(MONEY Magazine) – Can we talk? This Internet thing has gotten crazy. Unknown little companies go public and bam!, their share prices shoot up like starbursts. Whoever heard of eBay or Ticketmaster Online-Citysearch a few months ago? Now they have market caps as big as K Mart's and B.F. Goodrich's, respectively. And they've made some people--some very lucky people--very, very rich. It's enough to make better- known Internet companies like America Online, Yahoo! and Amazon.com seem like solid blue chips. But they're not. Even these stocks have been rising and falling more in a single day than most stocks do in a year. And like their newer IPO brethren, they've produced gains that are simply astonishing. Had you been wise or daring enough to purchase 100 shares each of AOL, Yahoo! and Amazon at the beginning of 1998--at a cost, then, of roughly $21,500--by the end of November the stake would have mushroomed to $111,800. That's a gain of 419%, or some 400% more than Standard & Poor's 500-stock index.

What exactly is going on? The irrational exuberance that Federal Reserve chairman Alan Greenspan questioned back at Dow 6400 is nothing compared with the Internet exuberance we're seeing now. Net start-ups with no history of earnings go public and their twentysomething founders watch investors bid their stocks up more than 600% (see "Inside an Internet IPO" on page 101). Companies that announce ventures having anything at all to do with the Internet or that simply have a .com in their names are almost guaranteed a big pop in price.

Trading is so frenzied that many online brokerages are restricting trading of certain Net stocks. Some are also requiring investors who buy Internet stocks with borrowed funds to keep more collateral on hand to cover their loans. For months, even before its merger with Netscape, America Online has been one of the most heavily traded stocks on the New York Stock Exchange.

The action has been so fast and furious that even devout disciples of the Internet are stunned. In the chat room of one financial Website--the cyberspace equivalent of an office watercooler discussion--a group of investors who follow Amazon (ticker symbol: AMZN) exchanged these reactions to the stock's remarkable trajectory:

"AMZNing," wrote one.

"Amazombies," scoffed another.

"Kick back and smell the tulips," posted a third, referring to the notorious speculative frenzy of 1634, when the price of tulip bulbs went sky-high among Dutch speculators--and then crashed.

The story driving up Net stocks is simple and compelling: The Internet is changing the way we shop, communicate and do business, and those companies that lead the transformation may well dominate the Web marketplace of the future. But there is also little doubt that, on the investment front, we are deep into a stock mania of historic proportions. "The frenetic buying in the Internet sector makes the tulip investors look like value players," says analyst Rick Berry of J.P. Turner & Co. "It's probably the most absurd thing I've seen in my life."

WHAT HISTORY TEACHES

Markets have been roiled by many a mania, but there has never been anything like this. Perhaps the closest parallel in modern times is the biotechnology boom of the late 1980s and early '90s, when stocks such as Genentech tore up and down the charts. In less than 18 months, Genentech's stock jumped 375% on the promise of a clot-busting heart-attack drug. The frenzy was so wild that, like today, brokerages raised their requirements for investors buying on margin. And sure enough, Genentech's bubble burst, falling about 75% from its peak.

But history also teaches a different lesson: Sometimes, when a technological revolution comes along, if you wait to buy until the outcome is secure and stock prices seem reasonable, you miss the big payday. Consider the legacy of software behemoth Microsoft. On its first day of public trading in 1986, Microsoft's stock jumped from $21 to $28 for a 33% gain. At the time, this was referred to in the press as "manic enthusiasm," and from that day to this the stock has always looked expensive compared with others, at least on a current earnings basis. "There was never a 'right time' to buy Microsoft," says Credit Suisse First Boston senior analyst Bill Burnham. But if you waited, you lost.

On the other hand, as Burnham notes, Microsoft may have seemed expensive through the years, but it never got near the valuation levels of Internet stocks today. That 33% first-day rise pales in comparison with the Internet companies stampeding to market now. Microsoft's price/earnings multiple back then, while considered pricey because it was above the market average, was still only 16. No one even dreamed of the outlandish multiples we see today--and that's for the few Net players that actually have earnings.

Microsoft is also an example of how, if past technology cycles are any guide, the hot pioneers aren't always the big winners. By the time Microsoft went public, the PC industry was well established. And several years earlier, the sector had been through a binge and bust of its own. Anybody remember Commodore? Atari? Tandy? After turning red hot, those stocks began to fall in mid-1983 and fell for the next year or so. In February 1984, a group of 24 PC stocks were, on average, 50% below their 52-week highs.

"When you have a new industry or sector," says Robert Farrell, senior investment adviser at Merrill Lynch, "they get their wildest sponsorship and speculation at the very beginning when nobody knows what the earnings will be and you can't place a valuation on them." Invariably, he says, that first stage goes to an excess that winds up with a correction.

The lessons, then, of manias gone by:

--Investors can pay too much for even the best stock in a hot industry. Consider biotech darling Amgen. Even as the company posted 27% annual earnings growth from 1991 to 1997, the stock fell 52% from year-end 1991 to mid-1993 and didn't really surpass its mid-1991 level until 1995.

--Industry pioneers don't always survive. Witness Commodore, Atari.

--Be suspicious of copycat IPOs. In the words of Edward M. Kerschner, chairman of PaineWebber's investment policy committee, "Beware of mediocre companies that come to market simply because there is a demand for them."

--No sure thing is as sure as it seems. "High-tech booms," says Kerschner, "occur when cyclical trends in the stock market and economy favor an industry, but these trends can change rapidly." That's what happened to PC stocks in the '80s when capital spending weakened.

--But you can't win if you don't play. Speculating on Atari proved costly. Speculating on Microsoft, however...

THE NEXT WAL-MART?

If there is one stock that captures the thrill, insanity and potential of Internet mania, it is Amazon.com, the online bookseller that is expanding into music and video. Since going public in May 1997, Amazon has racked up a 2,283% price gain--even though it has never earned a cent.

Amazon is the brainchild of Jeff Bezos, who at age 30, while working for a New York hedge fund, was struck by a report on the phenomenal growth expected of the Internet. A few months later Bezos, his wife and their dog headed west in a Chevy Blazer, and while his wife drove, Bezos wrote a business plan on his laptop. Ultimately, the company would feature no actual stores, impressive inventory efficiency and this remarkable concept: Customers would pay Amazon well before Amazon paid its wholesalers. Four years later, Amazon is the fastest-growing Internet brand, and Bezos' stake in the company is worth about $4 billion (at least on paper). The company has been heralded by some as a potential "Wal-Mart of the Internet."

And that's undoubtedly part of what investors find appealing about the company. On Oct. 28, Amazon reported an operating loss of $21 million, or 49[cents] per share. Yet investors sent the stock up 8.1% the next day, from $117 to $126.50. Seventeen trading days later, the stock was at $218--up 86% since the earnings report--and the company had announced a three-for-one split.

Why such optimism? The earnings report (or, we should say, nonearnings report) contained some dramatic figures: Revenue was up 306% from the previous year. Cumulative customer accounts grew to 4.5 million, up a record 1.2 million from the previous quarter and up 377% from the previous year. And in its first full quarter selling music, Amazon outsold all the established online music sellers--strong evidence that the Amazon brand has meaning.

Building that brand has required huge marketing expenditures, which helps explain why Amazon remains unprofitable. An e-business consulting firm, IceGroup, of Wakefield, Mass., caused a stir in August when it applied what it calls an "everyman's point of view" analysis to Amazon's financials. IceGroup calculated that it costs the company $55.91 to process its average order, but Amazon receives just $48.76 for that order--"essentially creating a loss of $7.15 for each order processed," says Jane Waller, IceGroup's president. Another analysis, done by analyst Henry Blodget of CIBC Oppenheimer, pegs the operating loss at $4.50 per order. (An Amazon spokewoman responds: "We don't disclose those figures.... That's pure guesswork.")

So you've got a brand-name business growing like gangbusters on the hottest economic frontier--yet the more it grows, the more money it loses. What's an investor to do? "As of today, I'd be more of a seller than a buyer of these stocks," says Keith Benjamin, BancBoston Robertson Stephens' Internet analyst. "They've been very volatile and are at all-time highs. I'd wait for a correction." But what's a correction? A $20-a-share fall? A $40 fall? In industries with a history, coming up with a buy price is easier, because valuations are more predictable. With Net stocks, nothing is easy.

THE VALUATION DANCE

To get a chuckle (or a groan) out of an analyst, ask for the best way to value Internet stocks. Because so few of these companies make money, the traditional method of weighing the stock's share price against the company's earnings--the good old price to earnings ratio--doesn't work. In a recent report, Shaun Andrikopoulos of BT Alex. Brown notes that the methodologies he's seen are "mind boggling" and "creative in their artistry." He's found analysts attempting to put a value on Internet stocks by comparing a company's market cap with its number of sales reps, its total employees, even the amount of electricity it consumes.

One factor that makes analyzing Net stocks so difficult is that an analysis that makes sense for one may not work for another. That's because some Net companies are retailers (Amazon), some are essentially media companies (Yahoo!) and some are software makers (Netscape) that started giving away software for free and tried to become media companies but are now merging with a very big Internet service provider (AOL) that is really a media company. Got that?

Still, analysts try to quantify the madness. Among the tools:

--Average cost to acquire a customer: For companies like AOL that have regular paying customers, this figure can help measure progress against competitors. But for Internet companies that rely solely on advertising or that are transaction-based retail businesses, this makes less sense. It also doesn't help compare an Internet stock with a non-Internet stock.

--Price-to-revenue ratio: If a company doesn't yet have earnings, analysts often focus on revenue growth, which at least is a hard number. CIBC's Blodget pegs Amazon at a price-to-revenue multiple of about 17. (Wal-Mart's, by comparison, is 1.4). Boldget warns, however, that price to revenue is meaningful only if you figure in what profits those revenues are likely to produce. The razor-thin margins in Amazon's current businesses, selling books and music CDs, mean that at best, "that revenue will give you a 10% operating profit," he says.

--Discounted future stock price: This method uses projections about how the number of customers, order size and order frequency will drive revenues. In October, Blodget concluded that Amazon could have $10 billion in annual revenue within five years, with a 12% operating margin. The company would then generate earnings per share of $10. After discounting those earnings back to the present year at a 15% discount rate and using a projected P/E multiple of 40, Blodget can justify a price of more than $300 for Amazon. A more conservative forecast of just $2.75 in earnings per share in 2003 produces a target valuation of just $75. "What this shows you is that valuations are very arbitrary," says Blodget. "You're really just buying a vision of the future."

HOW TO PLAY THE NET

In many ways, of course, these valuation methods are merely rationalizations, attempts to explain why today's Net prices aren't out of line. But they are out of line. The key for most investors is deciding whether they want to invest or speculate. If you're okay with the risks, go ahead and enter the virtual casino. Take your chances with a new IPO (see "What's Left in the IPO Pipeline?" on page 97). Or play it more cautiously by buying stock in a profit-making market leader such as Yahoo! or AOL. Or you can try to hedge your bets by buying a sector fund (see "Internet Funds: Looking for a Safety Net," on page 98). But with all those choices, understand the risk you're taking--and don't venture anything you can't lose. As with other manias, this bubble may well burst, and if it does, things will get ugly.

So does that mean a prudent investor should sit back and watch so many others getting rich on this promising new industry? Not necessarily. One other option is to go in through the side door--to latch on to the already profitable and established franchises making money from the Net. One top choice is MCIWorldCom, which we recommend in "12 Best Investments for 1999," beginning on page 86. You can also look at Cisco Systems, Sun Microsystems, even Microsoft or Dell (see page 38), all of which should benefit from the Internet revolution. These stocks aren't cheap in an absolute sense, but relative to the pure-play Net stocks, they're a bargain. (Dell, for instance, still sells at only a modest premium to its unbelievable--and real--profit growth, as does Cisco.) With these investments, you simply bet that use of the Web will continue to grow. You don't have to figure out which company is going to win the battle for eyeballs. Says PaineWebber's Kerschner: "You probably won't have as much of a thrill, but you probably won't feel as much of a kill." Hey, it's your nickel.