In crafting any policy, two equally important questions should guide the process. First, and rather obviously: “what is the desired goal?” This can be aspirational, although ultimately has to be informed by pragmatic considerations of getting the policy adopted and getting stakeholders to buy in to it.
The second question is much less intuitive, and is too frequently absent from the process: “how do we want this policy to fail?” Any system designed and implemented by humans will include errors and even deliberate exploitation. It is inevitable. With that in mind, we should be designing policies that tilt those errors in the least damaging direction. Perhaps the most famous example of this analysis is William Blackstone’s commentary on criminal procedure noting that “the law holds that it is better that ten guilty persons escape than that one innocent suffer.” We are not going to get it right all the time, so how do we want to get it wrong?
In the context of voting laws, the answer to the first question should be obvious. The goal of all election laws should be to capture the maximum number of eligible votes. It is certainly true that every fraudulent vote nullifies a legitimate vote, which of course is the rallying cry of conservatives complaining about election integrity. But when this allegation is made, the first response should be to highlight that such fraud is only a problem if we agree that the goal is to capture the maximum number of eligible votes.
That’s where the second question becomes dispositive. Of course it is possible to create election laws that make fraud less possible. But if the outcome of such a law is to prohibit more eligible votes than it does fraudulent votes, it is self-evidently a bad policy. For example, this election featured multiple complaints about matching signatures on mail-in ballots to those on the registration lists. If the purpose of such a requirement is to ensure that someone does not cast a ballot in someone else’s name, then its effectiveness has to be measured by determining whether the number of fraudulent ballots excluded this way is greater than or less than the number of legitimate ballots excluded. It should be just that simple, and the answer is clear — strict enforcement of signature matching will exclude more valid votes than fraudulent votes.
Of course, this analysis does not address any of the underlying moral questions. After all, modern elections simply do not see coordinated voter fraud efforts. By contrast, restrictions on voting are routinely targeted at marginalized communities. But despite the overwhelming evidence, it’s still pretty challenging to get conservatives to acknowledge that their position is racist as well as undemocratic. If election policy is to be driven by moral considerations, the appropriate election law policy is even more obvious. Increase access to the ballot box, even at the cost of increasing the opportunity for election irregularities.
Thus, if the new Congress does its job, and passes meaningful election reform, I sincerely hope that the 2022 election will see a small uptick in election irregularities and even a small uptick in actual election fraud, accompanied by a significant increase in the participation of eligible voters. If so, the result will be an even more accurate election than we just completed.