Skill mismatch EU-LFS dataset

The dataset provides time series of skill mismatch data and underutilisation rates in all EU-LFS countries for years 2014-202. The dataset is derived for the active labour force and focuses on indicators of skill mismatch measured through the underutilisation of workers' human capital, following the conceptual framework proposed by Marois et al. (2026). Definitions and classifications of the variables are consistent with Marois et al. (2026. Definitions, file descriptor, varables and missigness are further detailed in the provided technical documentation. The dataset consists of five aggregate level tables with point estimates of the underutilisation rates for population subgroups of the active labour force for years 2014-2023. Data source: EU-LFS 2014-2023, all countries available Coverage: EU-27 Member States, Switzerland, Norway and United Kingdom (2014-2019) Years: 2014-2023 Technical documentation: L4S_deliverable_2.2_supporting_documentation. pdf Output flies (abulations for total active labour force across all ages) underutil_by_edu.csv Variables: YEAR, COUNTRY, SEX, EDU, underutil_w, active_w, underutil_rate, underutil_n, active_n underutil_by_ageg.csv Variables: YEAR, COUNTRY, SEX, EDU, ageg, underutil_w, active_w, underutil_rate, underutil_n, active_n underutil_by_pob.csv Variables: YEAR, COUNTRY, SEX, POB, underutil_w, active_w, underutil_rate, underutil_n, active_n underutil_by_pob_edu.csv Variables: YEAR, COUNTRY, EDU, POB, underutil_w, active_w, underutil_rate, underutil_n, active_n The following tabulation is restricted to working age (20-64): underutil_workage_by_country_year.csv Variables: Variables: YEAR, COUNTRY, underutil_w, active_w, underutil_rate, underutil_n, active_n Replication code: LFS_underutilisation_code.do LFS_underutilisation_code.txt