Reading data · 5 min read · 5 steps · Updated 19 October 2026

How to interpret PCE inflation data

The Personal Consumption Expenditures (PCE) price index is the Federal Reserve's preferred inflation gauge. It is published monthly by the Bureau of Economic Analysis — but it is less well understood than CPI. Here is how to read it correctly.

The PCE price index is published by the Bureau of Economic Analysis (BEA) as part of the monthly Personal Income and Outlays report. It measures the average change in prices paid by US households across all goods and services — similar to CPI, but with important methodological differences that cause the two measures to diverge by 0.3–0.5 percentage points on average.

The Federal Reserve's 2% inflation target is explicitly defined in terms of the PCE price index, not CPI. This matters enormously: when the Fed says inflation is "2.7%," it is referring to the PCE rate, not the CPI rate (which is typically 0.3–0.5 pp higher). Understanding PCE is essential for following the Fed. The BEA publishes the monthly PCE report at 08:30 ET, about 2 weeks after the month's CPI release.

Start with core PCE — the Fed's policy rate anchor

Core PCE — all items excluding food and energy — is the primary metric the Federal Reserve uses for policy decisions. This is the number that appears in the dot plot projections, in Chair Powell's press conferences, and in the FOMC statement. Before looking at anything else in the PCE report, find the core PCE 12-month rate and the seasonally-adjusted month-on-month rate. In late 2026, core PCE is running at approximately 2.7% YoY — about 0.5 pp below core CPI (3.2%), due to structural differences between the two measures.

The monthly annualised rate is equally important: if the 3-month average of monthly SA core PCE prints is running at a 2.2% annualised pace, that is more relevant to the forward inflation path than the current 2.7% YoY rate (which is still influenced by high prints from 12 months ago). The BEA publishes both the MoM change and the 12-month change — track both. See the PCE vs CPI explainer for the full comparison.

Find the "supercore": services ex-housing PCE

Federal Reserve officials — including Chair Powell since 2023 — have emphasised a sub-component called "core services ex-housing PCE" or "supercore PCE." This strips out housing costs (owners' equivalent rent and rent of primary residence) from core PCE services to isolate the component most directly driven by domestic wages. The rationale: housing costs in PCE adjust slowly (similar to CPI shelter lag), so they can dominate the core reading even as the truly current wage-driven inflation is moving differently.

"Supercore PCE" is not published as a named line in the BEA report — you need to calculate it. The BEA's interactive data tool (BEA.gov) allows you to construct it: take PCE services, subtract housing costs (NIPA Table 2.4.5, lines 68–70). In September 2026, supercore PCE was running at approximately 3.2% YoY — higher than headline PCE (2.8%) but falling gradually. When supercore breaks below 3.0% on a sustained basis, the Fed will have strong evidence that wage-driven inflation is under control.

Compare PCE and CPI: understand the structural gap

PCE typically runs 0.3–0.5 pp below CPI for structural reasons:

Basket composition: PCE covers a broader scope of consumption (including healthcare paid by employers and government, which accounts for a large share of healthcare spending). CPI covers only out-of-pocket spending. PCE healthcare is less volatile than CPI healthcare.
Weighting formula: PCE uses a chain-weighted (Fisher ideal) formula that adjusts for consumer substitution. CPI uses a fixed Laspeyres formula. Because substitution always runs "downhill" (consumers switch from expensive to cheaper options), the Laspeyres formula in CPI overestimates the cost of living change, producing a higher rate than PCE's chain formula.
Shelter weight: CPI's shelter component (OER + rent) has a weight of ~35%. PCE's shelter weight is ~16%. Since shelter inflation is currently elevated (~4.9%), this weighting difference alone accounts for roughly 0.3 pp of the CPI-PCE gap today.

When CPI and PCE diverge more than the typical 0.3–0.5 pp, it is usually due to the shelter component moving differently from other categories. In 2024–25, the CPI-PCE gap widened to ~0.8 pp as CPI shelter (OER-heavy) stayed elevated while PCE shelter diverged less. This gap should narrow as CPI shelter disinflation accelerates in 2027. See our shelter CPI outlook for the timeline.

Read the category breakdown for wage-driven components

The PCE report includes a detailed breakdown by major category. For inflation analysis, the most policy-relevant categories to track are:

Healthcare services: ~17% of PCE. Moves slowly; primarily driven by administered prices and Medicare/Medicaid rate changes rather than market dynamics. A healthcare PCE spike in a given month is often administrative rather than inflationary.
Financial services and insurance: ~10% of PCE. Highly volatile month-to-month because of how the BEA imputes financial services prices (based on interest rate margins and fee income). This volatility is why supercore can jump in months with financial market activity changes.
Food away from home (restaurants): A key domestic wage-driven category. Restaurant price inflation closely tracks food service labour costs. As wage growth in the food service sector eases from its 2022–23 peak, restaurant PCE should moderate.
Transportation services: Includes airfares, auto insurance, and auto repair. Airfares are volatile month-to-month; auto insurance has been a persistent upside contributor in 2024–26 following a surge in claims and repair costs.

Time the PCE release relative to CPI and FOMC meetings

The PCE report is published approximately 2 weeks after the month's CPI release — which means analysts already know CPI when PCE comes out. Because PCE and CPI are correlated (they measure overlapping consumption), the CPI release often "front-loads" the information in PCE. If CPI surprised upward, markets often already partially price in a higher PCE reading by the time PCE is released. The PCE surprise relative to the post-CPI expectation is what matters for market reaction, not the surprise relative to a blind pre-CPI forecast.

For the FOMC November 5 meeting, the September PCE (released October 31) is the last major inflation data point available. It arrives 5 days before the decision, making it highly market-sensitive. If September PCE shows core running at 2.7% YoY and 0.25% MoM SA — in line with the trend — it reinforces the case for a 25bp cut. If it surprises significantly higher, it could tip the balance toward a hold. Track the PCE data alongside the Fed's inflation target on our inflation explorer.

Sources & further reading

Terms in this guide

core inflation consumer price index cpi pce price index inflation target seasonal adjustment

Frequently asked questions

Why does the Fed use PCE rather than CPI as its inflation target?

The Federal Reserve switched to PCE as its policy benchmark in 2000, after years of using CPI. The rationale: PCE is methodologically superior for measuring the true cost-of-living change because it uses a chain-weighted formula that accounts for consumer substitution (reducing "substitution bias"), covers a broader scope of consumption including third-party payers (employer-provided healthcare), and is revised in a more systematic and transparent way than CPI. The net effect is that PCE runs about 0.3–0.5 pp below CPI on average, so the Fed's 2% PCE target is roughly equivalent to CPI running at 2.3–2.5%. See our PCE vs CPI explainer for the full comparison.

How often is PCE data revised?

PCE data is revised more extensively and more regularly than CPI. The BEA publishes three revisions: the initial estimate (month T+1), a revised estimate (month T+2), and then incorporates annual revisions in July of each year. The annual revisions can affect data going back 3–5 years and sometimes produce meaningful changes to the level and trend of core PCE — including revisions that change the policy narrative significantly. The July annual revision is closely watched by Fed economists. By contrast, BLS revises CPI only rarely (seasonal factor revisions in February) and never goes back more than 5 years. This means PCE is a more accurate but less stable measure than CPI.

What is PCE deflator vs PCE price index — are they the same?

Yes, the PCE price index and the PCE deflator are the same thing. The term "deflator" comes from the national accounts context: PCE is published as part of the NIPA (National Income and Product Accounts) where it is used to "deflate" nominal spending to real spending. The formula for the deflator is: nominal PCE / real PCE × 100. The resulting index (the PCE price index / PCE deflator) measures the average price change across the entire basket of personal consumption expenditures. Core PCE deflator = core PCE price index. You may see both terms used in academic papers and Fed communications; they are interchangeable.