CPI Ran Hot. The Real-Yield Band Is the Test.
Both numbers accelerated. That is the first thing to say, because the headline and the core are usually read as rivals and here they moved the same way: the all-items index rose 3.7% over 12 months, and the index that removes food and energy rose 2.6%. A month ago those reads were 3.4% and 2.4%. Nothing in the release softens. The energy complex is doing the visible work - gasoline alone rose 9.1% on the month and accounted for more than half of the headline increase - but core rose on its own terms too. What follows is what printed, where it lands on this column's tape, and whether it is bullish.
What printed
September CPI, released 14 October at 12:30 UTC. Headline +0.6% on the month and +3.7% over 12 months (NSA) against +0.6% and +3.6% expected. Core, all items less food and energy, +0.3% on the month and +2.6% over 12 months (NSA) against +0.2% and +2.5% expected.
The beat is the point. Three of the four reads came in hotter than consensus, and the one that matters most is core m/m: +0.3% against +0.2%. That is the figure closest to the price pressure the central bank reacts to, and it is the one that turns an energy spike into something broader. A hot headline was expected; a core that beat was not, and it forces a repricing rather than a shrug.
For comparison, August printed 3.4% headline and 2.4% core. Both measures stepped up, so this is not one series pulling the other.
The composition is the energy story. Gasoline rose 9.1% in September and supplied more than half of the monthly all-items increase; energy is up 20.4% over the year. The 10yr breakeven had been printing 2.33% as of 13 October, near the low end of its recent range, against a West Texas barrel at $94.39 - up 6.19 on the week. The inflation compensation priced into the bond market was, until this print, not reflecting what the energy tape had already done.
Why both numbers matter here
The distinction is not academic. The headline is where energy shows up; the core is closer to what the central bank reacts to, because it is the part that persists once a commodity move has passed through.
A hot headline with a quiet core is a pass-through question. A hot headline with a rising core - which is what printed - is a demand and services question, and it removes room to argue the strength is only a barrel of oil.
That is why the two are printed side by side on this site. The gap between them is the story; either one alone hides half of it.
Where it lands on this column's tape
The 10yr real yield is 2.887% - the 10yr cash note at 5.217% less inflation compensation of 2.33% (13 October). That rate is the discount on every long-duration claim, and it has printed inside a 3 basis point band for three weekly reads: 2.905% on 30 September, 2.909% on 7 October, 2.887% on 13 October.
This morning's Scarcity piece named its invalidation explicitly: the real yield breaking back above 2.95% with copper still at rank 95. Copper closed at percentile rank 98.8, a new high for the sequence, with the dollar firmer - metal bought for its own tightness rather than for currency.
An inflation print that lifts the breakeven tests exactly that line. If the breakeven rises and the nominal holds, the real yield falls and the hurdle gets easier - the scarcity complex keeps its bid. If the nominal rises with it, the real yield pushes toward 2.95% and the first casualty is the physical end that just made a rank high. The print alone does not decide it; the two legs' response does.
What we are watching
One: the breakeven. It was 2.33% on 13 October. A move back toward the mid-2.40s without a matching nominal move is the benign path; a breakeven that does not rise on a 3.7% headline would say the market treats this as an energy spike, not a regime.
Two: the real yield against 2.95%. That is the number this column set, not a round figure chosen after the fact.
Three: the next long-end Treasury auction. A tail steepens the curve and compresses the premium on the levered bitcoin claim; a stop-through reverses it.
Credit is quiet into it - high-yield OAS 3.11% and investment-grade 0.83% as of 12 October. The composite reads +0.218, NEUTRAL, on 5 of 5 sub-indices, with Rates & Policy dominant.
Bullish or not
Rates: bearish. Both inflation measures accelerated, and a hotter core is the harder of the two to argue away. It narrows the room for the front end to come down.
The dollar: mildly bearish. A stickier inflation print against a central bank with less room to ease is not a dollar-negative story on its own, and the index firmed to 102.645 on the day. This is the weakest of the four reads.
Gold and bitcoin: mixed, tilting constructive. Higher inflation compensation helps the case for assets with no cash flow; a higher real yield hurts it. Which of the two dominates depends entirely on the breakeven response in point one above.
The scarcity complex: neutral until the real yield answers. The call this morning stands until 2.95% is breached, and the physical legs carry it.
The honest summary: this is a print that keeps the current thesis on probation rather than settling it. The level to watch was named before the number arrived, and it has not been crossed.