Summary
reversing district court order that stayed Michigan's state ban on paying someone to provide transportation to polls and crediting state's interest in preventing fraud
Summary of this case from Texas Democratic Party v. ScottOpinion
No. 20-1931
10-21-2020
ORDER
The district court enjoined Michigan's enforcement of a state statute designed as a prophylactic measure against voter fraud. After the state attorney general declined to challenge the injunction, the two houses of the Michigan Legislature jointly sought an emergency stay of the injunction from the district court—which denied their motion—and now from this court. Because the legislature has standing to appeal the order granting the injunction, the state statute is likely not preempted by federal law, and the balance of equities weighs in favor of staying the district court's order, we grant the legislature's motion.
I. PROCEDURAL SUMMARY
Three voter-advocacy organizations challenged two Michigan election statutes in the district court, one regulating absentee ballots—not at issue here—and another mandating that no one "hire a motor vehicle or other conveyance or cause the same to be done, for conveying voters, other than voters physically unable to walk, to an election." Mich. Comp. Laws (MCL) § 168.931(1)(f), which we denote as the voter-transportation law. While Michigan Attorney General Dana Nessel was the named defendant in the district court, four other parties moved to intervene in the case as defendants: both houses of the Michigan Legislature, the Michigan Republican Party, and the Republican National Committee. All four were granted permissive intervenor status.
The district court later denied the voter-advocacy organizations’ motion for a preliminary injunction against enforcement of the absentee-ballot statute but granted their motion to preliminarily enjoin enforcement of the voter-transportation law. The four intervenors appealed—the legislative parties in this docket and the Republican-Party parties in Docket No. 20-1940, not currently before us—but Nessel did not. When all four intervenors moved in the district court for an emergency stay of the injunction pending appeal, Nessel declined to take a position on the motion. The district court denied the intervenors’ motion for a stay, and the Michigan House of Representatives and Senate now move this court for an emergency stay of the district court's injunction. The voter-advocacy organizations have responded in opposition, and we have granted several election-law scholars leave to appear as amici curiae in support of none of the parties.
II. ANALYSIS
A. Standing
A party seeking to invoke a federal court's jurisdiction must have standing to do so. Lujan v. Defs. of Wildlife , 504 U.S. 555, 560, 112 S.Ct. 2130, 119 L.Ed.2d 351 (1992). The familiar requirements for standing from Lujan —a "concrete and particularized" injury that is "actual or imminent"; a "causal connection between the injury and the conduct complained of," and a likelihood "that the injury will be ‘redressed by a favorable decision,’ " id. at 560–61, 112 S.Ct. 2130 —usually describe the requirements for a plaintiff who seeks to bring a complaint in federal district court. But standing doctrine also applies to appeals. Va. House of Delegates v. Bethune-Hill , ––– U.S. ––––, 139 S. Ct. 1945, 1950–51, 204 L.Ed.2d 305 (2019). Although a party may generally intervene in a district court proceeding without showing that it would have standing, if an adversely affected plaintiff or defendant does not appeal the district court's ruling, then the intervenor must satisfy the Lujan requirements to bring an appeal itself. Id. at 1951. Here, the two houses of the Michigan Legislature claim that the legislature, as an institution, is injured by the district court's suspension of enforcement of the voter-transportation law.
The law of legislative standing touches both the separation of powers and principles of federalism. As this case shows, there is unfortunately still not a wealth of guiding precedent on the ability of a state legislature to defend a law when no one else will. Such guidance would be especially useful to deal with the instances where a single state executive official (and a single plaintiff and a single trial judge) could nullify the people's will, as expressed through its democratically elected legislature, without the possibility of a means of review.
One useful data point we do have is United States v. Windsor , 570 U.S. 744, 133 S.Ct. 2675, 186 L.Ed.2d 808 (2013). In Windsor , the district court had held (and the Second Circuit had affirmed) that section 3 of the Defense of Marriage Act, Pub. L. 104-199, 110 Stat. 2419, 2419–20 (1996), was unconstitutional. 570 U.S. at 754–55, 133 S.Ct. 2675. Even before the district court had ruled, however, the government had conceded that section 3 was unconstitutional and notified the Speaker of the U.S. House of Representatives, as required by statute. Id. at 753–54, 133 S.Ct. 2675. The Bipartisan Litigation Advisory Group (BLAG) of the House, a committee consisting of five House members, voted 3–2 to recommend that the House intervene in the case. Br. on Jurisdiction for Resp't the Bipartisan Legal Advisory Group of the U.S. House of Representatives at 5, Windsor , 570 U.S. 744, 133 S.Ct. 2675, 186 L.Ed.2d 808 (2013) (No. 12-307), 2013 WL 701229. The district court granted BLAG leave to intervene as an interested party. Windsor , 570 U.S. at 754, 133 S.Ct. 2675. The government appealed the district court's ruling to give Congress the opportunity to defend the statute before the Supreme Court and "to ‘recogniz[e] the judiciary as the final arbiter of the constitutional claims raised.’ " Ibid. (alteration in original) (quoting the U.S. Attorney General's message to Congress). And after the Court granted certiorari, the House voted to authorize BLAG to defend the statute's constitutionality in the case. Br. on Jurisdiction at 8.
The Court ultimately held that the United States, despite its agreement with the plaintiff on the underlying merits, maintained a sufficient interest for standing to appeal because the district court had ordered it to pay the plaintiff a tax refund. Windsor , 570 U.S. at 757–58, 133 S.Ct. 2675. Without formally reaching the question of BLAG's standing, the Court allowed BLAG, as an agent of a single House of Congress, to defend the interests of the United States when the Executive refused to. See id. at 761, 133 S.Ct. 2675. In doing so, the Court underscored that, although allowing Congress to defend a law in lieu of the Executive should not be a "routine exercise," to disallow it altogether would present "grave challenges to the separation of powers"—an Executive, acting alone, could nullify a legislative enactment without any judicial determination. Id. at 762–63, 133 S.Ct. 2675. Justice Scalia, joined by just two other justices, railed against both the Court's holding that there was a case or controversy at all, id. at 778–88, 133 S.Ct. 2675 (Scalia, J., dissenting), and the idea that Congress could have standing to defend laws in its own right, arguing that Congress should instead use its power of the purse to confront the President directly, not litigate through the courts, id. at 788–91, 133 S.Ct. 2675. But his views did not prevail.
Here, as in Windsor , the State of Michigan is injured in its sovereign capacity by its inability to enforce its duly enacted statute. The state executive has acceded to the district court's injunction of the voter-transportation law and declined to appeal. And, whereas the Court permitted an agent of just one House of Congress to defend the law in Windsor , both houses of the Michigan Legislature now act in concert to defend the voter-transportation law. Denying the legislature standing to defend its own law would allow the state executive to nullify a state statute without any ultimate judicial determination.
And, although we are not concerned here with the federal separation of powers as the Court was in Windsor , federalism requires us to respect the separation of powers a state has adopted for itself. See Tennessee ex rel. Tenn. Gen. Assembly v. U.S. Dep't of State , 931 F.3d 499, 515 (6th Cir. 2019) (examining Tennessee law to determine whether its legislature was authorized to speak on behalf of the state). Indeed, Michigan has adopted the same basic division of legislative, executive, and judicial power as the federal government, indicating the state has similar interest as the federal government against allowing the executive to nullify a law unilaterally. Mich. Const. Art. III, § 2 ("No person exercising powers of one branch shall exercise powers properly belonging to another branch except as expressly provided in this constitution."). The appellees point to a Michigan statute stating that the attorney general "shall" defend the state in court. MCL § 14.28. Of course, the United States has arguably stronger language in its regulations that "assigned" the Office of the Solicitor General to defend cases against the United States, 28 C.F.R. § 0.20, but the Office did not really discharge that duty in Windsor , and the U.S. Supreme Court permitted an agent of one house of Congress to do so. MCL § 14.28 also says that the attorney general "shall" intervene and appear in a case in which the state is interested upon request from either house of the legislature. And the Michigan House and Senate's appeal here is effectively such a request. Even though state law ultimately controls who may speak on behalf of the state in court, see Tenn. Gen. Assembly , 931 F.3d at 515 (citing Bethune-Hill , 139 S. Ct. at 1951–52, Hollingsworth v. Perry , 570 U.S. 693, 712–13, 133 S.Ct. 2652, 186 L.Ed.2d 768 (2013), and Arizonans for Official English v. Arizona , 520 U.S. 43, 65, 117 S.Ct. 1055, 137 L.Ed.2d 170 (1997) ), it is not clear from Michigan's statute what happens if its attorney general refuses to defend a case that Michigan law decrees that she "shall" defend. Looking beyond the statute at least gives some more guidance.
Michigan assigns to its legislature the responsibility of guarding election integrity. Mich. Const. Art. II, § 4 (2) ("[T]he legislature shall enact laws to ... preserve the purity of elections, to preserve the secrecy of the ballot, to guard against abuses of the elective franchise...."). Michigan's court of appeals has also recently recognized that the state legislature is "an entity that certainly has an interest in defending its own work," finding that the legislature had standing to "defend[ ] the constitutionality of several of its statutes, as well as the manner in which future elections are to be conducted" in Michigan. Mich. All. for Retired Ams. v. Sec'y of State , No. 354993, ––– N.W.2d ––––, ––– N.W.2d ––––, ––––, 2020 WL 6122745, at *3 (Mich. Ct. App. Oct. 16, 2020) (quoting League of Women Voters of Mich. v. Sec'y of State , 948 N.W.2d 70, 75 n.4 (Mich. 2020) (McCormack, C.J., dissenting from denial of motion to reconsider)). Much as the Windsor Court permitted BLAG (as an agent of the House) to defend the Defense of Marriage Act when the President would not, Alliance for Retired Americans permitted the legislature to defend one of its laws on appeal when Michigan's executive officers "declined to appeal" the decision in the lower court—which too had allowed the legislature to intervene after "the Executive Branch abdicated its role" in the case. Ibid. These authorities demonstrate at least that Michigan law authorizes its legislature, both houses acting in concert, to defend a state election law in court when the attorney general will not.
Even if the state legislature lacked authority to defend the law on behalf of Michigan in lieu of the attorney general, the district court's injunction does the legislature institutional injury in its own right. As the appellees agree in their brief, we have recognized that a legislature body suffers an injury sufficient to confer standing if that body's specific powers are disrupted. See Tenn. Gen. Assembly , 931 F.3d at 511. While the injunction is in effect, Michigan's legislature cannot enact any enforceable laws that even regulate hired voter transportation for federal elections. And even if the injunction lasts only a relatively brief time, the injury would be no less severe. Laws governing conduct related to polling places are effective only while polls are open—election days—and what promises to be a significant election day draws near. The legislature has lost the ability to regulate that election in a particular way. Because its powers to regulate elections have been disrupted, the Michigan Legislature has suffered a sufficient injury for standing.
And once the injury-in-fact is established, Lujan ’s two remaining requirements follow quickly here: the district court's injunction is certainly the cause of the legislature's injury, and we can redress that injury by staying the injunction.
B. Staying the Injunction
Because the district court denied the legislature's motion for a stay pending appeal, we may consider that motion now. Fed. R. App. P. 8(a)(2)(A)(ii). We consider the motion de novo because "we are not reviewing any district court decision or order." A. Philip Randolph Inst. v. Husted , 907 F.3d 913, 917 (6th Cir. 2018). To grant a stay, we must be satisfied that the balance of four factors—the Michigan Legislature's likelihood of showing that the voter-transportation law is enforceable, the likelihood and degree of irreparable injury to the legislature if we do not grant a stay, the prospect that the stay would substantially injure other parties interested in the proceedings, and the interest of the public in granting the stay—is sufficient to justify a stay. Mich. Coal. of Radioactive Material Users, Inc. v. Griepentrog , 945 F.2d 150, 153–54 (6th Cir. 1991).
1. Appellants’ Likelihood of Prevailing on the Merits
The district court found the plaintiffs likely to prevail in showing that the Federal Election Campaign Act (FECA), as amended, preempts the voter-transportation law. We disagree.
As first enacted in 1972, 52 U.S.C. § 30143 (formerly 2 U.S.C. § 453 ) read:
(a) Nothing in this Act shall be deemed to invalidate or make inapplicable any provision of any State law, except where compliance with such provision of law would result in a violation of a provision of this Act.
(b) Notwithstanding subsection (a), no provision of State law shall be construed to prohibit any person from taking any action authorized by this Act or from making any expenditure (as such term is defined in section 301(f) of this Act) which he could lawfully make under this Act.
Federal Election Campaign Act of 1971, Pub. L. 92-225, § 403, 86 Stat. 3, 20 (1972). It was amended in 1974 to read, in its entirety:
The provisions of this Act, and of rules prescribed under this Act, supersede and preempt any provision of State law with respect to Federal office.
Federal Election Campaign Amendments Act of 1974, Pub. L. 93-443, § 301, 88 Stat. 1263, 1289 (1974). Since then, it has been amended only once, to create an exception for state or local parties’ use of funds for office buildings. Bipartisan Campaign Reform Act of 2002, Pub. L. 107-155, § 103(b)(2), 116 Stat. 81, 87–88 (2002).
On the one hand, the statute, as currently written, contains broad preemption language. If that were all, it might be enough to support the plaintiffs’ argument. It is a bit strange, of course, that in the nearly 50 years since FECA was enacted, no one has tried to use it to challenge Michigan's statute or many other state statutes related to nonmonetary election expenditures. Still, alone, that fact might not move the needle enough to warrant staying the injunction.
For example, local bans on election-day alcohol sales, see, e.g. , Ky. Rev. Stat. Ann. § 244.480(3)(a) ; Ga. Code Ann. § 3-3-20(b)(2)(B), or state laws cited by state election officials to threaten companies offering free or discounted coffee to voters, see, e.g. , Cal. Elec. Code § 18521 ; see also Freebies for Voters May Break the Law , Long Beach Press-Telegram (Nov. 2, 2008), https://www.presstelegram.com/2008/11/02/freebies-for-voters-maybreak-the-law.
On the other hand, § 30143 also specifies that the "rules prescribed under" the FECA also preempt state law. The statute contemplates that a court will consider the regulations promulgated under it. If we turn to those regulations—specifically, 11 C.F.R. § 108.7 —the scope of FECA's preemption becomes less clear. Subsection (a) basically restates the statute in equally sweeping language. But immediately afterward, subsection (b) specifies three kinds of state laws that are preempted. Such a clarification would be wholly unnecessary if (a) truly were as sweeping as is claimed. And the three types of laws mentioned there are about campaign finance : the sources of funding and reporting on its collection and distribution. By ejusdem generis , the kind of state regulations contemplated as preempted likely do not include restrictions on selling alcohol on election day, treating voters to coffee, and transporting voters to the polls.
Subsection (c) then specifically sets out types of state laws that are not preempted. By expressly allowing many types of election regulations, subsection (c) contradicts a sweeping interpretation of subsection (a). And the kinds of state laws that are not preempted occupy some of the broad ground left open by subsection (b). In particular, (c)(4) specifically allows state laws prohibiting "false registration, voting fraud, theft of ballots, and similar offenses " (emphasis added).
The Michigan statute, enacted in 1895 and prohibiting hiring carriages to take ambulatory voters to the polls, is assuredly aimed at preventing a kind of voter fraud known as "vote-hauling." Vote-hauling can be a classic form of bribery—paying a voter to "haul" himself or herself (and maybe immediate or extended family) to the polls to vote. It is also a usual sink for election-day "street money" or "walking-around money," as shown in several Kentucky federal vote-buying cases. See, e.g. , United States v. Adams , 722 F.3d 788 (6th Cir. 2013) ; United States v. Turner , No. CRIM. 05-02, 2005 WL 4001132 (E.D. Ky. Dec. 16, 2005). Tracy Campbell, a professor of history at the University of Kentucky, wrote about vote-hauling in his book about the history of American election fraud. See Tracy Campbell, Deliver the Vote 276 (2005) ("While cast as a way to get voters to the polls, it was often little more than an efficient vote-buying operation that provided ‘walking-around money’ to those willing to sell their votes."); see also id. at 279, 337.
Other states have or have had laws forbidding "expenditures" for alcohol on election day for similar reasons. See, e.g. , Ky. Rev. Stat. Ann. § 244.480(3)(a) (authorizing local governments to ban alcohol sales on election days); Ga. Code Ann. § 3-3-20(b)(2)(B) (same) ; Act of June 21, 1954, No. 633, § 1, 1954 La. Acts 1145, 1145 (requiring bars to be closed until one hour after polls close); Iowa Liquor Control Act, ch. 24, § 18(d)–(e), 1934 Iowa Acts 38, 46 (banning sales and delivery of alcohol on election days). Saloonkeepers often served as poll officers. Allowing bars to remain open on Election Day created the opportunity for a proprietorelection official to give alcohol to a voter—in exchange for voting a certain way? One would never know.
Of course, by raising the potential for fraudulent vote-hauling, we do not cast any aspersions on the appellee organizations or their motives. Not all vote-hauling payments are fraudulent, after all. A campaign might in all innocence pay a volunteer for his or her time and gasoline spent hauling voters to the polls. But a statute can be a prophylactic rule intended to prevent the potential for fraud where enforcement is otherwise difficult. Michigan's ban on paid voter transportation is one provision among several others in the statute intended to prevent fraud and undue influence. The statute forbids, for example, paying people for votes. MCL § 168.931(1)(a). Or threatening to fire workers for not voting a certain way. MCL § 168.931(1)(d). It also prevents religious leaders from using undue divine influence over their flocks. MCL § 168.931(1)(e). Moreover, the law was enacted in a way and at a time such that we can infer no invidious intent on the legislature's part.
We also make one last point regarding the district court's analysis of the statute as originally enacted. Act of May 13, 1895, No. 135, § 13, 1895 Mich. Pub. Acts 264, 267. The court read the 1895 statute to forbid only a quid pro quo—no hiring transportation in exchange for a vote —and found that the revised, current language was different and therefore had no connection to fraud or election integrity. (R. 79, Order, PageID# 1617–18.) The court read the phrase "for the purpose of securing such voter's vote, support, or attendance at such primary or convention" to apply to all the types of conduct prohibited by the act: hiring carriages or other conveyances for voters, soliciting persons to cast unlawful votes at primaries, offering voters money or a reward, treating or furnishing entertainment to voters, or promising voters a place or position.
The full text of this provision reads:
Any person who shall hire any carriage or other conveyance, or cause the same to be done, for conveying voters, other than those physically unable to walk thereto, to any primary conducted hereunder, or who shall solicit any person to cast an unlawful vote at any primary, or who shall offer to any voter any money or reward of any kind, or shall treat any voter or furnish any entertainment to any voter, or shall promise any place or position for the purpose of securing such voter's vote, support or attendance at such primary or convention, or shall cause the same to be done, shall be deemed guilty of a misdemeanor.
But, as the legislature points out, that reading does not make sense: It would be redundant to solicit a person to cast an unlawful vote at a primary "for the purpose of securing such person's vote, support, or attendance at a primary." The phrase is better read as modifying only behavior proscribed after the last "who shall," preserving the grammatical parallelism. So no quid pro quo was required to outlaw the paid provision of transportation in 1895. Nor is it meaningful to this case that the legislature allegedly omitted such a requirement in 1982, when it replaced "carriage" with "motor vehicle" and inserted the vote-hauling ban into a long list of other antifraud provisions. Act of July 1, 1982, No. 201, § 1, sec. 931(1)(k), 1982 Mich. Pub. Acts 574, 578.
2. Balancing the Equities
As described above, the legislature's likelihood of success on appeal is high. We now consider the remaining factors relevant to granting a stay. The harm to the legislature without a stay would be irreparable: November 3, 2020, will only happen once, and the legislature would lose its ability to regulate paid voter transportation for that election. Although prosecutions for illicit vote-buying would still be possible, enforcement would be far more difficult, requiring proof of a quid pro quo. And any vote-hauling fraud that does occur would still have affected the election itself.
On the other side, the harm to the voter-advocacy organizations appears modest. There are other ways, without violating Michigan's statute, to take voters to the polls. Volunteers can drive voters for free. Generally paid campaign workers—ones who are not specifically paid to take voters to the polls—may also fall outside the statute's ban, as might using cars that are commercially rented for many different campaign purposes, only some of which are to haul voters. So the organizations’ resources will likely not go to waste. And with the expansion of mailed ballots in Michigan this year, there are likely fewer voters who need to be driven to the polls at all.
These injuries also track the public interest, which lies in both fair elections—conducted with a minimum of fraud—as well as free elections—in which as many eligible voters can vote as desire to. A stay benefits the public interest more than harms it.
We also consider the potential for confusion coming from a change in election rules on the eve of an election. Purcell v. Gonzalez , 549 U.S. 1, 4–5, 127 S.Ct. 5, 166 L.Ed.2d 1 (2006) (per curiam). But we note—as amici urge—that this consideration is only one of many and is neither dispositive nor establishes a presumption against enjoining election rules close to election day. Although the injunction may delight some voters who had hoped to receive paid transportation to their precinct on November 3, it would not likely result in the voter confusion that Purcell cautions against or incentivize the electorate not to vote. So this factor does not weigh heavily in our decision to issue a stay.
III. CONCLUSION
For the reasons above, we GRANT the Michigan Legislature's emergency motion to stay the district court's injunction of the voter-transportation law.
COLE, Chief Judge, dissenting.
DISSENT
Today we eschew the constitutional limits of our power in holding that a state Legislature that suffered no injury may ask us to opine about the validity of a state law.
I. Standing
Standing is not a one-and-done issue. It must exist at every stage of the litigation Hollingsworth v. Perry , 570 U.S. 693, 705, 133 S.Ct. 2652, 186 L.Ed.2d 768 (2013). And a party seeking relief must show it has standing for every form of relief sought. City of Los Angeles v. Lyons , 461 U.S. 95, 109, 103 S.Ct. 1660, 75 L.Ed.2d 675 (1983). "When a private party has a judicial decree safely in hand to prevent his injury, additional judicial action requires that a party injured by the decree seek to undo it ." United States v. Windsor , 570 U.S. 744, 784, 133 S.Ct. 2675, 186 L.Ed.2d 808 (2013) (Scalia, J., dissenting). Here, the district court enjoined the state law and the Michigan Legislature must prove that it was injured by the injunction to obtain "extraordinary relief" in the form of a stay. Winston–Salem/Forsyth Cnty. Bd. of Educ. v. Scott , 404 U.S. 1221, 1231, 92 S.Ct. 1236, 31 L.Ed.2d 441 (1971).
A legislative body suffers an injury sufficient to confer standing where there is a "disruption to that body's specific powers." Tennessee ex rel. Tenn. Gen. Assembly v. U.S. Dep't of State , 931 F.3d 499, 511-12 (6th Cir. 2019). Legislatures have successfully shown institutional injury where the challenged law interferes with their constitutionally vested powers. Ariz. State Legislature v. Ariz. Indep. Redistricting Comm'n , 576 U.S. 787, 800, 135 S.Ct. 2652, 192 L.Ed.2d 704 (2015) (holding that a claim that a proposition interfered with the Legislature's constitutionally vested power over redistricting established standing). But a Legislature cannot make out a case of institutional standing where the injury instead arises from an "abstract dilution of legislative power." Tenn. Gen. Assembly , 931 F.3d at 512 (quoting Raines v. Byrd , 521 U.S. 811, 826, 117 S.Ct. 2312, 138 L.Ed.2d 849 (1997) ).
And yet, the Michigan Legislature alleges only an "interest in the enforcement and constitutionality of the paid-transportation ban." (Reply at 3.) That is the definition of an "abstract dilution of legislative power." Raines , 521 U.S. at 826, 117 S.Ct. 2312. At issue here is not the Legislature's power to legislate, but rather the enforcement of one of its laws. No legislative authority has been usurped, either temporarily or permanently. At this stage of the litigation, the district court has only preliminarily enjoined the state law, meaning the law will not be enforced for a short period of time while the court resolves the merits of plaintiffs’ claims. The preliminary injunction deprives the Legislature of nothing more than the temporary satisfaction of seeing one of its laws enforced.
And the Legislature has no cognizable interest in the law's enforcement. That interest belongs to the Attorney General, whom the State of Michigan tasks with enforcing the laws. The Attorney General would have standing in this case if she chose to appeal because she is being ordered to temporarily halt enforcement, a concrete injury to her powers. But unlike the Attorney General, the Legislature is not being ordered to "do or refrain from doing anything." Hollingsworth , 570 U.S. at 705, 133 S.Ct. 2652. The preliminary injunction only pauses the law's enforcement. To say that a brief pause in enforcement strikes at the core of the Legislature's constitutionally vested power strains credulity. Perhaps that is why the Michigan Legislature and the majority fail to cite a single case where the mere act of enjoining the enforcement of the law causes the Legislature such injury that it may invoke the judicial power of the United States.
What's more, finding the Legislature is injured any time a law is not fully enforced, simply by merit of having passed the law, would represent a sea change in legislative standing jurisprudence. If the Legislature had such an interest in the constant enforcement of its laws, the Executive branch would inflict a judicially cognizable injury on the Legislature every time it declines to enforce a law or enforces a law with less rigor than the Legislature would prefer. See United States v. Windsor , 570 U.S. 744, 789 n.3, 133 S.Ct. 2675, 186 L.Ed.2d 808 (2013) (Scalia, J., dissenting). A Legislature could seek an injunction any time the Executive declines to prosecute an offense against its laws. Surely that is not a coherent standing doctrine. The Legislature's interest in the constitutionality of its laws is doubly abstract in this case. Even if the Legislature were appealing from a final decision on the merits, it would have to show that invalidating a single law affects legislative power, not just legislative interests. That alone is a dubious proposition. The Supreme Court explained in Bethune-Hill that "[t]his Court has never held that a judicial decision invalidating a state law as unconstitutional inflicts a discrete, cognizable injury on each organ of government that participated in the law's passage." Virginia House of Delegates v. Bethune-Hill , ––– U.S. ––––, 139 S. Ct. 1945, 1953, 204 L.Ed.2d 305 (2019). In this case, we are one step removed from that question because the district court has not even issued a final ruling on the validity of the state law.
No court has ever extended such a sweeping invitation to Legislatures to call upon the powers of the courts without suffering an injury. Despite the majority's suggestion, Windsor certainly does not resolve this case. Indeed, the majority's exclusive reliance on Windsor is surprising given that the Supreme Court expressly declined to decide in that case "whether [the Legislative representatives] would have standing to challenge the district court's ruling" on its own authority. Windsor , 570 U.S. at 761, 133 S.Ct. 2675. The Court determined that the initial parties—the United States and Windsor—still had an active stake to meet Article III's standing requirements because the Executive refused to pay the money ordered by the district court. Id. at 757-58, 133 S.Ct. 2675 ("Windsor's ongoing claim for funds that the United States refuses to pay thus establishes a controversy sufficient for Article III jurisdiction."). Therefore, the Court did not require the Congressional representatives in Windsor to establish standing on their own behalf and allowed them to continue as intervenors. Id. at 761, 133 S.Ct. 2675. For Windsor to be even remotely analogous to this case, the Attorney General would need to remain involved. As such, the Windsor opinion offers no guidance in the instant case.
The dissenting Justices in Windsor —who disagreed that Article III requirements were met by the initial parties and had to reach the question of legislative standing—shed better light on our present dispute. According to Chief Justice Roberts and Justices Scalia and Thomas, Congress has standing "only to vindicate its own institutional powers to act," not "to correct a perceived inadequacy in the execution of its laws." Id. at 788-89, 133 S.Ct. 2675 (Scalia, J., dissenting). Justice Scalia thus proposes the same rule for legislative standing as proposed here in dissent: Where the "validity of a mode of congressional action" is challenged, the Legislature has standing to reassert its "institutional powers." Id. But a Legislature may not assert a concrete injury every time one of its laws is not fully enforced. Id.
Finally, a note on the majority's invocation of federalism. The majority assumes that because "Michigan has adopted the same basic division of legislative, executive, and judicial power as the federal government," that indicates that "the state has similar interest as the federal government against allowing the executive to nullify a law unilaterally." Maj. Op. ––––. That is a bold assumption that proves to be unsupported by Michigan's law. Only a matter of months ago, the Michigan Court of Appeals considered "whether or not the Senate and House of Representatives had standing to seek to uphold the constitutionality of a validly enacted [election law] statute" in light of the Attorney General's agreement with plaintiffs that the Act was unconstitutional. League of Women Voters of Mich. v. Sec'y of State , No. 350938, ––– Mich.App. ––––, ––– N.W.2d ––––, ––––, 2020 WL 423319, at *1 (Mich. Ct. App. Jan. 27, 2020). The Michigan Court of Appeals held that "[w]hile the Legislature asserts that it is the only real party in interest in ensuring that Michigan laws are enforced and upheld when the Attorney General will not do so ... we find the legislature's position unavailing. " Id. at ––––, 2020 WL 423319, at *6. In other words, the state court explicitly considered and rejected the fundamental basis of the majority's opinion. See id . at –––– n.10, 2020 WL 423319, at *7 n.10 (considering and rejecting the Michigan Legislature's argument that the opinion "will result in a single member of the executive branch being able to exercise unchecked veto power over a bill that has already been passed and enacted into law").
The Michigan Court of Appeals continued to analyze the Michigan Legislature's "interest in upholding the legislation that it has passed," finding "that injury is not personal or unique to the Legislature." Id. at ––––, 2020 WL 423319, at *7. The court explains "[t]his is particularly so, given that once the votes of the legislators have been counted and the statute enacted, ‘their special interest as lawmakers has ceased.’ " Id. (quoting Killeen v. Wayne Co. Road Comm'n , 137 Mich.App. 178, 357 N.W.2d 851, 855 (1984) ). Finally the court expresses concern that "[t]o accept the Legislature's argument that it has standing here would open the door for the Legislature to seek a declaratory judgment whenever the constitutionality of a statute was challenged." Id.
This leaves the majority's conclusions on shaky ground, particularly the notion that "Michigan law authorizes its legislature, both houses acting in concert, to defend a state election law in court when the attorney general will not." Maj. Op. ––––. Most of all it shows that while the majority pays lip service to federalism, its opinion actually takes sides in a controversial state debate. Far from respecting federalism, the majority uses an extraordinary remedy to throw federal judicial weight behind one branch of state government against the express disagreement of the other two branches.
A principled system of jurisprudence cannot treat the injury requirement as malleable clay, made to bend into whatever shape might be needed to reach the merits. I would stay within the limits of this court's constitutional authority and deny the stay for lack of standing.
II. Likelihood of Success
Though I would not reach the merits of this issue, given that the majority has offered its view of the merits, I offer mine here as well. I conclude that the Legislature would be unlikely to prevail on the merits because FECA preempts the voter-transportation law and there is doubt about the law's constitutionality.
FECA includes an express preemption provision, which states that "the provisions of this Act, and of rules prescribed under this Act, supersede and preempt any provision of state law with respect to election to Federal office." 52 U.S.C. § 30143(a). Notably, the federal law has broad preemptive scope. It preempts any state law "with respect to election to Federal office." Id. And Congress intended to occupy the field of federal election law. The House Committee drafted the preemption provision "to make certain that the Federal law is construed to occupy the field with respect to elections to Federal office and that the Federal law will be the sole authority under which such elections will be regulated." H.R. Rep. No. 1239-93, 93 at 10 (1974). To clarify FECA's broad statement of preemption, the FEC issued regulations that clearly delineate the kinds of state laws that are and are not preempted. The FEC lists three categories of state laws that are definitely preempted by FECA. 11 C.F.R. § 108.7(b). As relevant here, FEC regulations state that "Federal law supersedes State law concerning the ... [l]imitation on contributions and expenditures regarding Federal candidates and political committees." 11 C.F.R. § 108.7(b)(3).
Under the majority's reading, these subsections of the regulation narrow the scope of FECA's broad preemption provision. Maj. Op. ––––. But the FEC may only clarify, not alter or narrow, express statutory language. A more natural reading (and a reading that keeps the FEC within its regulatory authority) is that subsections (b) and (c) clarify, by way of concrete examples, which state laws are definitively covered or excluded by FECA's comprehensive scheme. But in no way do FEC regulations change the preemption language drafted by Congress nor does they absolve our responsibility as a federal court to interpret the statute's text.
The Michigan law limits expenditures regarding Federal candidates. Plaintiffs, a Super PAC and two other non-profits, believe that encouraging voter turnout would help their candidates of choice and want to spend their money to transport voters to the polls. But Michigan law prevents organizations from spending money to support federal candidates if that money goes to pay for transportation. The law definitionally limits expenditures, which are defined broadly as "anything of value" provided to "any [ ] person in connection with any election to" federal office. 11 C.F.R. § 114(a)(1). The Michigan law effectively sets a spending limit of $0 on transporting voters to the polls. This directly contravenes federal law which precludes states from placing limitations on expenditures regarding federal candidates. 11 C.F.R. § 108.7(b)(3).
Finally, federal preemption becomes even clearer when we look to other FEC regulations. FECA allows corporations and unions to "provid[e] transportation to the polls." 11 C.F.R. § 114.4(d)(1). Regulations note that "providing transportation to the polls" is an expenditure that is regulated by FECA unless it is conducted in accordance with paragraphs (d)(2)(i)-(v). Id. These provisions require, among other things, that the organization make rides available regardless of support for particular candidates.
These regulations demonstrate two things. First, as an interpretive matter, providing transportation to the polls is generally considered an expenditure. Congress and the FEC evidently thought that providing transportation was a contribution or expenditure because they developed detailed rules to exempt corporations and labor organizations from federal laws governing contributions. Second, as a preemption matter, FECA and subsequent regulations govern providing transportation to the polls and supersede contrary laws like Michigan's.
Perhaps this analysis clarifies why it is no great mystery that other laws such as "local bans on election-day alcohol sales" have not been challenged. Maj. Op. –––– n.1. Buying alcohol on election day is not an expenditure regarding a federal candidate. The bright line rule is clear: If a person makes a contribution or expenditure in support of a federal candidate, those expenditures are covered by FECA.
I now turn to address the majority's contention that the voter-transportation law falls within the carve-out for state laws that provide for the "[p]rohibition of false registration, voting fraud, theft of ballots, and similar offenses." 11 C.F.R. § 108.7(b)(3). The majority finds that the voter-transportation law is a similar offense. But paying for rides to the polls is not in any sense similar to theft of ballots or voter fraud unless the similarity is defined at such a level of generality as to allow the exception to swallow the rule. Apart from the majority's speculation that [t]he Michigan statute "is assuredly aimed at preventing a kind of voter fraud," there is no actual evidence that the law has any fraud-prevention purpose. Maj. Op. ––––.
The majority points to the fact that the voter-transportation law is "one provision among several others in the statute intended to prevent fraud and undue influence." The voter-transportation law is found in Act 116 which consolidates all Michigan election laws under section 931. That section lists all "prohibited conduct" that qualifies as a misdemeanor. MCL § 168.931. It is unsurprising that Michigan also prohibits bribing voters. The majority's argument can hardly be that because both acts are prohibited during elections, thus codified in the same section on prohibited conduct, both must be fraudulent.
Without any evidence of an anti-fraud purpose, we would need to conclude that voter-transportation fundamentally promotes voter fraud. This proposition beggars belief given that the activity is expressly permitted under FECA regulations and allowed in 49 other states. The majority's invocation of vote-hauling is unpersuasive. Plaintiffs want to rent buses to help people get to the polls; companies like Uber want to provide discounted rides to the polls in Michigan as they have in every other state. These prohibited activities are a far cry from the majority's specter of vote-hauling.
I think it important to note that the majority does not mention plaintiffs’ constitutional argument. Had they reached the issue, they may well have decided that the stay should be denied. Instead, without spending a single sentence on the merits of the constitutional issue, the majority grants a stay, reimplementing a potentially unconstitutional law days before the election.
I therefore respectfully dissent.