At 8:30 p.m. last night, two things were in place at the same time.
Warsh’s new policy framework as Fed chair, along with clearly softer inflation data.
As it turned out, the market delivered a third answer.
The 30-year Treasury yield closed at 5.64%, the highest since 2002.
The Dow Jones index fell 443.87 points that day.
First, explain these two things, and then explain why the market isn’t buying it.
1. What exactly is Warsh’s “new deal”
Kevin Warsh took over as chairman of the Federal Reserve on May 22.
The first thing he did after taking office was to dismantle the Horizon guidance toolset.
On August 28, at his first speech at Jackson Hole, he said it very plainly.
This briefing can be called an outline, or a hiking route map—just don’t call it forward guidance.
His reason is that this setup would tie the Fed’s own hands.
The former chair is used to telling everyone in advance which direction rates will go next.
Wojc believes that in this way, the market will no longer focus on the real data.
The second thing happened at the July 29 meeting.
That meeting decided to keep the interest rate unchanged; the vote result was three against.
Three regional Fed presidents demanded a rate hike on the spot.
This is the first time in many years that three votes went against the same direction.
That day’s market reaction is worth remembering for a long time.
The 30-year yield jumped 14 basis points that day, touching a high of 5.23%.
And yet the 2-year yield is actually moving downward.
The short end is pricing the Fed’s patience; the long end is pricing other things.
The third thing happened on September 16.
He pushed through a full-vote rate hike, raising the policy rate into a new range.
The reason can be summed up in one sentence: inflation is too high—and it’s been too high for too long.
This step was made in defiance of the president’s public opposition.
There’s one more thing that’s rarely cited.
He set up five working groups to specifically assess the Fed’s policy framework.
One of these sets of remarks may suggest lowering the weight of PCE as the official inflation target.
His own phrasing is that, at this stage, they are still using the PCE measure.
But after next January, he said he can’t be sure how the policy framework will be adjusted.
II. In the “new data,” there are actually two halves.
Core PCE rose 0.2% month-over-month in August; the market expected 0.3%.
Up 3.0% year over year; the market expected 3.3%.
Overall month-over-month rose 0.3%, in line with expectations.
Up 3.4% year over year, below the expected 3.7%.
Looking at just these lines, inflation is indeed softening.
But in the same report, two other numbers are hidden as well.
The core month-over-month for July was revised to 0.1%; it was originally reported as 0.2%.
The overall month-over-month figure was also revised down from 0.2% to 0.1%.
Real PCE rose 0.6% month-over-month in August, and July was revised up to 0.1%.
That is, the “price” half is soft, while the “quantity” half is rising.
This is exactly the opposite of what it looked like a month ago.
That same night, there were two other growth datasets, and the directions were completely一致.
The final estimate for US Q2 real GDP was revised up to 2.2%; the market originally expected only 1.5%.
ADP employment for September was 90,000, versus the market’s forecast of 70,000; last month was only 38,000.
The Atlanta Fed’s model reported 5.1% actual GDP growth for the third quarter.
Inflation is cooling, while growth is heating up.
Put these two sentences together—that’s the real information from last night.
III. Why the market won’t buy it
First, it cut the odds of a rate hike by a notch.
According to CME data, the probability of a 25-basis-point rate hike in October fell from 51% to 35%.
With the second action, it pushed long-end yields higher.
The 10-year ended at 5.29%, and during the session it even briefly broke above 5.3%.
The 30-year settled at 5.64%, and the next move was another five basis points higher on the day.
The directions of these two actions are exactly opposite.
It shows that what’s priced on the long end isn’t actually what the Fed will do at its next meeting.
Break down the 30-year rate at 5.64%.
The yield on 30-year inflation-protected Treasurys is 3.33%.
Subtract the two, and the market prices only 2.31% inflation compensation for the next thirty years.
The 5-year is 2.36%, the 10-year is 2.36%, and the three tenors are squeezed onto the same level line.
Among the three tenors, the maximum difference is only five basis points.
It shows the market isn’t worried about inflation getting out of control.
What the market is worried about is something else: how much real return is needed for inflation to return to target.
So within that 5.64%, it’s the half made up of the real rate that’s rising.
The half made up of inflation expectations is firmly anchored.
And it’s the same night’s GDP upward revision and ADP report for 90,000 jobs that pushed up real rates.
IV. Who is paying this bill?
US equities have diverged; that matters more than whether the indices are up or down.
The Dow fell 443.87 points, the S&P 500 dropped 0.25%, and the Nasdaq closed up 0.24%.
The intraday gains of these three indices look even better.
The S&P 500 was up as much as nearly 0.7%; the Nasdaq was up more than 1% at one point.
By the close, all of these gains were completely unwound.
What got killed was the Dow, not the Nasdaq.
The long-end rate is collecting the tax on the portion of assets sensitive to interest rates.
Gold’s reaction is even more complete—worth watching piece by piece.
After the data was released, it surged to $4,227.
By late night, it returned to around $4,165, and it gave up all of that day’s intraday highs.
Binance’s XAU/USDT perpetual contract open interest is 142,311 lots.
That number was higher before the data was released, and it has not come back at all.
Prices surged but positions didn’t rise—this suggests the money entering isn’t new long money.
Bitcoin’s daily chart left a very long upper wick.
On September 30, it hit a high of $85,604 and then closed at $83,564.7.
Brent crude rose nearly two percentage points that day.
V. Where is Wojc’s real problem?
His problem isn’t that the market doesn’t believe he can keep inflation under control.
Inflation compensation is anchored at 2.3%, suggesting the market believes inflation will return near the target.
His problem is that he simultaneously dismantled two anchors.
One is forward guidance, and the other is the definition of the target metric itself.
After the anchor is dismantled, every data point has to be priced by the market itself.
The volatility gets booked on his head, but the direction doesn’t belong to him.
After the meeting in July, JPMorgan’s Feroli wrote one sentence.
He said these two points make the market question whether the new chair can deliver on commitments to lower inflation.
The rate hike on September 16 regained some credibility.
But last night’s revision rewrote the basis for his case for the rate hike itself.
On September 16, he said inflation has been above the 2% target for 65 straight months.
Three weeks later, the same indicator’s history was revised by the statistical agencies to 3.0%.
This isn’t the market’s judgment—it’s an official revision.
Six: for your position, there’s only one switch.
Don’t use year-over-year PCE figures to go long on risk assets.
This transmission chain is wrong.
When the inflation numbers improve, it puts pressure on the half of inflation compensation.
And inflation compensation is already anchored between 2.3% and 2.4%.
What truly prices your position is the real interest rate on the long end.
On September 30, the official 10-year real yield was 2.93%.
Remember two signals on the upside.
The 30-year real interest rate broke above 3.4%, meaning the term premium has started to run out of control.
Only after it falls back below 3.0% can you say the long end has truly turned.
Last night the market moved in the previous direction.
There’s also one counterintuitive discipline.
The day when inflation data improves is often the most dangerous day for people who’ve added leverage.
Because expectations move faster than prices.
Finally
Wojc’s new policy isn’t wrong—new data is actually helping him.
But the answer the market gives is 5.64%.
Until real rates on the long end fall, the trade that inflation cools down can only be considered a bounce.
Tomorrow is the September nonfarm payroll report.
—MK keeps his word
