Paul Hernnson and Charles Stewart have posted this draft on SSRN. Here is the abstract:

COVID-19 caused worldwide disruption to virtually every aspect of human life, including elections. This study assesses the impact of COVID death rates, convenience voting policies, and partisanship on voter behavior in the 2020 U.S. general election. Using a new data set comprising county and some state data, we demonstrate that countywide COVID-death rates depressed turnout somewhat from 2016 levels, and it contributed to increased use of mail and early-in-person voting options. We also show that the availability of different options structured the methods voters used to cast a ballot. Our results reveal that the emergence of voting policies as a salient issue contributed to a new partisan gap in voter behavior.