Is voter fraud a problem in the US?

Is Voter Fraud a Problem in the US?

The overwhelming consensus from academic research, nonpartisan institutions, and even conservative-leaning databases is that voter fraud occurs, but it is extremely rare and not a widespread or systemic problem that affects election outcomes. There is a notable gap between the actual incidence of fraud and public perception of it.

What the Data Shows

Extremely low documented rates. Multiple independent studies converge on similar findings. Research conducted during the past two decades has concluded that there is no evidence of significant election fraud in contemporary American national elections.

The Brennan Center's foundational research found that most reported incidents of voter fraud are actually traceable to other sources, such as clerical errors or bad data matching practices, and incident rates fell between 0.0003 percent and 0.0025 percent. The report famously concluded it is more likely that an American "will be struck by lightning than that he will impersonate another voter at the polls."

Even using conservative, Republican-aligned sources, rates are tiny. A 2024 Brookings analysis reviewed the Heritage Foundation's own "Election Fraud Database"—compiled by a conservative think tank skeptical of election security—and still found vanishingly small numbers. In Arizona, 42.6 million votes were cast with only 36 cases of fraud in total, while Pennsylvania had over 100 million votes cast across 32 elections over 30 years with only around 39 cases of voter fraud. Brookings scholar Elaine Kamarck put it directly: American election fraud is "miniscule," and the share of reported cases over 13 to 38 years across several key swing states is less than 1 percent. In Pennsylvania specifically, that amounted to just 39 cases of voter fraud identified out of more than 100 million ballots cast—0.000039 percent.

Studies specifically on impersonation fraud (the type voter ID laws target) find almost nothing. A comprehensive review found 31 credible instances of impersonation fraud from 2000 to 2014, out of more than 1 billion ballots cast—and even this tiny number is likely inflated since it counted any credible claim, not just prosecutions or convictions. Similarly, academic election-forensics research using survey list experiments on the high-stakes 2012 election concluded researchers found no evidence of systematic voter fraud or vote buying, which they found particularly encouraging given the closeness and high stakes of the election.

Mail-in voting specifically. Despite frequent claims that mail voting is especially vulnerable, research comparing states before and after adopting vote-by-mail found VBM states experienced a 57% decline in fraud cases per million voters, compared to a 36% decline in non-VBM states, suggesting no meaningful fraud increase from mail voting.

The Perception Gap

Despite this evidence, a large share of Americans believe fraud is common. A 2024 national survey of registered voters found substantial proportions of U.S. registered voters believe these types of election fraud are common, with partisanship correlating strongly with endorsement of fraud claims—Republicans were consistently more likely to state that types of election fraud are common, even controlling for other factors. Researchers also found conspiratorial thinking is strongly correlated with belief in the occurrence of types of election fraud, even when controlling for partisanship.

This gap matters because, as researchers note, when prominent political elites continue to press false claims of election fraud, millions of American voters may continue to believe election fraud exists, which might affect their confidence in election administration and their willingness to participate in the electoral process.

Bottom Line

Voter fraud exists in isolated, individual cases—typically involving confused voters, clerical/data-matching errors, or rare deliberate acts by individuals—but it is not a systemic problem that threatens election integrity at scale. The real-world rates documented across multiple independent studies, using different methodologies and even partisan-leaning data sources, consistently land in the range of a few cases per million or tens-of-millions of votes cast—nowhere near a level capable of altering election outcomes.