Tom Williams/CQ Roll Call via AP Images
In its most recent term, the Supreme Court punted two cases about gerrymandering back to lower courts. These cases questioned the district lines in Wisconsin and Maryland that state legislators drew after the 2010 census. Plaintiffs claimed that the lines, drawn by officials up for re-election, unfairly favored specific political parties and incumbents, prevented competitive elections, and misrepresented the state's political demographics.
Without those Supreme Court decisions, the redistricting process remains vulnerable to gerrymandering. Proposed changes to the upcoming census could also influence the redistricting process after 2020, but individual states have taken action to make the redistricting process fairer. The American Prospect spoke to Brennan Center for Justice Counsel Michael Li about potential state-level solutions and how the census could impact redistricting in 2020. This interview has been edited and condensed.
TAP: Each state handles redistricting differently, as you detail in your “50 State Guide to Redistricting.” What are some key differences among the states?
Michael Li: Traditionally, state legislatures were the ones who did redistricting. But in the 20th century, a number of states began taking power out of the hands of lawmakers who have a self-interest in the process and gave it to a more independent body. These bodies tend to be people who were appointed by lawmakers directly. You see that, for example, in Washington [state]. There’s another model used in some states like Arkansas, where actual lawmakers are elected to serve on the commission—the legislature or the governor or the attorney general draw the maps. Some states like Texas use a backup commission if the legislature deadlocks, then the backup commission kicks in and draws maps. And then there are what we call citizen [or independent] commissions. California and Arizona both have independent screening processes. But particularly California is the gold standard at this point in time.
Can you explain what led California to establish its nonpolitical independent commission in 2008? How does this system work?
In 2001, Democrats and Republicans got together and protected one another's seats [and] kept the seats they had. The map was so effective that between the U.S. House and the State Legislature, only one incumbent lost [over the next] decade. It was really a pretty aggressive gerrymander. There was a lot of interest among citizens in California in coming up with a better process and better maps, and a lot of interest in participating.
Members of the commission have to apply—it's a fairly extensive process. You have to submit an application and write essays, and you have to go in front of a panel of three auditors: one Democrat, one independent, one Republican, and they all have to agree that you move on. The interview is 90 minutes and it's broadcast on public access television.
They ask the same questions of everybody, but they are very good, imposing questions. That's supposed to screen out bad actors or rogue actors but also make sure that people understand what they are getting into and have the ability to be fair. It's much like a jury. We don't expect jurors to come in and not have opinions or life experience, but we don't want people who are biased, or who come in with an end goal in mind. We want people who are going to hear the evidence, and reason together, and the screening process that's in place in California is designed to get a pool of people who can work in that way. In California, something like 30,000 people applied in 2010 to be on the California commission, for 14 spots.
Would California's system work in other states, or should every state figure out its own system based on its population and demographics?
There's an argument that, especially for congressional redistricting, there should be uniform rules around the country. But our federal system has been described as a “laboratory of democracy,” [which allows] states to experiment with what works best for them and what is politically feasible in a given state. [For example,] the people in Iowa would tell you that the system in Iowa works really well because Iowa is a particularly uncomplicated state; its population is fairly evenly spread; it's one of the whitest states in the country.
The good news is [that] states have lots of models to look at, and there is an increasing amount of evidence about what works and what doesn't. There are a lot of really good proposals that will be on the ballot this fall in places like Michigan and Utah that would create commissions and strengthen the rules around redistricting, and I think those have a good chance of passing. Ohio has also been really innovative in prohibiting partisan gerrymandering through an amendment that voters passed in May of this year. There is definitely a growing reform movement around the country.
Pennsylvania had a really good redistricting bill and the legislative leaders managed to kill it, despite the fact that it had overwhelming bipartisan support. It just shows you that redistricting is really important to people in power.
How might things change after the 2020 census for the next redistricting cycle?
The Census Bureau is planning right now to add a citizenship question to the census, which has many people concerned that it will help drive an undercount in states with large immigrant communities.
That would be an undercount not only of non-citizens, but potentially of citizens. Because if you have undocumented parents with U.S.-born citizen children but the parents don't fill out the form, the children don't get counted either. So the reallocation of congressional seats that occurs after the census could affect redistricting in the states, the ability to draw districts under the Voting Rights Act—it could have a lot of impact.
Also, there's a move to change that so the districts are redrawn so that they have the same number of eligible voters or citizens. Right now, every ten years they redistrict so each district has the same number of people. If you did it by eligible voters, you would be excluding non-citizens as well as children. Places that would lose representation would be cities and suburbs—any place where you have a large number of immigrants and children. If you did it to equalize the number of citizens, the suburbs wouldn't be as impacted by that, but cities would be, because cities are the place where immigrants tend to go. That would have the effect of shifting power out of cities and potentially suburbs into more rural areas. And those rural areas tend to be more Republican.
Do you think that will happen?
People are expecting that there will be a run on that. Missouri came close to passing a bill that would've apportioned based on citizens, and people are expecting that some states will do that potentially as early as next year. In 2015, there was a challenge saying [Missouri] had to redistrict in order to balance the number of eligible voters; [it couldn't] use total population. The Supreme Court said it's fine to use total population, leaving open for another day whether a state could choose to use eligible voters or citizens, and that opening is something we expect people to try and take advantage of the next time. And it will likely go back up to the Supreme Court, but it might be a very different Supreme Court, potentially.
When most people think of gerrymandering, they think about it with regard to congressional districts, but in your 50 State Guide, you also mention state legislative districts. Are those equally as gerrymandered?
They can be, absolutely. Sometimes they're gerrymandered to give one party a particular advantage, but sometimes they're gerrymandered to protect incumbents or for other sorts of purposes. In fact, sometimes [they're subject to] even more egregious gerrymandering, it's just that fewer people watch that than national politics, so there are fewer eyes on the process.
How does incarceration gerrymandering—when prisoners are counted in the census at the site of the prison instead of their home address—affect local representation?
When prisoners are released, they tend to go back to where they came from, where they have family. All the prisons in New York are in upstate New York, but most of the prisoners are from New York City. Nobody says, “Well I'm just going to stay upstate because it was so lovely up here.” And usually these prison terms are very short. In between censuses, you could be back home within a matter of years. Not only does it hurt representation to count them at the site of the prison, but their [home] communities are losing money because the census is used to allocate funding for education, transportation, and other things.
It's a particularly severe problem at the local level and at the state legislative level. But even lower than that, at the school board, the city council, the county level. It doesn't really affect Congress because congressional districts are so large. But for example, Harris County—which is where Houston is—would have an additional state county seat if you counted incarcerated persons from Houston, opposed to at the site of the prison in East Texas.
There was an effort to get the Census Bureau to change the rule on counting prisoners. The Census Bureau decided not to do so, but they said they would voluntarily help states if states wanted to do that. New York reformed the process, passing a law saying we're going to count prisoners at their home addresses, not where the prison is. And California has done something similar, so they could fix the problem. Whether lawmakers have the incentive to do so is another question.
