Today, the Supreme Court ran away from the question of partisan gerrymandering. Chief Justice Roberts’s majority decision cited the problem of “proportional representation” as an impediment to establishing a standard of fairness. His argument has a major logical problem which is easily fixed – though at this point, it will have to be done in state courts. Here’s how.

Roberts asserts that claims of partisan gerrymandering rely on an assumption about what would be appropriate levels of representation. Specifically, he says that the number of seats is not guaranteed for a given number of votes cast statewide – a quota of sorts. (Note that although he says “proportional representation,” he doesn’t mean literal proportionality. He means any diagnostic tool that uses the number of seats won by either party, compared with their vote share.)

To be sure, some tests count wins and losses. The efficiency gap comes to mind, as well as a well-established measure called “partisan bias.” But not all tests do that.

In several law articles [Harvard Law Review Blog]  [Election Law Journal] I argue that the tests fall into two categories:

1. Inequity of opportunity: did both parties have similar opportunity to win seats?
2. Inequity of outcome: did the parties get a fair share of seats?

Roberts has focused on category (2). But category (1), inequity of opportunity, fits very well with previous doctrine – which has focused on race as a factor. Was a group packed? Then they would have lopsidedly large wins. Was a group cracked between districts? Then their opponents would have carefully arranged, narrow wins.

Here are some tests that probe for inequity of opportunity:

• The lopsided-wins test: in a closely-divided state like North Carolina, does one side have more lopsided wins than the other? The t-test does this. As students of science and engineering know, this is literally the oldest test in the book. [Stanford Law Review]
• In a party-dominated state like Maryland, were the wins engineered to be very uniform, thus protecting the majority party? Again, it’s a simple test – and again, it doesn’t count the number of wins. [Election Law Journal]

Finally, if the Court didn’t want to get into math, they could have done what they’ve always done in racial gerrymandering cases: inspect individual districts for evidence of packing and cracking.

So basically their reason for not addressing partisan gerrymandering is at best a misunderstanding of the detection methods; at worst, it’s a clever dodge.

Would pointing all this out to the Supreme Court have swayed them? I doubt it. Fundamentally, I suspect Roberts didn’t want to address these cases at all. If in her dissent Justice Kagan had cited the arguments I’ve written about, I am sure Roberts would have found another excuse.

Although federal courts are now closed, state courts are wide open – indeed, Roberts leaned on this point fairly hard. And all the data-based arguments for partisan gerrymandering can still work in state court. We’ve built a legal doctrine that finds relevant principles in all fifty state constitutions. [U. Pennsylvania Journal of Constitutional Law, forthcoming] In North Carolina, where plaintiffs lost in today’s Supreme Court decision, the state supreme court is friendly to voting rights. There’s a lot of hope for the Tar Heel State.

The dream of fair districting lives on!