Chance Elections, Social Distancing Restrictions, and Kentucky’s Early COVID-19 Experience

Estimates an event-study model to link adoption of several common social distancing measures – public school closures, bans on large gatherings, closures of entertainment-related businesses such as restaurants, and shelter-in-place orders (SIPOs) – to the growth rate of cases across counties in the Midwest and South in the early stages of the pandemic. These policies combined to slow the daily growth rate of COVID-19 cases by 9 percentage points after 16 days, with SIPOs and entertainment establishment closures accounting for the entire effect.

In order to obtain results with more direct applicability to Kentucky, we then estimate a model that interacts the policy variables with a “white working class” index characterized by political conservatism, rurality, and high percentages of white, evangelical Christian residents without college degrees. We find that the effectiveness of early social distancing measures decreased with higher values of this index. The results imply that the restrictions combined to slow the spread of COVID-19 by 12 % points per day in Kentucky’s two largest urban counties but had no statistically detectable effect across the rest of the state.