Although non-citizen voting in federal elections is illegal, commentators and politicians have voiced concern that the law may not be adequately enforced. Unadjusted for potential undercount, Census Bureau data collected earlier this year indicates that there are 23.2 million voting-age non-citizens in the country, roughly half of whom are illegal immigrants.1 In most states the non-citizen population is not large enough to flip the state in 2024, assuming the state is as close as it was in 2020. However, our analysis does show that in a dozen states, plus the Nebraska 2nd district, the non-citizen population in 2024 is potentially large enough to exceed the state winner’s margin in 2020. Of course, just because it is mathematically possible for non-citizens to decide a state’s presidential outcome does not mean it will actually happen, but it illustrates the potential seriousness of the non-citizen voting issue.2
Link to Excel Spreadsheet with Adult Non-Citizens by State
Among the findings:
- Mathematically, there are 12 states, plus the Nebraska 2nd congressional district, where the voting-age non-citizen population is larger in 2024 than the state winner’s margin of victory in the last presidential election.3
- In Georgia, 1.5 percent of adult non-citizens in 2024 would need to vote to equal the state winner’s margin of victory in the last presidential election. In Arizona, 1.71 percent of non-citizens would have to vote to potentially change the outcome.
- Other places where the non-citizen population is large enough in 2024 to equal the winner’s margin of victory in the last presidential election are: North Carolina where 10.26 percent would have to vote, in Nevada it would require 12.98 percent, in Pennsylvania 15.61 percent, in Wisconsin 15.96 percent, in Texas 18.01 percent, in Florida 19.23 percent, in Michigan 56.87 percent, in Nebraska’s 2nd district 63.63 percent, in New Jersey 72.96 percent, in Virginia 79.54 percent and in New Mexico 82.65 percent.
- All of the above percentages come with the important caveat that they assume that all non-citizens would vote the same way.
- Exit polling in 2016 indicated that two-thirds of naturalized citizens voted for the Democratic presidential candidate and one-third voted for the Republican when looking at the two-party vote.4 Assuming non-citizens would have the same voting preferences as naturalized citizens in 2016, then it would require 4.49 percent of non-citizens to vote in Georgia and 5.13 percent to vote in Arizona in 2024 for the net gain in votes for Harris to exceed the margin of victory in the last presidential election.
- If non-citizens voted two-thirds for Harris versus one-third for Trump, it would require 30.79 percent of non-citizens to vote in North Carolina for the net gain for Harris to equal the margin of victory in the 2020 presidential election. In Nevada it would require 38.97 percent to vote, in Pennsylvania 46.84 percent, in Wisconsin 47.89 percent, in Texas 54.06 and in Florida 57.71 percent to vote in Florida.5 These percentages seem extremely implausible.
- Alternatively, there is some evidence naturalized citizens may prefer Harris over Trump by roughly a 60 percent to 40 percent margin.6 If non-citizens had this voting preference it would require 7.48 percent of non-citizens in Georgia and 8.55 percent of non-citizens in Arizona to vote in 2024 for the net gain in support for Harris to exceed the margin of victory in the last presidential election.7
- Assuming a 60 percent to 40 percent Harris v. Trump split among non-citizens would require 51.30 percent of non-citizens to vote in North Carolina, 64.94 percent to vote in Nevada, 78.05 percent to vote in Pennsylvania, 79.81 percent to vote in Wisconsin, 90.08 percent to vote in Texas and 96.18 percent to vote in Florida to change the 2024 outcome.8 These percentages seem extremely implausible.
- We know of no polling to indicate that foreign-born voters are likely to favor former President Trump over Vice President Harris.
 |
Endnotes
1 Data for this analysis comes from a combined three-month average of the public-use January, February, and March Current Population Survey, which is collected by the Census Bureau.
2 Media accounts based in part on prosecutions and a database on non-citizen voting maintained by the Heritage Foundation show that non-citizen voting has occurred, but only in small numbers. Perhaps the most widely cited study on the topic was in the academic journal Elections Studies in 2014. That study estimated did find that 6.4 percent of non-citizens voted in the 2008 presidential election, which if correct would be significant. However, this finding is much debated.
3 Maine and Nebraska allocate their electoral votes in part based on congressional districts. We do not have 2024 data for the 2nd Nebraska district from the CPS. We instead use the 2023 American Community Survey data for that district found at data.censu.gov.
4 Unfortunately, the exit polls in 2020 did not ask if voters were born outside of the United States.
5 In no other states could non-citizen voting exceed the 2020 margin of victory, assuming a 2:1 partisan breakdown in favor of Harris.
6 A report by the U.S. Immigration Policy Center University of California, San Diego based on a survey of naturalized citizens done by the National Partnership for New Americans finds that on average in the battle ground states of Arizona, Florida, Georgia, Michigan, Nevada and Pennsylvania naturalized citizens showed that of the two-party vote, 60 percent of respondent preferred Vice President Harris over former President Trump. https://usipc.ucsd.edu/publications/usipc-npna-new-american-voters-poll-september.pdf Of course, it is certainly possible that non-citizens would have a different partisan break down in the upcoming election. It is worth pointing out that the Current Population Survey Annual Social and Economic Supplement from March 2024 shows that the adult non-citizen population is much more likely to be Hispanic and Asian, groups that tend to vote Democratic, than is the naturalized citizen population. The same data also shows that non-citizens have significantly lower average incomes than naturalized citizens. This also suggests that non-citizens might be more likely to vote for Vice President Harris than naturalized citizens. Further, a very large share of non-citizens are in the country illegally, or are legal immigrants who live with illegal immigrants, and the vice president has indicated her strong support for an amnesty for illegal immigrants, while former President Trump has argued for robust enforcement of immigration laws. This too seems likely to increase her popularity among non-citizens if they were to vote.
7 To calculate these percentages, we take the percentage of adult non-citizens who would need to vote in 2024 to exceed the margin of victory in 2020 and multiply it by 5.001. For example, 1.496 percent of the non-citizen population in Georgia would equal the 2020 margin of victory in the state. If we take that value and multiple it by just slightly more than 5 it produces a percent of the non-citizen population of 7.479 percent or 58,907, with 35,344 for Harris and 23,563 for Trump giving Harris a gain that just slightly exceeds the 2020 margin of victory for President Biden. Again, this assumes a Harris advantage of 60 percent to 40 percent and the margin of victory in 2024 would be the same as in 2020.
8 In no other states could non-citizen voting exceed the 2020 margin of victory, assuming a 60 percent to 40 percent partisan breakdown.