Holding Analysts Accountable

SAN FRANCISCO (06/05/2000) - On Jan. 17, Merrill Lynch analyst Henry Blodget appeared on CNN's Moneyline and said how much he liked Internet Capital Group.

Shares of the business-to-business incubator, which had fallen to $133 from $171 the previous week, have since fallen below $30.

Blodget couldn't have foreseen the subsequent carnage in Internet stocks, but he could have disclosed an important fact: Merrill was the lead underwriter of ICG's 1999 IPO. Anyone who took the analyst's words to heart was unaware that Blodget was praising an important client of his employer. It's one of Wall Street's best-known secrets, and a dirty one. Research analysts writing recommendations of closely watched Internet stocks routinely face conflicts of interest.

Securities firms stand to rake in huge fees from underwriting the stock offerings of Internet companies, often the same companies that the firms' analysts recommend. In addition, analysts sometimes hold shares of companies they cover. And yet the analysts give only the flimsiest of disclosures of their conflicts of interest. For the veteran fund manager who has direct access to a research analyst, it's just the way business is done. For less-sophisticated retail investors, who see the blizzard of analyst comments in the media, the conflicts aren't as apparent.

Merrill says its analysts' comments on TV aren't intended to supplant research.

They are "the verbal version of what they have previously written in a research report, which has been fully complianced," says Merrill spokeswoman Susan McCabe. For decades, research analysts had little to do with retail investors, working with the firm's sales team or talking only to the big-money managers.

But as the Internet has introduced a new generation of individual investors, many analysts have sought the limelight. Even those who shun the press see their recommendations reported in newspapers, on CNBC and on financial Web sites.

And concerns are mounting about how such conflicts affect an unsuspecting public. While Wall Street downplays the risk of analyst conflicts, a growing body of evidence is emerging from business schools to show their impact on investors. The Securities and Exchange Commission, for its part, is signaling its discomfort with the situation, a sign that regulators might require brokerages to increase disclosure of analyst conflicts.

At the SEC's prompting, the New York Stock Exchange and Nasdaq are considering ways to warn retail investors of the risks of analyst conflicts. Amy Highland, a spokeswoman for NASD Regulation, said the exchanges have met with the SEC several times and are addressing the concerns. But it's not yet clear what changes, if any, will result. SEC Chairman Arthur Levitt, who has targeted conflicts of interest at accounting firms, is turning his attention to research analysts. For retail investors, the threat is real: Individual investors often trade stocks based on an analyst's recommendation. Retail investors "are the ones most likely to be influenced by free research," says Greg Corso, general counsel to the SEC chairman. "It is particularly acute in the financial news, CNBC and the like. So many people are watching, but you never hear [the analysts] disclose they are related."

Most worrisome to the SEC is that analysts give stronger recommendations to companies that their firms underwrite, which has led to charges that analysts are a sales tool masquerading as an independent voice. For proof, regulators don't have to look further than the business schools that train many analysts.

In recent years, several academic studies have shown that securities firms produce earnings estimates and recommendations that are higher for clients than for nonclients. The standard response to these studies was that underwriters had better access to a company's fundamentals. In one recent study, Roni Michaely of Cornell University and Kent Womack of Dartmouth College discovered that stocks recommended by underwriters would pop after the recommendation and then drop like a stone in the following months.

Those recommended by nonunderwriters, however, would rise over the long haul.

After six months, stocks recommended by underwriters rose a median 15 percent, compared with a 20 percent gain by nonunderwriter recommendations. After two years, the underwriter-recommended stocks were down 52 percent while those recommended by nonunderwriters were up 23 percent.

"There is a bias in brokers' recommendations when they have an underwriting affiliation with a company," says Michaely. "Even more important, the public doesn't recognize it." Investment banks are reluctant to discuss any conflicts regarding their analysts. Several firms declined to talk about the issue. "I don't think we are going to comment on this one," said a Goldman Sachs spokesman. But Michaely and Womack surveyed 13 investment bankers who, when asked to explain the discrepancies in the study, chalked it up to conflicts of interest. They also surveyed 13 money managers, who agreed that the differences were caused by conflicts of interest.

"They know what they are doing, and they still recommend it," Michaely adds.

Investment banks have fine print in their research reports that mention the possibility of a conflict of interest by the banks' analysts or other employees. But rarely are investors informed on an individual basis. And investors grow desensitized to boilerplate disclosures, if they notice them at all, Corso said. What's more, many novice investors don't realize that such conflicts are encouraged. In a recent speech, Levitt pulled out an internal memo circulated at a Wall Street investment bank. "We do not make negative or controversial comments about our clients as a matter of sound business practice," the memo read. "The philosophy and practical result needs to be 'no negative comments about our clients.' " Analysts, like their banks, don't like to broach the topic of conflicts.

Former research analysts aren't so reserved: They bristle at the notion of plugging a stock they don't believe in. "It's not a question of conflict. It's a question of everybody tries their best and sometimes they fail," says Keith Benjamin, a former analyst at Robertson Stephens and a current partner at Highland Capital Group, a venture capital firm. "Do you think anybody would purposefully try to recommend a company that isn't a good company, and thereby look stupid?"

Benjamin and Danny Rimer, a partner at the Barksdale Group who covered Net stocks at Hambrecht & Quist (and is a contributor to The Standard), suggest that the studies showing the existence of conflicts are tainted with negative assumptions about investment banking. Banks don't underwrite stocks they think are bad investments. "We do a lot of work up front to understand whether we think this company is a good company," says Rimer, who offers another reason why recommendations from nonunderwriters do so well: They're just piggybacking on the successes of the underwriters. "Let's be clear, how many companies have independent analysts covering the stock unless it's a really, really strong company?"

Whatever the reason, it's clear that the world of Wall Street research is built to favor the banks and their clients at the expense of the retail investor - and that regulators are moving toward more accountability from analysts. Even those who welcome such a development say regulators will need to be careful not to bring about an onerous crackdown. "I think one would want to be very careful not to dry up coverage," says Maureen McNichols, a professor at Stanford University's Graduate School of Business who has authored studies on analyst biases. "A research report may be overly optimistic, but it still provides a fair amount of information that investors may miss out on."

Join the newsletter!

Error: Please check your email address.

More about Capital GroupCNBCCNNGoldmanICGInternet Capital GroupNASD RegulationSECSecurities and Exchange CommissionStanford UniversityWall Street

Show Comments

Market Place