Endogenous mobility in pandemics: theory and evidence from the United States

We study infectious diseases in a spatial epidemiology model with forward-looking individuals who weigh disease environments against economic opportunities when moving across regions. This endogenous mobility allows regions to share risk and health resources, resulting in positive epidemiological externalities for regions with high R0s. We develop the Normalized Hat Algebra to analyze disease and mobility dynamics. Applying our model to US data, we find that cross-state mobility controls that hinder risk and resource sharing increase COVID-19 deaths and decrease social welfare. Conversely, by enabling "self-containment" and "self-healing," endogenous mobility reduces COVID-19 infections by 27.6% and deaths by 22.1%.