Market Pulse
Mixed observation times, Oct 20–Oct 25, 2026 (5 days apart)
Economy · 7 min

Same 3.4% headline, but the Fed's core gauge ran at 3.0% in August

PCE and CPI inflation were both 3.4% in the year to August 2026. Ex food and energy, the Fed's PCE gauge was 3.0% and the CPI 2.4%. The weights explain why.

◆ The Signal
30-second read
What happened
The BEA reported on September 30, 2026 that the PCE price index rose 3.4% in the 12 months to August 2026, the same as the CPI published by the BLS on September 11, 2026. Excluding food and energy, PCE prices rose 3.0% and the CPI 2.4%.
Why
The two indexes cover different spending, use different formulas and weight the same items differently. Shelter carried 35.6% of the CPI in December 2025; housing was 15.8% of 2025 consumer spending in BEA's accounts; health care services weigh far more in PCE.
The numbers that matter
3.4%PCE and CPI, 12 months to August 2026
3.0%Core PCE, 12 months to August 2026
2.4%Core CPI, 12 months to August 2026
2%FOMC longer-run goal, measured by PCE
Why it matters
Social Security and the federal tax structure are adjusted with CPI measures; the Fed sets interest rates against a 2% goal written in PCE. A household's income adjustments and its borrowing costs are tied to different yardsticks.
What happens nextas we saw it on 7 Oct 2026
The September 2026 CPI is scheduled for October 14, 2026 at 8:30 a.m. ET. The FOMC meets October 27–28, 2026. The BEA's September 2026 personal income and outlays release is scheduled for October 29, 2026 at 8:30 a.m. ET, the day after the meeting ends.
Go deeper ↓
Two inflation gauges
Inflation over the 12 months to August 2026: CPI and PCE price indexes, headline and core

The headlines match at 3.4%; underneath, the Fed's preferred core measure ran 0.6 points above core CPI and a full point above the 2% goal.

Inflation over the 12 months to August 2026: CPI and PCE price indexes, headline and core
PCE price index3.4%
CPI-U, all items3.4%
Core PCE (ex food and energy)3.0%
Core CPI (ex food and energy)2.4%
% change, August 2026 on August 2025Source: BEA, Personal Income and Outlays, August 2026 (September 30, 2026); BLS, Consumer Price Index, August 2026 (September 11, 2026)
Go deeper

On September 30, 2026 the Bureau of Economic Analysis reported that the price index for personal consumption expenditures rose 3.4% in the 12 months to August 2026. On September 11, 2026 the Bureau of Labor Statistics had reported that the Consumer Price Index rose 3.4% over the same 12 months. Two agencies, two methods, one number.

Take out food and energy and they split. Core PCE prices rose 3.0%. Core CPI rose 2.4%. The PCE index is the one the Federal Reserve's 2% goal is written in, so the gauge the Fed is judged against was running 0.6 percentage points hotter underneath than the one most households hear quoted, and a full point above its target.

The August numbers, side by side

Data
PCE prices rose 0.3% in August 2026 and 3.4% over 12 months
BEA, September 30, 2026: the PCE price index rose 0.3% in August, and 0.2% excluding food and energy. From August 2025 it rose 3.4%, and 3.0% excluding food and energy.
Data
The CPI rose 0.4% in August 2026 and 3.4% over 12 months
BLS, September 11, 2026: the CPI-U rose 0.4% in August, seasonally adjusted, and 3.4% over 12 months; excluding food and energy, 2.4%. Shelter rose 3.0% over the year, medical care 1.6%.
Two measures of the same 12 months
CPI-U (BLS)PCE price index (BEA)
All items, 12 months to August 20263.4%3.4%
Excluding food and energy, 12 months2.4%3.0%
All items, August on July0.4%0.3%
PublishedSeptember 11, 2026September 30, 2026
Whose spendingOut-of-pocket spending by urban householdsSpending by and on behalf of households and nonprofits serving them
Formula across itemsModified LaspeyresFisher-Ideal
Source: BLS, Consumer Price Index, August 2026; BEA, Personal Income and Outlays, August 2026; BEA FAQ on differences between the two indexes.

Why the Fed's target is written in PCE

The Committee reaffirms its judgment that inflation at the rate of 2 percent, as measured by the annual change in the price index for personal consumption expenditures, is most consistent over the longer run with the Federal Reserve's statutory maximum employment and price stability mandates.FOMC, Statement on Longer-Run Goals and Monetary Policy Strategy, adopted effective January 24, 2012; as amended effective August 22, 2025

The choice of index predates the 2% number. In its Monetary Policy Report of February 17, 2000, the Fed switched its inflation projections from the CPI to the PCE chain-type index and gave three reasons in a footnote.

Fact
The Fed's stated reasons: formula, coverage and revisability
The 2000 report said the PCE index “is constructed from a formula that reflects the changing composition of spending”, avoiding the “upward bias associated with the fixed-weight nature of the CPI”; that its weights “are based on a more comprehensive measure of expenditures”; and that its history “can be revised to account for newly available information and for improvements in measurement techniques”, giving “a more consistent series over time”. The Board's explainer: PCE “more quickly adapts to changes in spending patterns.”

Scope, formula, weights

BEA answers the question in an FAQ titled “What accounts for the differences in the PCE price index and the Consumer Price Index?” It sorts the gap into four effects.

  • Scope. In BEA's words, “PCE measures spending by and on behalf of the personal sector, which includes both households and nonprofit institutions serving households; the CPI measures out-of-pocket spending by households.” A 2011 BLS note gives the main example: PCE medical care includes what employers, Medicare and Medicaid pay on consumers' behalf.
  • Formula. PCE uses a Fisher-Ideal formula; the CPI uses a modified Laspeyres formula across its 7,776 item-area cells, with a geometric mean inside most item strata and weights refreshed each January from the Consumer Expenditure Survey.
  • Weight. The same item carries a different share of each basket.
  • Other. Seasonal adjustment, price differences, residuals.
How much of each basket goes to housing and health care
ComponentShare of CPI-U, December 2025Share of 2025 PCE (calculation)
Shelter / housing35.6%15.8%
Owners' equivalent rent / imputed rent of owner-occupied housing26.2%12.0%
Rent of primary residence / rental of tenant-occupied housing7.8%3.7%
Medical care services / health care services6.9%16.9%
Source: BLS, relative importance of CPI components, December 2025; BEA annual PCE by type of product via FRED (owner-occupied rent updated September 30, 2026; other series April 9, 2026). PCE shares are this article's calculation.
Wealth Signal calculation
Housing is less than half as heavy in PCE as shelter is in the CPI
2025 PCE on housing $3,310.951bn ÷ total PCE $20,954.857bn = 15.8%. Imputed rent of owner-occupied housing $2,521.870bn ÷ $20,954.857bn = 12.0%. Tenant rent $776.768bn ÷ $20,954.857bn = 3.7%. Health care services $3,551.570bn ÷ $20,954.857bn = 16.9%. Against the CPI's December 2025 relative importance: shelter 35.625, owners' equivalent rent 26.204, rent 7.840, medical care services 6.935.
Data
When shelter ran hot, the weight gap pushed the CPI above PCE
BLS reconciled the two indexes in a May 2011 note, using BEA's method. From 2002 to the first half of 2007, PCE prices rose 2.5% a year and the CPI 2.9%. The formula effect added 0.17 points, the weight effect 0.66 and other effects 0.04; the scope effect subtracted 0.53. BLS: “Shelter has a significantly higher relative weight in the CPI than the PCE, and shelter inflation generally outpaced overall inflation from 2002–2007.”

Why core PCE was higher in August

In both periods BLS reconciled in 2011, the weight and formula effects put the CPI on top: by 0.4 points a year from 2002 to mid-2007, and 2.6% against 1.7% in the fourth quarter of 2010. August 2026 does not fit that pattern, and the shelter arithmetic shows why the weight difference cannot be the explanation this time.

Wealth Signal calculation
Shelter was pulling core CPI up, not down
Shelter's December 2025 relative importance of 35.625 out of 79.919 for all items less food and energy puts it at 44.6% of core CPI. Shelter rose 3.0% in the 12 months to August 2026, so it contributed 0.446 × 3.0 = 1.34 points of the 2.4% core rise. The rest of core CPI, 55.4% of the weight, rose (2.4 − 1.34) ÷ 0.554 = 1.9%. Weights drift during the year, so this uses fixed December weights.
Interpretation
The gap has to come from what PCE covers that the CPI does not, or weights more
Shelter rose faster than core CPI, so PCE's smaller housing weight cannot explain why core PCE came out higher. The candidates are items PCE weights more heavily or the CPI leaves out: health care paid on households' behalf, at 16.9% of PCE against 6.9% for medical care services in the CPI, and spending by nonprofits serving households. BEA publishes a monthly decomposition in its NIPA Table 9.1U; this article does not assign the 0.6-point gap to any single item.

Income, spending, saving and a large revision

Data
Spending outpaced income in August 2026
Personal income rose $66.6bn, or 0.2%, and disposable income $68.6bn, or 0.3%. Consumer spending rose $190.8bn, or 0.9%, to an annual rate of $22,308.5bn. After prices, spending rose 0.6% and disposable income was unchanged. Personal saving was $990.2bn, a saving rate of 4.1% of disposable income.
Fact
The September 30 release rewrote five years of history
BEA says the September 30 release “presents monthly results from the annual update of the National Economic Accounts”, with revisions to personal income and outlays beginning “with January 2021.” July 2026 had been published on August 26, 2026 with PCE inflation of 3.7% and core of 3.3%, and a saving rate of 3.0%.
Wealth Signal calculation
On the revised data, July and August inflation were the same
FRED's PCE index levels, updated September 30, 2026: headline 131.172 in July 2026 against 126.906 in July 2025, a rise of 3.4%; core 130.133 against 126.363, 3.0%. August: 131.579 against 127.229 (3.4%) and 130.455 against 126.646 (3.0%). The move from 3.7% to 3.4% between the two press releases is a revision, not a slowdown.

Set the two press-release headlines side by side and July-to-August looks like a deceleration the revised series does not contain. The same caution applies to the saving rate. This is the Fed's third reason from 2000 at work: PCE history is rewritten when better source data arrive. The CPI works the other way. BLS says the unadjusted CPI-U and CPI-W “are final when issued”, because they are written into pensions, contracts and tax brackets; only seasonally adjusted CPI figures are revised, for up to five years. What banks pay on household saving is measured in a separate piece.

What it means for your money

The two indexes reach a household through different pipes. According to BLS, the CPI adjusts “consumers' income payments (for example, Social Security)”, eligibility levels for government assistance and “the Federal income tax structure”. The Social Security cost-of-living adjustment is built from the third-quarter average of the CPI-W. The Fed's interest-rate decisions are made against a goal measured in PCE. On September 16, 2026 the FOMC raised its target range to 3.75%–4.00%.

Wealth Signal calculation
What 3.4% means for a $50,000 year of spending
A household spending $50,000 a year on a basket that rose 3.4% needs $50,000 × 0.034 = $1,700 more to buy the same things 12 months later. At the Fed's 2% goal the figure would be $1,000. The $700 difference is what the gap between inflation and target costs that household in a year.

Neither index measures any one household's prices. A renter spending a third of the budget on housing has a basket closer to the CPI's 35.6% shelter weight than to PCE's 15.8%. Employer-paid health insurance counts in PCE but not in the CPI. The piece on how the CPI is built shows how the CPI's weights are set.

This piece does not claim that either index is the correct measure of inflation, that core PCE stays above core CPI, or anything regarding the FOMC's decision on October 28, 2026. The PCE weight shares are this article's calculations from BEA annual data, with category definitions close to but not identical with the CPI's. The August core gap is not decomposed here.

Sources

Wealth Signal prefers primary sources — regulators, statistical agencies and company filings. Named secondary sources are used where a primary document does not exist or is not public. Our source standards.

Editor and publisher · Markets, economics, personal finance

Wealth Signal is researched, written, edited and published by one person, under the Wealth Signal name. Every figure links to the document it came from, and errors are corrected on the page and logged publicly.

Related analysis

Cite this: Wealth Signal NYC (October 7, 2026). “Same 3.4% headline, but the Fed's core gauge ran at 3.0% in August.” https://www.wealthsignalnyc.com/stories/pce-versus-cpi-august-2026/ — the figures are the issuing agency’s, listed with their dates in the sources above; quote the agency, link the explainer.

This article is general information and analysis, not investment, tax or legal advice; read the financial disclaimer. Spotted an error? Corrections are made on the page and logged on the corrections page — tell the editor.