, said insurers still are wary about writing this business because of lingering ambiguity about its accounting treatment.

“Even if you want to do it properly...it's not clear exactly how you should be accounting for it,” said Mr. Paterson. Insurers could go to several different auditing firms and “get different advice from each.”