U.S. September CPI rises 3.7% annually as gasoline lifts inflation and core pressure persists

The rise in U.S. inflation in September was led by a sharply higher gasoline bill, but the report's price pressures extended beyond the pump. The Bureau of Labor Statistics reported on October 14 that the Consumer Price Index for All Urban Consumers, or CPI-U, rose 0.6% in September on a seasonally adjusted basis, following 0.4% in August. The unadjusted annual increase climbed from 3.4% to 3.7%. Core CPI, which excludes food and energy, rose 0.3% over the month, unchanged from August, while its annual increase rose from 2.4% to 2.6%.¹

Those movements need to be read together. Gasoline explained most of the overall index's monthly increase, but the monthly pace of core inflation did not ease either. A higher annual core rate does not, however, mean its monthly rate also accelerated. For business operations and long-term investment, the important questions are how long the energy shock lasts and whether other prices continue rising, rather than which interest-rate decision can be inferred from a single release.

Historical photograph, June 26, 2025, Crawfordville, Florida, United States. Photograph by The Bushranger: Gas pump at Circle K, Crawfordville, Florida, used under CC BY-SA 4.0. The historical prices and signage do not represent 2026 fuel prices.

Gasoline led the rise, but energy prices did not move together

The gasoline index rose 9.1% in September after seasonal adjustment. BLS attributed more than half of the monthly increase in overall CPI to gasoline. Energy prices rose 5.2% over the month, up from 2.1% in August, and 20.4% over the year; gasoline was 34.2% higher than a year earlier. These figures show why the 0.6% headline increase should not be read as a similar rise across all goods and services.¹

There were also marked differences within energy. Electricity prices rose a seasonally adjusted 0.2% in September, while utility piped-gas prices fell 0.3%. Their annual increases were still 4.4% and 5.0%, respectively. Sharp gasoline movements therefore coexisted with much smaller monthly utility-price changes. A business dependent on electricity cannot use gasoline inflation as a direct substitute for its own power bill, while gasoline may loom much larger in the monthly spending of a household that drives frequently.¹

Price growth and contribution to the total are different concepts. CPI combines categories using expenditure weights, so the fastest-rising price does not necessarily have the largest impact on the overall index. Gasoline's importance this month is supported by BLS's contribution statement. A few component growth rates alone cannot allocate the entire headline increase or establish that rising oil-related costs caused every other category's increase.

Core pressure persisted without uniform acceleration

Shelter rose a seasonally adjusted 0.4% in September, following 0.3% in August, and was 3.2% higher over the year. Within shelter, owners' equivalent rent increased 0.3% over the month, rent rose 0.2% and lodging away from home rose 2.0%. This category combines different uses and pricing arrangements; its 0.4% increase does not mean every tenant's rent rose by that amount in September.¹

The broader services index excluding energy services rose 0.4% monthly and 3.1% annually. Commodities excluding food and energy commodities rose 0.3% monthly and 0.7% annually. Among other components, used cars and trucks rose 0.7% over the month, recreation 0.5%, education 0.6% and airline fares 1.2%; communication prices fell 0.2%. These monthly figures are seasonally adjusted.¹

The composition supports a narrower, more reliable conclusion: prices were rising in several categories beyond energy. It does not establish equally strong inflation in every industry. Used cars and trucks illustrate the distinction particularly well. Despite their 0.7% monthly increase, prices were still 1.5% lower than a year earlier. A monthly rebound and an annual decline can coexist because they answer different questions.¹

Flat food prices did not mean an unchanged grocery receipt

The food index was unchanged over the month as movements in different categories offset each other. Food at home fell 0.1% after seasonal adjustment. Food away from home rose 0.2%, but BLS marks that particular series as unadjusted, so the two monthly rates should not be treated as exactly comparable measures. Over the year, food overall rose 2.5%, food at home 1.8% and food away from home 3.5%.¹

“Food prices did not rise” therefore describes the month's aggregate food-index result. It does not mean prices returned to their year-earlier level, or that every household's bill was unchanged. Different shopping baskets, purchasing locations and quantities can produce different spending outcomes even when households face the same set of prices.

Three distinctions behind the 0.6% and 3.7% figures

The first is seasonal adjustment. BLS uses it to reduce the influence of price patterns that recur at similar times each year, making short-term trends easier to examine. Gasoline's unadjusted monthly increase was 6.5%, compared with 9.1% after adjustment. These are different statistical treatments of the same data, not two successive price increases or a claim that every filling station raised prices by 9.1%. Seasonal factors are updated annually, with related seasonally adjusted data revised for the previous five years.¹

The second is the comparison period. The 0.6% figure compares September with August; 3.7% compares September with the same month a year earlier. They cannot be added or substituted for each other. Annual changes reflect both recent prices and the year-earlier comparison base. The increase in annual core inflation from 2.4% to 2.6% must be considered alongside two consecutive monthly increases of 0.3%. Nor can one month's increase simply be extrapolated into the inflation actually experienced over the next year.

The third is the spending basket. CPI-U represents a population covering more than 90% of U.S. residents and combines price changes using that population's expenditure weights. It remains a statistical estimate rather than a census of every price. The national average basket is neither an individual company's input list nor an individual household's fixed bill. Reliance on rent, driving or restaurant meals changes how a household experiences the month's price movements.¹

Business and financing effects depend on transmission

If fuel increases persist and businesses cannot offset them through contracts, efficiency improvements or changes in consumption, related costs such as transportation may rise. Whether a company can pass those costs on depends on demand, competition and its pricing arrangements; inability to do so could squeeze margins. This is a conditional account of cost transmission. September's CPI itself provides no evidence of corporate profits, wage changes or cost increases already passed through to customers.

Financing is a separate channel. If persistent inflation changes lenders' and investors' assessments of future interest rates and risk, new borrowing costs or the discount rates used to assess projects could increase. For capital-intensive projects such as data centers and energy storage, where returns arrive later, a higher discount rate reduces the present value of future cash flows, other things equal. Revenue contracts, debt maturity, fixed versus floating rates and energy procurement arrangements can all change the practical effect. A consumer-price report alone cannot establish that those projects have already lost value.

Conversely, a quick reversal in gasoline prices and slower monthly gains elsewhere would weaken the case for persistent transmission. The same energy-price movement can also affect the revenue and costs of producers and users differently. Describing it as a loss for every business, or equating the CPI release with action already taken by the Federal Reserve, would go beyond the evidence.

What would test the implications?

Subsequent evidence needs to answer three questions. Do monthly gasoline and energy increases continue or retreat? Do shelter and services excluding energy keep rising over several months, and do goods-price increases become more widespread? Do actual financing quotes, corporate profit disclosures, employment and wage data support the inference that cost pressure is changing investment and operations? One component cannot validate that entire chain, and the headline annual rate cannot replace those tests.

The release schedules October CPI for November 10 at 8:30 a.m. Eastern Time.¹ Until then, September's clearest message is that gasoline dominated the monthly rise in headline inflation while core price pressure persisted. Its durability and commercial consequences still need confirmation from subsequent price, business and financing evidence.

Source notes

1. October 14, 2026, 8:30 a.m. Eastern Time · Consumer Price Index — September 2026, including Table A and the Technical Note · U.S. Bureau of Labor Statistics. ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶ ↩⁷ ↩⁸ ↩⁹ ↩¹⁰