(This is the second post in a two-part series on gerrymandering. To read about different types of gerrymandering, including cases where it’s legally required, read the first post here.)
Although there are several motivations behind and methods to gerrymandering, identifying specific instances of it and agreeing on appropriate regulations to stopping it pose significant hurtles to finding a lasting solution.
To identify gerrymandering, journalists, opposing political parties, and advocates have traditionally pointed to the distorted and bizarre shapes that often arise. Lists of the top 10 most gerrymandered districts circulate in the media and contests are held to identifying what shapes the districts most resemble. For instance, looking at 114th U.S. Congressional Districts we can see Pennsylvania’s famous 12th Congressional District, nicknamed “Goofy Kicking Donald”:
Texas’s Second Congressional District, which winds through Houston:
And Illinois’s Fourth Congressional District, which looks very much like a horseshoe:
Unfortunately, while these clearly feel “wrong,” they leave the definition as “you know it when you see it” which is not enough to hold up in court. Two main approaches have arisen to create a more objective measure to gerrymandering. The first involves quantifying the distortion of a district through a measure called compactness. The second focuses on comparing voter power across party lines.
Compactness tries to quantify the degree to which congressional districts are unnaturally stretched and misshapen, a frequent byproduct of extreme gerrymandering. It is measured mathematically by calculating a district’s area in proportion to its perimeter to give a measure of how efficiently the district contains its area. The index is created by comparing a district area to perimeter ratio with that of a circle. A measure of 1 indicates that the district was a perfect circle, and thus the most compact of all shapes. The smaller the measure the more likely the district is extremely extended and misshapen. For instance, Maryland’s Third Congressional District has one of the worst compactness scores at .0333.
Compactness also helps prevent districts from unnecessarily dividing existing geographic or political boundaries, which many state constitutions prohibit.
However, not all gerrymandering involves the use of strange district shapes, and many oddly shaped districts are instead the result of geographic features or political boundaries. For these and other reasons, the method has received much criticism and has not been used in any decisive court cases to set precedence.
An alternative measure of gerrymandering gaining popularity is the “efficiency gap.”Instead of focusing on individual gerrymandered districts, the method compares the voting power of each political party in a state. Devised by Nicholas Stephanopoulos, Professor at the University of Chicago Law School, and Eric McGhee, Research Fellow at the Public Policy Institute of California, the efficiency gap calculates the wasted votes in each party and compares these counts to see if either party had an advantage in the number of votes they got compared with the number of seats they won.
The efficiency gap is first calculated by finding wasted votes. Wasted votes are considered any votes cast for the losing party, but also any votes cast for the wining party that constitute more than a simple majority. The total number of wasted votes in each party are calculated for the state, and the net difference is calculated between the two parties. For instance, if Democrats wasted 100 votes and Republicans wasted 75 votes, Democrats would have a net wasted value of 25. The final step in calculating the efficiency gap is to divide the net wasted voters by the total number of votes cast for that state in the specific election. The final efficiency gap measure tells if one party won a larger number of seats as compared to a neutral district plan. For instance, a 40 percent efficiency gap for Party A would mean that Party A won 40% more of the seats than it should have based on a proportionally fair share of its supporters. To see the efficiency gap calculated for each state, check out Azavea’s recent gerrymandering project.
By focusing on unused votes, the efficiency gap method is able to capture both cracking and packing gerrymandering methods, because both techniques try to increase the number of unused votes for the opposing political party while minimizing their own party or candidate’s unused votes. With cracking, a voting block is split between districts so they are not a majority in any district, and thus their votes never count towards electing a representative. With packing, all of a party’s supporters are concentrated into a few districts, where any votes greater than a simple majority are wasted on electing a representative that already won. Both of these methods are captured in the efficiency gap because it counts unused votes as both votes for the losing political party as well as votes in excess of a majority for the winning party. Further, by focusing on the voting power of each political party, it demonstrates which voters are unable to contribute to the election outcome based on their district characteristics.
Currently, most voter regulations focus on preventing racial gerrymandering (although, there is a fine line between minority-majority districts and trying to remove the voting power of minority voters). By contrast, there is no clear precedence to stop partisan gerrymandering. The Supreme Court is set to hear Wisconsin’s gerrymandering appeal, which could change this. The original suit against Wisconsin used the efficiency gap measure to demonstrate the presence of political gerrymandering. The lower courts ruled against Wisconsin. If the Supreme Court decides to deny Wisconsin’s appeal, the efficiency gap is likely to be used in many other states to force redistricting.
The Brennan Center provides a list of current Partisan Gerrymandering Court Cases, for reference.
Even if there is agreement on which districts are being politically gerrymandered, there is little agreement on a solution.
The most common proposal is using independent or non-partisan commissions for re-redistricting, which is done in several states including Alaska, Arizona, California, Idaho, Montana, and Washington. Nevertheless, many opponents say it is difficult to find “non-partisan” committee members.
Other advocates propose removing human influences altogether and relying on computer algorithms to “fairly” create districts. For instance, Brian Olson, has created an algorithm that optimized the creation of new districts based on equal numbers of voters and compactness. This method relied heavily on compactness as the ultimate goal in creating a district, ignoring the current stipulation that districts focus on “communities of interest.” While this specific method may be overly simplistic, it demonstrates the opportunity for more sophisticated algorithms to manage redistricting in the future. Nevertheless, deciding what priorities would be used for the creation of districts would involve significant debate and would likely still run into issues of partisan disagreement.
Political Voting System
Further advocates propose changing the political electoral system altogether to remove the ability for any gerrymander. Instead of focusing on individual gerrymandered districts, they instead advocate for a switch away from our voting system to a proportional representation or ranked choice runoff method. The existing single-member district method was fully implemented after the Civil War by progressive legislators that feared southern states would use it to prevent freed slaves from influencing voting elections at the local level. It forced states to be divided into roughly equal population districts, and elect one representative for each district. For more information on the existing political system see FairVotes’s history of single-member districts in the US.
With proportional representation a party holds as many seats as are proportional to their voting support. For instance, if 40% of the voters in a state voted for Party A, then 40% of the representatives of that state would go to Party A. While decidedly fair regarding the final representation of each part, this system does have critics that argue that the representatives are not as closely connected to their constituents and local issues. Many countries around the world use this system, and the U.S. Constitution already requires proportional representation based on population. However, the switch to a proportional representation voting method without the use of single-member districts would require changing the 1967 law passed by congress which prohibited at-large and other multi-member elections by states with more than one House seat.
With ranked choice voting, voters rank all the candidates in order of preference. If no candidate has a clear majority of votes (often greater than 51%), then the candidate with the least number of first place votes is eliminated. Voters whose first choice was the eliminated candidate are counted towards their second favorite candidate and the totals are recounted. This recounting continues until there is a clear majority supported candidate. Many advocates say that this method would make voters more open to a third party and encourage candidates to be more cordial to their rivals as they would be hoping to be those candidates’ supporters’ second choice. Ranked choice voting is already used in several local cities in California, Colorado, Maine, Massachusetts, Maryland, and Minnesota. Other states already use the method for overseas votes in runoffs. Ranked Choice voting was also approved state-wide by Maine voters in fall of 2016, however the decision was advised to be in conflict with their state constitution and there is some controversy over whether it will be implemented for 2018. The outcome in Maine is likely to influence the likelihood of other states to pass similar state-wide measures.
Regardless of the solution taken, tackling gerrymandering will be an important goal to ensure that voters continue to have the power to choose their representatives, instead of letting representatives chose their voters.