Good night, finite? While regulators step up their scrutiny of past finite reinsurance transactions, industry watchers say there's still a valid place for the nontraditional transactions, if insurers toe the regulatory line.

A rose by any other name would still smell as sweet. But if Shakespeare's roses had come under regulatory investigation, he, too, might have considered changing the name.

Finite reinsurance, a nontraditional form of reinsurance, has been the subject of recent investigations. Companies have been accused of using the insurance tool its a balance sheet shell game, a convenient way to hide bad news from investors, auditors, accountants and regulators, without actually transferring any risk. Today you won't find many people in the insurance industry trying to sell "finite reinsurance" products. Instead, you'll find business cards offering such things as "hybrid solutions," "alternative risk transfer," "structured finance" or the ever ambiguous "reinsurance solutions."

It's not the first time the complicated transactions have fallen under regulatory scrutiny, nor the first time the transactions have been renamed. After Hurricane Andrew, and various accounting changes, "financial reinsurance" became known as "finite reinsurance."

Finite reinsurance transactions can be used to smooth the results of primary insurers as well as the earnings of non-insurance companies. Traditional reinsurance allows insurers to pass along a portion of their premiums and risk to reinsurers, who promise to pay if future losses reach a set level. Finite reinsurance is intended to carry a portion of the ceding company's risk for a set period of time, and can include coverage for losses that already have occurred.

Even though some companies already have paid fines for alleged bad deals that bore no risk, and subpoenas into other transactions continue to fly, industry experts say there is a legitimate place for finite reinsurance--even if no one but regulators and rating agencies call it by that name any more.

They say the examples of "bad" transactions are not the norm, and insurers must become more vigilant in insuring that every transaction they make is valid.

Fraud or Form?

The insurance industry has been around for hundreds of years, but finite or financial reinsurance is a fairly new insurance product, so it makes sense that creative insurance leaders, who are the first to try new products, are now in the hot seat for these transactions, some said. Call it financial reinsurance, hybrid solutions or alternative risk transfer, these transactions tend to be tailored to match each client's specific need.

"That's the heart of the point," said Stephan Christiansen, director of research at Conning Research & Consulting. "This is an evolutionary product, and it has been for more than a decade. At one time, it wits referred to as 'time and distance' policies out of London and Bermuda, when it was strictly reinsuring the time that it would take for something to happen, as opposed to covering the amount. The reinsurance level of risk was based on the pace of payment, not the amount they'd pay."

What concerns regulators is the question of when should a transaction be considered a transfer of risk--which is insurance--compared with a simple loan. Each is accounted for differently.

Dan Malloy, executive vice president of ART Solutions for Benfield, said there may have been some deals that did not transfer risk. But that doesn't mean the whole system should be scrapped.

"Because someone bounces a check, do you say no one should be allowed to write a check?" Malloy said. "The vast majority of transactions are valid. There's a long list of companies that wrote this business from 1997 to 2001 and lost billions and billions of dollars. Given the poor underwriting results that are now apparent, there wits significant underwriting risk transferred to finite reinsurers, probably much more so that was originally expected."

No Risk, No Insurance

The test, Malloy said, should be what does it take to have a legitimate transfer of risk?

"In retrospect, it's easy to look back and say this deal made money or that deal lost money: That, however is not how to test for risk transfer. The process of considering whether the deal should be considered reinsurance or not is a family complicated actuarial and mathematical process which looks at discounted cash flows of a range of potential outcomes, which is performed at the inception of a contract. Auditors who review these transactions have to come to their own conclusions based on these projections, but they also make other 'substance based' opinions, such as potential impairment of intangible assets," Malloy said.

This isn't the first time that financial reinsurance or finite reinsurance has come under regulatory concern, said Frank Cacchione, a member of PA Consulting Group's management group. In 1992-93, in the wake of Hurricane Andrew, the Financial Accounting Standards Board stepped up regulation to prevent companies from undertaking financial reinsurance transactions that were essentially designed to smooth their earnings without any risk transfer. "True, good valid finite reinsurance hits its place, but bad behavior is bad behavior," Cacchione said.

After Hurricane Andrew, when the term "finite reinsurance" was born, EASB and other post-Andrew accounting changes established a benchmark to define risk transfer as a 10% chance of having a 10% loss, Christiansen said.

"It's very arbitrary. A lot of finite reinsurance was set up to be around whatever the definition was, based on the actuarial losses and the risk of loss," he said. "My guess will be that this will be a readjustment, one more time. People are pushing the envelope in terms of creative solutions to very complex corporate risk problems, and they are competing with very creative solutions from other segments of the industry, capital markets, investment banks and the like. That's why it's hard to regulate."

If a company experiences a loss, it can borrow from a bank to finance it, or enter into some other transaction that pays for it, Christiansen said. "Borrowing is very clear, but the other transaction is not on the balance sheet, even though you might still have an obligation to pay it back. That's distorting the balance sheet, hiding things from the marketplace," he said.

Some transactions aren't designed to hide anything, but to spread out the loss. "It seems to me there should be legitimate purposes for smoothing risks, just as insurance and reinsurance itself is a smoothing of unexpected losses," Christiansen said.

"One of the challenges to the sellers and buyers being examined would be the concern over whether those different purposes would be understood. I think in most cases, it's the customer who's potentially misleading the financial markets; the reinsurer is providing capital market solutions to things."

A Question of Disclosure

The National Association of Insurance Commissioners' Property and Casualty Reinsurance Study Group is currently examining the issue. Joseph Fritsch of the New York State Insurance Department noted the difficulty in defining or identifying finite contracts, and he cited one published report that listed some features common, but not exclusive, to finite reinsurance. These include using the time value of money to bring the funds backing the contract in line with expected losses; provisions that confine the reinsurer's potential losses to a relatively narrow range; refunding or additional payment of premiums based on loss experience under the contract; slim underwriting margins coupled with fees paid to the reinsurer; and commutation clauses under which the reinsurer refunds the premium but keeps the fee.

Fritsch said regulators often see a pattern of repeated commutations over several years, suggesting that as a practical matter the reinsurer merely is holding deposited funds under the guise of a contract that allows the cedant to take credit for reinsurance on its financial statements--with no real transfer of risk.

The regulators suggested there are large gaps in definitions, disclosure requirements and industry practices, all of which make it difficult for regulators and others to get a clear picture of how the contracts operate and what their purpose is. "We've got this vague risk-transfer test," said Mike Moriarty, director of the Capital Markets Bureau for the New York department, referring to the definitions under statutory accounting principles. Many contracts pass the test but in reality are designed to aid the cedant's financial reporting, he said.

Fritsch said regulators sometimes see the same contracts reported differently in filings with the U.S. Securities and Exchange Commission or with rating agencies than they are under statutory accounting.

It also can be hard to find the documentation behind a reinsurance contract, Fritsch said. He suggested that if a cedant can't produce an underwriting file for a contract, it shouldn't be credited as reinsurance.

The regulators also noted that finite reinsurance sometimes comes with "side agreements" or verbal understandings that hide the true intent of the parties. Fritsch stressed that finite reinsurance isn't necessarily bad, but that greater disclosure would better equip regulators to make an informed judgment.

One example of the alleged use of side agreements is a lawsuit that regulators in Virginia and Tennessee have filed against Berkshire Hathaway's General Re, saying it was complicit in the demise of Richmond, Va.-based malpractice insurer Reciprocal of America. The suit claims that Gen Re did not disclose secret side agreements with the reciprocal that reduced the amount of risk that Gen Re actually shouldered. Gen Re has denied any wrongdoing.

While not commenting on any specific case directly, Christiansen said any side letter or agreement that isn't included in the formal documentation "strikes me as much more questionable."

A.M. Best Co. asks companies to reveal details of finite reinsurance in their filings.

"The key issue in the use and evaluation of finite reinsurance is disclosure," A.M. Best Co. said in a written statement. "Through our Supplemental Rating Questionnaire and meetings with insurance company management, A.M. Best requests information relating to finite reinsurance contracts being used by rated companies. In those cases where the information is disclosed, A.M. Best considers these transactions in our rating analysis. A.M. Best does not audit an insurer's financial records or statements."

Benfield's Malloy said A.M. Best "has done a great job of getting this information."

He said those who are charged with monitoring the financial solvency of insurers, including regulators, are likely to continue to push for disclosure. "I think there's going to be more attention paid to disclosure, and more questions asked about the substance of a transaction," Malloy said.

Also, he said he expects insurers will be more vigilant in writing the contracts, and that auditors also will step up their reviews.

"In the past, the more complicated a contract got, the more likely it was to be approved," Malloy said. "There's been a whole market shift. When you had auditors taking shares in new companies instead of getting paid, and an extremely soft market, and the meter was more pegged toward greed than fear. Then along comes 9/11, Enron and Arthur Anderson, and the environment has changed completely. Underwriting discipline is much more apparent and auditors are much more conservative. What you have is a healthy environment now."

He estimated that finite reinsurance may represent just 5% to 10% of total reinsurance premium in the industry. An active finite reinsurer may do to or 15 transactions a year, where traditional reinsurers do hundreds, if not thousands of transactions. Finite reinsurance remains a highly customized industry, however, and there's no single product that's widely sold, Malloy said.

"But it still remains, and will continue to be, a very important option to the traditional reinsurance product," Malloy said. "It will continue to be a valuable tool in our tool kit, because there will always be people looking for alternatives."

Learn More

General Reinsurance Corp. (A Berkshire Hathaway Company)

A.M. Best Company # 02198

Distribution: Direct

American International Group 

A.M. Best Company # 05953

Distribution: Direct, agents and brokers

For ratings and other financial strength information about these companies, visit www.ambest.com.

Key Points

* Finite reinsurance represents perhaps 5% to 10% of the total volume of reinsurance premium in the industry.

* A number of industry leaders have been subpoenaed by regulators investigating the legitimacy of certain finite reinsurance agreements.

* Regulators are currently evaluating existing rules to see if the guidelines concerning the transactions should be tightened.

Types of Transactions

Finite reinsurance transactions, also referred to as financial reinsurance or structured reinsurance, can be cost-effective alternatives to traditional covers and can cover either past or future losses. While each transaction is unique, they can include the following types:

* Adverse Development Cover/ Loss Portfolio Transfer: Addresses old liabilities and permits management to focus on ongoing business. Can include transfer of claims management.

* Structured Quota Share: Allows access to traditional pro rata protection while allowing the client to retain a share of the positive economics.

* Catastrophe Excess: Uses multiple years of coverage to reduce reinsurers' risk charge.

* Aggregate Stop Loss: Provides whole account protection against both frequency and severity of loss.

Source: Benfield Group 

Innovators on the Front Line

A virtual who's who of insurers has come under scrutiny for their work in finite reinsurance. American International Group, Ace Ltd., Platinum Underwriters Holdings Ltd., MBIA Inc., St. Paul Travelers Cos., Swiss Re, Zurich Financial Services Group and Berkshire Hathaway Inc.'s General Re have all received subpoenas or requests for information from regulators.

During the past few years, Berkshire Chairman Warren E. Buffett has addressed the company's participation in financial reinsurance in letters to company shareholders. In his 2000 letter, Buffett said the practices involve "transactions in which we assume past losses of a company that wants to put its trouble behind it."

"To illustrate, the XYZ insurance company might have last year bought a policy obligating us to pay the first $1 billion of losses and loss-adjustment expenses from events that happened in, say, 1995, and earlier years," Buffett wrote. "These contracts can be very large, though we always require a cap on our exposure."

Buffett said the program, which the company calls "retroactive insurance," neither benefits nor penalizes Berkshire's current earnings under generally accepted accounting principles.

"Instead," he wrote, "we set up an asset called 'deferred charges applicable to assumed reinsurance,' in an amount reflecting the difference between the premium we receive and the (higher) losses we expect to pay (for which reserves are immediately established).We then amortize this asset by making annual charges to earnings that create equivalent underwriting losses. ... By their nature, these losses will continue for many years, often stretching into decades. As an offset, though, we have the use of float--lots of it."

While many investigations or lawsuits are ongoing, AIG resolved one set of claims by agreeing to a $126 million settlement with the U.S. Securities and Exchange Commission, the U.S. Department of Justice and federal prosecutors over structured finance deals involving Brightpoint Inc. and PNC Financial Services Corp.

As part of the settlement, AIG agreed to appoint an independent consultant who will review transactions entered into between 2000 and 2004 to determine whether the transactions were used by a counterparty to violate generally accepted accounting principles or to obtain a specified accounting or reporting result. The settlement also requires AIG to establish a transaction review committee, and the independent consultant will review the policies and procedures of the transaction review committee. The company admitted no wrongdoing in agreeing to the settlement.

In announcing the settlement in November, AIG Chairman Maurice R. Greenberg said in a statement: "This proposed settlement will be comprehensive and bring finality to the claims raised by the SEC and the DOJ. It is in the best interests of our shareholders, customers and employees to reach this settlement."

Those specific claims may have been resolved, but in February, AIG said it received additional subpoenas requesting more information on some of its finite reinsurance transactions from the SEC and New York state Attorney General Eliot Spitzer.

Test for Risk

For a structured transaction to count as reinsurance, it must demonstrate:

* That the reinsurer has assumed significant insurance risk. An examination of the contract wording should find that the amount and timing of reinsurer payments are not the same as those of the cedant.

* That the reinsurer has a reasonable probability of a significant loss. Mathematical and actuarial analysis should find there's a 10% chance of a 10% loss.

Source: Benfield Group

Why Undertake Finite Reinsurance?

Companies may turn to finite reinsurance as an alternative to traditional products for several reasons, including:

* The company's loss experience is much better than average, making reinsurance "too expensive."

* The company's loss experience is much worse than average, making reinsurance unavailable.

* The company wants to exit lines of business.

* The company wants to increase writings or take a larger retention in existing portfolio, but is constrained by limited capital.

Source: Benfield Group

--Brendan Noonan contributed to this article.