Infectious Greed

by Frank Partnoy

  The media portrayed the case as a significant victory for securities regulators, and any criticism seemed to come from an undistinguished peanut gallery. Even John Gutfreund—the chairman of Salomon Brothers who had been forced to resign during the Paul Mozer scandal—emerged from the shadows to complain about CS First Boston’s settlement, saying he was “dismayed” and that the fine was just a “slap on the wrist.”59 (It was less than the fine Salomon paid in 1992, after Gutfreund resigned.)

  More than a thousand private lawsuits were filed, by investors in 263 companies.60 Those cases were a thorn in John Mack’s side, although CS First Boston would benefit from the legal changes of the mid-1990s that made such lawsuits more difficult. CS First Boston suffered from a decline in IPO fees, which plunged almost 100 percent, but that was primarily due to a widespread decline in the market. During the first six months of 2002, the firm did just one IPO, for Simplex Solutions, Inc., and made just $3.4 million, barely one week’s worth of compensation for Frank Quattrone.

  Some commentators have called Frank Quattrone the Michael Milken of the 1990s. There were a few similarities, but in some respects the comparison wasn’t fair to Milken. Both men were Wharton grads who quickly rose to the top of their firms and negotiated similar compensation packages. Quattrone symbolized the mania associated with Internet stocks in the same way Milken symbolized the 1980s obsession with corporate raiders and takeovers. Regardless of whether Quattrone “measured up” to Milken, Milken’s punishment arguably was too harsh, while Quattrone’s treatment arguably was too generous.

  The scheme involving securities analysts was closely related to the IPO scheme, but was much simpler to describe. In fact, three words were enough: “pump and dump.” Securities analysts at investment banks pumped up stocks, especially those recently issued in IPOs, with overly optimistic reports. Individual investors believed the hype and stock prices remained high until the 180-day lockup period during which corporate insiders were prohibited from selling shares. After the lockups expired, insiders dumped their shares.

  It wasn’t always this way. Before 1990, securities analysis was a respected profession, and investors valued the quality of analysts’ research and the independence of their opinions. Analysts told investors not only when to buy but, even more important, when to sell. Today, business journalists uncover most financial fraud; ten years ago, that was the job of securities analysts.

  As markets became more efficient, and information began flowing more quickly and inexpensively, it became difficult for research analysts to add much value. Investors weren’t willing to pay for research reports if the market price of a stock already reflected the information. Consequently, analysts faced pressure to add value in other ways: by helping investment bankers solicit business from the companies they covered.

  In 1990, Clayton J. Rohrbach III, a senior officer of Morgan Stanley, suggested in an internal memo that analysts’ compensation should be tied to how much business the companies they rated gave to Morgan Stanley. Rohrbach suggested that analysts themselves receive explicit ratings of A to C. Morgan Stanley never formalized this policy, because it presented obvious conflicts of interest, but analysts at the firm reported that their pay was based on this grading. Most Wall Street firms, including CS First Boston, adopted similar systems of linking investment banking and research.61

  These systems remained secret for a decade, until the market finally collapsed and analysts began confessing their sins. In 2002, Gretchen Morgenson of the New York Times persuaded several analysts to tell her about their experiences on condition of anonymity. They spoke of being “not so much an analyst as a marketing machine” and said analysts’ pay was tied to the investment-banking business they brought in.62 Charles Gasparino of the Wall Street Journal uncovered documents that offered analysts “1% to 3% of the firm’s net profit per transaction” or 8.5 percent of revenues that “the analyst is clearly instrumental in obtaining,” or that stated “Banking Related Compensation: You will be paid banking related compensation. . . .”63

  Bankers and clients also pressured analysts to avoid making negative or controversial comments. For example, when David Korus, an analyst at Kidder Peabody (just before the firm went under after Joseph Jett’s losses), questioned some of Dell Computer’s currency trading, the company threatened to sue Kidder, and barred Korus from meetings.64 It is not surprising, then, that according to a May 1998 survey by the First Call Corporation, just under two-thirds of analyst recommendations were “buy” or “strong buy,” one-third were “hold,” and just one percent were “sell.” In 1990, there were fifteen times more “sell” recommendations than “buys.”65 It wasn’t that analysts had some evil intent; they were simply responding to an incentive system that rewarded them financially for making positive comments, and punished them for making negative ones. Analysts began inflating ratings for the same reason children acquire good manners.

  In addition, as financial statements became more complex, analysts—like investors—didn’t have time to scrutinize financial statements. By the late 1990s, even an experienced stock analyst would need at least a day to read an annual report carefully. But a typical analyst covered fifteen or more companies, and spent most of the day on the telephone or in meetings with investors and clients. The easy way out was to accept a company’s own profit estimates and label it a “buy.”

  By 1998, most investment bankers recognized that a high-profile securities analyst could help them get business. After Frank Quattrone arrived at CS First Boston, he noted, “We don’t compete against Morgan Stanley—we compete against Mary Meeker.” Mary Meeker was an Internet analyst who consistently received top rankings from Institutional Investor magazine, which rated analysts based on a poll of fund managers, just as U.S. News & World Report ranked colleges and graduate schools. Experts criticized these rankings, but they were the most credible and simple rankings available, and individuals trusted them. Just as Princeton University attracted students based on its number-one ranking, Mary Meeker attracted investors.

  A cozy relationship with a company helped a top analyst, too. Company officials “selectively disclosed” certain information to their favorite analysts, but not to the general public. Companies also prepared their favorite analysts in advance of public announcements, so that they could adjust their expectations and accurately predict corporate-earnings announcements,66 almost like magic.

  In short, corporate clients wanted a high-profile analyst to do the “pumping” before they did the “dumping.” This scheme didn’t bother investors as long as stock prices continued to increase. When the analysts said “buy,” they generally were giving good advice, at least in the short run.

  The key difference was that now it was the analysts who were lying, instead of the companies. Years earlier, analysts had accurately predicted the “numbers” for Cendant, Waste Management, Sunbeam, and Rite Aid, too. But in those cases, companies had lied to both the analysts and to investors. Investors could continue to trust analysts, even if they did not trust the companies.

  Now, the greed that earlier had consumed corporate executives was spreading to securities analysts. If CEOs and investment bankers could make tens of millions of dollars per year, why couldn’t analysts do the same? All they had to do was rate companies higher than they otherwise would and persuade those companies to do business with their banks. With the links between banking and research, they could then claim the right to a ten-million-dollar-plus bonus. The downside was limited: criminal liability seemed out of the question, and they could always claim they genuinely believed in a company and were surprised—no, shocked—to see a company go under. Besides, federal regulators had shown no interest in pursuing analysts, who were, according to traditional economic theory, providing an important service to financial markets.

  The analysts hadn’t counted on Eliot Spitzer, the attorney general of New York. Although federal securities regulators had primary jurisdiction over financial markets, state regulators also
had the ability to prosecute financial fraud. Federal and state prosecutors in New York had fought turf battles for decades, and often negotiated parallel investigations, with the U.S. attorney for Manhattan taking one case and the New York district attorney taking another.

  This time, there were no negotiations. Spitzer was politically ambitious and the federal regulators—led by Harvey Pitt, the new chairman of the Securities and Exchange Commission—were not aggressively pursuing many cases. Pitt was an able lawyer, but was constrained by perceived conflicts of interest from his previous job representing the major accounting firms and Wall Street banks. That left an opening for Spitzer.

  Spitzer issued subpoenas for the e-mail records of analysts at several Wall Street banks, including Merrill Lynch, where Henry Blodget—of Amazon fame—was the top Internet analyst, with a guaranteed annual compensation in 2001 of $12 million.67 Blodget was one of the most popular Internet analysts, and Merrill had done numerous high-profile Internet deals, including eToys, Excite@Home, InfoSpace, Internet Capital Group, iVillage, Pets.com, Quokka Sports, and Webvan. Needless to say, these companies hadn’t done very well, and were the subject of numerous lawsuits.

  Spitzer uncovered documents confirming that the greed infecting corporate executives had spread to securities analysts. In short, the e-mails were the “smoking gun.”

  Spitzer submitted an affidavit to a New York court excerpting some of these e-mails. Investors read edited versions of some of them in the newspapers; others were not fit to print. But, as a whole, the lengthy affidavit was a powerful indictment of how analysts were behaving at Merrill Lynch. If other firms were the same, the system of securities ratings was totally rotten.

  Like most banks, Merrill Lynch had a stock-rating system, ranging from 1 to 5. A rating of 1-1 was the highest, then 1-2, 1-3, and so forth. A rating in the general category of 1 or 2 was positive. A 3 was neutral. Merrill’s Internet group never ranked companies a 4 or 5; instead, they simply stopped covering the stock.68

  The e-mails showed that analysts were publicly issuing high ratings while privately ridiculing the same stocks. For example, analysts privately described stocks with a neutral 3 rating as “crap” or a “dog” or “going a lot lower.”69 Stocks with a positive 2 rating were repeatedly described as a “piece of shit” or “such a piece of crap.”70 Internet Capital Group, Inc., with a 2-1 rating, was described as “Going to 5.” InfoSpace, which Merrill gave its highest rating of 1-1, was nevertheless described in e-mails as a “piece of junk” and a “powder keg.”71

  On April 8, 2002, just after the affidavit was filed, Merrill issued a statement that “E-mails are only one piece of a continuous conversation, isolated at a single point in time—not an end conclusion.” A Merrill spokesperson cautioned investors not to take these e-mails out of context, and disputed Spitzer’s allegations.

  But in context, the e-mails made Merrill look even worse. Consider the “Going to 5” comment about Internet Capital Group. Merrill had maintained a positive 2 rating for Internet Capital Group and kept the stock on its “top-ten” list of technology stocks, even though Henry Blodget was predicting privately that “there really is no floor to the stock.”72 But as the stock’s price fell from $200 to $15, Blodget was upset by the pressure to maintain a high rating, and he threatened to start “calling the stocks like we see them, no matter what the ancillary business consequences are.”73

  Or consider the “piece of junk” and “powder keg” comments about InfoSpace, which Merrill rated a 1-1. Blodget was pressured to maintain the high rating even as InfoSpace’s stock fell 80 percent, and Blodget expressed “enormous skepticism” about the stock and complained that “I’m getting killed on this thing.” Why? InfoSpace was planning to buy another Internet company, Go2Net, for more than a billion dollars, and Merrill—which represented Go2Net—would earn substantial investment-banking fees from the deal. After the deal was done, Blodget finally reduced his rating.

  Perhaps the most damning example—in context—was Merrill’s coverage of GoTo.com, an Internet search company now known as Overture Services, Inc. In 1999, Merrill lost its pitch to do GoTo’s IPO, and Merrill’s analysts did not issue a rating for GoTo’s stock. In 2000, Merrill again solicited investment-banking business from GoTo, and dangled a stock rating from Henry Blodget as a carrot. In September 2000, GoTo finally agreed to give Merrill some business, arranging for a European deal, and Merrill promised that Blodget would begin covering GoTo. When a fund manager asked Blodget, “What’s so interesting about GoTo except banking fees?” Blodget responded, “Nothin’.”74

  It would have been very time consuming for Merrill to prepare its initial research report on the company from scratch, so GoTo executives supplied the data and comments, and actually typed changes into the draft report. They even supplied the full text of particular sections. When Kirsten Campbell, a junior Merrill research analyst, questioned whether GoTo would become profitable before 2003, and suggested it deserved only a 3 rating, she became embroiled in a dispute with GoTo’s executives, who argued that the company would become profitable in 2002 and deserved a 2 rating. Campbell e-mailed Blodget that she did not “want to be a whore for f-ing management” of GoTo.

  A decade earlier, Campbell would have been respected for her interest in ensuring an accurate rating. But in 2000, she was merely causing trouble by raising concerns: “We are losing people money and I don’t like it. John and Mary Smith are losing their retirement because we don’t want Todd [Tappin, GoTo’s chief financial officer] to be mad at us.” No one wanted to hear Campbell’s complaints that “the whole idea that we are independent from banking is a big lie.”75 And no one supported her opinion that GoTo should be rated 3-2 at the highest.

  Meanwhile, GoTo’s stock had fallen below $10 a share. There was a brief circus as Merrill—which was prohibited by law from issuing new ratings on stocks with a price below $10—tried to decide what to do. Blodget, showing off his Yale education, said, “Waiting for $10 is waiting for Godot.” On January 10, the price briefly hit $10, and Blodget immediately announced coverage of GoTo at a rating of 3-1, higher than Campbell had recommended.

  Campbell left Merrill a few months later and, after she left, Blodget upgraded GoTo to a 2-1, and the stock rose 20 percent. Blodget then joined a group of investment bankers sponsoring a road show for a new stock issue GoTo was planning. Merrill hoped to be named lead manager for the deal, which would mean substantial investment banking fees.

  When GoTo indicated it was leaning toward CS First Boston and Frank Quattrone, instead of Merrill, Blodget was irate. He immediately began preparing to downgrade GoTo from the rating of 2-1. A Merrill banker complained, “Not only did Henry Blodget show leadership by initiating on the stock near its low point but he recently upgraded it and sponsored a set of investor and Merrill sales force meetings for management in New York, which dramatically moved the stock price.” One of Blodget’s assistants began drafting a memorandum explaining that Merrill was downgrading the stock on valuation, meaning that the price had risen too much (he noted that “I don’t think I’ve downgraded a stock on valuation since the mid-90’s”). The downgrade was held in reserve until GoTo decided who to name as lead manager for its offering. During the morning of June 6, 2001, GoTo filed its stock-offering documents, listing CS First Boston as lead manager. A few hours later, Henry Blodget issued a notice that he was downgrading GoTo to a 3-1.

  Blodget left Merrill a few months later, with a severance package reportedly worth $2 million.76 Merrill ultimately agreed to pay a $10 million fine and to reform its approach to research. But many major investment banks and analysts continued to assert that their high ratings may have been wrong in hindsight, but were perfectly legal. For example, Mary Meeker, the Morgan Stanley analyst, had not been accused of any wrongdoing and continued to receive big bonuses for 2002 (although perhaps not the eight-figure bonus she had received at the peak of the Internet boom).

  The final group of people profiting from th
e technology bubble was the most obvious: corporate executives. CEOs in 1999 were no different from CEOs a few years earlier, and all responded to the incentives created by earlier changes in law. Executive compensation rules continued to encourage stock-option grants, and CEOs with millions of options sought to increase share prices in the short run, even if they knew their actions jeopardized the long-term health of their companies. Likewise, limits on securities lawsuits continued to insulate executives, and their accountants and bankers, from liability. Not surprisingly, technology firms aggressively manipulated their financial statements, just like their predecessors at Cendant, Sunbeam, Waste Management, and Rite Aid.

  There were numerous “new-economy” examples of fraud allegations, ranging from Lernout & Hauspie to W. R. Grace, from Livent to Yahoo!, from Lucent to Navigant, from MicroStrategy to Xerox. But the stories were essentially the same. To the extent the problems at these companies differed from previous frauds, SEC lawyers tried to put the problems within one of the simple categories of fraud SEC chair Arthur Levitt had outlined in his 1999 “Numbers” speech.

  Consider MicroStrategy Inc., one of the many high-flying companies that did an IPO in 1998 and then crashed in 2001. Michael Jerry Saylor and Sanjeev Bansal cofounded MicroStrategy to provide custom software for companies to use in analyzing large databases. Their products were successful, and they had large corporate clients such as Kmart and NCR Corporation.

  MicroStrategy’s stock price told an interesting story. In June 1998, six months before Henry Blodget’s Amazon prediction, MicroStrategy did an IPO at $6 per share. The stock price went up 75 percent the first day—a huge increase, but merely average compared to other IPOs at the time. For more than a year, the stock price traded in a range of $10 to $18. Suddenly, in October 1999, the shares took off like a shot, doubling within a month and reaching $110 by mid-December. By March 2000, the peak of the Internet bubble, MicroStrategy was worth $333 per share.77 Then the stock collapsed, and by summer 2002 was back down to its IPO price of $6.