Earnings Torpedoes
Forbes.com
By David Dreman
Ebay trades at 2,171 times trailing earnings and RealNetworks at 1,214 times. Yahoo is comparatively cheap: You can get it for a mere 1,195 times trailing earnings. Investors buying these stocks are looking beyond current results, relying instead on exponentially higher earnings estimates five to ten years out. What are the chances that such long-range expectations will come true? This question is particularly important in a market where high valuations are more the rule than the exception.
The answer is, poor.
The director of research for the Dreman Foundation, Eric Lufkin, and I studied 108,000 analysts' quarterly consensus estimates of widely followed companies from the beginning of 1973 to the end of 1998. Normally, each consensus forecast was made up of from 10 to 40 individual analysts' estimates.
The results were not encouraging. The consensus forecasts over the entire 25-year period were off from reported earnings by a startling 42% on average. (To be fair, analysts have gotten better lately at picking the right numbers: In 1998 they were off by only 36%.) Most of the analysts erred on the more optimistic side.
The forecasting should not have strained the resolution on their crystal balls. We took the forecasts published two weeks before release of actual results.
If analysts missed that much for current quarters, what happens when they project earnings well into the future? After all, market valuations in the billions for the high-flying Internet and large technology companies are predicated on the accuracy of five-year-out earnings (or at least revenue) projections.
--------------------------------------------------------------------------------
Our figures show that analysts were wrong just as often in a good economy as in a bad one.
--------------------------------------------------------------------------------
There is no complete database of five-year forecasts, but you can get a sense of how shaky all this is by looking at the odds against accuracy over shorter periods. Yes, there are some companies that get high Value Line ratings for "earnings predictability"--Interpublic, Genuine Parts and Johnson & Johnson, for example. But do you really think you could pinpoint earnings five years out for an Internet retailer? I think you have a better chance of winning the jackpot in the New York State Lottery.
Some of these earnings shocks provide cruel jolts to stock prices. IBM recently dropped 15% in a day, after it said that the upcoming quarterly report would be unpleasant.
Negative surprises, or earnings torpedoes (as they are called on Wall Street), are devastating to your capital. Compaq dropped 60% in the past ten months on earnings disappointments. Gillette went down 40%, and Coca-Cola, the bluest of blue chips, 34%.
Headlines recently celebrated Nasdaq's penetration of the 3,000 barrier. They didn't say much about the relation of Nasdaq prices to earning power. The top 100 stocks in this index trade at 87 times trailing earnings, making this group extremely vulnerable to negative earnings surprises. With increasing evidence of rising inflation, which will trigger higher interest rates, these stocks are even more vulnerable to selloffs from bad-earnings tidings.
Though the disconcerting numbers on analysts' accuracy have been published in FORBES and in my books since the early 1980s, they are widely ignored on The Street. Analysts and money managers always believe these odds apply to other forecasters, not themselves, until a torpedo hits.
Does it matter what the overall economy is doing? No. Our figures show that analysts were wrong just as often in a good economy as in a bad one.
You would only be prudent then to avoid issues trading at sky-high valuations. Buy low-P/E or other contrarian stocks instead. Although they have been temporarily trampled by momentum and growth issues, they have proved to be a particularly good defense against earnings torpedoes over time. If earnings disappoint on these stocks, they move down very little, because the poor expectations are already built into the price. When earnings surprises are positive, this group crisply outperforms the market. Certainly, nobody expects these also-rans to exceed consensus forecasts by a wide margin.
Here are a number of contrarian stocks that make sense in this volatile and expensive market. Philip Morris (25, MO), trading at 10 times trailing earnings, yields 7.5%; R.J. Reynolds (21, RJR), P/E 8 (on 1999 estimates), yields 14.7%; McKesson-HBOC (20, MCK), P/E 15 (on this year's estimates), yields 1.2%; Bank One (39, ONE), P/E 14, yields 4.3%; Humana (8, HUM), P/E 14.
David Dreman is chairman of Dreman Value Management of Red Bank, New Jersey. His latest book is Contrarian Investment Strategies: The Next Generation. |
