Carbon tax policy adoption and energy transition in the European Union: A quasi-experimental analysis of environmental and socio-economic dividends

Abstract

Carbon taxes are increasingly promoted as market-based instruments for advancing the energy transition, yet evidence on their joint environmental and socio-economic effects remains limited. This study provides a quasi-experimental evaluation of the double dividend hypothesis by assessing the effects of carbon tax adoption on energy intensity, renewable energy consumption, and the Human Development Index across 24 European Union member states over 1990-2019. Exploiting staggered adoption timing, the analysis applies the Callaway and Sant’Anna (2021) heterogeneity-robust difference-in-differences estimator and the Wooldridge (2025) extended two-way fixed effects framework, providing mutual robustness across two independent estimation strategies, dual control group specifications, and doubly robust inverse probability weighting. The results indicate that carbon tax adoption reduces energy intensity by 0.278 units on average (ATT = −0.278, p < 0.01), increases renewable energy consumption (ATT = 0.070, p < 0.01), and improves the Human Development Index (ATT = 0.021, p < 0.01), with effects that strengthen cumulatively for countries with longer policy exposure. Cohort-specific estimates reveal substantial heterogeneity: early adopters with over two decades of implementation exhibit welfare gains up to three times larger than late adopters, indicating that the socio-economic dividend materialises only under sustained exposure. These findings are corroborated by Driscoll-Kraay robustness checks, randomisation inference (p ≤ 0.014 for all three outcomes), and leave-one-out sensitivity analysis confirming that no single treated country drives the results. The study provides integrated quasi-experimental evidence that carbon taxation is directionally compatible with both environmental improvement and broader welfare gains within the EU context, supporting the double dividend hypothesis conditional on sustained policy exposure and effective institutional implementation.