International Evidence on Vaccines and the Mortality to Infections Ratio in the Pre-Omicron Era

Prior to the appearance of the Omicron variant, observations on countries like the UK that have accumulated a large fraction of inoculated individuals suggest that, although initially, vaccines have little effect on new infections, they strongly reduce the share of mortality out of a given pool of infections. This paper examines the extent to which this phenomenon is more general by testing the hypothesis that the ratio of lagged mortality to current infections is decreasing in the total number of vaccines per one hundred individuals in the pre-Omicron period. This is done in a pooled time-series, cross-section sample with weekly observations for up to 208 countries. The main conclusion from the statistical analysis is that vaccines moderate the share of mortality from a given pool of lagged infections. This is essentially a favorable shift in the tradeoff between life preservation and economic performance. Controlling for income per capita, stringency of containment measures, and the fraction of recovered and old individuals, estimation is carried out by linear least squares, with standard errors clustered by country and region. The main result is robust to sensitivity analysis with a logarithmic specification. The practical lesson is that, in the presence of a sufficiently high share of inoculated individuals, governments can shade down containment measures, even as infections are still rampant, without significant adverse effects on mortality.