The Impact of Differing COVID-19 Mitigation Policies: Three Natural Experiments Using Difference-in-Difference Modelling

This paper presents a comparison of the Covid-19 infections between select pairs of border counties in neighboring states. These adjacent border county groups (regions) are where the policies of the Non-Pharmaceutical Interventions (NPI) such as Lockdown/Stay-at-Home differ. These analyses represent a form of a natural experiment and using a Difference-in-Differences (Diff-in Diff) model test the effectiveness of NPI in mitigating COVID-19 infections between border county regions in neighboring states. The border counties are in the states on the Iowa and Illinois border, the Dakotas (North and South) and Minnesota border and the Arkansas and Mississippi border. In each case the policies on each side of the border differ and the border is clearly designated by a river separation. Based on the Diff-in-Diff model output, state policies appear to make a significant difference between some of these specific adjacent border regions, at least early in the pandemic (March-August 2020). This report is an outcome of a short (quick hit) Schar Foundation Grant focused on the first wave of the Covid pandemic in the US. (Feb. 2020-August 2020). Professor Kingsley E. Haynes was the Director and the Principal Investigator with Rajendra Kulkarni as the lead researcher and Abu Siddique and Meng-Ho Li as the primary research assistants with significant research and report writing responsibilities.