Statistics Explained

Unemployment statistics and beyond

Data extracted in May 2022.

Planned article update: May 2023.


In 2021, the EU unemployment rate was 7.0 %, down from 7.2 % in 2020, but above the rate of 6.8 % in 2019.
In 2021, in the EU, the unemployment rate ranged from 2.8 % in Czechia to 14.8 % in Spain.
The unemployment rate in the EU stood at 6.7 % for men and at 7.4 % for women in 2021.

Unemployment rate (total, female, male, youth and senior), 2009-2021
Source: Eurostat

This article focuses on annual statistics on unemployment in the European Union (EU) and its individual Member States, as well as 3 EFTA countries (Iceland, Norway and Switzerland) and one candidate country (Serbia).

A separate article on unemployment statistics presents the unemployment figures on a monthly basis. An article on Labour market slack – annual statistics on unmet needs for employment is also available, focusing on the 3 measures of labour market attachment that supplement the unemployment rate.

Unemployment levels and rates move in a cyclical manner, largely related to the general business cycle. However, other factors such as labour market policies and demographic changes may also influence the short and long-term development of unemployment. In addition to that, the COVID-19 crisis also greatly affected unemployment in 2020 and 2021.

Please note that rates shown in the dynamic tool may differ in some cases from those mentioned in the text, due to continuous revision of the source data. The tool refers to the most recent data (as shown in the Eurostat database under Employment and unemployment (Labour Force Survey), while the text refers to data from May 2022.

Full article

How did the COVID-19 crisis affect unemployment ?

The interactive line chart (see Tool 1) shows how each country's unemployment rate has changed since 2009. By clicking on the icons at the bottom of the tool, you can view the development for specific breakdowns of the unemployment rate: from left to right, you can switch from the total active population to women, men, young people and seniors, respectively.

After falling for 6 consecutive years, the unemployment rate for the total EU population aged 15-74 increased in 2020, the year the COVID-19 crisis began in Europe, and reached 7.2 % of the population in the labour force, up from 6.8 % in 2019.

In 2021, the unemployment rate fell to 7.0 % (lower rate than in 2020 but still above the value from the pre-pandemic 2019). In terms of numbers, unemployed people amounted to 15.0 million in 2021, 0.2 million lower than in 2020 but 0.5 million higher than in 2019.

Tool 1: Unemployment rate (total, female, male, youth and senior), 2009-2021
Source: Eurostat

From 2019 to 2020, the unemployment rate increased in the vast majority of EU Member States (23 out of 27 countries), with the largest increases found in Estonia (+2.4 percentage points, or pp.) and Lithuania (+2.2 pp.) (see Tool 1 or Figure 1).

In addition, increases of between 1 and 2 pp. were recorded in Latvia, Sweden, Spain, Luxembourg, Austria, Romania and Slovakia. By contrast, the unemployment rate fell from 2019 to 2020 in Poland, Greece, France and Italy, with the largest decrease recorded in Italy (-0.6 pp.).

Figure 1: Unemployment rate, 2019 - 2021
Source: Eurostat (une_rt_a)

It is worth noting that the unemployment rate in 2020 increased for the first time in Latvia and Malta after 9 consecutive years of decrease, and in Hungary, Ireland and Denmark after 7 consecutive years of decrease.

From 2020 to 2021, the unemployment rate fell in 16 EU Member States, with the largest cuts recorded in Greece (-2.9 pp.), Luxembourg (-1.5 pp.) and Lithuania (-1.4 pp.). In 9 other Member States, the share of unemployed among the population in the labour force increased over the same period. Belgium stood out with the largest increase (+0.5 pp.). In Finland and Hungary, the unemployment rate remained stable between 2020 and 2021.

Note that Belgium, Sweden, Ireland, Austria, Czechia, Slovakia and Croatia had an increase in the unemployment rate from 2019 to 2020, which continued also from 2020 to 2021.

Among the EU countries which experienced an increase in the unemployment rate from 2019 to 2020, only 4 (Malta, Portugal, the Netherlands and Luxembourg), recorded a lower rate in 2021 than in 2019. In 19 Member States, the unemployment was still higher in 2021 than in 2019, with Sweden, Estonia, Austria, Latvia, Ireland, Slovakia and Croatia having rates in 2021 at least 1 pp. above their 2019 rate.

It might be useful to remember that to be considered unemployed according to the ILO's criteria, a person should not have worked during the reference week, be available to start working within the next two weeks and have actively sought employment at some point during the last four weeks. However, due to the nature of the COVID-19 crisis, which led to people’s confinement in their homes and the lockdown of some businesses, actively searching for work and availability to work were hampered in 2020 and 2021, leading people to be classified as outside the labour force and not as unemployed (given that they did not meet the two criteria of job search and availability). Therefore, the impact of the COVID-19 pandemic on unemployment might not be as strong as the impact on employment.

In 2021, among the EU Member States, the unemployment rate ranged from more than 8 % in Spain (14.8 %), Greece (14.7 %), Italy (9.5 %) and Sweden (8.8 %) to less than 4 % in Germany (3.6 %), Malta (3.5 %), Poland (3.4 %) and Czechia (2.8 %).

Differences between men and women

The male and female unemployment rate for each year between 2009 and 2021 for each country is available through the second and third icons of Tool 1. In 2021, in around half of EU countries (14), women recorded a higher unemployment rate than men. This is also true for the EU as a whole, where the gender unemployment gap has even widened from 0.3 pp. in 2009 to 0.7 pp. in 2021; the unemployment rate in the EU stood at 6.7 % for men and at 7.4 % for women in 2021.

Looking at the particular situation of each country, differences are visible as regards the unemployment gender gap (see Figure 2). Greece is the EU Member State where the unemployment rate of women is the highest compared with that of men: in 2021, unemployed women represented almost one-fifth of the Greek women in the labour force (unemployment rate of 18.9 %), whereas the corresponding share for men was 11.4 %, resulting in a gender gap of 7.5 pp.

Spain and Italy followed, with a female unemployment rate that surpassed the male unemployment rate by 3.6 pp. and 1.9 pp. respectively. By contrast, in Poland, Ireland, France, Austria and Denmark, the unemployment gender gap was 0.2 pp. or less in 2021. The unemployment rate for men exceeded that for women in 12 EU countries in 2021. Among these, Latvia stood out with the widest gap (1.9 pp.).

Figure 2: Unemployment rate by sex, 2021
Source: Eurostat (une_rt_a)

Are young people more exposed to unemployment than seniors ?

Tool 1 shows that the youth unemployment rate (people between 15 and 24 years old) has always been higher than the total unemployment rate (people aged 15-74) for all countries since 2009 (comparing data via the fourth and the first icon of Tool 1).

For people aged 55-74, the opposite can be observed: in most countries, the senior unemployment rate has been lower than the total unemployment rate (comparing data via the fifth and the first icon of Tool 1); only the Netherlands recorded higher unemployment rates for older people over several consecutive years (from 2014 to 2018).

High youth unemployment rates reflect, to some degree, the difficulties faced by young people in finding jobs. However, this does not necessarily mean that the group of unemployed people aged between 15 and 24 is large, as many young people, as opposed to older people, are studying full-time and are therefore neither working nor looking for a job (so they are not part of the labour force which is used as the denominator for calculating the unemployment rate). For this reason, an alternative indicator is calculated for the purpose of analysis: the youth unemployment ratio. This presents the share of unemployed young people among the whole youth population.

As displayed in Figure 3, the EU youth unemployment ratio shows that 6.5 % of all people aged 15-24 were unemployed in 2021. This share amounts to 16.6 % if only the young people in the labour force are taken as a denominator (youth unemployment rate). The youth unemployment ratio in the EU ranged from 2.2 % in Czechia to 13.4 % in Sweden. And the youth unemployment rate ranged from 6.9 % in Germany to 35.5 % in Greece.

Figure 3: Youth unemployment rate and ratio, 2021
Source: Eurostat (une_rt_a)

How does educational level affect unemployment ?

The unemployment rate also varies considerably by level of educational attainment (see Figure 4, results for people aged 25-74). Generally, the higher the educational attainment level, the lower the unemployment rate. This can be observed in all countries. Exceptions were Denmark, with a higher unemployment rate for people with a high educational level than for those with a medium educational level, and Portugal with equal rates for people with a low and medium educational level.

The highest unemployment rates in the EU were found among the labour force with low educational level in Slovakia (38.5 %), Sweden (22.3 %) and Spain (19.6 %), and the lowest among the labour force with high educational level in Bulgaria, Romania, Poland, Hungary, Malta and Czechia (rates below 2 % for all).

Figure 4: Unemployment rate by educational attainment level, 2021
Source: Eurostat (une_educ_a)

What about long-term unemployment ?

Long-term unemployment is one of the main concerns of policy-makers. Figure 5 shows the breakdown of the unemployed by unemployment duration in 2021. At EU level, one-fifth (20.6 %) of those unemployed have been unemployed for more than 2 years (i.e. classified as very long-term unemployment). However, this average hides large differences between EU Member states. In Greece, 39.9 % of the unemployed people had been looking for a job for over 2 years, followed by 36.0 % in Italy, 35.9 % in Slovakia and 27.3 % in Bulgaria. Denmark was at the other end of the ranking with 7.1 % of unemployed people in this group, followed by 7.3 % in Sweden, 7.4 % in Malta and 8.6 % in Poland.

Figure 5: Unemployment by duration, 2021
Source: Eurostat (une_ltu_a)

Source data for tables and graphs

Methods and definitions

Data sources

All figures in this article are based on the European labour force survey (EU-LFS).

Source: The European Union Labour Force Survey (EU-LFS) is the largest European household sample survey providing quarterly and annual results on labour participation of people aged 15 and over as well as on persons outside the labour force. It covers residents in private households. Conscripts in military or community service are not included in the results. The EU-LFS is based on the same target populations and uses the same definitions in all countries, which means that the results are comparable between the countries. The EU-LFS is an important source of information about the situation and trends in the national and EU labour markets. Each quarter around 1.8 million interviews are conducted throughout the participating countries to obtain statistical information for some 100 variables. Due to the diversity of information and the large sample size, the EU-LFS is also an important source for other European statistics like Education statistics or Regional statistics.

Please note that Eurostat provides two sets of indicators linked to the annual unemployment rate, which serve different purposes and which in some cases differ from each other:

1) The LFS main indicators, which contain seasonally adjusted series. They include the labour market headline indicators used e.g. in the Macroeconomic Imbalance Procedure Scoreboard or the European Statistical Recovery Dashboard and are consequently used for monitoring policy. They have only a few breakdowns and normally refer to the age group 20-64.

2) The detailed results, which contain series that are not seasonally adjusted. They have a large number of breakdowns and can therefore be used for more detailed analysis. For France, only one data series is published. This series contains data for metropolitan France until the fourth quarter of 2013, and from 2014 on, also the French overseas departments.

This article presents annual results for indicators from the "LFS main indicators" set.

Reference period: Yearly results are obtained as averages of the four quarters in the year.

Coverage: The results from the EU-LFS currently cover all European Union Member States, the EFTA Member States Iceland, Norway and Switzerland, as well as the candidate countries Montenegro, North Macedonia, Serbia and Turkey. For Cyprus, the survey covers only the areas of Cyprus controlled by the Government of the Republic of Cyprus.

European aggregates: EU and EU-27 refer to the sum of the 27 EU Member States. If data are unavailable for a country, the calculation of the corresponding aggregates takes into account the data for the same country for the most recent period available. Such cases are indicated.

Country notes

In Germany, from the first quarter of 2020 onwards, the Labour Force Survey (LFS) is part of a new system of integrated household surveys. Technical issues and the COVID-19 crisis has had a large impact on data collection processes in 2020, resulting in low response rates and a biased sample. For more information, see here.

In the Netherlands, the 2021 LFS data remains collected using a rolling reference week instead of a fixed reference week, i.e. interviewed persons are asked about the situation of the week before the interview rather than a pre-selected week.

Spain and France have assessed the attachment to the job and included in employment those who, in their reference week, had an unknown duration of absence but expected to return to the same job once health measures allow it.


The concepts and definitions used in the EU-LFS follow the guidelines of the International Labour Organisation (ILO).


Eurostat publishes unemployment statistics based on a definition of unemployment provided by the International Labour Organisation (ILO) for which there are three criteria, namely:

  • being without work;
  • actively seeking work;
  • and being available for work.

The ILO definition of the unemployment rate is the most widely used labour market indicator because of its international comparability and relatively timely availability. Besides the unemployment rate, indicators such as employment and job vacancies also give useful insights into labour market developments.

Monthly unemployment figures are published by Eurostat as rates (as a percentage of the labour force) or levels (in thousands), by sex and for two age groups (persons aged 15 to 24, and those aged 25 to 74). The figures are available as unadjusted, seasonally adjusted and as a trend series. The time series for data for the EU and the euro area (EA-19) aggregates start in 2000; the starting point for individual EU Member States varies.

Quarterly and annual unemployment figures from the EU-LFS are published with more detailed breakdowns (for example, a wider range of age groups, by nationality, or by educational attainment); there are also figures available on long-term unemployment (unemployed for more than 12 months) and very long-term unemployment (unemployed for more than 24 months).

Unemployment rates are also presented according to the educational attainment level of the population, i.e. the highest level of education successfully completed. The different levels of educational attainment are defined by the United Nations International Standard Classification of Education (ISCED 2011). Low level of educational attainment refers to ISCED levels 0-2 (less than primary, primary and lower secondary education), medium level refers to ISCD levels 3 and 4 (upper secondary and post-secondary non-tertiary education) and high level refers to ISCED levels 5-8 (tertiary education).

Underemployment and potential additional labour force

Many persons only partially fulfil the three unemployment criteria above and are therefore not considered as unemployed. In order to provide information on people who are not unemployed, Eurostat also publishes indicators on the following groups.

  • Underemployed part-time workers: persons working part-time who wish to work additional hours and are available to do so.
  • The potential additional labour force: jobless persons who want to work and are either available to work or are searching for work but not both at the same time. This group includes, among others, discouraged job seekers and persons prevented from job seeking due to personal or family circumstances. This group is split into two groups: persons seeking work but not immediately available and persons available to work but not seeking work.

Time series

Regulation (EU) 2019/1700 came into force on 1 January 2021 and induced a break in the EU-LFS time series for several EU Member States. In order to monitor the evolution of employment and unemployment despite of the break in the time series, Member States assessed the impact of the break in their country and computed impact factors or break corrected data for a set of indicators. Break corrected data are published on the Eurostat website for the LFS main indicators.

Additional methodological information

More information on the EU-LFS can be found via the online publication EU Labour Force Survey, which includes eight articles on the technical and methodological aspects of the survey. The EU-LFS methodology in force from the 2021 data collection onwards is described in methodology from 2021 onwards. Detailed information on coding lists, explanatory notes and classifications used over time can be found under documentation.


The unemployment rate is an important indicator with both social and economic dimensions. Rising unemployment results in a loss of income for individuals, increased pressure with respect to government spending on social benefits and a reduction in tax revenue. From an economic perspective, unemployment may be viewed as unused labour capacity.

Time series for unemployment are used by the European Commission, other public institutions, and the media as an economic indicator, while banks may use the data for business cycle analysis. Finally, there is interest among the general public for information concerning unemployment.

The unemployment rate is considered to be a lagging indicator. When there is an economic downturn, it usually takes several months before the unemployment rate begins to rise. Once the economy starts to pick up again, employers usually remain cautious about hiring new workers and it may take several months before unemployment rates start to fall.

Male, youth and long-term unemployment appear to be more susceptible to cyclical economic changes than overall unemployment. Indeed, social policymakers often face the challenge of remedying these situations by designing ways to increase employment opportunities for various groups of society, those working in particular economic activities, or those living in specific regions.

Globalisation and technological developments appear to have an ever-increasing effect on daily life, and the demand for different types of labour and skills changes, sometimes at a rapid pace. While enterprises try to improve their productivity and become more competitive and innovative, they may well seek to pass on risk to the labour force through greater flexibility — both in relation to those already in employment, as well as those searching for a new job. Within the context of the European employment strategy (EES), there are a number of measures that are designed to help encourage people to remain in work or find a new job, including: the promotion of a life-cycle approach to work, encouraging lifelong learning, improving support to those seeking a job, as well as ensuring equal opportunities.

Direct access to

Other articles
Dedicated section

LFS main indicators (t_lfsi)
Unemployment - LFS adjusted series (t_une)
LFS series - detailed annual survey results (t_lfsa)
Unemployment rates of the population aged 25-64 by educational attainment level (tps00066)
LFS main indicators (lfsi)
Unemployment - LFS adjusted series (une)
LFS series - detailed quarterly survey results (from 1998 onwards) (lfsq)
Total unemployment - LFS series (lfsq_unemp)
LFS series - Detailed annual survey results (lfsa)
Total unemployment - LFS series (lfsa_unemp)