We are not aware of any documented cases in which individual noncitizens have either intentionally registered to vote or voted while knowing that they were ineligible. Given that the penalty (criminal prosecution and deportation) is so severe, and the payoff (one incremental vote) is so minimal for any individual voter, it makes sense that extremely few noncitizens would attempt to vote, knowing that doing so is illegal.
Although there are a few recorded examples in which noncitizens have apparently registered or voted, investigators have concluded that they were likely not aware that doing so was improper. In one highly publicized case, for example, noncitizens were given voter registration forms by a group helping them through the naturalization process, as immigration officials stood and watched. These individuals most likely mistakenly thought it their obligation and privilege to complete the paperwork, and most likely did not intentionally fabricate their citizenship status in front of federal officials who knew that they were noncitizens.
Far more common than these incidents of noncitizen voting are allegations of noncitizen voting that prove unfounded. Other claims are premised on matching lists of voters from one place to another, but upon closer inspection, the match process shows error. Sometimes the interpretation is flawed: two list entries under the same name—even the same name and birthdate—indicate different individuals. Sometimes the lists themselves are flawed, with an individual listed with an outdated or inaccurate citizenship status, or marked due to a clerical error as voting when she did not in fact cast a ballot. Sometimes it’s both.
The following analyze some recent allegations of voter fraud by noncitizens:
- In Washington in 2005, an individual asked county offices to investigate the citizenship status of 1668 registered voters based on their “foreign-sounding names.” There are no reports of which we are aware that any individual on the submitted list was actually a noncitizen.
- In Washington in 2004, documentation appears to show that two votes were cast in King County by noncitizens. Given these votes, the rate of documented noncitizen votes in King County was 0.0002%.
- In Milwaukee in 2001, journalists analyzed 370,000 voting records from 1992–2000, and found four instances in which voters’ names matched a list of naturalized city residents but appeared to have voted before their naturalization dates. Even if all four of the matched records accurately represent noncitizen votes, the rate of noncitizen voting among the city records examined would have been 0.001%.
- In Hawaii in 2000, 553 apparent noncitizens were alleged to have registered to vote. On further investigation, 144 documented that they had become citizens. At least sixty-one individuals affirmatively asked to cancel their registration; the others were stopped at the polls and specifically asked about their citizenship before voting. There are no reports of which we are aware that any noncitizen actually voted. To the extent that noncitizens were actually represented on the rolls, officials attributed the registrations to mistake rather than fraud.
- In Hawaii in 1998, 30–40 percent of the registered voters in certain precincts were alleged to be noncitizens. The INS investigated more than 10,000 names, and identified fewer than twelve noncitizens whose names matched those on the voter rolls; none of those registered voters were proven to be noncitizens. A separate proceeding uncovered three noncitizens who were proven to have voted, and three others who were reported to be under further investigation. Even if all six had voted, the overall noncitizen voting rate would have been 0.001%.
- In California in 1996, the Task Force of the U.S. House of Representatives investigating the Dornan/Sanchez election alleged that 624 registered noncitizens cast votes in the 46th Congressional District; the allegations were based largely on attempts to match immigration lists to voter rolls (71 voters matched the individual’s name, date of birth, and signature; other matches were less reliable). A minority report claimed that about 400 of the individuals had registered as noncitizens but became citizens by election day. State officials reportedly found that one independent nonprofit group in particular registered 632 noncitizens, 364 of whom cast votes. The noncitizens registered by this group were proceeding through the naturalization process; many were registered immediately after passing the INS citizenship interview and receiving INS letters stating, “Congratulations, your application for citizenship has been approved” – though the actual swearing-in ceremony did not take place for several months. Assuming that 624 individuals were in fact noncitizens when they cast their votes, the overall noncitizen voting rate would have been 0.6%.
In part, federal law now addresses the mistaken registration problem: the Help America Vote Act has changed the federal registration form so that registrants are specifically and prominently asked whether they are citizens. To the extent that additional safeguards are thought necessary, such solutions should reflect that any problem supported by facts seems to be misinformation rather than fraud.