NEWBY, Justice.

Following the 2010 Decennial Census, the General Assembly of North Carolina enacted redistricting plans for the North Carolina Senate and House of Representatives, and for the North Carolina districts for the United States House of Representatives. Plaintiffs challenge the legality of these plans, arguing that they violate the Constitutions of the United States and of North Carolina, controlling federal statutes, and applicable decisions of the Supreme Court of the United States (the Supreme Court) and the Supreme Court of North Carolina. The three-judge panel1 reviewing the plans unanimously concluded that the General Assembly applied traditional and permissible redistricting principles to achieve partisan advantage and that no constitutional violations resulted. On plaintiffs' direct appeal, this Court affirmed the three-judge panel's ruling. Dickson v. Rucho, 367 N.C. 542, 766 S.E.2d 238 (2014). Thereafter, the Supreme Court vacated this Court's opinion and remanded the case to this Court for further consideration in light of its recent decision in Alabama Legislative Black Caucus v. Alabama, ___ U.S. ___, 135 S.Ct. 1257, 191 L.Ed.2d 314 (2015) (Alabama). Dickson v. Rucho, ___ U.S. ___, 135 S.Ct. 1843, 191 L.Ed.2d 719 (2015) (mem.).

In compliance with the Supreme Court's mandate, we have reconsidered this case in light of Alabama. Specifically, Alabama requires a district-by-district analysis in which the federal equal population requirement is simply a "background" rule that does not influence the predominant motive analysis. Alabama, ___ U.S. at ___, 135 S.Ct. at 1271, 191 L.Ed.2d at 332-33. After rebriefing and a careful review of the record in this case, we observe that the three-judge panel conducted the required detailed district-by-district analysis without giving improper weight to population equalization. See id. at ___, 135 S.Ct. at 1271, 191 L.Ed.2d at 332-33. The panel detailed its extensive findings and conclusions in a one hundred seventy-one page Judgment and Memorandum of Decision. Our careful review of that document leads us to conclude that, as to the twenty-six districts drawn to comply with the federal Voting Rights Act of 1965 ("Voting Rights Act" or "VRA"), the three-judge panel erred when it applied strict scrutiny prematurely; however, because these districts survive this most demanding level of review, plaintiffs were not prejudiced by the three-judge panel's error. As to the remaining challenged districts, we affirm the ruling of the three-judge panel that the predominant factors in their creation were the traditional and permissible redistricting principles encompassed within the mandatory framework as established by precedents of the Supreme Court and this Court.2

I. Procedural Background

The Constitution of North Carolina requires decennial redistricting of the North Carolina Senate and North Carolina House of Representatives, subject to several specific requirements. The General Assembly is directed to revise the districts and apportion Representatives and Senators among those districts ("House Districts" and "Senate Districts" or, collectively, "State House and Senate Districts"). N.C. Const. art. II, §§ 3, 5. Similarly, consistent with the requirements of the Constitution of the United States, the General Assembly establishes North Carolina's districts for the United States House of Representatives (Congressional Districts) after every decennial census. U.S. Const. art. I, §§ 2, 4; 2 U.S.C. §§ 2a, 2c (2012).

Redistricting in North Carolina has been challenged in this Court on multiple occasions.3 As a result, redistricting in this State does not proceed upon preferences or guidelines determined by the General Assembly. Instead, the legislature's priorities in drawing new district lines must be implemented within the mandatory framework recognized by this Court as required by federal law, federal and state constitutional mandates, and prior decisions of this Court. Pender County v. Bartlett, 361 N.C. 491, 493, 649 S.E.2d 364, 366 (2007) (Pender County), aff'd sub nom. Bartlett v. Strickland, 556 U.S. 1, 129 S.Ct. 1231, 173 L.Ed.2d 173 (2009) (plurality) (Strickland); Stephenson v. Bartlett, 355 N.C. 354, 562 S.E.2d 377 (2002) (Stephenson I).

The North Carolina Constitution "enumerates several limitations on the General Assembly's redistricting authority." Pender County, 361 N.C. at 493, 649 S.E.2d at 366. In particular, Sections 3 and 5 of Article II of the North Carolina Constitution, which address State House and Senate Districts, both include an equal population requirement and a Whole County Provision (collectively referred to as the "Whole County Provision"). Specifically, those sections of the constitution provide:

N.C. Const. art. II, §§ 3, 5.

While the federal one-person, one-vote standard addresses every district statewide, our state law instructs that the state constitution's equal population requirement must be read in the context of the geographic boundaries of counties, the state-recognized political subdivisions. In other words, the Whole County Provision, as recognized by this Court, requires that each State House and Senate District be confined to a single county or minimum grouping of contiguous counties. Stephenson I, 355 N.C. at 383-84, 562 S.E.2d at 397. In effect, North Carolina's Whole County Provision, of which equal population is a component, establishes a framework to address the neutral redistricting requirement that "political subdivisions" be respected.4 Shaw v. Reno, 509 U.S. 630, 646-47, 113 S.Ct. 2816, 2826-27, 125 L.Ed.2d 511, 528-29 (1993) (Shaw I); Stephenson I, 355 N.C. at 364, 371, 562 S.E.2d at 385, 389 (recognizing "the importance of counties as political subdivisions of the State of North Carolina" and "observ[ing] that the State Constitution's limitations upon redistricting and apportionment uphold what the United States Supreme Court has termed `traditional districting principles' ... such as `compactness, contiguity, and respect for political subdivisions'" (citation omitted) (quoting Shaw I, 509 U.S. at 647, 113 S.Ct. at 2827, 125 L.Ed.2d at 528)). Our state constitution's Whole County Provision establishes requirements not just for the number of voters, but for their identity as well. Thus, the approach to redistricting used here, required by the state constitution's Whole County Provision, is fundamentally different from the federal one-person, one-vote requirement addressed in Alabama, which spoke only to the number of voters. See Alabama, ___ U.S. at ___, 135 S.Ct. at 1271, 191 L.Ed.2d at 332 (explaining that equal population goals play a role in determining the number of persons placed in a district, but do not necessarily control "which persons were placed in appropriately apportioned districts"); see also Stephenson I, 355 N.C. at 371, 562 S.E.2d at 389 (distinguishing the "traditional districting principles" found in the North Carolina Constitution, including the Whole County Provision, from the federal "one-person, one-vote" standard).

In addition, the General Assembly followed the mandatory framework of our decision in Stephenson I, which harmonized the requirements of federal and state law and set out nine criteria that the General Assembly must follow in drawing new district lines. 355 N.C. at 383-84, 562 S.E.2d at 396-97. These nine criteria may be summarized as follows: First, "legislative districts required by the VRA shall be formed" before non-VRA districts are created. Id. at 383, 562 S.E.2d at 396-97. Second, "[i]n forming new legislative districts, any deviation from the ideal population for a legislative district shall be at or within plus or minus five percent" to ensure "compliance with federal `one-person, one-vote' requirements." Id. at 383, 562 S.E.2d at 397. Third, "[i]n counties having a... population sufficient to support the formation of one non-VRA legislative district,... the physical boundaries" of the non-VRA district shall "not cross or traverse the exterior geographic line of [the] county." Id. at 383, 562 S.E.2d at 397. Fourth, "[w]hen two or more non-VRA legislative districts may be created within a single county, ... single-member non-VRA districts shall be formed within [the] county," "shall be compact, and shall not traverse the [county's] exterior geographic boundary." Id. at 383, 562 S.E.2d at 397. Fifth, for non-VRA counties that "cannot support at least one legislative district," or "counties having a non-VRA population pool, which, if divided into [legislative] districts, would not comply with" one-person, one-vote requirements, the General Assembly should "combin[e] or group[] the minimum number of whole, contiguous counties necessary to comply with the at or within plus or minus five percent `one-person, one-vote' standard." Id. at 383, 562 S.E.2d at 397. Moreover, "[w]ithin any such contiguous multi-county grouping, compact districts shall be formed, consistent with the [one-person, one-vote] standard, whose boundary lines do not cross or traverse the `exterior' line of the multi-county grouping." Id. at 383-84, 562 S.E.2d at 397. "[T]he resulting interior county lines created by any such groupings may be crossed or traversed in the creation of districts within said multi-county grouping but only to the extent necessary to comply with the at or within plus or minus five percent `one-person, one-vote' standard." Id. at 384, 562 S.E.2d at 397. Sixth, "only the smallest number of counties necessary to comply with the at or within plus or minus five percent `one-person, one-vote' standard shall be combined." Id. at 384, 562 S.E.2d at 397. Seventh, "communities of interest should be considered in the formation of compact and contiguous [legislative] districts." Id. at 384, 562 S.E.2d at 397. Eighth, "multi-member districts shall not be" created "unless it is established that such districts are necessary to advance a compelling governmental interest." Id. at 384, 562 S.E.2d at 397. Ninth, "any new redistricting plans ... shall depart from strict compliance with" these criteria "only to the extent necessary to comply with federal law." Id. at 384, 562 S.E.2d at 397. Within this mandatory framework, the General Assembly may consider permissible and traditional redistricting principles such as compactness, contiguity, and respect for political subdivisions and communities of interest. See Miller v. Johnson, 515 U.S. 900, 916, 115 S.Ct. 2475, 2488, 132 L.Ed.2d 762, 779-80 (1995).

Following the 2010 census, leaders of the North Carolina House of Representatives and the North Carolina Senate independently appointed redistricting committees. Each committee was responsible for recommending a plan applicable to its own chamber, while the two committees jointly were charged with preparing a redistricting plan for North Carolina's Congressional districts for the United States House of Representatives.

Guided by the United States Supreme Court's redistricting principles, in addition to the state constitution and the mandatory framework of this Court's prior decisions, the redistricting committees sought information and suggestions from numerous sources, including the North Carolina Legislative Black Caucus and the North Carolina delegation to the United States Congress. In addition, these committees solicited input from various constituencies; invited public comment and conducted public hearings in multiple counties, including twenty-four of the forty counties then covered by section 5 of the Voting Rights Act;5 heard both lay and expert testimony regarding such matters as racially polarized voting; solicited and received advice from the University of North Carolina School of Government; commissioned reports from independent experts to fill gaps in the evidence; and considered written submissions, including proposed redistricting maps submitted by the Southern Coalition for Social Justice.

The General Assembly convened on 25 July 2011 to deliberate the redistricting plans drawn by the House and Senate committees. That same day, the leaders of the Democratic Party and the Legislative Black Caucus submitted other alternative maps. On 27 July, the General Assembly ratified the 2011 North Carolina Senate redistricting plan and the 2011 plan for the federal House of Representatives districts. On 28 July, the General Assembly ratified the 2011 North Carolina House of Representatives redistricting plan. On 2 September 2011, the General Assembly submitted the three plans to the United States Department of Justice (USDOJ) for preclearance under section 5 of the Voting Rights Act. That same day, the General Assembly filed a suit also seeking preclearance in the United States District Court for the District of Columbia. The General Assembly dismissed this suit upon receiving preclearance from the USDOJ on 1 November 2011.6

On 3 November 2011, Margaret Dickson and forty-five other registered voters filed a complaint seeking to have the three redistricting plans declared invalid on both constitutional and statutory grounds. These plaintiffs filed an amended complaint on 12 December 2011. On 4 November 2011, the North Carolina State Conference of Branches of the NAACP, joined by three organizations and forty-six individuals, filed a complaint seeking similar relief. These plaintiffs filed an amended complaint on 9 December 2011. Following the filing of the original complaints, then-Chief Justice Sarah Parker of the Supreme Court of North Carolina appointed a panel of three superior court judges to hear these actions, pursuant to N.C.G.S. § 1-267.1. On 19 December 2011, the three-judge panel consolidated both cases for all purposes.

Plaintiffs argue that the redistricting violated their federal and state equal protection rights as well as the state constitution's Whole County Provision. Underlying all of plaintiffs' complaints is the implicit argument that the Supreme Court incorrectly decided Strickland and that the General Assembly impermissibly utilized a fifty percent plus one black voting age population in the challenged VRA districts.

On 6 February 2012, the three-judge panel allowed in part and denied in part defendants' motion to dismiss. Plaintiffs filed a motion for partial summary judgment on 5 October 2012, and defendants filed a motion for summary judgment on 10 December 2012. The three-judge panel heard arguments on these motions on 25 and 26 February 2013.

While a ruling on the motions for summary judgment was pending, the three-judge panel issued an order determining that genuine issues of material fact existed as to two issues that could not be resolved by summary judgment.7 The panel conducted a trial on these two issues on 4 and 5 June 2013. On 8 July 2013, the three-judge panel issued its unanimous Judgment and Memorandum of Decision denying plaintiffs' motion for partial summary judgment and entering judgment for defendants on all claims asserted by plaintiffs, including those related to the issues addressed at trial.

In rendering its ruling, the three-judge panel conducted a district-by-district review of the constitutionality of each challenged district. After considering thousands of pages of evidence and testimony from numerous witnesses, the panel produced a detailed, one hundred seventy-one page document setting out its findings of fact and conclusions of law. In upholding the General Assembly's redistricting plans, the panel recognized that:

The three-judge panel first considered plaintiffs' claims that the General Assembly's redistricting plans violated the equal protection guarantees of the United States and North Carolina Constitutions. The panel's first step was to determine which level of scrutiny to apply to each challenged district. It recognized that while generally "all racial classifications [imposed by a government]... must be analyzed by a reviewing court under strict scrutiny," see Johnson v. California, 543 U.S. 499, 505, 125 S.Ct. 1141, 1146, 160 L.Ed.2d 949, 958 (2005) (alterations in original) (quoting Adarand Constructors, Inc. v. Pena, 515 U.S. 200, 227, 115 S.Ct. 2097, 2113, 132 L.Ed.2d 158, 182 (1995)), mere "consciousness of race" is insufficient to trigger strict scrutiny in redistricting cases, Bush v. Vera, 517 U.S. 952, 958, 116 S.Ct. 1941, 1951, 135 L.Ed.2d 248, 257 (1996) (plurality). Instead, the three-judge panel explained that strict scrutiny is only appropriate when plaintiffs establish that "all other legislative districting principles were subordinated to race and that race was the predominant factor motivating the legislature's redistricting decision." See Cromartie v. Hunt, 133 F.Supp.2d 407, 425 (E.D.N.C.2000) (citing, inter alia, Miller, 515 U.S. at 916, 115 S.Ct. at 2488, 132 L.Ed.2d at 779-80), rev'd sub nom. Easley v. Cromartie, 532 U.S. 234, 121 S.Ct. 1452, 149 L.Ed.2d 430 (2001) (Cromartie II).

The three-judge panel determined that twenty-six8 of the thirty districts challenged by plaintiffs were created by the General Assembly to be VRA districts. The General Assembly intended to draw each of these districts so as to include at least fifty percent Total Black Voting Age Population (TBVAP). The three-judge panel concluded that, "even though legislative intent may have been remedial and the districts may have been drawn to conform with federal and state law," these VRA districts were "predominantly determined by a racial objective." Therefore, the three-judge panel determined that strict scrutiny was the appropriate level of review for these twenty-six VRA districts. The panel acknowledged, however, "that a persuasive argument can be made that compliance with the VRA is but one of several competing redistricting criteria balanced by the General Assembly and that a lesser standard of review might be appropriate." Nonetheless, the three-judge panel employed strict scrutiny because that standard provides a "convenient and systematic roadmap for judicial review," and because, if the plans survive strict scrutiny, in which the evidence is considered in a light most favorable to the non-prevailing party, then the plans would necessarily survive a lesser level of scrutiny, such as rational basis review.

The three-judge panel made specific findings of fact for each of the twenty-six VRA districts. Based on its findings, the three-judge panel concluded that the twenty-six VRA districts survive strict scrutiny because they were narrowly tailored to achieve a compelling governmental interest in "avoiding future liability under § 2 of the VRA and ensuring future preclearance of the redistricting plans under § 5 of the VRA." See Shaw v. Hunt, 517 U.S. 899, 915-16, 116 S.Ct. 1894, 1905-06, 135 L.Ed.2d 207, 225-26 (1996) (Shaw II).

The three-judge panel concluded that avoiding section 2 liability was a compelling governmental interest because, based upon the panel's exhaustive review of the entire record, "the General Assembly had a strong basis in evidence to conclude that each of the Gingles preconditions was present in substantial portions of North Carolina," see Thornburg v. Gingles, 478 U.S. 30, 106 S.Ct. 2752, 92 L.Ed.2d 25 (1986), "and that, based upon the totality of the circumstances, VRA districts were required to remedy against vote dilution."9 In considering whether compliance with section 5 provided a compelling governmental interest, the three-judge panel explained that "the newly-enacted plan may not undo or defeat rights afforded by the most recent legally enforceable redistricting plan in force or effect in the covered jurisdiction (the `benchmark' plan)." See Riley v. Kennedy, 553 U.S. 406, 128 S.Ct. 1970, 170 L.Ed.2d 837 (2008) (cited by the panel in support of this statement). Because "the General Assembly had a strong basis in evidence to conclude that [its plans] must be precleared" under section 5, the three-judge panel determined that preclearance under section 5 provided "a compelling governmental interest."

The three-judge panel next concluded that each of the twenty-six VRA districts was narrowly tailored to avoid section 2 liability and to ensure section 5 preclearance. See Shaw I, 509 U.S. at 645, 113 S.Ct. at 2831, 125 L.Ed.2d at 534 (quoted by the panel and providing that in responding to the compelling interests in complying with sections 2 and 5, the General Assembly is not granted "carte blanche to engage in racial gerrymandering"). The panel recognized "that the `narrow tailoring' requirement of strict scrutiny allows the States a limited degree of leeway in furthering such interests." Vera, 517 U.S. at 977, 116 S.Ct. at 1960, 135 L.Ed.2d at 268.

First, the unanimous panel found that the enacted plans do not contain a greater number of VRA districts than are reasonably necessary to comply with the VRA because "the General Assembly had a strong basis in evidence for concluding that `rough proportionality' was reasonably necessary to protect the State from anticipated liability under § 2 of the VRA and ensuring preclearance under § 5 of the VRA." See League of United Latin Am. Citizens v. Perry, 548 U.S. 399, 438, 126 S.Ct. 2594, 2621, 165 L.Ed.2d 609, 643-44 (2006) (LULAC); Shaw II, 517 U.S. at 915-16, 116 S.Ct. at 1905-06, 135 L.Ed.2d at 225-26; Johnson v. De Grandy, 512 U.S. 997, 1000, 114 S.Ct. 2647, 2651, 129 L.Ed.2d 775, 784 (1994) ("[N]o violation of § 2 can be found ... where, in spite of continuing discrimination and racial bloc voting, minority voters form effective voting majorities in a number of districts roughly proportional to the minority voters' respective shares in the voting-age population.").

Second, the panel found that the General Assembly did not unnecessarily "pack" VRA districts with black voters when it endeavored to create all VRA districts with at least fifty percent TBVAP in order to avoid liability under section 2. See Strickland, 556 U.S. at 13, 129 S.Ct. at 1242, 173 L.Ed.2d at 183 (plurality) (opinion of Kennedy, J.) (stating that compliance with section 2 allows creating majority-minority districts that contain "a numerical, working majority of the voting age population" of a specific minority group and that it does not mandate creating or preserving crossover districts). The three-judge panel explained that under Strickland, "the State must be afforded the leeway to avail itself of the `bright line rule' and create majority-minority districts, rather than cross-over districts, in those areas where there is a sufficiently large and geographically compact minority population and racial polarization exist[s]." As a result, the three-judge panel found that, "notwithstanding the racial classification inherent in the creation of > 50% TBVAP VRA districts, the Enacted Plans substantially address the threat of anticipated § 2 liability and challenges to preclearance under § 5 of the VRA."

Third, the three-judge panel heard evidence on the following issue:

Based on this evidence the panel made numerous detailed findings of fact, including one hundred eighty-eight findings on this issue set out in Appendix A of its judgment. The three-judge panel conducted an individualized analysis of each of the VRA districts, setting out how racially polarized voting was found in the locales. For example, the court noted that a study conducted by Thomas Brunell, Ph.D., found "statistically significant racially polarized voting" in fifty out of fifty-one counties examined.10 The three-judge panel then determined that "the General Assembly had a strong basis in evidence for concluding that [] each of the VRA districts in the Enacted Plans were placed in a location that was reasonably necessary to protect the State from anticipated liability under" sections 2 and 5 of the VRA.

Finally, the three-judge panel found that the VRA districts are sufficiently compact and regular in shape to meet the requirement that they be narrowly tailored. Quoting Justice Kennedy, the panel stated: "`Districts not drawn for impermissible reasons or according to impermissible criteria may take any shape, even a bizarre one,' provided that the bizarre shapes are not `attributable to race-based districting unjustified by a compelling interest.'" Vera, 517 U.S. at 999, 116 S.Ct. at 1972, 135 L.Ed.2d at 284 (Kennedy, J., concurring). The three-judge panel further found that plaintiffs' retained expert testified that the shape of a district is irrelevant, as are traditional notions of communities of interest.

Ultimately, the three-judge panel concluded that plaintiffs failed to produce alternative plans that (1) contain VRA districts in rough proportion to the black population in North Carolina, (2) comply with the General Assembly's decision, as supported by Strickland, to populate each VRA district with more than fifty percent TBVAP, or (3) comply with the state constitution's Whole County Provision.

Accordingly, the three-judge panel concluded that

Because it found that the twenty-six VRA districts were narrowly tailored to achieve a compelling governmental interest, the three-judge panel concluded that these districts survive strict scrutiny. Alternatively, the panel noted "that under a lesser standard of review, such as a rational relationship test, the creation of the VRA districts as drawn was supported by a number of rational bases."

Next, the three-judge panel considered the constitutionality of the four remaining challenged districts, which were non-VRA districts.11 The panel explained that if these non-VRA districts were "unexplainable on grounds other than race" and "the legislature neglected all traditional redistricting criteria such as compactness, contiguity, respect for political subdivisions and incumbency protection," then strict scrutiny would apply. See id. at 976, 116 S.Ct. at 1959-60, 135 L.Ed.2d at 268 (plurality). Otherwise, if the legislature was not motivated predominantly by race in drawing these four districts, the three-judge panel must apply rational basis review. The panel stated that whether race was the General Assembly's predominant motive is a factual question. See Hunt v. Cromartie, 526 U.S. 541, 549, 119 S.Ct. 1545, 1550, 143 L.Ed.2d 731, 740 (1999) (Cromartie I) (citing Shaw II, 517 U.S. at 905, 116 S.Ct. at 1900, 135 L.Ed.2d at 218-19). Thus, the three-judge panel held a trial to determine the following issue: "For six specific districts (Senate Districts 31 and 32, House Districts 51 and 54 and Congressional Districts 4 and 12 — none of which is identified as a VRA district), what was the predominant factor in the drawing of those districts?" Although Senate District 31 and House District 51 were not challenged by plaintiffs as racial gerrymanders, the three-judge panel heard evidence on these two districts as well because they are neighboring districts that are part of the same contiguous-county grouping required under the North Carolina Constitution's Whole County Provision. Therefore, these two unchallenged districts are necessarily intertwined with those that were challenged, making consideration of the motivation for the creation of each unchallenged district relevant to a determination of the motivation for the creation of the counterparts, the challenged districts.

As it did for the twenty-six VRA districts, based upon the evidence received, the three-judge panel made specific findings of fact as to each non-VRA district, including Senate District 31 and House District 51. After conducting a detailed, district-by-district analysis, the panel made numerous specific findings of fact on whether race was the General Assembly's predominant motive in drawing these districts. The three-judge panel found, in addition to complying with federal and state law and applying nonracial traditional redistricting criteria, the General Assembly desired to create districts "more competitive for Republican candidates." The panel noted the "undisputed" fact "that in North Carolina, racial identification correlates highly with political affiliation." But, the goal to create State House and Senate Districts more competitive for Republicans could only be realized while following the requirements of the state constitution's Whole County Provision. Thus, while the three-judge panel recognized the General Assembly's desire "to equalize population among the districts," for state redistricting purposes, this finding must be viewed in the context of the Whole County Provision requirement.

Based upon its findings, the three-judge panel concluded that rational basis review was the appropriate level of scrutiny for each of the non-VRA districts and that "the General Assembly has articulated a reasonably conceivable state of facts, other than a racial motivation, that provides a rational basis for creating the non-VRA districts." Moreover, the three-judge panel determined that plaintiffs failed to proffer, as required by Cromartie II, "any alternative redistricting plans that show that the General Assembly could have met its legitimate political objectives in alternative ways that are comparably consistent with traditional districting principles." See Cromartie II, 532 U.S. at 258, 121 S.Ct. at 1466, 149 L.Ed.2d at 453. Accordingly, the three-judge panel concluded that plaintiffs' racial gerrymandering claims failed.

The three-judge panel next addressed plaintiffs' claim that the Senate and House plans violated the Whole County Provision of the North Carolina Constitution. The panel concluded that the enacted plans conform to the Whole County Provision, as interpreted and applied by this Court in Stephenson I and Stephenson v. Bartlett, 357 N.C. 301, 582 S.E.2d 247 (2003) (Stephenson II), and that plaintiffs' alternative proposed plans failed to comport with the Whole County Provision.

Plaintiffs entered timely notice of appeal pursuant to N.C.G.S. § 120-2.5. On 19 December 2014, this Court affirmed the three-judge panel's decision, holding "that the General Assembly's enacted plans do not violate plaintiffs' constitutional rights." Dickson, 367 N.C. at 575, 766 S.E.2d at 260. Plaintiffs petitioned the Supreme Court for a writ of certiorari on 16 January 2015. On 25 March 2015, while plaintiffs' petition was pending, the Supreme Court issued its decision in Alabama, ___ U.S. ___, 135 S.Ct. 1257, 191 L.Ed.2d 314. The Supreme Court subsequently vacated our decision and remanded the case to this Court for reconsideration in light of Alabama. Dickson, ___ U.S. ___, 135 S.Ct. 1843, 191 L.Ed.2d 719. The parties filed briefs, and we heard oral arguments regarding the applicability of Alabama to this case.

Our review of the three-judge panel's unanimous decision is limited. Though "[o]ur standard of review of an appeal from summary judgment is de novo," In Re Will of Jones, 362 N.C. 569, 573, 669 S.E.2d 572, 576 (2008), unchallenged findings of fact are binding on appeal, see, e.g., Keeter v. Town of Lake Lure, 264 N.C. 252, 257, 141 S.E.2d 634, 638 (1965). Likewise, regarding issues tried by the panel, its findings of fact are binding on this Court if not challenged at trial or on appeal, see e.g., Koufman v. Koufman, 330 N.C. 93, 97, 408 S.E.2d 729, 731 (1991), or if supported by competent evidence found by the three-judge panel, e.g., In re Estate of Trogdon, 330 N.C. 143, 147-48, 409 S.E.2d 897, 900 (1991). Conclusions of law are reviewed de novo to determine if they are supported by the findings of fact. E.g., N.C. Farm Bureau Mut. Ins. Co. v. Cully's Motorcross Park, Inc., 366 N.C. 505, 512, 742 S.E.2d 781, 786 (2013). After a careful review of the Alabama decision and the record in this case, we conclude that the three-judge panel's Judgment and Memorandum of Decision complied with the standards articulated in Alabama as well as other pertinent federal and state laws.

II. Overview of Alabama and Its Impact Here

Like the case before us, the Alabama case involved a challenge to the state legislature's redistricting plans following the 2010 decennial census. Alabama, ___ U.S. at ___, 135 S.Ct. at 1262-63, 191 L.Ed.2d at 323-24. The Alabama Code requires the creation of a legislative committee known as the Permanent Legislative Committee on Reapportionment "to prepare for and develop a reapportionment plan for the state." Ala.Code 1975 § 29-2-50 to -51 (2015). Unlike North Carolina, where the General Assembly's priorities can only be implemented in accordance with the federal and state constitutional requirements as specified by this Court, the Alabama legislative committee adopted "guidelines for drawing the new district lines." Ala. Legislative Black Caucus v. Alabama, 989 F.Supp.2d 1227, 1245 (2013), vacated, ___ U.S. ___, 135 S.Ct. 1257, 191 L.Ed.2d 314. These guidelines provided that new districts should be drawn in a manner to achieve numerous traditional redistricting objectives, including compactness, not splitting counties or precincts, minimizing change, and protecting incumbents. Alabama, ___ U.S. at ___, 135 S.Ct. at 1263, 191 L.Ed.2d at 324; Legislative Black Caucus, 989 F.Supp.2d at 1245. The guidelines acknowledged, however, "that not all of the redistricting goals could be accomplished," Legislative Black Caucus, 989 F.Supp.2d at 1245, and the guidelines placed the greatest emphasis on two of these goals: (1) minimizing the extent to which any district deviated by more than one percent from the theoretical precisely equal population ideal; and (2) avoiding retrogression under section 5 of the VRA by maintaining roughly the same black population percentage in existing majority-minority districts, Alabama, ___ U.S. at ___, 135 S.Ct. at 1263, 191 L.Ed.2d at 324. To achieve population equalization while avoiding retrogression, the legislature adjusted existing, underpopulated, majority-minority districts by adding massive numbers of minority voters. Id. at ___, 135 S.Ct. at 1263, 191 L.Ed.2d at 324. For example, to maintain a 72.75% black population in Senate District 26, the legislature added 15,785 individuals, all but 36 of whom are black. Id. at ___, 135 S.Ct. at 1263, 191 L.Ed.2d at 324.

The plaintiffs in Alabama claimed that the legislature's new districts created "racial gerrymanders" that violated the Equal Protection Clause of the Fourteenth Amendment. Id. at ___, 135 S.Ct. at 1262, 191 L.Ed.2d at 323. A federal district court panel ruled in favor of the State. Id. at ___, 135 S.Ct. at 1262, 191 L.Ed.2d at 323. In vacating that decision and remanding the case for further proceedings, the Supreme Court identified four problems with the district court's ruling. Id. at ___, 135 S.Ct. at 1264, 191 L.Ed.2d at 325.

First, the Supreme Court in Alabama held that the district court erred by considering the state "as a whole," rather than conducting a "district-by-district" analysis of the racial gerrymandering claims. Id. at ___, 135 S.Ct. at 1265-68, 191 L.Ed.2d at 326-30. This ruling does not affect our case because North Carolina's three-judge panel conducted the required individualized, district-by-district analysis.

Second, the Supreme Court held that the district court erred in ruling that one of the plaintiffs lacked standing without giving that plaintiff an opportunity to prove that it had standing. Id. at ___, 135 S.Ct. at 1268-70, 191 L.Ed.2d at 330-31. This is a fact-specific issue that has no relevance to or effect on our case.

Third, as previously noted, the Alabama legislative committee placed great emphasis on ensuring that no district deviated by more than one percent from the theoretical equal population ideal. The Supreme Court held that the district court improperly concluded that "[r]ace was not the predominant motivating factor" in the creation of any of the challenged districts because the court "placed in the balance, among other nonracial factors, legislative efforts to create districts of approximately equal population." Id. at ___, 135 S.Ct. at 1270, 191 L.Ed.2d at 331. The Supreme Court observed that equal population goals might explain the number of persons placed in a district, but they do not address "which persons were placed in appropriately apportioned districts." Id. at ___, 135 S.Ct. at 1271, 191 L.Ed.2d at 332. As a result, an equal population goal is not a factor "to be weighed against the use of race to determine whether race `predominates'"; rather, equal population "is part of the redistricting background" that should be "taken as a given, when determining whether race, or other factors, predominate" in the creation of new districts. Id. at ___, 135 S.Ct. at 1270, 191 L.Ed.2d at 332. The Supreme Court concluded that "had the District Court treated equal population goals as background factors, it might have concluded that race was the predominant boundary-drawing consideration," id. at ___, 135 S.Ct. at 1272, 191 L.Ed.2d at 333, given that the record contained "strong, perhaps overwhelming, evidence that race did predominate as a factor," id. at ___, 135 S.Ct. at 1272, 191 L.Ed.2d at 333.

This portion of the Alabama decision supports our holding here. The legislative committee in Alabama adopted guidelines that included compliance with the federal one-person, one-vote standard. In contrast, North Carolina's constitutional equal population criteria are a component of and intertwined with the state constitution's Whole County Provision, as explained above. See N.C. Const. art. II, §§ 3(1), 5(1); Stephenson I, 355 N.C. at 383, 562 S.E.2d at 397 (providing, for example, that for non-VRA counties that "cannot support at least one legislative district," or counties "having a non-VRA population pool which, if divided into [legislative] districts, would not comply with" one-person, one-vote requirements, the General Assembly should "combin[e] or group[ ] the minimum number of whole, contiguous counties necessary to comply with the at or within plus or minus five percent `one-person, one-vote' standard," and that "[w]ithin any such contiguous multi-county grouping, compact districts shall be formed, consistent with the [one-person, one-vote] standard, whose boundary lines do not cross or traverse the `exterior' line of the multi-county grouping"). Unlike the situation in Alabama, the General Assembly here did not place special emphasis on compliance with federal one-person, one-vote standards; rather, equal population was a "background" criterion that entered into formulating the challenged congressional and state legislative districts in conjunction with meeting the Whole County Provision of the state constitution. Nevertheless, to ensure compliance with Alabama, we have carefully reviewed the record, the transcripts, and the three-judge panel's findings to ensure that federal population equalization is treated as a "background rule" in conducting the predominance test. Alabama, ___ U.S. at ___, 135 S.Ct. at 1271, 191 L.Ed.2d at 332-33. The three-judge panel's finding that race was a predominant factor in forming the VRA districts is unaffected. In its predominance analysis of the non-VRA districts, the three-judge panel properly weighed the relevant facts and made its findings, and the findings support its conclusion that race was not the predominant factor in drawing these districts.

Last, in Alabama the Supreme Court held that the district court misinterpreted the requirements of section 5 of the VRA in finding that the challenged districts were "narrowly tailored" to satisfy strict scrutiny.12 Id. at ___, 135 S.Ct. at 1272-74, 191 L.Ed.2d at 334-36. The Court instructed that section 5 "does not require a covered jurisdiction to maintain a particular numerical minority percentage" in order to avoid retrogression. Id. at ___, 135 S.Ct. at 1272, 191 L.Ed.2d at 334 (pointing to evidence in the record reflecting Alabama's belief that section 5 "forbids, not just substantial reductions, but any reduction in the percentage of black inhabitants of a majority-minority district" (quoting id. at ___, 135 S.Ct. at 1289, 191 L.Ed.2d at 337 (App. B))). Instead, "[section] 5 is satisfied if minority voters retain the ability to elect their preferred candidates." Id. at ___, 135 S.Ct. at 1273, 191 L.Ed.2d at 334. As an example, the Court explained that "Congress did not mandate that a 1% reduction in a 70% black population district would be necessarily retrogressive." Id. at ___, 135 S.Ct. at 1273, 191 L.Ed.2d at 335. The Court concluded that, in "rel[ying] heavily upon a mechanically numerical view as to what counts as forbidden retrogression," id. at ___, 135 S.Ct. at 1273, 191 L.Ed.2d at 335, the district court failed to ask the right question with respect to narrow tailoring: "To what extent must [the legislature] preserve existing minority percentages in order to maintain the minority's present ability to elect the candidate of its choice?" Id. at ___, 135 S.Ct. at 1274, 191 L.Ed.2d at 336.

The three-judge panel's conclusion here that the twenty-six challenged VRA districts survive strict scrutiny is consistent with the Supreme Court's clarification of the section 5 narrow tailoring analysis. Our conclusion that the VRA districts are constitutional is not dependent on a section 5 analysis. Each of the challenged VRA districts subject to strict scrutiny was created because the State had a compelling interest in compliance with section 2, and each was narrowly tailored to accomplish that goal; therefore, each of the VRA districts is constitutional on the basis of a section 2 analysis alone. Regardless, as explained below, the Supreme Court and this Court have stated that the legislature's requirement that each of the challenged districts consist of a TBVAP exceeding fifty percent of the total voting age population in that district is permissible. See Strickland, 556 U.S. at 23, 129 S.Ct. at 1248, 173 L.Ed.2d at 190. The TBVAP was not greater than necessary to avoid retrogression, while also avoiding liability under section 2, even considering the Supreme Court's warning against a "mechanical interpretation" of section 5. Therefore, the challenged VRA districts survive strict scrutiny under either a section 2 or section 5 analysis.

Significantly, the United States Supreme Court in Alabama did not modify its prior holding in Strickland, where it made clear that a state legislature may create majority-minority VRA districts with a fifty percent plus one TBVAP. Id. at 23, 129 S.Ct. at 1248, 173 L.Ed.2d at 190. In the case sub judice plaintiffs persistently argue that the General Assembly must create crossover or coalition districts and that the General Assembly violated section 2 by drawing districts with a fifty percent plus one TBVAP. Essentially, plaintiffs argue that the Supreme Court wrongly decided Strickland, in which Justice Kennedy stated for the plurality that "[section] 2 does not require the creation of influence districts." Id. at 13, 129 S.Ct. at 1242, 173 L.Ed.2d at 183 (citing LULAC, 548 U.S. at 445, 126 S.Ct. at 2625, 165 L.Ed.2d at 648 (opinion of Kennedy, J.)). In fact, none of the alternative plans proposed by plaintiffs or supported by them complied with Strickland. Accordingly, plaintiffs' arguments implicitly premised upon revisiting the Supreme Court's decision in Strickland are without merit.

III. Plaintiffs' Federal Claims

We now consider plaintiffs' claims brought under federal law. If a redistricting plan does not satisfy federal requirements, it fails even if it is consistent with the law of North Carolina. See U.S. Const. art. VI, § 2; N.C. Const. art. I, § 3. Nonetheless, as emphasized by Stephenson I, in making redistricting decisions, federal and state law must be read in harmony. 355 N.C. at 363, 562 S.E.2d at 384. Plaintiffs argued first to the three-judge panel, and now to us, that the redistricting plans violate the Equal Protection Clause of the Fourteenth Amendment to the Constitution of the United States because they impermissibly classify individuals based upon their race. In other words, plaintiffs contend that the redistricting plans constitute impermissible racial gerrymandering that has denied them equal protection under the law.

A. Levels of Scrutiny

A court considering allegations of racial gerrymandering first must determine the appropriate level of scrutiny. Strict scrutiny, the highest tier of review, applies "when the classification impermissibly interferes with the exercise of a fundamental right or operates to the peculiar disadvantage of a suspect class." White v. Pate, 308 N.C. 759, 766, 304 S.E.2d 199, 204 (1983) (citations omitted). "Race is unquestionably a `suspect class,'" Phelps v. Phelps, 337 N.C. 344, 353, 446 S.E.2d 17, 23 (1994), and if a court finds that race is the "predominant, overriding factor" behind the General Assembly's plans, the plans must satisfy strict scrutiny to survive, Miller, 515 U.S. at 920, 115 S.Ct. at 2490, 132 L.Ed.2d at 782. "Under strict scrutiny [review], a challenged governmental action is unconstitutional if the State cannot establish that it is narrowly tailored to advance a compelling governmental interest." Stephenson I, 355 N.C. at 377, 562 S.E.2d at 393 (citation omitted). If, on the other hand, the plans are not predominantly motivated by improper racial considerations, the court applies the rational basis test. See Nordlinger v. Hahn, 505 U.S. 1, 10, 112 S.Ct. 2326, 2331, 120 L.Ed.2d 1, 12 (1992) ("[U]nless a classification warrants some form of heightened review because it jeopardizes exercise of a fundamental right or categorizes on the basis of an inherently suspect characteristic, the Equal Protection Clause requires only that the classification" satisfy rational basis review.). Under rational basis review, "[t]he general rule is that legislation is presumed to be valid and will be sustained if the classification drawn by the statute is rationally related to a legitimate state interest." City of Cleburne v. Cleburne Living Ctr., Inc., 473 U.S. 432, 440, 105 S.Ct. 3249, 3254, 87 L.Ed.2d 313, 320 (1985) (citations omitted).

A party challenging a redistricting plan has the burden of establishing that race was the predominant motive behind the state legislature's action. Miller, 515 U.S. at 916, 115 S.Ct. at 2488, 132 L.Ed.2d at 779-80. In Miller the Supreme Court stated that

Id. at 916, 115 S.Ct. at 2488, 132 L.Ed.2d at 779-80 (quoting Shaw I, 509 U.S. at 647, 113 S.Ct. at 2827, 125 L.Ed.2d at 529). In Alabama the Court clarified that federal "equal population" requirements cannot serve as a "traditional race-neutral districting principle[]" in the predominant motive analysis. Alabama, ___ U.S. at ___, 135 S.Ct. at 1270-71, 191 L.Ed.2d at 332-33.

As a court considers which level of scrutiny is appropriate, it should be mindful of the Supreme Court's observation that "courts must `exercise extraordinary caution in adjudicating claims that a State has drawn district lines on the basis of race.'" Cromartie II, 532 U.S. at 242, 121 S.Ct. at 1458, 149 L.Ed.2d at 443 (quoting Miller, 515 U.S. at 916, 115 S.Ct. at 2488, 132 L.Ed.2d at 779). At least three factors lie behind this admonition. First, in light of the interplay detailed below between the Fourteenth Amendment, which virtually forbids consideration of race, and the VRA, which requires consideration of race, the Supreme Court has acknowledged that the existence of legislative consciousness of race while redistricting does not automatically render redistricting plans unconstitutional. Miller, 515 U.S. at 916, 115 S.Ct. at 2488, 132 L.Ed.2d at 779-80 ("Redistricting legislatures will, for example, almost always be aware of racial demographics; but it does not follow that race predominates in the redistricting process."); see also Shaw I, 509 U.S. at 646, 113 S.Ct. at 2826, 125 L.Ed.2d at 528 ("[T]he legislature always is aware of race when it draws district lines.... That sort of race consciousness does not lead inevitably to impermissible race discrimination."). Second, the Supreme Court has recognized the importance of the states' own traditional districting principles, holding that states can adhere to them without being subject to strict scrutiny so long as those principles are not subordinated to race. Vera, 517 U.S. at 978, 116 S.Ct. at 1961, 135 L.Ed.2d at 269. Finally, the Supreme Court has accepted that some degree of deference is due in light of the difficulties facing state legislatures when reconciling conflicting legal responsibilities. Miller, 515 U.S. at 915, 115 S.Ct. at 2488, 132 L.Ed.2d at 779 ("Electoral districting is a most difficult subject for legislatures, and so the States must have discretion to exercise the political judgment necessary to balance competing interests."); Vera, 517 U.S. at 1038, 116 S.Ct. at 1991, 135 L.Ed.2d at 308 (Stevens, Ginsburg & Breyer, JJ., dissenting) ("[I]n fashioning a reapportionment plan or in choosing among plans, a district court should not preempt the legislative task nor `intrude upon state policy any more than necessary.'" (quoting White v. Weiser, 412 U.S. 783, 795, 93 S.Ct. 2348, 2355, 37 L.Ed.2d 335, 346 (1973)); see also Page v. Va. State Bd. of Elections, 58 F.Supp.3d 533, 542 (E.D.Va.2014) (recognizing that redistricting is "possibly `the most difficult task a legislative body ever undertakes'") (quoting Smith v. Beasley, 946 F.Supp. 1174, 1207 (D.S.C.1996))), vacated and remanded, Cantor v. Personhuballah, ___ U.S. ___, 135 S.Ct. 1699, 191 L.Ed.2d 671 (2015) (mem.) (remanded for further consideration in light of Alabama, ___ U.S. ___, 135 S.Ct. 1257, 191 L.Ed.2d 314 (2015)).

A court's determination of the predominant motive underlying a redistricting plan is factual in nature. Cromartie I, 526 U.S. at 549, 119 S.Ct. at 1550, 143 L.Ed.2d at 740 (citations omitted). Factual findings are binding on appeal if not challenged at trial or on appeal, e.g., Koufman, 330 N.C. at 97, 408 S.E.2d at 731, or if supported by competent evidence found by the three-judge panel, e.g., In re Estate of Trogdon, 330 N.C. at 147-48, 409 S.E.2d at 900. Conclusions of law are reviewed de novo. E.g., Cully's Motorcross Park, 366 N.C. at 512, 742 S.E.2d at 786 (citation omitted). Here, before the three-judge panel, plaintiffs challenged thirty House, Senate, and Congressional Districts.13 The three-judge panel concluded that the twenty-six VRA districts were predominantly motivated by race and thus subject to strict scrutiny review. The three-judge panel concluded that the remaining four challenged non-VRA districts were not predominantly motivated by race and thus were subject to rational basis review. We consider each group in turn.

B. The VRA Districts

We turn first to the twenty-six state legislative VRA districts that the three-judge panel subjected to strict scrutiny. As to these districts, the panel reached two significant conclusions. First, the panel unanimously found that "it is undisputed that the General Assembly intended to create 26 of the challenged districts to be `Voting Rights Act districts'" that would include a TBVAP of at least fifty percent. This unchallenged finding of fact is binding on us. Koufman, 330 N.C. at 97, 408 S.E.2d at 731. The three-judge panel then reached a second conclusion that drawing VRA districts "necessarily requires the drafters of districts to classify residents by race," that "the shape, location and racial composition of each VRA district was predominantly determined by a racial objective," and that the process of creating such districts resulted in "a racial classification sufficient to trigger the application of strict scrutiny as a matter of law." While the three-judge panel recognized that a "persuasive argument" could be made that other non-racial factors could have predominated, it chose to apply strict scrutiny because (1) strict scrutiny provides a "convenient and systematic roadmap for judicial review," and (2) if the plans withstand strict scrutiny, when the evidence is considered in a light most favorable to plaintiffs, then they would necessarily withstand a lesser level of scrutiny, such as rational basis review. Although the three-judge panel's determination that race predominated is neither purely factual nor purely legal, we are mindful that federal precedent cited above instructs that the General Assembly's consideration of race to the degree necessary to comply with section 2 does not rise to the level of a "predominant motive" as a matter of course. Accordingly, before reviewing the three-judge panel's application of strict scrutiny, we believe it necessary to review its conclusion as to the General Assembly's predominant motive.

1. Predominant Motive

The challenges faced by the General Assembly while redistricting are easy to express but persistently difficult to resolve. The Fourteenth Amendment, by guaranteeing equal protection for all citizens regardless of race, essentially prohibits consideration of race during redistricting. U.S. Const. amend. XIV, § 1. Yet the Voting Rights Act, passed "to help effectuate the Fifteenth Amendment's guarantee that no citizen's right to vote shall `be denied or abridged ... on account of race, color, or previous condition of servitude,'" Voinovich v. Quilter, 507 U.S. 146, 152, 113 S.Ct. 1149, 1154-55, 122 L.Ed.2d 500, 510 (1993) (alteration in original) (citations omitted), specifically requires consideration of race. For instance, section 2 "prohibits the imposition of any electoral practice or procedure that `results in a denial or abridgement of the right of any citizen .. . to vote on account of race or color.'" Id. at 152, 113 S.Ct. at 1155, 122 L.Ed.2d at 510 (alteration in original) (quoting 42 U.S.C. § 1973(a) (effective 1 September 2014, recodified as 52 U.S.C.S. § 10301(a) (LexisNexis 2014))). At the same time, the General Assembly must ensure that, to the greatest extent allowed under federal law, the state legislature's redistricting plans comply with the Whole County Provision of the North Carolina Constitution, including the "background rule" of plus or minus five percent as required by our constitution. Stephenson I, 355 N.C. at 382-84, 562 S.E.2d at 395-97. Moreover, the Supreme Court has acknowledged other legitimate considerations, such as compactness, contiguity, and respect for political subdivisions, see Miller, 515 U.S. at 916, 115 S.Ct. at 2488, 132 L.Ed.2d at 780; Shaw I, 509 U.S. at 646, 113 S.Ct. at 2826, 125 L.Ed.2d at 528; Reynolds v. Sims, 377 U.S. 533, 578, 84 S.Ct. 1362, 1390, 12 L.Ed.2d 506, 537 (1964); political advantage, see Cromartie I, 526 U.S. at 551, 119 S.Ct. at 1551, 143 L.Ed.2d at 741; and accommodation of incumbents, see Karcher v. Daggett, 462 U.S. 725, 740, 103 S.Ct. 2653, 2663, 77 L.Ed.2d 133, 147 (1983). Thus, "[t]he courts, in assessing the sufficiency of a challenge to a districting plan, must be sensitive to the complex interplay of forces that enter a legislature's redistricting calculus." Miller, 515 U.S. at 915-16, 115 S.Ct. at 2488, 132 L.Ed.2d at 779.

Despite this cat's cradle of factors facing the General Assembly, the three-judge panel found that no factual inquiry was required regarding the General Assembly's predominant motivation in forming the twenty-six VRA districts beyond the General Assembly's concession that the districts were drafted to be VRA-compliant. In light of the many other considerations potentially in play, we do not believe that this concession established that race ipso facto was the predominant motive driving the General Assembly. The three-judge panel assumed that because federal law requires racial considerations, race then predominates. Yet, it appears from the three-judge panel's findings that the General Assembly was concerned with compliance with federal law more than addressing race per se. In other words, race was only a factor insomuch as required by federal law. Because of the three-judge panel's truncated findings of fact on this issue, we do not know which additional factors may have influenced the creation and shape of these twenty-six districts and the extent of any such influence. As a result, we do not know whether race fairly can be described as the predominant factor in the formation of these districts and whether, in turn, strict scrutiny was the appropriate level of scrutiny. Moreover, in future cases such an assumption — that deliberate creation of VRA-compliant districts equates to race as the predominant motive in creating the districts — may well shortcut the fact-finding process at which trial courts excel, resulting in scanty records on appeal. Accordingly, we hold that the three-judge panel erred in concluding as a matter of law that, just because the twenty-six districts were created to be VRA-compliant, the General Assembly was motivated predominantly by race.

Nonetheless, this error is not fatal and does not invalidate the three-judge panel's decision. First, the panel itself concluded that, "[t]o the extent that the most exacting level of review, strict scrutiny, is not warranted... [,] under a lesser standard of review, such as a rational relationship test, the creation of the VRA districts as drawn was supported by a number of rational bases." Further, a similar scenario occurred in Cromartie I, in which the courts reviewed the General Assembly's creation of North Carolina's Twelfth Congressional District. 526 U.S. at 543, 119 S.Ct. at 1547, 143 L.Ed.2d at 736. The plaintiffs filed suit in federal court, arguing that the district was the result of an unconstitutional racial gerrymander. Id. at 544-45, 119 S.Ct. at 1548, 143 L.Ed.2d at 737. The three-judge panel of the United States District Court heard arguments pertaining to pending motions, but did not conduct an evidentiary hearing. Id. at 545, 119 S.Ct. at 1548, 143 L.Ed.2d at 737. The panel majority, finding that the General Assembly used race-driven criteria in drawing the district and that doing so violated the Equal Protection Clause of the Fourteenth Amendment, granted the plaintiffs' motion for summary judgment and entered an injunction. Id. at 545, 119 S.Ct. at 1548, 143 L.Ed.2d at 737. On appeal, the Supreme Court reversed, finding that the General Assembly's motivation in drawing district lines is a factual question that, when contested, should not be resolved by summary judgment. Id. at 549, 553, 119 S.Ct. at 1550, 1552, 143 L.Ed.2d at 739, 742.

The posture of the litigants here is distinguishable because plaintiffs, unlike their counterparts in Cromartie I, lost at summary judgment and are the appealing party; however, even if we were to follow Cromartie I's lead and reverse, plaintiffs could gain nothing on remand. The basis for our reversal would be that the three-judge panel erred in applying strict scrutiny before making adequate findings of fact. As the panel noted in its Judgment, if defendants' plans survived strict scrutiny, they would surely survive a less rigorous review. On the other hand, if the three-judge panel on remand found facts and determined once more that strict scrutiny is proper, the panel has already conducted its analysis under that standard. If these plans survive strict scrutiny, they survive rational basis review.

2. Compelling Governmental Interest

We begin this analysis by considering the factors that defendants contend constitute a "compelling governmental interest." See Stephenson I, 355 N.C. at 377, 562 S.E.2d at 393 (citation omitted). Defendants argue that the General Assembly drafted the twenty-six districts both to avoid liability under section 2 of the VRA and to obtain preclearance under section 5 of the VRA by avoiding retrogression, which has been defined as "a change in voting procedures which would place the members of a racial or language minority group in a less favorable position than they had occupied before the change with respect to the opportunity to vote effectively." Id. at 363-64, 562 S.E.2d at 385 (citations omitted). Defendants' brief acknowledges that three principles guided the General Assembly: (1) Compliance with the Whole County Provision of the Constitution of North Carolina, as set out in Stephenson I and Stephenson II; (2) Where possible, establishment of VRA districts having a TBVAP above fifty percent, in accord with Pender County; and (3) Exploration of "the possibility of establishing a sufficient number of VRA legislative districts to provide [black] voters with rough proportionality in the number of VRA districts in which they have a reasonable opportunity to elect their candidates of choice."

Although the Supreme Court has never held outright that compliance with section 2 or section 5 can be a compelling state interest, that Court has issued opinions that expressly assumed as much. To be specific, the Supreme Court in Shaw II assumed arguendo that compliance with section 2 could be a compelling state interest, 517 U.S. at 915, 116 S.Ct. at 1905, 135 L.Ed.2d at 225, and adopted a similar approach in Miller, where the issue was the State's desire to comply with section 5 of the Voting Rights Act, 515 U.S. at 921, 115 S.Ct. at 2490-91, 132 L.Ed.2d at 783. In addition, the Supreme Court has observed that "deference is due to [States'] reasonable fears of, and to their reasonable efforts to avoid, § 2 liability." Vera, 517 U.S. at 978, 116 S.Ct. at 1961, 135 L.Ed.2d at 269. The three-judge panel here, citing several federal cases addressing the issue, stated that "[i]n general, compliance with the Voting Rights Act can be a compelling governmental interest." Faced squarely with the issue, we agree with the three-judge panel. The Equal Protection Clause of the Fourteenth Amendment requires equal treatment regardless of race, while the Voting Rights Act requires consideration of race. Because a federal statute may not violate the Constitution of the United States, a State's efforts to comply with the Voting Rights Act create tension with the Fourteenth Amendment. An alleged violation of the Fourteenth Amendment based on race triggers strict scrutiny, mandating that the State demonstrate a compelling interest. Because the Supreme Court and the United States Congress have indicated without ambiguity that they expect states to comply with the Voting Rights Act, state laws passed for the purpose of complying with the Act must be capable of surviving strict scrutiny, indicating that such compliance is a compelling state interest.14 This analysis applies equally to a State's efforts to comply with sections 2 and 5 of the Voting Rights Act.

Moreover, the General Assembly's desire to comply with the Voting Rights Act is justifiable for other reasons. Holding elections is a core State function, fundamental in a democracy. Establishing voting districts is an essential component of holding elections. In doing so, a State is subject to federal mandates in addition to those found in the Voting Rights Act and the Fourteenth Amendment, such as the "one-person, one-vote" requirement. Stephenson I, 355 N.C. at 363-64, 383, 562 S.E.2d at 384-85, 397. A determination that the State does not have a compelling interest in complying with federal mandates would invite litigation by those claiming that the State could never satisfy the requirements of strict scrutiny, undermining the General Assembly's efforts to create stable districts between censuses and citizen expectations that existing election districts are valid. On a level no less practical, we also assume that North Carolina, and all states for that matter, would prefer to avoid the expense and delay resulting from litigation. Accordingly, we hold that compliance with sections 2 and 5 of the Voting Rights Act may be a compelling state interest.

We next consider whether compliance with either section 2 or section 5 constitutes a compelling state interest under the facts presented here. Those goals may reach the level of a compelling state interest if two conditions are satisfied. First, the General Assembly must have identified past or present discrimination with some specificity before it could turn to race-conscious relief. Shaw II, 517 U.S. at 909, 116 S.Ct. at 1902, 135 L.Ed.2d at 221 (citing City of Richmond v. J.A. Croson Co., 488 U.S. 469, 504, 109 S.Ct. 706, 727, 102 L.Ed.2d 854, 889 (1989)). Second, before acting, the General Assembly must also have "had `a strong basis in evidence'" on which to premise a conclusion that the race-based remedial action was necessary. Id. at 910, 116 S.Ct. at 1903, 135 L.Ed.2d at 222 (quoting Wygant v. Jackson Bd. of Educ., 476 U.S. 267, 277, 106 S.Ct. 1842, 1849, 90 L.Ed.2d 260, 271 (1986) (plurality)).

a. Compelling Interest Under Section 2 of the Voting Rights Act

Before we turn our attention to consideration of individual districts, we consider the application of section 2 of the VRA in the instant case. "The essence of a § 2 claim is that a certain electoral law, practice, or structure interacts with social and historical conditions to cause an inequality in the opportunities enjoyed by black and white voters to elect their preferred representatives." Gingles, 478 U.S. at 47, 106 S.Ct. at 2764, 92 L.Ed.2d at 44; see 52 U.S.C.S. §§ 10301-10702 (LexisNexis 2014). The question of voting discrimination vel non, including vote dilution, is determined by the totality of the circumstances. Gingles, 478 U.S. at 43-46, 106 S.Ct. at 2762-64, 92 L.Ed.2d at 42-44 (discussing section 2(b) of the VRA, now codified at 52 U.S.C.S. § 10301(b)). Under Gingles, however, a reviewing court does not reach the totality of circumstances test unless the challenging party is able to establish three preconditions. Id. at 50-51, 106 S.Ct. at 2766-67, 92 L.Ed.2d at 46-47. First, a "minority group must be able to demonstrate that it is sufficiently large and geographically compact to constitute a majority in a single-member district." Id. at 50, 106 S.Ct. at 2766, 92 L.Ed.2d at 46. Second, the minority group must "show that it is politically cohesive." Id. at 51, 106 S.Ct. at 2766, 92 L.Ed.2d at 47. Finally, the minority group must "be able to demonstrate that the majority votes sufficiently as a bloc to enable it ... usually to defeat the minority's preferred candidate." Id. at 51, 106 S.Ct. at 2766-67, 92 L.Ed.2d at 47. Although Gingles dealt with multi-member districts, the same preconditions must be met when a claim of vote dilution is made regarding a single-member district. Growe v. Emison, 507 U.S. 25, 40-41, 113 S.Ct. 1075, 1084, 122 L.Ed.2d 388, 403-04 (1993); see also De Grandy, 512 U.S. at 1006-07, 114 S.Ct. at 2654-55, 129 L.Ed.2d at 788.

Unlike cases such as Gingles, in which minority groups use section 2 as a sword to challenge districting legislation, here we are considering the General Assembly's use of section 2 as a shield. Defendants argue that, because the Gingles test considers race, the State has a compelling interest in preemptively factoring race into its redistricting process to ensure that its plans would survive a legal challenge brought under section 2. To establish that this state interest is legitimate, defendants must show a strong basis in evidence that the possibility of a section 2 violation existed at the time of the redistricting. See Shaw II, 517 U.S. at 910, 916, 116 S.Ct. at 1903, 1905-06, 135 L.Ed.2d at 222, 225-26. But because this inquiry addresses only the possibility of a section 2 violation, and because a totality of the circumstances inquiry is by its nature fact-specific, defendants' evidence need only address "the three Gingles preconditions" to establish a compelling governmental interest. See Vera, 517 U.S. at 978, 116 S.Ct. at 1961, 135 L.Ed.2d at 269 (citing Growe, 507 U.S. at 40, 113 S.Ct. at 1084, 122 L.Ed.2d at 403-04).

Thus, to establish a compelling interest in complying with section 2 when the redistricting plans were developed, the legislature at that time must have had a strong basis in evidence that the TBVAP in a geographically compact area was fifty percent plus one of the area's voting population. Such evidence would satisfy the first Gingles precondition. Pender County, 361 N.C. at 503, 649 S.E.2d at 372. In addition, a strong basis in evidence of racially polarized voting in that same geographical area would satisfy the second and third preconditions set out in Gingles. LULAC, 548 U.S. at 427, 126 S.Ct. at 2615, 165 L.Ed.2d at 637 (majority). Against this background, we consider the three-judge panel's application of these standards in discerning whether defendants here could legitimately claim a compelling interest in complying with section 2.

The three-judge panel's decision included two extensive appendices. In the body of the Judgment and Memorandum, the panel described the legislative record that existed when the plans were enacted, then referred to Appendix A, where this information was presented in detail. Appendix A, titled "Findings of Fact Relevant to the Issue of Racial Polarization in Specific Locations where Voting Rights Act Districts were Placed in the Enacted Plans," is incorporated by reference into the three-judge panel's decision.

Appendix A is broken into three parts. Part I, titled "General Findings of Fact," opens with a summary of the background of the case, then notes results of recent elections. For instance, the three-judge panel observed that all black incumbents elected to the North Carolina General Assembly or the United States Congress in 2010 were elected in districts that were either majority black or majority-minority coalition districts. In addition, no black candidate elected in 2010 was elected from a majority white crossover district, and two black incumbent state senators running in majority white districts were defeated in that election. No black candidate for the United States Congress was elected in a majority white district between 1992 and 2010, while from 2004 through 2010, no black candidate was elected to office in a statewide partisan election.

In this Part I of Appendix A, the court also considered an academic study of racially polarized voting conducted by Ray Block, Jr., Ph.D. This study, prepared for the Southern Coalition of Social Justice, is titled "Racially Polarized Voting in 2006, 2008, and 2010 in North Carolina State Legislative Contests." Dr. Block employed Justice Brennan's conclusion in Gingles that racially polarized voting occurs when there is a consistent relationship between the race of the voter and the way in which that person votes, and found that such a relationship existed in the areas examined. He added that he also found evidence that "majority-minority districts facilitate the election of [black] candidates." The court determined that Dr. Block's study provided "substantial evidence regarding the presence of racially polarized voting in almost all of the counties[15] in which the General Assembly enacted the 2011 VRA districts."

Nevertheless, the three-judge panel observed that the North Carolina General Assembly identified a few limitations in Dr. Block's study. For instance, the study did not pinpoint the percentage of white voters in majority black or majority-minority districts who voted for the candidate of choice of black voters. In addition, his study could analyze a legislative election only when the black candidate had opposition. As a result, the General Assembly commissioned Dr. Brunell to prepare a supplementary report. Dr. Brunell's study, titled "Report on Racially Polarized Voting in North Carolina," examined the forty North Carolina counties covered by section 5 of the Voting Rights Act, plus Columbus, Duplin, Durham, Forsyth, Jones, Mecklenburg, Richmond, Sampson, Tyrrell, Wake, and Warren Counties. Dr. Brunell found "statistically significant racially polarized voting" in fifty of these fifty-one counties.16

The three-judge panel made additional findings of fact in Part I of Appendix A that we believe would be pertinent to a Gingles totality of circumstances test and that, by extension, indicate a strong basis in evidence that the Gingles preconditions existed. At the beginning of the redistricting process, the General Assembly noted that North Carolina had been ordered to create majority black districts as a remedy for section 2 violations in Bertie, Chowan, Edgecombe, Forsyth, Gates, Halifax, Martin, Mecklenburg, Nash, Northampton, Wake, Washington, and Wilson Counties. See Gingles v. Edmisten, 590 F.Supp. 345, 365-66, 376 (E.D.N.C.1984), aff'd in part, rev'd in part sub nom. Thornburg v. Gingles, 478 U.S. at 80, 106 S.Ct. at 2782, 92 L.Ed.2d at 65. Faculty at the North Carolina School of Government advised the chairs of the General Assembly's redistricting committees that North Carolina is still bound by the holding in Gingles. In addition, the United States District Court noted on remand from the decision in Cromartie I that the parties there had stipulated that legally significant racially polarized voting was present in North Carolina's First Congressional District. Cromartie v. Hunt, 133 F.Supp.2d at 422-23. The three-judge panel found that consideration of race in the construction of the First District was reasonably necessary to protect the State from liability under the Voting Rights Act. Id. at 423. This finding by the three-judge panel was not appealed and thus is not affected by the holding in Cromartie II and remains valid.

In addition, the three-judge panel found as fact that the documents submitted by plaintiffs included a law review article prepared by an attorney for the North Carolina NAACP. Anita S. Earls et al., Voting Rights in North Carolina 1982-2006, 17 S. Cal. Rev. L. & Soc. Just. 577 (2008). The court observed that this article "also provided evidence of racially polarized voting as alleged or established in voting rights lawsuits filed in many of the counties[17] in which 2011 VRA districts were enacted." The court added as a finding of fact that no witness testified that racial polarization had disappeared either statewide or in those areas in which the General Assembly previously had created VRA districts.

In Part II of Appendix A, the three-judge panel conducted an individualized analysis of each of the VRA districts created by the General Assembly in 2011. Generally, each finding of fact relates to one district. While four of the findings of fact deal with more than one district, in each such instance those districts are situated within the same county. Each finding of fact in this Part II follows a similar pattern. The finding of fact begins with data that explain how the information in Part I of the Appendix applies to the district under examination. The finding of fact lists the counties included in the district, along with that district's TBVAP. This information is pertinent to the first Gingles precondition, that the minority group is able to demonstrate that it is sufficiently large and geographically compact to constitute a majority in a single-member district. See Pender County, 361 N.C. at 503, 649 S.E.2d at 372 (discussing Gingles, 478 U.S. at 50, 106 S.Ct. at 2766, 92 L.Ed.2d at 46). Subsequent sections of each finding of fact set out how racially polarized voting was found in many of the counties contained within the district or districts, under either Dr. Block's analysis or Dr. Brunell's analysis, or both. This information is pertinent to both the second and third Gingles preconditions: that the minority group is politically cohesive and that the majority votes sufficiently as a bloc to enable it usually to defeat the minority's preferred candidate. LULAC, 548 U.S. at 427, 126 S.Ct. at 2615, 165 L.Ed.2d at 637. Additional information in the findings of fact conveys how many counties within the district or districts are affected by Gingles or Cromartie II, or both. This information is useful in determining the totality of circumstances.

Plaintiffs have not challenged any of the three-judge panel's findings of fact relating to the twenty-six VRA districts, and thus those findings are binding on appeal. Koufman, 330 N.C. at 97, 408 S.E.2d at 731. The three-judge panel's findings of fact indicate that each of the challenged districts had a TBVAP exceeding fifty percent, thus satisfying the first Gingles precondition. See Pender County, 361 N.C. at 503, 649 S.E.2d at 372. The facts found by the three-judge panel also indicate that the maps are sufficient to satisfy the second and third Gingles preconditions, as each district demonstrates racially polarized voting according to Dr. Brunell's analysis. See LULAC, 548 U.S. at 427, 126 S.Ct. at 2615, 165 L.Ed.2d at 637. Although Dr. Block's analysis did not cover some of the counties in some of the challenged districts, they reached the same conclusions where the two studies overlapped.

Moreover, the three-judge panel made additional findings of fact, recited above, that would be relevant to the Gingles totality of circumstances test for twenty-two of the challenged VRA districts.18 Specifically, of the twenty-six VRA districts challenged here, fifteen include counties lying within the area where the Gingles court found section 2 violations; nine include counties lying within the area which the parties in the Cromartie litigation stipulated to have racially polarized voting; and thirteen include counties that were subject to various section 2 lawsuits filed between 1982 and 2006 in which plaintiffs alleged or established racially polarized voting.19 While we assume from the Supreme Court's language in Vera, 517 U.S. at 978, 116 S.Ct. at 1960-61, 135 L.Ed.2d at 269, that satisfaction of the Gingles preconditions is sufficient to trigger a State's compelling interest in avoiding section 2 liability, we believe that this additional evidence, while pertaining to only some of the covered districts, is consistent with and reinforces the three-judge panel's conclusions of law.

Based upon the totality of this evidence, we are satisfied that the three-judge panel correctly found that the General Assembly identified past or present discrimination with sufficient specificity to justify the creation of VRA districts in order to avoid section 2 liability. See Shaw II, 517 U.S. at 909, 116 S.Ct. at 1902, 135 L.Ed.2d at 221. In addition, we see that the General Assembly, before making its redistricting decisions, had a strong basis in evidence on which to reach a conclusion that race-based remedial action was necessary for each VRA district. Id. at 910, 116 S.Ct. at 1903, 135 L.Ed.2d at 222. Accordingly, we conclude that the three-judge panel's findings of fact as to these VRA districts support its conclusion of law that defendants established a compelling state interest in creating districts that would avoid liability under section 2 of the Voting Rights Act. While avoiding liability under section 2 is sufficient to establish a compelling state interest, we now turn to the State's additional desire to comply with section 5.

b. Compelling Interest Under Section 5 of the Voting Rights Act

As noted above, forty of North Carolina's one hundred counties were covered by section 5 at the time of redistricting. This section, which prevents retrogression, forbids "[a]ny voting qualification or prerequisite to voting, or standard, practice, or procedure with respect to voting that has the purpose of or will have the effect of diminishing the ability of any citizens of the United States on account of race or color ... to elect their preferred candidates of choice." 52 U.S.C.A. § 10304(b).20 Section 5 requires preclearance, either by the USDOJ or by a three-judge panel of the United States District Court for the District of Columbia, of any election procedure that is different from that in force on the relevant coverage date. See Perry v. Perez, ___ U.S. ___, ___, 132 S.Ct. 934, 939, 181 L.Ed.2d 900, 904 (2012) (per curiam) (citing Nw. Austin Mun. Util. Dist. No. One v. Holder, 557 U.S. 193, 198, 129 S.Ct. 2504, 2509, 174 L.Ed.2d 140, 147 (2009)). The Supreme Court has left no doubt, however, that in fashioning its redistricting plans, a State must comply with the substantive requirements of section 5, not merely obtain preclearance from the Department of Justice. Miller, 515 U.S. at 922, 115 S.Ct. at 2491, 132 L.Ed.2d at 783. As the Supreme Court intimated in Miller, the Department of Justice is not infallible, so courts have "an independent obligation in adjudicating consequent equal protection challenges to ensure that the State's actions are narrowly tailored to achieve a compelling interest." Id. at 922, 115 S.Ct. at 2491, 132 L.Ed.2d at 783.

We concluded above that compliance with section 5 was a compelling state interest at the time the plan was adopted. Turning then to the facts of this case, we take into account the evidence recited above in our discussion regarding the State's concern about possible section 2 liability. In addition, the appendices to the three-judge panel's Judgment and Memorandum of Decision indicate that all of North Carolina Senate Districts 5, 21, and 28, and all of North Carolina House Districts 5, 7, 12, 24, 42, and 57 are in counties covered by section 5. Also, section 5 covers most of the territory contained in Congressional District 1, Senate Districts 4 and 20, and House Districts 21, 32, and 48. Moreover, all of the twenty-six challenged districts contain areas that previously have been part of majority-minority districts. As a result of their connection with counties covered under section 5, these districts may become subject to nonretrogression analysis. Accordingly, we conclude from the totality of the evidence that a history of discrimination justified the General Assembly's concern about retrogression and compliance with section 5. We further conclude that the General Assembly had a strong basis in evidence on which to reach a conclusion that race-based remedial action was necessary.

3. Narrow Tailoring

Having determined that defendants had a compelling interest both in avoiding section 2 liability and in avoiding retrogression under section 5, we now consider whether the redistricting was sufficiently narrowly tailored to advance those state interests as to the twenty-six districts created to comply with the Voting Rights Act. See Stephenson I, 355 N.C. at 377, 562 S.E.2d at 393.

a. Narrow Tailoring Under Section 2 of the Voting Rights Act

In the context of redistricting,

Vera, 517 U.S. at 977, 116 S.Ct. at 1960, 135 L.Ed.2d at 268 (internal citations omitted). Thus, while a State does not have a free hand when crafting districts with the intent of avoiding section 2 liability, the Supreme Court has acknowledged that "[a] § 2 district that is reasonably compact and regular, taking into account traditional districting principles such as maintaining communities of interest and traditional boundaries, may pass strict scrutiny without having to defeat rival compact districts designed by plaintiffs' experts in endless `beauty contests.'" Id. at 977, 116 S.Ct. at 1960, 135 L.Ed.2d at 269.

As discussed above, the three-judge panel found that the General Assembly designed each of the challenged districts to consist of a TBVAP exceeding fifty percent of the total voting age population in that district. The Supreme Court and this Court have held that doing so is permissible as a method of addressing potential liability under section 2. Strickland, 556 U.S. at 19-20, 129 S.Ct. at 1246, 173 L.Ed.2d at 187; Pender County, 361 N.C. at 503, 649 S.E.2d at 372. Unlike redistricting plans that have been faulted for setting arbitrary thresholds for TBVAP, see, e.g., Smith v. Beasley, 946 F.Supp. 1174 (D.S.C.1996) (order declaring certain majority-minority districts required to have to least 55% TBVAP unconstitutional), the target of fifty percent plus one of the TBVAP chosen by North Carolina's General Assembly is consistent with the requirements of the first Gingles precondition. See Strickland, 556 U.S. at 19-20, 129 S.Ct. at 1246, 173 L.Ed.2d at 187 ("[A] party asserting § 2 liability must show by a preponderance of the evidence that the minority population in the potential election district is greater than 50 percent."). Nevertheless, because section 2 limits the use of race in creating remedial districts by allowing race to be considered only to the extent "reasonably necessary" for compliance, the question arises whether the percentages of TBVAP in each of North Carolina's challenged districts are higher than "reasonably necessary" to avoid the risk of vote dilution. See Vera, 517 U.S. at 979, 116 S.Ct. at 1961, 135 L.Ed.2d at 269.

The TBVAP percentage ranges from a low of 50.45% to a high of 57.33% in the twenty-six districts in question; however, the average TBVAP of the challenged districts is only 52.28%. Twenty-one of the twenty-six districts have TBVAPs of less than 53%, and only two of these districts, Senate District 28 and House District 24, exceed 55% TBVAP. We are mindful that a host of other factors were considered in addition to race, such as the Whole County Provision of the Constitution of North Carolina, protection of incumbents, and partisan considerations, and these factors were considered against a backdrop of the state constitutional requirement of plus or minus five percent population deviation. As a result, we are satisfied that these districts are sufficiently narrowly tailored. They do not classify individuals based upon race to an extent greater than reasonably necessary to comply with section 2 of the VRA, while simultaneously taking into account traditional districting principles.

Plaintiffs argue that creating districts with a TBVAP percentage exceeding fifty percent constitutes impermissible racial packing, citing id. at 983, 116 S.Ct. at 1963, 135 L.Ed.2d at 272; Missouri v. Jenkins, 515 U.S. 70, 88, 115 S.Ct. 2038, 2049, 132 L.Ed.2d 63, 80 (1995); and Shaw I, 509 U.S. at 655, 113 S.Ct. at 2831, 125 L.Ed.2d at 534. Plaintiffs also argue that districts with a TBVAP exceeding fifty percent are not automatically necessary because minority voters in cross-over and coalition districts have elected candidates of their choice where the TBVAP was between forty and fifty percent. This Court previously has considered, but declined to adopt, similar arguments. Pender County, 361 N.C. at 502-04, 649 S.E.2d at 371-73. We concluded in Pender County that applying a bright line rule — that the presence of more than fifty percent of the TBVAP satisfied the first Gingles prong — was logical and gave the General Assembly "a safe harbor for the redistricting process." Id. at 505, 649 S.E.2d at 373; see Strickland, 556 U.S. at 13, 129 S.Ct. at 1242, 173 L.Ed.2d at 183 ("This Court has held that § 2 does not require the creation of influence districts." (citation omitted)).

Although the burden is upon the State under strict scrutiny, the parties challenging the redistricting must also make a showing.

Cromartie II, 532 U.S. at 258, 121 S.Ct. at 1466, 149 L.Ed.2d at 453. Here, when the evidence is undisputed that racial identification correlates highly with party affiliation, plaintiffs have failed to meet this obligation. The General Assembly's plans fall within the safe harbor provisions of Pender County while respecting, to the extent possible, the Whole County Provision, as mandated by Stephenson I. In contrast, plaintiffs' proposals would effectively invite the type of litigation over section 2 claims envisioned in Pender County, see 361 N.C. at 505-06, 649 S.E.2d at 373, while failing to provide for the legitimate nonracial political goals pursued by the General Assembly in its plans.

We are aware of the Supreme Court's warning that "if there were a showing that a State intentionally drew district lines in order to destroy otherwise effective crossover districts, that would raise serious questions under both the Fourteenth and Fifteenth Amendments." Strickland, 556 U.S. at 24, 129 S.Ct. at 1249, 173 L.Ed.2d at 190 (citations omitted). In addition, we acknowledge the Supreme Court's caution in Alabama against a "mechanical interpretation" of the VRA. Alabama, ___ U.S. at ___, 135 S.Ct. at 1273, 191 L.Ed.2d at 335. In addressing these concerns, we note that the legislature adopted a TBVAP threshold of fifty percent plus one, a baseline number that allowed for flexibility within each district. The average TBVAP in the twenty-six VRA districts is 52.28% of the total voting age population, and no district's TBVAP is higher than 57.33%. This figure indicates that minority voters were moved out of other districts only to the extent necessary to meet Pender County's safe harbor provision, while simultaneously pursuing other legitimate nonracial political goals, including those mentioned above. Unlike in Alabama, where it was "difficult to explain just why ... maintain[ing] the black population at 70% is `narrowly tailored' to" ensure that the minority community can maintain its ability to elect its candidate of choice, id. at ___, 135 S.Ct. at 1273, 191 L.Ed.2d at 335, the TBVAP here was well within a reasonable range to ensure that a section 2 district was drawn where a minority community was actually a majority of the population. Where racial identification correlates highly with party affiliation, placing additional Democratic voters in districts that already vote Democratic is not forbidden as long as the motivation for doing so is not primarily racial. See Cromartie I, 526 U.S. at 551-52, 119 S.Ct. at 1551, 143 L.Ed.2d at 741; see also Strickland, 556 U.S. at 20, 129 S.Ct. at 1246, 173 L.Ed.2d at 188 ("Section 2 does not guarantee minority voters an electoral advantage."). Accordingly, we conclude that plaintiffs have failed to demonstrate improper packing or gerrymandering based upon race and that the districts are narrowly tailored to comply with section 2.

b. Narrow Tailoring Under Section 5 of the Voting Rights Act

We first note that, as discussed above, the twenty-six challenged VRA districts survive strict scrutiny on the basis of section 2 alone. Nevertheless, we conclude that the challenged VRA districts are also narrowly tailored to advance the State's compelling interest in avoiding retrogression under section 5.

The "purpose of § 5 has always been to insure that no voting-procedure changes would be made that would lead to a retrogression in the position of racial minorities with respect to their effective exercise of the electoral franchise." Beer v. United States, 425 U.S. 130, 141, 96 S.Ct. 1357, 1364, 47 L.Ed.2d 629, 639 (1976). Section 5, however, does not "give covered jurisdictions carte blanche to engage in racial gerrymandering in the name of nonretrogression. A reapportionment plan would not be narrowly tailored to the goal of avoiding retrogression if the State went beyond what was reasonably necessary to avoid retrogression." Shaw I, 509 U.S. at 655, 113 S.Ct. at 2831, 125 L.Ed.2d at 534.

In Alabama, the Supreme Court made it clear that section 5 "does not require a covered jurisdiction to maintain a particular numerical minority percentage" in covered jurisdictions. ___ U.S. at ___, 135 S.Ct. at 1272, 191 L.Ed.2d at 334. But section 5 does require a covered "jurisdiction to maintain a minority's ability to elect a preferred candidate of choice." Id. at ___, 135 S.Ct. at 1272, 191 L.Ed.2d at 334. The Court explained that "we do not insist that a legislature guess precisely what percentage reduction a court or the Justice Department might eventually find to be retrogressive." Id. at ___, 135 S.Ct. at 1273, 191 L.Ed.2d at 335. Rather, "the legislature must have a `strong basis in evidence' in support of the (race-based) choice that it has made." Id. at ___, 135 S.Ct. at 1274, 191 L.Ed.2d at 336. As an example, the Court noted that a "[one percent] reduction in a [seventy percent] black population district" is not "necessarily retrogressive." Id. at ___, 135 S.Ct. at 1273, 191 L.Ed.2d at 335. The Court cautioned against heavy reliance "upon a mechanically numerical view as to what counts as forbidden retrogression," as opposed to "the more purpose-oriented view reflected in the statute's language." Id. at ___, 135 S.Ct. at 1273, 191 L.Ed.2d at 335.

Plaintiffs argue that by increasing the TBVAP in the challenged VRA districts to at least fifty percent plus one, the legislature improperly relied upon section 5 to unnecessarily augment, not just maintain, black voters' ability to elect their preferred candidate of choice. See Vera, 517 U.S. at 983, 116 S.Ct. at 1963, 135 L.Ed.2d at 272 (concluding that an increase in TBVAP from 40.8% to 50.9% was unnecessary to ensure nonretrogression). Plaintiffs argue that such a substantial increase in TBVAP in districts where black voters were already able to elect their candidates of choice goes beyond what was necessary to ensure nonretrogression. Plaintiffs essentially ask us to force the State to "unpack" majority-minority districts and replace them with coalition, crossover, and influence districts. In enacting amendments to the VRA in 2006, however, Congress made clear that section 5 was designed to prevent legislatures from "unpacking majority-minority districts" and that it did not "lock into place coalition or influence districts." S.Rep. No. 109-295, at 18-19 (2006); see Miller, 515 U.S. at 923, 115 S.Ct. at 2492, 132 L.Ed.2d at 784 ("Wherever a plan ... increas[es] the number of majority-minority districts, it `cannot violate § 5 unless the new apportionment itself so discriminates on the basis of race or color as to violate the Constitution.'" (quoting Beer, 425 U.S. at 141, 96 S.Ct. at 1364, 47 L.Ed.2d at 639)); see also Strickland, 556 U.S. at 23, 129 S.Ct. at 1248, 173 L.Ed.2d at 190 (stating that the VRA "does not mandate creating or preserving crossover districts"). "[T]he various provisions of an act should be read so that all may, if possible, have their due and conjoint effect without repugnancy or inconsistency, so as to render the statute a consistent and harmonious whole." Walker v. Am. Bakeries Co., 234 N.C. 440, 442, 67 S.E.2d 459, 461 (1951) (quoting 50 Am. Jur. Statutes § 363 (1944)); see FDA v. Brown & Williamson Tobacco Corp., 529 U.S. 120, 133, 120 S.Ct. 1291, 1301, 146 L.Ed.2d 121, 134 (2000). As discussed above, the legislature adopted a fifty percent plus one threshold to avoid liability under section 2. That threshold is the exact number that the Supreme Court approved in Strickland for a State to use in creating districts to comply with section 2. If, on the one hand, a TBVAP exceeding fifty percent is required to avoid section 2 liability, we cannot, on the other hand, conclude that this percentage is higher than necessary to avoid retrogression under section 5. In other words, section 5 cannot forbid what section 2 requires. While the legislature may have chosen a threshold of fifty plus one percent, such a deliberate decision was not arbitrary and was made with the "purpose-oriented view" of complying with section 2. See Alabama, ___ U.S. at ___, 135 S.Ct. at 1273, 191 L.Ed.2d at 335.

Plaintiffs' argument seeks to undo the Supreme Court's holding in Strickland, which affirmed this Court's decision in Pender County. In Strickland the Supreme Court explicitly rejected State election officials' claim that "[section] 2 required them to override state law and split Pender County, drawing District 18 with [a black] voting-age population of 39.36 percent rather than keeping Pender County whole and leaving District 18 with [a black] voting-age population of 35.33 percent." 556 U.S. at 14, 129 S.Ct. at 1243, 173 L.Ed.2d at 184. The Supreme Court stated that "[t]here is a difference between a racial minority group's `own choice' and the choice made by a coalition," id. at 15, 129 S.Ct. at 1244, 173 L.Ed.2d at 185, and the Court made it clear that the creation of crossover districts is "a matter of legislative choice or discretion," to be exercised within the bounds of state law, id. at 23, 129 S.Ct. at 1248, 173 L.Ed.2d at 190.

As the Supreme Court stated in Alabama, "legislators may have a strong basis in evidence to use racial classifications in order to comply with a statute when they have good reasons to believe such use is required, even if a court does not find that the actions were necessary for statutory compliance." ___ U.S. at ___, 135 S.Ct. at 1274, 191 L.Ed.2d at 336 (citation and quotation marks omitted). Here the legislature had a strong basis in evidence of racially polarized voting to justify creating majority-black districts with a TBVAP in excess of fifty percent even though that same action resulted in an increase in the TBVAP in former coalition districts. Accordingly, we conclude that the challenged VRA districts were narrowly tailored to achieve a compelling interest in complying with both section 2 and section 5 of the VRA. Therefore, they survive strict scrutiny.

4. Proportionality

Finally, because plaintiffs challenge the General Assembly's consideration of proportionality, the three-judge panel analyzed whether the legislature used proportionality in the enacted plans improperly to "link[] the number of majority-minority voting districts to minority members' share of the relevant population." See De Grandy, 512 U.S. at 1014 n. 11, 114 S.Ct. at 2658 n. 11, 129 L.Ed.2d at 792 n. 11. The three-judge panel found as fact that "the General Assembly acknowledges that it intended to create as many VRA districts as needed to achieve a `roughly proportionate' number of Senate, House and Congressional districts as compared to the Black population in North Carolina," adding that each VRA district had to be at least fifty percent black in voting age population. The three-judge panel specifically found that the General Assembly's enacted plans

Based on these and other findings, the three-judge panel concluded that "the General Assembly had a strong basis in evidence for concluding that `rough proportionality' was reasonably necessary to protect the State from anticipated liability under § 2 of the VRA and ensuring preclearance under § 5 of the VRA."

Plaintiffs argue that this conclusion is erroneous as a matter of law because racial proportionality is neither a compelling governmental interest nor a requirement of the VRA. They contend that, because "[t]he VRA was not designed to guarantee majority-minority voting districts, but to guarantee that the processes, procedures, and protocols would be fair and free of racial discrimination," the legislature's redistricting was based upon an unconstitutional premise. Plaintiffs contend that, by focusing on proportionality at the statewide level, the General Assembly necessarily predetermined how many VRA districts to draw without first considering where potential liability existed for section 2 violations. Plaintiffs maintain that, as a result, the General Assembly's process sought "`outright racial balancing,'" which is "patently unconstitutional" under such cases as Fisher v. University of Texas at Austin, ___ U.S. ___, ___, 133 S.Ct. 2411, 2419, 186 L.Ed.2d 474, 486 (2013); Parents Involved in Community Schools v. Seattle School District No. 1, 551 U.S. 701, 729-30, 127 S.Ct. 2738, 2757, 168 L.Ed.2d 508, 529 (2007) (plurality); and Grutter v. Bollinger, 539 U.S. 306, 330, 123 S.Ct. 2325, 2339, 156 L.Ed.2d 304, 333 (2003), and thus can neither be required by section 2 nor constitute a compelling state interest.

The VRA provides that "nothing in this section establishes a right to have members of a protected class elected in numbers equal to their proportion in the population." 52 U.S.C.A. § 10301(b). Consistent with this proviso, the Supreme Court has repeatedly held that proportionality does not provide a safe harbor for States seeking to comply with section 2. LULAC, 548 U.S. at 436, 126 S.Ct. at 2620, 165 L.Ed.2d at 642-43 (citing De Grandy, 512 U.S. at 1017-21, 114 S.Ct. at 2660-62, 129 L.Ed.2d at 794-97). Such a rule "would be in derogation of the statutory text and its considered purpose ... and of the ideal that the Voting Rights Act of 1965 attempts to foster," De Grandy, 512 U.S. at 1018, 114 S.Ct. at 2660, 129 L.Ed.2d at 795, and could allow "the most blatant racial gerrymandering... so long as proportionality was the bottom line," id. at 1019, 114 S.Ct. at 2661, 129 L.Ed.2d at 796. Even so, the Court has also held that proportionality can be an element of the "totality of circumstances" test under Gingles. Id. at 1000, 114 S.Ct. at 2651, 129 L.Ed.2d at 784. When considered in this manner, the Court has instructed that the "probative value assigned to proportionality may vary with other facts" and "[n]o single statistic provides courts with a shortcut to determine whether a set of single-member districts unlawfully dilutes minority voting strength." Id. at 1020-21, 114 S.Ct. at 2661-62, 129 L.Ed.2d at 797; see also LULAC, 548 U.S. at 436, 126 S.Ct. at 2620, 165 L.Ed.2d at 642.

In light of these standards, the record here demonstrates that the General Assembly did not use proportionality improperly to guarantee the number of majority-minority voting districts based on the minority members' share of the relevant population. We believe that such an effort, seeking to guarantee proportional representation, proportional success, or racial balancing, would run afoul of the Equal Protection Clause. See De Grandy, 512 U.S. at 1017-22, 114 S.Ct. at 2658-62, 129 L.Ed.2d at 794-98. Instead, the General Assembly considered rough proportionality in a manner similar to its precautionary consideration of the Gingles preconditions, as a means of protecting the redistricting plans from potential legal challenges under section 2's totality of the circumstances test. Proportionality was not a dispositive factor, but merely one consideration of many described in the materials and other contributions from numerous organizations, experts, and lay witnesses. The General Assembly's consideration of rough proportionality was a means of ensuring against voter dilution and potential section 2 liability, not an attempt to trade "the rights of some minority voters under § 2 ... off against the rights of other members of the same minority class." Id. at 1019, 114 S.Ct. at 2661, 129 L.Ed.2d at 796. Accordingly, we conclude that this factor does not constitute grounds for a violation of section 2.

Thus, with regard to the VRA districts, we hold that, while the General Assembly considered race, the three-judge panel erred by concluding prematurely that race was the predominant factor motivating the drawing of the districts without first performing adequate fact finding. Nonetheless, because we held above that the three-judge panel correctly found that each of the twenty-six districts survives strict scrutiny, we need not remand the case for reconsideration under what may be a less demanding level of scrutiny.

C. Non-VRA Districts

We now turn to the four districts that the three-judge panel found were not drawn as VRA districts but which were challenged by plaintiffs as being the result of racial gerrymandering. For trial, the three-judge panel characterized the issue as follows: "For six specific districts (Senate Districts 31 and 32, House Districts 51 and 54 and Congressional Districts 4 and 12 — none of which is identified as a VRA district), what was the predominant factor in the drawing of those districts?" Although the plaintiffs did not challenge Senate District 31 or House District 51, the three-judge panel found an examination of Senate District 32 and House District 54 to be intertwined with the associated districts because of the groupings required under the state constitution's Whole County Provision. Thus, the three-judge panel recognized that, because of the Whole County Provision, a determination of the predominant motive necessarily included the motivation in creating the required county pairings.

After receiving evidence, the three-judge panel made numerous specific findings of fact, district by district, as to whether race was the General Assembly's predominant motive in drafting these districts. Looking first at the challenged Congressional Districts, the court found that race was not a factor in drawing Congressional Districts 4 or 12. In fact, the record indicates that the drafters of these two districts did not consider any racial data. The panel found that political goals were a factor in drawing these two Congressional Districts, as well as compliance with Cromartie II.

Turning to the two challenged state legislative districts, Senate District 32 and House District 54, the three-judge panel received evidence and made a number of uncontested specific factual findings regarding the creation of each of these non-VRA districts. The three-judge panel found that, in addition to complying with federal and state law, the General Assembly desired to create districts "more competitive for" the majority party. The panel noted the "undisputed" fact that in North Carolina "racial identification correlates highly with political affiliation." The desire to create districts more competitive for the majority party had to occur within the requirements of the state constitution's Whole County Provision. Thus, while the three-judge panel noted the General Assembly's desire "to equalize population among the districts," for state redistricting purposes, this finding must be viewed in the context of the Whole County Provision, which recognizes political subdivisions. See Stephenson I, 355 N.C. at 366, 562 S.E.2d at 386 (recognizing "the importance of the county to our system of government" and that "[i]t is through [counties], mainly, that the powers of government reach and operate directly upon the people" (quoting White v. Comm'rs of Chowan Cty., 90 N.C. 437, 438 (1884))). In light of the Whole County Provision, equalizing population is not simply a matter of deciding how many voters are placed in each district; the districting proceedings also determine which voters are included in the joined counties. In addition, the Whole County Provision establishes a hierarchy regarding the required number of contiguous counties to be joined.

Ultimately, regarding the four challenged non-VRA districts, the three-judge panel concluded "that race was not the predominant motive in the creation of [these] districts." The panel then applied rational basis review and concluded that the General Assembly's creation of the non-VRA districts was constitutional. The three-judge panel noted that plaintiffs

The Supreme Court has recognized that compliance with federal law, incumbency protection, and partisan advantage are all legitimate governmental interests, see Shaw I, 509 U.S. at 654, 113 S.Ct. at 2830, 125 L.Ed.2d at 533 (compliance with federal law); Karcher, 462 U.S. at 740, 103 S.Ct. at 2663, 77 L.Ed.2d at 147 (incumbency protection); Cromartie I, 526 U.S. at 551, 119 S.Ct. at 1551, 143 L.Ed.2d at 741 (partisan interests), as is honoring traditional districting principles, such as compactness, contiguity, and respect for political subdivisions, Miller, 515 U.S. at 916, 115 S.Ct. at 2488, 132 L.Ed.2d at 779-80, which is embodied in our state constitution's Whole County Provision, Stephenson I, 355 N.C. at 371, 562 S.E.2d at 389 ("We observe that the State Constitution's limitations upon redistricting and apportionment [the Whole County Provision] uphold what the United States Supreme Court has termed `traditional districting principles.' These principles include factors such as `compactness, contiguity, and respect for political subdivisions.' The United States Supreme Court has `emphasize[d] that these criteria are important not because they are constitutionally required — they are not — but because they are objective factors that may serve to defeat a claim that a district has been gerrymandered on racial lines.'" (internal citations omitted)). In light of this authority and the three-judge panel's findings of fact, we agree that plaintiffs failed to establish that race was the dominant factor in drafting these districts and conclude that the panel's application of the rational basis test was appropriate. The court's findings of fact support its conclusions of law. The General Assembly's actions in creating these districts were rationally related to all its expressed goals. Accordingly, we affirm the three-judge panel as to these non-VRA districts.

IV. Plaintiffs' State Claims

We now consider plaintiffs' claims brought under state law. Initially, we note that our analysis here is unaffected by the holding in Alabama. Plaintiffs argue that the three-judge panel erred when it failed to find that the enacted Senate and House plans violate the Whole County Provision of the North Carolina Constitution. Article II, Section 3(3) of the Constitution of North Carolina provides that "[n]o county shall be divided in the formation of a senate district," while Article II, Section 5(3) contains a similar provision with regard to each representative district.

The tension between the Whole County Provision and federal requirements is apparent. In 1983, a three-judge panel of the United States District Court for the Eastern District of North Carolina held that the Whole County Provision was unenforceable anywhere in the state. Cavanagh v. Brock, 577 F.Supp. 176, 181-82 (E.D.N.C.1983). This Court subsequently rejected Cavanagh's analysis and held that the Whole County Provision remained enforceable to the extent that it could be harmonized with federal law. Stephenson I, 355 N.C. at 374, 562 S.E.2d at 391. This Court provided a roadmap to compliance with the Whole County Provision, requiring each created district to be contained within a single county or be part of a pairing with other districts containing as few contiguous counties as possible. Id. at 383-84, 562 S.E.2d at 396-97. As a result, the Whole County Provision remains in effect but must accommodate both the one-person, one-vote mandate and the requirements of the VRA. Because the Constitution of North Carolina requires that each senator and each representative represent "as nearly as may be" an equal number of inhabitants, N.C. Const. art. II, §§ 3(1), 5(1), the former federal requirement is met by definition. Thus, we consider plaintiffs' contentions that the challenged House and Senate Districts violate the Whole County Provision, as harmonized with the VRA.

As previously noted, this Court has set out nine criteria for ensuring that House and Senate Districts satisfy both the Whole County Provision and the Voting Rights Act. Stephenson I, 355 N.C. at 383-84, 562 S.E.2d at 396-97. These criteria may be summarized as follows: First, "legislative districts required by the VRA shall be formed" before non-VRA districts. Id. at 383, 562 S.E.2d at 396-97. Second, "[i]n forming new legislative districts, any deviation from the ideal population for a legislative district shall be at or within plus or minus five percent" to ensure "compliance with federal `one-person, one-vote' requirements." Id. at 383, 562 S.E.2d at 397. Third, "in counties having a ... population sufficient to support the formation of one non-VRA legislative district," "the physical boundaries" of the non-VRA district shall "not cross or traverse the exterior geographic line of" the county. Id. at 383, 562 S.E.2d at 397. Fourth, "[w]hen two or more non-VRA legislative districts may be created within a single county," "single-member non-VRA districts shall be formed within" the county, "shall be compact," and "shall not traverse" the county's exterior geographic line. Id. at 383, 562 S.E.2d at 397. Fifth, for non-VRA counties that "cannot support at least one legislative district," or counties "having a non-VRA population pool" that, "if divided into" legislative "districts, would not comply with" one-person, one-vote requirements, the General Assembly should combine or group "the minimum number of whole, contiguous counties necessary to comply with the at or within plus or minus five percent `one-person, one-vote' standard." Id. at 383, 562 S.E.2d at 397. Moreover, "[w]ithin any such contiguous multi-county grouping, compact districts shall be formed, consistent with the [one-person, one-vote] standard, whose boundary lines do not cross or traverse the `exterior' line of the multi-county grouping." Id. at 383-84, 562 S.E.2d at 397. "[T]he resulting interior county lines created by any such groupings may be crossed or traversed in the creation of districts within said multi-county grouping but only to the extent necessary to comply with the at or within plus or minus five percent `one-person, one-vote' standard." Id. at 384, 562 S.E.2d at 397. Sixth, "only the smallest number of counties necessary to comply with the at or within plus or minus five percent `one-person, one-vote' standard shall be combined." Id. at 384, 562 S.E.2d at 397. Seventh, "communities of interest should be considered in the formation of compact and contiguous [legislative] districts." Id. at 384, 562 S.E.2d at 397. Eighth, "multi-member districts shall not be" created "unless it is established that such districts are necessary to advance a compelling governmental interest." Id. at 384, 562 S.E.2d at 397. Ninth, "any new redistricting plans ... shall depart from strict compliance with" these criteria "only to the extent necessary to comply with federal law." Id. at 384, 562 S.E.2d at 397.

In their discussion of the Whole County Provision, plaintiffs contend that the test of a plan's compliance with Stephenson I's fifth and sixth criteria is the number of counties left undivided. They argue that the current plan violates Stephenson I because it divides counties and traverses county lines to an unnecessary extent. In support of their argument, plaintiffs submit charts indicating that their suggested "House Fair and Legal" plan results in five fewer divided counties and six fewer county line traversals than the enacted House plan, while maintaining the same number of groupings. Similarly, plaintiffs' charts indicate that their suggested "Senate Fair and Legal" plan divides five fewer counties and contains eleven fewer traversals of county lines than the enacted Senate plan.

Defendants respond that plaintiffs have misinterpreted the requirements of Stephenson I. According to defendants, Stephenson I is satisfied by minimizing the number of counties contained within each multi-county grouping. In other words, a proper plan maximizes the number of possible two-county groupings before going on to create three-county groupings, maximizes the number of possible three-county groupings before creating four-county groupings, and so on. Defendants argue that plaintiffs have misread Stephenson I because, under Stephenson I, divisions of counties and traversals of county lines are relevant only if plaintiffs' alternative maps are comparable to the State's maps in terms of the number of counties within each grouping. In support of its argument, the State provides charts showing that the enacted House and Senate plans result in a greater number of groupings that contain fewer counties, as compared with the various proposed alternative plans, all of which create groupings that contain more counties than the enacted plans. To illustrate, the enacted House district plan contains eleven groupings consisting of one county and fifteen groupings consisting of two counties. The closest comparable alternative plan proposed by plaintiffs, House Fair and Legal, also contains eleven groupings consisting of one county but only nine groupings consisting of two counties. Similarly, while both the enacted Senate plan and plaintiffs' proposed Senate Fair and Legal contain one grouping consisting of one county and eleven groupings consisting of two counties, the enacted plan contains four districts consisting of three counties while Senate Fair and Legal contains only three groupings consisting of three counties.

While we are conscious of the efforts of the litigants to interpret Stephenson I's requirements faithfully, after careful review of our opinions in Stephenson I and Pender County, we are satisfied that defendants' interpretation is correct. Stephenson I's fifth factor states that, when combining two or more counties to comply with the one-person, one-vote standard, "the requirements of the [Whole County Provision] are met by combining or grouping the minimum number of whole, contiguous counties necessary" for compliance. 355 N.C. at 384, 562 S.E.2d at 397. Only after these groupings have been established does Stephenson I state that "the resulting interior county lines ... may be crossed or traversed ... only to the extent necessary to comply with the ... `one-person, one-vote' standard." Id. at 384, 562 S.E.2d at 397. Thus, the process established by this Court in Stephenson I and its progeny requires that, in establishing legislative districts, the General Assembly first must create all necessary VRA districts, single-county districts, and single counties containing multiple districts. Thereafter, the General Assembly should make every effort to ensure that the maximum number of groupings containing two whole, contiguous counties are established before resorting to groupings containing three whole, contiguous counties, and so on. As shown by the charts provided by defendants, plaintiffs have not produced an alternative plan that complies with a correct reading of Stephenson I's fifth and sixth factors better than the plans enacted by the General Assembly. Because the enacted plans result in groupings containing fewer whole, contiguous counties than do any of plaintiffs' plans, we need not discuss the number of counties divided or county lines traversed.

In addition, the maps that plaintiffs employ to support their arguments regarding the Whole County Provision are not helpful because they are inconsistent with our holding in Pender County as affirmed by the Supreme Court in Strickland. In Strickland the Supreme Court held that the first Gingles precondition can be shown only where the minority population is fifty percent plus one of the TBVAP. Strickland, 556 U.S. at 19-20, 129 S.Ct. at 1246, 173 L.Ed.2d at 187 ("[A] party asserting § 2 liability must show by a preponderance of the evidence that the minority population in the potential election district is greater than 50 percent."); Pender County, 361 N.C. at 502, 649 S.E.2d at 371 (The "minority group must constitute a numerical majority of the voting population in the area under consideration before Section 2 of the VRA requires the creation of a legislative district to prevent dilution of the votes of that minority group."). Here, like the plaintiffs in Strickland, plaintiffs argue that we should adopt a standard that allows VRA requirements to be satisfied by other forms of minority districts, such as coalition and crossover districts. Not only is plaintiffs' argument inconsistent with the Supreme Court's holding in Strickland, but this flawed approach adversely affects the first step of the process required by Stephenson I, the formation of VRA districts. As a result, plaintiffs' maps are distorted ab initio and the distortion is compounded at each subsequent step. Consequently, even if plaintiffs' proposed alternative plans were comparable to the enacted plans in terms of the number and composition of county groupings, their incompatibility with Pender County means that they cannot serve as an adequate basis for comparison with the enacted plans.

Plaintiffs have also compared the General Assembly's enacted plans with earlier redistricting plans approved in North Carolina; however, those plans were tailored to a particular time and were based upon then-existing census numbers and population concentrations. The requirement that the State maintain its one-person, one-vote standard as populations shift makes comparisons between current and previous districting plans of limited value. The utility of prior plans is further diminished by subsequent clarifications of the legal standards in effect when these earlier plans were promulgated. See, e.g., Pender County, 361 N.C. at 503-04, 649 S.E.2d at 372 (explaining the requirements of the first Gingles precondition). As a result, no meaningful comparisons can be made in this case.

Separately, plaintiffs argue that this Court should consider the purported lack of compactness of the districts created by the General Assembly and the harm resulting from splitting precincts. While these may be valid considerations, neither constitutes an independent legal basis for finding a violation, and we are unaware of any justiciable standard by which to measure these local factors. See Vera, 517 U.S. at 999, 116 S.Ct. at 1972, 135 L.Ed.2d at 284 (Kennedy, J., concurring) ("Districts not drawn for impermissible reasons or according to impermissible criteria may take any shape, even a bizarre one," so long as "the bizarre shape ... is [not] attributable to race-based districting unjustified by a compelling interest.").

Finally, plaintiffs argue that the enacted plans violate the "Good of the Whole" clause found in Article I, Section 2 of the Constitution of North Carolina. We do not doubt that plaintiffs' proffered maps represent their good faith understanding of a plan they believe to be best for our State as a whole; however, the maps enacted by the duly elected General Assembly also represent an equally legitimate understanding of legislative districts that will function for the good of the whole. Because plaintiffs' argument is not based upon a justiciable standard, and because acts of the General Assembly enjoy "a strong presumption of constitutionality," Pope v. Easley, 354 N.C. 544, 546, 556 S.E.2d 265, 267 (2001) (per curiam) (citation omitted), plaintiffs' claims fail.

V. Conclusion

We agree with the unanimous three-judge panel that the General Assembly's enacted plans do not violate plaintiffs' constitutional rights. We hold that the enacted House and Senate plans, as well as the federal Congressional plan, satisfy state and federal constitutional and statutory requirements and, specifically, that the three-judge panel's decision fully complies with the Supreme Court's decision in Alabama. Accordingly, we reaffirm the three-judge panel's Judgment and Memorandum of Decision.

AFFIRMED.

Justice BEASLEY, concurring in part and dissenting in part.

I agree with the judgment of the Court as to plaintiffs' challenge under the "Good of the Whole" Clause in Article I, Section 2 of the North Carolina Constitution; however, because I would conclude that the parties and the jurisprudence of this State would be better served by vacating the trial court's Judgment and remanding this case to the trial court for more complete findings of fact consistent with the guidance provided in Alabama Legislative Black Caucus v. Alabama, I respectfully dissent.

The order of the United States Supreme Court (the Supreme Court) vacating and remanding this Court's judgment in Dickson v. Rucho, 367 N.C. 542, 766 S.E.2d 238 (2014), directed this Court to reconsider the case in light of Alabama Legislative Black Caucus v. Alabama, 575 U.S. ___, 135 S.Ct. 1257, 191 L.Ed.2d 314 (2015) (ALBC). See Dickson v. Rucho, 575 U.S. ___, 135 S.Ct. 1843, 191 L.Ed.2d 719 (2015). The majority reads ALBC so narrowly that its implications for the case before this Court are negligible at best. In my view, if the Supreme Court saw fit to vacate and remand the previous judgment for reconsideration in light of ALBC, this Court would do well to give credence to the legal principles imparted in that decision.1

To that end, I am of the opinion that ALBC bolsters all the points made in my previous dissent, particularly with respect to the General Assembly's use of proportionality as a benchmark, and provides authoritative guidance on how the trial court should have viewed the record of evidence before it. I stand by everything espoused in that dissent, and for that reason, portions of my previous dissent appear in this opinion.

The ALBC plaintiffs, like plaintiffs here, contended that their state legislature's enacted redistricting plans constituted racial gerrymanders in violation of the Equal Protection Clause of the Fourteenth Amendment. ALBC, 575 U.S. at ___, 135 S.Ct. at 1262, 191 L.Ed.2d at 323. Alabama defended its enacted plans by arguing that the plans reflected its goals of coming close to a one-person, one-vote ideal, with no more than a 1% deviation from that ideal, and avoiding retrogression under § 5 of the Voting Rights Act of 1965.2 Id. at ___, 135 S.Ct. at 1263, 191 L.Ed.2d at 324. The federal district court held that the plaintiffs "had failed to prove their racial gerrymandering claims." Id. at ___, 135 S.Ct. at 1264, 191 L.Ed.2d at 325. The Supreme Court vacated the district court's judgment, citing errors in the district court's application and interpretation of the pertinent law. Id. at ___, 135 S.Ct. at 1264, 191 L.Ed.2d at 325-26. Although the arguments made by the parties in ALBC are somewhat different from the arguments made by the parties here, ALBC illuminates errors in the trial court's analysis of plaintiffs' racial gerrymandering claims in this case, and as stated above, provides guidance on how the trial court should approach the analysis on remand. This Court's majority does not give the trial court the opportunity to get it right and fails to require the trial court to make conclusions of law resting on adequate findings of fact that reflect a correct application of the law. I cannot agree with this decision.

I.

"Classifications of citizens solely on the basis of race `are by their very nature odious to a free people whose institutions are founded upon the doctrine of equality.'" Shaw v. Reno, 509 U.S. 630, 643, 113 S.Ct. 2816, 2824, 125 L.Ed.2d 511, 526 (1993) (Shaw I) (quoting Hirabayashi v. United States, 320 U.S. 81, 100, 63 S.Ct. 1375, 1385, 87 L.Ed. 1774, 1786 (1943)). "One of the principal reasons race is treated as a forbidden classification is that it demeans the dignity and worth of a person to be judged by ancestry instead of by his or her own merit and essential qualities." Rice v. Cayetano, 528 U.S. 495, 517, 120 S.Ct. 1044, 1057, 145 L.Ed.2d 1007, 1026 (2000). "Furthermore, `[i]t is axiomatic that racial classifications do not become legitimate on the assumption that all persons suffer them in equal degree.' These principles apply to the drawing of electoral and political boundaries." Johnson v. De Grandy, 512 U.S. 997, 1029-30, 114 S.Ct. 2647, 2666, 129 L.Ed.2d 775, 802 (1994) (Kennedy, J., concurring in part and concurring in the judgment) (alteration in original) (internal citations omitted) (quoting Powers v. Ohio, 499 U.S. 400, 410, 111 S.Ct. 1364, 1370, 113 L.Ed.2d 411, 425 (1991)). Accordingly, the Supreme Court has held that "the Fourteenth Amendment requires state legislation that expressly distinguishes among citizens because of their race to be narrowly tailored to further a compelling governmental interest." Shaw I, 509 U.S. at 643, 113 S.Ct. at 2825, 125 L.Ed.2d at 526 (citations omitted). "Applying traditional equal protection principles in the voting-rights context is `a most delicate task' ... because a legislature may be conscious of the voters' races without using race as a basis for assigning voters to districts." Shaw v. Hunt, 517 U.S. 899, 905, 116 S.Ct. 1894, 1900, 135 L.Ed.2d 207, 218 (1996) (Shaw II) (quoting Miller v. Johnson, 515 U.S. 900, 905, 115 S.Ct. 2475, 2483, 132 L.Ed.2d 762, 772 (1995)). Therefore, race must be "the predominant factor motivating the legislature's decision" in order to trigger strict scrutiny. Miller, 515 U.S. at 916, 115 S.Ct. at 2488, 132 L.Ed.2d at 779.

The burden to make this showing falls to the plaintiff:

Id. at 916, 115 S.Ct. at 2488, 132 L.Ed.2d at 779-80.

If the plaintiff satisfies this initial burden of production,3 the redistricting legislation "cannot be upheld unless it satisfies strict scrutiny, [the] most rigorous and exacting standard of constitutional review." Id. at 920, 115 S.Ct. at 2490, 132 L.Ed.2d at 782. Once strict scrutiny review is triggered, the burden shifts to the State to "show not only that its redistricting plan was in pursuit of a compelling state interest, but also that `its districting legislation is narrowly tailored to achieve [that] compelling interest.'" Shaw II, 517 U.S. at 908, 116 S.Ct. at 1902, 135 L.Ed.2d at 220-21 (alteration in original) (quoting Miller, 515 U.S. at 920, 115 S.Ct. at 2490, 132 L.Ed.2d at 782).

a.

In my earlier dissent, I questioned this Court's handling of the apparent deficiencies in the trial court's findings of fact with respect to whether race was the predominant motivating factor for the General Assembly in the creation of the twenty-six VRA districts. See Dickson v. Rucho, 367 N.C. 542, 578-79, 766 S.E.2d 238, 262-63 (2014) (Beasley, J., concurring in part and dissenting in part). The trial court found that

(Footnote call numbers omitted.) Accordingly, the trial court "conclude[d] ... that in drawing [the] VRA districts ... [,] the shape, location and racial composition of each VRA district was predominantly determined by a racial objective and was the result of a racial classification sufficient to trigger the application of strict scrutiny as a matter of law."

On remand, this Court holds, just as it did before, that the trial court's finding that "no factual inquiry was required regarding the General Assembly's predominant motivation in forming the twenty-six VRA districts beyond the General Assembly's concession that the districts were drafted to be VRA-compliant" was insufficient to show "whether race fairly can be described as the predominant factor." The majority then proceeds under the assumption that race was the predominant motivating factor and, accordingly, embarks on a strict scrutiny analysis. I maintain that, instead of perpetuating the trial court's mistake in making "truncated findings of fact," this Court should remand this case to the trial court to correct the deficiency because the citizens of this state would be better served if we held to our usual course and vacated the trial court's Judgment and remanded the case to the trial court for proper findings of fact and conclusions of law based upon a correct interpretation of the law. When viewed in light of the guidance provided in ALBC, the deficiencies in the trial court's findings with respect to predominance and the error of this Court in ignoring the same are clear.

One of the Alabama legislature's goals in developing its redistricting plan was to come as close as possible to the ideal one-person, one-vote population with a deviation of no more than 1%. ALBC, 575 U.S. at ___, 135 S.Ct. at 1263, 191 L.Ed.2d at 324. In other words, Alabama endeavored to create districts of approximately equal population. The district court concluded that race was not a predominant motivating factor because, on balance, the plans reflected nonracial motivations, such as an attempt to equalize population. Id. at ___, 135 S.Ct. at 1270, 191 L.Ed.2d at 331. The Supreme Court took issue with the district court's "calculat[ion]" of predominance, reasoning that "an equal population goal is not one factor among others to be weighed against the use of race to determine whether race `predominates.' Rather, it is part of the redistricting background, taken as a given, when determining whether race, or other factors, predominate in a legislator's determination as to how equal population objectives will be met." Id. at ___, 135 S.Ct. at 1270, 191 L.Ed.2d at 332. Further, the Supreme Court explained that predominance has a special meaning in the context of racial gerrymandering claims because "[i]t is not about whether a legislature believes that the need for equal population takes ultimate priority. Rather, it is, as we said, whether the legislature placed race above traditional districting considerations in determining which persons were placed in appropriately apportioned districts." Id. at ___, 135 S.Ct. at 1271, 191 L.Ed.2d at 332 (citation, emphasis, and quotation marks omitted). The Supreme Court noted that, if the district court had applied the predominance test correctly, its conclusions may have been different, id. at ___, 135 S.Ct. at 1271, 191 L.Ed.2d at 333, and, hence reconsideration on remand was appropriate.

While the precise issue identified in ALBC is not present in the case before us, the takeaway is relevant: the predominance test — whether "the legislature subordinated traditional race-neutral districting principles... to racial considerations," Miller, 515 U.S. at 916, 115 S.Ct. at 2488, 132 L.Ed.2d at 779-80 — must be applied correctly. The lower court in ALBC balanced race against traditional redistricting principles, but improperly included equal population as a traditional principle, and the Supreme Court took exception to the court's misapplication of the law. Here the trial court skipped the balancing of race against traditional redistricting principles entirely and instead concluded that it could "by-pass this factual inquiry" for the twenty-six VRA districts. The Supreme Court acknowledged that a misapplication of the law can lead to erroneous conclusions and, therefore, remand under those circumstances is necessary to ensure proper application. Remand is also necessary here. But, as I noted in my previous dissent, there is ample evidence that the General Assembly subordinated traditional redistricting principles to racial considerations, and thus, the record contains sufficient evidence to support a predominance finding.

Plaintiffs and amici point to evidence showing that State Senator Robert Rucho and State Representative David Lewis, the respective chairs of the Senate and House Redistricting Committees, instructed Dr. Thomas Brooks Hofeller, the "chief architect" of the redistricting plans, to draw the plans to provide "substantial proportional[ity]" between the percentage of the state's population that is Black and the percentage of districts that would be majority-Black. Dr. Hofeller was also told to "draw a 50% plus one district wherever in the state there is a sufficiently compact black population" to do so. The public statements released by Senator Rucho and Representative Lewis also reflect these legislative goals, saying that, in order to comply with VRA § 2, the VRA districts are designed to provide Black voters with "substantial proportionality" and "must be established with a BVAP of 50% plus one." Because the Supreme Court has held that similar evidence demonstrated that race was the predominant motivating factor, the trial court would have ample justification for determining that strict scrutiny is the appropriate level of review to apply to the Enacted Plans.4

Assuming that the trial court makes a proper predominance finding on remand, the trial court must properly apply the strict scrutiny standard. In its decision the trial court states that, if plaintiffs meet the threshold burden of establishing that "race was the overriding consideration behind a redistricting plan," then

In support of this proposition, the trial court quotes the district court's decision in Shaw II. In Shaw II, however, the Supreme Court reversed the trial court and, in doing so, held that under strict scrutiny, "North Carolina... must show not only that its redistricting plan was in pursuit of a compelling state interest, but also that `its districting legislation is narrowly tailored to achieve [that] compelling interest.'" Shaw II, 517 U.S. at 908, 116 S.Ct. at 1902, 135 L.Ed.2d at 220-21 (alteration in original) (emphasis added) (quoting Miller, 515 U.S. at 920, 115 S.Ct. at 2490, 132 L.Ed.2d at 782). This language from Shaw II clearly places the burden of proof on the State once strict scrutiny is triggered.

This conclusion is bolstered by the Supreme Court's earlier statement in Miller that, "[t]o satisfy strict scrutiny, the State must demonstrate that its districting legislation is narrowly tailored to achieve a compelling interest." 515 U.S. at 920, 115 S.Ct. at 2490, 132 L.Ed.2d at 782 (emphasis added) (citations omitted). More recently, in the affirmative action context, the Supreme Court has been more explicit on this point: Under strict scrutiny, "it remains at all times the [government]'s obligation to demonstrate, and the Judiciary's obligation to determine," that the challenged action is narrowly tailored to achieve a compelling governmental interest. Fisher v. Univ. of Tex. at Austin, ___ U.S. ___, ___, 133 S.Ct. 2411, 2420, 186 L.Ed.2d 474, 486-87 (2013) (emphasis added).

Here, the trial court attempted to distinguish Fisher on the ground that the General Assembly is entitled to some degree of deference given that redistricting is "an inherently political process." The Supreme Court, however, has declined to defer to political decision makers and apply something less than strict scrutiny to race-based classifications:

Johnson v. California, 543 U.S. 499, 512, 125 S.Ct. 1141, 1150, 160 L.Ed.2d 949, 962-63 (2005) (citations omitted); accord Parents Involved in Cmty. Schs. v. Seattle Sch. Dist. No. 1, 551 U.S. 701, 744, 127 S.Ct. 2738, 2766, 168 L.Ed.2d 508, 539 (2007) (plurality) (explaining that "deference is fundamentally at odds with our equal protection jurisprudence" and that courts "put the burden on State actors to demonstrate that their race-based policies are justified" (citations and quotation marks omitted)). Moreover, to whatever extent the legislature may be entitled to deference, that "limited degree of leeway in furthering [its] interests" in complying with the VRA relates to whether the state has met its burden of establishing "the `narrow tailoring' requirement of strict scrutiny." Bush v. Vera, 517 U.S. 952, 977, 116 S.Ct. 1941, 1960, 135 L.Ed.2d 248, 268 (1996) (plurality). Nonetheless, this deference does not relieve the State of "the burden to prove `that the reasons for any [racial] classification [are] clearly identified and unquestionably legitimate.'" Fisher, ___ U.S. at ___, 133 S.Ct. at 2419, 186 L.Ed.2d at 485 (alterations in original) (emphasis added) (quoting City of Richmond v. J.A. Croson Co., 488 U.S. 469, 505, 109 S.Ct. 706, 728, 102 L.Ed.2d 854, 889 (1989)).

Even though the evidence, in my view, provides ample justification for the conclusion made by the trial court — that race predominated — it is important that the trial court do the work, as required by Miller and now ALBC, and properly determine whether the General Assembly subordinated traditional redistricting principles to racial considerations. Moreover, the trial court's misunderstanding and misapplication of the strict scrutiny analytical framework provides additional justification for a decision to vacate the trial court's decision and remand this case to the trial court for reconsideration in light of correct principles. See Fisher, id. at ___, 133 S.Ct. at 2421-22, 186 L.Ed.2d at 487-89 (remanding after determining the trial court and court of appeals misapplied strict scrutiny standard so that challenged admissions policy could be "considered and judged under a correct analysis"). Failure to apply properly the operative constitutional test is, in itself, a sufficient basis for overturning the trial court's decision. See id. at ___, 133 S.Ct. at 2421-22, 186 L.Ed.2d at 487-89. This Court should not forgive the misapplication of the law because, as the Court in ALBC confirmed, a misapplication of the law upon which a conclusion depends, infects the remainder of the analysis.

b.

After the trial court concluded that race predominated in the General Assembly's decision to create twenty-six VRA districts and that strict scrutiny applied, the trial court also side-stepped its compelling state interest analysis with respect to § 2. Specifically, the trial court's findings were insufficient as they relate to determining whether the challenged districts met all three Gingles preconditions. The trial court concluded "that the General Assembly had a strong basis in evidence to conclude that each of the Gingles preconditions was present in substantial portions of North Carolina and that, based upon the totality of circumstances, VRA districts were required to remedy against vote dilution."

At the outset, the trial court's conceptualization of the Gingles preconditions analysis was faulty. The ALBC opinion explains that "[a] racial gerrymandering claim, however, applies to the boundaries of individual districts. It applies district-by-district. It does not apply to a State considered as an undifferentiated `whole.'" ALBC, 575 U.S. at ___, 135 S.Ct. at 1265, 191 L.Ed.2d at 326. The trial court's conclusion that the preconditions were present "in substantial portions of North Carolina" suggests that the trial court did not engage in a district-by-district analysis as required by law. The trial court must make findings of fact addressing the presence of each of the Gingles preconditions in each of the challenged districts.

Moreover, the trial court's findings of fact in the Judgment and Memorandum of Decision and in Appendix A of the Judgment related to the third precondition are deficient. In Thornburg v. Gingles the Supreme Court held that, in order to establish a § 2 voting dilution claim, the minority group must demonstrate that (1) "it is sufficiently large and geographically compact to constitute a majority in a single-member district"; (2) "it is politically cohesive"; and (3) "the white majority votes sufficiently as a bloc to enable it ... usually to defeat the minority's preferred candidate." Thornburg v. Gingles, 478 U.S. 30, 50-51, 106 S.Ct. 2752, 2766-67, 92 L.Ed.2d 25, 46-47 (1986) (citations omitted); see De Grandy, 512 U.S. at 1006, 114 S.Ct. at 2654, 129 L.Ed.2d at 788 (majority) (confirming that vote dilution in the case of single-member districts likewise requires proof of the three preconditions (citing Growe v. Emison, 507 U.S. 25, 40-41, 113 S.Ct. 1075, 1084, 122 L.Ed.2d 388, 403-04 (1993))). Furthermore, when defendants use § 2 as a defense against claims that redistricting plans are unconstitutional, defendants bear the burden of demonstrating the existence of all three mandatory preconditions. Pender County v. Bartlett, 361 N.C. 491, 496, 649 S.E.2d 364, 367 (2007) (Pender County), aff'd sub nom. Bartlett v. Strickland, 556 U.S. 1, 129 S.Ct. 1231, 173 L.Ed.2d 173 (2009) (Strickland) (plurality).

The trial court summarized the evidence relevant to the three preconditions that was before the General Assembly when it enacted the plans. Most of the evidence related to the existence of racially polarized voting in North Carolina. To a certain extent, explaining the evidence of racial polarization was appropriate because, in Gingles the Supreme Court explained:

Gingles, 478 U.S. at 56, 106 S.Ct. at 2769, 92 L.Ed.2d at 50 (emphasis added) (citations omitted). The trial court found as fact that experts and community members had concluded that racially polarized voting exists in the challenged districts. Yet, neither the Judgment and Memorandum of Decision nor Appendix A to the Judgment makes a finding as to whether the majority usually bloc votes to defeat the minority's preferred candidate, the third precondition under Gingles. To the contrary, the third part of Appendix A is dedicated to describing the election results over time in the 2003 version of Senate Districts 14, 20, 21, 28, 38, and 40, the 2009 version of House Districts 12, 21, 29, 31, 48, 99, and 107, and the 2001 version of Congressional Districts 1 and 12.5 The trial court found that each highlighted district was a majority-minority district in its pre-2011 form. The trial court also found that in each of these districts, an African-American Democratic candidate had been successful over a Republican opponent. This is evidence that voters in these majority-minority districts tended to elect African-American Democratic candidates, thereby suggesting that the minority group is politically cohesive in those districts. The findings do not establish, however, that the majority votes in a manner that usually defeats the minority group's preferred candidate of choice in those same districts.

The closest the findings come to demonstrating that the majority bloc votes in a way that usually defeats the minority group's preferred candidate is by recalling that

(Numeral and internal citations omitted.) These generalized findings have limited value. There is no indication that this set of data applies to the challenged VRA districts or reflects voting patterns over a period of time rather than the results of a single election. It is unlikely that the data would apply to the challenged VRA districts because they are all majority-minority districts, and many have been so under previous plans. In addition, generalized findings of fact do not constitute a district-by-district analysis as required by ALBC.

The trial court's findings clearly indicate racially polarized voting, and Supreme Court decisions establish that evidence of racially polarized voting is relevant to the second and third preconditions, see Gingles, 478 U.S. at 56, 106 S.Ct. at 2769, 92 L.Ed.2d at 50; however, the fact remains that Gingles requires a showing of all three preconditions. Without adequate findings of fact related to the third precondition, the trial court's conclusion that all three preconditions have been met is unsupported by its findings. Therefore, I would remand this case to the trial court for adequate findings regarding the third precondition. It is worth noting that when one precondition is not satisfied, the General Assembly is not required to create a § 2 district. See Pender County, 361 N.C. at 499, 649 S.E.2d at 369. If the third precondition is not satisfied with respect to any district in a case such as this in which there is substantial evidence to suggest that race predominated the General Assembly's redistricting decisions for the twenty-six VRA districts, it follows that the General Assembly developed a race conscious redistricting plan that was not justified by § 2 as a compelling state interest.

c.

The trial court's narrow tailoring analysis also misses the mark in other respects. First, because the trial court failed to provide adequate findings of fact related to the third Gingles precondition, the trial court cannot rely on Strickland to conclude that creating VRA districts with a TBVAP greater than 50% was necessary to avoid liability under § 2. Second, ALBC disapproves of the use of mechanical numerical targets to avoid retrogression under § 5. Third, the General Assembly's use of racial proportionality to establish the total number of VRA districts was impermissible under De Grandy. I will discuss each in turn.

Plaintiffs argued at trial that drawing the VRA districts such that each would contain a TBVAP greater than 50%, thereby creating a majority-minority district, could not be narrow tailoring. The trial court rejected plaintiffs' argument and determined that "the General Assembly had a strong basis in evidence for concluding that it was reasonably necessary to endeavor to create all VRA districts within the Enacted Plans with 50% TBVAP to protect the state from anticipated liability under § 2 of the VRA and to ensure preclearance under § 5 of the VRA." The trial court relied upon this Court's decision in Pender County v. Bartlett.

In Pender County this Court considered "whether [the first Gingles] precondition, that a minority group must be `sufficiently large and geographically compact to constitute a majority in a single-member district,' requires that the minority group constitute a numerical majority of the relevant population, or whether a numerous minority can satisfy the precondition." Pender County, 361 N.C. at 499, 649 S.E.2d at 370 (citation omitted). The majority of this Court answered that question by creating a bright line rule: "[A] minority group that otherwise meets the Gingles preconditions [must] constitute a numerical majority of citizens of voting age...." Id. at 504, 649 S.E.2d at 373. The Court reasoned that "[a] bright line rule for the first Gingles precondition `promotes ease of application without distorting the statute or the intent underlying it'" and serves as a "safe harbor" for the General Assembly in the redistricting process. Id. at 505, 649 S.E.2d at 373 (quoting McNeil v. Springfield Park Dist., 851 F.2d 937, 942 (7th Cir.1988), cert. denied, 490 U.S. 1031, 109 S.Ct. 1769, 104 L.Ed.2d 204 (1989)). Affirming the majority's conclusion, the Supreme Court in Strickland reiterated that "the dispositive question is: What size minority group is sufficient to satisfy the first Gingles requirement?" Strickland, 556 U.S. at 12, 129 S.Ct. at 1242, 173 L.Ed.2d at 183 (plurality). The Court reasoned, as this Court did in Pender County, that a majority-minority rule provided "workable standards and sound judicial and legislative administration." Id. at 17, 129 S.Ct. at 1244, 173 L.Ed.2d at 186.

The trial court explained the Strickland holding as follows: "[W]hen the State has a strong basis in evidence to have a reasonable fear of § 2 liability, the State must be afforded the leeway to avail itself of the `bright line rule' and create majority-minority districts, rather than cross-over districts, in those areas where there is a sufficiently large and geographically compact minority population and racial polarization exists." Writing for the plurality in Strickland, Justice Kennedy made the following crucial observations:

Id. at 23-24, 129 S.Ct. at 1248-49, 173 L.Ed.2d at 190-91 (emphases added) (internal citations omitted). Justice Kennedy's qualification of the holding suggests that it is a mistake to interpret Strickland as requiring majority-minority districts in order to comply with § 2 under all circumstances. Taking into consideration the limitations of Strickland's holding, it is not possible to determine whether all the majority-minority districts created by the General Assembly are required in this case because, as explained above, the trial court did not make adequate findings of fact regarding whether the third Gingles precondition was satisfied, necessitating the determination that there are inadequate findings of fact to support the conclusion that each of the VRA districts was justified.

Even assuming that the State had a compelling interest in avoiding liability under VRA § 2 and obtaining preclearance under VRA § 5,6 and assuming that the factors set forth in Thornburg v. Gingles have been properly addressed, the trial court's findings with respect to the greater than 50% TBVAP threshold and proportionality do not support its ultimate conclusion that the redistricting plans pass strict scrutiny. Therefore, this Court should vacate and remand regarding the twenty-six VRA districts.

As explained in ALBC, "Alabama believed that, to avoid retrogression under § 5, it was required to maintain roughly the same black population percentage in existing majority-minority districts." 575 U.S. at ___, 135 S.Ct. at 1263, 191 L.Ed.2d at 324. The Supreme Court rejected this view by establishing that, to avoid "forbidden retrogression," setting fixed percentages or pursuing a "mechanically numerical view" differs significantly from a "more purpose-oriented view," id. at ___, 135 S.Ct. at 1273, 191 L.Ed.2d at 335, which inquires whether "minority voters retain the ability to elect their preferred candidates," id. at ___, 135 S.Ct. at 1273, 191 L.Ed.2d at 334. As the Supreme Court explained:

Id. at ___, 135 S.Ct. at 1274, 191 L.Ed.2d at 336.

Similarly, the North Carolina General Assembly did not ask the right question. The trial court here found that "it is undisputed that the General Assembly intended to create 26 of the challenged districts to be [VRA districts] and that it set about to draw each of these VRA districts so as to include at least 50% [TBVAP]." Although Alabama and North Carolina sought to avoid retrogression differently — by drawing the challenged districts so as to maintain the same percentage of minority voters in Alabama, and to ensure at least 50% TBVAP in North Carolina — both legislatures used a mechanical numerical target in that effort. This is precisely what ALBC forbids.

Under a § 5 retrogression analysis conducted in accordance with ALBC, the trial court must consider to what extent the number of minority voters within each existing majority-minority district must change, if at all. If, within an existing majority-minority district, the minority group is able to elect the candidate of its choice, based on the record of evidence related to voting patterns and other indicia, there are no grounds to increase the minority voter percentages under § 5. Conversely, if the minority group within an existing majority-minority district is no longer able to elect the candidate of its choice because of demographic shifts or other changes, then § 5 requires an increase in the percentage of minority voters within that district to avoid retrogression. Without asking the correct question — "To what extent must we preserve existing minority percentages in order to maintain the minority's present ability to elect the candidate of its choice?" — the trial court authorized the legislature to move minority voters into certain districts based solely on their race without justification.

Finally, the General Assembly endeavored "to create as many VRA districts as needed to achieve a `roughly proportionate' number of Senate, House and Congressional districts as compared to the Black population in North Carolina." The General Assembly reasoned that, because 21% of North Carolina's voting age population identified as any part Black, roughly 21% of the Senate, House, and Congressional seats should be filled by candidates elected by voters in VRA districts, i.e., majority-minority districts. The trial court found that the General Assembly used rough proportionality as a "benchmark" for the number of VRA districts it would create, and the court concluded that this methodology was appropriate to avoid any potential § 2 liability. But as I noted in my previous dissent, this conclusion is based on a misapprehension of De Grandy.

In De Grandy the State of Florida argued "that as a matter of law no dilution occurs whenever the percentage of single-member districts in which minority voters form an effective majority mirrors the minority voters' percentage of the relevant population." 512 U.S. at 1017, 114 S.Ct. at 2660, 129 L.Ed.2d at 795. The Supreme Court rejected this safe harbor rule because such a rule "would be in derogation of the statutory text and its considered purpose, however, and of the ideal that the Voting Rights Act of 1965 attempts to foster. An inflexible rule would run counter to the textual command of § 2, that the presence or absence of a violation be assessed `based on the totality of circumstances.'" Id. at 1018, 114 S.Ct. at 2660, 129 L.Ed.2d at 795 (citation omitted). In addition, such a rule would lead to a tendency on the part of the State to create majority-minority districts even when they may not be necessary to achieve equal opportunity for a minority voter to elect his or her preferred candidate. Id. at 1019-20, 114 S.Ct. at 2661, 129 L.Ed.2d at 796.

Here, however, defendants' public statements undermine their adherence to the applicable standards and demonstrate the central role proportionality played in the 2011 redistricting plan. On 17 June 2011, defendants announced a public hearing concerning redistricting issues, in which defendants expressed the intention to propose redistricting plans containing a sufficient number of majority-minority districts to provide substantial proportionality. Defendants proposed "that each plan include a sufficient number of majority African American districts to provide North Carolina's African American citizens with a substantially proportional and equal opportunity to elect their preferred candidate of choice." Defendants explained that "proportionality for the African American citizens in North Carolina means the creation of 24 majority African American House districts and 10 majority Senate districts.... Unlike the 2003 benchmark plans, the Chairs' proposed 2011 plans will provide substantial proportionality for North Carolina's African American citizens."

In light of its misreading of De Grandy, the trial court cites approvingly defendants' use of proportionality as the "benchmark" for creating the Enacted Plans — beginning with proportionality as the goal and then working backwards to achieve that goal. Similarly, the trial court reasoned: "When the Supreme Court says `no violation of § 2 can be found' under certain circumstances, prudence dictates that the General Assembly should be given the leeway to seek to emulate those circumstances in its Enacted Plans." (Quoting De Grandy, 512 U.S. at 1000, 114 S.Ct. at 2651, 129 L.Ed.2d at 784.) But this approach is precisely what the Supreme Court rejected in De Grandy: proportionality is relevant as a means to an end (compliance with the VRA), but it is not an end in itself and it does not — contrary to the trial court's reasoning — provide a safe harbor for redistricting plans premised on race. The Court in De Grandy centered its analysis on the role of proportionality in the totality-of-the-circumstances analysis in a § 2 claim. See id. at 1013-14, 114 S.Ct. at 2658, 129 L.Ed.2d at 792-93 ("The court failed to ask whether the totality of facts, including those pointing to proportionality, showed that the new scheme would deny minority voters equal political opportunity." (footnote call number omitted)). The Court made clear that equal political opportunity is the focus of the inquiry, not proportionality. By not focusing on whether the Enacted Plans denied minority voters equal political and electoral opportunity, the trial court misapplied the law, resulting in an erroneous conclusion that defendants' use of proportionality as an end is constitutionally permissible.

The majority concludes that "the record here demonstrates that the General Assembly did not use proportionality improperly to guarantee the number of majority-minority voting districts based on the minority members' share of the relevant population." The majority is only able to draw this conclusion by overlooking the trial court's determination — based upon "the undisputed evidence" — that the General Assembly used proportionality as a "benchmark." The majority's conclusion becomes more confusing given its statement that "[w]e believe that such an effort, seeking to guarantee proportional representation, proportional success, or racial balancing, would run afoul of the Equal Protection Clause." (Citing De Grandy, 512 U.S. at 1017-22, 114 S.Ct. at 2658-62, 129 L.Ed.2d at 794-98.) I agree "that such an effort ... would run afoul of the Equal Protection Clause," and its use in this instance has that effect.

By characterizing the General Assembly's consideration of race as a "precautionary consideration" used "as a means of protecting the redistricting plans from potential legal challenges under section 2's totality of the circumstances test," the majority appears to join the trial court in using race as a legislative safe harbor in derogation of the clear prohibition against reliance upon such criteria set forth by the Supreme Court of the United States. De Grandy, 512 U.S. at 1018-20, 114 S.Ct. at 2660-61, 129 L.Ed.2d at 795-97. In light of these errors, this Court should vacate the trial court's Judgment and remand the case for reconsideration under a correct understanding of the law.

Based on the foregoing, I would remand this case to the trial court for more complete findings of fact on the VRA districts with respect to whether the General Assembly subordinated traditional redistricting principles to racial considerations, with respect to the third Gingles precondition, and with respect to the proper application of the non-retrogression requirement. Even if race predominated the General Assembly's motivations and §§ 2 and 5 constitute compelling state interests, the trial court's findings of fact do not suffice to support a conclusion that the Enacted Plans were narrowly tailored to achieve those interests and did not violate the Equal Protection Clause of the Fourteenth Amendment.

On remand, the trial court should begin by determining how many majority-minority districts, if any, need to be created and where those districts should be located in order to comply with § 2 and § 5. In addition, after determining the total number of majority-minority districts needed to comply with the VRA, the trial court should determine, where necessary, the percentages of TBVAP in each district needed to ensure the minority group's present ability to elect its candidate of choice and to avoid retrogression. In answering these questions on remand, the trial court should engage in the following district-by-district analysis in accordance with the directives provided in ALBC and existing law.

As to the former inquiry, the trial court must look to § 2 and consider whether defendants have established the existence of the three mandatory Gingles preconditions for each majority-minority district that defendants created, and if so, whether the totality of the circumstances establishes that each of these districts was required by § 2 (unless the creation of that district was mandated by a prior court order that remains in effect7). Again, proportionality may be considered in the totality-of-the-circumstances determination but must not be the starting point for a redistricting plan. To the extent that defendants fail to establish that any of the majority-minority districts were required by § 2 based on the existence of the three mandatory Gingles factors and the totality of the circumstances, or a valid court order, the creation of such a majority-minority district is not justified.

As to the latter inquiry, the trial court must look to § 5 and determine on a district-by-district basis which actions, if any, were necessary to ensure non-retrogression in any district created pursuant to § 5. In doing so, the court should look to the districts as they existed under the prior redistricting plan to discern the TBVAP of each district and whether the minority group had the ability to elect its candidate of choice in that district. In the event the answer to that question is in the negative, then the court needs to determine what TBVAP is needed to permit the election of the minority group's candidate of choice. In the event that the answer to that question is in the affirmative, then the court must determine whether defendants impermissibly increased the TBVAP in that district.

II.

Plaintiffs also challenged four non-VRA districts, which are districts with a TBVAP of less than 50% — Congressional Districts 12 and 4, Senate District 32, and House District 54. The discussion in ALBC concerning the proper analysis for determining whether race was the predominant motivating factor in drawing the districts supports the conclusion that the trial court viewed equalizing population among the districts as a traditional redistricting principle rather than as "part of the redistricting background, taken as a given." 575 U.S. at ___, 135 S.Ct. at 1270, 191 L.Ed.2d at 332.

The Court in ALBC was clear in its instruction that "an equal population goal is not one factor among others to be weighed against the use of race to determine whether race `predominates.' Rather, it is part of the redistricting background, taken as a given, when determining whether race, or other factors, predominate in a legislator's determination as to how equal population objectives will be met." Id. at ___, 135 S.Ct. at 1270, 191 L.Ed.2d at 332. The majority, in its attempt to show that the principles articulated in ALBC do not apply to the case at bar, reasons that, "[i]n effect, North Carolina's Whole County Provision, of which equal population is a component, establishes a framework to address the neutral redistricting requirement that `political subdivisions' be respected." In its framing of the Whole County Provision, the majority essentially concludes that equal population is a traditional redistricting principle in North Carolina and, in the process, ignores the Supreme Court's holding that equal population is not a traditional redistricting principle to be weighed among other such principles.

In concluding that the non-VRA districts were not drawn with race as the predominant motivation, the trial court explained:

(Emphases added.) This statement reflects the trial court's understanding that equalizing population was as relevant to its predominance analysis as other legislative motivations, including compliance with the Whole County Provisions, protection of incumbents, and creating districts in which Republicans would be more competitive. The trial court's more specific findings of fact in Appendix B of the Judgment related to Congressional Districts 4 and 12 provide further evidence of the trial court's view of equal population as a factor to be weighed. The trial court found that, in Congressional District 4, "[a]ll of the divisions were done to equalize population among the Fourth Congressional District and the adjoining Congressional districts, to make the district contiguous, or for political reasons. None of the [Vote Tabulation Districts] were divided based upon racial data." Similarly, the trial court found that, in the Twelfth Congressional District, "[a]ll of [the] divisions were done to equalize population among the Twelfth Congressional District and other districts or for political reasons." If the trial court were to remove equalizing population as a traditional redistricting factor in accordance with ALBC, it would be left to consider only whether these two factors were subordinated to racial considerations: making the Fourth Congressional District contiguous and political considerations. In the case of the Twelfth Congressional District, the trial court would be left with only political reasons to weigh against race. This balance is particularly troublesome in the case of Congressional District 12.

The shape of Congressional District 12 has been the subject of much litigation over the last two decades, and for good reason. See, e.g., Easley v. Cromartie, 532 U.S. 234, 121 S.Ct. 1452, 149 L.Ed.2d 430 (2001) (Cromartie II); Hunt v. Cromartie, 526 U.S. 541, 119 S.Ct. 1545, 143 L.Ed.2d 731 (1999); Shaw II, 517 U.S. 899, 116 S.Ct. 1894, 135 L.Ed.2d 207; Shaw I, 509 U.S. 630, 113 S.Ct. 2816, 125 L.Ed.2d 511. In Cromartie II the Supreme Court observed that "racial identification correlates highly with political affiliation" in North Carolina, 532 U.S. at 258, 121 S.Ct. at 1466, 149 L.Ed.2d at 453, and that the plaintiffs in that case "ha[d] not successfully shown that race, rather than politics, predominantly account[ed] for" the shape, location, and composition of the 1997 version of Congressional District 12, id. at 257, 121 S.Ct. at 1466, 149 L.Ed.2d at 453. Here the trial court found as fact that Congressional District 12 (and District 4) were drawn based on the locations where President Obama received the highest voter totals during the 2008 presidential election. Even if these voter totals were "[t]he only information on the computer screen," we cannot ignore the fact that race played an extraordinary role in that election. See Bob Hall, 2008 Recap: Same-Day Registration & Other Successes, Democracy North Carolina (Dec. 26, 2008, updated Mar. 19, 2009), http://www.democracync.org/downloads/WrapUpYearofVoterPR2008.pdf ("[I]n 2008, a record 72% of registered blacks voted, which surpassed the rate of whites (69%) for the first time.... That record level of participation proved crucial for many candidates, beginning with Obama."). To justify this serpentine district, which follows the I-85 corridor between Mecklenburg and Guilford Counties, on partisan grounds allows political affiliation to serve as a proxy for race and effectively creates a "magic words" test for use in evaluating the lawfulness of this district. Because race and politics historically have been and currently remain intertwined in North Carolina, the record contains evidence tending to suggest that politics are no more than a pretext for this excruciatingly contorted district. Therefore, given the inadequacy of its findings of fact, the trial court erred by concluding that "the shape, location and composition of [this district] ... included equally dominant legislative motivations ... to protect incumbents[] and to ... enact redistricting plans that were more competitive for Republican candidates." Upholding this district's tortured construction creates an incentive for legislators to stay "on script" and avoid mentioning race on the record, and in this instance, it is disingenuous to suggest that race is not the predominant factor. As such, this Court should vacate and remand as to Congressional District 12.

With respect to Senate District 32, plaintiffs contend that the trial court's findings actually undermine its conclusion that strict scrutiny does not apply because the non-VRA districts are not race-based. The trial court found the following relevant facts:

As discussed above, the Supreme Court has held that when redistricting plans drawn in an attempt to preempt VRA § 2 litigation or obtain VRA § 5 preclearance are predominantly race-based, such plans attract strict scrutiny. See Vera, 517 U.S. at 959, 116 S.Ct. at 1952, 135 L.Ed.2d at 257; Shaw II, 517 U.S. at 906, 116 S.Ct. at 1901, 135 L.Ed.2d at 219; Miller, 515 U.S. at 920, 115 S.Ct. at 2490, 132 L.Ed.2d at 782.

The trial court acknowledged that compliance with the VRA was a motivating factor behind the enacted plans, but concluded that "comply[ing] with the Whole County Provision,... equaliz[ing] population among the districts, ... protect[ing] incumbents, and... satisfy[ing] the General Assembly's desire to enact redistricting plans that were more competitive for Republican candidates" were "equally dominant legislative motivations." But, in the section of its fact-finding Judgment addressing Senate District 32, the trial court made no findings regarding these other considerations. While the evidence might support such a conclusion, the trial court's actual findings do not. Accordingly, this Court should vacate the trial court's Judgment and remand this case to the trial court to address whether race was the predominant motivation behind the shape, location, and composition of Senate District 32.

With respect to House District 54 and Congressional District 4, the trial court also found that race was not the predominant motivating factor. Plaintiffs do not contest these determinations, and they are binding on appeal. Koufman v. Koufman, 330 N.C. 93, 97, 408 S.E.2d 729, 731 (1991). As stated above, however, because the shapes and compositions of the four non-VRA districts are necessarily affected by the VRA districts, it would be impossible to vacate and remand piecemeal.

Because I conclude that the issue of whether race was the predominant motivating factor in drawing the non-VRA districts should be remanded to the trial court for more complete findings of fact taking into account the guidance provided by ALBC, I do not find it necessary to address the trial court's application of the rational basis test or the majority's approval of it.

III.

Plaintiffs contend that the trial court erred in concluding that the enacted House and Senate plans do not violate the provisions of the state constitution, which dictate that "[n]o county shall be divided in the formation of a senate district," N.C. Const. art. II, § 3(3), and "[n]o county shall be divided in the formation of a representative district," id. § 5(3). In Stephenson v. Bartlett, 355 N.C. 354, 562 S.E.2d 377 (2002) (Stephenson I), this Court construed the Whole County Provisions in light of federal law and "mandated that in creating legislative districts, counties shall not be divided except to the extent necessary to comply with federal law, including the `one-person, one-vote' principle and the VRA." Stephenson v. Bartlett, 357 N.C. 301, 309, 582 S.E.2d 247, 251-52 (2003) (Stephenson II) (citing Stephenson I, 355 N.C. at 363-64, 562 S.E.2d at 384-85). To ensure complete compliance with federal law and to provide maximum enforcement of the Whole County Provisions, this Court "out-lined in Stephenson I the following requirements that must be present in any constitutionally valid redistricting plan":

Stephenson II, 357 N.C. at 305-07, 582 S.E.2d at 250-51 (alterations in original) (emphases added) (quoting Stephenson I, 355 N.C. at 383-84, 562 S.E.2d at 396-97).

The majority concludes that its analysis of the Enacted Plans under the Whole County Provisions remains "unaffected" by ALBC. Yet, a Whole County Provisions analysis conducted in accordance with the framework set forth in Stephenson I, requires application of the nine rules listed above, the first of which is premised on designing any legislative districts required by the VRA. Stephenson I, 355 N.C. at 383, 562 S.E.2d at 396-97. One cannot sever compliance with the VRA from compliance with the Whole County Provisions. Because ALBC provides legal principles that must be applied in determining the constitutionality of VRA districts, ALBC affects an analysis under the Whole County Provisions, albeit indirectly.

In view of the necessity for a remand to the trial court to address the equal protection claim, the trial court must also address the Whole County Provisions issue on remand given that the General Assembly, in attempting to comply with Stephenson I's Rule 1, drew the VRA districts before applying Rules 2 through 9. Because I conclude that the trial court's findings of fact fail to establish that the VRA districts are constitutional, the trial court must, after making a valid determination relating to the VRA districts, and to the extent necessary, revisit the Whole County Provisions issues as well. Simply put, to the extent that the VRA districts are unconstitutional, that fact would necessarily affect the result reached with respect to the General Assembly's application of the rubric set forth in Stephenson I. See Pender County, 361 N.C. at 508-09, 649 S.E.2d at 375 (concluding that a house district created with the intent to comply with VRA § 2 was not required by the VRA and thus, "must be drawn in accordance with the WCP and the Stephenson I requirements"). As such, I would vacate and remand on this issue.

IV.

When addressing the parties' arguments in support of and in opposition to the Enacted Plans, we cannot lose sight of the purpose of the VRA. The House Report accompanying the original Voting Rights Act of 1965 noted:

H.R.Rep. No. 89-439, at 8 (1965) (titled "Voting Rights Act of 1965"), as reprinted in 1965 U.S.C.C.A.N. 2437, 2439-40 (citations omitted). In Gingles the Supreme Court noted the "historical pattern of statewide official discrimination" in North Carolina.9 478 U.S. at 39, 106 S.Ct. at 2760, 92 L.Ed.2d at 39. The VRA was intended to remove barriers to enfranchisement and ensure that new barriers did not arise in their place; however, "[s]ince the adoption of the Voting Rights Act [some] jurisdictions have substantially moved from direct, over[t] impediments to the right to vote to more sophisticated devices that dilute minority voting strength." De Grandy, 512 U.S. at 1018, 114 S.Ct. at 2660, 129 L.Ed.2d at 795 (second and third alterations in original) (citation and internal quotation marks omitted). Although those new barriers may not look the same as the old barriers, it is this Court's duty to identify and invalidate policies or tactics that effectively impede the minority group's ability to elect its candidate of choice in compliance with the VRA and the Equal Protection Clause.

Here, even if the legislature considered traditional redistricting principles, such as compactness and the protection of incumbents or other political motivations, the fact remains that the General Assembly started with the premise that African-American voters in North Carolina should only be guaranteed the opportunity to elect candidates of their choice to 21% of the seats in each chamber. From there, the legislature worked backwards to avoid liability under § 2 and ensure preclearance under § 5. The implications of such a premise reach beyond the challenged VRA districts, affecting the non-VRA districts as well.

When the legislature purposely carves out majority-minority districts, increasing or decreasing the TBVAP by a few percentage points while maintaining a greater than 50% TBVAP, the district-drawing process necessarily requires the identification of voters by race and the movement of the district lines to incorporate or exclude those voters accordingly. This scheme compels the question: Is the ability of the minority voters who are suddenly no longer represented by their preferred candidate of choice in a VRA district unimportant? If the only way to ensure that the minority group has the ability to elect the candidate of its choice is to create majority-minority districts, the General Assembly has the power to determine which of the voters in the minority group will be represented by the candidate of their choice, and which voters will not.

Writing separately in De Grandy, Justice Kennedy warned:

Id. at 1029, 114 S.Ct. at 2666, 129 L.Ed.2d at 802 (Kennedy, J., concurring) (internal citation omitted). In my view, the trial court's decision to uphold the Enacted Plans in the absence of adequate findings demonstrates that Justice Kennedy's concerns may be wellfounded.

For all the complexity of VRA jurisprudence, the bottom line is that the manipulation of district lines based on race to a greater extent than necessary to comply with the VRA is unconstitutional. The record in this case contains evidence tending to show that the General Assembly used numerical targets formulated by racial considerations to avoid liability under § 2 and ensure preclearance under § 5 without fully considering whether the decisions made were necessary to enable the minority group to elect its preferred candidate of choice in compliance with the VRA.10 Any such scheme would be unconstitutional. The trial court's findings are not adequate to support a conclusion that this unconstitutional scheme did not occur. Any impermissibly racially gerrymandered districts affect the entire state under the Whole County Provisions of the North Carolina Constitution. For any of these errors, this Court would do well to vacate and remand rather than prematurely affirm a defective districting plan.

Accordingly, I concur in that part of the majority's opinion regarding plaintiffs' remaining state claims related to the "Good of the Whole" Clause in Article I, Section 2 of the Constitution of North Carolina, and respectfully dissent from those parts of the opinion affirming the trial court's erroneous judgment.

Justices HUDSON and ERVIN join in this opinion.