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.

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.

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 *5 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

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.

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 *8 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).

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.

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”).

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), *12 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

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.

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. *15 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 *16 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.

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 *18 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.

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

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

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).

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 *23 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

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.

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.” *26 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).

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 *28 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.

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 *30 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 *31 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 *32 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.

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.

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).]

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.

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.

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.

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

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.

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 *41 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.

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 *3 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.

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, *5 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).

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,

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).

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 <https://www.ncsl.org/research/elections-and-campaigns/chart-of-the-initiative-states.aspx> (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.

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).

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).

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).

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).

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

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.”).

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

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).

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

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).

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.

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.

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) <https://www.michigan.gov/documents/sos/Initia_Ref_Under_Consti_12-08_339399_7.pdf> (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.

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).

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, wellorganized 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.

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.

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).

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.).

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.

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,

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-

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).

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

§ 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).]

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.”).

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

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).

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.

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.

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.

(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.

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.

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).

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).

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.

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.

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.

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

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).

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.

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.

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

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 *3 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).

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

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 *3 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