(FORTUNE Magazine) - There's the slightest insinuation of warmth in the New York air; the whisper of a promise of crocuses; the unmistakable sound of the crack of the bat.

Ah, yes, the first-quarter earnings season is almost upon us. Just as April showers bring May flowers, earnings season brings earning guidance -- quite a lot of it, in fact.

More from Fortune Will Mmmhops be a hit? NBA confirms L.A. Clippers sale to ex-Microsoft CEO Steve Ballmer FBI and SEC probe into Carl Icahn and golfer Phil Mickelson FORTUNE 500 Current Issue Subscribe to Fortune

Wal-Mart: Desperately seeking ethics The world's largest retailer is trying to get out in front of critics and project a reputable image. (Read the column) More Plugged In • Forget oil. Now the Gulf is exporting money. • Letters: Energy independence is a disaster in the making • Google nudes • Stumped about Google? So are the experts. Plugged In Archive

The National Investors Relations Institute found that 71 percent of the companies it surveyed in 2004 provided earnings guidance; 61 percent did so on a quarterly basis. And yet, the evidence continues to build that the theory of earnings guidance is all wrong. It imbeds perverse incentives (to manage the numbers rather than the business); relies on dubious assumptions (that earnings should be stable); and feeds into a short-term mentality.

For the worst-case scenario of guidance gone amok, look no further than a certain courtroom in Houston, where the Enron trial is producing such bon mots as former CEO Jeff Skilling's alleged demand to "give me all the juice you can" (i.e., the means to match earnings targets).

Now, let it be said that Wall Street loves guidance, and because analysts want it, companies feel they have to give it. But the stock market existed for centuries before guidance became common practice in the early 1990s; it can certainly function without it.

It is more than a little bit telling that CEO John Thain indicated that analysts could expect no earnings guidance from the New York Stock Exchange (Research), which went public on March 8. If the engine of Wall Street disdains the practice, it can hardly be considered essential.

The justifications are weak. One is that providing guidance increases liquidity for companies that might otherwise be ignored -- and liquidity helps to reduce volatility. Recent research by McKinsey, however, suggests that for large companies earnings forecasts actually increase volatility because the fact of hitting or missing creates trading action.

And besides, what is wrong with some ups and downs? Business is, after all, inherently unpredictable.

Defenders of guidance also argue that earnings are a key part of corporate performance. Such information is important. How could knowing less be better? The thing is, earnings guidance is not knowledge; it is just an educated guess -- or rather, an aspirational one that far too many companies will fold, spindle or mutilate themselves to meet.

And we're not just talking the likes of Enron and WorldCom here. Burying expenses under the rubric of "restructuring"; timing asset sales; piling up debt; or squirreling money away in reserve funds are among the means desperate managers have used to hit their targets.

None have anything to do with running a good business.

Gillette, for example, used to go to extraordinary lengths to boost revenues late each quarter to ensure it would meet its earnings targets. In effect, it was putting things on sale that it could have sold for full price. Is that really in the interests of the company or its shareholders? Of course not, which is why one of the first things CEO Jim Kilts did on taking over in 2001 was to kill the guidance goose.

Before she waded into a flawed acquisition, Carly Fiorina made mistaken earning guesses in 2000 that lost $23 billion of Hewlett-Packard's (Research) market value in three days. Now that's volatility!

To be fair, it is true that some analysts will drop a stock if the company drops guidance. But does that really mean the death of liquidity? There was once a small Midwestern company that refused to give guidance, but it kept pumping out good returns, and is now a very big one known as Berkshire Hathaway (Research). The lesson: If you build a record of profits, the investors will come -- even to Omaha, of all places.

Finally, not giving guidance does not mean not giving information. JP Morgan Chase (Research) reveals medium-to long-term goals for each of its businesses -- return on equity, say, or default ratios. Earnings are part of the equation but not the only variable. The gurus of Gen X at Google (Research) also eschew guidance, but their recent presentation at "Analysts Day" was extensive, touching on the company's priorities, strategy, issues and finances.

Good economic performance is a process, not a number. Even the Chinese Communist Party gets this concept. The National People's Congress, which is having its annual rubber-stamping session, is about to consider the newest five-year plan. But this one, unlike the previous ten, does not feature an economic growth target.

Beijing found that if it set a number, it got it -- by hook or, more commonly, by crook. People lied to get the "right" number -- call it Enron with Chinese characteristics. This not only made it difficult to know what was really going on, but also meant that growth was pursued regardless of the consequences.

Like Warren Buffett and the gang at Google, China has come to recognize good management requires that numbers be servants, not masters. Mao, whose knowledge of economics could fit neatly on the head of a pin, said something pertinent once: "Many things may become baggage, may become encumbrances if we cling to them blindly and uncritically."

Earnings guidance is such baggage.

Plugged In is a daily column by writers of FORTUNE magazine. Today's columnist, Cait Murphy, can be reached at cmurphy@fortunemail.com.