Europe’s unmet need for work is twice its unemployment.

The number built to count the people the rate misses

For two editions this desk has printed a ratio and then spent the piece explaining who is missing from it. Britain: 2.5 unemployed people per vacancy. America: 1.0 unemployed person per job opening. Both times the list of the absent was the same - applications, applicants who already have a job, and everyone outside the labour force. That third group is the largest, and in most countries nobody puts a number on it.

Eurostat does. Its measure is called labour market slack, and the definition is worth quoting because the word suggests something vaguer than the statistic is: slack is all persons who have an unmet need for employment, including unemployed people. In the first quarter of 2026 it stood at 23.7 million people across the European Union, or 10.9% of the extended labour force.

In the same quarter, 12.0 million of those people were unemployed. Subtract, and 11.7 million people in the EU had an unmet need for work while not being counted as unemployed - a figure that is arithmetic on two published numbers rather than a third published number, and is labelled as such below. Slack is not a rival to the unemployment rate. It is the unemployment rate plus the three groups it was never built to see.

Why these four numbers cannot be added

Here is the trap, and it is the reason this edition exists rather than a summary of one press release. The EU unemployment rate is 5.7%, and the components of slack are published as percentages too - 2.3%, 2.3%, 0.7%. Four percentages on one page invite one operation, and that operation is wrong. The unemployment rate is measured against the labour force. The slack components are measured against the extended labour force, which is larger, because it also contains the people who want work but fall outside the unemployment definition.

The arithmetic proves it rather than merely asserting it. If 23.7 million people are 10.9% of the extended labour force, that base is about 217 million. Twelve million people measured against 217 million is about 5.5% - not the published 5.7%. The gap is not an error in either figure; it is two different denominators doing their jobs. A reader who adds 5.7 to the rest gets a number that corresponds to no population at all.

The published total behaves the same way, and Eurostat prints it honestly enough that the seam shows: the four components as published sum to 11.0%, while the total slack figure is 10.9%. Nothing is wrong. Each figure is rounded independently before publication, so a sum of rounded parts need not equal a rounded total. It is worth naming because a desk that quietly corrects the 11.0 to 10.9, or the 10.9 to 11.0, has smoothed away the only visible evidence that these are separately computed statistics.

Three ways to want work and not be unemployed

The three groups are not a residual category and they are not variations on one another. Underemployed part-time workers - 2.3% of the extended labour force - have jobs. They are working fewer hours than they want to work, which is a different problem from having no job at all, and it is invisible to every headline rate because the person is employed.

People available to work but not seeking - also 2.3% - are the group most often misread. They are not people who have chosen not to work. They want employment and could start; they are not currently searching, which is what the unemployment definition requires. The reasons are ordinary: a search abandoned after a long run of silence, caring responsibilities that make a search pointless until something changes, a region where the person already knows what is posted. Discouragement is a labour market outcome, not a personality.

People seeking work but not available - 0.7% - are the mirror image: the search is active, the start is not yet possible. A notice period, a course to finish, a care arrangement that begins next month. They are doing the thing the unemployment definition asks for and failing the other half of it.

Put together, these three are why "the unemployment rate is low" and "I cannot find enough work" are both true statements about the same economy, spoken by different people. 11.7 million of them, on the arithmetic above.

What moved, and what it means for hiring

On the quarter, slack fell slightly - from 11.0% of the extended labour force in the fourth quarter of 2025 to 10.9% in the first quarter of 2026 - while the employment rate for people aged 20 to 64 rose from 76.2% to 76.3%. Both moves are one tenth of a percentage point. Read them as a market holding its position, not as a recovery.

The aggregate hides the spread, which is where the useful part is. Slack increased in 11 countries, was stable in 6, and decreased in 10. The largest increase was Italy, up 0.5 points, with Belgium, Cyprus, Lithuania, Slovakia and Sweden each up 0.4. The largest decreases were Latvia, down 0.8, and Ireland, down 0.7, with Slovenia and Finland each down 0.3. A continental average of minus 0.1 is the arithmetic residue of movements an order of magnitude larger in both directions.

For anyone hiring in Europe, the operative fact is not the direction but the size of the pool. The people in the three non-unemployed groups are, by construction, not visible in the channels built around unemployment - they are not on the register, not in the flow of active applications, and in one case already employed elsewhere. A hiring process that only ever meets active searchers is meeting roughly half of the people who want more work than they currently have. For anyone looking: if you are in one of those three groups, the statistics have counted you even when the headline rate did not, and the reason you are hard to reach is structural rather than personal.