CPI vs PCE - How Inflation Is Measured
The United States has two primary measures of consumer inflation, and they frequently tell different stories. The Consumer Price Index, published by the Bureau of Labor Statistics, is the most widely recognized and the one that generates the biggest market reactions on release day. The Personal Consumption Expenditures price index, published by the Bureau of Economic Analysis, is the Federal Reserve's preferred inflation gauge and the one that actually determines monetary policy. Between 2022 and 2024, the gap between CPI and PCE ran as wide as a full percentage point, meaning that the inflation rate investors saw on the news was meaningfully different from the inflation rate driving the Fed's rate decisions.
Understanding both measures, their methodological differences, their respective strengths and weaknesses, and when each matters most, is a prerequisite for interpreting inflation data accurately. Investors who trade on CPI without understanding why PCE tells a different story frequently misjudge the monetary policy outlook, which is among the costliest analytical errors in equity and fixed income markets.
What CPI Measures
The Consumer Price Index measures the average change in prices paid by urban consumers for a fixed basket of goods and services. The BLS defines the basket based on the Consumer Expenditure Survey, which tracks how households actually spend their money. The basket is updated periodically but held relatively stable between updates. Prices for approximately 80,000 items in 75 urban areas are collected monthly by BLS economic assistants who physically visit stores and service establishments or collect prices online.
CPI is reported in several variants. CPI-U (All Urban Consumers) covers approximately 93% of the U.S. population and is the most commonly cited version. CPI-W (Urban Wage Earners and Clerical Workers) covers about 29% and is used for Social Security cost-of-living adjustments. Core CPI excludes food and energy prices, which are volatile and can obscure the underlying trend. CPI-E is an experimental index focused on the spending patterns of the elderly.
The CPI basket weights are based on what consumers purchased in the base period. The current major categories and approximate weights are: housing (approximately 36%), food (approximately 13%), transportation (approximately 16%), medical care (approximately 7%), education and communication (approximately 7%), recreation (approximately 5%), apparel (approximately 3%), and other goods and services (approximately 3%).
Housing is by far the largest component, and the way CPI measures housing costs is one of the most consequential methodological choices in all of economics.
What PCE Measures
The Personal Consumption Expenditures price index measures the change in prices of goods and services consumed by individuals and nonprofit institutions serving individuals. It is derived from a broader data set than CPI, incorporating not just consumer survey data but also business survey data, Medicare and Medicaid records, and other administrative sources.
The Federal Reserve explicitly targets PCE inflation at 2% over the medium term. This choice was formalized in January 2012 under Chairman Ben Bernanke and reflects the Fed's view that PCE is a more comprehensive and methodologically sound measure of consumer price changes.
PCE covers a broader scope of expenditure than CPI. It includes spending by and on behalf of consumers, such as employer-provided health insurance and government-provided healthcare (Medicare and Medicaid). CPI only includes out-of-pocket spending by consumers. This broader scope means PCE captures the full cost of healthcare consumption, not just the portion that consumers pay directly.
Core PCE, which excludes food and energy, is the specific metric the Fed focuses on most intensely. When Fed officials discuss whether inflation is "returning to target," they are almost always referring to year-over-year core PCE.
The Key Methodological Differences
Several technical differences between CPI and PCE explain why they produce different readings.
Weighting methodology. CPI uses fixed weights based on a base-period consumer spending survey, updated every two years. PCE uses chain-weighted methodology that adjusts weights every month based on actual spending patterns. When the price of one good rises, consumers tend to substitute toward cheaper alternatives. PCE captures this substitution effect; CPI does not, at least not between major categories. This is why CPI tends to overstate inflation slightly compared to PCE. Economists refer to this as "substitution bias."
Housing weight. Shelter costs account for approximately 36% of the CPI basket but only about 15% of the PCE basket. This difference has enormous practical significance. Owners' equivalent rent (OER), a measure of what homeowners would pay to rent their own homes, is a large component of both indices but carries far more weight in CPI. When shelter inflation runs hot, as it did in 2022-2023, CPI rises faster than PCE. When shelter inflation cools, CPI falls faster.
Healthcare measurement. CPI measures out-of-pocket healthcare costs paid by consumers. PCE includes all healthcare consumption, including spending paid by employers and the government. In practice, PCE healthcare inflation tends to be lower than CPI healthcare inflation because Medicare and Medicaid reimbursement rates often grow more slowly than out-of-pocket costs.
Scope. CPI covers out-of-pocket spending by urban consumers. PCE covers all personal consumption, including items purchased on behalf of consumers by third parties. Financial services, for example, receive a higher weight in PCE because PCE imputes the value of financial services consumed (like checking account services provided for free in exchange for maintaining a balance).
Formula effect. CPI uses a modified Laspeyres (fixed-basket) formula. PCE uses a Fisher-Ideal index that averages the Laspeyres and Paasche indices. The Fisher-Ideal formula better accounts for substitution between goods and generally produces a lower inflation reading.
The Practical Gap
The cumulative effect of these methodological differences is that CPI consistently runs higher than PCE, typically by 0.3 to 0.5 percentage points per year on a core basis. During periods of rapid shelter inflation, the gap can widen to 0.8 to 1.0 percentage points because of the much larger housing weight in CPI.
In the 2022-2023 inflationary period, this gap was particularly consequential. Core CPI was running above 5% while core PCE was closer to 4.5%. The difference of half a percentage point may seem small, but at the margin of monetary policy decisions, it mattered. A Fed that targets PCE at 2% was dealing with inflation roughly 2.5 percentage points above target, not 3.0 or more.
The gap also created confusion in public discourse. Media coverage focused on CPI because it is released first and is more familiar to the public. But the inflation number that determined whether the Fed would raise rates, hold, or cut was PCE. Investors who positioned based on CPI readings without adjusting for the PCE differential were sometimes surprised by the Fed's policy signals.
Market Reactions: CPI vs. PCE Release Days
CPI is released approximately two weeks before PCE and receives significantly more market attention. The CPI release at 8:30 AM Eastern on its designated day routinely produces the largest intraday moves in Treasuries, equities, and currencies of any data release on the monthly calendar.
The market reaction to CPI is amplified by several factors. Options expirations are sometimes structured around CPI release dates. Algorithmic trading systems are calibrated to react to CPI surprises. And the media coverage of CPI is far more extensive than PCE, meaning that retail investor attention and sentiment are more affected by CPI.
PCE, released roughly two weeks later as part of the Personal Income and Outlays report, typically generates a smaller market reaction because much of the new information has already been incorporated from the CPI release. The components of CPI and PCE overlap sufficiently that analysts can estimate PCE with reasonable accuracy once CPI is known. A large CPI surprise that has already been digested by the market will not produce a comparable surprise in the subsequent PCE release.
However, when PCE diverges from what CPI implied, the market reaction can be significant. If CPI runs hotter than expected but the components that drive the CPI-PCE gap (particularly housing) are responsible, informed investors will anticipate that PCE will not be as hot. When PCE confirms this, the market may actually rally despite the earlier CPI scare.
Using Both Measures in Investment Analysis
For portfolio management, the practical approach is to use both measures for different purposes.
CPI is more useful for assessing consumer purchasing power, real wage growth, inflation-linked securities (TIPS are indexed to CPI-U, not PCE), cost-of-living adjustments for Social Security and some pension benefits, and rent-sensitive analysis.
PCE is more useful for predicting Federal Reserve policy, understanding the inflation trajectory from the Fed's perspective, and assessing whether inflation is genuinely returning to target or merely appearing to moderate due to methodology differences.
The gap between CPI and PCE is itself informative. A widening gap driven by shelter inflation suggests that the housing component is distorting CPI relative to the broader inflation picture. A narrowing gap suggests that inflation is becoming more uniform across the economy. The housing component of CPI has a well-documented lag, reflecting rental agreements signed months earlier. This lag means that real-time improvements in housing affordability may take 12 to 18 months to fully appear in CPI data.
For equity sector analysis, both inflation measures have sector-specific implications. Healthcare stocks are more affected by PCE healthcare inflation (which captures total spending) than CPI healthcare inflation (which captures only out-of-pocket costs). Real estate and housing-related stocks are more sensitive to the shelter component, where CPI and PCE diverge most significantly.
The takeaway is not that one measure is right and the other wrong. Both capture real aspects of price changes in the economy. The investor's job is to understand what each measures, where they differ, and which is most relevant for the specific investment question at hand. For the most consequential question in macro investing, where is the Fed headed, PCE is the answer that matters.
Put these principles into practice. Track fundamentals, build portfolios, and analyze stocks with AI-powered insights.
Start Free on GridOasis →