Is 2026 going to turn out like 1999?

Only if the market fails to come to its senses about SpaceX and Tesla

I asked Google Gemini AI to make an image for this article - I rather like this, even though it’s obviously AI propaganda

I am getting all nostalgic about 1999, the year when I first started paying somewhat obsessive attention to the share prices of technology companies, a habit which has endured to this day. As a BBC TV business reporter, watching the stock market had been part of my job for a decade by then, but this was different. In the US, and to a much smaller extent in the UK, companies suddenly needed to have one simple thing at the end of their name to send their shares rocketing - the suffix .com.

All the traditional ways of valuing a company had been blown away by the arrival of the Internet as a consumer proposition. Suddenly, you did not need profits or even substantial revenues to be a valuable business. You just needed eyeballs - in other words, an online audience whose attention you could monetise. It was an exhilarating time to be a journalist covering the startup scene, and eventually I got my first book, Dot Bomb, out of the experience.

As share prices grew and grew, the warnings from those who had been watching the stock market for longer than a couple of years got ever louder. But the true believers scoffed at these Cassandras, claiming they just didn’t get it - the “dot coms” were going to change the world and the old rules of investment did not apply.

And among those evangelists for the dot coms were many of the analysts at investment banks. Somehow, their lengthy reports on tech companies who were being wined and dined by another branch of the bank, always seemed to end with a “buy” recommendation.

Then, of course, in the Spring of 2000, it all went pop. Enough of the true believers stopped believing to take the air out of the bubble and share prices collapsed. Some of the dot coms survived - Amazon’s business model was strong enough to weather a 94% fall in its stock price - others, such as pets.com and just about any business that had not got their IPO away before March 2000, did not.

Afterwards, there was a reckoning. When it was revealed that some of the analysts who had publicly slapped a “buy” label on companies had advanced a very different view in private - calling them variously “a piece of junk”, “dogs” and “a piece of shit” - they found themselves facing retribution from the regulators and the courts. In the United States a tough new law, the Sarbanes-Oxley Act, was passed to ensure investors were not duped with false information from banks with conflicts of interest. The regulator the SEC made it clear that it was going to be much tougher on any organisation or individual breaking its rules.

So flash forward to 2026 and we appear to have a similar bubble inflating, this time around just two letters, A and I. Once again, we are told by the true believers that this time it is different, that this technology is going to be so transformational that the old ways of valuing a business do not apply.

The comforting thing is that this time we have got strong regulation in place, which should put a brake on the irrational exuberance we saw back at the turn of the century. Or have we?

The SEC chairman Gary Gensler resigned the day after Donald Trump won the 2024 election, knowing that the new President had promised to sack him the moment he was back in the White House, apparently unhappy with his tough stance on cryptocurrency regulation. Since then, several other senior SEC officials involved in enforcing its rules have resigned.

That has led to a sense that something did not smell quite right about the bubbliest event of the current bubble, the recent IPO of Elon Musk’s SpaceX. The space, robotics, AI and social media conglomerate was allowed to have a corporate structure that put extraordinary powers in the hands of its founder and was quickly ushered into the Nasdaq index while other newcomers have to wait a while.

But it was the ratings given by the Wall Street analysts after the IPO that told a really 1999 kind of story. Almost all gave SpaceX a buy recommendation, projecting that what was already a ludicrous valuation for a company expected to lose over $4 billion this year would rocket deep into the furthest reaches of the galaxy.

Meanwhile, the wider market seems to have regained contact with Planet Earth. Having soared briefly above $200 from the $135 IPO price, SpaceX stock has headed steadily downwards. This is healthier behaviour than we saw in the dot-com bubble, with reality gradually seeping in rather than arriving in a vivid flash of enlightenment that sends everyone scurrying for shelter.

There are two other big differences compared with 1999. There are a whole bunch of well-established tech giants whose shares may be overvalued but which are not going to disappear in a hurry. That may sound positive but with the so-called Magnificent Seven - Alphabet, Amazon, Apple, Meta, Microsoft, Nvidia, and Tesla - accounting for 33% of the value of the S&P 500, the impact on the wider economy of a share price collapse will be much greater.

But more worrying is the ever greater role in the markets of retail investors, people hunched over a computer in their basement rather than in a dealing room. Back in 1999 many of the professionals lost their marbles for a while, and it became a courageous position to keep on forecasting that the bubble would burst. One gloom-monger the fund manager Tony Dye - known as Doctor Doom to headline writers - was eventually “let go” by his employers because his views were costing them customers and money. Just a couple of months on, the Nasdaq collapsed and Dye was vindicated, proof of the old saying that the market can be wrong longer than you can be solvent.

The trouble is that with a new generation of investors who act more like fans, whether it is “hodling” Bitcoin or backing meme stocks such as GameStop, the market is wrong for ever longer about certain companies.

Tesla’s share price took off just as the business was facing strong headwinds. Is this really a $1.2 trillion business?

The prime example is Elon Musk’s other public company Tesla. Back in 2020 when its stock market valuation soared above Toyota’s to make it the world’s most valuable car company, the price could just about make sense if you believed that within five years or so it would have left its rivals in the dust when it came to sales and profits while building a dominant position in the robotaxi industry. Instead, disappointment has followed disappointment. Just about every forecast made by Elon Musk about progress towards autonomous driving has proven to be wildly optimistic, Toyota still sells six times as many cars as Tesla, and China’s BYD has swept past Musk’s firm to take first place in EV sales.

But you would not know that from Tesla’s share price, which rose to levels making it worth as much as the major Western car companies combined. Wednesday’s second quarter earnings report, with its nasty surprise about profits way short of market expectations, sent the shares down 10% in early trading on Thursday but that still left Tesla with a price/earnings ratio of nearly 300.

Just to put that in context, investors are paying $300 for every $1 of Tesla earnings whereas the standard P/E ratio in the car industry is between 10 and 15 - the market is not expecting much growth from a Toyota or a Volkswagen or a General Motors but reckons the electric car pioneer will deliver profits on a heroic scale. But a more meaningful comparison is with other tech companies. Alphabet, which owns Google and, through Waymo, the world’s leading robotaxi operation, has a P/E of just 16 while Facebook’s owner Meta is not that far ahead with a score of 22.

Tesla, to sum up, is still grotesquely overvalued and so is SpaceX, with fervent Musk devotees doubtless seeing the current weakness in the shares of both companies as a buying opportunity. One day, they will finally wake up but we have to hope that is not the result of some sudden catastrophic event which leads the Muskites to all sell at once, triggering a wider market meltdown.

No, if 2026 is not to end up being the precursor to the Great AI Crash of 2027, just as the magical thinking of 1999 led to the dot com collapse of 2000, we need to start being more honest about what the new technology can and can’t do. That includes investment banks being more transparent about the huge risks involved in putting your savings into a business like SpaceX and regulators being far more active in policing their rules. But, seeing as a booming Wall Street is the only thing that Donald Trump has going for him at the moment, I can’t see that happening.