In and out of lockdown: Propagation of supply and demand shocks in a dynamic input-output model

TL;DR

This paper develops a dynamic input-output model to analyze how Covid-19 related supply and demand shocks propagated through industries, emphasizing the importance of input necessity and accurately predicting economic impacts during lockdowns.

Contribution

It introduces a novel modeling approach that incorporates industry-specific input necessity based on expert surveys, improving the understanding of shock propagation during Covid-19.

Findings

The model accurately predicted aggregate and sectoral dynamics during the UK lockdown.

Input substitutability assumptions significantly affect model predictions.

Output multipliers are only mildly useful for shock impact prediction.

Abstract

Economic shocks due to Covid-19 were exceptional in their severity, suddenness and heterogeneity across industries. To study the upstream and downstream propagation of these industry-specific demand and supply shocks, we build a dynamic input-output model inspired by previous work on the economic response to natural disasters. We argue that standard production functions, at least in their most parsimonious parametrizations, are not adequate to model input substitutability in the context of Covid-19 shocks. We use a survey of industry analysts to evaluate, for each industry, which inputs were absolutely necessary for production over a short time period. We calibrate our model on the UK economy and study the economic effects of the lockdown that was imposed at the end of March and gradually released in May. Looking back at predictions that we released in May, we show that the model predicted aggregate dynamics very well, and sectoral dynamics to a large extent. We discuss the relative extent to which the model's dynamics and performance was due to the choice of the production function or the choice of an exogenous shock scenario. To further explore the behavior of the model, we use simpler scenarios with only demand or supply shocks, and find that popular metrics used to predict a priori the impact of shocks, such as output multipliers, are only mildly useful.