While the members of the Tucson City Council play fast and loose with the residency requirements for candidates and councilmembers, on Friday the Ninth Circuit Court of Appeals found that the City’s hybrid election system is constitutional.
“We hold that Tucson’s system does not violate the Equal Protection Clause of the Fourteenth Amendment to the U.S. Constitution and so affirm the district court’s order awarding judgment in favor of the City and its co-defendants,” concluded Judge Marsha S. Berzon, who wrote the opinion for the Court.
Berzon wrote, “A divided three-judge panel of this court reversed, holding that by denying out-of-ward voters the ability to vote in the primary elections of other wards, the hybrid system violates the one person, one vote guarantee embedded in the Equal Protection Clause. Pub. Integrity All., Inc. v. City of Tucson, 805 F.3d 876, 883 (9th Cir. 2015). We took the case en banc and now affirm the district court. Tucson’s hybrid voting system for its city council elections does not violate the Equal Protection Clause.”
The ruling reads:
The structure of municipal governments and methods of selecting municipal officials vary greatly across the country. Such diversity is a manifestation of our federal structure, which ideally, though not always, “allows local policies ‘more sensitive to the diverse needs of a heterogeneous society,’ permits ‘innovation and experimentation,’ enables greater citizen ‘involvement in democratic processes,’ and makes government ‘more responsive by putting the States in competition for a mobile citizenry.’” Bond v. United States, 564 U.S. 211, 221 (2011) (quoting Gregory v. Ashcroft, 501 U.S. 452, 458 (1991)). This case requires us to consider the constitutional validity of one municipality’s chosen election system.
Public Integrity Alliance, a nonprofit corporation, and four Tucson voters (collectively referred to as “Public Integrity Alliance”) challenge as unconstitutional the City of Tucson’s system for electing members of its city council. We hold that Tucson’s system does not violate the Equal Protection Clause of the Fourteenth Amendment to the U.S. Constitution and so affirm the district court’s order awarding judgment in favor of the City and its co-defendants.
BACKGROUND
I.
Tucson is one of nineteen charter cities in Arizona. City of Tucson v. State, 229 Ariz. 172, 174 (2012) (en banc). Under Arizona’s constitution, charter cities are municipalities of more than 3,500 people that have elected to “adopt a charter—effectively, a local constitution — for their own government without action by the state legislature.” Id. Charter cities enjoy enhanced autonomy with regard to government structure and the selection of their city officials. See id.; Strode v. Sullivan, 72 Ariz. 360, 368 (1951).
Since adopting its current city charter in 1929, Tucson has used a “hybrid election system” for electing members to its city council. City of Tucson, 229 Ariz. at 175; Tucson City Charter, ch. XVI, § 9. Tucson’s city council election system operates as follows: Tucson is divided into six wards of approximately equal populations. Id. ch. XVI, § 8. Each ward is allotted one seat on the six-member city council. Id. ch. III, § 1. Council members serve four-year terms and are elected on a staggered basis, with three council members elected every odd-numbered year. Id. ch. XVI, §§ 3, 4. For example, elections for the seats allotted to Wards 1, 2, and 4 were held in 2015, and elections for the seats allotted to Wards 3, 5, and 6 will be held in 2017. A candidate for city council must reside in the ward from which she seeks to be nominated. Id. ch. XVI, § 5.
Council members are elected through a hybrid system involving a ward-level partisan primary election an an at-large partisan general election. First, each ward with a city council seat up for election conducts a partisan primary to select one nominee from each recognized political party. Persons who reside within that ward and are registered with a political party qualified for representation on the ballot may vote in their party’s ward-level primary. Ariz. Rev. Stat. § 16-467(B); Tucson City Charter, ch. XVI, § 9. A person registered as an independent, as having no party preference, or as a member of a party not entitled to representation on the Rev. Stat. § 16-467(B).
The top vote-getter from each party eligible for inclusion on the ward-level primary ballot then advances to an at-large general election, where she competes against the other candidates nominated from the same ward. Every Tucson voter may vote for one candidate from each ward that held a primary—that is, all voters may vote for one candidate for each of the three council member seats appearing on the general election ballot. Tucson City Charter, ch. XVI, § 9. Thus, when city council seats for Wards 1, 2, and 4 were up for election in 2015, residents of Ward 1 were permitted to vote in the primary only for a candidate from Ward 1, but then were permitted to vote for candidates from Wards 1, 2, and 4 in the general election. Once elected, council members represent the entire city. See City of Tucson, 229 Ariz. at 179.
Tucson’s voters twice have affirmed their commitment to the system. They rejected a proposal to change from at-large to ward-based general elections in 1991 and disapproved a proposal to change from partisan to non-partisan elections in 1993. Id. at 175.
Analogous election systems can be found in at least two other states in our circuit. Washington employs a similar system to elect county commissioners in 32 of its 39 counties and has done so for nearly a century. See State v. Bd. of Comm’rs of King Cty., 146 Wash. 449, 463 (1928), overruled on other grounds by Lopp v. Peninsula Sch. Dist. No. 401, 90 Wash. 2d 754 (1978) (en banc); Wash. Rev. Code §§ 36.32.040, 36.32.050, 36.32.0556. Several Washington cities, school districts, and special purpose districts also use similar hybrid election systems. See Wash. Rev. Code § 35.18.020 (cities); § 28A.343.660 (school districts); § 53.12.010 (port districts); § 54.12.010 (public utility districts); 52.14.013 (fire protection districts); § 57.12.039 (water-sewer districts). In Nevada, at least two cities, Sparks and Reno, conduct “hybrid,” albeit nonpartisan, city council elections, with the primary election by ward and the general election city-wide. See Reno City Charter, Art. V, §§ 5.010, 5.020; Sparks City Charter, Art. V, §§ 5.010, 5.020.
Public Integrity Alliance alleges that Tucson’s hybrid system runs afoul of the Equal Protection Clause of the Fourteenth Amendment1 because it violates the “one person, one vote” principle, relying mainly for their analysis on Gray v. Sanders, 372 U.S. 368, 380–81 (1963). The core of their argument is that Tucson voters currently are denied the right to participate in primary elections for all but one of their representatives on the city council. Because city council members represent Tucson as a whole, Public Integrity Alliance contends either (1) every Tucson voter must be permitted to vote in each ward’s primary, or (2) Tucson must switch to a purely ward-based system, in which voters for both the primary and general elections for a given council seat are limited to voting for the representative from their own ward and have no voice in selecting candidates from other wards. In other words, Public Integrity Alliance’s position is that an entirely ward-based or entirely at-large system of voting would be permissible, but the combination of the ward-based primary and at-large general is constitutionally fatal.
Public Integrity Alliance filed a complaint in federal district court seeking to enjoin the operation of Tucson’s hybrid system and secure a declaration that the scheme is unconstitutional. The district court held Tucson’s system constitutional and so denied Public Integrity Alliance’s request for relief.
A divided three-judge panel of this court reversed, holding that by denying out-of-ward voters the ability to vote in the primary elections of other wards, the hybrid system violates the one person, one vote guarantee embedded in the Equal Protection Clause. Pub. Integrity All., Inc. v. City of Tucson, 805 F.3d 876, 883 (9th Cir. 2015). We took the case en banc and now affirm the district court. Tucson’s hybrid voting system for its city council elections does not violate the Equal Protection Clause.
STANDARD OF REVIEW
“[T]he Constitution grants to the States a broad power to prescribe the ‘Times, Places and Manner of holding Elections for Senators and Representatives,’ Art. I, § 4, cl. 1, which power is matched by state control over the election process for state offices.” Tashjian v. Republican Party of Conn., 479 U.S. 208, 217 (1986). “This power is not absolute,” however. Wash. State Grange v. Wash. State Republican Party, 552 U.S. 442, 451 (2008). “[V]oting is of the most fundamental significance under our constitutional structure,” Ill. State Bd. of Elections v. Socialist Workers Party, 440 U.S. 173, 184 (1979), and state and local government election laws that violate the Constitution are impermissible. See Wash. State Grange, 552 U.S. at 451; Moore v. Ogilvie, 394 U.S. 814, 818 (1969).
The Supreme Court delineated the appropriate standard of review for laws regulating the right to vote in Burdick v. Takushi, 504 U.S. 428 (1992). Burdick recognized that governments necessarily “must play an active role in structuring elections,” and “[e]lection laws will invariably impose some burden upon individual voters.” Id. at 433. Consequently, not every voting regulation is subject to strict scrutiny. Id.
Instead, . . . a more flexible standard applies. 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 burden imposed by its rule,” taking into consideration “the extent to which those interests make it necessary to burden the plaintiff’s rights.”
Id. at 434 (quoting Anderson v. Celebrezze, 460 U.S. 780, 789 (1983)).
Under Burdick’s balancing and means-end fit framework, strict scrutiny is appropriate when First or Fourteenth Amendment rights “are subjected to ‘severe’ restrictions.” Id. (quoting Norman v. Reed, 502 U.S. 279, 289 (1992)). “But when a state election law provision imposes only ‘reasonable, nondiscriminatory restrictions’ upon the First and Fourteenth Amendment rights of voters, ‘the State’s important regulatory interests are generally sufficient to justify’ the restrictions.” Id. (quoting Anderson, 460 U.S. at 788).
Applying these precepts, “[w]e have repeatedly upheld as ‘not severe’ restrictions that are generally applicable, evenhanded, politically neutral, and protect the reliability and integrity of the election process.” Dudum v. Arntz, 640 F.3d 1098, 1106 (9th Cir. 2011) (citation and alterations omitted).2 Our “respect for governmental choices in running elections has particular force where, as here, the challenge is to an electoral system, as opposed to a discrete election rule.” Id. at 1114.
Despite Burdick, the City of Tucson asks that we apply traditional rational basis review, rather than a balancing and means-end fit analysis. Public Integrity Alliance agreed at oral argument that if we rejected its position that primary and general elections must involve identical electorates, traditional rational basis was the appropriate standard of review.
Our case law has not always accurately described the Burdick test. In Libertarian Party of Washington v. Munro, 31 F.3d 759 (9th Cir. 1994), we stated that where plaintiffs can demonstrate only a “slight” or “de minimis” impairment of their rights, they bear “the burden of demonstrating that the regulations they attack have no legitimate rational basis.” Id. at 763. But Burdick calls for neither rational basis review nor burden shifting. See Ariz. Libertarian Party v. Reagan, 798 F.3d 723, 732 n.12 (9th Cir. 2015) (noting the “tension” between Munro and Burdick); id. at 734–36 (McKeown, J., concurring) (same). To the extent Munro prescribed a different standard from the one articulated by the Supreme Court in Burdick, it is now overruled.
DISCUSSION
I.
Public Integrity Alliance argues that Tucson’s hybrid system severely burdens the Fourteenth Amendment by denying Tucson voters the right to vote in the primary elections for five out of six of their representatives on the city council. Central to Public Integrity Alliance’s articulation of the alleged burden is their interpretation of Gray v. Sanders.
According to Public Integrity Alliance, the case before us “is controlled by a single, simple maxim of equal protection” from Gray: “Once the geographical unit for which a representative is to be chosen is designated, all who participate in the election are to have an equal vote . . . wherever their home may be in that geographical unit.” Gray, 372 U.S. at 379. Public Integrity Alliance interprets this language as a requirement that primary and general elections use identical geographical units. Because members of the city council represent the entire city, Public Integrity Alliance reasons, the relevant “geographical unit” is the city as a whole. So, Public Integrity Alliance maintains, Tucson cannot constitutionally designate individual wards as the geographical units for the primary elections and limit participation in a given ward’s primary election to that ward’s residents, and then designate the whole city as the geographical unit for the general election.
Gray establishes no such principle. A vote dilution case, Gray involved a challenge to Georgia’s system of primary elections for statewide officers, a system wholly different from Tucson’s hybrid system of primary and general elections. Instead of counting individual votes, Georgia employed a “county unit system.” 372 U.S. at 370–71. Candidates who received the most votes in a county were considered to have won the county primary and, with respect to the statewide primary, were awarded “county units” in proportion to the number of representatives the county had in Georgia’s lower legislative body. Id. at 371. Georgia’s primary election system was thus similar to the electoral college used to elect our President, with counties’ representation substituted for the state representation in the electoral college. The county units were not proportionate to the county population, giving residents in one county dramatically more influence in the nomination of candidates than residents in another county. Id.
Gray held Georgia’s county unit system violative of the one person, one vote principle, because it diluted the voting power of certain voters based only on where they happened to live. Id. at 379–80.3 But Gray concerned only the primary election, not a comparison of the geographical units used in the primary and general elections. Gray therefore did not hold that the same geographical unit must apply to both primary and general elections; no issue regarding the relationship between the voting basis in the primary and in the general election was before the Court. And Gray has never been cited for the proposition Public Integrity Alliance puts forward. Instead, Gray has uniformly been construed as an unequal vote weighting case for a single election stage. See Williams v. Rhodes, 393 U.S. 23, 52 n.5 (1968) (Stewart J., dissenting) (Gray “sustained the right of a voter to cast a ballot whose numerical weight is the equal of that of any other vote cast within the jurisdiction in question.”); Fortson v. Morris, 385 U.S. 231, 235 (1966) (“The Gray case . . . did no more than to require the State to eliminate the county-unit machinery from its election system.”); Lucas v. Forty-Fourth Gen. Assembly, 377 U.S. 713, 744 (1964) (Stewart, J., dissenting) (noting that Gray was irrelevant to a case Gray held Georgia’s county unit system violative of the one person, one vote principle, because it diluted the voting power of certain voters based only on where they happened to live. Id. at 379–80.3 But Gray concerned only the primary election, not a comparison of the geographical units used in the primary and general elections. Gray therefore did not hold that the same geographical unit must apply to both primary and general elections; no issue regarding the relationship between the voting basis in the primary and in the general election was before the Court. And Gray has never been cited for the proposition Public Integrity Alliance puts forward.
Instead, Gray has uniformly been construed as an unequal vote weighting case for a single election stage. See Williams v. Rhodes, 393 U.S. 23, 52 n.5 (1968) (Stewart J., dissenting) (Gray “sustained the right of a voter to cast a ballot whose numerical weight is the equal of that of any other vote cast within the jurisdiction in question.”); Fortson v. Morris, 385 U.S. 231, 235 (1966) (“The Gray case . . . did no more than to require the State to eliminate the county-unit machinery from its election system.”); Lucas v. Forty-Fourth Gen. Assembly, 377 U.S. 713, 744 (1964) (Stewart, J., dissenting) (noting that Gray was irrelevant to a case “hav[ing] nothing to do with the ‘weighting’ or ‘diluting’ of votes cast within any electoral unit”). We decline to take a single sentence in a decades-old vote dilution case concerning a single stage of an election, read it without regard to the issue before the Court in that case, and transform it into a new voting rights principle requiring a two-stage election to cover the same geographical base at each stage.
Indisputably, primary elections are state action subject to the same constitutional constraints as general elections. See Smith v. Allwright, 321 U.S. 649, 661–62 (1944); United States v. Classic, 313 U.S. 299, 318–19 (1941). And primaries and general elections have an obvious and strong interconnection; that relationship is why the Supreme Court has described them as “a single instrumentality for choice of officers.” Allwright, 321 U.S. at 660. But the recognition that primaries are of great significance to the ultimate choice in a general election and thus directly implicate the right to vote does not mean that primaries and general elections must be identically structured and administered.
In fact, that contention is belied by decades of jurisprudence permitting voting restrictions in primary elections that would be unconstitutional in the general election. See, e.g., Clingman v. Beaver, 544 U.S. 581, 584 (2005) (permitting a semiclosed primary, in which only people who are registered as party members or independents may vote in a party’s primary); Am. Party of Tex. v. White, 415 U.S. 767, 786 (1974) (providing that states may establish waiting periods before voters may be permitted to change their registration and vote in another party’s primary); Ziskis v. Symington, 47 F.3d 1004, 1004–05 (9th Cir. 1995) (holding that a law requiring participants in primaries be registered with a political party did not violate the challenger’s Fourteenth Amendment right to vote). These voting restrictions are constitutionally permissible in primaries because primaries serve a different function than general elections: A primary determines which candidates will compete in the general election, a critical stage and one fully subject in its own right to constitutional scrutiny under Burdick, but a stage as to which the legitimate state interests are not identical with those pertinent to the general election, as the partisan primary cases illustrate.
II.
Having concluded that Gray does not require that the primary and general elections use identical geographical units, we now apply the Burdick balancing approach, assessing first the burden imposed on Tucson voters by its hybrid system.
All voters in Tucson have an equal right to vote, both during the primary election and during the general election. Each voter may vote for the candidate of her choice in her ward’s primary election. No one may vote in another ward’s primary. And each voter may vote in the general election for one candidate from each ward with a council member position on the ballot.
That the city council elections are staggered is immaterial to the vote denial claim at issue, as Public Integrity Alliance admits in its opening brief. Although half of Tucson’s residents are unable to vote in a primary in a given election year, that burden quickly evens out over time, as the other half of Tucson’s residents will not be able to vote in a primary in the next election year. Ultimately, every voter has an equal opportunity to vote in their own ward’s primary every four years and in the general election every two years.
As is constitutionally required, then, every voter in Tucson has the same voting power as every other voter in the primary and general city council elections. See San Antonio Indep. Sch. Dist. v. Rodriguez, 411 U.S. 1, 35 n.78 (1973) (noting that the constitution protects the right “to participate in state elections on an equal basis with other qualified voters whenever the State has adopted an elective process for determining who will represent any segment of the State’s population”). There is no unequal weighting of votes, no discrimination among voters, and no obstruction or impediment to voting. See Holshouser v. Scott, 335 F. Supp. 928, 933 (M.D.N.C. 1971) (rejecting, in the context of judicial elections, a challenge to a state law providing that judges should be nominated from their respective districts and elected by statewide vote in a general election); Stokes v. Fortson, 234 F. Supp. 575, 578 (N.D. Ga. 1964) (per curiam) (same). The burden on Public Integrity Alliance’s Fourteenth Amendment rights is far from severe. If a burden exists at all, which we doubt, it is at best very minimal.4
As to the governmental interest justifying whatever minimal burden may exist, Tucson has asserted that the hybrid system serves to promote local knowledge and legitimacy, geographic diversity, and city-wide representation on the city council:
Having nominations through primary elections in each ward, using separate ballots for each party, allows the party electorates in each of those wards to make their own choice of a nominee, and simultaneously acts as a guarantee for the City electorate as a whole that each ward’s nominee actually has support among the party members within that ward. Moreover, since nominees compete in the general election only against other candidates nominated in the same ward, . . . ward nominations also help assure that each ward has a local representative on the council, and, conversely, that the Mayor and Council has members who are aware of each ward’s issues, problems, and views.
There is no question that Tucson’s interests are important. The Supreme Court has approved requirements that a city council candidate elected at-large reside in the district with which her seat is affiliated. See Dallas County v. Reese, 421 U.S. 477, 481 (1975) (per curiam) (upholding an election regime providing for countywide balloting for county commission members but requiring that one member reside in and be elected from each district); Dusch v. Davis, 387 U.S. 112, 117 (1967) (same). Candidate-residency requirements promote a similar interest to the one Tucson has articulated: ensuring local representation by and geographic diversity among elected officials. By holding ward-based primaries in addition to maintaining a candidate-residency requirement, Tucson is working to ensure that the candidates nominated in a given ward actually have the support of a majority of their party’s voters in that ward, a conclusion that may not always follow from a candidate-residency requirement alone.
Tucson’s hybrid system represents a careful, longstanding choice, twice affirmed by voters, as to how best to achieve a city council with members who represent Tucson as a whole but reflect and understand all of the city’s wards. It is, in other words, the product of our democratic federalism, a system that permits states to serve “as laboratories for experimentation to devise various solutions where the best solution is far from clear.” Ariz. State Legislature v. Ariz. Indep. Redistricting Comm’n, 135 S. Ct. 2652, 2673 (2015) (quoting United States v. Lopez, 514 U.S. 549, 581 (1995) (Kennedy, J., concurring)).
CONCLUSION
Tucson’s hybrid system for electing members of its city council imposes no constitutionally significant burden on voters’ rights to vote. And Tucson has advanced a valid, sufficiently important interest to justify its choice of electoral system. On the facts alleged herein, the system does not violate the Equal Protection Clause’s one person, one vote commitment.
AFFIRMED.
1 Public Integrity Alliance’s complaint also alleged that Tucson’s system violates the Equal Privileges and Immunities Clause and the Free and Equal Elections Clause of the Arizona Constitution. Ariz. Const. art. II, §§ 13, 21. Because these state-law claims were not developed in the appellate briefing, we consider them abandoned. See Greenwood v. F.A.A., 28 F.3d 971, 977 (9th Cir. 1994).
2 Restrictions that block access to the ballot or impede individual voters or subgroups of voters in exercising their right to vote receive different treatment from rules establishing an overall, generally applicable electoral system. Controversies concerning laws allegedly designed to impede voting are not a historical artifact. See, e.g., Brennan Center for Justice, Voting Restrictions in Place for 2016 Presidential Election (last updated Aug. 10, 2016), http://www.brennancenter.org/sites/default/files/analysis/New_Restrictions_2016.pdf; U.S. Gov’t Accountability Office, GAO-14634, Elections: Issues Related to State Voter Identification Laws 44–56 (2014), http://www.gao.gov/assets/670/665966.pdf. Under Burdick, courts are to assess the “character and magnitude” of the asserted burden, the proven strength of the state’s interest, and whether the extent of the burden is “necessary” given the strength of that interest, so as to ferret out and reject unconstitutional restrictions. 504 U.S. at 434. Recently, in Crawford v. Marion County Election Board, 553 U.S. 181 (2008), a majority of the Supreme Court agreed that in so doing, courts may consider not only a given law’s impact on the electorate in general, but also its impact on subgroups, for whom the burden, when considered in context, may be more severe. Id. at 199–203 (plurality opinion) (recognizing that a voter identification law may have disproportionately burdened certain persons, but holding that petitioners’ evidence was insufficient to permit the Court to quantify the burden imposed on the subgroup); id. at 212–17 (Souter, J., dissenting) (disagreeing as to the sufficiency of evidence in the record regarding the burden imposed on subgroups of voters).
3 The Supreme Court later clarified that the unit system violated equal protection not only because it diluted votes, but because aggregating county units rather than individual votes meant that votes for losing candidates were effectively discarded, solely because of the voter’s county residence. See Gordon v. Lance, 403 U.S. 1, 4–5 (1971).
4 We note that no geographically based vote dilution allegation is before us on appeal, nor has minority or other subgroup vote dilution been alleged.