The 10-year Treasury yield exceeded 5.3 percent yesterday—its highest intraday level since 2002. It has risen by roughly half a percentage point this month. Given the scale of the US debt that needs refinancing and projections for ongoing annual federal deficits, such a surge in government borrowing costs could be a big deal for US interest spending levels and, ultimately, the path of the debt itself.
The debate is ongoing about why the federal government’s borrowing costs have risen so sharply of late. This phenomenon is seen across many countries, as Robin Brooks notes, which suggests common global forces at work. Perhaps long-term yields are climbing almost everywhere as markets price in the likelihood of central banks using higher rates to tame inflation? Yet the stress is most acute where public finances are weakest, especially in France. That Switzerland, whose constitutional “debt brake” has helped keep debt low relative to GDP, has seen much less of a spike hints that concerns about the debt outlook are a contributing factor, though it hardly proves that “bond vigilantes” are afoot.
Market-implied inflation expectations are flat (essentially all of September’s rise in the 10-year Treasury yield reflects higher real yields). That weighs against a sharp rise in expected inflation—including fears of debt monetization—as a core explanation for the recent increase. Higher real yields may instead reflect a combination of the current US AI investment boom or simply a bigger term premium as the supply of US bonds grows and fiscal and monetary outlooks are judged more uncertain.
Why High Bond Yields Pose a Threat to the Federal Budget
Fortunately, we don’t need to know precisely why yields have risen to see why interest-rate risk is a key fiscal vulnerability.
The federal government had accumulated $32.4 trillion of debt held by the public through Monday, up from $27.5 trillion in April 2024 when I published a report highlighting bond yield shocks as a major budget risk. The government must continuously refinance a large proportion of that accumulated debt. At the end of August, around one-third of publicly held marketable debt—more than $10 trillion—was scheduled to mature within the next 12 months. Over half matures within three years.
That doesn’t mean the Treasury needs $10 trillion of additional borrowing this next year. Most of that financing will replace maturing securities rather than fund new deficits. But that is precisely where interest-rate risk manifests: Maturing debt must be refinanced at whatever rates investors demand today. Previously “cheap” debt is becoming more expensive.
The average interest rate on outstanding marketable Treasury debt is currently just under 3.5 percent (Figure 2). If today’s much higher yields persist, then refinancing will push up the government’s average borrowing cost, with effects that build over time as more debt is rolled over. That would mean more interest spending even if Washington were merely replacing maturing debt. In reality, it keeps adding vast quantities of new debt too.
The Congressional Budget Office (CBO) expects a $2.1 trillion federal deficit for 2026, with deficits totaling $24.4 trillion over 2027–2036. As a result, debt held by the public is projected to rise from around 100 percent of GDP today to 120 percent by 2036 and 175 percent by 2056 (Figure 3). Remarkably, CBO projects these large deficits while expecting unemployment to remain below 5 percent throughout—in other words, assuming benign economic conditions and no recession.
Interest costs are consequently becoming a major budget item. CBO expects net interest spending to rise from around $1 trillion this year to $2.1 trillion in 2036. Even before the recent yield spike, rising interest costs accounted for more than the entire projected increase in the deficit—from 5.8 percent of GDP to 6.7 percent over that period (Figure 4).
How Much Worse Could Things Get if Yields Remain This High?
CBO’s February baseline assumes rates well below today’s. It has the three-month Treasury bill rate drifting down to around 3.1 percent by the 2030s, while the 10-year Treasury yield averages roughly 4.3–4.4 percent. At 5.3 percent today, the 10-year yield is already more than a percentage point above CBO’s projected 2026 average of 4.1 percent. The 2026 average up to September 29 was already 4.46 percent.
CBO’s own rules of thumb illustrate the potential fiscal consequences. If all Treasury rates, from three-month bills to 10-year notes, were just 0.1 percentage point higher each year than CBO assumes, deficits would be $379 billion larger over 2027–2036.
CBO says those estimates are roughly scalable for interest-rate deviations of up to one percentage point. If borrowing rates therefore averaged a percentage point above CBO’s baseline for the whole decade—think a 10-year yield averaging closer to 5.3 percent than 4.3 percent, with short-term rates a point higher too—cumulative borrowing would be about $3.8 trillion higher. For context, that’s around the same cost as the more optimistic forecasts of how much the One Big Beautiful Bill Act would add to deficits over a decade.
It’s true that if yields have risen partly because of an AI boom that fuels faster growth, then the net fiscal consequences could be much less troubling than any analysis of the increase in debt-service costs alone suggests. In that scenario, we’d also see higher incomes and tax revenues.
But the converse matters too. If borrowing costs stay higher without meaningful improvement in growth— because investors demand larger term premiums or compensation for uncertainty—then Washington faces a higher interest bill without the revenues needed to pay it.
That was the interest-rate risk I worried about in 2024. It has only become more salient since. The case for significant deficit reduction through spending cuts remains strong.
