Simulating the economic impacts of crises using large-scale firm-level production network data

The COVID-19 pandemic and vast natural disasters revealed how the complex interactions of firms' supply chain relations can amplify the economic damages of crises.

Swiftly implementing targeted economic recovery packages to reduce the costs to society, requires reliable economic impact prediction (EIP) tools.

Reliable EIP models need to consider how supply chains and their interdependencies evolve in response to crises and, hence, aggravate economic downturns and support the subsequent recovery.

However, most existing EIP models rely on input-output table (IOT) data - with strong resolution limits regarding economic agents, time, and geography - preventing realistic modelling of supply chain interdependencies.

Newly available large-scale production network data - containing the supply links between all firms of entire countries - has the potential to substantially improve EIP by calibrating new economic models to unprecedented economic detail. In this project (IMPACTSIM) we leverage a unique firm-level production network data set to improve EIP of crises.

First, we will analyse the empirical firm-level shock propagation and recovery dynamics during COVID-19 with state-of-the-art tools from econometrics and network science.

Second, we will de-velop a data-driven economic simulation model with firm-level production network dynamics to better predict the economic impacts of future crises.