Abstract
Supply chain networks (SCN) form the structural backbone of any society. They constitute the societal metabolism that literally produces everything for everybody by coordinating practically every single person on the planet. SCNs are by no means static but undergo permanent change through the entry and exit of firms and the rearrangement of supply relations. Here, we use a unique dataset to explore the temporal evolution of firms and their supplier-buyer relations of a national SCN. Monthly reported value added tax data from Hungary from 2014 to 2022 allows us to reconstruct the entire economy with 711,248 companies and 38,644,400 connections, covering practically every restructuring event of an entire economy at firm-level resolution. We find that per year about 25% of firms exit the SCN while 28% new ones enter. On average, 55% of all supply-links present in 1 year will not be present in the next. We report the half-life time of supply-links to be 13 months. New links attach super-preferentially to firms with a probability, p(i) ∝ ki1.08 , with ki firm i's number of supply connections. We calibrate a simple statistical network generation model that reproduces the stylized characteristics of the dominant Hungarian SCN. The model not only reproduces local network features such as in- and out-degree distributions, assortativity, and clustering structure but also captures realistic systemic risk profiles. We discuss the present model in how rewiring dynamics of the economy is essential for quantifying its resilience and to estimate shock propagation.