Traders work on the floor of the American Stock Exchange (AMEX) at the New York Stock Exchange (NYSE) in New York, US, Monday, March 9, 2026.
Michael Ngale | Bloomberg Getty Images
BANGKOK – The International Monetary Fund warned on Tuesday that government bond yields could rise in an erratic manner, as rising inflation, rising debt and AI-related debt pressure sovereign bond markets.
Government bond yields in several advanced economies have risen to their highest in decades, driven by higher energy costs due to the Middle East conflict and expectations for tighter monetary conditions as central banks look to curb rising inflation again.
In its latest Global Financial Stability Report, the fund warned that sovereign bond risks could worsen, warning of a “rapid rise in sovereign bond yields.” The report was released Tuesday as part of the IMF-World Bank annual meeting in Bangkok, Thailand, this week, where more than 18,000 delegates gathered to discuss the state of the world economy and financial system.
High debt burdens and rollover risks make the bond market “vulnerable to more frequent fiscal and inflationary shocks,” the report said. The IMF has warned last week that global public debt has reached its highest level since World War II and will soon exceed 100% of GDP.
“The impact could be severe given the fundamental importance of sovereign bond markets,” the IMF said Tuesday. A sharp rise in yields will mark down asset values through higher discount rates, including for AI-related assets, and erode the value of government bonds posted as collateral.
The IMF also noted growing borrowing by AI companies – which compete with sovereign bonds for investor demand – as a source of additional tension in the bond market. Hyperscalers’ share of global investment grade bond issuance will rise to 10% from less than 1% by 2024, the report said.
Global AI builders have seen tech giants, such as Alphabet Inc google, Meta platformhyperscalers and AI labs raised hundreds of billions of dollars in debt this year.
“AI-related issues – many of which are longer – are already large enough to compete with government bonds for investor capital, although they do not appear to be the dominant driver,” according to Marc Seidner, CIO of non-traditional strategies at Pacific Investment Management.
Hedge funds are worried
Hedge funds’ large holdings of Treasuries leave the market “vulnerable to disorderly unwinds,” the IMF said. Partially leveraged-based trading was not eliminated without significant pressure on the short-term loan market, but the impact is still a concern.
The hedge fund’s share of US Treasuries will rise to around 9% in 2025 from around 4% in 2022, mostly through arbitrage strategies, the IMF said, citing data from the Federal Reserve.
Caio Ferreira, deputy head of the IMF’s division for financial supervision and regulation, said the concern stems from the rising influence of hedge funds and increasingly overlapping positions, meaning that some funds hold the same trade, which could increase the impact of collective unwinding.
On the bright side, liquidity has been maintained in the US, Europe and the UK, the IMF said. Japan is an exception: liquidity in the government bond market has “continued to deteriorate” as yield volatility has increased since the end of the yield curve control.
The fund urged debt managers to extend maturities, warning that rollover risk could be magnified by strategies that replace long-term debt with short-term debt on a large scale.
‘Circular’ AI Funding
The IMF sounded the alarm over the web of “circular” financing offers among AI companies.
Since 2023, 14 publicly announced transactions channeled $300 billion in equity, led by hyperscalers, to weaker AI companies, along with $650 billion in debt. In exchange, investors get $1.4 trillion in “future revenue commitments” through services or chip orders, the IMF said.
The revenue promises set the deal apart, Jason Wu, assistant director of the IMF’s monetary and capital markets department, said at a press conference on Tuesday. “What makes this a bit unusual compared to previous deals is the future revenue commitment, hence the circularity.”
If commercialization of AI is low, losses can come together. “If there is any question about monetization in AI, equity investment will decrease in value, debt may be difficult to service, and future revenue commitments may not be met,” Wu said, explaining what he calls a “triple whammy.”
Heavy funding and revenue like that is promising for groups in some AI companies, the report said. Because of this concentration, “there are certain companies and nodes that stand out in particular,” Wu said. “If there are some problems with these entities, then circularity will lead to market risk.”