Could an AI market crash rival 2000 or 2008? Unlikely: McGeever

By Jamie McGeever ORLANDO, Florida, July 29 (Reuters) - With chip stocks tumbling and AI bubble fears spiraling, it's legitimate to ask how bad this stock market volatility could get. Yet it also might seem an odd question, given that Wall Street still appears remarkably resilient. The Dow and S&P 500 are only 1% and 2% below their all-time highs, respectively, and the Russell 2000 small cap index is up 20% this year. But a storm is brewing around the tech stocks that drove the equity rally in recent years. The Nasdaq is flirting with a 10% correction, and while the Philadelphia Semiconductor Index is still up 55% on the year, it has recently slipped into a technical bear market. Many are inevitably drawing parallels with the dotcom crash a quarter of a century ago, when the Nasdaq plunged by 75% and took 15 years to recover. What might be most unnerving now about that crash is that it was so severe even though the root cause of the frenzy, the internet, completely changed the world. Fast forward to today, and this suggests that an investor might be right on AI over the long run and still lose their shirt. But 2000 was nothing compared to the 2008 Global Financial Crisis, and some more high-octane voices on financial social media are claiming that the AI crash they say is inevitably coming could rival or even exceed that credit crunch. They point to the build-up of debt and leverage today – particularly the arrangements where firms in the AI value chain are funding each other, so-called “circular financing” – making comparisons with the complex leveraged products in the U.S. sub-prime housing market in the mid-2000s. Just as the U.S. housing crash led to the global credit crunch and ultimately the Great Recession, these bears contend that the circular financing involved in the trillion-dollar AI buildout is so interconnected that if one pillar falls, the whole edifice will come tumbling down, bringing the wider market and economy down along with it. This argument partly rests on the high concentration of today’s equity indices in AI, chips and tech. For example, semiconductor companies' weighting in the S&P 500 index is a record 19%, more than double what it was in 2000. Meanwhile, investors' speculative tendencies are arguably being encouraged from the very top. "I think we have the ability as an industry to double each year,” Nvidia CEO Jensen Huang told Bloomberg last week. Is 100% annual growth over a sustained period likely? If that weren't enough, warning signs are flashing elsewhere. Leverage is soaring, the cost of insuring against some of the hyperscalers defaulting is at record highs, and private credit is an opaque $3 trillion tinder box at the mercy of rising bond yields. Put it all together, and an unnerving picture emerges. Could the recent ructions in various pockets of Wall Street be just the tip of the iceberg? As worrying as this all is, comparisons with both 2000 and 2008 appear wide of the mark. First, the Nasdaq’s current valuation – and those of many large tech companies – aren’t particularly stretched, especially when compared to the ludicrous heights in the dotcom bubble. The index's 12-month forward price/earnings ratio is around 30, compared with 70 in March 2000. Many companies in today's line of fire are highly profitable, established firms – a far cry from the unprofitable online newbies that drove the dotcom boom. When it comes to claims that a coming AI crash could rival the GFC, it’s possible that some people have forgotten how perilous that situation actually was. It's not hyperbolic to say that the global financial system was on the brink of collapse in 2008. The S&P 500 and Nasdaq each lost 50% of their value in just seven months before bottoming out in early March 2009, with the S&P 500's low on March 6 famously clocking 666. The U.S. economy also contracted by 5% peak to trough between 2007 and 2009, the worst recession since World War Two. Only government and central bank intervention on an unprecedented scale prevented the country from plunging into what could have been a dystopian Depression. If the 2020 COVID recession was a controlled explosion as governments shut down their economies, what the world faced in 2008 was more akin to a potential nuclear explosion. That's unlikely to be repeated. For starters, the GFC took root in the housing market, a sector that represents 16% of U.S. GDP. Housing plays a pivotal role in the U.S. economy, not only through the buying and selling of homes but also through construction and related industries. Housing has a major impact on consumer spending because real estate is most people's biggest asset, making it a fundamental driver of either positive or negative “wealth effects”. In short, the housing market affects everyone. Would another 25% retreat in chip stocks, or a 50% downturn in private credit have equally catastrophic economic or wider market impacts? Almost certainly not. Finally, the GFC dramatically changed the financial system itself. Targeted regulation, stronger capital rules and tighter supervision since 2008 mean the chances that the U.S. banking system will completely freeze as it did in 2008 are now extremely remote. That's not to say real risks don’t exist today around market concentration, sky-high return expectations and irrational exuberance. They do. But if the stock market is in an AI bubble that pops, it almost certainly won't take 15 years to recover or result in a 5% economic contraction. Not every bear market is a crisis. (The opinions expressed here are those of the author, a columnist for Reuters) Enjoying this column? Check out Reuters Open Interest (ROI), your essential new source for global financial commentary. Follow ROI on LinkedIn, and X. And listen to the Morning Bid daily podcast on Apple, Spotify, or the Reuters app. Subscribe to hear Reuters journalists discuss the biggest news in markets and finance seven days a week.

Could an AI market crash rival 2000 or 2008? Unlikely: McGeever

AI selloff raises comparisons with past crashes

Why the comparison with 2000 and 2008 may be overstated

Why an AI bubble would probably not become a systemic crisis

Related News

Replimune Announces Favorable Outcome Of FDA's Cellular, Tissue, And Gene Therapies Advisory Committee Meeting For Rp1 In Advanced Melanoma

Coastal Financial Corp Appoints Christopher D. Adams As Executive Chair Effective July 29, 2026 - SEC Filing

US FDA advisers back Replimune skin cancer drug trial data

Apple disappoints with forecast dogged by supply chain struggles

Copart CEO Jeffrey Liaw sells 27,745 shares for $846,342.05