How Broward's Ballot Design Caused 25,000 Undervotes but Didn't Flip 2018 Race

📚 A New Typology of Election Harms

This article introduces a clear typology that separates election administration harms into three categories: individual, electoral, and systemic. The typology demonstrates why ballot design is a particular vulnerability in administering elections and helps clarify the different stakes when errors occur.

🧾 Examining Florida's 2018 Senate Race with Ballot- and Precinct-Level Records

The analysis revisits the 2018 United States Senate race in Florida, focusing on Broward County's ballot, which violated federal guidance and was criticized as pivotal to the outcome. The study uses both ballot-level and precinct-level data to measure the ballot's effects on voter choices and undervotes.

📈 Key Findings

• The ballot design in Broward County is estimated to have induced roughly 25,000 excess undervotes in the Senate race.
• The statewide margin in the race was about 10,000 votes, meaning the excess undervotes exceeded the margin but did not change the official outcome.
• Excess undervotes were concentrated among low-information voters, indicating differential impact across voter groups.
• Despite the large number of excess undervotes, the empirical evidence shows Broward's ballot did not alter the election result.

⚖️ Why This Matters and the Policy Response

Flawed ballot designs remain a significant concern given heightened voter distrust and the potential for broader, systemic harms. Because ballot problems can produce harms at the individual, electoral, and systemic levels, procedural reforms are warranted. The article concludes by offering a roadmap for procedural reforms aimed at improving ballot design to reduce these risks.