Justice TIMMER

Joseph A. Kanefield (argued), Brunn W. Roysden III, Ballard Spahr LLP, Phoenix, for Arizona Citizens Clean Elections Commission. Mary R. O'Grady, Timothy J. Eckstein, Christina C. Rubalcava, Osborn Maledon, Phoenix; and Timothy M. Hogan, Arizona Center for Law in the Public Interest, Phoenix, for Louis J. Hoffman, Victoria Steele, and Arizona Advocacy Network.
Richard Rice, Acting Attorney General, David Weinzweig, Senior Litigation Counsel, Daniel P. Schaack (argued), Assistant Attorney General, Phoenix, for Ken Bennett.

Michael T. Liburdi (argued), Kelly A. Kszywienski, Snell & Wilmer LLP, Phoenix; Gregrey G. Jernigan, Office of the President, Arizona State Senate, Phoenix; Peter A. Gentala, Pele Peacock, Office of the Speaker, Arizona House of Representatives, Phoenix, for Andy Biggs and Andrew M. Tobin.

Andrew S. Gordon, Roopali H. Desai, Melissa A. Soliz, Coppersmith Brockelman PLC, Phoenix, for Amici Curiae Arizona Chamber of Commerce and Industry, et al.

James E. Barton II, Torres Law Group, PLLC, Tempe, for Amici Curiae League of Women Voters, et al.

Paul V. Avelar, Timothy D. Keller, Institute for Justice, Tempe, for Amicus Curiae Institute for Justice.

, opinion of the Court.

¶ 1 In 1998, Arizona voters enacted the Citizens Clean Elections Act to establish public funding for political candidates in statewide and state legislative elections. The Act prohibits a candidate who opts not to receive public funding from accepting contributions greater than eighty percent of the campaign contribution limits specified in A.R.S. § 16–905. The issue here is whether the Act fixes campaign contribution limits at eighty percent of the amounts that existed in 1998 or instead provides a formula for calculating limits. We hold that the Act provides a formula for calculating contribution limits.

I. BACKGROUND

¶ 2 Both Arizona voters and the legislature have taken an active role in developing campaign financing laws. In 1986, voters enacted by initiative A.R.S. § 16–905, which established campaign contribution limits for state, county, and local elected officials. The legislature amended § 16–905 in 1993, 1994, 1997, and 2007 to increase those limits. 1993 Ariz. Sess. Laws, ch. 226, § 4 (1st Reg.Sess.); 1994 Ariz. Sess. Laws, ch. 379, § 2 (2d Reg.Sess.); 1997 Ariz. Sess. Laws, ch. 201, § 6 (1st Reg.Sess.); 2007 Ariz. Sess. Laws, ch. 277, § 1 (1st Reg.Sess.).

¶ 3 In 1998, voters passed an initiative to create the Citizens Clean Elections Act, A.R.S. §§ 16–940 to –961 (“CCEA” or “Act”), which established an alternative campaign financing system for primary and general elections and created the Citizens Clean Elections Commission to administer it. Under this system, candidates for statewide and state legislative offices who agree to limit fundraising and campaign spending (“participating candidates”) receive public campaign financing. Eligible candidates who choose not to participate (“nonparticipating candidates”) can accept private campaign contributions up to eighty percent of the limits established by A.R.S. § 16–905(A)–(E), as adjusted periodically for inflation. A.R.S. § 16–941(B). In an apparent effort to “level the playing field,” the Act also originally provided that once expenditures by or on behalf of a nonparticipating candidate exceeded a publicly funded opponent's initial funding allotment, that opponent would be given roughly one dollar for every additional dollar spent by or on behalf of the nonparticipating candidate, capped at three times the initial public funding allotment. A.R.S. § 16–952 (1998); see Bennett, 131 S.Ct. at 2813. Candidates for countywide and municipal offices are not eligible to participate in the Clean Elections system.

¶ 4 In 1998, the voters also passed another initiative, unrelated to the CCEA, which adopted the Voter Protection Act (“VPA”). Ariz. Const. art. 4, pt. 1, § 1(6). The VPA limits the legislature's authority to modify laws enacted by voters at or after the November 1998 general election. Cave Creek Unified Sch. Dist. v. Ducey, 233 Ariz. 1, 4 ¶ 9, 308 P.3d 1152, 1155 (2013); Ariz. Const. art. 4, pt. 1, § 1, Historical Notes (West 2014).

¶ 5 In April 2013, the legislature passed and the Governor signed House Bill (“H.B.”) 2593, which amended § 16–905 by increasing campaign contribution limits for statewide, countywide, and local offices, eliminating restrictions on the aggregate amount of money candidates can receive from political committees, and eliminating restrictions on the amount of money individuals can contribute to political committees that give money to candidates. 2013 Ariz. Sess. Laws, ch. 98, § 2 (1st Reg.Sess.). The effective date for H.B. 2593 was September 13, 2013.

¶ 6 In July 2013, the Citizens Clean Elections Commission and others (collectively, the “Commission”) sued Arizona's Secretary of State, asking the superior court to declare H.B. 2593 unconstitutional, as applicable to nonparticipating candidates, and to enjoin the Secretary from implementing it. The Commission alleged that the CCEA fixed campaign contribution limits as they existed in 1998 for nonparticipating candidates, and that the legislature could not alter those limits by amending § 16–905 without complying with the VPA. The court permitted the President of the Senate, Andy Biggs, and the Speaker of the House, Andrew M. Tobin, (collectively, “Intervenors”) to intervene and join in defending H.B. 2593.

¶ 7 The superior court denied the Commission's motion to preliminarily enjoin implementation of H.B. 2593, finding that the Commission did not have a strong likelihood of success on the merits. See Shoen v. Shoen, 167 Ariz. 58, 63, 804 P.2d 787, 792 (App.1991) (setting forth factors to consider in ruling on a motion for preliminary injunction). The court reasoned that § 16–905 is probably not subject to the VPA because (1) the CCEA referred to § 16–905 only as part of a formula for computing contribution limits, (2) subjecting cross-referenced statutes to the VPA could “create havoc,” and (3) it was uncertain whether a majority of voters in 1998 intended that the VPA apply to the CCEA. The court also found that, “in light of ... First Amendment issues presented” by Intervenors, it could not conclude that irreparable harm would occur or that the balance of hardships or public interest favored a preliminary injunction.

¶ 8 In an ensuing special action, the court of appeals accepted jurisdiction and granted relief to the Commission. Ariz. Citizens Clean Elections Comm'n v. Brain, 233 Ariz. 280, 282 ¶ 1, 311 P.3d 1093, 1095 (App.2013). The court did not address the parties' VPA-related or First Amendment arguments. Instead, it held that, “as a matter of statutory construction, when the voters enacted the [CCEA] in 1998, they fixed campaign contribution limits as they existed in 1998 ... [and] did not adopt a mere formula that would allow the Legislature to easily amend the § 941 limits.” Id. at 288 ¶ 31, 311 P.3d at 1101. Based on that determination, the court of appeals further concluded that because § 16–941(B) applies “[n]otwithstanding any law to the contrary,” it “preempts” and renders “ineffective” those provisions of H.B. 2593 that altered campaign contribution limits applicable to nonparticipating candidates. Id.

¶ 9 In light of its interpretation of § 16–941(B) as providing fixed limits rather than a formula, the court of appeals vacated the superior court's order denying the Commission's motion for a preliminary injunction. Id. at 290–91 ¶ 41, 311 P.3d at 1103–04. Because the superior court acknowledged but did not sufficiently address Intervenors' arguments that fixed limits violate the First Amendment, however, the court of appeals instructed the superior court to revisit that issue. Id. Pending that determination, the court of appeals maintained the preliminary injunction against H.B. 2593's implementation, as applicable to nonparticipating candidates. Id. at 292 ¶ 46, 311 P.3d at 1105.

¶ 10 Pursuant to our jurisdiction under Article 6, Section 5(3) of the Arizona Constitution, and as a matter of statewide importance, we granted review to decide this single statutory issue: whether A.R.S. § 16–941(B) provides a formula for calculating campaign contribution limits for nonparticipating candidates or instead fixes those limits. We previously vacated the court of appeals' opinion and lifted the preliminary injunction, indicating that this opinion would follow.

II. DISCUSSION

¶ 11 Our primary objective in interpreting a voter-enacted law is to effectuate the voters' intent. See Ariz. Early Childhood Dev. & Health Bd. v. Brewer, 221 Ariz. 467, 470 ¶ 10, 212 P.3d 805, 808 (2009). If the statute is subject to only one reasonable interpretation, we apply it as written without further analysis. Id. But if the statute is ambiguous, we consider secondary principles of statutory interpretation, such as “the context of the statute, the language used, the subject matter, its historical background, its effects and consequences, and its spirit and purpose.” Wyatt v. Wehmueller, 167 Ariz. 281, 284, 806 P.2d 870, 873 (1991) (citation omitted). We review the interpretation of statutes de novo as an issue of law. State v. Gutierrez, 229 Ariz. 573, 576 ¶ 19, 278 P.3d 1276, 1279 (2012).

A.

¶ 12 Before addressing the statutory issue before us, we make clear that two factors have no bearing on determining the voters' intent when they passed the CCEA in 1998: the VPA and testimony from plaintiff Louis J. Hoffman, who primarily drafted the CCEA and served as a member of the Citizens Clean Elections Commission. Although voters approved both the VPA and the CCEA in the 1998 general election, the acts' subjects are not linked, and nothing in the CCEA's publicity pamphlet or ballot language mentions the VPA. In short, those who voted to enact the CCEA might or might not have supported the VPA and could not have counted on its simultaneous enactment. And just as a legislator, lobbyist, or other interested party lacks competence to testify about legislative intent in passing a law, Golder v. Dep't of Revenue, 123 Ariz. 260, 265, 599 P.2d 216, 221 (1979), the drafter of a voter initiative is not competent to testify about the voters' intent in passing that initiative.

B.

1.

¶ 13 We begin our statutory analysis with the language of A.R.S. § 16–941(B):

Notwithstanding any law to the contrary, a nonparticipating candidate shall not accept contributions in excess of an amount that is twenty per cent less than the limits specified in § 16–905, subsections A through E, as adjusted by the secretary of state pursuant to § 16–905, subsection H. Any violation of this subsection shall be subject to the civil penalties and procedures set forth in § 16–905, subsections J through M and § 16–924.
That subsection can be reasonably read as either providing a formula for calculating campaign contribution limits for nonparticipating candidates, applicable as the amounts prescribed in § 16–905 change (as Intervenors argue and the superior court determined) or fixing those limits at eighty percent of the amounts listed in § 16–905 at the time of the 1998 election (as the Commission argues and the court of appeals held). Therefore, the statute is ambiguous.

2.

¶ 14 Application of secondary principles of statutory construction reveals support for each competing interpretation of § 16–941(B). For the reasons stated below, however, we are convinced that the statute is most reasonably interpreted as establishing a formula.

¶ 15 First and foremost, the voters used a percentage for calculating contribution limits for nonparticipating candidates. Application of a percentage to a given amount is characteristic of a formula. See Random House Webster's Unabridged Dictionary 753 (2d ed.2001) (defining “formula” in part as “a set form of words ... for indicating procedure to be followed”). Had voters intended to fix static contribution limits, they could have easily and clearly done so by specifying dollar amounts. Cf. McElhaney Cattle Co. v. Smith, 132 Ariz. 286, 290–91, 645 P.2d 801, 805–06 (1982) (reasoning that if the electorate had intended to include cattle raisers or farmers within the meaning of “wholesaler” in tax exemption, “it is difficult to believe that (they) would have attempted to carry it into effect in such an uncertain and doubtful manner, when (they) could have done so easily and naturally”) (citation and internal quotation marks omitted).

¶ 16 Indeed, the voters fixed monetary amounts in other parts of § 16–941 and elsewhere in the CCEA. SeeA.R.S. §§ 16–941(A)(2) (personal expenditure limits); –941(D) (independent expenditure reporting); –942(B) (civil penalties); –944 (lobbyist fees); –945(A) (early contributions limits); –948(C) (petty cash restrictions); –954(B) (tax reduction); –955(G) (commissioner compensation); and –961(G) (primary election spending limits). Even assuming, as the dissent suggests, that the Act's drafters might have chosen to refer to § 16–905 for ease of reference rather than listing contribution limits in § 16–941(B), Dissent ¶ 41, it defies common sense to conclude that voters, who had no role in the drafting process, were similarly motivated. In short, no sound reason exists to conclude that voters intended to establish fixed contribution limits in § 16–941(B) by using a percentage formula that expressly incorporates another, existing statute, § 16–905.

¶ 17 Second, voters treated the § 16–941(B) limits differently from fixed amount limits specified elsewhere in the Act. Specifically, voters included an inflation-adjustment mechanism for monetary amounts fixed in other provisions of the Act, including § 16–941(A)(2) and (D), but they did not do so for § 16–941(B). SeeA.R.S. § 16–959. Instead, § 16–941(B) provides that the inflation-adjustment mechanism in § 16–905 applies. The fact that voters treated the § 16–941(B) limits differently from fixed amount limits further suggests that the voters did not intend to fix monetary limits in § 16–941(B). Cf. Fidelity Nat. Fin. Inc. v. Friedman, 225 Ariz. 307, 310 ¶ 12, 238 P.3d 118, 121 (2010) (acknowledging that statutes relating to the same subject should be construed together); Farmers Co-op. Co. v. DeCoster, 528 N.W.2d 536, 539 (Iowa 1995) (“[W]here a statute with respect to one subject contains a given provision, the omission of such provision from a similar statute is significant to show a different intention existed.”) (citation omitted).

¶ 18 Third, interpreting § 16–941(B) as fixing contribution limits for nonparticipating candidates at eighty percent of 1998 levels would widen the voter-approved gap between these limits and those for candidates not subject to the Act. Voters in 1998 constructively knew that the legislature would at some point likely amend § 16–905 to increase campaign contribution limits, as it had done three times in the preceding twelve years, including the year before the election. Because § 16–941(B) applies “notwithstanding any law to the contrary,” fixed limits for nonparticipating candidates would remain in place, and increased limits would apply only to candidates not subject to the Act. Although the voters intended a twenty-percent gap between limits for nonparticipating candidates and candidates not subject to the CCEA, nothing indicates that voters wanted to widen this gap.

¶ 19 To illustrate, under § 16–905's current limits, an individual could contribute $5,000 to a city council candidate for primary and general elections. But if § 16–941(B) fixes limits at eighty percent of 1998 levels for nonparticipating candidates, that individual could not contribute more than approximately $1,850 to a nonparticipating gubernatorial candidate who must run a more expensive statewide campaign. CompareA.R.S. § 16–905(A)(2), with id. §§ 16–905(B)(1), –941(B) (1998). Under this scenario, a sixty-seven percent gap is created between contribution limits for nonparticipating candidates and those for candidates not subject to the CCEA. Interpreting § 16–941(B) as providing a formula that increases contribution limits for nonparticipating candidates as § 16–905 is amended prevents such anomalies and maintains the voter-approved twenty-percent gap. Cf. Gutierrez v. Indus. Comm'n of Ariz., 226 Ariz. 395, 396–97 ¶ 6, 249 P.3d 1095, 1096–97 (2011) (acknowledging preference to interpret a statute to give it “a fair and sensible meaning”) (citation and internal quotation marks omitted).

¶ 20 Fourth, interpreting § 16–941(B) as providing fixed limits creates a needlessly confusing system, which the voters also likely did not intend. Section 16–941(B) provides that the eighty percent contribution limits are calculated from the amounts specified in § 16–905 “as adjusted by the secretary of state pursuant to § 16–905, subsection H.” Section 16–905(H) requires the Secretary of State to adjust for inflation the amounts in § 16–905 biennially. Because nothing requires the Secretary to adjust 1998–established limits for nonparticipating candidates following the legislature's changes to the § 16–905 limits, it is uncertain whether and how the § 16–941(B) limits would be adjusted for inflation. Interpreting § 16–941(B) as prescribing a formula eliminates that uncertainty.

¶ 21 Fifth, and finally, nothing in the ballot or attendant publicity pamphlet for the 1998 election informed voters that § 16–941(B) permanently fixed contribution limits at eighty percent of 1998 levels. See Ruiz v. Hull, 191 Ariz. 441, 450 ¶ 36, 957 P.2d 984, 993 (1998) (“In construing an initiative, we may consider ballot materials and publicity pamphlets circulated in support of the initiative.”) (citation omitted). The ballot stated that one effect of the CCEA would be to “reduc[e] the current contribution limits by 20% for non-participating candidates.” Similarly, the pamphlet advised that the Act would “reduce by twenty percent the amount per individual that can currently be contributed to a candidate.” These statements accurately described the immediate impact of the CCEA, no matter whether § 16–941(B) prescribed a formula or a fixed amount. Significantly, nothing alerted voters that the percentage reduction in § 16–941(B) would apply only to then-existing limits in § 16–905, or that any future increases in contribution limits under § 16–905 would not apply to nonparticipating candidates. Were that intended, we would expect a clear explanation of such a feature in light of its significant impact on campaign financing.

3.

¶ 22 We do not address every argument supporting a contrary interpretation of § 16–941(B). Even if some of those arguments are persuasive, the most reasonable interpretation remains that § 16–941(B) prescribes a formula. Nevertheless, we address the main points advanced by the Commission and adopted by the court of appeals and, to some extent, our dissenting colleagues.

¶ 23 The directive that § 16–941(B) apply “[n]otwithstanding any law to the contrary” does not evidence an intent to fix contribution amounts for nonparticipating candidates at 1998 levels regardless of any future alterations to § 16–905's limits. See Brain, 233 Ariz. at 286 ¶ 19, 311 P.3d at 1099 (reasoning that inclusion of this phrase reflects the voters' intent that § 16–941(B) “stand on its own” and “have significance separate and apart from ... § 16–905”). Rather, by providing that § 16–941(B) applies “notwithstanding” other laws, voters ensured that the twenty-percent reduction to § 16–905's limits applies regardless of any future adjustments to those limits and regardless of any contrary laws.

¶ 24 Inclusion of the inflationary-adjustment language in § 16–941(B) does not evidence the voters' intent to permit only the Secretary of State to adjust limits for nonparticipating candidates. Nor is that language superfluous if, as we conclude, § 16–941(B) prescribes a formula. See Brain, 233 Ariz. at 286 ¶ 18, 311 P.3d at 1099. Absent that language, the twenty-percent reduction would apply to the limits established by § 16–905(A)–(E) without periodic inflationary adjustments. Inclusion of the language clarifies that the adjustments required by § 16–905(H) apply. Significantly, § 16–941(B) neither states that inflationary adjustments are the only changes that may be made to contribution limits nor forecloses calculations of new limits if the legislature amends § 16–905.

¶ 25 Section 16–941(B) is not rendered “illusory” or “virtually meaningless” if the voters intended to prescribe a formula any more than if they intended to fix limits. Dissent ¶ 39; Brain, 233 Ariz. at 287 ¶ 23, 311 P.3d at 1100. When voters enacted the CCEA, pre-VPA law in effect at that time prohibited the legislature from repealing or modifying a voter-approved law only if it was enacted by a majority of all registered voters, not merely a majority of those who voted on the measure—an unlikely scenario. See Cave Creek Unified Sch. Dist., 233 Ariz. at 4 ¶ 9, 308 P.3d at 1155. Thus, as far as 1998 voters knew, just as the legislature could have responded to the twenty-percent reduction in § 16–941(B) by increasing the limits in § 16–905 by twenty percent, under then-current law it could have amended § 16–941(B) to increase any fixed amount or to completely remove any limits. By providing a twenty-percent reduction formula rather than a fixed amount, voters increased the chances that § 16–941(B) would remain viable after any increases to § 16–905's limits and consequently would continue to serve as an incentive for candidates to participate in the CCEA.

¶ 26 We reject the dissent's assertion that voters likely intended to fix contribution limits in § 16–941(B) because “[p]ublic funding allotments under the CCEA are not tied to the contribution limits for non-participating candidates,” and so a formula would make public funding “less appealing as it becomes easier for candidates to receive large private contributions.” Dissent ¶ 45. At the time voters passed the CCEA initiative, the now-defunct matching-funds provision existed to ensure that public funding kept pace with private contributions up to “three times the original spending limit” for participating candidates in both the primary and general elections. A.R.S. § 16–952(E) (1998). In light of this generous matching provision, the voters did not have to fix contribution limits to entice candidates to opt into the Act and accept public funding.

¶ 27 Nelson Machinery Co. v. Yavapai County, 108 Ariz. 8, 491 P.2d 1132 (1971), on which the court of appeals relied, does not persuade us to interpret § 16–941(B) differently. See Brain, 233 Ariz. at 287 ¶ 24, 311 P.3d at 1100. In Nelson Machinery, this Court cited a statutory construction canon providing that when a statute adopts another statute by specific reference, the adopted statute is taken as it then exists and does not include subsequent amendments, unless the enactors of the adopting statute expressly intended otherwise. Nelson Mach., 108 Ariz. at 9, 491 P.2d at 1133. Apparently recognizing the canon's limitations, the Nelson Machinery Court neither embraced nor applied it, turning instead to other factors to determine a contrary legislative intent. Id. at 9–10, 491 P.2d at 1133–34. Similarly, because the canon does not help ascertain the voters' intent, and in light of the evidence indicating that the voters intended § 16–941(B) to provide a formula, we likewise decline to apply the canon here.

III.

¶ 28 For the foregoing reasons, the most reasonable interpretation of A.R.S. § 16–941(B) is that voters intended to prescribe a formula to calculate campaign contribution limits for nonparticipating candidates for statewide and state legislative offices. Because the court of appeals reached a contrary conclusion, we vacate its opinion and affirm the superior court's order denying the Commission's motion for a preliminary injunction. Vice Chief Justice BALES, with whom Chief Justice BERCH joins, dissenting.

¶ 29 Seeking to reform the financing of campaigns for the legislature and certain state offices, Arizona's voters approved the CCEA in 1998 to provide public funding for participating candidates and to reduce the contribution limits for non-participating candidates to “twenty per cent less than the limits specified in § 16–905.” A.R.S. § 16–941(B). Since the CCEA was adopted, many have questioned its premises and effectiveness, and some of its provisions have been challenged as contrary to the First Amendment. Whether the CCEA is desirable as a policy matter or vulnerable to further constitutional challenges are not issues before this Court, and I express no view regarding them.

¶ 30 The issue we must decide is whether the voters in 1998 intended to reduce the contribution limits to eighty percent of the limits then specified in § 16–905 (i.e., the limits reduced by twenty percent) or instead to adopt a formula that would automatically ratchet up the limits to eighty percent of whatever amounts the legislature might set. The majority adopts the latter interpretation, thereby approving the legislature's recent increases to the individual contribution limits (among others) by more than 400 percent for statewide candidates (from $912 to $4000 for each election cycle) and more than 900 percent for legislative candidates (from $440 to $4000). I respectfully dissent because construing A.R.S. § 16–941(B) to refer to the “limits specified in § 16–905” when the CCEA was adopted better effects the intent of the voters, who plainly sought to reduce the impact of large campaign contributions in Arizona elections.

¶ 31 In analyzing the issue presented, I agree with many of the conclusions reached by the majority, although sometimes for different reasons. I agree that the voters' approval of the VPA in the 1998 election is not relevant to the issue presented. Op. ¶ 12. But I disagree with the majority's observations that the VPA is “not linked” to the CCEA because the latter's supporters “might or might not have supported the VPA.” Id.

¶ 32 The two Acts are indeed “linked.” The VPA applies to measures “decided by the voters at and after the November 1998 general election.” Proposition 105, § 2, 1999 Ariz. Sess. Laws 1937, 1941. Thus, by approving the VPA, a majority of the voters contemplated that it would apply to other measures approved at the 1998 election, even though they may not have then known that particular measures such as the CCEA would be approved.

¶ 33 The VPA is irrelevant here for a different reason. If the majority is correct that the voters intended a formula, then the legislature's increasing the limits by amending § 16–905 would comport with the CCEA, and the VPA would not be implicated. Conversely, if the voters intended to adopt limits based on the then-existing amounts set in § 16–905, any change in the contribution limits for non-participating candidates would have to comply with the VPA's requirements, including approval by at least three-fourths of each house of the legislature. (Indeed, the legislature arguably complied with these requirements when it increased the contribution limits previously in 2007.) But the existence of the VPA does not help in resolving whether the voters contemplated that the CCEA, by referencing the § 16–905 limits, would adopt fixed limits or instead a formula.

¶ 34 I also agree with the majority that § 16–941(B) is ambiguous, Op. ¶ 13, contrary to the parties' conflicting contentions that the “plain meaning” or “explicit” language of § 16–941(B) supports their respective positions. The “plain meaning” rule guides our analysis when statutory language is “subject to only one reasonable meaning.” See Arizona Early Childhood Dev. & Health Bd. v. Brewer, 221 Ariz. 467, 470 ¶ 10, 212 P.3d 805, 808 (2009). When, as is true here, a statute may reasonably be interpreted more than one way, determining its meaning is not advanced by assertions that one plausible interpretation must be right because it reflects the “plain meaning.”

¶ 35 The majority also correctly concludes that our analysis is not aided by the “specific reference” canon discussed in Nelson Machinery Co. v. Yavapai County, 108 Ariz. 8, 9, 491 P.2d 1132, 1133 (1971). Although Nelson Machinery noted that statutes that specifically reference another are presumed to refer to the other statute in its then-existing version, the Court did not apply the canon in that case, id., and it is unclear whether we have otherwise endorsed it. Rather than distinguish Nelson Machinery, Op. ¶ 27, I would disclaim the specific reference canon entirely.

¶ 36 “Our primary objective in construing statutes adopted by initiative is to give effect to the intent of the electorate.” See Brewer, 221 Ariz. at 470 ¶ 10, 212 P.3d at 808 (quoting State v. Gomez, 212 Ariz. 55, 57 ¶ 11, 127 P.3d 873, 875 (2006)); Op. ¶ 11. Accordingly, rather than relying on any canon, we should look to the intent of the voters to determine if § 16–941(B) refers to § 16–905 in its 1998 version or instead as it may later be amended. Cf. United States v. United Mine Workers of Am., 330 U.S. 258, 314, 67 S.Ct. 677, 91 L.Ed. 884 (1947) (“[A] canon, like other generalities about statutory construction, is not a rule of law. Whatever persuasiveness it may have in construing a particular statute derives from the subject matter and the terms of the enactment in its total environment.”) (Frankfurter, J., concurring).

¶ 37 The voters clearly stated their intent in enacting the CCEA: to reduce the influence of large campaign contributions. The CCEA noted that the then-existing system of private financing “[a]llow[ed] Arizona elected officials to accept large campaign contributions from private interests over which they have governmental jurisdiction,” permitted disproportionate influence by “a small number of wealthy special interests,” and “[drove] up the cost of running for state office.” A.R.S. § 16–940(B).

¶ 38 Also relevant is the context in which the voters approved the CCEA. After the voters had first approved contribution limits and other campaign finance regulations in 1986, and the so-called “AzScam” scandal in 1991 had revealed troubling instances of misconduct and outright corruption in connection with campaign contributions, the Legislature increased the contribution limits three times between 1993 and 1997. 1993 Ariz. Sess. Laws, ch. 226, § 4 (1st Reg.Sess.); 1994 Ariz. Sess. Laws, ch. 379, § 2 (2d Reg.Sess.); 1997 Ariz. Sess. Laws, ch. 201, § 6 (1st Reg.Sess.).

¶ 39 Against this backdrop, the most plausible conclusion is that the voters in 1998 intended to reduce the limits to eighty percent of the amounts the Legislature had just set the previous year. It strains belief, and contradicts the voters' stated purpose, to instead conclude that the voters intended to tie the contribution limits prospectively to eighty percent of whatever amounts the legislature might choose to set in § 16–905. Such a formula would make the reduction in § 16–941(B) illusory, since the legislature could raise the limits to any desired amount merely by increasing the limits in § 16–905 to 125 percent of that amount.

¶ 40 The majority points out that neither the 1998 ballot nor the publicity pamphlet explicitly informed voters that the CCEA would set fixed limits. Op. ¶ 21. But neither did these materials tell voters that the CCEA would create merely a formula tied to future increases. Given that the CCEA is ambiguous and the ballot measure materials are inconclusive on the specific issue, we should be guided by the voters' more general purposes in enacting the CCEA. See Hayes v. Cont'l Ins. Co., 178 Ariz. 264, 270, 872 P.2d 668, 674 (1994).

¶ 41 The majority justifies its conclusion by squinting to find clues of the voters' intent from textual arguments that are at best inconclusive. For example, the majority states that if the voters had wanted to set specific limits, rather than adopt a formula, they could have specified the limits in § 16–941(B) rather than referencing § 16–905. Op. ¶ 15. The majority notes that the CCEA provides specific amounts in other sections. Op. ¶ 16. This argument, however, ignores the fact that the amounts stated in the other sections did not have counterparts in pre-existing law, and thus there was no reasonable possibility of incorporation by reference. Section 16–941(B), in contrast, could refer to a twenty percent reduction in the limits already specified in § 16–905. This made clear that the CCEA was reducing pre-existing limits and avoided the need to specify dollar amounts for each of the ten limits taken from § 16–905. SeeA.R.S. § 16–905 (1997), amended by 2007 Ariz. Sess. Laws, ch. 277 (1st Reg.Sess.) (listing ten limits with specific dollar amounts). Similarly, there is no inconsistency between the voters referencing the inflation-adjustment provision in § 16–905 while separately providing an inflation-adjustment provision for other amounts fixed by the CCEA. Because § 16–905 had a pre-existing adjustment mechanism, it could be referenced in the CCEA and there was no need to create a new one.

¶ 42 In response, the majority states that even if the drafters of the CCEA might have chosen to refer to § 16–905 for ease of reference, “it defies common sense to conclude that voters, who had no role in the drafting process, were similarly motivated.” Op. ¶ 16. But the voters never collectively “draft” ballot measures. Common sense supports construing a measure's ambiguous language to further the general purposes the voters expressly approved. It is unconvincing to say the voters must have been “motivated” to adopt a formula because they approved language in § 16–941(B) referencing § 16–905. This begs the question whether the voters intended the reference to adopt the existing limits or instead a formula.

¶ 43 The majority also posits two purposes the voters might have intended a formula to serve, but neither withstands scrutiny. First, the majority observes that if § 16–941(B) adopts a formula, the CCEA would allow the legislature to ensure the “gap” between the CCEA-adjusted limits for state candidates and the limits for local candidates does not exceed twenty percent. Op. ¶¶ 18–19. The CCEA did result in lower contribution limits for legislative candidates than for local candidates. (For example, as a result of the CCEA, the individual contribution limits were $256, while the limit for local candidates was $320.)

¶ 44 The CCEA did not affect the existing contribution limits for local candidates, and nothing in the CCEA or its history suggests that the voters intended to preserve a “twenty-percent gap” between state and local limits. Since the CCEA did not affect local limits at all, it is unclear why the voters would have had any expectations regarding them. By attributing to the voters a purpose nowhere stated, the majority interprets the CCEA to achieve a result (increasing the role of large campaign contributions) that the voters expressly sought to avoid.

¶ 45 The majority also contends that the voters chose a formula in order to “increas[e] the chances that § 16–941(B) would remain viable” as an incentive for candidates to participate in the public funding system, even in the face of increases to the § 16–905 limits. Op. ¶ 25. Insofar as the voters sought to provide incentives to participate in public funding, however, this purpose is better advanced by construing § 16–941(B) as adopting fixed limits rather than a formula. Public funding allotments under the CCEA are not tied to the contribution limits for non-participating candidates, so public funding would generally become less appealing as it becomes easier for candidates to receive large private contributions.

¶ 46 The majority incorrectly states that “[a]t the time voters passed the CCEA initiative, the now-defunct matching-funds provision existed to ensure that public funding kept pace with private contributions.” Op. ¶ 26. Matching funds could not exceed twice the amount of the initial public funding, A.R.S. § 16–952(E) (1998), so a publicly funded candidate could not “keep pace” with a privately funded candidate who spent more. A formula allowing dramatic increases in contribution limits like those approved by the majority would make it easier for privately funded candidates to outpace their publicly funded opponents, even when matching funds existed.

¶ 47 Because the voters in 1998 sought to reduce the role and influence of private contributions in political campaigns, the CCEA is more plausibly construed as lowering the limits as they then existed under § 16–905 rather than tying them to increases the legislature might later enact. Thus, I respectfully dissent.