What is the term premium and why has it been negative?
The term premium is the extra yield investors demand for holding long-duration bonds instead of rolling short-term ones. It compensates for the risk that interest rates rise unexpectedly during the holding period. From 2016 to 2021, the term premium turned negative, reflecting massive central bank balance sheet expansion and structural demand for safe duration. Its return to positive territory in 2022-2023 marked a regime shift.
In this article
The short answer
Imagine choosing between a 10-year Treasury yielding 4% and rolling 1-year Treasuries for 10 years. If short-term rates average 3.5% over that period, the rolling strategy returns less than the 10-year. The 50 bp difference is the term premium — compensation for committing capital and bearing rate volatility.
Theoretically, the term premium should be positive: investors prefer flexibility and need to be paid for giving it up. But empirically, it can compress to zero or even turn negative when massive demand for long bonds (from central banks, pension funds, foreign reserves) overwhelms the supply.
The Adrian-Crump-Moench (ACM) model decomposes the 10-year yield into expected short rates plus the term premium. From 2016 to 2021, this decomposition showed persistently negative term premia — investors were effectively paying to hold duration. That regime ended in 2022 as QT, inflation surprises and supply expansion reversed the trend.
→ New to bonds? Investing for beginners hub
What the data shows
NY Fed ACM model data on the 10-year term premium (1961-2024):
- The term premium averaged 150 bp from 1961 to 2008
- It compressed steadily during quantitative easing programs, turning negative in 2016
- Reached a trough near -100 bp in 2020 amid pandemic-era policy
- Returned to roughly 0 to +50 bp by late 2023 as quantitative tightening proceeded
The exception worth noting: term premium estimates are model-dependent. Different specifications (Kim-Wright, ACM) can produce term premium estimates differing by 50-100 bp at any given moment. The directional message is more robust than the precise level.
→ Dataset: 10y-3m yield curve spread
Why it happens — the macro mechanism
The term premium emerges from three structural forces.
Risk compensation for duration. Long bonds expose holders to inflation surprises and unexpected rate moves. The compensation required varies with inflation volatility regimes — high inflation uncertainty raises the term premium; stable inflation compresses it.
Central bank balance sheet operations. When central banks buy long bonds (QE), they remove duration from the market, compressing the term premium. The reverse (QT) restores it. QT and the term premium is a direct empirical link.
Structural demand from price-insensitive buyers. Pension funds, life insurers and foreign central banks demand long duration for liability matching or reserve management, often regardless of yield level. This demand floor compresses the term premium. Macro-financial regimes determine when this demand intensifies.
Synthesis by regime: in QE regimes the term premium compresses; in QT or inflation-surprise regimes it normalizes upward.
The term premium is the price of patience — what you must be paid to wait through uncertainty.
What it means for different economic actors
Savers rarely interact with term premium directly, but it affects every long-duration product they purchase, from 30-year mortgages to long-bond funds.
Investors use term premium signals to identify when long bonds are richly priced (negative premium) versus reasonably compensated (positive premium).
Central banks view the term premium as a transmission channel. Compressing it via QE eases financial conditions; allowing it to rise via QT tightens them.
A common error is treating term premium as a fixed parameter rather than a regime-dependent variable that reflects ongoing macro and monetary conditions.
Practical observation
What the data suggests for understanding your situation:
- Question to ask yourself: When I evaluate long bonds, am I considering term premium or only the headline yield?
- Data to monitor: The NY Fed ACM term premium estimate, updated monthly
- Historical parallel: The 1980s saw term premia above 400 bp; the 2010s saw them turn negative — a 600 bp swing
- What the literature documents: Adrian, Crump, Moench (2013); Kim & Wright (2005)
This is descriptive information to help you frame your own analysis. Eco3min does not provide investment advice.
Go deeper
📊 Full study: Yield curve inversion and credit channel
📁 Datasets: 10y-3m spread · 10-year Treasury yield
Questions liées
Frequently asked questions
The most widely cited model is the Adrian-Crump-Moench (ACM) approach, which uses a no-arbitrage affine term structure model to decompose Treasury yields. The model fits cross-sectional yield data daily and produces estimates of expected future short rates and the residual term premium. Other approaches (Kim-Wright, surveys, market-based) produce different but generally correlated estimates. The methodology is transparent but model-sensitive.
Empirical research strongly supports this. Studies by D’Amico and King (2013) and Krishnamurthy and Vissing-Jorgensen (2011) document that Federal Reserve bond purchases compressed term premia by 50-100 bp at peak. The mechanism operates through duration removal: by holding long bonds on its balance sheet, the Fed reduces the supply of duration available to private investors, who then accept lower term premia.
The evidence points to ongoing normalization, though the trajectory is uncertain. Continued QT reduces Fed holdings; rising US debt issuance increases duration supply; inflation volatility has structurally increased; foreign demand from China and Japan has plateaued. Empirical research suggests these factors collectively support a higher equilibrium term premium than the post-2009 era. However, episodic flight-to-quality and pension demand can produce temporary compression even within an upward trend.
Last updated — 5 May 2026
Disclaimer – Financial Information: The analyses, commentary, and content published on eco3min.fr are provided for informational and educational purposes only. They do not constitute investment advice or a solicitation to buy or sell financial instruments. Past performance is not indicative of future results. All investment decisions involve risk and are the sole responsibility of the reader.
