Voter ID: How Security Theatre Can Win Elections

As a number of people have observed, there is no obvious means by which taking nail clippers away from little old ladies prevents healthy young men from using weapons already secreted aboard an airplane to try to hijack it. Similarly, all the aggressive increases in ID checking seem irrelevant; none of the 9/11 hijackers were using assumed names. And yet, for the most part, people have put up with the TSA’s hijinks. Why?

The answer is best characterized by the classic syllogism:

• Something must be done.
• This is something.
• We must do this.

When confronted with a problem, people are inclined to accept a proposal that purports to do something about it, whether or not the proposed response will actually solve the problem. However, there is more to it; not only is it unnecessary that the proposed action solve the problem, it is quite possible to use this to get people to accept a course of action which has entirely unrelated, or even contrary, effects.

The push for “Voter ID” laws is a brilliant example of this. Everyone agrees that vote fraud is bad; the question is, how shall we stop it? Enter “Voter ID” laws. The theory is that, by requiring people to show their ID, we will verify their eligibility to vote. This sounds, on the surface, plausible.

Since I’ve gotten in a few discussions of this over the last few months, I thought I’d gather together some points on the issue. I propose to demonstrate that “Voter ID” laws will not measurably reduce fraud, will disenfranchise many legitimate voters, and have been drafted primarily in terms of that effect.

Point 1: Voter ID laws can’t prevent the fraud which is actually occurring.

There is a fair amount of dispute about how much vote fraud is really happening. One source pushing for voter ID laws claims that over a thousand felons voted in the 2008 Minnesota senate race. On the other hand, the League of Women Voters report that only 14 people out of approximately 2,800,000 voters fraudulently cast ballots during the 2004 election in Minnesota.

However, a closer look reveals the essential bait-and-switch tactic: Even if we grant that felons are voting, who shouldn’t have been, they are not voting by pretending to be someone else. An ID requirement would not prevent them from voting. In fact, the claim that over a thousand felons voted is itself false.

And that, really, is where the whole thing breaks down: None of the recorded fraud is of a kind where requiring ID would change the outcome of the fraud. A more detailed review of vote fraud does turn up examples of impersonation; roughly ten cases in the entire country in twelve years. So voter ID requirements wouldn’t change even one percent of the fraud cases identified. So why bother? What outcomes might it change?

Point 2: ID requirements will disenfranchise people selectively

Obviously, in principle, everyone can have ID with a correct address. However, some groups of people are much less likely to have such ID in practice. Poor people, for instance. Students.

Defenders of these laws argue that people could get IDs, but this is not quite on point. Even assuming everyone attempts to get a suitable ID, the barriers to success are much higher for some people to others. It’s hardly sane to dispute that, whatever your proposed task, it will typically be much harder for people who are very poor to accomplish it. People who have never been in these situations tend to underestimate the degree to which it can be exceptionally hard for people working menial jobs to accomplish bureaucratic tasks. There is no guarantee that an employer will grant someone time off to go get an ID. You could in theory mandate that by law, but laws giving poor people rights do not have much impact in practice; it turns out that if you’re poor enough to have to work a job where the boss won’t let you go to the DMV during business hours, you are probably not able to afford a lawyer. (And you probably don’t even know how to go about trying to find someone to take a case pro bono, for that matter.)

Students who live at a college campus frequently have ID that reflects their summer address, not bothering to update it during the school year — when voting happens.

In short, the real objection isn’t “it costs money”, it’s “there are many people who do not have the time or energy to spare to update their ID.” Something which only disenfranchises ten or twenty percent of a narrowly-targeted group is less destructive than something which disenfranchises all of them, but it’s still serving the same purpose. Overall, nationwide, about 11% of voters don’t have suitable and current ID.

No matter how elaborate the attempts to make ID reliably and cheaply available, the reality is that not everyone will successfully get ID; if this prevents them from voting, it has disefranchised them.

Point 3: That is, in fact, the goal.

There’s a fair amount of evidence that the selective disenfranchising of people who are likely to vote Democrat is not an accident, but is the underlying motivation of these laws. The most obvious source, of course, would be the former Republican party chair of Florida stating directly that this was the intent.

There’s no doubt that what the Republican led legislature in Florida and Governor Scott are trying to do is make sure the Republican party has an advantage in this upcoming election by reducing early voting and putting roadblocks up for potential voters, Latinos, African Americans to register and then to exercise their right to vote. There’s no doubt. I was in the room. It’s part of the strategy.
[…]
In three and a half years as Chairman in Florida, I never had one meeting where voter fraud was discussed as a real issue effecting elections. Never one time.

Admittedly, this from a man who is now disgraced and in a great deal of trouble — but that doesn’t create a particularly coherent motive for him to suddenly start lying about an unrelated topic.

Or consider Pennsylvania’s voter ID law. Of note is that the Republican House majority leader for Pennsylvania characterizes the law as “gonna allow Governer Romney to win the state of Pennsylvania.” How can they be so sure? It helps that they already have the statistics in place.. They already know that they have hundreds of thousands of voters who lack ID that would allow them to vote, and that the proportions are much higher in areas that have traditionally-Democratic demographics. During promotion of the proposed law, the Republicans claimed that fewer than 1% of voters lacked suitable ID; it’s actually around 9% in Pennsylvania.

And this is, after all, the point. The goal is to disenfranchise people.

When looking at the rhetoric on this topic, it’s important to note that there’s a lot of emphasis on the emotional importance of preventing fraud from “stealing” elections — and none at all on the importance of preventing selective disenfranchisement from “stealing” elections. There’s a lot of language about how the high costs of implementing these laws are the “price of a free democracy”, but there is no evidence at all that they will make elections any more representative of the intent of legitimately eligible voters, or that they are even intended to. They are aimed at people who are already a little bit marginalized, trying to marginalize them enough further to keep their votes from being cast or counted.