In 2011 Alabama passed a voter ID law that would go into effect in 2014. This sort of thing was usually subject to approval from the Justice Department, but the Supreme Court struck down a key part of the Voting Rights Act. Some regarded this as reasonable, since voting seemed to be going along reasonably well. This is the same sort of reasoning that indicates that a patient with diabetes should stop taking her insulin on the grounds that her disease is now under control.
Critics of voter ID laws, who are most often Democrats, contend that they are aimed at disenfranchising minorities and the poor. These are the people who generally tend to vote for Democrats. Proponents of voter ID laws, who are most often Republicans, contend that voter ID laws are critical for preventing voter fraud. Since I have written extensively on this matter before, I will simply note that the best evidence shows that voter ID laws do have a negative impact on poor and minority voters. I will also note that voter fraud does occur, but at an incredibly low rate.
Now that Alabama’s voter ID law is in effect, the state seems to have upped its game by stating that driver’s license examiners would no longer be working at thirty one offices in the state. As might be guessed, Alabama officials claim that this is the result of budget cuts and is not intended to make things harder for minority and poor voters (who tend to vote for Democrats) in upcoming elections. It is also most likely a coincidence that this is occurring prior to the 2016 presidential election.
In what must surely be another coincidence eight of the ten counties with the highest percentage of non-white voters will have the license offices closed. These eight include the five counties that voted most strongly for Democrats in 2012. John Merrill, Alabama’s Secretary of State, counters that the state is ensuring that voters can get IDs. All the counties still have Board of Registrars offices and they issue voter ID cards. The state also has a mobile ID office that is supposed to visit all the counties.
While these IDs are available, only 29 IDs have been issued by the mobile offices since the start of 2015 and only 1,442 have been issued in total from all sources. In response to concerns about these low numbers, Merrill insists that the fault lies with the voters, noting that “you can lead a horse to water. But you can’t make him drink.” He points to the existence of an advertising campaign to inform voters and the availability of the above mentioned IDs.
On the one hand, it is certainly tempting to agree with Merrill. As he noted, voters can get an ID other than a driver’s license and can do so in each county. There as, as he claimed, been a public awareness campaign.
If someone wants to vote in Alabama, it can be argued, then that person should take the effort to learn what she needs to do and make sure that she has the requisite ID. To use an analogy, for each class with a paper, I have a detailed paper guide that shows step-by-step how to do the paper and how it will be graded. I also have three videos on the paper and spend about 45 minutes in a class going over the paper. Despite all that, I always get at least 10% of the class who make it clear (usually by asking things like “so, what is this paper you mentioned?”) they have no idea about the paper. As such, Merrill’s replies have some merit.
On the other hand, there is the concern that the efforts to inform voters are not adequate. People who voted before the new voter ID law went into effect and did not happen to see the advertising campaign are likely to have no idea of the existence of this requirement. Those who are aware of the requirement for an ID might believe that a driver’s license is required and might have no idea that there is even such a thing as a special voter ID available. Even those who are aware of the law and the special IDs might face difficulties in getting an ID. Transportation could be an issue as could making the time to go get the ID.
Some people counter these claims by referencing their own experience. They already have a driver’s license, so they find it hard to believe that others would not have them. They have TV and the internet and free time to watch shows in which the advertising appears. They have their own car and time to do things, so they assume the same is true of other people. This is a natural psychological tendency, but the beliefs based on it can easily be in error. For example, when I was in grad school, I found it easy to get by without a car. It was fairly easy to walk two miles to the grocery store and walk back with a week’s worth of groceries. It was easy to just run or bike to campus. It was easy to run or bike to stores, the bank and other places. So, it would be natural for me to think this would be easy for everyone based on my own experience. However, I was well-aware that what is easy for me could be very hard for someone else in different circumstances.
Some refute these claims by arguing that even if it is not easy or convenient to learn about the special IDs and acquire them, people who want to vote should take the effort to check before every election to make sure of what rule changes might have occurred. These people should then be willing to take the steps needed to be able to vote and then take the steps needed to actually vote—no matter how challenging or inconvenient these things might be.
A reasonable reply to this is that since voting the basic foundation of democracy, the process should be made as easy and accessible as possible. To do otherwise is to disenfranchise people unjustly. As such, people should not need to keep up with rule changes nor should they have to have an ID to vote.
The usual counter to this takes us back to the start: the concern about voter fraud. It is, I certainly agree, right to take steps to prevent voter fraud. However, as has been established beyond all rational doubt, the amount of voter fraud in the United States is miniscule. The fraud that does occur is also of the sort that voter ID would not prevent. I also accept the principle that it is better to allow a voter to vote fraudulently than to disenfranchise a legitimate voter—especially given that even if a method of fraud prevention did work, it would be preventing an incredibly low number of cases of fraud while most likely disenfranchising a vastly larger number of people.
Since I do like to think well of people, I am willing to accept that the officials in Alabama are acting from the most noble of intentions and, despite the evidence to the contrary, are not trying to take steps to increase the chances of Republican victories. That said, the methods they have chosen will have no real impact on fraud—both because it barely exists and because the voter fraud that occurs is generally not the sort that can be prevented by IDs. These methods will, however, have a negative impact on voters and that is certainly wrong—at least if democracy is accepted as a good.
One essential aspect of a democracy is the right of each citizen to vote. This also includes the right to have her vote count. One aspect of protecting this right is to ensure that voter fraud does not occur. After all, voter fraud can rob legitimate voters of their right to properly decide the election. Another aspect of protecting this right is to ensure that voter suppression does not occur. This is because voter suppression can unjustly rob people of their votes.
Many Republicans have expressed concerns about voter fraud and have worked to enact laws aimed, they claim, at reducing such fraud. In response, many Democrats have countered that these laws are, they claim, aimed at voter suppression. Naturally, each side accuses the other of having wicked political motives. Many Democrats see the Republicans as trying to disenfranchise voters who tend to vote for Democrats (the young and minorities). The Republicans counter that the Democrats are supporting voter fraud because the fraud is in their favor. In many cases, these beliefs are no doubt quite sincere. However, the sincerity of a belief has no relevance to its truth. What matters are the reasons and evidence that support the belief. As such, I will look at the available evidence and endeavor to sort out the matter.
One point of contention is the extent of voter fraud. One Republican talking point is that voter fraud is widespread. For example, on April 7, 2014 Dick Morris claimed that over 1 million people voted twice in 2012. If this was true, then it would obviously be a serious matter: widespread voter fraud could change the results of elections and rob the legitimate voters of their right to decide. Democrats claim that voting fraud does occur, but occurs at such a miniscule level that it has no effect on election outcomes and thus does not warrant the measures favored by the Republicans.
Settling this matter requires looking at the available facts. In regards to Dick Morris’ claim (which made the rounds as a conservative talking point), the facts show that it is false. But the fact that Morris was astoundingly wrong does not prove that voter fraud is not widespread. However, the facts do. For example, in ten years Texas had 616 cases of allegations of voter fraud and only one conviction for double voting. In Kansas, 84 million voter records were analyzed for fraud. Of these, 14 cases were referred to prosecution with, as of this writing, zero convictions.
Republicans have argued for voter ID laws by contending that they will prevent fraud. However, investigation of voter fraud has shown only 31 credible cases out of one billion ballots. As such, this sort of fraud does occur—but only at an incredibly low rate.
In general, significant (let alone widespread) voter fraud does not occur although the myth is widespread. As such, the Republican claims about voter fraud are based on a myth and this would seem to remove the foundation for their claims and proposals regarding the matter.
It could be countered that while voter fraud is insignificant, it must still be countered by laws and policy changes, such as requiring voter IDs and eliminating early voting. This does have some appeal. To use an analogy, even if only a fraction of 1% of students cheated, then professors should still take steps to counter that cheating for the sake of academic integrity. Unless, of course, the measures used to counter that cheating did more damage than the cheating. The same would seem to apply to measures to counter voter fraud.
One rather important matter is the moral issue of whether it is more important to prevent fraud or to prevent disenfranchisement. This is analogous to the moral concern about guilt in the legal system. In the United States, there is a presumption of innocence on the moral grounds that it is better that a guilty person goes free than an innocent person is unjustly punished. In the case of voting, should it be accepted that it is better that a legitimate voter be denied her vote rather than an illegitimate voter be allowed to get away with fraud? Or is it better that an illegitimate voter gets away with fraud then for a legitimate voter to be denied her right to vote?
My own moral conviction is that it is more important to prevent disenfranchisement. Obviously I am against fraud and favor safeguards against fraud. However, given the minuscule rates of fraud if attempts to reduce it result in disenfranchisement, then I would oppose such attempts on moral grounds. Naturally, another person might take a different view and contend that it is worth disenfranchising voters in an attempt to reduce the minuscule rates of fraud to even more miniscule levels.
Returning to the matter of facts, one rather important concern is whether or not the laws and policies in question actually result in voter suppression. If they do not, even if they do nothing to counter voter fraud, then they would be tolerable (assuming they do not come with other costs).
Unfortunately, the evidence is that the laws that are allegedly aimed at preventing voter fraud actually serve as voter suppression measures, mostly aimed at minority voters. Keith Bentele and Erin E. O’Brien published a study entitled “Jim Crow 2.0? Why States Consider and Adopt Restrictive Voter Access Policies.” Based on their analysis of the data, they concluded “the Republican Party has engaged in strategic demobilization efforts in response to changing demographics, shifting electoral fortunes, and an internal rightward ideological drift among the party faithful.” The full study, from the journal Perspectives on Politics, is available here. Since this is a factual matter, those who disagree with these findings can counter this by providing an analysis of equal or greater credibility based on supported facts.
Interestingly, it is a common talking point among Republicans that professors are tools of the Democrats and that academic experts should not be trusted. While this is a marvelous ad homimen, what is needed is actual evidence and arguments countering the claims. If professors are tools of the Democrats and academic experts are not to be trusted, then it should be rather easy to provide credible, objective evidence and analysis showing that they are in error. In terms of specifics regarding voter suppression, I offer the following evidence based discussion.
One of the best-known methods proposed to counter voter fraud is the voter ID law. While, as shown above, the sort of fraud that would be prevented by these laws seems to occur 31 times per 1 billion ballots, it serves to disenfranchise voters. In Texas 600,000-800,000 registered voters lack such IDs with Hispanics being 40-120% more likely to lack an ID than whites. In North Carolina 318,000 registered voters lack the required ID and one third of them are African-American (African-Americans make up about 13% of the US population).
Another approach is to make it harder for citizens to register. One example is restrictions on voter registration drives—Hispanics and African-Americans register to vote at twice the rate of whites via drives. It is not clear how these methods would reduce fraud. The restrictions mostly do not seem to be aimed at making it harder for people to register fraudulently—just to make it more inconvenient to register.
A third tactic is to reduce the available early voting times and eliminate weekend and evening voting. This would seem to have no effect whatsoever on fraud, but seems aimed at minority voting patterns. In 2008 70% of African-American voters in North Carolina cast their ballots early. Minority voters are more likely than white voters to vote on weekends and in the evening. For example, 56% of the 2008 weekend voters in Cuyahoga County in Ohio were black.
A fourth tactic is to make it harder for people with past convictions to regain their voting rights. This impacts African Americans the most: 7.7% of African-Americans and 1.8% of the rest of the population have lost their right to vote in this manner. This tactic does not prevent fraud—it merely denies people the right to vote.
It would seem that the laws and policies allegedly aimed at voter fraud would not reduced the existing fraud (which is already miniscule) and would have the effect of suppressing voters. As such, these laws and proposals fail to protect the rights of voters and instead are a violation of that basic right. In short, they are either a misguided and failed effort to prevent fraud or a wicked and potentially successful effort to suppress minority voters. Either way, these laws and policies are a violation of a fundamental right of the American democracy.