Is voter fraud a problem in the US?

Short answer: No, not in any significant or widespread way

Extensive research across multiple decades, methodologies, and political affiliations (including conservative-leaning sources) consistently shows that voter fraud in the U.S. is exceedingly rare and has no meaningful impact on election outcomes.

What the research shows

It's statistically tiny. A Loyola Law School professor found 31 credible allegations of fraud among roughly 1 billion ballots cast between 2000 and 2014. Similarly, a comprehensive 2014 study found 31 credible instances of impersonation fraud from 2000 to 2014, out of more than 1 billion ballots cast—and even that tiny number is likely inflated.

Even conservative-leaning data confirms this. The Heritage Foundation maintains an "Election Fraud Database" often cited by fraud proponents, but when researchers examined it: an analysis of the Heritage Foundation's database found the share of reported fraud cases over 13-38 years across key swing states is less than 1%, and in Pennsylvania, 30 years of data across 32 elections showed just 39 cases of voter fraud—0.000039% of over 100 million ballots cast.

Voter impersonation (the type ID laws target) is almost nonexistent. The Department of Justice found "no apparent cases of in-person voter impersonation...anywhere in the United States, from 2004 through July 3, 2014." A review by Arizona State University's News21 program similarly found that out of 2,068 tallied cases of alleged election fraud between 2000 and 2012, only 10 could have been prevented with voter-ID legislation.

Mail-in voting fraud is also negligible. A 2023 MIT Election Data and Science Lab paper found that studies consistently show negligible evidence of fraud in state elections, and Brookings Institution researchers examining 2016–2022 elections estimated about four cases of fraud for every 10 million mail ballots cast.

2024 election data confirms the trend continued. Heritage's Election Fraud Database found 1,561 instances of voter fraud resulting in 1,325 convictions overall, with only 20 cases brought in 2024 — a minuscule number relative to the roughly 155 million votes cast that year.

Noncitizen voting is also essentially nonexistent. A 2014 paper concluded that "the likely percent of non-citizen voters in recent US elections is 0." In Washington state, for example, 15 cases of noncitizen voting have been prosecuted from 1982 through 2025, or about 0.000006% of ballots cast.

The disconnect: perception vs. reality

Despite this overwhelming evidence, public belief in widespread fraud remains high and has grown in some categories. In 2024, 40% of surveyed voters believed noncitizen voting is a common occurrence, up from 34% in 2016. Similarly, despite fraud being extremely rare, six in ten U.S. adults were concerned about it ahead of the 2024 presidential election, with concern reaching 86% among Republican-leaning voters versus 33% among Democratic-leaning voters.

Bottom line

Individual, isolated cases of fraud do occasionally happen and are prosecuted, but every rigorous study—regardless of who conducted it—finds that fraud occurs at rates so low (often a tiny fraction of 1%) that it could never swing an election outcome. The gap between actual fraud rates and public perception is itself a well-documented phenomenon, often linked to political rhetoric and media coverage rather than empirical evidence.