No Panic in Pandemic: The Impact of Individual Choice on Public Health Policy and Vaccine Priority
48 Pages Posted: 14 Jan 2021 Last revised: 30 Jun 2021
Date Written: January 10, 2021
Infectious disease outbreaks such as COVID-19 pose significant public health threats and challenges worldwide due to their high transmissibility and potentially severe symptoms and complications. Although public health interventions such as social distancing and lockdown can slow the disease spread, the disruption to regular economic and social activities caused by these interventions have caused significant financial losses. Strategic planning is required to optimize the timing and intensity of these public health interventions by considering individual response. We derive insightful structural properties of the optimal public health interventions and conduct numerical studies based on representative COVID-19 data in Minnesota. We find that the individual equilibrium activity level is higher than the socially optimal activity level due to an individual’s ignorance of the negative externality imposed on others, with the largest difference at a middle-range disease prevalence. As a result, lockdowns and social distancing policies are more effective when the disease prevalence is not at its peak level. Social distancing is more effective than lockdowns based on the representative COVID-19 data from Minnesota. Moreover, due to the limited vaccine capacity, vaccination priority strategy needs to consider the trade-off between the higher mortality rate of the less active group and the higher negative externality imposed by the more active group. Changes in vaccine production capacity, mortality rate ratio and infection rate may affect vaccination priorities. Lastly, while the vaccine priority to the elderly group is most effective in reducing total deaths, it has to be accompanied with more stringent social distancing policies.
Keywords: COVID-19, dynamic compartmental model, public health policy analysis, game theory, multinomial logit model
Suggested Citation: Suggested Citation