Loading today’s stories
Invest1 distinct publisher3 min readPublished
Friday's 11 to 14 basis point jump in one-to-five-year yields leaves the three-year at 4.41 per cent, so anyone modelling a 2027 refinancing off a curve that priced cuts is now using the wrong base rate.
The Investor · Invest desk
Compiled by The InvestorSomething wrong?How this is made
Subtract the two numbers this auction week hands you: three-year paper at 4.41 per cent standing 78 basis points above effective fed funds implies a funds rate near 3.63 per cent [5][6][1]. A three-year yield is roughly the average overnight rate expected over three years plus whatever the market demands for waiting, so a gap that wide on the positive side means the expected path sits above today's setting, and Wolf Street reads it as multiple hikes being priced in that segment [7]. Or rather, the more careful version: nobody can say how much of the 78 basis points is term premium extracted in a week when the government moved $797 billion of paper through ten auctions [1], so the claim worth carrying into a budget is the narrow one, that three-year money now embeds an average policy rate above the current one with no net easing inside it.
The cost of that repricing landed on whoever bought the calendar. Tuesday's $78 billion of two-years cleared at 4.20 and the same maturity closed Friday at 4.34 [9][8], Wednesday's $79 billion of five-years went about nine basis points cheap to Friday's 4.48 [11], and Thursday's $50 billion of sevens about eight cheap to 4.59 [12]. Size-weight those three and the $207 billion of coupons sold this week is marked 10.6 basis points worse than the level the government sold it at [2][7], inside four days, on a press event rather than a data print [4].
Which is why the funding mix is the more interesting document. Bills were $562 billion of the $797 billion, 70.5 per cent of the week [2][4], and the residual $28 billion of the note calendar was a two-year floater priced at 5.5 basis points over the most recent 13-week bill auction, resetting weekly [3][10]. Bessent, whose job is to move $1 trillion of paper every three to five months at the lowest yield available [16], is not terming out into a thirty-year at 5.22 per cent [13]; he is funding short, at the precise point on the curve that just repriced, and he has spent three interventions in a month on the long end instead (the yen operation confirmed August 3 [17], the doubled buybacks announced August 19 [18], the leaked General Account story of August 24 [19]), each of which bought a day or two before yields resumed. Druckenmiller's line in the WSJ, that debt management which even appears to follow the political calendar spends credibility two centuries in the making [20], is the expensive version of the same observation.
This is probably wrong, but the front end is the number to plan from, because the long end has held above 5.20 through all three of those operations [13] while the three-year walked up to meet it, leaving 81 basis points of curve between three years and thirty [5]. Three ways it plays differently: Friday was an event repricing and the three-year drifts back toward Tuesday's 4.20 clearing level [9]; the 78 basis points is supply premium rather than policy expectation, in which case a lighter coupon calendar fixes it with no help from the Fed; or the hikes arrive and 4.41 for three years was cheap. The falsification is easy to diary. If the three-year trades back inside 4.20 within a month with the funds rate unchanged, this desk mistook a press conference for a level. Until then, the transactable number is 4.41 [5].
Ranked by verification strength, evidence, and original report placement.
The US government sold $797 billion of Treasury securities in the week, spread over 10 auctions held Monday through Thursday, with no auctions scheduled on Friday.
Of the week's total, $562 billion were Treasury bills with maturities from 4 weeks to 26 weeks across six auctions, most of which replaced maturing T-bills.
Of the week's total, $235 billion were Treasury notes across four auctions, including a regular 2-year note with a fixed coupon and a 2-year Floating Rate Note.
On Friday, Fed Chair Warsh refused to offer markets reassurance, which caused yields across the Treasury curve to rise, with maturities of 1 year to 5 years rising the most, by 11 to 14 basis points.
The 3-year Treasury yield rose 11 basis points on Friday to 4.41%, the highest since a few days in January 2025 and before that the highest since 2024.
The 3-year Treasury yield is now 78 basis points above the Effective Federal Funds Rate, which the Fed targets with its policy rates.
Distinct publishers with included, body-backed reporting in this cluster.
Follow any of these and your For You feed starts watching them — no settings page required.
invest
Treasury moved $742 billion in a week. The price was a 5.216% thirty-year.1 distinct publisher
invest
The 30-year is back at 5.27% and Treasury's buyback rally lasted two days1 distinct publisher
invest
Yields rose, the dollar fell: Bessent's buybacks broke the offset allocators price1 distinct publisher
invest
Central banks have sold $233 billion of Treasuries since February. Private money did not blink1 distinct publisher
Evidence-backed comparisons of source perspectives and observed adoption signals. Read the methodology
Which Builder, Operator, and Investor concerns the observed source mix emphasized—not a truth score.
Evidence, demonstrated adoption, hype gap, incentives, and confidence are assessed independently, each on its own current evidence. How these are measured.
Checkable numbers, one transcriber
Almost every figure that carries weight here — auction sizes, clearing yields, the floating-rate spread, Friday's closes — is public Treasury data that anyone can verify, and the internal arithmetic holds when you test it. The weakness is not the numbers but the joins between them: that Warsh's silence produced the front-end move, and that the 3-year gap to the funds rate means multiple hikes, are asserted by Wolf Street and checked by nobody.
Repriced in executed trades
This is not a forecast about where rates might go; $797 billion changed hands in four days, and the repricing shows up in what buyers paid. The 2-year cleared at 4.20% on Tuesday and closed at 4.34% on Friday. Across all $207 billion of fixed-coupon notes, the week's buyers were about 10.6 basis points underwater within days. The August 13 long-bond auction at 5.216% puts the same signal on the other end of the curve.
Attribution outruns the tape
The gap runs in one direction only. The tape in this story is sober and, if anything, conservative — the 10.6 basis point average loss on the week's coupon auctions is stated flatly and never spun. The overreach is in the causal packaging: one Fed appearance credited with a curve-wide move, three Treasury actions filed as 'hocus-pocus' that 'fizzled', and a 78 basis point spread read as the market pricing multiple hikes. Strip the verbs and the numbers still stand.
Issuer, critic and author all positioned
Everyone in this story wants yields to do something. The issuer's incentive is stated outright — sell a trillion every three to five months at the lowest yield he can get — which is also the motive for the buyback and intervention headlines the story dismisses. The critic, Druckenmiller, is Bessent's former boss making his case in the Wall Street Journal. And the narrator tells you he is not buying 30-year bonds at a 2.25% breakeven, so a story about long yields staying high is a story about his own book being right.
One desk, internally consistent
Confidence is capped by arithmetic rather than raised by it. Nothing corroborates this account — a single publisher, no Fed or Treasury voice, the CNBC report relayed second-hand. What earns the middle of the range is that the figures cross-check: 3-year less 78 basis points gives a plausible funds rate, the three coupon sizes sum to the stated note total, and the 3s30s spread of 81 basis points falls out cleanly. A writer inventing numbers rarely leaves them this consistent.