“Can I See Some ID?”

If the vote in the United States was intended to give everyone a voice in matters of government, it has rarely been successful. Since the founding of the Republic there have been times when you couldn’t vote if you didn’t own land, if you were a woman, or if you were black or poor—overt value judgements on who had the capacity or right to decide the future of the Nation. But the manipulation of the vote has taken a number of other forms as well, from racist Jim Crow laws intended to disenfranchise blacks to allegations of ballot box stuffing in the 1960 presidential election to the party-issued ballots of the progressive era. Today, the battle over the margins of the vote manifests in a suite of laws in place across the Union designed to ensure the integrity of the vote. Proponents argue that such laws, which take the form of voter ID requirements or restrictions on registration and early voting, ensure the principle of “one man, one vote,” and prevent illegal immigrants or criminals from unduly influencing elections of crucial importance. Some, however, claim that many such laws actually injure election integrity by disproportionally suppressing the vote of young and minority voters, and that anti-voter-fraud techniques are designed solely with partisan aims.

As a general rule, the United States votes less than its economically developed counterparts. Voting and registering to vote are largely an individual responsibility, as the United States does not have automatic voter registration procedures, compulsory voting laws, or holiday for national elections. Data from the last century also indicate that the U.S. votes less in midterm than in presidential elections, that older Americans are more likely to vote than younger Americans, and that white Americans are most likely to vote, followed by black Americans and Hispanic Americans. Overall, voter participation has dropped since the 1970’s in both midterm and Presidential election years, and the current generation of young voters—Generation X, Millennials, and post-Millennials—are much less likely to vote than their older counterparts were. Research strongly indicates that factors like voter engagement, income levels, and the ease of registration can help determine voting rates. Political parties spend tens of millions of dollars a year trying to manipulate turnout in order to advance their own election prospects. Voter suppression is one way to do that. And yet, the proposition that who votes is nearly as important as who is on the ballot itself remains undebated.

Since the 1965 Voting Rights Act (VRA), states with a history of voter suppression or racist voter-registration policies (defined in Section 4b) had been forced to obtain “preclearance” from the Department of Justice or a federal court before changing any of their voting laws or procedures (Section 5). Over the course of its history, the Voting Rights Act had blocked 86 potentially discriminatory laws from implementation. In addition to overt blockings, 262 voting changes were withdrawn or altered upon DOJ recommendation between 2006 and 2013 alone; but in 2013, a Supreme Court ruling in Shelby County v. Holder held Section 4b, which held these states accountable, unconstitutional. The ruling deemed the concerns addressed by the Voting Rights Act as having “no logical relationship to the present-day.”

U.S. Attorney General Eric Holder held a press conference announcing Department of Justice plans to sue North Carolina over Voter ID regulations September 30, 2013 in Washington, DC. Under the new law North Carolina residents are required to show a photo ID at polling places which some believe threatens the voting rights of minorities

In response to the ruling, several states immediately went forward with various pieces of voting legislation or rule changes once-held up by the VRA. Just hours after the Shelby ruling, Texas announced the implementation of a strict photo ID law, projected by the Department of Justice to disenfranchise hundreds of thousands of registered Texas voters, a disproportionate number of whom were Latino. Days later, North Carolina passed a law which cut back on early voting, reduced the window for registration, and implemented strict voter ID requirements. 30% of those affected by the ID requirement were African American, and 70% of African Americans who cast their ballots in the 2008 election used early voting. A report by NYU Law’s Brennan Center documents these changes as well as and others through 2014, one year after Shelby was decided.

Voter ID laws today exist in 34 states, and take various forms—a comprehensive resource for voting requirements is kept here. For example, Ohio has a wide range of acceptable forms of identification—you can use your phone bill—while states like Wisconsin and Virginia demand photo identification. Texas, among other states, requires photo ID, but precludes student IDs from the mix.

Since the first enactment of voter identification law, to the first time they were tested in the Supreme Court (Crawford v. Marion County, 2008), voter ID has been about a state’s interest in protecting itself from voter fraud, at least prima facie. Variations of the voter ID requirement are justified by the ‘threat’ that people may vote multiple times or may use the name of a deceased voter, according to the legislatures who pass such measures. Notably, however, every voter ID law on the books (with the exception of one in Rhode Island) has been passed by a state legislature controlled by Republicans. And their interest, according to their Democratic colleagues, is voter suppression, not the protection of the vote.

To begin with, it is worth noting that voter fraud does exist. The Heritage Foundation, which keeps a database of voter-fraud convictions, cites exactly 1,145 instances of proven voter fraud, 992 criminal convictions, and 84 civil penalties. Their analysis illustrates a key point of the debate, however, because it documents all proven instances of voter fraud as far back as the 1982 midterms, as well as some cases which occurred in state and local elections. To give some perspective, over 128 million votes were cast in the last presidential election alone. Even if we consolidate every proven instance of voter fraud since 1982 into the 2016 election and pretend that the fraud occurred in the presidential election, the fraudulent ballots make up 0.0008% of the vote. In an ironic turn, the only two proven instances of voter fraud in the 2016 presidential elections were the convictions of a Texas man and an Iowa woman, both of whom voted twice for President Trump to combat his election-rigging worries. All that aside, the obvious reality is that while voter fraud does exist, the degree to which it affects presidential or congressional elections is absolutely zero. In fact, many of the instances of voter fraud cited by the Heritage Foundation were conducted using absentee ballots, violations which would not have been prevented by requirements that voters show identification at the polls.

The likely rationale for voter ID laws and the regular purging of voter rolls, absent that stated, is that burdening voters with a requirement to re-register to vote every few years or obtain a some new form of identification will lower turnout or suppress their vote. And, to the benefit of the Republican legislatures that enact such restrictions, time and again, those voter ID laws and the purging of voter rolls have been shown to disproportionately disenfranchise black and Latino voters. For example, that same strict North Carolina voter ID law, mentioned above, was recently denied its day in the Supreme Court after a federal judge found it unconstitutional because it targeted blacks “with almost surgical precision.” That law, similar to voter ID laws being challenged in court today, rejected forms of identification used disproportionately by blacks, including student IDs, IDs issued to government employees, and IDs used to collect public assistance.

The midterm elections of 2018 have, continuing upon the precedent of recent election years, already seen the implementation of voter identification laws and further voter-suppression methods. North Carolina, for example, responded to the 2017 legal defeat in time for the 2018 midterm elections by enacting early voting restrictions, a policy that particularly affects blacks due to the higher likelihood that they are hourly-wage workers who will be unable to reach the polls on Election Day. In a nationally known scandal, Georgia Secretary of State and Republican nominee for Governor, Brian Kemp, has invoked a Georgia law requiring an exact match between voter registration data and information kept by other State administrations in order to suspend 53,000 pending voter registrations. Incidentally, while blacks make up 32% of the State’s population, they represent 70% of the suspended applications. Secretary Kemp’s office also has a record of purging naturalized Asian-Americans from voter rolls without notifying them of any changes.

But until recently, questions have remained about whether we can state conclusively that voter ID laws are specifically responsible for lower turnout. In rare cases, there is data available to indicate exactly how many people were turned away from polls due to lack of proper identification, but we are otherwise left to guess whether it was ID requirements, lack of engagement, or unfavorable weather that kept voters at home. Fortunately, a new working paper by Bernard Fraga and Michael Miller promises to conclusively name voter ID laws as the culprits of discriminatory suppression by examining Texas election data. In a case challenging Texas’ 2014 voter ID law, a federal court forced Texas to offer an alternative voting method to those who traveled to a polling location without an acceptable form of ID. In response, Texas allowed voters to file a “reasonable impediment declaration” and vote after providing some other type of ID, like phone bill or birth certificate. Fraga and Miller analyzed more than 16,000 such declarations and uncovered evidence of the Texas law’s bias, finding that those voting without ID were more likely to be black or Latino than who brought ID to the polling place. A similar study conducted by Marc Meredith and Michael Morse found that 28,000 Michigan voters lacked identification in the 2016 election, and that those who would have been denied a vote under a stricter law were disproportionately brown, black, and Democratic. Unfortunately for voting-rights advocates, such laws are incredibly popular. Up to 80% of Americans support a photo-identification voting requirement, bolstered by 95% of Republicans and 83% of Independents. Notably, however, up to 63% of Americans support automatic voter registration.

Not only are voter ID laws often designed to suppress the vote, but also they often work to elect Republicans: recent research suggests that overall turnout is decreased by up to 2% under the strictest laws, far exceeding President Trump’s margins of victory in Wisconsin and Michigan. In other words, not only do many of these laws have racial bias, but their effects are wide-ranging and extremely consequential.