The Treasury buyback is doing what it was designed to do — to Treasuries. It is not a fixed income bid; it is a UST bid. Munis sat out the bounce, ratios cheapened into it, and the long end of the tax-exempt curve is now being paid to wait rather than to chase. Three catalysts land before Friday's close.
The tape
Front end firm, back end still heavy on the level even as the shape improves. 2Y 4.195 (+1.2bp), 5Y 4.349 (+0.9), 10Y 4.639 (+0.6), 30Y 5.173 (+0.6). Bills sit at roughly 3.79 against a 3.50–3.75 target — the front end is not pricing a cut, it is pricing a small hike premium. The 1Y at 3.990 carries about 24bp of it.
Tuesday's close was the real move: TLT +1.10%, TLH +0.86%, IEF +0.54%, LQD +0.64%, AGG +0.47%. The long-duration complex now sits at Z-scores of 1.7–2.1. That is a stretched tape into three scheduled catalysts, which is a poor place to be adding beta.
The muni read
This is the part the rates desks will not write up. MUB returned +0.16% on the session against TLT +1.10%, AGG +0.47% and LQD +0.64%. SUB returned +0.02%. Z-scores tell the same story: the Treasury and IG complexes print 1.7–2.1 while MUB sits at 0.84 and SUB at 0.32.
Munis did not participate. That is not a mispricing to fade — it is mechanical. The buyback is a Treasury-specific technical with no transmission into tax-exempt. Every basis point the long bond rallies on Bessent's bid is a basis point of muni/UST ratio cheapening, because the denominator moves and the numerator does not.
For anyone running long muni against a Treasury benchmark, this cuts one way: relative value is improving passively, without you doing anything. The discipline is to let it. Chasing the Treasury rally with tax-exempt duration means paying for a technical you do not receive.
The second-order risk is the front end. If the twist is bills-funded, short UST yields stay pinned, and SIFMA-linked and short-call tax-exempt paper stays under pressure. SUB going nowhere at a 0.32 Z-score is the early read on that.
The third-order risk is the one that actually matters for long muni credit: corporate supply. Hyperscaler issuance ran roughly $132B through July 31 against a $35B annual average over the prior four years, with S&P putting AI-related supply near $225B through mid-2026. Amazon printed $25B in July at approximately 120bp over the 30-year. For a crossover buyer sizing long-duration exposure, that is a direct substitute for a high-grade 30-year revenue bond — same duration, better liquidity, no tax-code dependency. Long muni is not competing with Treasuries for that dollar. It is competing with Amazon.
The curve: 20s30s at one basis point
Curve shape is where the buyback shows up, but it needs reading carefully. 20Y at 5.163 against 30Y at 5.173 — one basis point. That is tight for a sector that spent most of the post-2020 period trading cheap to the long bond. What it is not is a fresh move. The same spread was already inside a basis point in late spring, well before the buyback was announced. The compression predates Bessent.
So the honest framing is that Treasury is defending an already-compressed sector rather than creating the compression. Operations concentrated in the 10–20Y and 20–30Y buckets put a floor under a spread relationship the market had closed on its own. That distinction decides how much survives the program: a spread the buyback manufactured would widen back out when operations end in November. A spread the buyback is merely holding may not.
5s30s at 82.4bp is 35bp flatter on the year; 10s30s at 53.4bp is 12bp flatter. Against that, 3m10s at 84.7bp has steepened 76bp over twelve months. The curve is being pulled from both ends at once: a policy-risk front end steepening on hike premium, and a supply-managed back end flattening on the buyback bid. Those are different trades with different sponsors, and they will not stay coincident.
The operational detail matters more than the announcement. Buybacks move from $2B to at least $4B per operation beginning September 9 and running through November, targeted at the 10–30Y sector. The funding question is unresolved — bills, or the roughly $950B Treasury General Account. If it is bills, the twist pins the front end higher exactly where the hike premium already sits.
Crack, not crude
The commodity tape is telling a different inflation story than the headline. WTI 82.36 (-3.1% on the day, -9.0% on the month), Brent 88.58 (-3.9%). But heating oil is +99.8% YTD and gasoline +90.0% YTD. Crude is falling while refined products sit at or near records.
That is a refining-capacity and crack-spread problem, not a crude problem, and it hits the CPI and PCE baskets directly through transport and utilities. Anyone reading falling crude as disinflation into this morning's print is reading the wrong contract.
Precious metals agree: silver +18.2% on the month, gold +14.2%, CRB +37.1% YTD. That is a real-rate and debasement bid running hard against a Treasury Secretary openly managing the long end.
Credit
IG broad OAS 81bp, BBB 100bp, IG 15+ 103bp at a 6.22% yield. HY broad 269bp, BB 157bp, B 285bp. Spreads did not do the work — the IG 15+ bucket returned +0.50% on the session against +0.21% for IG broad and +0.09% for HY broad. That is a pure duration rally with no credit beta underneath it.
The tail is the tell. CCC at 1,036bp against BB at 157bp is a 6.6x ratio, and CCC has produced a +0.5% total return over twelve months against +5.1% for HY broad. Whatever is being repriced in this market, it is not compensation for distressed risk.
On the tape
08:30 ET today — July PCE and the second estimate of Q2 GDP. Last official inflation read before Jackson Hole. Core at or above 0.3% MoM puts the September 15–16 meeting live; 0.1% or below takes it off the table.
After the close today — Nvidia fiscal Q2 2027, consensus near $92B revenue. This is a fixed income event whether or not it is framed as one. It sets the hyperscaler capex trajectory, which sets corporate long-duration supply, which is the term-premium input Treasury cannot buy back.
Friday 10:00 ET — Warsh delivers his first Jackson Hole keynote as Chair, nineteen days before the September FOMC. September hike odds sit near one-in-three. The July meeting held at 3.50–3.75% on a 9-3 vote with three regional presidents dissenting for immediate tightening. Warsh has been actively reducing forward guidance, so the risk is a speech that resolves nothing and leaves the August payroll and CPI prints to do the work.
What I'm watching
20s30s. It has been pinned near zero since spring, so the signal is a decisive break, not the level. Through zero says the buyback is dominating the sector and the 20Y is better sold into the operation than owned. Back out through 5bp says the defense is failing.
Muni/UST 10Y and 30Y ratios. Passive cheapening is the setup; the entry is a ratio that stops cheapening while the Treasury bid persists.
SIFMA and the short tax-exempt curve for evidence on how the buyback is funded.
Long-dated high-grade corporate concessions versus long muni. The crossover buyer is the marginal bid in both, and right now the corporate side is paying more for it.
Source: Koyfin — curve, commodity and credit levels as of 08:12 ET, August 26, 2026; ETF price and Z-score panel as of the August 25 close. Curve spreads derived. Late-spring 20s30s reference from a Koyfin yield-curve export dated May 27, 2026. Event calendar, buyback terms and issuance figures per public reporting.
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.