A Cross-Country Analysis of the Determinants of Covid-19 Fatalities

By Toya, Hideki / Skidmore, Mark

My coauthor (Hideki Toya) and I recently posted a working paper that offers an examination of the factors that explain differences across countries in COVID-19 infections and fatalities over the past year. Our analysis explains a high proportion of variation in both infections and fatalities across more than 150 countries. We find that high-income countries and countries with older populations have more infections and fatalities. The high-income result is unusual….epidemics/outbreaks are typically far more common in low-income countries were sanitation, access to clean water, etc. are a challenge. Island countries and countries that restricted international travel also had fewer infections and fatalities. Recommended (but not mandatory) lockdowns were associated with fewer infections. However, we find little evidence that any type of lockdown reduced fatalities…if anything the opposite was the case for mandatory nationwide lockdowns. Perhaps the two most important findings are: 1) Countries that relied on PCR testing had more infections but not more fatalities. This result is consistent with evidence that PCR testing generates a high rate of false positives, resulting in more recorded infections but not more fatalities; and 2) access to hydroxychloroquine (HCQ) substantially reduced fatalities. The estimates suggest that if the U.S. had made HCQ widely available early on, 80,000 to 100,000 lives could have been saved. While the paper has been submitted to a journal, it is not yet peer reviewed. You can access the paper here.

Download the PDF

Lighthouse Economics