In 2000, Arizonans voted to take the state legislature out of the redistricting process. They hoped to curb partisan gerrymandering by creating an independent, bipartisan commission that would handle the duty of redrawing state and federal voting districts after every Census.
The commission’s redistricting plan led to Democratic wins in 2012, which upset the GOP-controlled legislature. Republicans claimed that the whole process was unconstitutional because the founding fathers specifically commanded state legislatures to run elections.
They were referring to the Constitution’s Elections Clause, which reads as follows:
The Times, Places and Manner of holding Elections for Senators and Representatives, shall be prescribed in each State by the Legislature thereof; but the Congress may at any time by Law make or alter such Regulations . . . .
Yes, this is another Supreme Court case that hinges on the meaning of a word. Here, the word is “legislature.” Does the Constitution refer solely to a state’s body of elected representatives? Or does the term “legislature” here lend itself to a broader interpretation, one that would allow citizens to have a direct say in how elections are conducted?
Writing for the majority, Justice Ruth Bader Ginsberg argued that the power of regulating elections should ultimately lie with the people — such is the spirit of democracy, after all.
“[I]t would be perverse to interpret the term ‘Legislature’ in the Elections Clause so as to exclude lawmaking by the people, particularly where such lawmaking is intended to check legislators’ ability to choose the district lines they run in,” she wrote.
In his dissent, Chief Justice John G. Roberts Jr. accused his fellow justices of fudging the law to bless a policy that they liked. “The majority today shows greater concern about redistricting practices than about the meaning of the Constitution,” he wrote.
Let’s stop there for a second. Regardless of how the decision was achieved, it protects an important attempt by citizens to ensure fair elections at a time when politics is become increasingly bitter and undemocratic.
One unfortunate fact of American politics is that election districts are becoming more and more misshapen. Recently, Stephen Ansolabehere, a professor at Harvard, and Maxwell Palmer, an assistant professor at Boston University, crunched the numbers on the past 200 years of congressional district shapes.
The researchers used several different ways to measure how convoluted a district’s borders are. All of the methods tend to agree that district complexity has substantially increased since the 1970s. The changes occurred across the board. Both simple and raggedy districts are starting to look weirder and weirder.
Of course, just because a district looks ugly doesn’t mean it’s unfairly drawn. It’s important to create districts that represent natural communities, and communities often have irregular boundaries. It’s equally important to preserve minority representation, and sometimes districts look like spaghetti in order to prevent the minority vote in a state from being diluted.
Partisan gerrymandering can’t be detected solely by examining geography and geometry. There’s culture and history to consider. That’s also why courts and lawmakers have struggled to define what is an unacceptable level of gerrymandering. The process is subjective.
For instance, Ansolabehere and Palmer also found that the weirdest-looking districts tend to vote Democratic. That doesn’t prove that Democrats are the more shameless gerrymanderers, though.
Besides, Democratic voters tend to concentrate in urban areas, which can make it hard to draw compact voting districts that are also fair.
A 2013 paper published in the Quarterly Journal of Political Science argued that the clustering of Democratic voters creates a natural Republican bias when districts are drawn up. Political scientists Jowei Chen, of the University of Michigan, and Jonathan Rodden, of Stanford, call the phenomenon “unintentional gerrymandering.”
To demonstrate, Chen and Rodden used a computer to randomly draw district plans using data from the 2000 presidential election in Florida. Famously, the popular vote in Florida split more or less evenly between George W. Bush and Al Gore. So in a fair, representative plan, you might expect half the districts to go for Bush, and half for Gore.
But across hundreds of versions of these randomly generated district maps — paragons of disinterested district making — a bias emerged. On average, three-fifths of districts went for Bush, even though half of the population was voting for Gore.
The explanation, according to Chen and Rodden, is that Democratic voters are at a disadvantage by dint of their living patterns.
Democratic voters concentrate either in big cities, or in Democratic enclaves (black suburbs, college towns) scattered across the state. The downtown districts are overwhelmingly Democratic, while the small Democratic outposts tend to get engulfed by their Republican neighbors.
More equal representation might result if planners could better distribute Democratic voters — perhaps, say, by appending some solidly Democratic urban neighborhoods to purplish suburban districts.
But at some point that starts to look like gerrymandering, doesn’t it?
The complex, cross-cutting considerations involved in redistricting provide cover for partisan shenanigans. When legislators start to get creative with district borders, are they adjusting for a state’s uneven political geography? Or are they angling for an advantage?
Putting the process in the hands of an independent body is one way to make matters a little more fair. Political scientists have generally observed that courts and independent commissions tend to design more competitive districts.
In a study of congressional elections from 1972–2012, researchers at the University of Georgia and the University of Texas at Austin found that districts drawn by legislators are significantly less likely to result in close elections, even after controlling for polarization, candidate quality and regional differences. But this effect is only prominent from 1992 onward:
The researchers suspect that computers deserve much of the blame. By 1991, all but four states were using geographic information programs to guide their redistricting efforts. Nowadays, any legislator with a laptop can call on software to concoct districts with whatever characteristics they like. Gerrymandering has never been this easy, or data-driven.
Perhaps, then, it is no coincidence that some of the most tortured-looking districts were designed in recent years, as these maps from Ansolabehere and Palmer show:
Independent redistricting commissions are supposed to curb gerrymandering but most lack full control of the redistricting process. Some are restricted to making suggestions, which legislatures can vote down or amend. Others step in only after the legislature has stalemated. Like the filibuster, the power of redistricting is a political tool that both parties are loath to give up. Legislatures in 42 states still have the final say when it comes to determining districts for House seats in Congress. And in 37 states, legislators control the redistricting process for their own state legislative districts.
In Arizona and California, citizens used ballot initiatives to wrest that power away from their legislators. The independent commissions in these states are completely in charge of redistricting. Their maps cannot be changed or overruled by the legislature. To keep politics at bay, these committees are kept strictly bipartisan, with reserved seats for Democrats, Republicans, and independents. Selection rules exclude politicians and those who work for them. California went as far to use a lottery to pick some of the commissioners.
Despite the cautionary measures, these committees have been the site of considerable partisan bitterness. There have been allegations that the non-partisan members on the commission secretly favor one side, and that the technical staff assisting the commissions introduced bias into the process. For these and other reasons, the resulting district plans have consistently been challenged in court. As political scientist Bruce Cain writes:
[R]edistricting is bedeviled by the sore loser problem: because new district lines can determine the electoral fates of candidates, political parties, and interest groups, it is usually worth their time and effort to overturn a plan that they do not like for the uncertain prospect of something better.
In fact, Monday’s case was merely the first of two GOP lawsuits before the Supreme Court involving redistricting in Arizona.
Now after the Court has established that Arizona’s independent commission is constitutional, it will consider whether the commission’s most recent redistricting map apportioned Republicans fairly. On Tuesday, the Court announced that it will hear this second lawsuit some time next session.
Many political scientists say that all the fuss over gerrymandering amounts to wasted breath. Recent decades have seen large decreases in the number of competitive House districts, and large increases in political polarization — but these twin trends can’t be blamed on gerrymandering alone.
In a 2006 paper published in the Journal of Politics, Emory researchers pointed out that House elections have been getting less competitive even during years when district boundaries haven’t changed.
If gerrymandering alone were the culprit, the number of competitive seats should plummet after each redistricting — between 1990 and 1992, or 2000 and 2002, or 2010 and 2012. But the changes that happen between redistricting years have been modest.
Furthermore, Princeton political scientist Nolan McCarty has noted that political polarization increased just as fast in the Senate as in the House. Since senators are elected in statewide races, they aren’t affected by gerrymandering. Yet, the Senate, too, has been paralyzed by partisan rancor.
Gerrymandering, it seems, is a symptom of increasing viciousness in politics — not the root cause.
But that’s even more of a reason for states to safeguard the district-making process from partisanship. The ongoing fight in Arizona has demonstrated that political parties will battle tooth and nail to tilt the electoral map in their favor.
Arizona and California’s experiments in independent redistricting attempt to de-politicize redistricting by appointing citizen committees. It’s possible to take that idea one step further. These days, software and cheap computing gives every citizen the power to create their own district maps. Opening up the redistricting process to public participation and review would offer one more check against legislators acting in their own self interest.
That tracks the spirit of Monday’s decision, in which Ginsburg borrowed liberally from founding father James Madison. “The genius of republican liberty seems to demand . . . not only that all power should be derived from the people, but that those intrusted with it should be kept in dependence on the people,” she quoted.
She wrote: “The people of Arizona turned to the initiative to curb the practice of gerrymandering and, thereby, to ensure that Members of Congress would have ‘an habitual recollection of their dependence on the people.'”
Independent redistricting commissions won’t heal the nasty rift in American politics. But at the very least, they would ensure that any nastiness be entirely representative of the American public.