Since the initial outbreak of COVID-19 in the United States, researchers from a variety of scientific disciplines have sought to understand the factors influencing the evolu- tion of cases and fatalities. This paper proposes a two-stage econometric modeling approach to analyze a range of socioeconomic, demographic, health, epidemiological, climate, pollution, and political factors as potential drivers of the spread of COVID- 19 across waves and counties in the United States. The two-step modeling strategy allows us to (i) accommodate the observed heterogeneity across waves and counties in the transmissibility of the virus, and (ii) assess the relative importance of the cross- sectional measures. We leverage the availability of daily data on confirmed cases and deaths of COVID-19 in counties across the 48 contiguous states and the District of Columbia, spanning a two-year period from March 2020 to March 2022. We find that socioeconomic and demographic factors generally had the greatest influence on the transmissibility of the virus and the associated mortality risk, with health and climate factors playing a lesser role.
