Short Rates Desk
Fri, Jul 31, 2026

Notes for Fri, Jul 31, 2026

Morning brief

· generated 11:44 ET

The hold paid; the back-end selloff is term premium and the front end stays anchored

The hold vindicated our fade; 2y rallied to 4.22% while the 30y ran to a 19-year high. We hold the 2s30s steepener.

FOMC

The Committee held at 3.50-3.75%[1] on Wednesday, and the fade we carried in paid. Our exit never came into play: the 2y rallied to 4.22%[2] by the close, well inside the 4.37% 7/23 high[3] we set as the out. The presser resolved the same way. Warsh leaned dovish on the path while three members dissented for a hike[4], and the front end took the hold rather than the dissent. Nine-to-three with Hammack, Kashkari and Logan pushing to tighten is a hawkish count, but it reads as September optionality rather than a July signal: futures now price a September hike near 57%, down from roughly 75%[5] before the meeting.

Duration

The overnight move sits at the long end, and we read it as term premium. The 30-year sold off 11bp[6] to 5.20% on the decision[7], a 19-year high[8], and extended to 5.24% Thursday[9]. The 2y went the other way. When the front rallies and the back sells off on the same headline, that is a duration move on term premium, not a Fed-path repricing; a path shock carries the front end with it. Warsh made the read easy by welcoming the long-end repricing[10] as policy working as intended. 2s10s at 45bp[11] confirms the shape.

We hold a 2s30s steepener, receiving the front end against the 30y, into the September FOMC. The alternate read is that the back-end selloff leads a September hike that eventually drags the front end into a bear-flattener; we reject it because the 2y richened to 4.22%[12] while the 30y made new highs, the reverse of a path repricing. We're wrong if a September hike prices toward certain and the 2y breaks and holds back above 4.37%; that flips us to a bear-flattener and we are out. This assumes Treasury holds its bill-heavy issuance pace; a coupon upsize at refunding would extend the steepener for a supply reason rather than the term-premium one we want.

Data

The data flow gives the front end no reason to sell. Core PCE cooled to 3.3% y/y from 3.4%[13], headline flat on the month. Q2 GDP missed at 1.5% against 2.0% consensus[14], but final sales to private domestic purchasers ran 3.9%[15], so the miss is inventory and net-export noise over firm demand. ECI printed 3.3% this morning[16], in line, no wage acceleration to force the Committee's hand. Claims at 197k[17] keep the labor read tight. Cooling inflation with steady wages and a tight labor market keeps the front end anchored and the hike dissents as optionality.

Funding

Funding is orderly through the TGA rebuild. The Treasury General Account climbed to $970bn[18] on the week, a $135bn build[19], and that drain pulled reserves down $78bn[20] to $2.985trln[21]. SOFR fixed 3.65% Thursday[22], level with the 3.65% IORB[23], and EFFR sat at 3.63%[24]; the drain has not touched the front end. The cushion is what we watch: RRP take-up is near empty at $1.1bn[25], so the next TGA build comes straight out of reserves rather than the facility. GCF Treasury repo edged up to 3.69%[26] into today's month-end turn, 2bp firmer than Wednesday[27], orderly for a turn. Long-end term premium is bleeding into the mortgage market, with the 30-year fix at 6.66%[28].

Risks

The stagflation tail is oil. If Hormuz disruption holds crude above $90[29], the supply shock forks the front end between an inflation bid and a growth-fear rally, and the hike dissents get louder. We re-baseline the steepener if core re-accelerates on energy passthrough into the next PCE.

Sources read

8 sources read

  • Commentary items: 8

Sources read

8 sources read

  • Commentary items: 8

Citations

  1. [1]held at 3.50-3.75% (3.75 upper bound, 3.50 lower bound) — FRED DFEDTARU · Jul 31, 2026
  2. [2]the 2y rallied to 4.22% (4.22000) — FRED DGS2 · Jul 29, 2026
  3. [3]the 4.37% 7/23 high (4.37000) — FRED DGS2 · Jul 23, 2026
  4. [4]three members dissented for a hike (3 dissenters (Hammack, Kashkari, Logan) voted hike, highest dissent count in years) — Commentary · zerohedge.com
  5. [5]a September hike near 57%, down from roughly 75% (reprices September hike odds down to ~57% from ~75%) — Commentary · zerohedge.com
  6. [6]sold off 11bp (+0.11 d/d) — FRED DGS30 · Jul 29, 2026
  7. [7]5.20% on the decision (5.20000) — FRED DGS30 · Jul 29, 2026
  8. [8]a 19-year high (30Y at 5.21% (19-year high)) — Commentary · twitter.com
  9. [9]extended to 5.24% Thursday (30Y at 5.24%) — Commentary · zerohedge.com
  10. [10]welcoming the long-end repricing (Warsh explicitly welcoming long-end repricing) — Commentary · twitter.com
  11. [11]2s10s at 45bp (0.45000) — FRED T10Y2Y · Jul 30, 2026
  12. [12]the 2y richened to 4.22% (4.22000) — FRED DGS2 · Jul 29, 2026
  13. [13]Core PCE cooled to 3.3% y/y from 3.4% (Core PCE cooled to +3.3% YoY (from +3.4%)) — Commentary · zerohedge.com
  14. [14]Q2 GDP missed at 1.5% against 2.0% consensus (GDP misses badly at 1.5% vs. 2.0% consensus, but final sales to private domestic purchasers print 3.9%) — Commentary · zerohedge.com
  15. [15]final sales to private domestic purchasers ran 3.9% (final sales to private domestic purchasers print 3.9%) — Commentary · zerohedge.com
  16. [16]ECI printed 3.3% this morning (ECI beat at 3.3% YoY) — Commentary · zerohedge.com
  17. [17]Claims at 197k (Claims at 197k reinforce tight labor market) — Commentary · zerohedge.com
  18. [18]climbed to $970bn (970,442M) — Treasury General Account · Jul 29, 2026
  19. [19]a $135bn build (+135,025M w/w) — Treasury General Account · Jul 29, 2026
  20. [20]reserves down $78bn (-77,579M w/w) — FRED WRESBAL · Jul 29, 2026
  21. [21]$2.985trln (2,984,570M) — FRED WRESBAL · Jul 29, 2026
  22. [22]SOFR fixed 3.65% Thursday (3.65000%) — NY Fed SOFR · Jul 30, 2026
  23. [23]the 3.65% IORB (3.65000%) — FRED IORB · Jul 31, 2026
  24. [24]EFFR sat at 3.63% (3.63000%) — NY Fed EFFR · Jul 30, 2026
  25. [25]RRP take-up is near empty at $1.1bn (1.076) — FRED RRPONTSYD · Jul 30, 2026
  26. [26]edged up to 3.69% (UST 3.68700) — DTCC GCF repo · Jul 30, 2026
  27. [27]2bp firmer than Wednesday (UST 3.66900) — DTCC GCF repo · Jul 29, 2026
  28. [28]the 30-year fix at 6.66% (6.66000) — FRED MORTGAGE30US · Jul 30, 2026
  29. [29]holds crude above $90 ($90+ locks in a persistent inflation floor) — Commentary · zerohedge.com

Generated by Short Rates Desk. Informational only. Not investment advice.

Close brief

· generated 16:20 ET

Front end pinned, long end at 2007 highs: we keep the term-premium read

Our September-hike break test did not trip; the front end stayed anchored through the presser aftermath. We hold the steepener.

FOMC and duration

Hike odds got pared during Warsh's press conference[1] while the front end stayed pinned to administered rates[2], and that divergence is the basis for the term-premium read. Our break test on record was a September hike pricing toward certain with the 2y breaking and holding above 4.37%. It did not trip. The standing read is intact: the long-end selloff is term premium and the front end holds.

Sahm flags that the 30y repriced to 2007 highs even as hike odds fell, and reads Warsh's hedging on the PCE target and the January 2027 strategy review as a credibility risk[3]. Warsh is steering the Fed from guidance toward market-mechanics tightening, a deliberately steeper curve with less jawboning and ample short-end liquidity[4]. Yesterday's ECI at 3.3% came in line and did not shift terminal pricing[5], so nothing in the wage data forces the front end off its anchor.

We weighed the counter. Brooks argues the bear steepening has no legs given benign inflation prints, and that the real trade is dollar weakness from rate-differential compression[6]. We keep the term-premium read anyway: the front-end anchoring alongside Warsh's stated pivot decides it, and the inflation path is secondary. Position: stay short the long end as a term-premium steepener over the coming weeks, sized in tens of basis points, not the 100-200bp realignment the regime bulls want. We're wrong if a September hike prices toward certain and the 2y breaks and holds back above 4.37%; that flips us to a bear-flattener and we're out.

Funding

The Treasury General Account climbed $135.0bn on the week to July 29[7], a reserve drain of that size hitting the front end. Reserve balances are down about $77.6bn since Wednesday[8]. None of it disturbed the corridor: SOFR printed 3.65% and EFFR 3.63%, both pinned[9], and the SOFR-FF basis held stable at +1.25bp for July and -2bp for August[10]. No acute funding stress.

Bill supply stayed heavy, $221.5bn gross today, with the TGA about $20bn above target[11]. Collateral is not scarce: the most special bill in SOMA lending cleared at just a 7.5bp fee versus the 5bp minimum[12], and the rest of the specials sit below that. We keep the collateral call separate from the duration call; on funding there is nothing to trade here beyond staying anchored to the administered-rate corridor.

Cross-currents and the refunding assumption

The live assumption is Treasury's issuance mix. We need the bill-heavy pace to hold, and next week's coupon auction announcements are the test[13]. A coupon upsize at refunding would extend the steepener for a supply reason rather than the term-premium one, and we'd re-baseline the trade accordingly.

Japan is the other cross-current. The US Treasury told banks through the New York Fed that it may intervene in the yen, with rate checks already run as a precursor[14]. BoJ September hike odds sit at 35-40%, and if the BoJ keeps delaying, the JGB selloff stalls and 10y Treasuries lose a marginal bid from the world's third-largest creditor[15]. That cuts toward higher long-end yields, the same direction as our term-premium read, but a sustained yen bid that forces BoJ hikes would land through JGBs first and carry-unwind flows second. We watch it, we don't size it yet.

Sources read

8 sources read

  • Commentary items: 8

Sources read

8 sources read

  • Commentary items: 8

Citations

  1. [1]Hike odds got pared during Warsh's press conference (markets repriced 30y to 2007 highs and pared hike odds during presser) — Commentary · stayathomemacro.substack.com
  2. [2]the front end stayed pinned to administered rates (SOFR 3.65%, EFFR 3.63%, both pinned despite $77.6bln drain in reserve balances since Wednesday) — Commentary · conks.plumbing
  3. [3]Sahm flags that the 30y repriced to 2007 highs even as hike odds fell, and reads Warsh's hedging on the PCE target and the January 2027 strategy review as a credibility risk (Sahm flags Warsh's hedging on PCE target and explicit hint at January 2027 strategy review as credibility risk) — Commentary · stayathomemacro.substack.com
  4. [4]Warsh is steering the Fed from guidance toward market-mechanics tightening, a deliberately steeper curve with less jawboning and ample short-end liquidity (Warsh Fed pivoting from guidance-heavy to market-mechanics tightening, deliberately steeper curve, less jawboning, ample short-end liquidity) — Commentary · capitalwars.substack.com
  5. [5]Yesterday's ECI at 3.3% came in line and did not shift terminal pricing (ECI print yesterday (3.3%, in line) did not shift near-term terminal-rate pricing) — Commentary · conks.plumbing
  6. [6]Brooks argues the bear steepening has no legs given benign inflation prints, and that the real trade is dollar weakness from rate-differential compression (Brooks argues bear steepening has no legs given benign inflation prints... the real story is Dollar weakness from rate-differential compression) — Commentary · robinjbrooks.substack.com
  7. [7]The Treasury General Account climbed $135.0bn on the week to July 29 (+135,025M w/w) — Treasury General Account · Jul 29, 2026
  8. [8]Reserve balances are down about $77.6bn since Wednesday ($77.6bln drain in reserve balances since Wednesday) — Commentary · conks.plumbing
  9. [9]SOFR printed 3.65% and EFFR 3.63%, both pinned (SOFR 3.65%, EFFR 3.63%, both pinned) — Commentary · conks.plumbing
  10. [10]the SOFR-FF basis held stable at +1.25bp for July and -2bp for August (SOFR-FF basis stable at +1.25bps (Jul) / -2bps (Aug)) — Commentary · conks.plumbing
  11. [11]$221.5bn gross today, with the TGA about $20bn above target (Bill auction sizes remain heavy ($221.5bln gross today), TGA $20bln above target) — Commentary · conks.plumbing
  12. [12]the most special bill in SOMA lending cleared at just a 7.5bp fee versus the 5bp minimum (912797UY1 (B 11/12/26) at 7.5 bp) — Observation · observation:seclend_observations:912797UY1:2026-07-31
  13. [13]next week's coupon auction announcements are the test (watch next week's coupon auction announcements for Treasury's financing plans and any signal on pace of TGA normalization) — Commentary · conks.plumbing
  14. [14]The US Treasury told banks through the New York Fed that it may intervene in the yen, with rate checks already run as a precursor (US Treasury informed banks through Federal Reserve Bank of New York that it may intervene Friday; NY Fed rate checks considered precursor to interventions) — Commentary · zerohedge.com
  15. [15]BoJ September hike odds sit at 35-40%, and if the BoJ keeps delaying, the JGB selloff stalls and 10y Treasuries lose a marginal bid from the world's third-largest creditor (September hike odds at 35-40%... if BoJ delays again, JGB selloff stalls and 10yr TSY finds less demand from the world's 3rd-largest creditor) — Commentary · peterboockvar.substack.com

Generated by Short Rates Desk. Informational only. Not investment advice.