Privacy violations in election results.

Area of Science:

• Political Science
• Computer Science
• Information Security

Background:

• Debate exists on whether releasing anonymous ballots post-election enhances transparency or compromises ballot secrecy.
• Policy-makers consider public disclosure to address public distrust in election integrity.
• Concerns arise that linking votes to voters could undermine the principle of secret ballots.

Purpose of the Study:

• To introduce and define the concept of 'vote revelation'—linking anonymous votes to voter identities.
• To analyze the theoretical mechanisms through which vote revelation could occur.
• To quantify the privacy risks associated with releasing individual ballot records versus aggregate data.

Main Methods:

• Case study analysis of the 2020 election in Maricopa County, Arizona.
• Modeling the potential for vote revelation under different data release scenarios.
• Comparison of individual ballot record release against current aggregate reporting practices.

Main Results:

• Releasing individual ballot records would result in no vote revelation for 99.83% of voters.
• Current aggregate reporting in Maricopa County protects voter privacy slightly better (99.95% no revelation).
• Vote revelation risk is concentrated among voters using provisional or federal-only ballots.

Conclusions:

• The risk of vote revelation from releasing individual ballot records is minimal for the vast majority of voters.
• A trade-off exists between election transparency and voter privacy, inherent in all reporting methods.
• Targeted remedies can mitigate privacy violations for specific voter groups, balancing transparency goals.