The release of the November CPI data Consumer Price Index has kicked off a huge discussion. Analysts and observers have been warning for literally years about the gaping holes in the math behind Owners’ Equivalent Rent (OER). OER has an outsize role in CPI, perhaps the most important price index. It is based on a question posed to homeowner respondents in the Consumer Expenditure Survey. So the U.S. Bureau of Labor Statistics (BLS) had a hard time figuring out the right data to collect in October 2025. Consequently, they used bad methodologies in the follow-up November report.
Because of a lapse in appropriations, the BLS was unable to collect survey data for October 2025. With this rollout, the agency was again met with a data drought. It did this by computing the November indexes based on the price change from November 2025 to October 2025, even in the absence of any data in November. Yet BLS officials justified their decisions based on the premise that this was the case – that October prices were like those of September. This decision has drawn criticism from many pro-market economists.
All of this was made worse by a surprising drop in OER, with September posting just a 0.13 percent increase. This decline was a record-setting 2.4 percentage-point decrease for the third month in a row. That set off red flags on the accuracy of the data. Over the last three months OER has been flat with an annualized rate of just 1.6 percent. This is a steep increase from the 4.1 percent average rate in the six months prior to the revised data from September.
OER is hugely influential on general inflation measures. It is 26 percent of the CPI, 33 percent of core CPI, and an even bigger 44 percent of the core services CPI. As it represents more than 40% of the OER, any miscalculations can cause dramatic cascades of misguided alarms about economic instability.
The November CPI data was overall quite cool as well — even beyond expectations — much of this driven by the OER outlier. The BLS was able to retroactively collect most of the nonsurvey data for October. They ran into challenges with OER calculations that led to analysts on the back end of the data calling the data “made up.”
“BLS did not collect survey data for October 2025 due to a lapse in appropriations. BLS was unable to retroactively collect these data. For a few indexes, BLS uses nonsurvey data sources instead of survey data to make the index calculations.” – source unspecified
Despite the challenges, BLS met the task head on and resumed CPI data collection on November 14, 2025. Their goal was to collect data for the full month of November.
“Collection began on Friday, November 14. By authorizing additional collection hours, BLS attempted to collect data for the entire month of November.” – source unspecified
The theory and process on which November’s indexes were derived has faced criticism. As the BLS observed, these indexes were calculated by comparing prices collected in November 2025 with those collected in October 2025.
“How were November indexes calculated? November 2025 indexes were calculated by comparing November 2025 prices with October 2025 prices.” – source unspecified
Critics claim that this approach missed key data quality checks and resulted in the wrong conclusion. The accuracy of OER numbers is very much in doubt. These figures are based on self-reported surveys by homeowners on what they think they would get if they rented it out.
“If someone were to rent your home today, how much do you think it would rent for monthly, unfurnished and without utilities?” – source unspecified
The reliance on previous months’ data without actual verification has only fueled skepticism regarding the integrity of the CPI figures. It’s those September data that watchdogs have argued were the foundation for the doubted math behind the bad October data. These estimates were subsequently compounded by a major outlier drop in that month.
“The September data, which was used as base for the made-up October data, was marred by the total outlier plunge of OER… And that outlier plunge in September was carried forward to October and November.” – source unspecified
Indeed, economic commentators have sounded alarm bells over these changes. This sharp drop in OER could lead one to doubt even further the accuracy with which inflation is being measured and communicated.
“[OER] has been at an annualized rate of 1.6 percent for the past three months, compared to an average 4.1 percent in the six months before the doctored September. That’s a sudden 2.4 percentage-point plunge out of nowhere for the third month in a row.” – WolfStreet