This section presents an overview of major data breaks in the data underlying the OECD Skills for Jobs Database (OECD, 2026[153]). The section first presents data breaks for the EU Labour Force Survey and the EU statistics on income and living conditions, followed by a breakdown by country.
Mapping Skill Shortages in OECD Countries
New Evidence from the Updated Skills for Jobs Database
Note: OJP stands for online job postings. Data from the European Labour Force Survey is taken from Eurostat (2006-2024[175]); and data from the EU statistics on income and living conditions is taken from Eurostat (2004-2025[176]). *underqualification is excluded for some occupations when the modal education level is the lowest level of education.
Source: OECD elaboration.
EU-Labour Force Survey
Change in measure of labour force status and methodology in 2021. The size and direction of the breaks vary by country. Based on visual inspection of the time‑series, the following countries and occupations are adjusted backwards using the country-level growth rate in the break year and the growth rates of the pre‑break series for earlier years:
Belgium: Occupation 32, 41 and 44
Denmark: Occupation 23
France: Occupation 24 and 31
Spain: Occupation 51 and 52
More information about the breaks is reported in the national documentation on Eurostat’s website, see:
Germany integrated the EU-LFS into a new microcensus as a subsample in 2020. This, combined with the impact of the COVID‑19 pandemic, led to significant breaks in the time series. All occupations were backwards adjusted from 2020. Read more about the break here:
https://ec.europa.eu/eurostat/documents/10186/6246844/LFS-2020-Note-on-German-data.pdf.
Education variables were affected by changes in the questionnaire and survey process in France in 2013. Underqualification was backwards adjusted.
In 2013, the Netherlands switched to online interviews which led to changes in the elicitation of occupation information. All occupations were adjusted backwards in 2013.
In 2012, there was a change in the national occupation classification in Germany. This led to visible breaks in the time series for occupations 41 and 43, which were adjusted backwards.
In 2011, many EU-labour force surveys underwent significant changes leading to breaks in the data series. All data series were adjusted backwards.
Improvements in the classification of occupations in Czechia in 2010 led to visible data breaks for underqualification of occupations in ISCO major group 2. These were backwards adjusted.
EU Statistics on income and living conditions
In 2024, there was a methodological break for the Slovak Republic due to using administrative data to complement the survey. This led to visible breaks in median hourly wage from 2023 and 2024. For this reason, 2024 data is excluded for the Slovak Republic.
EU-SILC switched from ISCO‑88 to ISCO‑08 in 2011, which led to breaks in the median hourly wage in 2010 as individuals are asked about their income in the year before the survey. The wage was backwards adjusted for all countries.
Argentina (Permanent household survey)
Due to errors in sampling framework, population weights from 2016 to 2018 were adjusted by coefficients provided by the Instituto Nacional de Estadística y Censos (indec). For more information see here: https://www.indec.gob.ar/ftp/cuadros/menusuperior/eph/EPH_nota_metodologica_1_trim_2019.pdf.
Chile (Encuesta Nacional de Caracterización Socio‑económica – CASEN)
There were significant changes to the CASEN survey in 2022. Size of break is not clear as data for 2024 was not available at time of analysis.
Korea (Korean Labor & Income Panel Study – KLIPS)
Included additional households in the survey in 2018 to improve representativeness. Based on visual inspection, the data series was not adjusted.
Mexico (The National Survey of Occupation and Employment – ENOE)
During the COVID‑19 pandemic, field surveys changed to telephone surveys from April to June 2020, whereafter the survey was replaced by the National Survey of Occupation and Employment, New Edition until 2022. From 2023, data collection reverted to the National Survey of Occupation and Employment. During the COVID‑19 pandemic, the sampling frequency increased with some individuals sampled once per month rather than every quarter. For these individuals, the last observation of each quarter was kept. Beyond this there were no visible breaks as a result of the change in survey vehicle.
In 2012, a new occupation classification was adopted, SINCO 2011. Based on visible inspection, underqualification was backwards adjusted in 2012.
Note: The most recent year available is 2024 for most countries. Exceptions are Chile and Norway (2022); Korea, the Slovak Republic, Switzerland and Türkiye (2023); and Australia and New Zealand (2025). The OECD average is the mean of the industry-level indicator in the latest year for OECD Member countries included in the Skills for Jobs Database, and the EU average is the mean for EU countries included in the database. The categorisation is based on the percentiles for the OECD average from 2019 to 2025, with large surplus being below the 20th percentile, small surplus being between the 20th and 40th percentile, in balance between the 40th and 60th percentile, small shortage between the 60th and 80th percentile and large shortage above the 80th percentile.
Source: OECD (2026[153]), OECD Skills for Jobs Database 2026, www.oecd.org/en/data/dashboards/skills-for-jobs.html.
Note: The most recent year available is 2024 for most countries. Exceptions are Chile and Norway (2022); Korea, the Slovak Republic, Switzerland and Türkiye (2023); and Australia and New Zealand (2025). The OECD average is the mean of the skill-level indicator in the latest available year for OECD Member countries included in the Skills for Jobs Database, and the EU average is the mean for EU countries included in the database. The categorisation is based on the percentiles for the OECD average from 2019 to 2025, with in balance being shortages below the 33 percentile, small shortage being shortages between the 33 and 66 percentile and large shortages being shortages above the 66th percentile.
Source: OECD (2026[153]), OECD Skills for Jobs Database 2026, www.oecd.org/en/data/dashboards/skills-for-jobs.html.
Note: The most recent year available is 2024 for most countries. Exceptions are Chile and Norway (2022); Korea, the Slovak Republic, Switzerland and Türkiye (2023); and Australia and New Zealand (2025). The OECD average is the mean of the skill-level indicator in the latest available year for OECD Member countries included in the Skills for Jobs Database, and the EU average is the mean for EU countries included in the database. The categorisation is based on the percentiles for the OECD average from 2019 to 2025, with in balance being shortages below the 33 percentile, small shortage being shortages between the 33 and 66 percentile and large shortages being shortages above the 66th percentile.
Source: OECD (2026[153]), OECD Skills for Jobs Database 2026, www.oecd.org/en/data/dashboards/skills-for-jobs.html.
Note: The most recent year available is 2024 for most countries. Exceptions are Chile and Norway (2022); Korea, the Slovak Republic, Switzerland and Türkiye (2023); and Australia and New Zealand (2025). The OECD average is the mean of the skill-level indicator in the latest available year for OECD Member countries included in the Skills for Jobs Database, and the EU average is the mean for EU countries included in the database. The categorisation is based on the percentiles for the OECD average from 2019 to 2025, with in balance being shortages below the 33 percentile, small shortage being shortages between the 33 and 66 percentile and large shortages being shortages above the 66th percentile.
Source: OECD (2026[153]), OECD Skills for Jobs Database 2026, www.oecd.org/en/data/dashboards/skills-for-jobs.html.
Note: The most recent year available is 2024 for most countries. Exceptions are Chile and Norway (2022); Korea, the Slovak Republic, Switzerland and Türkiye (2023); and Australia and New Zealand (2025). The OECD average is the mean of the skill-level indicator in the latest available year for OECD Member countries included in the Skills for Jobs Database, and the EU average is the mean for EU countries included in the database. The categorisation is based on the percentiles for the OECD average from 2019 to 2025, with in balance being shortages below the 33 percentile, small shortage being shortages between the 33 and 66 percentile and large shortages being shortages above the 66th percentile.
Source: OECD (2026[153]), OECD Skills for Jobs Database 2026, www.oecd.org/en/data/dashboards/skills-for-jobs.html.