In 2000, General Electric posted its 100th consecutive quarter of growth in continuing operations. That's 25 years. Raise your hand if that sounds just a bit suspicious. Whatever business you're in, that feat just isn't possible unless your company's managing its reported earnings.

According to a 1998 survey, 78% of CFOs attending a given conference said they'd been asked to "cast financial results in a better light" without running afoul of GAAP. Half said they'd done it. Nearly half said they'd been asked to misrepresent their company's numbers, and 38% admitted they'd done so. Another survey at a different conference found that more than half of the CFOs attending had been asked to juice their numbers, and 17% had agreed to do so.

It's easy to understand why companies succumb to the incredible pressure to make it look like they've met or beaten targets or Wall Street expectations. Consistent growth is a feather in any CEO's cap, and a rising stock price often increases many executives' compensation, especially from stock options. But when companies stray from merely managing their numbers within GAAP into outright fudging them -- Enron, Sunbeam, we're looking at you here -- they can ruin themselves and their shareholders.

How can we spot suspicious earnings patterns soon enough to save ourselves? We can track how closely a company meets earnings expectations, monitor its frequency of year-over-year growth, and compare those stats to numbers from a few competitors, which should be affected similarly by changes in the business cycle. Any company that lands eerily close to earnings-per-share (EPS) expectations, and grows earnings year-over-year with unusual reliability, should raise a yellow flag and invite us to look closer.

Here's a look at what AK Steel (NYSE: AKS), the metal products supplier, has done over the past few years. I've also included a couple of other businesses playing in the same space for comparison.

Company

Reported EPS Within $0.02 of Estimates?

How Close to Estimates, on Average

How Often It Reported Growth

AK Steel Holding

7 times in last 26 quarters.

$0.09

13 times in last 22 quarters.

Carpenter Technology (NYSE: CRS)

0 times in last 24 quarters.

$0.01

10 times in last 20 quarters.

Steel Dynamics (Nasdaq: STLD)

9 times in last 26 quarters.

$0.01

13 times in last 22 quarters.

Source: www.earnings.com and author's calculation. Difference in number of quarters counted due to data source.

This set of companies doesn't even cause me to blink. Well under half the time do they hit close to estimates (and Carpenter managed to miss every time for the past six years); about half the time they reported yearly growth, and their averages to estimates don't concern me because the spread is so wide. Carpenter had misses on both sides ranging from $0.04 to $0.25, while Steel Dynamics had 12 "misses" of $0.04 or greater. No yellow flags, here, Fool.

Note that I'm not concentrating on managing estimates here -- though management does that, too. However, if a management team always seems to deliver on estimates time and time again, you should probably dig a bit deeper, to see whether its interpretation of GAAP is getting a bit too fast and loose.

Investors crave consistency. That's one reason why its string of reliable results spurred GE's stock price to rise so much in the 1980s and 1990s. But the real world isn't consistent, and Foolish investors should account for that. If a company's results seem too steady to be true, Fools should proceed with caution.

Fool analyst Jim Mueller is a beneficial owner of General Electric, but doesn't have a position in any other company mentioned. He works with the Stock Advisor newsletter service. Try any of our Foolish newsletter services free for 30 days. The Fool is all about investors writing for investors.

True to its name, The Motley Fool is made up of a motley assortment of writers and analysts, each with a unique perspective; sometimes we agree, sometimes we disagree, but we all believe in the power of learning from each other through our Foolish community.