October prices rose.
Real pay stood still.
October CPI rose 0.4% as average real pay held flat. Grocery and shelter pressure persisted, while missing 2025 data limited annual comparisons.
The Bureau of Labor Statistics released the October consumer-price and real-earnings reports on November 10. Nominal hourly pay increased 0.4%, matching the seasonally adjusted rise in consumer prices for the broad private-payroll measure. For production and nonsupervisory employees, a shorter workweek produced a weaker result: inflation-adjusted weekly earnings declined even as purchasing power per hour edged higher.
There is also an unusual hole in the report. BLS cannot publish the customary October-to-October inflation comparison for most categories because it did not collect the relevant survey data during the October 2025 lapse in appropriations. That does not erase the new monthly figures or mean prices stopped being measured this October. It means the usual year-earlier reference point is missing. For readers trying to decide whether pay is finally catching up with expenses, this release supplies a current monthly answer while withholding much of the familiar annual one.
A slower increase is still an increase
The headline CPI-U rose 0.4% in October, down from September's 0.6% increase. The distinction is between two positive rates of change, not between rising and falling prices. The national basket became more expensive again, just at a slower seasonally adjusted pace. The movement also was not confined to one volatile category. Food increased 0.5%, shelter 0.3%, and energy 0.7%. The index excluding food and energy rose 0.3%, the same monthly increase recorded in both August and September.
Energy's slowdown was especially large: its October increase followed a 5.2% September jump. A small number after a large increase can make a headline look calmer without reversing the earlier rise. Using the rounded published changes purely as an illustration, a $100 energy index level before September would become $105.20 after that month and about $105.94 after October. This is compound arithmetic, not a forecast or an estimate of a particular utility bill. It explains why a household may see little practical relief when a monthly inflation rate falls sharply but remains positive.
Core inflation answers a different question from the food-and-fuel-inclusive headline. Removing those two categories helps analysts examine price movements beyond often volatile necessities; it does not remove those necessities from a household's budget. October's repeated 0.3% core increase is evidence of continuing monthly pressure outside food and energy.
The grocery aisle has more than one story
Food at home rose 0.4% after declining 0.1% in September. Four of the six major grocery groups increased in October, one declined, and one was unchanged. The breadth matters because the result was not simply an increase in one unusually volatile product. Fruits and vegetables and dairy products each rose 0.6%. The combined meat, poultry, fish and eggs category increased 0.5%, with beef up 0.9%. Other food at home rose 0.7%, including a 1.0% increase in snacks.
The grouped meat, poultry, fish and eggs increase does not establish that eggs alone rose 0.5%. In the opposite direction, nonalcoholic beverages declined 0.3% after rising 0.9% the previous month. Cereals and bakery products were flat. A household buying different proportions of these goods can reasonably experience a checkout change different from the overall grocery index.
Meals away from home increased 0.5%, compared with 0.2% in September. Full-service meals rose 0.7%, while limited-service meals increased 0.4%. There is an important table note: food away from home is not seasonally adjusted. It should not be described as though every number in the monthly comparison received an identical seasonal adjustment. The practical signal is still clear within that published series: restaurant prices increased more rapidly in October than in the preceding month.
The grocery price index resumed rising while restaurant prices also increased. But the smaller grocery percentage does not measure savings from cooking: ingredients, portions and starting prices differ. The new evidence is a rise in both series, with offsets in only some grocery groups.
Why gasoline can be down and up in the same report
Gasoline illustrates why readers need the label attached to a percentage. The seasonally adjusted gasoline index rose 1.0% in October after a 9.1% September increase. Before seasonal adjustment, gasoline prices declined 0.7% in October. Both figures appear in the BLS release. They describe the same month through different comparisons, rather than two contradictory claims about whether every filling station raised its price.
The unadjusted measure preserves the month's observed price pattern. Seasonal adjustment removes recurring seasonal influences to make consecutive months more comparable. An observed decline can therefore coexist with a positive adjusted change if the month's movement is stronger than the seasonal pattern would otherwise imply. For someone checking a fuel receipt, the unadjusted figure may be more recognizable; for an analyst comparing underlying month-to-month momentum, the adjusted series has a different purpose.
Electricity increased 0.9% after seasonal adjustment, while utility piped gas fell 0.8%. Fuel oil rose 2.0%. The opposite directions are important as heating needs come into focus: the energy total does not describe all fuels equally. A household's bill also reflects how much energy it used, not simply the price per unit. Colder weather or a change in occupancy can raise total spending even when a relevant price index declines.
Gasoline is also one of the exceptions to the missing annual comparisons. Its unadjusted index was 37.4% above October 2025. That is a category-specific annual result, not an all-items inflation rate. Treating it as the economy-wide number would dramatically change what BLS actually reported. Conversely, saying that no annual price data exist would discard information the agency did publish. The right approach is to preserve each available comparison and leave unavailable categories explicitly unfilled.
Shelter's size matters as much as its rate
Shelter rose 0.3% and accounted for more than 30% of the monthly increase in the all-items index, according to BLS. This is a contribution statement, not a claim that shelter prices rose by 30%, nor that every household spends exactly that share on housing. The difference between a category's percentage change and its contribution to the overall index is central to this report. A relatively modest change in a heavily weighted expense can matter more to the aggregate than a much larger movement in a smaller category.
Within shelter, rent increased 0.2% and owners' equivalent rent rose 0.3%. Lodging away from home increased 0.6%, following a 2.0% September rise. These are distinct measures, not interchangeable estimates of a new apartment lease, a home purchase or a hotel reservation. The CPI is a consumption-price index; BLS excludes investment assets such as real estate from the basket. Reading the shelter result as a direct measure of home-sale prices would confuse the cost of housing services with the purchase price of an asset.
Other services and goods moved in different directions. Motor vehicle insurance increased 0.7%, recreation 0.8%, and household furnishings and operations 0.2%. Medical care rose 0.3%, with hospital services up 0.8%, physicians' services 0.4% and prescription drugs 0.3%. These separate movements help explain why a family with an insurance renewal or substantial medical spending can feel pressure that a simple grocery-and-gas discussion misses.
Used cars and trucks supplied a notable offset, declining 1.2% after a 0.7% September increase. New vehicles rose 0.2%. Unlike many other categories, both also have annual comparisons in the published table: new vehicles were up 0.7%, while used vehicles were down 3.3%. A prospective buyer therefore receives a more complete price comparison than is available for the overall CPI. But a vehicle price index is not a total ownership-cost index. Insurance, financing terms, taxes and the particular vehicle's condition can change a purchase decision even when the broad used-car index falls.
The raise that did not buy more
For all employees on private nonfarm payrolls, average hourly earnings increased from $37.86 in September to $38.00 in October. Average weekly hours remained at 34.4, and average weekly earnings rose from $1,302.38 to $1,307.20. After adjusting for the CPI-U, BLS reported no monthly change in either average hourly or average weekly purchasing power. The extra nominal dollars broadly kept pace with the price index rather than creating a real gain.
The word “average” carries substantial weight. These figures are not the median worker's wage, an hourly minimum, or the raise received by every employee. BLS derives the hourly estimate from payroll divided by paid hours and calculates weekly earnings using average hourly earnings and average weekly hours. The mix of jobs matters: changes in employment across higher- and lower-wage industries can move the average, as can changes in pay rates. A worker whose wage did not change should not read the aggregate increase as a statement about their own paycheck.
The weekly average combines full-time and part-time jobs, including jobs held by multiple jobholders. It measures pay per job, not total household income. BLS also cautions against turning one weekly average into typical annual earnings by multiplying by 52: turnover, layoffs and the changing job mix affect that comparison.
October's real hourly average, $11.26 in 1982–84 dollars, expresses purchasing power in older reference-year units rather than an actual paycheck. BLS shows real weekly earnings edging from $387.27 to $387.35 while the rounded percentage change remains 0.0%. Levels and rounded rates can therefore look different without contradicting one another.
For many workers, hours weakened the weekly result
Production and nonsupervisory employees had a different outcome. Their nominal hourly average increased 0.4%, while the CPI-W used to adjust this group's earnings increased 0.3%. BLS reported a 0.1% rise in real hourly earnings. Yet real weekly earnings fell 0.3%. The average workweek declined from 33.9 hours to 33.8, reducing the benefit of the higher hourly average. A slightly stronger hourly purchasing-power measure therefore did not translate into more purchasing power over a whole week.
The table shows nominal weekly earnings increasing only from $1,106.16 to $1,106.95 for this group, while the hourly average rose from $32.63 to $32.75. This is a useful contrast with the all-employee series, where weekly hours were unchanged. The underlying question is not simply whether hourly wages rose, but whether paid hours rose, fell or stayed level alongside them. Because the displayed rates and levels are rounded, summing the published hourly and hours percentage changes is not a substitute for the agency's separately calculated weekly result.
This group is substantial but precisely defined. It covers production employees in mining, logging and manufacturing, construction employees in construction, and nonsupervisory employees in service industries. BLS says these groups account for approximately four-fifths of employment on private nonfarm payrolls. They should not be relabeled as all low-income workers, all hourly workers or all Americans. The use of CPI-W rather than CPI-U also means the two real-pay series have different inflation adjustments as well as different payroll populations.
The distinction matters when assessing a job offer or a new schedule: a higher hourly rate can coexist with less weekly purchasing power. October's combination makes that relationship visible in a large payroll group, while the all-employee average shows what happens when hours hold steady.
The missing annual number is not a license to invent one
Normally, readers can place monthly results beside a twelve-month change. October 2026 is unusual because most of the necessary October 2025 CPI indexes are unavailable. BLS attributes that absence to the lapse in appropriations that prevented survey collection. The missing input is the year-earlier price measure, so the problem also affects inflation-adjusted annual earnings comparisons. It does not make every annual payroll statistic unavailable: the earnings tables still publish nominal hourly and weekly earnings changes, which do not require the missing price index.
For example, all-employee nominal hourly earnings were 3.1% above October 2025, and nominal weekly earnings were 3.7% higher. Those figures cannot establish the corresponding annual real gain without a valid inflation adjustment. The annual inflation comparison shown for September belongs to a different period and cannot simply be carried forward into October. Doing so would silently mix time windows and produce a result that BLS did not publish. The absence is particularly consequential when the question is whether workers are better off than a year earlier rather than merely whether they kept pace last month.
BLS does describe an approximation procedure for some users who need a missing index value, such as parties to an escalation contract. If all parties agree, they can use the geometric mean of the immediately preceding and following month's index values. The agency explicitly warns that such a calculation is not an official CPI series and that percentage changes derived from it are unofficial estimates. That is a narrowly described workaround for a missing observation, not a newly released national annual inflation rate or a direction to alter a contract unilaterally.
Monthly measures still show the latest direction and its components; the available annual gasoline and vehicle comparisons add a separate view. Preserving those specific results is more informative than discarding the release or treating an interpolated value as the missing official annual headline.
What October changes in the household picture
The strongest conclusion is not that all costs accelerated or that inflation has been defeated. October combined a slower headline increase with renewed grocery pressure, continued shelter increases and divergent energy prices. Average all-employee real pay was flat, while the production and nonsupervisory weekly measure declined. These results make the difference between a pay-rate improvement and a spending-power improvement concrete: prices determine what a dollar buys, and hours help determine how many dollars arrive.
The CPI's expenditure weights describe its reference population, not every family's budget. A renter without a car and a homeowner with a long commute face different combinations of October's shelter, fuel and other price movements. Their exposure helps explain why a flat real-pay average can coexist with very different experiences.
The next CPI and real earnings releases are scheduled for December 10 at 8:30 a.m. Eastern and will describe November. The questions to carry forward are specific: whether grocery increases persist, whether energy's monthly pace settles, whether core price growth changes, and whether paid hours strengthen or weaken alongside hourly earnings. Those comparisons will show whether October's mixed price-and-pay picture persists.
Comments
Post a Comment