Late last week, the Missouri Supreme Court blocked the use of new congressional districts—drawn by state lawmakers in September of 2025 in a statute known as HB1—for the upcoming fall 2026 congressional elections. HB1 was part of a movement by a number of states throughout the nation to redraw each of their congressional district lines, seemingly to enhance the prospects for either the Republican or Democratic Party, depending on the redness or blueness of the state in question. In Missouri, HB1 altered the previous district lines (adopted in 2022) in ways that appear to benefit the Republican Party, adding one more likely Republican seat among the state’s eight congressional districts.
A few months after HB1 was enacted but prior to the date it was to become effective, opponents of HB1 submitted (on December 9, 2025) 300,000 signatures of Missouri voters to the Missouri Secretary of State, along with a petition to place a referendum on HB1 on the state’s November 2026 ballot. Under the Missouri Constitution, if a valid referendum petition is made and supported by enough signatures, then the legislative measure in question is in effect blocked—or stayed—as of the date of the petition, until after the referendum vote is held, and “shall take effect [only] when approved by a majority of the votes cast thereon, and not otherwise.”
Accordingly, if the referendum petition and signatures filed on December 9, 2025 were legally adequate under Missouri law, then the new district lines provided for in HB1 were prevented as of that date from becoming law, and remain legally ineffective until they are approved by voters this November. This would mean that the old (2022) lines would have to be used until then, including in this fall’s election to pick congresspersons from Missouri to serve in the next Congress.
I said “if” in the previous paragraph because not all referendum filings meet state requirements to trigger a referendum vote. Under Missouri constitutional and statutory law, the Missouri Secretary of State is charged, in the first instance, with “examin[ing] the petition to determine whether it complies with the Constitution of Missouri,” and then “issu[ing] a certificate of sufficiency or insufficiency as to the petition within… [a statutorily prescribed] timeframe.” Right before that statutory timeframe expired, some eight months after the petition was filed, Missouri’s Secretary of State (Denny Hoskins) rejected the petition on the ground that Missouri’s referendum device is not available in the context of congressional redistricting. Per the Missouri Supreme Court:
At about 4:00 p.m. on August 4 [the date of the state’s primary election], the secretary issued a “Certificate of Insufficiency of Petition” finding “the referendum petition is insufficient under the Missouri Constitution because the Missouri Constitution does not authorize a referendum on congressional redistricting plans passed by the General Assembly.” The secretary directed the referendum…should not be placed on the November 2026 general election ballot.
Thereafter (that is, a month ago) one of the referendum petition proponents, Richard von Glahn, sued in state court to challenge the Secretary’s Certificate of Insufficiency, arguing that the Secretary misunderstood the Missouri Constitution.
The Cole County Circuit Court ruled against Mr. von Glahn, but last Thursday the Supreme Court of Missouri unanimously embraced his challenge. (Each member of the Supreme Court of Missouri is appointed by the Governor acting on input from a nonpartisan commission; for whatever it might be worth, five members of the current court were appointed by Republican governors, and two by Democratic governors, so, happily, the unanimous ruling cannot easily be attacked as partisan.) As the state high court observed, “[t]here is no dispute the referendum petition (2026-R004) seeking a referendum on…HB1 was timely filed [and] there is no dispute the referendum petition has the number of signatures required by the Missouri Constitution. The only issue is whether the Missouri Constitution authorizes the referendum petition as a matter of law.”
And on that question, the court found the text of the Missouri Constitution clear and dispositive: “Because the plain language of article III, section 49 of the Missouri Constitution authorizes a referendum as to ‘any act of the general assembly’ and no exception applies, the referendum petition was legal, sufficient, and timely.”
As to remedy, the court prescribed that:
the secretary [is ordered] to issue a certificate of sufficiency as to the referendum petition…, to place the referendum…on the November 2026 general election ballot, and to take any and all additional steps necessary to place the referendum…on the ballot at that election before September 8, 2026. The secretary and all of those acting in concert with him are restrained and enjoined from implementing, using, or mandating the use of the congressional redistricting in HB 1 for the November 2026 general election or at any other time thereafter unless and until HB 1 is approved by the voters at the November 2026 general election and takes effect…thereafter.
As straightforward as the Supreme Court of Missouri’s reasoning and command were, the court did not wade into the practical issues that will arise concerning compliance with the court’s edict. To be sure, the court declined to take account of “the confusion, expense, and practical difficulties that may result from not utilizing the congressional redistricting in HB 1 for the November 2026 election [that the Secretary relied on, because these arguments are] irrelevant to the secretary’s…basis for declaring the referendum petition insufficient [and because it was] the secretary’s delay [in issuing the Certificate of Insufficiency that] created the confusion, expense, and practical difficulties of which he complains.” But the fact remains that, because of the Secretary’s delay, the primary elections held last month used HB1’s lines. And employing different district boundaries for the general election than were used in the primary elections undoubtedly complicates matters.
Meanwhile, according to the New York Times, the Attorney General of Missouri was harshly critical of the ruling after it came down, vowing to take the matter up to the U.S. Supreme Court. According to the Times, Missouri Attorney General Catherine Hanaway characterized the ruling as an “unprecedented decision [that] clearly violates federal law and has thrust our state into a full-fledged constitutional crisis.” Late on Friday, AG Hanaway did indeed file on behalf of Missouri an application for a stay to U.S. Supreme Court Justice Brett Kavanaugh (who handles emergency requests from all the states, including Missouri, located within the United States Court of Appeals for the Eighth Circuit). Justice Kavanaugh has asked the referendum proponents to file a response by noon today.
Notwithstanding these harsh words from AG Hanaway, it seems rather unlikely the U.S. Supreme Court will get involved. Start with the so-called Purcell principle—under which federal courts should not wade into election matters so as to disrupt the status quo shortly before elections are held. But beyond that, there is the simple matter that Missouri does not seem to have a strong federal claim.
The stay application makes a handful of arguments. The one Missouri leads off with and spends the most time on is that the Missouri high court ruling runs afoul of the Elections Clause of the U.S. Constitution (Article I, section 4), which provides that the “times, places and manner of holding elections for [U.S.] Senators and Representatives, shall be prescribed in each state by the legislature thereof [subject to congressional override].” Missouri acknowledges, as it must, that the U.S. Supreme Court has repeatedly (and as recently as 2023, in Moore v. Harper) made clear that the word “legislature” in the Elections Clause does not prevent a state from vesting power to regulate congressional elections in persons or institutions other than the regularly elected state lawmakers. That is why state supreme courts can invoke state constitutions that regulate federal elections, why governors can be involved in vetoing congressional districting measures, and why states can make use of independent redistricting commissions, instead of elected lawmakers, to draw district lines. But Missouri argues that allowing approximately 3.3% of Missouri’s voters (the threshold currently required for triggering a referendum) empowers a tiny minority of Missourians to effectively block the will of the majority (as reflected in enacted legislation) in a way that does violence to the Elections Clause. According to Missouri’s stay application, “the Elections Clause forbids permitting a small fraction of a State’s voters from suspending a congressional map—and imposing a new map—before a statewide vote.”
Putting aside the wisdom vel non of having referendum devices in state constitutions, the claim that Missouri’s referendum device violates the Elections Clause is weak. For starters, as long as a state respects fundamental guarantees of republican government (which require that the people of the state can alter their forms of government when they choose), the Tenth Amendment permits each state to empower institutions that do not necessarily abide by transient majority preferences to regulate federal elections. That is the lesson of the Supreme Court’s decision in 2015 to uphold Arizona’s use of an independent districting commission to draw congressional districts, a lesson and result reaffirmed in Moore eight years later.
But even putting the Arizona case to one side, Missouri’s stay application misframes the alleged majoritarian/minoritarian conflict. The referendum provisions of the Missouri constitution reflect the will not just of the 3.3% (or more) of voters who trigger any particular referendum, but the will of the people of Missouri more generally to proceed with prudence and caution when tens of thousands of voters formally weigh in via referendum signatures. After all, ultimately the will of a majority of Missouri voters will determine the fate of HB1; the referendum is simply a brake on the process of lawmaking for some period of time. It is quite possible that a majority of Missouri voters will reject HB1, and the referendum device will be seen as having vindicated, rather than frustrated, majoritarian sentiment; in those states where it exists, the referendum device is itself a reminder that elected legislatures do not always reflect the wishes of the majority.
Nor is the referendum device the only kind of brake on the legislative process that is permissible. Imagine the Missouri constitution provided that all bills relating to districting (or all bills generally!) had to be approved by the legislature twice—once before and then again after a general election had been held—before being enacted. Such a requirement (which might be justified on the ground that the voters need a chance to vote out of office legislators who supported bills unpopular with the people) could be seen as having the effect of frustrating the will of a majority of lawmakers for some period of time, but states have the latitude to do that. As I explained in a previous column, this scenario is similar to what happened recently in Virginia, and the U.S. Supreme Court did not interfere with that setup. (The 27th Amendment to the U.S. Constitution does something similar for laws increasing congressional pay.)
In these respects, the referendum is not unlike a filibuster device (which certainly states could employ in their own legislative bodies just as the U.S. Senate does). While it may seem “minoritarian” for 41 Senators to be able to block the will of the other 59, this minoritarian objection dissolves once we recognize that 51 Senators can, if they choose, get rid of the filibuster and impose the will of a simple majority. But the filibuster is designed to slow down the process and encourage deliberation before action. So too, Missouri can eliminate the referendum altogether (or in certain settings, as it has) if it desires, but in the meantime the existence of the referendum device reflects the sense by the people of Missouri to proceed cautiously in passing laws to change the status quo, where there is non-trivial, organized pushback by voters.
These analytic critiques of Missouri’s position seem overwhelming, but probably the biggest problem with Missouri’s claim is that it is foreclosed by Supreme Court precedent squarely on point that goes back over 100 years. One of the seminal cases on which Moore relied—and that Moore emphatically reaffirmed—unanimously upheld states’ ability to use the referendum device in the context of congressional-district drawing. This 1916 case, Ohio ex rel. Davis v. Hildebrant, involved Ohio’s use of the referendum in federal-election regulation, and while Ohio law and Missouri law can obviously differ, the Missouri court last week reasoned that “[t]here is no meaningful difference between [the text of the Missouri constitution] and the Ohio constitutional provision the Supreme Court upheld in Hildebrandt [sic] in 1916.”
In its stay application papers, Missouri argues that the referendum upheld in Hildebrant did not empower minority dissenters in the same way that Missouri’s referendum does, but never meaningfully explains why. Ohio’s referendum device at issue in Hildebrant required the signature of 6% rather than 3.3%, but certainly that numerical difference is beside the point if the claim is that a very small minority should not be able to slow down an enactment that had majority support among lawmakers. And Ohio’s referendum device, just like Missouri’s, mandates that once a petition with the requisite number of signatures has been filed, the measure in question is put on hold until after the referendum election. Missouri’s stay application seems to have no answer at all to Hildebrant, a case extensively relied on by the Supreme Court of Missouri.
Beyond this Elections Clause claim, Missouri makes a few other arguments in its request for a stay. First, Missouri argues that to use different districts for the general election than were used in the primary election (an aspect of the current case that does make it different from the scenario in Hildebrant) effectively nullifies the primary election and disenfranchises the people who cast ballots in the primary, violating their constitutional right to vote in House elections that is protected under Article I, section 2, which provides that members of the U.S. House shall be “chosen…by the people.” To be sure, using different district lines for the general election than were used in the primary is suboptimal. But primary election results have been disregarded in other settings, and Missouri cites no U.S. Supreme Court case suggesting that such a scenario necessarily violates the right of the people to select House members.
And there is a reason for that lack of judicial precedent: to use HB1’s district lines instead of the 2022 lines for the November elections would be, according to the Supreme Court of Missouri, to hold an election employing district lines that lack and have never had the force of law (insofar as the legally sufficient referendum petition blocked, under Missouri law, HB1 from ever taking legal effect until the voters ratify it). If it violates the constitutional right to vote to have the primary voters’ wishes nullified, how could it not violate the right to vote to have the November election conducted pursuant to rules that are, essentially, made up in the sense that they have no legal legitimacy? Missouri in its stay application papers asserts that “this case is not about the Missouri Supreme Court’s interpretation of the Missouri Constitution,” but if the Missouri Supreme Court’s interpretation of the state constitution is to be taken as valid (as it must, see below), then the “right to vote” challenge seems to be a loser.
None of this means that it will be logistically easy to run a general election that uses different district lines than were used in an already-held primary. One hopes that the election results in the primary election may offer good predictions about what the results would have been had the 2022 lines been used, as (according to the Missouri high court) they should have been. Perhaps also, given that ballots must be finalized soon even if the deadline for doing so might be modified somewhat, state political parties can quickly be consulted to help determine which candidate should represent each party in the general election.
But the big point here is that an ostensibly unlawful primary (that used lines that lacked the force of law) should not justify, much less compel, the use of similarly unlawful lines in a general election—especially where, even if the Secretary of State did not intentionally sit on the referendum petition until the primary election was imminent or had begun, he could have avoided much of the dislocation of which Missouri today complains had he issued the Certificate of Insufficiency months earlier, something easily doable in light of the purely legal grounds on which he ultimately relied. The case against using HB1’s district lines that lack the force of law for the general election boils down to this: Two wrongs do not make (or vindicate) a right (to vote). (Missouri also gestures in the direction of the Equal Protection Clause, saying it would be arbitrary to move voters into districts that are different from the ones used in the primary election, but there is surely nothing arbitrary about using lines that have the force of law under the state constitution.)
Finally, Missouri contends that state courts, when issuing rulings that affect federal elections, do not have “free rein,” and must respect the “deliberate choice” the U.S. Constitution makes to confer federal-election-regulation power to the “legislatures” of the states. This language is plucked from (and this argument doubles back to) Moore, the seminal ruling in 2023 by the U.S. Supreme Court rejecting the so-called “Independent State Legislature” theory but reminding that state courts, in interpreting state laws that regulate federal elections, are still subject to some limits.
But this language from Moore will not likely help Missouri obtain the relief it seeks. As I have explained in detail in previous writings, the language from the end of Moore v. Harper—admonishing state courts not to get too ambitious or creative when construing state laws that relate to federal elections—must be understood in light of what the Court said in the rest of Moore. Most importantly, the language at the end of the Moore opinion must make its peace with the emphatic declaration earlier in the ruling (that Missouri purports to accept) that the U.S. Constitution “does not preclude a State from vesting congressional redistricting authority in a body other than the elected group of officials who ordinarily exercise lawmaking power.” As the Missouri high court observed, Moore “rejected ‘the contention that the Elections Clause vests state legislatures with exclusive and independent authority when setting the rules governing federal elections.’”
In other words, under the Tenth Amendment a state can vest power to regulate federal elections in whatever body and pursuant to whatever lawmaking process the state chooses, provided the state respects the requirement of republican forms of government; the U.S. Constitution’s “deliberate” (Moore’s language) mention, in Article I and Article II (concerning congressional and presidential selection, respectively), of the “legislature[s]” of the states is not a conferral of power by the U.S. Constitution to any particular state entity such as Missouri’s elected legislature, or any particular process of lawmaking, but is instead simply a command that a state honor whatever system the state constitution itself has chosen to deal with federal election regulations; the federal Constitution merely requires that states follow their own laws—no more and no less.
But centuries of Supreme Court rulings make clear that the meaning of state laws is almost always a matter to be decided by state, rather than federal, courts. (Again, Missouri purports not to be challenging the Missouri Supreme Court’s construction of state law, but in effect Missouri seems to be doing just that.) As I have previously suggested, so long as state courts are interpreting state constitutions in ways that would not violate the U.S. Constitution if they are applied to state (as distinguished from federal) elections, then those same state court rulings can apply to federal elections:
[If a federal court believes] that state courts have misbehaved by flouting state law in violation of the Elections Clause [of Article I], then the federal court would [also] have to be prepared to ensure that state elections are free from the same egregious state-court action. Lawlessness so severe as to violate the Elections Clause would also seem inevitably to run afoul of many federal constitutional provisions when applied to state elections. A state-court ruling sufficiently corrupt, aberrant or irrational to warrant federal-court oversight under the Elections Clause would also fail, for use in state elections, to survive even rational basis review under the Equal Protection Clause (which requires a rational fit to a permissible government purpose), much less the heightened scrutiny that is implicated on account of the fundamental “right to vote” in any election, state or federal. And state-court rulings concerning state separation-of-powers or state-law individual rights that are so far-fetched and unforeseeable as to not count as good-faith interpretations of state law would also violate the due process, fair notice, and rule-of-law concerns embodied in the Fourteenth Amendment (and, I would argue, also in the Guarantee Clause and elsewhere in the Constitution). Put conversely, if a state judicial interpretation is rational enough to pass federal constitutional review for purposes of state elections, it would be hard if not impossible to conceive of a basis for saying the state courts misinterpreted state law for purposes of [federal elections under Moore v. Harper].
To state this standard is to make plain how weak Missouri’s claim of federal violation is. The correctness of the Supreme Court of Missouri’s ruling might not be obvious to all observers (although it seems at first blush to be very convincing as a matter of textual interpretation), but surely the ruling is precisely the kind of thing that state and federal courts do all the time when exercising judicial review. To call this ruling—based tightly on the text of the state constitution—something other than “ordinary” would be to condemn hundreds if not thousands of judicial rulings by state and federal courts. And can anyone possibly think that the Missouri high court’s interpretation of the scope of state referendum powers would violate the U.S. Constitution if only state (and not federal) elections were at issue? Of course not. All of that is why the U.S. Supreme Court will not likely do anything to interfere with the Missouri high court’s decision.