US September CPI to be released tonight: Overall inflation may rise to 3.6%. Can slowing core inflation dispel the Federal Reserve's October rate hike expectations?

The energy shock is accelerating overall inflation, but with marginal easing of core pressures, providing the Federal Reserve with a window to remain on hold.

The U.S. Bureau of Labor Statistics will release the September CPI data at 8:30 p.m. Beijing time tonight. Wall Street investment banks such as Goldman Sachs and Citigroup predict that the overall CPI year-on-year increase is expected to jump from 3.4% in August to the 3.6%-3.7% range, mainly driven by rising energy prices; meanwhile, the month-on-month increase in core CPI is expected to narrow slightly from 0.29% in August to between 0.18% and 0.24%, with stabilizing housing inflation and declines in certain service sub-items being the main drivers of core cooling.

Before the data release, the latest statements from Federal Reserve officials have sent a clear signal to the market. The FOMC minutes from September released last week show that the "majority" of officials consider one more rate hike this year appropriate, but several officials later emphasized there is no need to act hastily—Fed Governor Waller said there is no need to hike at consecutive meetings, while Vice Chair Williams added that one more hike this year is "possible," but there will be no further tightening in 2027. As a result, the market-implied probability of a 25 basis point rate hike in October has dropped sharply from about 70% before the employment report to about 20%, while the probability for December remains around 72%.

Tonight’s data will directly influence policy expectations for the FOMC meetings on October 28 and in December. If core CPI lands mildly, it will bolster the confidence of the "wait-and-see" camp; if there is a broad-based, stronger-than-expected core inflation, it could reignite the October rate hike debate. However, market analysts note that the Fed has never raised rates at an October meeting before U.S. midterm elections in history.

According to Goldman Sachs, September’s overall CPI is expected to rise 0.59% month-on-month and climb to 3.70% year-on-year, above the market consensus of 3.6% and noticeably higher than August’s 3.40%. The energy component is expected to rise sharply by 5.0% month-on-month, with surging retail gasoline prices as the main driver; the food component is expected to increase 0.2% month-on-month.

Citigroup’s forecast is relatively modest, projecting overall CPI year-on-year at 3.6% for September, but also confirming that energy prices are the core factor driving overall inflation. Citigroup’s data shows energy is expected to rise by approximately 21.6% year-on-year, the highest level so far this year.

Airfare is a key indicator to observe the secondary transmission effect of the energy shock. According to Goldman Sachs, September airfare is expected to rise 3.0% month-on-month, directly reflecting the pass-through effect of persistently high jet fuel prices. Goldman Sachs data shows that since the outbreak of the Iran war, U.S. jet fuel spot prices have roughly doubled, with airlines fully passing the increased costs onto ticket prices.

Goldman Sachs predicts September core CPI to rise 0.24% month-on-month, corresponding to 2.46% year-on-year, both slightly below consensus (0.2% MoM, 2.5% YoY); Citigroup gives a milder forecast with core CPI up only 0.18% MoM, and points out the probability of drifting down to a rounded 0.1% is higher than rising to 0.3%.

Goldman Sachs outlined three key subcomponent trends: First is a rebound in auto inflation, with used car prices expected to increase 0.5% MoM, new cars up 0.2%, and car insurance prices up 0.2%; second is continued mild housing inflation, with Owners’ Equivalent Rent (OER) and primary residence rent anticipated to rise 0.20% and 0.17% MoM, respectively, extending their downward trend; third, airfare up 3.0%, remains the most worthy pressure point to track in core inflation.

Citigroup, on the other hand, emphasizes that wireless communication service prices surged abnormally in August, contributing about 10 basis points to core CPI, but this subcomponent’s volatility pattern has become unstable since the upcoming methodology change in July 2025—historically, a sharp increase is usually followed by a larger pullback. Citigroup forecasts this component to decline by -0.5% in September, which is a key reason its core CPI forecast is lower than Goldman’s.

CPI data is also a critical input for market forecasts of the Federal Reserve’s preferred measure, core PCE. Goldman Sachs estimates September core PCE will rise 0.20% month-on-month, slightly below the core CPI forecast, mainly because new and used cars have a lower weighting in the PCE basket, and PCE data sources for airfares and medical services will show relatively mild increases.

Citigroup also forecasts September core PCE will rise 0.21% MoM, staying at 3.0% YoY. Citigroup currently tracks Q4 2026 core PCE YoY at 2.9%, about 50 basis points below the Fed’s latest projection. Citigroup economists Veronica Clark and Andrew Hollenhorst note that even though core PCE may still exceed core CPI in the near term, monthly inflation data this year are likely to continue displaying signs of core inflation moderation, with limited evidence of broad inflationary pressures spreading.

Looking ahead over the coming months, both institutions expect core inflation to continue falling, but differ on the pace and extent.

Goldman Sachs forecasts October core CPI MoM increase of about 0.2%, mainly driven by seasonal distortions from missing data during last year’s government shutdown; decreases to about 0.1% are expected in both November and December, as slowing housing inflation and gradually fading tariff-related price pressures drive further moderation. Based on this trajectory, Goldman Sachs economists are more optimistic about 2027’s inflation outlook compared to market consensus.

Citigroup also forecasts core CPI YoY to decline to a low near 2.3% in the next two months. However, both institutions explicitly note that if oil price shocks prove more persistent than expected, inflation risks will skew markedly to the upside. Brent crude remains above $105—whether the energy shock will stabilize by year-end remains the biggest uncertainty.

In summary, the current market consensus expects: overall inflation running hot due to higher gasoline prices, with core inflation staying mild thanks to cooling housing costs—a combination that gives the Fed grounds to maintain a "patient" stance while reserving the option to hike in December.

Goldman Sachs believes that a mild core CPI print will reinforce the "wait-and-see" camp’s arguments; if core inflation turns out broadly stronger than expected, it could reignite calls for a rate hike discussion in October—however, with just a week to the midterm elections and in keeping with Fed history, the likelihood of an October hike remains extremely low.

Citigroup is more optimistic about the path of core inflation by year-end, forecasting Q4 2026 core PCE at 2.9%, well below the Fed’s official forecast of 3.4%. However, Citigroup also warns that persistently high energy prices, stronger corporate pricing power compared to pre-pandemic levels, and rising household inflation expectations are the major downside risks to achieving this forecast.

At the market pricing level, there is a clear divergence between asset classes.

According to Goldman Sachs’ equity volatility trading desk, on this Wednesday’s CPI release day, the S&P 500 straddle options priced at just 52 basis points—the lowest for CPI release days this year and well below the recent average of 78 basis points; the Nasdaq 100 straddle options are about 85 basis points, again below the average of 111. The implication: equity markets expect this CPI release to have virtually no impact on risk assets.

The forex market is a different story. Goldman Sachs notes that EUR/USD’s implied CPI event move is roughly 32 basis points, in the 77th percentile of the past year; USD/JPY’s implied volatility is over four times its historical median. Three major asset classes are diverging sharply in their pricing for the same data—one side is bound to be proven wrong.