In the last few weeks there has been a huge crash in the stock prices of the new information technology companies. Until then, the great new economy of computers, mobile phones, digital TV and, above all, the internet has been greeted by capitalist investors around the world as an unstoppable avenue to untold wealth. Every day we have been told in the newspapers of yet another 20-something internet entrepreneur, who, with a small bunch of others, has come with an idea to sell something on the internet, and then becomes a multi-millionaire overnight, thanks to a launch on the stock market.

Investors, like lemmings, have been rushing to throw their money at any dot.com company they can find. The stock prices of these companies have reached the stratosphere. Take the handheld computer maker, Palm. On March 2, it went 'public' by issuing 23m shares at $38 each. Right away, the price jumped to $160, making the company worth $90bn, or equal to the annual output of Indonesia with its 200m people! But Palm has sales of less than $1bn a year and makes no profit - pretty good for a product that's no more than a pocket calculator with an inbuilt modem for the internet. Palm was worth more than General Motors or Boeing who employ hundreds of thousands and make billions of profit.

But now the bubble is bursting. In just a few weeks, the stock price of these high-tech stocks, as measured by the NASDAQ stock market in the US, has fallen 40%. And there's much more pain to come. The whizz-kid internet entrepreneurs may have got their paper millions from gullible and greedy investors but that paper wealth is beginning to burn to ash. Within the next five years, 90% of these upstart companies will have disappeared. It's going to be short-lived for lastminute.com.

It was the same in previous technological revolutions under capitalism. There is an initial boom or rush as investors pour in money to take advantage of supposedly undreamt of riches generated by huge increases in the productivity (and profits) of the new industries. That's what happened in the steam and railway revolution of the 19th century and in the electrical/motor vehicle revolution of the 1920s. But under capitalism, investment is made in a giant casino where lady luck decides where investment goes. It is anarchic, not planned. Investment, inevitably, exceeds the profit created.

Take the example of one company. Sothebys is a famous international auction house. In its recent annual report for 1999 released in March, it reported record auction sales, up 16% to $2.3bn. The company made $443m in revenues from these auctions. But that was down 1% from 1998. Sothebys made even less profit, down from $45m to $33m. Why did it do so badly from such huge turnovers? It had spent $43m on launching a new internet service. So the investment cost in the new technology had actually lowered profits not raised them. Sure, Sothebys hopes to make better profits from the new service in the future. But there's no guarantee of that. On the contrary, it's unlikely. The reality is that capitalist firms everywhere must plough huge amounts into the new technology because everybody else is. If you don't invest, you will be undercut by your rival. But because everybody invests, nobody gets an advantage (at least not for very long) that gains extra profit.

And worse, the huge increase in overall investment starts to outstrip the gains in productivity from the new technology. As Marx explained 150 years ago, technology is dead labour. It's the product of previous efforts by human beings, but it does not create any new value. That depends on living labour using it. As more and more computers, modems and websites are built and used, they drive up costs more than the concomitant increase in productivity from their use by workers. The result is eventually a fall in overall profitability.

We are not quite at that point yet. The great boom in new technology stock prices in the US seen in the last quarter of 1999 coincided with very low interest rates and rising profits as US workers increased the hours they worked with hardly any increase in wages. Profits in US businesses rose 8.3% in the last quarter of last year and labour costs rose just 0.7%. That compares with a fall of 1.8% in profits in the last quarter of 1998. Indeed, the more you look at the profit figures of US companies in the second half of the 1990s, the more you become convinced that the end of 1999 was exceptional. Between 1994-96, US company profits rose massively at nearly 14% a year. In 1997, they were still up 11%. The new technology was hugely productive. But in 1998, profits rose only 1%. In 1999 it was better at 5% up, but only because of that last quarter.

And investment is continuing to race upwards. It is rising faster in relation to the growth of annual output that at any time since 1945. The only comparable period is ominously in the 1920s. And companies and investors have been borrowing hugely to finance this expansion. Without the borrowing, US companies would have enough. Company debt levels in relation to assets have never been higher. It will only take a sustained fall in stock market prices or a slowdown in the spending of American households to bring down the whole house like a pack of cards.

Now the stock market is falling. The bears (sellers) are now on top of bulls (buyers). And the Federal Reserve Bank in the US is preparing to raise interest rates to cool an 'overheating' economy. So by the end of this year, instead of growing at over 4% a year, the US economy will manage only a 2% rate. That is going to put a lot of companies in trouble with their debts. Then it won't be just the internet companies that will be losing money.

That does not mean the information technology revolution will not happen. The Great motor vehicle transformation went ahead in the 1930s, but only after a Great depression that put millions out of work and led to world war. Under capitalism, scientific progress is combined with horrific cost to human wellbeing. First, the small capitalist investors will eat dirt. Then millions of workers in old and new industries take the cosh.