How Much Do Election Law Changes Affect Voter Turnout? Featuring Justin Grimmer

Transcript

Tom Church:
Howdy. I'm Tom Church of the Hoover Institution, and this is Factual Foundations of Policy. Today, I'm trying to figure out how voting reforms like voter ID or no excuse, mail-in voting affect turnout, especially on partisan lines. And to do that, I'm speaking with Justin Grimmer. Justin, thanks for talking with me today.

Justin Grimmer:
Thanks for having me.

Tom Church:
So Justin, I'm talking to you for many reasons. One, you're a political scientist here at Stanford University. You're a senior fellow at the Hoover Institution, and I've been following a lot of your research following the contested election of 2020 research you've done with Eitan Hersh, investigating claims of election fraud or general Tomfoolery in that election and 2022. So can you tell me what the focus of your research has been since then on that topic?

Justin Grimmer:
Sure. Yeah. There's two broad parallel tracks that we've progressed along. With the research with Eitan, we've examined what happens when states change their election rules, and there's a lot of debate about this. In fact, you might hear from the left that various election law changes are harbinger of the death of democracy or could completely foreclose elections from one side winning or not. It turns out those claims are vastly overstated. In a paper we just published in the Journal of Legal Analysis, we show that in fact, most election law changes have minuscule if no effect on the partisan balance in an election. The reason for that's quite simple, it's because most laws will target only a small share of the population. And then even among those folks who are targeted, the laws have a small effect. And even among those on their turnout, and even among those people who have the small effect on their turnout, it turns out the laws tend to target Republicans and Democrats in a pretty even manner.
So we're talking in large electorates, you might have policies that are quite controversial that affect 100 people, for example. So that's not going to swing even the closest elections in the United States. In a parallel track, I've done a series of papers and research investigating prominent claims about election fraud in the 2020 election and subsequently. And what's interesting there, I think there's a number of interesting things. One particularly interesting point is that experts who are advancing these claims of fraud end up duping themselves into believing that they've uncovered some evidence of fraud or source of fraud. But then when you look at it with a truly professional lens, you apply basic statistical analyses and some core scientific principles, you find that in fact, all of these claims fall apart under scrutiny. And so there's actually no good evidence of broad systematic voter fraud after the 2020 election or in subsequent elections.

Tom Church:
All right, so two avenues for us to talk about today then. So it's those three things, the framework, the three reasons why these policies that people worry about on the left and the right don't have the effects that you really think. And then of course, we'll go through some 2020 fraud accusations. So again, those three things are voting reforms apply to a small number of people, the actual relationship between who is affected and the laws is quite small as well. And then turnout effects left and right equally. So I mean, is that true for things like voter ID? I feel like most people will hear these things and think, okay, that is going to benefit Republicans, that is going to benefit Democrats. So how does that work?

Justin Grimmer:
Yeah, I think there's some basic intuition that leads people to believe, oh, a voter ID law is going to really benefit Republicans, or no excuse absentee voting is going to be huge for Democrats. That intuition is in part because of how people think about voting and because of a small logical error that's very easy to make as we reason about these policies. That error is that we might confuse a group being disproportionately affected by a law with the law resulting in a big difference in that group's turnout. So let's just walk through an example with voter ID, which I think will really illustrate the principle nicely.
I published a paper actually back in 2020, well, it's back in 2021, but I did the work in 2018 with my former grad student, Jesse Yoder. And we examined the North Carolina voter ID law that was in place for a primary election and then actually taken off the books for the general. And North Carolina published a series of just really good data sets for figuring out the effect of the law. Not only did we know everyone who has registered to vote North Carolina and whether they turned out to vote, we also knew whether the state of North Carolina thought that individual had a driver's license or some sort of valid piece of identification, something that would comply with the law. So using this, we're able to get what I believe to be the most precise estimates of the effect of photo identification law. So the first thing to note is that in a big state like North Carolina, something like three to 4% of people don't have valid identification, that's an overestimate because among those individuals, they could have a passport or other kind of identification the state doesn't know about. Then within that three to 4%, we less likely did the law make people without identification to turn out relative to people who had identification. And there we find there's about a 0.7 percentage point decline in turnout.
Now that might be surprising. You might say, well, if I don't have ID, how do I vote? Part of the reason the effect is so small is that individuals without identification weren't turning out to vote very often already. So there's a pretty low baseline to work from. The second thing is, of course, they could have other identification, and the third thing is they could find a way around the law, for example, by going out and getting valid identification, including going and picking up a different sort of state issued ID. Then we can dig into who benefits from this law. One of the great things about the North Carolina dataset is that we have partisan registration. So we know if people registered as a Democrat or Republican. And if I'm remembering the research correctly, when the primary election is in place, we're looking at something, the total of this law is something like 400 more Republican votes on the margin than democratic votes. So that's 400 real people. That's a difference that we might care about, but even the very closest statewide elections are not decided by 400 votes. So even though it's the case that Democrats were disproportionately targeted, it's not the case that this manifests in some big difference.

Tom Church:
And is it fair to say that that's just one first order effect? Because I know down in Georgia, they were also generally worried about voter ID, making fewer Democrats be eligible to vote, and then turnout went up there as well because they used it as an issue. So how do you disentangle first and then second order effects of something like an unpopular or popular voting reform?

Justin Grimmer:
I think this is a exceedingly interesting issue, and that's why a lot of these designs that we use are trying to make comparisons say within the state. So if we're in North Carolina, everyone within North Carolina would have heard that information about this being an issue or this being important. Of course, we could imagine that individuals with and without identification might hear it at different rates or get differentially excited about it. But that would in part explain why we'd see the decline among individuals without identification. So if individuals with I were more likely to because they were excited about the law, then you see a bigger gap.
There has been some work to examine this. And there's other published research that makes the claim that what's going on with photo identification laws, the reason they don't have a big effect is that there's what they call a psychological counter mobilization. People just get so gosh darn angry about it that they decide I need to turn out to vote. Unfortunately, that research doesn't really support that conclusion when you scrutinize it a bit more closely. So that work is based largely upon a series of experiments that was conducted in a survey context. And in that survey context, the individuals running this experiment gave people different messages associated with photo identification requirements. And the author's report that one message was particularly effective. And that was a message that said, photo ID disenfranchised Black voters. But when you dig into what the outcome was, the outcome there included five different components, only one of which was about turnout. There were also components about whether you'd like to help educate people about photo identification laws or if you'd like to donate or otherwise volunteer in an election. Obviously those have nothing to do with turnout.
So to the credit, the authors of this study, Eitan and I, contacted them and said, we'd love to know the results just for the turnout DV, and they weren't able to share the data with us, but they were able to run that analysis, and they found that even that message that they thought would be very effective had no effect on individuals reported intention to turn out, let alone whether they actually turned out. So this idea that people could get excited, it's intuitive, it's of course logically possible. We just don't have very good evidence that it happens all that often.

Tom Church:
People like to fight about politics and elections, but over the last, I want to say 30, 40 years voting reforms have made it easier to vote. In general, we've really lowered the bar to register to vote to go to the polls to even get a ballot in the mail, right?

Justin Grimmer:
On nearly every front. It is easier to vote. I wouldn't say lower the bar, I would say expanded access. I think it's something to be celebrated, and it's unfortunate the rhetoric that we sometimes hear about reforms closing off democracy from certain groups, it couldn't be farther from the truth even compared to 20 years ago. It is so much easier to cast a ballot in America today.

Tom Church:
And so what has that done to voter turnout?

Justin Grimmer:
That's a good question. We are at a high point of voter turnout. It doesn't necessarily coincide with these reforms. So there's some evidence that some of these reforms will drive up turnout. So for example, no excuse, absentee voting seems to increase turnout maybe one percentage point.

Tom Church:
It's not that much.

Justin Grimmer:
It is not that much. And even some of the best work there would suggest that might be an overestimate. So my colleague in political science, and actually just in the Hoover Institution now, Andy Hall, has published the best papers on, no excuse absentee voting and its effect on turnout. They have one field defining paper that came out in 2020 that suggested there was something on the order of a one percentage point increase in turnout, but no real effect when it comes to the partisan balance and who's elected. They did work after the 2020 election using 2020 data, and it was a really creative design. So in Texas and Indiana, when you turn 65, you no longer need an excuse to vote absentee. So before that, you have to say, "I'm sick or I'm out of town." But once you turn 65, you can just say, "Sounds like a better way to vote." And what they find is that there's a big switch in the way people vote. So people 65 and over really like voting absentee. That's how they decide to cast their ballot.

Tom Church:
And so how much more likely are they to vote in general, right?

Justin Grimmer:
0% percentage points more likely. So that doesn't change the overall turnout rate at all, just changes the way people decide to cast their ballot. So what's going on? How could this explain why people are turning out or not? We might think... Just separately before answering that question. There's lots of reasons why we want voting to be a thing that makes sense for people. That's relatively easy. I personally, on a normative grounds, like the idea of an expansive electorate, I think it helps people believe in the system. But the big things that seem to drive people to turn out to vote are when they're excited about a candidate or they really don't like another candidate.

Tom Church:
Hence why we're in very high turnout mode right now.

Justin Grimmer:
Yeah, exactly. So 2018 Democrats collectively have their hair on fire about Donald Trump. That carries over to 2020 where both sides were staring into the abyss, but very excited about their candidate, but very excited to vote against the other candidate. So we'll see in 2024 where some of the passion seems to certainly have died off, at least from the democratic side, we'll see if Republicans can muster excitement for Trump again, and we'll see if that continues. But that really seems to be what's pushing turnout around.

Tom Church:
So as someone who studies elections, whether they're run freely, fairly, how to measure if people are voting illegally, are you at all worried about this enormous quick shift from most people voting on election day to voting before election day, whether early in person or mail-in voting?

Justin Grimmer:
I'm not about it in a free and fair sense, there's of procedures in place to insure ballot that are cast are being cast by the person who is claimed to be casting them. So we could go through the safeguards. From the application, to the delivery, to the return, to the validation. There are so many safeguards in place. And even if there weren't, the coordinated effort to do anything that would be meaningful for fraud is so extensive and would require such a large conspiracy that it almost certainly would be detected. So that doesn't concern me.
But what does concern me is that when we make this move, people are seeing very different sorts of election night results. So they're seeing it take much longer to count ballots or it's drifting into a week after the election before we have a determined winner. And that I think undermines confidence in the election, gets people very worried. So in a paper I wrote with number other coauthors, including Sean Westwood, we asked people who were skeptical about the election, well, why are you skeptical? One of the most cited reasons was that it just took so long to count the ballots. And because individuals don't understand those ballots have to already be returned or already be postmarked depending on the locality you're in, it gives the impression that these ballots appear out of nowhere even though we've known that they've been in the basket the whole time. So from an actual safety reason, I'm not concerned. But in making the case to the public that they should trust the election, that becomes concerning and that becomes a place where I think we could do a lot of work.

Tom Church:
I talked to Ben Ginsberg about this, and we went state by state in terms of procedures to verify votes and then actually count them and how states are not uniform. Why states all uniform? Why doesn't the federal government step in and say you need to all do it this way?

Justin Grimmer:
Well, the United States Constitution probably has a lot to do with that, allowing state legislatures to set the time and manner of a number of elections. So I'd say that's the primary reason. And that local control states really like. It is tempting to think that the federal government could come in and solve this problem.

Tom Church:
Big picture. They could say, "Listen, you have to be able to count votes before election day that come in three days before that," something along those lines.

Justin Grimmer:
So the federal government could come in and say that. I think we know less about what the optimal way to run an election is than we might think, and that even though this diversity and outcome might be perceived to be a problem, there's also some advantages there. We're also getting a lot of mini little tests of different ways to run these elections. So certainly the federal government could say, you can at least prepare it for counting, which is what they do in Georgia. You can get ready to count the ballots. I would support that reform. I think there's, for example, very good evidence that not a lot of people take advantage of postmarking your ballot by election day. If you just tell people to get it in by election day, they'll get it in by election day. So I think that that's another just reasonable policy. But we know a lot about for when people return ballots, even when they can return them late, because we have states trying out these different policies. And so I think even the federal government coming in and offering some very strong legislation also would get dragged down in the courts.

Tom Church:
So before we turn to 2020 election and a lot of the claims that were made without much backing, I just want to back up and ask, how did you meet Eitan Hersh? How did you end up coming to work together? How does that happen with professors who are at different universities or maybe even think a little differently about issues like this?

Justin Grimmer:
So we are very good friends from graduate school.

Tom Church:
Okay.

Justin Grimmer:
So we met way back when. And we do think about problems very differently. Eitan is much more policy focused. He does a very good job of being tuned into the exact political implications of policies. And I tend to be a bit more methody in the way I think about things. But our difference in skills compliment each other, I think pretty well, and we really enjoy working together and having these long conversations where we can argue with each other a lot to reach a conclusion. So it's been a very fruitful collaboration. We're writing a book together right now.

Tom Church:
Wow.

Justin Grimmer:
So spending a lot of hours on the phone figuring out what that book's going to say, so.

Tom Church:
The book writing's a long hard process.

Justin Grimmer:
Long hard process.

Tom Church:
Yes. So let's turn to the 2020 election because there were a lot of claims made about what happened, who voted where votes were counted, all the rest of it. Maybe I want to start with illegal aliens, non-citizens voting. This has come up actually just even recently today based on a paper written over a decade ago or about a decade ago saying suggesting maybe 10, 20% of non-citizens were illegally registered to vote and can go through voting. So how do you approach that to say, how do you even start to investigate whether that might be true?

Justin Grimmer:
Yeah, so one way to do it is just to look at the actual evidence that's proffered for that sort of claim. So this paper is by Professor Jesse Richman. It was published in a good journal electoral studies, and it used a reputable data set called the Cooperative Election Study to reach this conclusion. And so that's a headline finding, that's going to make a lot of people very concerned. But when you dig into that, it turns out there's a very interesting survey response issue. And actually this same survey response issue came up in some other work that I've done. And just before I give that, I'll talk about what this other work is.
So there were a bunch of claims being made that some outlandish percentage of the US population supported using political violence against their partisan enemies. 10. 20% of the US public supported that sort of engagement and action, which is alarming.

Tom Church:
Seems high.

Justin Grimmer:
And seems high for a country that many people don't know there's three branches of government or can't name the speaker of the house.

Tom Church:
Right.

Justin Grimmer:
Right.

Tom Church:
Seems like-

Justin Grimmer:
Well, most Americans aren't really paying attention to politics.
Yes. But nevertheless, they're going to take a gun and shoot their neighbor. It seems improbable. It's the same basic issue. So here's the issue. Suppose I'm trying to use a survey to measure some very low prevalence population. Let's suppose it's the population of people who committed a very serious crime in the past week, and even let's just suppose that those people who committed that crime are going to truthfully tell us that they did. But suppose that that is a 10th of a percent of the population. Fortunately, not a very high crime rate for this very serious crime. But we are going to ask this question in an online survey, and our online survey is going to be filled with people who are doing a good job and paying attention. But unfortunately, some folks are just going to click around or maybe they'll misunderstand the response and they'll accidentally click that in fact, they committed this serious crime.
What you find is that when these folks are clicking around, that's going to bias the reported rate way up. For example, if there are four options. Among the people clicking around, they're going to have a 25% rate of performing this serious crime. Or if the same group of people said that and that they voted, that could be 25% if there's four options. The result of that is it's going to cause this massive upward bias in our estimate, and we're going to reach a conclusion that is not actually supported by the data.
So there's a very nice follow-up paper to the non-citizen voting claim by Steven and [inaudible 00:20:39] and Brian Schaffner, where they went back and asked the people who said they were non-citizens. They said, "Well, are you really a non-citizen?" They didn't ask it like that. Just ask, "Are you non-citizens?" Ask them again. And it turns out most of them are citizens and the people who are not non-citizens didn't actually vote. And so as a result, that number is much, much higher.
This isn't to say that there aren't ways in which non-citizens accidentally do get registered to vote. One policy area that I've done some research on with Jonathan Rodden, who's also at the Hoover Institution, is on this policy called automatic voter registration. You go to the DMV and either you opt in on the front end, you say, "I'd love to be registered to vote today," or on the back end. If you go in with the documentation for your clear ID, they will automatically put you on the registration roles. And what's interesting, I've been on panels about this policy quite a bit, and there will be very lefty pro-immigration groups on these panels advocating for the backend because individuals who don't have English as a first language or just trying to comply with what's being asked of them at the DMV who are also non-citizens may accidentally opt in on the front end, which you can't do on the back end because the state's verifying that you have all that identification.

Tom Church:
I see.

Justin Grimmer:
So it is possible that some people get registered to vote. But it's exceedingly unlikely that people are being registered to vote and then voting in federal elections, the consequences are dire. If you're an immigrant and you apply for a green card, you won't get the green card. You can't become a citizen. It's a very serious issue. And it's going to be one vote. It's not going to decide anything. So you don't have much incentive to commit the crime. A lot of punitive incentive to avoid committing the crime.

Tom Church:
Well, and states and counties and cities frequently go over their voting roles and have to cure them. I mean, isn't that something that would show up, you would presume along those lines?

Justin Grimmer:
It's a good question. So the curing or the maintenance of the roles, states will do that based on inactivity. So the Help America Vote Act regulates how often you can do that. So if an individual inadvertently registered and then doesn't act on that, doesn't make contact with the state, then they will be removed from the voter roll. So that's one way that could happen. But in states where you require photo identification, many of those photo identifications will also reveal whether you're a citizen or not.

Tom Church:
So you distinguish, in some of the papers you've written, between voting irregularities and false claims about the election. How do you disentangle those two? Because irregularities, I guess they always happen. I expect them to happen, but people assume them to be nefarious.

Justin Grimmer:
Yeah, it's a good question. So part of what goes on in this broad literature where people are looking for an anomaly is that it's very easy for humans to detect a pattern even when there isn't one. And so if we were to go into the 2024 election and we were to sit in a room like this, and we were to say, over the course of this evening, let's find the weirdest stuff that happened tonight. We would find weird stuff. We'd find stuff that had never happened in American elections before. By definition, there will be a county that changed their partisan vote composition more than any other county. And then the question is, when you see, that's just-

Tom Church:
It's math. Right. Yes.

Justin Grimmer:
Yeah. It must be the case that somebody does the biggest change. And so what the task of a statistician is, is going in and saying, okay, what actually is consistent with evidence of something nefarious, of some manipulation happening in this election or some sort of intentional manipulation amounting to fraud. And that's where there's a disconnect between a lot of these claims and whether the claims actually support the conclusion that the proponents are offering. And so every election, something will look weird, but everything that looks weird is not evidence that somebody is meddling.

Tom Church:
Right. Sure, sure. We're talking about voting reforms on the front end, but you also say that it's important to look at post-election policies and they deserve a lot of scrutiny. This is counting a ballot certification. What can we do to shine a light on that and make that better? Because it seems like we've lowered the bar, not the bar. What did you say?

Justin Grimmer:
Expanded access.

Tom Church:
Expanded access to voting for everyone in America. That's terrific. But on the other side of things, which is often more invisible, like you said, it's people just see vote counts start to tick up and go from there. What can we do to make that more trustworthy?

Justin Grimmer:
That's a great question. I think one of the biggest things we can do is educating the public about what is happening in those rooms. A lot of it seems opaque. Not because election officials haven't been trying to communicate to the public, but because people, most of the time are not all that interested in very marginal issues, like how is it a potentially contentious signature adjudicated, that is not going to make primetime TV most of the time? So educating the public about what actually is going on is very important. Then there are a number of small policy proposals that folks make. You are floated one that people like, which is counting the ballots earlier. That seems like it would just give less opportunity for people to be-

Tom Church:
Surprised.

Justin Grimmer:
On the counting. A policy proposal I'd like would be to get rid of a company called Edison or these instantaneous vote reports. The reason I don't like that, really two reasons. So first off, they make mistakes, and these mistakes fuel a lot of conspiracy theories because people don't understand the difference between the media company's report and the official vote totals. So because they don't understand the difference between those two, you have a conspiracy theories online where folks will say, "Isn't interesting that million ballots got deducted from President Trump's vote total?" It was just a spreadsheet error at Edison.

Tom Church:
Interesting.

Justin Grimmer:
And then the second reason I'm not a huge fan is it creates this impression that you're watching almost like a basketball game. In fact, the charts look exactly like a good summary of a basketball game where you'll have two lines going up, and of course, the line that ends up highest is the one that wins. When in reality, all of those ballots have already been cast. They're all there. So it's not as if ballots are being counted or cast after election day, they're all there already. It's just a matter of when we've logged and then reported in this case to Edison. And so, one small reform would be that we only report on the hour from an official data source. Secretaries of State could put Edison out of business by doing something like that. I think that would go a long way to increasing trust.

Tom Church:
Are there any states that do it really well that you wish other states would look at and adopt?

Justin Grimmer:
Florida.

Tom Church:
Florida. Yeah.

Justin Grimmer:
Yeah, Florida does a great job. That wasn't always the case, as you might remember. I think they learned a lot from that. Now, Florida just does a really good job with its post-election ballot counting. Gets it done very quickly. Not a lot of issues.

Tom Church:
It stops early voting three days early too, right? So is this the trade-off you're going to have to negotiate and argue with that, right? That oh, well, if you went to a Florida model from other states that allow early voting all the way up to it, well, you're taking away a few days of early voting from people, right?

Justin Grimmer:
That's interesting. So I think the biggest trade-off would be about this issue, about when absentee ballots have to be postmarked, and then how you go about counting them. So if you say they have to be received, you're going to be able to do it very quickly, received by election day. It is the case. You could end up losing two days of early voting. The evidence that those two days of early voting are going to affect turnout very much is not very high. It's also, there's not great evidence it's going to lead to longer lines of the polls or anything like that.
So in fact, one thing we know about when polls open, it's like when a new iPhone used to be released, there's just a dedicated group of people who are so excited that they're going to be in line first, and it'll create this appearance of a huge line. And that's because these people really just want to vote. They're very excited to either vote for their candidate or vote against a candidate. And then that dissipates, and then you'll have people use the early voting, but I don't think there'll be a huge trade-off. And I think the gain in trust would be huge if we move to a Florida model.

Tom Church:
Yeah. So other areas that made people skeptical, I think of elections or call them rigged, stolen. Have you looked into evidence on super PAC? So-called dark money on elections.

Justin Grimmer:
I have not spent a lot of time on Super PACs. No.

Tom Church:
How about felon disenfranchisement? Is that another one that disproportionately affects one of the other, or is it just fall under your, here's three reasons why this doesn't happen?

Justin Grimmer:
It falls under the three reasons. So we talk about it in our paper. And it turns out across the states, sometimes refranchising felon will help Republicans and sometimes it'll help Democrats. And the reason for this is that one of the most steadfast Republican groups right now are low education whites, in particular, low education White males, are very pro-Republican. And in some states, they comprise a large share of the felon population who can't cast their vote. So if those individuals are allowed to be enfranchised, then you would see some vote move towards the Republicans. At a baseline, people who have their voting rights restored don't tend to vote at a very high rate. So if you're released from having served a felony conviction, you got a lot of other things going on, it's going to be hard to get to the polls. So our best evidence seems to be that when felons get the right to vote, that causes a 10 percentage point increase in turnout. So it goes from zero to 10%.

Tom Church:
Oh, sure. Yeah.

Justin Grimmer:
And so that's a pretty big jump. Among a group who's targeted by law, that's one of the biggest that we'll see. But the composition of that group is going to be balanced. Even though in every state, black individuals are disproportionately likely to have a felony and therefore would disproportionately have their rights restored.

Tom Church:
So it varies geographically state to state, area to area?

Justin Grimmer:
Right, exactly. Yeah.

Tom Church:
What has been the best way you have found to try to open people's minds a little bit, who think on either side, the election is rigged, stolen, we don't have a chance, the other side is cheating. How do you get them to start to think, okay, how do I test that? What do I look into here? Where do I go?

Justin Grimmer:
This is the ever present challenge. So one of the things that I've actually found is most effective is walking them through this very simple arithmetic about how we go from having a policy and how that policy could possibly affect the results. It turns out that arithmetic would also affect voter fraud as well. So suppose you think there's this much more opportunity for fraud and only one party's engaged with it that you can make people see very quickly. It's very hard to get any meaningful amount of ballots actually counted. It's also helpful that people will offer specific explanations. So they'll say, well, I think certain events in Georgia lead me to believe that the vote was fraudulent. And then it's helpful just to walk through there and say, either that didn't happen, or again, the arithmetic just does not support that conclusion. I get people, they go through the arithmetic, they see it, but in the end, they sometimes will just rely upon this intuition where they say like, "Well, I just think that maybe there's something amiss here," which is fine, but can have that intuition. But usually we can dispel the specifics with this very simple math.

Tom Church:
Sure, sure. I wanted to ask about redistricting because it fits into I think a general response to this notion of, all right, well, these voting issues that you worry about don't have much effect because they affect people equally. So redistricting does or doesn't show up in the literature as having much of an effect on benefiting one side of the other?

Justin Grimmer:
Nationally, it doesn't benefit one side or another.

Tom Church:
Okay. So what do you mean nationally?

Justin Grimmer:
So for example, for the House of Representatives, gerrymanders in states end up compensating for each other, and so you end up with no real advantage because states are allowed to draw their own congressional districts.

Tom Church:
Sure. Okay. So that's nationally.

Justin Grimmer:
Yeah.

Tom Church:
But what do you say to the people who live in those states or districts, intra-district who now have less of a chance to reclaim a seat? Because we can say, listen, it doesn't affect politics in general, but all politics is local.

Justin Grimmer:
Sure. Thinking about redistricting and gerrymandering is actually really tricky to think about because there's two competing forces. One force could be, I want to draw districts that make my incumbents super secure. And so you would draw very partisan districts, and those incumbents would face very easy reelection. A different force you might have is to say, "I would like to maximize my party's political power." And to do that, you want to draw a large number of very competitive districts where you have a very marginal victory. So if you're in those competitive districts, that actually, I don't want to say gerrymandering's good or whatever, but in those districts, it's actually quite likely that it could be very competitive. Parties can make mistakes, and it can be likely that they elect a candidate of their partisan choice or a candidate that they like.
The other thing I would say is that, and this is a thing I talk to undergrads about a lot. Sometimes we look at a district that looks like a gerrymander who might actually be excited about it or think that it accomplished some other political goal. So one of the best examples of this is a district whose number, I'm not going to remember, but it's on the outskirts of Chicago. And if you looked at the district, it looks almost like the letter C. And you'd say, well, this is an obviously gerrymandered district, but that district has enabled Latino representation for decades. And so that's enabled for prominent Latino leaders to be elected to the US House. I think a lot of people would agree that that was good for representation, not just for the constituents of that district, but for a population in this country that are underrepresented within the US house. They can point to at least some number of co-ethnics within the house. And so that could be good. So it's always more complicated. And it's hard to know what the right district would be. There's probably no answer to that.

Tom Church:
Yeah.

Justin Grimmer:
Yeah.

Tom Church:
That's a classic professor answer. It's a trade off. It's what you decide.

Justin Grimmer:
I try to avoid those. That is a classic professor answer, isn't it? Yeah.

Tom Church:
Fair. Fair. All right. Well here, last one for you.

Justin Grimmer:
Yeah.

Tom Church:
What advice would you give to people who are still worried about elections in general? 2024 is going to be contentious again, who think, okay, we've had time to set things up, to rig it on our way, or to make sure that one candidate wins or another on the other side to stack the deck. What would you tell them to go and put their mind at ease?

Justin Grimmer:
I would actually tell them the best remedy would be for them to engage in the political process. So if they're worried that one side has done something to make it much easier, much harder for a group of people to vote, the best remedy is to volunteer, go out, get people who you want to see turn out, turn out to vote. Or even better, to volunteer as an election administrator. So work at a polling place. When people do that, I think they'll feel a lot better about the way elections are run. And then engaging with the political process actually helps individuals to feel efficacious. So I'll plug Eitan's other book. It's called Politics is for Power. And the key idea there is that we shouldn't think about politics as sitting around and watching Fox or MSNBC. Politics is about getting coalitions together and enacting things that you want to see. And so when people want to see candidates elected, they should work to elect that candidate. I think they'll feel a lot better about the process if they do that.

Tom Church:
Thank you for that positive note.

Justin Grimmer:
Thank you. Thanks for having me.

Tom Church:
That'll do it for Factual Foundations of Policy. If you'd like to hear more from Justin Grimmer, you can go to justingrimmer.org. If you'd like to follow me, you can do that at Twitter @TomvChurch, and you can follow the Hoover Institution @HooverInst. That's I-N-S-T. If you had any questions that you'd like to see answered, please send them in and we will try to get to them on future episodes. And as always, check the show notes where I'll make sure to put in all the sources citations that Justin, the professor, cited a lot of them. We'll get them in there for you. Thanks for tuning in. Talk to you next time.