The structural interdependencies of industries: An agent-based model

We develop an agent-based stock-flow consistent macroeconomic model with multiple industries and supply chains to analyse the propagation of sectoral shocks. The model features five industries with heterogeneous firms producing final goods, intermediate inputs, and capital goods. Key innovations are the distinction between homogeneous intermediate goods (produced on stock) and tailor-made capital goods (ordered in advance), reflecting differences in production processes and the usage of the Almost Ideal Demand System (AIDS) for modelling household consumption behaviour. Calibrated to Austrian data using Eurostat sources and neural posterior estimation, the model is used to analyse the economy’s response to a sector-specific supply shock, illustrated through the example of a flooding event affecting the primary sector. Our results demonstrate that inventory levels critically determine economic resilience a 100-year flood has limited impact regardless of the industry setup, but under a 1,000-year flood, low inventory ratios trigger a vicious circle in which supply shortages cascade across industries, preventing reconstruction and causing a prolonged GDP contraction. High inventory buffers, by contrast, enable rapid recovery. Hence, the structural decomposition into industries becomes decisive when inventories are low, revealing that interdependencies matter most during supply-constrained crises. These findings highlight the importance of explicitly modelling industry interdependencies and inventory dynamics for understanding shock propagation.