Return to main story


New methods show why the bets are so risky

I have some Internet stocks...I'd sure like to buy more. But I'll think to myself, these stocks are ridiculously high, they CAN'T go higher. BUT THEY DO...I can't bear missing out on the excitement. But I also can't connect reality to what I see in the stock valuations. -- A recent E-mail at the Internet Stock Report at

Maybe you're not trading Internet stocks, but if you're watching them fly, you know what that E-mailer is talking about. Trying to get your arms around the value of an Internet stock is like trying to hug the air. By time-tested valuation methods, such as price-to-earnings or price-to-sales ratios, they trade at ethereal levels, leading most Wall Streeters to knock them, only to see them go higher and higher.

Are we missing something? Or is Internet investing a mania fated to come to a bad end? That's the $64 million--or $64 billion--question of the day. And to answer that, analysts and portfolio managers are coming up with new ways of valuing Net stocks. Some border on the ridiculous, others are serious attempts to get a handle on the value of companies operating in an industry that is revolutionizing the economy. No method is the panacea, but some will give you insight into the bet you are making when you buy these stocks.

It's not easy to develop a valuation model that explains why a Net stock that trades at 300 times next year's earnings is a steal. Certainly, there's evidence to suggest that the people trading these stocks are not the type to dissect financial statements (page 122). ''I bought eBay (at $125) because I thought it would go up fast,'' says Rod Puckett, 27, a contractor in Rockford, Ill. ''These stocks are like McDonald's when it first came out, I hope.''

Still, in the quest for the Holy Grail of Internet valuation, analysts are slicing, dicing, and torturing numbers until they can be molded into what might pass for a rationale to back up a table-pounding investment recommendation.

Perhaps you can't be persuaded to buy Yahoo! for 320 times projected 1999 earnings. Is there a way to make the stock more compelling? Shaun G. Andrikopoulos, who follows Net stocks for investment bank BT Alex. Brown Inc., came up with Theoretical Earnings Multiple Analysis, or TEMA. By that measure, which projects future earnings by estimating revenue growth and the operating margins that the company would hope to achieve when it has matured, Andrikopoulos says the p-e on year 2000 earnings is just 192. But factor in a revenue growth that's 55% higher--and he thinks Yahoo!, with a history of higher-than-expected sales growth, deserves that--and the TEMA p-e drops to 124. He's recommending the stock.

The analyst claims that his model gives a ''feel for the realistic theoretical earnings power of Yahoo!'' Realistic theoretical earnings? That might fly in an investment bank, but in English it's oxymoronic.

VISITOR COUNTS. Net stocks move so fast these days that the recommendations become stale in no time. On Oct. 29, analyst Steven R. Horen of NationsBanc Montgomery Securities Inc. recommended Inc. at 117, about 8.5 times 1999 sales, and set a target of 150 in 12 months. The stock hit 150 in three weeks. Horen still rates a buy, as do 14 other analysts. ''I'm not in the business of changing my recommendations every three days,'' he says.

The problem with the valuation methods that result in earnings forecasts is that they require estimates of revenue growth, profit margins, and other important inputs. That's especially problematic when you have a company with little history and an inchoate business plan. Even thoughtful estimates are still guesses, and a bad one will throw a valuation off-kilter--making the stock seem either too cheap or too dear--so an an ideal valuation method would be one that minimized the guesswork.

That's why some followers of the Net stocks, such as Steve Harmon, the senior investment analyst at, also look at figures that don't have to be guessed, such as the number of individuals who visit a Web site. That data is reported by Media Metrix Inc., a Web research firm. The numbers show that 25.2 million people used Yahoo! at least once in October. By dividing the number of Yahoo! visitors into the market cap of $19.7 billion, you get a cost of $782 per visitor. Compare that to Lycos Inc., which works out to $135 per head, and Excite Inc. at $165. ''When you look at these companies' user values, you appreciate the disparities,'' says Harmon, ''and that leads to good questions.'' Is a Yahoo! visitor more valuable than one at Excite? Perhaps. If Yahoo! can continue to outdraw the others, it deserves a premium. But is it worth nearly five times as much?

The problem with such relative value measures is that while they may help identify undervalued and overvalued companies in similar businesses, they can't address the most important question: whether the whole industry is overvalued.

A different approach to valuation turns the whole issue on its head and is on the frontier of investment analysis. Instead of forecasting all the inputs to come up with a number for what the stock should be worth, this method starts with the stock price and works backwards to answer the question: What kind of growth does this company have to deliver to justify this price? This method does not avoid making some assumptions, but it's a way to perform a reality check.

Look at At the recent price of 214, the market is implying that Amazon's revenues will increase 59.6% a year over the next 10 years (table). That's the conclusion of veteran securities analyst Charles R. Wolf of Warburg Dillon Read.

To come to that conclusion, Wolf uses a valuation method built on Economic Value Added (EVA), a concept developed by Stern Stewart & Co., a management consulting firm. The idea behind EVA is that in the long run, it's not accounting profits--taking in one more dollar than you put out--but economic profits that matter. And simply put, a company earns an economic profit only if it has earned more than its cost of capital, which is not found on an income statement. EVA was originally designed as a management tool, not for investing. But some, including Steve O'Byrne, a former Stern Stewart consultant who now has his own firm, and Wolf have adapted EVA principles for investment analyses.

COMMON SENSE COUNTS. The idea in EVA analysis is that the market value--the stock price times number of shares--has two components. One, the Current Operations Value (COV), is a measure of the worth of the company as it now operates. The second, Future Growth Value (FGV), measures the company's expected growth. Once you determine the COV--and that's the easier of the two--you can figure out the implied future growth value. And once you know that, you can determine the implied revenue growth rate. Then you can make a judgment as to whether that growth rate is achievable.

Critical to the analysis is the cost of capital. For, Wolf estimates a cost-of-capital charge of 15%, a figure derived from such variables as the risk-free rate of return, the extra return that equities historically return over bonds, and Amazon's ''beta,'' or price volatility as compared with the stock market.

Can achieve a nearly 60% average annual revenue growth rate over 10 years? That's where investors must turn to industry fundamentals and old-fashioned common sense. For instance, using Wolf's calculations, Amazon should reach sales of $63 billion in 10 years. Is that realistic?

Not by a long shot. U.S. retail book sales in 1997 were $11.8 billion, and they're not expected to be much higher in 1998. Even if the book market expanded at 3% a year, it would be only around $16 billion 10 years out. True, Amazon is selling recorded music, but that market is no larger than books, with growth prospects no better. ''Amazon has to sell a lot more than books, CDs, and videos if it's ever going to reach the revenue growth implied in the price,'' says Wolf. True, that's in the company's plans. But as it changes from a bookstore to a mass marketer, it will run up against competitors. Says Wolf: ''Barriers to entry are low, and others can easily underprice them.''

The Internet is probably the most sweeping and potentially powerful medium to come along since television. There's no question that enormous growth is there, but how much should investors pay for that potential? Whether you're weighing revenue projections for Amazon or any other Net stock, it might be useful to note that at Microsoft Corp., perhaps the most successful company in recent history, revenue growth averaged 43% a year since it went public in 1986. Keep that in mind when you're tempted to buy a Net stock that reQuires a 60% growth rate to justify its price.

To read a letter to the editor about this story, click here. By Jeffrey M. Laderman in New York, with Geoffrey Smith in Boston


COVER IMAGE: The Wild World of E-Commerce

TABLE: Amazon's Growing Universe

TABLE: Amazon's Amazing Valuation

TABLE: Stacking Up the Amazon Advantage



RESUME: Jeffrey P. Bezos


TABLE: Valuing an Internet Stock


TABLE: Small Investors Drive the Net Stocks


Return to main story


Updated Dec. 3, 1998 by bwwebmaster
Copyright 1998, Bloomberg L.P.
Terms of Use