Why does PCE inflation differ from CPI inflation?

The short answer

CPI and PCE both measure inflation, but they answer slightly different questions. CPI asks: how much more does it cost a typical urban household to buy the same basket as last year? PCE asks: how much have prices changed across all consumption-related transactions in the economy, including those paid by third parties on behalf of households?

The scope difference matters most for healthcare. CPI counts only what households pay out of pocket for medical care; PCE counts what employers, Medicare, and Medicaid pay too. Because medical inflation typically runs higher than overall inflation, PCE captures more of it.

The two indices generally move together but diverge during periods of significant relative price shifts. In 2021-2022, CPI rose faster than PCE; in 2008-2009, the gap narrowed sharply during the recession.

→ New to inflation measurement? Financial education hub

What the data shows

Key figures (BEA, BLS, FRED, 2000-2024):

• Average gap CPI minus PCE on YoY basis: approximately +35 basis points over 2000-2024
• CPI peaked at 9.1% in June 2022; headline PCE peaked at 7.2% in the same month — a 190bp gap
• Shelter weight: ~36% in CPI vs ~16% in PCE — the largest single source of methodological divergence
• Healthcare weight: ~9% in CPI vs ~17% in PCE — the second largest divergence
• Core PCE has remained the Fed’s primary policy gauge since the 2000 Greenspan-era pivot

The 2022 gap of 190bp at the peak was historically large, driven primarily by shelter weight differences during a period of rapid rent acceleration.

→ Dataset: US PCE inflation

Why it happens — the macro mechanism

Three structural choices separate PCE from CPI methodologically.

Scope of expenditure. CPI uses the Consumer Expenditure Survey to weight what households directly pay. PCE uses national accounts data covering all transactions on behalf of households, including employer-provided health insurance and government-paid medical care. This makes PCE roughly 25-30% larger in nominal coverage than CPI for any given period.

Weighting methodology. CPI uses a modified Laspeyres formula with weights updated annually (since 2023) or biennially before that. PCE uses a Fisher chain-weighted index that updates monthly to reflect substitution. When beef prices rise and households shift to chicken, PCE captures this shift quickly; CPI captures it with a lag. This is documented as the main reason for the persistent CPI-PCE wedge — see how CPI is calculated.

Component weights. The most consequential differences are in shelter (heavier in CPI) and healthcare (heavier in PCE). When shelter inflation diverges from overall inflation — as in 2022-2023 — the two indices tell substantially different stories. When healthcare costs accelerate relative to other components, PCE rises faster than CPI. See owners’ equivalent rent explained.

CPI and PCE are not rivals — they are two lenses on the same phenomenon, calibrated for different policy questions.

→ Framework: Inflation regimes pillar

What it means for different economic actors

Savers tracking purchasing power erosion typically use CPI because it more closely matches direct household out-of-pocket experience. PCE’s broader scope can understate the inflation actually felt by households who do not benefit from third-party-paid healthcare.

Investors in TIPS receive coupons indexed to headline CPI, regardless of which index the Fed targets. Equity and bond traders nonetheless watch PCE closely because it is the Fed’s policy reference, and policy shifts move asset prices. See breakeven inflation rates explained.

Policymakers at the Fed have targeted headline PCE since 2000, formalized at 2% in January 2012 under Bernanke. The choice reflects PCE’s broader scope, faster substitution adjustment, and lower vulnerability to CPI-specific volatility from shelter measurement lags. See why the Fed targets 2%.

A common misreading is treating the gap between CPI and PCE as evidence that one index is “wrong.” The gap reflects deliberate methodological choices, each appropriate for different questions. Markets need to know which index drives which policy or contractual mechanism.

Practical observation

Go deeper

Related questions

• How is CPI actually calculated?
• Why does the Fed have a 2% inflation target?
• Why does the Fed watch median and trimmed-mean CPI?
• What is the difference between core and headline inflation?

Frequently asked questions

Why did the Fed switch to targeting PCE instead of CPI?

The Fed shifted its primary inflation gauge from CPI to PCE around 2000, with formal codification in the 2012 statement on longer-run goals. The reasoning, articulated by Greenspan and later Bernanke, centered on PCE’s broader scope, more flexible methodology, and reduced vulnerability to substitution bias. PCE also incorporates revised data over time, allowing methodology improvements to propagate through historical series — a feature CPI lacks because of its institutional role in indexing contracts and entitlements.

Is the CPI-PCE gap stable over time?

The gap is generally positive (CPI > PCE) but varies substantially with economic conditions. During shelter-driven inflation episodes, the gap widens because CPI weights shelter more heavily. During healthcare cost accelerations, the gap narrows or inverts because PCE weights healthcare more heavily. The 2021-2022 episode produced one of the largest gaps on record, peaking near 200bp.

Which index is better for measuring real wages?

The choice depends on the analytical question. For measuring household purchasing power on direct out-of-pocket spending, CPI deflators are typical. For measuring economy-wide real consumption, PCE deflators are standard in national accounts. The BLS publishes real average hourly earnings using CPI-W (urban wage earners) — a narrower index than CPI-U typically cited in news media.

Last updated — 1 May 2026