Is voter fraud a problem in the US?

Based on extensive research from academics, government investigations, and think tanks across the political spectrum, voter fraud in the U.S. is documented to occur, but at extremely low rates that make it a minor issue rather than a widespread problem affecting election outcomes.

What the research shows

Academic and government studies consistently find fraud rates are minuscule:

Even conservative-leaning data confirms this pattern: The Heritage Foundation's own database — often cited by those concerned about fraud — found 1,561 instances of voter fraud that resulted in 1,325 convictions across decades. When Brookings analyzed this data, they found that in Arizona, around 42.6 million votes cast with only 36 cases of fraud in total over 25 years, and in Pennsylvania, over 100 million votes cast in 32 elections across 30 years and only around 39 cases of voter fraud.

Other independent reviews reach similar conclusions: The investigative mapping for federal cycles found roughly one fraudulent ballot per 2.1 million ballots cast across 24 years—an incidence far below thresholds that could plausibly flip statewide or national outcomes.

The gap between belief and evidence

There's a notable disconnect between documented reality and public perception. Belief in widespread fraud remains common among the public even as empirical studies find it rare; survey research in 2024 documents persistent public concern despite decades of research showing no significant scale of fraud. Research also shows conspiratorial thinking is strongly correlated with belief in the occurrence of types of election fraud, even when controlling for partisanship.

Bottom line

Voter fraud does happen — isolated cases are prosecuted almost every election cycle, involving things like false registrations, absentee ballot schemes, or ineligible voting. However, the overwhelming weight of evidence from studies spanning multiple decades, political viewpoints, and methodologies shows it occurs at rates measured in millionths or ten-thousandths of a percent of ballots cast — nowhere near the scale needed to be considered a systemic problem or to alter election outcomes. The debate over voter fraud remains politically charged, often used to justify either stricter voting laws or arguments against voter suppression, but the empirical evidence itself is remarkably consistent across studies.