What was the dot-com crash?
Between March 2000 and October 2002, the NASDAQ shed nearly four-fifths of its value — and the internet’s first age of innocence ended.
Last updated: October 2026
On March 10, 2000, the NASDAQ Composite touched 5,048.62 in intraday trading — the highest point the index had ever seen, and the summit of a delirium that had convinced the world the future could be bought on credit and never have to pay.
The dot-com crash was the collapse of a speculative bubble in internet-related stocks that had inflated through the late 1990s and burst in 2000. Between March 2000 and October 2002, the NASDAQ Composite index, the American stock market index most heavily weighted toward technology companies, shed nearly four-fifths of its value, falling from its peak of 5,048.62 in March 2000 to a low of about 1,114 in October 2002. Trillions of dollars of paper wealth vanished. Hundreds of internet companies failed. And yet the internet itself survived, and the companies that survived the crash, among them Amazon, eBay and Google, went on to build the modern digital economy. The dot-com crash is therefore a story of two things at once: a financial mania that ended in ruin for many, and a technological revolution that was real even when the valuations were not.
The boom began in the middle of the 1990s, when the World Wide Web moved from a research tool to a public medium. The Netscape browser, released in 1994, made the web easy to use, and Netscape’s initial public offering in August 1995, in which the company’s shares more than doubled on their first day of trading, announced that the internet could make people rich. Investors, journalists and entrepreneurs began to speak of a new economy, in which the old rules of business no longer applied: growth mattered more than profit, market share mattered more than revenue, and a company with a clever name and a large number of users was worth billions even if it had never made a cent. Venture capital flooded into internet start-ups, and the stock market rewarded companies for spending heavily to acquire users, on the theory that profits would follow later.
The boom was fed by a favourable economic environment. The United States economy grew steadily through the 1990s, inflation was low, and the Federal Reserve kept interest rates moderate. The Telecommunications Act of 1996 deregulated the American communications industry and encouraged investment in networks and services. The approach of the year 2000 led companies to spend heavily on computer upgrades to avoid the Y2K bug, the feared failure of older software to handle the change of date. And the stock market itself became a feedback loop: rising prices attracted more investors, whose buying pushed prices higher, until the belief that internet stocks could only rise became, for a time, self-fulfilling. In December 1996, the chairman of the Federal Reserve, Alan Greenspan, warned of irrational exuberance in asset markets, a phrase that became famous. The market ignored him, and kept rising for three more years.
The logic of the bubble was visible in the way companies behaved. Internet companies measured success in eyeballs, the number of visitors to a website, and in mindshare, rather than in sales or profits. They spent lavishly on advertising, on office space, on staff and on acquisitions, and they financed that spending with venture capital and with the proceeds of stock market offerings. The burn rate, the speed at which a company spent its cash, became a standard measure of a start-up’s aggression, and companies boasted of how fast they were burning money. The reasoning was that the internet was a winner-takes-all market, and that the company that moved fastest and grabbed the most users would dominate its category, after which profits would follow. For a handful of companies, that reasoning turned out to be right. For most, it was not.
The symbols of the mania are still remembered. Pets.com, an online pet supply retailer, spent heavily on advertising, including a famous sock-puppet mascot, and went public in February 2000, only to liquidate in November of the same year. Webvan, an online grocer, raised enormous sums and built vast automated warehouses before it had the customers to fill them, and collapsed in 2001. TheGlobe.com, a social networking site, saw its shares rise by more than 600 per cent on their first day of trading in 1998. At the Super Bowl in January 2000, fourteen dot-com companies bought expensive advertising spots, spending tens of millions of dollars to reach a television audience for companies that most viewers had never heard of. The advertisements were a sign of the times: companies that had no profits were spending like companies that had. In the background, the market’s professional watchdogs were often compromised. Analysts at investment banks issued enthusiastic recommendations for companies whose stock offerings their own banks were underwriting, a conflict of interest that later led to enforcement actions and settlements.
The bubble was not only an American story, though its centre was the NASDAQ in New York. Internet manias appeared in Europe, in Israel, in India and elsewhere, and telecommunications companies around the world spent vast sums laying fibre-optic cable and buying licences for third-generation mobile networks. The fibre-optic build-out was so excessive that the price of bandwidth collapsed, bankrupting many of the companies that laid the cables, but the cables themselves remained, and the glut of cheap bandwidth became the infrastructure on which the next generation of internet services was built. The bubble’s excesses, in other words, left behind real assets, even as they destroyed the companies that had paid for them.
The crash began in March 2000. The proximate causes are still debated, but several forces converged. The Federal Reserve had been raising interest rates through 1999 and into 2000 to cool an economy it feared was overheating, which made borrowing more expensive and made safer investments more attractive. The Y2K spending that had inflated technology budgets in 1999 came to an end, and companies cut back on orders. The lock-up periods that prevented insiders and early investors from selling shares after an initial public offering expired for many companies in early 2000, and a wave of selling followed. And the market’s mood turned: investors who had been willing to overlook the absence of profits began to ask when profits would arrive, and for most dot-com companies the answer was never.
The fall was brutal. The NASDAQ peaked at 5,048.62 in intraday trading on 10 March 2000, and then began a decline that lasted two and a half years. By October 2002 it had fallen by nearly 80 per cent. The damage was not confined to the stock market. Hundreds of internet companies failed, and the failures spread through the economy: advertising dried up, technology spending collapsed, and the telecommunications companies that had borrowed to build networks defaulted on their debts. WorldCom, the telecommunications giant, collapsed in 2002 amid an accounting scandal and filed what was then the largest bankruptcy in American history. Enron, the energy company whose rise had been bound up with the same culture of hype and accounting manipulation, collapsed in late 2001. The American economy entered a recession in 2001, and unemployment rose. The crash also reached beyond the United States: stock markets in Europe and Asia fell, and the global technology industry entered a long winter.
The human cost of the crash was real but unevenly distributed. Employees who had been paid in stock options saw their wealth evaporate. Investors who had bought at the peak lost heavily, and small investors who had been drawn into the market by the boom’s publicity were often the worst hit. But the crash also had a paradoxical mercy: it hit a rich country in a boom, and the broader American economy, though bruised, was not devastated. The Federal Reserve cut interest rates sharply, the government cut taxes, and the economy recovered. The companies that failed left behind engineers, managers and ideas that were quickly reabsorbed into new ventures. The crash destroyed fortunes, but it did not destroy the technology, and it did not destroy the talent.
The most important fact about the dot-com crash is that the internet was real. The bubble was a financial phenomenon, but it was built around a technology that was genuinely transformative, and the companies that survived the crash went on to become the most valuable in the world. Amazon, which had been mocked as a bookseller that would never make money, survived by cutting costs, diversifying into new product categories, and eventually building a cloud computing business that became the backbone of much of the modern internet. eBay survived by turning its auction marketplace into a durable business. Google, founded in 1998, was still private during the crash, and it built its advertising business in the years after the bubble burst, becoming one of the great companies of the twenty-first century. The crash did not kill the internet. It killed the companies that had no path to profit, and it taught the survivors to find one.
The crash also changed the rules. The accounting scandals of the period, above all Enron and WorldCom, led to the Sarbanes-Oxley Act of 2002, which tightened the regulation of corporate accounting and executive responsibility in the United States. Regulators moved against the conflicts of interest that had led analysts to recommend stocks they privately doubted. Investors became more sceptical of companies that promised profits in some distant future, and the venture capital industry became more disciplined, at least for a time. The phrase due diligence acquired a harder edge. But the deeper lesson of the dot-com crash is that bubbles and revolutions are often the same event seen from different angles. The railway mania of the nineteenth century, the radio boom of the 1920s and the dot-com boom of the 1990s all followed the same pattern: a transformative technology, a wave of speculation, a crash, and then a mature industry built on the wreckage.
The dot-com crash is still the standard against which later bubbles are measured. When the housing bubble burst in 2008, when cryptocurrency prices collapsed in 2022, and when the markets for artificial intelligence stocks wobble, commentators reach for the same vocabulary: eyeballs, burn rates, irrational exuberance, and the question of whether this time is different. The answer that the dot-com crash gives is a double one. The technology is usually real, and the valuations are usually wrong. The internet did change the world, just as its promoters promised, but it did so on a timetable and through companies that the mania of 2000 could not have predicted. The crash is a reminder that the future arrives, but rarely in the form in which the speculators bet on it.
- Encyclopaedia Britannica, “Dot-com bubble.”
- Cassidy, John, Dot.con: How America Lost Its Mind and Its Money in the Internet Era (HarperCollins, 2002).
- Federal Reserve History, “The Great Moderation” and related essays.
- NASDAQ, historical index data for the NASDAQ Composite.