Duration Dashboard

The July 2026 CPI report landed Wednesday and did what the front end wanted: headline inflation eased to 3.4% year over year from 3.5%, core moderated to 2.5%, and both matched consensus. Thursday's producer price index was softer still, unchanged on the month against expectations of a 0.2% gain. Under a Committee that spent the summer debating whether to raise rates, that combination is meaningfully dovish, and the 2-year duly declined to 4.17% — its lowest Friday close since July 3. The long end declined to participate. From five years out the yield curve cheapened, leaving 2s30s eight basis points wider on the week and the 30-year at the 99th percentile of its five-year range.

The mechanism was supply, not data. Treasury brought $125 billion of coupon issuance across three sessions, and the quality of the takedown deteriorated as the tenor lengthened. The 3-year note stopped through its when-issued level; the 10-year tailed marginally; the 30-year tailed 0.4 basis points and left primary dealers holding 11.6% of the issue against a 10.6% average. Awarded at 5.22%, it was the most expensive 30-year sale since August 2001. Set against our July 12 report, when the comparable 30-year cleared at 5.06%, the long end has repriced roughly 16 basis points in five weeks with no change in the policy rate.

Treasury Auction Scorecard

The refunding week is worth reading as a single sequence. Front-end demand was genuinely strong: the 3-year drew a 2.71x cover, comfortably above its six-month average, with indirect bidders — the customary proxy for foreign accounts — taking 64.2%. The 10-year cleared at 4.68%, the highest financing cost at that tenor since 2007, but on above-average cover and 76.7% indirect participation. Only the 30-year showed genuine indigestion, and even there the deterioration was modest rather than disorderly: cover slipped to 2.39x from 2.44x at July's reopening, and dealers absorbed roughly a percentage point more than usual. Treasury's outstanding coupon stock averages 3.44%, so long money issued this week costs approximately 177 basis points more than the average bond it refinances. Interest expense reached $1.170 trillion through the first ten months of fiscal 2026, with total federal debt at $39.913 trillion as of August 12.

Credit Pulse — Spreads Unmoved by the July 2026 CPI Report

Credit spreads moved in the opposite direction from the long end. High yield spreads narrowed two basis points to 251 basis points, settling at the 5th percentile of their five-year distribution against a window low of 228; investment grade widened a single basis point to 78, the 30th percentile. Implied equity volatility declined to 14.25, the 16th percentile. The absence of any spread response to a 30-year auction at a twenty-five-year high in yield is the notable observation, and it has a straightforward technical explanation: corporate credit is an intermediate-duration asset class, and a steepening driven almost entirely beyond ten years leaves the five- to ten-year segment where most index risk sits largely unaffected.

Primary markets confirmed the tone. Investment grade issuers brought one of the heaviest August calendars on record — nineteen high-grade borrowers priced on Monday alone, the busiest single session since January 5 and the largest issuer count in seven months, spanning utilities, overseas banks and Tyson Foods, with dealers projecting roughly $40 billion more across the balance of the week. That volume followed the prior week's approximately $80 billion, the third-heaviest week of 2026. Borrowers were opportunistic in the plain sense: the payroll contraction detailed in our August 9 report on the July employment miss had pulled intermediate yields lower, and issuers moved to lock funding ahead of the July 2026 CPI print. No deals were pulled or postponed, and no fallen-angel or rising-star migrations of consequence were reported.

Federal Reserve Policy Outlook After the July 2026 CPI Report

The Committee held the target range at 3.50%–3.75% on July 29 by a 9–3 vote, with all three dissents favoring a hike — the configuration detailed in our August 5 report on the divided July FOMC. That vote framed the week's central question, and the July 2026 CPI report answered it in the doves' favor. By Friday, CME FedWatch implied a 69.4% probability that the Committee holds the range at the September 15–16 meeting and 30.6% for a 25 basis point increase — with hike odds down from 33.9% on Thursday and 50.0% a month earlier. Goldman Sachs chief economist Jan Hatzius wrote over the weekend that a September hike now looks very unlikely, arguing that two consecutive months of softer employment and inflation data leave little basis for a dove to move toward tightening; his base case is a hold through year-end with cuts deferred into 2027. Chicago Fed President Austan Goolsbee, speaking during the week, characterized output and labor market conditions as basically stable.

The relevant tension for allocators is that Fed policy has become considerably more legible over the near term while its long-run credibility question has not. Two disinflationary prints reduce the case for insurance tightening, and the September meeting carries an updated Summary of Economic Projections that will show whether the nine participants who penciled in at least one 2026 hike in June have revised. But the long end is not pricing the next twenty-five basis points; it is pricing the compensation required to hold duration through a fiscal path that neither a CPI print nor a policy hold alters. Wednesday's release of the July minutes will be the week's substantive event on this front, given how little detail the post-meeting statement and press conference provided about how the Committee intends to return inflation to target.

Energy, the Dollar, and the Japanese Channel

Crude reasserted itself as the marginal driver of the long-end inflation premium. West Texas Intermediate rose roughly 5% on Monday to $82.13 and settled Friday at $82.40, up more than 5% on the week, with Brent settling at $88.52. The catalyst was the extension of a US naval blockade of Iranian ports on an indefinite basis, alongside Treasury Secretary Scott Bessent's stated intent to pursue Iranian economic isolation. This matters directly for duration: with energy still contributing the bulk of headline inflation, an oil price that responds to Hormuz headlines rather than to demand fundamentals keeps a floor under long-dated breakevens irrespective of what core services do. The same dynamic was visible in our July 26 report on the oil-driven bear flattener, though the curve response this week ran the opposite way.

The Japanese channel deserves monitoring for its Treasury-specific implications. Following the coordinated US–Japan yen purchase in early August — the first joint intervention of its kind in more than a decade — the yen had retraced roughly half its gains, trading beyond 159 to the dollar by August 12 as rate differentials reasserted. Persistent yen weakness raises the possibility that Japanese official accounts fund further intervention through Treasury sales, a mechanical source of long-end supply that arrives independent of US fundamentals. Japan's Ministry of Finance has signaled it would use the Federal Reserve's FIMA repo facility instead, which is designed precisely to neutralize that channel, and the Bank of Japan left policy unchanged on August 1 while signaling a possible near-term move. A BOJ hike would narrow differentials and relieve pressure; disorderly yen weakness would do the reverse. The dollar traded softer through the week as hike expectations faded. Equities registered none of the bond market's caution, with the S&P 500 setting a 52-week high of 7,816.70 on August 13.

Week Ahead: FOMC Minutes and the First Read on August Activity

• FOMC Minutes (August 19): The week's principal event. The July 29 statement was unusually terse and Chair Warsh's press conference offered little detail on the path back to target, leaving the minutes as the first substantive look at how the three dissenters framed their case and how close the majority came to joining them.
• Housing Starts and Building Permits (August 18): July data arrives with the 30-year fixed mortgage average at 6.67% as of August 13. Permits are the forward-looking series and will indicate whether builders are responding to the long-end repricing.
• Industrial Production (August 18): July output follows a retail sales decline and a sentiment reading at 51.0. A soft print would extend the demand-side deterioration from consumption into production.
• Initial Jobless Claims and Philadelphia Fed (August 20): Claims for the week ended August 15 follow a 209,000 reading. Given how much of the September debate now rests on labor data, a sustained move above 210,000 would carry disproportionate weight for front-end pricing.
• Flash S&P Global PMIs (August 21): The first August activity reading for manufacturing and services, and the cleanest available look at whether July's demand softness carried into the current month.
• Further out: The Jackson Hole symposium runs August 27–29 on the theme of financial innovation and payments, with Chair Warsh delivering his first keynote in that role on August 28 — nineteen days before the September 16 decision.

US Economic Positioning and Global Context

The week clarified where the United States now sits in the global rate structure and at what cost. A 30-year Treasury clearing at 5.22% is not a distress signal — the sale drew reasonable participation, with indirect bidders taking 66.8% — but it does establish that the world's benchmark sovereign borrower now pays a twenty-five-year-high rate for thirty-year money while its policy rate sits at 3.50%–3.75%. That gap is term premium, and it has been widening in a specific pattern: the front end responds to data, the long end responds to the financing arithmetic. For allocators, the practical implication is that duration and policy exposure have partially decoupled. A September hold, which the July 2026 CPI report made the strong base case, does not mechanically support the long bond.

The intermediate curve is where the risk-reward has improved. The 10-year at 4.69% sits at the 98th percentile of its five-year range while the 2-year sits at the 57th — an asymmetry along the yield curve that has widened steadily through the summer. Real yields near 2.4% represent a genuine competing claim on capital, and the equity market's record highs alongside a bond market repricing term premium is the cross-asset disconnect worth watching rather than the credit spread level in isolation. The five- to ten-year segment offers most of the available yield with materially less exposure to the fiscal and energy inputs concentrated beyond twenty years. The dollar's softness through the week, alongside gold's continued strength on sovereign-debt concerns across major economies, suggests the term premium story is not exclusively American. What distinguishes the US case is the size of the financing requirement and the absence, so far, of any policy mechanism directed at it.

Frequently Asked Questions

What did the July 2026 CPI report show?

The July 2026 CPI report showed headline consumer prices rising 0.1% on the month and 3.4% year over year, down from 3.5% in June. Core inflation rose 0.2% monthly and 2.5% annually, the softest core reading since March 2021. Both figures matched consensus, with shelter contributing roughly two-thirds of the headline monthly gain.

Why did Treasury yields rise if the July 2026 CPI report was soft?

Because the front end and long end price different questions. Softer inflation lowered September hike odds and pulled the 2-year down 3 basis points. The 30-year rose 6 basis points on fiscal supply, an energy-driven inflation premium, and term premium — inputs a single CPI print does not address.

How significant was the 30-year Treasury auction clearing at 5.22%?

It was the highest awarded yield at that tenor since August 2001. Demand was softer than average rather than absent: bid-to-cover slipped to 2.39x from 2.44x in July, the stop tailed 0.4 basis points past the when-issued level, and dealers absorbed 11.6% against a 10.6% average.

What are credit spreads signaling in August 2026?

Very little concern. High yield narrowed 2 basis points to 251, the 5th percentile of its five-year range, and investment grade widened 1 basis point to 78, the 30th percentile. Spreads at these levels leave minimal compensation for the higher refinancing costs the long-end repricing implies.