Dr. Moon Duchin: Redistricting and gerrymandering
SciLine conducts interviews with experts and makes the footage available to journalists for use in their stories.
What is Experts on Camera?
Expert on Camera:
This year, voters in Ohio will decide whether to use an independent commission—rather than a partisan one—to draw the state’s political maps. Other states are considering their own approaches to countering gerrymandering—the periodic redrawing of political maps to favors one party or group.
On Wednesday, October 30, 2024, SciLine interviewed: Dr. Moon Duchin, a professor of mathematics and public policy at Cornell University who studies the art, science and political consequences of gerrymandering. She has served as an expert witness in a range of redistricting lawsuits.
Journalists: video free for use in your stories
High definition (mp4, 1280x720)
Introduction
[0:00:20]
MOON DUCHIN: My name is Moon Duchin. I’m a professor of mathematics and public policy at Cornell University, and I study what I call data science for democracy. So—math and mathematical modeling interventions in thinking about elections and civil rights.
Interview with SciLine
What is redistricting? And what is gerrymandering?
[0:00:43]
MOON DUCHIN: So, where redistricting can be very general—it’s just when you take some territory and divide it up into pieces—but when we talk about it in the context of elections, we’re thinking about political redistricting, or you might take a state and divide it up into districts that are going to conduct elections. So, that’s redistricting, and in the U.S., we have to do it at least every 10 years after there’s a new census out, so we know how many people live in every place. Gerrymandering is when you have control of the redistricting process and you use that control to get advantage for one group over another group. So, it’s sort of—you can think about it as abuse of redistricting. It can be really hard to identify, and that’s one of the things that I study.
How can gerrymandered maps skew election outcomes?
[0:01:32]
MOON DUCHIN: The basic idea is: If you know where people live and how they’re likely to vote, then you can draw the lines in a careful way—kind of think about this as like designer districts—so that your group has efficient majorities, so that it wins without wasting a lot of votes in some districts. And then in the districts that your side is going to lose, you don’t want to have any votes at all. So, sometimes this is called packing and cracking, and these are two strategies that go together to get more control for your group and to marginalize or dilute the vote of other groups.
Can you give us any examples of gerrymandered maps?
[0:02:13]
MOON DUCHIN: So, the classic first example—the thing that gave us the word gerrymander—comes from right after the 1810 census. So yeah, that’s just about as old as the country. And there in Massachusetts, there’s a senate district north of Boston that was thought to advantage one political party over the other. At the time, the advantaged political party was the Democratic Republicans—all one party—over the Federalists. And that district was satirized in a famous political cartoon, where it was made to look like a dragon with wings and teeth. The idea was that it’s really snaky, so that it captures this population and skips over that population. The governor at the time was Elbridge Gerry, so that was Gerry’s salamander, and that’s what gave us the word gerrymander. And I mean that story is interesting for a lot of reasons, but one of them is how old partisan gerrymandering is in the history of the U.S. And the other is that today, if we look at that particular district, it’s just a collection of towns that goes right up to the border—the northern border of the state—it no longer looks, you know, it looks kind of benign from a modern point of view. So, that’s one classic historical example. And then there’s another famous example from just the last census cycle, just a few years ago, where in North Carolina, at the time, there were 13 congressional districts, and North Carolina had a really tight 50-50 voting pattern for Democrats and Republicans. But the district drawing was controlled by Republicans, and the head of the subcommittee pursued really aggressive lines that were thought by observers to lock in a 10-3 outcome favoring Republicans, even though the voting was more like 50-50. What’s notable about that one is that when asked in the press, why did you draw a 10-3 plan, he gave the answer, we couldn’t find an 11-2 one. We had to settle for 10 to 3. So, that’s a case where you don’t have to speculate about intent, because the intent was just completely spelled out.
How can gerrymandered maps affect voters’ perception of the integrity of elections?
[0:04:24]
MOON DUCHIN: So, this is something that the court has been fairly sensitive to. It’s particularly thought that if you live in a district that looks bad, like it has a lot of twists and turns, and the boundary maybe looks like a reptile, like the original gerrymander, that’s going to undermine people’s confidence in the basic functioning of their democracy. And that can have trickle down effects on people’s willingness to turn out to vote, on how the campaigns are run, on who runs for election, and so on. So, the idea is that we need kind of confidence in the legitimacy of that piece of the process in order for the whole rest to be healthy and well functioning.
Can you tell us a bit about independent redistricting commissions, like the one being considered for Ohio?
[0:05:13]
MOON DUCHIN: So, independent commissions are an extremely popular idea with voters. So, just the first thing to say is that, as far as I’m aware, every time it comes up for a vote, it passes. And that’s true in states across the political spectrum. So, in 2018 five different states put some form of independent commission in front of their voters, and all five passed. And that included Colorado, Ohio—again at the time, we’ll talk about that—Michigan, Missouri and Utah. So you see, it’s a whole variety of different kinds of states. And in all the states, voters went for the idea of an independent commission to take the line drawing out of the hands of the people who’d be running in those districts. So interestingly, what happened in Ohio at the time, a commission passed that would be bipartisan. It was going to have members from both major political parties, and some thought that that would be just as good as an independent commission, which doesn’t have political figures on it. But it worked out, as it did in other states, with a bipartisan design, that you had kind of a log jam, and the two sides had a hard time agreeing. And so there tend to be two sets of legal teams, two sets of technical line drawers, and it produced a lot of gridlock. So this time, Ohio voters are getting another bite at that and looking to make an independent citizens’ commission, and those are thought generally to function pretty well.
What might better laws and policies look like?
[0:06:40]
MOON DUCHIN: It can be pretty hard even to identify a gerrymander. And that means it can be really difficult to write rules and regulations that would prevent some of the worst ones from going into place without also blocking—having unintended consequences. So, writing rules around redistricting is going to be very difficult, even though it’s so important, because the whole detection problem is actually pretty subtle. I think you can see a good model of what that might look like if you go back a few congressional terms and you look at what was ultimately called the Freedom to Vote Act. So, that was an act that actually had some pretty significant support and ended up one Senate vote short of passage, but it contains—among many other provisions that have to do with conducting elections—it contains a test for gerrymandering. The only thing I’ll say about the details of that test is that—one of the things I think is really good about the way it’s written is that it gives you a test for what would be a presumptive, what we would presume to be a gerrymander, and then it puts a suspect map into kind of a penalty box, and then it puts the requirement on the people who drew the lines to defend why it’s not a gerrymander. So I actually like that structure. It says you might be a gerrymander, tell me why you’re not, and it gives people a chance to tell the story of the map.
Can you tell us a bit about your work using mathematical models to flag gerrymandering, and to help draw fair maps?
[0:08:24]
MOON DUCHIN: I run a lab—we’re an interdisciplinary lab, but we’re rooted in mathematics and mathematical modeling. And here, by mathematical modeling, all I mean is using math to build a structure that reflects the properties of the system that you want to understand. So, I think our biggest contribution to the study of gerrymandering was in building some powerful and efficient algorithms that make lots and lots of alternative maps. So, if a particular state has some rules in place for how they want redistricting to be carried out—what are the priorities and maybe in what order—what you can do is you can sort of tailor an algorithm to generate thousands, millions, even billions of alternatives that are all made just according to those rules. What’s useful about that—and courts have actually found this to be quite useful. I mean, of course, it has a mixed track record, but over and over again, courts have found this kind of idea to provide useful evidence, because what it does is it gives you a glimpse of what maps might look like if they were made only according to the rules, because the algorithms don’t implement some kind of hidden agenda to favor one group over another. So, it can be a really nice way to see just how where people live and the rules for districts combined to create a kind of baseline of representational outcomes.
