Chapter Three

Joe Jett and Kidder Peabody, 1994

Born in 1847, he only attended school for three months—the victim of what his teacher called a “wandering mind.” In fact, the young man was so often distracted that his instructor referred to his brain as being “addled.” It was, to say the least, a less-than-complimentary way to describe the pupil, and that student would be saddled with it throughout his academic career. Then, to make matters worse, our young hero was stricken with a case of scarlet fever that left him partially deaf. His mother decided that the only remedy for his wayward thinking was to teach him at home, relying heavily on readings from School of Natural Philosophy by R. G. Parker.

His family later moved to Port Huron, Michigan, where the young man got a job working on a railroad car, selling candy and newspapers to travelers as a way of making a living. During his free time on the train, he’d perform various experiments with chemicals that were available to him—an indication that perhaps his earlier diagnosis of “addled” was misguided. It seemed that he was more curious than mentally confused.

That innate streak of curiosity—paired with an entrepreneurial instinct—led the young man to become an inventor, where he achieved a marked degree of success. By the time of his death, he held 1,093 U.S. patents, making him the fourth most prolific inventor in history. In addition to inventing the phonograph, the stock ticker, and the carbon microphone used in telephones, he is also credited with inventing the first commercially feasible incandescent lightbulb.

That student who had so much trouble paying attention was named Thomas Edison, and would go on to found fourteen different companies, including today’s commercial giant called General Electric.

Just four years after Edison’s death in 1931, a child was born to parents of Irish descent living in Peabody, Massachusetts. His parents named him after his father, John Welch, but they decided to call him Jack. John Sr. was a railroad conductor, his wife a homemaker. Jack attended Salem High School where, unlike Edison, he excelled. He went on to the University of Massachusetts at Amherst, from where he graduated in 1957 with a bachelor’s degree in chemical engineering.

Jack Welch joined General Electric in 1960 after earning a PhD in chemical engineering from the University of Illinois at Urbana-Champaign. After twelve years at the company, Welch rose to the position of vice president in 1972, then senior vice president in 1977, and then to vice chairman in 1979. Finally, in 1981, Jack Welch was named the youngest CEO in General Electric’s history at age forty-five, making him also the youngest CEO of his generation.

During his tenure, which would last until 2001, Welch led General Electric through one of the most far-reaching transformations ever experienced by a major U.S. corporation. Under Welch’s guidance, the company’s market value skyrocketed from $14 billion to over $410 billion, and the company became the world’s most valuable company. In 1999, Welch was even named “Manager of the Century” by Fortune .

Getting to the ultimate zenith of managerial success, however, was not an easy road. When Welch took over the reins at GE, the company was a major provider of lightbulbs, home appliances, aircraft and locomotive engines, and countless other pieces of equipment—manufacturing products, which were quickly becoming the remnants of a bygone era. Competition from foreign companies was increasing, and that new competition was able to make higher quality products at a cheaper cost.

Since his first day at work, Welch was disturbed by how much bureaucracy there was around GE, and as CEO, he set out to change it. Within a year, he had trimmed the firm’s structure, streamlining and consolidating as he went. But that was just the first step; Welch also knew that he had to change the way GE did business in general, reinventing the company from the top down in order to make it successful in the modern world.

Most of those changes were fueled by Welch’s own management principles, many of which were adopted by other CEOs around the world following the success they saw at GE. His style was immortalized in a book, Jack Welch and the GE Way , a title that led to the phrase “the GE Way” as a way of describing his managerial strategies.

Welch’s primary focus was on leadership, specifically leading more and managing less. He was known to lead by his own example, a trait he expected other leaders to embody. His feeling was that in order to spark others to perform to their fullest potential, a leader had to show the example he wanted them to follow. He surrounded himself with people who were qualified, and he trusted them to do what he had hired them to do. He believed that close supervision, control, and bureaucracy were the easiest way to destroy the competitive spirit. “Weak managers are the killers of business,” he said. “They are the job killers.”

Welch welcomed great ideas from anyone who might have one, no matter what the individual’s status within the company. “The hero is the person with the new idea,” he was fond of saying. New ideas, he argued, were the lifeblood of an organization. Change is a reality in business, and it should be welcomed as an opportunity rather than a negative. To that end, he eliminated what he saw as boundaries to the free flow of ideas that might help move the company forward.

The execution of his grand strategy manifested itself in Welch cutting out what he saw as the fat in the company. He closed factories, reduced payrolls, cut outdated units, and fired a vast number of employees that were unnecessary. His mantra was that any GE business must be either number one or two in its particular industry, otherwise it needed to be improved or cut altogether. That sort of draconian policy earned Welch the nickname of “Neutron Jack” with some in the professional world. The name was a reference to a neutron bomb, a weapon that kills nearly every living thing within the blast zone, while still leaving the structures intact. Prior to Welch’s taking over GE, approximately 412,000 employees were on the payroll. During his time as CEO, that number was cut to under 229,000.

He applied the same hard-line philosophy to his management team. Every year, managers who were fortunate enough to find themselves in the top 20 percent in terms of performance were rewarded with bonuses and stock options. Those who were unfortunate enough to find themselves in the bottom 10 percent were fired.

Despite the criticism—or perhaps because of it—Welch managed to create an extremely profitable company. He successfully and relatively seamlessly transformed GE from the outdated model of a nineteenth-century manufacturing company to a modern company that was a paradigm of financial success for the twenty-first century.

One component of that evolution was his shift into financial services through acquisitions. It was one of the rockiest evolutionary roads that Welch traveled. One of those acquisitions was Kidder Peabody, a financial firm GE acquired in 1986. It would turn out to be a blemish on an otherwise nearly impeccable management record. A trading scandal that rocked the financial world would cause an implosion of Kidder Peabody in 1994, the result of the actions of a single trader. True to form, however, Welch stuck to his guns, and, when he realized that the financial firm was bound for destruction, he acted on his principles: face reality, then act decisively. It was a hard, expensive lesson to learn.

*   *   *

In 1985, another brilliant young man, cut from the same sort of cloth as both Thomas Edison and Jack Welch, graduated with a degree in chemical engineering from MIT. That young genius was named Joe Jett, and he would go on to work for Jack Welch at GE not just once, but twice.

Orlando Joseph Jett was born in 1958 near Cleveland, Ohio. As an African American man, he wasn’t the prototypical Wall Street hotshot, an esteemed group that is only about 1 percent black. He wasn’t a big man, yet he was an imposing figure. He had a very large personal presence. He had the drive and the brain that understood the most complex mathematical conundrums; that talent led him to the Massachusetts Institute of Technology, where he studied chemical engineering. He excelled as an undergraduate, and was accepted into MIT’s master’s program.

Jett graduated with his master’s in science in 1984 and strode into a $36,000-a-year salary working as a research engineer at a subsidiary of General Electric in Albany, New York. For reasons that were entirely his own, Jett soon decided to accept an offer to interview for the MBA program at Harvard. He was accepted and enrolled in the fall of 1985. While at Harvard, Jett cultivated an interest in Wall Street after listening to presentations given by some of the major players, including Drexel Burnham, Salomon Brothers, and Shearson Lehman. That interest blossomed further during a summer job at Ford Motor Company in Dearborn, Michigan, where Jett assisted the company’s treasury department in investing the firm’s cash. The job allowed him to converse with a variety of brokers, traders, and salespeople on Wall Street, and from that moment forward, he was hooked.

He graduated from Harvard with an MBA in 1987 and accepted a position with Morgan Stanley in New York City. Joe Jett was twenty-nine years old and living on Manhattan’s Upper West Side. The drug- and alcohol-soaked culture on Wall Street was a long way from Cleveland, but Jett would later admit that women “were [his] only vice.” But before he could soar with the Wall Street eagles, he had to get in the door, which meant a four-month stint in Morgan Stanley’s in-house training program. After successfully completed the requirement, he landed a position on the mortgage-backed securities desk, due in no small part to his superb math skills, where Jett was assigned the less-than-authoritative title of “trading assistant.” The professional culture at that time was such that trading assistants were effectively the lowest of the low. They were often treated as subhuman, primarily because they had been given an opportunity that the other traders didn’t yet think they had earned.

In short time, Jett proved his mettle and was promoted to the level of junior trader, which afforded him a small trading book as sort of training wheels. The kid from Cleveland was on his way, and he started to live the life of a successful and freewheeling financial wunderkind. He garnered a reputation as a partier, sleeping briefly after leaving the office and then hitting the nightclubs until four in the morning.

Then, in what was to become something of a pattern, Jett started having conflicts with other traders on the desk. He grew argumentative and confrontational, which led to his trading trade book being reduced by the senior trader. Then, as the conflicts continued, Jett found his trade book eliminated completely and his position reduced back to the rank of trading assistant. He would later say that his “critical mistake was in failing to understand how highly Morgan Stanley valued team-playing.”

He was laid off from Morgan Stanley in April 1989, an action he blamed on the firm’s distaste for his late-night social habits. As a bizarre form of self-flagellation, Jett removed all of the furniture from his apartment and began sleeping on the floor. He had failed and saw the need to punish himself for it. As he liked to tell himself, “Failure is not an option.”

He wasn’t on the rolls of the unemployed for long; in September 1989 he was hired to work on the mortgage-backed securities desk at First Boston. The firm was no longer doing anything cutting edge in the MBS field anymore and operated as a straightforward Collateralized Mortgage Obligation (CMO) packaging operation. Jett was hired as a CMO structurer, and in that capacity he was tasked with putting together the different classes and structures of the investments. Specifically, he worked to find new ways to price CMOs, especially the equity portion—that part of the CMO that was the final tranche and carried the majority of the risk, the part that traders affectionately refer to as “the nuclear waste tranche.”

The term “tranche” is thrown around by those in finance, and oftentimes the concept is lost on those outside that world. The word is derived from the French word for “slice,” and that is what a financial tranche is. When a mortgage-backed security is converted into a CMO, the investment bank divides it into individual pieces that can be sold off as separate securities. In the case of CMOs, those pieces—called tranches—typically range from low-risk to high-risk, with the latter offering the promise of potentially higher returns. It is the lowest tranche that carries the highest risk, and that was what Jett focused his mathematical skill on.

While at First Boston, Jett kept quiet about his social life. He’d learned that lesson the hard way at Morgan Stanley and had no interest in repeating his past mistakes. He kept his head down and didn’t socialize at work, instead directing his energies toward developing a new, computer-based system that assisted in calculating CMO prices. But First Boston, despite that it had been a pioneer in the MBS industry and effectively invented the CMO, had turned conservative in the business, which sent many of its best traders to seek greener pastures. To replace them, First Boston brought in a new team from Salomon Brothers—just like many other Wall Street firms had done in the 1980s—and Jett found himself a remnant of the old guard, even though he had really just arrived. Clearly, the clock was ticking on his time at First Boston.

While still there, Jett had spent a great deal of time working with traders at Kidder Peabody, a subsidiary of General Electric—the same company where Jett’s chemical engineering career was both born and died. What made Kidder exciting is that it was one of two firms—the other being Bear Stearns—to further expand and complicate the MBS market by dividing its CMOs into significantly more tranches, as many as twelve. By dividing the product into so many slices, the risk in the lowest tranches was concentrated to extreme levels, and the new structure increased both the risk, and the return, for CMO investors.

With his status rapidly diminishing at First Boston, and his connections increasing around the Street, Jett sought out a meeting with Mike Vranos, the legendary head trader of the MBS desk at Kidder Peabody. Vranos was impressed enough with Jett to offer him a job starting at $180,000 a year, and, at least according to Jett, the pair shook hands on the deal.

Vranos, however, saw it a different way. After the rounds of interviews at the MBS desk, he claims that Jett was sent to the government bond desk because he didn’t understand mortgage-backed securities well enough, despite having worked in the business at Morgan Stanley and First Boston.

When Jett returned to solidify what he thought was a done deal, evidently Vranos backed off, saying that all he could offer Jett was as assistant job that carried with it a $35,000-a-year salary. Jett balked at the sudden change of heart and walked away. With still no future at First Boston, Jett slinked back in April 1990 to see if that offer was still good, and it wasn’t. He was stuck.

But Jett wasn’t one to take no for an answer, and he returned again the next month to interview for a different job with then-CFO Richard O’Donnell. Someone was needed to investigate and clean up what O’Donnell termed the “aged inventory” of the fixed income department. GE was nervous about Kidder’s positions, and the traders’ seeming inability to sell it. The “GE Way” was clearly about cleaning up old inventory positions. As senior management saw it, an outsider was needed who could come in and make the tough decisions that Kidder Peabody’s traders were unwilling to. Jett saw it as a foot in the door into one of the Street’s largest fixed-income trading departments at the time. Jett was hired and arrived at Kidder Peabody in July 1990 to clear out approximately $1.3 billion in old securities. He was to sell them off to the highest bidders in hopes of flipping them for a profit, or at least moving them out the door.

*   *   *

Kidder Peabody was founded in 1865 by Henry Kidder and Francis Peabody who took over the Boston-based private bank known as J. E. Thayer & Brother when that bank’s founder, John Eliot Thayer, retired. The firm quickly established a reputation as a major investment banking powerhouse in the Boston area, and part of that reputation was built on the firm’s close relationship with Baring Brothers, the premier merchant bank in London. Kidder Peabody quickly joined the ranks of the “Yankee Bankers,” due to their northeastern U.S. concentration, numerous Harvard alumni, and white-shoe heritage.

There was more to Kidder Peabody than its Harvard connections. The firm had close ties to both the American Sugar Refining Company and American Telephone and Telegraph (AT&T), as the premier financier for those companies’ bond issues. The investment bank was also a part of the consortium known as “The Money Trust”—the bankers who financed the megalithic trusts of the day, including such stalwarts as Standard Oil and U.S. Steel.

In 1905, Francis Peabody, the last of the original partners, died, and control of the firm shifted to Frank Webster and Robert Winsor. In the early 1920s, the firm’s operating capital sat at about $7 million, which was enough to run its operations at the time, but the industry was changing. Robert Winsor, who had an inherent distrust of stocks because of their volatility, steered the firm on a course dedicated to bond trading and financing; that strategy, however, cost the firm heavily when they missed the bull stock market of the 1920s. Profits continued to decline, and then when the Italian government withdrew its $10 million deposit, it left the firm drastically undercapitalized and incapable of surviving the stock market crash of 1929 without outside help.

That help came in the form of a financial bailout orchestrated by J. P. Morgan. It was determined that Kidder Peabody needed a cash infusion of $15 million to survive, and several banks—many from the Boston area—agreed to help fund the bailout. In the end, the package totaled $10 million, leaving Kidder Peabody to raise $5 million on its own.

The woes associated with the lack of operating capital would continue to fester throughout the twentieth century. In 1964, due to the firm’s growth in size, yet still small capital base, the partners collectively felt that the business was too large to continue as an unlimited liability partnership. The decision was made to convert the investment bank into a corporation, though it would remain privately owned and run by the partners under the articles of incorporation.

Ralph DeNunzio was the leader who emerged to guide Kidder through much of the latter half of the twentieth century. He had joined the firm immediately after graduating from Princeton in 1953 and had risen to the position of senior partner by 1969. He also held the distinction of being the chairman of the New York Stock Exchange, clearly a major figure on Wall Street for a number of years. Kidder remained a well-known name in the bond markets but still did not have much luck breaking out of that mold. Under DeNunzio’s leadership, however, the firm continued to move forward and grow, expanding into new business lines, including mergers and acquisitions, and stock trading. By 1985, the capital issue once again reared its head, and Kidder was under pressure from both the Securities and Exchange Commission and the New York Stock Exchange to bring more capital into the firm, citing the fact that far too much money was tied up in its bond inventory. Both worried that a major bond market sell-off could affect Kidder’s ability to meet its obligations to its customers.

DeNunzio sent his deputy, Max Chapman, to act aggressively and broker a deal with Jack Welch at General Electric. Under the arrangement, GE would buy an 80 percent stake in Kidder Peabody in exchange for $650 million, a much-needed cash infusion for the struggling investment bank. That money was a godsend and would put Kidder on the same level as Goldman Sachs and Lehman Brothers in terms of capitalization. DeNunzio and Kidder expected to retain control of the business and finally have all the capital they needed to grow. GE was expecting something else, however—to continue its financial services transformation and diversification by entering into the investment banking world through a well-known firm, then convert it to the “GE Way” of doing things. It would be a purchase that all involved would grow to regret, and it came as no surprise to the many on Wall Street who said the deal was doomed from the start.

Given Welch’s experience and business esteem, it seems somewhat surprising that he would pay so much for a firm that was clearly experiencing difficulties. Perhaps he believed the GE management style would lead the company to new financial heights, a suggestion further buoyed by Welch’s decision to shake up Kidder’s management almost immediately. DeNunzio was out, replaced by a GE director named Silas Cathcart. Kidder Peabody was about to learn what it meant to be a GE company and to be run the “GE Way.”

But before Welch could get Kidder Peabody into the GE mold, a much-publicized scandal rocked the firm to its core. A Kidder executive was publicly linked to Ivan Boesky, the man who became the public face of the 1980s scandals in insider trading. Laws had existed for a long time that made insider trading illegal, but until Boesky’s prosecution in 1986, those laws were seldom enforced. As it turned out, much of Boesky’s insider information had been provided by a Kidder Peabody man in the firm’s mergers and acquisitions department, a banker named Martin Siegel. Siegel had originally been seeking to supplement his salary by doing freelance work for Boesky, and hoping to obtain a more lucrative job; Boesky didn’t hire him, but instead paid him cash for takeover tips. The first payment was rumored to be $150,000 in cash, delivered in a briefcase that required a secret code to open. Siegel only served two months in prison, but his reputation was forever tarnished, and the public relations nightmare for General Electric had just begun.

*   *   *

By 1991, Kidder had further expanded its operations to include asset management, equity research, futures, and a major push into investment banking, but it was the fixed income department at Kidder Peabody that was the mule pulling the financial cart. Fixed income was the only profitable business in the entire firm, accounting for 110 percent of Kidder’s income, and it was Mike Vranos, the head of mortgage-backed securities, who dominated the department.

Joe Jett started at Kidder Peabody—by now under the full control of General Electric—on June 24, 1991, to begin sweeping its old securities out the door. Once Jett arrived, however, the head of fixed income, Edward Cerullo, had another idea for him. By chance, Robert Dickey, who had been the head STRIPS trader, left the firm just before Jett’s entrance onto the grand stage. Joe Jett was available for the job, and certainly no one in fixed-income trading wanted an outsider micromanaging its inventory positions anyway. Cerullo asked Jett if he’d be interested in that $75,000-a-year trader position, and Jett accepted. Suddenly, the former mortgage trader and CMO structurer was the head STRIPS trader at Kidder Peabody. (STRIPS, a Treasury security, is the acronym for “Separate Trading of Registered Interest and Principal of Securities.”)

At the time, the STRIPS desk was staffed by three junior traders and was mostly overlooked by the other members of the fixed-income trading department. STRIPS trading was not producing the same revenue as the other desks, which meant it didn’t get a great deal of respect around the trading floor. Joe Jett planned to change that immediately.

STRIPS was the name given to the divided portions of U.S. Treasury bonds after the interest payments had been separated from the principal payment, much like what many firms were doing with mortgage-backed securities. Prior to 1985, however, the practice was not supported by the U.S. Treasury, so traders who wanted to get separated coupon payments and principal payments had to do so by setting up their own private conduits. Despite the fact that it was a thinly traded market, investor demand was so strong that investment banks started pulling apart Treasury bonds with zeal, creating such conduit names as LIONs (Lehman Investment Opportunity Notes) and TIGRs (Treasury Investment Growth Receipts).

Finally, in 1985, the U.S. Treasury officially started the STRIPS program, an acronym that stands for Separate Trading of Registered Interest and Principal Securities. Regardless of the name ascribed, the stripping process is relatively uniform across the board. Remember that a bond is a debt, and the bondholder is entitled to periodic payments of interest as well as the original principal amount when the bond matures. Every bond is essentially a series of semiannual coupon payments, with the original loan amount paid back in full on the maturity date. A $1 million bond with a five-year maturity and a 10 percent coupon means that the bondholder gets a coupon payment equal to 5 percent every six months, plus the full million dollars in principal back on the day of the bond’s maturity.

In a STRIPS security, however, someone can buy the series of coupon payments or the one lump-sum principal payment (the part sometimes referred to as a zero-coupon bond). Under the new STRIPS program, when a U.S. Treasury security was stripped, a primary dealer informed the Federal Reserve Bank of New York of the intention to do so, and then sent the bond into the Fed the next day. Upon receipt, the Fed sent back both the coupon payments and the principal amount as brand new securities.

The opposite of stripping a bond is called reconstituting, or simply a recon. In a recon transaction, the primary dealer sends in the individual stripped components—both the principal and coupons—and the Fed returns the original bond. Either way, STRIPS require a single day’s notice to the Federal Reserve; and this is important—there is no way to book a transaction with the Fed to strip a bond or reconstitute it with more than one day’s notice.

The fact that a bond can be stripped down into components in one transaction and then put back together means there is always an arbitrage possibility between the stripped component parts and the actual underlying bond. Determining the actual STRIPS arbitrage is a complicated matter, but for those that can master it, there’s some money to be made. Many investment banks assign full-time trading desks on these arbitrage opportunities. If the coupon and principal STRIPS are ever trading at abnormally high prices, the trading desk can send a bond to the Fed to be stripped, and then sell off the STRIPS pieces in the market. Alternatively, if components are trading at abnormally low prices, the same trading desk could buy them in the market and send them into the Fed to be reconstituted.

What this means for STRIPS traders is that they’re constantly on the lookout for STRIPS that were trading at prices either too high or too low—then either stripping existing bonds and selling off the pieces, or buying the pieces and reconstituting the bond. The primary goal is to arbitrage both the STRIPS and underlying bonds, a task made much easier by technological advances and computer pricing models.

Kidder began its first foray into STRIPS trading in 1984, when it moved Andy Ansel into the new role of full-time trader on the newly created Zero Desk. At the time, the market was still trading LIONs and TIGRs, but the following year when the Treasury officially entered the market, STRIPS trading took off. After Ansel left, Bob Dickey, his junior trader, assumed his position, and the business escalated. The STRIPS market grew and so did the profits on the desk. Dickey made over $10 million trading STRIPS in 1988 and $15 million in 1989, making STRIPS the most profitable U.S. Treasury trading desk at Kidder Peabody.

But when bonus numbers started being discussed in 1990 for the previous year, Dickey was not happy. Despite the fact that he’d generated an additional $5 million in revenue from the previous year, Cerullo wanted to pay him the same bonus as the last year. Dickey would have stayed on for only $100,000 more, but Cerullo was adamant about the number. Cerullo’s refusal to negotiate opened the exit door for Dickey, and his departure created an open seat on the trading floor just before Jett’s arrival. Ironically, a dispute over $100,000 in bonus money would ultimately lead to the entire firm’s destruction.

The atmosphere around Kidder Peabody prior to Jett’s arrival was one of constant worry. There was talk about GE trying to sell off the firm because it never managed to fully integrate the investment bank into the GE culture. Jack Welch clearly did not like the world of investment banking, and Kidder’s financial shortcomings were forcing him to spend time enmeshed in them. And that time was not being rewarded with significant gains. Welch actually did try to sell Kidder in 1992 to Primerica, but the deal fell through. When he couldn’t sell it, he turned his attention to making the best of it, which meant focusing on the fixed-income trading department, the one profitable entity in the whole firm. Welch issued a mandate that the fixed-income trading department needed to be built up and strengthened.

Jett had the brains and the drive to be a leader on Kidder’s trading floor, but the firm was under intense pressure to perform, so he didn’t have a lot of time to show results. Kidder was considered a weak member of the primary dealer community and was perpetually in danger of having its status revoked by the Federal Reserve. The major issue was that the firm rarely bought enough new issues at the Treasury bond auctions, a requirement for maintaining primary-dealer status. That led to constant pressure to build up the government bond trading desks, which were constantly trying to compete with the mortgage-backed securities desk within the firm. That pressure was transferred to Jett, who was already putting tremendous pressure on himself. This was, after all, his third shot at a Wall Street career; if he blew this one, there probably wouldn’t be another chance.

Within a month of starting at Kidder, Cerullo laid it out for him. “I expect this operation to generate a million dollars each month,” he said to Jett. “I need improvement in the STRIPS ledgers, and I need it now. Or there will be changes.” That thinly veiled threat was more than enough to get Jett’s attention, and he set out to meet Cerullo’s expectations. At first he failed miserably.

Within a month, Jett single-handedly lost $90,000 over the course of a single day. That disaster resulted in a meeting between Cerullo, Jett, and Jett’s immediate boss, Mel Mullin, the head of government bond trading at Kidder. Cerullo scolded Jett for his losses, adding, “Trading STRIPS has nothing to do with accounts. It is an arbitrage market, and it’s fairly simple. Does the sum of the parts equal more than the whole?” The condescension was a hit to Jett’s ego to be sure, but the directive was clear: STRIPS trading had nothing to do with customer “accounts,” it was all about building up a trading book. Cerullo even upped the ante when he added, “I want profits in short order.”

Jett didn’t need it to be repeated; he got the message the first time. Still, he had little experience with trading STRIPS, though luckily, he had a mathematical mind that made learning the specifics of the job easier. But he lacked any practical experience with buying and selling STRIPS. The status as a novice was no excuse on a trading floor, a fact that Cerullo hammered home for him when he passed by a few weeks later: “I’m expecting improvement this month,” he warned Jett. “Just a reminder.”

That improvement didn’t show up, and by the end of December 1991, Jett’s year-end bonus reflected his lack of performance. Mel Mullin broke the news in a meeting; Jett would be taking home a bonus of $5,000, an amount that was effectively a joke on Wall Street. When your bonus is $5,000, the message is pretty clear: keep your desk clean because you’re not going to be around much longer.

As the calendar page flipped into January 1992, Jett was reminded yet again of the incredibly high expectations placed on him by management. “You’ve had six months,” Mullin told him. “The training period is over. It’s time to produce.” This time, though, Jett would be finally able to live up to those expectations. Within just a few months, Jett would be making money, and a lot of money. His trading book was built up to $8 billion in open STRIPS positions, clearly a part of his aggressive style. His trading account showed him to be making $3 million a month for the firm—though that number would later be disputed by SEC filings—and he began commanding the respect of the other traders on the trading floor. Either Jett had finally gotten the hang of STRIPS trading and immediately become one of the best STRIPS traders on the Street, or he had found some shortcut to make it look that way.

The STRIPS desk was suddenly a big part of Kidder Peabody’s fixed-income trading business, and Mullin credited Jett with raising the desk’s status. He rewarded Jett with a raise, doubling Jett’s salary to $150,000. The quick elevation of status and a promotion was intoxicating, as Jett would later say: “The effect of money on people at work is striking. Immediately, I became popular at work.”

There’s an old saying about a tiger not being able to change his stripes, and by June 1992, Jett’s old stripes were shining through again. He began arguing with traders at other desks, and he clashed with his superiors in the same way he’d done at previous Wall Street stints. One source of disagreement came from the firm’s government bond trading desk, the group that controlled the buying and selling of the Treasury bonds that Jett was stripping and reconstituting. The government bond traders argued that they needed to coordinate the flow of all government bond trading through a single location, specifically their desk. Naturally, it meant they made money off the internal business, which boosted their own revenue numbers.

And while it might make sense for them to feel that way, the MBS trading desk was allowed to buy and sell in the government market as it deemed necessary. It was exempt from trading through the government desk, a right acquired purely as a result of its size and profitability. The message was clear: if you made enough money, you got to call your own shots. It wasn’t too different on every trading desk on Wall Street. Jett wanted the same freedom for his STRIPS desk, and breaking out led to frequent arguments with Bill Glaser, the head of the government desk.

His repeated confrontations would become something of a source of pride to Jett, who would later write about his experiences at Kidder Peabody: “I’d already thrown down the gauntlet, defying Mullin at several key junctures. I got Hugh Bush fired, I refused to deal with regional brokers, I circumvented Glaser, I balked at reducing my trading position from $8 billion to $6 billion.” It says a great deal about the mindset Jett possessed at the time—that he was boasting about professional confrontation with his superiors—but apparently it didn’t bother anyone at Kidder enough to cut ties with Jett. Jett was now bringing in steady money each month, money his bosses desperately needed.

Jett claimed to be making profits from three different sources: arbitrage opportunities in the STRIPS market, making markets for clients by taking advantage of the bid/offer spreads in STRIPS, and maintaining a long position in bonds while simultaneously hedging it by selling bond futures contracts. The last part is what is known as “bond basis” trading.

But the reality was quite different. Jett had noticed a flaw in the Kidder Peabody trade processing system that allowed him to book what amounted to phantom profits. He came upon this flaw when he had booked a “forward trade,” and now the smart guy was flexing his mental muscles.

A forward trade is simply a trade that settles at some point in the future, generally further in the future than normal trade settlement. For example, if two Wall Street traders decide to book a forward trade, they’ll decide all of the details—the security, the price, the quantity, etc.—on the trade date. That’s when the details are agreed to, but not when the actual securities change hands. The securities get transferred on the day called the settlement date. Government securities typically settle on either the same day, called cash settlement, or the next day, called regular settlement.

To be clear, the Federal Reserve doesn’t allow STRIPS to settle any longer than regular settlement, but as it turns out, nobody told that to the Kidder Peabody computer system. Jett discovered that he could book recon trades in Kidder’s trading system with settlements as far as thirty days in the future. The best part about those forward trades: the profits generated by the trade were booked immediately in Jett’s trading account. Even when no money had yet changed hands, Jett showed profits in his account, so he began booking these trades as frequently as he could. When the settlement dates neared, he simply rolled the trade forward, rebooking the settlement date of the original trade for thirty more days in the future and still retaining those imaginary profits.

Overall, 1992 was a very good year for bond traders. Prices were on a continual climb upward as the Fed had been cutting interest rates all year long. Jett was dealing with a variety of different STRIPS, and that position had paid off nicely by October. Jett had booked $17 million in profits from January 1, 1992, to October 31, 1992. Only there was one catch—all of those profits didn’t really exist. When the SEC did a little checking years later, those profits were booked with forward settlement dates that had yet to be reached. Going from a trading desk expected to make several million dollars a year to $17 million was a big leap in the minds of all those around Jett at Kidder. What was even better, his whole system for booking phantom profits from forward trades was about to get even easier.

At the end of 1992, Kidder decided to upgrade its trade processing system with a new computer program called Government Trader. One of the snazzy new elements of the new system was the fact that it allowed traders to book forward trades at any future settlement date. So whereas the original system had limited traders to a thirty-day window, the new system had no such limitations. Unknowingly, by improving it trade processing system, Kidder just made Jett’s trading scheme even easier.

As it turns out, there are two versions of what happened next, following the implementation of the new system. According to the SEC, Jett immediately picked up on the new loophole and began to exploit it. The first trade he booked under the new system didn’t settle for 203 days in the future. Additionally, the system had a new flaw too. It didn’t take into account the cost of borrowing the money to finance the trades—called the cost of carry—when calculating the profits, all of which were ignored by the system until trade settlement. So Jett was now able to book up to forty times the profit he had booked before on the same trades, thanks to the new limitless forward settlement date window. That sounds pretty cut and dried, of course, and casts Jett as acting in a clearly fraudulent manner.

Jett, however, had a different version of what happened. He claimed the new processing system at Kidder Peabody wasn’t even operational until the last week of December 1992. He remembered it being the day after Christmas, in fact. The office was quiet, as most traders were on vacation. Jett, the hardworking trader, was in the office manning the fort while the other traders enjoyed their vacation. And the market was quiet too, for much the same reason. According to Jett, when he input his first recon trade on that December morning, he noticed that something was immediately odd. Whereas he’d been expecting the trade to show a $30,000 profit, the computer system credited him with a $300,000 profit.

There was clearly a computer system malfunction, and Jett claims to have brought the matter to the attention of his boss, Mel Mullin, and the computer programmer, Moishe Benatar. Benatar told him that the profits were unrealized, meaning they shouldn’t be counted toward actual profits until the actual settlement day. Mullin, a PhD in mathematics, later claimed not to have understood the accounting method or the program flaw himself. One former trader who worked for Mullin even pointed out, “Mel never understood the bond market business. He had no idea.” Robert Dickey, the former head STRIPS trading at Kidder, later said that the traders were all well aware of the flaws and how to exploit them.

The way Jett exploited the new system was pretty similar to the way he did it with the old system. When Jett input a forward recon trade, the system showed a bond being delivered to Kidder from the Fed, and Kidder delivering the coupon and principal STRIPS to the Fed in return. All of this was scheduled to happen on that future date. In Kidder’s system, it then showed that Jett had a long position in a bond and a short position in the STRIPS. But it wasn’t a real trade with the Federal Reserve; it was just what appeared in Kidder’s trade processing system: receiving a bond from the Fed and delivering the STRIPS back. Essentially, by booking a recon forward trade Jett was announcing his intention  to execute the trade on the forward date. But no trade was done yet, and, obviously, no money had yet to change hands.

As the settlement date neared for the booked recon trade, Jett would cancel that trade, then book another one with a settlement date way off in the future. The cost of carry, or the cost to borrow money to finance those trades, wasn’t factored in by the system either, so he didn’t even have to pay any expenses on the “profit” he was taking. As long as he kept pushing the forward dates further into the future, he could keep on booking increasingly larger trades, showing increasingly larger profits. As long as the size of the make-believe trades kept increasing, so too did the amount of make-believe profit.

The repo desk was the part of the trading floor responsible for financing Jett’s short and long positions, which also left it privy to what he was doing. Or at least privy to what was actually happening; whether or not the repo desk knew what he was doing is anybody’s guess. If Jett was long a STRIPS, the repo desk would loan it into the repo market to raise the requisite cash to pay for it; if he was short, they’d borrow the security to cover the short position. The desk was quick to report to management that Jett was trading unusually large amounts of STRIPS against forward settlements that oftentimes never settled at all. The trades were being closed out just before the settlement dates with the Fed. Regardless of whether or not they understood the true nature of his trades, the repo traders noticed a pattern, specifically that Jett had very large STRIPS positions, oftentimes greater than the amount of the actual Treasury bond outstanding. That is, if you added up the size of all of the particular STRIPS that Jett was short or long on, it was more than could have possibly been stripped, because the actual bond size wasn’t that large. Other STRIPS traders around the Street knew something was wrong a Kidder too. Jett’s oversized positions were distorting the STRIPS market, and other firms were making a killing on it.

Regardless of who knew what, in the space of two months Jett’s reported profits jumped to a total of $40.4 million—that’s money he made in sixty days, mind you—which was more than double the profits he’d booked all of the previous year. It is equally surprising that neither Cerullo nor Mullin thought anything was amiss when Jett’s profits suddenly skyrocketed. But again, it was  a good year for bonds, so maybe they thought they’d just found a gold mine of a trader. So as long as the trading desk made money, they didn’t quite care about the details of how it was made. Given that the other traders on the STRIPS desk had lost $10 million in 1992, the new rock star kept the STRIPS desk profitable.

Given the mechanics of Jett’s trade, it’s surprising that a rational trader would keep it going without restraint. It was a classic example of a snake eating its own tail. In order for the scheme to continue, his positions had to get larger and larger. At some point, Kidder, or GE, or the ability to borrow cash would prevent the business from growing any larger. If Jett ever stopped rolling his fraudulent trades and growing their size, the profits would cease, and the snake would have caught up to itself. However, at the particular point in time, Jett had no problem continuing the deception by increasing the size of his trades as forward settlement dates approached.

By the end of 1992, Jett’s profits were, by all accounts, pretty impressive. Of course, they’d be a lot more impressive had they been actual profits as opposed to merely imaginary ones, but that didn’t stop his superiors from taking notice and rewarding Jett for his activities. Jett was given an outstanding review for his year-end performance and was promoted to the level of senior vice president; Mullin told colleagues that “Jett has become one of the top STRIPS traders in the industry.” The young trader’s year-end bonus was $1.6 million, which was a far cry from the $3 million he had been expecting, but still a sizable Christmas present for a trader who’d been one hair away from begin fired just twelve months earlier.

In January 1993, Jett formalized his trading strategy, such as it was. This would become his truth, his version of what he’d been doing all along. Despite the fact that it wasn’t really a viable strategy after all, Jett would never acknowledge that he’d done anything wrong, basically because his plan hinged on the computer glitch he’d happened upon. Remember that the Kidder system treated recon and STRIPS trades with the Fed as real trades, despite the fact that there were no actual trades involved in the quasi-transaction. It was simply an internal booking for a possible future trade settlement. No more, no less.

That said, Jett needed an explanation as to why, and how, he was all of a sudden making a lot of money. Inside and outside the company, during and after his time at Kidder, he would call this his three-part arbitrage strategy or sometimes just plainly referred to it as his trading strategy. Keep in mind, the internal bookings, shown as trades with the Fed, were the cornerstone to his three-part plan. The so-called “strategy” started with the forward recon trade with the Fed; remember, that showed up as a forward purchase of a bond plus a short position in both the coupon and principal STRIPS, but those trades didn’t start until sometime in the future. So step one appeared as a long position in a bond and a short position in the coupon and principal STRIPS. Next step: Jett went out into the market and purchased the same STRIPS, but settling immediately this time as a way of hedging the forward recon trade. He actually owned those STRIPS, as of the current day when he booked the trade. Finally, he would sell the bond futures contract, as he claimed, “to hedge the price of the reconstituted forward position [I] was holding in bonds.”

As long as you accept the forward recon trade with the Fed as a real trade, it appears to be a sound strategy. However, when you take out the phantom trades with the Fed, the somewhat complicated STRIPS strategy isn’t much of a strategy at all. Let’s start with Jett buying coupon and principal STRIPS on the day of the booking. Then, he short-sells the bond futures contract, which is the equivalent of selling bonds to someone at some point in the future. For purposes of explanation, let’s assume those contracts settle ninety days in the future. It’s important to point out here that if a trader owns all of the coupon payments and the principal payment, in essence, it’s the same as owning the actual bond itself. Now, since we know the forward recon trades with the Fed don’t really exist, they’re just an intention  that exists in the Kidder Peabody trading system, we just erase those trades from the strategy. What we have left is Jett owning all the coupon and principal payments that make up a full bond and a future contract to sell bonds ninety days in the future. In other words, Jett’s tried-and-true, radical new strategy for STRIPS trading was simply owning the equivalent components of a bond and selling that bond at a future settlement date.

The genius part of his plan, if you call it that, was the fact that he booked and rebooked so many trades that he was able to effectively hide that he was exploiting the flaw in Kidder’s trade processing system. When the accounting team came to audit his trades—including the differences in trade and settlement dates, different bonds, recons, coupon STRIPS, principal STRIPS, and bond futures—the team was easily confused by the sheer chaos inherent in any kind of flow chart it could produce to trace his actions. In a typical recon or STRIPS trade with the Fed, there can be as many sixty-one separate components. Think of a thirty-year bond: it includes sixty semiannual coupon payments and one principal payment at the end, all trading as separate securities. Perhaps the accounting folks could be forgiven for their confusion after all.

In standard market parlance, Jett’s strategy was nothing more than a “basis trade.” The only real risk he assumed in his strategy was that the bonds he owned (held in the separate coupon and principal components) were not the same bonds required to be delivered as part of the bond futures contract on the delivery date. That’s what they call basis risk, and it’s generally very small. Regardless of risk, however, there is one important linchpin to this whole scheme. There is no mathematical way possible in which a trader can amass $150 million annually by trading the bond basis. It just can’t happen that way. But nobody at Kidder Peabody seemed to pay attention to that harsh reality.

One of the reasons Jett got away with everything was in part to his skills at office politics. Obviously, it wasn’t his skills with coworkers or the other desk heads, but rather, his skills with his superiors. He kept himself very close to Ed Cerullo, close enough to get invited to Cerullo’s vacation home in Aspen, Colorado. Perhaps Cerullo’s fatal flaw was what one of his traders pointed out: “Ed trusted everyone who worked under him to be honest.” And then there were all of the astronomical profits Jett continued to bring in. All that income led to suggestions that Jett should also be running the whole government area—the Holy Grail for a trader in his position—as opposed to Mullin. Jett pushed Cerullo to promote him and was told that the entire department would eventually be reorganized. When that time came, Jett would find himself promoted to head of all government bond trading, and Mullin would be moved to a new position as head of what was to be called Derivative Products.

The reorg came in February 1993. Jett was, as promised, promoted to head of all government bond trading, which gave him oversight of the STRIPS desk, as well as the U.S. Treasury and federal agency trading desks, the latter including such items as the securities issued by Fannie Mae and Freddie Mac. A crew of fourteen traders was now reporting to Jett. The aggressive newcomer had just taken charge, and Kidder Peabody was about to realize the magnitude of the mistake.

*   *   *

Right around the time when Jett assumed control of the entire group, Vranos’s mortgage desk had taken a $70 million loss—a loss that completely shocked Jack Welch. The loss was large enough to wipe out Kidder’s entire profits for the previous year, and Welch responded quickly by moving a new man, David Bernstein, to head up risk and compliance at Kidder. Not surprisingly, Bernstein’s first assignment was to audit the fixed-income trading department. It was also a clear shift in focus for Welch. Whereas previously he’d been worried about the firm’s $86 billion balance sheet, he had failed to fully understand the risks inherent in Kidder’s business, which was now the center of his attention.

Welch had previously seen a limited balance sheet and asset size as a way to limit risk, but those days had changed. He was now worried that Kidder’s business contained more risks than he originally thought, and of course he was right in that belief. Ironically, he chose to focus on losses that the company had booked properly on the mortgage desk instead of profits that were booked improperly on the STRIPS desk. He knew something was wrong, but he was looking in the wrong place.

Another problem was now at Cerullo’s feet. Whereas before he’d been able to rely on the mortgage desk for profits to grow the firm’s income statement, that avenue was now basically shut. The risk limits imposed by Bernstein would tie up the mortgage desk in terms of a source of revenue, so Cerullo had only one other option, which was the government desk. And that meant his only way of growing the firm’s profits relied solely on Joe Jett.

By the middle of 1993, however, it was clear there was not going to be a problem with profits the year. The long end of the bond market had continued its spectacular rally, and the Fed had kept short-term interest rates down at historic lows. Jett was showing a staggering $66.7 million in profits for his forward STRIPS trades, though $58.5 million of that amount was imaginary. David Bernstein was now comfortable enough in his new job to challenge Jett on his profits, but Jett was prepared. Because everyone knew that Kidder’s system did not properly handle trades that were booked so far in advance, Jett had kept a meticulous diary of his forward transactions in what would come to be known as the Red Book. It was, he explained, a way of helping the repo desk keep track of his forward trades. To Bernstein, that sent up a warning flare, as he expected that all trades should be accounted for in the firm’s trading system in order to accurately reflect profits.

Upon further examination of the Red Book, it was discovered that as of July 1993, Jett had booked so many forward recon trades in one bond that the size of the trades was again greater than the amount of the bond being reconstituted. In other words, Jett was again projected to put back together more bonds than could possibly have been stripped—more of a certain bond than even existed in the market. The result, though, was that Bernstein simply told Jett that he could no longer book forward recon trades settling more than three months in the future because the mainframe couldn’t properly handle such long periods of time. Bernstein felt he had to do something about the problem, but his action really didn’t change anything.

Bernstein’s opportunity for making changes would come back at the end of the third quarter in September 1993, when GE wanted to cut back on the size of Kidder’s trading positions and instructed its subsidiary to reduce its balance sheet significantly. Whereas GE imagined the balance sheet was supposed to be kept below $50 billion, Kidder was routinely carrying assets in excess of $80 billion, and during the summer of 1993, with Jett’s growing STRIPS positions, that figure had swelled to over $100 billion. In order to achieve the $50 billion goal and keep GE happy but not limit the revenues rolling in, Kidder needed a good plan. The result was found in a little creative accounting. Kidder first switched its accounting system to do away with recognizing unsettled trades; that meant all of Jett’s forward recon trades were suddenly off the firm’s balance sheet. Of the $30 billion in STRIPS positions that he was carrying at the time—three times his authorized limit—$24 billion were forward-settling trades.

To further efforts to keep GE off their backs, Cerullo wanted to cut existing assets by another $26.2 billion, an amount that included the outright elimination of $5 billion of Jett’s long positions—the coupon and principal STRIPS. Jett, as was to be expected, was not happy about the move. After some back and forth with Cerullo, Jett was outright ordered to do so. As we know, Jett’s scheme involved rolling his trades into larger and larger sizes, and cutting back did not figure into his game. In order to close down his STRIPS trades, he had to cancel the forward recon trades that were juicing up his profits for so long. And by cutting back the forward recon trades, he was forced to give back profits that he’d been perpetually been rolling forward.

In theory, if Jett had merely been closing out long “basis trades” by selling bonds and buying back the futures contracts, his losses would be insignificant. But there’s nothing a mathematician hates more than theory. His excuse for the massive losses that suddenly appeared was that “to get the STRIPS we needed in such a short period, we paid exorbitant amounts.” This neglected the fact that government securities trade with razor tight spreads, often at times with bids and offers one sixty-fourth of a point apart.

Despite Jett’s claim that he was up $70 million for the year, he was forced to book an immediate $48 million loss on his positions. Truthfully, both the accounting team and the management at Kidder should have realized immediately that something was not right. When a trader can’t close positions without taking a big loss, there’s clearly something wrong with what he’s doing. But Jett didn’t let the quarter’s end slow him down for long. The day the next quarter began, Jett went to work booking all of his STRIPS and recon back into the system. All of the forward recon trades and the phantom profits that had disappeared were suddenly back in his trading account. Incredibly, management ignored his sudden turnaround: trades disappeared and profits disappeared; then trades reappeared and profits reappeared. Perhaps those above Jett truly wanted to believe their star STRIPS trader was just that good. Cerullo’s bonus was partially a function of Jett’s success, so maybe it was easy to turn a blind eye.

By the end of 1993, the government bond desk had overtaken the mortgage desk in terms of profits, led in no small part by Jett’s imaginary revenues. The STRIPS profits rolled in at $150.7 million for the year, but there was $198.2 million in imaginary money involved. The result was that the profit for the STRIPS desk really wasn’t a profit at all. In total, Jett’s trading desk delivered a net loss for the year of $47 million, according to SEC documents. But Jett’s star wasn’t diminished at that time because no one knew about the inflated numbers. The firm even named him “Man of the Year” due to the fact that he alone had accounted for 27 percent of the entire profit generated by the fixed-income trading department, a division of seven hundred people. For his efforts, Jett took home a $9.3 million bonus and a promotion to managing director.

It seemed like everything was going well for Jett, and he was living the life he had always dreamed about; however, nothing continues forever, and all hell broke loose in the bond market in February 1994. In what’s remembered as a complete surprise to the market, the Fed announced a surprise twenty-five-basis-point tightening of interest rates. The result was a massive sell-off across the entire bond market, leaving trading desks everywhere nursing huge losses. Kidder Peabody was no exception.

For months after the September quarter-end debacle and in order to win some independence from GE, Kidder was trying desperately to secure outside funding from a variety of banks, including UBS. In order to get that lifeline, however, the unequivocal message was that Kidder needed to tidy up its balance sheet by the end of the quarter, March 31. Jett saw the writing on the wall and realized a repeat of the September quarter end was in the balance. But this time he had his own strategy for subterfuge. He continued to book the forward recon trades, but also started booking the outright purchases of coupon and principal STRIPS as forward trades too. Knowing that Kidder’s accounting system, as of six months prior, did not recognize the forward settling trades, it was just the next loophole to exploit. By doing this, he kept almost all of his trades off of the firm’s balance sheet, with the assumption he wouldn’t have to go through the fire drill of closing everything out again. The result, however, was an operations mess, with the creation of a massive number of tickets for transactions that were simultaneously being closed out, settled, and rebooked to be settled in the future.

Cerullo, despite his apparent ignorance of Jett’s malfeasance the previous year, was growing suspicious. He finally took notice that once Jett sold off his positions to reduce his balance sheet in September; Jett was showing a fraction of the profit that he’d been showing before. Why this suddenly became suspicious in March 1994 but not in September 1993 is anyone’s guess. Bernstein shared Cerullo’s suspicions too, and when Jett was in London on a business trip, Bernstein interrogated one of Jett’s traders.

It was the first opportunity that anyone ever had to speak with individuals who worked on the STRIPS desk. Up to that point, not only would Jett not allow it—he had never taken so much as one day of vacation since he’d started working at Kidder, and his physical presence prevented that type of communication. But with Jett away on business, there was finally an opportunity.

Bernstein started by asking a trader, whose name was Dave, to value the forward trades at current prices and report back to him when finished. The result, suffice it to say, wasn’t encouraging. When Jett returned from London, he was told by Bernstein, “Dave here is afraid that when we allow your trades to settle, you’re going to have to write a big check to the Fed.”

Jett shrugged off the suggestion as he was staring down the barrel of a loaded financial gun, but Bernstein didn’t give up. Bernstein next went to Cerullo and pressured him to close down all of Jett’s trades, telling Cerullo, “I’m certain. I’m certain that if these trades settle, he’s going to have to pay up.” Bernstein’s fear was palpable, and Cerullo’s own anxieties fed off that fear. He authorized Bernstein to investigate Jett’s trades further, but again, he was finding the right answer but looking for the wrong reason. Bernstein feared that Jett was taking too much risk, and that’s what scared him. He sensed that the profits might be somehow inflated, but he assumed it was due to risk taking, not an accounting flaw in the computer system.

The financial shit hit the fan just days later when Askin Capital Management, the largest customer of Kidder’s mortgage desk, declared bankruptcy as a result of losses in its mortgage-backed IO and PO trading positions. Cerullo immediately ordered Jett to unload his STRIPS trades, telling him, “This Askin Capital situation has forced us to have as simple and understandable a balance sheet as we can, so I want you to liquidate your positions.” Jett’s world was about to come crumbling down. He told Cerullo that they’d be taking a loss, just like he had the previous September, but it fell on deaf ears.

The trader who had once commanded hero status at Kidder Peabody knew his time was up, and he began to panic. But he reasoned that if he got out while his reputation was still intact, he had time to write his own ticket at another firm. He tried, unsuccessfully, to recruit some of the other traders under his purview to abandon Kidder Peabody with him and head to another firm.

On March 28, Jett’s assets started off at $29 billion on paper, but then the liquidation began, and Cerullo was stunned at what happened. The major problem with all of Jett’s STRIPS positions, from a risk standpoint, is that they weren’t completely hedged. Through miscalculation or just misunderstanding, Jett was actually net long $1.5 billion worth of securities. Being long in a declining market is not the position a trader wants to be in .

As Jett kept selling his positions, cash was leaving the firm like water through a funnel. Cerullo called him on the phone directly. “Do you have any idea how much liquidating your position has cost us?” he asked incredulously. At that moment, Cerullo had determined that approximately $300 million in nonexistent profits were about to disappear, generating a massive loss, and he explained that painful fact to Jett.

Jett was backed into a corner and he knew it, but he wasn’t going down without a fight. Defiant to the very end, he replied to Cerullo flatly, “Not if you return my position to me.” The argument again fell on deaf ears.

A few days later Jett was called into a meeting. In attendance were some of the more powerful figures at Kidder, including CFO Richard O’Donnell, the firm’s chief legal counsel, and the head of human resources. At that meeting, Jett was apprised of the situation. “We’re looking at a loss of $300 million.” That loss was all Jett’s, due to the fact that the profits he had booked didn’t really exist. There was no hiding from that fact.

On April 17, 1994, Joe Jett received a letter at his home, delivered by a messenger service. It was a pretty straightforward message: “Your employment with Kidder Peabody is terminated.” That same day, GE announced a one-time charge on its earnings of $210 million after taxes; the pretax amount had been $350 million. GE was forced to inject $200 million into Kidder’s operating capital, an action that ruffled more than a few feathers at the top of the org chart. “We’d be having great numbers,” Jack Welch said, “without Kidder.”

During his tenure at Kidder Peabody from July 1991 to April 1994, Jett had booked a total of $338.7 million in phantom profits. His net loss to the firm turned out to be $74.7 million. And though Jett’s fraud had been the metaphorical rest of the iceberg, further investigation into the firm’s operations showed that there were smaller icebergs that nobody knew existed. There were two other traders holding dishonest trading positions. An options trader named Neil Margolin, it was discovered, had hidden $11 million in losses on French and Spanish government bonds. Another trader named Peter Bryant had hidden $6 million of his own losses. Both men received the same fate as Jett: they were fired from the firm.

As the truth began to come to light, Jett’s confrontational side began to shine. He initially called the whole situation nothing more than “window dressing,” saying it was an attempt by Kidder Peabody to disguise its balance sheet and hide the true size of its assets from GE. Ironically enough, Jett’s line of attack was itself nothing more than window dressing, as he merely attempted to shift the blame for the firm’s losses from his own phantom profits. And then, in a leap of logic that could only make sense to him, he claimed that because Kidder’s management—the men he referred to as his “sophisticated supervisors”—had believed his strategy to be legitimate, it couldn’t be considered a fraud. In other words, because he’d managed to disguise what he was doing well enough to convince his superiors that he wasn’t doing anything wrong, he truly wasn’t doing anything wrong.

Kidder froze all of Jett’s brokerage accounts that were being held at the firm, which was expected of an employee terminated under the circumstances. The stakes got higher when his home was raided by FBI agents, and when he was called in front of an arbitration panel set up by the National Association of Securities Dealers (NASD). He received a subpoena from the SEC and was banned by the New York Stock Exchange in December 1994. It was certainly not a good way for Jett to end the year.

Jett retained two lawyers—one a specialist in securities law, the other a criminal attorney—and invoked his Fifth Amendment right against self-incrimination when questioned by the SEC. During that time, he found work with a furniture moving company in Manhattan, saying that no Wall Street firm would even talk to him. He would eventually get his day in court, however.

There were still other players who had to answer for their actions before the Jett case was ever heard; Michael Carpenter, the CEO of Kidder Peabody, had the unfortunate duty of telling Jack Welch in person during the second week of April 1994. Carpenter was forced to resign his position two months later, and was replaced by Dennis D. Dammerman, a senior vice president and the CFO of GE. Dammerman retained his GE titles, but also added that of CEO of Kidder Peabody to his resume.

Edward Cerullo was forced to resign from Kidder Peabody one month after that, in July 1994, but he didn’t go without a parting gift, of sorts. He was fined a measly $5,000 for his failure to supervise Jett and was allowed to keep his bonuses, even those inflated from Jett’s phantom profits. He kept a total of $35.6 million that he’d been paid in bonus money during Jett’s time at Kidder, theoretically a reward for his leadership of the firm’s rainmaker. He claimed absolutely no knowledge of fraudulent activity on Jett’s part. He did, however, acknowledge that he had certainly “failed to supervise” Jett’s activities.

Mel Mullin also contended that he knew nothing of what Jett was doing, but that ignorance didn’t help him. He was fired on August 3, 1994. Charging that Mullin had failed to adequately supervise Jett, Kidder kept his deferred compensation of $2.7 million. He maintained his ignorance, but he eventually paid a $25,000 fine to the SEC and served a three-month regulatory suspension from any association with a broker-dealer.

As for Kidder Peabody itself, the firm was fined $40,000 by the SEC, and GE unloaded its albatross of an investment bank on Paine Webber in October 1994. Welch acted quickly and decisively, following another one of his famous GE philosophies: any business GE owned had to be an industry leader or else it had to be “fixed, closed, or sold.” GE certainly tried everything it could to fix its problematic subsidiary but in the end, it was clearly time for GE to get out of the investment banking business.

Where GE had initially paid $650 million for Kidder Peabody in 1986, its investment ballooned to over $2 billion during the eight years that it owned the firm, pumping in approximately $1.4 billion in additional capital as it hopped from one scandal or loss to another. By the time GE finally sold Kidder, Paine Webber agreed to pay only $670 million, meaning that GE had lost just shy of a $1.5 billion on its failed experiment in investment banking. And just to add insult to injury, as part of the sale agreement, GE was forced to keep the mortgage-backed securities inventory. And it had to manage the positions without the expertise of Michael Vranos. After the Jett fiasco, Vranos resigned and moved to a hedge fund, taking approximately forty of his mortgage traders with him.

Paine Webber, for its part, mostly wanted Kidder’s retail brokerage business, which employed about 1,150 brokers in fifty offices. Kidder had catered to high-net-worth individuals, and its retail broker network was extremely lucrative. Paine Webber did what it could to integrate Kidder into its existing business structure, but there were many changes. For starters, the vast majority of former Kidder employees were laid off immediately, and the Kidder Peabody name was dropped altogether. One of Wall Street’s most esteemed names in investment banking ceased to exist right then and there.

As for Joe Jett, the drama did not end. The popular TV crime series Law and Order picked up on the Jett story, running an episode about a rogue trader who was involved in a murder. In the end, the trader was found guilty and convicted, but that was the television world; in the real world, Joe Jett’s epic story would continue for years.

At the end of 1995, Jett received word from the SEC that he was “not a target” of the Justice Department, meaning there would be no criminal charges pursued against him. But then in January 1996, the SEC officially filed civil charges against him. That same year, the National Association of Securities Dealers—the self-regulatory agency overseeing the securities industry—handed down its own ruling. The arbitration panel found Jett innocent of fraud, and ordered Kidder to return the $5 million in Jett’s brokerage account. The victory, however, was bittersweet, since Jett was unable to get his legal bills covered and was forced to pay $4 million to his lawyers.

The SEC took its time with the investigation and didn’t announce a ruling until March 2004. Much of that time was taken to sort out witness statements, many of which were contradictory. In several cases, witnesses gave conflicting stories over the years, which forced their removal from the witness lists. With many of the key witnesses to the SEC’s case gone, the case was close to falling apart.

But the facts were the facts, witnesses or not, and the SEC eventually found that Jett “deceived the firm about his trading performance, and caused, and aided and abetted record-keeping violations.” They also found that he “recognized the computer glitch that made posting profit from forward reconstitutions possible, but no one else understood the implications.” In other words, Jett was guilty of being the smartest guy in the room. However, that wasn’t itself a crime, and no criminal charges were ever filed against him.

In a semantic twist to this whole case, coupon STRIPS and principal STRIPS are not legally considered securities under the definition of the word, so there could be no securities fraud. The SEC was convinced that Jett “had every intention of defrauding Kidder,” and he was ordered to pay a disgorgement of $8.21 million and a $200,000 fine as a civil penalty. Additionally, he was “barred from association with a broker or dealer” thereafter.

True to his character, Jett never wavered from his insistence that he had done nothing wrong. “I was desperate to argue that I wouldn’t have lost a cent if Cerullo had not forced me to liquidate my inventory,” he said after the dust had settled. The problem with that impassioned plea is that the whole idea is itself absurd. Worse than absurd. It’s nonexistent. His defense is predicated on the false idea that the profits were real—kind of like the trader who refuses to sell a losing position because he doesn’t want to take a loss. They money was already lost; refusing to sell doesn’t make the loss go away.

As of this writing, Jett has moved on to become his own boss. He is the CEO of Jett Capital Management, which bills itself as performing asset management, advisory, and private equity services. The firm’s website highlights the fact that “the NASD found no basis for Welch’s assault against Jett and declared Jett innocent of fraud.” The information conveniently leaves out any mention of the SEC’s findings. The site does, however, highlight the fact that Jett “pioneered the use of off-balance sheet financing,” though whether or not that is a good selling point is up for argument. He is periodically an invited guest on financial television shows, for which his appearance fee is rumored to be between $4,000 and $8,000 a show. As of 2010, he claimed to be living in the dingy basement of a former girlfriend’s house, from which he filmed at least one YouTube video.

*   *   *

Joe Jett was never found criminally liable for his actions, so at the core of the whole case, he is in one regard innocent. But did he commit fraud in the true sense of the word? On the one hand, he didn’t try to hide anything from anyone. He submitted to audits and explained exactly what he was doing, though he did so in ways that were, let’s call it, less than forthcoming. He booked all of his trades in Kidder’s trade processing system, had the “Red Book” for trades that were not in the system, and freely discussed everything with his superiors and auditors. So could it still be fraudulent?

The fact of the matter is that Jett found a flaw in the system at Kidder, and he exploited that flaw to the fullest extent for his own gain. Yes, his activity was in the open; he was claiming profits that he knew were not real. He was clearly and purposely taking full advantage of the flaw he discovered, plain and simple.

Being the mathematician that he always was, in an “aha” moment, Jett stressed an equation to demonstrate that what he did was completely legitimate. The equation, which he credited to Mullin, claiming that Mullin had written it on a blackboard during a meeting, reads: “c + p = bond + acc.” The “c” is for the coupons, “p” is for the principal portion; “bond” is the bond’s price, and “acc” is accrued interest. It’s basic math and, in Jett’s mind, it proves Jett was right all the time. Again, though, he is adept at using fuzzy math to justify his actions. His proof is based on a fallacy, because his equation doesn’t take into account the settlement date. And remember, Jett was doing his trades as forward recons trades. The value of securities between now and ninety days from now is much different.

Jett’s formula should really have the first half of the equation being today and the second half being three months from now; it was missing the cost of carry all the time. Jett’s formula should read: “c(today) + p(today) + cost of carry = bond (today + ninety days) + acc (today + ninety days).” Therein lies the true secret to Jett’s strategy. Remember, the coupon and principal payments are the equivalent of the bond, and Jett was able to keep ninety days’ worth of accrued interest instantaneously upon booking the trade. Kidder’s accounting system took all profits up front on the future settlements; it just didn’t incorporate the cost to carry.

Jett was never charged by the government with losing money; after all, a trader who loses money in an honest fashion isn’t guilty of a crime. If losing money on a trade were a criminal offense, there would be a lot of Wall Street traders going to jail. No, losing money is not a crime, but hiding losses and booking false profits clearly is.