The major elections legislation currently being considered by the Senate—the Safeguard American Voter Eligibility (SAVE) America Act—is a bill in flux. At times it has required in-person voter registration (as it currently does). At times it hasn’t. At times it has sought to include everything from a ban on mail voting to a prohibition on transgender athletes in women’s sports. As it currently stands, neither of those provisions are included.
But the legislation’s core feature has always been a requirement for voters to provide proof of citizenship when registering to vote. Some states already have this requirement. But federal law (the National Voter Registration Act of 1993) currently requires voters only to attest under penalty of law that they are United States citizens.
The House passed its version of the bill in February, but the legislation has proved far more contentious in the Senate, where a handful of Republicans don’t support the bill in its current form. President Donald Trump has called the SAVE America Act’s passage “his No. 1 priority.”
Many pugilists in the debate over the SAVE America Act assume that a proof-of-citizenship requirement would yield a Republican electoral advantage. Trump said that the Democrats “know if we get [the SAVE America Act], they probably won’t win an election for 50 years.” Sen. Mike Lee of Utah warned that Republicans will lose power—“likely for a long time”—if Congress doesn’t pass it (although Trump won in 2016 and 2024 without the legislation, and Utah, Lee’s home state, does not require documented proof of citizenship for voter registration).
For Republicans, the purported political advantage of making the SAVE America Act law wouldn’t stem from the sudden termination of large numbers of noncitizens voting for Democrats, as some in the GOP have claimed. There is no evidence that this has happened or would happen. Out of the billions of ballots cast in the last 40 years, there have been only a few prosecutions, in any state, of voting by noncitizens. And recent affirmative investigations taken by states have confirmed this virtual nonexistence of noncitizens on the voter rolls.
Nor is there any evidence that the limited amount of election fraud that is committed is done exclusively in favor of Democrats. A recent case of successfully prosecuted election fraud involved a man who voted twice for President Donald Trump in the 2020 election, once by mail-in ballot in Pennsylvania and once after moving to and registering to vote in Florida. When I was the Maricopa County recorder, I would often tell people that, on a partisan basis, the limited instances of voter fraud probably balance each other out to net zero. Nobody liked hearing that.
Another popular theory is that the SAVE America Act will favor Republicans because Republicans are predominantly white, and Democrats are less so, therefore Republicans will more easily meet the new registration requirements. This theory also squares nicely with the long-held—but false—notion that higher turnout benefits Democrats. But it’s a theory that doesn’t meet the underlying statistics of who has documented proof of citizenship.
If the SAVE America Act becomes law, the vast majority of Americans will use either a passport or a birth certificate to prove citizenship. Although more Americans than ever before have a passport, only about half of U.S. citizens have one—namely the affluent and the educated. Most (58 to 64 percent) Americans with a bachelor’s degree have a passport, and 71 percent of Americans with a graduate or professional degree have one. Only 25 percent of those with only a high school diploma have a passport, and 39 percent with some college or an associate’s degree have the document. Members of households making more than $100,000 a year are likely to have a passport (64 percent), those in households with incomes between $50,000 and $100,000 are not (44 percent), and people with household incomes of under $50,000 definitely are not (20 to 21 percent).
This might sound advantageous to the Republican Party of Mitt Romney and Paul Ryan. But that’s not Trump’s GOP. CNN’s exit poll from the 2024 election found that voters in households earning more than $100,000 per year favored Kamala Harris over Trump, 51 to 47 percent, and people in households earning more than $200,000 went for Harris 52 to 46 percent. Trump narrowly won among members of households earning less than $50,000 per year.
The difference in party preference based on education level in voting patterns in the 2024 election is even more pronounced. According to Pew Research, Harris won by 32 percent among voters with a graduate or professional degree, and she won college graduates by 5 percent. Trump won those with “some college” by 9 percent and those with a high school degree or less by 20 percent.
When you look at the geographic breakdown of passport holders, other variations become clear. Two-thirds of residents in California, New Jersey, New York, and Massachusetts have passports. Fewer than one-third of adults have a passport in Alabama, Arkansas, Kentucky, Louisiana, Mississippi, Oklahoma, and West Virginia. Nobody could look at that map and think it gives an advantage to Trump Republicans.
And indeed, that’s the conclusion of researchers from the University of Maryland. “Democrats are more likely to have access to a passport across all surveys,” said Sam Novey, the chief strategist for the school’s Center for Democracy and Civic Engagement. And in the state-specific work they’ve done, Novey and his colleagues found that in Georgia, 53 percent of Democrats, 43 percent of Republicans, and 36 percent of independents have a passport. In Texas, “Democrats are somewhat more likely to be able to use a passport than Republicans (60 percent and 54 percent, respectively).”
There is less data for the other prove-your-citizenship document: a birth certificate. The most commonly cited data point is an aggregated assessment from the same team at the University of Maryland, which worked with voting access advocacy and research groups VoteRiders, Public Wise, and the Brennan Center for Justice. Their assessment found that “more than 9 percent of American citizens of voting age, or 21.3 million people, don’t have … a passport, birth certificate, or naturalization papers … readily available.” Here, on a partisan basis, Republicans have a slight advantage. But that advantage could be nullified by other variables. For example, women who are Republican or lean Republican are more likely to take the last name of their husbands; the name change creates an additional burden under the SAVE America Act because they would also have to provide a marriage certificate that documents the name change. Therefore, registering to vote under the SAVE America Act might be less burdensome for women who haven’t changed their name, and those women are likely more liberal.
There are legitimate public policy interests in play with the SAVE America Act; we should have conversations about proof of citizenship. But there is no evidence to suggest that the SAVE America Act is a silver bullet for Republican electoral success. I suspect that issues such as gas prices, the war in Iran, security lines at airports, and the cost of goods and services will all be much more significant in determining who wins control of Congress in November 2026.
