On Tuesday, the U.S. Supreme Court will hear oral arguments in a major new case about partisan gerrymandering. The case began just days after the Nov. 8 election, when a federal court struck down a Republican-drawn legislative map in Wisconsin for being too partisan. Because of special rules for some voting rights cases, the Supreme Court is required to hear the case.
One of the key decisions the court must make is whether it’s even possible to identify an extreme partisan gerrymander. The answer of the many political scientists who have weighed in is: “Yes.”
What is the case about?
In what is now called Gill v. Whitford, a panel of three federal judges in Wisconsin found that the Wisconsin State Assembly map adopted by Republicans in 2011 violated the First and 14th amendments to the Constitution. The court’s majority declared that the map “constitutes an unconstitutional partisan gerrymander” because it intentionally “dilutes the voting strength of Democratic voters statewide” and “is not explained by the political geography of Wisconsin nor is it justified by a legitimate state interest.” This was the first time in decades that a federal court had struck down a legislative map based on excessive partisanship.
In the last big partisan redistricting case, Vieth v. Jubelirer, decided in 2004, a plurality opinion written by Justice Antonin Scalia declared that partisan gerrymanders are not the purview of courts — that is, they are “non-justiciable.”
The swing vote in Vieth was Justice Anthony Kennedy. Kennedy agreed that the Pennsylvania districts at issue in the case should not be struck down by the court. But he remained open to the idea that an extreme partisan map could be justiciable.
What was missing, Kennedy wrote, was a clear standard for deciding when a little partisan mischief became unconstitutional. As Kennedy put it, “That no such standard has emerged in this case should not be taken to prove that none will emerge in the future.”
What standard is being used in this case?
The plaintiffs in Gill base their argument on one particularly promising measure, which is known as the “efficiency gap.” First developed by Eric McGhee and then elaborated in work with his collaborator Nicholas Stephanopoulos, the measure captures the relative number of votes that are “wasted” among Democratic and Republican voters in a set of legislative elections.
Votes are wasted in two cases. The first is when there are more votes than are needed to elect a candidate. For example, if 70 percent of voters in one district vote for the Democratic candidate, then almost 20 percent of those votes are “wasted,” because it would take just over 50 percent of the vote for the Democrat to win.
Second, votes can be wasted when there are too few votes to elect a candidate. In this same hypothetical district, all of the votes for the Republican (30 percent) are wasted because they did not do anything to help the party elect a candidate.
The “efficiency gap” captures whether a redistricting plan is potentially biased against a party by making that party waste more votes than the other. Positive values of this measure indicate an advantage for the Democrats, and negative values indicate an advantage for the Republicans. For example, a value of +4 percent indicates that the Democrats got four percentage points more in seats than they “should have” given their popular vote.
How biased is the Wisconsin map?
In the Wisconsin State Assembly elections at issue in Gill, the efficiency gap was computed as -13 percent in 2012 and -10 percent in 2014. This indicates that Republicans got a big seat surplus. In fact, in 2012, Democrats won 53 percent of the votes but got only 39 percent of the seats.
The political scientist Simon Jackman calculated the efficiency gap for every state legislative map from 1972 to 2014 and found that Wisconsin in 2012 and 2014 was “virtually without historical precedent.” He suggested a threshold of +7 percent or -7 percent as being sufficiently extreme to be judged as statistically meaningful.
A key question is whether partisan bias in a map arises from redistricting itself or from a state’s political geography. If one party’s voters are concentrated in a few areas — as Democrats are often concentrated in metropolitan areas — then their votes may be “wasted” naturally, even without a purposeful gerrymander. At the federal district trial, Sean Trende argued that the efficiency gap cannot distinguish intentional bias from geographic clustering.
On the other hand, the political scientist Kenneth Mayer argued that it was possible to draw a map that satisfied other criteria for redistricting and still had a less extreme efficiency gap — suggesting that the partisan bias in the assembly map was not simply due to geography.
Are there other ways to measure a gerrymander?
Yes, there are. The efficiency gap is not the only measure, even if it has figured prominently in the Wisconsin case.
In the district court trial, the scholar Nicholas Goedert critiqued the efficiency gap for assuming an unrealistic relationship between the votes won by a party and the seats awarded to the party. In place of the familiar “S-curve” connecting seats and votes, the efficiency gap assumes a linear relationship.
If justices are concerned about any limitations of the efficiency gap, they will have plenty of other measures to choose from.
For example, Anthony McCann and his co-authors have argued for a standard of “partisan symmetry.” Symmetry means that the party with the most votes gets a majority of seats.
Other scholars have advocated a “mean-median” test. The measure compares the average percentage of a party’s vote across all districts to the median percentage of that party’s vote. If a party’s supporters have been “packed” into a few districts, the median will fall below the mean, indicating an asymmetry that advantages the other party.
Another test involves simulating districts that were drawn based on neutral criteria and comparing those with the actual districts. Wendy Tam Cho and Yan Liu have developed one such test to judge the bias of existing maps.
What have political scientists told the Supreme Court?
Since the Supreme Court agreed to hear this case, political scientists have submitted several amicus curiae briefs.
- Bernard Grofman and Keith Gaddie argued that partisan maps should be justiciable and that political scientists’ tools enable the court to determine if a map advantages one party.
- Several political scientists and other scholars concluded that Wisconsin’s map violates any measure of partisan symmetry, which they contend is a “workable standard,” easy to compute and legally appropriate.
- Jowei Che, Jonathan Rodden and others argued that three different techniques all point to unconstitutional partisan gerrymander in Wisconsin.
- A team of political scientists at Binghamton University conducted simulations suggesting that roughly half of any partisan bias in the Wisconsin map is due to political geography while the other half is due to intentional discrimination by mapmakers.
- A brief penned by me and 17 other election scholars contended that the Wisconsin map exhibits a particularly sophisticated kind of partisan bias that avoids drawing districts that are not “compact.” This is important because previous court decisions often used non-compact districts as evidence of a possible legal or Constitutional violations.
What are the implications of the court’s decision?
The Gill case is likely to be a watershed in election law jurisprudence. If the court ruled against the plaintiffs, then it could mean that no amount of partisan bias is truly unconstitutional. But the court could establish that Wisconsin and a slew of other states overreached in drawing extreme partisan districts.
Either way, the case has galvanized scholars to expand the study of redistricting. As a result, Kennedy and the rest of the Supreme Court have even more guidance on what to do.
Barry Burden is professor of political science and director of the Elections Research Center at the University of Wisconsin at Madison. He is co-editor with Charles Stewart III of “The Measure of American Elections.” Find him on Twitter @bcburden.