Those changes in methodology knocked the PCE price index down, but even the new-and-improved version remained far above the Fed’s target.
By Wolf Richter for WOLF STREET.
Today was the day for the annual revisions of the PCE price index by the Bureau of Economic Analysis (BEA). This is the inflation index that the Fed favors for its 2% inflation target. The revisions are undertaken every year, and the data is revised each year for the past five years.
What’s different today is that the BEA also included changes of its methodology for three subcategories of the PCE price index. The first two slashed the “core services” PCE price index; and the third slashed the “core goods” PCE price index:
- Portfolio management and investment advice services
- Legal services
- Computer software and accessories.
The year-over-year “core services” PCE price index was slashed by 34 basis points for July as a result of these methodological changes and other revisions, from an old increase for July of +3.69% year-over-year, to a new-and-improved increase for July of +3.35% year-over-year. And in August, based on this new methodology, the core services PCE price index also increased by +3.35%. Blue = old year-over-year inflation rates; red = new-and-improved year-over-year inflation rates:
Core services account for over 60% of consumer spending. They include housing, healthcare, travel, lodging, transportation services (airline fares, etc.), insurance of all kinds, auto repair and maintenance, communication (cellphone services, broadband, etc.), subscriptions, financial services, recreational activities, memberships, etc.
The “core goods” PCE price index rose by 1.97% year-over-year in August.
But July’s year-over-year inflation rate was slashed by 43 basis points to +1.85%, from the originally reported +2.28%, due to the changes of the methodology for the subindex, “Computer software and accessories.”
Under the old method, the subindex for “computer software and accessories” had spiked by 21.2% year over year in July, driven by the side-effects of the AI investment boom. Today, the new-and-improved July inflation rate for “computer software and accessories” was +12.3%.
Blue = old year-over-year inflation rates; red = new-and-improved year-over-year inflation rates.
Despite those changes in methodology, inflation remained hot.
These reductions of the “core services” PCE price index and the “core goods” PCE price index due to the changes in methodology then pulled down the “core” PCE price index, and the all-items PCE price index, which are the inflation measures that the Fed uses as yardstick for its 2% inflation target.
Despite those changes in methodology, inflation remained substantially above the Fed’s 2% target, with the revised core PCE price index at 3.01% and the all-items PCE price index at 3.4%.
The PCE price index is an alternative to the Consumer Price Index (CPI) by the Bureau of Labor Statistics, whose August data was already released on September 11.
The core PCE price index – which excludes energy and food – rose by 0.25% in August from July.
But July’s previously reported 0.25% month-to-month increase was cut in half today by the new methodology to +0.12%.
Year-over-year, the core PCE price index rose by 3.01% in August. And July’s year-over-year increase was cut by 37 basis points, to +2.98% from the originally reported +3.35% (in the chart below, blue = old year-over-year inflation rates; red = new revised year-over-year inflation rates).
The Fed uses the core PCE price index as one of the yardsticks for its 2% inflation target (dotted purple line) because it provides a view of inflation beyond the spike and subsequent plunge of energy prices. The core PCE price index has been above target since March 2021. The closest it got to target was the new-and-improved +2.6% in April 2025.
The all-items PCE price index rose by 0.31% in August from July, and July’s original month-to-month increase of +0.16% was cut to +0.05%.
Year-over-year, the new-and-improved PCE price index accelerated to 3.42% in August, from the revised 3.36% in July. But the revisions slashed July’s increase by 34 basis points from the originally reported +3.70%.
It is almost funny how inflation, by even this new-and-improved measure, has moved away from the Fed’s 2% inflation target (dotted purple line) since April 2025 while the Fed ignored it and cut its policy rates three times later that year (blue = old year-over-year inflation rates; red = new revised year-over-year inflation rates).
The food & beverage PCE price index was not revised. In August, it rose by 1.9% year-over-year.
This chart shows the price level, not the year-over-year percentage change.
The energy PCE price index was not revised. It spiked by 16.8% year-over-year in August. These are energy goods and services that consumers pay for directly, such as gasoline, utility natural gas, electricity, heating oil, etc.
The subindex for gasoline spiked by 27.8% year-over-year in August.
This chart shows the price level, not the year-over-year percentage change.
Enjoy reading WOLF STREET and want to support it? You can donate. I appreciate it immensely. Click on the mug to find out how:
I trust Wolf’s articles which are data and evidence based. Wolf is clearly calling inflation is still hot.
CNBC headline says – “Fed’s preferred inflation much lighter than expected”.
Stock market is rallying today. I honestly do not believe we will ever see data driven recession or bull markets ever again. Feels like, everything is manipulated and big banks and firms will move stock in a direction they want.
We have inflation through the roof, bond yields have reached decades high and stock market is absolutely unfazed because it is last day of the quarter and market has to reach certain point to close the books and take profit.
Wolf,
Keep doing what you are doing!! News/Articles we can trust.
What are your thoughts on the changes they made? A better measurement or a lame attempt?
There were real issues with the methodology of those three categories, which caused the PCE price index to separate fairly dramatically from the CPI recently, and run hotter than the CPI, when it normally runs a little cooler since it’s a broader index that includes more stuff and has much less emphasis on OER.
Those changes brought the PCE price index roughly in line again with the CPI.
But I would rather that the BLS fixes the CPI by replacing the huge weight of OER with the actual costs of homeownership, and by replacing the methodology of the health insurance index by something that actually tracks health insurance premiums.
Just paid $12 for 12 pack of pop. Inflation going to make me skinny again. No more pop. I’m finally cutting out some off this over valued junk.
I just paid $1.49 for a dozen grade A large eggs, down from the $6 to $8 range not long ago. My favorite coffee got less expensive too. And ground beef has come down a little, bought the first special in a long time. You’re fooling yourself by using one item to judge inflation, because according to the three items I just mentioned we’d have massive deflation 🤣
But yes, I totally agree, stop drinking pop. It’s an unhealthy waste of money. Tap water is good. If it tastes funny where you are, squirt a little lemon into it.
My well water has a consistency of 90% water. 8% iron. 2% Radon. Maybe I’ll develop super powers and my super power will be Deflation-Man. I can schwakathoonkle the inflation rates for the Fed. Ker-CHOW!
Rather than do the right thing and stop inflation, they – all of the a$$clowns running the government – are going to redefine what inflation is. It’s a sick, sick county.
Monkeys outta the bottle bro
The government is cooking the books again.
The problem with this statement is that it’s dumb and makes rational discussions of the real issues impossible.
There were real issues with the methodology of those three categories, which caused the PCE price index to separate fairly dramatically from the CPI recently, and run hotter than the CPI, when it normally runs a little cooler since it’s a broader index that includes more stuff and has much less emphasis on OER.
Those changes brought the PCE price index roughly in line again with the CPI.
Yes Wolf, but what if we are not in normal times (which I would suggest we are not due to the inflationary pressures of tariffs, higher interest rates and AI) and in those adjusted categories, companies are finding it difficult to pass on their higher input costs to their customers. If that is true, it’s important information and government statistics should not try to cover that up.
Neither the adjustments nor what I said has anything to do with what you said. Re-read my comment.
the Fed’s target…we hear that statement as if its proven model. Seems like we have a great deal of comfort in assuming that 2% – has somehow analytically been arrived at through decades of economic success with this target. It would seem that this is an arbitrary target that a group of individuals in the 80’s and 90’s decided was a perfect place for the economy to stay. In reality, in order to achieve 2% the unemployment rate needs to be very high – over 5% or 6% or more…which creates a need for rate reductions and then the hike towards more growth and inflation. Could it be that we need to rethink this the next decade and use our newly gained knowledge (since 1990) to better approach the balancing of the growth and inflation cycles?
2% is a compromise: It’s low enough to where it doesn’t change behavior in the economy by businesses and consumers; and high enough to stay away from deflation and help a little dealing with the national debt.
The 2nd qtr. gross domestic product has been revised higher, now @8.5%.
The FED is still running the economy hot.
yes, nominal GDP, sizzling. Inflation, sizzling.