Every Redistricting Map Is A "Gerrymander" From Somebody's Perspective

How do you tell a perfectly fair and neutral congressional redistricting map from a “gerrymander”? You might think that the answer to the question can be found by looking at the map. If you see sprawling and oddly-shaped districts, whose outlines in some cases resemble the form of a salamander, then obviously this is a “gerrymander.” But then you look at the maps that were at issue in yesterday’s Supreme Court decision, and you are not so sure. In the case of North Carolina, the outlines of the districts in question — drawn by partisan Republicans allegedly to maximize their advantage — look relatively compact and normal, mostly following county boundaries, and not particularly unusual in any way. Here is the North Carolina map that was at issue:


On the other hand there’s the Maryland map, this time drawn by partisan Democrats to maximize their advantage. This map has several very oddly-shaped and non-compact districts, most famously Maryland’s third district, known as the least-compact district in the country:


You probably know by now that the Supreme Court on Thursday let stand both the North Carolina and Maryland maps on the ground that the issue of partisan gerrymandering is non-justiciable. But the bloc of four liberals, led this time by Elena Kagan, issued an impassioned dissent. Justice Kagan et al. would have found both the North Carolina and Maryland redistricting plans equally unacceptable, and would have empowered federal judges to intervene to block them.

Suppose you agree with the liberal bloc that the judiciary ought to be able to intervene, at least in extreme cases, to strike down partisan gerrymanders and impose a result that is somehow “fair” to all voters. But how are you supposed to tell which among the many redistricting plans out there meet this test of fairness, and which do not? Let me suggest that there is no better answer than this one: If you perceive that your side is coming out on top with this particular map, then you can acknowledge that it is perfectly fair and neutral. If you perceive that this map is disadvantaging your side, then obviously it is a “gerrymander.”

Let’s analyze Justice Kagan’s dissent. Perhaps you might expect to find in there a principled attack on the nefarious practice of drawing oddly-shaped districts to append disparate populations into a district and thereby make sure that disfavored groups get outvoted. But actually, that argument is not here. Which is not surprising, because that argument would likely have resulted in the Maryland (Democratic) gerrymander getting struck down, while the North Carolina (Republican) gerrymander got upheld.

So instead, Justice Kagan goes with a different argument, which is that the improper gerrymander can be spotted by the degree of “dilution” of the value of the votes of the disfavored voters. Here’s the phenomenon at work in North Carolina:

In 2016, Republican congressional candidates won 10 of North Carolina’s 13 seats, with 53% of the statewide vote. Two years later, Republican candidates won 9 of 12 seats though they received only 50% of the vote.

And here are the comparable numbers for Maryland:

In the four elections that followed [the gerrymander at issue] (from 2012 through 2018), Democrats have never received more than 65% of the statewide congressional vote. Yet in each of those elections, Democrats have won (you guessed it) 7 of 8 House seats . . . .

And the good Justice goes on to cite comparable statistics from some other states. For example, there is Pennsylvania:

Take Pennsylvania. In the three congressional elections occurring under the State’s original districting plan (before the State Supreme Court struck it down), Democrats received between 45% and 51% of the statewide vote, but won only 5 of 18 House seats.

Or Ohio:

Or go next door to Ohio. There, in four congressional elections, Democrats tallied between 39% and 47% of the statewide vote, but never won more than 4 of 16 House seats.

Do these statistics prove that an extreme level of partisan gerrymandering is going on? Actually not at all. This is just the usual level of innumeracy that we have come to expect from seemingly “smart” progressives with strong verbal skills, the very type of people who might end up on the U.S. Supreme Court. In fact, there is no inherent reason why the results of elections from first-past-the-post single-member districts should in any way resemble what would occur in a proportional representation system. As just one obvious example, suppose that a state has twenty representatives, and consists of 55% Democratic voters, and 45% Republican voters, all distributed equally everywhere in the state. Should there then be a “fair” districting plan, where the Democrats will win 11 of the seats, and the Republicans 9? In fact in the hypothetical that would actually not be possible. The Democrats would win every single district by the 55/45 margin, and that would occur no matter how the district lines were drawn. Why is there anything “unfair” about that? And how are you going to change it?

Justice Kagan selected Pennsylvania and Ohio as her additional examples because in those states there had been litigation brought by Democrats challenging Republican redistricting plans, and using the statistics cited by the Justice. But there are plenty of other states with even more extreme numbers that have not recently been the subject of litigation. Funny, but the most extreme examples prove to be deep blue. For instance, in the 2018 election in Massachusetts, there were five contested congressional seats. The Democrats got an aggregate of about 67% of the vote in those districts, yet they won every one of the seats. Anything wrong with that?

And then there’s California, with its 53 seats. In 2018, in the 50 seats contested by Republicans, the Democrats got an aggregate of about 55% of the vote, to 45% for the Republican candidates. But the Democrats won 43 of the 50 seats, or 86%. Outrageous! Oh, the lines were drawn by an “independent” commission with equal numbers of Republicans and Democrats. I’m not sure that there is any possible map that could be drawn for California that could get the Republicans to 45% of the seats, but likely there is a map that could get them to around 40%, 20 out of 50. Should that now be required? Of course Justice Kagan does not raise this issue. To do so would work to the disadvantage of her side.

Pennsylvania is one of the states cited by Justice Kagan for having an outrageous gerrymander created by a Republican-controlled legislature, which led to Republicans controlling 13 of 18 congressional seats through three elections (2012, 2014 and 2016) despite very narrow margins in the aggregate popular vote. In 2018 the Democrat-controlled Pennsylvania Supreme Court imposed a new districting plan in time for the 2018 elections. Result: Pennsylvania’ congressional delegation is now evenly split, 9-9. But compare the two maps, the Republican one that was struck down here, and the one imposed by the Supreme Court here. Neither has particularly oddly-shaped or non-compact districts. Look at the two and you will see what is going on. Certain small areas in Pennsylvania — notably the city of Philadelphia and to a lesser degree Harrisburg and Pittsburgh — have very dense concentrations of Democratic voters. In the Republican map, those voters were concentrated in districts to run up huge margins for a small number of candidates. In the Democratic/Supreme Court map, districts are drawn, particularly in the Philadelphia area, to divide up the city and spread those voters into districts shared with suburban and rural areas, so that the city voters will outnumber and prevail.

Now, how exactly do you tell which one of those maps is more “fair”? If a court should impose redistricting on Pennsylvania to force the allocation of seats to more nearly resemble the percentages of votes, then why not the same treatment for California? The fact is that pretty much everyone will perceive the map that is more to their own political advantage to be the more fair. As far as I can see, there is no objective way to make this determination.