America Has Two Inflation Rates. The Fed Is Bound to the Higher One.

EYES ONLY

SUBSCRIBER ONLY // DO NOT FORWARD WITHOUT PERMISSION

BLACKOPS
FINANCE

COVERT FINANCIAL INTELLIGENCE. INTERCEPTED DAILY.

MISSION BRIEF

Updated 1115 ET

The Federal Reserve votes eight days from now. There are two official measures of American inflation, they disagree by 80 basis points, and the committee is bound to the higher one.

Start with the numbers, both published by the United States government for the same month. The consumer price index for July, released by the Bureau of Labor Statistics on August 12, put core inflation at 2.5% over twelve months. The personal consumption expenditures index for July, released by the Bureau of Economic Analysis on August 26, put core inflation at 3.3%.

The gap is not a rounding difference. It is eight tenths of a percentage point on a target of two, and it decides the meeting, because the Fed’s mandate is not written on the index the public reads. In the Federal Reserve’s own words, the goal is “inflation of 2 percent over the longer run, as measured by the annual change in the price index for personal consumption expenditures.”

Now the calendar. August CPI arrives this Friday, September 11, at 8:30 Eastern, five days before the decision. August PCE does not arrive until September 30, fourteen days after it. So when the committee votes, its most recent reading of the index its own target is defined on will be July’s, forty-seven days past the end of the month it measures.

The policy rate is 3.50% to 3.75% and has not changed since December 11 last year. By meeting day that is nine months without a move. The ten-year Treasury note yield topped 4.8% today, the highest since October 2023, and roughly 58% of the futures market is now positioned for a quarter-point increase next week.

Consider what each index tells a different audience. A household reading the CPI headline sees core inflation at 2.5% and concludes the problem is nearly solved. A committee reading its own mandate sees 3.3% and concludes it is more than a full point above target. Neither is wrong and neither is unofficial. The disagreement is not about the facts, it is about which set of prices you are required to count, and only one of those definitions carries a vote.

SPONSORED

Forget SpaceX, this is Elon’s Next Breakthrough

While everyone was distracted with the recent SpaceX IPO…

Elon Musk quietly filed a patent with the U.S. Patent and Trademark Office to protect what I believe will be his next breakthrough…

Something he called “the greatest tech invention in history.”

Elon is predicting this new AI breakthrough will unleash…

A $1 quadrillion new wealth wave.

That’s more than 30 times bigger than the entire U.S. economy.

And just to give you an idea of how much wealth we’re talking about...

That would be enough to send a check for $2.8 million to every single American.

Click here to see the details because I believe this invention will make a lot of people rich.

THE OPERATION

Why the two numbers disagree

The two indexes are built differently on purpose. The PCE index counts all consumption, including spending made on households’ behalf rather than out of their pockets, which is why employer-paid health insurance sits inside it. Its weights are updated monthly. The CPI is built around what urban consumers pay directly, and shelter carries a far larger share of it. Two honest measures of the same economy, weighted for two different questions.

What matters this week is the direction of the gap, because it is the wrong way round. Historically the PCE reading runs below the CPI reading. PIMCO put the normal spread at negative 30 to negative 40 basis points, and in May it recorded the spread as having flipped to positive 60, calling it one of the largest reversals since 1985. The July prints widen it to positive 80. Measured from the historical middle, that is a swing of about 115 basis points in the relationship between the two numbers.

The components doing it are the ones weighted more heavily in PCE than in CPI. PIMCO’s list runs through technology hardware and software, where memory and computing prices accelerated on chip and server demand, then information and communication services, then healthcare, with streaming and cellphone pricing running above normal seasonal patterns and tariff pass-through arriving unevenly across categories.

Read that list again and notice what it implies. The AI buildout has become large enough to move a national price index, and it moves the one the Federal Reserve is mandated to hit rather than the one that makes the evening news. The same demand that shows up as capital expenditure in a datacentre budget shows up as core inflation in the Fed’s target series.

Not everyone expects it to persist. Capital Economics argued in April that the forces pushing PCE above CPI should reverse by the turn of the year, with core PCE falling while core CPI edges up. That is the reasonable counter-case, and it is a forecast about the fourth quarter rather than about next Wednesday.

There is one more wrinkle in next week’s meeting. September 15 to 16 carries a Summary of Economic Projections, which means the committee will publish its own forecasts for core PCE inflation. It will do that while its latest actual core PCE reading is July’s.

This is why Friday’s CPI print is a weaker guide to the decision than it looks. A soft core CPI number does not move the index the committee is judged against, and a hot one does not either. CPI matters next week only through what it implies about the PCE components that overlap with it, which is most of goods and rather less of services. The parts of PCE currently running hottest, healthcare and the services measured indirectly, are precisely the parts Friday will not tell you about.

RULES OF ENGAGEMENT

Three dates before the vote

Thursday, September 10. August producer prices at 8:30 Eastern. July final demand ran at 4.7% over twelve months. Producer prices feed into the PCE calculation more directly than into CPI, so this is the release that speaks to the Fed’s index rather than the public’s.

Friday, September 11. August CPI at 8:30 Eastern. The number to watch is core against July’s 2.5%. Read it as information about goods prices and about the overlap with PCE, not as a verdict on the meeting.

Do not trade the headline as a policy signal. If core CPI comes in soft on Friday and the market rallies on a dovish read, remember that the index the committee answers to does not get updated until September 30, two weeks after the vote is cast.

Watch the mortgage transmission. Freddie Mac put the thirty-year fixed average at 6.71% in its September 3 survey, a thirteen-month high, up from 6.66% the week before and 6.50% a year earlier. The fifteen-year sits at 6.04% against 5.60% a year ago. With the ten-year through 4.8% today, Thursday’s survey is where a three-year high in the cost of long money starts showing up in household borrowing.

Mark September 30. That is the first update to the index the Fed’s target is written on, and it lands after the decision it should have informed. If the gap is still near 80 basis points then, the October meeting starts from the same problem.

The exposure is a decision taken on a stale reading of the only index that officially counts. The Federal Reserve’s 2% goal is defined on the PCE price index. Its latest PCE reading is July’s 3.3%, published on August 26. The next one arrives on September 30. The vote is on September 16. In between, the committee will receive a CPI report measuring a different index that currently reads 2.5%, publish its own projections for the index it cannot yet see updated, and set the price of money for an economy where the ten-year note already yields more than at any time since October 2023.

END OF TRANSMISSION.

EYES ONLY

EYES ONLY

SUBSCRIBER ONLY // DO NOT FORWARD WITHOUT PERMISSION