A country-by-country unemployment map becomes misleading if the data year is hidden. This article uses the International Labour Organization’s ILOSTAT national-source unemployment rate series for total sex and ages 15+, keeping the latest annual observation available for each country or area within the supplied 2018–2026 extraction window. The resulting table contains 177 country or area rows, but their latest reference years range from 2018 to 2025.
That means this is not a single-year 2025 ranking. The first map answers a different question: what is the latest national-source unemployment observation available for each place in this ILOSTAT series? A second map shows the reference year behind each value. For a synchronized comparison, the article then isolates the 81 observations that are actually dated 2025 and calculates their median, distribution and selected-country values separately.
Table of Contents
Unemployment rate uses the labour force as its denominator
The ILO defines the unemployment rate as unemployed persons expressed as a percentage of the labour force. The labour force is the sum of employed and unemployed persons. In the standard framework, unemployment concerns people who are not in employment, are available for work and are seeking employment. The ILOSTAT labour force statistics methodology provides the formal concepts and also explains why national-source statistics are not always perfectly comparable across countries.
This denominator matters. An unemployment rate of 5% does not mean that 5% of everyone aged 15 and older is jobless. People outside the labour force—such as some students, retirees, caregivers or discouraged potential workers—are not simply counted as unemployed. Unemployment should therefore be read alongside labour-force participation, employment-to-population ratios, underemployment, youth NEET rates, wages and other measures when the goal is to assess the broader labour market.
Only 81 of the 177 latest observations are from 2025
The reference-year distribution is uneven. Eighty-one rows, or about 45.8%, are dated 2025. Another 31 are from 2024, so 112 of 177 observations (63.3%) are from 2024 or 2025. The remaining 65 rows (36.7%) are latest values from 2018 through 2023: four from 2018, six from 2019, nine from 2020, nine from 2021, 22 from 2022 and 15 from 2023.
That timing difference changes what the latest-value map can support. South Africa’s latest retained value is 32.279% from 2024, while China’s is 4.569% from 2022. Korea is 2.789% in 2025 and the United States is 4.282% in 2025. Putting all four into a precise current ranking would mix labour-market differences with reference-year differences. The latest-value map is useful for locating the most recent national observation, but a same-year comparison must filter the dates first.
The synchronized 2025 subset has a median of 5.17%
Restricting the data to the 81 rows dated 2025 produces a much cleaner time comparison. The median unemployment rate is 5.166% and the simple, unweighted mean is 6.046%. Each country or area receives one equal observation, so the mean is not the unemployment rate of the world labour force and should not be interpreted as a population- or labour-force-weighted global statistic.
The distribution is concentrated in the middle. Four observations are below 2%, 36 are from 2% to under 5%, 35 are from 5% to under 10%, four are from 10% to under 20%, and two are at 20% or more. In other words, 71 of the 81 synchronized observations fall between 2% and 10%. The two very high values above 20% help pull the mean above the median.
Djibouti and Palestine are above 20% in the 2025 national observations
Djibouti has the highest 2025 value in this retained national-source subset at 38.782%, followed by Palestine at 28.710%. North Macedonia is 11.532%, Bosnia and Herzegovina 10.732%, Spain 10.504% and Angola 10.391%. These are the only six observations at or above 10% among the 81 rows dated 2025.
| Country or area | 2025 unemployment rate |
|---|---|
| Djibouti | 38.782% |
| Palestine | 28.710% |
| North Macedonia | 11.532% |
| Bosnia and Herzegovina | 10.732% |
| Spain | 10.504% |
| Angola | 10.391% |
| Finland | 9.609% |
| Chile | 8.894% |
| Greece | 8.838% |
| Sweden | 8.740% |
This ranking deliberately excludes high latest values that belong to other years. Eswatini is 34.265% in 2023 and South Africa is 32.279% in 2024, so placing them in a 2025 table would hide a real timing difference. The cost of synchronizing the year is a much smaller sample: the latest-value table has 177 rows, but only 81 can be compared on a 2025 basis in this particular national-source extract.
Korea, the United States, Japan and other major economies in the same 2025 subset
Among major economies with a 2025 observation in this dataset, Korea is 2.789%, Japan 2.500%, Mexico 2.542%, Germany 3.824%, Australia 4.170%, the United States 4.282%, India 4.506% and the United Kingdom 4.896%. Brazil is 5.832%, Italy 6.133%, Canada 6.819%, France 7.700% and Spain 10.504%.
| Country | 2025 unemployment rate |
|---|---|
| Korea | 2.789% |
| Japan | 2.500% |
| United States | 4.282% |
| India | 4.506% |
| Germany | 3.824% |
| United Kingdom | 4.896% |
| Brazil | 5.832% |
| Canada | 6.819% |
| France | 7.700% |
| Spain | 10.504% |
China illustrates the date issue again: its latest retained national-source observation is 4.569% from 2022, so it is not inserted into the 2025 table. A latest-national-statistics question and a synchronized-2025 question are different tasks. Filling a missing 2025 national observation with an assumption or silently substituting another series would blur that distinction.
National-source observations and ILO modelled estimates should not be merged
ILOSTAT publishes statistics from national sources and also a separate set of ILO modelled estimates. The ILO labour force statistics documentation explicitly distinguishes the two. National-source observations preserve the reporting structure and timing of country statistical systems, while modelled estimates are designed to provide broader and more consistent international coverage using nationally reported information together with imputation and harmonization where needed.
This distinction is important because Green Map already has a separate 2025 unemployment article based on ILO modelled estimates. The two products answer different questions. National-source data are valuable when the user wants the latest observed figure reported through country statistical systems and is willing to inspect the reference year. Modelled data are often easier for broad same-year comparisons. Neither should be treated as a corrected version of the other, and values from the two series should not be stitched into one ranking without a methodological investigation.
A low unemployment rate is not a complete score of labour-market health
The lowest 2025 values in this retained subset are Thailand at 0.604%, Moldova at 1.279%, Vanuatu at 1.412% and the United Arab Emirates at 1.744%. These figures describe one specific relationship: unemployed persons as a share of the labour force. They do not directly measure wages, hours, job security, informality, social protection, productivity or living standards.
The same caution applies to high unemployment. This map does not identify why Djibouti, Palestine, Spain or another country has its observed value. Economic conditions, demographic structure, industry mix, migration, labour-market institutions, survey design and statistical definitions can all matter. Geographic clustering can suggest where to investigate, but it cannot establish causation by itself.
Data source and mapping method
The statistical source is the International Labour Organization’s ILOSTAT annual Unemployment rate by sex and age (%) dataset. The extraction keeps total sex (SEX_T), ages 15+ (AGE_YTHADULT_YGE15), and one latest observation per country within the supplied 2018–2026 window. The source query is available through the ILOSTAT SDMX endpoint, while the broader catalogue is available at ILOSTAT Data.
The latest-value map uses all 177 retained country or area rows. The reference-year map visualizes the year attached to those same observations. The 2025 distribution, summary statistics and ranking tables use only the 81 rows whose year equals 2025. Missing countries are not converted to zero, and older observations are never relabeled as 2025. ISO-3 codes are joined to a low-resolution world boundary layer; small islands and separately reported areas may have a statistical row without a visible polygon at this scale.
Frequently Asked Questions
Are all unemployment rates on this map from 2025?
No. The 177 latest observations span 2018–2025. Only 81 are dated 2025, and the 2025 ranking and distribution use only those 81 rows.
How is this different from an ILO modelled unemployment-rate map?
This article uses the ILOSTAT national-source observation series, whose latest year varies by country. ILO modelled estimates are a separate series designed for broader harmonized international coverage.
What is the median unemployment rate in the synchronized 2025 subset?
The median across the 81 observations dated 2025 is 5.166%, and the unweighted mean is 6.046%. The mean is not a labour-force-weighted global unemployment rate.
Does a low unemployment rate automatically mean a strong labour market?
No. Unemployment measures joblessness within the labour force. It does not directly capture participation, wages, hours, informality, job security, underemployment or living standards.
Related Articles
These related Green Map articles separate national observations, modelled estimates and employment ratios so the measures are not treated as interchangeable.
- Global Unemployment Rate Map – Country Patterns in 2025
- Employment-to-Population Ratio Map – Latest National Estimates
- Global Employment-to-Population Ratio Map – Ages 15+ in 2025
Green Map creates custom-edited map images using open geographic data sources such as geoBoundaries, Natural Earth, OpenStreetMap, and government open data.
These maps are edited visual materials, not raw data files, and are provided for education, documents, presentations, and graphic reference.
