The US Supreme Court gerrymandering case

I do not have time to dissect the arguments before the US Supreme Court in the case concerning the permissibility of the partisan gerrymander in Wisconsin. It clearly is a case of great importance to issues we care about at this blog. So, feel free to discuss here.

I highly recommend two pieces by Michael Latner:

Sociological Gobbledygook or Scientific Standard? Why Judging Gerrymandering is Hard (4 OCt.)

Can Science (and The Supreme Court) End Partisan Gerrymandering and Save the Republic? Three Scenarios (2 Oct.)



19 thoughts on “The US Supreme Court gerrymandering case

  1. I wonder if the efficiency gap, proposed in the suit to measure level of gerrymandering, is consistent with the seat-vote equation. It is probably too premature to predict which way that the Supreme Court chooses to go will bring the US House of Representatives any closer to the adoption of PR, or even any state legislature, given how little mention it is getting in the media, even in relation to this lawsuit.

    • The Efficiency Gap test is best applied when each of two parties has close to 50% of the votes. According to the 7% threshold proposed in Whitford v Gill, if two parties divided both votes and seats in a 58:42 ratio, that probably would get flagged as a gerrymander in favour of the minority party! PR is partisan symmetry with responsiveness 1; EG=0 is partisan symmetry with responsiveness close to 2.

      Should the US Supreme Court accept the EG as one of a battery of tests for gerrymandering, it would be a step forward in limiting the practise. Whether that would be a step on the road to PR or merely regulating the rules of engagement between two dominant parties is another matter. If SCOTUS rules instead that gerrymandering is not a matter for the courts, that almost certainly isn’t the road to PR.

    • I’ve thought of trying to apply the S-V equation to gerrymandering, but have never seriously tried it. And I am not sure it is up to that task, frankly. If I had tried and found anything, I could have submitted an amicus brief!

  2. Today the US Supreme Court heard arguments concerning a Maryland gerrymander suit (Benisek v Lamone) in which Republican appellants asked for reversal of a lower court order.

    In 2010 the congressional seats for Maryland divided 6-2 in favour of the Democrats. (They would have divided 5-3 under PR.) In 2012, after redistricting under the control of state Democrats, the division was 7-1. Republicans from the flipped district (#6) claimed in a suit that they were targeted/punished for the way they had voted in 2010, in a manner that was deliberate and effective, in violation of the First Amendment.

    The claim is based primarily on a comparison in a single district between results obtained under the challenged map and under the map it replaced. In contrast, Wisconsin Democrats challenged the districting for the state assembly (Whittaker v Gill) by considering the ensemble of results statewide without direct reference to the previous map.

    The Justices gave the counsel of each side a rough ride during oral presentation. They saw many problems with manageability of the plaintiffs’ proposed test: would there be a suit every time a seat flipped following redistricting? Nor did they have any patience with the defendants’ position that the Maryland redistricting was not a gerrymander. It’s hard to say what this means for the Wisconsin case.
    There is a link to a transcript (‘Tr’) at

  3. The US Supreme Court continues to wrestle with the two cases of gerrymandering — Wisconsin’s state assembly districts and Maryland’s congressional districts. I have a modest proposal to help SCOTUS out: declare gerrymandering to be justiciable but kick the Maryland case down the road without ruling on whether that map is a gerrymander; uphold the lower court decision that the Wisconsin map is a gerrymander and must be redone. But …

    State that the Wisconsin decision did not rest on a single definitive test, but rather on the confluence of many factors: (1) means, through control of legislature and executive branches by a single party, (2) intent, through hiring an expert to confirm a model of partisan bias, (3) secrecy and absence of public input, (4) plaintiffs’ and defendants’ experts finding similar maps of partisan bias, allowing calculation of the (small) effect of political geography on the Efficiency Gap test, (5)poor adherence to traditional criteria such as compactness and respect for boundaries of other administrative units.

    Future drawers of gerrymandered maps might take care to conceal Intent, give token public consultation and weaken the Efficiency Gap test by denying that maps of partisan bias were used while challenging plaintiffs’ bias maps. In anticipation, SCOTUS could strengthen item #5 with a strong suggestion that a draft map should be offered for public input, accompanied by quantitative scorings for apportionment, compactness and respect for administrative boundaries. The implicit threat is that a map might be disallowed if plaintiffs presented alternatives having substantially better scores.

    This probably is too much like sociological gobbledygook or a steak rub recipe to be accepted by the present Court, however.

    • This almost reads like an attempt to say that gerrymandering is okay for Democrats, but not Republicans. Gerrymandering is wrong and should be considered wrong. Absolutely and always. Either it is okay for politicians to pick their constituents or boundaries must be drawn fairly to allow fair elections.

      The only real questions should be around which of blindly compact districts, respecting local boundaries, or drawing truly competitive districts should be the best option

  4. Mark,
    I completely agree with you that gerrymandering is wrong, whoever is doing it.

    In my view, the single-district test proposed by the Maryland plaintiffs was flawed: it would enable many claims where there was no gerrymander but could not identify Wisconsin-style gerrymandering based on detailed mapping of partisan bias. Instead, a SCOTUS ruling along the lines I suggested would immediately open the door to a challenge of the Maryland gerrymander for failure to observe “traditional criteria” (e.g. gross non-compactness of districts 3, 6, et al.).

    The issues in Benisek v. Lamone were whether the lower court erred in placing certain requirements on a claim of injury (punishment for having voted a certain way in a previous election) based on the First Amendment. By “kick the Maryland case down the road” I meant that SCOTUS could simply instruct the lower court on how to handle that First Amendment claim, without at this time saying whether the claim was well founded. The Court could then focus more sharply on the issues of the (better founded) Wisconsin case.

  5. The US Supreme Court has ruled that the Democrat plaintiffs in the Wisconsin gerrymander case had not proven their standing as individual voters in individual districts. In a Nelsonian move, the Court turned a blind eye on the plaintiffs’ claims that statewide results, in the form of a large Efficiency Gap, provided clear evidence of gerrymandering. Plaintiffs’ interest in “their collective representation in the legislature” was dismissed as “undifferentiated, generalized grievance about the conduct of government.”

    Instead, SCOTUS insisted that plaintiffs had to prove that they suffered vote dilution because each was in either a cracked or a packed district. Because they had failed to make that case, the plaintiffs were deemed to lack standing and the district court judgment was vacated.

    The Justices appear to be asking a plaintiff to square the circle in proving a single district is gerrymandered, given that a gerrymander involves at least two (and typically many) districts. Twenty-first century gerrymanders of the sort perpetrated in Wisconsin leave no blatant evidence (no “salamander” or “rabbit on a skateboard”). They may be distinguishable from “neutral” maps only when compactness or number of split municipalities is averaged over an entire state. It may be impossible to find a smoking gun in a single district.

    The Court did not dismiss the case — the Democrat plaintiffs will have a chance to re-present their case in the district court. Based the Court’s opinion, I think their chance of success is slim.

  6. Today the US Supreme Court heard argument in Rucho v Common Cause, the gerrymandering case from North Carolina (transcript at )

    Much of the discussion circled around the question why was it “unfair” if a 50:50 split in popular vote resulted in a 10:3 split in seats: was it because of a deviation from proportional representation? The conservative justices thought it relevant that the framers of the US Constitution made no mention of proportional representation. Counsel for Common Cause asserted that the proposed statistical tests didn’t use PR as a baseline.

    Why is the concept that a 1% increase in popular vote should give a 1% increase in seats a bête noire at SCOTUS? I wonder what they would make of Missouri’s recently passed state constitutional Amendment 1. Amendment 1 calls for a state “demographer” to design district boundaries such that the two leading parties have equal numbers of “wasted votes” and that the difference should remain close to zero for a +/-5% variation in popular vote. As a mathematical consequence, it has responsiveness that would cause about 2% of seats to flip for every 1% change in a party’s popular vote. (Many incumbent legislators are strongly opposed to this amendment!)

    Could SCOTUS accept responsiveness 2%/% as a design criterion but continue to baulk at 1%/%?

  7. Today the Supreme Court of the US ruled that gerrymandering was none of the Court’s business. Dividing 5-4, SCOTUS threw out the cases from Maryland and North Carolina in which lower courts had found districting maps to be impermissible gerrymanders.

    Since the Vieth decision 15 years ago, the Court’s position (a 4-1-1-2-1 split) was that courts might consider future gerrymandering cases, but that no manageable test had been presented. In the last two years some mathematical heavy artillery (Markov Chain Monte Carlo, Evolutionary Algorithms) has made it possible to put an exact number to how much a disputed map is an outlier from neutral computer-generated maps. Lower courts in Pennsylvania, Ohio and Michigan, as well as North Carolina, found the outlier analysis to be a manageable test.

    The SCOTUS majority repeated Justice Scalia’s argument in Vieth for gerrymandering being non-justiceable and dismissed the outlier test because, of itself, it could not set a value to protecting incumbency vs competitiveness vs proportionality. They said that a political solution to gerrymandering could come either from Congress or from people changing their state constitutions.

    By refusing to “get involved in politics”, has SCOTUS in fact made a political decision?

    • The offer of changing their state constitutions is particularly weak when John Roberts has a dim view of state’s taking Congressional redistricting away from state legislators

    • Yep. Absolutely. Moreover its a deeply political decision made by a 5-4 majority that includes the stolen seat of Merrick Garland. The Kagan dissent is both ferocious and well-reasoned:

      So the only way to understand the majority’s opinion is as follows: In the face of grievous harm to democratic governance and flagrant infringements on individuals’ rights—in the face of escalating partisan manipulation whose compatibility with this Nation’s values and law no one defends—the majority declines to provide any remedy. For the first time in this Nation’s history, the majority declares that it can do nothing about an acknowledged constitutional violation because it has searched high and low and cannot find a workable legal standard to apply.

      But in throwing up its hands, the majority misses something under its nose: What it says can’t be done has been done.

      • I find myself particularly disturbed by this new theory that courts cannot fix problems. It opens the door for new constitutional violations to go unpunished and unrestrained.

  8. A striking thing about the Majority opinion is how dated it sounds – as if large sections had been cut and pasted from Vieth (2004). You would never know the US is living through a revolution in generation and detection of gerrymanders by computer analysis of big data sets.

    For instance, the Majority questions the durability of gerrymanders by reference to two ill-fated dummy-manders of the 1980s, rather than considering the manifest durability of the two 2010s gerrymanders in the present case.

    They bang on for four paragraphs about proportional representation and how the Constitution doesn’t call for PR, even though the “outlier” test of Rucho in no way depends on PR. (Not even the single point that Symmetry test or the Efficiency Gap have in common with PR for a 50:50 vote split.) See

    • If they had addressed the gerrymanders actually before them, they would also have had to address the judicial standards crafted by lower federal courts to address those gerrymanders. And that would have made nonsense, as Kagan J notes, of their claim that there are no judicial standards. The hollowness of the majority reasoning is most blatantly shown where they mention state-vel independent commissions as a solution, even though three of them have voted against those commissions in the past.

      The minority should really have done what the minority did in Bush v Gore. They wrote ‘I dissent’, not the traditional ‘I respectfully dissent’.Apparently that was earthshaking. This decision is more poorly argued than Bush v Gore and its effects will outlast that decision by many years.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.