On Wall Street Analysts' Forecasting

Yesterday's post on Bob Doll's CNBC equivocation performance pushed this post back to today. Appearing in the 'Ahead of the Tape' column in the Wall Street Journal's Money & Investing section was a piece entitled "Analysts Botch Profit Forecasts On Home Turf."

In this piece, Journal staffer Scott Patterson observes that Wall Street's analyst corps seemed particularly blind to their own sector's weakness this year. He begins with noting that in July, analysts forecast an average third-quarter earnings gain of 9% for the sector, when, in reality, it came in at -27%. At that same time, these analysts projected the sector's fourth-quarter earnings as increasing 10%. Now, those same analysts expect a -45% earnings change for the year-ending period.

Patterson spotlights Merrill's Guy Moszkowski as retaining a buy rating on various large brokers/investment banks, including Bear Stearns, through late August.

Mike Mayo, now at Deutsche Bank, recommended selling various second-tier banks, but got Merrill wrong and maintained a buy rating through early November.

Finally, longtime commercial bank analyst and Citigroup apologist Dick Bove is regarded by Patterson as "better than most." But while he

"started getting bearish on banks and brokerages in late 2006, though he waited until July to downgrade the broader sector. He is the only analyst with a "sell" rating on Bear Stearns, Morgan Stanley and Lehman, according to Zacks Investment Research."

Bove is a frequent guest analyst on CNBC, and happens to have been a personal friend of my business partner when Bove still lived up north, near New York City. Bove has, in an odd omission by Patterson, been steadfastly bullish on Citigroup throughout the year. Back in the spring, he held forth on how Citigroup would be forging ahead with earnings growth. Even now, Bove won't throw in the towel.

If this is 'better than most,' it's a sad commentary, indeed, on sell side analysts rating their own sector's members, isn't it?

Patterson observes that Besty Graseck of Morgan Stanley broke ranks to make Citigroup her top short stock recommendation for 2008.

Of course, it's worth considering this in light of the recent year's performance of Citigroup stock. Even with its dividend added back, the stock's price is down around 40% this year, while the S&P is just barely positive.

So you have to ask yourself, can Citigroup's stock price really fall, again, in 2008, more than any other large cap?

Don't you think most of the surprise is gone now? I mean, a near-50% haircut is pretty serious. So Graseck must be looking for something like another 20% drop in the firm's share price.

At that point, I'd think we are talking a Fed-brokered acquisition by Citi, or its pieces, to avoid truly serious financial system damage.

So perhaps Graseck's call is more for spite, directed Vikram Pandit's meteoric rise, noted in this posts, here and here, from cashiered Morgan Stanley employee to Citi CEO in only two years.

In any event, Patterson's article goes a long way toward explaining why I pay absolutely no attention to analysts' calls on stock price behavior. While an analyst may know a lot about the internal mechanics of their subject companies, this article points up their long, inherent weakness in forecasting things like earnings and stock prices.

Years ago, a one-time partner in my equity portfolio strategy business sent me an article which compared analysts' earnings projections deviations from actual earnings, over time. A decade ago, they were already becoming progressively worse each year. I guess the trend has continued.

Maybe this is a fine example of 'lizard brain' behavior that is the theme of Terry Burnham's excellent book, "Mean Markets and Lizard Brains." Perhaps even institutional investment managers seek safety in the advice of highly paid analysts, even though their demonstrable track record, as a group, are so deplorably bad.