Election administration harms and ballot design: A study of Florida's 2018 United States Senate race

Abstract We introduce a typology of election administration harms and apply it to empirically study the consequences of ballot design. Our typology distinguishes between individual, electoral, and systemic harms. Together, it clarifies why ballot design can be a particular vulnerability in election administration. Using both ballot‐level and precinct‐level data, we revisit Florida's 2018 United States Senate race, in which Broward County's ballot design flouted federal guidelines and, according to critics, was pivotal to the outcome. We estimate that Broward's ballot design induced roughly 25,000 voters to undervote in a race determined by about 10,000 votes and that these excess undervotes were concentrated among low‐information voters. Broward's ballot did not, however, affect the outcome of the election. Nonetheless, flawed ballot designs are still concerning in an age of voter distrust. Given the risk that flawed ballots can cause systemic harm, we offer a roadmap for procedural reforms to improve ballot design.