More than just transition: Uncovering heterogeneous socio-economic outcomes of climate policies

Achieving sustainability in sociotechnical systems requires transition processes supported by far-reaching policy action. By their very nature, these processes generate diverse and uneven consequences across territories, affecting key socio-economic dimensions such as unemployment, GDP, average disposable income, and the gender employment gap. Moreover, the web of linkages on which policy actions work eludes traditional analytical frameworks. Taking the European Union as a case study of a just transition process, we propose an empirical setting based on a machine learning technique with three main objectives. First, we develop a model that better captures the complexity of a sustainable and just transition process. Second, we compare results with standard quantitative assessment models, and we test whether the resulting insights derived from the implementation of ML-based approach align with economic theory. Third, we explore the potential of a controlled ML framework in supporting context-specific policymaking for a case study of a just and sustainable transition process. • A sustainable human system requires transformative shifts. • Just transition processes need complex and non-linear analysis. • A machine learning empirical setting is implemented for dynamic analysis. • Context-specific policymaking is preferred for reducing uneven distribution.