The worldwide spreading of the coronavirus in 2020 is a large risk for human lives and for the world economy. It causes severe economic disruptions in both the production and the service sectors due to mortality and morbidity, quarantines, travel restrictions, as well as changes in investor and consumer behavior. Practitioners and policy makers face difficulties in understanding the pandemic and forecasting its economic fallout. These difficulties reflect a lack of empirical evidence on the global economic dynamics of epidemics.
In microeconomics, there is a tradition of empirical work on the economic impact of viral diseases. However, the studies largely focus on individual outcomes or single countries. In macroeconomics, the literature on epidemics is thin and largely confined to model-based analyses. In Kholodilin and Rieth (2020) we construct a news-based viral disease index for the monthly frequency. We analyze the text of over 500 million newspaper articles by counting words like `coronavirus' or `swine flu', following a popular approach for measuring economic policy uncertainty (Baker et al. 2016). We use our index in time-series models (structural vector autoregressions) and estimate the effects of epidemic shocks on the world economy. We analyze the global macroeconomic effects and propagation of such shocks, and complement this evidence with detailed country studies and a historical analysis of epidemics in the U.S. since 1920.
The estimates provide three stylized facts. First, the macroeconomic damage of epidemic shocks is about 20 times larger than that of seasonal influenza shocks. Second, the adverse impact of epidemic shocks lasts for about three quarters and is roughly four times larger for epicenter countries than for countries indirectly affected through supply chain disruptions, global trade, or sentiment. Third, epidemic shocks lead to a simultaneous fall of consumer prices with economic activity.
The results have several implications. First, they suggest that epidemics are disparately more economically damaging than regular influenza, thus questioning a popular view that Covid-19 is just another flu. Furthermore, they indicate that epidemics are costly tail events that are roughly comparable to financial or political crises. The costs also reflect containment measures, such as school closures, lockdowns, and other means of social distancing, which are taken to reduce the number of cases because the public health system has limited shock absorption capacities. In the long-run, these costs must be weighed against the short-run gains of trimming public health systems as more resilient systems would potentially allow for softer forms of disease control in the case of tail events.
Second, the results indicate that preventing the spreading of viral diseases across borders is key as the direct costs of epidemic shocks to countries is substantially larger than the indirect effects. Transparency in the testing and reporting of cases worldwide and multilateral containment policy coordination, for example, through the World Health Organization, could help mitigate international contagion.
Third, the findings suggest that the negative demand effects of epidemic shocks on businesses, consumers, and investors are larger than the negative supply effects. Epidemic shocks are both contractionary and deflationary. This, in turns, indicates that the expansionary monetary and fiscal policies currently implemented around the world are an appropriate response to the Covid-19 shock.
Malte Rieth (DIW Berlin)
Kholodilin, K.A., Rieth, M., 2020. Viral shocks to the world economy. DIW Discussion Paper 1861.
Baker, S.R., Bloom, N., Davis, S.J., 2016. Measuring economic policy uncertainty. The Quarterly Journal of Economics 131, 1593-1636.
