Local effects of energy decoupling in EU-27 on labor: a spatial investigation

In order to achieve its target of 55% GHG emission reduction by 2030, the EU needs to bridge a 24-percentage point between 2022 and 2030. This final push will concentrate employment losses in specific regions, and not spread them uniformly across space. Using spatial econometric decomposition on 1,131 NUTS-3 regions (2000-2022), we quantify regional employment exposure to this remaining decarbonization burden. We decompose the employment-CO2 elasticity by isolating two distinct mechanisms: labor-output elasticity and output-CO2 elasticity, capturing carbon lock-in. This decomposition reveals which channel drives vulnerability, and enables targeted policy response. Estimating via Spatial Durbin Models with Bartik shift-share instrumentation for endogeneity, we find an overall employment elasticity to CO2 of φ = 0.169 (SE = 0.0025). The central finding breaks down conventional wisdom: inland regions face 49% higher employment burden from decarbonization, but not because labor markets are rigid. Labor-output elasticities are statistically identical across locations, and the vulnerability gap is entirely driven by sectoral carbon intensity. For the remaining 24 pp emissions cut, inland regions face roughly 7% average employment losses versus 5% for coastal regions. We estimate employment impacts under two policy scenarios: uniform (15.5M jobs, 7.3%), differentiated by CO2/worker (14.5M jobs), and differentiated by raw CO2 (19.0M jobs). These estimates provide quantitative foundation for allocating €100+ billion in Just Transition Mechanism support, targeting regions facing more than 2% employment loss.