Policy Brief on the Truth About “Voter Fraud”
* Fraud by individual voters is both irrational and extremely rare.
* Many vivid anecdotes of purported voter fraud have been proven false or do not demonstrate fraud.
* Voter fraud is often conflated with other forms of election misconduct.
* Raising the unsubstantiated specter of mass voter fraud suits a particular policy agenda.
* Claims of voter fraud should be carefully tested before they become the basis for action.
Fraud by individual voters is both irrational and extremely rare. Most citizens who take the time to vote offer their legitimate signatures and sworn oaths with the gravitas that this hard-won civic right deserves. Even for the few who view voting merely as a means to an end, however, voter fraud is a singularly foolish way to attempt to win an election. Each act of voter fraud risks five years in prison and a $10,000 fine - but yields at most one incremental vote. The single vote is simply not worth the price.
Because voter fraud is essentially irrational, it is not surprising that no credible evidence suggests a voter fraud epidemic. There is no documented wave or trend of individuals voting multiple times, voting as someone else, or voting despite knowing that they are ineligible. Indeed, evidence from the microscopically scrutinized 2004 gubernatorial election in Washington State actually reveals just the opposite: though voter fraud does happen, it happens approximately 0.0009% of the time. The similarly closely-analyzed 2004 election in Ohio revealed a voter fraud rate of 0.00004%. National Weather Service data shows that Americans are struck and killed by lightning about as often.
Many vivid anecdotes of purported voter fraud have been proven false or do not demonstrate fraud. Although there are a few scattered instances of real voter fraud, many of the vivid anecdotes cited in accounts of voter fraud have been proven false or vastly overstated. In Missouri in 2000, for example, the Secretary of State claimed that 79 voters were registered with addresses at vacant lots, but subsequent investigation revealed that the lots in question actually housed valid and legitimate residences. Similarly, a 1995 investigation into votes allegedly cast in Baltimore by deceased voters and those with disenfranchising felony convictions revealed that the voters in question were both alive and felony-free.
Many of the inaccurate claims result from lists of voters compared to other lists - of deceased individuals, persons with felony convictions, voters in other states, etc. These attempts to match information often yield predictable errors. In Florida in 2000, a list of purged voters later became notorious when it was discovered that the “matching” process captured eligible voters with names similar to - but decidedly different from - the names of persons with felony convictions, sometimes in other states entirely. A 2005 attempt to identify supposed double voters in New Jersey mistakenly accused people with similar names but whose middle names or suffixes were clearly different, such as “J.T. Kearns, Jr.” and “J.T. Kearns, Sr.,” of being the same person. Even when names and birthdates match across lists, that does not mean there was voter fraud. Elementary statistics students are often surprised to learn that it is more likely than not that among just 23 individuals, two will share a birthday. Similar statistics show that for most reasonably common names, it is extremely likely that at least two people with the same name in a state will share the same date of birth. The ostensible “matches” may not represent the same person at all.
Other allegations of fraudulent voting often turn out to be the result of common clerical errors, incomplete information, or faulty assumptions. Most allegations of voter fraud simply evaporate when more rigorous analysis is conducted.
Voter fraud is often conflated with other forms of election misconduct. It is extremely rare for individuals to vote multiple times, vote as someone else, or vote despite knowing that they are ineligible. These rare occurrences, however, are often conflated with other forms of election irregularities or misconduct, under the misleading and overbroad label of “voter fraud.u201D Some of these other irregularities result from honest mistakes by election officials or voters, such20as confusion as to whether a particular person is actually eligible to vote. Some irregularities result from technological glitches, whether sinister or benign: for example, voting machines may record inaccurate tallies. And some involve fraud or intentional misconduct perpetrated by actors other than individual voters: for example, flyers may spread misinformation about the proper locations or procedures for voting; thugs may be dispatched to intimidate voters at the polls; missing ballot boxes may mysteriously reappear. These more common forms of misconduct are simply not addressed by the supposed “anti-fraud” measures generally proposed.
Raising the unsubstantiated specter of mass voter fraud suits a particular policy agenda. Voter fraud is most often invoked as a substantial problem in order to justify particular election policies. Chief among these is the proposal that individuals be required to show photo ID in order to vote - a policy that disenfranchises up to 10% of eligible citizens. But the only misconduct that photo ID addresses is the kind of voter fraud that happens as infrequently as death by lightning. Therefore, it suits those who prefer photo ID as a policy to lump as much misconduct in with “voter fraud” as possible, to create the impression that the problem is far more significant than it actually is. Moreover, to the extent photo ID is suggested as a solution to the perception that voter fraud occurs, it behooves those who prefer photo ID to reinforce the unsubstantiated perception that voter fraud exists.
Claims of voter fraud should be carefully tested before they become the basis for action. Researchers, reporters, public figures, and policymakers confronted with claims of potential fraud should carefully examine these claims before calling for action. Do the claims depend on matching information from one list to another? Is the matching process accurate? Does a match indicate an illegal vote, or is there a more plausible explanation? Is corroborating evidence available? If there actually appears to be a problem, can it be addressed by existing practices, or is a new solution necessary? If so, will the solution proposed - usually either a mass purge or photo identification - really solve the problem? Is the solution sufficiently burdensome that it becomes a greater problem than the problem itself? These basic questions are crucially important to evaluating claims of voter fraud, but are all too often unasked and unanswered.