There is little doubt that Voter ID requirements are a hot-button issue in this election cycle, and the Commonwealth of Virginia has just thrown itself into the fray, becoming the 31st state to pass such a provision through its legislature. On Thursday, the House of Delegates passed SB 1 [Full Text] with a 66-31 vote. Supporters of the bill, mostly Republicans, assert that such a measure is necessary to combat voter fraud that is already widespread and influencing elections. Detractors, mostly Democrats, are labelling such measures as a thinly veiled attempt to suppress voter turnout or a return to “Jim Crow.” With all of this rhetoric, it’s hard to know what the real effects of such a bill are.
SB 1, and an identical House bill named HB 9, are in a category of ID laws known as “Non-photo ID laws.” This means that they remove the option to cast a full ballot with a sworn statement affirming the identity of the voter, but also expands the list of acceptable forms of ID. Any voter who does not present an acceptable form of ID at the polls will instead be given a provisional ballot, which will only be counted if the voter presents acceptable ID by email, fax or in person before the vote count is finalized on the sixth day after the election. The list of acceptable forms of ID is as follows: A Virginia voter registration card, social security card, Virginia driver’s license, any other ID card issued by any state or by a municipality within the Commonwealth of Virginia, a valid student ID from a 4-year institution within the Commonwealth, a valid employee ID card containing a photo of the employee, or a copy of a current utility bill, bank statement, government check or paycheck that shows the voter’s name and address.
This extensive list of acceptable identification makes Virginia’s bill one of the more lenient of this trend. 6 states require photo ID only (with 2 more states pending Department of Justice approval for the same) and 7 states require photo ID unless the voter meets another requirement such as being recognized by a poll worker or answering personal questions.
With such an extensive list, comes a number of opportunities for a competent fraudster to evade the new requirement, but the evidence is limited that such a problem exists. In Georgia, Politifact reports that between 2009 and 2011, the Secretary of State investigated 647 cases of election or voter fraud. The number of convictions, however, is not available in the public record – even if it were, 647 extra votes are highly unlikely to change the outcome of an election. Florida’s numbers are significantly lower, with 33 cases investigated during the same time frame.
As for the effects on the voter turnout, a study by the Election Assistance Commission found that there was a 2.9% lower overall voter turnout in states with voter ID laws, though the effect was different on varying subsets of the population. Less educated voters were 5.1% less likely to show up to the polls, while racial minorities had a turnout that was reduced by 6-10%. The numbers from other studies are somewhat less conclusive, with comparisons of voter turnout before and after a voter ID in the same state showing no significant difference in turnout. Both of these methodologies, however, fail to take into account other factors, such as voter enthusiasm for candidates in those states for those election years. A study by the Harvard Law & Policy Review, however, does take these factors into account and finds a small – though significantly significant – drop in voter turnout after the implementation of such a law. The effect that it found was that of a 1.6-2.2% reduction in turnout [Full study results] (PDF).
When it comes to Voter ID laws, then, it comes down to the lesser of two evils. The state legislatures now face the choice of a few cases of voter fraud or voter disenfranchisement. Virginia’s General Assembly has chosen, and the Commonwealth is waiting for Governor McDonnell to decide whether or not he agrees.