Financial markets have been thrown into renewed turmoil as fighting in the Middle East intensifies without a clear resolution, a slowdown looms in the AI investment race, and tinderbox conditions in government bond markets fuel alarm. The question investors are now asking is whether stocks could be next in line for a fall.
- 5%+US 10-year Treasury yield this week, highest since 2007
- ≈41S&P 500 CAPE ratio, vs a ~17 long-run average
- $20tn+Combined value of the "magnificent seven" tech stocks
- 30%Fathom Consulting's odds the AI bubble pops in 2027
Why borrowing costs are the trigger
In the past week, US government borrowing costs climbed to their highest level since 2007, with knock-on consequences for households, businesses and other governments worldwide. The worry is that the war in the Middle East is pushing oil above $100 a barrel, igniting inflation, while Donald Trump's tax and spending plans — which have pushed Washington's debt above $40tn — leave investors uneasy.
"These are febrile times," Albert Edwards, a senior analyst at Société Générale, wrote in a note to clients. "The key worry for investors and policymakers alike is the extent to which the current oil price 'shock' will ripple through the global economy and whether it will necessitate sharply higher, recession-inducing, interest rates."
The scale of the worry has already moved central banks. The US Federal Reserve raised interest rates this week for the first time since 2023, defying pressure from President Trump. Markets expect the Bank of England to raise rates four times before the end of next year, the European Central Bank has already raised rates, and the Bank of Japan lifted its policy rate to a 31-year high.
A stretched AI trade
The bigger concern for many investors is that the main hope for economic redemption — artificial intelligence — could itself be a bubble. One popular valuation gauge, the CAPE (cyclically adjusted price-to-earnings) ratio, has climbed to its highest level since 2000. For the S&P 500 it now stands at almost 41 points, more than double its roughly 17-point long-term average and edging toward the 44.19 record set in December 1999, just before the dotcom crash.
Research by Fathom Consulting suggests that for the AI boom to turn a profit, AI-related sales at the companies involved would need to rise by between $600bn and $800bn within two years — growth the consultancy calls unlikely. It puts the odds of the AI bubble bursting next year at 30%.
"For all the impressive advances in AI technologies in recent years, the economics behind the current capex boom do not work," said Brian Davidson, an economist at Fathom. "Sales of AI models need to increase by hundreds of billions of dollars per year over the next two years to justify the current spend. Such growth appears unlikely."
More than 1,000 investors registered for an analyst call held by Jefferies this week on "AI extinction warnings", after several top tech executives called for a slowdown in what they termed "reckless" development. Oracle's shares, which surged after it announced a cloud-computing deal with OpenAI, have since halved as investors worry the company is borrowing too heavily to fund datacentres.
Echoes of 1929 and a margin-call warning from Seoul
Historians of markets see uncomfortable parallels with the run-up to the 1929 crash, when many small US investors bought shares "on margin" — with borrowed money — and were wiped out when prices turned. A modern version has already played out this year in South Korea, where retail investors piled into AI-linked chipmakers on margin, roughly doubling the value of the blue-chip Kospi index before a sharp reversal triggered a wave of margin calls. Goldman Sachs estimates 1.2 million South Korean retail accounts were affected.
Credit markets are also flashing warning signs. The Bank of England reported this summer that the gap between the riskiest and safest high-yield debt has widened since the war in the Middle East began, a sign investors are growing warier of risk. Government bonds have sold off in recent weeks, pushing yields higher and making bonds relatively more attractive than shares.
The US 10-year Treasury yield climbed above the psychologically important 5% level this week. That threshold is "seen by some as a level above which financial markets might go into meltdown," said John Higgins, chief economic adviser at Capital Economics, though he added: "While we aren't convinced that 5% is that 'magic' number, higher Treasury yields would certainly pose a risk to the sustainability of the US public finances as well as threaten equities." Bloomberg macro strategist Simon White has pointed to 5.25% as a more meaningful "inflection point" beyond which stocks and bonds have historically reinforced each other's losses.
The case against a crash
Not everyone expects markets to fall. "A major downturn would need a trigger. Further geopolitical instability could be the catalyst, but we've long argued that the impact of geopolitical shocks on economic activity is overstated," analysts at Oxford Economics wrote in a client note, adding that they see market inflation expectations as overstating the risk of further central bank tightening.
There is also evidence that AI investment is starting to pay off in the real economy: US productivity growth has picked up, and AI is cited as one reason the UK economy grew faster than any other G7 country in the first half of the year. A genuine productivity turnaround could help justify today's high valuations.
Do I think there's a significant dose of reality though in that productivity miracle in AI? Yes. I don't think outright collapse in a dotcom bubble type fashion. But could I see a slow release of air that doesn't collapse the world economy, but slows it down? Yes, I could.
What to watch
For now, the balance of risks rests on three moving parts: whether the war in the Middle East keeps pushing oil and inflation higher, whether the 10-year Treasury yield stays anchored near or breaks decisively above 5%, and whether AI companies can start showing real revenue to match their spending. Any one of the three tipping further could be the trigger investors are watching for.
