August PCE due Wednesday at an expected 3.7%, but BEA revisions could cut July's core to about 3%

The Commerce Department's Bureau of Economic Analysis publishes August personal income and spending on Wednesday, September 30, including the personal consumption expenditures price index, the Fed's preferred inflation measure. Economists polled by Dow Jones expect prices to rise 0.3% on the month at both the headline and core level, leaving the annual rates at 3.7% and 3.3%, unchanged from July, CNBC reported. Consumer spending is forecast to rise 0.8% after 0.2% in July, helped by higher gasoline prices.

This release is not a normal month, and the difference matters for anyone reading the headline.

The revision nobody should mistake for progress

Wednesday is also the first day of the BEA's annual update, which this year revises most figures from the first quarter of 2021 through the first quarter of 2026, the agency said in an August notice. It is changing how it prices three parts of the PCE index:

• Portfolio management and investment advice. The BEA will stop deflating spending with a producer price index and instead estimate the quantity of service from employment in the industry.
• Legal services. The BEA said the unpublished consumer price index it had been using "exhibited erratic changes that cannot be corroborated." It is replacing it with a composite of producer price indexes, starting with 2024.
• Computer software and accessories. A new composite that adds producer prices for game software and IT hosting services to the consumer index.

Wall Street estimates cited by CNBC suggest the changes lower July's annual PCE readings by two to three tenths of a percentage point, possibly to about 3%. Applied to core, that is 3.3% becoming roughly 3.0% to 3.1%.

Here is the trap: if Wednesday's report shows a lower annual rate than the 3.7% and 3.3% everyone has been quoting, much of that gap will be the new method rewriting history, not prices cooling in August. The number that says something about the trend is the monthly change. A 0.3% monthly rise, the consensus for August, works out to about 3.7% at an annual pace. The Fed's 2% goal requires monthly readings of about 0.17%.

Why the Fed is watching the month, not the year

New York Fed President John Williams said on Tuesday that inflation is running at 3.7% and that he expects 3.5% for 2026, according to his speech. He and Governor Michael Barr both signalled more tightening, with Williams putting one more hike "late this year." A revised-lower history does not change those forecasts much on its own; a hot August month would. We covered what Williams said here.

Who it hits

The expected 0.8% jump in spending is also less cheerful than it looks for a lot of Main Street. If a large part of it is gasoline, that is money flowing to fuel rather than to the restaurant, the salon or the hardware store. Owners who track their own card volume against last year will see this first: nominal household spending can rise while the share left over for discretionary purchases shrinks. Consumer confidence fell to 81.9 in September, as we reported this morning.

For borrowers, a hot monthly print would add to the case for an October hike; a soft one would support Williams' "no urgency" line. Either way, the annual number alone will not say which one happened.

What to watch Wednesday

• Monthly core PCE: 0.3% is the consensus; 0.2% or lower would be the first real sign of relief.
• Revised July annual rates: how far they fall tells you how much of any drop is methodology.
• Real spending, after inflation: whether the 0.8% nominal gain holds up once prices are stripped out.

Sources: CNBC; U.S. Bureau of Economic Analysis; BEA release schedule; Federal Reserve Bank of New York. Annualized figures are our calculations. This is market information, not investment advice.