By Kresmion Research, July 24, 2026
Brent crude settled at 100.69 dollars on Thursday, up about 7 percent, after Houthi forces attacked two Saudi oil tankers in the Red Sea. US stocks fell hard. The Nasdaq Composite lost 2.15 percent and the S&P 500 lost 1.21 percent, while interest rate markets moved further toward pricing something they have not priced in this cycle: a Federal Reserve rate increase.
The reading that followed almost everywhere was straightforward. Oil shock, therefore inflation, therefore a Fed that has to tighten.
Before accepting that, it is worth asking what the repricing actually consists of, because most of it happened before Thursday. The move that put the two year Treasury above the Fed's own target range has been building since March, and when you split it into its parts, the component that prices expected inflation is going the other way.
Two things about dating, because they shape everything below. Treasury yields, real yields and credit spreads publish with a lag, so the latest complete readings are Wednesday July 22. That is not a clean pre shock cutoff: the tanker attack was reported on July 22, and Brent had already risen about 5 percent from 89.22 dollars on July 20 to 94.07 by the Wednesday close, before Thursday added another 7 percent. So Wednesday's readings sit in the middle of the oil move. Every figure below is labelled with the date it belongs to.
Key takeaways
| Measure | Latest | One year earlier | Where it sits |
|---|---|---|---|
| Fed target range, upper bound | 3.75 percent | 4.50 percent | Three cuts of 25bp since September |
| 2 year Treasury yield | 4.31 percent | 3.83 percent | Highest since February 12, 2025 |
| 2 year minus target range top | 56 basis points | negative 67 basis points | Widest since November 9, 2022 |
| 10 year real yield (TIPS) | 2.39 percent | 1.96 percent | 99.5th percentile since 2015 |
| 10 year breakeven inflation | 2.28 percent | 2.39 percent | 32nd percentile of the trailing year |
| High yield credit spread | 268 basis points | 290 basis points | 4th percentile of the trailing year |
All rows are dated July 22, 2026 against July 22, 2025. Breakeven inflation also has a July 23 print, unchanged at 2.28 percent.
The gap between policy and the two-year is the widest since 2022
Over the twelve months to July 22 the Federal Reserve lowered the upper bound of its target range from 4.50 percent to 3.75 percent, in three 25 basis point steps in September, October and December of last year. The effective fed funds rate fell from 4.33 percent to 3.63 percent.
The two year Treasury yield went the other way. It rose from 3.83 percent to 4.31 percent, an increase of 48 basis points.
That leaves the two year sitting 56 basis points above the top of the Federal Reserve's own target range. Across 2,889 trading days since the start of 2015 the gap has been wider on 449 of them, about 15 percent, so this is no record. But it is the widest reading of the trailing year, and the last day it was strictly wider was November 9, 2022, in the middle of the fastest tightening cycle in four decades. A year ago the same measure was negative 67 basis points, with the two year sitting below the policy rate, which is what a market expecting cuts looks like. It has been positive continuously since March 12 of this year.
The move has also been quick. The two year rose 18 basis points in the five sessions to July 22, against a median absolute five session move of 7 basis points and a 90th percentile of 17 over the trailing two years. It is a top decile move by size and no more: 40 of the last 500 five session windows produced a larger absolute change, the biggest being a 48 basis point fall in August 2024.
What moved is the real yield, not the inflation component
A nominal Treasury yield splits into a real yield, quoted directly by the inflation protected market, and a breakeven inflation rate, which is the difference between the two. If an oil shock were feeding a genuine inflation problem, the breakeven is where it should appear.
Over the year to July 22 the 10 year real yield rose from 1.96 percent to 2.39 percent, a gain of 43 basis points. Over the same dates the 10 year breakeven inflation rate fell from 2.39 percent to 2.28 percent, a decline of 11 basis points. The nominal 10 year rose 32 basis points, and 43 minus 11 is 32, so the decomposition closes exactly.
The three month picture is the same shape. Since April 22 the real yield has risen 47 basis points and the breakeven has fallen 10, against a 37 basis point rise in the nominal.
The levels agree. The 10 year real yield at 2.39 percent sits at the 99.5th percentile of every observation since 2015, and the last strictly higher reading was October 31, 2023. The 10 year breakeven at 2.28 percent sits at the 32nd percentile of the trailing year. At the five year point, where an oil passthrough should bite hardest, the real yield rose 70 basis points over the year while the breakeven fell 17, from 2.47 percent to 2.30 percent.
What Thursday itself shows, and what it does not
Breakeven inflation is the series that saw the oil move most directly. On Thursday, as Brent rose about 7 percent, the 10 year breakeven finished unchanged at 2.28 percent and the five year fell one basis point, from 2.30 percent to 2.29 percent.
That is one session and it admits more than one reading. The obvious one is that expectations are anchored and the market does not think an oil spike changes the medium term inflation path. But the S&P 500 fell 1.21 percent and the Nasdaq 2.15 percent the same day, and a growth scare pushes breakevens down too. A flat breakeven on a sharp risk off session is consistent with both stories, and a single print cannot separate them. It is worth reporting because it is the most direct evidence available, and it is worth not over reading.
Corporate credit, and a correction to our own first take
The Treasury curve, its inflation protected counterpart and the breakeven derived from them are all one market. Splitting a single curve into two components is arithmetic, and we are not going to present it as a second opinion.
Corporate credit is a genuinely separate market. The ICE BofA high yield option adjusted spread closed at 268 basis points on July 22, 3 basis points tighter than a week earlier and 22 basis points tighter than a year earlier. Against its own history that is the 4th percentile of the trailing year and the 7th percentile of 787 observations since July 2023. Credit spreads are the price of the risk that companies fail to pay, and during an inflation shock that forces a central bank to tighten into a slowing economy, that price rises.
Two qualifications. The 268 reading is 3 basis points wider than a month ago, when the spread touched 265, so the calm reads better against a one year window than a one month one. And it is a Wednesday figure.
We initially treated one further item as evidence against the credit reading, and checking it properly showed the opposite, so it is worth setting out. Kresmion's correlation engine flagged the high yield credit ETF breaking its relationship with the energy and healthcare sectors in Thursday's run, which looked like credit and equities parting company. It is not. On that run energy broke with 16 separate partners and healthcare with 10, together accounting for 26 of the 55 breaks, so both sectors were decoupling from the whole tape on an oil shock day. The credit ETF's correlations with the S&P 500, the Nasdaq 100, the total market and the Dow all held. Its break with healthcare had also fired in seven of the previous ten sessions, so it is not new. Credit did not separate from the broad equity market on Thursday.
What argues against this read
The evidence is mid shock rather than pre shock, and thinner than it looks. The 18 basis point move at the two year point, the 268 basis point credit spread and the real yield level are Wednesday readings taken after Brent had already risen about 5 percent from Monday. They cannot describe the market's response to Thursday, and they are not a clean picture of the market before the escalation either.
Breakevens have been rising recently. Over the week to July 22 the 12 basis point rise in the nominal 10 year splits into 7 basis points of real yield and 5 basis points of breakeven, so the breakeven contribution was about 40 percent of that week's move. That is a long way from the 11 basis point annual decline this article rests on. The real yield story holds clearly at the three month and one year horizons and much less well over the last week.
This article does not identify what repriced the policy path. The two year moved 18 basis points in five sessions ending before the largest oil headline, and we can show that the move was real rather than inflationary in composition without being able to say what caused it. That is a gap in the account and not a small one.
The two year yield at 4.31 percent is a 17 month high, and across the full period since 2015 it sits at the 87.6th percentile. As a level it is elevated rather than extreme. What stands out is the distance from the policy rate.
What the rates market is actually pricing
The two year yield is close to an average of expected overnight rates over the next two years. When the policy rate falls 75 basis points and the two year rises 48, the arithmetic points somewhere specific: the market has erased the further easing it used to expect and begun pricing the reverse.
Event markets agree on direction and differ on magnitude. On Polymarket, read live at 10:38 UTC on July 24, the July 29 meeting prices at 74.75 percent for no change and 23.15 percent for a 25 basis point increase, with 0.55 percent on anything larger. Its cumulative markets price a hike by the July meeting at 22.05 percent, a hike by the September meeting at 60.5 percent, and any increase during 2026 at 72 percent. Those numbers are not perfectly consistent with each other: the July per outcome prices sum to 23.7 percent of some increase against 22.05 percent on the cumulative market, a gap of about 1.6 points inside a single venue.
CNBC reported fed funds futures on Thursday afternoon at roughly 38 percent for July and about 82 percent for September. Comparing those with Polymarket takes care. On the standard futures convention the September figure is cumulative, covering rates being higher after September including a July move carried forward, which would make the comparable Polymarket number the 60.5 percent cumulative market rather than the 51.5 percent priced on a September move specifically. The two sets were also read about a day apart. Even after matching definitions a gap remains, and the CME September reading of 82 percent sits above Polymarket's 72 percent for any increase across all of 2026, which cannot both describe the same world.
What they agree on is the direction of travel. A week ago the Polymarket July markets traded between roughly 3.7 and 5.1 percent through the session, and CNBC put the CME equivalent below 12 percent.
What would change the read
One test can resolve quickly. The high yield spread widening past roughly 286 basis points, the top of its range across the 64 observations since April 27, would mean credit has started to price the tightening as a growth risk. That is the live one, and it could happen inside a week.
Two are slower. The five year breakeven, at 2.29 percent on July 23, moving back above the 2.61 percent it printed on April 22 would mean inflation expectations are driving the curve and this read is wrong, but that is a 32 basis point move and not a near term test on any realistic horizon. And next week's Treasury and credit prints will be the first to contain Thursday, which is less a condition than a wait.
Two calendar items frame it. The Commitments of Traders report released this afternoon at 15:30 Eastern covers positions as of Tuesday July 21, before Brent crossed 100 dollars, so it cannot speak to the oil move either. The report on July 31 is the first that can. The Federal Open Market Committee meets on July 28 and 29, with the decision at 14:00 Eastern on Wednesday, and the Federal Reserve has been in its communications blackout since July 18. That meeting has since arrived, and the market's own pricing of it moved a long way in the days after this note was written: what the rate markets were pricing going into the decision.
A note on method. This was a directed test rather than a scan. The hypothesis came from the news, and it was checked against three channels chosen in advance: breakeven inflation, real yields and credit spreads. It is not the largest anomaly pulled from a wide search, which matters when judging whether a result means anything.
Frequently asked questions
Does a two year yield above the Fed's target range mean the Fed will definitely raise rates?
No. The two year yield reflects the average expected policy rate over two years, so it can sit above the current target range simply because the market expects rates to be higher later. It describes an expected path. It does not forecast what any particular meeting will deliver.
Why use breakeven inflation instead of the published inflation rate?
The consumer price index reports what already happened. Breakeven inflation is derived from the gap between nominal Treasury yields and inflation protected yields, so it reflects what the market expects going forward and it updates every day, while the index is monthly. US headline CPI ran at 3.5 percent year over year in June, down from 4.3 percent in May.
Is the yield curve inverted?
No. The 10 year yield less the 2 year yield stood at 36 basis points on July 22, positively sloped. It is flat relative to the past year, in the flattest 7 percent of the last 249 sessions, and both ends rose over the week, with the 2 year rising faster than the 10 year.
What is a high yield option adjusted spread?
It is the extra yield investors require to hold below investment grade corporate bonds instead of Treasuries, adjusted for early repayment options. It rises when investors think defaults are more likely and falls when they think the opposite.
Where do these numbers come from?
Treasury yields, real yields, breakeven inflation and the high yield spread come from the Federal Reserve Bank of St Louis FRED database, which publishes the underlying Treasury and ICE BofA series. Correlation figures come from Kresmion's own engine, which tracks 843 pairs across 42 symbols. Event market prices were read live from Polymarket at the timestamp stated above.
Sources
Treasury and credit series publish with a one to two day lag, so the most recent available observation is July 22 for yields and credit and July 23 for breakeven inflation. Every figure above is labelled with the date it belongs to.
Kresmion tracks the macro regime and rates picture daily, and yesterday's brief covered a related question about what gold has been trading with.