Unpacking Structural Polarisation: Economic Complexity and Productivity across Italian Territories

This paper investigates the structural foundations of regional productivity divergence in Italy through the lens of economic complexity. Leveraging a newly constructed Economic Complexity Index (ECI) at the NUTS-3 level, we examine how the sophistication and diversity of local productive structures shape long-run productivity trajectories of Italian provinces over the period 2000–2021. Empirical approach combines panel data models with instrumental variable (IV-GMM) techniques, spatial econometrics, and simultaneous equation systems (3SLS) to capture the direct, spatial, and bidirectional relationships between complexity and productivity. The findings reveal that economic complexity is a robust and consistent predictor of regional labour productivity. This association is particularly strong in Northern provinces, where institutional density and in- novation ecosystems amplify the returns to complexity, and where spatial spillovers from neighbouring territories enhance local outcomes. In contrast, Southern regions experience lower returns and limited externalities, reflecting persistent development traps. Crucially, I provide the first integrated empirical evidence of a cumulative, self-reinforcing loop between complexity and productivity: more complex regions become more productive, and more productive regions are better equipped to diversify into complex activities.