Is John Oliver Right About Voter ID Laws Being Problematic?

While many are focused on the death of the late Supreme Court Justice Antonin Scalia and what his death means for the Supreme Court, I thought that I would look at something different. Earlier this week, John Oliver started his new season of Last Week Tonight. I don't agree with John Oliver on everything he says, but he's generally more spot-on with political and social commentary than Stephen Colbert or John Stewart have been. With which topic did John Oliver start his season? Voter identification laws. The argument for voter identification laws is that it's a common-sense law not simply because voter identification is used to preserve the integrity of the electoral system, but also because identification is used for a number of other transactions, including renting a car, acquiring a gun, applying for food stamps, getting married, buying alcohol or cigarettes, or getting a mortgage. It is common-sense enough where 36 states in the United States have some form of voter ID laws. Oliver argues that a) voter fraud is incredibly rare, and b) it is more difficult to acquire a valid form of voter identification than one would think. Is John Oliver right? Are voter identification laws worse than intuition would suggest?

Let's touch upon "dead people voting," which is a type of voter fraud where a dead person remains registered, and a living person fraudulently uses that person's name to vote. Proving the prevalence of this form of fraud is difficult not only because it's expensive or inconvenient, but also because the number of studies are few and between. The Pew Center found that as of 2012, 1.8 million dead individuals were still registered as voters. With a 2012 voter-age population of about 235 million, that has the potential to sway a national election by a maximum of 0.8 percent. With that being said, there are instances in which "dead people vote." According to Rutgers University professor Lorraine Minnite, voter-impersonation fraud is quite rare and not on the rise. I can point to the South Carolina case study in which there were 953 "dead voters" at first glance, but after a closer look, there were only five actual cases of fraud. I can point to a 2006 New York case study where there were 2,600 cases of fraudulent voting. Even with the anecdotal evidence that exists, we cannot definitely state the extent to which voter fraud is a problem, although this 2014 Washington Post article points out some good academic literature showing that concerns for voter fraud are largely overblown. Also, there is some intuition as to why it wouldn't be so prevalent. As Oliver put it, it's pretty stupid to "stand in line, risk five years and a $10,000 fine, all to cast one inconsequential vote."

Even if we cannot definitively say that voter fraud is so prevalent, we can still assess different issues with voter ID laws. One is Oliver's claim that it is more difficult to acquire a form of identification to satisfy legal requirements. The Brennan Center found back in 2011 that 11 percent of Americans did not have a valid voter ID. Some like to challenge the issues with acquiring a voter ID, such as those at the Heritage Foundation or Right-leaning National Review, but the Brennan figure seems to be a pretty well-accepted figure. That's the bad news behind voter identification laws. Oliver made a point about racial disparities in voter identification, which if true, would violate the Fourteenth Amendment. A 2014 report from the Government Accountability Office (GAO) found that while there was some decrease in voter turnout due to voter identification laws, there was not a sizable disparity between races. The same GAO report also pointed out that there were no apparent cases of voter-impersonation fraud, either. A report from Political Research Quarterly (Rocha and Matsubayashi, 2013) also confirms that there is no discernible difference in terms of voter turnout by race.

Ultimately, this creates some mixed feelings on my end. One the one hand, a valid form of identification is necessary for other interactions in life, and we should encourage individuals to better integrate into society. On the other hand, voter identification laws make it more difficult to vote. In this respect, it is like so many other government regulations in that it makes it more difficult for people to exercise their freedoms. If we are to consider the United States a representative republic, we should encourage each state to find the way to balance electoral integrity and voter freedom that works best for that state, although addressing the fraud at its source, i.e., during the voter registration process, would help more. However, that's how I feel in theory, and that's what the ideal should be. If proponents of voter identification laws are to justify such legislation, they need to prove, at the very least, that a) there is a significant prevalence of voter identification fraud, and b) voter identification laws are a significant deterrent to the problem. Prevalence rates can vary from state to state, which is why it should be handled on a state level, but proponents have a ways to go before proving a real need for the voter identification laws currently enacted. In the meantime, I'm going to maintain my skepticism about voter identification laws.


2-27-2019 Addendum: The National Bureau of Economic Research recently released a study on strict ID laws. The contention of the Left is that strict voter ID laws suppress enough voters from showing up to the poll booth and exercise their right to vote. This study (Cantoni and Pons, 2019) shows that strict voter ID laws do not have a negative effect on voter turnout or registration.