In the early 1990s, when the chief executive of a Fortune 50 company came by the Wall Street Journal offices in New York for lunch with a dozen of us, he complained bitterly that articles written about his company’s earnings were always judged against securities analysts’ expectations. "Who died and made the analysts king?" he asked.

Being a smart aleck at heart, I assured him that if he would supply us with his internal projections we would be happy to use those as the basis for comparison and forget about Wall Street. He didn’t find that funny.

That conversation would be almost impossible today—and not because I left the Journal a few years ago; there are loads of wiseguys still there. The conversation wouldn’t happen because CEOs have become masterful at using Wall Street expectations to their advantage.

Some 80% of articles on earnings that mention expectations say the company met or exceeded them. Corporate America now resembles Lake Wobegon, the fictional Garrison Keillor town where all the children are above average.

The change is no accident. Securities analysts have always been in the business of pushing stocks and generating commissions. More recently, securities analysts have seen their bonuses come to depend on how well they generate investment-banking business, so they are even more inclined to be nice to the companies they follow. CEOs, being a bright lot, have taken full advantage and now manage expectations to their liking.

Did anybody really believe Microsoft when, after reporting blowout earnings quarter after quarter, executives said analysts should moderate their earnings expectations? Of course not. But analysts were happy to play along, so the software concern could keep providing pleasant surprises to investors and the stock could maintain a nice, steady climb.

Some CEOs also have figured out ways to push the envelope on how they calculate profits, drawing from the techniques that dot-coms have used to make earnings look as rosy as possible. For instance, online services provider America Online used to account for its marketing costs as though they were spread out over several years, so that it could report profits, rather than treat the costs as immediate expenses and report large losses. The Securities and Exchange Commission eventually fined AOL and forced it to change its practices, but only after AOL had built a stock-market following and had become a sustainable business. Companies, including software concern MicroStrategy, sometimes book the full revenue from a contract as soon as it is signed, even though the revenue will be generated over several years. Others, including chip maker Intel, have lumped gains from investments in other companies in with the rest of their earnings, rather than treating them as one-time events. Intel argues that it will have a steady stream of such gains, but, especially with the fervor disappearing from Internet stocks, I think not.

While the evidence is sketchy and anecdotal, I am convinced that more companies are stretching the bounds of propriety than in the past. With heightened expectations built into stock prices these days, and with the certainty that a disappointment will send the stock plummeting, there is ample reason even for long-established businesses to be aggressive about reporting earnings.

The irony is that the relentless focus on profitability can hurt not just the investor but also the company. At least in recent years, companies have needed to try to innovate in areas such as e-commerce even without a clear timetable for a return on the investment. But many companies have been timid about investing real money, unless they can find an accounting trick that leaves their earnings story intact.

The chief executive of Bertelsmann, the German media conglomerate, reportedly told business heads in the early days of the Internet that they were free to test e-commerce ventures—as long as they turned profitable within a year. Right. Not surprisingly, Bertelsmann has done nothing of note online.