In Pa. gerrymander case, experts can't defend the indefensible | Opinion

Map of Pennsylvania’s Seventh Congressional District.

Pennsylvania is no stranger to partisan gerrymandering disputes. In a blockbuster 2004 case, the Supreme Court declined to strike down the congressional map then in effect. The court didn’t quite hold that the map was lawful; rather, it couldn’t think of a workable standard for evaluating the map’s validity.

Another gerrymandering suit is now making its way through the Pennsylvania courts, with a decision expected by the end of the year. But unlike its predecessor, this suit is based on a manageable test as well as a mountain of damning evidence. Perhaps for this reason, it has thoroughly flummoxed the state’s lawyers and experts.

What’s changed since 2004? In recent years, political scientists have developed powerful new tools for evaluating district maps. These tools indicate how skewed a map is in a party’s favor, how likely this skew is to persist in the future, and whether the skew has a legitimate explanation (like a state’s political geography). In the Wisconsin case now pending before the Supreme Court, the plaintiffs used these tools to craft a simple test for gerrymandering:

  • First, was a map designed to benefit a particular party?
  • Second, has the map actually exhibited a large and durable tilt in this party’s direction?
  • Third, is this tilt bigger than we’d expect given the spatial patterns of the state’s voters?

It’s virtually undisputed that Pennsylvania’s congressional map satisfies this test. The map was drawn in secret by Republicans using reams of sophisticated data. It was then jammed through the General Assembly, largely on party-line votes, in a matter of days. In the 2012, 2014, and 2016 elections, the map set a modern record for partisan unfairness. According to one metric, it was more skewed, on average, than any other congressional map in the country since 1972.

This enormous Republican advantage is likely to endure for the rest of the decade. It would take a five-point swing for Democrats to capture just one more seat. For the map’s skew to disappear, Democrats would need to improve on their 2016 showing by 14 points — a bigger wave than Pennsylvania has seen in generations. And the map’s tilt can’t be explained by the clustering of Democratic voters in Philadelphia and Pittsburgh. In hundreds of computer-simulated maps, Democrats usually win nine or 10 of Pennsylvania’s 18 districts. The state’s underlying geography is thus close to neutral.

This evidence doesn’t just satisfy the proposed test. It also confirms that the test is manageable. If it weren’t, we’d have no way of knowing whether Pennsylvania’s congressional map passes or fails the test. But we do know that the map falls on the wrong side of the constitutional line —it intentionally, severely, durably, and unjustifiably discriminates against Democratic candidates and voters. The political scientists’ tools make the legal analysis easy and reliable.

Precisely because the map is so egregious, the state’s lawyers and experts have resorted to increasingly desperate tactics in their efforts to defend it. The lawyers first tried to stop the litigation until the Supreme Court decides its gerrymandering case. After they were rebuffed, they sought to move the proceedings to federal court — even though the plaintiffs only allege violations of Pennsylvania law. When this ploy failed too, they asked the court to exclude information (all publicly available) about how the map was designed.

The state’s experts, similarly, make a series of weak arguments in the reports they filed earlier this week. Professor Nolan McCarty, for example, claims that Republicans should only be expected to win 10 of Pennsylvania’s House districts. But they actually won 13 in 2012, 2014, and 2016. And they did so because of a variable — incumbency — that McCarty ignores. If every incumbent suddenly retired, Democrats might have a shot at fairer representation. If wishes were horses, beggars might also ride.

Next, Professor Wendy Tam Cho observes that the hundreds of computer-simulated maps might not be representative of all possible lawful maps. This is true but irrelevant. Academics continue to explore the properties of redistricting algorithms. But no legal issue hinges on these features. Even if the simulated maps aren’t perfectly representative, they still prove the existence of hundreds of fairer alternatives to the enacted map. They still prove, that is, that Pennsylvania’s geography didn’t cause the enacted map’s pro-Republican skew.

The state’s final expert, Professor James Gimpel, points out that the enacted map achieves certain nonpartisan goals. Its districts are equal in population, it splits three fewer counties than its predecessor, it creates a black-majority district in Philadelphia, and so on. The map’s drafters deserve a gold star for not violating any other legal requirements. But their compliance with nonpartisan criteria obviously doesn’t mean that their creation isn’t a partisan gerrymander. A map can be valid on one ground yet clearly unlawful on another.

Why are the state’s lawyers attempting one unlikely gambit after another? And why are the state’s experts aiming for the capillary rather than the jugular? The questions answer themselves. Pennsylvania’s congressional map may be the most extreme gerrymander of the last half-century. One’s only choice is to distract when one is trying to defend the indefensible.

Nicholas Stephanopoulos, a law professor at the University of Chicago, will be among the panelists discussing “Gerrymandering and the Future of American Democracy” at 6:30 p.m. Monday at the National Constitution Center. To register, visit or call 215-409-6700.