OPINION AND ORDER DENYING PLAINTIFFS' MOTION FOR SUMMARY JUDGMENT AND GRANTING DEFENDANTS' MOTION FOR SUMMARY JUDGMENT; APPENDICES "A" & "B"

PER CURIAM:

The Hawaii Constitution specifies the use of permanent residents as the relevant population base in apportioning state legislative seats. In a 2012 decision, the Hawaii Supreme Court laid out the appropriate method for determining permanent residents by extracting non-resident military personnel, their dependents, and non-resident students from the total population count. The Hawaii Reapportionment Commission adopted a new legislative apportionment plan to comply with that directive.

This suit asks us to consider the constitutionality of Hawaii's 2012 Reapportionment Plan under the Equal Protection Clause of the United States Constitution. Previously, we considered a motion for a preliminary injunction seeking to halt implementation of the 2012 Reapportionment Plan and to enjoin conducting the 2012 elections under that plan. On May 22, 2012, we denied that request, concluding that the citizens' group seeking the injunction had not established a likelihood of success on the merits of its claim that the permanent resident population basis violates equal protection. Nor did the equities and public interest weigh in favor of an injunction that risked jeopardizing the 2012 primary and general elections. See Kostick v. Nago, 878 F.Supp.2d 1124 (D.Haw.2012).

We now consider the equal protection challenges on cross motions for summary judgment — the citizens' group asks us to declare that the 2012 Reapportionment Plan violates equal protection, and the government seeks judgment in its favor as to those questions. Following extensive briefing and a January 14, 2013 hearing on the cross motions, we DENY Plaintiffs' Motion for Summary Judgment and GRANT Defendants' Motion for Summary Judgment.

For the reasons that follow, we conclude that the 2012 Reapportionment Plan does not violate the United States Constitution. The Commission's reliance on a permanent resident population base, as ordered by the Hawaii Supreme Court, is permissible under the Equal Protection Clause. Likewise, the disparities in the size of the Commission's legislative districts pass constitutional muster.

I. INTRODUCTION

In our May 22, 2012 Order Denying Plaintiffs' Motion for Preliminary Injunction, we extensively reviewed the historical and evidentiary record at that stage. The current record has not changed appreciably, and the cross motions for summary judgment ultimately turn on legal arguments applied to undisputed facts. Accordingly, we draw heavily on the May 22, 2012 Order in explaining the background and context for this apportionment challenge. Where appropriate, we incorporate parts of the May 22, 2012 Order in addressing the cross motions.

Hawaii reapportions its state legislative and federal congressional districts every ten years, after the decennial United States Census (the "Census"), based upon changes in population. See Haw. Const. art. IV, § 1. The Hawaii Constitution as amended in 1992 requires that reapportionment of Hawaii's state legislative districts be based upon "permanent residents," id. § 4, as opposed to the Census count of "usual residents." Any resulting reapportionment is subject to the constitutional principles of "one person, one vote." Reynolds v. Sims, 377 U.S. 533, 557-58, 84 S.Ct. 1362, 12 L.Ed.2d 506 (1964) (citing Gray v. Sanders, 372 U.S. 368, 381, 83 S.Ct. 801, 9 L.Ed.2d 821 (1963)).

In this action, Plaintiffs Joseph Kostick, Kyle Mark Takai, David P. Brostrom, Larry S. Veray, Andrew Walden, Edwin J. Gayagas, Ernest Laster, and Jennifer Laster (collectively, "Kostick" or "Plaintiffs") challenge aspects of the March 30, 2012 Supplement to the 2011 Reapportionment Commission Final Report and Reapportionment Plan (the "2012 Reapportionment Plan"), which Hawaii implemented in 2012 and utilized in its recent 2012 primary and general elections. The Defendants are the members of the 2011 Reapportionment Commission in their official capacities; the Commission itself; and Scott T. Nago, in his official capacity as secretary to the Commission and Hawaii's Chief Elections Officer (collectively, "the Commission" or "Defendants").

The 2012 Reapportionment Plan — fulfilling a mandate from the Hawaii Supreme Court in Solomon v. Abercrombie, 126 Haw. 283, 270 P.3d 1013 (2012) — "extracted" 108, 767 active-duty military personnel, military dependents, and university students from Hawaii's reapportionment population base. Kostick claims that this extraction by itself, and the 2012 Reapportionment Plan's subsequent apportionment of the resulting population base, violate the Equal Protection Clause of the Fourteenth Amendment and "one person, one vote" principles.

Kostick asks the court to (1) declare the 2012 Reapportionment Plan unconstitutional; (2) order the 2011 Hawaii Reapportionment Commission (the "Commission") to formulate and implement a reapportionment plan using the 2010 Census count of "usual residents" of Hawaii as the population base; and (3) order the use of an August 2011 proposed reapportionment plan, which utilized a population base that includes the now-extracted 108,767 people. In addition, Kostick seeks an order requiring an apportionment of state legislative districts that are "substantially equal in population."1

As in our May 22, 2012 Order, we again emphasize that this Opinion addresses only the legal considerations underlying the challenged actions — not whether extracting certain "non-permanent" residents from Hawaii's reapportionment population base is good public policy and not whether Hawaii could or should use "usual residents" as that base. Hawaii has long debated these important and difficult questions, which involve political judgments and require consideration and balancing of competing legislative interests — tasks for which courts are ill suited. See, e.g., Perry v. Perez, 565 U.S. ___, 132 S.Ct. 934, 941, 181 L.Ed.2d 900 (2012) (per curiam) ("[E]xperience has shown the difficulty of defining neutral legal principles in this area, for redistricting ordinarily involves criteria and standards that have been weighed and evaluated by the elected branches in the exercise of their political judgment.") (citations omitted).

In short, we express no opinion as to how Hawaii should define its reapportionment base, but instead examine only the challenged aspects of the 2012 Reapportionment Plan itself. We certainly do not pass on what no one here disputes: Hawaii's military personnel constitute a significant and welcome presence in Hawaii's population.

II. BACKGROUND2

This reapportionment challenge raises issues that are best understood by first examining the historical context. We begin by reviewing the historical and legal factors that the Commission faced in crafting the 2012 Reapportionment Plan. We then set forth the details of Kostick's challenge to the Plan and recount the procedural history of this case.

A. Historical and Legal Context

1. The Census as Population Baseline

The Census counts the "usual residents" of a state. See, e.g., Franklin v. Massachusetts, 505 U.S. 788, 804, 112 S.Ct. 2767, 120 L.Ed.2d 636 (1992) ("`Usual residence'... has been used by the Census Bureau ever since [the first enumeration Act in 1790] to allocate persons to their home States.").

The Census defines "usual residence" as "the place where a person lives and sleeps most of the time" and "is not necessarily the same as the person's voting residence or legal residence." Doc. No. 26, Parties' Stipulated Facts re: the Motion for Preliminary Injunction in Response to Court Order ("Stip. Facts") ¶ 1; Doc. No. 68, Pls.' Separate and Concise Statement of Facts ("CSF") No. 2. The definition thus excludes tourists and business travelers. Stip. Facts ¶ 5; Doc. No. 28-16, Pls.' Ex. H ("Ex. H") at 3. The 2010 Census counted people at their usual residence as of April 1, 2010. Stip. Facts ¶ 2; Pls.' CSF No. 1. Active duty military personnel who were usual residents of Hawaii on April 1, 2010 were or should have been counted by the 2010 Census as part of its count for Hawaii. Stip. Facts ¶ 3; Pls.' Ex. H at 8-9. Similarly, students attending college away from their parental homes are counted where they attend school (i.e., where they "live and sleep most of the time"). Pls.' Ex. H at 5. Students enrolled at a Hawaii university or college who were usual residents of Hawaii on April 1, 2010 were or should have been counted by the 2010 Census as part of the 2010 Census count for Hawaii. Stip. Facts ¶ 4. According to the 2010 Census, Hawaii has a population of 1,360,301 usual residents. Doc. No. 32, First Am. Compl. ("FAC") ¶ 30; Stip. Facts ¶ 32.

After each Census, Hawaii establishes a Reapportionment Commission to implement a reapportionment. See Haw. Const. art. IV, § 2; Haw.Rev.Stat. ("HRS") § 25-1 (2012). The Commission uses the Census's "usual residents" figure as Hawaii's total population for purposes of apportioning Hawaii's federal congressional districts. See Haw. Const. art. IV, § 9; HRS § 25-2(b) (2012) (requiring use of "persons in the total population counted in the last preceding United States census" as the relevant population base). But the Commission does not use the Census figure as the population base for state legislative districts. Instead, Hawaii uses a "permanent residents" count as the relevant population base.

2. Hawaii's Reapportionment Population Base Dilemma

Defining the reapportionment population base for Hawaii's legislative districts has long presented a dilemma, primarily because Hawaii's population has historically contained a large percentage of military personnel — many of whom claim residency in other states and do not vote in Hawaii elections. See, e.g., Burns v. Richardson, 384 U.S. 73, 94, 86 S.Ct. 1286, 16 L.Ed.2d 376 (1966) (referring to "Hawaii's special population problems" stemming from "the continuing presence in Hawaii of large numbers of the military"). The Supreme Court in Burns noted that "at one point during World War II, the military population of Oahu constituted about one-half the population of the Territory." Id. at 94 n. 24, 86 S.Ct. 1286. More recently, well after statehood, the 1991 Reapportionment Commission found that non-resident military personnel constituted "about 14% of the population of Hawaii" with "[a]bout 114,000 nonresident military and their families resid[ing] in this state, primarily on the Island of Oahu." Doc. No. 65-9, Defs.' Ex. G at 6, State of Hawaii 1991 Reapportionment Comm'n, Final Report and Reapportionment Plan at 23; Solomon, 270 P.3d at 1015.3

The vast majority of military and their families live on Oahu because of its many military installations, including Joint Base Pearl Harbor-Hickam, Schofield Barracks, and Kaneohe Marine Corps Air Station. Regardless of whether these individuals claim residency in Hawaii, Hawaii's elected officials still represent them — it is a fundamental constitutional principle that elected officials represent all the people in their districts, including those who do not or cannot vote. See, e.g., Garza v. Cnty. of L.A., 918 F.2d 763, 774 (9th Cir.1990).

A dilemma thus arises because imbalances of potential constitutional magnitude are created whether or not Hawaii's non-resident military and family members are factored into the apportionment base.

If the group is included in the population base but votes elsewhere, Oahu voters potentially have greater "voting power" than residents of other counties. See, e.g., Reynolds, 377 U.S. at 568, 84 S.Ct. 1362 ("[A]n individual's right to vote for state legislators is unconstitutionally impaired when its weight is in a substantial fashion diluted when compared with votes of citizens living in other parts of the State."). That is, the vote of an Oahu voter could count more than that of a non-Oahu voter. See, e.g., Bd. of Estimate v. Morris, 489 U.S. 688, 698, 109 S.Ct. 1433, 103 L.Ed.2d 717 (1989) ("[A] citizen is ... shortchanged if he may vote for ... one representative and the voters in another district half the size also elect one representative."); Chen v. City of Houston, 206 F.3d 502, 525 (5th Cir.2000) ("If total population figures are used in an area in which potentially eligible voters are unevenly distributed, the result will necessarily devalue the votes of individuals in the area with a higher percentage of potentially eligible voters.").

But if this group is excluded, then Oahu residents (particularly residents in an Oahu district with large concentrations of non-resident military) may have diluted representation. See, e.g., Garza, 918 F.2d at 774 ("Residents of the more populous districts ... have less access to their elected representative. Those adversely affected are those who live in the districts with a greater percentage of non-voting populations...."); Chen, 206 F.3d at 525 ("[T]he area with the smaller number of voters will find itself relatively disadvantaged. Despite the fact that it has a larger population — and thus perhaps a greater need for government services than the other community — it will find that its political power does not adequately reflect its size.").

3. The Population Base's Impact on Basic Island Unit Autonomy

The Commission was also driven by a geographic constraint — grounded in Hawaii's history and its Constitution as explained below — to apportion among "basic island units," which correspond to Hawaii's counties. See Appendix B (Hawaii map from National Atlas of the United States, March 5, 2003, http://nationalatlas.gov). And Hawaii's choice of a reapportionment population base has the potential to affect the distribution of political power among these basic island units. Excluding large numbers of non-residents, most of whom live on Oahu, from the population base can — as it did in this instance — result in a gain or loss of legislators (here, Hawaii County gained a State Senate seat that the City and County of Honolulu lost). Stip. Facts ¶ 40. Thus, including or excluding nonpermanent residents could contribute to a subtle shift in power among the counties.

Historically, residents of each basic island unit "have developed their own and, in some instances, severable communities of interests" resulting in "an almost personalized identification of the residents of each county — with and as an integral part of that county." Burns v. Gill, 316 F.Supp. 1285, 1291 (D.Haw.1970) (three judge court). County residents "take great interest in the problems of their own county because of that very insularity brought about by the surrounding and separating ocean." Id. And forty-three years after Gill, many individuals still identify themselves in relation to their island. See, e.g., Doc. No. 66-3, Defs.' Ex. Y, Solomon Decl. ¶ 9 (noting "socio-economic and cultural differences ... that predated statehood" between parts of Maui and the Big Island (as Hawaii Island is often called)).

The integrity of the basic island units reaches back centuries. A three judge court explained in 1965:

Holt v. Richardson, 238 F.Supp. 468, 470-71 (D.Haw.1965) (internal citation omitted), vacated by Burns, 384 U.S. 73, 86 S.Ct. 1286. Likewise, at the 1968 Hawaii Constitutional Convention when implementing apportionment provisions in the Hawaii Constitution, committee members incorporated the concept that:

Doc. No. 65-13, Defs.' Ex. K at 26-27, Standing Comm. Rpt. at 261-62. See also Doc. No. 66-14, Defs.' Ex. KK, McGregor Decl. ¶¶ 5-11 (explaining that each basic island unit's history indicates each was a separate society or community with unique identities and indicating that by the year 1700 each unit was a separate kingdom).

Besides considering the long history of the basic island units in addressing apportionment, Hawaii's 1968 Constitutional Convention also considered the effect of Hawaii's centralized state government, which performs many functions that other states have delegated to local government units. The Convention's apportionment committee explained:

Doc. No. 65-13, Defs.' Ex. K at 27, Standing Comm. Rpt. at 262. The committee gave examples of centralized services such as (1) public education; (2) highways, harbors, and airports; (3) administration and collection of taxes; (4) health and welfare activities; (5) the judicial system; (6) land use districts; (7) fishing, forestry, minerals, agriculture, and land; and (8) labor and industrial relations. Id. These examples of state-wide control largely still exist today.

The committee's conclusion was "obvious and inescapable: if a voter of the State of Hawaii is to have meaningful representation in any kind of government, he must have effective representation from his own island unit in the state legislature." Id. at 28, Standing Comm. Rpt. at 263.

4. The Hawaii Constitution

Crafted to protect basic island unit autonomy, the present-day Hawaii Constitution requires that the population be apportioned on the basis of permanent residents. It also requires that "[n]o district shall extend beyond the boundaries of any basic island unit." Haw. Const. art. IV, § 6. This second requirement is often described as a policy of avoiding "canoe districts," a term that describes legislative districts spanning two basic island units (Counties) separated by ocean. See Doc. No. 65-24, Defs.' Ex. V, Masumoto Decl. ¶ 3.4

Specifically, the Hawaii Constitution provides:

Haw. Const. art. IV, § 4 (emphasis added). After such allocation, the Commission is then required to apportion members of the Hawaii Legislature within those basic island units as follows:

Haw. Const. art. IV, § 6 (emphasis added).

The basic island units correspond to Hawaii's Counties: Hawaii County Hawaii Island); Kauai County (the islands of Kauai and Niihau); Maui County (the islands of Maui, Molokai, Kahoolawe, and Lanai); and the City and County of Honolulu (the island of Oahu).5 See Appendix B. Hawaii's Constitution provides for a bicameral Legislature consisting of 25 senators and 51 representatives. Haw. Const. art. III, §§ 1-3.

The Hawaii Constitution's apportionment provisions have stood since 1992, when Hawaii voters approved a constitutional amendment substituting as the relevant apportionment population base for Hawaii's legislative districts the phrase "the total number of permanent residents" in place of "on the basis of the number of voters registered in the last preceding general election" in Article IV, § 4. See 1992 Haw. Sess. L. 1030-31 (H.B. No. 2327); Solomon, 270 P.3d at 1014-15.

Prior applications of a "registered voter" population base were the subject of litigation and, as analyzed further in this Opinion, ultimately entail many of the same fundamental questions that arise in this action.6 See, e.g., Burns, 384 U.S. at 97, 86 S.Ct. 1286 (upholding a Hawaii apportionment plan based on registered voters that approximated a plan based on population); Travis v. King, 552 F.Supp. 554, 572 (D.Haw.1982) (three judge court) (striking down a Hawaii apportionment plan based on registered voters, primarily because of insufficient justifications for wide disparities in allocation). Indeed, the 1991 Reapportionment Commission utilized a population base of "permanent residents" (extracting — similar to the present action — 114,000 non-resident military members and their families), despite the Hawaii Constitution's (pre-1992 amendment) provision to use "the number of voters registered in the last preceding general election" as the base. This approach was apparently adopted at least in part because of equal protection concerns. See Doc. No. 65-9, Defs.' Ex. G at 4-7, State of Hawaii 1991 Reapportionment Comm'n, Final Report and Reapportionment Plan at 21-24; Solomon, 270 P.3d at 1014-15.

Likewise, the 2001 reapportionment, to which we now turn, extracted nonresident military personnel, their dependents, and non-resident college students as "non-permanent" residents. Solomon, 270 P.3d at 1016-20.

B. Steps Leading to the 2012 Reapportionment Plan

1. The August 2011 Plan

The Commission was certified on April 29, 2011, and promptly began the 2011 reapportionment process. The Hawaii Supreme Court in Solomon describes in exacting detail the process the Commission took in formulating initial and revised apportionment plans. Solomon's description is consistent with the record before this court, and we thus draw extensively from Solomon here:

270 P.3d at 1016 (internal footnote omitted).

The data obtained in May and June 2011 from the military on Hawaii's nonresident military population were apparently deemed insufficient. "The Commission, at its June 28, 2011, meeting, voted 8-1 to apportion the state legislature by using the 2010 Census count — without exclusion of nonresident military and dependents and nonresident students — as the permanent resident base." Id. at 1017.

The Commission staff explained:

Id. at 1018 (brackets in original).

Because of the gaps in the DMDC data, the Commission's August 3, 2011 apportionment plan ("August 3, 2011 Plan") was based on 2010 Census figures without any extractions. Stip. Facts ¶ 27. The Chair of the Commission explained that this August 3, 2011 Plan was "preliminarily accepted for the purpose of public hearings and comment," because of the impending September 26, 2011 statutory deadline for a final plan and the statutory requirement of conducting public hearings. Doc. No. 65-18, Defs.' Ex. P, Marks Decl. ¶ 7.

2. The September 26, 2011 Plan

Further proceedings followed the Commission's initial decision to use the 2010 Census figures without extractions. The Commission was provided with additional data from military sources on Hawaii's "non-permanent military resident population and from Hawaii universities on nonpermanent student resident population." Solomon, 270 P.3d at 1017.

Id. at 1018 (brackets in original). The Commission staff's "Extraction A" listed 16,458 active duty military, their dependents, and out-of-state university students (mostly on Oahu); its "Extraction B" listed 73,552; and its "Extraction C" listed 79,821. Id. Additionally, an "August 17, 2011 `Staff Summary' show[ed] a state population of 47,082 non-permanent active duty military residents, 58,949 military dependents, and 15,463 out-of-state university students" totaling 121,494 "nonpermanent" residents. Id. at 1019.

The Commission held a September 13, 2011 public hearing in Hilo, Hawaii. It received testimony from State Senator Malama Solomon ("Solomon") and three members of the Hawaii County Democratic Committee, advocating extraction of the 121,494 "non-permanent" residents from the apportionment population base. Such an extraction would increase Hawaii County's Senate seats from three to four. Id. Hawaii Governor Neil Abercrombie also supported that extraction, indicating that based upon the State Attorney General's preliminary view, "counting nonresidents is not warranted in law." Id.

On September 19, 2011, after much debate, "[t]he Commission adopted a final reapportionment plan that computed the permanent resident base by excluding 16,458 active duty military and out-of-state university students from the 2010 census population of 1,330,301." Id. at 1020; Stip. Facts ¶ 32.7 That is, it chose "Extraction A," primarily because of the certainty of that data. The resulting apportionment allocated "as to the senate, 18 seats to Oahu County, 3 seats for Hawaii County, 3 seats for Maui County, and 1 seat for Kauai County." Solomon, 270 P.3d at 1020. The Commission filed this plan on September 26, 2011 ("the September 26, 2011 Plan"). Id.; Stip. Facts ¶ 32.

3. Challenges to the September 26, 2011 Plan: Solomon v. Abercrombie; and Matsukawa v. State of Hawaii 2011 Reapportionment Commission

On October 10, 2011, Solomon and the three members of the Hawaii County Democratic Committee filed a petition in the Hawaii Supreme Court, challenging the September 26, 2011 Plan. Solomon, 270 P.3d at 1020. The next day, Hawaii County resident Michael Matsukawa filed a similar petition in the Hawaii Supreme Court. Id.; Stip. Facts ¶ 33. Among other claims, these petitions asserted that the Commission violated the Hawaii Constitution's requirement to base a reapportionment on "permanent residents" by failing to extract all nonresident military, their dependents, and non-resident students. Solomon, 270 P.3d at 1020. Solomon's petition asserted that the Commission knew that extracting only 16,000 non-residents would not trigger the loss of an Oahu-based Senate seat, and that "the fear of Oahu's loss of this senate seat was the driving force" for the extraction. Id. They sought an order requiring the Commission to prepare and file a new reapportionment plan for the State legislature that uses a population base limited to "permanent residents" of the State of Hawaii. Stip. Facts ¶ 33.

On January 4, 2012, the Hawaii Supreme Court issued orders in the Solomon and Matsukawa proceedings that invalidated the September 26, 2011 Plan as having disregarded Article IV, § 4 of the Hawaii Constitution. The Hawaii Supreme Court, among other things, ordered the Commission to prepare and file a new reapportionment plan allocating members of the State legislature among the basic island units using a permanent resident population base. Id. ¶ 34. On January 6, 2012, the Hawaii Supreme Court issued an opinion covering both the Solomon and Matsukawa proceedings. Id. ¶ 35.

As for the requirement in Article IV, §§ 4 and 6, for the Commission to apportion the State legislature by using a "permanent resident" base, the opinion held that the requirement "mandate[s] that only residents having their domiciliary in the State of Hawaii may be counted in the population base for the purpose of reapportioning legislative districts." Solomon, 270 P.3d at 1022 (quoting Citizens for Equitable & Responsible Gov't v. Cnty. of Hawaii, 108 Haw. 318, 120 P.3d 217, 221 (2005)). To determine "the total number of permanent residents in the state and in each county," the Commission was required "to extract non-permanent military residents and non-permanent university student residents from the state's and the counties' 2010 Census population." Id. It directed that,

Id. at 1024. It appears that the parties did not raise, and the Hawaii Supreme Court did not address, equal protection concerns.

4. The 2012 Reapportionment Plan

Soon after Solomon was issued, the Commission commenced a series of public meetings and obtained additional information regarding military personnel, their family members, and university students. The Commission eventually extracted 42,332 active duty military personnel, 53,115 military dependents, and 13,320 students from the 2010 Census population of "usual residents." Stip. Facts ¶¶ 8, 10, 14, 36. This extraction totaled 108,767 persons, resulting in an adjusted reapportionment population base of 1,251,534. Id. ¶ 37.

Active duty military included in the 2010 Census were extracted if they "declared a state other than Hawaii as their home state for income tax purposes." Doc. No. 28-12, Pls.' Ex. D at 8. That is, they were extracted "based on military records or data denoting the personnel's state of legal residence." Stip. Facts ¶ 8.

The extracted military family members were identified by associating them with their active duty military sponsor. In other words, the Commission extracted military dependents who were associated with or attached to an active duty military person who had declared a state of legal residence other than Hawaii. Id. ¶ 10. The military did not provide the Commission with any data regarding the military dependents' permanent or non-permanent residency other than their association or attachment to an active duty military sponsor who had declared a state of residence other than Hawaii. Id. ¶ 12.

The students were extracted solely on the basis of (a) payment of nonresident tuition or (b) a home address outside of Hawaii. Id. ¶¶ 14, 18-19. The students were from the University of Hawaii System, Hawaii Pacific University, Chaminade University, and Brigham Young University Hawaii. Id. ¶ 15.

After extraction, the Commission reapportioned the adjusted population base of 1,251,534 "permanent residents" by dividing the base by the constitutionally-defined 25 Senate seats and 51 House seats. Id. ¶ 37. This resulted in an ideal Senate district of 50,061 permanent residents, and an ideal House district of 24,540 permanent residents. Id. The Commission then reapportioned within the four basic island units as set forth in Article IV, § 6 of the Hawaii Constitution, and as guided by the criteria set forth in that provision.

As for the Senate districts, under the 2012 Reapportionment Plan: (a) the largest Senate district (Senate district 8, Kauai basic island unit) contains 66,805 permanent residents, which is 16,744 (or 33.44 percent) higher than the ideal Senate district of 50,061 permanent residents; and (b) the smallest Senate district (Senate district 1, Hawaii basic island unit) contains 44,666 permanent residents, which is 5,395 fewer (or 10.78 percent less) than the ideal. Id. ¶ 38. Thus, the maximum deviation for the Senate districts is 44.22 percent. The 2012 Reapportionment Plan resulted in one Senate seat moving from the Oahu basic island unit to the Hawaii basic island unit. Id. ¶ 40.

As for House districts: (a) the largest House district (House district 5, Hawaii basic island unit) contains 27,129 permanent residents, which is 2,589 (or 10.55 percent) higher than the ideal House district of 24,540 permanent residents; (b) the smallest House district (House district 15, Kauai basic island unit) contains 21,835 permanent residents, which is 2,705 fewer (or 11.02 percent less) than the ideal. Id. ¶ 39. The maximum deviation for the House districts is 21.57 percent.8

As explained more fully when we address Kostick's malapportionment claim, the extent of the deviations is driven primarily by the Commission's decision to continue to avoid canoe districts. See Doc. No. 65-22, Defs.' Ex. T at 32, 2012 Reapportionment Plan at 21. Canoe districts were eliminated in the 2001 reapportionment, after being imposed in 1982 when a three judge court found a 1981 reapportionment plan to be unconstitutional and ordered use of an interim plan that utilized canoe districts. See Doc. No. 65-4, Defs.' Ex. C-1 (April 27, 1982 Final Report and Recommendations of Special Masters in Travis v. King). The 2001 Reapportionment Commission eliminated canoe districts, concluding after experience and public input that such districts were ineffective. See, e.g., Doc. No. 65-15, Defs.' Ex. M at 11, 2001 Reapportionment Plan at 25; id. at 14, 2001 Reapportionment Plan at A-209.

The 2012 Reapportionment Plan was adopted and filed on March 8, 2012, with notice published on March 22, 2012. Stip. Facts ¶ 36.

C. Procedural History

This action was filed on April 6, 2012. The Complaint requested a three-judge district court pursuant to 28 U.S.C. § 2284. On April 10, 2012, Judge J. Michael Seabright granted the request for a three-judge district court, determining that the constitutional claims were "not insubstantial," as necessary to convene such a court. See, e.g., Goosby v. Osser, 409 U.S. 512, 518, 93 S.Ct. 854, 35 L.Ed.2d 36 (1973). On April 17, 2012, the Chief Judge of the Ninth Circuit Court of Appeals appointed the present panel, Ninth Circuit Judge M. Margaret McKeown, and District Judges J. Michael Seabright and Leslie E. Kobayashi.

Kostick filed a Motion for Preliminary Injunction on April 23, 2012, which we heard on May 18, 2012, and denied on May 22, 2012. See Doc. No. 52 (Kostick, 878 F.Supp.2d 1124). The cross motions for summary judgment were filed on October 1, 2012. Doc. Nos. 64, 67. Oppositions were filed on October 29, 2012, Doc. Nos. 72, 74, and corresponding Replies were filed on November 19, 2012, Doc. Nos. 76, 77. The court heard oral argument from the parties on January 14, 2013.

III. STANDARD OF REVIEW

Summary judgment is appropriate where there is no genuine issue of material fact and the moving party is entitled to judgment as a matter of law. Fed. R.Civ.P. 56(a). Rule 56(a) mandates summary judgment "against a party who fails to make a showing sufficient to establish the existence of an element essential to the party's case, and on which that party will bear the burden of proof at trial." Celotex Corp. v. Catrett, 477 U.S. 317, 322, 106 S.Ct. 2548, 91 L.Ed.2d 265 (1986); see also Broussard v. Univ. of Cal. at Berkeley, 192 F.3d 1252, 1258 (9th Cir.1999).

As noted earlier, the relevant historical facts are undisputed: "Where a case turns on a mixed question of law and fact and, as here, the only disputes relate to the legal significance of undisputed facts, `the controversy collapses into a question of law suitable to disposition on summary judgment.'" Blue Lake Rancheria v. United States, 653 F.3d 1112, 1115 (9th Cir.2011) (quoting Thrifty Oil Co. v. Bank of Am. Nat'l Trust & Sav. Ass'n, 322 F.3d 1039, 1046 (9th Cir.2003)).

IV. DISCUSSION

Kostick makes a bifurcated equal protection challenge to Hawaii's reapportionment plan.9 He first protests the extraction of non-resident military personnel, their dependents, and non-resident students. He argues that using a population base that does not include the extracted individuals violates equal protection. Next, even if such an extraction is allowed, Kostick claims that deviations in the 2012 Reapportionment Plan exceed constitutional limits.

Before turning to these claims, we address the threshold issue of standing. The Commission argues that Plaintiffs lack standing to assert either claim because they have suffered no injury.10 It is enough, for justiciability purposes, that at least one party with standing is present. See Dep't of Commerce v. U.S. House of Representatives, 525 U.S. 316, 330, 119 S.Ct. 765, 142 L.Ed.2d 797 (1999).

With regard to Count One, the result of the challenged extraction of military and other residents was the loss of an Oahu Senate seat. Stip. Facts ¶ 40. All Plaintiffs were "usual residents" of Oahu with a military connection (aside from Walden). FAC ¶¶ 1-8; Doc. No. 38-4, Gayagas Decl. ¶ 5. Some of those with military connections, such as Jennifer Laster, were or may have been "extracted" from the reapportionment base despite being permanent residents of Hawaii. Because these individuals have suffered the injury of losing a representative, Plaintiffs have standing to bring Count One.

With regard to Count Two, the Commission argues that Plaintiffs lack standing to challenge the apportionment deviations because no Plaintiff resides on Kauai, the island that is most under-represented in the State Senate. See Doc. No. 72 at 19, Defs.' Opp'n at 12. Id. What the Commission overlooks is that three of the Plaintiffs — Kostick, Walden, and Veray — do live in underrepresented districts, albeit not on Kauai. They have standing to challenge the Commission's apportionment plan, which disadvantages them compared to residents of over-represented districts. Although the decisions cited by the Commission support the proposition that residents of overrepresented districts cannot challenge reapportionment plans, the same logic does not support the Commission's argument that residents of an underrepresented district cannot challenge a reapportionment plan as a whole. See Fairley v. Patterson, 493 F.2d 598, 603-04 (5th Cir. 1974) (holding that an intervenor from an underrepresented district "had standing to attack the original malapportioned districts," including two others in which he did not reside) (emphasis added).

A. Count One (Equal Protection Challenge: Population Basis)

Count One centers on Hawaii's apportionment of its population on a permanent resident basis, extracting non-resident military, their dependents, and non-resident students. At the preliminary injunction stage, we found that Kostick was unlikely to succeed on the merits of this issue. Kostick proffers very little new evidence in support of his position on summary judgment, and the facts are not in dispute.

We conclude that Hawaii's choice of a permanent resident population base is constitutionally permissible. There is no evidence that Hawaii discriminated unreasonably among non-resident groups; rather, the State extracted all nonpermanent populations that exist in sufficient numbers to affect the apportionment of districts and about which it could obtain relevant, reliable data. Neither is there evidence that Hawaii's method of extraction was irrational. The Commission reasonably relied upon available statistics. Nothing suggests that the methods resulted in the exclusion of permanent residents from the population basis in numbers sufficient to affect legislative apportionment.

1. Standard Governing Choice of Population Basis

The Supreme Court has emphasized that "`the Equal Protection [Clause's requirement] that the seats in both houses of a bicameral state legislature must be apportioned on a population basis' ... requires only `that a State make an honest and good faith effort to construct districts... as nearly of equal population as is practicable,' for `it is a practical impossibility to arrange legislative districts so that each one has an identical number of residents, or citizens, or voters.'" Brown v. Thomson, 462 U.S. 835, 842, 103 S.Ct. 2690, 77 L.Ed.2d 214 (1983) (quoting Reynolds v. Sims, 377 U.S. 533, 568, 577, 84 S.Ct. 1362, 12 L.Ed.2d 506 (1964)) (emphasis added). By recognizing the alternative population bases of "residents, or citizens, or voters," id., the Court contemplated that a state's redistricting efforts would entail not only the line-drawing necessary to create districts, but also the choice of how to define the population.

Kostick's argument that the governing standard is "close constitutional scrutiny," requiring a "substantial and compelling reason" for extracting segments of the total population, finds no support in precedent. Doc. No. 74 at 9-10, Pls.' Opp'n at 1-2. He draws this requirement from Dunn v. Blumstein, 405 U.S. 330, 335, 92 S.Ct. 995, 31 L.Ed.2d 274 (1972). The Dunn decision, however, is inapposite because it considered a challenge not to state reapportionment, but to a state's durational residency requirement for the right to vote. Similar voting rights cases upon which Kostick relies are likewise inapt. See, e.g., Evans v. Cornman, 398 U.S. 419, 90 S.Ct. 1752, 26 L.Ed.2d 370 (1970) (applying close constitutional scrutiny to Maryland's denial of voting rights to residents of a National Institutes of Health enclave).11

The Supreme Court applies this higher standard to cases alleging infringement of the fundamental right to vote, in contrast to equal representation or equal voting power challenges in the context of reapportionment. In practice, the standard for this latter category approximates rational-basis review. See Brown, 462 U.S. at 844, 103 S.Ct. 2690 (upholding a Wyoming reapportionment plan because it resulted from "the consistent and nondiscriminatory application of a legitimate state policy") (emphasis added). We invoke the Brown standard here.

2. Use of Permanent Resident Population Base

In considering Kostick's claim, we have the benefit of longstanding Supreme Court precedent, including the 1966 decision stemming from Hawaii's earlier apportionment plan — Burns v. Richardson, 384 U.S. 73, 86 S.Ct. 1286, 16 L.Ed.2d 376 (1966). Just two years earlier, in Reynolds v. Sims, the Court decided a seminal case on the "right of a citizen to equal representation." 377 U.S. at 576, 84 S.Ct. 1362. The Reynolds decision reasoned that under the Equal Protection Clause, "an individual's right to vote for state legislators is unconstitutionally impaired when its weight is in a substantial fashion diluted when compared with votes of citizens living in other parts of the State." Id. at 568, 84 S.Ct. 1362. The Court accordingly held that "the seats in both houses of a bicameral state legislature must be apportioned on a population basis," id., but "carefully left open the question what population was being referred to." Burns, 384 U.S. at 91, 86 S.Ct. 1286.

This question did not remain unaddressed for long. In Burns, the Court considered whether it was permissible for Hawaii to use registered voters rather than a broader population as the basis for districting. In discussing Reynolds, the Court "start[ed] with the proposition that the Equal Protection Clause does not require the States to use total population figures derived from the federal census as the standard by which ... substantial population equivalency is to be measured." Id.

The Burns decision explained what constitutes a "permissible population basis." Id. at 91-93, 86 S.Ct. 1286. One such permissible population basis, discussed in Reynolds, was the total population. Had Burns left the matter there, Kostick might have a different case. However, in Burns the Court went on to acknowledge the power of states to "[ex]clude aliens, transients, short-term or temporary residents" from "the apportionment base," noting that "[t]he decision to include or exclude any such group involves choices about the nature of representation with which we have been shown no constitutionally founded reason to interfere." Id. at 92, 86 S.Ct. 1286.

Although Hawaii earlier chose to use the registered voter base, the Court foreshadowed Hawaii's later decision to shift to a permanent resident base: "Hawaii's special population problems might well have led it to conclude that state citizen population rather than total population should be the basis for comparison." Id. at 94, 86 S.Ct. 1286. The Court went on to quote the district court's finding that "[i]f total population were to be the only acceptable criterion upon which legislative representation could be based, in Hawaii, grossly absurd and disastrous results would flow." Id. Such results derived from Hawaii's "large numbers of the military" as well as "tourists" — both of which "tend to be highly concentrated on Oahu and, indeed, are largely confined to particular regions of that island." Id. Accordingly, "[t]otal population figures may thus constitute a substantially distorted reflection of the distribution of state citizenry." Id. The Court concluded that "[i]t is enough if it appears that the distribution of registered voters approximates distribution of state citizens or another permissible population base." Id. at 95, 86 S.Ct. 1286. In short, the Court specifically sanctioned the use of an "approximate[] distribution of state citizens" as a "permissible population base." Id.12

Kostick argues that the 2012 Reapportionment Plan is not sanctioned by Burns because it does not identify the "permissible population base" that the "permanent residents" standard approximates, and, even if "state citizens" is the permissible comparable basis, it is not a substantial duplicate of a plan constructed on that basis. Doc. No. 67 at 38, Pls.' Mot. at 27. Neither of these arguments is persuasive.

Because Burns recognizes Hawaii's prerogative to exclude the temporary populations of non-resident servicemembers, their dependents, and non-resident students from the definition of "permanent residents," Hawaii's definition of "permanent residents" constitutes "state citizens" by another name. The State need not demonstrate that its plan under the "permanent residents" standard is a duplicate of a plan made on another permissible basis. Burns explicitly benchmarked the registered voter population basis against a "state citizen population," which was extrapolated by effectively deducting the "military population of Oahu" from the "total population." Burns, 384 U.S. at 95, 86 S.Ct. 1286. The plan before us does the same thing, but in a manner more finely tuned than the plan considered in Burns — it does not deduct the entire "military population" but only non-resident military personnel and dependents, as well as nonresident students, to approximate the permanent resident base.13

Kostick argues that "state citizen" is defined under the Fourteenth Amendment as ordinary residents of a state, which Kostick contends includes military and excludes aliens — the opposite of the 2012 Reapportionment Plan. Doc. No. 74 at 31-32, Pls.' Opp'n at 23-24; see U.S. Const. amend. XIV (providing that "[a]ll persons born or naturalized in the United States, and subject to the jurisdiction therefore, are citizens of the United States and of the State wherein they reside") (emphasis added). Nowhere in Burns, however, did the Court suggest that the "state citizen population" it considered a permissible basis for apportionment was that defined by Kostick's interpretation of the Fourteenth Amendment. To the contrary, when Burns approved of "state citizenry" as a permissible population base, it understood that the term could exclude the military stationed in Hawaii. Burns, 384 U.S. at 94, 86 S.Ct. 1286.

The Ninth Circuit's decision in Garza further confirms that a state need not apportion on the basis of total population. Kostick characterizes Garza as "holding that if there is a conflict between voting equality and representational equality, the latter prevails." Doc. No. 74 at 21, Pls.' Opp'n at 13. Although the court noted and discussed the tension between these two principles, the decision upholding a judicially-imposed plan for Los Angeles County based on Census population did not mandate use of total population in all circumstances. Notably, the court stated that while Burns permitted states to consider the distribution of the voting population as well as that of total population, "[i]t does not, however, require states to do so." Garza, 918 F.2d at 774; see also Daly v. Hunt, 93 F.3d 1212, 1225 (4th Cir.1996) (explaining that "[t]he more important lesson that may be gleaned from Burns is that courts should generally defer to the state to choose its own apportionment base, provided that such method yields acceptable results"). In Garza, California law expressly required Los Angeles County to redistrict on the basis of total population. Garza, 918 F.2d at 774 (citing California Elections Code § 35000). By contrast, as discussed above, the Hawaii Constitution, as interpreted by the Hawaii Supreme Court, requires use of a "permanent resident" population basis rather than total population.

3. Discrimination Among Non-Resident Groups

To be sure, if Hawaii's exclusion was carried out with an eye to invidiously targeting only certain non-resident groups, it would raise serious constitutional concerns. See Carrington v. Rash, 380 U.S. 89, 95, 85 S.Ct. 775, 13 L.Ed.2d 675 (1965) (holding that discrimination against the military in provision of the right to vote is unconstitutional); Burns, 384 U.S. at 95 & n. 25, 86 S.Ct. 1286 (suggesting that Carrington required equal treatment of the military for the purpose of reapportionment). Kostick, however, provides no evidence that Hawaii's exclusion of non-resident servicemembers, their dependents, and non-resident students was carried out with any aim other than to create a population basis that reflects Hawaii's permanent residents. Notably, the Hawaii Supreme Court's decision that prompted the current plan faulted the Commission, not for failing to exclude specific groups in the redistricting effort, but for failing to exclude all "non-permanent residents" for which the State had data. Solomon, 270 P.3d at 1021.

The Commission's reapportionment efforts over the years reflect its general concern with excluding non-permanent residents from the population basis, rather than with invidiously targeting certain groups. For example, in 1991, the Commission initially excluded minors as well as the military and their dependents. Doc. No. 34-20, Defs.' Ex. 30 at 3, 1991 State of Hawaii Reapportionment Comm'n, Final Report and Reapportionment Plan at 21. The Commission also sought to exclude aliens, but was informed that no data was available to do so. Id. at 22. The Commission noted at that time that "[o]ther groups, such as nonresident students, are statistically insignificant and cannot be easily placed in specific census blocks. The Commission, therefore, decided to eliminate those transients which could be identified to a particular census block and which constituted the vast majority of transients included in the census counts: nonresident military." Id. at 23.

Since the efforts of the 1991 Commission, the State has diligently considered how and whether other non-permanent resident groups could be removed from the population base. Subsequent commissions have considered excluding aliens, but have been unable to do so because of lack of data. See Doc. No. 34-21, Defs.' Ex. 30 at 22, 2001 State of Hawaii Reapportionment Comm'n Reapportionment Plan at A-226; Doc. No. 33-5, Rosenbrock Decl. ¶ 15 (discussing 2011 Commission). Although data regarding aliens was in short supply, the Commission in 2011 conscientiously renewed contacts with university officials and successfully obtained data to exclude non-resident students. Doc. No. 33-6, Marks Decl. ¶¶ 18, 20.

Kostick nonetheless criticizes the fact that the State extracted military personnel, their dependents and students, but not illegal aliens, minors, federal workers, and prisoners, institutionalized persons, and even the unemployed. Doc. No. 74 at 11-12, Pls.' Opp'n at 3-4. Several of these comparator groups are not relevant: Kostick does not seriously suggest that minors, the unemployed, and prisoners are not generally Hawaii residents who lack the "present intention of establishing [their] permanent dwelling place" in Hawaii. HRS § 11-13(2) (2012).14 The Commission tried — but was unable — to get information regarding aliens, as discussed above. Doc. No. 65-18, Defs.' Ex. P, Marks Decl. ¶ 4; see also Doc. No. 65-16, Defs.' Ex. N, Rosenbrock Decl. ¶ 8, 15 (noting the Commission's understanding that prior efforts had shown that "reliable information that identified the number or census block location of aliens in Hawaii" was lacking). Kostick's passing argument with reference to federal workers is unavailing: he presents no evidence as to the number of federal workers in Hawaii, nor does he seriously contend that the vast majority of these workers are anything but bona fide State residents. The record provides no indication that these aliens, minors, or incarcerated populations are concentrated in areas of the State in such a way as to affect the apportionment of districts.

To summarize, the 2012 Reapportionment Plan resulted from a careful and comprehensive process free from any taint of arbitrariness or invidious discrimination against minority groups or the military. And the record is likewise clear that the Commission faced a mathematical reality — the inclusion or exclusion of non-permanent military and military dependents causes an equal imbalance in either representational equality or electoral equality.

Over and over, the Supreme Court has explained that reapportionment involves fundamental choices about the nature of representation, where states have discretion (absent discrimination) to exercise political judgment to balance competing interests. See, e.g., Gaffney v. Cummings, 412 U.S. 735, 749, 93 S.Ct. 2321, 37 L.Ed.2d 298 (1973); Karcher v. Daggett, 462 U.S. 725, 740, 103 S.Ct. 2653, 77 L.Ed.2d 133 (1983); Brown, 462 U.S. at 847-48, 103 S.Ct. 2690; Miller v. Johnson, 515 U.S. 900, 915, 115 S.Ct. 2475, 132 L.Ed.2d 762 (1995); Perry, 132 S.Ct. at 941; and Tennant v. Jefferson Cnty. Comm'n, ___ U.S. ___, 133 S.Ct. 3, 5, 183 L.Ed.2d 660 (2012). Given the record presented to us, we simply have "no constitutionally founded reason to interfere." Burns, 384 U.S. at 92, 86 S.Ct. 1286.

In the absence of discrimination, this principle of deference is dispositive. Daly was apt in applying this "overriding theme in the Court's prior apportionment cases weighing against judicial involvement," when it reiterated that "[t]his is a decision that should be made by the state, not the federal courts, in the inherently political and legislative process of apportionment." 93 F.3d at 1227 (citing Burns, 384 U.S. at 92, 86 S.Ct. 1286). See also Chen, 206 F.3d at 528. The choice facing the Commission — between representational or electoral equality — was quintessentially political, requiring "the sort of policy judgments for which courts are, at best, ill suited." Perry, 132 S.Ct. at 941.

4. Implementation of Extraction

Kostick claims that even if the chosen permanent resident base is permissible, the extraction mechanism does not pass constitutional muster because it also eliminates from the reapportionment basis some Hawaii citizens, such as plaintiff Jennifer Laster. But Hawaii's methods need not have "`[m]athematical exactness;'" rather Hawaii must simply employ procedures that "make an honest and good-faith effort to construct ... districts" in such a way that the number of permanent residents in each district are as "`equal ... as is practicable.'" Gaffney, 412 U.S. at 743, 93 S.Ct. 2321 (quoting Reynolds, 377 U.S. at 577, 84 S.Ct. 1362). The method adopted must yield "a reasonable approximation of," and track the distribution of, a permissible population basis. Burns, 384 U.S. at 92-93, 95, 86 S.Ct. 1286. In other words, it is not enough for Kostick to show that the extraction excluded some State citizens: he must also show that the exclusion was egregious enough to result in an unequal distribution of the citizen population base among the various districts. He has not done so.

a. Military

To extract non-resident military, Hawaii used the servicemember's chosen state for taxation to determine residency. Doc. No. 28-9, Pls.' Mot. Ex. A at 10-11, Office of Elections, Non-Permanent Population Extraction for 2011 Reapportionment and Redistricting — Addendum D-8 to D-9. Servicemembers are not automatically excluded from residency. They are given an opportunity to identify their state of residence for purposes of taxation. See Doc. No. 34-7, Defs.' Ex. 17 at 1 ("Instructions of Certification of State of Legal Residence."). By designating a state other than Hawaii as their state of taxation, servicemembers avoid paying Hawaii resident state taxes. HRS § 235-7 (2012). Servicemembers are informed that state residency requires "physical presence ... with the simultaneous intent of making it your permanent home and abandonment of the old State of legal residence/domicile." Doc. No. 34-7, Defs.' Ex. 17 at 1. This language tracks the residency requirement under Hawaii law, which requires a "present intention of establishing the person's permanent dwelling place" in the State. HRS § 11-13(2) (2012). By indicating a different state for the purposes of taxation, a servicemember declares that he or she has no present intention of establishing his "permanent dwelling place" in Hawaii.

Reliance on this declaration is a rational means of determining a servicemember's residence under Hawaii law. See Burns, 384 U.S. at 92 n. 21, 86 S.Ct. 1286 ("The difference between exclusion of all military and military-related personnel, and exclusion of those not meeting a State's residence requirements is a difference between an arbitrary and a constitutionally permissible classification.") (emphasis added). Hawaii does nothing to prohibit members of the military from establishing residency in the State. Because, on this record, Hawaii does not resort to overbroad means to exclude non-resident servicemembers, the extraction is permissible. See id. at 95, 86 S.Ct. 1286 (noting that there was no attempt to disenfranchise the military by preventing them from becoming State residents).

Kostick's criticism of the means by which the Commission identified nonpermanent servicemembers repeats arguments made at the preliminary injunction stage and fares no better this time. He asserts that "[t]here may be little correlation between the place where a servicemember pays state taxes, and where she is actually located. Nor does the DD2058 form ask the servicemember to declare where they are located, or where they intend to remain." Doc. No. 67 at 28, Pls.' Mot. at 17. No one disputes that many extracted servicemembers are actually located in Hawaii. That fact alone does not establish that they are permanent residents. And, as noted above, the form does in fact inquire regarding a servicemember's intent to remain in Hawaii.

Kostick also attacks the information from the Defense Manpower Data Center that was utilized to extract military personnel as being not detailed enough to provide the State with a reasonable basis for determining who to extract. Id. Specifically, he contends that the information provided no way to confirm that the servicemembers extracted based on the data were in Hawaii on Census day and included in the "usual residents" count from which the extractions were taken. Id. The possibility that some servicemembers extracted on the basis of the Defense Manpower Data Center statistics were not present on Census day does not render the Commission's methodology unreasonable. In this context, Hawaii's extraction methods need not have "`[m]athematical exactness.'" Gaffney, 412 U.S. at 743, 93 S.Ct. 2321.

For similar reasons, Kostick's criticism of the extraction based upon its effect in other states is unavailing. Because every state other than Hawaii and Kansas uses the actual Census count for reapportionment, he contends "those individuals who were counted by the Census as Hawaii residents, but extracted from the Hawaii population for reapportionment purposes, are not counted anywhere for state reapportionment." Doc. No. 74 at 30, Pls.' Opp'n at 22. This observation is an insufficient reason to conclude that Hawaii's reapportionment methods were unreasonable. Inherent in the Supreme Court's reasoning that "the decision to include or exclude [groups such as transients, short-term or temporary residents]" is generally a question "about the nature of representation with which we have been shown no constitutionally founded reason to interfere," Burns, 384 U.S. at 92, 86 S.Ct. 1286, is that different states may provide different answers to that question.15

b. Military Dependents

The Commission presumed that all dependents of non-resident servicemembers are also non-residents. Kostick points to plaintiff Jennifer Laster — and only to Jennifer Laster — to argue that this approach improperly eliminates residents and registered voters from the population base. Doc. No. 35-13, Defs.' Ex. 44 at 15. This evidence fails to show that Hawaii's exclusion is overbroad. The record shows otherwise — the military informed Hawaii in 1991 that 98 percent of families of nonresident servicemembers had the same residency as that of the servicemember. Doc. No. 34-20, Defs.' Ex. 30 at 3, 1991 State of Hawaii Reapportionment Comm'n, Final Report and Reapportionment Plan at 21. Defendants submit that a 2012 Defense Department paper on the concerns that military spouses possess with respect to state occupational licensing laws notes that military spouses were ten times as likely as their civilian counterparts to have moved across state lines in the past year. Doc. No. 66-16, Defs.' Ex. MM at 3. Kostick presents no new evidence since the preliminary injunction stage that the status quo has changed. Given this failure, the record supports finding that Hawaii's assumptions about dependents were rational and support the extraction.

c. Students

Hawaii extracted students from Brigham Young University Hawaii, Hawaii Pacific University, Chaminade University, and the University of Hawaii System. Doc. No. 33-5, Rosenbrock Decl. ¶ 9. Other than noting that students from other universities were not included, the record is bereft of evidence to suggest that the number of students at any remaining universities was substantial enough to make any difference. Rather, the evidence indicates that these universities are the four "major colleges in Hawaii." Id. As Gaffney suggests, the Commission need not have considered small institutions that are attended by too few non-resident students to affect the allocation of state residents.

The tests established for excluding non-resident students within the four universities were reasonably designed to meet the goals of identifying nonresidents. For Brigham Young University Hawaii, Hawaii Pacific University, and Chaminade University, a student is considered a non-resident if the student lists a "home address" outside Hawaii. It falls within the State's discretion to use this method to determine which individuals are transient residents. Identifying a home address in Hawaii fairly reflects a "present intention of establishing the person's permanent dwelling place" in the State, as required for residency under Hawaii law. HRS § 11-13(2).

For the University of Hawaii System, any student paying out-of-state tuition is considered a non-resident. The essential requirements for establishing residency for tuition purposes for the University of Hawaii System are (1) bona fide residency, shown by various methods, most importantly, registering to vote and paying state taxes, (2) for a period of twelve months. Haw. Admin. Rules § 20-4-6. Kostick takes issue with the year-long residency requirement: a student is not counted as a Hawaii resident for the purposes of redistricting unless he has been a resident for one year. Doc. No. 36 at 20, Pls.' Rep. at 15 & n. 5. He reminds us that in Dunn, the Supreme Court held that imposing a year long durational requirement for the purposes of voting was constitutionally impermissible. Dunn, 405 U.S. at 360, 92 S.Ct. 995.

As we explained above, the standard applicable to impediments on the fundamental right to vote differs from that applicable to the right to equal representation for purposes of state legislative apportionment. Kostick provided no evidence at the preliminary injunction stage of even a single student who had become a resident of Hawaii within the one-year period but was excluded from the population basis. The same evidence is missing at this stage. As with the military extraction, the possibility of some mathematical imprecision does not render the methodology constitutionally unsound, particularly in the absence of any evidence that the discrepancy would affect apportionment. For the same reason, the fact that Hawaii based the extraction on spring 2010 enrollment, rather than enrollment on Census day, is of no moment.

In sum, Hawaii's decision to extract three categories of non-permanent residents was legitimate and its methods of extraction were reasonable. Defendants are therefore entitled to judgment as a matter of law on Count One.

B. Count Two (Equal Protection Challenge: Mal-Apportionment)

In Count Two, Kostick contends that the Commission violated the Equal Protection Clause by apportioning Hawaii's legislative districts unequally — leading to a maximum deviation of approximately 44 percent for Hawaii's Senate districts and approximately 22 percent for its House districts. We first summarize applicable apportionment standards and then analyze the 2012 Reapportionment Plan in that light.

1. Legal Requirements for Apportionment

The "basic aim of legislative apportionment" is achieving "fair and effective representation for all citizens." Reynolds, 377 U.S. at 565-66, 84 S.Ct. 1362. "[I]t was for that reason that [Reynolds] insisted on substantial equality of populations among districts." Gaffney, 412 U.S. at 748, 93 S.Ct. 2321. Thus, equality of population among state legislative districts is the ideal. See Reynolds, 377 U.S. at 560-61, 84 S.Ct. 1362 ("[R]epresentative government in this country is one of equal representation for equal numbers of people...."). For state and local elections, "substantial" (not exact) equality is required. Gaffney, 412 U.S. at 748, 93 S.Ct. 2321. Recognizing that legislative districts with an "identical number of residents, or citizens, or voters" are "a practical impossibility," the Supreme Court has held that the Equal Protection Clause requires only "that a State make an honest and good faith effort to construct districts ... as nearly of equal population as is practicable." Reynolds, 377 U.S. at 577, 84 S.Ct. 1362.16

This "honest and good faith effort" standard recognizes that some deviations from population equality may be necessary to allow states to pursue other legitimate objectives, such as "maintain[ing] the integrity of various political subdivisions" and "provid[ing] for compact districts of contiguous territory." Id. at 578, 84 S.Ct. 1362. And so, "minor deviations from mathematical equality among state legislative districts are insufficient to make out a prima facie case of invidious discrimination." Gaffney, 412 U.S. at 745, 93 S.Ct. 2321. The Court in Brown v. Thomson reiterated that an "unrealistic overemphasis on raw population figures ... may submerge these other considerations ... that in day-to-day operation are important to an acceptable representation and apportionment arrangement." 462 U.S. at 842, 103 S.Ct. 2690 (quoting Gaffney, 412 U.S. at 749, 93 S.Ct. 2321).

In this regard, "as a general matter ... an apportionment plan with a maximum population deviation under 10% falls within this category of minor deviations." Voinovich v. Quilter, 507 U.S. 146, 161, 113 S.Ct. 1149, 122 L.Ed.2d 500 (1993) (quoting Brown, 462 U.S. at 842, 103 S.Ct. 2690). On the other hand, a "plan with larger disparities in population ... creates a prima facie case of discrimination and therefore must be justified by the State." Id. The burden thus shifts to the Commission to demonstrate legitimate considerations "incident to the effectuation of a rational state policy." Reynolds, 377 U.S. at 579, 84 S.Ct. 1362. The policy must be applied in a manner "free from any taint of arbitrariness or discrimination." Roman v. Sincock, 377 U.S. 695, 710, 84 S.Ct. 1449, 12 L.Ed.2d 620 (1964).

In Mahan v. Howell, the Supreme Court laid out a two-part test for evaluating the constitutionality of a reapportionment plan for which a state must justify the deviations. First, can the legislature's plan "reasonably be said to advance [a] rational state policy"? Mahan, 410 U.S. at 328, 93 S.Ct. 979. Second, if so, do "the population disparities among the districts that have resulted from the pursuit of this plan exceed constitutional limits"? Id. In addition to the size of the population disparities, courts should also consider the "consistency of application and the neutrality of effect of the nonpopulation criteria." Brown, 462 U.S. at 845-46, 103 S.Ct. 2690.

The Supreme Court has repeatedly struck down reapportionment plans for which states failed to justify the population disparities among the districts. For example, in Swann v. Adams, 385 U.S. 440, 445, 446, 87 S.Ct. 569, 17 L.Ed.2d 501 (1967), the Court disapproved a Florida plan with variations of 30 percent in one house and 40 percent in the other because the state made "no attempt to justify any particular deviations, even the larger ones." The Court explained that "[d]e minimis deviations are unavoidable, but variations of 30% among senate districts and 40% among house districts can hardly be deemed de minimis and none of our cases suggests that differences of this magnitude will be approved without a satisfactory explanation grounded on acceptable state policy." Id. at 444, 87 S.Ct. 569 (emphasis added). Similarly, Kilgarlin v. Hill held that "it is quite clear that unless satisfactorily justified by the court or by the evidence of record, population variances of the size and significance evident here [26.48%] are sufficient to invalidate an apportionment plan." 386 U.S. 120, 122, 87 S.Ct. 820, 17 L.Ed.2d 771 (1967) (emphasis added). Although Texas asserted a justification of respecting county boundaries where possible, the Court was "not convinced that the announced policy ... necessitated the range of deviations between legislative districts which is evident here." Id. at 123, 87 S.Ct. 820. Two rejected plans also respected county lines but "produced substantially smaller deviations." Id. at 124, 87 S.Ct. 820.

In contrast, in Mahan, the Court upheld a Virginia reapportionment plan with a maximum deviation of 16.4 percent where the state asserted a justification of maintaining the integrity of political subdivision lines. 410 U.S. at 319, 325, 93 S.Ct. 979. Unlike the Texas plan at issue in Kilgarlin, the Virginia plan "produce[d] the minimum deviation above and below the norm, keeping intact political boundaries." Id. at 326, 93 S.Ct. 979 (citation and internal quotation marks omitted). The court-imposed plan created by the district court (which struck down the Virginia plan as unconstitutional) made "readily apparent" that applying a test of absolute equality "may impair the normal functioning of state and local governments." Id. at 323, 93 S.Ct. 979. Under the district court's plan, Scott County was divided between two districts. Its "representation was thereby substantially reduced in the first district, and all but nonexistent in the second district. The opportunity of its voters to champion local legislation relating to Scott County [was] virtually nil." Id. at 323-24, 93 S.Ct. 979. Virginia Beach "saw its position deteriorate in a similar manner under the court-imposed plan," and the residents transferred to another district in which they amounted to only 8.6 percent of that district's population complained that they were "effectively disenfranchised." Id. at 324, 93 S.Ct. 979.

Similarly, in Brown, the Court held constitutional the additional deviations from population equality caused by granting a representative to Wyoming's least populous county, Niobrara County, where Wyoming offered as a justification its "longstanding and legitimate policy of preserving county boundaries" in drawing representative districts. 462 U.S. at 846-47, 103 S.Ct. 2690. Recognizing the "peculiar size and population of the State and the nature of its governmental structure," this policy had "particular force." Id. at 844, 103 S.Ct. 2690. The Court also found it "noteworthy" that the policy was applied nondiscriminatorily. Id. at 848, 103 S.Ct. 2690. Were the alternative plan, which called for combining Niobrara County with a neighboring county, to be implemented, Niobrara County would be deprived of its own representative "even though the remainder of the House of Representatives would be constituted so as to facilitate representation of the interests of each county." Id. And "considerable population variations" in the state would remain even under the alternative plan. Id. at 847, 103 S.Ct. 2690.

The Court has emphasized that there must be some limit to permissible population disparities. See id. at 845, 103 S.Ct. 2690 ("Even a neutral and consistently applied criterion such as use of counties as representative districts can frustrate Reynolds' mandate of fair and effective representation if the population disparities are excessively high."). But, beyond the prima facie 10 percent threshold, the Supreme Court has never "define[d] the precise point at which a state may no longer justify its plan — the point at which population inequalities undermine the `substantial equality' standard."17 Gorin v. Karpan, 775 F.Supp. 1430, 1438 (D.Wyo.1991) (three judge court). "Neither courts nor legislatures are furnished any specialized calipers that enable them to extract from the general language of the Equal Protection Clause of the Fourteenth Amendment the mathematical formula that establishes what range of percentage deviations is permissible, and what is not." Mahan, 410 U.S. at 329, 93 S.Ct. 979.

So far, Mahan's 16.4 percent maximum deviation is the high-water mark for a plan approved by the Supreme Court. The Court noted that "this percentage may well approach tolerable limits," but "we do not believe it exceeds them." 410 U.S. at 329, 93 S.Ct. 979. Given the Court's case-by-case approach, this figure is helpful but not determinative. Brown, too, provides limited guidance as to the maximum possible deviation. Although the Wyoming plan in Brown resulted in an 89 percent maximum deviation, the Court emphasized that it was not deciding the question of whether the 89 percent deviation as a whole was justified, but only whether the additional deviation caused by granting a representative to Niobrara County was justified. 462 U.S. at 846, 103 S.Ct. 2690. And in Morris, the Court cited to the limiting language in Brown in support of its statement that "no case of ours has indicated that a deviation of some 78% could ever be justified." 489 U.S. at 702, 109 S.Ct. 1433.

The only clear rule that emerges from the Supreme Court's cases as to the permissible population deviation is that "each case must be evaluated on its own facts, and a particular population deviation from the ideal may be permissible in some cases but not in others." Chapman, 420 U.S. at 22, 95 S.Ct. 751. "[T]he fact that a 10% or 15% variation from the norm is approved in one State has little bearing on the validity of a similar variation in another State." Swann, 385 U.S. at 445, 87 S.Ct. 569. A state's particular circumstances factor into what deviation may be permissible. See id. (citing Reynolds, 377 U.S. at 578, 84 S.Ct. 1362). It appears, too, that the greater the deviation, the stronger the justification required. See Morris, 489 U.S. at 702, 109 S.Ct. 1433 ("At the very least, the local government seeking to support [a 78 percent deviation] between electoral districts would bear a very difficult burden....").

2. Constitutionality of Apportionment in 2012 Reapportionment Plan

We have carefully reviewed the Commission's reasons for the challenged population deviations, and the sizable (and uncontested) evidentiary record supporting the Commission's choices. After considering the entire record before us (with no contradicting evidence), we conclude that the Commission's justifications embody rational, legitimate, and substantial State policies. The 2012 Reapportionment Plan reasonably advances those policies in a neutral and nondiscriminatory manner. See Mahan, 410 U.S. at 326, 93 S.Ct. 979. We recognize that the maximum deviations here are significant. We do not suggest that any other state could justify deviations of this magnitude — in fact, it is possible that no other state could do so. Hawaii's geography, history, culture, and political structure set it apart. Given Hawaii's unique circumstances, the deviations here are justified. The Commission has met its burden to demonstrate the 2012 Reapportionment Plan's constitutionality.

a. Deviations Stemming from Maintaining the Integrity of Basic Island Units

The Commission offered numerous explanations for the deviations in the 2012 Reapportionment Plan, but the policy that drove the bulk of the deviations was maintaining the integrity of the basic island units, which also make up Hawaii's four counties. Unlike the counties of any other state, Hawaii's basic island units are all separated by 30 to 70 miles of open ocean. Doc. No. 65-13, Defs.' Ex. K at 26, Standing Comm. Rpt. at 261. Creating districts of equal population would require canoe districts spanning the ocean and comprised of different basic island units.

i. Background of Decision to Maintain Basic Island Unit Integrity

As summarized earlier, from its inception in April 2011, the Commission recognized that one of the overriding concerns and goals of the 2011 Reapportionment was to comply with the Hawaii Constitution's criterion that "no district shall extend beyond the boundaries of any basic island unit" as provided in Article IV, § 6. See, e.g., Doc. No. 65-24, Defs.' Ex. V, Masumoto Decl. ¶¶ 3-8. On June 9, 2011 — after considering past experience and Hawaii's Constitution — the Commission formally declared its intent to avoid "canoe districts" between basic island units. See id. ¶ 5. It made this decision after hearing much public testimony against canoe districts. See, e.g., Doc. No. 66-6, Defs.' Ex. CC at 3 ("The [Kauai Island Advisory] Council strongly recommends AGAINST the use of canoe districting.... Our island's past experience with the use of canoe districts has shown that representation in this manner does not work."); Doc. No. 66-7, Defs.' Ex. DD at 2 ("Members of the [Hawaii Island Advisory Council] voted NO to canoe district[s].").

The adherence to districts contained within basic island units comported with the elimination of canoe districts as a major revision in the prior reapportionment in 2001. See Doc. No. 65-15, Defs.' Ex. M at 11, 2001 Final Report and Reapportionment Plan at 25 (setting forth justification for elimination of canoe districts in 2001). The 2012 Reapportionment Plan summarizes:

Doc. No. 65-22, Defs.' Ex. T at 32, 2012 Reapportionment Plan at 21.

With the State constitutional standard of protecting the integrity of basic island units as a starting point, and factoring in the historical realities of island autonomy, the challenged disparities in district population sizes are driven by a single and unalterable fact: the permanent resident population of Kauai is 66,805. Given a mathematically ideal population for a State Senate district of 50,061 (with the non-permanent resident extraction), Kauai necessarily has an additional 16,744 residents in its single Senate district. Alternatively, Kauai could have two State Senate seats, but this would cause the opposite imbalance: two districts of about 33,400 residents, each deviating by 16,661 below the mathematical ideal. Without being part of a canoe district, Kauai is either underrepresented or overrepresented in the State Senate.

As it stands, Kauai's single Senate district is responsible for 33.44 percent of the challenged deviation and is alone enough to shift the burden to the Commission to demonstrate valid reasons for the deviation. See Doc. No. 65-22, Defs.' Ex. T at 31, 2012 Reapportionment Plan at 20. The 33 percent figure cannot be changed without using canoe districts.18 (The other 11 percent of the challenged 44 percent deviation comes from a Senate district seat on the Big Island with a deviation of 10.78 percent.) Id. As drawn, all four Big Island Senate districts have populations from approximately 6 to 11 percent below the Statewide mathematical ideal — disparities that likely result from the additional Senate seat created by extracting the large block of non-permanent residents from Oahu. Id.

Similarly, much of the challenged 21.57 percent disparity in State House districts results from Kauai's House districts 14, 15, and 16, which are all underpopulated as compared to the mathematically ideal State House district size. See id. at 30, 2012 Reapportionment Plan at 19. In particular, State House district 15 contains a population of permanent residents that is 2,705 below the mathematically ideal size of 24,540 — or 11.02 percent below that benchmark. Id. These underpopulated districts result from the Commission's decision to allot an additional House seat to Kauai given the overpopulated Senate district — that is, to help address concerns about equal representation that arise from a State Senate seat that is 33 percent above the ideal. The Commission explained:

Id. at 32, 2012 Reapportionment Plan at 21. The Commission, when measuring disparities among all legislators (the 76 House and Senate seats combined) by basic island unit, indicates a statewide deviation from an ideal of only 5.62 percent. Id. at 34, 2012 Reapportionment Plan at 23.

In short, the choice is straightforward: Either keep Kauai as a single Senate district (with correspondingly large deviations) or require canoe districts (to balance populations more equally).

This is not to say canoe districts are impossible. Prior to their elimination in the 2001 reapportionment, such districts were used in the 1982 and 1991 reapportionments, primarily to equalize populations among districts. In 1982, the Travis three judge district court ordered the use of an interim plan that utilized canoe districts. See Doc. Nos. 65-4, 65-5, Defs.' Exs. C-1, C-2, April 27, 1982 Final Report and Recommendations of Special Masters. Travis's court-ordered interim plan, however, had no choice but to use canoe districts. The court plan did not have the flexibility to exceed even de minimis population deviations that legislatively-initiated plans ordinarily have. See, e.g., Connor v. Finch, 431 U.S. 407, 415, 97 S.Ct. 1828, 52 L.Ed.2d 465 (1977). Connor reiterated that a "court-ordered reapportionment plan of a state legislature ... must ordinarily achieve the goal of population equality with little more than de minimis variation.'" Id. at 417, 97 S.Ct. 1828 (quoting Chapman, 420 U.S. at 26-27, 95 S.Ct. 751). "Court-ordered districts are held to higher standards of population equality than legislative ones." Abrams v. Johnson, 521 U.S. 74, 98, 117 S.Ct. 1925, 138 L.Ed.2d 285 (1997). This is because "a state legislature is the institution that is by far the best situated to identify and then reconcile traditional state policies within the constitutionally mandated framework of substantial population equality," whereas federal courts "possess no distinctive mandate to compromise sometimes conflicting state apportionment policies in the people's name." Connor, 431 U.S. at 414-15, 97 S.Ct. 1828.

Even the Travis court, which adopted recommendations of special masters, recognized that canoe districts were contrary to Hawaii's history, tradition, and its Constitution.19 The special masters recommended,

Doc. No. 65-5, Defs.' Ex. C-2 at 13 (April 27, 1982 Final Report and Recommendation of Special Masters at 41).

The 1991 Reapportionment Plan continued to use canoe districts to balance populations, mindful of equal protection concerns and Article IV, § 6, of the Hawaii Constitution. The 1991 Reapportionment Commission reported that it was "constrained to breach the boundaries of certain basic island units and create `canoe' districts as necessary in order to comport with the demands of the Equal Protection Clause." 1991 State of Hawaii Reapportionment Comm'n, Final Report and Reapportionment Plan, at 18 (available at Hawaii Legislative Reference Bureau).

The evidence presented leads to but one conclusion — Hawaii's two-decade canoe-district experience was perceived as a failure by constituents and representatives alike. See, e.g., Doc. No. 65-24, Defs.' Ex. V, Masumoto Decl. ¶¶ 10-12; Doc. No. 65-16, Defs.' Ex. N, Rosenbrock Decl. ¶ 12; Doc. No. 65-23, Defs.' Ex. U (articles documenting canoe district experiences). That is, the uncontested evidentiary record establishes that using canoe districts to balance populations creates, at the very least, a perception of unfair and ineffective representation for Hawaii's citizens.

Harold Masumoto, a Commission member both in 2001 and 2011, listed the primary reasons that he heard for people disliking canoe districts:

Doc. No. 65-24, Defs. Ex. V, Masumoto Decl. ¶ 11.

The experiences of legislators who represented canoe districts bear out these concerns. Malama Solomon (a current State Senator from the Big Island and a Senator from a canoe district from 1983 to 1990) stated that it was "impossible to properly represent all of [her] constituencies with a `canoe' district." Doc. No. 66-3, Defs.' Ex. Y, Solomon Decl. ¶ 11. For one thing, because "capital improvement project (CIP) funds are allocated in caucus on a per legislator basis, rather than on a per island basis," constituents on one or both islands of a canoe district "were disappointed because only half of two roads could be fixed, or only one school or other community facility in the district could be repaired." Id. ¶ 5. In addition, Solomon, who lived on Hawaii, was unable to spend as much time on Maui as on her home island. Id. ¶ 7. When in Maui, she did not have staff or a fixed place to meet constituents. Id. Solomon suspects that it would have been difficult for a representative to be elected from Maui because "twice as many people live in the Hawaii portion of the district as in the Maui portion." Id. ¶ 8. Constituents approached Solomon to "express their disappointment that [she] was unable to fully represent them." Id. ¶ 11.

Former representative Hermina Morita, who represented a Kauai/Maui canoe district from 1996 to 2002, tells a similar story. In her district, Kauai had a voter ratio of two to one over Maui, and no one from Maui ran against her or the other candidates from Kauai between 1996 and 2002. Doc. No. 70-1, Defs.' Ex. Z, Morita Decl. ¶ 5. Like Solomon, Morita spent little time on Maui, and she also "came to the conclusion that [she] was not representing [her] constituents from Maui effectively." Id. ¶¶ 6-8. She faced difficulties related to securing grants-in-aid for two counties that "may have taken conflicting positions on the same issues, or prioritized and focused on different aspects of the same issue." Id. ¶ 9. According to Morita, the constituents were disappointed with a representative "who was unable to devote all of his or her time and energy to advancing and protecting their interests and needs because they were from two different island communities with different wants, interests and resources." Id. ¶ 12. Other representatives had similar experiences. See, e.g., Doc. No. 39-13, Apo Decl.

Although the people of Kauai are arguably those most affected by "underrepresentation" in the Senate, they strongly disfavor canoe districts, at least as measured by the record before the court.20 For instance, the current Kauai State Senator, Ronald Kouchi, who was a Kauai County Councilmember when Kauai was represented in a canoe district, noted that the 2012 Reapportionment Plan's statements that there was "`overwhelming public sentiment'" against the use of canoe districts during the 2011 reapportionment proceedings, with "`universal dissatisfaction'" with canoe districts "echoe[d][his] personal experience." Doc. No. 72-3, Kouchi Decl. ¶¶ 7-8. Kouchi "cannot recall receiving any input in support of canoe districts." Id. ¶ 8. He explained that there "are numerous issues facing residents of Kauai County that are unique to Kauai County" and that there is an "impression that the Kauai residents' needs would not be adequately addressed and/or protected by a canoe district legislator as it is very difficult for one legislator to physically be on three different islands (i.e., Kauai, Maui, and Oahu.)." Id. ¶ 9.

ii. Analysis of Basic Island Unit Integrity Justification

Under Mahan, we must determine whether the 2012 Reapportionment Plan "may reasonably be said to advance [a] rational state policy." 410 U.S. at 328, 93 S.Ct. 979. The first question, then, is whether maintaining the integrity of the basic island units is a rational state policy. We conclude that it is. Hawaii's goal of maintaining the integrity of its basic island units is precisely the kind of "legitimate objective" articulated in Reynolds. See 377 U.S. at 578, 84 S.Ct. 1362 (noting that "maintain[ing] the integrity of various political subdivisions" and "provid[ing] for compact districts of contiguous territory" are legitimate aims); cf. Tennant, 133 S.Ct. at 8 (reiterating that avoiding splitting of political subdivisions is a "valid, neutral state districting polic[y]"). Reynolds explained that "insuring some voice to political subdivisions, as political subdivisions," is important because in many states, "much of the legislature's activity involves the enactment of so-called local legislation, directed only to the concerns of particular political subdivisions." 377 U.S. at 580-81, 84 S.Ct. 1362. As in Brown, the policy of keeping the counties intact "has particular force, given the peculiar size and population of the State and the nature of its governmental structure." Brown, 462 U.S. at 844, 103 S.Ct. 2690.

Because Hawaii's government is highly centralized and the State legislature controls many matters typically left to local governments, ensuring that political subdivisions have a voice in the State legislature is particularly important in Hawaii. Like Virginia Beach and Scott County under the court-imposed plan in Mahan, the constituents in the less-populated portion of canoe districts felt effectively disenfranchised. See 410 U.S. at 323-24, 93 S.Ct. 979. Former canoe district representatives felt unable to meet the needs of multiple islands, and it appears that the home islands of these representatives benefitted at the expense of the other islands.

While ensuring a voice to political subdivisions may be a legitimate policy for any state, Hawaii's unique geography, history, and culture give additional weight to the Commission's decision to maintain the integrity of the basic island units. Reynolds stated that political subdivisions "have been traditionally regarded as subordinate governmental instrumentalities created by the State to assist in the carrying out of state governmental functions." 377 U.S. at 575, 84 S.Ct. 1362. But Hawaii's basic island units were not created for the convenience of the State; rather, they predate statehood. Culturally and historically, the island units are stand-alone geographic and physical units and were distinct Kingdoms at the time of Western Contact in 1778, having been ruled by different chiefs and possessing distinct language variations and traditions. Doc. No. 66-14, Defs.' Ex. KK, McGregor Decl. ¶¶ 5-10. "[I]t is reasonable to posit that each basic island unit has an innate sense of individuality and separateness that is traceable to antiquity because each was organized independent of each other, in response to each island's geography, resources, and how their communities were governed." Id. ¶ 14.

Hawaii's Constitution was amended in 1968 specifically to preserve this historical and culturally-based basic island unit autonomy: "The local issues in any island unit are unique and typically bear very little similarity to those in the next island unit. Our conclusion [in the 1968 Constitutional Convention] was that for an island resident in Hawaii to have meaningful representation in the State Legislature, the representative must be from that resident's island unit." Doc. No. 66-12, Defs.' Ex. II, Schulze Decl. ¶ 14 (statement of Richard Schulze, Delegate to the 1968 Hawaii Constitutional Convention, and Chair of the Committee on Legislative Apportioning and Districting). Basic island units have been and continue today to be "separate societies or communities, with aspects and identities unique to themselves and distinct from each other." Doc. No. 66-14, Defs.' Ex. KK, McGregor Decl. ¶ 5.

Of course, other states have districts that struggle to meet the different needs of distinct communities — for example, districts that combine both rural and urban populations. But in no other state is each county separated from the others by 30 to 70 miles of ocean. Creating multi-county districts presents very different challenges for Hawaii than it would for any other state. It creates logistical, as well as political, problems. Former canoe district representatives were unable to spend as much time on their non-home islands as on their home islands. In addition, Ronald Kouchi, the current Senator from Kauai, notes that, were Kauai to become part of a canoe district, "the rising cost of air travel and lodging ... would put an additional burden on limited state resources" because he would have "to work in Honolulu and travel to two islands that comprised [his] district." Doc. No. 72-3, Kouchi Decl. ¶ 14.

Like the Court in Mahan, we "are not prepared to say that the decision of the people of the [State] to grant the [legislature] the power to enact local legislation dealing with the political subdivisions is irrational." 410 U.S. at 325-26, 93 S.Ct. 979. The question, then, is "whether it can reasonably be said that the state policy urged by [Hawaii] to justify the divergences in the legislative reapportionment plan of the House is, indeed, furthered by the plan adopted by the legislature." Id. at 326, 93 S.Ct. 979. By maintaining basic island unit integrity, the 2012 Reapportionment Plan clearly furthers the goal of giving a voice to the political subdivisions.

Given Hawaii's geographical constraints, the deviations do not appear to be significantly "greater than necessary to preserve" the basic island unit. Brown, 462 U.S. at 844, 103 S.Ct. 2690; cf. Karcher, 462 U.S. at 741, 103 S.Ct. 2653 (considering lack of "availability of alternatives" as a consideration). As explained earlier, the bulk of the deviations among both the House and Senate districts stems from Kauai, which is underrepresented in the Senate and overrepresented in the House. And 33.44 percent of the 44.22 percent Senate deviation is entirely unavoidable without resorting to canoe districts. The 2012 Commission believed that "the district boundary lines contained in the 2012 Supplemental Report were the best lines that could be drawn, given the dual goals of providing appropriate representation to the residents of the State of Hawaii and minimizing the population variance between the various State legislative districts." See Doc. No. 72 at 45, Defs.' Opp'n at 38.

Finally, the Commission emphasizes that it balanced the underrepresented Kauai Senate district by allocating an additional House seat to Kauai, thus slightly over-representing Kauai in the House. Although not a constitutional solution standing alone, such an apportionment scheme can factor into our determination of whether an honest and "good faith effort to establish districts substantially equal in population has been made." Lucas, 377 U.S. at 735 n. 27, 84 S.Ct. 1459. Lucas reasoned that "a court must necessarily consider a State's legislative apportionment scheme as a whole." Id.

Id.; see also Reynolds, 377 U.S. at 577, 84 S.Ct. 1362 ("Simply because the controlling criterion for apportioning representation is required to be the same in both houses does not mean that there will be no differences in the composition and complexion of the two bodies.... [A]pportionment in one house could be arranged so as to balance off minor inequities in the representation of certain areas in the other.").

Thus, the 33.44 percent deviation in Kauai's Senate district is tempered by the Commission's allotment of an additional House seat. This creates a deviation of 5.62 percent among basic island units when considering total legislators (House and Senate). See Doc. No. 65-22, Defs.' Ex. T at 34, 2012 Reapportionment Plan at 23. Hawaii has used this scheme since 1968 as at least one means of equalizing representation among the State's residents. See Gill, 316 F.Supp. at 1298 (approving of such a plan, reasoning that "Kauai's senatorial voters, at first glance, seem more seriously shortchanged, but... [a]ny such `loss' however was deliberately and meaningfully compensated for by providing 3 representatives for those same Kauai voters"); see also Blair v. Ariyoshi, 55 Haw. 85, 515 P.2d 1253, 1255-56 (1973) (approving of the assignment of an additional House seat to Kauai to account for underrepresentation in the Senate).21 Standing alone, this equalization effort would be insufficient to justify the deviation. However, this strategy to equalize total representation, in combination with the other factors analyzed above, demonstrates that the Commission made an "honest and good faith effort to construct districts ... as nearly of equal population as is practicable." Reynolds, 377 U.S. at 577, 84 S.Ct. 1362.

In sum, Hawaii's choice of basic island unit autonomy (i.e., no canoe districts) is grounded in the Hawaii Constitution, Hawaii's unique geography, cultural history (running back to before Hawaii was a Kingdom), governmental organization (with centralized statewide services in many areas of traditionally municipal-level control), and a two-decade failed experiment with canoe districts that has proven antithetical to the basic aim set forth in Reynolds of achieving "fair and effective representation for all [Hawaii's] citizens." 377 U.S. at 565, 84 S.Ct. 1362. This choice is not only rational — it is substantial and has considerable force. And where an apportionment plan is justified by a "longstanding and legitimate policy of preserving county boundaries," Brown, 462 U.S. at 847, 103 S.Ct. 2690, and absent any taint of discrimination, "substantial deference is to be accorded the political decisions" of the people of Hawaii. Id.

b. Other Disparities

The policy of keeping basic island units together does not explain all of the deviations in district size. Because the total deviations exceed 10 percent, the "entire plan is thus suspect and all deviations substantially adding to the maximum deviation must be justified with expressed reasons." Travis, 552 F.Supp. at 561.

In this regard, Kostick challenges the maximum population deviations within Oahu's districts of 8.89 percent (for Oahu's House seats) and 9.53 percent (for Oahu's Senate seats). See Doc. No. 65-22, Defs.' Ex. T at 26-28, 2012 Reapportionment Plan at 16-17.22 He argues that the deviations within Oahu should be "minimal," and contends the 2012 Reapportionment Plan "provides no other reasons for these intraisland deviations." Doc. No. 67 at 67, Pls.' Mot. at 56 (citation, internal quotation marks, and brackets omitted); see also Doc. No. 74 at 61, Pls.' Opp'n at 53.

In Travis, the court struck down a plan with similar ranges of intra-island deviations. 552 F.Supp. at 561. But there, Hawaii provided "no other reason for these deviations." Id. There was no evidence showing that Hawaii could not have drawn districts of equal populations on Oahu. Id. The State wrongly contended that, because the intra-island deviations were under 10 percent, they were de minimis and needed no justification. Id. As the court in Travis explained, "there is no support for the state's proposition that this standard can be used to compare and justify deviations between or within the geographical or political subdivisions of a state." Id.

In contrast, here the Commission has amply justified the deviations within Oahu. These deviations — like those on the other islands — are due in large part to "drawing district boundaries that adhere to permanent and easily recognized geographical features," "avoiding the submergence of areas in larger districts with different socio-economic interests," and "trying to maintain existing district boundaries" to "avoid disruption and confusion." Doc. No. 72 at 57, Defs.' Opp'n at 50; see also Doc. No. 65-22, Defs.' Ex. T, 2012 Reapportionment Plan at 11 (describing effort with regard to Oahu, complicated by the shift in population).

With regard to Oahu specifically, the Commission explained that the census block sizes posed a major challenge. Because Hawaii relies on data from the United States Census Bureau in apportioning districts, the Commission could not split census blocks when drawing the district lines. Doc. No. 72-2, Rosenbrock Second Supp. Decl. ¶¶ 13. Many census blocks on Oahu contain more than 2,000 residents — approximately 10 percent of the number of permanent residents in the average House district — and moving a single census block could cause the variances to exceed 10 percent. Id. In addition, drawing districts in Oahu was complicated by the significant population shifts that have occurred over the past ten years. Doc. No. 66-4, Defs.' Ex. AA, Nonaka Decl. ¶ 7 (describing how growth in west and central Oahu "required that two State House seats and one State Senate seat shift from urban Honolulu to west Oahu").

Although Kostick does not specifically mention deviations outside of Oahu that were not based on canoe districts, the Commission justified each county, house by house, explaining the choices it made in drawing district lines. See Doc. No. 72 at 39-61, Defs.' Opp'n at 32-54. We conclude that these justifications embody rational state policies, and that "the state polic[ies] urged by [Hawaii] to justify the divergences" are, "indeed, furthered by the plan adopted by the legislature." Mahan, 410 U.S. at 326, 93 S.Ct. 979; see also Brown, 462 U.S. at 843, 103 S.Ct. 2690; cf. Bush v. Vera, 517 U.S. 952, 977, 116 S.Ct. 1941, 135 L.Ed.2d 248 (1996) (plurality opinion) (noting traditional districting principles such as maintaining communities of interest and traditional boundaries).

Kostick's challenge to the other deviations therefore fails.

c. Size of Maximum Deviations

The final step in the Mahan test is whether "the population disparities among the districts that have resulted from the pursuit of this plan exceed constitutional limits." 410 U.S. at 328, 93 S.Ct. 979. Here we find ourselves in uncharted waters, but we look to the Supreme Court's guidance that "each case must be evaluated on its own facts." Chapman, 420 U.S. at 22, 95 S.Ct. 751. We recognize that deviations of this magnitude have yet to be countenanced by the Court. But we also recognize that the Court has yet to deal with this issue vis-à-vis Hawaii. We conclude that, given Hawaii's unique history, culture, and geography, the deviations of 44.22 percent in the Senate and 21.57 percent in the House do not exceed constitutional limits. We emphasize that our holding is specific to the facts before us. We do not hold that Hawaii's documented rationales — strong as they are — could justify any deviation, no matter how large. Nor do we suggest that Hawaii's state constitutional mandate trumps the Equal Protection Clause.

This court has intervened before in Hawaii's legislative reapportionment, to little benefit and much dissatisfaction. Perhaps such intervention was warranted in 1982 on the record before the court in Travis. But on the extensive record before us, which evidences Hawaii's thoughtful and deliberative attempt to adequately represent its citizens in the face of unique challenges, we come to a different conclusion. Crediting the strength of the Commission's rationales and the uncontradicted evidentiary support in the record, the 2012 Reapportionment Plan's deviations pass constitutional scrutiny. The Commission created a reapportionment plan that was implemented in a manner consistent with principles of equal representation. The 2012 Reapportionment Plan complies with Reynolds's ultimate aim — to achieve and assure "fair and effective representation for all citizens." 377 U.S. at 565-66, 84 S.Ct. 1362.

Thus, Defendants are entitled to judgment as a matter of law on Count II.

V. CONCLUSION

For the foregoing reasons, Plaintiffs' Motion for Summary Judgment, Doc. No. 67, is DENIED; and Defendants' Motion for Summary Judgment, Doc. No. 64, is GRANTED. Judgment shall issue in favor of Defendants on Counts One through Four of the First Amended Complaint. There being no remaining claims, the Clerk of Court shall issue final judgment and close the case file.

IT IS SO ORDERED.

2011 Reapportionment Commission Final Report and Reapportionment Plan 2012 Supplement

Table 9 — House Statewide Targets and Deviations

                                                 DEVIATION      DEVIATION %
                   STATEWIDE                     FROM STATE      FR STATE
                     TARGET        DISTRICT        TARGET         TARGET
HOUSE DISTRICT     POPULATION     POPULATION     POPULATION     POPULATION

House 1                24,540         26,553           2013           8.20%

House 2                24,540         25,652           1112           4.53%

House 3                24,540         25,935           1395           5.68%

House 4                24,540         26,990           2450           9.98%

House 5                24,540         27,129           2589          10.55%

House 6                24,540         25,239            699           2.85%

House 7                24,540         26,098           1558           6.35%

House 8                24,540         26,857           2317           9.44%

House 9                24,540         26,976           2436           9.93%

House 10               24,540         24,541              1           0.00%

House 11               24,540         24,705            165           0.67%

House 12               24,540         25,509            969           3.95%

House 13               24,540         25,956           1416           5.77%

House 14               24,540         22,718          -1822          -7.42%

House 15               24,540         21,835          -2705         -11.02%

House 16               24,540         22,252          -2288          -9.32%

House 17               24,540         23,468          -1072          -4.37%

House 18               24,540         23,382          -1158          -4.72%

House 19               24,540         23,221          -1319          -5.37%

House 20               24,540         23,798           -742          -3.02%

House 21               24,540         23,451          -1089          -4.44%

House 22               24,540         23,395          -1145          -4.67%

House 23               24,540         23,259          -1281          -5.22%

House 24               24,540         23,524          -1016          -4.14%

House 25               24,540         23,134          -1406          -5.73%

House 26               24,540         23,209          -1331          -5.42%

House 27               24,540         23,129          -1411          -5.75%

House 28               24,540         23,277          -1263          -5.15%

House 29               24,540         23,178          -1362          -5.55%

House 30               24,540         23,625           -915          -3.73%

House 31               24,540         23,507          -1033          -4.21%

House 32               24,540         23,261          -1279          -5.21%

House 33               24,540         23,495          -1045          -4.26%

House 34               24,540         25,101            561           2.29%

House 35               24,540         24,076           -464          -1.89%

House 36               24,540         25,209            669           2.73%

House 37               24,540         25,128            588           2.40%

House 38               24,540         25,190            650           2.65%

House 39               24,540         25,272            732           2.98%

House 40               24,540         25,239            699           2.85%

House 41               24,540         25,217            677           2.76%

House 42               24,540         25,280            740           3.02%

House 43               24,540         25,076            536           2.18%

House 44               24,540         25,219            679           2.77%

House 45               24,540         24,133           -407          -1.66%

House 46               24,540         25,037            497           2.03%

House 47               24,540         25,175            635           2.59%

House 48               24,540         25,238            698           2.84%

House 49               24,540         25,206            666           2.71%

House 50               24,540         24,498            -42          -0.17%

House 51               24,540         23,982           -558          -2.27%

 Total                             1,251,534

Statewide Deviation House-All                                        21.57%

Table 10 — Senate Statewide Targets and Deviations

             STATEWIDE                     DEVIATION      DEVIATION %
              TARGET        DISTRICT      FROM TARGET        FROM
DISTRICT        POP        POPULATION         POP           TARGET

Senate 1        50,061         44,666          -5,395         -10.78%

Senate 2        50,061         46,808          -3,253          -6.50%

Senate 3        50,061         47,218          -2,843          -5.68%

Senate 4        50,061         44,904          -5,157         -10.30%

Senate 5        50,061         53,833           3,772           7.53%

Senate 6        50,061         49,246            -815          -1.63%

Senate 7        50,061         51,465           1,404           2.80%

Senate 8        50,061         66,805          16,744          33.44%

Senate 9        50,061         51,322           1,261           2.52%

Senate 10       50,061         51,745           1,684           3.36%

Senate 11       50,061         51,900           1,839           3.67%

Senate 12       50,061         52,195           2,134           4.26%

Senate 13       50,061         51,206           1,145           2.29%

Senate 14       50,061         48,386          -1,675          -3.35%

Senate 15       50,061         52,090           2,029           4.05%

Senate 16       50,061         48,778          -1,283          -2.56%

Senate 17       50,061         47,729          -2,332          -4.66%

Senate 18       50,061         51,689           1,628           3.25%

Senate 19       50,061         47,450          -2,611          -5.22%

Senate 20       50,061         47,556          -2,505          -5.00%

Senate 21       50,061         48,311          -1,750          -3.50%

Senate 22       50,061         47,729          -2,332          -4.66%

Senate 23       50,061         47,993          -2,068          -4.13%

Senate 24       50,061         51,053             992           1.98%

Senate 25       50,061         49,457            -604          -1.21%

 Total                      1,251,534

Statewide Deviation Senate-All                                 44.23%

APPENDIX A

APPENDIX B