Inflation is a broad, sustained rise in prices across an economy. When inflation is positive, the purchasing power of a dollar falls: you need more dollars to buy the same goods and services than you did before. Economists don’t track the price of “everything.” They track price indexes—statistical measures designed to approximate a “typical basket” of purchases.

Deflation is the opposite: a broad, sustained fall in prices. A dollar buys more over time. It sounds pleasant until you notice what tends to come with it. In modern advanced economies, deflation is rare, and when it appears it often coincides with weak demand, wage pressure, or debt stress. Falling prices can encourage consumers and businesses to wait, suppressing spending and investment when the economy needs it most.

Then there’s the word that would save half the internet from misunderstanding: disinflation. Disinflation means inflation is still positive—prices are still rising—but the rate is slowing. If inflation falls from 4% to 2%, that doesn’t mean prices fell. It means prices rose more slowly.

Central banks generally prefer low, stable, predictable inflation. The European Central Bank is unusually candid about why: it frames 2% inflation as a buffer against deflation and measurement problems, and as a help for economic adjustment over time. The ECB’s emphasis on symmetry—aiming for 2% “over the medium term”—signals that a small cushion beats flirting with a zero-inflation edge where deflation risk rises. (ECB)

In the United States, the Federal Reserve’s latest policy statement (Jan. 28, 2026) said inflation remained “somewhat elevated” as it held the fed funds target range at 3.5%–3.75%. (Federal Reserve) That line—somewhat elevated—matters because it signals how officials view the direction of travel, not just the level of prices people already face.

Households don’t buy the same basket. A renter in a city, a homeowner in the suburbs, a remote worker, and a commuter can live in the same economy while feeling different inflation. Official indexes try to measure a representative average, not your personal ledger.

Inflation headlines usually come from a simple idea: compare today’s prices with last year’s. The hard part lies in choosing what to compare. Price indexes approximate a basket of goods and services, assigning weights based on what households buy in aggregate. That basket is a model, not a literal shopping list.

The benefit of a model is consistency. If you want to know whether prices are rising broadly—not just whether avocados or airline tickets spiked—you need a common yardstick. The drawback is obvious: the yardstick doesn’t fit every home.

Consider two households with similar incomes:

- A young renter whose largest expense is rent and who spends heavily on commuting.
- A retired homeowner who owns their home outright and spends more on healthcare services.

Even if overall inflation is the same, the renter is more exposed to shelter changes and fuel. The retiree may feel inflation most sharply where services rise. Official figures aren’t “wrong” for either household; they’re averages that can miss intensity.

That gap explains why public trust frays when inflation cools on paper but budgets stay tight in practice. High prices can persist even when inflation declines, because the index measures the rate of change, not the level. Affordability doesn’t reset simply because the climb slows.

In the U.S., most people meet inflation through the Consumer Price Index (CPI), produced by the Bureau of Labor Statistics (BLS). CPI is the familiar headline, the one that moves markets on release morning and shapes political talking points by lunchtime.

Policymakers at the Federal Reserve, however, anchor their 2% goal to a different measure: the Personal Consumption Expenditures (PCE) price index, produced by the Bureau of Economic Analysis (BEA). PCE covers a broader set of expenditures and uses different weights than CPI. It is also typically revised more—an advantage if you want the most accurate history, a frustration if you want clean, fixed headlines.

A concrete data point helps: the BEA reported that in January 2025, the PCE price index rose 2.5% year-over-year, and core PCE rose 2.6%. (BEA) Even when CPI and PCE point in the same direction, they can differ in level.

The short version is composition and methodology:

- Different weights: CPI and PCE don’t assume households allocate spending the same way.
- Different scope: PCE includes a wider range of expenditures.
- Different revision practices: PCE tends to be revised more, which can smooth or alter prior readings.

None of that means one is honest and the other is propaganda. It means each was built for a slightly different purpose—CPI for a consumer-facing snapshot, PCE for a broader, policy-focused view.

When inflation feels personal, “core” can feel like an insult. Core inflation strips out food and energy because those prices can swing sharply from month to month. The goal isn’t to dismiss grocery bills or gas prices; it’s to see whether the underlying trend is persistent.

Defenders argue that monetary policy shouldn’t whip around every time energy prices jump, especially when those moves can be driven by global shocks outside a central bank’s control. Critics respond with a fair point: households can’t exclude food and energy. People live in the headline economy, not the core economy.

The tension is real, and it helps explain why policymakers reference both measures. Headline inflation tells you what is happening now to total costs. Core inflation tries to tell you what is likely to keep happening once temporary spikes fade.

A central bank’s job is not to subsidize pain; it is to steer demand and expectations. Food and energy can be volatile; policy tools are blunt and slow. Filtering volatility can help officials avoid overreacting and accidentally causing more economic damage than the original spike.

At the same time, ignoring lived experience can erode legitimacy. A sensible reading is that headline inflation drives public sentiment, while core inflation informs policy strategy.

If core inflation is one filter, economists have built others. Some approaches aim to remove the most extreme price changes—up or down—so that a handful of wild categories don’t dominate the overall reading.

The Cleveland Federal Reserve publishes two widely cited measures:

- Median CPI
- 16% trimmed-mean CPI

For December 2025, Cleveland Fed data show Median CPI at 3.1% year-over-year and the 16% trimmed mean at 3.0% year-over-year. (Cleveland Fed) Those figures matter not because they are magic, but because they illustrate persistence. When headline inflation swings due to a few volatile categories, median and trimmed-mean measures can reveal whether inflation pressure is broad.

A trimmed-mean measure lops off the biggest price moves—both increases and decreases—then averages what remains. Median CPI finds the middle category: half of prices are rising faster, half slower.

The point isn’t to create a better headline for public relations. The point is diagnostic. If inflation looks high because a narrow set of prices is surging, the policy response might differ from a world where inflation is embedded across many categories.

Inflation reporting has another layer that confuses even attentive readers: seasonal adjustment. Many prices follow predictable annual patterns—holiday travel, back-to-school clothing, winter heating. Seasonal adjustments attempt to remove those recurring patterns so month-to-month changes better reflect underlying movement.

The BLS recalculates seasonal factors each year with the January CPI. That process can revise seasonally adjusted indexes for the previous five years. For the January 2026 CPI, revisions covering Jan 2021–Dec 2025 were scheduled to be published alongside the release. (BLS)

Revisions don’t mean the data were fabricated and later “fixed.” They mean statisticians updated the model as new information accumulated—like improving the calibration of an instrument.

Because of a funding lapse/partial shutdown, the BLS delayed the January 2026 CPI release to Friday, Feb. 13, 2026 at 8:30 a.m. ET. (BLS) Readers looking for the “latest inflation number” should know that timing, and should also know what comes with it: the annual seasonal-factor update and revisions.

A common public frustration sounds reasonable: why not aim for zero inflation? If inflation erodes purchasing power, wouldn’t zero be ideal?

Central banks have answered that question in practice by choosing a small positive target, commonly 2%. The ECB explicitly frames 2% as a buffer against deflation and measurement issues and as helpful for macroeconomic adjustment. (ECB) In other words, a little inflation grease helps wages and prices adjust without requiring outright cuts that can be economically and politically painful.

In the U.S., the Federal Reserve’s January 28, 2026 statement held rates at 3.5%–3.75% and described inflation as “somewhat elevated.” (Federal Reserve) The level of rates isn’t the same as the inflation rate, but the message matters: policy remains restrictive enough to lean against inflation that officials still view as above comfort.

Skeptics argue that any steady inflation is a quiet tax on savers, and they’re not wrong about the arithmetic. Defenders argue that the alternative—regular flirtation with deflation—can be worse, especially in debt-heavy economies where falling prices raise the real burden of what people owe.

Both sides share a hidden premise: inflation is never just a statistic. It reshapes bargaining power, expectations, and politics. That’s why a 2% target is less a moral claim than a risk-management choice.

Inflation literacy isn’t about memorizing acronyms. It’s about asking better questions than the headline invites.

When you see an inflation number, read it in three passes:

1. Level vs. direction: Are prices still rising, just more slowly (disinflation), or actually falling (deflation)?
2. Measure: Is the headline CPI or the Fed’s preferred PCE? Are you looking at headline or core?
3. Breadth: Are filtered measures (median, trimmed mean) telling a similar story about persistence?

Then connect it back to your own household basket. If your largest expense is rent, you will care about shelter dynamics more than a commuter who cares about fuel. If you recently renewed insurance or bought a car, your personal inflation can be unusually high even in a cooling economy.

When inflation falls, relief often arrives as slower increases—not as a rewind to old prices. That can still matter. Slower inflation gives wages a chance to catch up. It allows planning. It reduces the pressure for aggressive rate hikes that can trigger job losses.

The price level is where you live. The inflation rate is where policymakers steer.

Inflation debates turn toxic when people argue past the same word. “Inflation” can mean a rate, a feeling, a political accusation, or a personal budget crisis. The official definition is narrower: a broad, sustained rise in prices, measured by imperfect but useful indexes.

Those indexes disagree for intelligible reasons. CPI and PCE weigh the world differently. Core and “supercore” measures strip volatility to find persistence. Seasonal adjustments revise the recent past because the economy isn’t a lab experiment—it’s a moving target.

Once you internalize that, the headlines get less hypnotic. You stop expecting falling inflation to mean falling prices. You stop treating one month’s number as a verdict. You start looking for what actually matters: whether price increases are broad, whether they’re persistent, and whether your own household basket is drifting away from the average.

The next time someone says inflation is “back to normal,” ask a sharper question: normal for which measure, over what time frame, and for whom?