(Bloomberg) — An age-old economics tenet posits that excessive government borrowing can leave little room for companies to tap financial markets and drive up their interest rates to punishing levels. It’s called the “crowding out” theory.
Most Read from Bloomberg
Now, as the so-called hyperscalers embark on a borrowing binge for artificial intelligence that’s hitting the market with a record flood of bonds, hoovering up hundreds of billions of dollars, some are wondering if the opposite is starting to happen.
That’s not to say that a reverse-crowding-out effect is forcing the US government to rein in its spending. In fact, it’s spending more than ever, running deficits that are unprecedented outside of a crisis. And that, along with the surprising resilience of the economy and inflationary jolt from the war in Iran, is largely seen as the primary force behind the relentless rise in bond yields that has pushed the Treasury’s long-term borrowing costs to a 25-year-high and elevated rates across the financial system.
But as the AI borrowing spree keeps accelerating, blowing past Wall Street’s forecasts, the spillover effect is, many investors and analysts agree, playing a growing role by taxing the market’s ability to absorb so much debt.
“Whoever’s issuing, be it a government or a hyperscaler or a non-hyperscaler credit, is now competing with more borrowers,” said Tony Rodriguez, head of fixed-income strategy at Nuveen Asset Management. “And therefore yields have to be higher.”
Already this year, investment-grade companies have sold nearly $1.5 trillion of bonds, a 36% jump from a year earlier, putting them on pace to eclipse the record from 2020, when businesses were rushing to seize on near-zero interest rates. Nomura Securities estimates the roughly $200 billion of borrowing by the biggest tech companies alone is equivalent to roughly 25% of the US Treasury’s net issuance of notes and bonds to private investors, five times more than it was in 2025.
There are no signs that the AI debt deluge is slowing. Amazon.com Inc. and Alphabet Inc. both recently increased their spending forecasts, while chipmaker Nvidia Corp. announced it is working with Wall Street firms to raise another $500 billion for AI.
As the slew of longer-dated bonds keeps hitting the market, some investors have sold US Treasuries. That freed up cash to buy higher-yielding debt from immensely profitable companies like Alphabet, whose recent 30-year debt was issued at a yield of nearly 6.4%, 1.15 percentage points more than comparable Treasuries. Last month, a bond that’s financing a data center tied to Meta Platforms Inc. paid over 7.5%.
“When these new issues come into the market, there’s a premium that comes into play,” said Olumide Owolabi, a senior portfolio manager for Neuberger Berman, whose $1.1 billion core bond fund has bought debt sold by Oracle Corp., Meta and Alphabet, among others. “Selling Treasuries to go buy corporates doesn’t mean I do not like Treasuries. It just means I have a better opportunity.”
According to Morningstar Inc., bond funds that are limited to investment-grade US debt dialed back their Treasury holdings this year and increased their average allocations to corporate bonds to 30%, a three-year high. A similar shift has been seen among overseas investors, long a key source of funding for the federal government.
That’s complicating the Trump administration’s effort to ease the pressure on the Treasury market and, in turn, pull down the rates on mortgages and consumer loans. While Treasury Secretary Scott Bessent once expressed confidence that Trump’s fiscal policies would do that by taming the deficit, the government has continued to spend nearly $2 trillion more than it brings in each year, keeping the national debt growing far faster than the economy.
To avoid pushing up longer-term rates even more, Bessent has relied heavily on sales of short-term debt to finance the deficit instead. That has allowed him to keep the size of 10-year and 30-year bond sales steady. Barclays last month estimated that the move will curb the net supply of new Treasury notes and bonds to about $1.2 trillion this year, about $440 billion less than in 2025.
But the surge in new AI debt — much of which is also long-dated — has more than filled the breach: According to Barclays, the net supply of corporate bonds is expected to swell by $474 billion, much of it because of sales by the tech giants.
Alex Payne, a senior portfolio manager at Vanguard Group Inc. who focuses on Treasuries and mortgages, said AI spending has become the top issue during his firm’s daily discussions because of how much it’s affecting the pace of economic growth and new debt sales.
“AI has been the biggest story in markets for the last couple years,” he said. “It just touches everything.”
Quantifying the precise impact on interest rates is difficult and estimates vary. But in a note to clients on Friday, Bank of America Corp.’s economists said the AI borrowing is “potentially crowding out long-end Treasury demand” and has played a major role in the rise of bond yields. They estimated that the surge in corporate-debt sales — along with a rise in issuance of mortgage-backed securities — pushed up 10-year rates by about 0.3 percentage point this year.
That is likely problematic for the Treasury, according to Jonathan Cohn, head of US rates desk strategy at Nomura.
A growing supply of long-dated debt from AI-related companies could force the Treasury to reduce the size of its own long-term debt sales to avoid higher borrowing costs. While such scenario is not Cohn’s base-case scenario, he said it has become a real possibility.
“For Treasury, the sheer amount of duration supply forced onto the market, notably at the long-end, should be a concern,” he said in a note to clients.
By some estimates, the AI boom is really just beginning. JPMorgan Chase & Co. strategists expect that spending on the infrastructure for it will total $5.5 trillion through 2030.
Greg Peters, the co-chief investment officer of PGIM, said the borrowing needed to pay for much of it will continue to keep longer-term rates elevated, regardless of what policymakers do.
“That is a crowding-out effect,” he said during an appearance on Bloomberg Television.
“It is important to remember that we are just starting,” he said. “This hyperscaler debt issuance story has really just begun.”
–With assistance from Gerson Freitas Jr..
Most Read from Bloomberg Businessweek
©2026 Bloomberg L.P.
