People routinely say inflation figures do not match their experience, and understanding how the figure is built explains why that is both true and not evidence of error.
The basket
Statistical agencies define a representative basket of goods and services and track its cost.
Which is weighted by how much households actually spend on each category.
The basket is updated periodically to reflect changing consumption, and those updates are published.
Price collection
Prices are gathered from thousands of outlets, increasingly supplemented by scanner data and web scraping.
Which produces a large sample rather than a few observations.
Collection methodology is documented publicly by most national agencies.
Why it differs from your experience
The index measures an average household's spending pattern.
Which nobody actually has.
Someone spending disproportionately on energy or rent experiences something different from the aggregate, and this is a measurement property rather than a flaw.
Quality adjustment
When a product improves, part of a price rise reflects the improvement rather than inflation.
Which agencies adjust for using defined methods.
The adjustment is genuinely difficult for technology products and is a legitimate area of methodological debate.
Housing costs
Handled differently between countries and between indices.
Which is one of the largest sources of divergence between measures.
Some indices include owner-occupier costs and others do not, which produces materially different headline numbers.
Core versus headline
Core measures exclude volatile food and energy prices.
Which is useful for identifying underlying trends and is not what households pay.
Central banks watch core; households experience headline.
Substitution
People switch to cheaper alternatives when prices rise.
Which some index formulas account for and others do not.
This is a genuine methodological choice with consequences for the reported figure.
Reading a release
Agencies publish category breakdowns alongside the headline.
Which shows what is actually driving the number.
The breakdown is generally more informative than the headline and is rarely reported.
Personal inflation rates
Some statistical agencies publish calculators letting you weight the basket by your own spending.
Which produces a figure closer to your experience.
Households spending heavily on rent, energy or childcare frequently find substantial divergence from the headline.
Producer prices
Measured at the factory gate, upstream of consumer prices.
Which can indicate coming consumer price movement, with an uncertain lag.
The relationship is looser than commonly assumed, since margins absorb some movement.
Wage comparisons
Real wages are nominal wages adjusted for inflation.
Which is the figure that matters for living standards.
Nominal increases below inflation are real decreases, and coverage frequently reports the nominal figure alone.
Index revisions
Figures are occasionally revised as more complete data arrives.
Which is normal statistical practice and receives far less coverage than the initial release.
Central bank targets
Monetary policy is generally set against a specified inflation target over a horizon.
Which means policy responds to expected future inflation rather than to the current figure.
Published minutes explain the reasoning and are more informative than commentary about them.
Why the basket changes
Items enter and leave as consumption patterns shift, and agencies publish the annual changes.
Which is a small news story every year and is genuinely informative about how households are living.
Items removed because they are no longer widely bought tell you as much as the additions.
Regional differences
National figures average across regions with different price levels and different rates of change.
Which means the experienced rate varies geographically.
Some agencies publish regional breakdowns and the differences can be substantial, particularly for housing.
Expectations
Surveys of household and business inflation expectations are published and are watched by central banks.
Which matters because expectations influence wage bargaining and pricing behaviour.
Anchored expectations are a stated policy objective for most central banks.
Historical comparisons
Methodology has changed over decades, which complicates long-run comparison.
Which is why agencies publish back-series constructed on consistent methodology.
Comparing a current figure to a historical one calculated differently is a common error in commentary.
What to read
The statistical release itself, which includes the category breakdown, the contributions to the change and the methodology notes.
These are written to be readable and take about ten minutes.
Why the gap between figures and feeling persists
People notice prices that change frequently and that they pay attention to — food, fuel, energy — and notice less the categories that are stable or falling.
Which produces a perceived rate systematically above the measured one, a finding replicated across countries.
Neither the perception nor the measurement is wrong; they are answering different questions.
What the figure is used for
Uprating benefits and pensions, indexing contracts, setting monetary policy and calculating real wages.
Which is why methodology attracts political attention.
Changes to which index is used for uprating have material distributional consequences and have been contested in several countries.
Which is why the choice of index for uprating is a political decision dressed as a technical one.