I see dead people -- and they're going to beat their numbers.
After a certain summer
hit made whispering all the rage, maybe it should come as no surprise that even august
First Call/Thomson Financial
has decided to get into the whisper-numbers business as another earnings season looms.
Whisper numbers, the fly-by-night earnings estimates that make the rounds before companies report results, always generate a bit of excitement. Ostensibly, they're what analysts
a company might hit -- a "between you and me" that gets intoned into the ears of favored clients. Passed around by trading desks -- and, more recently, Web sites with various "scientific" means of gathering inflated estimates -- whisper numbers can help give a company's stock a nice goose before earnings come out. And if a company beats that higher number, it's on the road to Dell-ville.
The problem comes when a company can't live up to its whisper (as
itself discovered in recent quarters). In some cases, the market gets so conditioned to the idea that a company will beat its whispered forecasts that it starts inventing its own estimates. Then, when results come in worse than expected, pain follows.
Though a company doesn't miss official analyst estimates, it still gets treated as a pariah. Now, you may weep for the poor, wronged outfit, but in a lot of ways, the company itself is to blame. In offering guidance to analysts, it lowballed and set itself up for a positive surprise.
Lowballing is "a tempting thing to do because, for the first three or four quarters, you look like a hero," says Chuck Hill, director of research at First Call -- the leading tracker of analyst estimates.
Yet though Hill thinks whisper numbers are ultimately damaging, First Call still has decided to begin publishing its own version of the things. In its new HISPER estimates (an unfortunate acronym that stands for "HIStorical SurPrise-based EaRnings"), First Call has come up with a list of companies that have historically beaten forecasts and figured out by how much they'll beat the current quarter's numbers based on that history. So while the consensus estimate is for
to earn 33 cents a share in its first quarter, the HISPER (ugh, there it is again) calls for it to earn 37 cents. And so on.
Hill's contention is that most whisper numbers don't come from analysts or from companies' investor-relations departments, but from people who are doing the exact same thing First Call is doing here -- looking at past performance, pulling out a calculator and trying to make a pretty good guess. "Once the pattern settles in, it becomes a bogey," says Hill, "whether we put out the number or not."
This is obviously a matter of conjecture. Microsoft, for example, is well known for discouraging whisper numbers. If management finds out an analyst is spreading them, that analyst doesn't get to talk to management. Go ahead. Call a trading desk on the day Microsoft reports and ask for the whisper number.
Better yet, save your quarter.
And this story has the potential to get even worse. First Call's new whisper numbers may actually
companies to continue to lowball. Since First Call is the most widely cited source for analysts' estimates, the company enjoys a status that other sources of whisper numbers do not. Miss the First Call whisper, and you may find yourself doing some real yelling.
"If I were a company, I would ignore it," says
chief strategist Jeffrey Applegate. "First Call collects what the analyst community is forecasting. First Call is not a forecaster."
Yet if the market is going to start trading off these numbers, and if the financial press starts using them as a guide in earnings season's pregame coverage, ignoring these whispers in the dark may be hard for a company to do.