As a practical matter, the issue in Louisiana v. Callais is not whether LA will have a second Voting Rights Act (VRA) district, in which black voters will have an equal opportunity to elect their candidate of choice. The issue is where in LA that district will be located. And if the Court invalidates the current district, that will cause significant political headaches for Republicans in LA and in Congress.

At an earlier stage of this litigation, the federal courts held that Section 2 of the VRA required LA to create a second VRA district. The plaintiffs had identified areas of the state in which a reasonably configured, second majority-black voting-age population could be created. But the LA legislature has no obligation to draw that district in the area the plaintiffs had identified; as long as it creates a second VRA district, it has complied with its VRA obligation.

Prior to this holding, Republicans held 5 of the 6 congressional seats. With the obligation to create a second VRA district, which inevitably would elect a Democrat, that meant one of these Republican incumbents would lose their seat. If the Republican legislature designed that district in the areas the plaintiffs had identified, it would have had to jeopardize the seat of the Speaker of the House, Mike Johnson, or the Majority Leader, Steve Scalise, or Rep. Julie Lettow.

Instead, the Republican legislature decided to eliminate an incumbent less popular among the Republican Party leadership, Rep. Garret Graves, from the Baton Rouge area. That’s why the legislature drew the current district that’s before the Supreme Court through the middle of the state. Graves was unpopular for many reasons, some of which this local story highlights. He opposed the current Republican Governor, Jeff Landry, in the last gubernatorial Republican primary, who was surely not unhappy to sign the bill eliminating Graves’ district; he was also a Kevin McCarthy supporter who was viewed as insufficiently supportive of Steve Scalise’s failed bid for House speaker.

If the Court strikes down LA’s attempt to create a second VRA district, while eliminating its least popular Republican incumbent, the LA legislature is still going to have to create that second VRA district. That will put the legislature back in the political position it was trying to avoid, in which it will have to figure out whether there is some other solution for its desire to protect its most powerful and most popular (with the party leadership) incumbents while also meeting its VRA obligations. Of course, if the Court had imposed similar constraints on partisan gerrymanders as it has on racial gerrymanders — as I argued it should have done in Principled Limitations on Partisan and Racial Redistricting — LA would not be able to defend against a racial gerrymandering claim with the defense that it was engaged in partisan gerrymandering.

Two caveats: at oral argument, some Justices suggested they wanted to re-visit the underlying issue of whether the VRA requires LA to create a second district at all. I’ll be surprised, though, if a majority of the Court decides to re-open that question from the earlier stage of this litigation. Second, if the Court strikes down the current LA district, that will have implications for racial gerrymandering doctrine more generally. How significant or limited those implications might be can’t be said in advance, given the intensely factual mix of race, politics, political geography, and other factors involved in the case.

At an earlier stage of Judge Griffin’s attempt to challenge the election results in the contest for the North Carolina Supreme Court seat, I had written that the state court of appeals decision to call into question over 61,000 ballots violated federal due process. The North Carolina Supreme Court ultimately agreed that those ballots were legitimately cast.

That left two much smaller categories of ballots being fought over, including ballots from military and overseas voters. Judge Griffin challenged the counting of those ballots in 6 counties, out of NC’s 100 counties. On this issue, the NC Supreme Court had agreed with Judge Griffin and held those ballots could not be counted unless those voters took additional steps to validate their ballots.

The federal district court has now rejected the North Carolina Supreme Court’s position. The federal court held that treating these military and overseas ballots differently in 6 counties from the rest of the state would violate Bush v. Gore’s Equal Protection holding. That decision is clearly correct, in my view. The identical ballot cannot be treated as a valid vote in one county but treated as invalid in another county in a statewide election. There is no legitimate reason for doing so. Rich Bernstein made that point earlier on this blog.

I want to highlight the application of Bush v. Gore here because some commentators argue that the case has little precedential value, since it has not often been applied in the lower courts or the Supreme Court. First, that overstates the facts; the lower federal courts have applied it a number of times, and the district court decision here relies on those prior applications. Second, I have seen very few contexts — and none in subsequent cases the Supreme Court has heard — in which the principle of Bush v. Gore is legitimately implicated. To say the case has not been applied often is not to say anything meaningful unless there are a significant number of cases in which the principle of Bush v. Gore is implicated. There haven’t been many such cases.

But this NC judicial election case is one of those unusual cases in which that principle is clearly implicated. As I said above, there is no legitimate state interest that would justify treating the same military and overseas ballots differently in different parts of the state. And with the principle of Bush v. Gore directly implicated, the federal district court rightly held that the North Carolina Supreme Court decision could not be implemented because doing so would violate Bush v. Gore. So this case was a good test (if unfortunate, since the entire litigation has created great turmoil over this seat) of whether Bush v. Gore has teeth in the lower courts.

I’m happy to be doing this event shortly. The link to watch the event is here: https://hopkinsathome.jhu.edu/watch-now

The recent 2-1 decision of the North Carolina Court of Appeals calls into question the legitimacy of over 61,000 ballots cast in the Griffin-Riggs battle over a critical seat on the North Carolina Supreme Court.  The court’s decision appears to violate the well-established federal due process principles that govern the resolution of state (and federal) election disputes, as reflected in the Roe v. Alabama line of cases, Roe v. Alabama, 68 F.3d 404 (11th Cir. 1995), as well as other federal courts of appeals decisions.  In the aftermath of Bush v. Gore, I wrote about these due process constraints in state election-dispute litigation that apply to the way state courts interpret and apply state election laws.  See Judging “New Law” in Election Disputes, 29 Fla. State U. L. Rev. 691 (2001).  

I’ll first describe the federal cases reflecting this due process doctrine, then suggest how that doctrine applies to the NC litigation.  As former federal judge and noted academic Richard Posner has written: “Nothing is more infuriating than changing the election rules after the outcome of the election, conducted under the existing rules, is known.”  That understandable feeling is reflected in this line of due process election jurisprudence.

The Roe v. Alabama litigation also involved election to a state supreme court (for the chief justice position).  After the initial tally, the margin of victory appeared to be about 200-300 votes.  Litigation then ensued over 1,000 to 2,000 absentee ballots, which one side alleged were  illegal under state law because they were improperly notarized or witnessed.  As is often the case (and is the case in NC) parallel litigation then began in both the Alabama courts and the federal courts.  Eventually, the Eleventh Circuit certified to the Alabama Supreme Court the question whether these were legal votes, and the Alabama court held that they were.  But that was far from the end of the matter.

The federal district court then found that, before this dispute, Alabama had consistently treated as illegal absentee ballots with these defects.  Thus, despite the state supreme court ruling that these were legal votes under state law, the district court concluded the state court interpretation had significantly changed Alabama election law in the guise of interpreting it.  In affirming the district court, the Eleventh Circuit held that this significant change in state election law violated Fourteenth Amendment principles of “fundamental fairness.” The Eleventh Circuit emphasized that (1) counting ballots that state law had previously excluded would dilute the votes of legal voters and (2) the change in the rules after the election would have the effect of disenfranchising those who would have voted but for the inconvenience imposed by the notarization/witness requirement.  The federal courts thus ordered the certification of the candidate who won without these absentee votes being included in the count.

The First Circuit reached a similar decision in Griffin v. Burns, 570 F.2d 1065 (1st Cir. 1978).  That involved a city-council primary race in Rhode Island, where the Secretary of State had concluded that absentee ballots could be used, as they were by statute for general elections.  State officials publicized the availability of these ballots.  Absentee ballots turned out to provide the margin of the victory, and the Rhode Island Supreme Court concluded that state law did not permit absentee balloting in primaries.  But the federal courts ruled that excluding absentee ballots in these circumstances violated due process.  The First Circuit held that where ”broad-gauged unfairness permeates an election, even if derived from apparently neutral action” an election process can reach a “point of patent and fundamental unfairness” triggering a due process violation.  “[D]ue process is implicated where the entire election process — including as part thereof the state’s administrative and judicial corrective process — fails on its face to afford fundamental fairness.”

The reliance interests of voters played a central role in Griffin.  The First Circuit did not hold that the state supreme court had misconstrued state law.  But the First Circuit found that the decision was inconsistent with the state’s longstanding prior practice and with the advice state administrative officials provided before the election.  The federal courts also found that voters casting absentee ballots had reasonably relied on past practice and the advice of these election administrative officials, and that many of these voters would have gone to the polls had they known that absentee ballots would not be valid.  Thus, excluding these absentee ballots violated due process.

In North Carolina, after the canvass Riggs is ahead by 734 votes out of 5,540,090 votes cast.  The state supreme court has stayed certification of the result until the courts resolve protest actions that Griffin has brought.  These protest actions challenge three categories of ballots for different reasons. 

The biggest category is voters with incomplete voter registration forms who voted in reliance on their county board’s notice that they were properly registered and on past practice (60,273 voters).  Second, military and overseas voters who did not provide a photo identification with their ballots (1,409 voters).  Third, children of military and overseas families whose NC registration is being challenged (267 voters).  I’ll focus on the first category.

In NC, the state Election Board is tasked by statute to develop a voter-registration form.  And in 2004, NC law was amended to require a voter seeking to register to provide a current, valid driver’s license number or the last four digits of their social security number; if they have neither, the Board is to assign them a unique identifier number. But the Board never amended the voter registration form until 2023.  The Board then amended its form to require only new voter registration applicants to include this information.

Griffin’s protest is thus to voters who registered using the Board form that existed before 2023 and fully complied with that form’s requirements.  The Board is required to notify voters who failed to provide the driver’s license or social security information, but the Board did not do that for these voters.  Indeed, all of these voters received notice in the mail from the Board at some point that they were properly registered to vote.  When they showed up to vote, they were on the registration rolls and no issue was raised.  In fact, many of these voters had voted for years without this information in their voter-registration record. 

So the Board made the mistake of not updating its voter registration form from 2004-2023 and voters had no way of knowing there was any issue about their registration.  Yet in casting doubt on over 60,000 votes that were cast, the majority’s substantive analysis of this issue is about one page long (pp.24-25).

Under Roe and Griffin, to exclude these points would almost certainly seem to violate due process.  The reliance interest of these voters is even stronger, in my judgment, than those at issue in Roe and Griffin.  As I noted above, many had been voting for years; the Board had notified them at some point that they were properly registered; no one had ever questioned their registration; and they accurately completed the form the Board required from 2004-2023.

None of the judges on the NC Court of Appeal appear to be aware of the federal due process election-law cases.  The dissenting opinion is very strong, in my view, and does highlight the way the majority opinion dramatically undermines the reliance interest of voters.  But he frames these interests in terms of the Purcell doctrine.  Purcell, though, applies to late in the day pre-election changes and is a rule of federal judicial practice, not one of constitutional law.  The real problem in the NC litigation over these so-called incomplete registrations is that to throw out the vote of these 60,000+ voters, under these circumstances, would be to violate the due process principles that apply to all elections.

Maire Le Pen’s National Rally party received the most votes in France’s recent elections and is the largest party in the National Assembly. Tomorrow, the judicial system is going to reach a verdict in a prosecution that seeks a criminal conviction and a five-year ban on running for office. Enormously important moment for both the courts and for politics in France.

This is from a paywalled article in the WSJ:

When judges reach a verdict in the embezzlement trial of Marine Le Pen on Monday, the fate of a politician with gravitational pull on France’s political system will be hanging in the balance.

Prosecutors are seeking a five-year prison sentence and a five-year ban from running for public office for Le Pen, a penalty that would exclude her from the next presidential race and thrust her party, National Rally, into limbo.

The question looming over France is whether the Paris tribunal—regardless of the underlying facts of the case against Le Pen—should deliver a ruling that shakes the country’s political order to its core.

“It’s my political death that’s being demanded,” Le Pen said in an interview on national TV, describing the proposed ban as “a very violent attack on democracy.”

For more than a decade Le Pen has been a mainstay presidential candidate. Her anti-immigration rhetoric has found growing support in France, positioning her as a front-runner for the presidential elections in 2027 when President Emmanuel Macron will reach his limit of two consecutive terms. 

The idea of a court decision eliminating Le Pen from contention has triggered a national debate over the reach of France’s fiercely independent judiciary branch. French prosecutors say no one, no matter their political status, should be above the law. But some of the country’s most prominent politicians, including avowed opponents of Le Pen, worry that a ban would sow distrust in the judicial system at a time when the institutions of France’s modern republic are increasingly fragile.

I’m struck by the similarity of the underlying factual contexts in these two cases. Only one Justice on the current Court, Justice Thomas, was on the Court that decided Shaw, and the similarity is not apparent from the opinion in Shaw. But for those of us who were engaged with the issues in Shaw at the time, the similarity is hard to miss.

The central legal question in both cases is this: when a State has justifiable reason to create a remedial VRA district, what constitutional constraints, if any, apply to the way that district is designed? 

In Callais, Louisiana believed it had to create a second VRA district to comply with federal court decisions that required such a district. The federal court had identified a region of the state in which, it found, a reasonably configured second VRA district could be created. But the Republican legislature preferred to draw that district in a somewhat different area of the state, in order to protect important Republican incumbents. The state designed a (less compact) district to be majority black, to comply with the VRA, but political reasons drove the decision to locate that district in specific areas of the state.

You can substitute North Carolina for Louisiana and Democratic for Republican and you’ll get the same story in Shaw. In Shaw, North Carolina believed it had to create a second VRA district to comply with federal rulings, this time from the Department of Justice (DOJ). The DOJ had identified a region of the State in which a reasonably configured second district could be drawn.  But the Democratic legislature wanted to protect an important incumbent who represented that area. So just as in Callais, the state designed a district a district to be majority black — the highway district, as it was called — to comply with the VRA, but political reasons drove the decision to locate that district in specific areas of the state.

It’s easy to miss the role that political considerations played in NC’s design of that district because the majority opinion does not address that issue and the dissent mentions it only in a footnote.  This is the relevant passage from note 10 in Justice White’s dissent:

This appears to be what has occurred in this instance. In providing the reasons for the objection, the Attorney General noted that “[f]or the south-central to southeast area, there were several plans drawn providing for a second majority-minority congressional district” and that such a district would have been no more irregular than others in the State’s plan. See App. to Brief for Federal Appellees lOa. North Carolina’s decision to create a majority-minority district can be explained as an attempt to meet this objection. Its decision not to create the more compact southern majority-minority district that was suggested, on the other hand, was more likely a result of partisan considerations. Indeed, in a suit brought prior to this one, different plaintiffs charged that District 12 was “grossly contorted” and had “no logical explanation other than incumbency protection and the enhancement of Democratic partisan interests …. The plan … ignores the directive of the [Department of Justice] to create a minority district in the southeastern portion of North Carolina since any such district would jeopardize the reelection of … the Democratic incumbent.” App. to Juris. Statement, O. T. 1991, No. 91-2038, p. 43a (Complaint in Pope v. Blue, No. 3:92CV71-P (WDNC)). With respect to this incident, one writer has observed that “understanding why the configurations are shaped as they are requires us to know at least as much about the interests of incumbent Democratic politicians, as it does knowledge of the Voting Rights Act.” Grofman, Would Vince Lombardi Have Been Right If He Had Said: ”When It Comes to Redistricting, Race Isn’t Everything, It’s the Only Thing”?, 14 Cardozo L. Rev. 1237, 1258 (1993). [italics added]

The one significant difference in the two contexts is that the remedial district in Shaw was in a completely different area of the state than the federal actor (there, the DOJ) had used as its basis for finding NC had violated its VRA obligations. The district at issue in LA includes about 70% of the population in the area that the federal actor (here, the federal court) had used as its basis for concluding LA had violated its VRA obligations.

I’m not going to go through here the whole analysis of how this difference does or should affect a full treatment of the doctrinal issues in Callais, which would require too much space to work through the quagmire of whether race “predominated” here and, if so, how the strict scrutiny analysis should be applied. Based on the oral argument, I also don’t have a confident view about how the Court is likely to decide the case.

But I wanted to note that the mix of race and politics in the two cases is very similar and that this might not be obvious to those who don’t recall the full context of Shaw. When federal authorities require partisan state legislatures to draw remedial VRA districts, politics ends up playing a significant role in where and how those districts are designed.

I’ll be participating in this important event tomorrow, which is being put on by the ABA Task Force on American Democracy.

The other participants are:

Judge Paul Grimm (Ret.) – David F. Levi Professor of the Practice of Law, Duke Law School; Director, Bolch Judicial Institute; Former United States District Judge, U.S. District Court for the District of Maryland

Judge Nancy Gertner (Ret.) – Senior Lecturer on Law, Harvard Law School; Former United States District Judge, U.S. District Court for the District of Massachusetts

Register here:

https://www.americanbar.org/events-cle/mtg/web/450088083

In light of the upcoming Supreme Court argument in Louisiana v. Callais, Travis Crum published a guest post at this blog arguing that racial gerrymandering should not be unconstitutional. Crum rejects the 30+ year line of cases that began with Shaw v. Reno, which the Court directly relied on in the 2010 round of redistricting to strike down Republican-enacted racial gerrymanders in Alabama, Virginia, and North Carolina.

I want to point out some of the consequences that would follow from Crum’s position, with which I disagree. It would permit legislatures to pack black voters by race into election districts at levels far higher than needed to ensure those voters have an equal opportunity to elect their preferred candidates. If black voters were able to elect their preferred candidates as long as the black voting-age population was 55%, for example, Crum’s position would permit a legislature to move voters by race to make that an 80% BVAP district instead. A legislature might seek to do that to minimize Democratic voting party in surrounding districts or to weaken black political influence.

The ban on racial gerrymandering prohibits a legislature from using race to create that result. That doctrine prohibits a State from the “excessive and unjustified use of race” in the districting process. If the Voting Rights Act requires the State to use race to design districts that comply with the VRA, that use of race is constitutionally permissible. But the State must have “a strong basis in evidence” for its position that the VRA plausibly requires that use of race. The State cannot unjustifiably pack black voters into districts by hiding behind the claim “The VRA made us do it.”

But without this constraint on racial gerrymandering, there would no legal means to challenge packing black voters by race into 80% districts. That cannot be challenged as illegal vote dilution under Sec. 2 of the VRA. To establish a Sec. 2 vote dilution claim, plaintiffs must show that black voters could be a majority in some other election district. Sec. 2 does not prohibit packing black voters by race into districts; packing becomes a violation only at the point at which the unpacked district could be split into two majority BVAP districts.

To put this another way, if black voters are entitled to one election district, due to racially polarized voting patterns and other factors, the fact that district might be 80% BVAP rather than 55% BVAP is legally irrelevant under Sec. 2 of the VRA. Either way, there is no vote dilution because black voters are a majority in the one district to which Sec. 2 entitles them.

This is exactly the problem we confronted in the first of the 2010 decade of Shaw cases, the Alabama Legislative Black Caucus v. Alabama case, which I argued before the Court in 2014. The Alabama legislature had packed black voters into districts at extremely high levels, on the claim the VRA required doing so. But none of those districts could be unpacked to create an additional VRA district.

Thus, the only hook we had for challenging excessive racial packing was to argue these were unconstitutional racial gerrymanders. The Court agreed with this analysis. In his dissenting opinion, Justice Scalia accused us (not entirely unfairly) of inventing “an entirely new argument” before the Supreme Court.

Race can be used in a variety of pernicious ways in the districting process. Without a constraint on racial gerrymandering, many of those means would not be capable of being legally addressed, as the 2010 round of cases from Alabama, North Carolina, and Virginia show.

Race can be used in districting when the VRA plausibly requires it, but not if race is being used in “excessive and unjustified” ways. To be sure, that can lead to difficult borderline cases to determine on which side of the line the design of a particular district falls. An amicus brief in Callais from a group of “Race and Democracy Scholars” details how the Court has navigated those doctrinal lines since the 1990s.

Of course, the Court could apply the racial gerrymandering doctrine in troubling ways. But that’s hardly unique to the constraints on racial gerrymandering. Crum’s position does have one current ally in the Court, Justice Thomas, who recently converted to a similar view. But I don’t think a majority of the Court is going to end the constitutional constraints on the “excessive and unjustified” use of race in the districting process.

In the NYT essay I wrote with Trevor Morrison on the tools courts have to deal with executive defiance of court orders, we concluded by noting that a major confrontation between the two branches might have significant effects on financial markets. These markets could provide a form of checks and balances in a world in which strong partisan attachments make Congress less likely to perform the role the Constitution’s designers imagined it would play.

The Financial Times quotes an analysis of this issue from Evercore ISI (no link to that report). Here’s the FT lead-in, then an excerpt from Evercore’s analysis:

Evercore’s broad conclusion is that the underlying case will affect whether and how much markets immediately react to the chaos, but that the real long-term danger is an insidious “erosion of market perceptions of US stability and safety”:

For example, if Trump defies courts in using the Alien Enemies Act of 1798 to deport suspected gang members without full due process protections, markets may not react. The issue at play is not fundamentally economic in nature, and while stripping of due process protections will certainly cause concerns, the Administration’s actions here will have been done on a relatively small scale and — the Administration would argue — only in response to the unprecedented circumstances of millions of excess immigration inflows relative to trend.

In contrast, if Trump defies a court order on a fundamentally economic or commercial issue — refusing a court order to pay a government contractor for work already completed, for instance markets might care more.

Even as certain norms around democracy and the rule of law have faced challenges in recent years, the U.S. judicial system has continued to function as an effective and independent mediator of economic and commercial disputes, providing an essential backbone to our free market system. Open defiance of a court order around payments or contracts would suggest that the U.S. government is no longer subject to the economic rule of law.

Next week, I’ll be delivering the 45th Sullivan Lecture at the Ohio Statehouse. Here are the details to register. I’m grateful to Jonathan Adler and Eugene Mazo, who will provide comments on the lecture, as well as to my host, Capital University.

I’ve posted this new essay, co-authored with Sam Issacharoff, at SSRN. The essay is forthcoming in 26 Theoretical Inquires in Law __ (2025). Here’s the abstract:

The declining ability of the state to deliver effective outcomes on the major issues of the day is among the greatest challenges democracies currently face. In this essay, we address two features of contemporary structures of democracy that contribute to this problem. One is excessive participatory requirements that have arisen based on the belief that government functioning must be more directly accountable to interested stakeholders. Process necessarily imposes its own costs and reformers too often think the problem with the functioning of government is always to expand participation even further. In the U.S., the way the practices of agency notice and comment have developed, as well as those of judicial review and other forms of accountability, now contribute to a long wind-up process for government action, too often followed by cost overruns, delay and, too frequently, abandonment of the project. This is the process that Francis Fukuyama aptly dubs the rise of “vetocracy.” 

A second means through which state capacity has declined reflects the weakening of the political parties. Absent strong parties to channel choice and enforce trade-offs between different constituencies, each proposed piece of legislation has to organize its own constituency out of disparate interest groups. The result is a decrease in legislation overall and a tendency for legislatures, even when in the hands of one party, to engage in largely symbolic legislation intended to signal virtue generally, even at the expense of accomplishing very much. These two features of contemporary practice contribute to the weakening of state capacity and the current dissatisfaction with democratic governments.

At Good Authority, the political scientist John Sides has a good summary of the findings in this important new study that’s received a lot of attention. The study, by Stanford’s Chenoa Yorgason, has the title above. Here is from Sides’ summary:

There is indeed evidence that small donors are not moderates. They tend to give more to ideologically extreme candidates. But maybe small donors are simply supporting candidates whose views they already share. The question is whether small donors are truly changing how politicians behave.

According to new research by Stanford’s Chenoa Yorgason, the answer is yes. Yorgason focuses on the adoption of ActBlue, the most prominent online fundraising platform for Democratic candidates and liberal groups.

ActBlue simplifies giving. You don’t have to repeatedly enter your personal information or credit card details. You can even schedule recurring contributions. In essence, you can donate with a single click. ActBlue takes a percentage of each contribution, but there’s no upfront fee or contract for candidates. Yorgason shows that from 2006 to 2020 more and more Democratic candidates adopted ActBlue. By 2020, the vast majority of Democratic candidates for U.S. House and Senate were using it. 

There is indeed evidence that small donors are not moderates. They tend to give more to ideologically extreme candidates. But maybe small donors are simply supporting candidates whose views they already share. The question is whether small donors are truly changing how politicians behave.

According to new research by Stanford’s Chenoa Yorgason, the answer is yes. Yorgason focuses on the adoption of ActBlue, the most prominent online fundraising platform for Democratic candidates and liberal groups.

ActBlue simplifies giving. You don’t have to repeatedly enter your personal information or credit card details. You can even schedule recurring contributions. In essence, you can donate with a single click. ActBlue takes a percentage of each contribution, but there’s no upfront fee or contract for candidates. Yorgason shows that from 2006 to 2020 more and more Democratic candidates adopted ActBlue. By 2020, the vast majority of Democratic candidates for U.S. House and Senate were using it. 

That line is from this tweet, by David Shor, who is one of the top political data analysts on the Democratic side. You will see that he references a new academic study that makes this point. As readers of this blog know, this is also something I have been saying in my work, such as this article in the Yale Law Journal Forum.

That’s one of the bottom-line findings that will surprise many in today’s Washington Post article that Nick Stephanopoulos co-authored and blogs about below. The five states most biased toward one party or the other are CA (+4.5 D seats); IL (+3.5 D seats); NC (+2.4 R seats); NY (+2.2 D seats); and NJ (+2.0 D seats).

Using the same metrics, in the first elections after the 2010 Census and round of redistricting, the House had a 27 net seat bias toward Republicans. All four subsequent elections in that decade maintained that R bias, though by lesser amounts: 2014: +7.4R; 2016 +17.6R; 2018 +15 R; 2020 +5 R.

These conclusions are all based on using one particular metric, known as the efficiency gap, for measuring the partisan bias in plans. There are a variety of other approaches to defining a “fair map” in partisan terms, and I have serious reservations about the efficiency gap. Nonetheless, it’s a commonly used metric and journalistic coverage often relies on this metric. So these are important findings and it’s good to see that Nick et. al. have taken their academic work and presented its findings for a general public audience.