CHELSEA COLLABORATIVE, INC., & others [Note 1] vs. SECRETARY OF THE COMMONWEALTH & others. [Note 2]

BUDD, J. We are asked to determine the constitutionality of a statutory scheme requiring registration at least twenty days prior to election day in order for an otherwise qualified voter to vote in that election. See G. L. c. 51, §§ 1F, 26, 34. For the reasons that follow, we conclude that the twenty-day blackout period for voter registration prior to an election does not violate the Massachusetts Constitution. However, we further conclude that, having chosen to impose a deadline for voter registration prior to an election, the Legislature has a continuing duty to ensure that the deadline is no further from election day than what the Legislature reasonably believes is consistent with the Commonwealth's interest in conducting a fair and orderly election. [Note 3]

Background. We summarize the history of voter registration requirements in the Commonwealth as well as the facts and procedural history of this case, reserving certain details for discussion of specific issues.

Page 29

1. The voter registration statute. Massachusetts law requires those planning to vote in an election to register in advance; as of 1993, prospective voters must do so at least twenty days prior to election day. G. L. c. 51, § 26, as amended through St. 1993, c. 475, § 6. [Note 4] See G. L. c. 51, §§ 1 (requiring voters to comply with requirements of G. L. c. 51 in order to vote), 1F (establishing voter registration deadline twenty days prior to election day for presidential and vice-presidential elections), and 34 (registrars may not register individuals to vote for upcoming election after voter registration deadline).

The Commonwealth has a long history of regulating the right to vote by way of voter registration laws. [Note 5] See St. 1822, c. 104, § 2. In 1874, the Legislature enacted a law that permitted qualified citizens to register to vote at a registration session the day before elections in cities and towns with more than 1,000 inhabitants and at a registration session within forty-eight hours of elections and

Page 30

again one hour before the election meeting in other towns. St. 1874, c. 376, §§ 8, 9, 13. See St. 1874, c. 60 (setting forth voter registration requirements for Boston).

In 1877 and 1879, the Legislature first established a longer blackout period between the deadline to register and election day for cities and for towns respectively. See St. 1877, c. 235, § 2; [Note 6] St. 1879, c. 37, § 1. [Note 7] Subsequently, the Legislature made numerous adjustments to the voter registration deadline before enacting a twenty-day blackout period for voter registration applicable to cities in 1894 and Statewide in 1928. See St. 1928, c. 103, § 1; [Note 8] St. 1894, c. 271, §§ 1-2; St. 1893, c. 417, § 40; St. 1892, c. 351, §§ 15-18; St. 1884, c. 298, § 37. Beginning in 1947, for a time, the registration deadline was over thirty days before election day. St. 1947, c. 34, § 1. [Note 9] In 1973, the blackout period was reduced to twenty-eight days before election day. St. 1973, c. 853, § 1. [Note 10]

In 1993, Congress enacted the National Voter Registration Act, commonly known as the "motor voter" law. [Note 11] Pub. L. 103-31, 103d Cong., 1st Sess., 107 Stat. 77 (1993). The Federal motor voter law provided that State voter registration blackout periods may not be longer than thirty days prior to any Federal election. 52 U.S.C. § 20507. Shortly thereafter, the Legislature enacted a State version of the motor voter law, in which it returned the voter registration deadline for all elections in the State to twenty days

Page 31

prior to an election, where it remains today. St. 1993, c. 475, § 6, amending G. L. c. 51, § 26.

In 2014, the Legislature authorized "early voting" for any biennial State election, permitting all voters who register by that deadline to vote earlier than election day. See St. 2014, c. 111, § 12, inserting G. L. c. 54, § 25B. [Note 12]

2. Factual and procedural history. The plaintiffs comprise two voter registration organizations and an individual who registered to vote less than twenty days before the November, 2016, election and sought to vote in that election. [Note 13] The plaintiffs filed a complaint on November 1, 2016, in the Superior Court against the Secretary of the Commonwealth (Secretary) and the election commissioner of Revere, the city clerk of Chelsea, and the chairman of the Somerville election commission (collectively, municipal defendants) for declaratory relief. [Note 14] The complaint sought a preliminary injunction allowing the three original individual plaintiffs to vote in the November, 2016, election.

Granting the request for a preliminary injunction, a Superior Court judge ordered the municipal defendants to accept and to count provisional ballots from the individual plaintiffs. After a bench trial, the judge declared G. L. c. 51, §§ 1, 1F, 26, and 34, to be "unconstitutional to the extent that their [twenty]-day deadline operates to deny constitutionally qualified voters the right to cast a ballot." [Note 15]

Page 32

The Secretary appealed, and this court granted the parties' joint application for direct appellate review. In nonjury cases, "[w]e accept the judge's findings of fact unless there is clear error." Silva v. Attleboro, 454 Mass. 165, 167 (2009). "However, 'we scrutinize without deference the legal standard which the judge applied to the facts.'" Id., quoting Kendall v. Selvaggio, 413 Mass. 619, 621 (1992).

Discussion. 1. The right to vote. "[V]oting has long been recognized as a fundamental political right and indeed the 'preservative of all rights.'" [Note 16] Massachusetts Pub. Interest Research Group v. Secretary of the Commonwealth, 375 Mass. 85, 94 (1978), quoting Yick Wo v. Hopkins, 118 U.S. 356, 370 (1886). The Constitution of the Commonwealth expressly protects the right to vote for qualified voters in both art. 9 of the Massachusetts Declaration of Rights [Note 17] and in art. 3 of the Amendments to the Massachusetts Constitution, as amended (art. 3). [Note 18], [Note 19]

We have established that the fundamental right to vote is also

Page 33

implicitly protected under other provisions of the Declaration of Rights. See Dane v. Registrars of Voters of Concord, 374 Mass. 152, 160 (1978) (right to vote is protected as "natural, essential, and unalienable right[]" under art. 1 of Declaration of Rights [citation omitted]); Swift v. Registrars of Voters of Quincy, 281 Mass. 271, 276 (1932) ("The right to vote is a precious personal prerogative to be sedulously guarded" under "[a]rts. 4, 7, 8, [and] 9 of the Declaration of Rights"); Attorney Gen. v. Suffolk County Apportionment Comm'rs, 224 Mass. 598, 601 (1916) ("The right to vote is a fundamental personal and political right" protected under arts. 1 through 9 of Declaration of Rights).

Simultaneously, the Constitution provides the Legislature with broad authority as part of the State's police power, to enact reasonable laws and regulations that are, in its judgment, appropriate. Part II, c. 1, § 1, art. 4, of the Constitution of the Commonwealth. See Massachusetts Comm'n Against Discrimination v. Colangelo, 344 Mass. 387, 395 (1962).

Because the right to vote is a fundamental one protected by the Massachusetts Constitution, a statute that significantly interferes with that right is subject to strict judicial scrutiny. See Cepulonis v. Secretary of the Commonwealth, 389 Mass. 930, 932, 935 (1983) (subjecting statutory scheme that did not permit qualified citizens to register to vote in State elections during their incarceration to strict scrutiny); Langone v. Secretary of the Commonwealth, 388 Mass. 185, 196, cert. denied, 460 U.S. 1057 (1983) ("Strict scrutiny is required if the interests asserted by the plaintiffs are fundamental and the infringement of them is

Page 34

substantial"). [Note 20] See also Doe No. 1 v. Secretary of Educ., 479 Mass. 375, 392 (2018), quoting Zablocki v. Redhail, 434 U.S. 374, 386 (1978) ("only a statute that 'significantly interfere[s] with' the fundamental right at issue burdens that right and justifies application of strict scrutiny"). By contrast, statutes that do not significantly interfere with the right to vote but merely regulate and affect the exercise of that right to a lesser degree are subject to rational basis review to assure their reasonableness. See McSweeney v. Cambridge, 422 Mass. 648, 656 (1996); Kinneen v. Wells, 144 Mass. 497, 499-500 (1887); Capen v. Foster, 12 Pick. 485, 490 (1832). See also Lee v. Commissioner of Revenue, 395 Mass. 527, 530 (1985) ("not every statute that affects [a fundamental right] must be supported by a compelling State interest").

The parties dispute the standard of review that applies in this case. The plaintiffs argue that the appropriate standard to apply to the voter registration deadline is a "necessity" test, to be applied in a manner functionally similar to strict scrutiny. The plaintiffs derived the necessity test from language in Kinneen, 144 Mass. at 499, where this court provided that any legislation diminishing the right to vote must be "defended on the ground that it is reasonable and necessary."

As discussed infra, we reaffirm the court's reasoning in Kinneen. However, we do not interpret the word "necessary" as used in Kinneen to be "fused with special meaning" from more modern jurisprudence such that strict scrutiny is always the applicable standard for reviewing regulations on the right to vote. Cf. United States Trust Co. of N.Y. v. New Jersey, 431 U.S. 1, 54 n.17 (1977) (Brennan, J., dissenting). Instead, the language from Kinneen contemplates a significant role for the Legislature in determining which regulations are appropriate. See Cole v. Tucker, 164 Mass. 486, 489 (1895) (use of official ballots as they "are such as may properly be deemed necessary by the Legislature"); 1 Op. Atty. Gen. 54, 55 (1899) (voter registration regulations "can be sustained only if it is a necessary or reasonable regulation for the purpose in view; and this is largely a question for the judgment of the Legislature"). Our inquiry recognizes different

Page 35

levels of scrutiny depending on the substantiality of the interference with the voting right. See Kinneen, 144 Mass. at 501, quoting Capen, 12 Pick. at 489 (statutes that "subvert or injuriously restrain the right" to vote "under the pretence and color of regulating" are beyond legislative power).

In arguing that a "sliding scale" standard of review applies, the Secretary relies heavily on Libertarian Ass'n of Mass. v. Secretary of the Commonwealth, 462 Mass. 538, 558 (2012), where we clarified that art. 9 does not extend any ballot access protections beyond the Federal constitutional requirements. But see Glovsky v. Roche Bros. Supermkts., Inc., 469 Mass. 752, 762 (2014) (clarifying holding in Batchelder); Batchelder v. Allied Stores Int'l, Inc., 388 Mass. 83, 88-89 (1983) (unlike Federal Constitution, infringements on electoral candidate rights under art. 9 do not require State action). Applying the Federal analysis, we said that "[r]egulations imposing severe burdens on a plaintiffs' rights must be narrowly tailored and advance a compelling state interest. Lesser burdens . . . trigger less exacting review, and a State's 'important regulatory interests' will usually be enough to justify 'reasonable, nondiscriminatory restrictions.'" Libertarian Ass'n of Mass., supra at 560, quoting Timmons v. Twin Cities Area New Party, 520 U.S. 351, 358 (1997).

In general, this "sliding scale" analytical framework is appropriate for cases that involve voting rights under the Massachusetts Constitution because that framework reflects both our Constitution's numerous provisions granting qualified citizens the fundamental right to vote and its grant of police power to the Legislature, which we have concluded authorizes the Legislature to regulate that right. See, e.g., Cole, 164 Mass. at 488 ("principal question then is, whether [elections law] . . . is a reasonable regulation of the manner in which the right to vote shall be exercised, or whether it subverts or injuriously restrains the exercise of that right"); Kinneen, 144 Mass. at 499-500; Capen, 12 Pick. at 490.

However, in this case and others, there may be circumstances where the Massachusetts Declaration of Rights and art. 3 require application of this analysis in a manner that "guard[s] more jealously against the exercise of the State's police power" than the application of the framework under the Federal Constitution. Blue Hills Cemetery, Inc. v. Board of Registration in Embalming & Funeral Directing, 379 Mass. 368, 373 n.8 (1979), quoting Coffee-Rich, Inc. v. Commissioner of Pub. Health, 348 Mass. 414,

Page 36

421 (1965). See Goodridge v. Department of Pub. Health, 440 Mass. 309, 328 (2003) ("The Massachusetts Constitution protects matters of personal liberty against government incursion as zealously, and often more so, than does the Federal Constitution, even where both Constitutions employ essentially the same language"); Planned Parenthood League of Mass., Inc. v. Attorney Gen., 424 Mass. 586, 590 (1997) ("we must accept responsibility for interpreting our own Constitution as text, precedent, and principle seem to us to require"). [Note 21]

Before making a determination as to the constitutionality of the twenty-day requirement, we must determine which level of scrutiny to apply.

2. The burden of registration requirements on the right to vote. Whether to apply the rational basis test or strict scrutiny to the requirement that a prospective voter register twenty days in advance of an election depends on whether that requirement significantly interferes with the fundamental right to vote. [Note 22] See

Page 37

Doe No. 1, 479 Mass. at 392; McSweeney, 422 Mass. at 656; Cepulonis, 389 Mass. at 932, 935.

Although we have not before directly reviewed the constitutionality of the length of a voter registration blackout period prior to an election, we have certainly acknowledged the existence of a preregistration system for voting. See Kinneen, 144 Mass. at 500 ("It is not an unreasonable provision that all persons entitled as voters shall be registered as such previously to depositing their ballots . . ."); Capen, 12 Pick. at 488. Thirty-five years ago, we implied in dicta that a voter registration blackout period longer than twenty days before election day would not likely violate the Massachusetts Constitution. See Cepulonis, 389 Mass. at 937 ("The time limit in [Rosario v. Rockefeller, 410 U.S. 752 (1973),] did not absolutely disenfranchise voters or deprive them of the right to vote for a lengthy period").

However, we acknowledge that, with the passage of time, voting regulations once considered constitutionally permissible may come to significantly interfere with the fundamental right to vote in light of conditions existing in contemporary society. Cf. Goodridge, 440 Mass. at 341 n.33 ("We are concerned with the operation of challenged laws on the parties before us, and we do not inhibit our inquiry on the ground that a statute's original enactors had a benign or at the time constitutionally unassailable purpose"). What was perhaps a reasonable regulation that insignificantly interfered with the right to vote thirty-five, one hundred, or 200 years ago may be considered to significantly interfere with the exercise of that right today in light of technological change and the reasonable expectations of Massachusetts citizens. Cf. Columbia Broadcasting Sys., Inc. v. Democratic Nat'l Comm., 412 U.S. 94, 102 (1973) (where technological change is dynamic, reasonable regulatory "solutions adequate a decade ago are not necessarily so now, and those acceptable today may well be outmoded [ten] years hence").

The question before us is whether, at this point in time, a twenty-day voter registration blackout period prior to an election is a substantial enough interference with the right to vote to justify the application of strict scrutiny. See Doe No. 1, 479 Mass. at 392 n.30. Because the deadline does not disenfranchise any voter, because the Commonwealth takes sufficient steps to minimize

Page 38

the number of qualified voters who miss it, and because registration itself is sufficiently simple and accessible, we conclude that the deadline does not require the application of strict scrutiny. [Note 23]

First, we give great weight to the fact that the registration deadline does not disenfranchise any voter. See Cepulonis, 389 Mass. at 937. Cf. Rosario, 410 U.S. at 758. Those in the individual plaintiff's position were free to register prior to the deadline and would have been eligible to vote in the 2016 election had they merely looked into what is required to register and done so. Importantly, the registration deadline does not apply to categories of qualified voters for whom the deadline is more likely to pose a burden. "Specially qualified voters," including Massachusetts residents who are absent from the Commonwealth during the seven days immediately preceding the voter registration deadline [Note 24] or who become a citizen of the United States after the registration deadline but before an election, are exempt. See G. L. c. 51, §§ 50, 51; G. L. c. 50, § 1; note 4, supra.

In addition, the record suggests that the Secretary undertakes

Page 39

sufficient actions to inform the public about the registration deadline. The Secretary mails every household in the Commonwealth an "Information for Voters" booklet that includes a voter registration form and information about the deadline with instructions in Spanish and Chinese as to how to obtain a translated version of the booklet in those languages. Copies of the booklet are provided to group homes, city and town halls, public libraries, senior centers, and various community organizations. Voter registration forms, which note the deadline, are available at municipal offices, post offices, and libraries; online; and via organizations conducting voter registration drives. Information is also disseminated through public service announcements on television stations and newspapers.

As for registration itself, the record contains ample evidence that the Commonwealth has taken great steps to ensure that the process is simple and accessible. A citizen can register to vote in multiple ways, including at his or her city or town hall, at any registry of motor vehicles location, or through any State agency "that provide[s] public assistance or assistance to people with disabilities." See G. L. c. 51, § 42F, 42G; G. L. c. 50, § 1. Mail-in voter registration forms are accepted for the next occurring election as long as they are postmarked before midnight on the deadline. See G. L. c. 51, § 26; 950 Code Mass. Regs. § 57.04(4)(b) (2017). In addition, online registration is possible up until midnight on the registration deadline. G. L. c. 51, § 33A. The registration process merely involves completing a brief form. [Note 25] Voter registration is free, and, once a voter is registered, the registration never expires. [Note 26]

We believe that a voter registration blackout period could be established that is so far from election day that, notwithstanding

Page 40

notification of that deadline and ease of registration, it nonetheless would significantly interfere with the right to vote. In 1973, the United States Supreme Court concluded that a "[fifty]-day registration period approaches the outer constitutional limits in this area" under the Federal Constitution. Burns v. Fortson, 410 U.S. 686, 687 (1973) (per curiam). Although today, at least under the Commonwealth's Constitution, the outer limits of a deadline that does not significantly interfere with the right to vote do not likely extend quite that far, we believe that the twenty-day period at issue falls within them. [Note 27]

Considering a totality of these factors, we conclude that a requirement that prospective voters register twenty days in advance of an election does not pose a significant interference with the fundamental right to vote under the Massachusetts Constitution so as to require the application of strict scrutiny. Therefore, we apply the rational basis test to the voter registration law to determine its reasonableness. See McSweeney, 422 Mass. at 656.

3. Rational basis review of G. L. c. 51, § 26. As a matter of due process, rational basis analysis requires that statutes "bear[] a real and substantial relation to the public health, safety, morals, or some other phase of the general welfare." Commonwealth v. Wilbur W., 479 Mass. 397, 403 (2018), quoting Goodridge, 440 Mass. at 330. As a matter of equal protection, "the rational basis test 'includes a requirement that an impartial lawmaker could logically believe that the classification would serve a legitimate public purpose that transcends the harm to the members of the disadvantaged class.'" Doe No. 1, 479 Mass. at 393, quoting English v. New England Med. Ctr., Inc., 405 Mass. 423, 429 (1989), cert. denied, 493 U.S. 1056 (1990).

The requirement that voters register before exercising their fundamental right to vote is supported by the legislative objective of conducting orderly and legitimate elections. See Capen, 12 Pick. at 492 (registration is reasonable method of determining constitutional qualifications of voters). See also Swift, 281 Mass. at 276 ("The regnant design of all election laws is to provide expeditious and convenient means for expression of the will of the voters free from fraud"). Further, such registration may be

Page 41

required prior to election day. See Capen, supra at 492 ("There is no express requirement, and we think there is no implication, arising either from the terms of the constitution or from the nature and purposes of the right of voting, which obliges [cities and towns] to [determine whether voters are constitutionally qualified], whilst in the actual performance of other positive duties" on election day).

The record contains ample evidence that it would be rational for the Legislature to conclude that some deadline is necessary prior to election day in order to achieve these legitimate public purposes.

Local election officials are responsible for processing voter registration applications, screening them for errors, and confirming voter qualifications. During the time immediately preceding an election, elections officials have a variety of tasks in addition to processing these forms. For example, these administrators must recruit and train poll workers, test elections-related machines, and print voter lists prior to the commencement of election day voting. [Note 28]

We do not minimize the harm that the deadline imposes on those members of our society who fail to register by the deadline. See note 28, supra. Even if individuals who fail to register by the deadline bear responsibility for their registration status on election day, the fact remains that these constitutionally qualified citizens are unable to exercise a core right under our Constitution. [Note 29] We recognize that, all else being equal, the closer the registration deadline is to election day, the fewer the number of

Page 42

individuals there are who are unable to exercise their right to vote because they missed that deadline.

Given the harm that the deadline imposes on those who miss it, and recognizing the need to balance the strong interests of voters with the limited permissible public purposes in regulating elections, this court has previously articulated the Legislature's specific duty in drawing a voter registration deadline substantially related to those limited legitimate public purposes:

"No system would be just that did not extend the time of registration up to a time as near that of actually depositing the votes as would be consistent with the necessary preparation for conducting the election in an orderly manner and with a reasonable scrutiny of the correctness of the list."

Kinneen, 144 Mass. at 502. [Note 30] A voter registration deadline that is not as near to election day as the Legislature reasonably believes is appropriate given the need for election officials to prevent fraud, prepare for election day, and conduct election day responsibilities does not "bear[] a real and substantial relation" to those legitimate purposes. Goodridge, 440 Mass. at 330, quoting Coffee-Rich, Inc., 348 Mass. at 422. Nor, in such a case, could lawmakers logically believe that the deadline prevents fraud and ensures an orderly election in a manner that "transcends the harm to" those who may not vote because they failed to register by the deadline. Doe No. 1, 479 Mass. at 393, quoting English, supra at 429.

Thus, we turn to whether the twenty-day voter registration deadline reflects a reasonable legislative determination that the deadline is set as near as possible to election day as consistent with the need to maintain an orderly election. See Kinneen, 144

Page 43

Mass. at 503. Cf. Marston v. Lewis, 410 U.S. 679, 680 (1973) (upholding State's voter registration deadline because it "reflects a state legislative judgment that the period is necessary to achieve the State's legitimate goals").

The legislative history suggests that the process for setting that twenty-day deadline was a deliberate one. In 1993, the Legislature considered a bill cosponsored by fifty legislators and supported by twenty organizations that would have lowered the voter registration deadline incrementally downward each year from the then-current twenty-eight days to seven days in 1996. See 1993 House Doc. No. 3931, § 3. See also the testimony of Rep. Marc Draisen in favor of House Doc. No. 3931 (Mar. 29, 1993), in the papers of Marc Draisen (on file at State Library, special collections). The Legislature referred that bill to the joint committee on election laws. One legislator's notes from a September 14, 1993, executive session of that committee suggest that, after considering alternatives, the committee's decision to reduce the deadline from twenty-eight days to twenty was based in part on a conclusion that twenty days were necessary at the time in light of the uncertainties regarding the other election law reforms that the Legislature was considering as part of the 1993 election law reform package. [Note 31] See the notes from the joint committee on election laws committee, executive session (Sept. 14, 1993), in the papers of Marc Draisen (on file at State Library, special collections). See also Voter Registration, State House News Service, Sept. 9, 1993 (reporting that House election laws "[c]ommittee chairman . . . said municipal clerks need time to adjust to the new law and shortening the deadline [beyond twenty days] would be an administrative burden").

Thus, the legislative history suggests a reasoned basis for concluding that a twenty-day blackout period between the end of

Page 44

voter registration and election day was necessary. [Note 32] See St. 1993, c. 475, §§ 6, 54. Importantly, however, in addition to changing the deadline, the statute created a "voter registration reform advisory commission" tasked with "prepar[ing] a report on the appropriateness of the deadline date for registration." [Note 33] St. 1993, c. 475, § 54.

Unfortunately, it appears that this commission never met or studied the appropriateness of the twenty-day deadline. See Report of the Senate Committee on Post Audit and Oversight, Driving Voters Away?: Implementation of Motor Voter Laws in Massachusetts, 1997 Senate Doc. No. 1775, at 26 (May 8, 1997) ("the [c]ommission has never held a single meeting").

In 2014, the Legislature enacted "An Act relative to election laws" which, among other things, implemented early voting in the Commonwealth. See St. 2014, c. 111. The act created an "elections task force" to "undertake a study of election issues," such as "more accessible voter registration, including, but not limited to, same-day registration." St. 2014, c. 111, § 16. The task force's report was due August 1, 2017. St. 2014, c. 111, § 16 (c). Although the Massachusetts Town Clerks Association spent considerable time preparing to provide input, here, too, the task force never met and a report was never submitted.

Although the Legislature appeared to have a reasoned basis for requiring voters to register twenty days in advance of an election

Page 45

in 1993, the mechanisms put in place for a periodic review of that requirement seem to have failed. [Note 34] Thus, we have a concern that, given the passage of time, the reasoned basis for the twenty-day blackout period may need to be reconsidered. Cf. Shelby County, Ala. v. Holder, 570 U.S. 529, 546, 554 (2013) (concluding, in controversial application, that statute that was rational when written had grown to be irrational and to bear "no logical relation to the present day"); Columbia Broadcasting Sys., Inc., 412 U.S. at 102 (where technological change is dynamic, reasonable regulatory solutions "adequate a decade ago are not necessarily so now, and those acceptable today may well be outmoded [ten] years hence"); Burns, 410 U.S. at 689 (Marshall, J., dissenting) ("Even if we would be inclined to defer to a recent and informed legislative determination of necessity" for voter registration deadline, "statutes here were adopted nearly a decade ago"). The Legislature has an ongoing duty to ensure that voters may register "up to a time as near that of actually depositing the votes as would be consistent with the necessary preparation for conducting the election in an orderly manner and with a reasonable scrutiny of the correctness of the list." Kinneen, 144 Mass. at 502.

At the same time, there is a strong presumption that the Legislature does not act arbitrarily. Carleton v. Framingham, 418 Mass. 623, 631 (1994) ("In any evaluation of reasonableness, the plaintiffs have a heavy burden to meet, and we will recognize every rational presumption in favor of the legislation"). This presumption is rooted in the separation of powers and need for judicial restraint. "It is not . . . easy for courts to step in and say that what was rational in the past has been made irrational by the passage of time, change of circumstances, or the availability of new knowledge. Nor should it be. Too many issues of line drawing make such judicial decisions hazardous. . . . What degree of legislative action, or of conscious inaction, is needed when that (uncertain) point is reached? These difficulties -- and many others -- counsel restraint, and do so powerfully." United States v. Then, 56 F.3d 464, 468 (2d Cir. 1995) (Calabresi, J.,

Page 46

concurring). There is nothing on the record before us to suggest that the Legislature's standing joint committee on election laws [Note 35] has not continued to take a hard look at the deadline. [Note 36] At the least, as the Legislature passed the statute imposing the current twenty-day blackout period, we can assume that the body has chosen, up to this point, not to change it. Therefore, we are inclined to conclude that, at least for the time being, an impartial lawmaker could logically believe that the voter registration deadline imposed twenty days prior to election day still serves legitimate public purposes that transcend the harm to those who may not vote. We assume that the legislative inaction in this area represents a conscious conclusion that the deadline remains "as near [election day] as would be consistent with the necessary preparation for conducting the election in an orderly manner and with a reasonable scrutiny of the correctness of the list." Kinneen, 144 Mass. at 502.

Conclusion. For the foregoing reasons, the judgment is vacated, and the case is remanded to the Superior Court for entry of a judgment consistent with this opinion.

So ordered.

Page 47

GANTS, C.J. (concurring, with whom Lenk and Gaziano, JJ., join). I agree with the court that the twenty-day voter registration requirement challenged here is not unconstitutional. I write separately only to articulate what I believe is the appropriate standard of review for laws that regulate the right to vote under the Massachusetts Constitution. I would review such laws using a "sliding scale" approach, similar to that articulated by the Federal courts, where the level of scrutiny is calibrated to match the burden imposed on the right to vote. However, because our Declaration of Rights is more protective of the right to vote than the Federal Constitution, I would adopt a version of that approach that is more protective than the Federal jurisprudence, applying strict scrutiny where a law imposes a substantial burden on the right to vote, and reviewing laws that impose lesser burdens under a sliding scale.

In reviewing voting regulations, courts are confronted with two significant but competing interests. On one side is the fundamental right to vote, which, being "preservative of other basic civil and political rights," cannot be restricted without "careful[] and meticulous[]" judicial scrutiny. Reynolds v. Sims, 377 U.S. 533, 562 (1964). On the other side, however, is the long-recognized authority of the State to regulate elections "to ensure that our democratic processes remain fair, honest, and orderly." Libertarian Ass'n of Mass. v. Secretary of the Commonwealth, 462 Mass. 538, 560 (2012) (LAM). Recognizing the need to balance these important interests, Federal courts have eschewed the usual predetermined levels of scrutiny when reviewing State election laws, adopting instead a more flexible sliding scale approach. See id. Under this approach, not every law that regulates the right to vote is subject to strict scrutiny. Rather, the level of scrutiny is calibrated to reflect the degree to which the law burdens the right.

This approach was first articulated in Anderson v. Celebrezze, 460 U.S. 780, 789 (1983), where the United States Supreme Court outlined the specific steps a court must take when reviewing a nondiscriminatory State election law. First, the court "must . . . consider the character and magnitude" of the burden imposed on the right to vote. Id. Then, it must "evaluate the precise interests put forward by the State as justifications for [that] burden" and determine "the extent to which those interests make it necessary to burden" that right. Id. In other words, the sliding scale approach requires courts to balance the burden on the plaintiffs' voting rights against the State's interest in regulating

Page 48

elections, so that the greater the burden on voting rights, the greater the State's interest must be -- and vice versa. "Only after weighing all these factors is the reviewing court in a position to decide whether the challenged [law] is unconstitutional." Id.

The purpose of the sliding scale approach is to reject simple categorization in favor of a more fine-tuned analysis. As the Supreme Court explained in Anderson, "[c]onstitutional challenges to . . . a State's election laws" -- because they implicate such strong competing interests -- "cannot be resolved by any 'litmus-paper test' that will separate valid from invalid restrictions." Id., quoting Storer v. Brown, 415 U.S. 724, 730 (1974). Instead, a court must engage in a careful balancing of the interests at stake. See Anderson, supra. Moreover, "[t]he results of this evaluation will not be automatic." Id. Where voting rights are implicated, "there is 'no substitute for the hard judgments that must be made.'" Id. at 789-790, quoting Storer, supra.

We adopted this sliding scale approach in LAM, 462 Mass. at 557, where we reviewed ballot access provisions under the Massachusetts Constitution. In describing this approach, we stated:

"Regulations imposing severe burdens on plaintiffs' rights must be narrowly tailored and advance a compelling state interest. Lesser burdens . . . trigger less exacting review, and a State's 'important regulatory interests' will usually be enough to justify 'reasonable, nondiscriminatory restrictions.'"

Id. at 560, quoting Timmons v. Twin Cities Area New Party, 520 U.S. 351, 358 (1997). We then went on to conclude that, because the ballot access provisions imposed only "modest burdens" on the plaintiffs' rights, they required "only a rational basis . . . in order . . . to pass constitutional muster." LAM, 462 Mass. at 567, quoting Barr v. Galvin, 626 F.3d 99, 110 (1st Cir. 2010).

In LAM, we described the sliding scale approach with reference to its extremes, explaining that at one end, regulations imposing only "modest burdens" are reviewed for rational basis, LAM, 462 Mass. at 567, whereas at the other end, regulations imposing "severe burdens" will trigger strict scrutiny. Id. at 560. Many regulations, however, will fall somewhere "between these two extremes." Obama for Am. v. Husted, 697 F.3d 423, 429 (6th Cir. 2012). Where they do, we must not fall back on the familiar, predetermined standards of review, but instead apply "a more

Page 49

flexible standard," under which "the rigorousness of our inquiry . . . depends upon the extent to which a challenged regulation burdens" voters' rights. Burdick v. Takushi, 504 U.S. 428, 434 (1992). [Note Gants-1]

In reviewing the voter registration requirement here, the court appears to reject the Federal sliding scale approach, stating that, although "[i]n general, this [approach] is appropriate for cases that involve voting rights," ante at 35, the Massachusetts Constitution requires a more protective standard. Id. The court then engages in an analysis that more closely resembles the usual two-track approach, where courts must choose between strict scrutiny or rational basis review, with no options in between. Indeed, the court assumes at the outset that it must apply one standard or the other. See ante at 36. The court first considers whether the registration requirement constitutes a significant interference with the right to vote and, having concluded that it does not, proceeds to review it under the rational basis test. See ante at 40.

As I discuss later, I agree with the court that the Massachusetts Constitution can and should be more protective of the right to vote than the Federal Constitution. This does not mean, however, that we must depart from the Federal jurisprudence entirely. In LAM, we applied the sliding scale approach in the context of ballot access rights, but this flexible standard of review is equally important where voting rights are regulated, because the resulting burdens will often resist simple categorization. Compare, for example,

Page 50

a twenty-day registration requirement to a ten-day registration requirement. Although neither requirement may warrant strict scrutiny, the twenty-day requirement is undeniably more burdensome than the ten-day requirement. To review both requirements under the same standard, without calibrating it to match their respective burdens, would ignore the important differences between the two.

But this is the approach the court appears to take today, applying the rational basis test to all voting regulations, no matter how different, as long as they do not present "a significant interference with the fundamental right to vote" and thereby trigger strict scrutiny. Ante at 40. The court's approach is not a sliding scale, but rather a binary switch, whereby the universe of possible regulations can be divided into just two categories -- those that result in a "significant interference" and those that do not -- with strict scrutiny applying to one and rational basis review to the other. Under the court's approach, any regulation that results in less than a "significant interference" will be reviewed only for a rational basis, inviting courts to avoid making those "hard judgments" that are so essential where voting rights are at stake. Anderson, 460 U.S. at 790, quoting Storer, 415 U.S. at 730. Cf. United States v. Alvarez, 567 U.S. 709, 731 (2012) (Breyer, J., concurring in the judgment) (some alternative standard is needed where challenged law "warrants neither near-automatic condemnation [as 'strict scrutiny' implies] nor near-automatic approval [as is implicit in 'rational basis' review]"). [Note Gants-2]

Page 51

I would instead hew more closely to the Federal sliding scale approach, insofar as it rejects the usual two-track analysis. I believe this approach is consistent with our own constitutional jurisprudence concerning voting rights, which has consistently sought to balance the fundamental right to vote against the Commonwealth's authority to regulate elections. See Cole v. Tucker, 164 Mass. 486, 488 (1895) ("The principal question . . . is whether [the challenged law] is a reasonable regulation of the manner in which the right to vote shall be exercised, or whether it subverts or injuriously restrains the exercise of that right"). See also Kineen v. Wells, 144 Mass. 497, 499-500 (1887); Capen v. Foster, 12 Pick. 485, 495 (1832).

I would not, however, adopt the Federal sliding scale approach in its entirety. I emphasize that, although we may borrow from Federal jurisprudence, we are not thereby bound to follow that jurisprudence when we interpret our own State Constitution. We are, after all, always "free to interpret state constitutional provisions to accord greater protection to individual rights than do similar provisions of the United States Constitution." Goodridge v. Department of Pub. Health, 440 Mass. 309, 328 (2003), quoting Arizona v. Evans, 514 U.S. 1, 8 (1995). I would adopt the Federal sliding scale approach only in a limited sense, as a methodological guide on how to tailor judicial review to the different burdens that may be imposed on the right to vote. And, as part of this sliding scale approach, I would also set a different, more protective threshold for triggering strict scrutiny. This is because, as stated, I agree with the court that the Massachusetts Constitution is more protective of the right to vote than the Federal Constitution. And I also agree with the court's position that, under our own Constitution, something less than a "severe burden" will suffice to trigger strict scrutiny. See ante at 36 & note 21. I do not agree, however, with the court's decision to articulate the threshold for strict scrutiny as "a significant interference with the fundamental right to vote." Ante at 40. As I read it, the term "significant interference" is vulnerable to differing interpretations. The word "interference" seems to suggest that the

Page 52

restriction must pose so great a burden on voting as to actually interfere with voting, perhaps intentionally so, and therefore creates the risk that the strict scrutiny threshold will be set too high. At the same time, the word "significant" implies that any restriction that does not trigger strict scrutiny is insignificant, potentially setting the strict scrutiny threshold too low. I prefer that the threshold be set at "substantial burden" rather than at "significant interference" because, by speaking in terms of "burdens," as in Federal jurisprudence, we make clear that we are more protective of the right to vote than our Federal counterparts, and that -- whereas under the Federal Constitution, strict scrutiny is triggered only where there is a severe burden -- under the Massachusetts Constitution, strict scrutiny will be triggered at a lower threshold, where there is a substantial burden. See LAM, 462 Mass. at 560.

In short, I would hold that strict scrutiny is triggered where a regulation imposes a substantial burden on the right to vote, with regulations that impose less substantial burdens reviewed, consistent with the sliding scale approach, under a more flexible standard.

Applying this approach here, I note first that the burden imposed by the twenty-day registration requirement is not a "modest" one. LAM, 462 Mass. at 567. The judge specifically found that more than 5,500 otherwise qualified Massachusetts voters were unable to vote during the 2016 presidential elections because, although they had registered before election day, they had missed the twenty-day deadline. The judge also found that the registration requirement deters many more citizens from voting, noting that, because public interest in elections peaks only after the twenty-day deadline, potential voters often do not learn about issues or candidates they care about until it is already too late. Indeed, as the court notes, almost twenty per cent of the qualified Massachusetts voters who did not vote in 2014 cited the registration requirement as the reason, compared to only ten per cent nationally. Ante at note 27.

Having said that, I agree that the twenty-day registration requirement does not impose so substantial a burden on voters as to warrant strict scrutiny. I reach this conclusion for two reasons. First, as the court notes, the registration requirement does not bar any qualified voter from voting who is in fact incapable of registering in time. Qualified voters who are absent from the Commonwealth during the seven days preceding the deadline, or

Page 53

who become citizens or turn eighteen years old after the deadline, are exempted. G. L. c. 51, §§ 50, 51. See ante at 38. Second, the Commonwealth has made significant efforts to reduce the obstacles to registration, for example by widely publicizing the deadline and by establishing a simple and flexible registration process. See ante at 38-39. Thus, the registration requirement only bars from voting those who could have timely registered but, for whatever reason -- whether because they did not take the time to register or because they did not think ahead -- failed to do so.

Where the burden is not substantial but also not modest, I would, under the sliding scale approach, calibrate our standard of review to match this burden, applying something less exacting than strict scrutiny but more searching than mere rational basis review. The touchstone of our inquiry is whether the State's interest in regulating elections adequately justifies the burdens imposed on the voters' rights. Like other courts that have followed this approach, I would consider whether the Commonwealth's asserted interests in imposing the restriction are legitimate, whether the restriction in fact serves those interests, and whether the restriction is "precisely drawn" to do so, weighing each of these factors in order to ascertain whether the burden on voting rights is justified. See Anderson, 460 U.S. at 796-806. See also Burdick, 504 U.S. at 439-440. Here, I believe the Commonwealth's legitimate interest in ensuring orderly elections outweighs the burden that is imposed by the twenty-day registration requirement. As to the Commonwealth's interest in orderly elections, I recognize that some States have abandoned pre-election registration requirements, instead allowing election day registration, but I also recognize that adopting such a system in Massachusetts at this time may entail substantial administrative challenges. As to the burden imposed by the twenty-day registration requirement, I reiterate that it is not a modest one, but I also recognize that the Commonwealth has taken considerable steps to mitigate that burden. Ante at 38-39. Thus, in this case, I reach the same conclusion as the court. But in other cases, how a court chooses its standard of review may very well tilt the balance and affect the outcome. I believe that a sliding scale approach, compared to the usual two-track analysis, gives courts the latitude needed to most effectively balance the competing interests at stake.

Another reason to adopt the sliding scale approach is that the balance between the State's interest in regulating elections and

Page 54

the burdens imposed on voters may change over time. As the court acknowledges, ante at 37, 44-45, a registration requirement that plays an important role in ensuring orderly elections today may become less important over time, as the technology of voter identification advances and as other States demonstrate the ability to run an orderly election without pre-election registration. Similarly, the burden posed by a registration requirement may increase over time, as more and more voters, accustomed to instantaneous transactions and decision-making, come to plan their lives on ever shorter timelines. Under the sliding scale approach, we need not wait for these burdens to become "severe" or even substantial in order to review the constitutionality of the regulations. Rather, this approach accommodates the possibility that the balance between voters' rights and the State's interest may shift over time, and empowers courts -- where an appropriate challenge is brought -- to take action when the burdens on those rights are no longer adequately justified. Cf. Anderson, 460 U.S. at 796-797 (early filing deadline that may have once been justified by State's interest in voter education no longer justified "[i]n the modern world," where information about candidates is instantly accessible).

Almost two centuries ago, in Capen, 12 Pick. at 489, this court noted the delicate balance that must be struck between the constitutional right to vote and the State's authority to regulate elections. "[W]here the [C]onstitution has conferred a political right," we wrote, "it is clearly within the just and constitutional limits of the legislative power" to regulate "the time and mode of exercising that right," so as "to secure and facilitate [its] exercise." Id. We cautioned, however, that this authority to regulate "afford[s] no warrant for such an exercise of legislative power" that would "subvert or injuriously restrain the right itself." Id. Then and now, the difficult task of balancing these two interests falls on the courts. By adhering to the traditional two-track analysis, the court today only makes this task more difficult. In this and future cases, I would review laws regulating the right to vote under a more flexible, more fine-tuned standard.