Headline vs. Core Inflation: Which Number Should You Watch?

In May 2026, two true numbers described the same economy and disagreed by a mile. The government said inflation ran 4.2 percent over the prior twelve months, a three-year high, released by the Bureau of Labor Statistics (BLS) on June 10. The same report said inflation ran 2.9 percent. Both came from the same agency, the same month, the same data. One of them is the number on the news ticker. The other is the number the Federal Reserve actually steers by. This article is about why they split, and which one you should keep an eye on.

The verdict, up front, before we unpack it: headline inflation is what you feel; core inflation is what the Fed acts on. If you want to know whether your paycheck is keeping up at the pump and the grocery register, watch headline. If you want to know where interest rates (your mortgage quote, your car loan) are heading next, watch core. They answer different questions, and in a month like May 2026 they answer them very differently.

This is for anyone who has read "inflation hit 4.2 percent" and wondered why the Fed didn't seem to panic. By the end, you'll be able to tell the two measures apart, explain why economists bother with both, and know which one to reach for depending on what you're trying to figure out. I'll define each term before I lean on it.

Two measures, one basket

Both numbers start in the same place: a giant imaginary shopping cart. Every month the BLS prices a fixed basket (the goods and services a typical urban household buys, from rent to ramen to gasoline) and tracks how much that basket's total cost changes. That change, measured against the same month a year earlier, is inflation.

Headline inflation, officially the "all items" Consumer Price Index, or CPI, counts the whole basket. Nothing left out. In May 2026 the all-items index was up 4.2 percent over the year. That is the figure that leads the broadcast, because it's the honest answer to "did life get more expensive?" Everything you actually buy is in there.

Core inflation takes that same basket and pulls two aisles out: food and energy. What's left, the "all items less food and energy" index, rose 2.9 percent over the year. Same cart, two items removed, and the number drops by more than a full percentage point. That gap is the whole story, so let's go find what's sitting in those two aisles.

What blew the gap open: one fuel pump

The thing that pried headline and core apart in May 2026 has a name, and it's gasoline. Over the twelve months ending in May, the gasoline index rose 40.5 percent. Not a typo: gas was up roughly two-fifths from a year earlier, the result of Middle East supply pressure, with the Strait of Hormuz throttled and transit costs spiking. Energy as a whole climbed 23.5 percent over the year.

Now the arithmetic clicks into place. The BLS reported that energy alone was responsible for more than 60 percent of the month's overall price increase. Strip energy (and food) out, and most of the 4.2 percent simply evaporates, which is exactly how you get from a 4.2 percent headline to a 2.9 percent core. The gap between the two numbers this May is, in large part, one fuel pump doing the heavy lifting.

Hold that image, because it explains the next question almost by itself: if one volatile item can swing the headline that hard, why would anyone setting policy trust the headline?

Why strip out food and energy at all?

It feels wrong at first. Food and energy are the most real expenses most households have, so leaving them out sounds like cooking the books. So let me defuse that worry directly: nobody is hiding food and energy. They're fully counted in the headline number, the one that came in at 4.2 percent. Core is not a replacement for headline. It's a second lens placed beside it.

The reason for that second lens is volatility. Volatile just means prices that jump around fast and far on short notice. Gasoline at +40.5 percent over a year is the textbook case: it can spike on a refinery outage or a conflict an ocean away, then fall back just as quickly. Food does the same on droughts and freezes. Those swings are loud, but they're often temporary, and the part that matters for policy is this: they're driven by shocks that interest rates can't fix. The Fed cannot raise rates to reopen a shipping strait or end a frost.

So economists built core inflation to hear the quieter, steadier signal underneath the noise: the rate of inflation likely to persist. The BLS itself notes that because the all-items index includes things whose prices rise and fall frequently, it's poorly suited to measure that underlying, medium-run rate. Strip the loudest, least-controllable items, and what remains tracks where inflation is genuinely settling. I'd argue that's the honest case for core: not that food and gas don't matter, but that they drown out the trend the Fed is actually trying to read.

Why the Fed leans on core

This is where core earns its keep. Research out of the Federal Reserve (the Philadelphia Fed's work on core inflation as a predictor of total inflation is the clean reference) finds that today's core inflation tends to forecast tomorrow's total inflation better than today's total inflation does. Read that twice: if you want to guess where overall prices are heading, the core number is the better crystal ball.

That matters because monetary policy works on a lag. When the Fed changes interest rates, the full effect on the economy takes months (often more than a year) to land. So a rate-setter can't steer by today's headline; by the time a rate change bites, that gasoline spike may already have unwound on its own. The Fed has to aim at where inflation is going, and core is the cleaner read on that destination. It can be assumed that's why, with headline at a scary 4.2 percent in May, the Fed didn't slam the brakes: it was looking at a 2.9 percent core and judging much of the headline to be a temporary energy shock.

Which number is yours?

So which do you watch? It depends entirely on the question you're asking, and here is a simple way to split it.

If your question is about your own budget (am I paying more for groceries, is filling the tank eating my paycheck, should I rethink this month's spending), watch headline. That 4.2 percent is the real cost of your real basket, gasoline very much included. Your wallet doesn't get to exclude the volatile stuff.

If your question is about the future (where are interest rates going, should I lock a mortgage rate now or wait, what's the Fed likely to do next), watch core. That 2.9 percent is the signal the Fed is steering by, and the Fed is what moves the rates on your loans. When core stays elevated, expect the Fed to stay tight. When core cools, rate cuts get easier to imagine.

The trap is using one number to answer the other's question. Reading a high headline as a guaranteed rate hike, or a calm core as proof your grocery bill is fine, gets you the wrong answer in a month exactly like this one, where the two figures sat 1.3 points apart and told opposite-feeling stories.

So the next time a headline shouts that inflation hit some alarming number, do one thing before you react: ask whether they mean the whole basket or the core. In May 2026 the difference between those two words was 4.2 percent versus 2.9 percent: a three-year-high panic, or a slow simmer, depending on which you read. Same economy. Two numbers. Now you know which one is talking to you.