A test of nerves: The new anatomy of government debt

The kindness of strangers

The ‘marketable’ US government bond (Treasury) market amounts to $31 trillion,² the largest amount of debt ever amassed in history. One of the cornerstones of US economic policy has been the nation’s ability to borrow at attractive rates – part of its so-called ‘exorbitant privilege’. For lenders, the fact that Treasuries are backed by the US government and therefore considered almost risk free has led to their position as the world’s largest and most liquid sovereign bond market. 

At the time of the Global Financial Crisis (GFC), foreign and private ownership of Treasuries was dominated by China and Japan, and their holdings formed a portion of their foreign exchange reserves. Since then, the share of Treasuries held by foreign official investors has declined steadily, with China’s reported Treasury holdings falling by about $400 billion between 2011 and 2024, even as the Treasury market grew.³ 

In contrast, the holdings of foreign private investors have increased. By mid-2025, foreign private holdings stood at about $7 trillion, outstripping the $3.9 trillion held by foreign institutions.⁴ Hedge funds alone now control 8.5% of the entire Treasury market, making them bigger players than Japan, China and Saudi Arabia combined.⁵

The US is not alone. By the end of 2024, NBFI assets reached $260 trillion globally.⁶ NBFIs have increasingly become a key source of government financing in the 30 largest economies as central banks have scaled back sovereign bond purchases. 

The anatomy of government debt markets

Why does this matter?

Ultimately, the change in ownership makes the market more uncertain. The US has “less of a safe-haven status by default if its holders’ focus is on value and they are more likely to switch at a dime”, suggests Purvi Harlalka, senior emerging market sovereign debt strategist. She adds that “the private sector is going after profits, where the public sector is more invested in stability and security.” This creates an entirely different environment and a Treasury market that is more vulnerable to shifts in global risk appetite and geopolitical downturns. 

Historically, one major appeal of Treasuries lay in their counter-cyclical nature: when other markets fell, investors flocked to Treasuries which drove their yields down. For example, during the GFC, the 10-year Treasury yield fell to 2.25% in December 2008 as investors sought the safety net provided by a stable sovereign.

However, more recent crises and turbulent episodes have seen Treasury yields spike. We first saw the breakdown of this traditional dynamic in 2020. As the COVID crisis intensified in March 2020, Treasury yields spiked sharply, a move which can be largely attributed to a negative demand shock, rather than traditional causes such as increased default risks, inflation and illiquidity. 

The cause was largely the sale of Treasuries by foreign institutions, mutual funds and the household sector (including hedge funds). These sales amounted to $287 billion, $266 billion and $196 billion for each group respectively in the first quarter of 2020.⁷ 

The effect of this sale was quickly suppressed by Federal Reserve (Fed) Treasury purchases. However, this ‘dash for cash’ episode is emblematic of how an unwinding of Treasuries could occur in a crisis. 

The Centre for Economic Policy Research (CEPR) has highlighted the persistent rise of US net external liabilities, as well as the country’s unsustainable fiscal trajectory, the gradual loss of its exorbitant privilege and its dependence on private capital inflows intermediated by NBFI as a key risk to global financial stability.⁸ 

Harlalka highlights the risk that as the private sector is invested in many areas of the market, a loss in one part of their portfolio could cause them to jettison the most liquid part of their portfolio – US Treasuries. She adds that the risk is exacerbated as these positions are often leveraged and can therefore be vulnerable to large forced unwinding. “This could destabilise markets far more quickly,” she concludes. 

Rollover risk

The overall government debt level and shifting ownership of the debt are just parts of the problem. Growing interest costs are adding to the burden, with the US now spending over $1 trillion a year to pay interest on its debt, triple the figure in 2020.⁹ Harlalka notes that as interest payments consume a larger portion of a government’s revenues, the risk of a fiscal crisis increases.

Governments are responding by shifting to issuing shorter dated maturities. “It’s classic – when a sovereign with high interest revenue ratio runs into trouble, its go-to tactic is to move to the shorter end – it’s a stop gap, not a long-term solution,” Harlalka says. She adds that having to rollover debt more frequently can cause a government to run into trouble if it finds itself unable to find buyers. 

Historically it has been an emerging market (EM)-style problem to have to issue at the short-end to keep the show on the road, while “a hallmark of developed markets (DM) has been the ability to issue long-dated debt at low rates because investors have confidence in governments' ability to manage the economy over the long-term”.
 

Imbalanced and unstable

A cornerstone of developed market policy is the continued appetite for global government bonds. If flows were to suddenly reverse, the situation would be exacerbated by the substantial imbalances upon which financial stability currently rests. Data from the International Monetary Fund shows that just four economies, the US, China, Germany and Japan, account for roughly two-thirds of global imbalances.¹⁰

According to Mark Carney, the former governor of the Bank of England, “From the Lawson boom to the Tequila and Global Financial Crises, a common lesson is that large current account deficits are one of the most trenchant early warning signs of financial instability.”¹¹

He likened today’s situation to the Great Depression of the 1930s when monetary policy was exhausted and fiscal policy was in abeyance. Meanwhile, the CEPR noted that global imbalances often end in financial crises.

Harlalka points out that  when there is a period of adjustment, the burden often falls to deficit countries. Given the largest deficits are concentrated in some core economies, such as the US and the UK, markets could find themselves in unchartered territory.

Risks or opportunities?

Markets are not complacent about these risks. This year, government bond yields have been rising globally at the long end, with more term premium required. Fed policymakers have commented on this change, noting that the shift from relatively price-insensitive official-sector holders to more price-sensitive private investors “could have implications for the term premium component of yields”.¹² Average bond yields across Group of Seven (G7) countries hit their highest level since 2008 in mid-August 2026.¹³

According to Eva Sun-Wai, fund manager of the Global Government Bond Strategy, the price dynamics reflect myriad considerations: “There are so many different factors in the valuation of a government bond. Investors do not put a price on fiscal risk alone, but they need a premium for political risk, geopolitical risk, for term premium, for supply risk and for liquidity risk.” However, she notes, fiscal risk is becoming an increasingly important element and this is reflected in the elevated term premium. “Bond deals look more interesting now that curves are offering some roll and some carry as the steepness has come back into the market.” This marks a change from the 2021/2022 period when yield curves were inverted  and investors couldn’t take advantage of tools such as roll-down, which is the return generated as a bond moves down the yield curve over time. 

On a typical, upward sloping curve, longer dated bonds offer higher yields because they embed more term premium. For example, a 10 year bond has historically yielded more than a seven year bond. After three years, that 10 year bond effectively becomes a seven year bond. If market conditions are unchanged, its yield could ‘roll down’ to the lower yield associated with seven year maturity. This decline in yield results in a higher bond price. As a result, the investor may earn not only the coupon income but also a capital gain from the roll down effect. 

Where is the diversification?

With global risks rising and traditional safe havens being called into question, investors are inevitably looking for a new source of relative safety. Harlalka highlights that the recent multi-year rise in gold prices speaks to investors’ search for alternatives. “As a finite resource, it is stable but as an asset, it yields nothing – you are banking entirely on price appreciation”, she notes. In her view, “when there is no alternative safe haven, diversification is the safe haven.”

Sun-Wai highlights the opportunity to look across government bond markets for relative value. She points to Australia, where long-dated government bond yields are higher than those in the US and France despite Australia’s stronger credit rating and fiscal position. She also highlights countries such as Norway and New Zealand which, while differing in their own fiscal metrics, generally have lower debt burdens and stronger fiscal profiles than more indebted developed markets. Yet, their government bond yields remain broadly similar. From a fiscal-risk perspective, these markets appear more attractive on a cross market basis. However, she adds “Is 5% ultimately good value for a bond? That’s a very different question and a much harder one to answer.”

This divergence in fiscal scenarios is not confined to developed markets. As a whole, emerging market (EM) debt has had to contend with non safe-haven status for longer, Harlalka points out. As a result, EM authorities have run policy accordingly and have strengthened their fiscal and monetary frameworks over the last decade.

Sun-Wai argues that historically EM debt was held by ‘faster money’, resulting in volatility in the instance of a risk-off event. But this is changing; the market is becoming less flighty as investors come to understand just how much diversification EM can offer. She adds that EM’s position is aided by its increased institutional credibility, its relative fiscal health and the responsiveness of central banks to crises, inflation, political and geopolitical events. “Investors are becoming more comfortable with the breadth and depth of EM debt as an asset class,” she notes. 

The corporate conundrum

If sovereigns are issuing vast quantities of debt at higher yields, does that leave corporate issuers vying for buyers? Despite the resurgence of the ‘higher-for-longer’ narrative, spreads on corporate bonds haven’t significantly widened, as traditional market patterns might have suggested they would. In part, Harlalka argues, this reflects “confidence in the fundamentals of higher yielding markets”. 

Sun-Wai notes that in some cases, individual corporate bonds can trade inside their sovereign equivalent, reflecting issuer-specific strengths or market dynamics. However, corporate bonds would generally be expected to compensate investors for additional risks, including default and liquidity. As she points out, “the mechanics of bonds mean that corporate debt remains at the mercy of the underlying government bond”: if the interest burden rises for the sovereign, it is likely to rise for companies as well. 

Harlalka points to the new dynamic entailed in the massive amount of issuance by AI-related companies. She notes that “when investment grade-rated tech companies’ capital needs become so high over time that they are issuing multi-billion dollar deals every few weeks, they could start to compete with the sovereign market.” Hyperscalers alone are expected to issue $317 billion of debt in 2026.¹⁴

The key question is whether investors are being sufficiently compensated for the risks they are taking. Rising debt levels alone do not cause crises, but how those debt levels are financed, their associated interest costs and the credibility of the policymaking that shapes their backdrop are instructive of the likelihood of a crisis. As those factors increasingly diverge across countries, diversification and relative value are likely to become increasingly important tools.

¹ Visual Capitalist, ‘The State of World Debt 2025’, (visualcapitalist.com), November 2025.
² Bipartisan Policy Centre, ‘Bond market tracker’, (bipartisanpolicy.org), July 2026.
³ Brookings, ‘Who’s buying US treasury debt and why?’, (Brookings.edu), August 2026.
⁴ Bipartisan Policy Centre, ‘Bond market tracker’, (bipartisanpolicy.org), July 2026.
⁵ Federal Reserve, ‘Decomposing hedge funds’ US treasury exposures’, (Federalreserve.gov), June 2026.
⁶ Financial Stability Board, ‘Global monitoring report on nonbank financial intermediation’, (fsb.org), December 2025.
⁷ Bank for International Settlements, ‘The Treasury market in spring 2020 and the response of the Federal Reserve’, (bis.org), October 2021.
⁸ Centre for Economic Policy Research, ‘The new global imbalances’, (cepr.org), April 2026.
⁹ Council for Foreign Relations, ‘The national debt hit $40 trillion’, (cfr.org), August 2026.
¹⁰ International Monetary Fund, ‘Understanding global imbalances’, (imf.org), April 2026.
¹¹ Bank of England, ‘A fine balance’, (bankofengland.co.uk), June 2017.
¹² Federal Reserve, ‘Minutes of the Federal Open Market Committee’, (federalreserve.gov), June 2026.
¹³ Bloomberg, ‘Why high yields on government bonds are causing alarm’, (Bloomberg.com), September 2026.
¹⁴ Bloomberg, ‘JP Morgan boosts tech bond sales outlook as AI debt binge expands’, (Bloomberg.com), August 2026.