My Finest Hour

I read this piece, from December 7, 1998, over again-and-again for this book, and in retrospect it's one of my best.

At the height of the Internet gold rush I correctly identified what was happening, fingered one of the more public culprits, and predicted that the game would end soon. Not bad for a struggling journalist...

Wall Street 1999

Usually I write about Internet Commerce, but this week I think we should all consider the impact of the Internet on commerce, and on the real economy.

So who'll play James Cramer when this sequel to the movie "Wall Street" comes out? He's obviously the Gordon Gecko of this age, but I'll predict right now his importance will be no more lasting than those of his 1980s' counterparts.

The fact is the Web has enabled a takeover of the world's largest asylum, Wall Street, and the inmates are running wild. From "day trading" rooms , and behind the screens at Web trading sites , individual investors are calling the shots, a few hundred shares at a time. This can be seen most clearly in the market for Internet-related issues. It started last summer, when professionals using conventional valuation techniques sold Yahoo and Amazon short, knowing they weren't worth what the "screen savers" were paying.

But here's what happened. While Yahoo and Amazon have a lot of stock, very little of it trades. So small changes in supply or demand for this "float" can have a big impact on price. Remember that 10,000 people buying 100 shares each represents 1 million shares of demand, and there are now hundreds of thousands of such investors. (The average trade in Yahoo and Amazon in the last few months has been just 300 shares, CNBC reporter David Faber has reported.) Thus the pros had to buy back shares and take huge losses. When the shorts retreated, and an autumn bear market turned into a mere correction, the way was clear for the real fun to begin.

When the threat of "shorting" - selling a stock now hoping to buy it back for less - is removed, many Internet issues become as easy to manipulate as penny stocks were a decade ago. (Cramer notes you can also short a stock by "buying a put," an option to sell at today's price in hopes the price will fall before you have to put out cash.) Since the signals that induce day traders to move come in the form of price changes on screens there's no conspiracy here. The game is legal. Without short sellers, however, there's no one placing bets that can make a target fall in price and provide discipline to the market.

Let's say you were in on the move in Yahoo. What was $10,000 is now $80,000, and it's now easy to borrow $40,000 on the position with which to buy eBay for $50/share. Within a few weeks you've now got $160,000, so you get downwind of its competitor, OnSale, trading two weeks ago around $20. (It went up to $100, then back down to $50 within a week.) You're a player now. Since the businesses are real, "news" (like a loss exceeding analysts' expectations) can be planned for. If you lighten up on your winners slowly, you're making real money. And if you're still at your day-trading screen, you can play this slot machine hundreds of times during any move.

If this game sounds familiar, you get our 'A' in history. The same game was played in the late 1920s by amateur investors. The game ended with the October, 1929 stock market crash. The problem, as with all speculative bubbles, is that if gains aren't driven by real profits, a market correction can become a collapse. The people who run the markets were doing their best to change that psychology, until they decided a market fall threatened the global recovery. But even if Alan Greenspan is in cahoots with you, a bubble is still a bubble. The game ends when there's a true rout that forces players out of the markets. Many thought that rout came last summer, and shares fell 25% in value on average, but the quick rally back to old highs emboldened the plungers. (This game will likely end as it did in 1929, with a recession. President Hoover denied the next year that recession was occurring. "It's just a small depression," he said.)

Over the last week the game has accelerated. It's also gotten more dangerous, moving from multi-billion dollar plays like Yahoo to smaller, more easily-manipulated issues like Books-A-Million. Even Davvix, a Christian community site, found itself used in this way when it announced it would sell gospel music online. (Cramer, to his credit, calls such plays "bad Internet," but it's the overvalued "good Internet" profits that fuel the game.) The constant churn has left those who didn't move quickly enough with big losses. Unlike the situation in a real casino, the wheel doesn't stop so you can cash out in comfort. The more you've got, the harder it is to get out the door with your gains.

The mania holds important lessons for Internet Commerce. First, cheap liquid markets are no protection against manipulation by a herd. Second, the Internet speeds up everything, both good and bad. Third, short sellers aren't all bad. Fourth, lots of stock trading sites will go under when the losers sue over what they did to themselves or just renege on bad trades. Fifth, we're all going to be hurt when the bubble bursts, and being on the Internet won't prevent that. Sixth, fulfilling orders, on both the buy and sell side, is the key to staying in an Internet business for the long haul. Finally, don't be confused by small hiccups in stock prices - when the bubble bursts, the explosion will be heard everywhere. (How bad will the crash be, whenever it comes? See the history of Japan, from 1987 to the present, for answers.)

There's a Pulitzer Prize for the reporter whose paper funds even two weeks in one of these trading rooms, profiling the traders, the systems, and the atmosphere. If you're a publisher, consider this a book proposal...

Prince Was Right

My predictions for 1999 were just a year off - most of them happened in 2000. But I did correctly identify the Nostradamus of our time - Prince Rogers Nelson of Minneapolis, Minesota. Princes' song "1999," written in the early 1980s, certainly did a better job of predicting what would actually happen in society than either George Orwell's "1984" or Arthur C. Clarke's "2001."

By 1984 the world wasn't entirely ruled by oligarchies, and I still don't have a zero-gravity toilet. But the Bard of Minneapolis was right - "2000-0-0 party over, out of time..."

This column ran on December 28, 1998:

Tonight We're Gonna Party Like It's...

Did you know "the artist formerly known as Prince" was a great computer engineer? It's true. He apparently discovered the Y2K bug way back in the early 80s. Listen to this lyric - "Year 2000, 0, 0, party over, out of time. So tonight we're going to party like it's 1999."

In that lyric, The Artist may have also been speaking of electronic commerce. Most large companies will be consumed this year by compliance testing for the Y2K bug. The bug itself isn't that hard to stamp out. (My wife, who writes transaction processing programs in Cobol, told me you can write a subroutine stating if a new date is 60 less than the most recent date, it's a valid date. I pointed out this creates the "Year 2060" problem. She replied. "None of these programs will be running then." My riposte: "That's what they said.") The problem is finding the little bugger, then making sure everything will keep running at the appointed time. You can't just run a test of January 1, 2000. You also have to run multiple months, and multiple years. (Don't forget that 2000 is a leap year, while 1900 was not.) This is what is earning your contract programming friends all that $150/hour overtime.

Assuming your small company has gone through all this tsuris, you have a huge opportunity. You can write, and build, new applications and new features, knowing the big boys can't match you. I know some Clueless stock analysts have predicted that "Wal-Mart is going to squash Amazon like a bug." My point is that won't happen right away.

Now that we have the Big Story down - the 1999 Lewinsky if you will - what other trends can we see in Internet commerce for next year? Here are several: