Invest1 publisher2 min readPublished
Three weeks put 27.4 basis points on the 10-year Treasury yield
The 10-year closed the week at 4.995%, eight tenths of a basis point under the 52-week high it set on Wednesday. About 28% of the yield's entire 2026 rise arrived in September, and Friday supplied 0.049 point of it.
The Investor · Invest desk
What happened
- The 10-year Treasury yield rose 0.021 percentage point this week to close at 4.995%, on 3 p.m. ET values from Tradeweb FTSE U.S. Treasury Closing Prices.
- The close sits 0.008 percentage point below the 52-week high of 5.003% set on Wednesday, September 16, and is the second-highest reading of the year.
- The yield is up 1.043 percentage points from its 52-week low of 3.952% on October 22, 2025, and 0.843 point since the start of 2026.
Compiled by The InvestorSomething wrong?How this is made
Why it matters
- cost Owners of long-dated cash flows absorb 9.5% of present value at ten years and 25.9% at thirty, with no change in the cash flows themselves, purely from the move off October's low.
- constraint September supplied about 28% of this year's rise, so hurdle rates set in January against roughly 4.152% now sit 0.843 point under the market and have to be rebuilt, not adjusted.
- decision The week netted 0.021 point while Friday alone moved 0.049, so a model's discount rate depends on which session's close it samples, and that choice is now worth more than the whole week.
A cash flow ten years out, discounted at 4.995% instead of 3.952%, loses 9.5% of its present value [1][12][6]. At thirty years the loss is 25.9% [7]. The 10-year has covered that distance since October 22, 2025 [12].
Most of the move is recent. Of the 0.843 percentage point the yield has added in 2026 [16], 0.238 came this month [15], about 28% of the year in three weeks of trading [4], and the three-week climb of 0.274 point is 32.5% of the year to date [4][5]. The Dow Jones Data Talk release calls it the largest three-week gain since the week ending May 15, 2026 [5].
The week itself was quieter than the close suggests. Net, the yield rose 0.021 point [1]; Friday alone added 0.049 [7], its largest one-day gain since September 10 [8], which puts the yield down roughly 0.028 point through Thursday before Friday took it back [8]. In price, the note finished at 97 4/32, down 5/32 for the week and 12/32 on the day [2][7], so Thursday's close was 97 16/32 [9].
Wednesday had already printed 5.003%, and Friday settled 0.008 point under it [11]. The yield has been up nine of the past twelve weeks and eight of the past ten sessions [6][9]. Anyone marking daily has had this level for two days, Friday's session covered six times the remaining gap to the high [11], and the present value of a long-dated cash flow moves smoothly through 5.00% with nothing mechanical happening at the round number. The release is one price series sampled at 3 p.m. ET from Tradeweb FTSE closing prices [17], and it does not report what any allocator did with it.
If the yield breaks 5.003%, January's discount rate of about 4.152% [1] is 0.843 point stale [16] for anything still modelled off it. The alternative is that September was a supply-and-data artifact and the 0.274 point unwinds inside a range that has already spanned 1.043 points since October [12][4]. The third case, sideways near 5%, is the one that changes capital allocation, because a hurdle rate that stops moving removes the reason to wait.
In my view the second case is the weakest while the series keeps closing higher eight days in ten [9]. The level I would hold onto is 4.721%, where the yield stood three weeks ago [3]: a close back under that erases the climb.
What to watch
- A close above 5.003% would set a new 52-week high and take the yield past Wednesday's mark.
- Whether next week's data extends the streak to a fourth week or starts giving back the 0.274 point.
- Whether another 0.049 point session appears, since that was the largest one-day gain since September 10.