Skip to content
TerminalOriginal observation

Consumer prices rose. Used vehicles fell.

Aug 2025 to Aug 2026

At publication ·

Change over this period

Consumer prices+10.84 index points
Aug 2025
323.291 index (1982–84 = 100)
Aug 2026
334.131 index (1982–84 = 100)
Used vehicles−4.29 index points
Aug 2025
185.259 index (1982–84 = 100)
Aug 2026
180.965 index (1982–84 = 100)
Source and recorded evidence →
The original test

Next evidence specified: Both series for Sept 2026, with the matching earlier observations. This is an observation period, not a release date.

What would change it: the next matching changes in Consumer prices and Used vehicles no longer have opposite signs.

See the recorded test →

The memo

The Weight of an Exception

Consumer prices stand at 334.131 for August 2026, up 10.84 index points from 323.291 in the same month a year earlier. Used-vehicle prices stand at 180.965 for August 2026, a change of -4.29 index points from 185.259 in the same month a year earlier. Both use the same CPI reference base. The same annual comparison points in opposite directions inside the same price family.

The exception defeats a claim that the supplied price measures are rising uniformly. It leaves much else unresolved. Opposite signs cannot establish that increases are becoming less widespread, that the aggregate is approaching a reversal, or that the relationship between these series has changed. The next matching annual comparisons could cease to have opposite signs. Even persistence would establish only that this particular disagreement had continued.

Franc wants the August 2026 change of -4.29 index points from the used-vehicle level of 185.259 a year earlier to establish that the broader price rise has begun to reverse. His position spends the representativeness of that component. If used vehicles remain below their year-earlier level while other categories remain above theirs, he must concede that his chosen evidence can persist without establishing his larger conclusion.

There is already evidence against that conclusion. For August 2026 against the same month a year earlier, core prices are 337.765 against 329.7, and shelter prices are 430.227 against 417.555. These are index levels on the same base as the opening comparison. Food, energy and medical prices also show positive annual changes over those months. The supplied categories have not joined used vehicles in a general decline.

Those comparisons establish direction within each series. They do not assign contributions to the aggregate increase. The supplied index-point changes are not percentage inflation rates, and the categories cannot be treated as independent votes: the aggregate and its components overlap. The evidence gives the used-vehicle exception a precise scope. Extending it into a claim about the whole basket requires something beyond its negative sign.

The week's editorial record shows how easily that scope could be lost. The desk considered the consumer-price and used-vehicle divergence in the opening two editions and declined it both times because the release cadence was incompatible with another treatment. The signal had not been released again since its previous appearance. Those refusals concerned whether there was another observation to examine, rather than whether the underlying argument had been disproved.

Its selection in this edition therefore cannot, by itself, establish that the divergence first appeared this week. The record identifies the earlier cadence problem, but it does not identify an intervening release date or provide the earlier price figures. The August comparison now available supports the opposing signs. The sequence of editorial decisions supplies no additional measurement of how those signs developed.

That distinction matters because the most persistent refusal concerned a different temptation: treating an unusual level as an argument on its own. Refinery utilisation was considered in every listed edition. Every time, the desk found no complicating evidence. Across the week's record, the proposed argument repeatedly lacked an observation that put its interpretation at risk. Repetition of the proposal did not supply that missing opposition.

The current evidence gives that refusal substance. Refinery utilisation is 97.8% for the week ended 2026-09-04, against a mean of 92.84% over the supplied window of fifty-two weeks ending 2026-09-04. Its position above that mean is established. The repeated editorial refusals do not establish that utilisation stayed unchanged throughout the week, and the comparison itself supplies no mechanism connecting refining to the opposing annual price changes.

Hours worked encountered a related boundary. The opening two editions declined its contextual extreme because the editorial move had just been made. This edition declined it because there was no complicating evidence. Those are different limits: an argument can be familiar, or it can lack a counterweight. Neither reason demonstrates a reversal in hours. A changed reason for leaving a subject aside is not another observation of the subject.

The work argument that did get selected twice also limits what can be inferred from the eventual change of focus. Both opening selections concerned opposing declared comparisons within work. This edition considered the comparison of construction and government employment and placed it behind prices. Being outranked records an editorial preference among available arguments. It does not record that the earlier work premise failed.

The current employment evidence preserves the distinction. Construction employment is higher in August 2026 than in the same month a year earlier, while government employment is lower across those same months; both estimates are provisional. That is a matching comparison with internal disagreement. Its absence from the lead cannot establish that employment has acquired a common direction. Prices occupying the lead tells us which argument the desk selected, without settling the other one.

Taken together, these decisions explain what persisted across the week more clearly than a succession of headlines could. The desk repeatedly encountered facts that could support a bounded statement but could not carry the proposed extension. A monthly observation could not become fresh through another edition. An elevated level could not acquire tension through repeated consideration. An available divergence could lose priority without losing its factual basis.

The price exception passes a different test: it directly challenges uniformity within matching annual comparisons. Yet the evidence that makes it useful also limits it. Used vehicles supply an opposing sign, while the aggregate and the other supplied categories retain positive signs. Franc's general reversal would require those observations to change in ways the current comparison does not establish. Selection gives his starting fact attention, not his conclusion additional support.

The monthly CPI release provides the next opportunity to compare consumer prices and used vehicles over another matching annual period. If their changes no longer have opposite signs, that would cut against persistence of this particular divergence. It would not automatically establish a general price reversal. If the signs remain opposed, the result would extend the observation without proving a cause, a changed relationship, or Franc's wider claim.

The week's record therefore supports an account of variation within the measures it examined, with specific limits on freshness, interpretation and persistence. It does not support turning a change of editorial subject into a change of economic regime. The used-vehicle decline matters because it qualifies what the aggregate can stand for. The positive changes elsewhere matter because they qualify what that exception can stand for.

Used vehicles have supplied an exception; the broader reversal still lacks evidence.

All observations from this edition

Observed 13 Sep 2026, 14:00 UTC

Public debt
40.05 T
10 Sept
Treasury cash
818.1 B
10 Sept
Average rate paid
3.49 %
31 Aug
Average rate — bills
3.79 %
31 Aug
Average rate — notes
3.35 %
31 Aug
Average rate — bonds
3.45 %
31 Aug
Consumer prices
334.1
Aug 2026
Core prices
337.8
Aug 2026
Food prices
350.3
Aug 2026
Shelter prices
430.2
Aug 2026
Employment
159.07 M
Aug 2026 provisional
Hourly pay
37.8
Aug 2026 provisional
Hours worked
34.4 h
Aug 2026 provisional
Output
1.5 %
2026Q2
Consumption
3.4 %
2026Q2
Investment
2.7 %
2026Q2
Crude output
13.95 Mb/d
4 Sept
Crude stored
424 Mbbl
4 Sept
Refining
97.80 %
4 Sept
Gasoline stored
207 Mbbl
4 Sept
Gas stored
3254 Bcf
4 Sept
Electricity made
355 TWh
May 2026
Debt owed outward
32.36 T
10 Sept
Debt owed inward
7.68 T
10 Sept
Exports
4.5 %
2026Q2
Imports
12.5 %
2026Q2
Government
-1.0 %
2026Q2
Energy prices
321.1
Aug 2026
Medical prices
592.9
Aug 2026
Used vehicles
181.0
Aug 2026
Factory jobs
12.64 M
Aug 2026 provisional
Building jobs
8.36 M
Aug 2026 provisional
Temp jobs
2.52 M
Aug 2026 provisional
Public jobs
23.32 M
Aug 2026 provisional
Long rate
4.96 %
11 Sept
Short rate
4.07 %
11 Sept

WORLDSTATE_2d46e3f8f137 · cd067cd58257