MW Bank of America strategists see a different historical parallel for the AI rally - and it isn't the dot-com boom
By Steve Goldstein
Strategists have doubts over whether there's really a golden age of profitability that markets seem to expect
Chinese Foreign Trade and Economic Cooperation Minister Shi Guangsheng signs his country's World Trade Organization $(WTO)$ accession documents during a ceremony in Doha in 2001. Investors may be hoping that AI boosts margins the way China's accession into the WTO did, say strategists.
The artificial intelligence-fueled rally is often compared with other capital expenditure-led build-outs, like that of the internet during the dot-com bubble.
Bank of America's European equity strategists accept that this may be the case, but also make a different comparison.
"The story of high profit expectations is not just a tech story, with global 12-month consensus margins excluding tech also hitting a fresh high of 12%," said strategists led by Sebastian Raedler.
"One plausible interpretation is that the market is pricing a broad step-up in corporate profitability in response to the accelerating AI rollout, similar to what happened when China entered the World Trade Organization in 2001."
Back then, the entry of more than 1 billion Chinese workers into the global economy reduced the wage share of gross domestic product - typically between 61% and 65% - to a lower range between 57% and 60%. Similarly, the corporate-profit share of GDP rose to between 10% and 12%, from 5% to 8%.
Earnings expectations are running far ahead of their traditional relationship to PMIs.
The Bank of America strategists noted that, in both the U.S. and Europe, earnings expectations for S&P 500 SPX and Stoxx 600 XX:SXXP companies have moved well ahead of their traditional relationship with the global economy, as shown in purchasing-manager indexes. Given this divorce from the underlying economy, much has to go right for what they call a "golden age of corporate profitability" to materialize.
"In shifting margin expectations to an even higher level now without the support of a clear-cut acceleration in global growth momentum, markets appear to be pricing in a scenario in which the rollout of AI tools could lead to a further structural increase in corporate productivity and profitability by allowing corporates to replace human capital with cheaper AI bots," they said.
The alternative argument about the AI build-out is buttressed by the fact that capital expenditure isn't immediately reflected in profit and losses.
"Capital spending filters through the balance sheet instead - and will show up on the P&L only slowly over time in the form of higher depreciation charges," they noted.
"This pattern of initially elevated margins at the start of an investment boom-driven by scarce capacity and strong demand-followed by booming investment, rising supply and a subsequent squeeze on pricing power and returns as depreciation catches up, arguably explains the boom-and-bust dynamics of many investment cycles in the past (including the railroad boom of the 1880s, the radio boom of the 1920s and the tech boom of the 1990s)."
Related: The AI capital-spending boom looks historic. Eleven other booms suggest a bust is inevitable.
Even without depreciation, the strategists flagged other risks. A slow re-opening of the Strait of Hormuz could force demand destruction, and some AI scenarios could hit profits.
AI could curtail demand through labor-market weakness. The expense of AI and surging token costs could slow adoption. Or simply, AI may not boost productivity much. A further risk comes from the political reaction to AI, already in evidence in the fight over data centers, which could lead to windfall taxes.
The strategists said they are negative on European equities and underweight cyclicals versus defensives.
-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
May 29, 2026 06:48 ET (10:48 GMT)
Copyright (c) 2026 Dow Jones & Company, Inc.
Comments