25 January 2017

Non-Citizen Voting Reality Check

How Many Non-Citizens Voted In 2016?

Voting by non-citizens is not a significant problem in the United States. Based upon the data below, a good ballpark estimate for the number of non-citizens who voted in the November 2016 Presidential election in the entire United States is about 1,500 (roughly 1 per 100,000 votes cast). 

Something on the order of 1 in 300 to 1 in 500 illegally cast ballots are cast by non-citizens. Most voters that are cast illegally are cast by felons on parole, disqualified felons in states where a felony conviction disqualifies you from voting, or by eligible voters who cast both an absentee ballot and also vote in person). Most cases of illegal voting of all types, moreover, are examples of honest mistakes and not concerted efforts to rig an election in favor of a particular candidate.

Roughly 1 in 14,000 adult non-citizens in the United States votes in any given Presidential election year.

Moreover, non-citizens who do vote are predominantly legal permanent residents of the United States who have met all or almost all legal requirements except a swearing in ceremony for obtaining U.S. citizenship, who were mistaken about when their U.S. citizenship status would become final. In 2014, the most recent year for which data are available, 654,949 people (a significant number of which are children) became naturalized U.S. citizens. So, something on the order of 1 in 200 to 1 in 400 adults who are on the verge of becoming U.S. citizens accidentally jump the gun and vote before their naturalization ceremony is complete. This percentage is a significantly smaller percentage than the number of felons on parole in the voter registration database who actually vote in states where they are not allowed to do so (which is about 2%).

The number of illegal immigrants who voted in 2016 was probably in the range of dozens to a few hundred, because legal permanent residents of the United States are much more likely to illegally vote than temporary immigrants or undocumented immigrants.

There is no evidence that voting by non-citizens has ever changed the outcome of an election in the United States in the last century. 

Also, it is not always been necessary to be a U.S. citizen to vote anyway. Until 1921, for example, it was legal for non-citizens to vote in the State of Texas.

Trump's estimate of 3 to 5 million non-citizen voters in 2016 is too large by a factor of roughly two thousand to three thousand.


There were 251.1 million people in the United States aged eighteen or older in the last election. Nationally, eligible voter turnout was 58.4%. 

About 8.4% of people in the United States who are eighteen years of age of older are not citizens (21.1 million people). There are about 11.3 million undocumented non-citizens in the United States, some of whom are eighteen years of age or older, perhaps 80% of whom (9 million) are adults. 

* A 2006 study found that 4 non-citizens voted out of 370,000 votes cast in one then recent Wisconsin election (a rate of roughly 1 per 100,000 votes cast). In each of the four cases, the individual had almost been granted citizenship and believed that the paperwork was final but was mistaken. By comparisons 12 dead people voted (two of whom were "juniors" who mistakenly voted under the names of their dead, and a third was a woman who had been legally declared dead in a death certificate but was, in fact, alive and trying to sort out the resulting mess), and 361 were felons who had been released from prison but were still on parole and didn't know that they weren't allowed to vote. 

In the 2004 election in Colorado, a study found that 122 people illegally voted twice, 120 felons on parole cast illegal ballots (2% of the roughly 6,000 who were registered to vote in the state), and no cases of non-citizens voting were identified.

* Few non-citizens voted in Utah according to a 2005 study:
In 2005, Utah’s legislative audit bureau attempted to undertake a systematic study of illegal immigrants who had obtained state identification cards – either driver’s license or state identification cards. Utah determined that some 383 possibly illegal immigrants were registered to vote. Utah asked ICE to review these registered voters to determine if, in fact, they were U.S. citizens. ICE examined a sample consisting of 135 of these individuals and determined that 5 were naturalized citizens, 20 were “deportable,” one was a permanent legal resident and the other 109 had no record and were likely in the United States illegally. Fourteen of these 383 individuals voted in a recent election in Utah, but ICE did not provide enough information to the state to allow it to determine whether these 14 individuals were in fact citizens.
In the 2004 election, 1,012,000 people voted in Utah, so the number of possible non-citizens who voted was again roughly 1 in 100,000. 

Studies in Harris County, Texas and Oahu, Hawaii, likewise found that the percentage of people registered to vote who were non-citizens was tiny and found no evidence that any significant number of non-citizens who were registered to vote actually did vote.

No comments: