Impact of COVID-19 on financial returns: a spatial dynamic panel data model with random effects

Abstract

Using a dataset including financial market returns and volatility proxies for several countries, we analyzed the impact of Covid-19 deaths on the financial economy. From the modeling perspective, we consider a spatial panel data model for returns and a spatial dynamic panel data for volatilities. Proper marginal effects are calculated to exploit information on short- and long-term effects. A Chow test is used to identify the existence of a structural break in each series. Our empirical evidence shows that in the first weeks of the Covid-19 outbreak, until mid-March 2020, the identified break date, the spatial effect of Covid-19 deaths was statistically significant, leading to a contraction in returns and an increase in risk. Moreover, the effects disappeared in the remaining months as the financial markets moved back to pre-crisis levels, causing a decrease in the overall risk. Our evidence supports the behavioral impact of the pandemic on financial markets.