I hate the way Paul votes, but what I hate more is that he wouldn’t have been able to vote in the 2016 election without the financial support of my sister. Yes, you read that correctly. He needed someone to give him money in order to vote. Let me explain.
Back in the late summer of 2016, my sister, Susan, invited Paul (not their real names), to live with her in Northern Virginia. Paul had been struggling significantly financially, so Susan and her husband invited Paul to stay with them to help him get back on his feet. Paul does not drive and that drastically narrowed the number of places he could work. Susan helped Paul find some local job opportunities to which he could walk or ride a bike. After a few weeks, he landed a job a the local grocery store. This is where things got very complicated. In the story that follows, Susan provided all of the transportation.
In order to be employed by the grocery store, Paul had to provide photo identification. As a non-driver, his North Carolina state ID had expired, and Paul hadn’t realized it. Neither Paul nor Susan understood the expensive and time-consuming journey that they would endure in order to obtain a photo ID for him. In the meantime, the 2016 election was quickly coming, and Paul realized he would need a photo ID in order to vote because Virginia had passed a Voter ID bill.
Over the past decade, several states have proposed Voter ID laws in one form or another. Some have been struck down by the courts, as in North Carolina, for being too obvious in their desire to disenfranchise many African-American voters, and others, as in Virginia, have remained on the books. On one side of the debate are those who believe that we have excessive voter fraud in our system. On the other side are those who are concerned that voter ID laws are merely vehicles for disenfranchising people of color and those who are poor. There there are necessary tradeoffs between accessibility and integrity when administering elections. When we make it easy to vote, we leave open the possibility that those with bad intents will game the system. But when we enact barriers to voting, many people effectively lose a fundamental, Constitutional right. There are several questions to be considered in this debate. What is an acceptable level of voter fraud knowing that no system can be foolproof? What is an acceptable level of voters who cannot vote to keep the system from corruption? And most importantly, is there presently a problem with voter fraud to cause all of this discussion?
We will address the third question first, as the answer to it drives answers to the other two questions. Depending upon whom you speak to, you can get very different answer as to whether there is a voter fraud problem. We will take a look at the research and methods used in several recent studies to figure this out. President Trump has stated and tweeted many times that there is a significant problem with voter fraud. However, that depends on what you consider a “problem” to be. The President appointed Kris Kobach, the Secretary of State of Kansas, to a commission to investigate voter fraud. To find voter fraud, the commission looked at the instances of people with the same name and birthdate across voter registration rolls calling it Crosscheck. Using this approach, the commission found that in the data for 21 states, there were an estimated 8,500 cases of duplicate voting. They extrapolated that data and came up with a figure of 40,000 cases of duplicate voting if all states were included. The commission was disbanded in January of 2018 due to increased litigation and many states not being willing to turn over their voter rolls. While these findings might seem quite concerning, there are problems with the data. In 2007 a study by McDonald and Levitt found that using only names and birthdates finds many false duplicates. They found that statistically among a group of 23 people, there is a greater than 50% chance that at least two of them will have the same birthday. If you increase the pool to 180, there is a 50% chance that two will have the same birthdate. This is relevant because if you take all of the William Smiths or Maria Rodriguez’s in the country, there are thousands, and there will be some with the same birth dates. Therefore, you need an additional identifier to be clear as to whether or not the voter is the same person.
Earlier this year, a group of researchers from Harvard, Stanford, University of Pennsylvania and Microsoft found that using only names and birthdates, as the Crosscheck system did, produced a list of many cases where there was no actual voter fraud. They found that by adding the last four digits of the voter’s social security number, the number of suspected duplicate votes plummeted to .02%. The study also looked at the effects of purging voter rolls on a regular basis. Based on the information they gathered on voters in Iowa, a purge would eliminate approximately 300 voters for every one fraudulent vote.
When we look at the data, we have to decide, is it worth disenfranchising 300 voters to get rid of one fraudulent vote? Is voter fraud really a problem when only .02% of votes are fraudulent? That is what the data tells us; you can draw you own conclusions.
Interestingly, in June of this year, a federal court judge found that the Kansas voter ID law which required residents to provide proof of citizenship before being allowed to register to vote, was unconstitutional and violated the Voting Rights act. Not only did the judge strike down the law, but the court concluded, “there isn’t a risk of noncitizens corrupting the electoral process in Kansas, but rather it is Kobach who has eroded confidence by disenfranchising voters with needless barriers to voter registration.”
A more promising development in identifying voter fraud is a non-profit, the Electronic Registration Information Center ((ERIC) which gathers information from the USPS, DMVs, and the Social Security Administration to keep track of voters and help states update their voter rolls. By using four unique identifiers, it is much more likely that the ERIC system will be able to identify real cases of double voting.
At this point, you are probably wondering what this all has to do with my relative, Paul. Please bear with me and the research so that you can get a full understanding of where I am going. When considering whether or not to implement voter ID laws, we also need to look at the number of voters who we are discussing. In a recent study, Stephen Ansalobehere of Harvard and Eltan Hersh of Tufts found that “across all registered voters in Texas, the researchers found 4.5 percent lack proper identification. For registered voters who actually showed up at the polls in 2012, it’s 1.5 percent.” While that is a small percentage, it represents 608,470 Texas citizens. In an article by Nate Cohn where he argues that enacting voter ID laws does not have a huge impact on election results except in the closest cases, he did note that “There’s no question that voter ID has a disparate impact on Democratic-leaning groups — those young, non-white, poor, immobile or elderly voters. The unmatched (voters who did not have a ID) North Carolina voters were registered as Democrats by a 37-point margin, compared with the 12-point Democratic margin statewide. They were 46 percent nonwhite, compared with 29 percent of all registered voters. But 22 percent of these voters were registered Republicans. The voters without an identification might be breaking something more like 70/30 for Democrats than 95/5.” While Mr. Cohn may feel that it is not that big of a deal, in our current system there are many elections that are decided by just a few votes (remember the Presidential election of 2000?), so it is likely that those who are advocating for voter ID laws understand that voter ID laws have meaningful chance to affect an election, and they are willing to disenfranchise some voters in order to increase their odds of winning.
In the Virginia voter ID decision, the Court of Appeals ruled that because the ID was free of charge there was no barrier for voters and let the law stand. Voters would get their picture taken, swear an affidavit, and be given a provisional ballot. However, they did need to bring a valid photo ID at a later time in order for their vote to count. However, the Court did not understand the ramifications of their decision. Let’s continue with Paul’s story.
When Paul went to the Virginia DMV, he learned that since his North Carolina ID had expired, he needed to present another form of photo ID in order to get a Virginia ID. He learned that a US passport was the only form of picture ID he could obtain that was acceptable. He needed a notarized copy of the birth certificate for the passport application as well as documents that had his signature and a date. He didn’t have his birth certificate, so he ordered one at a cost of $40 with expedited service. Remember, he had a job waiting for him. He waited. Once the birth certificate arrived, he purchased a passport for $60 plus $65 for expedited service. The passport photo cost $10.
The paperwork for such a passport application is unusual for the passport agents at the USPS, and the agent didn’t fill out the form the first time correctly requiring another application after 6 days. He went to the post office again and luckily had another photo, completed the paperwork again and resubmitted. The agent waived the application fee since it was his mistake in the application, and he waited again.
After about a week the passport agent contacted Paul requesting documents with his signature and date, such a rental agreements. They required at least five of these documents but suggested he send everything he could find. He collected a large packet and sent it off. After another 4–5 weeks he received his passport.
At this point, Paul still did not have a state ID because he had had to wait for the passport in order to get the state ID. By this time, the election was coming. Paul was lucky that he was able to vote in Virginia because he still did not have an ID that listed his address. Because Virginia had made the allowance for the sworn affidavit for voting, he was able to vote. However, he had to return with a valid photo ID after the election in order for his vote to count. Since Paul didn’t drive, Susan drove him to the center to have the picture taken. Paul finally received his Virginia state ID after bringing his passport to the DMV. To recap, it took $185, six or seven weeks and four trips to the DMV, passport office and Virginia voter’s registration office for him to be able to vote. If Susan had not been able to drive Paul around to all of these places and lend him the money, he would not have been able to vote.
Presently, the state of North Carolina is planning to put a referendum on the November ballot where voters will decide whether or not to add an amendment to the state constitution to require a photo ID for voting purposes. The conservative-leaning Heritage Foundation has kept track of voter fraud cases and found that from 2003–2017 in North Carolina there were a total of 10 illegal voters, 2 cases of individuals buying votes, 4 fraudulent voters, and 3 elections in Pembroke and Lumberton that were overturned due to irregularities in on precinct. That is total of 19 voter issues over 14 years in a state with a population of 10.27 million. Many who want the ID law say that since the ID is free, it’s not a big deal. However, they either ignore or don’t understand the cost of documentation for getting ID. Some of our older residents who live in rural areas may not even have a birth certificate on file at all. What will they do? Others do not have transportation or money to take care of getting updated identification. Paul was lucky to have had a family member who was devoted enough to make sure that he could vote. But there are many others who are not so lucky. Are we willing as a country to essentially take a Constitutional right away from the poor in order to ensure that an inconsequential number of votes do not get counted? If you were a North Carolina voter, like me, facing the voter ID amendment issue in November, how would you vote?