By DANIEL ALTMAN

Published: July 21, 2002

IN the stock listings in many newspapers, only one calculated statistic appears next to the trading activity for every issue: the price-to-earnings ratio. But lately, an avalanche of earnings restatements has made the denominator of the ratio, earnings, a flimsy indicator of real money-making at a growing number of the country’s biggest businesses — tarnishing the ratio, too.

Little wonder that some mutual fund managers are thinking twice about using P/E ratios even as a first cut for picking stocks. But what should they — and individual investors — use instead?

There are no easy answers. »The valuation process is an art as much as it’s a science,» said Thomas J. Dillman, co-chief of global research at Deutsche Asset Management. He and his colleagues have spent more time lately questioning the assumptions of their formulas.

»There’s a lot more conversation about how you really got to the numbers that you’re using to determine whether a stock is overvalued or undervalued,» he said.

The P/E ratio, typically calculated using the latest 12 months’ earnings but often using year-ahead projections, was once the undisputed champion of stock eyeballing. As Robert J. Shiller, the contrarian Yale economist, wrote in 1996, »The simplest and most widely used ratio used to predict the market is the price-earnings ratio.»

Many stock pickers used it as late as the mid-1990’s as the first measure of a share’s prospects. High P/E ratio? That meant either a growth stock, worth the risk for participation in a bright future — or maybe one that was just overvalued. Low P/E ratio? That meant a value stock — possibly good for steady dividend income.

In the Internet boom, that logic went out the window. Companies’ prospects were measured in page views, clicks and how fast the businesses were burning capital. After the start-ups unwound in 2000 and 2001, investors may have thought it safe to trust the ratio again. No such luck. Those reported earnings could vanish with the wave of an accounting whistle-blower’s hand.

SOME stock pickers still stand by the P/E ratio as a useful tool. »The first issue from my perspective is generally valuation, and the major valuation measure from my perspective is price-to-earnings ratios,» said Douglas D. Ramos, a senior equity portfolio manager at Dreyfus.

But even Mr. Ramos does not always use P/E ratios. Companies with few or no earnings are not rejected out of hand but are subjected to other metrics.

There are alternatives. »We’re using cash flow available for dividend payments,» said Marilyn G. Fedak, the chief investment officer for American value stocks at Sanford C. Bernstein. She said her company had never used P/E ratios. Looking at cash flow, she said, »helps you get behind all the accounting issues that can cloud how much you understand, as an outside investor, of what a company is really doing.»

Mr. Dillman’s company prefers to gauge stocks’ prospects with a statistic called cash flow return on investment, a measure that compares a company’s cash receipts to its cost of capital and the return on competing investments. But can ordinary investors make sense of such a mouthful? »Some of the caveats and assumptions that underlie all of these things are pretty arcane,» he allowed. »If we started to put cash flow return on investment numbers instead of P/E’s, no one would know what they meant. I can’t think of anything that would be necessarily better in terms of what belongs in the newspaper.»

Accounting shenanigans can render substitutes for P/E ratios based on cash flow just as useless, analysts warn. »The problem with cash flow,» said Burton G. Malkiel, a professor of financial economics at Princeton, is that it »is subject to exactly the same kinds of problems that the ‘E’ is subject to.»

Analysts often use earnings measured before a company pays taxes and interest on debt, or deducts depreciation and amortization, as a proxy for cash flow. But this measure, called Ebitda, also ran into problems recently, said Mr. Ramos at Dreyfus. »WorldCom has sort of showed you that you can’t necessarily trust cash flow either, as defined by Ebitda,» he said.

Even an apparently transparent measure of a company’s potential, like the price-to-sales ratio, can be misleading, according to Professor Malkiel.

»Think of the telephone companies that would swap their fiber optic networks,» he said. »They’ve got all this excess fiber optic capacity in the ground. At an inflated price, I sell you some fiber optic capacity; you sell me some fiber optic capacity. They tell you there’s a business purpose for this, but then the sales are overstated.»

A certain amount of analysis of a company’s books will always be based on trust, Ms. Fedak at Sanford C. Bernstein observed. »Unless you’re going to sit there and you’re going to count — looking through all the invoices and receipts — you can’t be 100 percent sure,» she said.

Mr. Ramos agreed that putting companies under multiple microscopes would help. »You look at the balance statement, you look at the footnotes, you do a lot more fundamental research, and you question things more in this market environment,» he said, »although you should have been doing that two years ago.»

When companies can be trusted to disclose their true earnings, Professor Malkiel said, one can again look at P/E ratios. »I don’t think there’s anything wrong with the valuation metrics that we use, including price-to-earnings multiples,» he said.

Even if all companies told the truth, there would still be some wiggle room. Many economists recommend using forecasts, rather than earnings from the past year, to calculate P/E ratios. A share’s price represents the value of profits yet to be earned, the thinking goes, so the appropriate ratio would use the company’s best guess about those profits. But that guess is inevitably subjective.

»A lot of what goes into that P/E is a lot of variables based on individuals’ and analysts’ expectations of the future,» Mr. Dillman said.

Those views can change with the market, Mr. Ramos added. »What you’re dealing with is investors’ time horizons,» he said. »They expand in bull markets and they contract in bear markets. The fact of the matter is that the reality is somewhere in between.»

The cautious mood on Wall Street is likely to affect those expectations and horizons. Even companies that have not had to restate earnings from previous years may rein in their forecasts. That could mean that investors will have to become accustomed to higher price-to-earnings ratios.

Indeed, P/E ratios may have been held down artificially for a long time. Ms. Fedak said she believes that earnings have been inflated for years before the recent periods for which companies have been revising results. »There have been all kinds of ways, over time, that companies have used for creative accounting,» she said, adding that she expects P/E ratios to creep upward.

This trend is likely to be accelerated by corporate decisions, like those by Coca-Cola and by a handful of other companies last week, to count the cost of options as expenses, which reduces earnings.

Trying to tell whether a share — or the whole market — is overvalued by comparing its past and present P/E ratios might make little sense. If investors’ expectations take time to catch up to more cautious earnings reports and higher P/E ratios, the market could be headed for more trouble.

»I just don’t believe all these earnings,» Professor Shiller said. »If you don’t believe them, then P/E ratios are really high. Investors are not going to tolerate that.»

Drawing (Phil Foster) Chart: »On Second Thought» How much annual earnings were reduced because of restatements by all companies publicly traded in the United States. Some announcements involve more than one statement year. Graph tracks earnings from 1997 through preliminary 2001. (Source: Min Wu, Stern School of Business, New York University)