I’ve spent 30 years testifying about numbers that decide outcomes, and juries always want a number they can trust. The number that just decided whether 709,841 Michigan voters get a say on their own state constitution never cleared that bar, and last week two courts confirmed it doesn’t matter this year.
On Aug. 24, the Michigan Board of State Canvassers pulled a random sample of 1,000 signatures from a citizen-led petition and found 626 valid. It needed 629. Three signatures out of a thousand decided the fate of a statewide ballot question backed by more people than live in Grand Rapids. The board split along party lines, two Republicans voting to certify and two Democrats voting no, and did nothing. Under Michigan law, nothing is a decision. The petition, which would require proof of citizenship to register and photo ID to vote, is now off the Nov. 3 ballot.
Sponsors asked the Michigan Supreme Court for relief on Aug. 27 and got silence. On Sept. 1, sponsors went over its head to the U.S. Supreme Court, and Justice Brett Kavanaugh gave Michigan officials until Wednesday to respond. On Sept. 3, he denied the emergency request in a one-line order with no explanation. Hours later, the Michigan Supreme Court issued its own 16-page ruling finding the canvassers had discretion, while suggesting the legislature write clearer deadlines so this doesn’t happen again. Americans for Citizen Voting says it will try again in 2028.
This isn’t really a story about citizenship voting. It’s a story about what happens to self-government when the machinery meant to run it grows slower and less accountable than the people it serves. Congress fixed the first Tuesday after the first Monday in November as the nation’s election date in 1845, mostly to stop states that voted early from swaying states that voted late. That date has held for 181 years. What hasn’t held is everything built around it. Michigan’s Bureau of Elections took more than five months to produce a staff report on a petition filed the week many farmers were still planting corn.
Do the math. Michigan required 446,198 valid signatures. Divide that by the 709,841 submitted and sponsors needed a validity rate of 62.9%, which is why the board set 629 out of 1,000 as the certification line. Bureau staff found 692 valid, then fielded 377 challenges and settled on 612, 17 short of the 629 needed. Sponsors then produced notarized affidavits for 17 more signers swearing the disputed signature was theirs, a method the campaign says the bureau’s own elections director calls acceptable. The board accepted 14 of them, moving the count to 626, three short. Reasonable people can disagree whether an old registration signature matches a shaky pen stroke. They shouldn’t disagree that a review this close and this slow has no business deciding whether 709,841 Michiganders get a vote on their own constitution.
The same fragility shows up every fall in how we actually vote. Michigan’s ballot had to lock 60 days before the election, on Sept. 4, so clerks could print and mail ballots on time, military and overseas voters first, everyone else 40 days out, on Sept. 24. Early voting isn’t the problem; a soldier overseas or a nurse on a double shift deserves that option. Every week added to the front of the calendar is a week where a signature review or a court fight can collide with ballots already mailed. Michigan’s canvassers had five and a half months to review this petition and finished in the one week state law left them no room to spare.
The people opposing this amendment aren’t making a fake argument. Strict ID rules can trip up voters who lack a driver’s license, can’t easily get a birth certificate, or move often enough that their address on file never catches up. That concern deserves an answer, not a shrug, and most states with ID laws build one in: free state identification, signed affidavits, provisional ballots, a cure period for a mismatched signature. Michigan’s own proposal paired photo ID with those safeguards. How generous that net should be is a legitimate debate. Whether three signatures out of a thousand should decide a ballot question, five months in, isn’t a partisan question. It’s a competence problem, and Michigan’s own Supreme Court just said as much.
The public settled the underlying question even if Lansing hasn’t. Seventy-seven percent of registered voters, including 62% of Democrats, told Fox News pollsters in July they want photo ID to cast a ballot, and 83% want it to register. Gallup found nearly identical numbers in 2024, 84% for voting and 83% for registration. That’s a landslide position a bureaucratic clock nearly buried before a single voter said yes or no.
DEMOCRATS SUBSIDIZE IDLENESS. REPUBLICANS REWARD HARD WORK
None of this happened because anyone at the Bureau of Elections is a villain; the incentives just point the wrong way. A review that takes five months and turns on a three-name margin isn’t corrupt; it’s unaccountable, and unaccountable is what a system produces when nobody answers for how long a review takes or how close a call it makes. Set a real deadline, publish the sampling methodology before the first signature gets collected, and give petitioners a meaningful appeal before the clock runs out, and the incentive fixes itself. The Michigan Supreme Court just told its own legislature the same thing.
Congress gave America one Election Day in 1845 because a farming country needed a date it could trust. We kept the date. We let everything around it drift: five-month reviews, three-signature margins, a ballot deadline that gave the courts one day to decide before printing began. Michigan’s 709,841 signers didn’t ask for a free-for-all. They asked for a vote, and they’ll ask again in 2028. The state that fixes its clock before the next petition and deadline collide is the state that keeps Election Day meaning what Congress intended, in 1845 and now.
Jay Rogers is a financial professional with more than 30 years of experience in private equity, private credit, hedge funds, and wealth management. He has a Bachelor of Science in criminal justice from Northeastern University and has completed postgraduate studies at UCLA, the University of Pennsylvania, and Harvard. He writes about issues in finance, constitutional law, national security, human nature, and public policy.