By Richard J. Batzler
In recent years, Wisconsin has been a battle ground over many controversial election law changes, including a voter ID requirement. I spoke with University of Wisconsin Professor Mayer about his research on the impacts of voter ID in Wisconsin and recent election law changes in Wisconsin and elsewhere.
Kenneth Mayer is a professor of political science at the University of Wisconsin-Madison. Professor Mayer’s election law scholarship includes campaign finance, voter identification, and election administration. Additionally, Professor Mayer has filed expert reports in cases involving voting rights, gerrymandering, and campaign finance, among other issues.
Could you talk about your participation in One Wisconsin Institute v. Nichol?
The One Wisconsin case was a challenge to a broad range of changes that Wisconsin made beginning in 2011, and the Voter ID law was part of that. I offered a conclusion about how many registered voters did not possess a driver’s license or a Wisconsin Department of Transportation ID, which are the two most common qualifying forms of ID under the Wisconsin law.
I went through a matching process where you have someone in the voter file and you have identifying information. A lot of this identifying information is provided under seal so it’s done in a secure way, and you determine whether a person in the voter file also appears in the Department of Transportation records of driver’s license and state ID holders. That provides an estimate of how many registered voters don’t have a driver’s license or a state ID.
The conclusion I reached was around 7-9% didn’t have ID, so there was a material percentage of registered voters who do not have a driver’s license, a state ID, or another form of identification. That number is analogous to results from other similar methods that have been used in litigation in other states.
After the 2016 election, you and Michael DeCrescenzo (Ph.D. Candidate) studied the impact of the Wisconsin Voter ID law on the 2016 election in Madison in Milwaukee. Could you discuss that study?
One of the challenges in trying to derive estimates of the effect of voter ID laws is that in epistemological terms you’re trying to estimate the prevalence of something that doesn’t happen, which is, how common is it for someone not to vote because they don’t have an ID.
We were interested in surveying registered voters who didn’t vote. We drew a sample of non-voting registrants in Dane and Milwaukee counties and sent a survey. It was a not a survey about voter ID. It was a survey about their experiences, trust in government, the reasons they didn’t vote, and forms of ID that they possessed. We used that to generate estimates of how many people who didn’t vote were either deterred or prevented by the law.
We found that about half the people who said they didn’t vote because they didn’t have ID actually had a qualifying form of ID. This has been increasingly recognized as one of the features of voter ID laws: voters are confused. These laws are complicated, and voters don’t fully understand what forms of ID qualify and what don’t. We estimated that around 14 to 15 thousand non-voters in Dane and Milwaukee counties were affected by the voter ID requirement and about 7 to 8 thousand were actually prevented from voting.
President Trump won Wisconsin by around 22,000 votes; I imagine that that point has been mentioned in response to your study.
Well, we were very careful in presenting the limits of what we did. This is a study of Milwaukee and Dane counties. You can’t take our numbers of 7% and apply that statewide to come up with a statewide estimate.
We did not ask people’s partisanship, we did not ask them who they would have voted for, because that’s not what we were doing. We were interested in the burdens that a voter ID law creates. But that’s certainly one of the things that people were interested in. Is it possible? Well, yeah, it’s possible, but that’s not what we’re saying.
The harm in my view is not a function of ‘does it help one party or hurt one party,’ but instead it’s erecting hurdles to a person’s access to the voting booth. It’s a harm to the person. It’s a harm to the integrity of the electoral process. I have testified in a number of voter ID cases and my conclusions are not offered in partisan terms. It’s not that the law is bad because it hurts Democrats and helps Republicans. It’s that the effect of these laws is that they impede access to the voting booth.
Well there are a couple of things. What has happened in Wisconsin is not unique. You have seen similar changes in other states. So it is in part a function of polarization, and it’s in part a function of achieving unified control of government and using that control to enact significant policy changes. There are a number of redistricting cases—Maryland, North Carolina, Michigan, here, and some other states—and I think one of the causes of that is that 2010 was a significant election in terms of creating in a number of states unified party control and a willingness to use the political power that that creates to enact significant changes.
This transcript has been edited and condensed for readability.