One peril of writing a column entitled ‘On Global Markets’ is that those global markets have an annoying tendency to move about. Sovereign bond yields are a case in point. Over the last few months, the yield on government bonds has been heading up – sometimes at an alarming rate. Only for them to suddenly plunge this week as your columnist began writing this column. At one point this week, the yield on a US government ten-year Treasury was north of 5.3%, whereas at the time of writing it was 5.1%. Of course, even that lower number is materially higher than the circa 4.5% levels of the early summer or the 4.3% level of the early months of the year.
So while the AI-investment-thesis wobble may have sent AI-exposed stocks lower and driven something of a flight into bonds in recent days, the bigger picture remains that of a large rise in government borrowing costs over the course of 2026. What, though, sits behind that change?
That was the topic addressed this week by the Clark Center’s Finance Experts Panel in a couple of related questions. The questions focused on two closely related arguments for why yields have risen over the course of 2026 (or indeed over the course of the 2020s as a whole) but, in their more expansive comments, various panel members brought up a much wider set of factors.
Firstly, the panel was asked whether ‘deteriorating fiscal outlooks have contributed substantially to the recent rise in long-term government bond yields across advanced economies’? This is one of the more common factors mentioned in the general press, and fits into an easy-to-understand and follow logic. Clearly, if the financial situation of any borrower is deteriorating, then one might expect anyone lending to that borrower to demand a higher interest rate to compensate them for the increased risk of their loan not being paid back.
Weighted by confidence, 31% of respondents strongly agreed with the statement, and another 50% agreed. But while that headline number of 80% agreement is useful, many of those agreeing were keen to point out this was not the only thing in play.
Secondly, the panel was asked whether ‘a substantial portion of the increase in yields is caused by the increased supply of government bonds, even after controlling for inflation and sovereign default risk’?
This requires some unpacking. Obviously, an increased supply of government bonds can be seen as part of a fiscal deterioration, but the panel was asked to leave aside concerns over default risk. Similarly, an increased supply of government bonds is suggestive of a widening government deficit which, all things being equal, would likely suggest an increase in economic demand possibly adding to inflationary pressure and again, prompting markets to require a higher return to compensate them for the increased risk that price rises will swallow up their returns.
The panel was asked to put these concerns to one side and focus on the simple dynamics of bond market supply and demand. In other words, is the rise in yields – at least partially – simply a case of the supply of government debt outstripping market demand for it?
Weighted again by confidence, 57% of respondents agreed, and 6% strongly agreed. Again, a fairly conclusive result but, also again, one that many panellists were keen to clarify in their comments.
As John Cochrane of the Hoover Institution, Stanford, noted in his response to the first question – with refreshing straightforwardness – ‘Face it, we don’t really know’.
When it came to specific worries about fiscal sustainability, more than half a dozen respondents specifically pointed out France as an example of a major economy where market concerns about the public finances were driving yields up. In the case of the United States and many other advanced economies, the panel was more divided. As Arvind Krishnamurthy of Stanford argued, ‘It’s more about markets requiring higher yields to accept more supply than default risk per se. The significant effect is that rates on both public and private debt are up. A second effect is that the convenience yield on public debt is down, pushing up public more’. Of course worries about never-ending deficits may well be putting upward pressure on yields, but that is not necessarily the dominant factor behind recent moves. As the Clark Center’s own Anil Kashyap argued, fiscal worries are ‘one of several factors’ at work.
What are those other factors? Several respondents mentioned the ongoing AI investment boom. As Campbell R. Harvey of Duke Fuqua put it, ‘The supply of government bonds is highly predictable. The surge in AI-related and data center bonds was unexpected. Nominal bond yields are up because expected real rates have increased due to higher expected economic growth’. There are two distinct channels at work here. First, increasing demand for borrowing to fuel the AI boom is raising competition for funds, and governments are having to pay more as a result. Secondly, if one believes that the result of the AI boom will be faster economic growth in the future, that too might be expected to increase borrowing costs.
Another potential cause, as identified by Stijn Van Nieuwerburgh of Columbia Business School, is the changing nature of the buyers of sovereign bonds: ‘The supply is large, but even more importantly, the buyer composition has changed with fewer foreign and central bank purchases. Price sensitive domestic buyers require higher yields’. To which several other panellists noted could be added the fact that domestic central banks are far less active in these markets than was the case in the 2010s.
Plus, of course, one cannot ignore the Strait of Hormuz shaped elephant in the room. The supply disruptions to global energy markets, and the knock-on impact on global inflation dynamics, might be expected to have a large impact both directly through a faster pace of price rises and from the anticipation of higher central bank policy rates as a potential result.
Any large market move will be accompanied by a variety of explanations (some of which will be more plausible than others), but pinpointing which of those factors – if any – is the ultimate driver is rarely straightforward. This week, for example, US inflation data came in hotter than anticipated, which – normally – might have been expected to send yields higher, but yields instead fell as investors sought cover from the AI trade in the relative safety of government bonds. The lack of a simple explanation can seem frustrating to the layperson who simply wants to know why their mortgage offer has become pricier, but it is one of the things that keeps global markets interesting.
