Treasury announced Sept 9 operations. It did not announce a number. The market is trading the blank.
DESK READ
Bessent gave the market a date — September 9 — and withheld the only parameter that matters: size. Monday's tape shows what happens when you hand a bond market a blank to fill in. A global, belly-led bull-flattening: UST 5s through 10s down 2.9–3.2bp, 30s −2.9 to 5.20, gilts, bunds and OATs down in sympathy, TLT +0.6% — with no data catalyst. JGBs, the one market with no buyback fantasy to price, were the only major sovereign to cheapen. This is not relief. This is speculation about an unsized operation, and every basis point of it is announcement premium being spent before the announcement is complete.
Ambiguity is doing real work here. An unsized operation lets every participant price their own preferred version — the desk that wants $50bn a month trades it like $50bn a month. That is a free option Treasury handed the market, and options decay: each session of pre-positioning raises the size required on September 9 to validate the move. 30s have gone from 5.26 Friday to 5.20 — six basis points of grade inflation before the exam.
WHAT THE PRECEDENTS SAY
Sixty-five years of curve management sorts into four buckets, and the sorting variable is not size — it is funding source plus commitment. Reserve-neutral duration swaps: Operation Twist 1961 (~$8.8bn) moved long yields roughly 15bp; the 2011–12 Maturity Extension Program moved ~25bp on $667bn. Selling the front to buy the back rearranges duration without creating money, and history prices that at 15–35bp — temporary. The 2000–02 Treasury buybacks ($67.5bn) sit at the top of that range (~35bp of scarcity premium), but they ran in a surplus, against shrinking supply. That precondition is inverted today; citing them as precedent gets the mechanism exactly backwards.
Fed QE with committed size: QE1 moved long yields on the order of 90bp — but on $300bn of Treasuries plus $1.25T of MBS in newly created reserves, in a crisis. QE2 ($600bn) managed ~20bp; QE3's ~25bp came from an open-ended commitment that was the opposite of ambiguity — a stated $85bn per month, every month, until conditions changed. The market knew the flow. The lesson buried in QE3 is the one that indicts Sept 9: open-endedness works as a floor when it commits to a number per unit time. A date without a number commits to nothing.
Crisis backstops: the BoE's 2022 gilt operation moved 30-year gilts ~100bp using only £19.3bn of a £65bn ceiling — the most capital-efficient intervention on the board, because it was capped, dated, credible, and aimed at dysfunction rather than levels. And the cautionary bucket: BoJ YCC held the curve for eight years on an unlimited commitment, then handed the entire suppressed premium back on exit — the +126bp one-year JGB repricing sitting on today's board is that bill being paid in real time. Level suppression is not a policy; it is a deferral with interest.
Now place Sept 9 on that map. To reverse the year's grade, Treasury needs a QE1-sized effect (75bp+) from a tool whose reserve-neutral form has never delivered more than ~35bp — and the one funding route that borrows QE's mechanism, the TGA, carries a built-in mechanical reversal when the account rebuilds through bill issuance. The liquidity ledger is equally unforgiving: Fed QE added reserves and left them; Twist and bill-funded buybacks are liquidity-neutral; TGA-funding adds reserves now and drains them later, on a schedule Congress's appropriations fights — not Treasury — will set. History's scoreboard says the announced structure can buy 15–35bp of temporary compression. The market has already advanced 6 of them on faith.
BEFORE SEPT 9 — WHAT THE SET-UP REWARDS AND PUNISHES
The pain trade between now and the operation is being short the long end into the ambiguity — Treasury has every incentive to let speculation do its compression work for free, and no incentive to cap it. But the entry degrades daily: chasing duration here means buying at 5.20 what traded at 5.26 three sessions ago, in front of a binary you cannot handicap because the counterparty controls the number and has told you nothing. For real-money buyers in our lane, the asymmetry argues for patience: let the event price, and treat any post-announcement backup — disappointment retrace, funding-math selloff, either flavor — as the entry, not this rally. The 15+ year IG bucket at 6.26% / 103 OAS and a muni curve grinding tighter (MUB 105.41) will still be there on September 10.
AFTER SEPT 9 — THREE BRANCHES
Big and TGA-funded: the rally extends, then the funding math bites. A general-account drawdown is a reserve injection today and a bill-issuance rebuild tomorrow — supply deferred, not destroyed. The second leg is the fade. Small and conventional: everything front-run comes back out; the year-long grade (30Y/3M +72bp) reasserts, and the announcement premium reverses at whatever level you chased. Unsized again, or deferred: the worst branch. Ambiguity stops functioning as an option and starts functioning as a credibility event — the market concludes there is no number because no number survives the funding arithmetic.
THE CREDIBILITY FIGHT GOES PUBLIC
This is no longer a technical dispute. Druckenmiller — Bessent's early mentor, per Bloomberg's framing — publicly calls the bond purchases a mistake. Warsh has been handed a high-profile Jackson Hole slot explicitly billed as a rebuttal to his critics, days before the operation runs. And the deficit plan that sits underneath all of it faces, in Bloomberg's words, grim prospects in Congress. Both sides of the argument now have microphones scheduled before September 9. For bond investors the meta-signal matters more than any speech: when the debate is about whether the intervention should exist, the intervention's half-life is already being priced.
THE WIRE
IRAN CAMPAIGN HINGES ON CHINA (Bloomberg) — A sanctions campaign that excludes China is unlikely to dent Iran meaningfully; Beijing defends its cooperation with Tehran and warns against secondary measures. The transmission channel for bond investors is unchanged: escalation to Chinese banks doesn't trade through oil, it trades through Treasuries — via the response function of the system's largest foreign creditor, in the same window Treasury is asking that market for patience.
CANADA ANSWERS WITH FISCAL, NOT JUST TARIFFS (Bloomberg) — Ottawa's response package reportedly includes loans and benefits — fiscal support layered onto an economy already printing negative GDP. Consistent with the tape: the Canadian front end rallied again (1Y −7.5bp to 2.68%), a rates market pricing both the growth hit and the policy answer to it. Canada still wears the trade war; now it's borrowing to pay for the privilege.
CRUDE, DAY TWO (Koyfin) — WTI −2.4% to 85.01, Brent 92.17 — second consecutive session lower post-sanctions. The escalation trade continues to live in products (heating oil +101% YTD) and gold (4,643, +14.5% 1M), not the barrel. One crack on the fixed-income sheet: senior loans at a −2.46 z-score, the only stretched row on the board — floating-rate paper quietly pricing the cut path.
PRESSURE GAUGE
Carried unchanged from Monday's letter: Front-End NEUTRAL · Term Premium PRESSURE · Credit CALM at the index, risk tiering at the CUSIP level · Inflation PRESSURE · Growth/Labor COOLING · Cross-Asset GROWING/ELEVATED · Composite: MODERATE — pressure broadening, not deepening.
WATCH
Any leak, hint, or trial balloon on Sept 9 size — the number is the whole trade · Warsh's rebuttal language at Jackson Hole · 30Y: 5.20 held into the event = premium building; back through 5.26 = premium refunded early · BKLN z-score persistence · Crack spreads vs. crude on day three.
Source: Koyfin market data as of August 25, 2026, 07:33 ET. Sovereign yields, FX, commodities, and index OAS as displayed; changes are 1-day and 1-year unless noted. Sept 9 operation date per Treasury statements as reported; operation size undisclosed at publication. Druckenmiller, Warsh, Iran, Canada, and Congressional items per Bloomberg reporting. Historical intervention impacts are approximate event-study estimates (Swanson; Gagnon et al.; Krishnamurthy & Vissing-Jorgensen; Bank of England); ranges vary by study.
Produced with AI assistance. All data selection, analysis, conclusions, and final editorial judgment are the author's. All content is reviewed and approved by Positive Carry LLC prior to publication.
The Bond Bro Dispatch is published by Positive Carry LLC. All content is general market commentary provided for informational and educational purposes only and does not constitute investment advice, a recommendation, an offer, or a solicitation to buy or sell any security. Nothing herein is tailored to the circumstances of any recipient. Data are drawn from sources believed reliable; accuracy and completeness are not guaranteed. [email protected] · © 2026 Positive Carry LLC. All rights reserved.