Banking analysts spend much of their year grading banks and predicting banks' earnings. As the year closes, it seems a good time to ask how the analysts themselves have done.
To discover who among the investment community's "sell-side" analysts have the best record of predicting earnings for banks, thrifts and related companies, American Banker asked Zacks Investment Research to survey the forecasts made since 1991.
Details of the study, which evaluates performance for the four quarters ended Sept. 30 and the four fiscal years ended last Dec. 31, are published today in the first annual Wall Street Sharp Shooters special edition.
The results of the American Banker survey show - for the first time - the top three analysts in earnings projection accuracy for each of the nation's 50 largest banks and 25 largest thrifts, as well as the largest mortgage-related, government-sponsored and specialty finance companies.
And the survey discloses the leading analyst overall in each of these categories - the champion Sharp Shooters of earnings prognostication.
The survey focuses on earnings forecasting because statistical measurement of success in this area is far easier than in the more subjective - some would even say political - realm of stock picking.
To be sure, earnings forecasting is only part of an analyst's role, and only one way among several of gauging their accomplishments. But it seems a good one.
Nearly all important factors that affect bank stocks are linked to earnings prospects, including interest rate expectations. Mergers, which have driven the stocks this year, are ultimately about earnings.
Rising or falling earnings expectations are a critical factor for institutional investors in screening banks and other stocks for either purchase or sale.
Among analysts covering traditional money-center banks - the big banks in New York City and Chicago - the winner was Norman Jaffe of Fox-Pitt, Kelton Inc., New York.
For regional banks, including superregional institutions, the leading forecaster was Sandra J. Flannigan of Merrill Lynch & Co., New York.
Ms. Flannigan edged out Chris Kotowski of Oppenheimer & Co., New York. Mr. Kotowski achieved the same level of forecasting accuracy as the Merrill Lynch analyst but follows fewer companies.
Among thrifts, the closest earnings estimates were made by Steven R. Schroll of Piper Jaffray Inc., Minneapolis.
For mortgage companies and the government-sponsored secondary market enterprises, the most accurate projections were made by Eric I. Hemel of Morgan Stanley & Co., New York.
Among credit card issuers and general finance companies, Michael R. Hughes of Merrill Lynch was the best. Indeed, he was the best by the widest margin of any of the category winners.
Each of the analysts is profiled in today's special edition.
The Sharp Shooter survey is designed to assess the accuracy of the analysts' earnings forecasts on both a short-term and long-term basis.
Quarterly projections for the four quarters ended Sept. 30 and annual forecasts for the years 1991-1994 were reviewed. For determining overall winners, results of the two studies were equally weighted.
Overall, several major factors seem apparent from the survey.
Over the last year, analysts with more bullish scenarios for the banking industry appeared to have the edge in forecasting accuracy, reflecting the stellar year the banks enjoyed in 1995. In a benign period of flat or declining interest rates and falling government deposit insurance premims, banking companies posted few earnings surprises except positive ones.
Recent concerns about credit quality that have jarred that happy mood were registered only in the third quarter. If concerns voiced by some analysts about credit and a few other issues are well-founded, negative surprises could begin surfacing more often and make those analysts' forecasts more accurate.
Another major finding apparent in the survey is that analysts need not be in New York or work at a large securities firm to be accurate earnings forecasters.
While analysts at securities industry giant Merrill Lynch took half the winners' spots overall, many bank and financial analysts from regional securities firms scored very well.
Over the last year, for instance, Richard X. Bove of Raymond James & Associates, St. Petersburg, Fla., was the best earnings forecaster among 19 analysts who track First Union Corp. and Michael Durante of McDonald & Co., Cleveland, was the most accurate of 15 analysts following Mellon Bank Corp.
Over the longer haul, Joseph A. Stieven of Stifel, Nicolaus & Co., St. Louis, was tops among 17 followers of Banc One Corp., while John Mason, who works in Atlanta for Interstate/Johnson Lane was the truest prophet regarding BankAmerica Corp.
Mr. Jaffe, the best money-center bank forecaster, works for a boutique firm specializing in banking, insurance and related financial stocks. He is also a regional bank analyst - this year he was first among 15 analysts in quarterly earnings forecasts for U.S. Bancorp.
In an interview this week, Mr. Jaffe offered a perspective on earnings forecasts.
"Earnings estimates are a very important guide in the shorter term for investors. They are very good up-front information for the ivnestor to have and give you a good idea of performance," he said.
"There is a role that earnings estimates play in the price action of bank stocks, but as you move further out in time, it becomes more art than science," Mr. Jaffe added.
He naturally cautioned that "however good the earning estimate for a bank may be, it certainly doesn't mean you have to own the stock."