Data-Driven Multivariate Assessment of Labour Market Institutions and Performance: A Partial Canonical Correlation Analysis of 25 European Countries

Ali Kemal NURDOĞAN

2026

Full text Open access: Hybrid Metadata (CSL-JSON)
DOI 10.17762/jidmis.v3.4944
Data confidence 70 medium
How this score is calculated

It is built from checkable properties of the record itself: whether the article has a DOI, how many independent sources confirmed its data, whether the required fields are filled in, and how recent the latest confirmation is. Disagreement between sources and a missing identifier lower it. A value of 80 or above means the record was published without further checking; 60 to 79 means it is published at reduced confidence, with a manual check advisable. A record verified by an editor by hand is never pushed below 90 by a recalculation. This scores our data about the article, not the article itself.

Abstract

This study develops a data-driven multivariate assessment of the joint association between a selected configuration of labour-market institutions and multidimensional labour-market performance rather than estimating isolated institutional effects. A balanced panel of 25 European countries for 2013–2019 links an institutional set—employment protection legislation (EPL), the net replacement rate of unemployment benefits (NRR), a statutory national minimum-wage regime (MW), and wage-bargaining coordination (COORD)—to a performance set comprising labour productivity per worker, employment, unemployment, and the NEET rate. Partial canonical correlation analysis is applied after residualising both sets for GDP growth, inflation, education, the working-age population share, gross fixed capital formation, and year effects. Although classical sequential tests identify two significant functions, a 19,999-replication country-block permutation test that preserves within-country dependence supports only the first (Rc1=.691; sequential permutation p=.047), while the second is not confirmed under cluster-aware inference (Rc2=.453; p=.235). The first function is a robust income-protection–productivity axis dominated by NRR (.852) and productivity (.870), with both loadings sign-stable under a 2,000-replication country-block bootstrap. A cross-sectional CCA based on the 25 countries’ residualised period means reproduces the same pattern (Rc1=.748; NRR=.851; productivity=.884). The second function is therefore treated as exploratory. The contribution is a decision-analytic framework for identifying shared institutional–performance profiles while explicitly accounting for clustered comparative data, rather than a causal ranking of individual policy instruments.