A new paper by an M.I.T. elections expert predicts that the outcome of this year’s presidential election — and the problem known as the “lost vote,” in which legitimate ballots go uncounted — could fuel post-election allegations of a rigged election. The paper, by Charles Stewart III, highlights how Mr. Trump’s claims about problems with mail balloting could drag on long after November if the election is close.
A “lost vote” occurs when a voter does everything necessary to vote but, thanks to administrative errors, the vote isn’t counted in the final tally, according to Mr. Stewart, a professor of political science. In Georgia’s June primaries, for example, ballot scanners did not count mail ballots when voters used check marks instead of filling in the ovals.
In the paper, released this week, Mr. Stewart concludes that, for a variety of reasons, while lost votes are rare, they occur more frequently when mail-in ballots are used.
In the 2016 election, he writes, approximately 4 percent of the mail ballots cast — or 1.4 million votes — went uncounted. With more states now embracing mail ballots, including a number of states with little experience with voting by mail, Mr. Stewart predicts a “disproportionate growth” in the number of lost votes in November.
“Despite the clear public health imperative that mail balloting be increased in the 2020 primary season and general election, the expansion of mail balloting comes with risks,” Mr. Stewart writes. “To be sure, these risks are small, and should not be sensationalized.”
In an interview Friday, Mr. Stewart said that some of his past writing on the topic had been misinterpreted, including in recent legal documents that used his research to sensationalize the risk of lost votes. “It’s not a basket of votes being hidden or mailed in from Siberia or name your fantasy story,” he said.