Short Rates Desk
Wed, May 13, 2026

Notes for Wed, May 13, 2026

Morning brief

· generated 07:47 ET

CPI is the trade today; the rest is positioning around the print.

Energy drag is priced, food is sticky, rates haven't caught equities' demand-destruction story, we stay defensive into the 8:30.

Into CPI

April CPI prints at 12:30 UTC against a 0.4% prior, and the tape is genuinely two-sided. Energy is a known drag, SPR releases now total 53M barrels with 172M committed[1], and Brent at $108 post-Iran is already baked into the headline[2]. That's backdrop, not the trade. The stickiness leaks in on the food side, where tomatoes are running +15.1% YoY and ground beef +14.5% YoY[3], categories that don't get bailed out by a strategic release. Our read: the swing factor is core ex-shelter, and the print most consistent with the current rates tape is one that keeps the front end pinned.

We weighed the alternate: a clean core miss of 5+ bp opens a bull-flattener and finally forces rates to catch the equity narrative on demand destruction. We're not there. Front-end OIS hasn't moved into the print, and the specials tape (below) tells us the marginal bid is still on bills, not duration. View breaks if core surprises soft AND OIS actually re-prices the June meeting, one without the other is noise.

Plumbing — reserves, TGA, specials

TBAC floated a structural TGA-repo mechanism for reserve adequacy management overnight[4]. Note it, don't trade it, reserves sit roughly $250bn above LCLoR[5], so the o/n corridor doesn't stress unless a tax inflow or sharp Treasury drawdown forces the math. CPI matters for the Fed path; the TGA mechanic matters only if reserves actually breach the zone. They won't this week.

The SOMA securities lending window came in heavy at the very front: 912797UM7 (Nov-26 bill) at 7.0 bp above floor[6], 912797UH8 (Sep-26 bill) at 6.3 bp[7], 912810TW8 (the 4.75s of Nov-43) at 6.0 bp[8]. Three of the top five specials are sub-12-month bills, that's a bill-scarcity signal, not a duration story, and it's consistent with where the Treasury issuance axis still sits.

What we'd do

10s at 4.5 with central banks on hold and vol elevated reads as risk-off stasis[9]: equities have already priced lower oil and demand destruction, rates haven't followed[10]. The CPI print is the forcing function. We'd carry into the number short belly vs. long the wings, the cleanest expression if core comes in hot and the long end has to take back some of last week's bid. On a soft core, we cover and reassess; we're not chasing the rally without OIS confirmation.

Gromen's gold trade, outperforms if real 10yr stays above 2% post-CPI[11], is the cleanest non-rates expression of the same sticky-print thesis. We're sympathetic to the framing but it's a metals trade, not ours to put on. Assumes the SPR cadence holds and no second-leg Iran escalation between now and Friday; if either breaks, energy stops being a drag and headline does the work for us.

Sources read

8 sources read

  • Commentary items: 8

Citations

  1. [1]SPR releases now total 53M barrels with 172M committed (SPR release (53M barrels now, 172M total committed) is demand-side oil supply management) — Commentary · zerohedge.com
  2. [2]Brent at $108 post-Iran is already baked into the headline (Brent at $108 post-Iran conflict is an input to tomorrow's CPI, energy will be a drag on headline ex-prior 0.4%) — Commentary · zerohedge.com
  3. [3]tomatoes are running +15.1% YoY and ground beef +14.5% YoY (Tomatoes CPI +15.1% yoy / Ground beef CPI up 14.5% yoy) — Commentary · twitter.com
  4. [4]TBAC floated a structural TGA-repo mechanism for reserve adequacy management overnight (TBAC floating TGA repo mechanism to manage reserve adequacy, structural plumbing fix, not a market mover tomorrow) — Commentary · conks.plumbing
  5. [5]reserves sit roughly $250bn above LCLoR (reserves sitting ~$250bn above LCLoR, so near-term repo stress unlikely) — Commentary · conks.plumbing
  6. [6]912797UM7 (Nov-26 bill) at 7.0 bp above floor (912797UM7 (B 11/05/26) at 7.0 bp) — Observation · observation:seclend_observations:912797UM7:2026-05-12
  7. [7]912797UH8 (Sep-26 bill) at 6.3 bp (912797UH8 (B 09/24/26) at 6.3 bp) — Observation · observation:seclend_observations:912797UH8:2026-05-12
  8. [8]912810TW8 (the 4.75s of Nov-43) at 6.0 bp (912810TW8 (T 04.750 11/15/43) at 6.0 bp) — Observation · observation:seclend_observations:912810TW8:2026-05-12
  9. [9]10s at 4.5 with central banks on hold and vol elevated reads as risk-off stasis (10s at 4.5 with central banks holding steady and vol elevated, classic risk-off stasis) — Commentary · twitter.com
  10. [10]equities have already priced lower oil and demand destruction, rates haven't followed (equities are already pricing lower oil and demand destruction, but rates haven't followed yet) — Commentary · twitter.com
  11. [11]Gromen's gold trade, outperforms if real 10yr stays above 2% post-CPI (Gromen betting gold outperforms if real rates stay elevated post-CPI tomorrow; implicit call that headline stays sticky enough to keep 10yr real yield above 2%) — Commentary · twitter.com

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

Close brief

· generated 16:54 ET

30Y prints a 5-handle, we read it as term premium, not regime change

Long-bond auction tailed to first 5%+ stop since 2007; front-end funding unfazed, curve bear-flattening into Warsh handoff Friday.

The session

Today's tape was the 30Y auction, not the CPI line in our calendar. CPI prints 333.02% in the feed, a tick we're treating as a data anomaly, not signal; sized risk off that number stays in the drawer until BLS clarifies. The real event was the long bond: the 30Y stopped at 5.046%, the first 5%+ high yield on a 30Y refunding since August 2007[1], tailing the WI by 0.5bp with a 2.303 bid-to-cover that came in below the 2.43 six-auction average[2]. Indirects took 66.6%, dealers were left with 11.7%[3].

We don't read this as a demand exodus. Indirects holding 66s into a tail says end-demand cleared the supply, just at a higher yield. It's term premium repricing, not a buyers' strike. Dealer takedown at 11.7% is on the heavy side of the recent range, worth watching, not yet a warning.

What the curve says

Bear-flattening, not bear-steepening, has been the trade through the week. 2Y closed Monday at 3.95%[4] versus 3.79% to start April; 30Y at 4.98%[5]. 2s10s compressed to 0.46[6] from 0.52 a month back. The NY Fed 10Y term-premium estimate sits at 0.70[7], running roughly 5bp above the April mean and the firmest read since spring. With Brent around $106 on Hormuz disruption[8] and Warsh sworn in Friday, the long end has reasons to stay heavy that aren't tied to today's CPI line.

Front-end / plumbing

Funding is undisturbed. SOFR fixed at 3.60% Monday[9], 5bp under IORB at 3.65%[10], the same corridor we've held all week. EFFR at 3.63%[11], GCF UST at 3.642%[12], ON RRP take-up $1.2B[13]. SOMA SecLend showed 912797TW7 (Aug 13 bill) at 7.2bp with $4.33B accepted[14], fail-prevention bid for a known on-the-run bill, not specialness. Through the front end there is nothing that contradicts the read that today's long-end move is duration repricing, not a plumbing event.

Into tomorrow

Warsh takes the chair Friday into a market already pricing zero 2026 cuts and a hike tail in 2027[15]. We're carrying a long-end short bias into the open: a 5-handle stop clears a psychological line, dealer takedown is on the heavy side, and the energy backdrop tilts inflation tails. Alternate read we considered and rejected: today's auction was a one-off concession and the long end mean-reverts on a soft PPI print. Weight against, the tail was modest and Indirects covered, so this isn't a failed auction, it's a market clearing at a new yield. View breaks if Hormuz reopens cleanly and the 30Y trades back through 4.85 inside a week. Assumes Treasury holds the May refunding bill-issuance pace and Warsh doesn't open dovish.

Sources read

8 sources read

  • Commentary items: 8

Citations

  1. [1]the first 5%+ high yield on a 30Y refunding since August 2007 (first 30Y auction to print with a high yield above 5%, and a coupon of 5%, since August 2007) — Commentary · zerohedge.com
  2. [2]tailing the WI by 0.5bp with a 2.303 bid-to-cover that came in below the 2.43 six-auction average (tailed the 5.041% When Issued by 0.5bps... bid to cover was 2.303, down from 2.385, below the 2.43 six auction average) — Commentary · zerohedge.com
  3. [3]Indirects took 66.6%, dealers were left with 11.7% (Indirects taking down 66.6%, up from 64.1% in April... Dealers were left with 11.7%) — Commentary · zerohedge.com
  4. [4]2Y closed Monday at 3.95% (3.95) — FRED DGS2 · May 11, 2026
  5. [5]30Y at 4.98% (4.98) — FRED DGS30 · May 11, 2026
  6. [6]2s10s compressed to 0.46 (0.46) — FRED T10Y2Y · May 12, 2026
  7. [7]NY Fed 10Y term-premium estimate sits at 0.70 (0.6999) — FRED THREEFYTP10 · May 8, 2026
  8. [8]Brent around $106 on Hormuz disruption (keep Brent crude prices elevated at ~$106 per barrel) — Commentary · zerohedge.com
  9. [9]SOFR fixed at 3.60% Monday (3.60) — NY Fed SOFR · May 12, 2026
  10. [10]IORB at 3.65% (3.65) — FRED IORB · May 13, 2026
  11. [11]EFFR at 3.63% (3.63) — NY Fed EFFR · May 12, 2026
  12. [12]GCF UST at 3.642% (UST 3.64200) — DTCC GCF repo · May 12, 2026
  13. [13]ON RRP take-up $1.2B (1.199) — FRED RRPONTSYD · May 12, 2026
  14. [14]912797TW7 (Aug 13 bill) at 7.2bp with $4.33B accepted (7.2 bp, $4.33B accepted) — Observation · observation:seclend_observations:912797TW7:2026-05-13
  15. [15]pricing zero 2026 cuts and a hike tail in 2027 (market prices zero cuts 2026 and eyes hike in 2027) — Commentary · zerohedge.com

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