10-Year Yield at 4.95%: Level, Slope, and That 20-Point Gap
The 10-year U.S. Treasury yield was 4.95% on September 10, 2026, according to FRED. Yield here means the yearly return for holding government debt to maturity.
The 5-year yield was 4.75% on the same trading date, per FRED. Together they imply a 5s10s spread of 20 basis points. A basis point is one-hundredth of a percentage point, so 20 points equals 0.20%. The longer tenor yielded more than the intermediate tenor.
FRED listed September 14, 2026 as the next release date for the daily 10-year series after its September 11, 2026 update. That dating sets the vintage of the observations. It does not revise the September 10 levels themselves.
The broader context here is what constant maturity means in practice. The series is not the transaction price of one specific bond, identified by a single CUSIP. It is a par curve reading, interpolated to fixed tenors to estimate a fresh 5-year and 10-year yield each day. That is why portfolio managers, risk officers and corporate treasurers reference it for benchmarking and attribution. It standardizes tenor. It allows clean tenor-to-tenor comparison across dates.
In my view, level and slope should be read separately. The level sets the discount anchor for long-duration cash flows. Small moves matter because duration is high and convexity builds with maturity, so present value shifts quickly. The slope isolates compensation between the belly and the long end. A positive 5s10s spread indicates additional yield for extending from five to ten years. That compensation bundles expectations for short-rate paths, term premium, supply absorption across tenors, and convexity demand. No single print disentangles those drivers.
Looking at what this means for books and balance sheets, the distinction matters. Rate hedgers calibrate DV01, the dollar change for a tiny rate move, by tenor bucket rather than by headline yield alone. Liability-driven portfolios map long-dated obligations against the long-end anchor. Issuers weigh tenor selection against all-in funding cost and refinancing risk. Mortgage and structured-product desks translate intermediate and long-end moves into prepayment and extension assumptions through their own models. The September 10 prints give each of those functions a common mark. They do not dictate positioning.
Looking at what this means for data discipline, process matters more than narrative. Daily constant-maturity data are point-in-time marks subject to release lag. Intraday price action can differ from the fixing. Revision methodology, holiday calendars and publication timing all affect vintage control. For that reason, the September 14 release date functions as a version marker. It tells users when to refresh models and attribution, not how to interpret direction. One observation establishes level. It does not establish momentum. Trend requires a sequence.
