Two days ago, the US Justice Department informed the state of Virginia that it would not challenge its new voter identification law. Utilizing the Voting Rights Act, the Justice Department has challenged recently the voter ID laws of a number of states. Virginia is one of sixteen states with histories of voter discrimination that must gain federal approval before changing state statutes regarding voting and voter eligibility. Newly approved, Virginia’s law should be in effect for the November election.
Republicans, who mostly championed the requirement, are giddy that the Justice Department upheld their efforts (rare for a southern state with a history of slavery and racial injustices). Virginia’s law is moderate in terms of what it allows as a legitimate form of legal identification (the law actually expanded the definition); nevertheless, the GOP sees this as imperative to ensuring the integrity of the voting process and ensuring the “one-person, one-vote” principle. Democrats, meanwhile, have cried foul in voter ID cases across the country. They claim that voter ID laws unnecessarily curtail turnout, especially amongst minority voters (who happen to lean heavily Democrat). This development—the implementation of a voter ID law in perhaps November’s second-most competitive and important swing-state—led me to revisit Nate Silver’s excellent piece on the New York Times’ Five Thirty Eight blog. How much do voter ID laws actually suppress turnout?
The short answer is “not by much.” To be sure, voter ID requirements affect some would-be-voters, but not all of them Democrats. Most studies focus on the percentage of registered voters that would be turned away with voter ID laws, or forced to cast provisional ballots, rather than the electoral consequences to particular political parties. But if, say, educational attainment was the key factor in deciding the likelihood of having a valid identification—as some studies indicate—white voters lacking college degrees, who tend to vote Republican, would also be affected.
Silver notes that most voter ID studies indicate that a stringent voter ID law may curtail, at most, about 2% of the registered voting population. Whether this is statistically significant is a different matter, depending upon the size of the sample or registered voting population. But it is well below the 9% proffered by many vociferous opponents of Pennsylvania’s new law. I think several truths mitigate the 9% figure:
- A vast majority of adults have some form of identification that can pass for voting purposes, and most states, with the exception of Pennsylvania, have moderate forms of voter identification requirements (Virginia’s new law expands the list of acceptable forms of ID to include utility bills, paychecks, bank statements, government checks, and college ID).
- Many people who do not have proper identification are not registered to vote and are unlikely to turn out to vote if they are registered.
- Poll workers may not enforce consistently voter ID laws at the local level.
- In most states with voter ID laws, voters without proper identification may cast provisional ballots, counted later in the event of a recount, dispute, or upon subsequent presentation of proper ID.
- Political campaigns may educate their supporters about voter ID requirements and engage in grass-roots operations to obtain proper documentation.
- Only two presidential swing-states have enacted voter ID laws to date: Pennsylvania, and now Virginia.
In all, Silver concludes, “it wouldn’t take that much in terms of increased voter engagement—and increased voter conscientiousness about their registration status—to mitigate them.”