Noncitizen Voting: What the Evidence Actually Shows
Election integrity is one of those topics where the temperature of the debate has almost nothing to do with the temperature of the evidence. Noncitizen voting has become a central issue in American politics. It drives legislation, shapes campaign messaging, and shows up in polls as a top concern.
So what does the evidence actually say? This is not an opinion piece. This is a look at the legal framework, the available data, and the policy tradeoffs. The data on this topic is clearer than the debate suggests.
It Is Already Illegal
This is the starting point that often gets lost in the conversation. Noncitizens voting in federal elections is already a crime. It has been since 1996.
The Illegal Immigration Reform and Immigrant Responsibility Act of 1996 made it a federal offense for any noncitizen to vote in a federal election. The penalty structure is serious:
- Fines and up to 5 years in federal prison under 18 U.S.C. 611
- Deportation and permanent inadmissibility under immigration law
- Additional penalties under 52 U.S.C. 10307(c) for anyone who votes fraudulently in a federal election
That last consequence is worth pausing on. For an undocumented immigrant, casting a single vote would create a permanent bar to ever obtaining legal status in the United States. For a green card holder, it would be grounds for deportation and loss of their legal residency. For someone in the naturalization process, it would end any chance of becoming a citizen.
When you register to vote in any state, you must sign an attestation that you are a U.S. citizen. That signature is made under penalty of perjury, which carries its own criminal penalties.
A Note on Local Elections
A handful of jurisdictions do allow noncitizens to vote in local elections. Several municipalities in Maryland, some towns in Vermont, and San Francisco have passed local ordinances allowing noncitizen residents to vote in school board or city council elections. These local policies are entirely separate from federal elections and do not change the federal law. No state allows noncitizen voting in state or federal elections.
How Often Does It Actually Happen?
This is the central factual question, and researchers have studied it from multiple angles. The findings are consistent across ideological lines.
The Brennan Center Study (2016)
The Brennan Center for Justice surveyed election officials across 42 jurisdictions covering 23.5 million votes cast in the 2016 election. They found approximately 30 suspected cases of noncitizen voting. That is an incident rate of about 0.0001%.
Critics sometimes point out that the Brennan Center is a left-leaning organization. That is fair context. But the methodology was straightforward: they asked the people who actually run elections how many cases they found.
The Heritage Foundation Database
The Heritage Foundation maintains the most comprehensive database of documented election fraud cases in the country. Heritage is a conservative organization, and its database was created specifically to document that voter fraud exists.
Their database contains roughly 100 documented cases of noncitizen voting over a period spanning more than 40 years. During that time, billions of ballots were cast in U.S. elections. Even using Heritage's own data, which is assembled by an organization with every incentive to find as many cases as possible, noncitizen voting represents a tiny fraction of total votes.
The Cato Institute Analysis
The Cato Institute, a libertarian think tank, analyzed the available data and concluded that noncitizen voting is "vanishingly rare." Their analysis looked at both the documented cases and the statistical evidence and found that even the most aggressive estimates produce numbers too small to change election outcomes.
This is notable because Cato is not an organization sympathetic to lax immigration enforcement. They reached their conclusion based on the data.
State-Level Audits
Several states have conducted their own audits, and the results follow a consistent pattern.
Texas (2019). The Texas Secretary of State flagged approximately 95,000 registered voters as potential noncitizens based on a comparison of voter rolls with Department of Public Safety records. The announcement generated national headlines. Then the actual investigation happened.
The comparison had matched people who were noncitizens when they got a driver's license or state ID but had since become naturalized U.S. citizens and legally registered to vote. After counties reviewed the flagged names, the vast majority were confirmed as eligible voters. The number of actual confirmed noncitizen registrations was a tiny fraction of the original 95,000 figure. The state faced lawsuits over the flawed list, and the Secretary of State's office acknowledged the methodology was problematic.
Georgia (2022). Georgia conducted an audit of its voter rolls and identified about 1,634 people who had attempted to register but were blocked by the state's verification system. Of those, the state found zero who had actually cast a ballot. The system worked as designed: it caught noncitizens trying to register and stopped them.
Pennsylvania (2024). Pennsylvania's audit of voter registrations found a small number of noncitizens on the rolls, most of whom had been added through clerical errors at motor vehicle offices. The state referred cases for investigation, but the total numbers were, again, extremely small relative to the state's millions of registered voters.
Ohio, North Carolina, and others have conducted similar reviews with similar results: very small numbers, often traceable to administrative errors, and very few cases of noncitizens actually casting ballots.
The 2024 Election Cycle
Multiple states conducted voter roll audits ahead of the 2024 election. Virginia removed approximately 6,300 people from its rolls who had been flagged as potential noncitizens, though civil rights groups challenged the process, arguing it swept up eligible citizens. Other states reported similar small-scale findings. No state found evidence of large-scale noncitizen voting.
The Incentive Problem
Beyond the data, there is a basic question of incentives that helps explain why the numbers are so small.
Consider the risk-reward calculation from the perspective of a noncitizen. Voting illegally means:
- Risking a federal felony conviction
- Risking up to 5 years in prison
- If undocumented, virtually guaranteeing deportation
- Creating a permanent bar to ever obtaining legal status
- Committing perjury on the registration form
The reward is: one vote in one election.
For someone who is building a life in the United States, whether as an undocumented immigrant, a green card holder, or someone in the naturalization pipeline, the consequences of getting caught are devastating and life-altering. The benefit is negligible. One vote has never decided a federal election.
This does not mean it never happens. The data shows it does happen in very small numbers. But the incentive structure helps explain why it does not happen at scale. The people who would supposedly be committing this crime are the people with the most to lose from getting caught.
The SAVE Act: What It Would Change
The Safeguard American Voter Eligibility (SAVE) Act, introduced as HR 22 in the 119th Congress, is the primary legislative response to concerns about noncitizen voting. You can read the full text of the bill to understand its specific provisions.
The bill would require documentary proof of U.S. citizenship to register to vote in federal elections. Acceptable documents would include a U.S. passport, a birth certificate, or a naturalization certificate.
The Case For
Proponents argue that the current system, which relies on a signed attestation of citizenship, is based on the honor system. They argue that requiring proof of citizenship is a common-sense verification step, similar to showing ID for other important transactions. If noncitizens are not voting, they argue, then the requirement will not burden anyone who should not be burdened. Supporters also point out that several democracies around the world require citizenship documentation to register.
The Case Against
Critics point to a specific number: an estimated 21 million U.S. citizens lack readily available documentary proof of citizenship. These are people who were born in the United States and are full citizens, but who do not currently possess a valid passport, birth certificate, or naturalization certificate.
This group is not randomly distributed across the population. Citizens without readily available documentation are disproportionately:
- Elderly, particularly those born in rural areas or at home before hospital births were standard
- Low-income, because passports cost money and replacing lost birth certificates requires fees and bureaucratic navigation
- Minority communities, reflecting historical disparities in access to documentation
- Rural residents, who may have been born in areas with less formal record-keeping
Critics argue that requiring documentation to solve an extremely rare problem would create a much larger problem: preventing millions of eligible citizens from registering to vote. They frame this as a tradeoff where the cure is worse than the disease.
The Real Tradeoff
This is where honest analysis requires acknowledging that both sides have a point, and the disagreement is about how to weigh competing risks.
On one side: any noncitizen voting in a federal election is a violation of law, and preventing it is a legitimate goal. On the other: creating barriers that prevent eligible citizens from voting is also a serious problem in a democracy.
The question is proportionality. The documented rate of noncitizen voting is extremely low. The number of citizens who could face barriers under a proof-of-citizenship requirement is much larger. Whether you think the SAVE Act is good policy depends partly on which risk you weigh more heavily.
The bill is currently working its way through Congress. Govbase tracks its status and the positions of your representatives on the issue.
Why the Perception Gap Exists
There is a significant gap between what the data shows and what many Americans believe. Polls consistently find that a substantial portion of the public thinks noncitizen voting is a widespread problem. This perception exists despite decades of data showing the opposite.
Several factors contribute to this gap.
High-profile political rhetoric. Noncitizen voting has become a central talking point in elections. When political leaders repeatedly describe something as a crisis, people tend to believe it is one. This is true across the political spectrum on different issues.
Individual cases get amplified. When a case of noncitizen voting is discovered, it gets enormous attention. When an audit finds that 99.999% of votes were cast by eligible citizens, that does not generate the same coverage. "System works as designed" is not a compelling headline.
Absence of evidence is hard to communicate. It is much easier to point to a specific case and say "this happened" than to point to a dataset and say "this almost never happens." Human brains are wired to respond to vivid examples more than to statistics. A story about one noncitizen who voted lands harder than a study showing 30 cases out of 23.5 million.
Social media dynamics. Individual cases get shared thousands of times. Corrections and context get shared far less. The information ecosystem favors alarming claims over reassuring data.
Distrust compounds. Once someone believes election systems are compromised, evidence that the systems are working can look like evidence of a cover-up. This creates a cycle that is very hard to break with data alone.
None of this means people who are concerned about election integrity are wrong to care about the issue. Election security matters. The integrity of the vote is fundamental to democracy. The point is that on this specific question, the data is quite clear, and the gap between perception and reality is driven by factors that have little to do with the evidence.
What Both Sides Actually Agree On
The debate over noncitizen voting is heated, but the areas of agreement are broader than the argument suggests.
Noncitizens should not vote in federal elections. There is no serious political movement to allow this. Both parties agree on this principle. The disagreement is not about whether it should be illegal. It already is.
Election integrity matters. Nobody is arguing that elections should not be secure. Both sides want elections that are accurate and trustworthy. They disagree about what threats are most serious and what solutions make sense.
Verification systems should work. The debate is about what kind of verification is appropriate, not about whether verification should exist. The current system includes registration checks, attestation under penalty of perjury, and post-election audits. The SAVE Act would add documentary proof on top of these existing measures.
The actual disagreement is narrower than it appears: given the evidence about how often noncitizen voting occurs, do the proposed solutions create more problems than they solve? That is a legitimate policy question, and reasonable people can weigh the tradeoffs differently.
What the Data Tells Us About the System
Step back from the political debate and look at what the evidence actually shows about the existing election infrastructure.
The current system catches noncitizens who attempt to register. Georgia's audit found 1,634 blocked registration attempts and zero successful votes. Texas's review found that its flagging system generated far more false positives than actual cases. States that have conducted audits consistently find that their verification processes are working.
Post-election audits, which compare voter rolls against citizenship databases, consistently turn up tiny numbers. When they do find noncitizen registrations, they are often the result of administrative errors at motor vehicle offices rather than intentional fraud.
The penalty structure creates a powerful deterrent. For people who are in the country without legal status, voting is one of the highest-risk activities imaginable. For legal residents working toward citizenship, it would destroy the goal they are working toward.
None of this means the system is perfect. No system involving millions of transactions is. But the evidence across multiple studies, conducted by organizations spanning the ideological spectrum, points in the same direction: noncitizen voting in federal elections is not occurring at a scale that affects outcomes.
Following This Issue
The SAVE Act and broader election integrity legislation will continue to move through Congress. How the bill progresses, what amendments are added, and how the debate evolves are all worth tracking.
Govbase monitors the SAVE Act's legislative status, committee actions, and floor votes in real time. If election integrity is an issue you care about, and it is an issue worth caring about regardless of where you fall on the policy specifics, you can follow the legislation and see how your representatives are engaging with it.
The evidence on noncitizen voting is not ambiguous. The policy response to that evidence is where the real debate lives. Understanding the data is the first step to having a productive conversation about what to do next.