The good news: Bubble predictions vastly outnumber actual episodes. The bad: There may be a 97% chance of one, right now.

By Steve Goldstein

Bubble warnings vastly outnumber actual episodes.

It turns out, the business media cries wolf about impending stock-market doom too frequently. But that doesn't mean that the warnings have no predictive value, and a new piece of research finds there's up to a 97% chance that we're in one right now.

That's the finding from a new paper by University of Cambridge and London Business School researchers, who trained artificial intelligence on 170 years worth of news articles from The Economist. The research was first written about in the Financial Times.

The Economist, it should be said, has an unfortunate track record of having its cover stories turn out to be contrarian indicators. That's been witnessed recently - on Oct. 10, The Economist's cover was "will bonds blow up?" which has been followed by a precipitous decline in the yield on the 10-year U.S. Treasury note BX:TMUBMUSD10Y.

The researchers looked at more than just cover stories - through AI, they examined 795,964 published articles, complemented with a catalogue of 22 bubble episodes across ten countries.

The researchers then evaluated each article on several dimensions, including the sentiment expressed toward financial assets, the intensity of speculative narratives and the stage of a potential cycle. In all, they created 14 monthly indicators and standardized them to their preceding 20-year histories.

The findings: articles imply the average probability of a major market collapse within six months is approximately 20%, but the realized frequency is just 5.7%.

That said, bubble warnings do tend to precede actual bubbles - including in the dot-com crash, the savings-and-loan crisis and the housing boom before the global financial crisis, the researchers found.

An analysis of news articles finds the probability of a stock-market bubble in September was as high as 97%.

Their conclusion was that the likelihood there was a bubble in Sept. 2026 was higher than its peak at the dot-com bubble - as high as 97%, with other indicators giving a probability of 40%.

The S&P 500 SPX has fallen 6% from its record close on Oct. 6, and the tech-heavy Nasdaq Composite COMP has dropped 12% from that date.

-Steve Goldstein

This content was created by MarketWatch, which is operated by Dow Jones & Co. MarketWatch is published independently from Dow Jones Newswires and The Wall Street Journal.

 

(END) Dow Jones Newswires

11-18-26 0623ET

Copyright (c) 2026 Dow Jones & Company, Inc.

The articles, information, and content displayed on this webpage may include materials prepared and provided by third parties. Such third-party content is offered for informational purposes only and is not endorsed, reviewed, or verified by Morningstar.

Morningstar makes no representations or warranties regarding the accuracy, completeness, timeliness, or reliability of any third-party content displayed on this site. The views and opinions expressed in third-party content are those of the respective authors and do not necessarily reflect the views of Morningstar, its affiliates, or employees.

Morningstar is not responsible for any errors, omissions, or delays in this content, nor for any actions taken in reliance thereon. Users are advised to exercise their own judgment and seek independent financial advice before making any decisions based on such content. The third-party providers of this content are not affiliated with Morningstar, and their inclusion on this site does not imply any form of partnership, agency, or endorsement.

Popular

Sponsor Center