Mascaretti, Andrea, Dell'Agostino, Laura, Arena, Marika, Flori, Andrea, Menafoglio, Alessandra, and Vantini, Simone
The identification of the channels through which a given shock spreads to the rest of the economy, determining its final impact, is essential to formulate effective policy interventions. Input–output tables (IOTs) are widely used to detect the network of intersectoral relations of a country – i.e., its sectoral technological structure or domestic supply chains – and the role of different sectors in the propagation of a shock. However, the heterogeneity that characterize the technological structures of different countries is inevitably a source of complexity for the development of supranational and timely coordinated policies because it requires to analyze and interpret a large amount of information. This paper proposes a unique problem setting that aims to deal with this complexity by facilitating the analysis and visualization of similarities and differences among the technological structures of countries, relying on the identification of a small number of archetypes and showing how their interpretation could be exploited to support the definition of coordinated policy interventions. Specifically, non-negative matrix factorization is used to extract the archetypal matrices of the technological structures of the 28 European countries from IOTs, revealing dense intersectoral relationships and a low degree of heterogeneity between them. Then, random walk indicators are applied to study shock propagation within these archetypes, uncovering sectoral centralities. Finally, COVID-19 lockdown restrictions are analyzed to exemplify the use of the proposed approach for coordinated policy action. • Archetypal matrices of technical coefficients synthesize multiple national IOTs. • Archetypes reduce multidimensionality related to the comparison of national IOTs. • Archetypes reveal low heterogeneity between the technological structures of EU IOTs. • Archetypes can be applied to inform coordinated and timely policies. [ABSTRACT FROM AUTHOR]