How Unemployment Figures Are Compiled

Unemployment figures are among the most closely watched economic releases. What they measure is narrower than the phrase suggests, and the definitions do most of the work.

The definition is specific and restrictive

Standard international definitions count someone as unemployed only if they are without work, available to start and have actively sought work in a recent period.

Someone who has stopped looking is not counted as unemployed. They are classified as outside the labour force entirely.

This means the unemployment rate can fall because people found work or because people stopped searching, and the headline figure does not distinguish between the two.

The rate has a shifting denominator

The rate is the number unemployed divided by the labour force, which is the employed plus the unemployed. It excludes everyone not participating.

Because the denominator changes as people enter and leave the labour force, the rate can move without any change in the number of jobs.

This is why the participation rate is published alongside it. The two together describe what the rate alone cannot.

Surveys and administrative data differ

Many countries produce both a survey-based measure and a count derived from benefit claims or payroll records. The two are constructed differently and rarely match.

Survey measures capture people regardless of eligibility for support, while administrative counts depend on the rules of the system generating them.

Changes to benefit rules can therefore move an administrative measure without any change in employment, which is one reason survey measures are preferred for comparison.

Sampling introduces uncertainty

Survey measures come from a sample rather than a census, so each estimate carries a margin of error that is published alongside it.

Month-to-month movements smaller than that margin cannot be distinguished from sampling variation, though they are frequently reported as meaningful changes.

Revisions occur as more responses arrive and as seasonal adjustments are updated, so initial figures should be treated as provisional.

The measure omits the quality of work

Someone working a small number of hours is counted as employed, so the headline rate says nothing about hours, pay or whether the work matches the person's skills.

Underemployment measures exist to address this, capturing people working fewer hours than they would like, and they often move differently from the headline rate.

Definitions, survey methods and published series differ between countries and are revised over time, so comparisons across jurisdictions require checking that the measures align.