Election authorities should remove ballots from confirmed noncitizens after reliable verification and due process

What's this about?

People disagree about whether staff should take out a vote from a person who is not a U.S. citizen.

The key question is whether strong proof and fair steps come first.

What supporters say

• Staff should not count a vote when strong, person-by-person proof shows that the voter could not vote.
• Special record checks may help staff spot clear cases, if people also check the facts by hand.

What critics say

• A record match can name the wrong person because of name changes, bad data, or typing errors.
• Voting by people who are not citizens seems so rare that wide list sweeps may harm more legal voters than they help.
• Taking out votes near an election may not give voters enough time to see the charge, fix a mistake, or get a fair review.

How to read this

The number of points on each side does not show who is right; the strength of the proof matters more.

The bottom line

The claim makes sense in rare, clear cases, after notice, review, and a fair chance to fix errors.

But the facts raise strong doubts about wide or late vote removals, so staff should not rely on one match or rush the choice.

The claim is that election authorities should remove a ballot when reliable evidence confirms that the voter was not eligible because they were not a U.S. citizen—but only after notice, review and a chance to correct mistakes. The evidence supports that principle in narrow cases, while raising serious doubts about broad or last-minute removals.

The case for

A confirmed ineligible ballot should not be counted. Federal election rules and ordinary voter-list maintenance give authorities a legitimate reason to identify and correct registration problems. If independent, individualized evidence establishes that a voter was a noncitizen, removing that ballot protects the votes of eligible citizens without requiring authorities to act against everyone whose records raise questions. 1

Specialized verification systems could make this process more careful than an automatic database purge. The federal government’s SAVE system, for example, is designed to help agencies check immigration status and citizenship for purposes including voter registration and list maintenance. Used as one part of a broader investigation—not as final proof from a single computer match—it could help authorities distinguish confirmed cases from uncertain ones. 2

That is the strongest version of the claim: not automatic removal, but removal after reliable confirmation and due process. On that narrower approach, authorities would rely on multiple records and human review before deciding that a ballot came from an ineligible voter.

The case against

The main danger is that a database match may identify the wrong person. Name changes, clerical mistakes, incomplete records and limitations in matching systems can create false positives. A government record that appears to conflict with a voter’s registration is therefore not, by itself, proof of noncitizenship. The National Academies has supported conditional or provisional procedures when matches are uncertain.

The available evidence also suggests that noncitizen voting is rare compared with the number of ballots cast. Investigations and reviews have generally found very few substantiated or suspected cases, although those reviews are incomplete and the underlying cases differ in quality. A peer-reviewed exchange has defended a nonzero estimate, but it relied on a small, self-reported sample and has faced methodological criticism. Other reviews conclude that many allegations cannot be accepted without individual confirmation. 4

There is also no independent, comprehensive assessment showing how accurately SAVE or similar systems work in election administration. One recent assessment found that election offices were using SAVE more extensively even though its accuracy and administrative burden remained largely unstudied. The central unanswered question is whether these systems can measure and control their false-positive rate before a ballot is permanently excluded. 3

Timing can make the safeguards ineffective. The Virginia litigation involved removing suspected noncitizens shortly before an election and sought restoration and notice for voters who may have been wrongly removed. The Supreme Court’s decision allowing the program to proceed did not establish that every flagged voter was actually a noncitizen. Late action can leave voters without meaningful time to receive notice, challenge the finding or correct the record. 5

The evidence therefore favors advance notice where feasible, individualized proof, human review, correction procedures and provisional voting while eligibility remains unresolved. These are not minor administrative details; they are essential protections against disenfranchising citizens.

The bottom line

The evidence weakly favors a tightly limited version of the claim: authorities should exclude a ballot when noncitizenship has been individually and reliably confirmed, with genuine notice and review. It does not support automated, bulk or late-stage removal based only on database flags.

Confidence in that conclusion is low. The basic principle—that a confirmed ineligible ballot should not count—is relatively clear. But the evidence remains incomplete on the questions that matter most in practice: how often verification systems wrongly flag citizens, whether voters can obtain timely review, and how widespread noncitizen voting actually is.