Term Premium Is Back: Why the 10-Year Hasn't Rallied

The Fed has been on hold for two meetings. The March 2026 dot plot median implies roughly two 25-basis-point cuts between now and year-end. And yet the 10-year Treasury yield has barely moved, sitting near 4.45% at the April 21 close per FRED’s DGS10 series, essentially unchanged from where it started the year.

If you anchor your duration view on the dot plot, this does not compute. Forward rate expectations should have pulled the long end lower. They have not.

The answer is in a series most retail investors have never heard of: the New York Fed’s ACM term premium model. And the story it is telling is the single most important input for how a 60/40 portfolio should think about duration in 2026.

What Term Premium Actually Is

A 10-year Treasury yield has two components. The first is the expected path of short rates over the next ten years: roughly, the average of where the Fed funds rate is expected to go. The second is the extra compensation investors demand for the risk of holding a long bond instead of rolling short-term paper. That second component is the term premium.

When term premium is positive, long-dated bonds are paying investors extra yield to take on duration risk. When term premium is negative, which it was for most of 2011 through 2022, the market is so hungry for safe long-dated paper that it accepts less yield than the expected-short-rate path alone would imply.

The ACM model, built by Tobias Adrian, Richard Crump, and Emanuel Moench at the New York Fed, decomposes the Treasury curve into these two components on a daily basis. It is the Fed’s house decomposition, cited directly by Fed governors including Christopher Waller in his December 2024 speech on balance sheet policy.

As of the most recent ACM update, the 10-year term premium is positive. That is a regime shift from the deeply negative readings of the 2020-2021 quantitative easing era.

Why the Regime Has Flipped

Three structural forces explain why term premium has rotated from deeply negative to positive.

Supply. Treasury net issuance remains high. The federal deficit, combined with the runoff of the Fed’s balance sheet, means the private sector must absorb a rising share of long-dated Treasury paper. SIFMA’s Treasury statistics page tracks the aggregate picture, and the Treasury’s Monthly Statement breaks down the fiscal flow side. When supply outpaces the bid, the bid asks for more compensation. Basic economics.

Demand. Foreign central bank demand for Treasuries has plateaued. Total foreign holdings, per Treasury’s TIC system, have grown much more slowly than nominal Treasury supply since 2014. Post-2023, domestic bank demand for duration has also been muted. The regional bank stress of 2023 made every bank treasury department more cautious about held-to-maturity portfolio risk, as visible in the Fed’s H.8 report.

Risk. The fiscal term structure looks different than it did five years ago. Investors holding a 10-year Treasury today are taking a view on where fiscal policy, inflation, and monetary policy will be in 2036. That is a genuinely harder bet than the same view was in 2018. When uncertainty rises, risk-neutral investors demand more compensation. The 2013 “taper tantrum” is the clean historical example: term premium spiked roughly 100 basis points in four months even though the Fed had not actually tightened.

What This Means for Forward Rate Expectations

The practical implication: if you are watching the Fed dots to forecast the 10-year yield, you are using a model that worked in 2015 and does not work in 2026.

In a low-term-premium regime, the 10-year yield is mechanically pulled toward the expected short-rate path. Cuts priced in → long yields fall.

In a positive-term-premium regime, the two components can move in opposite directions. The Fed can cut and term premium can rise at the same time, leaving the 10-year yield roughly where it started. This is arguably what has happened in 2026: the expected short-rate path has moved down (two cuts now priced), and term premium has moved up (supply-demand shift), with the net effect close to zero.

This is not a prediction. It is a description of the decomposition. The forward direction of the 10-year depends on which component wins from here.

The 2s10s Curve Story

You can see this at work in the 2s10s spread. Per FRED’s T10Y2Y series, the curve has been positively sloped for the full year, a change from the inverted regime of 2022-2024. A steepening curve in a pause-to-cut cycle is what most Fed-watchers expect. But the steepener this time is being driven less by the short end and more by the long end refusing to rally. Term premium, not expected rates, is the marginal driver.

Fixed-income investors who anchor duration decisions solely to Fed dots are fighting the wrong model. The better question in this regime: where is term premium going, and why?

Portfolio Implications

For a classic 60/40 portfolio, the duration exposure sits almost entirely in the 40. The Bloomberg U.S. Aggregate Bond Index carries a duration of roughly six years. A 40% allocation to AGG contributes about 2.4 years of portfolio duration. Every 10 basis point move in 10-year yields translates to roughly a 0.24% move in the bond sleeve.

That is not a catastrophic number. But it is the number to watch if term premium expands further. A 50-basis-point move in term premium alone, the kind of move the 2013 taper tantrum produced in four months, would generate a 3% drawdown in the bond sleeve, which equates to a 1.2% drag on the 60/40. Not the end of the world. Not nothing either.

Three framings we think are useful for sophisticated readers:

First, duration is not a free lunch in this regime. The reflexive “bonds will rally when the Fed cuts” framework assumes a stable term premium. That assumption has not held in 2026.

Second, the short end is doing more work. Historical pause-to-cut cycles where term premium was well-behaved saw long bonds carry the rally. This cycle, if term premium stays elevated, the front end of the curve will capture most of the benefit from cuts. That argues for shorter-duration Treasury exposure, not longer.

Third, watch the refunding announcements. Treasury’s quarterly refunding statements tell you how much long-duration paper will hit the market over the next three months. The next announcement is May 7, 2026. If the mix shifts toward more long-dated issuance, expect term premium pressure to intensify.

The Counterarguments We Respect

Term premium models are not gospel. The ACM decomposition and the Kim-Wright decomposition (the Federal Reserve Board’s alternative model) do not always agree. Point estimates from any single model should be treated as a range, not a number. Both the NY Fed ACM series and the Federal Reserve Board’s Kim-Wright estimates should be triangulated for serious analysis.

Forward rate expectations are also still the dominant driver in normal cycles. The argument here is not that term premium has replaced rate expectations as the story of the long end. The argument is that in an issuance-heavy, foreign-demand-plateaued regime, term premium has moved from being a residual to being a first-order driver. The two components are finally both doing meaningful work, and investors who watch only one are missing half the picture.

Inflation expectations matter too. 5-year, 5-year forward breakeven inflation rates, the market’s forecast of average inflation from five to ten years out, sit around 2.3% to 2.5%, close to the Fed’s target. That is important context: the term premium story is not being driven by runaway inflation expectations. It is being driven by supply, demand, and uncertainty about the path, not by a wholesale loss of faith in the 2% inflation anchor.

What We’re Watching

Three markers will tell us whether the term premium regime is the new normal or a transient adjustment.

The May 7 refunding announcement. Issuance mix and Treasury Borrowing Advisory Committee commentary are the most direct signals on supply.

Foreign demand at auction. The indirect bidder share at 10-year and 30-year auctions (we’ll be covering auction mechanics in a separate piece this week) is the cleanest real-time read on whether foreign demand is stabilizing or continuing to fade.

The ACM series itself. The NY Fed updates it daily. If term premium drifts back toward zero, the story reverts to the old model. If it keeps climbing, the regime shift is durable.

Closing Thought

The most cited piece of institutional commentary in 2026 has been “the bond market doesn’t believe the Fed.” That framing is lazy. The bond market, per the ACM decomposition, does believe the Fed. Expected short rates are tracking the dot plot reasonably well. What the bond market is telling you is that the long-duration premium has gone from negative to positive, and the cuts that are priced in are being offset by investors finally demanding compensation for duration risk.

That is a different story. It is also the right one to base a 2026 duration view on, in our view.

---

Ferrante Capital LLC is a registered investment adviser. Information presented is for educational purposes only and does not constitute investment advice, a solicitation, or a recommendation to buy or sell any security. All investing involves risk, including the possible loss of principal.

FC and its principals may hold positions in securities or asset classes discussed in this article. This analysis is for educational purposes only and does not constitute a recommendation to buy, sell, or hold any security.

Forward-looking statements reflect Ferrante Capital’s current analysis and involve assumptions and estimates. Actual results may differ materially. Past performance is not indicative of future results.

Please consult a qualified financial professional before making investment decisions.