Working Paper: NBER ID: w26987
Authors: Sydney C. Ludvigson; Sai Ma; Serena Ng
Abstract: The outbreak of COVID-19 has significantly disrupted the economy. This paper attempts to quantify the macroeconomic impact of costly and deadly disasters in recent US history, and to translate these estimates into an analysis of the likely impact of COVID-19. A costly disaster series is constructed over the sample 1980:1-2020:04 and the dynamic impact of a disaster shock on economic activity and on uncertainty is studied using a VAR. While past natural disasters are local in nature and come and go quickly, COVID-19 is a global, multi-period event. We therefore study the dynamic responses to a sequence of large disaster shocks. Even in a fairly conservative case where COVID-19 is a 5-month shock with its magnitude calibrated by the cost of March 2020 Coronavirus relief packages, the shock is forecast to lead to a cumulative loss in industrial production of 20% and in service sector employment of nearly 39% or 55 million jobs over the next 12 months. For each month that a shock of a given magnitude is prolonged from the base case, heightened macro uncertainty persists for another month.
Keywords: COVID-19; disaster; uncertainty
JEL Codes: E17; I0; I3
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| shock duration (C41) | heightened macro uncertainty (E66) |
| COVID-19 (I15) | persistent effects on economic activity (F69) |
| shock duration (C41) | cumulative loss in industrial production (E23) |
| shock duration (C41) | cumulative loss in service sector employment (J68) |
| COVID-19 (I15) | cumulative loss in industrial production (E23) |
| COVID-19 (I15) | cumulative loss in service sector employment (J68) |