An inflation observation has a reference month and a release history. The Bureau of Labor Statistics publishes both seasonally adjusted and unadjusted CPI data, and these versions serve different analytical purposes. Its annual recalculation of seasonal factors can revise recent adjusted history without turning the original month into a new economic event. BLS: Using seasonally adjusted and unadjusted CPI data
Follow the evidence
Trace how the event could reach markets, then inspect a competing explanation.
Compare explanations
Switch lenses to see what each account explains—and what remains uncertain.
WFN analysis: a historical research chart usually asks what the latest consistent series says about a period. A study of reactions at the original release asks what information was available at that moment. Downloading the current series answers the first question. It cannot, on its own, reconstruct the second. Both views can be useful when they are labelled accurately. Suppose a hypothetical adjusted monthly change was first shown as 0.2% and later as 0.3%. The difference is 0.1 percentage point. A chart of the latest vintage should show the revised figure, while a release-day review should retain the original observation and its timestamp. These numbers are illustrative and are not claims about a particular actual CPI month. A separate mistake is to switch adjustment basis mid-comparison. An adjusted monthly change and an unadjusted annual change describe different intervals and treatments of the underlying data. Putting both into a briefing can be informative, but only if each value retains its full label. A shortened heading such as inflation change can conceal what is actually being compared.
WFN analysis: a historical research chart usually asks what the latest consistent series says about a period. A study of reactions at the original release asks what information was available at that moment. Downloading the current series answers the first question. It cannot, on its own, reconstruct the second. Both views can be useful when they are labelled accurately. Suppose a hypothetical adjusted monthly change was first shown as 0.2% and later as 0.3%. The difference is 0.1 percentage point. A chart of the latest vintage should show the revised figure, while a release-day review should retain the original observation and its timestamp. These numbers are illustrative and are not claims about a particular actual CPI month. A separate mistake is to switch adjustment basis mid-comparison. An adjusted monthly change and an unadjusted annual change describe different intervals and treatments of the underlying data. Putting both into a briefing can be informative, but only if each value retains its full label. A shortened heading such as inflation change can conceal what is actually being compared.
For a reproducible data file, save the series identifier, reference month, adjustment status, retrieval date and release document. If studying a surprise, also retain the forecast’s original source and timestamp. A revised actual figure should not silently replace the observation used in the original surprise calculation. That would change the historical question being tested. Connecting the observation to the dollar requires another piece of analysis: the expected policy response and the corresponding outlook for the other currency. A seasonal revision can alter a description of past momentum. It does not by itself demonstrate a particular future exchange rate. Explicit vintages make that distinction visible and allow another reader to reproduce the inflation comparison.
The February 2026 revision record
BLS’s notice says recalculated seasonal indexes and factors for January 2021 through December 2025 became available on 13 February 2026. The annual recalculation accompanies release of the January CPI and reflects price movements during the completed calendar year. This is a documented revision process, not a newly announced October inflation reading. BLS: Updated seasonal factors introduced February 13, 2026
BLS explains that seasonally adjusted data remove recurring seasonal influences and are useful for analysing short-term movements. Unadjusted data are commonly used for annual comparisons and escalation arrangements. Adjusted indexes can be revised for up to five years; series outside that window are treated as final under the stated seasonal-revision policy. BLS: Using seasonally adjusted and unadjusted CPI data
Two questions require two data views
WFN analysis: a historical research chart usually asks what the latest consistent series says about a period. A study of reactions at the original release asks what information was available at that moment. Downloading the current series answers the first question. It cannot, on its own, reconstruct the second. Both views can be useful when they are labelled accurately.
Suppose a hypothetical adjusted monthly change was first shown as 0.2% and later as 0.3%. The difference is 0.1 percentage point. A chart of the latest vintage should show the revised figure, while a release-day review should retain the original observation and its timestamp. These numbers are illustrative and are not claims about a particular actual CPI month.
A separate mistake is to switch adjustment basis mid-comparison. An adjusted monthly change and an unadjusted annual change describe different intervals and treatments of the underlying data. Putting both into a briefing can be informative, but only if each value retains its full label. A shortened heading such as inflation change can conceal what is actually being compared.
Keep the release history beside the dollar thesis
For a reproducible data file, save the series identifier, reference month, adjustment status, retrieval date and release document. If studying a surprise, also retain the forecast’s original source and timestamp. A revised actual figure should not silently replace the observation used in the original surprise calculation. That would change the historical question being tested.
Connecting the observation to the dollar requires another piece of analysis: the expected policy response and the corresponding outlook for the other currency. A seasonal revision can alter a description of past momentum. It does not by itself demonstrate a particular future exchange rate. Explicit vintages make that distinction visible and allow another reader to reproduce the inflation comparison.