PCE vs. CPI: Why the Fed Watches One and Markets Watch the Other

Two inflation reports. Two different numbers. One Federal Reserve policy decision. Understanding why the Fed uses PCE while the financial media leads with CPI is not just a technical footnote — it's central to understanding how monetary policy actually gets made and why markets sometimes react differently to the same underlying inflation data.

The Consumer Price Index, published by the Bureau of Labor Statistics, is the older and more widely cited of the two measures. It tracks the prices paid by urban consumers for a fixed basket of goods and services, updated periodically to reflect changing spending patterns. The Personal Consumption Expenditures price index, published by the Bureau of Economic Analysis, is the Fed's preferred gauge. It covers a broader population — all U.S. households, not just urban consumers — and uses a methodology that adjusts the basket in real time as spending patterns change.

The most important structural difference between the two is how they handle substitution. When the price of one good rises sharply, consumers typically shift their spending toward cheaper alternatives. PCE captures this substitution effect immediately, because it uses a chain-weighted methodology that updates the basket each period based on actual spending data. CPI uses a fixed basket that is only updated every two years, which means it can overstate inflation during periods when consumers are actively trading down. This is one reason PCE tends to run about 0.3 to 0.5 percentage points below CPI in any given year.

The weighting of housing costs is another significant difference. CPI assigns roughly 33% of its weight to shelter costs, which are measured using a concept called "owners' equivalent rent" — an estimate of what homeowners would pay to rent their own homes. This measure is notoriously slow to reflect real-time changes in the housing market, which is why CPI shelter inflation remained elevated well into 2024 and 2025 even as actual rents were cooling. PCE assigns a lower weight to shelter — around 15% — and includes a broader range of housing-related expenditures. This makes PCE less distorted by the lag in shelter measurement.

Medical care costs are weighted very differently as well. CPI measures what consumers pay out of pocket for healthcare. PCE includes all healthcare spending, including the portion paid by employers and government programs like Medicare and Medicaid. Since employer and government healthcare spending tends to grow more slowly than out-of-pocket costs, PCE's healthcare component typically shows lower inflation than CPI's.

In practice, the gap between the two measures has been meaningful in the current cycle. As of early 2026, headline CPI was running approximately 0.4 to 0.5 percentage points above headline PCE on a year-over-year basis. Core CPI — which strips out food and energy — was running roughly 0.3 to 0.4 percentage points above core PCE. For the Fed, which is targeting 2% PCE inflation, this means the economy can have CPI running at 2.4% to 2.5% and still technically be at the Fed's target. Investors who only track CPI sometimes misread the Fed's reaction function as a result.

The practical implication for portfolio management is this: when the Fed says it needs to see inflation return to 2%, it means 2% PCE, not 2% CPI. If core PCE is at 3.0% and core CPI is at 3.4%, the Fed is further from its target than the CPI headline suggests. Conversely, if PCE cools faster than CPI — which can happen when shelter costs are distorting CPI upward — the Fed may be closer to cutting rates than the CPI headline implies. Tracking both measures, and understanding the gap between them, gives investors a more accurate read on where monetary policy is actually headed.

Not financial advice. This article is for informational and educational purposes only.