Rating the Analysts' Confusion

The Merrill Lynch case is over, but research departments are no wiser about how best to say if a stock is worth buying, keeping, or dumping

By Jane Black

These days, the toughest questions on Wall Street are simple ones: What's your rating on a stock? And what does that rating mean? Ever since New York State Attorney General Eliot Spitzer made public his investigation of Merrill Lynch (MER ) on allegations that one of its analysts, Henry Blodgett, was promoting some stocks he was privately trashing in e-mail, analysts appear genuinely flummoxed.

Take a recent phone conversation I had with an analyst about beleaguered telecom Qwest (Q ). When I asked how he rated the stock, he said: "It's a hold. No, it's a market perform. Wait, hold on, let me ask someone what the hell we're calling this now."

This analyst -- whom I won't identify lest he lose his job -- is no fool. His insight into companies usually is on target. So it was to my surprise that the conversation only got worse. He didn't know if the rating, which we eventually established was market perform, was part of a four- or five-point scale. Nor was he sure what "risk qualifier" should be attached to it. And he admitted that market perform was a pretty dumb label for a stock he thinks has little to no chance of rallying.


  Welcome to the mixed-up world of equity research, a field that still cries out for standardized reform, even after Merrill's settlement with Spitzer was announced on May 21 (see BW Online, 5/22/02, "Will Merrill's Deal Reshape the Street?"). Wall Street's research analysts -- some 2,000 of them -- are clearly still wandering in the dark. Conversations that once were businesslike increasingly have an Alice in Wonderland flavor, as analysts struggle to figure out how to represent their opinions in a way that's objectively accurate but also keeps investment-banking clients happy.

Publicly, of course, Wall Street firms declare that their ratings systems are straightforward -- both for analysts and investors. But everyone knows that's hogwash. For now, only Merrill is instigating new ratings (after agreeing on May 21 to pay a $100 million fine and apologizing to clients for any research misrepresentations). Deutsche Bank (DB ), Goldman Sachs (GS ), and Credit Suisse First Boston (CSR ), to name a few, say they're sticking with their nomenclature -- though they also say they would be happy to make them conform to any new regulations from the Securities & Exchange Commission or stock exchanges.

Government or stock-exchange regulation may not be the answer in this case. Privately, official and around-the-water-cooler debates are raging about just what ratings should say and how analysts should use them. Analysts can't agree on how to reform their systems and practices, which many admit are flawed.


  Smart firms won't wait for the SEC to act. It's up to the management of Wall Street brokerages to take the lead in fixing the problem. The chaos is making analysts fearful, defensive, and sometimes angry. And that serves only to further confuse investors.

Some analysts argue that any new system should have just two ratings: buy and sell. "After all, that's all you can do with a stock," quips one analyst. "You can't market perform it."

Putting that advice into practice, however, would be very difficult. Although the Merrill settlement requires Wall Street's biggest firm to separate its research arm from its investment-banking division, analysts continue to be scared of dividing the world into a black-and-white stock-rating system. The problem: At most firms, analysts must continue to work within the confines of a system that all but requires them to be positive about the stocks of client companies, whether or not the fundamentals merit a bullish outlook.


  One benefit of having a tiered system is that analysts can rate a company favorably and still indicate that it may not be the best buy out there. Says one Wall Street analyst: "You need to protect your [investment] 'banking buys' -- the ones that show you're being nice but not pounding the table and saying it's your best idea."

Merrill has agreed to add new disclosure to research reports to reveal the names of companies for which the firm provides banking services. It's far too early to tell whether the rest of the Street will follow suit.

One big advantage to a simple buy-or-sell rating system would be the ease with which investors could figure out when analysts made mistakes. "Analysts always want to hedge," says one who covers the volatile telecommunications sector. "Before, when they were just part of the promotion machine, the [biggest Wall Street] firms were more willing to stick their necks out and say 'buy, buy, buy.' Now, they need to have a real assessment -- and they desperately want some cover." So the analysts turn to ratings such as "market perform," "accumulate," and "underweight."


  The debate has clearly divided the normally clubby analyst community. One analyst at a midsize investment firm described a two-hour meeting that focused on how to redefine ratings. As she described it, the session was a battle between the "traditionalists" and the "blamers."

Traditionalists are analysts who have never covered technology, media, or telecom -- the sectors hardest hit by the stock market's burst bubble. They believe in the school of equity research that focuses on getting inside details and providing innovative analysis of a company's stock.

Blamers, on the other hand, are analysts who either made the wrong calls or, through no fault of their own, have seen stocks blow up in their faces. Blamers fear that a radical change in current practices will reflect badly on them -- perhaps even costing them their jobs.


  And so the debate rages, and confusion reigns. At Deutsche Bank, analysts freely admit that they have "no fixed method" for assigning a rating. "At the end of the day, it's up to individual analysts," one says, adding: "There's no absolute system that says you have to operate in a certain way."

He's right. A strong-buy rating in aerospace or manufacturing may mean a stock should jump by 25% over 12 months. Yet to deserve the same rating in so-called growth industries, a stock might need to double. The official Deutsche Bank guideline simply says a stock with a strong-buy rating is "our best pick.... Expect significant out-performance against the market. The time to act to buy the stock is now."

Whatever the guidelines say and the theories posit, now is the time to clarify stock ratings. On May 16, online broker Charles Schwab (SCH ) announced it would avoid subjective analyses in favor of computer-generated statistical measures to rate 3,000 stocks with the letter grades A through F. Other than that, the only thing that has been agreed upon is that somehow, something must change. Until it does, investors will end up the poorer.

Black is a reporter for BusinessWeek Online in New York

Edited by Beth Belton

    Before it's here, it's on the Bloomberg Terminal.