Noncitizen voters in the US 2026 are rare in every completed state audit, and no official national total exists. Five audits found 358 confirmed noncitizen registrants and 62 who voted among about 13.5 million voters, rising to 402 voters if New Jersey’s 2026-disclosed coding error is included. Georgia found 20 among 8.2 million registrations, Utah confirmed 27 among 2.1 million, and Iowa confirmed 277 among 2.3 million, of whom 35 voted in 2024. Federal data point the same way, with 41 noncitizens charged between January 2025 and May 2026, while larger federal lists, such as 20,000 “likely” noncitizens in California, remain unverified.
Noncitizen Voters in US 2026 – Introduction
Noncitizen voters in the US 2026 are the subject of the most detailed set of voter-roll audits in recent years. States including Georgia, Utah, Iowa, Montana, Michigan and Arizona have checked registration lists against citizenship records, and the results are public. In most cases the number of confirmed noncitizens is a small fraction of 1% of registered voters, and the number who actually cast ballots is smaller still.
The numbers are not identical everywhere, and they are disputed. Iowa found 277 confirmed noncitizen registrants, a rate about 50 times higher than Georgia’s, and New Jersey disclosed a coding error that wrongly registered about 6,600 people. Federal officials have also circulated much larger lists that states say are unreliable. This report sets out the audit results, voting rates, long-term data, federal lists, and the laws and bills that now shape the debate. Figures run through October 30, 2026, and claims that are not confirmed are labeled as claims.
Interesting Facts About Noncitizen Voters in US 2026
TOTAL NONCITIZEN VOTERS, DOCUMENTED (1 block = 10 people)
5 audits + New Jersey error, voted | ████████████████████████████████████████ 402
Heritage database, 1999 to 2023 | ███████▌ 77
5 audits, voted | ██████▏ 62
Federal charges, Jan 2025-May 2026 | ████ 41
NONCITIZENS FOUND ON VOTER ROLLS, BY STATE AUDIT (1 block = 10 people)
Iowa, confirmed registered | ████████████████████████████ 277
Yavapai County AZ, registered | ███ 30
Utah, confirmed registered | ███ 27
Montana, possible (unconfirmed) | ██ 23
Georgia 2024, registered | ██ 20
Macomb County MI, apparent | ▌ 4
| Fact | Confirmed Data |
|---|---|
| Official national total | None published; no federal or state agency tracks one |
| Total noncitizen registrants confirmed, 5 audits | 358 among about 13.5 million voters, about 0.0026% |
| Total noncitizens who voted, 5 audits | 62 (Iowa 35, Utah 13, Georgia 9, Yavapai 4, Macomb 1) |
| Total including New Jersey’s coding error | About 6,958 registrants and 402 voters among about 20.2 million |
| Heritage database, 1999 to 2023 | 77 instances, per the Bipartisan Policy Center |
| Georgia 2024 audit | 20 noncitizens among 8.2 million registrations; 9 had voted |
| Utah audit, April 2025 to May 2026 | 27 confirmed among 2.1 million voters; 99.72% confirmed citizens |
| Iowa audit, 2025 | 277 confirmed registered; 35 voted in 2024; 5 attempted to vote |
| Yavapai County, Arizona | 30 registered among about 200,000 voters; 4 had voted |
| Macomb County, Michigan | 4 “apparent noncitizens” among about 725,000 voters; 1 voted |
| Federal charges, Jan 2025 to May 2026 | 41 noncitizens charged with illegal voting or related crimes |
| Federal law | 18 U.S.C. § 611, up to 1 year in prison |
Source: Georgia Secretary of State, October 2024; Utah Lieutenant Governor’s Office, May 2026; Iowa Secretary of State and Attorney General, 2025; Cato Institute testimony, September 15, 2026; ProPublica; Michigan Secretary of State, January 2026.
Added together, the five audits with confirmed counts found 358 noncitizen registrants and 62 who voted, among about 13.5 million registered voters. Including New Jersey’s coding error, the voter count rises to 402. These totals are a tally of published audits, not an official national figure, because none exists.
The chart shows that the audits with a confirmed count land between 4 and 277 people. Iowa is the highest, and it also reported that 35 noncitizens voted successfully in the 2024 election. Georgia, Utah and Arizona’s Yavapai County found between 20 and 30 each. Montana’s 23 are possible cases, not confirmed ones.
The table shows the second pattern: most noncitizens found on the rolls never voted. Georgia’s 9 voters were 45% of its 20 registrants, Iowa’s 35 were about 13% of its 277, and Yavapai’s 4 were about 13% of its 30. Registering is not the same as voting, so the audits count the two separately.
Total Number of Noncitizen Voters in the US 2026
DOCUMENTED NONCITIZEN VOTERS, BY MEASURE (1 block = 10 people)
Audits incl. New Jersey error (voted) | ████████████████████████████████████████ 402
Heritage database, 1999 to 2023 (BPC) | ███████▌ 77
Audits excl. New Jersey error (voted) | ██████▏ 62
Federal charges, Jan 2025 to May 2026 | ████ 41
| Measure | Total |
|---|---|
| Official national total | None published; no federal or state agency tracks one |
| Noncitizen registrants confirmed in 5 audits (IA, UT, GA, Yavapai AZ, Macomb MI) | 358 among about 13.5 million voters, about 0.0026% |
| Noncitizens who voted in those 5 audits | 62 (Iowa 35, Utah 13, Georgia 9, Yavapai 4, Macomb 1) |
| Adding New Jersey’s coding error | About 6,958 registrants and 402 voters among about 20.2 million, about 0.034% and 0.002% |
| Heritage database, 1999 to 2023 | 77 instances, per the Bipartisan Policy Center |
| Federal charges, Jan 2025 to May 2026 | 41 noncitizens |
| Possible, unconfirmed (Montana) | 23 possible registrants |
Source: Totals calculated by this report from the state audit figures above (Iowa Secretary of State; Utah Lieutenant Governor; Georgia Secretary of State; Cato Institute, September 2026; NJ Spotlight News, August 2026); Bipartisan Policy Center, 2026; ProPublica.
There is no official total for the number of noncitizen voters in the US. The figures above are a tally of published audit counts, and they cover only the places that audited, so they are a floor for those places, not a national estimate. The five audits that confirmed noncitizens found 62 who voted, and the Heritage database lists 77 instances over 24 years. New Jersey’s coding error raises the audit total to 402, which is why the table shows both versions.
The tally should be read with care. The audits cover different years and methods, some counts overlap with statewide reviews (Michigan’s statewide comparison reported about 15 to 16 votes in 2024), and Iowa’s Secretary of State also cited 87 earlier-election votes that are not in the 62. Supporters of tighter rules say audits cannot catch everyone, and critics say the totals show the problem is rare. Neither side can point to a verified national figure above these counts.
Noncitizen Voter Rates by State: How Common Is It?
NONCITIZENS PER MILLION REGISTERED VOTERS (1 block = 10 per million)
Yavapai County AZ | ███████████████ 150
Iowa | ████████████ 120
Montana (possible) | ███ 29
Utah | █ 13
Macomb County MI | ▌ 5.5
Georgia | ▌ 2.4
| State or County | Registered Voters | Noncitizens Found | Rate |
|---|---|---|---|
| Georgia (2024) | 8.2 million | 20 | About 0.00024% |
| Utah (2026) | 2.1 million | 27 confirmed | About 0.0013% |
| Iowa (2025) | 2.3 million | 277 confirmed | About 0.012% |
| Montana (Jan 2026) | About 785,000 | 23 possible | About 0.003% |
| Yavapai County, AZ | About 200,000 | 30 | About 0.015% |
| Macomb County, MI | About 725,000 | 4 apparent | About 0.0006% |
Source: Georgia Secretary of State; Utah Lieutenant Governor; Iowa Secretary of State; Daily Inter Lake, January 14, 2026; Cato Institute, September 2026. Rates calculated from published counts.
Every rate in the table is below 0.02% of registered voters, meaning fewer than 2 in 10,000. The spread between 2.4 per million in Georgia and 150 per million in Yavapai County comes partly from method. Georgia checked its whole list against federal citizenship data, while county audits and driver’s-license matches can catch different people.
A rate is not a verdict. Supporters of tighter rules say that any noncitizen vote is too many, and that audits can miss people who were never checked. Critics of the federal effort say the same audits show the problem is rare and that flagged lists overcount. Utah’s audit also left 5,007 voters whose citizenship could not be confirmed either way, which is why officials planned follow-up letters.
How Many Noncitizens Actually Voted? Ballots Cast in 2024 and Earlier
NONCITIZEN VOTERS IDENTIFIED BY AUDIT (1 block = 10 people)
New Jersey, coding error (voted once+) | ██████████████████████████████████ 340
Iowa, 2024 election | ███▌ 35
Utah, certain elections | █▎ 13
Georgia, prior elections | ▉ 9
Yavapai County AZ | ▌ 4
Macomb County MI | ▏ 1
| Source | Noncitizens Who Voted | Share of Noncitizen Registrants |
|---|---|---|
| Iowa, 2024 election | 35 | About 13% of 277 |
| Utah, certain elections | 13 | About 48% of 27 |
| Georgia, prior elections | 9 | 45% of 20 |
| Yavapai County, AZ | 4 | About 13% of 30 |
| Macomb County, MI | 1 | 25% of 4 |
| New Jersey, coding error | 340 | About 5% of 6,600 |
Source: Iowa Attorney General, 2025; KSL News Radio, May 28, 2026; Georgia Secretary of State, October 2024; Cato Institute, September 2026; NJ Spotlight News, August 2026.
Votes are the most important number because they are the ones that can affect an election. In Iowa, 35 votes were cast out of the millions in the 2024 election. In Georgia, 9 of the 20 noncitizens had voted, mostly before the state added a citizenship check in 2012. Utah reported that 13 confirmed noncitizens cast ballots in certain elections, though all 27 came forward during the audit and asked to be removed.
Counts of votes differ from counts of registrants, and some figures overlap in time. Iowa’s Secretary of State also said 87 noncitizens had voted in previous state elections, in addition to the 35 in 2024. Michigan’s statewide comparison of motor vehicle and voter files reported about 15 to 16 noncitizen votes in the 2024 general election, according to published summaries. These are the highest documented figures for single states.
New Jersey Coding Error: The Largest Single Count
NEW JERSEY IMPROPER REGISTRATIONS (1 block = 500 people)
Improperly registered (coding error) | █████████████ 6,600
Voted in at least one election | ▌ 340
| Measure | Figure |
|---|---|
| People improperly registered | About 6,600 |
| Voted in at least one election | 340 |
| Registered voters in New Jersey | 6.68 million |
| Registrants share of voters | About 0.1% |
| Voters share of registered voters | About 0.005% |
| Cause | A coding mistake, per reporting on the state’s disclosure |
Source: NJ Spotlight News, August 2026; Cato Institute testimony, September 15, 2026; The Daily Signal, August 11, 2026.
New Jersey is the biggest single count in the public record, and it came from a system error, not an audit of fraud. About 6,600 people were improperly registered because of a coding mistake, and 340 of them voted in at least one election. The Daily Signal reports the figure as about 400 noncitizens who voted in state elections, which shows how totals differ by source and period.
Both sides use the New Jersey case. Critics of the SAVE America Act say the 340 voters equal about 0.005% of 6.68 million registered voters and came from a clerical bug that tighter documents would not have caught at registration. Supporters say the case shows that noncitizens can end up on the rolls and vote, which is their argument for stronger checks. The state said it would disclose records on how foreign nationals registered.
Long-Term Noncitizen Voting Data: Studies, Databases and Prosecutions
DOCUMENTED CASES OVER TIME (1 block = 10 cases)
Georgia "possible" noncitizens, 30 yrs | ████████████████████████████████████████ 407
Ohio flagged, 2019 to Sept 2023 | ████████████████████████████████████████████████████ 521
Heritage database, 1999 to 2023 (BPC) | ███████▌ 77
Federal charges, Jan 2025 to May 2026 | ████ 41
Convictions over 10 years (Heritage) | ████ 40
Brennan Center, 2016 election | ███ 30
| Source | Finding |
|---|---|
| Brennan Center, 2016 election, 42 jurisdictions | About 30 suspected cases among 23.5 million votes, about 0.0001% |
| Bipartisan Policy Center review of the Heritage database | 77 noncitizen voting instances from 1999 to 2023 |
| Heritage Foundation, 10-year view | 40 noncitizens convicted of ineligible voting across several states |
| Ohio, 2019 to September 2023 | 521 potential cases flagged; one charge found by Ohio Capital Journal review |
| Georgia Department of Audits, September 2026 | 407 possible noncitizen voters over 30 years, unconfirmed |
| Federal charges, Jan 2025 to May 2026 | 41 noncitizens, per ProPublica |
Source: Brennan Center for Justice; Bipartisan Policy Center, 2026; The Daily Signal, August 11, 2026; Ohio Capital Journal; Cato Institute, September 2026; ProPublica.
Long-run numbers are small too. The Bipartisan Policy Center found 77 instances of noncitizen voting in the Heritage Foundation’s election fraud database over 24 years, and each was investigated. The Heritage Foundation’s own count is 40 convictions in a decade. Ohio’s 521 flags led to very few charges, which shows the gap between a flag and a case.
The flagged lists are larger because they include possible matches. Georgia’s 407 are possible, not confirmed, and span 30 years, so they did not all vote in the same election. The federal government stepped up enforcement in 2025 and 2026, and 41 charges in about 17 months is the most recent national figure available. Critics say it shows how rare the offense is, while supporters say prosecutions are still rising.
Federal Noncitizen Lists in 2026: SAVE Flags, California and Washington
FEDERAL AND STATE LISTS OF "LIKELY" NONCITIZENS (1 block = 500 people)
DHS July letter, "as many as," California | ██████████████████████████████████████ 190,832
DOJ California list, October 13 | ████████████████████████████████████████ 20,000
SAVE flags, 7 states, 35M voters | ████████ about 4,200
LA County addresses on DOJ list | ██████████████ 7,035
Washington State list | ▌ 400
| List | Figure | Status |
|---|---|---|
| SAVE checks, 7 states, about 35 million voters | About 4,200 flagged, about 0.01% | DHS corrected data for at least 5 states |
| DHS letter to California, July 16 | “As many as” 190,832 | Disputed by California |
| DOJ California list, October 13 | More than 20,000 “likely” noncitizens | Only 7,035 have Los Angeles County addresses |
| DOJ Washington State list | 400 registrants; alleged 1,500 votes in past elections | Not confirmed by Washington |
| Court ruling, October 26 | Ninth Circuit dismissed DOJ’s emergency appeal | Procedural |
| Court ruling, October 28 | District court denied request to segregate ballots | Election continues |
Source: Texas Tribune, February 13, 2026; Bipartisan Policy Center, 2026; Ninth Circuit opinion, No. 26-7070; CNN, October 28, 2026; NBC News, October 29, 2026.
Federal lists are much larger than audit counts, and they are unverified. SAVE flagged about 4,200 people in seven states, or about 0.01% of roughly 35 million voters, and DHS later corrected data after the tool misidentified some voters. Los Angeles County said only 7,035 of the 20,000 California names have county addresses, and in a county sample of 200 records with a birthplace, 38.9% indicated birth in the United States.
The courts have so far ruled on timing, not on the lists’ accuracy. The Ninth Circuit said on October 26 that it lacked jurisdiction because the government gave the district court only about 15 hours to rule, and the district court declined on October 28 to order the ballots set aside. DOJ has opened a criminal investigation and sent subpoenas to election officials, which means the numbers may change before the November 3 election.
Noncitizen Voting Law and Policy in 2026: SAVE America Act and State Rules
SAVE AMERICA ACT HOUSE VOTE, FEBRUARY 11, 2026 (1 block = 10 votes)
Yes | █████████████████████▌ 218
No | █████████████████████▏ 213
| Rule or Measure | Detail |
|---|---|
| 18 U.S.C. § 611 | Unlawful for a noncitizen to vote in federal elections; up to 1 year in prison, plus fines |
| SAVE America Act | Passed the House 218-213 on February 11, 2026; needs 60 votes in the Senate |
| Requirement under the bill | Documentary proof of citizenship to register for federal elections, plus photo ID to vote |
| Florida | Governor signed a proof-of-citizenship law in March 2026; main provisions start in 2027 |
| South Dakota and Utah | Proof-of-citizenship bills take effect before the 2026 midterms |
| Public support | 83% backed proof of citizenship for first-time registrants (Gallup, October 2024) |
| Citizens without readily available proof | More than 20 million, per voting experts |
Source: 18 U.S.C. § 611; CBS News; NPR, March 2026; Voting Rights Lab; Gallup, October 2024; NAWBO SAVE America Act FAQ.
Everyone agrees that noncitizens may not vote in federal elections. The dispute is about the fix. Supporters of the SAVE America Act say proof of citizenship is needed to maintain election security, and polling shows broad support for the idea. Opponents say it would make registration harder for millions of citizens, noting that almost half of Americans do not have a passport and that many licenses do not show citizenship.
The bill’s path is narrow. It passed the House by 5 votes, and the Senate needs at least 7 Democratic votes to reach 60, so passage this year looks unlikely. States are acting on their own, and the Center for Election Innovation and Research found in 2025 that 18 states, across the political spectrum, had not publicly disclosed finding or removing any noncitizen voters.
Frequently Asked Questions About Noncitizen Voters in US 2026
What is the total number of noncitizen voters in the US?
No official national total exists. Five state and county audits found 62 noncitizens who voted, and adding New Jersey’s coding error brings the documented count to 402. The Heritage database lists 77 instances from 1999 to 2023.
How many noncitizens vote in US elections?
Audits find very few. Iowa identified 35 noncitizens who voted in 2024, Georgia found 9 who had ever voted among 20 registered noncitizens, and Utah reported 13 in certain elections.
What percentage of registered voters are noncitizens?
Completed audits put it below 0.02%. Georgia’s rate was about 0.00024% and Utah’s about 0.0013%, while Iowa’s was about 0.012%.
How many noncitizens did Georgia find on its voter rolls?
20 out of 8.2 million registered voters in its 2024 audit, and 9 of them had voted, mostly before a citizenship check was added in 2012.
How many noncitizens did Utah find?
27 confirmed among about 2.1 million voters, and 99.72% of the rolls were confirmed as citizens. Another 5,007 voters could not be confirmed either way.
How many noncitizens voted in Iowa in 2024?
35, according to Iowa’s Secretary of State, out of 277 confirmed noncitizen registrants. 5 more attempted to vote.
What happened in New Jersey?
A coding error improperly registered about 6,600 people, and 340 voted in at least one election, out of 6.68 million registered voters.
How many noncitizens have been charged with illegal voting?
41 were charged between January 2025 and May 2026, according to ProPublica. The Heritage Foundation counts 40 convictions over a decade.
What is SAVE?
DHS’s Systematic Alien Verification for Entitlements database, used by states to check citizenship. It flagged about 4,200 people in seven states, and DHS corrected some errors.
What is the 20,000 list in California?
A Justice Department list of “likely” noncitizens registered in California. Los Angeles County says only 7,035 names have county addresses, and courts have not ruled on its accuracy.
Is it a crime for a noncitizen to vote?
Yes. 18 U.S.C. § 611 makes it unlawful in federal elections, with up to 1 year in prison, and deportation can follow.
Did the SAVE America Act become law?
No. It passed the House 218-213 on February 11, 2026 and is waiting on the Senate, where it needs 60 votes.
When is the election?
Tuesday, November 3, 2026.
Disclaimer: The data research report we present here is based on information found from various sources. We are not liable for any financial loss, errors, or damages of any kind that may result from the use of the information herein. We acknowledge that though we try to report accurately, we cannot verify the absolute facts of everything that has been represented.
