League Of Women Voters Of Michigan V Secretary Of State

Michigan Supreme Court
Lansing, Michigan
Chief Justice: Justices:

Syllabus Bridget M. McCormack Brian K. Zahra
David F. Viviano
Richard H. Bernstein
Elizabeth T. Clement
Megan K. Cavanagh
Elizabeth M. Welch

This syllabus constitutes no part of the opinion of the Court but has been Reporter of Decisions:
prepared by the Reporter of Decisions for the convenience of the reader. Kathryn L. Loomis

LEAGUE OF WOMEN VOTERS OF MICHIGAN v SECRETARY OF STATE

Docket Nos. 163711, 163712, 163744, 163745, 163747, and 163748. Decided
January 24, 2022.

The League of Women Voters, Progress Michigan, the Coalition to Close Lansing
Loopholes, and Michiganders for Fair and Transparent Elections brought an action in the Court of
Claims against the Secretary of State, challenging the constitutionality of 2018 PA 608, which
changed the procedures by which the people of Michigan can circulate petitions to invoke the
referendum, initiative, and constitutional-amendment processes set forth in Michigan’s
Constitution and statutory election laws. Specifically, 2018 PA 608 amended MCL 168.471 to
state that no more than 15% of the signatures used to determine the validity of a petition could be
from any one congressional district; it amended MCL 168.482 by adding Subsection (7), which
required petitions to include checkboxes that would indicate whether the circulator of the petition
was a paid signature gatherer or a volunteer; and it added MCL 168.482a, which provides that
signature gatherers who are being paid must, before circulating any petition, file a signed affidavit
to that effect with the Secretary of State. The Department of the Attorney General intervened to
defend the laws, and the Michigan House and Senate participated as amici curiae. The Court of
Claims, CYNTHIA D. STEPHENS, J., struck down the geographical limitation in MCL 168.471 as
well as the checkbox requirement of MCL 168.482(7); however, it ruled that the affidavit
requirement, MCL 168.482a, was constitutional. The parties appealed, and the Court of Appeals
consolidated the appeals. Plaintiffs filed an application to bypass the Court of Appeals under MCR
7.305(C)(1)(a), which the Supreme Court denied. League of Women Voters of Mich v Secretary
of State, ___ Mich ___; 963 NW2d 377 (2021). The Court of Appeals, RONAYNE KRAUSE, P.J.,
and K. F. KELLY, J. (CAMERON, J., concurring), affirmed in part the Court of Claims’ decision,
striking as unconstitutional the geographic limitation in MCL 168.471 and the requirement in MCL
168.482(4) that petitions include language identifying the signer’s congressional district. The
Court of Appeals also reversed the Court of Claims as to the checkbox and affidavit requirements,
holding that the checkbox requirement in MCL 168.482 was constitutional but the affidavit
requirement in MCL 168.482a overly burdened the free-speech rights of the petitions’ sponsors.
League of Women Voters of Mich v Secretary of State, ___ Mich App___ (2021) (Docket Nos.
357984 and 357986). Plaintiffs sought leave to appeal, arguing that the checkbox requirement,
MCL 168.482(7), was unconstitutional. The Department of the Attorney General sought leave to
appeal the Court of Appeals’ holdings as to the 15% geographic requirement, MCL 168.471, and
the affidavit requirement, MCL 168.482a. Defendant Secretary of State sought leave to appeal in
order to request that, regardless of the outcome, the decision be applied only prospectively.

In an opinion by Justice CAVANAGH, joined by Chief Justice MCCORMACK and Justices
BERNSTEIN (except as to Part IV(A)), CLEMENT (except as to Parts IV(B) and V), and WELCH, the
Supreme Court, in lieu of granting leave to appeal and without hearing oral argument, held:

The 15% cap on signatures from any one congressional district and the precirculation
affidavit requirement for paid circulators violated the Michigan Constitution. The checkbox
requirement, however, passed constitutional muster. In light of the chaos and injustice that would
ensue were the opinion to be applied retroactively, the decision was given prospective effect only.
Any signature gathered after January 24, 2022, must be on a petition that conforms to the
requirements of MCL 168.482(7).

1. Const 1963, art 2, § 9 reserves to the people the ability to approve or reject legislation
that the Legislature has already adopted (the referendum) and to propose laws to the Legislature
and enact them if the Legislature refuses (the initiative). The initiative provision set forth in Article
2, § 9 serves as an express limitation on the authority of the Legislature. Although the Constitution
also directs the Legislature to implement these provisions, the Legislature’s power does not extend
to an ability to enact the 15% geographic-distribution requirement. The word “implement,” which
means “to put into practical effect” or “carry out,” carries the connotation that some received set
of rules is being carried out, not that a new set of rules is to be created. Const 1963, art 2, § 9
provides in part that the power of referendum must be invoked in the manner “prescribed by law.”
The committee in charge of drafting the Constitution used the phrase “prescribed by law” if only
the details were left to the Legislature and not the overall planning, whereas it used the phrase
“provided by law” when it intended that the Legislature do the entire job of implementation.
Accordingly, the Legislature is empowered only to adopt rules that further the principles already
set forth in Const 1963, art 2, § 9, which has no geographic-distribution requirement. The original
referendum and initiative provisions in Michigan were amendments of the legislative vesting
clause, Const 1908, art 5, § 1. While this provision originally did no more than vest the Legislature
with the legislative power of the state, as a result of political parties’ continually making and
breaking promises to pass legislation for which there was a popular demand, a 1913 amendment
took back from the Legislature the right of the people themselves to initiate legislation and to
approve legislation enacted by the Legislature. At the convention that produced the 1963
Constitution, much of the language added by amendment in 1913 was eliminated, but while
matters of legislative detail were left to the Legislature, the remaining language makes it clear that
this section is self-executing. Unless otherwise expressly indicated, the Legislature may not pass
laws supplementary to self-executing constitutional provisions that curtail or place undue burdens
on the rights guaranteed by those provisions. When implementing the direct-democracy provisions
of the 1963 Constitution, the Legislature may adopt the sorts of requirements that were formerly
provided in the Constitution of 1908—type size, the timeline for circulating petitions, the duties
of state officials in processing petitions that have been submitted, and so on. By contrast, the 15%
requirement in MCL 168.471 does not merely fill in necessary details, but rather adds a substantive
obligation. Further, by choosing a statewide minimum number of signatures without a geographic
cap, the people demonstrated their intent to allow a relatively small coalition of voters from a
concentrated geographic area to propose changes to the law, with the understanding that such
proposals would not become law without the approval of the Legislature or a majority of the voters
in a statewide election. A cap on how many signatures can come from each part of the state would
undermine these intentions by making it more difficult and expensive to gather the required
number of signatures within the time frame required by Michigan’s election laws, and the related
enforcement provisions would effectively ensure that some voters’ signatures would be rendered
void merely because they were obtained after the 15% cap for that district had been reached. Such
disenfranchisement would weigh most heavily on those residing in more densely populated parts
of the state, and it would run directly contrary to the clear intention that nothing more than a
minimum number of signatures from the statewide population is necessary to propose changes to
Michigan’s laws. For these reasons, the 15% geographic-distribution requirement is
unconstitutional.

2. Const 1963, art 12, § 2 provides that a petition to initiate a constitutional amendment
must be in the form, and signed and circulated in such manner, as “prescribed by law,” a phrase
used by the constitutional drafters when only the details were left to the Legislature and not the
overall planning. It is more than a detail to require that signatures be distributed geographically in
addition to requiring an overall raw total. This requirement does not merely restrict where
signatures of Michigan citizens may be collected; it limits from whom they may be gathered, and
is thus a limitation of a substantive right rather than a mere procedural requirement. While the
Legislature has some power to prescribe laws not specified in the four corners of the constitutional
text, the constitutional text and convention debates point to a limited role for the Legislature, and
the principle that the Legislature may not unduly burden the self-executing constitutional
procedure applies equally to both initiated legislation and initiated constitutional amendments. As
with legislative initiatives, the constitutional requirement for a minimum number of signatures
ensures a threshold level of support before a proposed change to our state’s Constitution can be
submitted to all voters for approval or rejection. But the 15% requirement exceeds the
Legislature’s authority to regulate a self-executing constitutional process by imposing a
substantive requirement that does not advance the express constitutional requirement. It does not
align with any of the aspects of statutory detail that were in the Constitution of 1908 and removed
when the Constitution of 1963 was proposed and ratified; rather, it aligns with proposals that the
convention specifically rejected, apparently, in no small part out of concerns that such
requirements would reduce or enhance the political power of Michiganders on the basis of the
location of their residence. Although Article 12, § 2 requires a proposed constitutional amendment
to be supported by “registered electors of the state equal in number to at least 10 percent of the
total vote cast” for Governor in the last election, the Constitution requires nothing more than this
minimum level of support from the electorate as a whole, and it does not require that such support
be evenly distributed in geographic terms. Accordingly, this requirement is an undue burden on
the Constitution’s self-executing voter-initiated constitutional-amendment process.

3. Plaintiffs’ challenge to the checkbox and affidavit requirements in 2018 PA 608 is
grounded in federal constitutional law. The right to solicit signatures to qualify an initiative
petition is protected by the constitutionally guaranteed rights of free expression, assembly, and
petition. Petition circulation is core political speech, and laws that directly or severely burden
political speech are generally subject to strict or exacting judicial scrutiny. Compelled disclosure
requirements for petition circulators may be reviewed under exacting scrutiny, which requires a
relevant correlation or substantial relation between the governmental interest and the information
required to be disclosed. That governmental interest must also be sufficiently important and must
reflect the seriousness of the actual burden on First Amendment rights. However, states have
considerable leeway to protect the integrity and reliability of the initiative process, and there is a
difference between regulations that directly affect core political speech and those that simply
regulate the mechanics of the electoral process. No bright line separates permissible election-
related regulation from unconstitutional infringements on First Amendment freedoms, and the
United States Supreme Court has left the determination of the appropriate level of scrutiny open-
ended. If there is no severe burden on political speech or the regulation is merely one of
mechanical electoral processes, courts generally apply a more flexible review such as the test set
forth in Anderson v Celebrezze, 460 US 780 (1983), and Burdick v Takushi, 504 US 428 (1992).
The Anderson-Burdick test requires a reviewing court to weigh the character and magnitude of the
burden that the state’s rule imposes on First Amendment rights against the interests that the state
contends justify that burden and consider the extent to which the state’s concerns make the burden
necessary.

4. Pursuant to 2018 PA 608, MCL 168.482(7) requires a petition that proposes a
constitutional amendment, initiation of legislation, or referendum on legislation to include
checkboxes that clearly indicate whether the circulator of the petition is a paid signature gatherer
or a volunteer signature gatherer. Under MCL 168.482(8), any signature obtained on a petition
that does not comply with the checkbox requirement is invalid and will not be counted. Similar to
the statute at issue in Buckley v American Constitutional Law Foundation, 525 US 182 (1999),
which required petition circulators to wear identification badges stating their names and whether
they were volunteers or being paid (and, if so, by whom), the checkbox requirement in MCL
168.482(7) compels the petition circulator to disclose their status as paid or volunteer at the same
time the political message is being delivered. However, in this case, circulators are not being
forced to reveal anything as personal as their identity or their employer, and they are therefore not
subject to the same sort of personalized heat-of-the-moment harassment that was present in
Buckley. Assuming that the checkbox imposes some direct but minimal burden on core political
speech, pursuant to exacting-scrutiny review, the state must still have an adequate interest in
creating the checkbox requirement, and the checkbox must bear a substantial relationship to that
interest. The Court of Appeals concluded that transparency in the political process, especially
transparency that permits voters to “follow the money,” is a compelling state interest, and the
United States Supreme Court has repeatedly held that increasing the amount of information
available to the voters is a legitimate state interest. Given the limited nature of the disclosure at
issue, the actual burden on First Amendment rights caused by the checkbox requirement was so
minimal that a governmental interest in increasing information for voters justified the requirement.
Therefore, MCL 168.482(7) survives exacting scrutiny.

5. 2018 PA 608 added MCL 168.482a, which requires paid signature gatherers, before
circulating any petition, to file a signed affidavit with the Secretary of State indicating that they
are paid signature gatherers. Any signature obtained on a petition by a paid circulator who has not
filed the precirculation affidavit is invalid and must not be counted. Unlike the checkbox
requirement, the affidavit requirement is a prerequisite to circulation of a petition that is a step
removed from the communicative aspects of petitioning. Because it does not directly burden core
political speech, the flexible Anderson-Burdick test applies. However, even under this more
relaxed standard, the affidavit requirement of MCL 168.482a does not pass constitutional muster,
given the burden it places on groups that rely on paid signature gatherers and the lack of an
apparent state interest served by the affidavit. Accordingly, the Court of Appeals’ holding that
MCL 168.482a is unconstitutional was affirmed.

6. Generally, judicial decisions are to be given complete retroactive effect. However,
where injustice might result from full retroactivity, holdings may be given limited retroactive or
prospective effect. Because the Court agreed with the lower courts as to the unconstitutionality of
MCL 168.471 and MCL 168.482a, there was no reason to depart from the Court’s standard practice
of providing retroactive effect to the Court’s decision regarding those two provisions. Instead, the
Court focused on whether enforcement of the checkbox’s constitutionality should be afforded
prospective-only effect. The constitutionality of the checkbox requirement in MCL 168.482(7)
was an issue of first impression that had been the subject of debate; therefore, the conclusion that
it was constitutional established a new principle of law. When a decision clearly establishes a new
principle of law, the Court considers three factors: (1) the purpose to be served by the new rule,
(2) the extent of the reliance on the old rule, and (3) the effect of retroactivity on the administration
of justice. Because the outcome of this case materially affected the right of the people to exercise
direct democracy, the test pointed toward prospective application of this decision. The purpose
served by the checkbox is generally to allow the public to be more informed about the paid or
volunteer status of circulators when solicited to sign a petition. Giving the ruling prospective
application will deprive voters who have already signed petitions without checkboxes of this
information, but the remaining two factors outweighed this countervailing consideration. As to
the extent of the reliance on the old rule, until the Court of Appeals upheld the checkbox
requirement, every court that had considered the question held that MCL 168.482(7) was
unconstitutional. Further, the Board of State Canvassers approved as to form the petitions of two
ballot-question committees whose petitions lacked paid circulator checkboxes, and the committees
reasonably relied on that approval under the old rule. Moreover, retroactive application would
almost certainly result in additional litigation with respect to petition signatures that have already
been gathered, causing serious confusion for voters who have already signed the petitions currently
in circulation. A person may sign a petition only once, MCL 168.482(5); therefore, invalidating
signatures already collected on checkbox-lacking petitions would make collecting them again
difficult, as signatories might refuse to engage with the circulator on the ground that they had
already signed the petition. This would infringe the electors’ rights to communicate and speak by
petition. Under these circumstances, the test for prospective application was satisfied.

Affirmed.

Justice ZAHRA, joined by Justice VIVIANO, concurring in part and dissenting in part,
concurred with the majority that the geographic-distribution requirement was unconstitutional as
to initiative and referendum petitions, that the checkbox requirement passed constitutional muster,
and that the Court’s opinion should apply prospectively only under the circumstances of this case.
However, he dissented from the majority’s conclusion that the affidavit requirement and the
geographic-distribution requirement as to voter-initiated constitutional amendments were
unconstitutional. He would have held that the geographic-distribution requirement was a matter
for the Legislature to prescribe, stating that this conclusion was consistent with the ratifiers’
understanding that voter-initiated constitutional amendments were to be reserved for substantial
matters worthy of constitutional elevation rather than routine policy matters normally addressed
through legislation and did not conflict with the self-executing nature of Article 12, § 2. He also
dissented from the majority’s conclusion that the affidavit requirement was facially
unconstitutional, explaining that, like the checkbox requirement, the affidavit requirement
imposed minimal burdens on petition circulators, did not bar paid petition circulation altogether,
did not chill or deter paid circulators from speaking, and served the important state interests of
detecting and deterring fraud, as well as assisting in the discovery of invalid signatures. He would
have held that the challengers to these two legislative provisions had failed to overcome the strong
presumption of constitutionality accorded all legislation duly passed through a bicameral
legislature and signed by the Governor, and he stated that holding the provisions unconstitutional
improvidently thwarted the will of the people as clearly expressed through their elected
representatives.

Justice BERNSTEIN, concurring in part and dissenting in part, agreed with the majority
opinion in large part but dissented with respect to Part IV(A), which addressed the checkbox
requirement. While he agreed that this requirement was subject to review under the exacting-
scrutiny standard, he would have held that the strength of the governmental interest was
insufficient to overcome even the minimal burden that the checkbox requirement imposed, given
that the state did not identify how the presence of a checkbox that imparted so little information
would advance its vaguely stated interest in transparency. He joined the majority opinion in all
other respects.

Justice CLEMENT, concurring in part and dissenting in part, agreed in full with the
majority’s analysis of why the 15% cap for direct-democracy signatures violated the Michigan
Constitution, as well as with its decision that the checkbox requirement complied with the First
Amendment. However, she dissented from its holding that the affidavit requirement violated the
First Amendment for the reasons stated by Justice ZAHRA. She also dissented from the Court’s
decision to give this opinion only prospective effect, given that neither of the petition sponsors that
would have been affected by the retroactive application of this decision had submitted its
signatures yet. She also noted that there was a serious question as to whether it was constitutionally
legitimate for the Supreme Court to render purely prospective opinions because they are essentially
advisory and do not come within the Court’s limited authority to issue advisory opinions under
Const 1963, art 3, § 8.
 Michigan Supreme Court
Lansing, Michigan
Chief Justice: Justices:

OPINION Bridget M. McCormack Brian K. Zahra
David F. Viviano
Richard H. Bernstein
Elizabeth T. Clement
Megan K. Cavanagh
Elizabeth M. Welch

FILED January 24, 2022

STATE OF MICHIGAN

SUPREME COURT

LEAGUE OF WOMEN VOTERS OF
MICHIGAN, PROGRESS MICHIGAN,
COALITION TO CLOSE LANSING
LOOPHOLES, and MICHIGANDERS FOR
FAIR AND TRANSPARENT ELECTIONS,

Plaintiffs-Appellants,

v Nos. 163711-2

SECRETARY OF STATE,

Defendant-Appellee,

and

ATTORNEY GENERAL,

Intervening Defendant-
Appellee.
LEAGUE OF WOMEN VOTERS OF
MICHIGAN, PROGRESS MICHIGAN,
COALITION TO CLOSE LANSING
LOOPHOLES, and MICHIGANDERS FOR
FAIR AND TRANSPARENT ELECTIONS,

Plaintiffs-Appellees,

v Nos. 163744-5

SECRETARY OF STATE,

Defendant-Appellee,
and

ATTORNEY GENERAL,

Intervening Defendant-
Appellant.

LEAGUE OF WOMEN VOTERS OF
MICHIGAN, PROGRESS MICHIGAN,
COALITION TO CLOSE LANSING
LOOPHOLES, and MICHIGANDERS FOR
FAIR AND TRANSPARENT
ELECTIONS,

Plaintiffs-Appellees,

v Nos. 163747-8

SECRETARY OF STATE,

Defendant-Appellant,
and

ATTORNEY GENERAL,

Intervening Defendant-
Appellee.

2
BEFORE THE ENTIRE BENCH

CAVANAGH, J.
This case involves the validity of amendments to the Michigan Election Law, MCL

168.1 et seq., which have the ability to affect how millions of Michiganders participate in

this state’s direct-democracy process. It, therefore, concerns issues that are of the utmost

importance to the constitutional rights of the voters in this state. It is this Court’s duty to

bring finality to these issues with a clear and decisive answer. To that end, we affirm the

Court of Appeals’ resolution of the constitutionality of the disputed provisions at issue.

We agree that the 15% geographic requirement in MCL 168.471, as amended by 2018 PA

608, is unconstitutional, as are the other provisions relating to this geographic requirement

in MCL 168.477 and MCL 167.482(4). Additionally, the affidavit requirement that 2018

PA 608 added to MCL 168.482a is unconstitutional. However, the checkbox requirement

that was added to MCL 168.482(7) passes constitutional muster. We further hold that our

decision today will be afforded prospective effect. 1

I. FACTS AND PROCEDURAL HISTORY

Our Constitution recognizes three forms of direct democracy. First, the people have

“the power to approve or reject laws enacted by the legislature, called the referendum.”

Const 1963, art 2, § 9. Second, the people have “the power to propose laws and to enact

and reject laws, called the initiative.” Id. Finally, “[a]mendments may be proposed to this

1
The issues presented in the instant litigation are well known to this Court, see, e.g., League
of Women Voters of Mich v Secretary of State, 506 Mich 561; 957 NW2d 731 (2020), and
the parties agree that an expeditious response is necessary to ensure that voters may be
heard. Accordingly, we decide this case without oral argument.

3
constitution by petition of the registered electors of this state.” Const 1963, art 12, § 2. To

avail themselves of these processes, proponents must gather petition signatures.

To invoke the initiative or referendum, petitions signed by a number of
registered electors, not less than eight percent for initiative and five percent
for referendum of the total vote cast for all candidates for governor at the last
preceding general election at which a governor was elected shall be required.
[Const 1963, art 2, § 9.]

To initiate a constitutional amendment, “[e]very petition shall . . . be signed by registered

electors of the state equal in number to at least 10 percent of the total vote cast for all

candidates for governor at the last preceding general election at which a governor was

elected.” Const 1963, art 12, § 2.

The Constitution gives the Legislature a role in regulating all three processes. It is

directed to “implement the provisions of” the section on referendum and initiative, Const

1963, art 2, § 9, while petitions for a constitutional amendment “shall be in the form, and

shall be signed and circulated in such manner, as prescribed by law,” Const 1963, art 12,

§ 2. The Michigan Election Law, MCL 168.1 et seq., designates the Secretary of State as

the official with whom such petitions are to be filed. See MCL 168.471. The Secretary of

State then submits the petitions to the Board of State Canvassers, MCL 168.475(1), which

“canvass[es] the petitions to ascertain if the petitions have been signed by the requisite

number of qualified and registered electors,” MCL 168.476(1), and then “make[s] an

official declaration of the sufficiency or insufficiency of a petition,” MCL 168.477(1).

In December 2018, the Legislature adopted and the Governor signed into law Public

Act 608. It made three relevant changes to the procedures regulating the referendum,

initiative, and constitutional-amendment process. First, it amended MCL 168.471 to

4
require that, when “determin[ing] the validity of a petition,” “[n]ot more than 15% of the

signatures to be used . . . shall be of registered electors from any 1 congressional district.” 2

Second, it amended MCL 168.482 by adding Subsection (7), which changed the required

form of petitions to include checkboxes “to clearly indicate whether the circulator of the

petition is a paid signature gatherer or a volunteer signature gatherer.” 3 Finally, it added

MCL 168.482a, which provides in Subsection (1) that paid signature gatherers “must,

before circulating any petition, file a signed affidavit with the secretary of state that

indicates he or she is a paid signature gatherer”; subsequent subsections set out

consequences for noncompliance.

In January 2019, the Secretary of State requested that the Attorney General advise

her on the constitutionality of these three provisions. 4 The Attorney General opined that

all three provisions violated the state and federal Constitutions. See OAG, 2019-2020, No.

7,310 (May 22, 2019). 5 Like in the instant case, a cohort of similar plaintiffs challenged

2
For conformity with this geographic-distribution requirement, it also amended MCL
168.477 to forbid the Board of State Canvassers from counting the signature of a registered
elector from a congressional district above the 15% limit, and it amended MCL 168.482(4)
to change the required form of petitions to include a declaration about the congressional
district in which the signing electors reside.
3
It also added MCL 168.482c, which made it a misdemeanor for a petition circulator to
“knowingly make[] a false statement concerning his or her status as a paid signature
gatherer or volunteer signature gatherer . . . .”
4
By statute, the Attorney General gives opinions on questions of law posed by state
officers. MCL 14.32.
5
Specifically, the Attorney General’s opinion was that the geographic-distribution
requirement violated the direct-democracy provisions of the Michigan Constitution and the
paid-circulator requirements violated the free-speech protections of the United States
Constitution. The Attorney General did opine, however, that certain other provisions of

5
the constitutionality of the provisions in the Court of Claims seeking a declaratory

judgment consistent with OAG, 2019-2020, No. 7,310. Litigation ensued, raising largely

the same issues we decide today, as well as other issues that are no longer relevant to the

instant dispute. Eventually, however, the lead plaintiff in that litigation abandoned its

ballot initiative in the wake of the COVID-19 pandemic, making the issues moot and

leaving no party with standing to pursue the appeal. Declining to issue an advisory opinion,

this Court vacated the lower-court decisions and remanded those cases to the trial court to

be dismissed. League of Women Voters of Mich v Secretary of State, 506 Mich 561, 571;

957 NW2d 731 (2020).

On February 8, 2021, the instant plaintiffs filed a new complaint in the Court of

Claims challenging the constitutionality of 2018 PA 608. The Department of the Attorney

General intervened to defend the laws, and the Michigan House and Senate participated as

amici curiae. Cross-motions for summary disposition were filed. On July 12, 2021, the

Court of Claims struck down the geographical limitation in MCL 168.471 as well as the

checkbox requirement of MCL 168.482(7); however, it upheld the affidavit requirement,

MCL 168.482a, as constitutional.

Both plaintiffs and defendants appealed, and the Court of Appeals consolidated the

appeals. Plaintiffs filed a bypass application in this Court under MCR 7.305(C)(1)(a), but

the Department of the Attorney General, as intervening defendant, 6 did not join in that

2018 PA 608—such as provisions in MCL 168.482a that invalidate signatures gathered
when various defects occur in the gathering process—were constitutional.
6
We note that the Department of the Attorney General is arguing both in favor of the
constitutionality of the provisions in its capacity as intervening defendant and against the
constitutionality of the statutes in its capacity as counsel for the Secretary of State.

6
request. This Court granted immediate consideration, but denied bypass and directed the

Court of Appeals to expedite its consideration. League of Women Voters of Mich v

Secretary of State, ___ Mich ___; 963 NW2d 377 (2021). On October 29, 2021, the Court

of Appeals affirmed in part the Court of Claims’ opinion striking as unconstitutional the

geographic limitation in MCL 168.471 and the requirement in MCL 168.482(4) that

petitions include language identifying the signer’s congressional district. The majority

opinion also reversed the Court of Claims as to the checkbox and affidavit requirements,

holding that the checkbox requirement in MCL 168.482 was constitutional, but that the

affidavit requirement in MCL 168.482a overly burdened the free-speech rights of the

petitions’ sponsors. League of Women Voters of Mich v Secretary of State, ___ Mich

App___; ___ NW2d ___ (2021) (Docket Nos. 357984 and 357986).

Plaintiffs applied for leave to appeal in this Court on November 2, 2021, arguing

that the checkbox requirement, MCL 168.482(7), was unconstitutional. They also moved

for immediate consideration and expedited briefing and argument and requested a decision

by December 31, 2021. Intervening defendant, the Department of the Attorney General,

applied for leave to appeal on November 15, 2021, asking this Court to reverse the Court

of Appeals as to the 15% geographic requirement, MCL 168.471, and the affidavit

requirement, MCL 168.482a. Defendant Secretary of State also applied for leave to appeal

on November 15, 2021, simply requesting that this Court apply any decision prospectively

in light of the fact that there are several petitions already in circulation that have sought to

comply with changing standards. The Court has also received amicus curiae briefs from

the Board of State Canvassers, the Michigan Senate and House of Representatives, and two

ballot-question committees, Secure MI Vote and Unlock Michigan. We now consider the

7
three applications together, as well as the other briefing submitted, affirm the Court of

Appeals as to the constitutionality of each provision, and hold that our decision shall be

applied prospectively only.

II. STANDARD OF REVIEW AND CONSTITUTIONAL STANDARDS

Whether a statute is constitutional is a question of law that we review de novo.

McDougall v Schanz, 461 Mich 15, 23; 597 NW2d 148 (1999). We presume that a statute

is constitutional “unless its unconstitutionality is clearly apparent.” Id. at 24.

Plaintiffs presented a facial challenge to the constitutionality of 2018 PA 608. “[A]

statute comes clothed in a presumption of constitutionality” because we presume that “the

Legislature does not intentionally pass an unconstitutional act.” Cruz v Chevrolet Grey

Iron Div of Gen Motors Corp, 398 Mich 117, 127; 247 NW2d 764 (1976). “The party

challenging the facial constitutionality of an act ‘must establish that no set of circumstances

exists under which the [a]ct would be valid. The fact that the . . . [a]ct might operate

unconstitutionally under some conceivable set of circumstances is insufficient . . . .’ ”

Council of Orgs & Others for Ed About Parochiaid, Inc v Governor, 455 Mich 557, 568;

566 NW2d 208 (1997), quoting United States v Salerno, 481 US 739, 745; 107 S Ct 2095;

95 L Ed 2d 697 (1987) (alterations in Council of Orgs). Our task, then, is to determine

whether 2018 PA 608 is unconstitutional in the abstract, rather than to “analyze the statute

‘as applied’ to the particular case.” Crego v Coleman, 463 Mich 248, 269; 615 NW2d 218

(2000).

8
 III. THE GEOGRAPHIC-DISTRIBUTION REQUIREMENT

We turn first to the requirements of MCL 168.471 and MCL 168.477 that no more

than 15% of the signatures required to invoke direct democracy can be gathered from one

congressional district. Plaintiffs challenged this requirement as a violation of the Michigan

Constitution. This Court has explained:

Our primary goal in construing a constitutional provision is to give
effect to the intent of the people of the state of Michigan who ratified the
Constitution, by applying the rule of “common understanding.” We locate
the common understanding of constitutional text by determining the plain
meaning of the text as it was understood at the time of ratification.
Interpretation of a constitutional provision also takes account of “the
circumstances leading to the adoption of the provision and the purpose
sought to be accomplished.” The Address to the People, which was
distributed to Michigan citizens in advance of the ratification vote and which
explained in everyday language what each provision of the proposed new
Constitution was intended to accomplish, and, to a lesser degree, the
constitutional convention debates are also relevant to understanding the
ratifiers’ intent. [Mich Coalition of State Employee Unions v Michigan, 498
Mich 312, 323-324; 870 NW2d 275 (2015) (citations omitted).]

We have previously held that both the section of our Constitution governing the

power of referendum and initiative, Const 1963, art 2, § 9, and the section providing for

citizen-initiated constitutional amendment, Const 1963, art 12, § 2, are self-executing.

Wolverine Golf Club v Secretary of State, 384 Mich 461, 466; 185 NW2d 392 (1971);

Citizens Protecting Michigan’s Constitution v Secretary of State, 503 Mich 42, 63; 921

NW2d 247 (2018). “It is settled law that the legislature may not act to impose additional

obligations on a self-executing constitutional provision.” Wolverine Golf Club, 383 Mich

at 466 (quotation marks and citation omitted). That said, the constitutional text for

referendums and initiatives is distinct from the text for constitutional amendments, so we

consider them separately.

9
 A. REFERENDUMS AND INITIATIVES

As indicated, the Michigan Constitution reserves to the people the ability to approve

legislation that the Legislature has already adopted (the referendum), and to propose laws

to the Legislature and enact them if the Legislature refuses (the initiative). Const 1963,

art 2, § 9. Direct democracy in Michigan is a series of powers that the people have reserved

to themselves from the Legislature. “The initiative provision set forth in art 2,

§ 9 . . . serves as an express limitation on the authority of the Legislature.” Woodland v

Mich Citizens Lobby, 423 Mich 188, 214; 378 NW2d 337 (1985). However, the

Legislature is directed to “implement the provisions” of Const 1963, art 2, § 9.

The Legislature’s power and duty to “implement” Const 1963, art 2, § 9 does not

support an ability to enact the 15% geographic-distribution requirement. Looking first to

the text of the provision, the word “implement” means “[t]o put into practical effect; carry

out[.]” American Heritage Dictionary of the English Language (5th ed). 7 It carries the

connotation that some received set of rules is being carried out, not that a new set of rules

is to be created. In keeping with this vision of a limited role for the Legislature, Const

1963, art 2, § 9 says that “[t]he power of referendum . . . must be invoked in the manner

prescribed by law within 90 days following the final adjournment of the legislative session

at which the law was enacted.” (Emphasis added.) This language has significance because,

as this Court has recognized:

7
Dictionaries that are more contemporaneous with the ratification of the Constitution give
essentially identical definitions. See Webster’s Seventh New Collegiate Dictionary
(defining “implement” as “to carry out : FULFILL; esp : to give practical effect to and ensure
of actual fulfillment by concrete measures”); Webster’s Third New International
Dictionary (“to carry out : ACCOMPLISH, FULFILL”).

10
 The committee on style and drafting of the constitutional convention of 1961
made a distinction in the use of the words “prescribed by law” and the words
“provided by law.” Where “provided by law” is used, it is intended that the
legislature shall do the entire job of implementation. Where only the details
were left to the legislature and not the over-all planning, the committee used
the words “prescribed by law.” [Beech Grove Inv Co v Civil Rights Comm,
380 Mich 405, 418-419; 157 NW2d 213 (1968).]

This text empowers the Legislature only to adopt rules that further the principles already

set forth in Const 1963, art 2, § 9—which has no geographic-distribution requirement.

Even so, the Legislature has adopted several valid rules for direct-democracy

petitions—deadlines, type-size requirements, and the like—that are not set out in the

Constitution. To truly make sense of the scope of this power to “implement,” we must

understand this section in the context of what it replaced. The original referendum and

initiative provisions in Michigan were amendments of the legislative vesting clause; while

Const 1908, art 5, § 1 originally did no more than vest the Legislature with the legislative

power of the state, an amendment ratified by the voters in 1913 added some 1,300 words

of additional detail clawing back from the Legislature the right of the people themselves to

initiate legislation and approve legislation enacted by the Legislature:

The initiative found its birth in the fact that political parties repeatedly made
promises to the electorate both in and out of their platforms to favor and pass
certain legislation for which there was a popular demand. As soon as election
was over their promises were forgotten, and no effort was made to redeem
them. These promises were made so often and then forgotten that the
electorate at last through sheer desperation took matters into its own hands
and constructed a constitutional procedure by which it could effect changes
in the Constitution and bring about desired legislation without the aid of the
legislature. [Hamilton v Secretary of State, 227 Mich 111, 130; 198 NW 843
(1924) (opinion of BIRD, J.).]

It was in this atmosphere of mistrust that the people ratified an amendment to the

Constitution that specified such details as the size of the type to be used (and its color),

11
various required aspects of the form of petition certificates, extensive deadlines, and so on.

We therefore held from early on that the provision was self-executing:

The section of the Constitution under consideration is not a mere
statement of principles. On the contrary, it points out in detail the various
steps to be taken in referring an act of the legislature to the electors, and
undoubtedly intends that the conduct of the election and the canvass and
return of votes shall be in accordance with the general laws of the State. And
the legislature in its session of 1915 made certain amendments to the general
election laws, with the evident purpose of adapting them more fully to the
requirements of the referendum. Perhaps further action by the legislature
may be advisable in aid of the constitutional provision; but, so far as the
proceedings are under review in this case, the course to be taken is plainly
pointed out, and this court should not, at the present time, enter into a minute
investigation for the purpose of discovering whether there may not be
somewhere in the election laws an inapplicable provision, or a step not
clearly provided for. [Thompson v Secretary of State, 192 Mich 512, 520;
159 NW 65 (1916).]

When this constitutional material was reviewed at the most recent constitutional

convention, it was substantially slimmed down to the language we have now. As the

Address to the People said, the revision “eliminat[ed] much language of a purely statutory

character.” 2 Official Record, Constitutional Convention 1961, p 3367.

Matters of legislative detail contained in the present section of the
constitution are left to the legislature. The language makes it clear, however,
that this section is self-executing and the legislature cannot thwart the
popular will by refusing to act. [Id.]

It is in this context, then, that the Legislature was directed to “implement the provisions of

this section,” Const 1963, art 2, § 9. 8 The Legislature was empowered to, in effect,

8
Intervening defendant relies heavily on a case from another jurisdiction: Utah Safe to
Learn-Safe to Worship Coalition, Inc v State, 94 P3d 217; 2004 UT 32 (2004). However,
the use of the word “implement” in our state Constitution significantly distinguishes this
case from that one. In Utah, the state constitution “grant[ed] the right to initiative,” but

12
prescribe the sorts of details that had previously been written directly into the

Constitution—deadlines, type sizes, requirements of form, and so on. As the chair of the

committee on legislative powers reported to the convention when the language was first

considered:

The committee is of the opinion there is much within the existing
section 1 of a purely legislative character and therefore several exclusions
and changes are suggested.

* * *

Removed from constitutional status are the provisions on content and
time of filing petitions, canvassing of names on petitions, type sizes, and right
of the legislature to prescribe penalties. . . .

All of these matters are left to the legislature in the last sentence.
However, the language of the last sentence also makes it clear that the section
is self executing and the legislature cannot thwart popular will by refusing to
act. [2 Official Record, Constitutional Convention 1961, p 2392.]

It is clear that the original referendum and initiative provisions of the Constitution

of 1908 were self-executing, and the Constitution of 1963 maintained that self-executing

status. The significance of designating a constitutional provision as self-executing is that,

while implementing legislation is to some extent inevitable and necessary, the courts will

protect a self-executing provision from legislative encroachment. See Wolverine Golf Club

“simultaneously circumscribe[d] that right by granting the legislature leave to regulate” the
process. Id. at 226. The language and history of our Constitution shows that it is the people
circumscribing the Legislature, not vice versa. We also note that intervening defendant’s
citation of state constitutions that do include a cap comparable to the one in MCL 168.471
is of no moment. That the people of Massachusetts or Mississippi, see Mass Const 48, Gen
Prov, Pt 2 and Miss Const, art 15, § 273, ratified a particular geographic requirement for
signature gathering into their respective constitutions bears no relation to the way that the
citizens of this state crafted our citizen-initiative and referendum provisions.

13
v Secretary of State, 24 Mich App 711, 728-729; 180 NW2d 820 (1970), aff’d 384 Mich

461 (1971) (“[T]he term ‘self-executing’ . . . cloak[s] the [constitutional] provision with

the necessary characteristics to render its express provisions free from legislative

encroachment.”). “The only limitation, unless otherwise expressly indicated, on legislation

supplementary to self-executing constitutional provisions is that the right guaranteed shall

not be curtailed or any undue burdens placed thereon.” Hamilton, 227 Mich at 125 (opinion

of BIRD, J.) (quotation marks and citation omitted).

Our inquiry, therefore, must be concerned with whether a particular law constitutes

an “undue burden” on voters’ exercise of their direct-democracy rights. We have said that

the Legislature’s power to implement direct democracy is “a directive to the legislature to

formulate the process by which initiative petitioned legislation shall reach the legislature

or the electorate.” Wolverine Golf Club, 384 Mich at 466. The clearest examples of

legislation that the Legislature can adopt to “implement” direct democracy are the sorts of

requirements that were formerly provided in the Constitution of 1908—type size, the

timeline for circulating petitions, the duties of state officials in processing petitions that

have been submitted, and so on.

The 15% requirement in MCL 168.471, by contrast, does not merely fill in

necessary details, but rather adds a substantive obligation. Soutar v St Clair Co Election

Comm, 334 Mich 258, 265; 54 NW2d 425 (1952) (When “the language of the Constitution

is self-executing,” “[o]bligations other than those so imposed may not be added.”). The

primary reason for a constitutionally required signature floor for referendums and

initiatives is to establish a minimum level of support for a proposed change in the law

before it is presented to the Legislature or to all Michigan voters for approval or rejection.

14
By choosing a statewide minimum number of signatures without a geographic cap, the

people demonstrated their intent to allow a relatively small coalition of voters from a

concentrated geographic area to propose changes to the law, with the understanding that

such proposals will not become law without the approval of the Legislature or a majority

of the voters in a statewide election.

A cap on how many signatures can come from each part of the state undermines the

intentions described above in several regards. The cap would undoubtedly make it more

difficult and expensive to gather the required number of signatures within the time frame

required by Michigan’s election laws because signature gatherers would need to cover a

larger swath of terrain. Perhaps more importantly, the related enforcement provisions

would effectively ensure that some voters’ signatures would be rendered void merely

because they were obtained after the 15% cap for that district had been reached, something

that voters are unlikely to know when signing a petition. Such disenfranchisement would

weigh most heavily on those residing in more densely populated parts of the state, and it

would run directly contrary to the clear intention that nothing more than a minimum

number of signatures from the statewide population is necessary to propose changes to

Michigan’s laws. As the Court of Appeals noted, this will “have the effect of reducing the

‘total quantum of speech,’ ” League of Women Voters, ___ Mich App at ___; slip op at 12,

during an election cycle rather than increasing it and would restrict the powers of direct

democracy that the people reserved to themselves.

Indeed, we concluded in Wolverine Golf Club that even a statutory recitation of the

requirements of the Constitution of 1908 is not proper if on-the-ground conditions do not

demand it. Under the Constitution of 1908, initiative proposals needed to be submitted at

15
least 10 days before the beginning of the legislative session at which they were to be

considered. The Michigan Election Law was drafted to track this constitutional

requirement. MCL 168.472. But circumstances have changed since 1908—the Legislature

has gone from meeting for a few months every other year to nearly the entire calendar year

of every year. As a result, the purpose of requiring that initiative proposals be submitted

before the legislative session begins has been obviated, and when the Constitution of 1963

removed the 10-day requirement, we struck down MCL 168.472 as unconstitutional,

because it “restricts the utilization of the initiative petition and lacks any current reason for

so doing.” Wolverine Golf Club, 384 Mich at 466. 9

In sum, we conclude that the 15% geographic-distribution requirement does not

merely “implement the provisions” of Const 1963, art 2, § 9. The constitutional text does

not provide for such a requirement, and the Legislature’s power (and duty) to prescribe

details in furtherance of the constitutional processes does not extend this far. The 15%

requirement goes beyond “formulat[ing] the process by which initiative petitioned

legislation shall reach the legislature or the electorate,” Wolverine Golf Club, 384 Mich at

466, and instead imposes an additional substantive requirement that does not advance any

of the express constitutional requirements.

9
Intervening defendant suggests that MCL 168.471 constitutes wise public policy seeking
to ensure that gathered signatures are “more evenly spread” across the state’s congressional
districts. We agree with the Court of Appeals’ explanation for why this justification is
suspect and further note that matters of public policy are of no concern to this Court when
reviewing the constitutionality of the provision at issue. The 15% cap is an additional
requirement not included in our Constitution that imposes “an obligation that restricts,
rather than furthers, the initiative process.” League of Women Voters, ___ Mich App at
___; slip op at 11, citing Soutar, 334 Mich at 265. Thus, it is unconstitutional.

16
 B. CONSTITUTIONAL AMENDMENTS

As noted, the 15% geographic-distribution requirement applies to all three forms of

direct democracy in Michigan. We have addressed the two provided for in Const 1963, art

2, § 9 and determined that the geographic limit is unconstitutional. However, the third

form of direct democracy—initiated constitutional amendments—is provided for in Const

1963, art 12, § 2, and the constitutional language is somewhat different. Intervening

defendant argues that even if the 15% requirement as applied to Const 1963, art 2, § 9 is

invalid, the language of Const 1963, art 12, § 2 allows for it to be validly applied as to

petitions for constitutional amendments. Specifically, Const 1963, art 12, § 2 provides that

a petition to initiate an amendment “shall be in the form, and shall be signed and circulated

in such manner, as prescribed by law.” We now consider whether this specific authority

to prescribe the manner of signing and circulating the petition gives the Legislature broader

authority to regulate initiated constitutional amendments.

Much as with Article 2, § 9, the text of Article 12, § 2 indicates that the Legislature

lacks this power. Critically, the text says that the Legislature may “prescribe[] by law” the

form of petitions and the manner in which they are to be signed and circulated. As already

noted, we held in Beech Grove Inv Co, 380 Mich at 419, that this “prescribed by law”

phrasing was used “[w]here only the details were left to the legislature and not the over-all

planning . . . .” It is more than a mere “detail,” or a matter of “functional detail” as Justice

ZAHRA suggests in his partial dissent, to require that signatures be distributed

geographically in addition to requiring an overall raw total. As we explained in Part III(A),

the 15% geographic-distribution requirement does not merely restrict where signatures of

Michigan citizens may be collected; it limits from whom they may be gathered. Thus, it is

17
a limitation of a substantive right that his partial dissent significantly undervalues by

dismissing it as a mere “procedural” requirement akin to the timing requirements at issue

in Consumers Power Co v Attorney General, 426 Mich 1; 392 NW2d 513 (1986). 10

That said, as with the legislative initiative and the referendum, the Legislature no

doubt has some power to prescribe laws not specified in the four corners of the

constitutional text. The evolution of Michigan’s constitutional text provides a clue to the

proper scope of the Legislature’s authority. The first direct-democracy process in

Michigan law was established in the Michigan Constitution of 1908, which allowed

citizens to initiate constitutional amendments. Const 1908, art 17, § 2. The 1908

10
While it is true that a majority of this Court has previously referred to the 10% minimum
signature requirement as a “procedural requirement of obtaining a certain number of
signatures,” Citizens Protecting Michigan’s Constitution, 503 Mich at 73, Justice ZAHRA’s
partial dissent takes this statement out of context. First, the issue before the Court in
Citizens Protecting Michigan’s Constitution was whether a particular proposal to modify
our Constitution was an “amendment” under Article 12, § 2, or a “general revision” under
Article 12, § 3. That is not at issue in this case. Secondly, the 10% minimum signature
language was part of an introduction to the Court’s review of debates at the constitutional
convention in 1961–1962 about whether to change or eliminate a minimum signature
requirement. The discussion provided context for analyzing the substantive limitations to
amendments proposed under Article 12, § 2. See Citizens Protecting Michigan’s
Constitution, 503 Mich at 73-75. Far from treating this as a mere procedural limitation,
the majority concluded that

the convention decided to keep voter-initiated amendments difficult because
amendments, like the Constitution itself, were intended to deal with serious
matters. The convention accomplished its goal by imposing what it viewed
as the clearest and most stringent limitation on initiative amendments: a
signature requirement. [Id. at 75 (second emphasis added).]

The debate in Citizens Protecting Michigan’s Constitution was about the initial threshold
required for a constitutional amendment to be placed on the ballot in the first place. As
discussed in more detail below, the 1963 drafters settled on a stringent signature
requirement. But at no time has this Court ever determined that a geographic limitation or
cap on that signature-gathering requirement was proper in any context.

18
Constitution contained some 500 words detailing the required content of petition forms,

the duties of the Secretary of State when petitions were submitted, the election processes

to use in submitting proposals to the electorate, and so forth. Id. And this constitutional

section was further expanded by amendment in 1913. Given this extensive level of detail,

we held that

[t]he constitutional provision contains procedural rules, regulations, and
limitations; it maps the course and marks the way for the accomplishment of
an end; it summons no legislative aid and will brook no elimination or
restriction of its requirements; it grants rights on conditions expressed, and
if its provisions are complied with and its procedure followed its mandate
must be obeyed. Its provisions are . . . self-executing. [Hamilton, 221 Mich
at 544 (emphasis added).]

It was in light of this background that the convention drafted Const 1963, art 12,

§ 2. Much as with initiatives and referendums, the convention affirmatively introduced a

role for the Legislature to play—prescribing by law the form of initiative petitions and how

the petitions must be signed and circulated. But as with initiatives and referendums, it used

the “prescribed by law” phrasing, signifying that “only the details were left to the

legislature”—which is exactly what the Address to the People said. See 2 Official Record,

Constitutional Convention 1961, p 3407 (“Details as to form of petitions, their circulation

and other elections procedures are left to the determination of the legislature.”). As the

committee report on the proposed language noted at the convention, the proposal had “the

aim of eliminating matters which we were convinced were statutory detail . . . .” Id. at

2459. A committee member noted:

The proposal that has just been read by the secretary eliminates a great
deal of material that was previously in the constitution. We have tried to
include the bare skeleton of the provision in order to still keep it self
executing without providing all the varied material as to how names are to

19
 be set forth and all of this type of thing which is presently provided for in the
statutes of this state. It was the opinion of the committee that in the event the
legislature refused to act to provide the things that are called for here by this
constitution that in one way or another it would still be possible to get an
amendment on the ballot with the amount of material which is still left, which
is still greatly statutory in nature. But since this is a provision in derogation
of the power of the legislature, so to speak, it seemed desirable that it be self
executing in nature, and that is why there is still a great deal of material here
but far less than there was before. [Id. at 2460.]

Indeed, the most extensive debate at the convention about this section was over the

number of signatures to require. Some delegates wanted an absolute cap on the number of

signatures necessary to put a measure on the ballot, while others preferred the 10%

threshold, meaning the qualifying figure would roughly increase with the state’s

population. Obviously, the sentiment in favor of an absolute cap did not prevail, see id. at

3199, but the discussion suggests that the delegates only contemplated the raw number of

signatures as the hurdle that amendment proponents would need to overcome. Indeed, on

two separate occasions, the convention voted down geographic-distribution requirements,

rejecting proposals to cap the share of signatures that any given county could contribute at

10%, id. at 2465-2469, and 25%, id. at 3200-3201. The criticism of these proposals was

not grounded in a sense that this was a detail that should be left to the Legislature, but

rather that they were substantively unacceptable. 11 It is hard to read this action by the

11
See, e.g., id. at 2466 (“I believe this is another one of those amendments that is based on
the theory that all people are equal, except that people in Wayne county are less equal than
other people.”); id. at 2468 (“[Y]ou could conceivably have a minority, an extreme
minority of the population in a large percentage of the thinly populated counties who would
have complete and absolute control over whether or not there would be any constitutional
amendments submitted to the people.”); id. at 2469 (“[W]e have seen various
attempts . . . to gerrymander this state but now we seem to be getting some ‘garrybrowning’
[a reference to the sponsor of the proposal, Garry Brown] in terms of the question of
petitions, and I think that gerrymandering or ‘garrybrowning’ is all the same; that it is

20
convention and conclude that a geographic-distribution requirement was contemplated as

the sort of detail the Legislature was empowered to fill in. 12

In sum, we have always understood the section on citizen-initiated constitutional

amendments to be self-executing—meaning the Legislature is constrained from

encroaching upon it just as the Legislature is constrained from encroaching upon the

statutory initiative and referendum. “Of the right of qualified voters of the State to propose

amendments to the Constitution by petition it may be said, generally, that it can be

interfered with neither by the legislature, the courts, nor the officers charged with any duty

in the premises.” Scott v Secretary of State, 202 Mich 629, 643; 168 NW 709 (1918). We

have confirmed that the alterations made in the 1963 Constitution did not change the self-

executing character of this section. See Ferency v Secretary of State, 409 Mich 569, 591

unjust and, therefore, we should defeat the [proposal].”); id. at 3200 (“It seems to me that
what you are saying here is that those of us who live in Wayne county, merely because we
do compose a third of the state’s population, are not going to count as heavily as those in
the rest of the state. I think this is a distinctly unfair and prejudiced position to take and I
object to it strongly.”); id. (“The one place where we thought we had one man, one vote
being equal was on getting petitions but now even this is being taken away by limiting the
number of petitions we can get. I think this is highly unreasonable and strenuously urge a
no vote to the amendment . . . .”). One delegate did note that “we are again trying to get
into a statutory position when we should not” and “urge[d] [the convention] strongly not
to get into this particular trap of putting this type of statutory language in the constitution,”
but he also suggested that he disagreed with the concept of a geographic-distribution
requirement because “[w]hen the proper signatures have been obtained, all of the people
of the state will have an opportunity, if they wish, to exercise their franchise.” Id.
12
Intervening defendant urges the Court to ignore the historical record. While we
recognize that the language of the Constitution controls, the constitutional convention
record may be particularly helpful and illuminating “when we find in the debates a
recurring thread of explanation binding together the whole of a constitutional concept.”
House Speaker v Governor, 443 Mich 560, 581; 506 NW2d 190 (1993) (quotation marks
and citation omitted).

21
n 9; 297 NW2d 544 (1980) (“While some of the legislation-like procedural detail was

eliminated in the 1963 Constitution, much was retained with the express purpose of

preserving the self-executing character.”); Citizens Protecting Michigan’s Constitution,

503 Mich at 63. While the 1963 Constitution did add a role for the Legislature to play, the

constitutional text and convention debates point to a limited role for the Legislature, and

we have said that “the principle that the Legislature may not unduly burden the self-

executing constitutional procedure applies equally to both” initiated legislation and

initiated constitutional amendments. Ferency, 409 Mich at 591 n 10.

We have already held that this geographic-distribution requirement is an undue

burden on the exercise of the legislative initiative (and referendum) power, and we likewise

conclude that it is an undue burden on the constitutional-amendment initiative power. We

have suggested that the clearest examples of requirements that the Legislature can provide

by statute under Const 1963, art 12, § 2 are of the sort that Const 1908, art 17, § 2 formerly

provided directly in the Constitution itself. See Citizens for Capital Punishment v

Secretary of State, 414 Mich 913, 914 (1982) (“The relevant provisions of the 1908

Constitution described in detail the form and manner for the signing of petitions. . . . [T]he

Legislature has provided those details [by statute] as contemplated by art 12, § 2 of the

1963 Constitution.”). It is true that the current constitutional language

“summons . . . legislative aid,” Hamilton, 221 Mich at 544, in a way that the Constitution

of 1908 did not, and it was on this basis that we upheld, in Consumers Power Co, the

constitutionality of 1973 PA 112, which established a rebuttable presumption that petition

signatures older than 180 days had been given by someone no longer registered to vote in

Michigan. But in that case, we noted that “[t]he purpose of the statute is to fulfill the

22
constitutional directive of art 12, § 2 that only the registered electors of this state may

propose a constitutional amendment.” Consumers Power Co, 426 Mich at 8. No such

constitutional directive is at issue here. 13

Much as with the legislative initiative, then, we hold that the 15% geographic-

distribution requirement goes beyond the Legislature’s power under Const 1963, art 12,

§ 2 to prescribe the “form” of petitions and the “manner” of their signing and circulation.

The constitutional text does not provide for such a distribution requirement, and in fact,

similar concepts were expressly rejected at the convention. As with legislative initiatives,

the constitutional requirement for a minimum number of signatures ensures a threshold

level of support before a proposed change to our state’s Constitution can be submitted to

all voters for approval or rejection. The 15% requirement exceeds the Legislature’s

authority to regulate a self-executing constitutional process by imposing a substantive

requirement that does not advance the express constitutional requirement—unlike the law

at issue in Consumers Power Co. 14 It does not align with any of the aspects of statutory

13
Given the language and history of this provision of our Constitution and its function as
a limitation on the Legislature, we conclude that our constitutional-amendment process is
also distinguishable from that in Utah Safe to Learn.
14
As specifically noted in Consumers Power Co, 426 Mich at 8, the statute at issue “does
not set a 180-day time limit for obtaining signatures,” and—prior to a more recent
amendment of the statute, MCL 168.471a,—“the presumption [that older signatures are
invalid could] be rebutted.” Conversely, the enforcement mechanisms for the geographic-
distribution requirement at issue irrefutably invalidate signatures collected beyond the
15% cap, making it a more onerous burden on the people’s right to propose amendments
to the state Constitution by obtaining support from at least 10% of the requisite voting
population. We reject Justice ZAHRA’s position that these mechanisms are similar because
it will “almost always be the case that some signatures affixed to a petition will be
invalidated in one way or another.” There is a significant difference between signatures
no longer being needed because the committee’s ballot drive has successfully obtained the

23
detail that were in the Constitution of 1908 and removed when the Constitution of 1963

was proposed and ratified. Rather, it aligns with proposals that the convention specifically

rejected, apparently, in no small part out of concerns that such requirements would reduce

or enhance the political power of Michiganders on the basis of the location of their

residence. We agree with Justice ZAHRA that Article 12, § 2 requires a proposed

constitutional amendment to be supported by “registered electors of the state equal in

number to at least 10 percent of the total vote cast” for governor in the last election. But

the Constitution requires nothing more than this minimum level of support from the

electorate as a whole, and it does not require that such support be evenly distributed in

geographic terms. Accordingly, we conclude that it is an undue burden on the

Constitution’s self-executing voter-initiated constitutional-amendment process. 15

IV. DISCLOSURE REQUIREMENTS

Although plaintiffs’ challenge to the 15% cap is grounded in the Michigan

Constitution, its challenge to the disclosure requirements in 2018 PA 608 is grounded in

federal constitutional law. While “[t]he Michigan Constitution has been interpreted as

affording broader protection of some individual rights also guaranteed by the federal

constitution’s Bill of Rights,” it “has never been so interpreted in the free expression and

minimum threshold of signatures and signatures being completely disregarded because of
the 15% cap. Electors who sign petitions in excess of the minimum threshold will still
have their voices heard by virtue of the ballot issue being placed before the voters; whereas
electors whose signatures are not counted because of the 15% cap may not ultimately see
their issues placed before the people. The latter are silenced by the 15% requirement in a
way that the former are not.
15
Because the 15% requirement in MCL 168.471 is unconstitutional, we agree with the
Court of Appeals that its accompanying provisions, MCL 168.477 and MCL 168.482(4),
are also unconstitutional.

24
petition context.” Id. at 202. Accordingly, while the “individual right to solicit signatures

to qualify an initiative petition is protected by the rights of free expression, assembly, and

petition, guaranteed in [Const 1963, art 1, §§ 3 and 5],” Woodland, 423 Mich at 215, we

look to federal jurisprudence to analyze the disclosure requirements.

Petition circulation is protected by the First Amendment because it is “core political

speech” that “involves both the expression of a desire for political change and a discussion

of the merits of the proposed change.” Meyer v Grant, 486 US 414, 421-422; 108 S Ct

1886; 100 L Ed 2d 425 (1988). Laws that directly or severely burden political speech are

generally subject to strict or exacting judicial scrutiny. Citizens United v Fed Election

Comm, 558 US 310, 340; 130 S Ct 876; 175 L Ed 2d 753 (2010). A plurality of the United

States Supreme Court recently clarified that “compelled disclosure requirements are

reviewed under exacting scrutiny.” Americans for Prosperity Foundation v Bonta, 594 US

___, ___; 141 S Ct 2373, 2383; 210 L Ed 2d 716 (2021) (opinion of Roberts, C.J.); but see

id. at ___; 141 S Ct at 2390 (Thomas, J., concurring in part and concurring in the judgment)

(“Laws directly burdening the right to associate anonymously, including compelled

disclosure laws, should be subject to the same scrutiny as laws directly burdening other

First Amendment rights.”); id. at ___; 141 S Ct at 2393 (Alito, J., concurring in part and

concurring in the judgment) (“I am not prepared at this time to hold that a single standard

applies to all disclosure requirements” because there was “no need to decide which

standard should be applied here or whether the same level of scrutiny should apply in all

cases in which the compelled disclosure of associations is challenged under the First

Amendment.”). Exacting scrutiny requires “a ‘relevant correlation’ or ‘substantial

relation’ between the governmental interest and the information required to be disclosed.”

25
Buckley v Valeo, 424 US 1, 64; 96 S Ct 612; 46 L Ed 2d 659 (1976) (citations omitted).

That governmental interest must also be “sufficiently important” and “must reflect the

seriousness of the actual burden on First Amendment rights.” Doe v Reed, 561 US 186,

196; 130 S Ct 2811; 177 L Ed 2d 493 (2010) (quotation marks and citations omitted).

However, states maintain “considerable leeway to protect the integrity and

reliability of the initiative process, as they have with respect to election processes

generally.” Buckley v American Constitutional Law Foundation, 525 US 182, 191; 119 S

Ct 636; 142 L Ed 2d 599 (1999). And there is a difference between regulations that directly

affect core political speech and those that simply regulate the “mechanics of the electoral

process.” McIntyre v Ohio Elections Comm, 514 US 334, 345; 115 S Ct 1511; 131 L Ed

2d 426 (1995); see also Buckley, 525 US at 207-208 (Thomas, J., concurring). “No bright

line separates permissible election-related regulation from unconstitutional infringements

on First Amendment freedoms,” Timmons v Twin Cities Area New Party, 520 US 351, 359;

117 S Ct 1364; 137 L Ed 2d 589 (1997), as even structural decisions may affect an

individual’s right to speak and associate with others for political ends, Anderson v

Celebrezze, 460 US 780, 788; 103 S Ct 1564; 75 L Ed 2d 547 (1983). The Supreme Court

has left the determination of the appropriate level of scrutiny open-ended, noting that there

is “no substitute for the hard judgments that must be made.” Buckley, 525 US at 192

(quotation marks and citation omitted). Where there is no severe burden on political speech

or the regulation is merely one of mechanical electoral processes, courts generally apply a

more flexible review such as the Anderson-Burdick test. 16 This test requires a reviewing

16
See Anderson, 460 US 780; Burdick v Takushi, 504 US 428; 112 S Ct 2059; 119 L Ed
2d 245 (1992).

26
court to “weigh the character and magnitude of the burden the State’s rule imposes on those

rights against the interests the State contends justify that burden, and consider the extent to

which the State’s concerns make the burden necessary.” Timmons, 520 US at 358

(quotation marks and citations omitted).

A. THE CHECKBOX REQUIREMENT

Pursuant to 2018 PA 608, MCL 168.482(7) now requires a petition that proposes a

constitutional amendment, initiation of legislation, or referendum on legislation to include

“at the top of the page check boxes and statements printed in 12-point type to clearly

indicate whether the circulator of the petition is a paid signature gatherer or a volunteer

signature gatherer.” Any signature obtained on a petition that does not comply with the

checkbox requirement in MCL 168.482(7) “is invalid and will not be counted.” MCL

168.482(8). If this requirement imposes an impermissible burden on political speech, then

it must be struck down.

Somewhat instructive is the Supreme Court’s decision in Buckley, 525 US 182. In

that case, the plaintiffs challenged the constitutionality of a Colorado statute that required

petition circulators to “wear identification badges stating their names, their status as

‘VOLUNTEER’ or ‘PAID,’ and if the latter, the name and telephone number of their

employer[.]” Id. at 188. The federal district court struck the provision, and the United

States Court of Appeals for the Tenth Circuit affirmed, explaining that “the badge

requirement force[d] circulators to reveal their identities at the same time they deliver their

political message,” at a time “when the reaction to their message may be the most intense,

emotional, and unreasoned.” American Constitutional Law Foundation, Inc v Meyer, 120

27
F3d 1092, 1102 (CA 10, 1997), aff’d sub nom Buckley, 525 US 182 (1999). 17 The Supreme

Court echoed those concerns: “The injury to speech is heightened for the petition circulator

because the badge requirement compels personal name identification at the precise moment

when the circulator’s interest in anonymity is greatest.” Buckley, 525 US at 199 (emphasis

added).

Similar to the badge requirement in Buckley, the checkbox requirement in MCL

168.482(7) compels the petition circulator to disclose their status as paid or volunteer at

the same time the political message is being delivered. It is not a “step removed from the

communicative aspect of petitioning . . . .” Doe, 561 US at 213 (Sotomayor, J.,

concurring). Instead it is inextricably linked to the circulator’s delivery of “core political

speech” during a one-on-one “discussion of the merits of the proposed change.” Meyer,

486 US at 421. Accordingly, we conclude that it is subject to exacting-scrutiny review.

Therefore, we must determine whether it is substantially related to a sufficiently important

governmental interest, weighing the strength of the interest against the seriousness of the

burden on First Amendment rights. Doe, 561 US at 196 (opinion of the Court).

We begin with the alleged burden imposed by MCL 168.482(7). Plaintiffs contend

that it will discourage participation in the petition-circulation process just as the name-

badge requirement in Buckley did. Similarly, they also argue that the checkbox’s “size and

location will discourage signing by triggering hostility to a paid circulator.”

17
Because the identification requirement alone rendered the statute unconstitutional, the
circuit court expressed no opinion on the constitutionality of the additional requirement
that the badge disclose whether a circulator was paid or a volunteer. Id. at 1104.

28
 First, we reject the argument that the checkbox requirement is analogous to the name

badge at issue in Buckley. In that case, the Colorado law required petition circulators to

wear a badge disclosing their name and whether they were paid or volunteer (and if paid,

by whom). Buckley, 525 US at 197. The Court, however, confined its review to whether

requiring that circulators disclose their names was constitutional. Id. at 197, 200. In

holding that the requirement was unconstitutional, the Court noted its agreement with the

lower courts that the requirement “force[d] circulators to reveal their identities at the same

time they deliver their political message,” which is when “reaction to the circulator’s

message is immediate and may be the most intense, emotional, and unreasoned,” exposing

circulators to “ ‘heat of the moment’ harassment.” Id. at 198-199 (quotation marks and

citations omitted). These risks are not nearly as apparent here—circulators are not being

forced to reveal anything as personal as their identity or their employer, and they are

therefore not subject to the same sort of personalized heat-of-the-moment harassment.

Moreover, as the concurring judge below observed, plaintiffs provide no evidence to

support the proposition that a checkbox disclosure would hinder petition-gathering

companies from recruiting and retaining paid circulators. League of Women Voters, ___

Mich App at ___ (CAMERON, J., concurring).

Second, plaintiffs also argue that the checkbox’s “size and location will discourage

signing by triggering hostility to paid circulators.” As the Court of Appeals recognized,

while a circulator’s paid status may cause some persons not to engage with the circulator,

it might also provide an incentive for others to sign a petition to help someone who is just

“ ‘doing their job.’ ” Id. at ___; slip op at 19 (opinion of the Court). That being said, the

Court of Appeals noted that the checkbox could create administrative burdens for petition

29
sponsors by making them ensure that checkboxes are adequately marked in order for them

to avoid being thrown out under MCL 168.482a(4). In sum, for the sake of argument, we

will assume that the checkbox imposes some direct but minimal burden on core political

speech.

Even if only minimal burdens are at stake, pursuant to exacting-scrutiny review, the

state must still have an adequate interest in creating the checkbox requirement, and the

checkbox must bear a substantial relationship to that interest. Intervening defendant

contends that the checkbox provides “valuable information to the electors,” explaining:

It is no secret that financial incentives can alter individuals’ priorities—it
may well be that electors see a volunteer as more committed to the cause they
are circulating petitions for, and thus more worthy of their support. On the
other hand, an elector could view a paid circulator as evidence that the
petition drive is well-funded, more likely to succeed, and therefore more
worthy of support. Either way, the result is more relevant information for
the electorate.

Similarly, the Court of Appeals concluded that “transparency in the political process,

especially transparency that permits voters to ‘follow the money,’ is a compelling state

interest.” League of Women Voters, ___ Mich App at ___; slip op at 18. The United States

Supreme Court has repeatedly held that increasing the amount of information available to

the voters is a legitimate state interest. See, e.g., McConnell v Federal Election Comm,

540 US 93, 196; 124 S Ct 619; 157 L Ed 2d 491 (2003) (holding that “providing the

electorate with information” is an “important state interest[]”), overruled on other grounds

by Citizens United, 558 US at 365; Eu v San Francisco Co Democratic Central Comm,

489 US 214, 228; 109 S Ct 1013; 103 L Ed 2d 271 (1989) (“Certainly the State has a

legitimate interest in fostering an informed electorate.”); Buckley, 424 US at 76 (upholding

30
a law whose purpose was “to insure that the voters are fully informed”). “In the election

context, disclosure requirements serve the important function of transparency . . . .”

Libertarian Party of Ohio v Husted, 751 F3d 403, 413 (CA 6, 2014); see also Citizens in

Charge v Gale, 810 F Supp 2d 916, 928 (D Neb, 2011) (upholding a provision similar to

MCL 168.482(7)).

In other contexts, such as whether personal or demographic information is disclosed,

it might be a closer call on whether a general interest in providing “information” to the

electorate can survive exacting scrutiny. As Justice Alito has observed:

Were we to accept respondents’ asserted informational interest, the State
would be free to require petition signers to disclose all kinds of demographic
information, including the signer’s race, religion, political affiliation, sexual
orientation, ethnic background, and interest-group memberships. [Doe, 561
US at 207 (Alito, J., concurring).]

“The simple interest in providing voters with additional relevant information does not

justify a state requirement that a writer make . . . disclosures she would otherwise omit.”

McIntyre, 514 US at 348. However, given the limited nature of the disclosure, we conclude

that the “seriousness of the actual burden” on First Amendment rights caused by the

checkbox requirement is so minimal that a governmental interest in increasing information

for voters justifies the requirement. Doe, 561 US at 196 (opinion of the Court) (quotation

marks and citations omitted). Therefore, we conclude that MCL 168.482(7) survives

exacting scrutiny, and we affirm the Court of Appeals.

B. THE AFFIDAVIT REQUIREMENT

An entirely new provision to the Michigan Election Law was added by 2018 PA

608: MCL 168.482a. This section requires a paid signature gatherer, “before circulating

31
any petition, [to] file a signed affidavit with the secretary of state that indicates he or she

is a paid signature gatherer.” MCL 168.482a(1) (emphasis added). Any signature obtained

on a petition by a paid circulator who has not filed the precirculation affidavit “is invalid

and must not be counted.” MCL 168.482a(2). Unlike the checkbox requirement, the

affidavit requirement is a prerequisite to circulation of a petition that is a “step removed

from the communicative aspects of petitioning.” Doe, 561 US at 213 (Sotomayor, J.,

concurring) (quotation marks and citation omitted). As it does not directly burden core

political speech, we apply the flexible Anderson-Burdick test. However, even under this

more relaxed standard, the affidavit requirement of MCL 168.482a does not pass

constitutional muster, and therefore we affirm the judgment of the Court of Appeals despite

rejecting some of its analysis.

First we consider the character and magnitude of the burden imposed. Timmons,

520 US at 358. We note that that the affidavit required by MCL 168.482a is only required

to be filed by paid signature gatherers and thus imposes additional hurdles on causes

furthered by groups who might rely on professional petition circulators. See Riley v Nat’l

Federation of the Blind of NC, Inc, 487 US 781, 799; 108 S Ct 2667; 101 L Ed 2d 669

(1988). Additionally, because signature gathering often operates on fairly short time

lines, 18 the requirement that a new employee file an affidavit before they may begin

gathering signatures is likely to hinder the ability to obtain the requisite number of

signatures. Paid circulators are often seasonal or temporary employees, and there is no

18
For example, the signature on a petition for a constitutional amendment or initiative for
legislation will not be counted if it was made more than 180 days before the petition is filed
with the Secretary of State. MCL 168.472a.

32
licensing requirement for this line of work. Circulators also might not reside in the county

(or even the state) where signatures are being collected. Completing an affidavit in front

of a local notary and then filing it with the Secretary of State before a single signature can

be gathered is a substantial burden and precondition on one’s ability to engage in political

speech. We therefore reject intervening defendant’s suggestion that the affidavit is merely

a “minor administrative burden.”

Second, we must consider the interests that the state contends justifies these

burdens. Timmons, 520 US at 358. According to intervening defendant, the precirculation

affidavit promotes greater transparency and it also ensures ballot integrity by allowing the

state to locate paid circulators and verify campaign-finance reporting. Like the Court of

Appeals, however, we struggle to understand how the affidavit serves any asserted state

interest.

To begin, unlike the checkbox requirement, the affidavit requirement of MCL

168.482 does nothing to increase transparency or provide information to the potential

signer of a petition. It is not immediately clear that the information in the affidavit would

be available to the public, 19 and it would not need to be, as the checkbox requirement would

readily inform the elector of the circulator’s status as paid or volunteer. Also, ballot-

question petitions must include information about the proponent of the petition drive under

the Michigan Campaign Finance Act, MCL 169.201 et seq. (stating in MCL 169.247(1)

that “printed matter having reference to . . . a ballot question, shall bear upon it an

identification that contains the name and address of the person paying for the matter”). By

19
It is unclear whether the Secretary of State would, or legally could, make these affidavits
available to the general public.

33
including this information along with the checkbox, petitions already adequately inform

voters about what special-interest groups are involved in gathering signatures for a

particular cause.

“Disclosure of the names of initiative sponsors, and of the amounts they have spent

gathering support for their initiatives,” is a state interest that we are willing to recognize

because it provides the electorate with information and exposes special interests who may

bankroll a petition drive. Buckley, 525 US at 202-203. However, as the Court of Appeals

recognized, the Michigan Campaign Finance Act already requires ballot-question

committees to report the names, addresses, and amounts contributed by financial

supporters, as well as whether they hire a firm that employs paid circulators. See MCL

169.226(1)(b) through (j); MCL 169.206. Intervening defendant suggests that the affidavit

requirement is rationally related to the state’s interest in “verifying campaign-finance

reporting”; however, we need not decide whether this justification could survive scrutiny

because the MCL 168.482a affidavit does not require the circulator to disclose whom they

are employed or paid by. 20

20
To the extent that intervening defendant suggests that the Legislature could have
rationally determined that paid circulators are more likely to commit fraud as a result of
financial incentive to produce signatures, we note that the United States Supreme Court
has stated:

[A]bsent evidence to the contrary, we are not prepared to assume that a
professional circulator—whose qualifications for similar future assignments
may well depend on a reputation for competence and integrity—is any more
likely to accept false signatures than a volunteer who is motivated entirely
by an interest in having the proposition placed on the ballot. [Buckley, 525
US at 203-204 (quotation marks and citation omitted).]

34
 More importantly, just as the precirculation timing of the affidavit requirement

works to impose a burden on First Amendment-related activity, its timing also serves to

demonstrate the lack of state interest. Intervening defendant suggests that the affidavit

allows the state to locate paid circulators, but the MCL 168.482a affidavit requires only the

circulator’s signature and the attestation that they are a paid signature gatherer. It does not

require any identifying information. Conversely, completed petitions are required to

provide the circulator’s signature, name, and residence as part of the certificate of circulator

located at the bottom of a petition sheet, MCL 168.544c(1), and a checkbox indicating

whether the circulator was paid for their service. MCL 168.482(7). 21 This raises questions

about how the affidavit would assist the Secretary of State in locating paid circulators. It

is not apparent why the state would need to, as suggested by intervening defendant, locate

paid circulators before signatures are even gathered and turned in or how the state would

be able to do so. The timing of the affidavit also distinguishes this case from the affidavit

mentioned in Buckley, 525 US at191 n 10, and upheld in Husted, 751 F3d at 413, both of

which required affidavits to be filed at the time the petition was submitted.

In sum, we conclude that the state’s asserted concerns do not make the burdens

added by the precirculation affidavit requirement necessary or appropriate. Timmons, 520

US at 358. In other words, when weighing the burdens on First Amendment rights with

the state’s asserted interest, we conclude that the burden eclipses the nominal interest. The

Additionally, “the risk of fraud or corruption, or the appearance thereof, is more remote at
the petition stage of an initiative than at the time of balloting.” Id. at 203 (quotation marks
and citation omitted).
21
MCL 168.482a(3) and MCL 168.482e also provide separate penalties for when any
circulator provides a false address or false information on the certificate of circulator.

35
affidavit does nothing to inform the general electorate about who may be funding petition

drives, the amount of money spent, or whether a circulator is paid or volunteer, nor does it

appear to serve the state’s interest in verifying campaign-finance disclosures. However, it

does have the potential to decrease the pool of potential circulators and hinder petition

drives that employ paid signature gatherers. For these reasons, we affirm the Court of

Appeals’ holding that MCL 168.482a is unconstitutional. 22

V. PROSPECTIVE APPLICATION

It is well established that “the general rule is that judicial decisions are to be given

complete retroactive effect.” Hyde v Univ of Mich Bd of Regents, 426 Mich 223, 240; 393

NW2d 847 (1986). “However, where injustice might result from full retroactivity, this

Court has adopted a more flexible approach, giving holdings limited retroactive or

prospective effect.” Lindsey v Harper Hosp, 455 Mich 56, 68; 564 NW2d 861 (1997).

The Secretary of State urges this Court to give our decision prospective effect. In

her application, she explains that the 15% requirement has never been enforced in reliance

on OAG, 2019-2020, 7,310 and subsequent court opinions. The affidavit requirement has

only been enforced since July 12, 2021, when the Court of Claims upheld it as

constitutional. And the checkbox requirement was not subject to enforcement until the

Court of Appeals upheld it as constitutional on October 29, 2021. Because we agree with

the lower courts as to the unconstitutionality of MCL 168.471 and MCL 168.482a, we need

not depart from our standard practice of providing retroactive effect regarding those two

22
Accordingly, accompanying provision MCL 168.482c (making it a misdemeanor to
provide a false statement on the affidavit) is no longer valid.

36
provisions. Instead our focus is on whether enforcement of the checkbox’s

constitutionality should be afforded prospective-only effect as, according to the Secretary

of State, two petitions currently in circulation do not contain the checkbox requirement of

MCL 168.482(7), while one petition that received preliminary approval does contain a

checkbox but has not yet begun circulation.

Our test for deciding retroactivity questions was most clearly laid out in Pohutski v

Allen Park, 465 Mich 675; 641 NW2d 219 (2002). In that case, we noted that there is a

“threshold question whether the decision clearly establishe[s] a new principle of law.” Id.

at 696, citing Riley v Northland Geriatric Ctr (After Remand), 431 Mich 632, 645-646; 433

NW2d 787 (1988) (opinion of GRIFFIN, J.). If a decision establishes a “new principle of

law,” we then consider three factors: “(1) the purpose to be served by the new rule, (2) the

extent of the reliance on the old rule, and (3) the effect of retroactivity on the administration

of justice.” Pohutski, 465 Mich at 696, citing People v Hampton, 384 Mich 669, 674; 187

NW2d 404 (1971).

The threshold question, then, is whether a decision amounts to a new rule of law.

“A rule of law is new for purposes of resolving the question of its retroactive

application . . . either when an established precedent is overruled or when an issue of first

impression is decided which was not adumbrated by any earlier appellate decision.”

People v Phillips, 416 Mich 63, 68; 330 NW2d 366 (1982). Here, while no precedent is

being overruled, this is an issue of first impression that has been subject to vigorous debate

essentially since 2018 PA 608 was enacted. We therefore conclude that the threshold

question has been satisfied.

37
 We then turn to the three-factor test. 23 We conclude that under the unique

circumstances presented in this case, which materially affect the right of the people to

exercise direct democracy, the test points toward prospective application of our decision

here. Generally speaking, the purpose served by the checkbox is to allow the public to be

more informed about the paid or volunteer status of circulators when solicited to sign a

petition. Giving our ruling prospective-only application, therefore, will deprive voters who

have already signed petitions without checkboxes of this information. In other words, the

voters who have signed those petitions will not be afforded the same amount of information

as those who sign petitions going forward. That said, the effect of this factor is somewhat

diminished by the fact that some voters who have signed petitions without a checkbox may

have done so knowing that the circulator was paid or would have signed the petition

regardless of the circulator’s status because they agreed with the petitions’ substance or

that it concerned a topic worthy of statewide consideration. Prospective application of our

decision allows signatures obtained on petitions without a checkbox—some of which may

nevertheless have been obtained on petitions with a checkbox—to remain valid.

Nonetheless, the remaining two factors outweigh the countervailing consideration

under the first factor. As to the extent of the reliance on the old rule, it is important to note

that until the Court of Appeals recently upheld the checkbox requirement every court that

had considered the question held that MCL 168.482(7) was unconstitutional. 24 Further,

23
Michigan derived its three-factor test from Linkletter v Walker, 381 US 618; 85 S Ct
1731; 14 L Ed 2d 601 (1965).
24
This includes two Court of Claims opinions and a Court of Appeals decision. League of
Women Voters v Secretary of State, 331 Mich App 156; 952 NW2d 491 (2020), vacated

38
the Board of State Canvassers, while not required to do so by statute, has long offered the

opportunity to ballot proposal committees to have their petitions preliminarily approved as

to form prior to circulation in order to prevent the late discovery of defects in those forms—

discoveries that, without preapproval, might not be detected until after circulation is

complete. Here, after the Court of Claims held that the checkbox requirement was

unconstitutional, under the law in effect at the time, the Board approved the petitions of

two ballot-question committees, amici curiae Unlock Michigan and Secure MI Vote, whose

petitions lack paid circulator checkboxes. The committees’ reliance on the “old rule” and

use of the now-defective petitions, therefore, was understandable given the Board’s

approval of their petitions. The committees did what they could to ensure compliance with

the law by seeking and obtaining preapproval of their petitions from the Board. 25 We reject

the argument that these committees assumed the risk that the form of their petitions without

checkboxes would be deficient when they began their petition drives before a final

resolution on the merits could be reached regarding the challenges in this litigation. Again,

506 Mich 561 (2020). The Attorney General also opined that the checkbox requirement
was unconstitutional; however, “[t]hose formal opinions . . . do not bind the courts,” and
“whether the formal opinions bind even other governmental agencies” is open to question.
League of Women Voters, 506 Mich at 597 & n 57, citing Danse Corp v Madison Hts, 466
Mich 175, 182 n 6; 644 NW2d 721 (2002).
25
In fact, after the Court of Appeals upheld the checkbox requirement as constitutional,
these committees sought the Board’s approval for petitions that included checkboxes. At
a meeting before the Board, the Director of Elections, on behalf of the Bureau of Elections,
recommended to the Board that it accept petition forms with or without checkboxes. The
Board, however, declined to approve these committees’ updated petition forms containing
checkboxes. Thus, these committees have been collecting signatures on petition forms
that, while lacking checkboxes, were nevertheless approved by the Board. These
committees should not be faulted for relying on that approval, which, at the time it was
given, was compliant with then-existing law.

39
the people reserved the right to govern themselves through the process of direct democracy.

Michigan’s citizens should not be expected to sideline their rights to participate in direct

democracy while waiting for a final determination on the legality of contested legislation

through the oftentimes long and arduous process of litigation.

Moreover, we have an interest in avoiding a disruption in the administration of

justice. Were this decision applied retroactively to these parties, we would at minimum

face the near certainty of additional litigation to test its application to petition signatures

that have already been gathered. A petition’s failure to strictly comply with MCL

168.482(7) would arguably be grounds to throw it out. See Stand Up for Democracy v

Secretary of State, 492 Mich 588, 601-602; 822 NW2d 159 (2012). This would cause

serious confusion for voters who have already signed the petitions currently in circulation.

A person may sign a petition only once, MCL 168.482(5); therefore, invalidating signatures

already collected on checkbox-lacking petitions would make collecting them again

difficult, as signatories may refuse to engage with the circulator on the ground that they

had already signed the petition. This would infringe the electors’ rights to communicate

and speak by petition.

Under these unique circumstances, we conclude that our test for prospective-only

application has been satisfied. We hold that our decision will not apply to direct-

democracy signatures gathered before the effective date of this opinion. That does not

mean that the proponents may continue to circulate defective petitions—any signature

40
gathered after January 24, 2022, must be on a petition that conforms to the requirements of

MCL 168.482(7). 26

VI. CONCLUSION

We affirm the Court of Appeals’ opinion. As to the merits of 2018 PA 608, we hold

that its geographic-distribution requirement for direct-democracy signatures (the 15% cap)

violates the Michigan Constitution. We also hold that its precirculation affidavit

requirement for paid circulators is unconstitutional. The checkbox requirement, however,

passes constitutional muster. Finally, in light of the chaos and injustice that would ensue

were we to give this opinion retroactive effect, we hold that today’s decision must be given

prospective effect only.

Megan K. Cavanagh
Bridget M. McCormack
Richard H. Bernstein (except as to
Part IV(A))
Elizabeth T. Clement (except as to
Parts IV(B) and V)
Elizabeth M. Welch

26
The Court of Appeals’ October 29, 2021 opinion would generally be applied
retroactively, rendering all the petitions currently in circulation defective in relation to
MCL 168.482(7). However, a decision by the Court of Appeals does not ordinarily become
effective until “after the expiration of the time for filing an application for leave to appeal
to the Supreme Court, or, if such an application is filed, after the disposition of the case by
the Supreme Court[.]” MCR 7.215(F)(1)(a). Thus, by virtue of the applications for leave
to appeal filed, until today the decision in the Court of Appeals remained “pending.”
Grievance Administrator v Fieger, 476 Mich 231, 248; 719 NW2d 123 (2006). We need
not, therefore, determine whether it would be appropriate to provide the Court of Appeals’
opinion with prospective effect; we need only ensure that our decision today will be applied
prospectively.

41
 STATE OF MICHIGAN

SUPREME COURT

LEAGUE OF WOMEN VOTERS OF
MICHIGAN, PROGRESS MICHIGAN,
COALITION TO CLOSE LANSING
LOOPHOLES, and MICHIGANDERS FOR
FAIR AND TRANSPARENT ELECTIONS,
Plaintiffs-Appellees,

v Nos. 163711-2

SECRETARY OF STATE,
Defendant-Appellee,
and

ATTORNEY GENERAL,
Intervening Defendant-
Appellee.

LEAGUE OF WOMEN VOTERS OF
MICHIGAN, PROGRESS MICHIGAN,
COALITION TO CLOSE LANSING
LOOPHOLES, and MICHIGANDERS FOR
FAIR AND TRANSPARENT ELECTIONS,

Plaintiffs-Appellees,

v Nos. 163744-5

SECRETARY OF STATE,

Defendant-Appellee,
and

ATTORNEY GENERAL,

Intervening Defendant-
Appellant.
 LEAGUE OF WOMEN VOTERS OF
MICHIGAN, PROGRESS MICHIGAN,
COALITION TO CLOSE LANSING
LOOPHOLES, and MICHIGANDERS FOR
FAIR AND TRANSPARENT ELECTIONS,
Plaintiffs-Appellees,

v Nos. 163747-8

SECRETARY OF STATE,
Defendant-Appellant,
and
ATTORNEY GENERAL,
Intervening Defendant-
Appellee.

ZAHRA, J. (concurring in part and dissenting in part).
In 2018, the Legislature enacted into law 2018 PA 608, which imposed new

requirements for gathering signatures on petitions for statewide ballot proposals, including

initiatives, referendums, and constitutional amendments. Three aspects of 2018 PA 608

are at issue in this appeal: (1) a provision allowing no more than 15% of petition signatures

for a statewide ballot proposal to be obtained in any one congressional district

(“geographic-distribution requirement”), (2) a provision requiring that petitions include

checkboxes to indicate whether the petition circulator is a paid or volunteer signature

gatherer (“checkbox requirement”); and (3) a provision requiring that all paid signature

gatherers file an affidavit with the Secretary of State indicating that the person is a paid

signature gatherer before circulating any petitions (“affidavit requirement”). PA 608

provides that signatures collected in violation of any of these provisions are invalid and

may not be counted. Plaintiffs argue that these aspects of 2018 PA 608 are facially

2
unconstitutional. Specifically, plaintiffs argue that the geographic-distribution requirement

violates Const 1963, art 2, § 9 and Const 1963, art 12, § 2 as unduly burdening their direct-

democracy rights and that the checkbox and affidavit requirements violate their rights of

free speech, association, and petition under the United States and Michigan Constitutions. 1

The Department of the Attorney General, intervening on behalf of the Secretary of State,

disagrees, arguing that plaintiffs have failed to meet their heavy burden of establishing that

these provisions are facially unconstitutional.

Although I concur with the majority opinion that the geographic-distribution

requirement in 2018 PA 608 is unconstitutional as to initiatives and referendums under

Const 1963, art 2, § 9, I respectfully dissent from the majority’s conclusion that this

requirement is unconstitutional as to voter-initiated constitutional amendments under

Const 1963, art 12, § 2. Rather, the requirement that no more than 15% of signatures on

petitions seeking to amend the Michigan Constitution, the supreme law of this state, be

from any one congressional district is a matter of circulation left for the Legislature to

prescribe. My conclusion is consistent with the ratifiers’ common understanding of Const

1963, art 12, § 2—that voter-initiated constitutional amendments are reserved for

substantial matters worthy of constitutional elevation rather than routine policy matters

normally addressed through legislation and, therefore, it should be more difficult to amend

the Constitution than to propose, approve, or reject legislation—and does not conflict with

the self-executing nature of Article 12, § 2. Therefore, I conclude that the geographic-

1
Plaintiffs include entities that, according to their verified complaint, intend to seek
constitutional and statutory changes through our Constitution’s direct-democracy
provisions, as well as voters who will either support or oppose such changes. Thus, it
appears plaintiffs have standing to bring the present challenge.

3
distribution requirement is a valid legislative measure under Const 1963, art 12, § 2 and

that plaintiffs have not demonstrated that the requirement clearly conflicts with the text of

Article 12, § 2 so as to render it constitutionally infirm.

Further, while I concur with the majority opinion that the checkbox requirement is

constitutional, I dissent from the majority’s conclusion that the affidavit requirement is

facially unconstitutional. Like the checkbox requirement, the affidavit requirement

imposes minimal burdens on petition circulators, it does not bar paid petition circulation

altogether, and it does not chill or deter paid circulators from speaking. The affidavit

requirement also serves the important state interests of preserving the integrity of the

electoral process by detecting and deterring fraud, as well as assisting in the discovery of

invalid signatures. Therefore, I conclude that both the checkbox and affidavit requirements

pass constitutional muster under the exacting-scrutiny standard of review.

Finally, I concur with the majority’s decision to give our opinion prospective effect.

Relying on the imprimatur of the Board of State Canvassers and following the Court of

Claims’ opinion that the checkbox requirement was unconstitutional, two ballot-question

committees began circulating petitions without checkboxes. After the Court of Appeals

upheld the constitutionality of this requirement, these committees sought the Board’s

approval for updated petition forms that included checkboxes. The Director of Elections

recommended that the Board approve the committees’ updated petition forms. But the

Board, after urging from plaintiffs’ counsel, inexplicably declined to approve them. This

created an untenable situation for these committees, who have attempted to conform their

conduct to the law at every stage of this litigation. To now require their petitions to contain

checkboxes—a requirement that, at the time they initially obtained the Board’s approval,

4
was deemed by the judiciary to be unconstitutional—is exactly the type of unique and

limited circumstance warranting prospective-only application.

I. JUDICIAL REVIEW

The constitutionality of a statute is a question of law that this Court reviews de

novo. 2 “In analyzing constitutional challenges to statutes, this Court’s ‘authority to

invalidate laws is limited and must be predicated on a clearly apparent demonstration of

unconstitutionality.’ ” 3 “[T]he burden of proving that a statute is unconstitutional rests

with the party challenging it.” 4 Here, plaintiffs bring forth a facial challenge to various

provisions of 2018 PA 608. “A party challenging the facial constitutionality of a statute

faces an extremely rigorous standard, and must show that no set of circumstances exists

under which the act would be valid.” 5 “The fact that the act might operate

unconstitutionally under some conceivable set of circumstances is insufficient.” 6

“We require these challenges to meet such a high standard because ‘[s]tatutes are

presumed to be constitutional, and we have a duty to construe a statute as constitutional

2
Taylor v Gate Pharm, 468 Mich 1, 5; 658 NW2d 127 (2003).
3
People v Skinner, 502 Mich 89, 99; 917 NW2d 292 (2018), quoting People v Harris, 495
Mich 120, 134; 845 NW2d 477 (2014).
4
In re Request for Advisory Opinion Regarding Constitutionality of 2005 PA 71, 479 Mich
1, 11; 740 NW2d 444 (2007).
5
Id. (quotation marks, citations, and brackets omitted).
6
Council of Organizations & Others for Ed About Parochiaid, Inc v Governor, 455 Mich
557, 568; 566 NW2d 208 (1997) (quotation marks, citation, brackets, and ellipsis omitted).

5
unless its unconstitutionality is clearly apparent.’ ” 7 The “legislative power is the power

to make laws,” and the “judicial power” is the power to “interpret[] the law . . . .” 8 “In

accordance with the constitution’s separation of powers, this Court cannot revise, amend,

deconstruct, or ignore the Legislature’s product and still be true to our responsibilities that

give our branch only the judicial power.” 9 However, because “the Legislature

cannot . . . ‘trump’ the Michigan Constitution,” 10 and “it is unquestioned that the judiciary

has the power to determine whether a statute violates the constitution,” 11 this Court can

nullify statutes to the extent that they are unconstitutional. 12 Nevertheless, the judiciary

7
Skinner, 502 Mich at 100, quoting In re Sanders, 495 Mich 394, 404; 852 NW2d 524
(2014), citing Taylor, 468 Mich at 6.
8
In re Complaint of Rovas Against SBC Mich, 482 Mich 90, 98; 754 NW2d 259 (2008).
9
Id. (quotation marks, citation, and brackets omitted). See also Const 1963, art 3, § 2
(“The powers of government are divided into three branches: legislative, executive and
judicial. No person exercising powers of one branch shall exercise powers properly
belonging to another branch except as expressly provided in this constitution.”).
10
Sharp v Lansing, 464 Mich 792, 810; 629 NW2d 873 (2001).
11
North Ottawa Community Hosp v Kieft, 457 Mich 394, 403 n 9; 578 NW2d 267 (1998).
12
See Marbury v Madison, 5 US (1 Cranch) 137, 178; 2 L Ed 60 (1803) (“If then the courts
are to regard the constitution; and the constitution is superior to any ordinary act of the
legislature; the constitution, and not such ordinary act, must govern the case to which they
both apply.”); The Federalist No. 78 (Hamilton) (Cooke ed, 1961), p 524 (“By a limited
constitution I understand one which contains certain specified exceptions to the legislative
authority; such for instance as that it shall pass no bills of attainder, no ex post facto laws,
and the like. Limitations of this kind can be preserved in practice no other way than through
the medium of the courts of justice; whose duty it must be to declare all acts contrary to
the manifest tenor of the constitution void. Without this, all the reservations of particular
rights or privileges would amount to nothing.”); The Federalist No. 81 (Hamilton) (Cooke
ed, 1961), p 543 (“[T]he constitution ought to be the standard of construction for the laws,

6
must exercise caution in striking down statutes enacted by the representatives of the people

and must do so only if it is clear that the statute violates the Constitution.

We exercise the power to declare a law unconstitutional with extreme
caution, and we never exercise it where serious doubt exists with regard to
the conflict. Every reasonable presumption or intendment must be indulged
in favor of the validity of an act, and it is only when invalidity appears so
clearly as to leave no room for reasonable doubt that it violates some
provision of the Constitution that a court will refuse to sustain its validity.[13]

This Court does not have the authority to strike down statutes merely because it

disagrees with their wisdom or prudence.

Our task, under the Constitution, is the important, but yet limited, duty to
read and interpret what the Legislature has actually made the law. We have
observed many times in the past that our Legislature is free to make policy
choices that, especially in controversial matters, some observers will
inevitably think unwise. This dispute over the wisdom of a law, however,
cannot give warrant to a court to overrule the people’s Legislature.[14]

Accordingly, it is a matter of significant constitutional moment when the judiciary

countermands, and renders null and void, the considered enactments of the Legislature.

and . . . wherever there is an evident opposition, the laws ought to give place to the
constitution.”).
13
Phillips v Mirac, Inc, 470 Mich 415, 422-423; 685 NW2d 174 (2004) (quotation marks
and citations omitted); see also Sears v Cottrell, 5 Mich 251, 259 (1858) (opinion of
CHRISTIANCY, J.) (“No rule of construction is better settled in this country, both upon
principle and authority, than that the Acts of a State Legislature are to be presumed
constitutional until the contrary is shown; and it is only when they manifestly infringe some
provision of the Constitution that they can be declared void for that reason. In cases of
doubt, every possible presumption, not clearly inconsistent with the language and the
subject-matter, is to be made in favor of the constitutionality of the Act.”).
14
Lansing Mayor v Pub Serv Comm, 470 Mich 154, 161; 680 NW2d 840 (2004).

7
 II. GEOGRAPHIC-DISTRIBUTION REQUIREMENT

A. DIRECT DEMOCRACY IN MICHIGAN

Although Michigan primarily follows the republican form of representative

lawmaking, 15 it also contains important aspects of direct democracy; that is, the people’s

power to bring matters of public policy directly to the people for a vote. Three forms of

direct democracy are at issue this appeal: the initiative and the referendum, in the legislative

context, and the ability to propose amendments to the Michigan Constitution. 16 The

people’s power to propose new laws by petition (the initiative) and to approve or reject

laws enacted by the Legislature (the referendum) is preserved in Const 1963, art 2, § 9,

which provides, in relevant part:

The people reserve to themselves the power to propose laws and to
enact and reject laws, called the initiative, and the power to approve or reject
laws enacted by the legislature, called the referendum. The power of
initiative extends only to laws which the legislature may enact under this
constitution. The power of referendum does not extend to acts making
appropriations for state institutions or to meet deficiencies in state funds and
must be invoked in the manner prescribed by law within 90 days following
the final adjournment of the legislative session at which the law was enacted.
To invoke the initiative or referendum, petitions signed by a number of
registered electors, not less than eight percent for initiative and five percent

15
Const 1963, art 4, § 1 (“[T]he legislative power of the State of Michigan is vested in a
senate and a house of representatives.”).
16
Twenty-five states currently have some degree of direct democracy, Michigan being 1
of 15 states to permit initiatives, referendums, and voter-initiated constitutional
amendments. See National Conference of State Legislatures, Initiative and Referendum States

(accessed January 11, 2022) [https://perma.cc/P7RT-B6SX]. Although not at issue in this
appeal, our Constitution also recognizes a constitutional referendum, see Const 1963, art
12, § 1, as well the people’s power to recall elected officials, see Const 1963, art 2, § 8. As
used in this opinion, the term “referendum” will only be used in its legislative context.

8
 for referendum of the total vote cast for all candidates for governor at the last
preceding general election at which a governor was elected shall be required.

* * *

The legislature shall implement the provisions of this section.[17]

The people’s power to propose constitutional amendments by petition (voter-

initiated constitutional amendments) is preserved in Const 1963, art 12, § 2, which

provides, in relevant part:

Amendments may be proposed to this constitution by petition of the
registered electors of this state. Every petition shall include the full text of
the proposed amendment, and be signed by registered electors of the state
equal in number to at least 10 percent of the total vote cast for all candidates
for governor at the last preceding general election at which a governor was
elected. Such petitions shall be filed with the person authorized by law to
receive the same at least 120 days before the election at which the proposed
amendment is to be voted upon. Any such petition shall be in the form, and
shall be signed and circulated in such manner, as prescribed by law. The
person authorized by law to receive such petition shall upon its receipt
determine, as provided by law, the validity and sufficiency of the signatures
on the petition, and make an official announcement thereof at least 60 days
prior to the election at which the proposed amendment is to be voted upon.[18]

These rights have a long history in Michigan, and “[t]his Court has a tradition of

jealously guarding” them against needless government encroachment. 19 Yet they are not

unfettered. The text of these constitutional provisions plainly requires a minimum number

of signatures from the registered electors of the state to invoke the people’s direct-

democracy rights, and each provision contemplates a specific role for the Legislature in

17
Emphasis added.
18
Emphasis added.
19
Ferency v Secretary of State, 409 Mich 569, 593; 297 NW2d 544 (1980).

9
regulating the processes. “Thus, just as the people have enacted the former authority, so

too have they enacted the latter constraint.” 20

B. 2018 PA 608

MCL 168.471 sets forth petition-filing deadlines for initiatives, referendums, and

voter-initiated constitutional amendments. 2018 PA 608 amended MCL 168.471 to

provide, in relevant part:

Not more than 15% of the signatures to be used to determine the validity of
a petition described in this section shall be of registered electors from any 1
congressional district. Any signature submitted on a petition above the limit
described in this section must not be counted. When filing a petition
described in this section with the secretary of state, a person must sort the
petition so that the petition signatures are categorized by congressional
district. In addition, when filing a petition described in this section with the
secretary of state, the person who files the petition must state in writing a
good-faith estimate of the number of petition signatures from each
congressional district.[21]

Plaintiffs contend that this requirement violates Const 1963, art 2, § 9 and Const

1963, art 12, § 2 because it exceeds the Legislature’s authority under the plain language of

those provisions and unduly burdens and restricts the people’s ability to initiate change by

way of initiatives, referendums, and voter-initiated constitutional amendments.

20
Protect Mich Constitution v Secretary of State, 492 Mich 860, 861 (2012) (MARKMAN,
J., concurring).
21
Before 2018 PA 608 was enacted, petition signatures were gathered and sorted on a
countywide basis; therefore, other provisions of the Michigan Election Law had to be
amended to comply with the geographic-distribution requirement in 2018 PA 608. See
MCL 168.477(1) (“The board of state canvassers may not count toward the sufficiency of
a petition described in this section any valid signature of a registered elector from a
congressional district submitted on that petition that is above the 15% limit described in
[MCL 168.471].”); MCL 168.482(4) (requiring petition forms to specify the congressional
district in which signatures are obtained).

10
Intervening defendant argues that the geographic-distribution requirement falls squarely

within the Legislature’s authority to regulate the process of direct democracy and that

plaintiffs have failed to carry their heavy burden of showing the requirement facially

violates Article 2, § 9 or Article 12, § 2. The Court of Claims granted summary disposition

to plaintiffs, striking down the geographic-distribution requirement as unconstitutional

under both provisions. The Court of Appeals affirmed. Our task is to determine whether

this requirement violates Article 2, § 9, Article 12, § 2, or both.

C. THE TEXT OF THE MICHIGAN CONSTITUTION DEMONSTRATES THAT THE
GEOGRAPHIC-DISTRIBUTION REQUIREMENT IS CONSTITUTIONAL UNDER
CONST 1963, ART 12, § 2 BUT NOT UNDER CONST 1963, ART 2, § 9

“The primary objective in interpreting a constitutional provision is to determine the

text’s original meaning to the ratifiers, the people, at the time of ratification.” 22 “The

lodestar principle is that of ‘common understanding,’ the sense of the words used that

would have been most obvious to those who voted to adopt the constitution.” 23

As the text of both Const 1963, art 2, § 9 and Const 1963, art 12, § 2 makes clear,

the drafters sought to differentiate between, on the one hand, the Legislature’s role in the

people’s power to enact, approve, or reject legislation, and on the other hand, the people’s

power to propose constitutional amendments. Article 2, § 9 states that the Legislature shall

“implement” its provisions, while Article 12, § 2 permits the Legislature to “prescribe[] by

law” the form of petitions seeking to amend our Constitution, as well as the “manner” by

which those petitions are “signed and circulated.”

22
Wayne Co v Hathcock, 471 Mich 445, 468; 684 NW2d 765 (2004).
23
Straus v Governor, 459 Mich 526, 533; 592 NW2d 53 (1999).

11
 At the time the 1963 Michigan Constitution was ratified and in the years that

followed, “implement” meant “to carry out” or “to give practical effect to and ensure of

actual fulfillment by concrete measures.” 24 Given this definition, I have no objection to

the majority opinion’s definition of “implement,” as I agree the term carries with it “the

connotation that some received set of rules is being carried out, not that a new set of rules

is to be created.” 25 The term “prescribe,” on the other hand, does contemplate the

Legislature’s ability to make new rules and does not, by itself, contemplate a set of rules

being received. At the time of ratification, to “prescribe” meant, in pertinent part, “to lay

down as a guide, direction, or rule of action” or “to specify with authority.” 26 Notably,

however, the drafters of our 1963 Constitution did not use the term “prescribe” in isolation.

Instead, they used the phrase “prescribed by law,” which does suggest some set of rules

being received by the Legislature. According to the official record of the 1961–1962

Constitutional Convention: “Where ‘provided by law’ is used, it is intended that the

legislature shall do the entire job of implementation. Where only the details were left to

the legislature and not the over-all planning, the committee used the words ‘prescribed by

24
Webster’s Seventh New Collegiate Dictionary (1963) (defining “implement” as “to carry
out” or “to give practical effect to and ensure of actual fulfillment by concrete measures”).
See also The American Heritage Dictionary of the English Language (1969) (defining
“implement” as “[t]o provide a definite plan or procedure to ensure the fulfillment of” or
“carry into effect”).
25
Ante at 10.
26
Webster’s Seventh New Collegiate Dictionary (1963).

12
law.’ ” 27 According to the drafters’ intended meaning of “prescribed by law,” then, the

Legislature must prescribe the details by which petitions for constitutional amendments are

to be signed and circulated under Article 12, § 2. 28

But while both Article 2, § 9 and Article 12, § 2 contemplate a set of rules being

received by the Legislature, it is important that only Article 12, § 2 expressly tasks the

Legislature with prescribing the details by which petitions proposing constitutional

amendments are to be signed and circulated. At the time the 1963 Michigan Constitution

was ratified and in the years that followed, to “circulate” meant, in relevant part, “to pass

from person to person or place to place: as . . . to become well known or widespread [or]

to come into the hands of readers; [specifically,] to become sold or distributed[.]” 29 The

27
Beech Grove Investment Co v Civil Rights Comm, 380 Mich 405, 418-419; 157 NW2d
213 (1968), citing 2 Official Record, Constitutional Convention 1961, pp 2673-2674
(emphasis added).
28
During the 1961–1962 Constitutional Convention, one delegate described the drafters’
use of the phrase “prescribed by law” as when “merely the details of some particular plan
were left to the legislature and not the overall whole planning, but merely the
implementation of a plan . . . .” 2 Official Record, Constitutional Convention 1961,
pp 2673-2674 (statement of Delegate Robert J. Danhof) (emphasis added). But while the
convention record helps guide us toward ascertaining the text’s original meaning as
commonly understood by the ratifiers, it cannot overcome the express language of the
Constitution. See Beech Grove Investment Co, 380 Mich at 427 (“[I]t is the Constitution,
not the debates, that was finally submitted to the people. While the debates may assist in
an interpretation of the Constitution, neither they nor even the Address to the People is
controlling.”). Therefore, it is significant that the word “implement” does not appear in
the text of Article 12, § 2. And given its absence, it is logical to conclude that the phrase
“prescribed by law” means simply that the drafters intended that “only the details were left
to the legislature and not the over-all planning”—nothing more. Beech Grove, 380 Mich
at 419.
29
Webster’s Seventh New Collegiate Dictionary (1963). See also The American Heritage
Dictionary of the English Language (1969) (defining “circulate” as “[t]o move around, as

13
constitutional text also uses the term “manner,” which was relevantly defined as the “mode

of procedure or way of acting[.]” 30 Accordingly, Article 12, § 2 authorizes the Legislature

to establish laws pertaining to the details and mode by which petitions proposing

constitutional amendments come into the hands of their signatories, the people.

The differences in language between Article 12, § 2 and Article 2, § 9 are

significant. Words matter; particularly words chosen for inclusion in the Michigan

Constitution. This Court has recognized that the call for legislative action in Article 12,

§ 2 is specific to the manner by which petitions proposing constitutional amendments are

to be signed and circulated. 31 No such specificity exists in Article 2, § 9 with respect to

the people’s right to engage in direct democracy in the area of legislation. Unlike Article

12, § 2, Article 2, § 9 simply directs the Legislature to “implement” the provisions of that

section with no reference to the manner by which initiatives and referendums are to be

circulated. These textual differences between Article 12, § 2 and Article 2, § 9 demonstrate

the validity of the geographic-distribution requirement in 2018 PA 608 under the former,

but not the latter. The geographic-distribution requirement answers the call for legislative

action in Article 12, § 2, setting forth the manner by which petitions proposing

from person to person, or place to place”; “[t]o spread widely among persons or places;
disseminate” or “[t]o cause to move about or be distributed”).
30
Webster’s Seventh New Collegiate Dictionary (1963).
31
Consumers Power Co v Attorney General, 426 Mich 1, 6, 9; 392 NW2d 513 (1986)
(“[Const 1963, art 12, § 2] clearly authorizes the Legislature to prescribe by law for the
manner of signing and circulating petitions to propose constitutional amendments. . . . The
Constitution of 1963, unlike that of 1908, does summon legislative aid in the area of the
form of these petitions as well as in the areas of circulation and signing.”).

14
constitutional amendments are to be signed and circulated. As applied to Article 2, § 9,

however, the geographic-distribution requirement goes beyond the Legislature’s authority

to implement the provisions of that section. Whereas Article 12, § 2 envisions a specific

role for the Legislature to regulate petitions proposing constitutional amendments by

establishing laws aimed at ensuring those petitions are distributed among the people, no

such role exists for the Legislature under Article 2, § 9. 32 This is not to say that the

Legislature lacks authority under Article 2, § 9 to provide functional detail. The question

before us relates to the type of detail. Certainly matters such as deadlines, form

requirements, and font sizes are the type of detail left for the Legislature. 33 But,

32
The generic legislative directive in Article 2, § 9 regarding the referendum—that “[t]he
power of referendum . . . must be invoked in the manner prescribed by law”—cannot be
viewed as a call for the Legislature to prescribe how referendum petitions are circulated
throughout the state. Again, words matter. The drafters of our Constitution included the
right to direct democracy in the areas of legislation and amending the Constitution. We
must presume that when language is included in one provision but omitted in another, that
omission is intentional; therefore, because the language in Article 12, § 2 expressly calls
for legislative action regarding the manner of circulation and the same or similar language
is absent from of Article 2, § 9, we must presume the Legislature lacks the authority to
impose geographic restrictions on the manner of petition circulations for matters under
Article 2, § 9. See People v Peltola, 489 Mich 174, 185; 803 NW2d 140 (2011)
(“Generally, when language is included in one section of a statute but omitted from another
section, it is presumed that the drafters acted intentionally and purposely in their inclusion
or exclusion.”). See also Scalia & Garner, Reading Law: The Interpretation of Legal Texts
(St. Paul: Thomson/West, 2012), p 170 (explaining that words are “presumed to bear the
same meaning throughout a text” and that “a material variation in terms suggests a variation
in meaning”) (boldface omitted). While this canon is typically applied when interpreting
statutory provisions, I see no reason why it would not apply with equal force when
interpreting the constitutional provisions at issue in this case.
33
See, e.g., MCL 168.471 (requiring initiative petitions to be filed with the Secretary of
State “at least 160 days before the election at which the proposed law would appear on the
ballot”); MCL 168.472a (requiring signatures on constitutional amendment petitions and
initiative petitions to be made 180 days or fewer before the petition is filed with the

15
significantly, nowhere in Article 2, § 9 does it leave to the Legislature the manner by which

initiative and referendum petitions are to be circulated.

The majority concludes that the geographic-distribution requirement is not within

the constitutional authority of the Legislature to prescribe, labeling it instead as a

“substantive requirement” that does not advance any of the express constitutional

requirements—“unlike the law at issue in Consumers Power Co [v Attorney General].”34

I disagree.

As for the difference in substance versus procedure, a majority of this Court recently

labeled the 10% minimum threshold in Article 12, § 2 as a “procedural requirement of

obtaining a certain number of signatures.” 35 In my view, the geographic-distribution

requirement is akin to that procedural requirement and does not alter the minimum number

of signatures needed to place a voter-initiated constitutional amendment on the ballot. The

minimum 10% threshold remains the same, while the geographic-distribution requirement

Secretary of State); MCL 168.473b (requiring signatures on constitutional amendment
petitions and initiative petitions to be made after the last November general election at
which a Governor was elected); MCL 168.482(1) (requiring petitions to be 8½ inches by
14 inches in size); MCL 168.482(2) (requiring the heading of the petition to be printed in
capital letters in 14-point boldfaced type); MCL 168.482(3) (requiring a “summary in not
more than 100 words of the purpose of the proposed amendment or question proposed” to
“be printed in 12-point type” on the petition); id. (requiring the full text of the proposed
amendment to be printed in 8-point type on the petition); id. (requiring the petition to
indicate whether the proposal would alter or abrogate an existing provision of the
Constitution and, if so, to include the provisions to be altered or abrogated); MCL
168.482(5) (requiring certain warnings to the electorate to appear on the petitions).
34
Ante at 23.
35
Citizens Protecting Michigan’s Constitution v Secretary of State, 503 Mich 42, 73; 921
NW2d 247 (2018).

16
simply directs how petitions are to be signed and circulated throughout the state in order to

meet that threshold. Although the former requirement is expressly contained within the

constitutional text while the latter is not, this Court has rejected the notion that the

Legislature’s prescribed details as to the form, signing, and circulation of petitions

proposing constitutional amendments are unconstitutional requirements under Article 12,

§ 2; instead, “those requirements, in essence, are authorized by the constitution

itself . . . .” 36 Because Article 12, § 2 requires a minimum number of signatures for

petitions proposing constitutional amendments and leaves to the Legislature the details as

to how those petitions are signed and circulated, I disagree with the majority opinion’s

conclusion that the 15% maximum per congressional district constitutes an impermissible

substantive obligation. It is just as much a procedural requirement as the 10% minimum

signature threshold.

Further, I fail to see how the geographic-distribution requirement at issue here is

any more substantive than other requirements the Legislature has legitimately imposed.

For example, in Consumers Power Co, we upheld the predecessor statute to MCL

168.472a, which stated, “ ‘It shall be rebuttably presumed that the signature on a petition

which proposes an amendment to the constitution or is to initiate legislation, is stale and

void if it was made more than 180 days before the petition was filed with the office of the

secretary of state.’ ” 37 The timing requirement at issue in Consumers Power Co pertained

36
Citizens for Capital Punishment v Secretary of State, 414 Mich 913, 915 (1982).
37
Consumers Power Co, 426 Mich at 2, quoting MCL 168.472a, as enacted by 1973 PA
112. The current version of MCL 168.472a states, “The signature on a petition that
proposes an amendment to the constitution or is to initiate legislation shall not be counted

17
to when the signatures must be collected, while the geographic-distribution requirement at

issue here pertains to where the signatures must be collected. Both are procedural in nature

because they both impose a process that must be followed while not altering the substantive

right granted to the people to propose constitutional amendments by petition. Moreover,

this Court in Consumers Power Co concluded that the purpose of the 180-day window for

signatures to be collected before the filing of the petition “is to fulfill the constitutional

directive of art 12, § 2 that only the registered electors of this state may propose a

constitutional amendment.” 38 Similarly, the geographic-distribution requirement fulfills

the directive in Article 12, § 2 that petitions proposing constitutional amendments “be

signed by registered electors of the state equal in number to at least 10 percent of the total

vote cast” for Governor at the last gubernatorial election. 39 The statute specifies how the

petitions must be circulated to and signed by “electors of the state”—that is, the whole

state, not one specific locale.

Put simply, the geographic-distribution requirement is a law that details the manner,

i.e., the mode or way, by which petitions are to be signed and circulated. This requirement

falls squarely within the Legislature’s authority to prescribe under Article 12, § 2 but goes

beyond implementing the provisions of Article 2, § 9.

if the signature was made more than 180 days before the petition is filed with the office of
the secretary of state.”
38
Consumers Power Co, 426 Mich at 8.
39
Const 1963, art 12, § 2 (emphasis added).

18
D. THE GEOGRAPHIC-DISTRIBUTION REQUIREMENT IS CONSISTENT WITH
THE RATIFIERS’ COMMON UNDERSTANDING OF CONST 1963, ART 12, § 2

The validity of the geographic-distribution requirement in 2018 PA 608 under Const

1963, art 12, § 2 but not under Const 1963, art 2, § 9 is also demonstrated through the

history of direct democracy in Michigan, the 1961–1962 Constitutional Convention

record, 40 and the textual clues within the 1963 Michigan Constitution itself.

As a majority of this Court recently discussed in Citizens Protecting Michigan’s

Constitution v Secretary of State, the process by which the people may propose

constitutional amendments through the exercise of direct democracy first appeared in the

1908 Michigan Constitution. 41 As ratified, Michigan’s first voter-initiated constitutional

amendment process required the total number of signatures on a petition to amend the

Constitution to “exceed twenty per cent of the total number of electors voting for secretary

of state at the preceding election of such officer.” 42 The provision also permitted the

Legislature to veto the people’s proposed amendment or submit an alternative or substitute

proposal covering the same subject matter. 43 In explaining the need for legislative

40
The constitutional convention debates and the Address to the People are useful
interpretive tools that aid in ascertaining the common understanding of the ratifiers.
Burdick v Secretary of State, 373 Mich 578, 584; 130 NW2d 380 (1964). Nonetheless,
these records, while relevant, are not controlling. People v Tanner, 496 Mich 199, 226;
853 NW2d 653 (2014).
41
Citizens Protecting Michigan’s Constitution v Secretary of State, 503 Mich at 71-75
(discussing the history of Article 12, § 2).
42
Const 1908, art 17, § 2 (as ratified).
43
Id.

19
oversight for voter-initiated constitutional amendments, the Address to the People in the

1908 Constitution stated:

The convention realized the far-reaching effect that each amendment to the
constitution may have beyond the immediate purpose intended by it, and it
was deemed essential in so important a matter as changing the fundamental
law of the state that the very greatest care should be required in both the form
and substance of amendments to it. Such care is secured by requiring the
amendments proposed to pass the scrutiny of the legislature.[44]

These features of Michigan’s first voter-initiated constitutional amendment process

proved too difficult, as demonstrated by the fact that the procedure was never used.

Recognizing this, the Legislature proposed constitutional amendments in 1913 that made

voter-initiated constitutional amendments more accessible. 45 These amendments, which

the voters ultimately approved, deleted the legislative veto, lowered the requisite number

of signatures for voter-initiated constitutional amendments to 10% of the total votes cast

for Governor at the most recent gubernatorial election, and added the initiative and

referendum processes that exist today. 46

44
Journal of the Constitutional Convention 1907–1908, p 1591.
45
McHargue, Direct Government in Michigan (Lansing: State of Michigan, 1961), p 22
(“The legislative veto and the high percentage required for petition qualification rendered
the provision for popularly initiated constitutional amendments ineffective. It was never
used. Nevertheless, it served as a stepping stone to the more liberal provisions adopted in
1913.”).
46
Const 1908, art 17, § 2 (as amended); Const 1908, art 5, § 1 (as amended). See also
Grossman, The Initiative and Referendum Process: The Michigan Experience, 28 Wayne
L Rev 77, 79 (1981), citing Pollock, The Initiative and Referendum in Michigan (Ann
Arbor: University of Michigan Press, 1940), pp 3-4.

20
 In the years that followed, petition drives demonstrated not only that voter-initiated

constitutional amendments were more accessible, but also more attractive. From 1913 to

1961, 35 voter-initiated constitutional amendments were proposed, compared to only one

initiative. 47 Even today, voter-initiated constitutional amendments continue to be the most

used of Michigan’s direct-democracy devices. Since the ratification of the 1963 Michigan

Constitution, 33 voter-initiated constitutional amendments were presented to the electorate,

compared to 14 initiatives and 10 referendums. 48 The inherent delays with initiatives first

being sent to the Legislature for approval, as well as the substantial interest generated from

proposed constitutional amendments, are among the reasons why voter-initiated

constitutional amendments have been the preferred direct-democracy device, despite the

heightened signature requirement. 49 Another reason is that there are no restrictions on the

47
Wolverine Golf Club v Secretary of State, 24 Mich App 711, 717; 180 NW2d 820 (1970),
aff’d 384 Mich 461 (1971).
48
State of Michigan Bureau of Elections, Initiatives and Referendums Under the
Constitution of the State of Michigan of 1963 (January 2019)

(accessed January 13, 2022) [https://perma.cc/9ZFX-FHR5]. Of those 33 voter-initiated
constitutional amendments, 12 were approved. Of the 14 proposed initiatives placed on
the ballot by petition, 8 were approved. And of the 10 referendums placed on the ballot by
petition, 1 was approved. Id.
49
Wolverine Golf Club, 24 Mich App at 718 (“ ‘Why has the indirect statutory initiative
been used so seldom? It would seem that the delay inherent in the process (and delay
occurs unless legislative acquiescence is forthcoming and even then if opponents can
gather sufficient signatures for a referendum petition) militates against the chance of
successful promotion of such a measure. Then, too, the direct constitutional initiative
requires only a slightly higher percentage of petition signatures and has the advantage of
attracting more interest and receiving a direct popular vote. Whatever the reasons, the
indirect initiative has been one of the least used of Michigan’s devices of direct
legislation.’ ”), quoting Direct Government, p 30.

21
subject matter of voter-initiated constitutional amendments; nothing prevents petition

drives from proposing what would otherwise be legislative matters as constitutional

amendments. 50 Further, proposed constitutional amendments that are ultimately passed by

popular vote have the effect of being elevated to the status of supreme constitutional law

of the state, making them difficult to remove once ratified. These procedural advantages

of voter-initiated constitutional amendments have been shown to outweigh the slightly

more difficult signature requirement. 51

The delegates discussed this preferred treatment toward voter-initiated

constitutional amendments at length during the 1961–1962 Constitutional Convention. A

great deal of the debate surrounding initiatives, referendums, and voter-initiated

constitutional amendments focused on ensuring that the Constitution could not be easily

amended, as well as encouraging the use of the initiative so that routine policy matters

normally addressed through legislation would not be elevated to the status of supreme

constitutional law. One example of these efforts was the delegates’ consideration of an

alternative to the minimum signature threshold for voter-initiated constitutional

50
City of Jackson v Comm’r of Revenue, 316 Mich 694, 710; 26 NW2d 569 (1947)
(“Nowhere in [Const 1908, art 17, §§ 1 to 3] or elsewhere in the Constitution do we find
any limitation to the effect that what might otherwise be considered as legislation cannot
be initiated by petition, under said sections, as an amendment to the Constitution.”). See
also The Michigan Experience, 28 Wayne L Rev at 107 n 176.
51
Id. at 107 (“[T]he constitutional initiative has been the preferred method of direct
legislation in Michigan. The greater signature requirements are far outweighed by [the]
procedural advantages: . . . [the] lack of restriction as to subject matter, immunity to
legislative change[,] and stronger legal protection in case of court challenge, especially in
state courts.”) (citations omitted).

22
amendments that would have changed the minimum number of signatures needed to

propose constitutional amendments by petition to either 10% of the total votes cast for

Governor at the last gubernatorial election “or 300,000 such registered electors, whichever

shall be less.” 52 Those in favor argued that a fixed-signature alternative would ensure that

voter-initiated constitutional amendments had sufficient support before being placed on the

ballot, while at the same time making the process accessible to more than just large, well-

organized special-interest groups. 53 Those opposed argued that as the population

increased, the minimum 8% of signatures needed to propose new legislation by petition

could surpass the 300,000 signatures that would be needed to propose constitutional

amendments by petition; in turn, this meant the 300,000 fixed-signature alternative could

make voter-initiated constitutional amendments progressively easier to place on the ballot

than initiatives. 54 The opponents eventually succeeded in striking the fixed-signature

alternative, the chief reasons being the significance of amending the Constitution and the

52
2 Official Record, Constitutional Convention 1961, p 2459 (capitalization altered).
53
Id. at 2460-2465.
54
See, e.g., id. at 2462 (“[A]s the state grows in the future in population, it will become
easier to put a constitutional amendment on the ballot by the initiative than it is to put an
ordinary statute on the ballot by use of the initiative.”) (statement of Delegate J. Harold
Stevens). Delegate Stevens also explained that his reasons for opposing the 300,000 fixed-
signature alternative were similar to his committee proposal to increase the minimum
number of signatures required for voter-initiated constitutional amendments from 10% to
15%—“discourag[ing] people from putting statutory matter into the constitution.” Id.

23
overuse of voter-initiated constitutional amendments for matters that were better suited for

legislation. 55

At the same time, the convention did not want to make the task of amending the

Constitution by petition insurmountable. For example, the delegates voted down a

proposal that would have required a majority of electors voting in the election (rather than

a majority vote on the amendment) to pass voter-initiated constitutional amendments; they

also voted down a proposal that would have required voter-initiated constitutional

amendments to pass by a ⅗ vote rather than by a simple majority. 56

Accordingly, the discussion of direct democracy during the 1961–1962

Constitutional Convention demonstrates a reverence for the Constitution as the supreme

law of the land and a recognition of the rightful impediments one should encounter when

55
Id. at 3199. The delegates also considered a proposal to lower the minimum number of
required signatures needed for initiatives from 8% to 5%, same as the referendum. Id. at
2392-2395. The proposal, if adopted, would have made voter-initiated constitutional
amendments twice as difficult as the initiative process. Supporters of the proposal
acknowledged that the initiative was seldom used and that the electorate frequently used
voter-initiated constitutional amendments to propose matters that were statutory in nature.
Thus, they argued that lowering the minimum threshold would encourage the use of the
initiative, making it less likely that the electorate would place routine matters of public
policy into the Constitution through the use of voter-initiated constitutional amendments.
Opponents of the proposal won the day, however, arguing that it should also be hard for
the people to initiate legislation and that the 8% minimum threshold would ensure that the
initiative process remained sufficiently difficult. Accordingly, while the delegates were
concerned with petition drives opting for voter-initiated constitutional amendments over
initiatives, they were also concerned with direct democracy supplanting traditional
republican lawmaking. See, e.g., id. at 2394 (“[The initiative process is] tough. We want
to make it tough. It should not be easy. The people should not be writing the laws. That’s
what we have a senate and house of representatives for.”) (statement of Delegate Richard
D. Kuhn).
56
Id. at 2469-2472.

24
attempting to amend the Constitution. The drafters deleted much of the detail previously

contained in what are now Article 2, § 9 and Article 12, § 2. But they also sought to ensure

that amending the Constitution by petition remained a difficult task—particularly as it

compared to proposing new legislation, or approving or rejecting the traditional

legislation—while also ensuring that the process remained accessible. As a majority of

this Court recently explained: “[T]he convention decided to keep voter-initiated

constitutional amendments difficult because amendments, like the Constitution itself, were

intended to deal with serious matters. The convention accomplished its goal by imposing

what it viewed as the clearest and most stringent limitation on initiative amendments: a

signature requirement.” 57 The heightened signature requirement for voter-initiated

constitutional amendments demonstrates the drafters’ intent to make it more difficult to

propose amendments to the Constitution (10%) than proposing legislation (8%) or

approving or rejecting legislation (5%). With the benefit of hindsight, however, we know

that the 10% minimum signature requirement in Article 12, § 2 has not accomplished the

convention’s goal of encouraging the use of the initiative rather than voter-initiated

constitutional amendments, particularly for matters that could be addressed through routine

legislation.

Read in light of this history and the surrounding context in the Constitution itself,

the Legislature’s role in the constitutional amendment process under Article 12, § 2 is more

robust than its role in the initiative and referendum processes under Article 2, § 9. The

Legislature’s authority under Article 12, § 2 to “prescribe[] by law” the way in which

57
Citizens Protecting Michigan’s Constitution, 503 Mich at 75 (emphasis omitted).

25
petitions proposing constitutional amendments are “signed and circulated” gives it more

control over the constitutional amendment process and thus reflects the greater significance

of a constitutional amendment than legislation brought about by the direct-democracy

provisions. Accordingly, this language in Article 12, § 2—along with the differing

signature requirements—reflects the convention’s aim of encouraging the use of the

initiative rather than voter-initiated constitutional amendments in order to avoid the

placement of routine policymaking into the Constitution and to ensure that voter-initiated

constitutional amendments are reserved for substantial matters worthy of constitutional

elevation. 58

The validity of the geographic-distribution requirement under Article 12, § 2 but not

Article 2, § 9 is ultimately “consistent with the underlying theme of the drafters, expressed

both during convention debate and in the difference in signature requirements[:] that it

should be more difficult for the people to change the constitution than to pass or reject a

58
See, e.g., 2 Official Record, Constitutional Convention 1961, p 3199 (“Michigan for
years has had a constitution which is one of the easiest to amend of any of the states. My
objection to this provision”—a 300,000 fixed-signature alternative to the 10% threshold
for voter-initiated constitutional amendments—“is simply to make it more difficult to
amend the constitution than to pass an ordinary statute.”) (statement of Delegate Stevens);
id. at 2394 (“I think often when people are trying to decide which to do, they may say:
well, let’s just get 2 per cent more and get a constitutional amendment. And that may be
one of the reasons that it’s said we have much legislation in our constitution.”) (statement
of Delegate Tom Downs); id. at 2395 (“Very, very seldom has th[e] . . . initiative been
used, because with the requirement of 10 per cent necessary to put a constitutional
amendment on, most groups have taken the alternative of putting the constitutional
amendment on, and thereby writing into the constitution many, many things which are
really legislative in detail, their theory being, we’ll put it in the constitution and the
legislature can’t change it.”) (statement of Delegate Clyne W. Durst, Jr.).

26
law.” 59 The requirement falls within the common understanding of the ratifiers, the people,

as to the meaning of Article 12, § 2. It prescribes the manner by which petitions proposing

constitutional amendments are to be signed and circulated throughout the state without

altering the minimum number of signatures needed to place voter-initiated constitutional

amendments on the ballot. It does not create an insurmountable hurdle for those seeking

to place constitutional amendments on the ballot by petition. Rather, it ensures that such

matters are worthy of constitutional elevation by securing some measure of statewide

support.

E. THE EFFECT OF THE 1961–1962 CONSTITUTIONAL CONVENTION’S VOTING
DOWN A COUNTY-DISTRIBUTION REQUIREMENT

This is not the first time a geographic-distribution requirement has been proposed

for voter-initiated constitutional amendments. The delegates at the 1961–1962

Constitutional Convention twice voted down a proposal that would have limited the

number of signatures on petitions proposing constitutional amendments per county, first at

10% and then at 25%. 60 Although the majority opinion relies on the Convention’s refusal

to adopt a county geographic-distribution requirement as being indicative of an aspect of

the voter-initiated constitutional amendment process that the Convention intended to

preclude from the Legislature, there are a number of reasons for why the delegates’

59
The Michigan Experience, 28 Wayne L Rev at 104.
60
2 Official Record, Constitutional Convention 1961, pp 2465-2469, 3200-3201. It should
be noted, however, that the 10% county-distribution requirement was packaged together
with other proposed revisions to Article 12, § 2 as a substitute. Id. at 2465-2469. The
delegates, therefore, did not vote down the 10% county-distribution requirement in
isolation, but voted down the substitute as a whole.

27
decision not to adopt a county-distribution requirement is not fatal to the congressional-

district-distribution requirement in 2018 PA 608.

First and foremost, the geographic-distribution requirements considered by the

Convention and those in 2018 PA 608 are fundamentally distinct from each other. A

county-distribution requirement would have impermissibly classified citizens on the basis

of population, something the United States Supreme Court expressly rejected on equal-

protection grounds just one year after the 1963 Michigan Constitution was ratified. 61 A

congressional-district-distribution requirement, on the other hand, does not create

impermissible classifications. Such a requirement sets out the percentage of signatures that

may be obtained for petitions on the basis of evenly divided congressional districts that are

reapportioned every 10 years. 62 As the majority opinion notes, many delegates opposed

61
Reynolds v Sims, 377 US 533, 567-568, 579; 84 S Ct 1362; 12 L Ed 2d 506 (1964)
(holding that apportionment of seats in both houses of a bicameral state legislature must be
done equally based on population, because “the overriding objective must be substantial
equality of population among the various districts, so that the vote of any citizen is
approximately equal in weight to that of any other citizen in the State”). A few years later,
the Supreme Court struck down on equal-protection grounds an Illinois statute requiring
presidential candidates seeking a place on the ballot to obtain 200 petition signatures from
at least 50 of the state’s 102 counties. Moore v Ogilvie, 394 US 814, 815-819; 89 S Ct
1493; 23 L Ed 2d 1 (1969). At the time, 93.4% of Illinois’s registered voters resided in 49
counties, with the remaining 6.6% spread over the remaining 53 counties. Because the
statute effectively allowed the rural minority in 53 counties to place a candidate on the
ballot while the urban majority in 49 counties could not, the Court held that the statute
“discriminate[d] against the residents of the populous counties of the State in favor of rural
sections.” Id. at 819.
62
See Evenwel v Abbott, 578 US ___, ___; 136 S Ct 1120, 1124; 194 L Ed 2d 291 (2016)
(explaining that “jurisdictions must design both congressional and state-legislative districts
with equal populations, and must regularly reapportion districts to prevent
malapportionment”); Utah Safe to Learn–Safe To Worship Coalition, Inc v State, 94 P3d
217, 229; 2004 UT 32 (2004) (“By basing the signature requirement on evenly divided,

28
the county-distribution requirement as “substantively unacceptable.” 63 That opposition

was due in large part to the concern that the county-distribution requirement would create

discriminatory classifications among rural and urban populations, allowing the rural

minority to act as a check on the urban majority and thus disproportionately affecting voters

in highly populated areas, like Wayne County. 64 Indeed, one delegate explicitly cited

population-based senate districts, the legislature has not created a discriminatory
classification or caused a disparate impact among classes or subclasses.”). See also Semple
v Williams, 290 F Supp 3d 1187, 1193-1194 (D Colo, 2018) (comparing courts that “have
uniformly struck down geography-based signature-gathering requirements when the
relevant geographic subdivision was the county” with those courts that “have uniformly
upheld geography-based signature-gathering requirements when the relevant geographic
subdivision [was] a congressional district or state legislative district, given that such
districts must (per Supreme Court precedent) be of approximately equal population”), rev’d
sub nom Semple v Griswold, 934 F3d 1134, 1138-1139, 1141-1142 (CA 10, 2019) (holding
that an amendment of the Colorado Constitution, which required petitions seeking to place
voter-initiated constitutional amendments on the ballot to be signed by at least 2% of total
registered electors in each state senate district, did not violate the Equal Protection Clause
because senate districts were roughly equal in total population).
63
Ante at 20, citing 2 Official Record, Constitutional Convention 1961, pp 2466-2469,
3200.
64
See, e.g., 2 Official Record, Constitutional Convention 1961, p 2468 (“I think everyone
is well aware of what the substitute means, what it is intended to do . . . [.] [Y]ou could
conceivably have a minority, an extreme minority of the population in a large percentage
of the thinly populated counties who would have complete and absolute control over
whether or not there would be any constitutional amendments submitted to the people.”)
(statement of Delegate William Marshall); id. at 3200 (“[W]hat you are saying here is that
those of us who live in Wayne county, merely because we do compose of a third of the
state’s population, are not going to count as heavily as those in the rest of the state. I think
this is a distinctly unfair and prejudiced position to take . . . .”) (statement of Delegate
Catherine Moore Cushman). See also Utah Safe to Learn, 94 P3d at 229 (explaining that
a distribution requirement based on geographically drawn counties “discriminate[s] against
urban voters by diluting the voting power of . . . urban counties” and allows “the rural
minority [to] act as a check and a balance on the urban majority” whereas a geographic-

29
equal-protection concerns as one of the reasons for his opposition. 65 Given the equal-

protection concerns with a county-distribution requirement, it is understandable that the

delegates voted it down.

Further, in reviewing the convention record, it appears that the delegates only

considered adding a county-distribution requirement to voter-initiated constitutional

amendments, not initiatives or referendums. This demonstrates that the delegates knew the

significance of amending the Constitution and understood that in order to have a

sustainable constitution, it should not be easily amended. 66 The delegates also recognized

that whereas legislation might affect only a part of the state, a constitution necessarily

affects the entire state; therefore, before a proposal to alter the state’s fundamental law is

distribution requirement based on evenly divided, population-based districts “does not
assign disproportionate power to any particular group of voters”).
65
Id. at 3200 (“I attack this particular amendment as being a violation of equal protection
of the laws.”) (statement of Delegate Melvin Nord).
66
A few delegates made this point when discussing whether to require a ⅗ vote to amend
the Constitution by petition rather than a simple majority, which was considered together
with the 10% county-distribution requirement. See, e.g., id. at 2466 (“[I]f we are right in
saying that we want to do away with statutory detail, that we want to write a good
constitution, a good fundamental law, let’s not leave it so it is going to be changed in
another year or two by everyone who has a little axe to grind getting aboard a petition[,]
putting it on the ballot[,] getting it to carry[,] and having our constitution sufficiently or
adequately detailed so that we will have to have another convention in another 10 years.”)
(statement of Delegate Garry Brown); id. at 2468 (“[W]hen we are dealing with something
as fundamental as a constitution, which is a protection against the imposition of the will of
the state, that we should be very careful in the allowance of those particular guarantees to
be changed because the constitution is a compact with the people. It represents not only
what the position of the people is for the present day but also for the future, for those yet
unborn children. I feel that it is very necessary to make it more difficult to change and alter
the basic law and constitution of the state.”) (statement of Delegate O. Lee Boothby).

30
placed on the ballot, it should have support from a wide geographic base. 67 As mentioned,

the geographic-distribution requirement in 2018 PA 608 is aimed at achieving broad and

generalized support for a proposal before allowing it to appear on a statewide ballot. 68

Moreover, one aspect of the floor debate surrounding the county-distribution

requirement was the removal of much of the detail from the predecessor of Article 12,

§ 2—Const 1908, art 17, § 2—that was statutory in nature. Part of the concern with the

county-distribution requirement was that its inclusion in Article 12, § 2 would run contrary

to the convention’s objective of eliminating detail that the Legislature could prescribe at a

later time. 69 Indeed, if a county-distribution requirement had been included in Article 12,

67
See, e.g., id. at 2467 (“The whole purpose of requiring that you get not more than 10 per
cent coming from any one county is that this is a statewide provision, that it will have
statewide effect, and that there should be more than a self starter in one county insofar as
any provision is concerned that is going to become part of our basic and fundamental law.”)
(statement of Delegate Brown); id. at 2467-2468 (“[A] law generally affects not a complete
state but, generally speaking, only a part of the state or a part of the whole. The constitution
affects the whole and, for that reason, it should reflect more of a general, all over policy
rather than a policy of one particular area.”) (statement of Delegate Boothby).
68
See, e.g., Utah Safe to Learn, 94 P3d at 229 (explaining that a geographic-distribution
requirement based on state senate districts “does not unduly burden the initiative right, but
is a reasonable means of achieving the legitimate legislative purpose of ensuring a
modicum of support for an initiative throughout the statewide population”). See also
House Legislative Analysis, HB 6595 (December, 13, 2018), p 2 (“A maximum percentage
from each congressional district would ensure that petitions destined for the ballot were
supported by a more representative geographic cross-section of Michiganders . . . .”).
69
In urging the convention to reject the 25% county-distribution requirement, one delegate
stated:

I think that we are again trying to get into a statutory position when we
should not. I don’t think that the amendment is consistent with what we want
in this constitution in this particular area. This amendment does not
determine the course of events. This merely creates an opportunity for the
people to vote on an issue and it does not determine the issue once the

31
§ 2, it would have been exceedingly difficult to near impossible to remove it after the

Constitution was ratified. Again, one reoccurring theme in the debate surrounding voter-

initiated constitutional amendments was ensuring that the Constitution could not be easily

amended. Thus, it stands to reason that the delegates rejected the county-distribution

requirement because it was detail better left to future legislative bodies. Regardless, the

mere fact that the drafters declined to make a geographic-distribution requirement a

constitutional mandate does not mean that they wanted to preclude the Legislature from

crafting such a requirement or, more importantly, that the constitutional text (as commonly

understood) precludes the Legislature from doing so.

Accordingly, even though the convention ultimately voted down a county-

distribution requirement, many factors and considerations went into that decision. While

the delegates understood the significance of amending the Constitution and its impact on

the entire state rather than particular locales, they also recognized the equal-protection

concerns arising from a county-distribution requirement. Rather than include that type of

statutory matter in the constitutional text, the ratifiers left it to the Legislature to decide

whether to prescribe a geographic-distribution requirement as a manner by which petitions

amendment is placed on the ballot. When the proper signatures have been
obtained, all of the people of the state will have an opportunity, if they wish,
to exercise their franchise. I would like to urge you strongly not to get into
this particular trap of putting this type of statutory language in the
constitution. I would like to urge that you defeat this amendment and leave
the language as we now have it. [2 Official Record, Constitutional
Convention 1961, p 3200 (statement of Delegate Arthur G. Elliott, Jr.)
(emphasis added).]

32
proposing constitutional amendments are signed and circulated—this being consistent with

the directive to the Legislature in Article 12, § 2. 70

F. THE GEOGRAPHIC-DISTRIBUTION REQUIREMENT IN 2018 PA 608 DOES
NOT CONFLICT WITH THE SELF-EXECUTING NATURE OF
CONST 1963, ART 12, § 2

Much of the opposition to the geographic-distribution requirement in 2018 PA 608

is focused on the notion that Const 1963, art 12, § 2 and Const 1963, art 2, § 9 are self-

executing; that is, the people need not wait for legislative enactment in order to invoke

direct democracy. 71 The aim of the drafters of the 1963 Constitution was to eliminate much

of the statutory detail that plagued the former direct-democracy provisions while, at the

same time, leaving enough guidance for the people to exercise their rights and allow the

Legislature to sort out the details. Any legislation, then, while not necessary to invoke the

70
Compare 2 Official Record, Constitutional Convention 1961, p 3367 (“Matters of
legislative detail contained in [Const 1963, art 2, § 9] are left to the legislature.”) with id.
at 3407 (noting with regard to Const 1963, art 12, § 2 that “[d]etails as to form of petitions,
their circulation and other elections procedures are left to the determination of the
legislature.”) (emphasis added).
71
Thompson v Secretary of State, 192 Mich 512, 520; 159 NW 65 (1916) (“ ‘A
constitutional provision may be said to be self-executing, if it supplies a sufficient rule, by
means of which the right given may be enjoyed and protected, or the duty imposed may be
enforced; and it is not self-executing when it merely indicates principles, without laying
down rules by means of which those principles may be given the force of law.’ ”), quoting
Cooley, Constitutional Limitations (7th ed), p 121. See also Wolverine Golf Club, 24 Mich
App at 725 (“Whether a constitutional provision is self-executing is largely determined by
whether legislation is a necessary prerequisite to the operation of the provision.”).

33
direct-democracy provisions, may supplement these self-executing provisions, but it may

not unduly burden the people’s right to participate in direct democracy. 72

The practical effect of the 15% geographic-distribution requirement is that petition

drives would need to satisfy the minimum threshold requirements for initiatives,

referendums, and voter-initiated constitutional amendments by obtaining signatures from

at least 7 of Michigan’s 14 congressional districts. This requirement would no longer

permit petition drives to gather the requisite minimum number of signatures from one

specific geographic location. Although this geographic-distribution requirement may

burden petition drives, there is no evidence proffered in the plaintiffs’ facial challenge that

the requirement will in fact unduly burden petition drives. Significantly, many of

Michigan’s congressional districts are located in the southeast part of the state. Thus, any

additional obligation that the geographic-distribution requirement places on petition drives

is not onerous. And although the geographic-distribution requirement may not lead to

petition drives circulating petitions from those congressional districts that cover a wider

physical area, like districts in northern Michigan, 73 the geographic-distribution requirement

72
Ferency, 409 Mich at 591 n 10 (explaining that “the principle that the Legislature may
not unduly burden the self-executing constitutional procedure applies equally to both
[Const 1963, art 2, § 9, and art 12, § 2]” because “both are procedures whereby the people
reserved to themselves the power to directly change the law, constitutional or statutory,
under which they live”). See also Constitutional Limitations, p 122 (explaining that
supplemental legislation to self-executing rights “may be desirable, . . . but all such
legislation must be subordinate to the constitutional provision, and in furtherance of its
purpose, and must not in any particular attempt to narrow or embarrass it”).
73
The legislators opposed to the geographic-distribution requirement in 2018 PA 608
pointed out that 7 of Michigan’s 14 congressional districts touch Wayne, Oakland, and
Macomb Counties; therefore, they questioned how the requirement would ensure greater
geographic representation if petition drives could gather the requisite number of signatures

34
does, strictly speaking, ensure wider geographic support. In any event, “when considering

a claim that a statute is unconstitutional, the Court does not inquire into the wisdom of the

legislation.” 74

We have said that any supplemental legislation to self-executing provisions,

including those “constitutional provisions by which the people reserve to themselves a

direct legislative voice,” must not limit or restrict the rights set forth in those provisions. 75

Yet we have also recognized that those limits or restrictions may be expressly provided

within the text of the self-executing provisions. 76 As this Court stated in Scott v Secretary

of State:

Of the right of qualified voters of the State to propose amendments to
the Constitution by petition it may be said, generally, that it can be interfered
with neither by the legislature, the courts, nor the officers charged with any
duty in the premises. But the right is to be exercised in a certain way and
according to certain conditions, the limitations upon its exercise, like the
reservation of the right itself, being found in the Constitution.[77]

Again, Article 12, § 2 reserves to the people the right to propose constitutional

amendments by petition. But it also provides a role for the Legislature to prescribe the

without leaving the tri-county area. See House Legislative Analysis, HB 6595
(December, 13, 2018), p 2.
74
Taylor, 468 Mich at 6.
75
Kuhn v Dep’t of Treasury, 384 Mich 378, 385; 183 NW2d 796 (1971).
76
Hamilton v Secretary of State, 227 Mich 111, 125; 198 NW 843 (1924) (“The only
limitation, unless otherwise expressly indicated, on legislation supplementary to self-
executing constitutional provisions is that the right guaranteed shall not be curtailed or any
undue burdens placed thereon.”) (quotation marks and citation omitted; emphasis added).
77
Scott v Secretary of State, 202 Mich 629, 643; 168 NW 709 (1918) (emphasis added).

35
manner, i.e., the mode or way, by which that right is to be exercised, including the manner

by which petitions proposing constitutional amendments are to be signed and circulated.

Thus, to the extent the geographic-distribution requirement in 2018 PA 608 incidentally

burdens the gathering of signatures for voter-initiated constitutional amendments, it does

so with explicit authorization from Article 12, § 2. 78 Because no such explicit authorization

78
The majority opinion fails to appreciate this point when it states that my position
“significantly undervalues” the geographic-distribution requirement as a substantive
limitation on the people’s right to propose constitutional amendments by petition. Ante at
18. As we held in Citizens for Capital Punishment, 414 Mich at 915, the Legislature’s
authority to prescribe the manner by which petitions proposing constitutional amendments
are signed and circulated is a textual limitation “authorized by the constitution itself” on
that right. In any event, as discussed, the geographic-distribution requirement does not
unduly burden the people’s right to engage in direct democracy. It does not alter the
minimum number of signatures needed to place a proposal on the ballot; it only specifies
where in the state those signatures must be collected. The majority opinion states that,
unlike the rebuttable presumption in Consumers Power, 425 Mich 1, the geographic-
distribution requirement “irrefutably invalidate[s] signatures collected beyond the 15%
cap, making it a more onerous burden on the people’s right to propose amendments to the
state Constitution . . . .” Ante at 23 n 14 (emphasis omitted). Yet it will almost always be
the case that some signatures affixed to a petition will be invalidated in one way or another.
Petition drives generally account for this by obtaining more than the minimum number of
signatures needed to place a petition on the ballot. Therefore, even after the 15% maximum
per congressional district is met, the requirement does not prevent people from signing the
petitions. In keeping with MCL 168.476(1), the geographic-distribution requirement
simply stops counting signatures once the Board of State Canvassers has “canvass[ed] the
petitions to ascertain if the petitions have been signed by the requisite number of qualified
and registered electors” in a single congressional district—just as the board does when the
10% minimum signature threshold used to place a proposal on the ballot. And, contrary to
the majority opinion’s position that electors whose signatures are collected after the 15%
geographic-distribution requirement is satisfied are “silenced” by its application, there is
no indication that these electors are prohibited from signing the petition even after the
requirement is met, nor are they otherwise prevented from expressing their support (or
opposition) to the ballot proposal.

36
exists in Article 2, § 9, the geographic-distribution requirement does not withstand scrutiny

under that provision.

Ultimately, the presumption that statutes are constitutional accords the Legislature

a measure of deference. Every reasonable presumption must be resolved in favor of the

statute’s validity, and this Court will not strike down a validly enacted law absent “a clearly

apparent demonstration of unconstitutionality.” 79 This Court exercises its power to declare

a law unconstitutional with extreme caution, never exercising that authority “where serious

doubt exists with regard to the conflict” between the statute and the constitutional provision

under review. 80 I conclude that there is no conflict between the geographic-distribution

requirement in 2018 PA 608 and Const 1963, art 12, § 2. The requirement is a valid

legislative measure under Article 12, § 2 that prescribes the manner by which petitions

proposing constitutional amendments are to be signed and circulated. Therefore, because

plaintiffs have not demonstrated that the geographic-distribution requirement clearly

conflicts with the text of Article 12, § 2, they have not met the rigorous standard necessary

to establish the requirement’s unconstitutionality under that provision.

79
Skinner, 502 Mich at 99 (quotation marks and citation omitted); see also Consumers
Power Co, 426 Mich at 10 (“A court will not declare a statute unconstitutional unless it is
plain that it violates some provisions of the constitution and the constitutionality of the act
will be supported by all possible presumptions not clearly inconsistent with the language
and the subject matter.”).
80
Phillips, 470 Mich at 422.

37
 III. DISCLOSURE REQUIREMENTS

A. 2018 PA 608

Under the new requirements of 2018 PA 608, the petition signature sheets must

indicate whether the circulator of the petition is a paid circulator or a volunteer circulator.

The circulator must check a box on the signature sheet to so indicate. These requirements

are codified in MCL 168.482(7) and (8), which provide:

(7) Each petition under this section must provide at the top of the page
check boxes and statements printed in 12-point type to clearly indicate
whether the circulator of the petition is a paid signature gatherer or a
volunteer signature gatherer.

(8) Each petition under this section must clearly indicate below the
statement required under subsection (7) and be printed in 12-point type that
if the petition circulator does not comply with all of the requirements of this
act for petition circulators, any signature obtained by that petition circulator
on that petition is invalid and will not be counted.

In addition, 2018 PA 608 requires paid petition circulators or “paid signature

gatherers,” 81 as opposed to volunteer signature gatherers, to file affidavits with the

Secretary of State indicating that the person has been paid to circulate a petition and gather

signatures. Signatures that are collected by people who have not filed the required affidavit

or have given false information or omitted certain details are invalid. These provisions

were codified at MCL 168.482a, which states:

(1) If an individual who circulates a petition under [MCL 168.482] is
a paid signature gatherer, then that individual must, before circulating any
petition, file a signed affidavit with the secretary of state that indicates he or
she is a paid signature gatherer.

81
A “paid signature gather” is defined as “an individual who is compensated, directly or
indirectly, through payments of money or other valuable consideration to obtain signatures
on a petition as described in [MCL 168.471].” MCL 168.482d.

38
 (2) Any signature obtained on a petition under [MCL 168.482] by an
individual who has not filed the required affidavit under subsection (1) is
invalid and must not be counted.

(3) If the circulator of a petition under [MCL 168.482] provides or
uses a false address or provides any fraudulent information on the certificate
of circulator, any signature obtained by that circulator on that petition is
invalid and must not be counted.

(4) If a petition under [MCL 168.482] is circulated and the petition
does not meet all of the requirements under [MCL 168.482], any signature
obtained on that petition is invalid and must not be counted.

(5) Any signature obtained on a petition under [MCL 168.482] that
was not signed in the circulator’s presence is invalid and must not be counted.

Plaintiffs argue that both the checkbox and affidavit requirements violate their rights

of free speech, association, and petition under the United States and Michigan

Constitutions by imposing undue burdens on paid petition circulators and failing to serve

any compelling state interest. Intervening defendant disagrees, arguing that the checkbox

and affidavit requirements impose minimal burdens on circulators’ speech and further the

state’s legitimate interest in transparency and accountability in the electoral process. The

Court of Claims struck down the checkbox requirement as unconstitutional, but upheld the

affidavit requirement. The Court of Appeals reached the opposite conclusion, holding that

the checkbox requirement satisfied “the more exacting strict scrutiny test” while striking

down the affidavit requirement under that same standard. 82 As explained below, I conclude

that both requirements are constitutional.

82
League of Women Voters of Mich v Secretary of State, ___ Mich App ___, ___; ___
NW2d ___ (2021) (LWV) (Docket Nos. 357984 and 357986); slip op at 16.

39
 B. FIRST-AMENDMENT PRINCIPLES AND THE EXACTING-SCRUTINY
STANDARD

The First Amendment, made applicable to the states through the Fourteenth

Amendment, provides that Congress “shall make no law . . . abridging the freedom of

speech, or of the press; or the right of the people peaceably to assemble, and to petition the

Government for a redress of grievances. 83 “[T]he circulation of a petition involves the type

of interactive communication concerning political change that is appropriately described

as ‘core political speech’ ” for which First Amendment protection is at its “zenith.” 84

However, “there must be a substantial regulation of elections if they are to be fair and

honest and if some sort of order, rather than chaos, is to accompany the democratic

processes.” 85 Accordingly, “[s]tates allowing ballot initiatives have considerable leeway

to protect the integrity and reliability of the initiative process, as they have with respect to

election processes generally.” 86 Among the measurers states may implement to protect

those interests are disclosure requirements. Generally speaking, disclosure requirements

83
US Const, Am I; see also Meyer v Grant, 486 US 414, 420; 108 S Ct 1886; 100 L Ed 2d
425 (1988). “The individual right to solicit signatures to qualify an initiative petition is
protected by the rights of free expression, assembly, and petition, guaranteed in [Const
1963, art 1, §§ 3, 5].” Woodland v Mich Citizens Lobby, 423 Mich 188, 215; 378 NW2d
337 (1985). While “the Michigan Constitution may afford broader free expression and
petition protections against government infringements” than the United States
Constitution, there is no contention here that it does. Id. at 202. Accordingly, it is
appropriate to review the free-speech rights at issue under both the United States and
Michigan Constitutions as coterminous.
84
Meyer, 486 Mich at 421-422, 425 (quotation marks and citations omitted).
85
Buckley v American Constitutional Law Foundation, Inc, 525 US 182, 187; 119 S Ct
636; 142 L Ed 2d 599 (1999) (ACLF) (quotation marks and citation omitted).
86
Id. at 191.

40
are aimed at promoting transparency and accountability in the electoral process, as well as

preserving the integrity of the process by combating fraud and assisting in the detection of

invalid signatures. 87 Notably, disclosure requirements do not “impose a ceiling on

speech.” 88 “Although they may burden the ability to speak, disclosure requirements do not

prevent anyone from speaking.” 89

“First Amendment challenges to disclosure requirements in the electoral context”

are reviewed “under what has been termed ‘exacting scrutiny.’ ” 90 “ ‘Exacting scrutiny,’

despite the name, does not necessarily require that kind of searching analysis that is

normally called strict judicial scrutiny; although it may.” 91 Rather, this standard requires

a “sliding-scale analysis,” in which “the strength of the governmental interest must reflect

the seriousness of the actual burden on First Amendment rights.” 92 The Supreme Court of

the United States has framed this “more flexible standard” of scrutiny as follows:

A court considering a challenge to a state election law must weigh “the
character and magnitude of the asserted injury to the rights protected by the
First and Fourteenth Amendments that the plaintiff seeks to vindicate”
against “the precise interests put forward by the State as justifications for the

87
Doe v Reed, 561 US 186, 197-198; 130 S Ct 2811; 177 L Ed 2d 493 (2010).
88
Libertarian Party of Ohio v Husted, 751 F3d 403, 413 (CA 6, 2014).
89
Id., citing Citizens United v Fed Election Comm, 558 US 310, 366; 130 S Ct 876; 175 L
Ed 2d 753 (2010).
90
Reed, 561 US at 196.
91
Husted, 751 F3d at 414.
92
Id. (quotation marks and citations omitted).

41
 burden imposed by its rule,” taking into consideration “the extent to which
those interests make it necessary to burden the plaintiff’s rights.”[93]

A few cases help illustrate how courts have applied the exacting-scrutiny standard.

In Meyer v Grant, the Supreme Court held that a statute that prohibited paying petition

circulators violated the First Amendment, explaining that the ban restricted political

expression in two ways:

First, it limits the number of voices who will convey appellees’ message and
the hours they can speak and, therefore, limits the size of the audience they
can reach. Second, it makes it less likely that appellees will garner the
number of signatures necessary to place the matter on the ballot, thus limiting
their ability to make the matter the focus of statewide discussion.[94]

The Court held that the “State’s interest in protecting the integrity of the initiative process

does not justify the prohibition because . . . we are not prepared to assume that a

professional circulator—whose qualifications for similar future assignments may well

depend on a reputation for competence and integrity—is any more likely to accept false

signatures than a volunteer who is motivated entirely by an interest in having the

proposition placed on the ballot.” 95

In McIntyre v Ohio Elections Comm, the Supreme Court held that a statute

prohibiting the distribution of anonymous campaign literature violated the First

Amendment. 96 The Court explained that the ban “indiscriminately outlaw[ed] a category

93
Burdick v Takushi, 504 US 428, 434; 112 S Ct 2059; 119 L Ed 2d 245 (1992), quoting
Anderson v Celebrezze, 460 US 780, 789; 103 S Ct 1564; 75 L Ed 2d 547 (1983).
94
Meyer, 486 US at 422-423.
95
Id. at 426.
96
McIntyre v Ohio Elections Comm, 514 US 334; 115 S Ct 1511; 131 L Ed 2d 426 (1995).

42
of speech”—i.e., anonymous political speech 97—and that it did not sufficiently advance

the state’s two asserted interests: providing the electorate with relevant information and

preventing fraudulent and libelous statements. The Court first rejected the state’s argument

that a more informed electorate justified the ban, explaining that “[t]he name and address

of the author add little, if anything, to the reader’s ability to evaluate the document’s

message.” 98 Next, although the Court recognized that the state’s interest in preventing

fraud and libel were legitimate and important interests, the Court noted that other

legislation already prohibited making or disseminating false statements during political

campaigns and that the ban at issue applied to documents that were not even false or

misleading. The Court ultimately concluded that while the state’s interest in preventing

fraud and libel “might justify a more limited identification requirement,” there was “scant

cause” for the complete prohibition of anonymous campaign literature. 99

In Buckley v American Constitutional Law Foundation, Inc (ACLF), the Supreme

Court held that a requirement that initiative-petition circulators be registered voters also

violates the First Amendment because it would “ ‘limit the number of voices who will

convey the initiative proponents’ message’ and, consequently, cut down ‘the size of the

audience proponents can reach.’ ” 100 It also held that the requirement that circulators wear

an identification badge bearing the circulator’s name violated the First Amendment

97
Id. at 357.
98
Id. at 348-349.
99
Id. at 353.
100
ACLF, 525 US at 194-195 (brackets omitted), quoting Meyer, 486 US at 422-423.

43
because it “inhibits participation in the petitioning process” by “expos[ing] the circulator

to the risk of ‘heat of the moment’ harassment.” 101 Finally, the Court held that a

requirement that proponents of an initiative report the names and addresses of all paid

circulators and the amount paid to each violates the First Amendment because it “ ‘forces

paid circulators to surrender the anonymity enjoyed by their volunteer counterparts[.]’ ”102

The Court distinguished this disclosure requirement from an affidavit requirement that

required circulators to attach to each section of the petition an affidavit containing the

circulator’s name and address, 103 stating that “[b]ecause the disclosure provisions target

only paid circulators and require disclosure of the income from circulation each receives,

the disclosure reports are of course distinguishable from the affidavit, which must be

completed by both paid and volunteer circulators, and does not require disclosure of the

amount paid individually to a circulator.” 104

Finally, in Libertarian Party of Ohio v Husted, the United States Court of Appeals

for the Sixth Circuit held that a statute requiring circulators of candidacy or nomination

petitions to disclose the name and address of the person employing them, if any, does not

violate the First Amendment. 105 The court found it significant that

101
ACLF, 525 US at 198, 199 (quotation marks and citation omitted).
102
Id. at 204 (citation and alteration omitted).
103
The lower courts in ACLF upheld the affidavit requirement, and those rulings were not
challenged in the Supreme Court.
104
Id. at 204 n 24 (citation and alteration omitted).
105
Husted, 751 F3d 403.

44
 the disclosure is not made by the circulator to the voter. Rather, the
disclosure is made by the circulator when the petition is filed, after the
signatures are gathered. So while the core First Amendment activity of
communicating with voters is occurring, the disclosure requirement plays no
part. The circulator does not directly lose anonymity with the voter whose
signature is being solicited.[106]

Further, the court noted that although “seeking any information about a circulator

has some potential, however small, to reduce willingness to engage in circulating . . . , little

else suggests that [any] chill has occurred or is likely to occur as a result of the

requirement.” 107 On the other hand, the court recognized that “the employer disclosure

requirement serves substantial and legitimate state interests” because it “helps deter fraud

and also to detect it” by “enabl[ing] the Secretary of State’s Office to cross-check [the

disclosures] with campaign expenditure reports and thus contributes to overall reporting

compliance.” 108

C. THE CHECKBOX AND AFFIDAVIT REQUIREMENTS IN 2018 PA 608 ARE
CONSTITUTIONAL

Applying the aforementioned principles, I conclude that both the checkbox and

affidavit requirements are constitutional. As an initial matter, although the Court of

Appeals correctly recognized that the exacting-scrutiny standard applies in this case, it

misapplied that standard in a few critical ways.

First, the Court of Appeals erroneously and unnecessarily analyzed the checkbox

requirement under a strict-scrutiny framework despite its recognition that the requirement

106
Id. at 417.
107
Id.
108
Id. at 417-418.

45
imposes “little to no burden on circulators.” 109 “Regulations imposing severe burdens” on

political speech in the electoral context warrant strict scrutiny, while “[l]esser

burdens . . . trigger less exacting review . . . .” 110 Therefore, it was inappropriate for the

Court of Appeals to hold the state to a higher level of scrutiny in its analysis of the checkbox

requirement. Instead, as the majority opinion concludes, the checkbox requirement

satisfies the exacting-scrutiny standard. It imposes only a minimal burden on petition

circulators and advances the important state interest of affording the electorate information

that may be helpful in making an informed electoral decision.

Second, the Court of Appeals incorrectly applied strict scrutiny to the affidavit

requirement. Under the exacting-scrutiny standard, “the rigorousness of our inquiry into

the propriety of a state election law depends upon the extent to which a challenged

regulation burdens First and Fourteenth Amendment rights.” 111 That is, it is only once

plaintiffs—as the parties challenging the statute’s constitutionality—have identified the

extent to which the law burdens speech that a reviewing court can then determine the

“rigorousness” of the scrutiny to be applied to the state’s asserted interest. 112 This

109
LWV, ___ Mich App ___; slip op at 18. Although the majority opinion assumes “for
the sake of argument” that the checkbox requirement “imposes some direct but minimal
burden on core political speech,” ante at 30, I agree with the Court of Appeals’ assessment
that the actual burden imposed by the checkbox requirement is de minimis.
110
Timmons v Twin Cities Area New Party, 520 US 351, 358; 117 S Ct 1364; 137 L Ed 2d
589 (1997) (emphasis added).
111
Burdick, 504 US at 434 (emphasis added).
112
See Anderson, 460 US at 789 (“[A reviewing court] must first consider the character
and magnitude of the asserted injury to the rights protected by the First and Fourteenth
Amendments that the plaintiff seeks to vindicate. It then must identify and evaluate the

46
framework fits within our well-established principles that legislation “is clothed in a

presumption of constitutionality” and that the challenger bears the burden of demonstrating

the statute’s constitutional infirmity. 113 Here, instead of concluding that the affidavit

requirement “is not an evenhanded restriction and the state has not shown how it protects

the integrity of the election process,” 114 the Court of Appeals should have recognized at

the outset that the affidavit requirement imposes only a minor burden on political speech,

and therefore, it is not subject to strict scrutiny. As a result, the usual presumption of

constitutionality afforded to the legislation is still controlling, and the burden remains with

the challengers to prove that the law is unconstitutional. 115

Although the majority opinion correctly recognizes that the affidavit requirement is

reviewed under the exacting-scrutiny standard, I disagree with its application of that

standard to plaintiffs’ facial challenge of that requirement. Plaintiffs have failed to show

how filing an affidavit with the Secretary of State imposes anything more than a de minimis

administrative burden on their ability to participate in the direct-democracy process. It is

clearly less of a burden than the absolute prohibition on paid petition circulators in Meyer

and the ban on distributing anonymous campaign literature in McIntyre. While those

precise interests put forward by the State as justifications for the burden imposed by its
rule.”) (emphasis added).
113
In re 2005 PA 71, 479 Mich at 11 (quotation marks and citation omitted).
114
LWV, ___ Mich App at ___; slip op at 20.
115
See Williams-Yulee v Florida Bar, 575 US 433, 444; 135 S Ct 1656; 191 L Ed 2d 570
(2015) (noting that, when strict scrutiny applies, the burden is on the state to show that a
restriction on speech is narrowly tailored to serve a compelling interest).

47
requirements outlawed specific categories of speech, the affidavit requirement at issue here

neither imposes a ceiling on speech nor prevents paid circulators from speaking. Nor have

plaintiffs shown that the affidavit requirement has any chilling effect on speech or that it

will deter paid circulators. Further, the affidavit requirement does not require disclosure

directly to the voter at the time petitions are circulated. The Court in ACLF observed that

an “affidavit is separated from the moment the circulator speaks,” and it therefore “does

not expose the circulator to the risk of ‘heat of the moment’ harassment.” 116 Given that the

affidavit must be filed before petitions are circulated, the circulator does not lose

anonymity with the voter and there is no risk of harassment.

“Election laws will invariably impose some burden upon” speech. 117 But where

plaintiffs “provide us scant evidence or argument beyond the burdens they assert disclosure

would impose,” and indeed, where “only modest burdens attend the disclosure . . . , we

must reject plaintiffs’ broad challenge” to the law. 118 At best, we are left to speculate about

the extent to which plaintiffs find the affidavit requirement unduly burdensome as to paid

circulators and their speech. Such speculation is insufficient to establish a facial challenge

to a statute that is presumed constitutional.

At the same time, the state undoubtedly has a legitimate and important interest in

preserving the integrity of the electoral process. As the Supreme Court has recognized,

this “interest in preserving electoral integrity is not limited to combating fraud. [It] extends

116
ACLF, 525 US at 198-199.
117
Burdick, 504 US at 433.
118
Reed, 561 US at 201.

48
to efforts to ferret out invalid signatures caused not by fraud but by simple mistake, such

as duplicate signatures or signatures of individuals who are not registered to vote in the

State.” 119 Here, the affidavits can be cross-checked with the petitions themselves to ensure

reporting compliance. And it can hardly be disputed that the state has a legitimate and

important interest in detecting and rooting out fraud in order to preserve the integrity of,

and promote accountability in, the electoral process. 120 Accordingly, I disagree with the

majority that plaintiffs have satisfied their heavy burden of showing that the affidavit

requirement is facially unconstitutional.

119
Id. at 198.
120
Although Meyer and ACLF held that there was no evidence that paid circulators are
more or less likely to commit fraud, those cases were decided in 1988 and 1999,
respectively. In 2010, the Supreme Court in Reed held that a statute that permits public
disclosure of the names and addresses of the signers of petitions does not violate the First
Amendment because it helps preserve the integrity of the electoral process by combating
fraud, detecting invalid signatures, and fostering government transparency and
accountability. Reed, 561 US at 197. In reaching that conclusion, the Court noted the
state’s citation of “a number of cases of petition-related fraud across the country,” thus
supporting the notion that “[t]he threat of fraud in this context is not merely hypothetical.”
Id. at 197-198. Also, in Husted, the circuit court in 2014 held that there was evidence that
paid circulators were more apt to commit fraud than volunteer circulators and, indeed, there
was evidence in Husted that paid circulators had committed fraud. Even more recently, in
Unlock Mich v Bd of State Canvassers, ___ Mich ___ (July 9, 2021) (Docket No. 162949),
intervening defendant Keep Michigan Safe challenged the certification of the initiative
petition at issue by alleging that signature-gathering efforts needed to be investigated by
the Board of State Canvassers for fraud. This Court ultimately ordered that the petition be
certified to the Legislature because the Board, lacking a majority willing to investigate
further, had a clear legal duty to certify the proposal. Our order did not address the merits
of the challengers’ fraud allegations, but the episode illustrates the fact that fraud remains
a recurring accusation in the context of petition drives.

49
 IV. CONCLUSION

For the reasons stated above, I concur with the majority that the geographic-

distribution requirement is unconstitutional as to initiative and referendum petitions. I also

concur with the majority that the checkbox requirement passes constitutional muster. And

given the unique circumstances presented in this case involving the rights of the people to

exercise direct democracy, I also concur that the Court’s opinion should apply

prospectively only. Nonetheless, I respectfully and vigorously dissent from the majority’s

conclusion that the affidavit requirement and the geographic-distribution requirement as to

voter-initiated constitutional amendments are unconstitutional. The challengers to these

two legislative provisions have utterly failed to overcome the strong presumption of

constitutionality accorded all legislation duly passed through a bicameral legislature and

signed by the Governor. Because the will of the people as clearly expressed through their

elected representatives has been thwarted by this Court’s improvident exercise of raw

judicial power, I dissent.

Brian K. Zahra
David F. Viviano

50
 STATE OF MICHIGAN

SUPREME COURT

LEAGUE OF WOMEN VOTERS OF
MICHIGAN, PROGRESS MICHIGAN,
COALITION TO CLOSE LANSING
LOOPHOLES, and MICHIGANDERS FOR
FAIR AND TRANSPARENT ELECTIONS,
Plaintiffs-Appellees,

v Nos. 163711-2

SECRETARY OF STATE,
Defendant-Appellee,
and

ATTORNEY GENERAL,
Intervening Defendant-
Appellee.

LEAGUE OF WOMEN VOTERS OF
MICHIGAN, PROGRESS MICHIGAN,
COALITION TO CLOSE LANSING
LOOPHOLES, and MICHIGANDERS FOR
FAIR AND TRANSPARENT ELECTIONS,

Plaintiffs-Appellees,

v Nos. 163744-5

SECRETARY OF STATE,

Defendant-Appellee,
and

ATTORNEY GENERAL,

Intervening Defendant-
Appellant.
 LEAGUE OF WOMEN VOTERS OF
MICHIGAN, PROGRESS MICHIGAN,
COALITION TO CLOSE LANSING
LOOPHOLES, and MICHIGANDERS FOR
FAIR AND TRANSPARENT ELECTIONS,
Plaintiffs-Appellees,

v Nos. 163747-8

SECRETARY OF STATE,
Defendant-Appellant,
and
ATTORNEY GENERAL,
Intervening Defendant-
Appellee.

BERNSTEIN, J. (concurring in part and dissenting in part).
I agree with the majority opinion in large part. I write solely to express my

disagreement with Part IV(A) of the opinion, as I would hold that the checkbox requirement

is also unconstitutional.

MCL 168.482(7) states, in relevant part, “Each petition under this section must

provide at the top of the page check boxes and statements printed in 12-point type to clearly

indicate whether the circulator of the petition is a paid signature gatherer or a volunteer

signature gatherer.” As a disclosure requirement that burdens political expression, I agree

with the majority opinion that the checkbox requirement is subject to review under the

exacting-scrutiny standard.

In order to survive the exacting-scrutiny standard, there must be “a substantial

relation between the disclosure requirement and a sufficiently important governmental

interest.” Doe v Reed, 561 US 186, 196; 130 S Ct 2811; 177 L Ed 2d 493 (2010) (quotation

2
marks and citations omitted). “To withstand this scrutiny, the strength of the governmental

interest must reflect the seriousness of the actual burden on First Amendment rights.” Id.

(quotation marks and citations omitted). I agree with the majority that the alleged burden

imposed by the checkbox requirement is minimal. 1

However, I disagree that the strength of the governmental interest is sufficient to

overcome even the minimal burden imposed by the checkbox requirement. As noted by

the Court of Appeals:

[T]he state has an interest in offering information regarding the paid status
of a circulator to voters when they decide whether to sign an initiative
petition. . . . Transparency in the political process, especially transparency
that permits voters to “follow the money,” is a compelling state interest.
Giving voters knowledge of whether they are being asked to sign a petition
by a volunteer or a paid circulator is valuable in its own right, but so is
knowing the extent to which the petition has the funds to pay circulators.

League of Women Voters v Secretary of State, ___ Mich App ___, ___; ___ NW2d ___

(2021); slip op at 18. The majority opinion similarly notes that “increasing the amount of

information available to the voters is a legitimate state interest.” Ante at 30.

This stands in stark contrast to the conclusion of the first Court of Appeals panel to

address the constitutionality of the checkbox requirement, as that panel noted that “no real

governmental interest has been asserted, let alone been proven . . . .” League of Women

Voters v Secretary of State, 331 Mich App 156, 191; 952 NW2d 491 (2020). I agree. It

continues to be the case that the state does not even attempt to assert with any specificity

1
Although the majority merely assumes for the sake of argument that the checkbox
minimally burdens core political speech, I agree with the Court of Claims and the previous
Court of Appeals panel that “[t]his type of compelled disclosure discourages participation
in the petition circulation process and inhibits core political speech.” League of Women
Voters v Secretary of State, 331 Mich App 156, 191; 952 NW2d 491 (2020).

3
an important governmental interest advanced by MCL 168.482(7), other than “generally

stated interests in transparency and accountability.” League of Women Voters, 331 Mich

App at 191. But “[t]he simple interest in providing voters with additional relevant

information does not justify a state requirement that a writer make statements or disclosures

she would otherwise omit.” McIntyre v Ohio Elections Comm, 514 US 334, 348; 115 S Ct

1511; 131 L Ed 2d 426 (1995).

Although the burden imposed by the checkbox requirement is admittedly minimal,

the state has entirely failed to identify a sufficiently important governmental interest that

is advanced by this requirement. Central to my conclusion here is the fact that the state has

not identified how the presence of a checkbox would advance even its vaguely stated

interest in transparency. MCL 168.482(7) only requires that there be a checkbox that

indicates whether a circulator is paid. I do not see how the presence of a single checkbox

would permit voters to “follow the money,” given that the checkbox imparts no information

as to who might be funding an initiative petition or how much money they have; more

importantly, the state offers no explanation as to why this might be the case.

In sum, even if the burden posed by MCL 168.482(7) is minimal, the state’s failure

to justify even that minimal burden renders the checkbox requirement unconstitutional. In

all other respects, I join the majority opinion.

Richard H. Bernstein

4
 STATE OF MICHIGAN

SUPREME COURT

LEAGUE OF WOMEN VOTERS OF
MICHIGAN, PROGRESS MICHIGAN,
COALITION TO CLOSE LANSING
LOOPHOLES, and MICHIGANDERS FOR
FAIR AND TRANSPARENT ELECTIONS,
Plaintiffs-Appellees,

v Nos. 163711-2

SECRETARY OF STATE,
Defendant-Appellee,
and

ATTORNEY GENERAL,
Intervening Defendant-
Appellee.

LEAGUE OF WOMEN VOTERS OF
MICHIGAN, PROGRESS MICHIGAN,
COALITION TO CLOSE LANSING
LOOPHOLES, and MICHIGANDERS FOR
FAIR AND TRANSPARENT ELECTIONS,

Plaintiffs-Appellees,

v Nos. 163744-5

SECRETARY OF STATE,

Defendant-Appellee,
and

ATTORNEY GENERAL,

Intervening Defendant-
Appellant.
 LEAGUE OF WOMEN VOTERS OF
MICHIGAN, PROGRESS MICHIGAN,
COALITION TO CLOSE LANSING
LOOPHOLES, and MICHIGANDERS FOR
FAIR AND TRANSPARENT ELECTIONS,
Plaintiffs-Appellees,

v Nos. 163747-8

SECRETARY OF STATE,
Defendant-Appellant,
and
ATTORNEY GENERAL,
Intervening Defendant-
Appellee.

CLEMENT, J. (concurring in part and dissenting in part).
I concur in full with the Court’s analysis of why the “15% cap” for direct-democracy

signatures violates the Michigan Constitution, as well as with its decision that the

“checkbox requirement” for petition circulators to indicate whether they are paid or

volunteer complies with the First Amendment of the United States Constitution. I dissent

from its holding that the “affidavit requirement” violates the First Amendment, however,

for the reasons stated by Justice ZAHRA in his dissent.

I also dissent from the Court’s decision to give this opinion prospective-only effect.

First, I disagree that the factors from Pohutski v City of Allen Park, 465 Mich 675; 641

NW2d 219 (2002), have been satisfied. In particular, I disagree that “the extent of the

reliance on the old rule,” id. at 698, justifies prospective effect here. Neither petition

sponsor that would be affected by retroactive effect of this decision has submitted its

signatures yet, meaning that they both have time to adapt to this decision. Moreover, “there

2
is a serious question as to whether it is constitutionally legitimate for this Court to render

purely prospective opinions, as such rulings are, in essence, advisory opinions.” Wayne

Co v Hathcock, 471 Mich 445, 485 n 98; 684 NW2d 765 (2004). This is because “to accord

a holding only prospective application is, essentially, an exercise of the legislative power

to determine what the law shall be for all future cases, rather than an exercise of the judicial

power to determine what the existing law is and apply it to the case at hand.” Devillers v

Auto Club Ins Ass’n, 473 Mich 562, 587 n 57; 702 NW2d 539 (2005). While this Court

has some limited authority to issue advisory opinions under Const 1963, art 3, § 8, those

circumstances are not satisfied here. I have previously written about my view of the

importance of this Court’s carefully observing the limitations on its authority to issue

advisory opinions. See In re House of Representatives Request for Advisory Opinion

Regarding Constitutionality of 2018 PA 368 & 369, 505 Mich 884 (2019) (CLEMENT, J.,

concurring). I would not evade those restrictions under the guise of giving decisions

prospective-only effect, and I dissent from the Court’s decision to do so in this matter.

Elizabeth T. Clement

3