Bank of England sounds alarm on financial stability risks due to AI bubble
The Bank of England (BOE) has expressed concern about the soaring valuations of AI companies as investors increasingly bet big on a future payoff that may never happen.
In a blog post published on Friday, the BOE pointed to the fact that AI stocks now account for 44% of the S&P 500’s market cap, and their financial impact now extends far beyond the technology sector.
In fact, Michael Cembalest, chairman of market and investment strategy for JPMorgan Asset Management, wrote in a report last month that “AI related stocks have accounted for 75% of S&P 500 returns, 80% of earnings growth and 90% of capital spending growth since ChatGPT launched in November 2022.”
And “AI related stocks” means the “asset price story” is not only being told by the biggest AI companies, but all the other players in the ecosystem, including the hyperscalers, cloud service providers, AI chip manufacturers and companies that specialize in data center infrastructure, according to the BOE.
The central bank said that “understanding the full impact of AI-driven events on asset prices and financial stability requires an understanding of the full AI stack, including key dependencies.”
The BOE notes how JPMorgan has developed an AI stock index with 30 names from the S&P 500 on it, which spans multiple sectors beyond tech, including utilities, real estate and consumer discretionary.
As investments continue to pour into the space, AI stocks have now pushed some of the valuation metrics for US stocks to their highest levels since the dot-com bubble, according to the BOE.
These soaring valuations will of course be justified if future earnings growth meets expectations.
And the growth projections are being driven by the expected continuation of substantial AI adoption, which would then also drive the need for massive amounts of computational power to meet the demand for compute and the need for more powerful AI models.
However, the BOE said that the valuations for AI companies could come under scrutiny for a number of different reasons, including a failure of these companies to monetize the users of their AI applications in a way that satisfies Wall Street.
“The speed of AI progress and economic impact is highly uncertain, as seen in the wide range of estimates of the future impact of AI on productivity by research economists and timelines to very powerful AI models by AI experts,” the BOE said.
Debt financing is raising red flags
The continued buildout of AI infrastructure is expected to cost trillions of dollars, which the BOE projects will be significantly financed by debt. Because the AI boom has so far been an “equity story,” the BOE does not expect any potential drop in AI-related asset prices to have a broad impact on financial stability.
However, that could change if financing for AI infrastructure continues to be heavily debt-driven.
“If the projected scale of debt-financed AI and associated energy infrastructure investment materialises over this decade, financial stability risks are likely to grow,” the BOE said.
“Banks would be exposed to this directly through their credit exposures to AI companies, as well as indirectly through their provision of loans and credit facilities to private credit funds and other financial institutions which are exposed to AI-impacted asset prices.”
Bloomberg reported last week that the BOE is looking into the expected growth of lending by banks to fuel the building of data centers. Much of the construction has thus far been paid for through equity, but with costs expected to balloon into trillions of dollars, lending is expected to pick up steam.
The BOE declined to comment, but a source told Bloomberg that one area the central bank is focused on is the “indirect exposure” caused by securitized loans, which are individual loans that have been pooled together and packaged into interest-bearing securities that are then sold to investors.
Securitized loans, especially subprime mortgages that were packaged into complex financial instruments, were a primary driver of the 2008 financial crisis.
But while the BOE is raising concerns about soaring US stock valuations, economists from the Bank of Italy presented model-based evidence in a report last month to cast doubt on the fears of an AI bubble forming, arguing that “the earnings growth rates required to justify current prices do not appear implausible given the underlying drivers of recent and expected performance.”
“We also argue that the adoption of generative AI technologies will be a major engine of growth, and it is already transforming markets and displacing jobs at scale,” they wrote. “Its rollout fuels rising demand for computing power, largely supplied by US tech companies, reinforcing their dominance.”