WhitmanTrading

What Is the Consumer Price Index?

Consumer price index measures the changing cost of a representative basket of goods and services bought by households, and is the standard measure of inflation. The basket is an average across the population, so the published rate rarely matches what any particular household experiences.

The consumer price index tracks what a representative basket of goods costs. It is the most consequential statistic most governments publish, and the representativeness is where every argument about it starts.

How it works

A price series representing a basket of goods.
The consumer price index measures a basket of goods. Illustrative chart - not real market data.

A basket is defined from household spending surveys — what people actually buy, in what proportions, across the whole population.

A steady series with prices collected repeatedly.
The standard measure of inflation. Illustrative chart - not real market data.

Prices are collected repeatedly from thousands of outlets, and the weighted total is compared to the same basket in a base period.

A rising series where the basket is a population average.
The basket is an average across everybody. Illustrative chart - not real market data.

The weights are population averages. Food might be 10% of the basket because it is roughly 10% of aggregate spending, not because it is 10% of anybody’s spending.

A falling series where individual experience differs.
So it matches almost no individual household. Illustrative chart - not real market data.

Why your experience differs

A choppy series where housing dominates the calculation.
Housing is the largest and most contested component. Illustrative chart - not real market data.

Spending patterns vary enormously. A pensioner, a student and a family with two cars have baskets that overlap only partially, and their inflation rates diverge accordingly.

A slow series where compounding runs over decades.
And different again over a long horizon. Illustrative chart - not real market data.

Housing is the biggest source of divergence. It is the largest component and the hardest to measure, since a house is partly consumption and partly an asset.

A calm series where prices are stable.
A quiet stretch hides what it measures. Illustrative chart - not real market data.

Different countries handle it differently — imputed rent, actual rents paid, or mortgage costs — and those choices produce materially different headline rates from identical underlying prices.

A worked example

Published inflation is 3%. A renter in a city where rents rose 9%, spending a third of income on housing, has personal inflation well above that before anything else is counted.

A homeowner with a fixed mortgage has a housing cost that did not move at all, and personal inflation below the published figure.

A falling series with a stop level marked.
A stop fills where the market is. Illustrative chart - not real market data.

Both are experiencing the same economy. The index is accurate about the aggregate and describes neither household, which is a property of averages rather than a defect in the statistic.

And the difference compounds. Two percentage points a year over a working life is an enormous divergence in real outcomes between two people reading the same headline.

The adjustments that generate arguments

Substitution. When beef becomes expensive people buy chicken, so a fixed basket overstates the cost of maintaining a standard of living. Adjusting for it is defensible and lowers measured inflation.

Quality adjustment. A computer costing the same as last year but twice as fast is recorded as a price fall. The reasoning is sound and the magnitude is a judgement.

New goods. Items enter the basket after they are already widespread, so the early period of falling prices for a new product is frequently missed entirely.

Each adjustment has a respectable rationale and each one lowers the measured rate, which is why critics argue the aggregate effect is systematic understatement — an argument that is genuinely unresolved rather than obviously right or obviously conspiratorial.

The original data

This site’s thirty-year fee measurement: 5 basis points costs 1.5% of the final balance, 20 costs 5.8%, 75 costs 20.2%, 150 costs 36.5%.

Inflation is that arithmetic applied to money itself. At 3% a year, money loses more than half its purchasing power over thirty years — a compounding drag larger than any fee on this table, working on everything not invested.

A candlestick chart annotated with the cost of a round trip.
A round trip costs a share of a bar. Illustrative chart - not real market data.

Which is why nominal returns mislead. A round trip on this site’s series costs 0.0098, about 2% of the median bar range — a cost investors watch carefully, and considerably smaller than the inflation drag most of them do not adjust for.

A price series with volume shown beneath.
Volume and price measure different things. Illustrative chart - not real market data.

Why the number matters so much

Central banks target it. Interest rate policy in most developed economies is set against an inflation target defined on a specific index, so the measurement decides the policy.

Payments are indexed to it. Pensions, benefits, wage agreements and index-linked bonds all reference it, which means a small measurement change moves very large sums.

And contracts reference it. Commercial leases, long-term supply agreements and royalty arrangements are frequently written against a named index series.

Which explains the intensity of methodological debate. Adjusting a statistic sounds technical; it changes government expenditure, monetary policy and the real value of millions of contracts simultaneously.

The variants worth distinguishing

Headline versus core. Core excludes food and energy, which are volatile and often reverse, so it shows the underlying trend more clearly and describes nobody’s actual spending.

CPI versus a retail price index. Some countries publish both, calculated differently, and the gap between them can exceed a percentage point — which matters when one indexes pensions and the other indexes debt.

And harmonised indices exist so countries can be compared on a common methodology, usually differing from each nation’s own headline figure.

Which measure a contract references is a substantive term. Two indices published by the same agency, for the same economy, in the same month, can differ enough to change a payment materially.

When it fails

The characteristic failure is planning a retirement on the headline rate. A saver assumes 2.5% inflation because that is the target and the published figure, and their actual spending is weighted toward healthcare, energy and housing — categories that have consistently risen faster than the aggregate. The index was accurate about the average basket and their basket is not the average one. The error compounds across decades, and it appears at the point when correcting it is least possible.

A candlestick series with a gap through a level.
A gap skips the level entirely. Illustrative chart - not real market data.

A second failure is comparing across countries without checking how each treats housing.

A third is using headline rather than core when looking at a trend, since food and energy are volatile and reverse.

A fourth is ignoring that weights are revised. The basket changes, so long series are not measuring quite the same thing at each end.

A declining series cut short at a decision point.
Inflation at 3%. Is that your rate? Illustrative chart - not real market data.

And a fifth is treating a nominal return as a return. What matters is the amount above inflation, and the two can point in opposite directions.

Inflation covers what the index measures and its effect on savings. Risk premium covers the return investors need above it. And market risk covers the exposure taken in pursuit of that return.

What I actually do

Your personal inflation rate is not the published one, and the gap can be large. If you rent in a city and drive a lot, you are experiencing a different basket from the one being measured — which matters, because that published number is what your pay rise and your savings are judged against.

— Michael Whitman

This page is educational, not financial advice. Test every idea on your own charts before risking money.