Is voter fraud a problem in the US?

NoSays it is contested

No, voter fraud is not a significant or systemic problem in US elections. It occurs in isolated instances, is illegal, and is prosecuted when detected, but extensive data from multiple sources shows it is exceedingly rare—far too infrequent to affect election outcomes at scale. Rates are typically measured in fractions of a percent (often 0.0001% or lower), making it rarer than many everyday risks. Claims of widespread fraud, particularly around recent presidential elections, have consistently lacked supporting evidence after audits, recounts, investigations, and court reviews.[1]

Key Data and Studies

•Heritage Foundation Election Fraud Database (conservative-leaning think tank): This tracks a sampling of proven cases (convictions, judicial findings, etc.) totaling around 1,500–1,600 instances spanning decades (roughly 1982–present, across all election types and levels). It includes categories like ineligible voting (e.g., noncitizens, felons, deceased), double voting, absentee ballot misuse, vote buying, and rare impersonation. Even this database—intended to highlight vulnerabilities—yields tiny rates when contextualized against total ballots cast (hundreds of millions to over a billion in federal elections alone over the period). For example, analyses of the data show rates like ~0.0000845% in Arizona over 25 years (36 cases amid ~42.6 million ballots) or similarly negligible figures in Pennsylvania and other swing states. No major federal election outcomes have been overturned due to such fraud. Heritage notes that small amounts could matter in extremely close local races and argues for stronger safeguards, but the numbers do not support systemic issues.[2][3][4]

•Academic and other analyses: Justin Levitt (Loyola Law School) documented only about 31 credible incidents of in-person voter impersonation (the type most relevant to photo ID debates) out of more than 1 billion ballots cast from 2000–2014. Later updates remained similarly low. The Brennan Center for Justice (progressive-leaning) and others describe fraud as "infinitesimally rare," with most allegations turning out to be clerical errors, data mismatches, or non-fraud irregularities rather than intentional misconduct. Noncitizen voting is "vanishingly rare" per multiple studies; one Brennan review of 23.5 million votes in 2016 found ~30 suspected incidents. Mail/absentee fraud rates are also tiny (e.g., ~0.000043% or about 4 cases per 10 million mail ballots in recent cycles).[5][6][7]

•2020 election specifically: An Associated Press investigation across six disputed swing states found fewer than 475 potential fraud cases out of more than 25 million ballots. Biden’s combined margin there was over 311,000 votes. Cases involved individuals (not organized schemes), were bipartisan, and often did not even result in counted ballots. Extensive audits, hand recounts (e.g., Georgia), lawsuits (nearly all dismissed for lack of evidence), and statements from officials of both parties, the DOJ (under Trump and Biden), and cybersecurity experts found no widespread fraud or outcome-altering issues. Similar patterns held in other cycles.[8][1]

•Prosecutions and recent efforts: Convictions number in the low hundreds over multi-year periods covering hundreds of millions of votes (e.g., ~300 around 2016–2020 elections). Recent DOJ and state efforts (including under Trump administrations) have charged noncitizens and others, but numbers remain small relative to scale (dozens of charges, not thousands of impactful illegal votes). State secretaries of state (including Republicans) routinely note that fraud is "exceedingly rare" even while referring cases.[9][10]

Brookings Institution analyses of Heritage data and other reviews reach the same conclusion: fraud is real but miniscule and has not altered major outcomes.[11]

Context and Caveats

Isolated fraud does happen and has overturned some local elections historically (e.g., certain mayoral or city council races involving absentee ballot schemes or vote-buying). Vulnerabilities exist in voter rolls (outdated registrations), absentee processes, or registration drives, which is why measures like voter ID (in most states), signature verification, risk-limiting audits, interstate crosschecks, and citizenship verification are used or debated. Deterrence works partly because fraud is a serious crime with high detection risk relative to the (usually one) vote gained, plus penalties including fines, jail, and (for noncitizens) deportation.

However, equating rare individual crimes with a "problem" that undermines overall election integrity or justifies broad claims of stolen elections is not supported by the evidence. Administrative errors, rejected ballots, or procedural issues occur more often than intentional fraud but are also managed through existing systems. Bipartisan election officials, courts, and statistical audits consistently affirm the security and accuracy of US elections.

In short, voter fraud warrants vigilance and prosecution, but data from conservative databases, progressive researchers, AP investigations, and official reviews show it is not a widespread or outcome-determinative problem. Exaggerated claims risk eroding public trust more than the fraud itself. For the most current figures, check sources like the Heritage database, state election offices, or DOJ reports directly, as numbers of documented cases slowly accumulate while ballot volumes remain enormous.

Citations (11)

1npr.orgNews and mediaquality 0.93
https://www.npr.org/2024/10/11/nx-s1-5147732/voter-fraud-explainer
Backs: “Claims of widespread fraud, particularly around recent presidential elections, have consistently lacked supporting evidence after audits, recounts, investigations, and court reviews.”
evidence, article, points to no. Claim: Audits, recounts, lawsuits and officials of both parties found no widespread fraud or outcome-altering issues in 2020.
2ballotpedia.orgEncyclopedias and referencequality 0.89
https://ballotpedia.org/Electoral_fraud
Backs: “Heritage notes that small amounts could matter in extremely close local races and argues for stronger safeguards, but the numbers do not support systemic issues.”
background, reference entry, neutral. Claim: Heritage database tracks proven fraud cases; contextualized against total ballots, rates are tiny and no major federal outcomes were overturned.
3newsweek.comNews and mediaquality 0.82
https://www.newsweek.com/voter-fraud-us-elections-data-research-1978342
Backs: “Heritage notes that small amounts could matter in extremely close local races and argues for stronger safeguards, but the numbers do not support systemic issues.”
statistic, article, points to no. Claim: Rates like ~0.0000845% in Arizona over 25 years (36 cases amid ~42.6 million ballots) from Heritage data.
4heritage.orgNonprofits and think tanks
https://www.heritage.org/press/heritage-releases-updated-election-fraud-database-featuring-new-fraud-map
Backs: “Heritage notes that small amounts could matter in extremely close local races and argues for stronger safeguards, but the numbers do not support systemic issues.”
attribution, official document, neutral. Claim: Heritage's database of about 1,500-1,600 proven cases; Heritage argues for stronger safeguards in close local races.
5washingtonpost.comNews and mediaquality 0.82
https://www.washingtonpost.com/news/wonk/wp/2014/08/06/a-comprehensive-investigation-of-voter-impersonation-finds-31-credible-incidents-out-of-one-billion-ballots-cast/
Backs: “Mail/absentee fraud rates are also tiny (e.g., ~0.000043% or about 4 cases per 10 million mail ballots in recent cycles).”
statistic, article, points to no. Claim: Levitt found about 31 credible in-person impersonation incidents out of over 1 billion ballots from 2000-2014.
6brennancenter.orgNonprofits and think tanks
https://www.brennancenter.org/topics/voting-elections/vote-suppression/myth-voter-fraud
Backs: “Mail/absentee fraud rates are also tiny (e.g., ~0.000043% or about 4 cases per 10 million mail ballots in recent cycles).”
evidence, article, points to no. Claim: Brennan Center describes fraud as infinitesimally rare; noncitizen voting vanishingly rare, ~30 suspected incidents among 23.5 million votes in 2016.
7brookings.eduNonprofits and think tanks
https://brookings.edu/articles/mail-voting-in-the-us-data-points-to-very-low-fraud-and-significant-benefits-to-voters
Backs: “Mail/absentee fraud rates are also tiny (e.g., ~0.000043% or about 4 cases per 10 million mail ballots in recent cycles).”
statistic, article, points to no. Claim: Mail/absentee fraud rates are tiny, about 4 cases per 10 million mail ballots.
8apnews.comNews and mediaquality 1.00
https://apnews.com/article/joe-biden-pennsylvania-arizona-nevada-wisconsin-de171efb89404a8bdefb6fb83a0b4ad1
Backs: “Extensive audits, hand recounts (e.g., Georgia), lawsuits (nearly all dismissed for lack of evidence), and statements from officials of both parties, the DOJ (under Trump and Biden), and cybersecurity experts found no widespread fraud or outcome-altering issues. Similar patterns held in other cycles.”
statistic, article, points to no. Claim: AP found fewer than 475 potential fraud cases among over 25 million ballots in six swing states, far below Biden's 311,000 margin.
9amarkfoundation.orgNonprofits and think tanks
https://archive.amarkfoundation.org/the2020election/voter-fraud-convictions-since-2016/
Backs: “State secretaries of state (including Republicans) routinely note that fraud is "exceedingly rare" even while referring cases.”
statistic, data page, points to no. Claim: Convictions number in the low hundreds (~300) around 2016-2020 elections covering hundreds of millions of votes.
10dailysignal.comNews and mediaquality 0.47
https://www.dailysignal.com/2026/09/06/trump-doj-charges-nearly-four-times-more-noncitizens-in-election-fraud-cases-than-biden-doj/
Backs: “State secretaries of state (including Republicans) routinely note that fraud is "exceedingly rare" even while referring cases.”
evidence, article, neutral. Claim: Recent DOJ efforts charged noncitizens, but numbers remain small relative to scale.
11archive.orgNonprofits and think tanks
https://web.archive.org/web/20250406020938/https://www.brookings.edu/articles/how-widespread-is-election-fraud-in-the-united-states-not-very/
Backs: “Brookings Institution analyses of Heritage data and other reviews reach the same conclusion: fraud is real but miniscule and has not altered major outcomes.”
evidence, article, points to no. Claim: Brookings analyses of Heritage data conclude fraud is real but minuscule and has not altered major outcomes.