Scaling of global input–output networks

Abstract

Examining scaling patterns of networks can help understand how structural features relate to the behavior of the networks. Input–output networks consist of industries as nodes and inter-industrial exchanges of products as links. Previous studies consider limited measures for node strengths and link weights, and also ignore the impact of dataset choice. We consider a comprehensive set of indicators in this study that are important in economic analysis, and also examine the impact of dataset choice, by studying input–output networks in individual countries and the entire world. Results show that Burr, Log-Logistic, Log-normal, and Weibull distributions can better describe scaling patterns of global input–output networks. We also find that dataset choice has limited impacts on the observed scaling patterns. Our findings can help examine the quality of economic statistics, estimate missing data in economic statistics, and identify key nodes and links in input–output networks to support economic policymaking.