Independent Redistricting Commissions

A while back, I wrote an article about gerrymandering and the performance of independent redistricting commissions, but due to a technical issue my writing was tragically obliterated. It's been hard for me to overcome writer's block on a rewrite, but here we go.

After each U.S. Census, the various states get to participate in America's favorite sport: redistricting our electoral maps! And, of course, the 2020 Census is no different: the 2020 redistricting cycle is currently in full swing in preparation for this year's elections, and with it comes the age-old practice of gerrymandering.

Gerrymandering is, simply put, the practice of drawing electoral districts to favor one party or another. For instance, state legislators can draw their own state legislative districts to make it easier for them to be re-elected, or draw congressional districts that make it easier for their party's candidates to win. The claim might sound outlandish, but just take a look at the congressional map Maryland's used for the last decade:

Maryland's Congressional Map (Credit: Wikimedia Commons)

This map has consistently elected seven Democrats and one Republican to office from a state where Republicans regularly get about a third of the vote. That's 12.5% of the seats for 34.8% of the vote, at least in the last election. This is done by packing a whole lot of Republican voters into one district -- the first -- and spreading just enough Democrats to make a majority across the rest of the districts. Believe it or not, the number of wasted votes for Republicans in 2020 was larger than the entire population of Wyoming! It's not that the map is drawn this way to better represent the individual parts of Maryland, either: every single district in this map covers at least part of the corridor between Washington DC and Baltimore, and connects them with far-away parts of the state that may have completely opposed priorities. The peninsulas of the third district aren't even fully connected on land.

It's generally accepted that gerrymandering is a problem, but there's a lot of debate over how we determine which maps are gerrymandered and which are fair, and what we can do to solve it. (One bipartisan anti-gerrymandering organization, RepresentUs, describes the problems of gerrymandering here.) One proposal to solve the problem is implementing Independent Redistricting Commissions (or IRCs), which are bodies of citizens chosen to draw the maps without having an incentive to pad their own elections. There are a variety of such commissions, with different ways of selecting commissioners, but the basic idea is that if Republicans, Democrats, and independents have to agree on a map, then it'll be fair to all sides. Both Republican and Democratic states use independent redistricting commissions, going back decades.

While there are some concerns about the commissions -- among other things, their being unelected (in many cases the commissioners are selected at random out of a set of nominees or applicants), and potentially unconstitutional for federal congressional maps (though the Supreme Court disagrees) -- it is not within the scope of this article to make a definitive recommendation on how to address gerrymandering. Instead, I simply want to present how independent redistricting commissions have performed across this year's redistricting cycle. What I present is just one person's view, and plenty of other resources are available online and elsewhere for you to read. Each and every reader has their own mind to make up, and you are of course free to make up your mind any way you please.

Method of redistricting in 2020 (Credit: Orser67/Wikimedia Commons)

Green states used independent redistricting commissions; Grey states did not redistrict

Red states had Republican control of redistricting; Blue states had Democratic control of redistricting

Yellow states had split control of redistricting

NJ and HI both have redistricting commissions made up of politicians, not voters.

That said, independent redistricting commissions performed very well in congressional redistricting this year. They were faster, fairer, and more reliable than state legislatures, and especially outperformed legislatures in cases where control of a state was split between parties. Unfortunately I don't have enough data to say all of this for the 2010 redistricting cycle, though I can say that independent redistricting commissions were at least more reliable in that year too. Let's break down the data.

Faster: This past year, the full Census data required to ensure that redistricting makes equally-sized districts was delayed, forcing states to redistrict faster than usual in order to be ready for this year's elections. Of the ten redistricting commissions tasked with congressional redistricting this year, all but two (80%) completed their congressional maps on time, and one of the two others finished just a few minutes past the deadline. Only one state with such a commission -- Hawaii, a very small state -- was left without a set map at the start of 2022. Meanwhile, of those states where state legislatures complete redistricting, only eighteen out of thirty-four (about 53%) finished their maps by the start of the year (some of which were later struck down and redrawn), while three states still have yet to pass congressional maps because of various power grabs. This is very important because delayed maps mean that potential candidates and voters don't know what district they'll be living in when the election comes around, and if the maps is especially delayed, primary elections might need to be delayed too. Overall, we don't need any more confusion with elections, and independent redistricting commissions have proven themselves quicker this cycle.

In short: 90% of IRCs finished their jobs by the end of last year, but only 53% of state legislatures did; three state legislatures still have yet to complete redistricting.

Fairer: It's often accused that the party in control of a state will try to use redistricting to gerrymander to their advantage, but it's very difficult to quantify exactly what constitutes a gerrymander and what doesn't. There are a whole lot of metrics, and I won't go into detail about them here, but one of the metrics used by the folks over at FiveThirtyEight is called the efficiency gap. You can read about the different metrics they use on their website if you like; the Princeton Gerrymandering Project also calculates and explains various metrics on their website here. To make this simple, however, I am going to focus on two simple metrics: efficiency gap, and seat projections. This is not a particularly advanced analysis, but it is enough to demonstrate the relative fairness of independent redistricting commissions.

The efficiency gap metric effectively measures how many votes a party needs to get in order to win seats. (For instance, in Maryland, the efficiency gap favored Democrats, meaning that the Democrats needed to win fewer votes per seat in Congress than the Republicans did.) It is a convenient measure of bias in congressional maps: why should one party have to work significantly harder than another in order to win the same number of seats? But politicians draw biased maps anyways: according to FiveThirtyEight, of the maps initially passed by state legislatures (which may not be the final maps, depending on court rulings), five of eight Democratic-drawn maps saw their efficiency gaps move in favor of Democrats (including in three of the four states where the old maps were not drawn by Democrats), while seven of sixteen Republican-drawn maps saw their efficiency gaps move in favor of Republicans. Every single state that had partisan control of redistricting passed a map where the efficiency gap benefited the ruling party. While there are reasons other than bias that would lead to an unbalanced efficiency gap to begin with, parties are ready and willing to draw maps that change this bias to benefit them, and much less willing to draw a map that negates it.

Seat projections are another measure of how biased maps are; Democrats generally draw maps that make it easier for Democratic candidates to win elections (and thus improve their projections), and the same goes for Republicans. According to projections by the Cook Political Report (for maps struck down, I used the Web Archive), six of the eight states where Democrats controlled redistricting saw solid gains in Democratic prospects (including in all four states where the Democrats did not control redistricting in 2010), whereas seven of the sixteen states where Republican legislatures have passed maps saw solid gains in Republican prospects (including in four of the seven states where the Republicans did not control redistricting in 2010). No Republican-controlled state saw a gain in Democratic prospects, and no Democratic-controlled state saw a gain in Republican prospects.

This means that thus far, in over 70% of those states where parties gained fresh control of redistricting that they didn't have last cycle, those states passed maps more biased towards said party. The following table shows how maps passed by state legislatures this cycle consistently benefit the party in power (keep in mind some of these maps have since been struck down in court):

Who drew the maps this time?	Who drew the maps last time?	How did the efficiency gap change?	Who benefits from the efficiency gap?	How did seat projections change?
Democratic legislature (8 states)	Democrats (4 states)	2/4 states pro-Democrats, 2/4 states no change	​4/4 states Democrats	​2/4 states pro-Democrats, 2/4 states no change
	Other (4 states)	​3/4 states pro-Democrats, 1/4 states pro-Republicans	4/4 states Democrats	​4/4 states pro-Democrats
Republican legislature (16 states)	Republicans (9 states)	​5/9 states pro-Democrats, 4/9 states pro-Republicans	​10/10 states Republicans	​5/9 states no change, 3/9 states pro-Republicans, 1/9 states shrunk
	Other (7 states)	​3/7 states pro-Republicans, 2/7 states pro-Democrats, 2/7 states no change	7/7 states Republicans	​4/7 states no change, 2/7 states pro-Republicans, 1/7 states shrunk

Meanwhile, independent redistricting commissions generally did not change their maps to benefit the ruling party of their state. While each commission's map is different, efficiency gaps got closer to the middle in four of the ten states, and only moved farther away in three; moreover, all of the changes in seat projections resulting from the new maps actually canceled one another out perfectly, and four commissions saw no changes at all. Furthermore, in another measure called median seat (which describes how far the median congressional seat leans to the right or left relative to the whole state), independent redistricting commissions performed very well, with seven of the nine successful commissions passing maps with median seats closer to the state average than before. The average median seat and efficiency gap measures for independent redistricting commissions are more neutral than the same average figures for state legislatures of either party or when control is split, and are even more neutral than the average figures for state legislature-drawn maps taken all together.

While none of these metrics, commissions, or maps are perfect, the performance of independent redistricting commissions certainly seems less predictably partisan and more fair on average than the performance of state legislatures, which regularly and consistently draw maps to their advantage.

In short: State legislatures consistently pass maps that advantage their own party across several measures, whereas independent redistricting commissions do not.

More reliable: Independent redistricting commissions are less likely to fail to pass a map, and more likely to have their maps upheld in court, than state legislatures. Nine of the ten independent redistricting commissions drawing congressional maps this year succeeded in their task, and every single map they passed has been held up in court thus far. (Further, of the six commissions operating in 2010, every single one succeeded in drawing a map, and not a single one of their maps was struck down.)

Meanwhile, state legislatures regularly have their maps struck down in court for various reasons (unfair gerrymandering is major one) -- of the twenty-nine state legislatures that passed maps in 2010, five had their maps struck down in court (in the case of North Carolina, more than once), whereas this cycle three states have already had maps struck down, with thirteen others in litigation that may lead to the map being struck down later this year. State legislatures regularly fail to pass maps as well, with eight of thirty-seven state legislatures (over 20%) tasked with redistricting failing their job in 2010. State legislatures that have control split between both parties (which may include a governor's veto), however, are the worst performers this cycle: fewer than half of them successfully even passed a map this year, meaning that courts have to draw the map instead. And with three states (all with one-party control) still missing maps -- all of them due to intra-party squabbles over how hard to gerrymander (FL, MO, NH) -- these numbers may soon grow. But, at least the split states that did pass maps haven't had them struck down in court, last cycle or this cycle.

In short: More than 90% of IRCs held since 2010 have successfully passed maps, of which 100% have been held up in court. Fewer than 80% of state legislatures failed to pass maps in 2010, and fewer than 80% will succeed this year as well, assuming the three legislatures yet to pass maps fail to do so entirely. State legislative maps are regularly struck down by courts.

My personal take is that elections should be decided by the voters, not by politicians and judges working behind the scenes. The fact that redistricting fights center around politicians determining election outcomes before an election even takes place is a threat to our republic. But moving on beyond my personal biases, the fact remains that independent redistricting commissions have simply performed better this redistricting cycle than state legislatures have. This doesn't necessarily mean that independent redistricting commissions are a positive force overall, or that they will solve any of the problems that their proponents say they will, but it does mean that performance-wise they are competitive with state legislatures, and a viable method of redistricting. One redistricting cycle is, in the end, just one redistricting cycle though, and I will be very interested to see how such independent commissions perform ten years from now.

I personally am a supporter of independent redistricting commissions. I believe that they are an important step to ensuring that our elections are fair and free from political tampering. I believe that they are a step in the right direction, though not all commissions are created equal and they most certainly don't solve all of our problems. But, while they certainly seem to perform well, that doesn't mean they're free from criticism. Due to their stellar performance this cycle, however, this criticism should probably be targeted more towards the legal and structural aspects of such commissions as opposed to their ability to redistrict competently.

You are free to make up your own mind based on the vast resources at your disposal, online and elsewhere. My views are my own, and although the figures listed above are based in fact, data frequently requires interpretation and my interpretation is my own.

I hope you have a wonderful week!

-- Seal with a Pen

Addendum: Sources

In addition to items linked above, my data comes from FiveThirtyEight's wonderful redistricting tracker, the awesome people over at Loyola Law School's All About Redistricting page, and the Cook Political Report's great coverage. In the interest of whole transparency, a spreadsheet I made and used to guide much of my writing is attached below. This is where my numbers come from. It's not necessarily formatted in the nicest or most approachable way, but it is available for anyone interested in my work.

Redistricting Commissions.xlsx

Download XLSX • 14KB