Last month, the Supreme Court dismayed opponents of partisan gerrymandering in two cases: Gill v. Whitford, a challenge to Wisconsin’s legislative map, and Rucho v. Common Cause, which challenged North Carolina’s congressional plan. In both cases, a Republican state legislature had drawn a map that, according to opponents, gave Republican candidates an unfair advantage in the state. The court remanded both cases back to the district courts for review under a new standard for challenging a partisan gerrymander.
What is that new standard? Chief Justice John G. Roberts Jr. made it clear in his majority opinion in Gill that the plaintiffs have to show — in specific districts in which some of them live — that they will have a harder time electing their preferred official because of the way the district has been either “packed” or “cracked.”
Identifying exactly which districts have been packed or cracked is challenging, but doable, as I will show below.
What is “packing” and “cracking”?
Before we go further, let’s explain packing and cracking, the two main terms of art in the gerrymandering literature. Packing means concentrating one party’s backers in just a few districts, so they win by overwhelming margins. Cracking means dividing a party’s supporters among several districts so that party has a harder time winning a majority in each one. There are other tools, such as separating minority-party incumbents from their previous supporters, but those two are the most common.
What are the consequences of having to prove partisan gerrymandering by district, not for the state as a whole?
First, successful challenges will not mean a whole state’s map must be redrawn. Rather, they will affect only a relative handful of districts — with some spillover effects, as adjacent districts will need to be redrawn.
Second, partisan gerrymandering opponents will have to come up with different types of evidence than they presented in these two cases. In these cases, expert evidence on partisan gerrymandering has involved statistical evidence about the effects of the plan as a whole, especially as compared with a politically neutral districting process. However, in two other states, Maryland and Michigan, partisan gerrymandering challenges are already being considered that do involve allegations of packing or cracking in particular districts.
Here’s the problem: If evidence about gerrymandering must be district-specific, it will be necessary to identify exactly which districts were (unconstitutionally) cracked and which were (unconstitutionally) packed. That is not easy.
So, how can we decide when an individual district has been packed or cracked?
Let me suggest two sources of insight.
First, the United States already has extensive case law on racial gerrymandering that can be borrowed to help assess partisan gerrymandering. On race, the Supreme Court has already taken a district-specific approach, as those interested can see in, say, Miller v. Johnson (1995). Borrowing from that line of jurisprudence makes sense, even though the court has always distinguished between racial and political gerrymandering.
Second, just this year, in League of Women Voters v. Commonwealth of Pennsylvania (2018), the Pennsylvania Supreme Court overturned the legislature’s congressional district map because of partisan gerrymandering and ordered its own plan used in 2018. While that case was decided on the grounds of state law, we can learn from it nonetheless.
Let’s look first at racial gerrymandering.
The majority opinion in Alabama Black Legislative Caucus v. Alabama (2015) in effect asserts there is no “magic number” that can uniformly be applied to all districts to determine when that district was drawn primarily because of race. Rather, the jurisdiction must maintain a minority’s ability to elect a preferred candidate.
In my report as special master in Personhuballah v. Alcorn (2015), where Virginia’s Third Congressional District was redrawn after race was found to be the preponderant motive in drawing the district, I reviewed the expert witness testimony and used this type of analysis to redraw the district in a constitutional fashion. Similarly, for partisan gerrymandering, a court could analyze each district separately to estimate how likely it might be that the minority party’s candidate could be elected. That would help determine whether a district has been illegally cracked or packed.
District geography is also relevant. In cases such as Miller v. Johnson and its offspring, such as Personhuballah, courts have used what are called “traditional factors” — such as whether the district unnecessarily fragments city or county boundaries — in analyzing the evidence. Similarly, we can look at whether there is a good explanation for why a district is tortuously shaped, or why particular pieces of electoral geography have been shifted from one district to another — or whether the only reason appears to be that the shift gives one party an advantage.
We can also learn from the Pennsylvania case.
The Supreme Court of Pennsylvania considered how much individual districts satisfied traditional factors such as preserving county boundaries, as shown in its review of testimony by Professor John J. Kennedy. The court, however, seemed to pay most attention to evidence about how much the legislature’s plan as a whole satisfied traditional factors as compared with politically neutral plans drawn by a computer, as you can see in its review of testimony by Professor Jowei Chen. That computer-simulation approach can be modified to help see whether any particular district or set of districts appears to be a statistical outlier, compared with what we would expect from a plan satisfying traditional factors.
No single measure will tell us everything we need to know.
In the Wisconsin and North Carolina cases, for example, the district courts also required plaintiffs to show that the legislature’s Republican majority intended to give itself an advantage over the Democrats; that those advantages could not be explained by chance; and that the gerrymander’s effects would last. Most courts will likely continue to expect challengers to show the same things when challenging individual district’s lines.
Many experts, including me, had hoped the Supreme Court would declare a standard for finding districts had been unconstitutionally gerrymandered for partisan advantage that looked at effects statewide. A district-specific standard is the next best thing, and completely manageable, if we adapt tools from the cases described above.
Bernard Grofman is the Jack W. Peltason Chair of Democracy Studies at the University of California at Irvine. He is a specialist on redistricting whose work has been cited in nearly a dozen U.S. Supreme Court cases. Most recently, he served as the special master to a federal district court responsible for the redrawing of county supervisor and school district lines in a small Utah County with a large Navajo population.