Commentary

Minnesota election administration explained: Recounts and election contests

September 8, 2022 7:30 am

A ballot that was challenged during the election recount in the Senate race between Al Franken and Sen. Norm Coleman is seen at the Minneapolis Elections Warehouse on November 19, 2008. Ballots circles are supposed to be filled in completely which is why this ballot was challenged. The key issue in that race wound up being improperly rejected absentee ballots. The Legislature subsequently established more consistent absentee voting procedures. Photo by Cory Ryan/Getty Images.

This is part of an occasional series on election administration. Read part 1, “Who does what?” Part 2, “Who can vote in Minnesota?” Part 3: “How and why polling places are computerized.” Part 4: “How absentee voting works.”  Part 5: “Reconciliation.” Part 6: “Post-election checks.” 

Elections don’t just identify winners. They also persuade losers that they have lost. A recent high-profile loser who was unwilling to be persuaded is a rare exception and should not distract from the mechanisms that support the more typical outcome.

Everything discussed earlier in this series, from registration through the post-election review, plays a role in convincing losers of the reality of their defeat. Two additional mechanisms are specifically designed to address any remaining doubts. 

A recount addresses doubts as to whether the votes were correctly counted, and a contest filed in court addresses any legal issues clouding the result.

Recounts are particularly important in very close elections. In the most extreme cases, the vote counts are exactly tied, leaving the winner to be decided by flipping a coin or some similar mechanism. If a recount finds even a single ballot the tabulator didn’t correctly interpret, that could be enough to decide such a race.

In recognition of this reality, Minnesota provides public funding for recounts of close elections. The losing candidate still needs to request the recount — it’s not automatic. But they don’t have to put up any money. (The only automatic recounts in Minnesota are for constitutional amendments.)

Except in very small elections, the threshold for public funding is one-half of a percentage point for state legislative seats and one-quarter of a point for other federal, state and judicial races. The statute governing recounts in county, school district and municipal elections is somewhat different but again establishes thresholds of one-half or one-quarter of a percentage point.

These thresholds are more than ample to ensure public funding whenever the recount stands a realistic chance of changing the outcome. Years of experience with both recounts and post-election reviews show that the number of ballots marked in ways that cause tabulation difficulties is measured in hundredths of a percent, not tenths.

But what if something completely outside normal experience happened in a particular election? Couldn’t the results be wrong, even in an election that wasn’t so close? That might worry a candidate whose emotional investment in victory was stronger than their faith in election administration. Maybe they even have seen potential signs of trouble, such as results far from their expectations in some specific precincts.

The good news is that Minnesota’s recount law allows a losing candidate to call for a recount no matter how large their margin of defeat. And it allows the candidate to specify up to three precincts to start with, so as to efficiently follow up on any particular suspicions.

In these recounts of less close elections, the candidate needs to provide “a bond, cash, or surety … for the payment of the recount expenses.” But even that comes with some good news. Depending on the recount’s outcome, the candidate might get their money back.

Unsurprisingly, if the recount overturns the election result, the public pays for it. We all have an interest in ensuring our public offices are held by the candidate who got the most votes. That’s not a narrow interest of the candidate, and there is no reason they should pay to vindicate it.

Less obviously, the public also pays if there’s any substantial change in the vote counts, even if not so large as to change the winner. Suppose, for example, someone who loses by 5 percentage points files for a recount, and the revised results show that they lost by only 4 points. That change by a full percentage point would be a huge red flag regarding the accuracy of the tabulation. And were it to ever occur, we the public would pay for having it pointed out to us.

One way to think about this is that the post-election review and publicly funded recounts are both based on the tabulation accuracy being better than a quarter of a percentage point. If anyone thinks we are wrong about that, they are welcome to show us. They can even focus in on particularly suspicious precincts. No one has ever pulled this off.

Regardless of who is paying for the recount, it is conducted the same way. This includes processes designed to ensure transparency — the recount must be open to the public and the procedures must be explained at the outset. The sealed ballot containers are only unsealed and resealed in public view. The actual counting is done by sorting the ballots into piles and then counting the number in each pile.

Unlike the post-election review, a recount includes the candidates or their representatives as active participants. They can challenge the recount official’s determination as to how any particular ballot is counted. Any challenged ballot is set aside, and if the challenge is not withdrawn after further discussion, then the canvassing board makes the final determination how that ballot is counted.

Once the recount is complete, the canvassing board certifies its results. Usually this is the end of the matter. Whether the original winner is re-certified or, more rarely, a new winner takes their place, the vote count is as accurate as it can be.

Losing candidates do also have the option to go to court, however, with a claim that the election outcome was changed by “an irregularity in the conduct of an election or canvass of votes,” or by “deliberate, serious, and material violations of the Minnesota Election Law.” They can also contest the election in court if — notwithstanding the recount — they claim that they received the “largest number of votes legally cast.”

As with recounts, election contests are primarily relevant when the margin of victory is small enough that it can easily be overcome. The most famous election contest is the one that ultimately declared Al Franken, rather than Norm Coleman, to be the winner of the 2008 election for U.S. Senate, having received the largest number of votes legally cast.

That contest was filed by Coleman, who was behind by 225 votes out of more than 2.9 million, and it succeeded only in widening his margin of defeat to 312 votes. At issue were improperly rejected absentee ballots. The Legislature subsequently established more consistent absentee voting procedures.

In an election to a legislative body, the election contest isn’t quite the final word. The legislative body itself has that prerogative under the federal or state constitution. Likewise, in elections of presidential electors, there’s a process for counting the electoral votes. But for other offices, an election contest is the end of the road. And in any case, it is the end of my tour of Minnesota election administration.

The previous six portions of that tour focused on the role of each level of government, as well as topics related to voter registration; the computerization of polling places; absentee voting; reconciliation; and post-election checks.

Did any of the seven columns leave you with questions? Or do you have questions about some eighth topic that I omitted entirely? The Reformer has agreed to cap the series off with an open question-and-answer forum. To participate, email your questions with the subject line: “Election administration.” 

And thank you for your engagement. I’m glad I’m not the only one who thinks election administration matters.

Our stories may be republished online or in print under Creative Commons license CC BY-NC-ND 4.0. We ask that you edit only for style or to shorten, provide proper attribution and link to our web site. Please see our republishing guidelines for use of photos and graphics.

Max Hailperin
Max Hailperin

Max Hailperin is a professor emeritus of mathematics, computer science, and statistics at Gustavus Adolphus College. He earned his Ph.D. at Stanford University and S.B. at the Massachusetts Institute of Technology. Since 2010, he has specialized in the intersection between election technology and election policy, and in 2014, he was awarded the National Association of Secretaries of State (NASS) Medallion Award “in recognition of his service and contributions to election-related technology and legislation.”

MORE FROM AUTHOR