Infectious Greed

by Frank Partnoy

  In reality, bankers and securities analysts had been corrupted by precisely the same forces that had corrupted accountants: pressure from corporate executives, limitations on liability, and conflicts of interest related to their own firms. Although investors hadn’t noticed, there were signs during the mid-1990s that Wall Street had begun to behave very badly. First were the various derivatives fiascos, with Bankers Trust leading the pack. Then there were a series of scandals in the NASDAQ market, where most technology stocks were traded. In 1994, economists William Christie and Paul Schultz reported that NASDAQ traders were conspiring to charge investors high commissions (the securities dealers later settled disputes related to the conspiracy for more than a billion dollars.) 34 In March 1995, several traders from Morgan Stanley allegedly manipulated the markets for several technology stocks—including Dell Computer, Novell, Sybase, and Tele-Communications Inc.—in order to profit from a complex scheme related to NASDAQ options (Morgan Stanley later was fined $1 million and the traders involved in the scheme were fined and suspended, although all of the fines were reduced substantially on appeal).35 Bankers also advised on technology-related mergers, which increased from $12 billion in 1992 to $150 billion in 1997,36 and wrote “fairness opinions”—that these mergers were fair for shareholders, even though the deals made no more sense than the disastrous conglomerate mergers of the 1960s.

  In the late 1990s, two schemes in particular—one involving IPOs, one involving securities analysts—enabled investment banks to make billions of dollars at the direct expense of their clients and investors. Through these two schemes, Wall Street hid important facts from investors, facts that might have sobered those drunk on animal spirits. Frank Quattrone was the key figure in the IPO scheme; Henry Blodget was the key figure in the analysts scheme.

  An Initial Public Offering is the most lucrative transaction any Wall Street firm can arrange for a corporate client. It is a special time in a company’s life: the moment it finally “arrives” in the public financial markets; the time when the corporate executives and venture capitalists who have invested money, time, and sweat finally raise money from large numbers of individual investors and establish a public market in which they later will be able to sell their shares. IPOs warranted high fees because they were risky and required more work than a typical deal. Securities regulators drew a bright line between private and public companies, and required that any company doing an IPO file an especially lengthy set of financial statements describing the company’s risks.

  The fees for almost all IPOs were set at exactly seven percent, ten times the fees from most structured derivatives transactions, and much more than the fees from any other investment-banking deals. Some economists were suspicious of these seven percent fees, and argued that they were evidence of an antitrust conspiracy among banks to charge high fees and avoid competition. They were right to be suspicious, but wrong about the argument. In reality, the seven percent fee was just a fraction of the fees the top banks made from IPOs. This was the IPO scandal. Here are some examples of how it worked.

  CS First Boston did more IPOs for technology companies in 1999 and 2000 than any other securities firm, thanks mostly to Frank Quattrone’s relatively small group. Consider one example: in July 1999, CS First Boston was the lead underwriter for the IPO of Gadzoox Networks, an Internet company.37 That meant that Frank Quattrone and his group were the bankers charged with primary responsibility for selling shares of Gadzoox to the public, including the road show during which they described the company to various institutional investors. Frank Quattrone didn’t directly call the little old lady in Peoria to solicit an order; individual brokers at CS First Boston did that. But Quattrone created the “story” for the company, which individual brokers would then use in selling shares.

  CS First Boston had about 3.4 million shares of Gadzoox to sell. The key issue, other than the story, was price. How should Quattrone advise Gadzoox about what the IPO price should be? The Gadzoox insiders wanted the price to be as high as possible, because all of the money raised in the IPO, less fees, went to the company. But CS First Boston’s clients wanted the price to be as low as possible, so they could buy at the cheap IPO price and then sell later at a higher price. CS First Boston was conflicted: it wanted a higher IPO price so that its seven percent fee would be higher, but it wanted a lower IPO price so that it would be sure to sell all of the shares. Historically, IPOs had been priced at a slight discount to the price the bankers believed the stock would trade at during its first day, as a compromise among the company doing the IPO, the bank, and its clients. As noted above, by 1999 this “IPO discount” had increased from a few percent to an average of 70 percent. It appeared that companies were agreeing to give up proceeds in return for rewarding the bank’s clients.

  The Gadzoox IPO was a mirror of the Netscape IPO in 1995, which had gone from $28 to $71 the first day. On July 20, 1999, CS First Boston sold Gadzoox shares in the IPO for $21, and by the end of the day, the shares were trading for $74, an increase of more than 250 percent. 38 The lucky clients who had bought in at the IPO price appeared to have made—in aggregate—more than $180 million. The Gadzoox IPO was not unusual for CS First Boston; the bank did a similar IPO the very next day for MP3.com, an Internet music distributor, in which it sold 7.2 million shares in the IPO at $28. Shares of MP3.com were trading at $63 by the end of the day, generating a one-day gain of more than $250 million for clients.

  During the days surrounding these two IPOs, some of CS First Boston’s clients began behaving strangely, and it was this type of behavior that tipped off a few reporters, and ultimately the Securities and Exchange Commission, to the IPO scheme. For example, one client who received 4,000 shares of Gadzoox and 10,000 shares of MP3.com, and therefore had a one-day profit of more than half-a-million dollars, began frenetically trading tens of thousands of shares in unrelated stocks, such as Allstate, Coca-Cola, Conoco, and Philip Morris, all at the very high commission rate of $1 per share, all at the same time as the IPOs.39 Another client who received 2,500 shares of Gadzoox and 5,500 shares of MP3.com traded 180,000 shares of various unrelated companies, paying commissions to CS First Boston on those trades of $124,000. Still another received 7,200 IPO shares and traded 210,000 shares with CS First Boston at high commissions. And so on.

  In addition, CS First Boston allocated shares to the venture-capital investment funds that sent their best corporate clients to CS First Boston. For example, Technology Crossover Ventures, a firm with close ties to Quattrone, was allocated 50,000 shares in the December 1999 IPO for VA Linux. Those shares went up by more than $10 million the day of the IPO.40

  This pattern was repeated in numerous IPOs arranged by CS First Boston, such as El Sitio, OTG Software, and Selectica. The most infamous example was the December 1999 IPO of VA Linux, which went from $30 to $239.25 the first day, an increase of nearly 700 percent. One “lucky” client received 13,500 shares of VA Linux at $30, for a total cost of $405,000. By the end of the day, those shares were worth $3.2 million. This grateful client immediately placed a trade for 2 million shares of Compaq at a total commission to CS First Boston of $1 million.

  Why were clients engaging in such strange behavior? In simple terms, clients of CS First Boston were buying underpriced IPO shares, pocketing substantial one-day profits, and then trading hundreds of thousands of unrelated shares with CS First Boston at very high commission rates. The commissions were as high as $3 per share—compared to the standard commission rate of six cents per share—and clients typically sold the stock they bought through CS First Boston immediately, sometimes simultaneously. In each case, the value of the commissions paid to CS First Boston was one-third to two-thirds of the client’s IPO profits. In aggregate, these payments represented as much as one-fifth of the firm’s total revenues from commissions.41

  It didn’t take the SEC long to conclude that clients were paying CS First Boston kickbacks through high commissions in return for shares in an IPO. In fact, CS First Bos
ton employees left little doubt that the IPO scheme was bribery, pure and simple. There was extensive evidence that CS First Boston employees told clients they were expected to repay a portion of the IPO profits in the form of excess commissions. CS First Boston tracked the percentage profits of various clients and told them they should maintain a ratio of 3-to-1 IPO profits to commissions. In early 2000, a senior executive at CS First Boston reportedly told a fund manager, “You get $3—we get $1.”42 Another senior executive described a client as being on “the 4 to 1 plan which is generous.” Other salesmen told customers they were expected to repay CS First Boston 50 percent of their profits from IPOs or they would not receive IPO shares in future deals.43

  Some clients were told to kick back to CS First Boston as much as 65 percent of the money they made on IPOs. One salesman reported to his boss that he had informed a client “that he was very far behind on his commissions and that we expect a 65% return on all money that we make him. I said that he still owes us for the VA Linux deal not to mention the deals that have come since then. I then stated that he can do trades to increase his commissions but he will be further cut from any syndicate in the future.” One especially slick salesman told a client, “Okay we got another screaming deal and I weasled [sic] you guys some stock we’ve yet to see any leverage out of your guys for the free dough-re-me does it make sense for me . . . to continue to feed your guys with deal stock or should I take the stock to someone who will pay us direct for the allocations.”44

  Some clients who received IPO shares in so-called “Friends of Frank” (Quattrone) accounts were CEOs of technology firms who were in a position to give those banks investment-banking business. These deals were not limited to CS First Boston; indeed, bankers later would defend them as standard custom and practice in the industry. Bankers regarded IPO allocations as no different from frequent flier miles: if you transacted enough times, you would receive a reward.

  In any other industry, federal regulators with evidence of these schemes would have impaneled a grand jury, shut the firm down, and prepared to send the perpetrators to jail. There was plenty of proof of a kickback scheme, and the victims were clear: individual investors who did not receive preferential treatment and, instead, bought at higher prices in the market. But the SEC did not have the authority to bring a criminal case; it had to persuade the Justice Department to do that. And the Justice Department didn’t have much of an appetite for complex financial cases, given the deluge of criminal activity in other areas, including the illegal-drug trade, health-care fraud, and terrorism.

  Instead of cooperating with the SEC, Allen Wheat, who had been running CS First Boston for three years at this point, casually dismissed the investigation. Anyone who had worked on Wall Street during the 1990s knew that the government rarely brought criminal cases alleging complex financial fraud; besides, given the aggressive ethos at CS First Boston, why would employees think they had done anything wrong? Wheat had created the same kind of mercenary culture that had pervaded his previous firm, Bankers Trust. He constantly asked employees, “How much did you make for us today?”45 (just as Charlie Sanford, Allen Wheat’s boss from Bankers Trust, had relentlessly questioned employees, “What’s next?”). Because the answer for compliance officers—the employees at banks who were supposed to police the traders and salesmen—was “nothing,” they were treated with disdain. On the other hand, CS First Boston paid bankers who generated profits more than any other bank, in part because its best employees constantly threatened to leave unless Wheat upped their pay. One managing director at CS First Boston said, “The only culture here is greed. You come to work 50 weeks a year for the one day when you are told how much your bonus will be.”46

  Wheat was a gambler and, in 1998, when CS First Boston lost about $3 billion in Russia, Wheat had bet on Quattrone, giving him not only a cut of profits, but also his own fiefdom, complete with salespeople, media relations, and research analysts. Until 2001, the bet on Quattrone had paid off, and was responsible for a good chunk of Allen Wheat’s personal fortune of several hundred million dollars. Wheat wasn’t about to remove his bet on Quattrone, now that a few regulators were sniffing around.

  By giving Quattrone special treatment, Wheat created an environment rife with conflicts of interest. For example, Wheat gave Quattrone and other members of his group $25 million a year to invest in potential clients. In June 1999, Quattrone and a few partners invested $126,000 in a private Internet software company called Interwoven, Inc. Four months later, they did an IPO for the company, and then later sold their shares—near the stock’s peak—for more than $2 million,47 a return that surpassed even Hillary Clinton’s profits on cattle futures.

  As news about the SEC’s investigation leaked, CS First Boston tried to blame three stockbrokers in the San Francisco office where Quattrone worked—Scott Bushley, Michael Grunwald, and John Schmidt. CS First Boston issued a press release on May 1, 2001, saying that Frank Quattrone was not responsible for overseeing IPO allocations, brokerage accounts, or commissions, contradicting Randall Smith, a top reporter for the Wall Street Journal, who had covered Quattrone for several years, and who called him a “control freak” who was involved in every step of the IPO process.48 The press release also was contrary to evidence later submitted by the Commonwealth of Massachusetts, indicating that Quattrone had the ability to “hire and fire” research analysts and determine their compensation.49

  In June 2001, instead of disciplining Quattrone or his bankers, CS First Boston fired the three brokers in the San Francisco office, and issued a press release saying the firings had followed an “extensive internal review” and noting that the firm had “informed the appropriate regulators of our action.”50 But CS First Boston was merely shooting the messengers: Scott Bushley was a junior employee; Michael Grunwald was a recent hire from Lehman Brothers, who was simply following past practices at CS First Boston; and John Schmidt was a conservative, bookish sales manager whose actions were approved by his bosses and by firm lawyers.51 It hardly seemed plausible that these three men had orchestrated the firm-wide scheme.

  When the National Association of Securities Dealers—the self-regulatory organization that governs securities dealers—began investigating the IPO scheme, CS First Boston submitted a 121-page argument in defense, claiming that the fees it made from the commissions—although perhaps high, at three percent—were perfectly appropriate, because they were less than a bright-line five percent rule for excessive commissions.52 The filing was made confidentially before the NASD, but two reporters from the Wall Street Journal—Susan Pulliam and Randall Smith—obtained a copy.

  In the filing, CS First Boston essentially admitted to the IPO scheme, but defended it as consistent with established rules and practice. Consider the following statement, from the filing: “There is absolutely nothing written in any guideline, rule, regulation, case or speech by a regulatory official that forbids the voluntary payment by clients of large commissions to CSFB to demonstrate that such clients are good enough customers to deserve being given IPO allocations.” Or this one: “The practice of rewarding better customers is neither new or unknown in the industry or investing public.”53 CS First Boston was using the classic teenager’s defense: everyone’s doing it. Average investors who were not “good enough” to deserve IPO allocations were not happy to learn about their lowly positions.

  Lukas Muehlemann, the chairman of Credit Suisse, which controlled CS First Boston, wasn’t happy about how Allen Wheat was handling the investigation. Wheat had brushed with regulatory authorities all over the world, and had paid little attention to the details of controls or compliance. For Credit Suisse, the IPO scheme was the last straw. Muehlemann told his managing directors, “We have been perhaps a little careless in the past. Now we will not tolerate that.”54 Allen Wheat was fired in July 2001.

  Wheat’s replacement was John Mack, who had battled with Frank Quattrone years earlier, and who had just lost his own battle for control of Morgan Stanley. Mack ca
ptured the culture Allen Wheat had created at CS First Boston when he called an early conference of the firm’s managing directors and showed a video clip of John Belushi from the movie Animal House.55 As Mack put it, “This firm has a history of tolerating cowboys. I don’t like cowboys.”56

  Not surprisingly, Mack committed to reinforce the firm’s compliance and legal departments. He immediately hired Gary Lynch, the former Milken prosecutor who had written the report for Jack Welch about Joseph Jett’s losses at Kidder Peabody. Lynch became the general counsel of CS First Boston, and a member of the firm’s board of directors. Mack and Lynch understood that if prosecutors filed criminal charges against CS First Boston for the IPO scheme, the firm probably would not survive (just as criminal charges brought against Arthur Andersen later assured the death of that firm). Lynch was charged with settling the case right away.

  After the terrorist attacks of September 11, 2001, and the collapse of Enron, the media coverage of the IPO scheme slipped from the front pages. On October 3, 2001, Mack and Quattrone sat down to have a little chat, over steaks in Kansas City, Missouri. The details of their conversation remain secret, but two conclusions were clear; first, Mack somehow persuaded Quattrone to give up his hefty pay package;57 second, Quattrone and CS First Boston would settle the investigation of the IPO scheme as quickly and quietly as they could. After that, Quattrone could find something else to do at the firm, or elsewhere.

  Gary Lynch negotiated the settlement beautifully. The public documents omitted the most serious allegations; instead, CS First Boston was charged with a technical violation of mislabeling the extra payments it had received from clients as “commissions” instead of as “IPO fees,” which they really were. The implication: if CS First Boston had simply disclosed that it was receiving a 65 percent IPO fee, its actions would have been perfectly legal.

  Prosecutors dropped the criminal case against CS First Boston. The firm’s $100 million fine represented roughly one year of pay for Quattrone, and only a fraction of the fees technology companies paid to CS First Boston the previous year.58 Lynch happily signed the settlement agreement in January 2002.