Impact of Gerrymandering on US elections

This is part 1 of Gerrymandering series in which we will discuss what gerrymandering is,how it impacts elections and how it can be detected.

Every ten years, the United States conducts a census, and from this population information, states create congressional districts, and from each of these districts, a representative is elected to represent the people of that district. However, these districts are not always drawn to be fair and truly representative. And that is caused by gerrymandering. But what is it actually? By definition,

It is a practice in which the politicians at the power will create advantage for their party by intentionally creating boundaries in such a way that they get the lead by distributing voters in their favor.

The term was used first in 1812 when a district looked like a salamander under the rule of Massachusetts governor Elbridge Gerry. Because of the way redistricting was done, the district with odd shape elected 3 Democratic-republicans that year. Although he did approve the plan, no evidence was found which showed whether he supported the maps. Nonetheless, the name just got stuck.

Elkanah Tisdale

Ever since then, gerrymandering has been an issue. For the 2012 election, redistricting was done with the census taken in 2010. In 2011, when the plan was made, Republicans were under control. And as you can imagine, the plan was favoring Republicans. But unlike the previous times where only few people took notice, more people became aware of what it is and how it impacts democracy. They even started to question where their vote is going.

The most notable states where gerrymandering was visible were North Carolina, Michigan, Ohio, Texas and Pennsylvania (but a remedial plan was made for PA in 2018).

The issue became so big that the reputation of election was at stake. Courts started taking more interest in this and started appointing people to tackle this issue. People from various domains such as political science, mathematics, law, computer science, etc. came together to find solutions.

But how is it done exactly?

There are two ways in which gerrymandering is done, namely:

1. Packing: In this, the main motive is to “pack” i.e. group as many opposing party’s voters as possible. This way, the incumbent party can win the neighboring districts where there is not much opposing party’s voters present.
2. Cracking: In this, the main motive is to “crack”, i.e. divide the opposing party’s voters in such a way that they are minority in those districts.

Azavea

How do we know whether a state is gerrymandered or not?

We have some metrics which can be used to check whether a state is gerrymandered or not. First we will discuss some compactness metrics. As the name states, compactness metrics checks by how much value, a district is “compact”. Higher the value, better is the compactness of that district. There are many compactness metrics but for now we will be focussing on 4 metrics namely:

1. Polsby-Popper: 4𝜋 * (A/P²), where A=Area of the district and P=Perimeter of the district.
2. Schwartzberg: 1/(P/C), where C = 2𝜋r and r = (A/𝜋)^(1/2)
3. Area/Convex Hull: ratio of area of district to the area of minimum convex polygon that can enclose the district.
4. Reock: area of district to the area of minimum bounding circle that can enclose the district.

Although compactness scores gives us an idea of how some districts which have a crooked shape must be investigated, it is not the ultimate factor. It has a lot of shortfalls such as it does not consider voter distribution which is an important criteria to be looked upon. For that, we will be using fairness metrics. In our case, we will focus on 3 metrics, namely:

1. Efficiency Gap: It checks how much votes have been wasted from each party due to packing and cracking.
2. Partisan-Bias: It is basically the difference between each party’s seat share and 50% in a hypothetical, perfectly tied election.
3. Mean-median Difference: It is the difference between a party’s median vote share and its mean vote share, across all of a plan’s districts. When the mean and the median diverge significantly, the district distribution is skewed in favor of one party and against its opponent.

Unlike compactness metric, we are are not going for higher scores. Rather it is better if a plan has a value closer to 0.

Other than that, it will be better if the districts have relatively equal population distribution.

This was just a short overview on what gerrymandering is and how it can be checked. In the coming blog, I will be doing more analysis, especially on the Pennsylvania state. We will be seeing how the 2012 plan performed on the fairness metrics and compactness metrics discussed above.

You can check my project here.

References:

1. https://www.nytimes.com/2019/06/27/us/what-is-gerrymandering.html
2. https://planscore.org/metrics/efficiencygap/
3. McGlone, D. (2020, January 09). Measuring District Compactness in PostGIS. Retrieved from https://www.azavea.com/blog/2016/07/11/measuring-district-compactness-postgis/