Inflation data and economy · Analysis
Inflation data revisions: why a changed comparison is not a new shock
Recent releases make data vintages important. Learn how to compare revised estimates and avoid mixing an old baseline with a new observation.
Why vintage matters now
Following September’s income and price releases, readers encounter tables that may differ from earlier articles. Economic estimates have publication dates as well as measurement periods. This analysis focuses on interpreting revisions; it does not allege a particular unreported correction. Some series are revised on different schedules, so verify the agency’s notes rather than assuming every index follows the same rules.
An invented acceleration mistake
Suppose last month’s growth was initially estimated at 0.4%, then revised to 0.2%, while this month comes in at 0.3%. Comparing with the stale estimate suggests a slowdown; comparing with the new baseline suggests an increase. The two descriptions use different information sets. These numbers are illustrative and are not a replacement for the actual release tables.
Keep two legitimate questions separate
What did investors know at the time of an announcement? What does the latest data suggest happened? An archived first estimate can answer the first question, while a revised series is often needed for the second. Revisions do not automatically imply misconduct, and a valid historical account should not silently rewrite what readers originally saw.
Build a consistent comparison
Save the release date, table identifier, units, and adjustment convention. When a baseline changes, explain the revision before describing the new trend. Our August PCE report highlights why matched vintages matter. This article remains a dated explanation; later revisions need later evidence, not a claim that old forecasts were always based on today’s information.