Why Equity Markets Go Up (Yes, Up) When There Is an Official COVID Announcement

CROCE, FARRONI AND WOLFSKEIL ANALYZE THE RELATIONSHIP BETWEEN THE DIFFUSION OF PANDEMIC INFORMATION AND MARKET TRENDS AND PROVIDE A PREDICTIVE TOOL BASED ON TWITTER, USEFUL TO AVOID BEING CAUGHT UNPREPARED AGAIN BY A FINANCIAL COLLAPSE OF THIS KIND

The COVID-19 pandemic has triggered an unprecedented destruction of financial wealth. In the last week of February alone, it is estimated that stock markets around the world lost 6 trillion and since then another 9 trillion disappeared. Yet, according to an analysis conducted by Mariano Max Croce, Full Professor of Finance at Bocconi and two PhD students, Paolo Farroni and Isabella Wolfskeil, in the moments immediately following the official daily briefings and updates about victims and contagion, stock markets rise.
 
 
In a paper published in the CEPR's COVID Economics series, the three authors analyze the relationship between pandemic information diffusion and financial markets, giving scientific evidence of some trends and making some counterintuitive observations. “The rise of prices in the moments immediately following the announcements”, explains Professor Croce, “seems to be due to the reduction of uncertainty. Prices are very low because they already incorporate the likely information on the pandemic's trend. When the announcement comes, they react positively because the element of uncertainty disappears. It is, however, a very short-lived effect, followed by a U-shaped trend: in the next 60 minutes, prices fall and then slowly pick up again”.
 
The risk of bad news about the pandemic has a cost to the markets, which scholars estimate at 1% per day. This risk has a global component (the pandemic trend as a whole) and a local component. The latter is measured by the relative share of deaths and infections in a given country against the total number of deaths and infections in the world.
 
On the other hand, government bond prices do not seem to be affected by the announcements, either in a positive or negative sense.
 
The three scholars take into consideration the markets of 15 countries and build two datasets on the spread of information. A low-frequency dataset includes official announcements and a high-frequency dataset includes the tweets of major media (a measure of the diffusion of information) and their retweets (a measure of market attention to the development of the pandemic).
 
Looking at the high frequency dataset, the important aspect is the tone of the tweets and retweets. When the tone is mostly positive, markets improve and when it is negative, they get worse. Deviation can be as much as 1% per hour.
 
“Although the observation may seem obvious,” explains Prof. Croce, “it is important because it confirms that our tweet and retweet-based indicator is a good predictor of market performance. If, in fact, in the event of a sovereign debt crisis we can monitor risk trends by observing credit default swap prices, but we had no instrument for pandemic risk”. Now, however, the Twitter-based dataset, which includes thousands of high-frequency observations and allows very reliable and fast statistical analysis, will be made public in the coming weeks. “It is intended as a tool that can be used to predict market trends as a function of pandemic trends,” says Prof. Croce, “and it could prove useful in the future to avoid getting caught off-guard by such emergencies”.
 
Mariano Max Croce, Paolo Farroni e Isabella Wolfskeil, “When the Markets Get COVID: Contagion, Viruses and Information Diffusion”, available in SSRN and in CEPR COVID Economics series.

by Fabio Todesco