September 10, 2026 · last fitted close

Treasury rates

Key rates from one month to thirty years, fitted from Treasury's own end-of-day prices. The ten-year zero rate is 4.953%, up 12.2 bp on the session.

Weekends and holidays have no fit. The nearest earlier trading day is shown instead.

Treasury rates by maturity on September 10, 2026, across the zero, par and money market curves, with the change from the previous fitted trading day in basis points.
MaturityZero couponPar yieldMoney marketReal (TIPS)Breakeven
Rate1dRate1dRate1dRate1dRate1d
1 month3.790%+4.8
3 month3.939%+4.5
6 month4.084%+6.8
1 year4.301%+10.74.347%+11.14.229%+11.7
2 year4.533%+13.84.584%+14.22.157%+6.42.377%+7.4
3 year4.633%+14.14.681%+14.32.141%+8.02.492%+6.2
5 year4.706%+13.14.749%+13.22.217%+9.42.489%+3.7
7 year4.776%+12.44.817%+12.92.359%+9.62.417%+2.8
10 year4.953%+12.24.978%+13.12.587%+9.12.366%+3.2
20 year5.484%+9.75.423%+10.23.055%+7.22.429%+2.4
30 year5.380%+7.05.367%+9.53.122%+6.9

A dash means that curve does not cover that maturity, which is a statement rather than missing data. The zero and par curves are fitted from coupon securities between one and thirty years; the money market curve is fitted from bills and stops at one year. Neither is extrapolated into the other's range, the front end is exactly where extending a coupon curve goes most wrong, which is why the money market curve exists separately. The 1d columns compare against September 9, 2026, the previous fitted trading day, so they are one-session moves rather than calendar-day ones.

3.51 bp root-mean-square fit error across 339 priced securities. Within the normal range for this family, which is flagged above 7.5 bp, its 95th percentile across the history.

Savings bonds

Treasury sets these on May 1 and November 1. They are not market rates: a savings bond cannot be traded, cannot be redeemed for twelve months, and forfeits its last three months of interest if cashed inside five years.

Series I

4.26%

Composite rate, for bonds issued from May 1, 2026

Fixed rate, for its life
0.90%
Inflation rate, six months
1.67%

The composite is not the sum of the two. It isfixed + 2×infl + fixed×infl, because the inflation rate is semiannual and compounds against the fixed rate.

Series EE

2.40%

Fixed for twenty years, for bonds issued from May 1, 2026

An EE bond is also guaranteed to double in value at twenty years, whatever its stated rate implies. At 2.40% the guarantee is worth more than the coupon, so the doubling is the real return.

Doubling over twenty years is 3.526% compounded annually. That is the number the stated rate has to beat.

How to read the breakeven column

Breakeven is the nominal zero rate less the real one at the same maturity: the inflation rate at which holding a nominal Treasury and holding an inflation-linked one would return the same amount. Above it the linker wins, below it the nominal does.

The subtraction is exact rather than an approximation, and only because both curves publish continuously compounded rates. On annually compounded yields the correct form is (1+n)/(1+r) − 1 and the plain difference drifts as rates rise.

It is not expected inflation, and nothing here calls it a forecast. It is expected inflation plus a premium for bearing inflation risk, less a liquidity premium on the linker, and that second term is largest exactly when a reader would most want to trust the number: inflation-linked securities traded far below fair value in late 2008, so breakevens collapsed further than any forecast did. Both premiums are unobservable, so separating them is a modeling exercise this does not attempt.

Six maturities, two to twenty years, because those are the ones both curves fit. A breakeven at a tenor where either side was not fitted would be a plausible figure whose error is invisible, so it is left out rather than interpolated. Only the 5 and 10 year points have an external series to check against, since those are the two the Federal Reserve publishes; the other four are computed identically and verified against nothing.

Index levels before September 30, 2026 are back-tested. They were computed after the fact by applying the rules to historical data, which benefits from hindsight in the choice of rules, and an index cannot be invested in directly. Methodology v1.0 takes effect at that rebalance, when levels begin to be struck on the day; the rulebook is identical either way, and the version is published on every row.

A fitted curve is a fit, not a quote. Daily error averages 3.8 basis points across the history and reaches about 20 on the worst days, in December 2008, when the market was genuinely hard to fit one smooth curve to. Every curve page publishes its own fit error rather than burying it.

Free to benchmark against. Paid only to track. Measuring anything against these curves and indices is free. No license, no fee, no registration, and no permission needed to say that you did, including in a prospectus. A fee applies to one thing: launching a product that tracks an index.

Safe Rate™, Safe Rate Indices™ and the Safe Rate US Treasury Index™ are trademarks of Safe Rate, claimed through use in commerce and not registered. Third-party marks are the property of their owners, which are not affiliated with and do not endorse this data.

Curves are fitted from public Treasury data and carry fit error; figures are not a record of trading, and an index cannot be invested in directly. No claim of compliance with the IOSCO Principles for Financial Benchmarks is made or implied. Not investment advice, not an offer, and not a recommendation to buy or sell any security.