When people insist that inflation feels worse than the official numbers suggest, economists often respond with a mixture of condescension and confusion. The data is the data, they say. Your memory is unreliable. You notice price increases but not decreases. You are suffering from recency bias.
They are not entirely wrong. But they are not entirely right either, and the persistent gap between measured inflation and experienced inflation reveals something important about the limits of macroeconomic statistics — and about whose economy those statistics are designed to describe.
The basket problem
Inflation indices work by tracking the price of a representative basket of goods and services over time. The composition of this basket is meant to reflect how an average household spends its money. But averages are treacherous things. A household earning in the top quintile spends roughly twelve percent of its budget on food; a household in the bottom quintile spends closer to thirty percent. When food prices surge, the official inflation rate — weighted toward that hypothetical average consumer — understates the impact on lower-income families while overstating it for the wealthy.
The same asymmetry applies to housing, healthcare, childcare, and education. These categories have seen price growth that consistently outpaces headline inflation over the past several decades, and they consume disproportionate shares of middle- and working-class budgets. Meanwhile, the prices of televisions, clothing, and consumer electronics have fallen in real terms — good news if you are in the market for a new screen, less relevant if you are trying to pay rent.
The quality adjustment controversy
Statistical agencies employ a technique called hedonic adjustment to account for quality improvements. If a new car costs more than last year's model but includes better safety features and fuel efficiency, some portion of that price increase is reclassified as a quality gain rather than inflation. The logic is defensible in theory. In practice, it means the official statistics can show modest inflation even as consumers face higher sticker prices for goods they have no choice but to buy.
The adjustments are not arbitrary — they follow rigorous methodologies developed over decades. But they embed assumptions about what counts as value that may not match how actual households experience their budgets. A parent buying a laptop for a child's schoolwork does not care that the processor is faster than last year's model. They care that they had to spend more money.
The frequency illusion
There is also a psychological dimension that economists are right to highlight. Humans notice losses more acutely than gains — a well-documented cognitive bias. We register the gallon of milk that now costs more but not the streaming subscription whose price has held steady. We remember the restaurant meal that shocked us but not the dozen unremarkable transactions in between.
This does not mean the feeling is wrong, only that it is not the whole picture. The official statistics and the lived experience are both partial truths, each capturing something real about an economy too complex to reduce to a single number.
Our take
The honest answer is that inflation is not one thing. It is a distribution of price changes experienced differently by different people in different circumstances. The headline number is useful for monetary policy, for comparing economies across borders and decades, for the serious work of macroeconomic management. But it was never designed to validate your grocery receipt, and expecting it to do so is asking the wrong question of the wrong tool. Your bills are not lying. Neither are the statisticians. They are simply measuring different things.
