OER weighs 26% of CPI, 33% of core CPI, 44% of core services CPI. It moves the needle. CPI inflation would have been a lot hotter without this outlier situation.
By Wolf Richter for WOLF STREET.
The delayed release of the Consumer Price Index today, cobbled together with perhaps not all the staff and means that the Bureau of Labor Statistics has normally available due to the government shutdown, was perhaps the best that could be done under the circumstances.
But there were a few things that were off, the most important of which was Owner’s Equivalent of Rent (OER), a huge component in CPI, accounting for 26% of overall CPI, for 33% of core CPI, and for 44% of core services CPI: It was a massive historic outlier.
OER rose by only 0.13% in September from August (blue line in the chart), according to the BLS today, compared to 0.38% in the prior month, and compared to the 12-month range between +0.27% (May) and +0.41% (July). Something went wrong there, and given its huge weight, OER significantly pushed down the month-to-month readings of overall CPI, core CPI, and core services CPI.
If this situation with OER hadn’t happened, the inflation readings today would have been a lot hotter than they were, particularly core services CPI where OER weighs 44% and core CPI where OER weighs 33%.
OER is not a measure of rent. The measure of rent is the Rent CPI. OER is a stand-in for the costs of homeownership. OER indirectly reflects the expenses of homeownership such as homeowners’ insurance, HOA fees, property taxes, and maintenance. It’s the only measure for those expenses in the CPI. It is based on what a large group of homeowners estimates their home would rent for, with the assumption that homeowners would try to recoup their cost increases by raising the rent.
Overall CPI rose by 0.31% (+3.8% annualized) in September from August. So not a benign inflation reading, but the second worst since January. And yet it was pushed down by the OER outlier situation (blue in the chart below).
Year-over-year, overall CPI rose by 3.01%, the worst increase since May 2024, and the sixth acceleration in a row.
It would have risen more without the OER outlier (red).
“Core” CPI, which excludes food and energy components to track underlying inflation, clearly showed the massive push-down from the OER outlier.
It rose by only 0.23% (2.8% annualized) in September from August (blue) which caused the year-over-year increase to decelerate to +3.0% (red).
Core services CPI, where OER weighs about 44%, was heavily impacted by the OER outlier: It rose by only 0.24% in September from August (+2.9% annualized).
Core services are about 60% of the overall CPI. They include many of the essentials that consumers cannot do without – housing (OER!), medical care, health insurance, auto insurance, tenant’s insurance, subscriptions; telephone, internet, and wireless services; lodging, rental cars, airline fares, education, movies, sports events, club memberships, water, sewer, trash collection, motor vehicle maintenance and repair, etc.
But with 44% of it gone haywire in some way, it doesn’t really indicate anything for September.
The CPI for durable goods dipped in September by 0.13%, and year-over-year, decelerated to +1.8%.
| Major durable goods categories | MoM | YoY |
| Durable goods overall | -0.1% | 1.8% |
| New vehicles | 0.2% | 0.8% |
| Used vehicles | -0.4% | 5.1% |
| Household furnishings (furniture, appliances, floor coverings, tools) | 0.2% | 3.0% |
| Sporting goods (bicycles, equipment, etc.) | 1.0% | 0.0% |
| Information technology (computers, smartphones, etc.) | -0.8% | -5.1% |
The chart below shows the price level, not the percentage change, of the durable goods. The index had spiked by 25% during the pandemic and then prices came down some from those highs, starting in mid-2022. The index is dominated by new and used vehicles. Used vehicle prices had spiked by 55% from mid-2020 to mid-2022, then declined. But 13 months ago, in August last year, used vehicle prices started zigzagging higher again on tight supplies.
Given how messed up the core services CPI is due to the outlier situation of OER, it doesn’t make much sense to delve into services inflation further at this point.
I will post a special and detailed analysis on food inflation and on energy inflation later today. Lots of interesting stuff happening there. Stay tuned.
Enjoy reading WOLF STREET and want to support it? You can donate. I appreciate it immensely. Click on the mug to find out how:
Bad situation. Either the government released data with a major error, or the government released manipulated data to get an inflation reading it wanted.
Let’s see what the “fake government data” hawks that have circled for a few years say about this one.
This really doesn’t look good.
There are other things that I noticed. For example, there were no revisions of prior data at all. None. I’m not sure I have ever seen that happen before. Since revisions go in both directions, this surely was due to a shortage of staff and time due to the government shutdown. They just did what they could to get something out.
It would be prudent for the government to disclose this frankly, right? All they need is a one-liner stating something along the lines of “Data provided in this CPI report will be revised in the future; due to shorter staffing during the government shutdown, various inflation-related measures and indices, in addition to evaluations of prior published CPI reports, should be considered preliminary.”
This could very well be a case of negligence rather than deliberate data manipulation. I am basing that off the absence of prior data revisions or seasonal adjustments. Tight deadlines with short staffing produce reports with these kinds of errors and omissions.
I looked for this one-liner but didn’t find one. But instead, I found that the BLS said that, due to staff shortages, contacting it and getting responses via email was limited. So it’s hard to reach out to them and get a reasonable answer. I have reached out to them before on a technical issue for durable goods, and they did reply back with a good answer. But I guess not today.
I would agree that the OER would not be accurate during the government shutdown. However, as a real estate broker in Central California, I would not be surprised to see OER declining. With the new home and resale markets being soft, it is historically true that rents will soften too. I am seeing asking rent reductions and special offers (free months rent.)
I had this consideration also. I am seeing a lot of price drops on the Zillow (somewhat unrelated).
Just looking at indicators of conditions and sentiment.
If the sentiment soured, could this be a real data point? Any “survey” based data is ultimately subjective (even if it is consistently so).
No wait a minute. OER has ALREADY cooled a lot over the past two years, reflecting what you say. That was already in it before today.
The housing market is slow moving. There are NO BIG CHANGES month to month. It’s like watching paint dry. Which is why today’s OER is such a historic outlier.
Perhaps they should change BLS to BSL and lose the L.
So if you just lose the L without rearranging you still arrive at the same place.
Wolf, couple typos I think. “ It rose by only 0.23% in September from August (blue) which caused the year-over-year increase to decelerate to +2.8% (red).” The graph shows 3.0% for the red line.
Also, CPi in the graph has a lowercase “I”.
As always, greatly appreciate your work and speedy reporting + analysis!
Using OER is the most ridiculous measure of rent there is. It should be removed entirely.
OER is NOT a measure of rent. The measure of rent is the Rent CPI.
OER is a stand-in for the costs of homeownership. OER indirectly reflects the expenses of homeownership such as homeowners’ insurance, HOA fees, property taxes, and maintenance. It’s the only measure for those expenses in the CPI. It is based on what a large group of homeowners estimates their home would rent for, with the assumption that a homeowner would try to recoup their cost increases by raising the rent.
Right, but it’s still a crap measure. Most homeowners have no intention of renting out their homes, and don’t have a clue what they would rent for. They literally just make shit up when asked this.
Here’s what the BLS says on the factsheet:
“The shelter service that a housing unit provides to its occupants is the relevant consumption item for the CPI. Most of the cost of shelter for renter-occupied housing is rent. For an owner-occupied unit, most of the cost of shelter is the implicit rent that owner occupants would have to pay if they were renting their homes, without furnishings or utilities.
Owned housing units themselves are not priced in the CPI Housing Survey. Like most other nations’ economic statistics programs, the CPI program views owned housing units as capital (or investment) goods distinct from the shelter service they provide, and therefore not as consumption goods. Spending to purchase and improve houses and other housing units is treated as investment and not consumption in the CPI. Interest costs (such as mortgage interest), property taxes, real estate fees, most maintenance, and all improvement costs are part of the cost of the capital good and are also not treated as consumption items. These non-consumption costs of owned housing are out of scope for the CPI under the cost-of-living framework that guides the index.”
While well-intentioned, I do not believe it makes sense. The bit about how interest, upkeep, property taxes and other “homeownership” costs are really “capital good” expenses and not consumer price items is farcical.
Alternatively, if the BLS really wants to distinguish “consumption” vs. “investment” costs for housing, it needs to recognize that everyone needs to maintain an ongoing “investment” in their housing, in order to have a place to live. So in addition to a “consumer price index” there would also need to be a “residential housing investment price index” to properly account for the real-world costs of housing.
improvements or increases in lifetime would be capex. simple maintenance gets treated as opex
This checks out. Anecdotally just in August and September my homeowners insurance increased 15% y/y, my property taxes went up 10% y/y and my HOA contribution went up 20% y/y. The increases are not relenting. I’ve had to up the amount I save each month to cover these payments twice now this year.
Maybe house prices are decreasing but ownership costs are going up.
Thanks for reporting this, this tid bit seems to have gone over everyone’s head on Wall Street today.
Please forgive if I missed it, but does OER include utilities (water, sewage, gas, electricity)…if not, then where does CPI capture it?
These are separate line items in CPI:
Under energy:
-Electricity service
-Utility natural gas to home
-Heating oil, propane, kerosene, firewood
Under core services:
-Water, sewer, trash collection services
Excellent find, I was skeptical when print originally released due to staff shortages, now I know why.
Yes b/c of Government Shutdown, accuracy of data is questionable.
But Wolf, if the OER did not show such an outlier low number, and came typically in the range or like it did in August, what do you think the Nominal CPI Inflation and Core CPI Inflation would have been ?
I would think that if the administration were going to fudge the numbers, it would be in ways difficult to notice — a nip here, a tuck there. Then again, when I think back to the ridiculously poor quality of the Liberation Day chart….
A lazy kid in junior high with an understanding of percentage and CPI weights would have done it exactly this way, if asked to manipulate CPI. A hard-working kid would have tried to hide it a little.
Wolf – we could borrow (no interest) ten clowns from the circus who could give us information that was just as, or more reliable and accurate.
The aforementioned clowns have a high school education at best, some of them dropouts.
You get the idea – no geniuses putting out these reports.
What is happening with all of those corporate profits?
I thought prices weren’t going to spike with tariffs and consumer pullback because record profits were going to eat the tariff costs.
Why are prices going up now? Are we back to profit padding?
Rich people and retirees are still spending like there is no tomorrow, and poor people don’t have anywhere else to cut and don’t really matter in the aggregate anyways due to wealth inequality, so what is going on here? Are the wealthy consuming so much that they are pushing prices up?
Services prices are going up. Services are not tariffed.
Since the July through September data sets the 2026 COLA, an inflation number skewed lower than actual inflation will shortchange millions of people with wages or benefits tied to the number.
Yes, but the good thing is the the COLA is based on a three-month average, so only one-third of whatever error was introduced today would apply to the COLA.
If one assumes OER was actually in the middle of its range for the past 12 months (+0.34%) instead of the reported outlier (0.13%), then:
(1) Core CPI (with 33% OER weight) would be up 0.30% instead of 0.23% over the month, and up 3.07% over the past year instead of 3.0%
(2) Core Services CPI (with 44% OER weight) would be up 0.33% instead of 0.24% over the month, and up 2.99% or 3.0% over the past year, instead of 2.9%.
And the Fed might not be so enthusiastic about rate cutting.
Bagehot’s Ghost 👍
In that scenario:
Core cpi mom annualized 3.6%
Core services mom annualized 3.96%
Hmm…
Lack of staff? Deliberate manipulation? Banana Republic stuff.
Hopefully some like minded individuals can get the word out. Won’t hold my breath. Fox News covering the important stuff on their front page, this time its the Clintons and Amy Schumer’s abs.
“Lack of staff?”
Didn’t you hear? government shutdown, government employees included at the BLS got furloughed.
So then the BLS hastily recalled some but not all of its employees to start belatedly working on the CPI report. But surveyors are still not in the field, and many roles that work on the CPI data are likely also unmanned currently. This has got to be a huge mess.
If the OER number is a mistake or a lie, then the Dow just hit a record high based on a mistake or a lie. If the OER number is real, and is in fact data accurately reflecting reality, then homeowners appear to be on their way to a world of hurt. The OER number needs to be explained by BLS, which apparently won’t happen. To see if this OER is indeed an outlier, or the beginning of a sharp downward trend, we will just have to wait for the next report, which, oh BTW, may not happen for quite while.
Without government-generated economic data, which wasn’t all that good to begin with, we are flying blind. The best course of action for the Fed is to stand pat until data starts flowing again. Powell’s favorite term, “data-dependent”, does not mean much if there is no data. Of course, the Fed will, in all its stupidity, lower rates at the next meeting.
They had to get this out so they could have a number for the COLA adjustment for next year. If it werent for that they wouldn’t have done it at all. And with about 40% of numbers imputed, along with these outliers, it looks like they could have easily found a way to bring that number in low to save a few bucks on next years SS outlays.
In the past shelter was raising the CPI calculation, this is the first time that shelter being doing pushed the CPI down.
After firing and/or furloughing all of the serious people at BEA, I’m not sure why much stock is still being put in these releases. That the topline number came in lower than expectations on account of the OER number being light (an imaginary metric to begin with) just reinforces the point.
The CPI is now little more than state-sponsored propaganda.
But leave it to the MSM to run with the government’s preferred takeaway… Inflation is under control, and the Fed should make another rate cut soon.
I’m sick of it.
The MSM is a tool of the Democrat Party and of Wall Street. Makes you wonder who composes the Democrats.
For those waiting with bated breath for the next inflation report, your breath is likely to get more bated:
“WASHINGTON, Oct 24 (Reuters) – The White House said on Friday it had learned there likely will not be a release of inflation data next month due to the U.S. government shutdown, which could leave a gap in a data series stretching back more than a century.
Because surveyors cannot deploy to the field, the White House has learned there will likely NOT be an inflation release next month for the first time in history,” the White House said in a statement.”
What a convenient time for the government to shutdown. I don’t think I’ve ever seen you question data before. So for you to question it, that really says something about what is going on.
I have “questioned” a lot of data, including cussing out publicly and multiple times the method for the health insurance CPI. To see that, google:
health insurance CPI chickenshit
My stuff on it comes up right at the top. Even AI is speechless on this topic 🤣
The health insurance CPI continues to be chickenshit.
Beginning next year, I believe household health insurance pemium costs will surpass annual rent. The PCE weights are FUBAR. They will never be able to close the gap in a way that maintains time averaged continuity with previous weights.
“household health insurance premium costs will surpass annual rent.”
For the overall US population on average, that’s just nonsense. Companies pay a big portion of health insurance for their employees and sometimes for their families. The portion that companies pay doesn’t go into CPI/PCE. People over 65, on Medicare, pay relatively little in terms of health insurance. People on Medical pay little in terms of health insurance. There are many cheaper plans under the ACA. People with high-deductible plans with Health Savings Accounts have relatively manageable premiums (we’ve had them since 2006, so don’t give me any shit about them). All this is what CPI/PCE health insurance attempts to capture.
CPI/PCE health insurance is chickenshit, but your comment is nonsense.
Enron math today! If we keep increasing debt and failing to control inflation we will end up like Argentine. Speaking of cryptocurrency ponzi Milei, his country midterms are the next data drop coming up. I read no cpi inflation data coming out in November, regardless of government shutdown or opening. The trend up is intact! Have a good weekend!
OER – a very strange and hard to measure connotation (Owners equivalent of rent).
If such a factor can be included for housing it can also be included for a car or a truck.
Pretty straightforward to calculate. It would increase the year over year inflation rates by approximately 0.05-0.1% and the annualized month over month rates by between 0.6 and 1.3% (smaller for the CPI overall, larger for CPI core services).
Didn’t Trump fire the head of the BLS a couple months ago? Why is anyone surprised the data is “good” now? Of course this was going to happen. In 6 months time inflation will be at .00005% annually according to the BLS, Powell will be gone, the Fed will have the money printer glowing red hot again.
The White House withdrew its nomination of E.J. Antoni, and so Trump’s man at the BLS is not at the BLS.
William Wiatrowski is the acting commissioner of the BLS. He is a long-time senor BLS employee who has been acting commissioner before.
So there has been no change of the guard at the BLS.
Wolf, that’s only half the picture. The acting commissioner has seen the previous boss get shafted for delivering bad news, so whether he’s Trump’s loyalist or not, the effect is still there.
i wish people quit spreading homemade BS here. You have no f**king idea why this happened. So don’t spread your homemade BS about it. I’m so sick of this homemade BS. This is not X.
The unemployment numbers from August were very suspect, not just shutdown issues at fault.
Whatever faults you had with the system before, it was at least consistent in the way it was reported. Now it’s just random numbers. Check with your county’s actual economic data this summer.
So based on this low-ball data, Social Security released the 2026 COLA today. So 40 million Americans will be underpaid throughout 2026 due to faulty data. A big savings for the federal government. Deliberate?
Thanks for the analysis. Everywhere I look I see contradictory headlines; up, down accelerating, decelerating. This is actual information that I can trade on. The Wolf dividend grows!
I have an idea for trading. Please do not laugh. $BND is going to go up anyways. Buy 400 shares of $BND, say for $30,000 at a price of $75. If it goes up to $75.25, then sell the ETF for a profit of $100. Now there will be fluctuations and just buy the BND whenever it dips. The feds are going to cut rates anyways. So BND will for sure go up! or even it goes down, just hold it for a while to get the dividends. A profit of $500 a month is a lot. What you all think?
At some point, the loss of illegal immigrants (2 million so far) is going to impact vacancies and rents, bleeding into the CPI numbers.
Bond market might become unhappy if good inflation data is hard to come by…
The 10 year didn’t drop today (actually went up a bit), so clearly it wasn’t as excited as the media was.