What Is Core Inflation?

In April 2026, the Consumer Price Index (CPI, the broadest measure of what Americans pay for goods and services) showed prices rising 3.8 percent from a year earlier.¹ That headline grabbed attention. But the number the Federal Reserve was actually watching came in lower: core Personal Consumption Expenditures (PCE), the price index stripped of food and energy, ran at 3.3 percent for the same month.¹ The gap between those two figures, 0.5 percentage points, is small enough to seem technical. It is not. It explains how central bankers decide whether to tighten borrowing conditions, which in turn determines the rate on your next mortgage and what your savings account pays.

Let's start with how core inflation actually works.

Why strip food and energy out

Food and energy prices move for reasons that have little to do with underlying monetary conditions. A drought in the Midwest pushes corn prices up. A hurricane takes a Gulf Coast refinery offline and gasoline jumps $0.40 a gallon overnight. OPEC cuts production and crude oil spikes. None of these are signals that the economy is running too hot or that the Fed's interest rate policy needs to change. They are supply shocks, temporary in most cases, and they fade without any policy response.²

If the Fed tightened monetary policy every time a supply shock pushed headline inflation up, it would be raising interest rates during the same periods when supply disruptions were already slowing the economy, which is exactly the wrong timing. Core inflation is designed to filter those spikes out and surface the underlying demand-side pressure that monetary policy can actually address.

The distinction showed up clearly in 2022. Headline CPI peaked at 9.1 percent in June of that year.⁶ Core CPI, at the same moment, sat at 5.9 percent.² The 3.2-point gap was mostly a commodity shock: the Russian invasion of Ukraine had disrupted global energy and grain markets earlier that year, sending gasoline and food prices into a steep climb that had nothing to do with American consumer spending patterns. Both measures were high, and both warranted concern. But a policymaker who anchored solely on headline inflation would have overestimated the persistence of the pressure. Core told the story with less noise.

The two versions: CPI versus PCE

This is where a nuance matters, and most coverage skips it. There are two main core inflation measures, and the Fed watches one while the public mostly hears about the other.

Core CPI, or CPI less food and energy, is calculated by the Bureau of Labor Statistics (BLS) and is what financial news typically leads with. It is familiar, widely reported, and published monthly. The Fed does not target it directly.

Core PCE is calculated by the Bureau of Economic Analysis (BEA) and is the Fed's official benchmark. The Federal Open Market Committee (FOMC), the body that sets the federal funds rate, targets "2 percent over the longer run, as measured by the annual change in the price index for personal consumption expenditures."³ That language is specific: PCE, not CPI. The two indexes differ in how they weight categories, with PCE giving more weight to healthcare (where government programs cover a significant share of spending) and less weight to housing costs than CPI does. PCE tends to run about 0.3 to 0.5 percentage points below CPI for structural reasons, which means the gap between 3.3 percent and the 2 percent target is real, not a measurement artifact.⁴

The chart above shows core CPI since the 1990s. The two features worth noting are the extended period of low, stable readings from roughly 2012 through 2021, where core held within a point of the Fed's target, and the sharp spike beginning in 2021. That spike, and the Fed's response to it, is what drove mortgage rates from the historic lows of 2020 to multi-decade highs by 2023 and 2024.

What the data does not show: the shelter problem

Core inflation has its own blind spot, and it is large enough to matter. The shelter component, which captures rent and equivalent costs for homeowners, carries a weight of roughly 36 percent within core CPI.⁵ It is the single biggest line item. And it is systematically late.

The BLS measures shelter costs using a methodology that lags actual market rents by 12 to 18 months. When apartment rents spike fast, the official shelter index takes more than a year to fully reflect that increase. The same lag applies on the way down: when actual market rents begin to soften, core CPI keeps recording elevated shelter costs well after the real-world condition has changed.

This matters because shelter lag is one reason core inflation looks stickier than it may actually be. Analysts who track real-time market rents, from sources like Zillow and Apartment List, have consistently shown rent growth cooling well before the BLS shelter index reflected it. So while core PCE at 3.3 percent is a genuine number from a legitimate methodology, it almost certainly overstates where underlying inflation will settle once the shelter measurement catches up. Roughly 0.5 to 0.8 percentage points of that figure, by some estimates, are an artifact of the lag rather than ongoing demand pressure.⁵

This is not a reason to dismiss core PCE. It is a reason to read it alongside real-time data and to understand what the Fed sees when it looks at the same dashboard.

Why this reaches your wallet

Now shift to where this connects to your money directly.

The Fed uses core PCE as its main gauge when deciding where to set the federal funds rate, the overnight rate at which banks lend to each other that serves as the anchor for most other borrowing costs in the economy.³ When core PCE is comfortably at or below 2 percent, the Fed has room to keep rates lower, which flows into cheaper mortgages, lower auto loan rates, and the rate your bank pays on savings accounts. When core PCE runs above 2 percent, as it has since early 2021, the Fed holds rates elevated to slow borrowing and spending, which eventually pulls inflation down but also makes credit more expensive across the board.

At 3.3 percent core PCE, the Fed remains in restrictive territory. While the trajectory has improved considerably from the peak of around 5.6 percent in early 2022, there is still meaningful distance between where core inflation sits and where the Fed needs it to be.¹ That gap is the direct reason the FOMC has held its policy rate at elevated levels through 2025 and into 2026, and why mortgage rates have not returned to the 3-percent range many buyers remember from 2020 and 2021.³

There is also a headline-versus-core dynamic in policy communication worth understanding. If headline inflation runs at 4 percent but core is near 2 percent, the Fed reads that as a commodity shock that will likely self-correct, and it is unlikely to tighten hard. If core is running above 3 percent with no clear downward trend, the Fed treats that as structural demand pressure that its tools can address, and it holds rates higher for longer. This asymmetry means the consumer price number that dominates the news is often the less relevant one from a rate-decision standpoint.

Overall, core inflation is not a trick played by economists to hide what food and groceries actually cost. It is a diagnostic tool designed to separate the signal from the noise in price data. The signal right now, at 3.3 percent core PCE, is that underlying inflation has come down substantially from its peak but has not reached the level where the Fed will ease up. When it does, credit gets cheaper. Until then, the cost of borrowing reflects the distance still left to close. Knowing which number the Fed is actually watching is the first step toward understanding why rates are where they are, and what it would take to move them.