Bye-bye, ‘BUY’?

Wall Street has a lot of explaining to do.

Wall Street has a lot of explaining to do.

Over the past couple of years, the securities business has come in for a storm of criticism for the poor quality of its research. Investment houses have been assailed with congressional inquiries, private lawsuits and media scorn. The main evidence of Wall Street’s mindless cheerleading: analysts who persisted in slapping buy recommendations on just about everything, even as the markets were tanking. The backlash continued last month when New York State Attorney General Eliot Spitzer alleged that Merrill Lynch & Co. analysts promoted to retail customers the stocks of investment banking clients, while deriding the same companies in internal e-mails. Now the Securities and Exchange Commission is “requesting” information on firms’ research practices.

Wall Street has been trying to battle back. Last month Merrill chairman and CEO David Komansky publicly apologized and took charge of negotiating a settlement with Spitzer. Goldman, Sachs & Co. reorganized its operation so that its head of research would report directly to the CEO. HSBC Securities declared last fall that henceforth for every buy recommendation it issued, it would make at least one sell call. And the Securities Industry Association has promulgated voluntary “best practices” standards designed for all research groups.

But for all this activity, one thing hasn’t changed. Based on an examination conducted by Institutional Investor , using data provided by First Call/Thomson Financial of analyst recommendations at 169 brokerage firms , Wall Street remains, astonishingly, as bullish as ever. Sell recommendations on Wall Street are as rare as a seven-figure bonus. In March 2000, at the peak of the market, the average brokerage house mix of stock ratings was 73 percent buy, 27 percent hold and less than 1 percent sell. After a year of market decline, the breakdown in April 2001 was 69 percent buy, 30 percent hold and 1 percent sell. Today, after all the recriminations and supposed reforms, that distribution has barely changed: On average, 62 percent of recommendations are rated buy, 35 percent hold and 3 percent sell (see table below).

“The big Wall Street firms are not interested in objective, independent research,” says Jack Hidary, co-founder of Vista Research, a new boutique that provides research to institutions by subscription. “They cover your stock because you’re a banking client, or because they want you to be a banking client.”

Wall Street’s byzantine ratings system can resist easy analysis. To perform this study, First Call consolidated sometimes excruciatingly subtle ratings into a uniform group of five ratings: strong buy, buy, neutral, sell and strong sell. Institutional Investor then further refined the ratings into three: buy, hold and sell. “Securities analysts go out of their way to obscure the real recommendation they are making to keep their sources of information and their new business,” says Samuel Hayes, professor emeritus of investment banking at Harvard Business School and a member of a congressional review board that last year recommended reforms for Wall Street research.

To be sure, there are some exceptions, notably at Morgan Stanley. Last year the firm’s analysts had 64 percent buys to 36 percent holds , and not a single sell recommendation. Today, after a major revamping of its ratings system, Morgan Stanley’s distribution is 33 percent buy, 45 percent hold and , hold your breath , 22 percent sell.

Why is Wall Street still clinging to its positive bias? For starters, investment banks don’t want to anger companies who pay high margin fees for banking business. Pressure from institutional investors on analysts gets less publicity, but it plays a role in keeping down sells as well. “I have definitely had institutional investors call me telling me to fire an analyst who put a sell recommendation on a stock they owned,” says Sallie Krawcheck, chairman and CEO of Sanford C. Bernstein & Co. “But I never had an investor call me and tell me to fire an analyst who put a buy rating on a stock they didn’t own.”

Analysts argue that one reason for their positive bent is the simple fact that they choose good , not bad , stocks to cover. And, they note, at this point in the market cycle, shares are more likely to rise than fall. “When an analyst makes a decision to cover a stock, there’s a natural bias toward companies with business models that will make money for the clients,” says Joseph Amato, director of global equity research at Lehman Brothers.

Whatever the reason, there will soon be nowhere to hide. The National Association of Securities Dealers and the New York Stock Exchange have proposed new rules that will require every brokerage firm to boil down the distribution of its recommendations into simple, if hypothetical, “buy, hold, sell” percentages. In research reports banks will be forced to graph the price of each stock and indicate at what point their analyst changed the rating or price target. And firms will also be required to disclose any corporate finance business they’ve performed for the company , or anticipate doing (see box, page 30).

Whether these new rules will alter Wall Street’s penchant for boosterism remains to be seen. But some research officers believe they may finally push the sell side to change. “Analysts will feel a little bit more pressure to be a bit more accurate on their recommendations,” says Alfred Jackson, global head of securities research at Credit Suisse First Boston. “I think you’ll see people willing to change their ratings more quickly.”

But sell remains a word that’s rarely heard on Wall Street. Only three firms , according to the analysis by II and First Call , rate more than 5 percent of the stocks they cover as sell. In addition to Morgan Stanley, they are newly formed research boutique Fulcrum Global Partners and Prudential Securities. Fulcrum does no investment banking; Prudential bowed out of that business in 2001. Morgan Stanley recently overhauled its four-tiered rating system (with two categories of buy) to a three-tiered one , overweight, marketweight and underweight , in which each stock is measured against the other stocks an analyst covers, rather than against a broad market benchmark, such as the Standard & Poor’s 500 index.

The findings contain some surprises. Take Bernstein, which prides itself on its independence because it has no investment banking. It rates only 1 percent of its stocks as sell. Institutional investors consider Bernstein, by a long measure, to be the research house with the most integrity on Wall Street (Institutional Investor, December 2001). “Typically, we have had around 10 percent sells,” explains Bernstein research director Lisa Shallet. “We’ve had a large number of upgrades this year, and we think that makes sense based on where we are in this cycle.” Indeed, last year 6 percent of Bernstein’s calls were sells, the highest of any firm.

Six firms have at least twice as many buys as they do holds, and almost no sells: Bear, Stearns & Co., CIBC World Markets, Citigroup/Salomon Smith Barney, Goldman Sachs, Lehman Brothers and SG Cowen.

What does Wall Street have to say for itself? Not even research directors try to justify all the Street’s ratings practices, but they point to some nuances that mitigate the problem. For example, they note that ratings aren’t everything. “We sell research, not recommendations,” says Krawcheck of Bernstein, which is now part of Alliance Capital Management.

The few firms with lots of sells see them as affirmations of their independence. By requiring analysts to compare the companies they cover with each other and not just the broad market, Morgan Stanley has sent sells soaring. A stock that might have been considered attractive or neutral relative to the broader market might be less so when compared only with its peer group covered by the same analyst.

“We think we’ve come up with a significant improvement on the ratings system,” says Dennis Shea, Morgan Stanley’s director of global equity research. “And we’re encouraged by the way all of our clients , both corporates and institutional investors , have reacted.”

Will other firms follow Morgan Stanley’s lead? Some are already actively considering changes, and others are sure to follow. But don’t expect any major investment banks to go as far as Fulcrum. The independent research group started in May 2001 and issues only buys and sells. “A hold to me is a foxhole that you can jump into and hide while you make up your mind,” says Mike Petrycki, the firm’s founder. “Fundamentally, unless you’re just focused on what’s going to happen tomorrow or next week, there are only two ways to go. And you should have an opinion one way or the other.”

One important critic with a clear opinion is New York State Attorney General Spitzer. Last month, wielding an 80-year-old law, he savaged Merrill Lynch, calling the brokerage firm’s research “tainted” by a conflict of interest that “jeopardizes the integrity of the marketplace.” Spitzer obtained a court order from the state supreme court requiring Merrill to disclose in each research report whether it has , or expects to have , an investment banking relationship with the company being covered. Spitzer has since served subpoenas on other Wall Street firms and has said he is considering filing civil and criminal fraud charges against Merrill. Spitzer noted that since late 1999 the firm’s Internet group has not issued a single reduce recommendation, never mind a sell. In that time the Internet sector sank by 64 percent.

Ironically, before Spitzer’s attack Merrill had been trying to clean up its research system. In June 2001 the firm added a strong buy to its ratings, eliminated accumulate and combined reduce and sell into one category. Other firms, including Deutsche Bank, Jefferies & Co. and Prudential made similar changes. Many adhered to the SIA’s voluntary guidelines: that analysts not report to investment banking; that they clearly disclose their personal holdings in stocks they cover; that they apply the sell rating more frequently; and that they not sell their own holdings before they downgrade a stock.

While research departments grumble, at least one person is hopeful, particularly since the SEC is expected to soon approve the NASD’s mandatory research disclosure regulations.

“I’m very encouraged by the NASD proposal,” says First Call director of research Charles Hill, who has been busy tallying buys and sells. “I think we’re going to get more ‘Say what you mean and mean what you say’ from Wall Street.”

A losing battle

Chastened Wall Street officials say they support the efforts of Congress and federal regulators to restore the integrity of equity research and, by extension, the stock market itself. But many are also seething over some of the less-publicized provisions of the reforms proposed by the National Association of Securities Dealers and the New York Stock Exchange.

Wall Street firms have several problems with the proposals, which were hammered out with input from research firms but under intense pressure from legislators and the Securities and Exchange Commission. A major sticking point is the requirement that firms disclose in research reports whether they have received compensation from corporate clients during the past year , or whether they expect to during the next three months. Investment banks as a matter of principle are always loath to air the details of client relationships. But some are especially concerned that the new rule will unfairly tip the market to pending mergers and acquisitions.

“If we have just taken on a new client, where the assignment is a hostile takeover bid, and all of a sudden we publish a research report on the company that contains this new disclosure that we expect to receive investment banking compensation from them in the future, doesn’t that signal to our analysts, other analysts, investors, that something is up?” asked John Faulkner, a capital markets lawyer at Morgan Stanley, during a conference about the new rules sponsored last month by the Securities Industry Association. “People are going to catch on.”

Another bone of contention: Firms must disclose ownership stakes of 1 percent or larger in any of the companies they research. For financial conglomerates with brokerage, asset management, merchant banking and other securities operations throughout the world, determining whether they exceed the disclosure threshold will require building new systems, which could take a year or longer.

Then there is the issue of what exactly constitutes a research report under the proposed rules. Some firms worry that research notes, updates and summary publications that supplement in-depth reports on individual stocks and sectors will be rendered unusable by reams of additional disclosure. Credit Suisse First Boston, for instance, says its “Monthly Review & Comment,” which took up 253 pages in April, would weigh in at an indigestible 1,500 pages. Its nine-page daily “Morning Highlights” would grow to 63.

But the most worrisome of the proposals’ provisions is the one calling for compliance officials to mediate and document virtually all communication between research analysts and investment bankers. Research officials complain that even with an army of such intermediaries, there is no way to monitor every phone call, e-mail or whisper in the elevator. Furthermore, they argue, this kind of chaperonage will discourage even innocent, beneficial contact between bankers and analysts.

“Every conversation that an analyst has with a banker is not an evil conversation about how to screw the retail investor,” says one big-firm research head. Often analysts use their depth of understanding about a company or sector to augment the advice an investment banker might give to management regarding capital markets transactions or mergers.

The additional printing and lawyering has led some on Wall Street to refer grumpily to the proposals as the “Lawyers and Lumberjacks Full Employment Act of 2002.”

Leading the charge for making the new rules less taxing to implement is the SIA. The group last month submitted a lengthy comment letter on the proposed rules, which must be approved by the SEC before taking effect. Among the recommendations: Exclude re-packaged, or summarized, earlier research from the disclosure requirements; limit disclosure of compensation to past investment banking fees only; and raise the 1 percent ownership threshold to 5 percent.

But significant changes aren’t likely. Regulators fear that in their attempt to accommodate even legitimate concerns, they might also create loopholes through which firms could creatively violate the spirit of the rules.

“We will carefully consider all the comment letters,” said NASD Regulation chief Mary Schapiro to an audience of about 200 Wall Streeters during last month’s SIA conference in New York. “But you need to understand that in Washington and in the country, there isn’t any sympathy for the burdens that these new rules are going to put on the industry.”

Put more bluntly, because of its recent performance, Wall Street has forfeited the right to the benefit of the doubt. , J.A., J.S.