Skip to main content

So how much exactly

did

Microsoft

(MSFT) - Get Microsoft Corporation (MSFT) Report

earn in the March quarter?

Of course, it depends on whom you ask. Microsoft on Thursday said that using generally accepted accounting principles, or GAAP, it earned 49 cents a share in the third quarter, which ended in March. But the earnings number that analysts are citing to analyze the state of business at the world's largest software maker runs the gamut from 46 cents to 58 cents a share.

The Thomson Financial/First Call consensus estimate was 51 cents a share, but what appeared to be a miss on first view was later called "nearly heroic performance" by one analyst and led to upgrades Friday by Merrill Lynch and First Albany.

The confusion made Microsoft shares bounce up and then back on Thursday and Friday. Shares rose 83 cents, or 1.5%, to close at $57.20 Friday.

The confusion stemmed from one-time gains and losses unrelated to the company's ongoing operations in the third quarter. The First Call estimate included a 10-cents-per-share gain Microsoft was expected to realize from the sale of

Expedia

(EXPE) - Get Expedia Group, Inc. Report

; Microsoft ended up realizing a 15-cent gain on the transaction. And the consensus estimate number didn't include a 14-cent investment impairment charge and a 2-cents-a-share investment gain that Microsoft included in its 49-cent-a-share earnings number.

Chuck Hill, director of research at First Call, said the analysts are selectively looking at numbers to paint a rosier picture of Microsoft. "Oftentimes, it's the company making numbers look better

and trying to make the analysts go along," Hill said. But "in all fairness, the bigger problem here is the analysts rather than the company

TheStreet Recommends

Microsoft."

He said analysts have painted themselves into a corner by inconsistently treating Microsoft's investments. In the first three quarters of fiscal year 2001, he said, analysts included Microsoft's venture capital gains in their earnings number. But they then excluded a writedown on those investments that Microsoft took in the fourth quarter.

"I'm sure the argument most gave was that, well, this isn't a realized transaction like the gains were," Hill said. "But the problem is, if they turn around and sell those things down the road, the gain is going to be based on that lower cost based on the writedown." And the analysts would recognize those gains, while they had not considered the writedowns that led to them.

Fall Into the GAAP?

The simple solution would be to do what most companies do, said Hill: Exclude asset gains and losses. But analysts in the past already have been including the gains, so that's how they've painted themselves into a corner, he said. "The analysts have behaved illogically on handling these items for a couple of years now, and now it's gotten so confusing I'm not sure they know."

The analysts already had planned to include the Expedia gain in their earnings estimate, and presumably would have excluded the investment writedown, Hill said. But with the higher-than-expected gain from Expedia, that would have led to earnings of 63 cents a share -- a 43% jump from the 44 cents a share in earnings reported by Microsoft a year earlier. "There was a sticker shock problem," Hill says.

Another scenario also led to sticker shock. To draw a fair comparison with his estimate of 50 cents a share, Robertson Stephens analyst Eric Upin figured Microsoft earned 58 cents a share -- representing 32% annual growth. That number is reached by adding the 14-cent investment charge that surprised analysts, and subtracting the 5-cent gain from Expedia -- neither of which Upin or others knew about when crunching their estimates.

But there's a problem with that number, said John McPeake, a senior enterprise software analyst with Prudential Securities. It still includes some of the one-time sale of Expedia, said McPeake, who has a buy rating on Microsoft. His firm hasn't done any banking with Microsoft, but McPeake has a material position in the company's stock.

So McPeake did what most analysts did: He cited an operating EPS number by deducting the one-time 15-cent Expedia gain and adding the 14-cent one-time investment impairment charge from Microsoft's GAAP number. That comes to operating EPS of 48 cents. Analysts then recalculated the company's prior guidance of 50 cents to 51 cents a share by subtracting the expected 10-cent gain from Expedia, leading to guidance of 40 cents to 41 cents a share. Under that scenario, the company exceeded guidance by about 7 cents.

"Everything above and including operating income is what the company is in the business

of doing," said McPeake. "Everything below that could be thought of as noise."

More Confusion

By other calculations, though, the company didn't perform quite that well. UBS Warburg analyst Don Young calculated operating EPS at 46 cents a share, by further subtracting the 2-cent gain from bonds. His firm hasn't done banking business with Microsoft.

Brendan Barnicle, an analyst at Pacific Crest Securities, is turning to the old mainstay -- GAAP numbers. "It's the only way you can get consistency," said Barnicle, who has a buy rating on Microsoft. (His firm hasn't done any banking business with Microsoft.) On a GAAP basis, Microsoft earnings rose nearly 12% from a year earlier.

But Barnicle suggested Microsoft could have reduced some confusion by making a practice of reporting pro forma numbers that exclude one-time charges, as other companies do. "I don't think Microsoft did a very good job presenting it," he said. The company should have anticipated there would be confusion because its GAAP number was so close to the consensus estimate yet included special items, he said.

Jeff Brotman, an adjunct professor of law at the University of Pennsylvania who teaches accounting, discouraged the use of pro forma earnings because it creates too many different ways of reporting results. Instead, he said Microsoft could have reported net income before extraordinary items and then net income after extraordinary items, such as the Expedia sale, to reduce confusion. He noted extraordinary items must be unusual and infrequent.

First Call still hasn't figured out which number it is going to go with, Hill said. "We'll do whatever the majority of the analysts want to do, no matter how stupid it may be."