What is PCE Inflation? The Federal Reserve’s Preferred Measure

Updated October 1, 2026, after the August PCE release. Figures below are from the Bureau of Economic Analysis and the Federal Reserve unless stated otherwise.

On September 30, the Federal Reserve’s favourite inflation gauge came in softer than almost anyone had forecast. Core PCE was expected to rise 3.3% from a year earlier; it printed 3.0%. Treasury yields fell within minutes. By the afternoon the 10-year yield had not only recovered but closed near 5.3%, around its highest level since 2007.

That round trip is the best short lesson there is on PCE inflation. The number matters enormously to the Fed. It matters to markets in a narrower, more conditional way than most explainers suggest. This guide covers both: what the index actually measures, why the Fed chose it over CPI, and how to read a PCE morning without being fooled by the headline.

What PCE inflation measures

PCE stands for personal consumption expenditures. The price index is published monthly by the Bureau of Economic Analysis, part of the Commerce Department, inside a release called Personal Income and Outlays. It tracks the prices of goods and services bought by people living in the United States, or bought on their behalf.

Those last four words do a lot of work. When an employer pays for a worker’s health insurance, or Medicare pays a hospital, the household never sees the bill, but it still consumes the care. PCE counts that spending. The consumer price index does not, because CPI is built around what households pay out of pocket.

Two versions get quoted. Headline PCE covers everything. Core PCE strips out food and energy, the two categories whose prices swing hardest month to month. The Fed’s target is written in terms of headline PCE, but policymakers lean on core to judge where inflation is heading, because a gasoline spike in one month says little about wage-driven price pressure six months out.

August shows why the distinction matters. Headline ran at 3.4% while core sat at 3.0%. Energy did most of the damage to the headline figure that month.

PCE vs CPI: why the Fed made the switch

The Fed began treating PCE as its main inflation measure in 2000, and when it adopted a formal 2% inflation target in 2012, it wrote that target in PCE terms. Three technical differences explain the choice.

Scope. PCE captures third-party spending on households, which is why health care carries a much heavier weight in PCE than in CPI.

Substitution. If beef gets expensive and shoppers buy chicken instead, PCE’s chain-weighted formula picks up that switch quickly. CPI updates its spending weights less often, so it tends to overstate the cost of living slightly when relative prices shift.

Housing. Shelter makes up roughly a third of CPI but only around a sixth of PCE. When rents and owners’ equivalent rent are running hot, CPI usually prints above PCE. When housing cools, the gap narrows.

For most of the past two decades those differences have left core PCE running a few tenths of a point below core CPI. Traders who watch both should expect the two to disagree, and should know which one the Fed is actually grading itself on.

Why the PCE number usually surprises less than CPI

Here is the part most guides skip. PCE for a given month is published roughly four weeks after the month ends, and by then the CPI and producer price index for the same month are already out. Economists feed those components into models and publish fairly precise PCE estimates days in advance.

The result: on most PCE mornings the market already knows the answer to within a tenth of a point. The release moves prices when it lands outside that narrow band, or when it revises earlier months.

September 30 was a revision story more than a cooling story. According to BEA’s own core series, core PCE has now printed 3.0% for June, July and August, after 3.2% in May. July had previously been reported at 3.3%. The downward revision to prior months, not a sudden drop in August, is what produced the headline-grabbing miss against consensus.

How a PCE release moves markets

The transmission runs through rate expectations. A hotter-than-expected core print raises the odds that the Fed tightens or holds rates higher for longer; short-dated Treasury yields usually react first, then the dollar, then rate-sensitive stocks such as homebuilders, small caps and long-duration growth names. A softer print does the reverse.

The 2026 backdrop sharpens that link. The Fed raised its target range to 3.75%–4.00% at its September 15–16 meeting, its first increase since 2023, and the question for every inflation release since has been whether a second hike follows. Traders pared bets on an October move after the August data.

Yet the 10-year yield still finished September 30 higher than it started. Long-dated yields answer to more than one month of inflation: the supply of Treasuries, oil prices, growth data and the extra compensation investors demand for holding long bonds. On a day when private hiring and growth figures also came in firm, those forces outweighed a friendly inflation print. A soft PCE number lowers the floor under rates; it does not guarantee they fall.

The October calendar quirk

This month the order of events is unusual. The Fed announces its decision on October 28. The September PCE report arrives the next morning, October 29. Policymakers will go into the meeting leaning on September CPI and staff estimates of PCE, but without the official number they say they target.

That makes the CPI release earlier in the month carry more weight than usual for the October decision, and it means the October 29 PCE print will be read as a verdict on whatever the Fed has just done.

How to read a PCE morning: a short checklist

1. Look at core month-on-month first, and to two decimals. Year-on-year figures move slowly and carry base effects. A 0.2% monthly core print can be 0.15% or 0.24% before rounding, and that difference annualises to roughly a full percentage point.
2. Check the revisions before the headline. As September 30 showed, a change to the prior two months can account for most of the surprise.
3. Compare against the estimates built from CPI and PPI, not just the survey consensus published a week earlier. The surveys can be stale.
4. Watch the 2-year yield, not the 10-year, for the cleanest read on what the data did to Fed expectations.
5. Read personal income and spending in the same release. Strong real spending alongside sticky core inflation is the combination the Fed likes least.

PreMarket Daily’s PCE coverage

For how earlier releases played out in real time this year, see our coverage of the May release and the June release. For the wider policy picture, our guides to how the Federal Reserve moves markets and why stocks care about the 10-year Treasury yield cover the mechanics referred to above.

Sources and method: PCE and core PCE data from the U.S. Bureau of Economic Analysis; policy rate and meeting schedule from the Federal Reserve Board; consensus figures as reported by CNBC (Dow Jones survey); intraday Treasury moves from market reports on September 30, 2026. This guide is reviewed and updated after each monthly PCE release.