Eve is here. As these headlines indicate, the mainstream financial media is openly skeptical of the latest US inflation reports. Let’s start with the Wall Street Journal.
Wolf Richter explains the data fabrication used by the BLS and how it began before this inflation report.
Keep in mind that this kind of statistical manipulation is not only bad in itself, but can also lead to, among other things, intentionally setting inflation adjustments too low or overstating GDP. This is the kind of thing that would happen before the financial crisis, when officials were making adjustments to keep the party alive. In the run-up to the financial crisis, analysts like Michael Shedlock and Barry Ritholtz criticized the “birth/death adjustment” of job creation data, but the BLS added this in light of the fact that its surveys do not capture business creations or failures that, too often, were unusually large and increased the level of reported profits. In 2007, we included fake official statistics in our list of Banana Republic indicators. Readers can add to this list:
Complex with private security for wealthy people
Economic/class mobility is restricted
Militarization, exaggeration of external threats
Income and asset ownership are highly concentrated at the top
Government policies were heavily biased towards the very wealthy. Looting the treasury
Freedom of the press is/is not restricted
Election fraud. In the extreme version, coup d’état, one-party rule
Judicial Independence/Assault on Kangaroo Courts
So while this deceptive data may seem like a lot of money when taken alone, it is yet another indicator of America’s continued decline.
Written by Wolf Richter, editor of Wolf Street. First appearance: Wolf Street
The Bureau of Labor Statistics today said in its November CPI report that most of the October data was missing and some of the November data was missing, and that it had filled gaps in the November data by “approximating missing data points,” such as rent equivalents (OER), the largest component of the CPI, which accounts for 26% of the total CPI, 33% of the core CPI, and 44% of the core services CPI.
OER experienced a suspicious outlier plunge in September, and that September suspicious outlier plunge carried over into October and November. And the BLS even explained some of it in a separate memo, so it’s not a secret.
This is a screenshot of the CPI summary table. You can see that almost all month-over-month entries for October and November are missing. The exceptions are entries where the BLS relies on “non-survey data” such as gasoline prices and new and used vehicle prices (vehicle data is purchased from J.D. Power).
What the BLS says about missing data and how to deal with it
“BLS did not collect survey data in October 2025 due to expiration of appropriations. BLS was unable to collect these data retrospectively. For some indexes, BLS did not collect survey data in October 2025. Instead, non-survey data sources are used for index calculations. BLS was able to obtain most of the non-survey data retrospectively. CPI data collection was summarized on November 14, 2025.
BLS briefly outlined some of the other shortcomings of this CPI release in a separate memo.
“What was the impact on November data collection? Collection began on Friday, November 14. By allowing additional collection time, BLS attempted to collect data for the entire month of November.”
It read, “Attempted recovery.”
And this is a bad joke. “How was the index for November calculated? The index for November 2025 is the same as the November 2025 price and ” However, prices for October are not present in the data… “Since BLS was unable to collect survey data for the October 2025 base period, the survey data was carried forward from September 2025 to October 2025 according to normal procedures.”
In other words, the BLS just made up the October data.
And the September data used as the basis for October’s fabricated data was marred by a total outlier plunge in OER, which accounted for 26% of total CPI, 33% of core CPI, and 44% of core services CPI. And the sharp drop in abnormal values in September carried over into October and November.
Specifically regarding OER, “BLS uses six months of panel collections to calculate rents and owner equivalent rents.” [surveys are sent to the same address every six months, instead of every month].
So there was this questionable outlier drop in September, and instead of recovering as it should have, it carried over into October and November, resulting in a very interesting graph like the one below.
Using OER’s BLS index data provided today…
August: 430.69
September: 431.27
October:
November: 432.44
…This is the annualized monthly percentage change in the clearly falsified OER. The annual rate for the past three months has been 1.6%, compared to an average of 4.1% for the six months to September for doctoral programs. This is the third straight month of sudden 2.4 point declines out of nowhere.
And this tampered component accounts for 26% of total CPI, 33% of core CPI, and 44% of core service CPI, making the entire CPI data a bad joke or worse.
But it’s not a bad joke, it’s much worse
Many things depend on CPI, including calculating the “inflation protection” for Treasury Inflation Protection Securities (TIPS), Series I savings bonds that the government sells to retail investors, Social Security COLAs, and other inflation adjustments paid to investors and beneficiaries, all of which will be undervalued relative to inflation.
This data also feeds into broader inflation-adjusted economic data, such as “real” consumer spending and “real” GDP. This is because BEA, which produces these overall economic indicators, uses some of this CPI data, including OER, to calculate the PCE price index and GDP deflator, among other things.
The BLS is currently causing serious problems for investors and beneficiaries, with inflation protections being ignored and inflation-adjusted economic indicators being inflated, all of which will, of course, play to the administration’s side.
