Christopher Graveline v. Jocelyn Benson

RECOMMENDED FOR PUBLICATION
Pursuant to Sixth Circuit I.O.P. 32.1(b)
File Name: 21a0072p.06

UNITED STATES COURT OF APPEALS
FOR THE SIXTH CIRCUIT

┐
CHRISTOPHER GRAVELINE; WILLARD H. JOHNSON;
│
MICHAEL LEIBSON; KELLIE K. DEMING,
│
Plaintiffs-Appellees, │
> No. 20-1337
v. │
│
JOCELYN BENSON, Secretary of State of Michigan; │
JONATHAN BRATER, Director of Michigan Bureau of │
Elections, in their official capacities, │
Defendants-Appellants, │
│
DANA NESSEL, Attorney General of Michigan, │
│
Intervenor-Appellant. │
┘

Appeal from the United States District Court
for the Eastern District of Michigan at Detroit.
No. 2:18-cv-12354—Victoria A. Roberts, District Judge.

Argued: December 16, 2020

Decided and Filed: March 29, 2021

Before: MOORE, GILMAN, and GRIFFIN, Circuit Judges.
_________________
COUNSEL

ARGUED: Erik A. Grill, OFFICE OF THE MICHIGAN ATTORNEY GENERAL, Lansing,
Michigan, for Appellants. Oliver B. Hall, CENTER FOR COMPETITIVE DEMOCRACY,
Washington, D.C., for Appellees. ON BRIEF: Erik A. Grill, Heather S. Meingast, OFFICE OF
THE MICHIGAN ATTORNEY GENERAL, Lansing, Michigan, for Appellants. Oliver B. Hall,
CENTER FOR COMPETITIVE DEMOCRACY, Washington, D.C., William P. Tedards, Jr.,
Washington, D.C., for Appellees.
MOORE, J., delivered the opinion of the court in which GILMAN, J., joined. GRIFFIN,
J. (pp. 30–38), delivered a separate dissenting opinion.
 No. 20-1337 Graveline et al. v. Benson et al. Page 2

_________________

OPINION
_________________

KAREN NELSON MOORE, Circuit Judge. This is the parties’ second time before our
court in regard to the constitutionality of a constellation of Michigan laws that, in combination,
govern an independent candidate’s ability to be placed on the ballot for statewide office.
Plaintiff-Appellee Christopher Graveline attempted to get his name on Michigan’s November
2018 general election ballot as an independent, non-partisan candidate for attorney general. The
other plaintiffs are registered Michigan voters who supported Graveline’s candidacy and who
intended to vote for him. Together, they challenge the Michigan laws that set forth the
requirements for Graveline’s name to appear on the general election ballot. Plaintiffs contend
that these laws deprive them of their rights to freedom of speech and association, equal
protection, and due process under the First and Fourteenth Amendments of the United States
Constitution.

Previously, this court denied Defendants-Appellants’ motion for an emergency stay
pending their appeal of the district court’s grant of a preliminary injunction in favor of Plaintiffs.
The district court preliminarily enjoined the enforcement of the provisions on the grounds that
Plaintiffs were likely to succeed in showing that the provisions violated their rights under the
First and Fourteenth Amendments. Subsequently, the district court granted Plaintiffs’ motion for
summary judgment, denied Defendants’ motion, issued a permanent injunction, and
implemented an interim requirement allowing independent candidates to qualify for statewide
offices by submitting a qualifying petition with 12,000 signatures. Defendants then filed a
motion to amend the district court’s findings to clarify the contours of the interim signature
requirement, which the district court denied. Defendants now appeal the district court’s grant of
summary judgment in favor of Plaintiffs and the denial of their motion to amend.

Applying the analytical framework set forth by the Supreme Court in Anderson v.
Celebrezze, 460 U.S. 780 (1983), and in Burdick v. Takushi, 504 U.S. 428 (1992), we hold that
the challenged provisions, applied in combination, impose a severe burden on the constitutional
rights of independent candidates and their potential voter-supporters. Because the provisions are
 No. 20-1337 Graveline et al. v. Benson et al. Page 3

not narrowly drawn to advance compelling state interests, Michigan’s statutory scheme for
qualifying independent candidates to be placed on the ballot violates the Constitution. Due to the
impermissible infringement on Plaintiffs’ constitutional rights, we also conclude that the district
court did not abuse its discretion in crafting its remedy. Thus, we AFFIRM the rulings of the
district court.

I. BACKGROUND

A. Factual Background

Michigan allows independent candidates running for statewide office to be placed on the
general election ballot if the candidate submits an affidavit and a “qualifying petition.” Mich.
Comp. Laws § 168.590c (2008). A qualifying petition must have at least 30,000 valid signatures
and must be submitted no later than “the one hundred-tenth day before the general election.” Id.;
id. § 168.544f. Additionally, a qualifying petition must be signed by at least 100 registered
voters in each of at least half of Michigan’s fourteen congressional districts (“geographic
distribution requirement”). Id. § 168.590b(4). Moreover, the signatures on a qualifying petition
must be obtained within 180 days of the filing deadline. Id. § 168.590b(3). The filing deadline
for the November 6, 2018 election was July 19, 2018. Therefore, the official process for an
independent candidate trying to run for attorney general in the November 6, 2018 general
election began in late January 2018—180 days prior to the July 19 deadline.

In contrast, the candidates for attorney general from the major political parties—
Republican, Democratic, and Libertarian—are nominated at their party’s convention rather than
elected in a primary. The party must hold its convention “not less than 60 days before the
general November election.” Mich. Comp. Laws § 168.591(1). In 2018, that deadline fell on
September 7, 2018. However, both the Republican and Democratic Parties held their nominating
conventions on August 25, 2018. R. 1-3 (Graveline Decl. ¶ 3 n.1) (Page ID #30). Given this
legally mandated timing sequence, an independent candidate likely would be ignorant of the
identities of the major party nominees before the independent candidate filing deadline.

Graveline’s efforts to qualify for the ballot reflect how the interplay of these deadlines
affect independent candidates. Graveline waited to begin his attempt to qualify for the ballot
 No. 20-1337 Graveline et al. v. Benson et al. Page 4

until June 4, 2018. R. 1-3 (Graveline Decl. ¶ 9) (Page ID #33). His lack of knowledge of the
major party nominees played a significant role in his delay. Graveline decided to enter the race
for attorney general only “when it became reasonably clear to [him] that the Democratic and
Republican Parties would be nominating candidates who d[id] not subscribe to [his] ideals.” Id.
¶ 7 (Page ID #32). Thus, Graveline decided to run after learning the identity of the candidate
that the Democratic Party informally “endorsed” in April 2018 and the two candidates of the
Republican Party who each “announced” at some undisclosed time. Id. ¶ 3 n.1 (Page ID #30);
id. ¶ 4 (Page ID #4). A second factor also led to some delay in the launch of Graveline’s
campaign: Graveline served as an Assistant United States Attorney, and the Hatch Act required
Graveline to resign from federal service before he could formally file as a candidate for an
elected office. Id. ¶ 8 (Page ID #32–33).

From June 5 until the July 19 deadline, Graveline, along with 231 volunteers and a
signature-gathering firm, collected 14,157 signatures. Id. ¶¶ 11, 15 (Page ID #34, 35). This
effort required 1,000 hours of volunteer time and the expenditure of $38,000. Id. ¶15 (Page ID
#35). Graveline attempted to file his petition on July 19, but the State rejected it because it did
not contain 30,000 signatures.1 R. 1-9 (Bureau of Elections Email) (Page ID #57).

B. Procedural History

Plaintiffs filed suit in the United States District Court for the Eastern District of Michigan
on July 27, 2018, and moved for a preliminary injunction on August 3. R. 1 (Compl.) (Page ID
#17); R. 4 (Pls.’ Mot. For Prelim. Inj.) (Page ID #91). After briefing and argument, the district
court granted the motion and ordered that “(1) Graveline must immediately present his
qualifying petition . . . to the Bureau of Elections; (2) The State must accept Graveline’s filing as
complete and determine the validity of the signatures in time to place Graveline on the ballot if
he has sufficient valid signatures; and (3) If Graveline has at least 5,000 valid signatures . . . [,]
his name must be placed on the November 6, 2018 general election ballot as an independent
candidate for the Office of Michigan Attorney General.” Graveline v. Johnson, 336 F. Supp. 3d
801, 817 (E.D. Mich. 2018).

1
Graveline was able to comply with the geographic distribution requirement.
 No. 20-1337 Graveline et al. v. Benson et al. Page 5

The district court issued its preliminary injunction after finding that Plaintiffs were likely
to succeed on the merits of showing that, under the Anderson-Burdick framework, “the
combination of Michigan’s ballot access regulations severely burdens their fundamental rights
under the First and Fourteenth Amendments.” Id. at 812. The district court also found that
Michigan “[fell] far short of satisfying its burden to show that the severe burdens caused by the
scheme are justified.” Id. at 815. Finally, the district court found that the remaining factors,
“whether Plaintiffs would suffer irreparable injury absent an injunction; whether an injunction
would cause substantial harm to others; [and] whether the public interest would be served by an
injunction[,] weigh[ed] in favor of an injunction.” Id. Defendants filed a notice of appeal on
August 29, 2018. The same day, Defendants filed a motion in the district court to stay the
preliminary injunction pending appeal. R. 14 (Defs.’ Mot. to Dist. Ct. for Stay) (Page ID #171).
The district court denied the motion for a stay on August 30.

On September 4, 2018, Defendants moved this court for a stay of the district court’s
injunction pending appeal. We denied the emergency motion to stay, holding, inter alia, that
based on the evidence in the record, Michigan’s laws, in combination, created a severe burden on
Plaintiffs’ rights, and that the laws were not “narrowly drawn to protect [the State’s] interests.”
Graveline v. Johnson, 747 F. App’x 408, 414–15 (6th Cir. 2018). Consequently, the Bureau of
Elections reviewed Graveline’s petition and placed him on the ballot for the November 2018
general election as an independent candidate for attorney general.

After discovery, both parties filed cross-motions for summary judgment, on which the
district court heard argument. R. 28 (Defs.’ Mot. For Summ. J.); R. 30 (Pls.’ Mot. For Summ.
J.). During the hearing, Plaintiffs clarified that, contrary to our statement in the denial of the
motion for an emergency stay, see Graveline, 747 F. App’x at 410 n.1, they maintain their
challenge to the geographic-distribution requirement because it “compound[s] the burden
imposed by the high signature requirement [and] the early filing deadline.” R. 60 (Summ. J.
Hr’g Tr. 30–32) (Page ID #1424–26). Proceeding with that clarification, the district court
granted Plaintiffs’ motion for summary judgment. The district court first determined that
Plaintiffs continued to have standing and that their claims were not moot, even though
Graveline was placed on the ballot and the 2018 election had ended. Graveline, 430 F. Supp. 3d
 No. 20-1337 Graveline et al. v. Benson et al. Page 6

297, 305–07 (E.D. Mich. 2019). Turning to the merits, the district court again applied the
Anderson-Burdick framework; however, it considered Defendants’ new evidence and found that
“none of it undermine[d] the Sixth Circuit’s conclusion” that the provisions, as applied in
combination, “severely burden[] independent candidates for statewide office and voters who
wish to support for [sic] them.” Id. at 314. After finding that strict scrutiny applied, the district
court then considered whether Defendants “set forth precise interests of compelling importance”
and showed that “the regulations are necessary and narrowly tailored to advance those interests.”
Id. at 315. The district court again found that Defendants had not met their burden. Id. at 317.

In granting summary judgment in favor of Plaintiffs, the district court permanently
enjoined Michigan from enforcing Mich. Comp. Laws §§ 168.590c(2), 168.544f, and
168.590b(4) in combination against independent candidates running for statewide offices. Id. at
318. Because the 2020 election year was swiftly approaching, the district court also instated an
interim measure allowing any independent candidate for statewide office to qualify for the ballot
by submitting a qualifying petition with at least 12,000 valid signatures. Id. However, the
statutory scheme’s filing deadline still applies, as does its geographic-distribution requirement.
Id. The court’s order states that the interim measure will expire when the Michigan legislature
enacts “a permanent measure replacing the invalidated scheme.” Id. Defendants filed a motion
to amend the court’s findings, asking the district court to clarify that the new signature
requirement was not a ceiling for any future legislative enactment. R. 45 (Defs.’ Mot. To Alter
Or Amend The Court’s Findings at 3) (Page ID #1035). Defendants also asked the district court
to establish a new maximum number of signatures that candidates may submit in conjunction
with the new minimum. Id. The district court denied both requests. Defendants timely appealed
the grant of summary judgment to Plaintiffs and the court’s denial of their motion for summary
judgment and motion to amend. R. 57 (Notice of Appeal) (Page ID #1383).

II. ANALYSIS

A. Jurisdiction

Defendants challenge the district court’s conclusion that Plaintiffs have standing and that
their claims are not moot. We review de novo a district court’s decisions regarding standing and
 No. 20-1337 Graveline et al. v. Benson et al. Page 7

mootness. Cleveland Branch, N.A.A.C.P. v. City of Parma, 263 F.3d 513, 523, 530 (6th Cir.
2001) (“Cleveland N.A.A.C.P.”), cert. denied, 535 U.S. 971 (2002).

1. Standing

Defendants argue that the district court erred in finding that Plaintiffs had “standing to
maintain their claims for declaratory relief” because after “the district court’s granting of
preliminary injunctive relief placing Graveline on the November 2018 ballot,” Plaintiffs no
longer have an actual or imminent injury. Appellants’ Br. at 34–35. Defendants are wrong.
Supreme Court and Sixth Circuit caselaw have consistently held that “the court must determine
whether standing exists at the time of the filing of the complaint only.” Cleveland N.A.A.C.P.,
263 F.3d at 524–26 (collecting cases). Defendants concede that Plaintiffs “arguably had
standing as of the initial filing of this case.” Appellants’ Br. at 34–35. We hold that they did. A
quick review of the facts supports our conclusion. To establish standing, “a plaintiff must show
(1) it has suffered an ‘injury in fact’ that is (a) concrete and particularized and (b) actual or
imminent, not conjectural or hypothetical; (2) the injury is fairly traceable to the challenged
action of the defendant; and (3) it is likely, as opposed to merely speculative, that the injury will
be redressed by a favorable decision.” Friends of the Earth, Inc. v. Laidlaw Env’t Servs., 528
U.S. 167, 180–81 (2000). At the time they filed the complaint, Plaintiffs submitted evidence to
show that Michigan’s statutory scheme operated to deny Graveline access to the ballot and
restricted Voter-Plaintiffs’ opportunity to vote for the candidate of their choice. R. 1-3
(Graveline Decl. ¶¶ 10–18) (Page ID #33–37) (describing the difficulties that Graveline faced in
securing the required signatures, especially in light of the early filing deadline). Plaintiffs plainly
allege a concrete actual injury that they traced back to Michigan’s ballot-access laws for
independent candidates. Furthermore, their injuries are redressable by a favorable decision
holding the laws, in combination, unconstitutional as applied and allowing Graveline to be
placed on the ballot. Consequently, we conclude that Plaintiffs had standing at the time they
filed their complaint.

Defendants’ attempt to argue that Plaintiffs must maintain standing throughout all stages
of the litigation rests on a misreading of Supreme Court precedent. Defendants cite
Hollingsworth v. Perry for the proposition that “Article III demands that an ‘actual controversy’
 No. 20-1337 Graveline et al. v. Benson et al. Page 8

persist throughout all stages of litigation.” 570 U.S. 693, 705 (2013) (quoting Already, LLC v.
Nike, Inc., 568 U.S. 85, 90–91 (2013)) (discussing mootness and not standing); Appellants’ Br.
at 34. In the very next sentence, the Supreme Court explains that its statement “means that
standing ‘must be met by persons seeking appellate review, just as it must be met by persons
appearing in courts of first instance.’” Hollingsworth, 570 U.S. at 705 (quoting Arizonans for
Official English v. Arizona, 520 U.S. 43, 64 (1997)). Far from requiring plaintiffs to maintain
standing throughout all stages of litigation, this narrow subset of standing caselaw requires
intervenors seeking “to defend on appeal in the place of an original defendant” to show that they
too “possess ‘a direct stake in the outcome.’” Arizonans for Official English, 520 U.S. at 64–65
(quoting Diamond v. Charles, 476 U.S. 54, 62 (1986)). Consequently, Hollingsworth does not
overturn long-established precedent holding that courts “must determine whether [a plaintiff’s]
standing exists at the time of the filing of the complaint only.” Cleveland N.A.A.C.P., 263 F.3d
at 526. In sum, Plaintiffs have standing to bring this lawsuit.

2. Mootness

Defendants also argue that the district court erred in finding that Plaintiffs’ claims were
not moot after Graveline’s placement on the 2018 ballot and the completion of the November
2018 general election. Appellants’ Br. at 26–28. Unlike standing, which a plaintiff does not
have to maintain throughout the entire litigation, a case may become moot at any stage of the
litigation. Libertarian Party of Ohio v. Blackwell, 462 F.3d 579, 584 (6th Cir. 2006). “[A] case
is moot when the issues presented are no longer ‘live’ or the parties lack a legally cognizable
interest in the outcome.” Powell v. McCormack, 395 U.S. 486, 496 (1969). However, an
“established exception to mootness” exists “for disputes capable of repetition, yet evading
review.” Fed. Election Comm’n v. Wis. Right to Life, Inc., 551 U.S. 449, 462 (2007). “The
exception applies where ‘(1) the challenged action is in its duration too short to be fully litigated
prior to cessation or expiration, and (2) there is a reasonable expectation that the same
complaining party will be subject to the same action again.’” Id. (quoting Spencer v. Kemna,
523 U.S. 1, 17 (1998)). Plaintiffs, as the party asserting that the exception applies, bear the
burden of establishing both prongs. Lawrence v. Blackwell, 430 F.3d 368, 371 (6th Cir. 2005),
cert. denied, 547 U.S. 1178 (2006).
 No. 20-1337 Graveline et al. v. Benson et al. Page 9

The district court properly determined that this case satisfies the first prong. “Challenges
to election laws are one of the quintessential categories of cases which usually fit this prong
because litigation has only a few months before the remedy sought is rendered impossible by the
occurrence of the relevant election.” Id. The facts of the instant case exemplify this problem;
less than four months elapsed between the filing of the complaint and the occurrence of the
election. Because the petition deadline falls so close to voting day, any future challenges will
face the same problem.

Furthermore, the district court also properly determined that Plaintiffs’ claims satisfied
the second prong. In Lawrence, a similar ballot-access case, we held that “the controversy [was]
capable of repetition” because “[t]he law at issue [was] still valid and applicable to both [the
candidate-plaintiff] and any independent candidate [the voter-plaintiff] might wish to vote for in
future election years.” Id. In so concluding, we held that the candidate-plaintiff need not state
that he planned to run again nor did the voter-plaintiff need to state explicitly that she “will wish
to vote for an independent candidate who did not decide to run until after the early filing
deadline passed.” Id. In the case before us, not only are the laws at issue still valid and
applicable to Graveline (and any other independent candidate who plans to run for statewide
office), but also Plaintiffs declared that they wish to vote for independent candidates for
statewide offices in future elections. R. 30-5 (Johnson Decl. ¶¶ 4–5) (Page ID #551); R. 30-6
(Leibson Decl. ¶¶ 4–5) (Page ID #554). Clearly, the controversy is capable of repetition.
Moreover, we have held that “[e]ven if the court could not reasonably expect that the controversy
would recur with respect to [the same parties], the fact that the controversy almost invariably will
recur with respect to some future potential candidate or voter in [the state] is sufficient to meet
the second prong because it is somewhat relaxed in election cases.” Lawrence, 430 F.3d at 372.
Because Plaintiffs allege that their injury directly resulted from Michigan’s current statutory
scheme, any future independent candidate attempting to run for statewide office in Michigan will
suffer the same harm. Plaintiffs have set forth evidence to establish that their challenge is
capable of repetition yet evading review. Thus, the district court properly exercised jurisdiction,
and so do we.
 No. 20-1337 Graveline et al. v. Benson et al. Page 10

Defendants’ argument that Lawrence is no longer good law after Wisconsin Right to Life
is meritless. A careful reading of Wisconsin Right to Life reveals that the Supreme Court did not
adopt a strict “same plaintiff same facts” requirement as Defendants suggest. See Appellants’
Br. at 30–34. The Court even expressly stated that the argument Defendants here put forth—
requiring that any future candidate’s run must have all the same characteristics of Graveline’s
candidacy— “asks for too much.” Wis. Right to Life, Inc., 551 U.S. at 463 (“Requiring repetition
of every ‘legally relevant’ characteristic of an as-applied challenge—down to the last detail—
would effectively [make the capable-of-repetition exception] unavailable for virtually all as-
applied challenges.”). The Court specifically referenced Storer v. Brown’s holding that even for
as-applied challenges, cases are not moot when the issues presented and their effects on
independent candidacies will persist because the state will apply the challenged statutes in future
elections. Id. (citing Storer v. Brown, 415 U.S. 724, 737 n.8 (1974)). Additionally, our Circuit
has continued to apply the “somewhat relaxed” repetition standard in election cases after
Wisconsin Right to Life. See, e.g., Libertarian Party of Ohio v. Husted, 831 F.3d 382, 394 (6th
Cir. 2016) (holding that an election law challenge was “capable of recurring, particularly given
the ‘“somewhat relaxed” repetition standard’ that our Circuit recognizes in election cases”
(quoting Libertarian Party of Ohio, 462 F.3d at 585)), cert. denied, 137 S. Ct. 651 (2017);
Libertarian Party of Mich. v. Johnson, 714 F.3d 929, 932 (6th Cir.) (same), cert. denied, 571
U.S. 1110 (2013). We therefore turn to the merits.

B. Constitutionality of Michigan’s Statutory Scheme

We review the district court’s grant of summary judgment de novo. Libertarian Party of
Ohio, 462 F.3d at 583. Summary judgment is appropriate only “if the movant shows that there is
no genuine dispute as to any material fact and the movant is entitled to judgment as a matter of
law.” Fed. R. Civ. P. 56(a). “Credibility determinations, the weighing of the evidence and the
drawing of legitimate inferences from the facts are jury functions, not those of a judge . . . .”
Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 255 (1986).

When deciding whether state election laws violate a plaintiff’s associational rights and
the right to vote effectively under the First and Fourteenth Amendments, we apply the
framework set forth by the Supreme Court in Anderson v. Celebrezze, 460 U.S. 780 (1983), and
 No. 20-1337 Graveline et al. v. Benson et al. Page 11

Burdick v. Takushi, 504 U.S. 428 (1992). The Anderson-Burdick framework requires us to
weigh the “character and magnitude of the asserted injury” to plaintiffs’ constitutional rights
against the “precise interests put forward by the State as justifications for the burden imposed by
its rule.” Anderson, 460 U.S. at 789. The appropriate balance is determined by the magnitude of
the burden. If the burden is severe, the regulation will be upheld only if it is “narrowly drawn to
advance a state interest of compelling importance.” Burdick, 504 U.S. at 434 (quoting Norman
v. Reed, 502 U.S. 279, 289 (1992)). If the regulations are minimally burdensome, the state’s
regulatory interests will likely justify “reasonable, nondiscriminatory restrictions.” Anderson,
460 U.S. at 788. Many regulations fall somewhere between these two extremes. When a
regulation imposes an intermediate burden, “courts engage in a flexible analysis, weighing the
burden on the plaintiffs against the state’s asserted interest and chosen means of pursuing it.”
Green Party of Tenn. v. Hargett, 767 F.3d 533, 546 (6th Cir. 2014).

In denying Defendants’ motion to stay the district court’s preliminary injunction, we
conducted a preliminary assessment of the magnitude of the burden imposed by Michigan’s
statutory scheme and held that it was severe. Graveline, 747 F. App’x at 413–14. On summary
judgment, the district court applied the Anderson-Burdick framework, assessed Defendants’
newly submitted evidence, and determined that the “combined effect of Michigan’s ballot access
regulations severely burdens independent candidates for statewide office and voters who wish to
support [] them.” Graveline, 430 F. Supp. 3d at 314. The district court also found that
Defendants had not put forth sufficient evidence to show that the provisions are narrowly tailored
to protect the interests Defendants put forth as justifications. Id. at 316–17. Defendants
challenge both determinations.

1. Character and Magnitude of Plaintiffs’ Asserted Injury

We begin our analysis by assessing the character and magnitude of the asserted injury
that Plaintiffs allege. Plaintiffs allege that Michigan’s statutory scheme burdens two of our most
fundamental rights—“the right of individuals to associate for the advancement of political
beliefs, and the right of qualified voters, regardless of their political persuasion, to cast their
votes effectively.” Williams v. Rhodes, 393 U.S. 23, 30 (1968). The Voter-Plaintiffs’ right to
cast an effective vote “is of the most fundamental significance under our constitutional
 No. 20-1337 Graveline et al. v. Benson et al. Page 12

structure.” Burdick, 504 U.S. at 433. “The rights of political association and free speech occupy
a similarly hallowed place in the constitutional pantheon.” Libertarian Party of Ohio, 462 F.3d
at 585. “[W]hen a candidate wishes to appear as one party’s standard-bearer and voters want to
exercise their constitutional right to cast a ballot for this candidate, the [Supreme] Court has
viewed state-imposed restrictions on this fundamental process with great skepticism.” Id. at 588.

The Supreme Court also has recognized the importance of protecting independent
candidates’ access to the ballot. In Anderson, the Court warned that when a state’s regulations
limit “the opportunities of independent-minded voters to associate in the electoral arena to
enhance their political effectiveness as a group, such restrictions threaten to reduce diversity and
competition in the marketplace of ideas.” 460 U.S. at 794. However, “[t]he existence of such
barriers does not of itself compel close scrutiny.” Bullock v. Carter, 405 U.S. 134, 143 (1972).
Courts must still “examine in a realistic light the extent and nature of their impact on voters.” Id.
Accordingly, we next assess the magnitude of the burden imposed by Michigan’s statutory
scheme.

The Supreme Court instructs us to distinguish between burdens that restrict political
participation equally and burdens that “fall[] unequally on new or small political parties or on
independent candidates.” Anderson, 460 U.S. at 793–94 (emphasis added) (noting that such a
discriminatory burden “impinges, by its very nature, on associational choices protected by the
First Amendment”). When a burden weighs more heavily on “an identifiable political group
whose members share a particular viewpoint, associational preference, or economic status,” we
must “focus on the degree to which the challenged restrictions operate as a mechanism to
exclude certain classes of candidates from the electoral process.” Id. at 793 (quoting Clements v.
Fashing, 457 U.S. 957, 964 (1982)). “The inquiry is whether the challenged restriction unfairly
or unnecessarily burdens ‘the availability of political opportunity.’” Id. (quoting Clements,
457 U.S. at 964). Plaintiffs contend that Michigan’s statutory scheme falls within the Supreme
Court’s disfavored category of restrictions. Specifically, they argue that Michigan’s filing
deadline for independent candidates imposes an unequal burden by requiring independent
candidates to conduct petition drives before major parties select their nominees and well in
advance of the general election.
 No. 20-1337 Graveline et al. v. Benson et al. Page 13

Furthermore, Plaintiffs argue that this unequal burden becomes heightened by the
combination of (1) the amount of signatures Michigan requires—30,000, and (2) the geographic
spread the 30,000 signatures must represent—at least 100 registered voters in each of at least half
of Michigan’s fourteen congressional districts. When plaintiffs allege that a statutory scheme, in
combination, imposes a burden on their rights, we must consider “the combined effect of the
applicable election regulations,” and not measure the effect of each statute in isolation.
Libertarian Party of Ohio, 462 F.3d at 586. Our precedent also directs us to consider several
factors, including “whether alternative means are available to exercise [the rights at issue]; the
effect of the regulations on the voters, the parties and the candidates; [and] evidence of the real
impact the restriction has on the process.” Id. at 587.

Beginning with Anderson, many courts have recognized the significant disadvantages
imposed on independent candidates and minor parties by early filing deadlines. First, early filing
deadlines serve to exclude or limit severely independent candidates “whose positions on the
issues could command widespread community support” because they often must begin their run
before the major parties’ nominees are known. Anderson, 460 U.S. at 792. An early filing
deadline restricts independent candidates’ opportunity to garner support from the most likely
subset of voters—the group that is dissatisfied with the choices within the two major parties—
because such subset rarely becomes “a cohesive or identifiable group until . . . [the] major parties
stake[] out their positions and select[] their nominees.” Id. at 791–92; see also Green Party of
Tenn., 767 F.3d at 548 (explaining that early filing deadlines weaken independent and minor-
party candidates’ ability “to position themselves as meaningful alternatives”); Libertarian Party
of Ohio, 462 F.3d at 590–91 (collecting cases). The Supreme Court also recognizes the
concurrent burden an early filing deadline places on independent-minded voters as well.
Anderson, 460 U.S. at 792 (noting that early filing deadlines “den[y] the ‘disaffected’ not only a
choice of leadership but a choice on the issues as well” (quoting Williams, 393 U.S. at 33)).

Second, early filing deadlines burden the efforts of independent candidates as they
attempt to gather petition signatures by striking at the core of their campaign efforts. Courts
have consistently documented the damage early deadlines inflict on an independent candidate’s
ability to (1) recruit and retain volunteers, (2) gain media exposure, (3) solicit campaign
 No. 20-1337 Graveline et al. v. Benson et al. Page 14

contributions, and (4) engage voters’ interest. See Anderson, 460 U.S. at 792 & n.14; see also
Libertarian Party of Ohio, 452 F.3d at 586 (noting that “[d]eadlines early in the election cycle
require minor political parties to recruit supporters at a time when the major party candidates are
not known and when the populace is not politically energized”; collecting cases). Consequently,
courts have routinely applied strict scrutiny and struck down laws that required independent
candidates (or minor parties) to file qualifying petitions substantially in advance of a primary
election or nominating convention. See, e.g., Anderson, 460 U.S. at 783 n.1 (deadline seventy-
five days before primary); New Alliance Party of Ala. v. Hand, 933 F.2d 1568, 1570 n.3 (11th
Cir. 1991) (deadline sixty days before primary); Cromer v. South Carolina, 917 F.2d 819, 822
(4th Cir. 1990) (deadline seventy days before primary); McLain v. Meier, 637 F.2d 1159, 1164
(8th Cir. 1980) (deadline ninety days before primary); Cripps v. Seneca Cnty. Bd. of Elections,
629 F. Supp. 1335, 1338 (N.D. Ohio 1985) (deadline seventy-five days before primary);
Stoddard v. Quinn, 593 F. Supp. 300, 306 (D. Maine 1984) (April 1 deadline, primary held on
second Tuesday in June); Bradley v. Mandel, 449 F. Supp. 983, 985 (D. Md. 1978) (deadline
seventy days before primary).

Our circuit is no different. For example, in Libertarian Party of Ohio, we held that an
Ohio law requiring minor parties to file a petition 120 days before the primary imposed a severe
burden. 462 F.3d at 590. In so holding, we emphasized the existence and effects of the same
disadvantages—burdens on signature-gathering efforts and an inability to respond to issues
raised later in the election year.2 Id. at 586–91. In keeping with the Supreme Court’s

2
The dissent attempts to distinguish Libertarian Party of Ohio from the instant case because it concerns a
minor party’s ability to access the ballot and not independent candidates. Dissent at 36. However, the Supreme
Court has recognized the importance of protecting independent candidates’ access to the ballot as well. As the Court
explained in Anderson:
A burden that falls unequally on new or small political parties or on independent
candidates impinges, by its very nature, on associational choices protected by the First
Amendment. It discriminates against those candidates and—of particular importance—against
those voters whose political preferences lie outside the existing political parties. By limiting the
opportunities of independent-minded voters to associate in the electoral arena to enhance their
political effectiveness as a group, such restrictions threaten to reduce diversity and competition in
the marketplace of ideas. Historically political figures outside the two major parties have been
fertile sources of new ideas and new programs; many of their challenges to the status quo have in
time made their way into the political mainstream. In short, the primary values protected by the
First Amendment—“a profound national commitment to the principle that debate on public issues
 No. 20-1337 Graveline et al. v. Benson et al. Page 15

instructions, we noted “the great weight of authority that has distinguished between filing
deadlines well in advance of the primary and general elections and deadlines falling closer to the
dates of those elections.” Id. at 590. Accordingly, we distinguished the case from Lawrence,
where we upheld an Ohio law requiring independent congressional candidates to obtain
signatures amounting to one percent of electors in the relevant congressional district. Lawrence,
430 F.3d at 370. There, we found the burden to be neither severe nor inherently unreasonable, in
part because “all candidates seeking place on the ballot in November must engage in substantial
campaign work before the early primary.” Id. at 373. In Lawrence, we expressly emphasized
that the filing deadline for independent candidates fell one day before the primary election. Id.
(noting that the one-day separation between deadlines created “a vital distinction” from the
“[t]he early deadline discussed in Anderson,” which “imposed such a significant burden because
it put independent candidates at a disadvantage vis-à-vis the major parties’ nominees who were
not named until nearly five months later”). We explained how the similar early deadlines for all
candidates ensured that “there is no particular group which feels the additional burden of being
placed at a disadvantage with respect to the rest of the field.” Id. Thus, unlike in Libertarian
Party of Ohio, we concluded in Lawrence that “the burden imposed by Ohio’s early deadline is
nondiscriminatory.” Id.; see also Wood v. Meadows, 207 F.3d 708, 713–14 (4th Cir. 2000)
(finding no severe burden when the filing deadline for independent candidates fell on the same
day as the primary); Council of Alt. Pol. Parties v. Hooks, 179 F.3d 64 (3rd Cir. 1999) (holding
no severe burden when the filing deadline for independent candidates fell a day before the
primary).

Michigan’s statutory scheme exhibits disparate treatment of independent candidates for
statewide office in a manner that is comparable to other early filing deadline cases. The
nominating conventions for major parties were required to be held by September 7; Graveline
was required to submit his petition on July 19, a 50-days-in-advance deadline. Crucially, unlike
in Lawrence, major party candidates and independent candidates do not shoulder a similar

should be uninhibited, robust, and wide-open,” New York Times Co. v. Sullivan, 376 U.S. 254, 270
(1964)—are served when election campaigns are not monopolized by the existing political parties.
Anderson, 460 U.S. at 793–94 (footnote and internal citations omitted) (emphasis added). In short, the effect of
Michigan’s statutory scheme on independent candidates and like-minded voters implicates the same institutional and
associational concerns as the ones raised in Libertarian Party of Ohio.
 No. 20-1337 Graveline et al. v. Benson et al. Page 16

burden. Here, the major party candidates are not required to obtain advance support during the
same time that an independent candidate must canvas for signatures. And the burden shouldered
by independent candidates is not reasonable because their filing deadline “[is] substantially in
advance” of the majority parties’ nomination conventions. Lawrence, 430 F.3d at 373–74 & n.2.
Unlike our concerns in Lawrence, here it is the major parties that have a “flexibility” that
independent candidates do not. Id. at 374; see also Anderson, 460 U.S. at 790–91 (noting that
early filing deadlines for independent candidates set in advance of major party nominations grant
“[c]andidates and supporters within the major parties . . . the political advantage of continued
flexibility” and impose “a correlative disadvantage” on independent candidates and their
supporters).

We acknowledge that, by late April, the Democratic Party had informally endorsed (but
not formally nominated) its candidate; however, the Republican Party nominee was still
unknown. R. 1-3 (Graveline Decl. ¶¶ 3–4) (Page ID #30–31). Thus, the early filing deadline
burdened Graveline by shortening the time he had to campaign effectively because voters had yet
to make up their minds about their satisfaction with the potential major party nominees. As
several courts have determined, attempts to begin collecting signatures before this dissatisfied
group coalesces often will be burdensome. Far from being a “general statement[] of the law,”
Dissent at 37, the Supreme Court, our court, and other courts have consistently recognized the
burdens imposed by early filing deadlines, with or without explicit factfinding in the record
before them, by relying on evidence of obstacles documented in prior cases. See, e.g., Anderson
460 U.S. at 791–92 & n.14 (citing a Maryland district court’s findings of fact as support for the
Court’s conclusion that an Ohio early filing deadline imposed a severe burden on “an
independent candidate’s organizing efforts”); see also Libertarian Party of Ohio, 462 F.3d at 586
(citing cases assessing other states’ filing deadlines to support its conclusion that an Ohio early
filing deadline negatively impacted a minor party’s ability “to recruit supporters” and to respond
to “contentious issue[s] raised” later in the election year); McLain, 637 F.2d at 1164 (same for
North Dakota); Council of Alt. Pol. Parties v. Hooks, 121 F.3d 876, 880 (3rd Cir. 1999) (same
for New Jersey). And, even if Graveline and the public were “at least aware of who the [major-
party] contenders were” before the filing deadline, Appellants’ Br. at 50, future elections might
have different circumstances, and the concerns highlighted in Anderson “remain significant” as
 No. 20-1337 Graveline et al. v. Benson et al. Page 17

applied to Graveline or any future candidate. Nader v. Brewer, 531 F.3d 1028, 1038 (9th Cir.
2008).

All told, Michigan’s system works to disadvantage independent candidates alone by
requiring them to seek a significant number of signatures from an electorate that is not yet
politically energized and to stake out positions in a race with yet undecided contours. Thus, we
break no new ground in concluding that the combination of the 30,000-signature requirement,
the geographic-distribution requirement, and the filing deadline imposes a severe burden on
independent candidates.

The evidence in the record exposes the severe burden that Michigan’s statutory scheme
and its inherent disadvantages impose on independent candidates’ access to the ballot. First,
Michigan’s history reveals the severity of this burden. The current statutory scheme is the only
way independent candidates can qualify for the ballot. No other alternative means exist. Since
its implementation in 1988, no independent candidate for statewide office has managed to
complete a qualifying petition.3 R. 1-2 (Winger Decl. ¶ 5) (Page ID #21). However, since 1997,
at least thirty candidates have formed the required committees to begin collecting signatures to
qualify for the ballot as an independent candidate for statewide office.4 Graveline, 430 F. Supp.
3d at 302; see R. 34-2 (Results For Mich. Att’y Gen. Comms.) (Page ID #653); R. 34-3 (Results
For Mich. Sec’y of State Comms.) (Page ID #659); R. 40-1 (Results For Mich. Governor
Comms.) (Page ID #786). Even though several candidates formed their committees after the
filing deadline, many more formed their committees before the deadline, and some well in
advance. See, e.g., R. 40-1 (Results For Mich. Governor Comms. at 21, 23, 42, 65) (Page ID
#805, 807, 826, 849). The dissent points out that these committees were formed for
gubernatorial candidates whose major party counterparts are not nominated at a convention but
instead are selected via the traditional primary process, which takes place much closer to the

3
In 1999, Michigan amended its signature requirement for independent candidates for statewide office by
changing it from one percent of the total vote for governor in the preceding election to the current requirement of
30,000 signatures. Neither party contends that this change was material.
4
The district court noted that these records “are incomplete in material ways.” Graveline, 430 F. Supp. 3d
at 302. For example, the records do not go back to 1988, which is when Michigan implemented the statutory
scheme. They also do not include the number of candidates who attempted to run for United States Senate in
Michigan.
 No. 20-1337 Graveline et al. v. Benson et al. Page 18

independent party filing deadline. See Dissent at 31. This difference does somewhat diminish
the relevance of the evidence, but Jenness v. Fortson counsels us not to forget basic principles.
403 U.S. 431 (1971). In Jenness, the plaintiffs challenged a statutory scheme with a filing
deadline for independent candidates that matched the filing deadline for parties participating in
the primary election. Id. at 433–34. But there was one important finding that distinguishes
Jenness from this case: record evidence showed that independent candidates had previously
qualified for the ballot. Id. at 439. This evidence allowed the Supreme Court to conclude that
Georgia “in no way freezes the status quo” and thus “[t]he open quality of the Georgia system
[was] far from merely theoretical.” Id. Whether hindered or unhindered by an early filing
deadline, independent candidates in Michigan have not appeared on the ballot. That is a reality
we cannot discount.5

The signature requirement plays an important role in explaining the absolute dearth of
independent candidates appearing on the ballot. When considered as a percentage of the state’s
population, there are fifteen states with higher requirements. R. 28-8 (Kelly Report at 4) (Page
ID #402). But, in terms of absolute numbers only five states have higher requirements. R. 28-5
(Winger Report at 3) (Page ID #375). The fact that no independent candidate for statewide
office has appeared on the Michigan ballot in thirty years indicates that a reasonably diligent
candidate could not meet the signature requirements within the signature-collection window. See
Storer, 415 U.S. at 742 (stating that “[p]ast experience will be a helpful, if not always an
unerring, guide: it will be one thing if independent candidates have qualified with some
regularity and quite a different matter if they have not”).

Second, Defendants’ petition expert, Lee Albright, established in his report that an
independent candidate’s compliance with Michigan’s statutory scheme requires substantial costs
of human and financial resources to qualify a petition-candidate for the ballot. R. 28-7 (Albright

5
The dissent raises another concern about the significance of this evidence; it argues that the formation of
campaign committees “says absolutely nothing about whether the statutory scheme inhibited [the candidates] from
satisfying the scheme.” Dissent at 32. We struggle to imagine how an absolute dearth of independent candidates
appearing on the ballot, even though several have received either donations or volunteer efforts to launch their
campaign, says nothing about whether the statutory scheme in which these campaigns operated inhibited their
efforts to qualify. But putting that aside, as discussed below, Plaintiffs highlighted evidence in the record that
substantiates what the absence of independent candidates already signals—a severe burden imposed by the statutory
scheme as a whole.
 No. 20-1337 Graveline et al. v. Benson et al. Page 19

Report). Albright explained that (1) successful all-volunteer efforts to qualify a petition-
candidate for the ballot are unusual; (2) most all-volunteer efforts fail and “professionals are used
to augment signature gathering efforts”; (3) had Graveline used the entire 180 days “his costs per
signature might have been lower”; and (4) mounting an all-volunteer effort requires significant
labor, planning, and costs. Id. at 1–3 (Page ID #390–92). Albright also clarified that the actual
number of signatures an independent candidate must gather is much higher than 30,000.
“Assuming a 75% validity rate, 40,000 signatures must be gathered.” Id. at 1 (Page ID #390).
However, a 75% validity rate is the industry standard. For volunteers, that rate can be “difficult
to maintain,” and requires a laborious verification process. Id. Albright’s report throws into
sharp relief the difficulties that independent candidates face in attempting to comply with
Michigan’s requirements. No matter when a candidate starts, efforts to comply with the
requirements will be arduous, time-consuming, and costly, even if professional signature-
gatherers aid in the attempt. Id. at 1–3 (Page ID #390–92).

Albright also raised several factors that impact the efficacy of volunteer efforts to qualify
a petition, even if professional signature-gatherers supplement their efforts. Id. Notably, many
of these factors weigh against an independent candidate’s success. For example, Albright stated
that a campaign would need “a significant group of volunteers ready to go on day one,”
consistent availability of good venues, and the financial ability to deal with any increases in
price-per-signature as efforts become more demanding. Id. at 1–2 (Page ID #390–91). For a
candidate to rally these resources without the backing of large organizations or an established
political machine is inherently difficult, as evidenced by Graveline’s petition-gathering efforts.
See, e.g., R. 1-8 (Byrd Decl. ¶ 6) (Page ID #54) (describing how public officials hampered
volunteers’ petition-gathering efforts at a prime venue). The early filing deadline for
independent candidates also interacts negatively these factors. As detailed earlier, it will be
difficult to recruit and retain a significant volunteer force and to gather financial resources in the
early months of the year long before major-party nominees are identified and the general election
is imminent. Anderson, 460 U.S. at 792.

In at least one other case, we have considered whether the incidental costs of
campaigning by themselves can “constitute exclusion or virtual exclusion from the ballot.”
 No. 20-1337 Graveline et al. v. Benson et al. Page 20

Libertarian Party of Ky. v. Grimes, 835 F.3d 570, 575 (6th Cir. 2016), cert. denied, 137 S. Ct.
2119 (2017). In Grimes, minor parties challenged Kentucky’s ballot-access scheme that required
them to obtain a certain number of signatures in order for their candidates to be placed on the
ballot. Id. at 572–73. The plaintiffs argued that the “hypothetical maximum cost of petitioning”
to “field a full slate of candidates for every state and county office during an actual four-year
term” would exclude the party from the ballot and thus imposed a severe burden on ballot access
for the party. Id. at 576–77. We held that such “incidental costs of gathering signatures on
petitions do not come close to exclusion from the ballot,” in part because the plaintiffs did not
demonstrate “any likelihood” that they “would actually field a full slate of candidates.” Id. at
577. However, Plaintiffs here do not argue that independent candidates’ exclusion rests on costs
alone, and the circumstances of Grimes are distinguishable in three other significant ways from
the case at hand. First, in Grimes, we dealt with the “high costs of gathering and filing petitions”
as applied to associations attempting to field a full slate of candidates for each office and “not as
applied to any individual candidate for a specific office,” which is the issue we face here. Id. at
573. Second, in Grimes, despite the plaintiffs’ arguments that the regulations’ requirements were
impossible to meet for third parties, third parties had qualified under the statutory scheme several
times in the past. Id. at 576. Not so here. Finally, the third parties in Grimes had a reasonable
alternative means of ballot access besides the challenged requirements. Id. at 575–76. No one
disputes that in Michigan, the challenged statutory scheme is the only means by which
independent candidates can access the ballot. The circumstances that supported a finding of an
intermediate burden in Grimes do not exist in the instant case.

Clearly, the Michigan statutory scheme places “a particular burden on an identifiable
segment of [Michigan’s] independent-minded voters.” Anderson, 460 U.S. at 792. The record
reveals that the challenged requirements inhibit independent candidates for statewide office from
recruiting supporters, disrupting the status quo, ensuring that existing political parties do not
monopolize elections, and engaging in one of our democracy’s most honored and essential
political activities—representing like-minded individuals in an election for the right to govern.
Consequently, we hold that Michigan’s statutory scheme for qualifying independent candidates
for the ballot imposes a severe burden on Plaintiffs’ associational and voting rights.
 No. 20-1337 Graveline et al. v. Benson et al. Page 21

Our recent decision in Kishore v. Whitmer does not counsel us to hold otherwise.
972 F.3d 745 (6th Cir. 2020). In Kishore, we reviewed an appeal from two independent
candidates for president and vice president seeking placement on the ballot in Michigan. They
challenged the same statutory scheme at issue in this case, arguing that, as applied, the statutory
scheme, in combination with Michigan’s orders restricting in-person gatherings during the
COVID-19 pandemic, was unconstitutionally burdensome under the First and Fourteenth
Amendments. Id. at 749. Applying the Anderson-Burdick framework, we held that the ballot-
access restrictions placed only an intermediate burden on the plaintiffs when enforced alongside
Michigan’s stay-at-home orders. Id. at 750. However, we began our opinion in Kishore by
reminding the reader that the district court in Graveline imposed an “interim measure” reducing
the “signature requirements for independent candidates” to 12,000 signatures and that the lower
requirement “is still in effect.” Id. at 748 & n.1. Accordingly, the burden we analyzed in
Kishore is in no way similar to the 30,000-required signatures and other requirements under
review in this case. The original 30,000-signature requirement is more than double the 12,000-
signature requirement applicable in Kishore, engenders different considerations, with or without
a pandemic, and necessitates a different conclusion.

That impactful fact also distinguishes Kishore from this case in another important aspect.
The interim 12,000-signature requirement foreclosed using the historical record at issue in this
case, i.e., the complete absence of any independent candidate ever qualifying for the ballot, to
demonstrate that the statutory scheme and stay-at-home orders, in combination, imposed a severe
burden on the rights of independent candidates and voters who might wish to vote for them. See,
e.g., Green Party of Tenn., 767 F.3d at 547–48 (finding a historical record that showed a
consistent absence of minor parties from Tennessee’s ballots less relevant after an amendment to
the challenged statutory scheme). Instead, the plaintiffs in Kishore could put forward evidence
of only their own struggles in gathering signatures under the statutory scheme and COVID
restrictions. And we found it relevant to note that during the period before the stay-at-home
orders went into effect, despite their campaign activities, the plaintiffs did not obtain “a single
signature on their qualifying petition.” Kishore, 972 F.3d at 749. This reality heavily influenced
our determination that the stay-at-home orders did not play a substantial role in burdening the
 No. 20-1337 Graveline et al. v. Benson et al. Page 22

Kishore plaintiffs’ ability to get onto the ballot. In the instant case, Plaintiffs challenge a
completely separate issue, and we assess considerably more burdensome statutory requirements.

We find unpersuasive Defendants’ arguments that the burdens imposed by the provisions
are not severe. Defendants begin by analyzing the effect of each of the challenged provisions
individually instead of assessing their combined effect as applied. Appellants’ Br. at 40. This
technique fails to heed our precedent, which instructs Defendants that the “inquiry is not whether
each law individually creates an impermissible burden but rather whether the combined effect of
the applicable election regulations creates an unconstitutional burden.” Libertarian Party of
Ohio, 462 F.3d at 586. Defendants also cite Lawrence, Storer, and De La Fuente v. California,
278 F. Supp. 3d 1146, 1155 (C.D. Cal. 2017), as cases where courts have held that similar
requirements, in particular higher signature requirements, were not a severe burden. Appellants’
Br. at 43–44. However, each case is easily distinguishable. In Lawrence, we did not find that
the requirements severely burdened independent candidates in large part due to the filing
deadline falling on the day before the primary for major parties, but here the deadline is at least
fifty days earlier. Lawrence, 430 F.3d at 373–74. In Storer, the Supreme Court remanded the
case because there was not enough evidence in the record to determine whether the requirements
severely burdened the plaintiffs, in part because no evidence existed showing whether
independent candidates regularly qualified for the ballot. Storer, 414 U.S. at 740–43. That
evidence is in the record for this case: independent candidates have never qualified for the
Michigan ballot. This also contrasts with De La Fuente, where the evidence showed that
independent candidates had qualified for the ballot in the past. De La Fuente, 278 F. Supp. 3d at
1154.

Defendants further argue that Graveline’s “dilatoriness” and unreasonable delay under
the circumstances “placed him in a race against the clock—not the statutes,” and thus, “there is
no way to assess the burden imposed by the signature requirement.” Appellants’ Br. at 52.
Defendants’ focus on Graveline and his unique circumstances is misplaced. The question is
“could a reasonably diligent independent candidate be expected to satisfy the signature
requirements, or will it be only rarely that the unaffiliated candidate will succeed in getting on
the ballot?” Storer, 415 U.S. at 742. The evidence shows that an unaffiliated candidate has
 No. 20-1337 Graveline et al. v. Benson et al. Page 23

never been on the Michigan ballot, even though Defendants’ records show that several
candidates formed their campaign committees well in advance of the filing deadline. According
to Albright’s report, “gathering 30,000 valid signatures . . . within the 180 days allowed can most
certainly be done.” R. 28-7 (Albright Report at 3) (Page ID #392). But he does not identify a
single successful petition drive. “Moreover, the fact that an election procedure can be met does
not mean the burden imposed is not severe.” Libertarian Party of Ohio, 462 F.3d at 592
(emphasis added). As Albright’s report detailed, efforts to comply with the requirements,
whether accomplished solely by volunteers or also by paying professional signature-gatherers,
will exact a high cost in human and financial resources. R. 28-7 (Albright Report at 1–3) (Page
ID #390–92). The early filing deadline reinforces the burdens imposed by these difficulties.
Anderson, 460 U.S. at 792.

In sum, by leveraging historical records, showing a lack of alternative means to qualify,
and identifying specific burdens imposed by the statutory scheme, Plaintiffs have satisfied their
burden to establish that the combined effect of Michigan’s statutory scheme imposes a severe
burden on their rights. For ballot-access cases, we have noted that “[t]he hallmark of a severe
burden is exclusion or virtual exclusion from the ballot.” Grimes, 835 F.3d at 574. Defendants
have not put forth any evidence to justify disregarding this telling hallmark. Thus, we hold that
the district court correctly determined that the combined effect of Michigan’s laws imposes a
severe burden on the First Amendment rights of independent candidates and their potential voter-
supporters. Therefore, we will uphold the provisions only if they are “narrowly drawn to
advance a state interest of compelling importance.” Burdick, 504 U.S. at 434 (quoting Norman,
502 U.S. at 289).

2. Strict Scrutiny of Michigan’s Asserted Interests

Defendants assert that the challenged provisions protect several legitimate interests.
Defendants correctly point out Supreme Court and Sixth Circuit caselaw recognizing that states
have “an important interest in ensuring that candidates demonstrate a ‘significant modicum of
support’ before gaining access to the ballot, primarily to avoid voter confusion, ballot
overcrowding, and frivolous candidacies.” Grimes, 835 F.3d at 577 (quoting Jenness, 403 U.S.
at 442). Defendants argue that the geographic requirement advances these interests “by requiring
 No. 20-1337 Graveline et al. v. Benson et al. Page 24

statewide candidates to show a modicum of support from voters within roughly half the state.”
Appellants’ Br. at 40. Similarly, Defendants contend that the 30,000-signature requirement
serves the same interests “by requiring unaffiliated candidates, who lack the support and interest
of a major or even minor party, to demonstrate they have a significant modicum of public
support for their candidacy.” Id. at 42. The Supreme Court has also recognized that “as a
practical matter,” States must regulate in “substantial ways . . . the time, place, and manner of
holding primary and general elections” for both state and federal elections in order to ensure that
elections “are to be fair and honest” and orderly. Storer, 415 U.S. at 730 (citing U.S. Const. art.
I, § 4, cl. 1); see also Timmons v. Twin Cities Area New Party, 520 U.S. 351, 364 (1997) (noting
that “[s]tates certainly have an interest in protecting the integrity, fairness, and efficiency of their
ballots and election processes as means for electing public officials”). Accordingly, we have
recognized that “easing administrative burdens on boards of elections are undoubtedly
‘important regulatory interests.’” Ohio Democratic Party v. Husted, 834 F.3d 620, 635 (6th Cir.
2016). Defendants assert that the filing deadline advances these interests because “it is part of a
carefully constructed set of election deadlines,” which the Director of Elections created to ensure
that candidates are certified in time for the election and absentee voters receive their ballots in a
timely fashion. Appellants’ Br. at 45–46.

Plaintiffs do not dispute, and we agree, that the challenged provisions serve compelling
interests. However, under strict-scrutiny review, Defendants are “obligated to demonstrate that
there was no less restrictive means by which [they] could achieve [their] important interest[s].”
Lawrence, 430 F.3d at 375. The Supreme Court observed that “it is especially difficult for the
State to justify a restriction that limits political participation by an identifiable political group
whose members share a particular viewpoint[ or] associational preference.” Anderson, 460 U.S.
at 793. We keep this observation firmly in mind as we apply strict scrutiny to each of
Defendants’ asserted interests.

At the outset, we note that Defendants again address each provision separately and make
no arguments to demonstrate how the provisions as applied in combination are the least
restrictive means by which they could achieve their compelling interests. Furthermore,
Defendants fail to discuss at all how the geographic-distribution requirement is narrowly drawn.
 No. 20-1337 Graveline et al. v. Benson et al. Page 25

Instead, they dismiss it as irrelevant to the analysis because Graveline satisfied the requirement.
Appellants’ Br. at 40–41. Defendants’ flawed understanding of how to defend against a
combined-effect challenge undermines their ability to satisfy their burden.

Defendants submitted uncontested evidence from the former Director of Elections that
demonstrates how the early filing deadline is a part of a carefully constructed scheme, such that
any alterations would necessitate changes to the whole system. R. 28-2 (Williams Aff.). The
former Director details how the deadline intersects with the Secretary of State’s duty to certify
qualified candidates, initiatives, and constitutional amendments for inclusion on the ballot so
that each county’s board of elections has sufficient time to print and distribute ballots for the
general election. Id. at 2–9 (Page ID #352–59). Plaintiffs do not contest that the filing deadline
is narrowly tailored to advance Michigan’s interest in administering a timely and orderly
election.

Plaintiffs also do not argue that the geographic-distribution requirement fails to survive
strict scrutiny. And we are reluctant to disturb the district court’s finding that the geographic-
distribution requirement so survives. Requiring 100 signatures from seven out of fourteen
districts appears to be the least restrictive means of ensuring that an independent candidate has a
modicum of statewide support.

But Defendants fail to provide any evidence to demonstrate that the 30,000-signature
requirement is narrowly drawn to protect Michigan’s interests in preventing voter confusion,
ballot overcrowding, or frivolous candidates, especially as combined with the early filing
deadline. In arguing that the requirement is narrowly tailored, Defendants rely solely on Dr.
Kelly’s analysis that shows that a requirement of 5,000 signatures could lead to nine or more
candidates on the ballot, R. 28-8 (Kelly Report at 9) (Page ID #407), and Albright’s statement
that it is possible for an independent candidate to collect 30,000 signatures. R. 28-7 (Albright
Report at 3) (Page ID #392); Appellants’ Br. at 54. But as the district court explained, those
conclusions establish “nothing about the current challenged 30,000 requirement” and whether
that amount is necessary for Michigan to prevent overcrowding and frivolous candidates.
Graveline, 430 F. Supp. 3d at 316. Furthermore, Defendants never disputed Winger’s evidence
that “no state that required more than 5,000 signatures has ever had more than eight candidates
 No. 20-1337 Graveline et al. v. Benson et al. Page 26

on the general election ballot for statewide office.” R. 30-3 (Winger Rebuttal Report at 2) (Page
ID #517). Nothing in the record demonstrates that a signature requirement below 30,000 and
above 5,000 has resulted in ballot overcrowding, voter confusion, or frivolous candidates in any
state. As the Supreme Court held in Williams, “[n]o such remote danger can justify the
immediate and crippling impact on the basic constitutional rights involved in this case.”
393 U.S. at 33. Defendants fail to show how requiring independent candidates to gather 30,000
signatures is the “least drastic means to achieve their ends” of preventing ballot overcrowding or
frivolous candidates from accessing the ballot. Ill. State Bd. of Elections v. Socialist Workers
Party, 440 U.S. 173, 185 (1979) (noting that for ballot-access cases, it “is particularly important”
to require “States adopt the least drastic means to achieve their ends”). Instead, in this case, the
30,000-signature requirement, when combined with the early filing deadline and the geographic-
distribution requirement, serves only to prevent the qualification of any independent candidate
for the ballot unless the candidate can mobilize substantial resources in adverse conditions.

We recognize that moving the filing deadline closer to the date of the nominating
convention or lowering the 30,000-signature requirement might impose some additional costs on
the State. But the First and Fourteenth Amendments’ guarantees often necessitate that states
shoulder some burdens. We acknowledge that Michigan has the power to enact statutes
governing the time, place, and manner by which independent candidates for statewide office may
qualify for a general election. In isolation, each of Michigan’s requirements may impose a
reasonable burden on constitutional rights. However, the combined effect of these provisions
imposes a severe and unequal burden on the associational rights of independent candidates and
their potential voter-supporters. Because Defendants have not shown that these provisions are
narrowly tailored to protect compelling state interests, we hold that Michigan’s system for
qualifying independent candidates violates the First and Fourteenth Amendments of the
Constitution.

C. Relief Granted

Finally, Defendants challenge the district court’s interim measure allowing independent
candidates to qualify to be placed on a ballot for statewide office with a minimum of 12,000
signatures. They also challenge the court’s denial of their motion to amend its findings, which
 No. 20-1337 Graveline et al. v. Benson et al. Page 27

asked the district court to set a new maximum for the signature requirement. Appellants’ Br. at
57–63. We review the scope of a district court’s grant of injunctive relief and denials of leave to
amend under the highly deferential abuse-of-discretion standard. See Sec’y of Labor v. 3Re.com,
Inc., 317 F.3d 534, 537 (6th Cir. 2003); GenCorp, Inc. v. Am. Int’l Underwriters, 178 F.3d 804,
832 (6th Cir. 1999); Fed. R. Civ. P. 52(b). “Abuse of discretion is defined as a definite and firm
conviction that the trial court committed a clear error of judgment.” Bowling v. Pfizer, Inc.,
102 F.3d 777, 780 (6th Cir. 1996), cert. denied, 522 U.S. 906 (1997).

As a court of equity, the district court was well within its discretion in choosing to reduce
the number of signatures required for a qualifying petition until the Michigan legislature “enacts
a permanent measure replacing the invalidated scheme for independent candidates for statewide
office.” Graveline, 430 F. Supp. 3d at 318; Brown v. Bd. of Educ., 349 U.S. 294, 300 (1955)
(“Traditionally, equity has been characterized by a practical flexibility in shaping its remedies.”);
see also Ohio Democratic Party, 834 F.3d at 623 (“Federal judicial remedies, of course, are
necessary where a state law impermissibly infringes the fundamental right to vote.”). In
fashioning its remedy, the district court first noted that because an election year approached, “the
[c]ourt fe[lt] compelled to make sure a procedure is in place now for independent candidates for
statewide office to access the ballot.” Graveline, 430 F. Supp. 3d at 318. As the district court
aptly observed, “having some ballot access requirements for independent statewide candidates—
even if temporary—is better than the alternative of none.” Id. The absence of any requirements
would serve only to actualize Defendants’ fears of overcrowding and frivolous candidates. Thus,
in reducing the number of signatures required, the remedy addresses the combined effect of the
statutory scheme but still accounts for the uncontested evidence that Defendants set forth to
defend the necessity of the early filing deadline.

Defendants take exception with the selection of 12,000 signatures. However, that
number is more than double the district court’s original 5,000-signature requirement. And it
considers the scant evidence that Defendants introduced showing that it was possible that a
requirement of 5,000 signatures may lead to ballot overcrowding and frivolous candidates. Id.
Furthermore, Defendants, despite ample prompting throughout the litigation, have never argued
for or suggested a different requirement. Nor is the number as arbitrary as Defendants attempt to
 No. 20-1337 Graveline et al. v. Benson et al. Page 28

make it appear. The district court selected the number from Michigan’s minimum requirement
for qualifying petitions in districts with populations up to 4,999,999, which is the rung just below
the statewide requirement. Mich. Comp. Laws § 168.544f. Considering that the Michigan
legislature has determined that 12,000 signatures adequately protect its interests for populations
approaching five million and the most recent gubernatorial election had a total electorate of
4,250,585, Appellants’ Br. at 60, we cannot say that the district court made a clear error in
judgment in requiring independent candidates to collect at least 12,000 signatures to qualify for
statewide offices. Finally, in granting this relief, the district court clarified that it “did not intend
to tell the Michigan Legislature what future changes to the ballot access framework would or
would not pass constitutional scrutiny.” Graveline v. Benson, No. 18-12354, 2020 WL 1303211,
at *2 (E.D. Mich. Mar. 19, 2020) (order). In short, “the [c]ourt has done its best to create a
workable interim provision without treading too far into matters best left to the State.” Green
Party of Ga. v. Kemp, 171 F. Supp. 3d 1340, 1374 (N.D. Ga. 2016) (entering a permanent
injunction with an interim requirement of 7,500 valid signatures in order to be “sensitive to the
State’s interests”). Thus, we hold that the district court did not abuse its discretion in crafting its
remedy.

Defendants’ argument that the district court abused its discretion by failing to set a new
maximum number of signatures is also unavailing. As the district court pointed out, Defendants
never raised the issue until after the district court entered final judgment. Graveline, 2020 WL
1303211, at *2. But setting that aside, Defendants have not shown how the district court’s
refusal would result in manifest injustice. GenCorp, 178 F.3d at 834. Defendants claim that in
allowing 60,000 signatures to remain the maximum while the minimum is lowered to 12,000,
Michigan will have to review many more signatures than is needed to qualify, which will be “a
considerable burden.” Appellants’ Br. at 62. This possibility seems very remote considering
that no candidate has ever qualified before and the evidence in the record establishing the
substantial amount of time, resources, and finances that candidates would have to expend in
order to reach even 30,000 signatures. See R. 28-7 (Albright Report at 2–4) (Page ID #390–92).
Candidates have no incentive to expend resources collecting far more signatures than is
necessary to qualify. Therefore, we also hold that the district court did not abuse its discretion
by declining to amend its findings.
 No. 20-1337 Graveline et al. v. Benson et al. Page 29

III. CONCLUSION

We recognize that “[t]he role of this court is not to impose our own idea of democracy
upon the [Michigan] state legislature.” Libertarian Party of Ohio, 462 F.3d at 587. Instead, it is
our responsibility to ensure that Michigan’s provisions for qualifying independent candidates for
statewide office fall within the bounds of what the First and Fourteenth Amendments require.
Our careful review of the facts and circumstances leads us to conclude that the 30,000-signature
requirement, geographic-distribution requirement, and filing deadline, when viewed in
combination, unconstitutionally burden Plaintiffs’ First Amendment rights.

For all of the reasons set forth above, we AFFIRM the judgment of the district court.
 No. 20-1337 Graveline et al. v. Benson et al. Page 30

_________________

DISSENT
_________________

GRIFFIN, Circuit Judge, dissenting.

Three years ago, the district court ordered that the State of Michigan place an unqualified
independent candidate for attorney general on the November 2018 general election ballot.
Graveline v. Johnson, 747 F. App’x 408, 418 (6th Cir. 2018) (Griffin, J., dissenting). We
refused to stay that decision, id. at 416 (maj. op.), and today double down in affirming the district
court’s rewriting of Michigan’s election laws governing ballot access. Specifically, under the
district court’s permanent injunction, independent candidates for statewide office may now
submit 3,000 fewer signatures (12,000) than partisan candidates for the same office (15,000) and
may now submit the same number of signatures as those required for an election district that is
only a fourth of Michigan’s population. M.C.L. § 168.544f. Percentages also tell the story of
the district court’s drastic remedy we now approve. The statute’s 30,000-signature threshold
represents just over seven tenths of a percent (.72%) of the 4,142,044 votes cast for attorney
general in the 2018 general election. The district court’s diminishing of the requirement to
12,000 signatures is even more miniscule, reducing an already low bar to a third of its former
height. Because I cannot agree that Michigan’s ballot-access laws unconstitutionally infringe
upon plaintiffs’ rights under the First Amendment, I respectfully dissent.1 I would reverse the
district court’s judgment.

I.

A.

In holding the combined effect of the laws at issue impose a severe burden on plaintiffs’
First Amendment rights, my colleagues—as they did when we previously denied defendants’
motion to stay—treat as definitive the historical fact that no independent candidate has submitted
a qualifying petition for a statewide office since 1988. See Graveline, 747 F. App’x at 414. That

1
I agree, however, that defendants’ standing and mootness arguments are not meritorious.
 No. 20-1337 Graveline et al. v. Benson et al. Page 31

is error. Historical evidence is “not conclusive in and of itself,” and represents instead just “a
helpful guide.” Libertarian Party of Ohio v. Blackwell, 462 F.3d 579, 589, 590 (6th Cir. 2006).
And, what was said earlier remains true today: “the mere absence of independent candidates, in
and of itself, says little about the effect of these restrictions; for all we know, Graveline could be
the first and only independent candidate to even attempt to qualify for the statewide ballot under
Michigan’s current scheme.” Graveline, 747 F. App’x at 417 (Griffin, J., dissenting). And as I
detail next, he plainly is the only such individual to try to do so.

The majority opinion tries to fill this gaping evidentiary hole with record evidence
establishing that thirty other individuals formed committees to run as independent candidates for
statewide offices. But this purported proof of disparate treatment operates in a different
paradigm. Michigan’s ballot-access scheme for independent statewide candidates mandates the
filing of signatures by no later than 110 days before the general election. M.C.L. § 168.590c(2).
My colleagues find this deadline severely burdensome given it occurs fifty days before partisan
nominating conventions for attorney general (and other statewide offices). See also Graveline,
747 F. App’x at 413.2 But all thirty of the committees relied on were those formed for
gubernatorial campaigns, and the major parties select their nominees for that position through
the traditional primary process and not via nominating conventions. M.C.L. § 168.53. And in
2018, for example, that occurred on August 7, 2018—about two weeks after Graveline’s July 19
signature deadline. These other committees are thus immaterial for purposes of plaintiffs’ claims
here.3

This factual distinction notwithstanding, there is a broader problem with finding
constitutional fault with Michigan’s signature, geographic, and durational requirements.

2
I note that bipartisan bills are currently pending in both chambers of Michigan’s legislature that would
fundamentally alter Michigan’s primary system. See, e.g., S.B. 130 and H.B. 4530, 2021-22 Leg., 101st Sess.
(Mich. 2021). Although the current version of the bills do not appear to materially alter the State’s nominating-
convention statute relied upon by my colleagues to hold in plaintiffs’ favor, see M.C.L. § 168.591(1), any change
should merit close review in the future as it relates to maintaining the district court’s permanent injunction now
approved by the majority.
3
Nor do the records demonstrate these other “candidates” did much, if anything, to advance their
campaigns. Take the seven who formed committees in 2018. Two were formed about two weeks before the general
election—well past the July signature-gathering deadline. The five that were timely formed reported little to no
contributions or expenditures, which undermines any argument regarding the genuineness of their “campaigns.”
 No. 20-1337 Graveline et al. v. Benson et al. Page 32

It equates proof that individuals formed committees to seek statewide offices (which must be
done to begin fundraising, collect signatures, and conduct other campaign activities) with proof
establishing these same individuals attempted to qualify for the ballot. The first is a requisite for
the second, but the same cannot be said in reverse. Simply, that several candidates formed
campaign committees says absolutely nothing about whether the statutory scheme inhibited them
from satisfying that scheme. There is nothing in the record that demonstrates the three elements
of the statutes plaintiffs challenge operated to preclude those other individuals from the ballot.
That is, we know nothing about how many, where from, and when they collected signatures.
Indeed, we do not even know if they collected any at all.

The only other record evidence relating to the burden imposed on plaintiffs comes from
the State’s expert, Lee Albright. As my colleagues note, Albright’s report makes clear that
mounting a signature-collection campaign requires significant effort, resources, and time. But
that is not a controversial proposition. See Libertarian Party of Kentucky v. Grimes, 835 F.3d
570, 577 (6th Cir. 2016) (states have “an important interest in ensuring that candidates
demonstrate a ‘significant modicum of support’ before gaining access to the ballot, primarily in
order to avoid voter confusion, ballot overcrowding, and frivolous candidates” (quoting Jenness
v. Fortson, 403 U.S. 431, 442 (1971))). Nor does it render a state’s ballot-access laws “severe.”
Id. (“[T]he incidental costs of gathering signatures on petitions do not come close to exclusion
from the ballot, and thus do not impose a severe burden on ballot access.”).

But more to the point, Albright was unequivocal with respect to whether Michigan’s
30,000-signature requirement within 180 days wholly precludes a candidate from the ballot:
“[G]athering 30,000 valid signatures of qualified Michigan voters within the 180 days allowed
can most certainly be done.” Plaintiffs did not challenge Albright’s it’s-hard-but-doable
conclusion below or offer their own rebuttal expert report. Sure, we have acknowledged that this
does not preclude a severe-burden conclusion per se, see Libertarian Party of Ohio, 462 F.3d at
592, but we have also made clear that history alone is not enough, id. at 589–90. And we have
also categorically rejected the proposition that “establishing . . . a petition deadline set before or
at the same time as a state’s primary election is necessarily unconstitutional.” Green Party of
 No. 20-1337 Graveline et al. v. Benson et al. Page 33

Tennessee v. Hargett, 767 F.3d 533, 548 (6th Cir. 2014). Simply, more evidence is required to
sustain a severe-burden finding than was put forth here.

In sum, the record facts establish only that one candidate has ever even attempted to
qualify as an independent candidate for a statewide office—plaintiff Christopher Graveline. He
collected nearly half the signatures required (14,157) in less than a quarter of the time permitted
(42 days). And neither the district court below nor my colleagues today contend his effort was
reasonably diligent given his significant delay in starting his signature-gathering effort. See
Storer v. Brown, 415 U.S. 724, 742 (1974); see also Graveline, 747 F. App’x at 417 (Griffin, J.,
dissenting). Accordingly, I cannot agree that plaintiffs satisfied their burden to establish
Michigan’s statutory scheme infringes upon their constitutional rights. See Hargett, 767 F.3d at
548–49 & n.4.4

B.

There is another problem with the majority opinion’s severe-burden conclusion—it is
inconsistent with our caselaw created in the wake of the COVID-19 pandemic. See Kishore v.
Whitmer, 972 F.3d 745 (6th Cir. 2020).

Like in many other states, the Governor of the State of Michigan ordered the public to
take drastic steps to mitigate the spread of COVID-19 beginning in March 2020, including
prohibiting most public gatherings. The State’s Stay-at-Home Order, combined with its
announced strict enforcement of election-related signature requirements, meant that those
seeking to collect signatures required to appear as candidates (or to have an initiative be placed
on a ballot) could not gather the requisite signatures. So they turned to the federal courts for
relief. See, e.g., id.; Esshaki v. Whitmer, 813 F. App’x 170 (6th Cir. 2020) (order); SawariMedia
v. Whitmer, 963 F.3d 595 (6th Cir. 2020) (order).

4
For this same reason, the district court wrongly faulted defendants for “fail[ing] to cite any authority for
the proposition that a plaintiff challenging ballot access laws must produce evidence of candidates who tried but
failed to qualify.” Graveline v. Benson, 430 F. Supp. 3d 297, 311 (E.D. Mich. 2019). As Hargett makes clear, that
shifting of the burdens—which my colleagues adopt by criticizing the State for “not put[ing] forth any evidence to
justify” the lack of independent candidates—is untenable. 467 F.3d at 548 & n.4 (discussing how plaintiffs could
develop a factual record when challenging a state’s ballot-access scheme).
 No. 20-1337 Graveline et al. v. Benson et al. Page 34

Kishore is directly on point. That matter dealt with two independent candidates for
president and vice president. 972 F.3d at 747. They faced the same signature-gathering
provisions Graveline challenges here: the collection of 30,000 signatures (as modified by the
district court’s injunction in this case), from at least 100 registered voters in at least half of
Michigan’s congressional districts, all within a 180-day window that closed in July. Id. at 747–
48. Even with limitations wrought by the pandemic and the State’s response to it, we held the
burden imposed by Michigan’s statutory scheme on an independent candidate could be no more
than intermediate. Id. at 750. Because there is no connection between those extraordinary
limitations and plaintiffs’ challenge today, no reading of Kishore supports finding a severe
burden here.

My colleagues primarily distinguish Kishore by pointing to a footnote where we
referenced the district court’s reduction to 12,000 signatures that we address today. Id. at 748 &
n.1. In their view, the 18,000 additional signatures Graveline had to collect makes all the
difference for why we should find a severe, and not intermediate, burden. I do not agree that the
statute’s 30,000-signature requirement that applies in ordinary times is somehow more
constitutionally oppressive than having to gather 12,000 signatures during the initial surge of the
COVID-19 pandemic in early 2020 when the State of Michigan (like other states) aggressively
restricted nearly every aspect of daily life. See Thompson v. DeWine, 959 F.3d 804, 806, 809
(6th Cir. 2020) (per curiam). As I read Kishore, its analysis turned not on numerical differences
in signature gathering, but rather on the timing of those collected signatures and the plaintiffs’
lack of reasonable steps—just like Graveline here—to collect those signatures. 972 F.3d at 750.
If its decision hinged on the 12,000-signature requirement (as my colleagues posit), one would
expect some mention of it in the burden analysis, yet we made only a fleeting reference to the
lower threshold while setting forth that case’s facts and procedural history. See id. at 748 & n.1.

Grimes further buttresses this intermediate-burden conclusion. There we considered
Kentucky’s election-law scheme that gives political candidates who are members of a political
party “blanket access” to the ballot—by way of nomination or primary election—if that political
party received two percent of the vote of the state in the last presidential election. 835 F.3d at
572. And if it did not, then candidates could file petitions to be placed on the ballot, which
 No. 20-1337 Graveline et al. v. Benson et al. Page 35

requires, among other things, 5,000 signatures for statewide offices, collected within a year of
the election, and filed by the second Tuesday in August preceding the general election. Id. at
572–73. We concluded this system imposed only an intermediate burden for several reasons—
the greater financial costs associated with minor parties achieving ballot status “did not
constitute exclusion or virtual exclusion from the ballot,” and third-party candidates had
appeared on the ballot in years past using the petition system. Id. at 575–76. Importantly, we
reasoned the “combined effect” of Kentucky’s system did not severely burden ballot access
because, among other reasons,

Kentucky allows political groups to file petitions for candidacy three months
before the general election, and provides such groups approximately nine months
to gather a quantity of signatures that amounts in almost all cases to less than
0.3% of the number of registered voters in the political unit that corresponds to
each office.

Id. at 576. Michigan’s scheme, although slightly more restrictive (.72% within six months), is in
accord. See also Jenness, 403 U.S. at 438 (approving of a 5% signature requirement to be
collected over a six-month period).

So too does Lawrence v. Blackwell. As here, that matter addressed a challenge to a
state’s scheme governing ballot access for independent (as opposed to minor party) candidates.
430 F.3d 368, 370 (6th Cir. 2005). The plaintiff there wanted to run as an independent candidate
for Congress, which, under Ohio law, meant that he had to file a nominating petition with at least
one percent of the electors in his congressional district on the day before the primary election.
Id. We found unpersuasive the argument that this early filing scheme unconstitutionally
burdened independent candidates. As here, the earlier filing deadline was “not so early that a
diligent candidate cannot meet the requirement,” and it was “reasonable because it prevents such
candidates from being able to make a decision to run for office after learning which candidates
will be representing the major parties.” Id. at 373–74. And we expressly noted “there is nothing
in the case law which suggests that a state is required to give independent candidates the
advantage of jumping into a race in response to late-breaking events which impact the political
landscape when major parties do not have the same flexibility.” Id. at 374.
 No. 20-1337 Graveline et al. v. Benson et al. Page 36

At the motion-to-stay stage of this litigation, we distinguished Lawrence (given the one-
day difference in filing deadlines for independent and primary candidates) and instead relied on
Libertarian Party of Ohio, 462 F.3d at 579. See Graveline, 747 F. App’x at 413–14. We follow
that blueprint again today. I remain unconvinced.

First, Libertarian Party of Ohio dealt not with a scheme governing independent
candidates, but a minor party’s ability to access the ballot. 462 F.3d at 586–91. A review of that
opinion reveals our analysis turned on our conclusion that those restrictions “serve[d] to prevent
a minor political party from engaging in the most fundamental of political activities—recruiting
supporters, selecting a candidate, and placing that candidate on the general election ballot in
hopes of winning votes and ultimately, the right to govern.” Id. at 590. We came to this
conclusion by emphasizing the Supreme Court’s preference for permitting the development of
“new political parties,” id. at 588, and political parties’ relationships with the voters’ First
Amendment rights, id. at 588–89. We then took those observations and found a severe burden
on those rights because Ohio’s scheme resulted in a monopolization by two political parties. Id.
at 589. The present matter, however, implicates no such institutional, political-party concerns.

Second, in today’s hyperpolitical environment, I cannot draw, as do my colleagues, an
inference that Michigan’s scheme burdens independent candidates differently by requiring them
to campaign when the public is not yet focused on election matters. One could argue Michigan’s
electorate was already politically motivated during the period in which Graveline was collecting
signatures to run as an independent candidate. By July 2018, Michigan voters had already
signed enough petitions to place one legislative initiative and two constitutional amendments on
the ballot. All three involved hundreds of thousands of signatures that were filed before
Graveline’s filing deadline (and two were filed in late 2017, before Graveline could even begin
collecting signatures in January 2018).5 Moreover, in that spring and summer, there were active
campaigns for two hotly contested statewide races, culminating in an August 7 primary election
for governor for three parties (with over two million voters choosing among eleven candidates)

5
See, e.g., Michigan Dep’t of State, State of Michigan Statewide Ballot Proposals, Sept. 19, 2018, available
at https://www michigan.gov/documents/sos/Bal_Prop_Status_560960_7.pdf.
 No. 20-1337 Graveline et al. v. Benson et al. Page 37

and the Republican nominee for Senate (with close to a million voters participating there).6 At
the very least, the record does not support the majority’s finding here. We should hold the
parties to evidence for this proposition, see Anderson v. Celebrezze, 460 U.S. 780, 792 & n.14
(1983), instead of just accepting general statements of the law.

For these reasons, I conclude Michigan’s statutory scheme imposes an intermediate
burden under Anderson-Burdick.

II.

Because the burden here is intermediate, Anderson-Burdick’s second and third steps merit
little consideration. At the second step, we have already approved Michigan’s justifications for
its ballot-access laws governing independent candidates: “the signature requirement ensures that
a candidate has a modicum of support before appearing on the ballot in order to further the
State’s interest in a fair and orderly election by avoiding ballot overcrowding, frivolous
candidates, and voter confusion” and the “filing deadline is essential to ensure an orderly, timely
election.” Kishore, 972 F.3d at 750–51; see also Grimes, 835 F.3d at 577 (similar). And to the
balancing point that is the third step, our review is “less exacting . . . , and a State’s important
regulatory interests will usually be enough to justify reasonable, nondiscriminatory restrictions”
given we deal here with “lesser burdens.” Timmons v. Twin Cities Area New Party, 520 U.S.
351, 358 (1997) (internal quotation marks omitted). Kishore commands that these interests
outweigh the burden imposed on independent candidates. 972 F.3d at 751; see also Lawrence,
430 F.3d at 375 (finding justifiable similar interests in relation to a one-percentage showing with
an early-submission deadline and commenting that “a wide range of methods of requiring
potential candidates to show support have been upheld as legitimate means to achieve this
important state interest”). It is not our job to “second-guess the legislative decisions of the
[Michigan Legislature] but only to evaluate whether those decisions pass constitutional muster.”
Grimes, 835 F.3d at 578. As set forth, “the strength of [Michigan]’s interests in avoiding voter
confusion, ballot overcrowding, and frivolous candidacies outweighs the modest burden of the
ballot-access regulations on [plaintiffs].” Id.

6
See Michigan Dep’t of State, 2018 Michigan Election Results,
https://mielections.us/election/results/2018PRI_CENR html.
 No. 20-1337 Graveline et al. v. Benson et al. Page 38

III.

This case illustrates once again why applying Anderson-Burdick’s grant of discretion to
the federal judiciary can lead to tension with the principles of federalism and separation of
powers. Cf. Daunt v. Benson, 956 F.3d 396, 424 (6th Cir. 2020) (Readler, J., concurring).
Before the 2018 general election, the district court entered a preliminary injunction mandating
that defendants place Graveline’s name on the ballot if he could demonstrate that he collected at
least 5,000 valid signatures with at least 100 of those from registered voters in each of at least
half of the State’s 14 congressional districts. Graveline v. Johnson, 336 F. Supp. 3d 801, 817
(E.D. Mich. 2018). But after Graveline’s name appeared on the 2018 general-election ballot, the
district court resolved plaintiffs’ claims on the merits and found 5,000 to be insufficient to
protect Michigan’s interests and then more than doubled it to 12,000 when devising a remedy.
430 F. Supp. 3d at 318. And today, we officially nullify Michigan’s election laws simply
because one dilatory candidate was unsuccessful in fulfilling Michigan’s reasonable statutory
requirements. Determining the appropriate signature threshold—whether 5,000, 12,000, 30,000,
or some other number—is best left to Michigan’s elected representatives, not federal judges. See
Esshaki, 813 F. App’x at 172 (concluding that lower court lacked authority to rewrite the same
ballot-access provisions at issue here because “federal courts have no authority to dictate to the
States precisely how they should conduct their elections”).

For these reasons, I respectfully dissent. I would reverse the judgment of the district
court.