(FORTUNE Magazine) – It was the year of the derivative. From last spring on, as if a blockade of ice had suddenly given way, bad news about these exotic financial innovations started to flow, and victims, corporate and public alike, began to wash ashore. In the wake of billions of dollars in losses since then, opinions about these new-age instruments have drastically hardened. "Derivatives," observes Richard Syron, chairman of the American Stock Exchange, which trades the species called puts and calls. "That's the 11-letter four-letter word."

The word's elevation to pejorative status is probably justified, but not simply because wild market swings turned many derivatives players into big losers last year. What magnified those losses and sent a troubling message to regulators was disturbing instances of managerial blindness, desperate behavior, even outright fraud. Among the most spectacular misadventures were those of Gibson Greetings, the Cincinnati card and wrapping paper company, which was the victim of lies that were subsequently exposed, Watergate style, by a taped conversation, which we'll listen in on in a bit. The preeminent purveyor of leading-edge derivatives, Bankers Trust, was censured and fined by regulators for its role in Gibson's loss. Enormous complexities delayed investigators in that case, just as the general confusion of derivatives has kept the world unsure of exactly what transpired in most of the other derivatives calamities. Only lately have the arresting details come to light.

The thread running throughout last year's disasters was the misuse of derivatives, now standard equipment in the financial quarters of many companies. When they are employed wisely, derivatives make the world simpler, because they give their buyers an ability to manage and transfer risk. But in the hands of speculators, bumblers, or unscrupulous peddlers, they are a powerful leveraged mechanism for creating risk. Last year the worst sort of crowd grabbed hold of the tool and took over the plant.

Like Gibson, Procter & Gamble was chewed up by derivatives that incorporated astounding leverage and confounding complexity. It is currently engaged in a court fight with Bankers Trust, which sold it the derivatives. Other large companies publicly acknowledged losses, and behind the scenes there were undoubtedly more slips and falls that never came to light. Then there was the nightmare of Orange County, California, which brought to life the dread of many Wall Streeters, a major loss of public money linked to derivatives. Many of these derivatives bear the fingerprints of Merrill Lynch.

In a way that the corporate disasters had not done, Orange County, with its mean effects on millions of citizens, triggered alarm in Washington. But at Senate Banking Committee hearings on derivatives in early January, a troop of top-level regulators were largely reassuring. In particular, they noted the absence of systemic risk last year. That is, no deep problem--extreme distress, say, at a major derivatives dealer--clutched the financial market and, by chain reaction, choked off the liquidity on which the system lives. Fears of such a crisis have ballooned with the prodigious growth of customized, over-the-counter derivatives. But a meltdown obviously didn't happen in the otherwise wretched year of 1994, and that has left regulators feeling relief. At the hearings, they went on to say they did not see themselves as needing new authority to deal with the hazards at hand. Good thing, since most of the committee's Republicans, newly ascended to power, were in no mood for legislation.

But to the ranking Democrat, Paul Sarbanes of Maryland, the Senate committee's attitude, in the wake of this dramatically troubled year, smacked too much of "complacency." Holding up a ten-month-old Fortune cover article about over-the-counter derivatives (March 7, 1994), he read its warnings, among them the fact that these things "make leverage all too easy to come by" and turn existing accounting rules into "hash." He stressed the article's title, "The Risk That Won't Go Away," and read out the subhead that followed: "Like alligators in a swamp, derivatives lurk in the global economy." What, he said in effect to the regulators, are you going to do about this?

Thanks for the plug, Senator, and we will readily accept some credit for our timing. Right after that article was published, the alligators crawled out of the swamp and got their jaws going. But if we may also offer an update, Senator, the add-on news is that horrendous trouble has, as usual, sped change. For one thing, those banking, securities, and commodities regulators who were sitting before you have emphatically toughened up their act. For another, new rules about disclosure have just gone into effect and will be adding a mixture of density and light to 1994 annual reports. On the clarity side, corporations must now state their purpose in using derivatives, a directive that just by its existence may deter speculation.

Even so, there remain plenty of reasons to worry. High on the list is the bedeviling variety and complexity of over-the-counter derivatives, a continuing pain for the accounting crowd and a true mind-bender for anyone trying to value these instruments. Commonly, the tools for valuations are complicated mathematical "models" that factor in, say, estimates of what interest rates will do and with what degree of volatility. The dealers who sell derivatives are customarily the experts on valuation, and many customers--including Gibson, to its sorrow--have simply taken the dealers' word as to what their derivatives are worth.

Then, of course, there is the tendency of markets to do what they wish, including becoming turbulent on occasion, and that means derivatives disasters will multiply. The bad guy last year was interest rates, which rose with a force and suddenness beyond the expectation of the treasurer of P&G, the treasurer of Orange County, and most of the world. The severe drop in the value of the Mexican peso (almost 40% as of late February) has already produced a few derivatives problems, and more are likely to surface. Do you think, perhaps, that the stock market could crater as it did in 1987? Equity derivatives were troublesome back then, and their numbers have since multiplied greatly.

Simon Lorne, general counsel of the Securities and Exchange Commission, recently spoke optimistically about all the steps that regulators have taken to get a grip on the derivatives market. But he also agreed the progress might not match the intensity of the problem: "That's the point Fortune made a year ago. This is the risk that won't go away."

The wrecks at Gibson, P&G, and Orange County point up everything about the risks, including the complexity and confusion of derivatives and their seeming determination to get out of control. But these incidents also raise the definitional problems that have come to both fuzz up this game and add to its mystery.

As the term is most commonly used, derivatives are contracts. They are written between two parties (the "counterparties") and have a value that is derived--that's the key word--from the value of some underlying assets, such as currencies, equities, or commodities; from an indicator like interest rates; or from a stock-market or other index. The options and futures traded on exchanges are derivatives contracts. So are the over-the-counter options and forwards sold by dealers (such as Bankers Trust) and bought by end users (such as Gibson and P&G). All of these instruments are "off balance sheet," a fact that tends to obscure the leverage and financial might they bring to the party.

The cousins of these contracts are derivatives securities, which show up on the balance sheet. In a pure sense, a security of this type has some sort of derivatives contract embedded within it--an option, for example. But as the notoriety of derivatives has increased, this pure definition has been extended to the point that all manner of securities that do not incorporate derivatives contracts are often referred to as derivatives. In short, if it's complex, it's apt to get the name.

To the extent that the Orange County mess was about derivatives, the villain was securities. (For an account of what happened there, see box). But the agent of doom at Gibson Greetings and P&G was totally derivatives contracts, sold in each instance by Bankers Trust, the seventh-largest bank holding company in the U.S. Over the years, Bankers has made itself a specialist in complex, "proprietary" derivatives, turning these into a highly lucrative business. In ads, Bankers has said, "Risk wears many disguises. Helping you see beneath its surface is the strength of Bankers Trust."

Those lines are quoted bitterly in the lawsuit that P&G has brought against Bankers, which has in turn launched a counterclaim against P&G. In another, still more dramatic set of legal actions, the Federal Reserve, the SEC, and the Commodity Futures Trading Commission have all levied penalties on Bankers Trust because of its behavior in dealing with Gibson.

The CFTC's order against Bankers describes the way in which Gibson gorged on a derivatives' diet. Between November 1991 and March 1994, the company entered into 29 derivatives transactions and amendments, managing in the process to deliver $13 million in revenues to Bankers. Many of the contracts, usually because they contained options, incorporated leverage that caused Gibson's losses to increase dramatically in response to small changes in interest rates. Many, according to the CFTC, also had lingo names, among them "the ratio swap, periodic floor, spread lock 1 and 2, Treasury-linked swap, knockout call option, Libor-linked payout, time swap, and wedding band 3 and 6."

"What is a wedding band?" Fortune recently asked Charles S. Sanford Jr., 58, chairman of Bankers Trust and a onetime trader. He suggested we get an answer from one of the bank's technical experts. Next day the company's investor relations director called to relay an expert's statement that a wedding band "was a swap containing a series of barrier options."

He went on, but this lurch into "barrier options" had shut down the listener's mind. After further research, Fortune submits this simplified explanation: A wedding band is typically a swap on which the client makes out well as long as interest rates stay within a relatively narrow range--"the band"--but that turns into a loser if rates move either below the band or above it. A gambler could get similar action by placing money on the total points to be scored in the Super Bowl. Gary Gastineau, head of derivatives research at S.G. Warburg, says that given long enough, he might be able to think of a risk management reason for entering into a wedding band swap: "But that's not really their purpose. These things are done by people who think they know better than the market where interest rates are headed."

GIBSON GETS TAKEN, ON TAPE

Gibson, which definitely thought it was smarter than the market, has nevertheless described itself in legal papers as "a conservatively managed company." This is also the corporation that William Simon, former Secretary of the Treasury, took private in a 1982 leveraged buyout and remarketed to the public a year later, thereby famously adding about $70 million to his net worth. Simon has been out of the company for years, and it is today run by Benjamin Sottile, 57, whose background is consumer products. In 1993, the latest year for which figures are available, Gibson had sales of $547 million and was not big enough to make the Fortune 500. The company's profits that year were $20 million.

In the manner of many other seers, Gibson's financial officers--the boss was then Ward Cavanaugh, now 65 and retired--believed from late 1991 on that interest rates were sure to drop, and they backed their belief by purchasing derivatives. But because of its small size and avowed conservatism, Gibson felt--and claims to have told Bankers Trust--that it could not tolerate a derivatives loss greater than $3 million. Originally that thought was irrelevant, because Gibson's first two derivatives contracts, both closed out within eight months of their inception, delivered Gibson $260,000 in profits. Or at least that is the profit accorded Gibson by Bankers Trust, whose valuation models were the source of all Gibson's knowledge about profits and losses. After Gibson's initial profits, the question of how much the company was making or losing on its derivatives becomes murky for two reasons. First, the two parties, Gibson and Bankers, began chain-linking transactions, so that just what was going on in the first vs. the next became obscure. Second, Bankers Trust people began lying to Gibson about how much money it was losing.

According to the CFTC and SEC, the lies began at the end of 1992 and caused Gibson to release inaccurate financial statements for 1992 and 1993. But more recent evidence of dishonesty is now emblazoned on the record because two Bankers Trust people holding an incriminating phone conversation were taped, caught in fact by an internal system that Bankers uses to monitor trades. In this smoking-gun conversation, which took place on Wednesday, February 23, 1994--about three weeks after the Fed had first begun to tighten interest rates--one of the two talkers worriedly discussed misinformation that Gibson was getting from Bankers and went on to consider strategies for getting out of the problem. Buried among thousands of tapes that Bankers Trust began listening to after various derivatives disasters hit the news last spring, this particular tape did not turn up for months. But when it did, it went to Washington's regulators and became the cornerstone of their disciplinary moves against Bankers Trust.

The essence of the problem described on the tape was that Bankers' data showed Gibson to have lost amounts that far exceeded what Gibson had been told was the case. Here, in sentences that include clarifying, parenthetical phrases inserted by Washington's regulators, is an unidentified managing director of Bankers Trust Securities discussing the "differential":

"I think that we should use this [a downward market price movement] as an opportunity. We should just call [the Gibson contact] and maybe chip away at the differential a little more. I mean, we told him $8.1 million when the real number was 14. So now if the real number is 16, we'll tell him that it is 11. You know, just slowly chip away at that differential between what it really is and what we're telling him."

Later the same day, the managing director spoke ominously of just what would happen if Bankers, looking at the large amounts Gibson now owed it, had to tell Gibson that it must stop the bleeding by "unwinding" its trades. In that case, of course, the truth would come out. Said the managing director: "We gotta try and close that gap...If the market hasn't changed at all, or was just kind of dottering around within a couple of ticks, then you know, there's nothing that we can really say...But when there's a big move...and he is down another 1.3, we can tell him he is down another two. And vice versa. If the market really rallies like crazy, and he's made back a couple of million dollars, you can say you have only made back a half a million."

Perhaps because interest rates pressed ever higher, the gradual return to reality that this conversation contemplates did not occur. On the Friday after the Wednesday of the taped coversation, Bankers Trust told Gibson that its losses had increased from the $8.1 million to $13.8 million. By the following Thursday, the figure was a remarkable $17.5 million, and Bankers, so says Gibson, was describing the company's losses as "potentially without limit." Gibson therefore took action: It signed up the next day for one amendment and one new derivative --the 28th and 29th of its transactions--that capped its loss at a maximum of $27.5 million but also held out the possibility that it could reduce its loss to $3 million. Grasp the implications of that move: At that moment, Gibson could have settled with Bankers Trust for less than $27.5 million-- perhaps something close to the $17.5 million. But gripped, no doubt, by desperation, Gibson took the chance of entering into new transactions that had the potential of increasing its losses. "I tell you," says a director of another company that reeled into derivatives trouble last year, "it's a lot like gambling. You get in deep. And you think, 'I'll get out of it with this one last trade.' "

After the new agreements, Gibson's losses in fact rose. As of September 30, 1994, they were up to $20.7 million. But just before that date, Gibson swerved in strategy, suing Bankers Trust in an effort to get out of paying on the deals. Gibson claimed it had been misinformed and misadvised by Bankers, which Gibson charged had a "fiduciary relationship" with its client. In October, Bankers denied those allegations and launched a counterclaim that presented Gibson--a "large and sophisticated corporation," said the legal papers--as simply the victim of its own rotten judgment.

And then Bankers Trust found the smoking-gun tape. Without knowing of Bankers' discovery, the world soon got the impression of wheels whirring. To begin with, the news leaked that Bankers was reassigning five executives, pending an internal investigation into certain derivatives deals. Next, Bankers announced that Gibson would pay it a settlement amount of $6.2 million, obviously a much-reduced figure.

The names of three of the executives reassigned have never been disclosed. The other two were managing directors of BT Securities: Jack A. Lavin, who headed the sale of corporate derivatives, and Gary S. Missner, who reported to Lavin and handled the Gibson account. Lavin, who continues to work for Bankers, was definitely not a participant in the taped conversation. Missner definitely was. He has left the bank and so has the other participant in the taped talks. Meanwhile, the internal investigation continues.

The tape's existence became publicly known only when the CFTC, the SEC, and the Fed came down hard on Bankers Trust in late December. The Fed's "written agreement"--a serious kind of sanction--requires Bankers to rework and greatly improve its procedures for selling leveraged derivatives and to hire an independent legal counsel to consider what disciplinary actions against Bankers' people might be appropriate. The agreement specifically directs the counsel to inquire into whether management failed to supervise its employees.

The CFTC's and the SEC's "orders" against Bankers charge it with both giving Gibson wrong information about its losses and causing Gibson to issue inaccurate financial statements. The SEC throws in a "failure to supervise" charge, and by its very presence on the scene makes the new point that certain kinds of over-the-counter derivatives are securities subject to its jurisdiction. Finally, in an allegation that has shocked many a derivatives dealer who never wanted to be a fiduciary of any kind, the CFTC finds Bankers to have been a "commodity trading adviser" that defrauded its client.

This allegation is tied to still other incriminating tape conversations, in which a Bankers managing director says, "From the very beginning, [Gibson] just, you know, really put themselves in our hands like 96%." The same executive acknowledged that Gibson was no derivatives whiz: "These guys [Gibson] have done some pretty wild stuff. And you know, they probably do not understand it quite as well as they should...And that's like perfect for us."

For all of these misbegotten deeds, the CFTC and the SEC jointly fined Bankers Trust $10 million. That is not a huge fine by SEC standards (its largest was the $300 million levied on Drexel Burnham Lambert in 1989 for insider trading and other violations of securities laws), but it is big stuff for the CFTC. Mary Schapiro, chairperson of the CFTC, says the $10 million "is not unrelated to the $13 million in revenues that Bankers Trust took in from Gibson." Could it also, she is asked, be roughly equivalent to Bankers' profits on the Gibson business? She says she doesn't know but "wouldn't be surprised" to find that true.

The SEC and CFTC orders state that Bankers has consented to the regulators' findings, without either admitting or denying these. Informally, the company has said wrongs were clearly committed. But CEO Sanford tends to believe the problem was simply "two guys" guilty of rogue behavior. In general, he says, "I think our people have been acting very fairly with customers."

He is asked to what extent he thinks top management bears responsibility for what happened with Gibson. "I don't think they-- That would be pretty hard. There's four or five levels of management below. I don't know what people are saying on the phone or doing until it's brought to our attention."

Some people who have worked at Bankers claim, nevertheless, that the Gibson behavior grew out of a culture that puts an almighty importance on profits. Says a former Bankers managing director who held a responsible position in its derivatives business: "This is very much a management issue. Any sales force is extremely sensitive to management. If you go for several years paying and promoting certain kinds of salespeople, the message gets across that what they do is acceptable behavior."

As of now, new rules about acceptability are in place at Bankers. Under the Fed agreement, Bankers is directed to make sure that its customers know about every wart, wrinkle, and whisker of their leveraged derivatives. Most significantly, Bankers is required to provide "transparency" about prices. If a client, for example, has entered into a highly leveraged contract, it is entitled to know on a daily basis what the contract's value is.

Along the way, that requirement seems likely to provide clients with new information about what Bankers is taking in on its trades. Imagine that a client asks on Day Two of a transaction what the contract's value is--in other words, what it would cost him to undo it. The client would be aware that he buys at an "offered" price and sells at a "bid" price. So when Bankers comes forth with an answer about the derivative's Day Two price, it will in effect, assuming that the market has been quiet, be divulging the "spread, " or revenues it booked on Day One. Revenues in this business certainly aren't equivalent to profits, since they must cover compensation and the sometimes sizable costs that a dealer will incur trying to hedge the risk it has just shouldered. But as a rough indicator of Bankers' take on the deal, the figures should be of keen interest to clients.

At the hearings in January, Fed Chairman Alan Greenspan said unequivocally that the Bankers Trust agreement should not be construed as setting general guidelines for the industry. Even so, other big derivatives dealers have been poring over the agreement, trying to understand what it means for them. Will they also be forced, perhaps even by customer pressure, into daily valuations? On a kind of ludicrous level, but with legal problems definitely in mind, should they do more taping or less taping? And what about that huge question of their fiduciary responsibility to the corporate clients with which they deal? Most dealers have thought of their customers as sophisticated types not needing handholding. But does the Gibson case suggest that the dealers must get deeply into what's appropriate for customers to own? That is, in fact, a question right now on the plate of Washington's regulators, who have so far not produced an answer.

P&G: A JURY'S NIGHTMARE

Any summation of the Gibson case would have to conclude that this company had no business signing up for leveraged derivatives. Beyond that, a match pitting Gibson against Bankers, with all its erudition in derivatives, was indisputably an unfair fight. But what should we think when the bout is between P&G and Bankers? Can P&G really be classed a victimized innocent? In a recent speech, Merton Miller, a well-known University of Chicago finance professor, got out the needle and delivered his answer: "You know Procter & Gamble? Procter is the widow, and Gamble is the orphan."

P&G is doggedly claiming, nonetheless, that in late 1993 and early 1994 it was cheated by Bankers into buying two swaps having "huge, concealed risks." One of P&G's complaints is that it was the prisoner of "a secret, proprietary, complex, multivariable pricing model" that Bankers would not share and whose inscrutability stood in the way of P&G figuring out how to mitigate its losses. Certainly P&G did not marvelously mitigate. Its derivatives loss, announced last April, was $157 million before taxes. Essentially, that was the present value of super-high payments that the company, by the terms of its two swaps, was obligated to make to Bankers Trust over the next several years.

P&G, however, has not yet paid a penny on the swaps, and in its lawsuit is asking that they be rescinded. It wants a jury trial, a mind-bending fact considering the complexities of this case--hardly the stuff for Court TV. Bankers, meanwhile, has denied wrongdoing of any kind and has countersued, asking that P&G be forced to hand over all it owes.

Here is a vastly simplified description of the morass that P&G got into--and that a jury would have to boot up for also. This is a company that over the years had done many kinds of business with Bankers Trust and that throughout the banking world had a reputation for aggressively managing its interest costs. In the early 1990s, for example, P&G swapped good-size quantities of its fixed-rate debt into floating-rate debt, thereby successfully betting on a drop in rates. At the wheel in this period was treasurer Raymond D. Mains, now 56, who from all appearances regarded his operation as a profit center.

In October 1993, still expecting rates to fall, P&G talked to Bankers Trust about ways of replacing a fixed-to-floating swap that was maturing. P&G's specific objective was to negotiate a new $100 million swap that would (a) again put it in the position of paying floating rates and (b) squeeze these to a minimum. Specifically, the company wanted to pay its standard, upper-crust commercial paper rate (then about 3.25% for six-month paper) minus 40 basis points--that is, 0.4 of 1%.

No problem, said Bankers Trust. In the way of big dealers and especially itself, Bankers was prepared to tailor-make a contract to fit the client's wishes and then to cover its own flanks by hedging the risk it had just taken on. In interest rates, however, as in life, there is no free lunch. If P&G were to do this deal, it would be required to take on extra risk, which would come in the form of reduced returns if interest rates did the unthinkable and went up. Keep talking, said P&G. Bankers then proposed several different deals, one after another, and to each P&G said no. This ping-pong had the feel of a never-give-up stock salesman saying, "Okay, if you don't like that one, how about this one?"

Finally a deal was struck. In early November, suddenly raising their sights to $200 million rather than half that, the two parties signed up for a five-year swap whose leverage sprang from an option included within it. For the first six months of the deal, P&G was to pay a floating rate not just 40 basis points, but 75, below commercial paper rates. For the 41/2 years after that, the floating rate was to be dictated by a brain-twisting formula whose components would include five-year and 30-year Treasury rates as of May 4, 1994, which was the six-month anniversary of the deal. Under the best case for P&G, the floating rate would continue at 75 basis points below commercial paper for the full term of the swap. Under the worst case--well, that's what the lawsuit is about.

For all the deal's complexity, the gist of it can be stated quite simply: The swap had a "notional," or principal, value of $200 million. Assume that P&G scored to the maximum, saving 75 basis points for five years. On the $200 million, that would be $1.5 million a year, for a total of $7.5 million. The annual savings would have cut P&G's interest bill, which runs around $500 million, by 0.3 of 1%.

And what risk did P&G accept in return? For this lure of $1.5 million annually--think of that as kind of an insurance premium payable to P&G --the company agreed in effect for the next six months to act as an insurer covering the risk of interest rate earthquakes. With remarkable fury, these quakes then occurred: Five-year Treasury rates rose from 5% in early November 1993 to 6.7% on May 4, 1994, a dramatic increase. P&G's other benchmark, 30-year Treasury rates, went from about 6% to 7.3%.

At an early point in this action-packed period, P&G went so far as to increase the money it had on the table. On February 14, 1994, just ten days after the Fed tightened rates, the company entered into another highly leveraged swap geared again to the idea that rates would not soar. This swap had a principal value of about $93 million, a term of 43/4 years, and was--ready?--a wedding band. Or perhaps, since it was denominated in deutsche marks, it should be called a Trauring.

We will not translate the terms of the swap into German, since they are difficult enough in English. This swap bestowed a very favorable floating interest rate on P&G for the first year of the swap and, over its full term, offered the promise of about $940,000 in total savings if everything went right with a certain German "swap rate." These savings would result if the rate, then 5.35%, did not fall below 4.05% or rise above 6.10% at any point before April 14, 1995.

On the other hand, if the rate popped out of that band, even for a day, another crazy-quilt formula took over. Under this formula, the level of the swap rate on the precise day of April 14, 1995, and its relationship to 4.50%--yes, we know that's a percentage you haven't seen before in these paragraphs--would determine what interest rate P&G paid for the last 3 3/4 years of the swap (are you getting all this, ladies and gentlemen of the jury?). The danger, then, for P&G occurred, first, if the swap rate jumped out of the prescribed band and, second, if the swap rate was above the 4.50% benchmark rate on April 14, 1995. In that event, P&G was to begin paying interest that included its base rate for the first year plus a "spread." And this spread was ten times the difference between 4.50% and the swap rate. So once again P&G had sold earthquake insurance.

As it happened, the magic date of April 14, 1995, never precisely came into play. Just 16 days after the contract went into effect--we are now in early March 1994--the swap rate flew out of the band on the upside. From all appearances, P&G then started to realize just how catastrophic things could get with its swaps if rates kept going up. It promptly began trying to mitigate the dangers by negotiating a "lock-in interest rate" for each of its swaps. In other words, it wanted to firmly set the rates that it would pay for the duration of the swaps, rather than leave itself at the mercy of interest rate movements.

Therein lies the crux of the legal dispute between P&G and Bankers. P&G claims that before the swaps were signed, Bankers repeatedly assured it that in the early stages of the swaps, the company would be able to do lock-ins at acceptable prices. Court papers, in fact, include letters from Bankers that make such assurances, though these consistently cite assumptions of stable or only slightly rising rates. P&G says, however, that on one occasion it "pointedly" asked the Bankers Trust person with whom it was dealing what the lock-in situation on the first swap would be if rates and volatility were not "stable." The answer, P&G says, was that "possible changes in rates or volatilities would not have a material or significant effect" on the company's lock-in position.

But when P&G commenced to negotiate lock-ins, so it says, it found itself frustrated by that "secret, proprietary, etc." pricing model and by loss figures that it found dismaying. Bankers Trust, says P&G, actually attributed part of the lock-in problem to the protective hedges the bank had itself set up after making the first swap contract. The hedge position was so large, P&G was given to understand, that unwinding it might itself drive up U.S. interest rates! Eventually, P&G came to believe that Bankers Trust, trying to cover all the leverage that had been packed into this one $200 million contract, had put on no less than $3 billion in hedges.

In the end, dismaying loss figures or not, P&G was forced to accept the deals that Bankers offered. By April 11, P&G had locked in rates on both swaps and confronted the horror of insuring quakes: For the duration of the first swap, which runs to late 1998, the company agreed to pay interest rates that are 1,412 basis points (14.12%) above the commercial paper rate. And as long as the wedding band binds--that's until late 1998 also--P&G must pay rates 1,640 basis points (16.40%) above the base rate specified by the swap.

Treasurer Mains left P&G last September. CEO Edwin Artzt, a man of a well-known and no doubt recently exercised temper, has called the swaps "a violation of the company's policy against speculative financial transactions" and banned all leveraged swaps henceforth. Hans Stoll, a finance professor and derivatives expert at Vanderbilt University, says the swaps are simply "not something that a corporation that manufactures things should be involved in." Imagine, in fact, an investor who bought P&G stock in early 1994. He would have thought he was buying soap, cosmetics, and a few cups of coffee. But wedding rings?

It is not yet clear what role tapes may play in the litiga- tion between P&G and Bankers Trust. Amending its lawsuit recently to include the wedding-band swap, P&G said that in the process of legal discovery it had secured tapes from Bankers Trust that "confimed" P&G's belief that it had been misled when entering into that swap. But were this another instance of a smoking-gun tape, Bankers would surely have settled, and it has shown no inclination to do that.

MORE TAPES, MORE TROUBLE

There is one other derivatives-troubled company that appears to have reaped a tape bonus: Federal Paper Board, a Montvale, New Jersey, company of Fortune 500 size. Last spring Federal emerged as one of several companies caught in a change that required certain highly leveraged derivatives to be shifted from "accrual accounting," which allows the gradual recognition of gains and losses, to "mark-to-market accounting," which requires prompt and full recognition. The change caused Federal to restate its 1993 profits, reducing these from $20.8 million to $6.4 million, and to begin taking mark-to-market losses in 1994 as well.

But recently it became clear that Federal was enswirled in other derivatives dramatics besides. In December the company announced that two foreign currency contracts written with Bankers Trust had been canceled, a move that resulted in a $12 million gain for Federal. Both sides avoided further explanations. But Fortune has learned from someone close to the situation that Bankers found "a Gibson-like interlude" in taped conversations about Federal, which probably explains the settlement.

If that took care of one derivatives problem, Bankers has plenty of others. It acknowledged as much in January by making special provisions for $423 million that it was owed on derivatives. It reclassified these debts as receivables in the loan account and started carrying them on a "cash" basis, a designation meaning that Bankers will book interest on the loans only when it is actually paid. Right after making that move, Bankers charged off $72 million of these debts to its reserve for credit losses, thereby treating these as uncollectible. The remaining $351 million, says Bankers, sweeps in both P&G and other leveraged contracts that "likely will not perform." These could, for example, include swaps that Bankers entered into with Air Products & Chemicals, which announced large derivatives losses in the middle of last year and is believed to be still negotiating with Bankers about their disposition.

The current state of the derivatives market may be suggested by the 1994 results that Bankers reported for the business it calls "client financial risk management." This operation had a splendid first quarter, earning $114 million. After that, things went south. By the fourth quarter, with the papers full of derivatives headlines and the financial markets jumpy, earnings were down to only $28 million. For the full year, this business had profits of $259 million against $336 million earned in 1993. Another 1994 casualty was Bankers' stock. The price fell from a February all-time high of $85 to a December low of $55. Recently the stock was about $63.

In the midst of this trouble, one of Bankers' competitors, Citicorp, ran an ad containing a definite dig: "You expect derivatives to solve problems, not create them." Asked about that ad, Bankers' Sanford looks pained, but says: "Well, they can do whatever they want." In January, Bankers itself ran multipaged year-end ads that focused on its skills in another part of its business, investment banking. But the ads ended with an admission of problems with leveraged derivatives. Bankers said it was both taking a "critical look" at its procedures and reaffirming its commitment to "uncompromising standards." Then came a wind-up mea culpa: "The loss of even one client is a stinging lesson. Lesson learned."

The future of derivatives is an uncertain matter, depending partly on whether the subject is leveraged derivatives or all others. In general, says Sanford, the use of derivatives will grow. He even sees the misadventures of the past year as having had a salutary effect, in the sense that many CEOs, learning about derivatives for the first time, came to understand their "positive effects." Asked for names of CEOs who might be representative of that breed, Sanford declines to give any: "Nobody's going to come out and say anything publicly, because they get skewered by the press and everybody else. You know, 'So-and-so's in derivatives.' " Fortune itself ran into evasions as it called companies to inquire what they were doing in derivatives. You would have thought we'd asked about incest.

On the other hand, it is possible to find burn victims who will themselves deliver testimonials to the matches that got them. Listen to P&G boss Artzt: "Straightforward derivatives--or, as the financial community calls them, plain vanillas--are a very effective way of managing interest rates and foreign exposures. So we plan to continue using them." Most regulators seem to agree with his view: It is the rare regulatory speech that does not stress the risk management benefits that derivatives offer.

Leveraged derivatives are a subject that regulators find harder to address, partly because they do not really wish to arbitrate the difficult question of what's risk management and what's speculation. Said Fed Chairman Greenspan at the January hearings: " 'Speculative' is a very fuzzy term."

In the end, leveraged derivatives may fall of their own weight. The commotion of the past year has, for the moment, made them a drug on the market. Many boards have indeed banned their use. In addition, the change last year in which mark-to-market accounting began to be required for highly leveraged contracts is itself a turnoff. Many companies despise mark-to-market accounting and will steer clear of those leveraged dudes if they come packing that kind of baggage.

The next question, though, is whether "mildly leveraged" derivatives will somehow be forced into the same mold. Many regional banks, among them Banc One and PNC Bank Corp., manage their interest rate risk by entering into contracts called "index amortizing swaps." These swaps include a gentle form of option and are therefore considered to be mildly leveraged. Normally, options must be marked to market, but these instruments have so far escaped that treatment. A change would be important. Based on information provided in footnotes to the banks' financial statements, it is clear that a switch to mark-to-market accounting would force many of these low-leverage derivatives players to recognize significant losses on their positions. So, is an accounting change coming? Says John T. Smith, head of Deloitte & Touche's financial instruments research group: "You just can't tell. But, in my opinion, it should. An option is an option, and I can't see why you should give the mild ones special treatment."

That's one derivatives surprise that might materialize, and there will surely be others. It would be reassuring to think that regulators could anticipate all the shocks that derivatives could deliver and minimize their impact. But the financial world, and especially its derivatives component, moves so fast that total foresight just isn't possible. One bank regulator understanding that well is Douglas Harris, senior deputy comptroller for capital markets at the Office of the Comptroller of the Currency and formerly a lawyer in the derivatives operation of dealer J.P. Morgan. Harris was brought into the OCC specifically to build up its expertise in derivatives, and he has been an important figure in their regulation. But he concedes a timing problem: "From the day I got here, I felt I was falling behind what was going on in the market. That doesn't mean we can't supervise. It just means we're always running to keep up."

Given the range of complications that derivatives present, outside directors cannot possibly achieve close communion with the contracts their companies hold. Most chief executives won't master the game, either. In the end, the choice of what risks to hedge, what derivatives to employ in doing it, and how to draw the bright line between risk management and speculation will be largely left to financial people down the corridor --some of whom, recent train wrecks notwithstanding, may think of themselves as running a profit center. And on the other end of their phones will be derivatives salespeople trying to sell the latest innovation, which assuredly will not be a plain-vanilla hedge.

It's not a particularly cheerful picture-not for a problem as big as derivatives. So maybe what we need is new thinking, a fresh approach, a suggestion so radical it goes off the page. Here's one: Warren E. Buffett, chairman of Berkshire Hathaway, says he'd deal with derivatives by requiring every CEO to affirm in his annual report that he understands each derivatives contract his company has entered into. Says Buffett: "Put that in, and I suspect you'll fix up just about every problem that exists." In a market that seems to thrive on complexity and obfuscation, such a solution won't happen. It's too simple. But he's right.