With artificial intelligence (AI) trading continues to drive U.S. stocks to new highs, prompting investors to increasingly question whether we are experiencing another financial bubble destined to burst.
The answer is not so simple, at least not historically.
In 2025, the S&P 500 is projected to rise by 16%, with Nvidia, a winner in the field of artificial intelligence , among the top performers. Alphabet, Broadcom and Microsoft They have made the largest contributions. However, at the same time, there is growing concern about the enormous expenditures of large tech companies in the field of AI. Data shows that Microsoft , Alphabet, and Amazon... Meta's capital expenditures are projected to grow by 34% over the next year, totaling approximately $440 billion.
Meanwhile, OpenAI has pledged to invest over $1 trillion in AI infrastructure, a staggering figure for a privately held company that is not yet profitable. Perhaps even more worrying is that the expansion in the AI field relies too heavily on “circular investment,” with investments and spending flowing back and forth between OpenAI and a few publicly traded tech giants.
Invesco Brian Levitt, the group's chief global market strategist, points out that throughout history, overinvestment has been a recurring phenomenon whenever societal technological advancements have emerged. He cites the development of the railroad, electricity, and internet industries as examples. This time may be no exception.
He said, "At times, infrastructure construction may exceed the short-term needs of economic development. But that doesn't mean the railways weren't built, or the internet didn't exist, right?"
Nevertheless, with stock valuations continuing to climb and the S&P 500 having just achieved double-digit percentage gains for the third consecutive year, investors are beginning to worry about how much more AI can drive US stocks up, and what kind of devastation the US stock market will suffer if the AI bubble bursts.
Gene Goldman, chief investment officer at Cetera Financial Group, said, "The bubble is likely to burst in the bear market," but he doesn't believe there's a bubble in artificial intelligence stocks. He added, "We don't see any signs of a bear market coming anytime soon."
The following is a comparison between the current AI boom and previous market bubbles:
Speed/Time Span
A simple way to gauge whether the surge in AI-driven tech stocks has gone too far or too fast is to compare it to previous bull markets. (Bank of America) Research by strategist Michael Hartnett shows that, looking back at the 10 stock market bubbles that have occurred globally since 1900, the average duration was slightly over two and a half years, with an average increase of 244%.
In contrast, the AI-driven rally has entered its third year, with the S&P 500 up 79% since the end of 2022, while the tech-heavy Nasdaq 100 has risen 130%.
While it's difficult to draw any conclusions from the data, Hartnett warns investors against selling stocks even if they believe there's a bubble, as the final stages of an uptrend typically see the largest gains, and missing out can be very costly. He suggests that one way to hedge against risk is to buy undervalued stocks, such as UK stocks and energy companies.
Concentration
The 10 largest stocks by market capitalization in the S&P 500 now account for about 40% of the index, a concentration unseen since the 1960s. This has worried some investors, including veteran Wall Street research expert Ed Yardeni, who stated last December that it was no longer meaningful to advise investors to overweight technology stocks.
However, market historians argue that while the current concentration may seem excessive compared to recent levels, it is not without precedent. Paul Marsh, a professor at London Business School who has studied global asset returns over the past 125 years, points out that the proportion of top-performing stocks in the US stock market reached similar levels in the 1930s and 1960s.
He stated that in 1900, 63% of the total market capitalization in the United States was related to railroad stocks, while by the end of 2024, that proportion had dropped to only 37%.
Fundamentals
TS Lombard economist Dario Perkins stated that asset bubbles are difficult to detect in advance, but relatively easier to detect afterward, because fundamentals are often the focus of debate, and the metrics that investors pay attention to can also be constantly changing.
He said, "Tech enthusiasts can easily claim 'things are different now,' and that fundamental valuations will never be the same again."
However, some fundamental factors remain crucial. For example, compared to the dot-com bubble era, today's AI giants still have lower debt-to-income ratios than companies like WorldCom Inc., which went bankrupt in 2003. Moreover, companies like Nvidia and Meta have reported strong profit growth from their AI businesses, something that was not the case during the speculative era 25 years ago.
Valuation
The S&P 500's valuation has reached its highest level since the early 21st century, at least according to the Shiller P/E ratio. This metric, also known as the cyclically adjusted P/E ratio (CAPE), calculates the price-to-earnings ratio using the average earnings over 10 years, adjusted for inflation, while the standard P/E ratio uses the earnings of the past year, thus smoothing out the impact of economic cycles on valuations. Historical data shows that a CAPE ratio exceeding 25 in the US stock market indicates a period of "irrational exuberance."
Bullish investors believe that while tech stocks have driven up market valuations, their growth rate is far lower than during the dot-com bubble. In 2000, Cisco... Nvidia 's price-to-earnings ratio once exceeded 200 times its earnings over the past 12 months, but now that figure is less than 50 times.
Richard Clode, a fund manager at Janus Henderson, said that in an environment where valuations are undisputed, stock prices will decouple from earnings growth.
“We haven’t seen that yet,” he said.
Investor sentiment
Discussions about a potential stock market bubble persisted throughout last year, then intensified significantly in November and December, fueled by warnings from "Big Short" Michael Bury and the Bank of England. Media compilations show that over 12,000 news articles mentioned the term "artificial intelligence bubble" in November, roughly equivalent to the total for the previous ten months.
A Bank of America survey in December showed that investors viewed the artificial intelligence bubble as the biggest "tail risk" event. More than half of the respondents said the "Big Seven" tech stocks were the most crowded trades on Wall Street.
This is very different from the dot-com bubble era, Barclays bank Venu Krishna, head of U.S. equity strategy, said that people were "incredibly excited" at the time that the internet would revolutionize everything. But now, with increased debt issuance, there is growing skepticism about whether investments in artificial intelligence will pay off.
“I won’t ignore this, but I generally think this kind of scrutiny (of investors) is beneficial,” he said. “In fact, it is this kind of scrutiny that prevents extreme situations like a crash.”
(Article source: CLS)
