After an election, should election officials release a copy of each anonymous ballot? Some policymakers have championed public disclosure to counter distrust, but others worry that it might undermine ballot secrecy.
A new study in Sciences Advances, co-authored by Penn Carey Law’s Michael Morse C’13, Assistant Professor of Law, along with Shiro Kuriwaki, Assistant Professor of Political Science at Yale, and Jeffrey B. Lewis, Professor of Political Science at UCLA, informs the ongoing public debate. It introduces the concept of vote revelation—the potential for a vote on an anonymous ballot to be linked to the voter’s name in the public voter file. Using the 2020 election in Maricopa County, Arizona, as a case study, the authors evaluate the extent of potential revelation across multiple methods of election reporting—from aggregating results at the precinct-level to releasing individual ballots. They find that the release of individual ballot records would lead to no revelation of any vote choice for 99.83% of voters as compared to 99.95% under Maricopa’s current practice of reporting results by precinct and method of voting. The study explores the tradeoffs between election transparency and voter privacy, offering insights into potential safeguards that could enhance public confidence while protecting the integrity of the secret ballot.
Michael Morse C'13, Assistant Professor of Law
“There will always be some tension between transparency and privacy in reporting the results of an election,” says Morse. “We want transparency—no one would trust an announcement from above that a given candidate won with no other details—but the push to publish ballots online has led to concerns about privacy, even though ballots themselves are anonymous.”
“Shiro Kuriwaki, Jeff Lewis, and I started this project about two years ago as election officials were inundated with public records requests for individual ballots, known as cast vote records,” he continued. “Now, I think we can offer election officials some theoretical and empirical guidance. First, the risk of vote revelation from cast vote records is quite small. Second, releasing individual, anonymous ballots do not actually reveal much more than the aggregate election results already published in many jurisdictions— that’s because vote revelation is a function of what we call quasi-identifiers, basically a voter’s precinct, vote method, and ballot style, which we explain in detail in the paper. Ultimately, different election officials may strike their own normative balance between voter privacy and election transparency. But at least as to the empirics, I think election officials should feel comfortable embracing transparency in election reporting and remedying the specific privacy risks through the targeted short and long-term solutions we offer.”
Original source can be found here.