Demonstrators rally with cutouts of congressional districts in front of the Supreme Court in Washington on March 28 before oral arguments in Benisek v. Lamone , a redistricting case in Maryland. (Joshua Robert/Reuters)

With the midterm elections only weeks away, the shape of U.S. House districts is again front and center in political conversation. A central question is how much Republican-drawn maps might disadvantage Democratic candidates. To tackle this issue and the debate about redistricting generally, I talked to two scholars: Jonathan Rodden of Stanford University and Richard Pildes of New York University Law School. A lightly edited transcript follows.

Q. The big question on people’s minds is whether the Democrats can retake the House in November. There has been no end of analysis arguing that Democrats are hamstrung by the maps drawn by Republican legislatures. So how much could a gerrymandered map hinder the Democrats in 2018?

A. Rodden: The Democrats clearly have a problem transforming their votes to seats. This is true both in Congress and in state legislatures. Without a doubt, gerrymandering is a big part of that problem. But it is not the only problem. Democrats are highly concentrated in cities, which gives them an inefficient distribution of support across districts. Gerrymandering exacerbates this underlying problem.

Q. Is there an easy way to put a number on it? How many more seats would the Democrats have if the maps were “fair”? Or is that too hard to estimate?

A. Rodden: I’ve never seen the wisdom in trying to produce such a number, in part because of the difficulty of defining “fair.” Do we want to say that the map produced by a nonpartisan simulation is a “fair” map, even if it is severely biased against one party? How do we think about the Voting Rights Act? What is a “fair” number of seats for the Republicans in overwhelmingly Democratic states like New York? What is a fair number of seats for the Democrats in Tennessee?

Q. Could “fair” simply mean “seats in strict proportion to statewide votes,” or in proportion to the statewide partisanship?

A. Rodden: If we adopt that definition, then any disjuncture between votes and seats gets attributed to gerrymandering. That can’t be right. Because of the way voters are typically distributed across districts, a party with 55 percent of the votes will almost always get well above 55 percent of the seats. As long as its support is reasonably dispersed across districts, a party with 70 percent of the votes will win 100 percent of the seats. In a majoritarian system, vote shares and seat shares will be different for reasons that have little to do with intentional gerrymandering.

A. Pildes: There are two ways we can think about “fairness” here. The first might be called a process-oriented understanding. The second is an outcome-oriented understanding.

In the first, we might consider the process fair if redistricting doesn’t take partisan considerations into account at all and instead is based on other appropriate considerations, such as equal population, keeping towns, cities, counties intact to the extent possible, keeping communities of interest together. But in such a partisan-blind process, one party (currently, Democrats) might still not do as well in translating its votes into seats as the other.

The outcome-oriented approach, which is what you asked about, instead aims to give each party a roughly proportionate seat-to-vote outcome. But to do that, you have to compromise certain other democratic values, such as keeping communities of interest together, since you would need to design districts that did things like put residents of cities and suburbanites together, as one example.

Q. It seems like the political momentum is toward the first approach — a process-driven approach. That is the logic behind independent redistricting commissions, correct?

A. Pildes: I think we are still in the midst of working these issues out. Some of the voter initiatives that are currently being considered, such as in Missouri, require that partisan fairness be taken into account. That reflects the more outcome-oriented approach. But when courts design districts, or hire a special master to assist them, they almost always take a process-based approach and expressly state that partisan outcomes are not to be taken into account.

A. Rodden: The “first generation” of redistricting reforms, like Florida’s constitutional amendment, were quite explicitly process-oriented. But my sense is that reformers are now taking seriously the notion that it is possible and desirable for a commission to be tasked with the job of achieving partisan symmetry in the transformation of votes to seats.

Q. For reformers, what seems to be the most promising route, especially after the SCOTUS decision in Wisconsin? Try to push through commissions, including via initiatives? Or is it worth trying to find an argument or a test for an unconstitutional partisan gerrymander that might pass muster with SCOTUS? And how would the appointment of Kennedy’s replacement affect this calculation?

A. Pildes: Nearly all legislatures refuse to relinquish redistricting power voluntarily. That is why we’re seeing a number of voter initiatives attempting to create independent commissions that take this power out of the hands of self-interested political actors. This November, such initiatives are on the ballot in five states.  But that option is limited to states that have an initiative process. So the efforts to rein in partisan gerrymandering will continue to pursue both tracks: judicial constraints through legal doctrine and efforts to create commissions. But even independent commissions must decide whether “fairness” should be defined in process or outcome terms.

A. Rodden: Rick is right. I think the judicial route will focus increasingly on state courts. And thus, appointments and elections to state supreme courts will become even more partisan and polarizing. I’d also note that redistricting reform often gets substantial bipartisan support when it makes its way to the ballot.

Q. Redistricting reform has been a big theme for Democrats — like Eric Holder’s initiative. Do you think that will translate into real reform? Or, if Democrats gain control of redistricting in more states after the 2018 and 2020 elections, would they resort to partisan gerrymandering?

A. Pildes: Historically, neither party has had a monopoly on political virtue or vice when it comes to redistricting. One of the most extreme and notorious gerrymanders of the 1980s was the “Burton” gerrymander of California, a product of a San Francisco Democrat, Rep. Phil Burton. When Republicans lose out in the process, I have no doubt they will turn to the courts to challenge Democratic gerrymanders. Right now, a Democratic gerrymander of Maryland is one of the major cases being litigated.

A. Rodden: In many states, if the Democrats gain control of the redistricting process, it will take a good deal of work and creativity just to neutralize their geographic disadvantage. The temptation to do this will be strong. Process-oriented redistricting reform may begin to look like a second-best alternative.

To Rick’s point, the recent experience of the Canadian Liberals might be instructive. Justin Trudeau campaigned on a promise that if the Liberals were elected, the 2015 election would be the last ever conducted under the existing first-past-the-post electoral system. But then he won, and rediscovered the benefits of that system.

Q. So it sounds redistricting is hard to do fairly, and it’s not clear that we’re headed for a redistricting cycle that’s any less partisan and litigious.

A. Pildes: I don’t agree completely. First, we can agree that certain factors in the districting process are legitimate ones and others are illegitimate or unconstitutional ones, even if we don’t agree on what an optimally fair map would look like. This happens all the time with legal doctrine: We condemn certain practices as unfair or unconstitutional without clear agreement on what is ideally fair. Second, we do have to figure out what “fairness” ought to mean, but if we can agree on that, then it’s not hard technically to draw a “fair” plan.

And yes, given our polarized politics and the slender margin between the parties in Congress and many state legislatures, the next redistricting cycle should produce intense partisan warfare unless the courts develop meaningful constitutional constraints or there are more independent commissions.

A. Rodden: Even in the states where Democrats are most concentrated, it is possible to draw “fair” plans. We just need to be clear about our definition of fairness and confront the trade-offs between fairness and other objectives. As for the next redistricting cycle? Let’s just say that partisan operatives and election lawyers will not be going hungry anytime soon.

This Monkey Cage Chat was produced in collaboration with the Media & Democracy program at the Social Science Research Council.