At The Motley Fool, we poke plenty of fun at Wall Street analysts and their endless cycle of upgrades, downgrades, and "initiating coverage at neutral." While the pinstripe-and-wingtip crowd is entitled to its opinions, we've got some pretty sharp stock pickers down here on Main Street, too. (And we're not always impressed with how Wall Street does its job.)

Given that, perhaps we shouldn't be giving virtual ink to "news" of analyst upgrades and downgrades. And we wouldn't -- if that were all we were doing. Fortunately, in "This Just In," we don't simply tell you what the analysts said. We also show you whether they know what they're talking about. To help, we've enlisted Motley Fool CAPS, our tool for rating stocks and analysts alike. With CAPS, we track the long-term performance of Wall Street's best and brightest -- and its worst and sorriest, too.

When to hold 'em?
It's turning into a busy week for tech stocks, as new ratings for Apple, Research In Motion, and Oracle vie for investors' attention. The busy week got even busier this morning when ace stock picker Sanford Bernstein pulled its buy rating on IBM (NYSE: IBM).

According to Bernstein, the reason for downgrading IBM to "market perform" (but not "sell") is really quite simple: After running up nearly 24% over the past 12 months, IBM now "appears increasingly fairly valued." And when an ace performer like IBM is only fairly valued, that's an argument for holding it -- not selling it.

But why does Bernstein consider the stock fairly valued? The analyst notes that IBM has beaten consensus earnings estimates for "13 quarters in a row." It's developed a reputation for "material beats and guidance raises." 

Alas, IBM has done so primarily by growing its gross margin, which has increased every year for the past seven years. At some point, IBM is bound to max out the number of profitable pennies it squeezes from every revenue dollar. At that point, growing revenue will become the only way IBM can grow earnings -- and Bernstein simply doesn't believe IBM is up to the task. While IBM management says it sees a route to 5% annual revenue growth for at least another five years, Bernstein thinks the company could hit the wall as early as this year, and decelerate to just 3.4% revenue growth.

Is Bernstein right?

Let's go to the tape
Unusually for a firm of its caliber, it's hard to judge based solely on Bernstein's record. Bernstein simply hasn't been very active in the fields where IBM operates -- computers and IT services. We only have the analyst on record for a handful of picks in these industries:

Company

Bernstein Rating

CAPS Rating
(out of 5)

Bernstein's Picks Beating
(Lagging) S&P by

Cognizant Technology (Nasdaq: CTSH)

Outperform

***

121 points

Lexmark International (NYSE: LXK)

Outperform

**

6 points

Hewlett-Packard (NYSE: HPQ)

Outperform

***

(14 points)

But overall, the analyst seems to be holding its own, and living up to its reputation for 55% accuracy on its picks -- a record that outperforms more than 92% of the investing public. What's more, I'm not certain you really need to be an IT expert to make this call. To me, the valuation at IBM today speaks for itself.

Right now, IBM shares fetch roughly 13.3 times trailing earnings, and 12.2 times free cash flow. That's not uber-expensive for a presumed 11% grower like IBM, especially when you have a tidy 1.6% dividend to help take up the slack. On the other hand, it doesn't leave much of a margin for safety, either. On balance, I'm inclined to side with Bernstein, based on IBM's valuation alone.

Up is down, and IBM's going up
Now here's where it gets tricky. What's really strange about Bernstein's rating switcheroo is that, at the same time as it downgraded the stock, Bernstein upgraded its target price on IBM to $173 per share. In making its valuation call, Bernstein argued that at a presumed 15 P/E valuation (that's where $173 would put it), IBM would sell for "its highest level versus the technology sector in over 15 years."

And I agree -- that would be getting pricey, even if IBM can grow at 11% going forward. On the other hand, if Bernstein is right, and revenue drives IBM's growth going forward, the picture changes. Rival IT specialists like Accenture (NYSE: ACN) and Wipro (NYSE: WIT) currently max out at gross margins in the low 30s. HP can't even reach that level, cresting at a 24% gross. The best companies in the business, IBM, Cognizant, and Infosys (Nasdaq: INFY), currently gross about $0.42 per revenue dollar.

All of which tells me that Bernstein is probably right about the revenue/earnings debate -- that IBM is at or near the peak of its profitability, and that profit margins are more likely to fall than rise going forward.

Foolish takeaway
To me, this argues strongly in favor of paring your risk on IBM. Yes, I know Bernstein says the stock could still "go to $173," and you might be tempted to go along for the ride, only to drop out once IBM hits Bernstein's target price. Still -- and you know it pains me to quote Jim Cramer here, but this time, he's right -- "Bulls make money. Bears make money. Pigs get slaughtered."

If you own IBM today, you've enjoyed 24% profits over the past 12 months, and more than doubled the S&P 500's performance. Now it's time to take chips off the table. To paraphrase one other wise man's advice: "The time to count your money ... is when IBM's share price no longer offers you a good deal."