Is voter fraud a problem in the US?
Is voter fraud a significant US problem? I'll gather the latest research and evidence to examine its scale.
In the United States, voter fraud exists but is documented to be extremely rare relative to votes cast - not at a scale that has been shown to alter statewide or national election outcomes.
What research finds
Empirical evidence across decades and multiple methodologies leads to one repeatable conclusion: proven voter fraud is extremely rare, measured in orders of magnitude of 0.0000X percent.
When reported as a rate, different studies find:
Government investigations and courts have reached similar conclusions:
Federal election security officials have stated:
What the largest database of proven cases shows
Even the Heritage Foundation—which maintains the most comprehensive database of documented voter fraud cases and has been among the most vocal advocates for stricter election integrity measures—has catalogued approximately 1,500 proven cases spanning multiple decades. Other tallies from the database cite:
Brennan Center analysis of Heritage data noted that in Arizona, Heritage had to search 25 years across 36 elections and over 42 million ballots to find 36 cases of fraud—a rate of 0.0000845 percent.
This illustrates the denominator problem: with well over a billion ballots cast since 2000, a few hundred or thousand cases over decades represents a microscopically small rate.
Does fraud never happen?
Yes, isolated cases of voter fraud have occurred. However, these are typically individual acts of misconduct, such as voter impersonation or illegal absentee voting, not evidence of widespread, organized schemes.
Common types that do appear in prosecutions include:
These cases often receive disproportionate media attention, contributing to the misconception of widespread fraud.
Some conservative analysts argue the Heritage numbers undercount fraud because not all fraud is detected or prosecuted, and describe voter fraud as a "real and pressing issue." However, nonpartisan academic reviews, court findings, and federal investigations have consistently concluded that while fraud is not zero, it is "vanishingly rare" and impersonating a voter at the polls is less common than being struck by lightning.
Why perception differs from data
Nearly half of Americans believe that there is a problem, when the research says there is none. Studies note that charges are usually based on anecdotes, and that inaccurate voter rolls (e.g., 24 million outdated or inaccurate registrations estimated by Pew) are often confused with actual fraudulent votes cast.
Most reported incidents of voter fraud are actually traceable to other sources, such as clerical errors or bad data matching practices.
Bottom line
Election officials therefore focus on safeguards — voter list maintenance, audits, signature verification for mail ballots, and post-election audits — to keep the already-low rates low, while researchers caution that overemphasis on fraud risk can lead to policies that create barriers for eligible voters.
Citations (1)
- Debunking the Voter Fraud Myth
Backs: “One widely cited Brennan Center for Justice analysis found that voter impersonation occurs at a rate between 0.0003% and 0.0025%. The Brennan Center's report reviewed elections that had been meticulously studied for voter fraud, and found incident rates between 0.0003 percent and 0.0025 percent”