June 19 (Bloomberg) -- The moment of clarity for Paul Miller came during a July 2007 conference call by the management of Countrywide Financial Corp., the biggest U.S. mortgage lender. Miller, an equity analyst at investment bank Friedman, Billings, Ramsey Group Inc., had thought until then that the crisis in the housing industry was confined to subprime loans.

On the call, Countrywide Chief Executive Officer Angelo Mozilo disclosed that delinquencies were also rising in the bank's prime home equity lines of credit. ``The leader of the mortgage industry is telling me I have a problem in my prime mortgages,'' Miller, whose firm is located in Arlington, Virginia, recalls thinking. He put a ``sell'' on Countrywide stock that day.

That call helped Miller finish first among bearish analysts, according to data compiled by Bloomberg in its ranking of the world's best stock pickers. From the time he first downgraded Countrywide to a ``hold'' in May 2007 the stock fell 76 percent.

The ranking is based on the stock recommendations of more than 3,000 analysts worldwide at 432 research firms and investment banks. The ratings stem from calls made by analysts on the share prices of 300 companies with a market value of $5 billion or more whose shares rose or fell at least 20 percent in the year ended on March 31. (See ``How We Crunched the Numbers.'')

In a separate ranking of the top 10 analysts who made bullish calls, the No. 1 spot was taken by Jack Xu of SinoPac Securities Asia Ltd. in Shanghai for his ``buy'' on shipper China Cosco Holdings Co. Four months after Xu's call, when he downgraded the company to ``hold,'' China Cosco was up 272 percent.

Morgan Stanley is No. 1

The No. 1 firm, based on the investment advice its analysts provided on the 300 companies, was New York-based Morgan Stanley. Its analysts made 94 accurate calls on the 300 stocks. Zurich- based Credit Suisse Group was No. 2, with 88 accurate calls, followed by New York-based Citigroup Inc., with 80.

The firm with the highest percentage -- as opposed to the highest number -- of accurate calls was New York-based Oppenheimer & Co., according to Bloomberg data. Almost 60 percent of its recommendations on the stocks it covers were accurate.

In a time of market turmoil, finding good stocks for investors has never been more important. ``We understand clients want trading offerings from the research department,'' says Stephen Penwell, head of U.S. equities research at Morgan Stanley. Since early 2007, the bank has linked its analysts' compensation in part to the number of good calls they make.

``We put a significantly higher rating on stock picking than we did in the past,'' Penwell says.

Tattered Image

A few good calls aren't enough to repair the tattered reputation of analysts who work for brokerage firms and investment banks. During the tumultuous 12 months ended on March 31, a minority of stock analysts accurately predicted the movement of the most volatile stocks. Of approximately 6,000 calls made by analysts on the 300 stocks during the year, fewer than 1,500 were accurate, Bloomberg data show.

Analysts did a particularly dismal job of perceiving what was going on in their own industry. The majority of the 17 analysts covering Bear Stearns Cos. had rated the stock ``sell'' or ``hold'' through 2006, when its price increased 42 percent. Most of the analysts who had had a ``buy'' on the shares failed to downgrade them before July 2007, when two Bear hedge funds heavily invested in mortgage-backed securities went bankrupt.

Bloomberg data show that New York-based Douglas Sipkin of Wachovia Securities, the investment banking unit of Wachovia Corp., made the best call on Bear Stearns for the year ended on March 31. He had a ``buy'' on the stock for 27 months before he downgraded it to a ``hold'' on May 16, 2007. And he downgraded it at a higher price -- $150 -- than any other analyst.

Bear Stearns's Fall

Sipkin says he lost faith in Bear Stearns shares when he decided the firm was too heavily invested in the housing market and not as diversified internationally as other big investment banks. ``We certainly didn't anticipate the mortgage collapse,'' Sipkin says, sitting among piles of computer printouts and bound reports in his office, which overlooks Manhattan's Park Avenue. ``But we felt that the spreading mortgage troubles would have a disproportionate impact on Bear Stearns.''

Sipkin says he had no inkling that Bear Stearns would ultimately implode. In March, the investment bank was saved from bankruptcy when, in a deal orchestrated by the Federal Reserve, JPMorgan Chase & Co. agreed to take it over for $10 a share. The 86-year-old firm disappeared into JPMorgan on May 30, at a cost of more than 12,000 jobs.

The majority of analysts who cover Merrill Lynch & Co. and Citigroup weren't astute at predicting trouble either. The banks' shares started to dive in July 2007 because, like Bear Stearns, they were deeply exposed to mortgage securities infected by subprime loans.

Getting Merrill Wrong

Sipkin, who was right on Bear Stearns, got Merrill wrong. He still had a ``buy'' on Merrill when its stock plunged 18 percent in the first three weeks of October following a writedown of $8.4 billion in assets and a third-quarter loss of $2.24 billion, the biggest quarterly loss in the bank's history.

``Merrill Lynch was the biggest and probably the most surprising writedown,'' Sipkin says. ``The absolute size of their exposure was difficult to ascertain by looking at their balance sheet in the first half.''

Most analysts were also slow to downgrade Citigroup. Two exceptions were Michael Mayo of Deutsche Bank AG, who made the best call by lowering his rating on Citigroup from ``buy'' to ``sell'' on Oct. 12, and Meredith Whitney of Oppenheimer, whose downgrade of Citigroup on Oct. 31 helped send its stock tumbling 8.1 percent the next day.

Morgan Stanley's Penwell, 46, says his firm's debt analysts saw the cracks forming in the housing market before the equity analysts did. ``Our credit guys understood the subprime crisis early, well before our equity guys,'' he says. ``It was frustrating. We underestimated the initial pain.''

Wessel Nails Ambac

Andrew Wessel, who covers mortgage finance and bond insurance at JPMorgan, saw trouble coming at bond insurer Ambac Financial Group Inc. as early as September 2006. Wessel, 29, started his coverage of the stock with a ``hold'' rating in that month. When Wessel renewed that rating on April 26, 2007, the stock sold for $92.65 and eight of the nine other analysts covering the company rated it a ``buy.''

In January, Ambac, the world's second-largest bond insurer, fired CEO Robert Genader and slashed its dividend 67 percent. A week later, it reported a $3.3 billion loss. The stock plunged to $5.75 on March 31 and $2.07 on June 18.

Wessel says his advantage came from starting his career as a credit analyst at investment firm BlackRock Inc. His job there was to dissect and analyze complex securities such as collateralized- debt obligations, which are packages of subprime loans and other debt that lost most of their value after the credit crunch hit.

Understanding CDO's

Standard & Poor's stripped Ambac of its AAA rating in June, casting doubt on the company's guarantee of $551 billion in municipal debt and CDOs.

``I actually understood the structure of the CDO payments,'' Wessel says. ``I knew a few large hits could wreck their ratings.''

In 2003, 10 Wall Street banks, under pressure from then New York Attorney General Eliot Spitzer and the U.S. Securities and Exchange Commission, agreed to pay $1.4 billion to settle charges that they misled investors with research designed to win underwriting and advisory business from corporate clients. The firms agreed to bar their analysts from working on investment banking deals, which was how the researchers generated income for their firms.

The Spitzer settlement did little to stimulate analysts' enthusiasm for making ``sell'' calls. They're still reluctant to pull the trigger on a stock. ``Sell'' recommendations represented just 5.5 percent of all calls as of mid-June, up from 2 percent in 2000, according to Bloomberg data.

Countrywide Beyond Help

Even Miller of Friedman Billings was quick to lift his July 2007 ``sell'' on Countrywide Financial to a ``hold'' when Bank of America Corp. said the next month it would buy a $2 billion stake in the struggling mortgage lender.

That turned out to be a mistake. While Countrywide stock rose in the first few days after Miller changed his rating, it plunged during the next six months. By May 5, when Miller lowered his rating to ``underperform'' again, the shares were selling for $5.36, down from $39 a year earlier. On June 18, they had fallen to $4.67, and Bank of America was poised to take over Countrywide.

``It's not easy making a `sell' call,'' says Miller, 47, a U.S. Navy veteran and former bank examiner for the Federal Reserve Bank of Philadelphia. He says analysts resist doing so out of fear of losing access to company executives. ``That's what people pay for, to get access and information,'' Miller says. Companies he's given ``sell'' ratings haven't cut him off completely. ``But calls don't get returned, and conversations are more stressed,'' he says.

Forced `Sell' March

Merrill Lynch research director Candace Browning is determined to cure her department of its allergy to ``sell'' calls. In early May, she announced she would require her 600 stock analysts to put ``underperform'' ratings on at least 20 percent of the companies they cover. On June 2, Merrill inaugurated the program by downgrading 487 stocks to ``underperform'' from ``neutral,'' including Duke Realty Corp., Lehman Brothers Holdings Inc. and PMC-Sierra Inc.

Even before Merrill took this step, it had ``sell'' ratings on 13 percent of the stocks its analysts cover, more than most Wall Street firms.

``We did a review, and about 40 percent of the stocks in the S&P decline each year,'' Browning says. ``The new system forces our analysts to rank their stocks top to bottom.''

Wall Street analysts insist they're quicker than ever to downgrade a stock. ``We've been more willing to be more aggressive and put a `sell' rating on a company than historically,'' says Deborah Weinswig, retail analyst at Citigroup.

Downgrading Macy's

She ranked No. 18 among the bearish analysts for her call on department store chain Macy's Inc. In November 2006, she downgraded Macy's to ``hold,'' warning investors the company was misfiring a year after its $11 billion acquisition of May Department Stores Co. Revenue was falling at former May stores, and the company wasn't selling enough private-label products, she said.

``There was a lot of hope for this merger, but they've struggled to get themselves back on track,'' Weinswig says. After a three-year ``buy'' rating on Macy's stock, she downgraded it to ``hold'' at $42.09 in late 2006 and kept that rating through mid- June. In the six months after Weinswig renewed her ``hold'' rating on Feb. 28, 2007, Macy's stock fell 32.8 percent.

The other 17 analysts that cover Macy's either had ``buy'' ratings on the stock going into 2007 or downgraded it at a lower price than Weinswig. On June 18 of this year, Macy's traded at $21.

A Big Pay Cut

Analysts are working harder today for less money. Senior researchers with at least 10 years of experience now earn about $750,000 a year, says Alan Johnson, managing director of Johnson Associates Inc., a New York-based compensation consultant.

That compares with multimillion-dollar salaries and bonuses in the years of the dot-com boom. In 2000, Jack Grubman, the former star telecommunications analyst at Citigroup's Salomon Smith Barney, was paid $20 million, according to his testimony before the U.S. Congress in July 2002.

Analysts struggle to attract the attention of their clients. ``Clients get frustrated by a wall of noise,'' says Dan Squires, a money manager at GSA Capital Services Ltd., a $2 billion London- based hedge fund firm, speaking of the flood of research reports that cross his desk. ``Some of the ideas are rubbish, some are good and then there's one or two gems.''

Fire Hose of Research

Christopher Yates, a senior research analyst at the Lord Abbett group of mutual funds, based in Jersey City, New Jersey, says he found 900 e-mails, half with a research note attached, when he returned to his office after two days of jury duty in May. ``It's like drinking from a fire hose,'' he says. ``To be successful, you have to do your own research.''

The investment banks are trying to win back investors like Yates with a host of services. ``If we analyze or interpret information that is generally available, like quarterly earnings, we could add, say, 10 percent value,'' says Stefano Natella, head of equity research at Credit Suisse, which ranked No. 2 overall. ``If we are able to find information that nobody else has, then we are 100 percent ahead of the competition.''

Natella says Credit Suisse analyst Scott Barry unearthed such information when he downgraded Carnival Corp., the world's largest cruise line operator, to ``neutral'' from ``outperform'' in February 2006, ahead of most analysts. Barry factored in rising fuel costs and did his own survey of travel agents and booking services, learning in the process that sales were sagging, Natella says.

Downgrading Carnival

As of June 18 of this year, Carnival stock had declined 32.5 percent following the call.

John Parks, director of research at Oppenheimer, says his 44 senior analysts, covering 725 companies, often do research that goes well beyond studying SEC 10-K filings. He cites the work done by analyst Carl McDonald to learn what was going on at WellCare Health Plans Inc.

In October, Federal Bureau of Investigation agents raided the Tampa, Florida, offices of the health insurance company in a probe of Medicare and Medicaid fraud. Its shares fell 80 percent in a week. McDonald got his hands on public documents including the search warrant and subpoenas connected to the investigation.

Based on his own research -- including a conversation with a former U.S. attorney -- he concluded the company would survive the federal criminal probe, and in November, he put a ``buy'' on the stock. In the seven months ended June 18, it had risen 40 percent. The criminal investigation is continuing.

There was some good news in the markets in the year ended on March 31. Analysts who bet on commodities and alternative energy fared well.

China Cosco's Big Gain

SinoPac's Xu says that deep research -- and a little luck -- guided his best pick. He put a ``buy'' on Shanghai-based China Cosco, the world's biggest dry-goods shipper, in June 2007. By late October, when he downgraded it to ``hold,'' the stock had shot up 272 percent.

What Xu, who covers the Chinese transportation industry, foresaw was a shortage of shipping capacity for commodities such as coal, iron ore and grain that China needed to feed its expanding economy. That would keep dry-bulk rates high and benefit the shipping companies, he predicted.

``There is limited capacity, and delivery dates for dry-bulk vessels are getting pushed back,'' Xu says from his office in Shanghai. Some companies that order new ships are waiting as long as a year for delivery.

Predicting that China Cosco's shares would rise, Xu says, was an easy call; 25 of 33 analysts who cover the stock gave it ``buy'' ratings in June. It was Xu's timing that set him apart.

Fleet of New Ships

Three months after he rated the shares ``outperform,'' China Cosco's state-controlled parent, China Ocean Shipping Group Co., offered to sell it 412 new vessels. The new ships enabled the company to take advantage of shipping rates that had doubled since the previous year, sending its shares soaring.

Solar power is what propelled Sanjay Shrestha of Lazard Ltd. to the top five of the bullish ranking. Shrestha, 34, first put a ``buy'' rating on solar module manufacturer First Solar Inc. in March 2007, when its stock was trading at $53. In July, he downgraded the stock to ``hold'' for a month, and it dropped 28 percent. He then rated it ``buy'' again. On June 18, the shares closed at $276.

Shrestha says that what makes Phoenix-based First Solar a hot stock is that its panels might make solar power competitive with oil, gas and coal within three years.

Solar vs. Oil

``The company is increasing the efficiency of its cells and cutting costs,'' he says.

Xu and Shrestha's savvy calls remain the exception -- especially when financial stocks are in question. On June 9, Lehman Brothers reported a $2.8 billion second-quarter loss, its first since it went public in 1994. That was also triple the most pessimistic estimate of 18 analysts surveyed by Bloomberg who follow Lehman.

On the day Lehman disclosed the red ink, its shares fell almost 9 percent, and nine of those analysts had a ``buy'' rating on the stock.

To contact the reporter on this story: Kambiz Foroohar in New York at kforoohar@bloomberg.net

Last Updated: June 19, 2008 00:15 EDT