Merchants work on the ground of the American Inventory Alternate (AMEX) on the New York Inventory Alternate (NYSE) in New York, US, on Monday, March 9, 2026.
Michael Nagle | Bloomberg | Getty Photographs
BANGKOK – The Worldwide Financial Fund warned Tuesday that authorities bond yields may rise in a disorderly style, as rising inflation, mounting debt and AI-related borrowings strain sovereign bond markets.
Government bond yields in several advanced economies have climbed to multidecade highs, pushed by greater vitality prices because of the Center East battle and expectations for tighter financial circumstances as central banks look to curb resurgent inflation.
In its newest World Monetary Stability Report, the fund warned that sovereign bond danger may deteriorate, warning of “a pointy rise in sovereign bond yields.” The report was launched Tuesday as a part of the IMF-World Financial institution annual conferences in Bangkok, Thailand, this week, the place greater than 18,000 delegates gathered to debate the state of the world economic system and the monetary system.
Excessive debt load and rollover dangers make the bond market “inclined to extra frequent fiscal and inflation shocks,” the report stated. The IMF had warned final week that world public debt had reached close to its highest level since World War II and was on observe to quickly exceed 100% of GDP.
“Influence might be extreme given the foundational significance of sovereign bond markets,” the IMF stated Tuesday. A pointy rise in yields would mark down asset values via greater low cost charges, together with for AI-related belongings, and erode the worth of presidency bonds posted as collateral.
IMF additionally famous the exponentially rising borrowing by AI corporations — that compete with sovereign bonds for investor demand — as an extra supply of pressure on bond markets. Hyperscalers’ share of worldwide investment-grade bond issuance has risen to 10% from lower than 1% since 2024, the report stated.
The worldwide AI buildout has seen tech giants, equivalent to Alphabet Inc’s Google, Meta Platforms, hyperscalers and AI labs elevating hundreds of billions of dollars in debt this 12 months.
“AI-linked issuance – a lot of it longer-dated – has grown massive sufficient to compete with government bonds for investor capital, even when it doesn’t seem like the dominant driver of yields,” in line with Marc Seidner, CIO of non-traditional methods at Pacific Funding Administration.
Hedge fund worries
Hedge funds’ rising holdings of Treasuries depart the market “inclined to a disorderly unwind,” the IMF stated. Leveraged foundation trades have partly unwound with out notable strain within the short-term mortgage market however the leverage that is still is a priority.
Hedge funds’ share of U.S. Treasuries rose to about 9% in 2025 from about 4% in 2022, largely via extremely arbitraged methods, the IMF stated, citing information from the Federal Reserve.
Caio Ferreira, IMF deputy division chief for monetary supervision and regulation, stated the concern stems from hedge funds’ rising leverage and their more and more overlapping positions, which means a number of funds are holding the identical trades, doubtlessly amplifying impression from a collective unwinding.
On the intense facet, liquidity has held up within the U.S., Europe and the U.Okay., the IMF stated. Japan is the exception: liquidity in its authorities bond market has “continued to deteriorate” as yield volatility has risen for the reason that finish of yield curve management.
The fund urged debt managers to elongate maturities, warning that rollover danger might be exacerbated by methods that substitute long-term debt with short-term borrowing on a big scale.
‘Round’ AI financing
The IMF sounded the alarm over an internet of “round” financing offers amongst AI corporations.
Since 2023, 14 publicly disclosed offers channeled about $300 billion of fairness, led by hyperscalers, into financially weaker AI corporations, alongside $650 billion of debt. In alternate, the traders secured $1.4 trillion of “future income commitments” via providers or chip orders, the IMF stated.
The income guarantees are what set the offers aside, Jason Wu, assistant director of the IMF financial and capital markets division, stated at a press briefing on Tuesday. “What makes this a bit of bit uncommon in comparison with earlier offers is the long run income commitments, therefore the circularity.”
If AI commercialization falls quick, the losses may arrive collectively. “If there are questions concerning the monetization in AI, the fairness investments will deplete in worth, the debt could also be troublesome to service, and the income commitments sooner or later could not materialize,” Wu stated, describing what he known as a “triple whammy.”
Such heavy financing and income guarantees cluster in a handful of AI corporations, the report stated. On account of that focus, “there are specific corporations and nodes which might be notably uncovered,” Wu stated. “If there are some sort of points with these entities, then the circularity would pose danger to the market.”
