Back in 1999, my friend Stephen bought shares in a company on the day it went public. He told his broker to buy 100 shares "at the market," which is to say at whatever price was required at that moment to get the number of shares he wanted. The company — it was an Internet company, of course — went public at $18, but the first trades were at three times that price, or $54. By the time Stephen's order made it through the system the price per share was up to $78, which is what he paid. But by the end of the day the price had dropped back to $48. While the IPO was a big success for the company founders and investment bankers and the story sounded great that night on the news ("$18 to $48, what a run-up!"), the picture wasn't nearly so pleasant for Stephen. By the end of the first day of trading he had not only lost $30 per share, but his broker had made a margin call and automatically sold the shares leaving Stephen with no shares at all and a $3,000 debt to his broker.

It is simple to look at Stephen's story and say he was crazy to buy at the market. But back in 1999, his behavior looked perfectly rational. What if the company he bought was another Netscape or Yahoo! that would seemingly just keep going up and up? In that case, Stephen's mistake would not have been in buying the shares, but in not having the resources to cover the margin call and hang onto those shares. And it is true that the company he bought that day eventually reached much higher prices and there were moments when he could have doubled his money even having bought at $78. So, if you squint just right and maybe hit your head against a wall two or three times, this kind of investing behavior makes some sense.

Or did make sense. Today, the stock that Stephen paid $78 per share for is worth less than a dollar. Internet stocks in general are in the toilet, Netscape is gone, and even mighty Yahoo! has been humbled. What is going on here?

Some people think what we have is a return to fundamentals, that is a return to sanity in investing. Companies are supposed to be in business to make money and their stock prices ought to reflect their profitability. The people who rightly spoke of the Internet Bubble of 1999-2000 and when it would pop are generally from this crowd. But I think there is more to it than that. I think what we are seeing today is just a different face of the same trading psychology.

There is investing, which is betting on the success of a company and its products or services, and there is trading, which is betting on whether a stock will go up or down in the next hour or day with little regard to products, services, or anything else in the real world. Trading, not investing, pushed those Internet stocks up so high and trading, not investing, has put them in the toilet, too. What's my proof of this? My proof is simply the number of Internet stocks that are now trading at prices substantially under their book values — WAY under. The book value is what we'd get for the company if we put everything up for sale. Sell all the assets, pay off all the liabilities and what you have is the book value. The book value of Covad Communications, the DSL service provider, is $7.17, yet as I am writing this Covad shares are selling for 29/32nds or around 90 cents per share. Does this make sense?

It makes sense if we think Covad is doomed, that there is no way for the company to even survive. This is thinking as an investor would think. But the current pricing and price-to-book value ratio of Covad makes sense to a trader simply because the trader thinks there are better places to put his money.

So what happens to a company like Covad is that downward price pressure from traders force the company to rethink its grand plan. That's not so bad, really, because upward price pressure from traders probably formed the foundation of that plan in the first place. The bad part of downward price pressure from traders is that it probably ends any chance Covad has for raising more money through stock sales or floating bonds. So Covad, with its $7.17 per share in book value, knows that $7.17 will be all it has to work with, maybe ever. Of course, this leads to lots of belt tightening, which is why we see companies with barrels of cash laying off employees and closing business units. They do this not only to impress Wall Street with how seriously they are taking their low stock price, but also because it is the only way they can hope to survive.

The result of all this is a dramatic slowdown in the deployment of certain technologies, but at the same time it creates a remarkable buying opportunity for companies that have loads of cash and don't feel threatened by the current market. That would be Microsoft, of course. Sure, the stock is down, but the company is immensely profitable and has $27 billion in its wallet. What should we buy today, Bill?

If I was Microsoft, I'd buy PSINet, the ISP wholesaler and Internet backbone provider. For around $40 million at today's prices, Microsoft (or Sun or IBM or any number of other companies) could buy PSINet's global infrastructure — an infrastructure that took years and more than $1 billion to build. That would probably cut MSN's network costs in half, saving more than the purchase price in the first year alone. It would put Microsoft's Net business one hop from 4,000 Internet Service Providers, who'd actually be paying Microsoft for that proximity. PSInet would be a good deal for Microsoft at $100 million or even $200 million. But if that's true, why is the price so low?

Enter professor Charles Plott from Caltech. Professor Plott teaches economics and political science to nerds and specializes in experimental economics. Among his experiments are simulations of market behavior that show how markets can drive share prices above what they are actually worth and how those shares (the ones that are vastly overinflated) eventually crash, taking prices lower than they are actually worth, which in some cases is where we are today.

Last year, while shooting my upcoming show "Electric Money," we visited a day trading operation in the UK. This was a big room filled with computers that were rented by the hour to day traders. They could have worked as easily, and probably cheaper, from home, but these traders liked the comraderie of the place and also liked the fact that it had a reputation for making money. They wanted to trade where the winners traded. And it's true, the place actually did make money where most day trading operations don't. The way they did this was by enforcing one rule: When you lost all the money you started with that day, the management made traders go home. This broke professor Plott's trading simulation by not allowing the unsuccessful traders to follow their inclination to double-up and make back their losses, which almost never works. It is a novel system that allows traders to ratchet-up their gains, but limits their losses to what they can afford. Try again another day.

Of course, it all works out in the end for the companies, too, though sometimes at the cost of careers and fortunes. A couple years ago, I wrote a column about Iridium, the satellite phone network, in which I predicted failure but also said that there was no way those 66 satellites would go unused no matter what happened to Iridium. I was right that time: Iridium failed spectacularly and then went through a number of months of threatening to "deorbit" the satellites and burn them up in the atmosphere. Of course it didn't happen. A group came along and bought Iridium's satellites for $25 million and is now taking in $72 million per year for providing satellite phone service to the U.S. government. And $72 million is a pretty good return on $25 million, though not on the $5 to 6 BILLION the Iridium system cost to build in the first place.

It's pretty easy to predict that many of these Internet companies are going to be picked up, like Iridium, for next to nothing and turned into going concerns. All the while, value investors like Warren Buffett just keep trudging along, ignoring the carnage and making a good living. We'll know the bloodbath is finally over when Buffett buys an Internet company.