Redistricting and Gerrymandering
This is the first part of a series on redistricting and gerrymandering in the United States, focusing on the process of redistricting, and how partisan gerrymandering can impact the outcome of elections.
Question: would you rather A) vote for the candidates you think would best represent you and your community, or B) have different candidates choose the citizens they wanted to vote for them? I think it’s safe to say that the vast majority of us would choose Option A.
In a small community, like the ancient Greek city-state of Athens, it was easy enough for a few thousand citizens to gather regularly to discuss and create policies. However, the United States is far too large for that. Even the smallest states have way too many people for this to be feasible. So instead, we have adopted a system of representative democracy, where citizens vote for people to represent them in government and make decisions on their behalf. We typically see congressional bodies of a few dozen to a few hundred people, which is far more manageable for policy discussions than it would be with millions of people.
To accomplish this, states divide themselves into smaller districts. They try to draw boundaries in a way that makes each district roughly the same size in terms of population, that way each of these smaller regions has an equal voice in the national and state legislatures. Every ten years the population is counted in the national Census, and states find out just how much their population has changed. If it goes up or down, states may gain or lose seats in the House of Representatives, which means they will gain or lose districts and have to change their maps. They may also find that different areas within the state have increased or decreased in population, and they may need to redraw the boundaries for their state legislatures, to make sure that those districts remain roughly equal in size. This process is known as redistricting.
But what happens when politics injects itself into this process? Let’s look at a hypothetical state, aptly named “Hypothetica,” to see what can happen during redistricting. Note: to make this easier to understand, I’m going to somewhat simplify it and use small population numbers (in reality, there are state and federal districts, and we’d likely be dealing with millions of people).
The 2010 Census counted 90 people in Hypothetica. This was also a midterm election year. Of these 90 people, 46 voted for the “Green Party” (not the actual Green Party) and 44 voted for the “Purple Party.” That means Green Party candidates won 51% of the vote, compared to the Purple candidates’ 49%. The state legislature had a Green Party majority, which means they were in charge of the redistricting process. Once the Census results were finalized, it was determined that Hypothetica should have nine seats in Congress, which meant they needed nine districts of ten people each. They drew up the following map in 2011.
This map was approved by the Green Party governor, and went into effect for the 2012 election. The results were the same as the 2010 midterm, with 46 people casting votes for the Green Party and 44 people casting votes for the Purple Party. However, even though the vote was so close, the way the Green Party drew the map ensured that they would win at least six of the nine seats in congress. The Purple Party won three, with one being extremely close. With 51% of the vote, Green got 67% of the seats easily.
Fast forward a few years. Hypothetica underwent an economic boom, focused around its central city. Plenty of good jobs became available, and the influx of tax revenue was invested in education and infrastructure, which attracted a lot of new people to the area. State officials projected that a significant increase in population would be recorded in the next Census.
The officials were correct: the 2020 Census found an 11% increase in the state’s population, which grew to 100 people. While the outer population remained the same, the central population greatly increased. When the 2020 election was held, there was a noticeable change from previous elections. Of the ten new residents, eight voted Purple, and two voted Green (seen in the center of the maps above). Now, with 52% of the vote, Purple was in the majority, which meant the state elected a Purple Party governor. However, the results for the congressional vote didn’t change at all.
The 2020 election was held using the district maps from 2011. All the new Purple voters moved into the central district, which already voted Purple. Their additional votes didn’t change anything, it just made that one district even more Purple. Notice that the districts on the left and right side of the map all still had ten people. The districts in the top-center and bottom-center each had 11 (one new Green resident moved into each), and the central district now had 18 (eight new Purple residents). This created an obvious imbalance in the state legislature, since those 18 people in the center only got one vote, while the smaller districts (with fewer people) also got one vote. Fortunately, this is fixed every ten years, when new district boundaries are drawn to make each region equal. But we have a problem.
Despite winning fewer votes statewide, the Green party still won six seats in Congress. This means that they’ll again be the ones in charge of redistricting. The Census results determine that Hypothetica, with its population growth, should be given a tenth seat in the legislature. The map needs to be redrawn with one more district, which will all still have ten people. Let’s say I’m a Green Party legislator in charge of drawing the new map. How should I do it? Let’s go through some possibilities.
I like to think of myself as a fair-minded individual. I draw up Possible Map 1. In this map, I create two districts that will likely vote Purple, two that will likely vote Green, and six with a pretty even split. These six districts will be highly competitive, which means each party will have to nominate good candidates to try and attract moderate voters, which I think will result in quality(ish) politicians running things. I draw up a second map too, just to give Congress options. I know my party got fewer votes in the election, and the majority of people supported a Purple Party governor. Still, the election was close, and a 52–48 vote is pretty close to 50–50. So, in Possible Map 2 I create five districts that will likely vote Purple, and five that will likely vote Green. I think this is pretty fair. But what if I wasn’t a fair-minded individual? Maybe all I’m interested in is holding onto power, and I’m worried if I act fairly, I might lose my power.
So I draw Possible Map 3. In this version, the boundaries are set so that the Green Party will have six seats, and the Purple Party will only get four. This means that we will still have the majority in Congress, and we can stop any Purple legislation we want. Then I realize this will also mean that any Green legislation we pass will likely be vetoed by the Purple governor. In order to override a veto, you need a two-thirds majority in Congress, which means we’d need seven seats. If I change just a few lines and make one of the central districts look really weird, I can pack more Purple people into one place and make sure we get seven seats. I draw Possible Map 4. Then I think, can I go any further? I play around with the map, and it turns out I can. If I tweak just a few more lines and draw a few more funky-shaped districts, I can actually make sure that, even with only 48% of the vote, my party will get eight (80%) of the seats in Congress. This leads to Possible Map 5. I know the new Purple Party governor isn’t going to like this, but then I realize something else: we currently have six out of nine seats in Congress. If the governor vetoes our map, we have the two-thirds majority to override his veto. This map is happening.
This is gerrymandering: drawing electoral districts in a way that benefits your own party. In the image below, I made two other Hypothetica maps that more clearly illustrate how this is done: through packing and cracking. Packing is when you “pack” one party’s voters into as few districts as possible. You can do this by drawing compact districts in areas that heavily favor your opponent’s party, or by drawing weird, squiggly shaped districts to try and get as many of your opponents into one as possible. You’ll notice in the two maps below that I have drawn a few snake-like districts that pack as many purple or green voters into one district as possible. Cracking is when you split up areas that favor your opponents and make them the minority in districts that favor your party. I did this in the maps below by drawing lines through a lot of purple areas to split them into green-majority districts, or through green areas to split them into purple-majority districts.
This is not just a hypothetical exercise. This is reality. Politicians have used gerrymandering to benefit their own party for centuries. The term originated in 1819 when Massachusetts Governor Elbridge Gerry oversaw redistricting efforts to limit the influence of the Federalist party by drawing weird, mythical salamander-shaped districts (Gerry + salamander = gerrymander) that packed Federalists into a few districts, and diluted the rest of them into Democratic-Republican-majority districts. While both parties today engage in this type of rather nefarious activity, one party has gotten really, really good at it, in both state and federal elections. I’ll explore the data related to this in the next article.
