Chief Justice BENDER delivered the Opinion of the Court.
¶ 1 This case involves the redistricting of Colorado's congressional districts following the results of the 2010 census. Redistricting is an incredibly complex and difficult process that is fraught with political ramifications and high emotions. This process is made all the more complicated by the depth and variety of Colorado's local and regional interests. In many ways, as the trial court observed, Colorado serves as a microcosm of our diverse and great nation, with its rich diversity of cultures and ethnicities, its eclectic and sharply-contrasting geographic features, its broad economic and recreational pursuits, and its combination of rural and urban populations. While this diversity is what makes Colorado great, each of these unique interests merely adds to the complexity and contentiousness of redistricting.
¶ 2 Judicial redistricting is truly an "unwelcome obligation." Connor v. Finch, 431 U.S. 407, 415, 97 S.Ct. 1828, 52 L.Ed.2d 465 (1977). Nevertheless, once it becomes clear that the General Assembly is unable or unwilling to amend constitutionally infirm boundaries, the task must necessarily be completed by the judiciary. See Beauprez v. Avalos, 42 P.3d 642, 648-49 (Colo.2002).
¶ 4 The 2010 census revealed that Colorado continues to be entitled to seven representatives in the United States House of Representatives but that the existing congressional districts are now of unequal population size due to population growth and movement since the last census. Thus, the old districts are malapportioned in violation of the one person, one vote principle mandated by Article I, Section 2 of the United States Constitution. See Reynolds v. Sims, 377 U.S. 533, 568, 84 S.Ct. 1362, 12 L.Ed.2d 506 (1964). Under our Constitution, the General Assembly was responsible for redrawing the state's congressional districts during the 2011 legislative session, but the legislature was unable to pass a new redistricting scheme.
¶ 5 Consequently, a number of plaintiffs sued the Secretary of State in Denver District Court ("the trial court") to enjoin the use of the existing, malapportioned congressional districts and to replace them with new districts of equal population that would satisfy all of the constitutional and statutory criteria. Although this action was proper under the procedure we have previously set forth, see Beauprez, 42 P.3d 642, we acknowledge that this forces the apolitical judiciary to engage in an inherently political undertaking. See Gaffney v. Cummings, 412 U.S. 735, 753-54, 93 S.Ct. 2321, 37 L.Ed.2d 298 (1973). On November 10, 2011, the trial court issued an order declaring the current congressional districts unconstitutional. The order also discussed in exhaustive detail how the map first proposed and later amended by the Moreno plaintiffs ("the Moreno/South Map") satisfied all mandatory and discretionary redistricting criteria. Hence, the trial court adopted the Moreno/South Map and ordered the Secretary of State to implement it in future congressional elections.
¶ 6 Petitioners here, the Hall Plaintiffs joined by Douglas County as an intervenor, appealed to the court of appeals, but asked us pursuant to C.A.R. 50 to issue a writ of certiorari before argument and judgment in that court. Due to the importance and time sensitive nature of this issue, we granted this request and ordered briefing and oral argument on an expedited schedule. The sole issue for our review is whether the adopted map satisfies all constitutional and statutory criteria.
¶ 7 On December 5, 2011, we issued an order and mandate affirming the trial court's adoption of the Moreno/South Map. We did so before issuing an opinion to give the appropriate officials adequate time to prepare for the upcoming elections. We stated that a written opinion would follow in the near future. This is that opinion.
¶ 8 The 2010 census revealed that Colorado's existing congressional districts were malapportioned in violation of Article I, Section 2 of the U.S. Constitution. See Wesberry, 376 U.S. at 18, 84 S.Ct. 526. Nevertheless,
¶ 9 As a result, in May 2011, plaintiffs Dominick Moreno, Christine Le Lait, William Patterson, Rita Mahoney, Roger Clark, Kristi Matsunaka, and Mikel Whitney (the "Moreno Plaintiffs")
¶ 10 The City of Aurora and Aurora Mayor Edward Tauer ("Aurora"), the Douglas County Board of County Commissioners ("Douglas County"), the Colorado Latino Forum and the Colorado Hispanic Bar Association (collectively, CLF/CHBA), and Bill Thiebaut intervened. The court also accepted and considered the amicus briefs of ten additional parties: Liane McFayden, El Paso County, Gerald Harrison, Larimer County, Weld County, Jefferson County, Clean Water Action, Shaffer for Colorado, Club 20, and Padres y Jóvenes Unidos. Club 20, Weld County, and Jefferson County also filed proposed maps along with their amicus briefs. Prior to trial, the Moreno Plaintiffs moved to strike these proposed maps, arguing that an amicus curiae cannot submit proposed maps without becoming an intervenor. However, the trial court accepted these proposed maps through the Hall Plaintiffs, who evidently adopted them as their own exhibits at a status conference prior to the trial.
¶ 11 Trial commenced on October 11, 2011, and spanned ten days. The court heard testimony from thirty-two witnesses (including five experts and five current U.S. Representatives from Colorado) and admitted over 1,000 pages of exhibits as evidence—including more than ten proposed maps. Seven different parties were represented at trial: the Secretary of State,
¶ 12 The trial court was flexible with all parties' witnesses throughout the trial court proceedings. Several witnesses were permitted to testify out of order to accommodate their schedules. The court agreed to permit witnesses to testify by phone if they were unable to be present, so that no party had to forego a witness due to scheduling issues.
¶ 13 The court demonstrated a conscious effort to be as inclusive as possible. Prior to the trial, the court permitted any party who so desired to intervene in the case and also permitted any party to file an amicus brief. During the trial, the court permitted all proposed testimony and the vast majority of objections were overruled. Virtually no exhibits were excluded. In addition, the court permitted the parties to submit final revised maps after closing arguments, as well as amended proposed findings of fact based on other parties' final submitted maps.
¶ 14 The trial process followed by the court to adopt a redistricting plan was unquestionably "thorough, inclusive, and non-partisan."
¶ 15 Paralleling the complexity and depth of the parties' positions and the endeavor presented, the trial court issued a detailed seventy-seven-page order articulating its findings of fact and conclusions of law, and ultimately adopting the Moreno/South Map as a lawful judicial remedy to the malapportioned holdover map.
¶ 16 In its order, the trial court acknowledged that the redistricting process is difficult and fraught with a variety of divergent interests. It stated that redrawing any district lines necessarily means disappointing citizens and interest groups, no matter how the lines are drawn. The court also stated that despite the accelerated nature of the proceedings, its decision was based on a full deliberation and consideration of all issues and that it had a full opportunity to review all of the evidence and testimony in the case, all written submissions, and all relevant jurisprudence. Finally, the trial court emphasized that no partisan or political position or consideration played a role in its task or ultimate decision.
¶ 17 The trial court accurately set forth the constitutional and non-constitutional criteria for congressional redistricting.
¶ 18 After making extensive findings of fact and reviewing the appropriate criteria for consideration, the court reviewed all of the submitted maps. The court found that the Moreno/South Map best met the constitutional and non-constitutional criteria for drawing congressional districts.
¶ 19 The trial court found that the Hall Minimum Disruption map was drawn without any regard to communities of interest based upon the testimony of the Hall Plaintiffs' map drawer stating that was the case. In addition, the court found that the Hall Plaintiffs' map relied on outdated planning and management regional maps that were unreliable for current redistricting purposes because they did not reflect changes in the state that have occurred since they were created.
¶ 20 The court ruled that the map's failure to consider communities of interest made it impossible to adopt. The court reasoned that adopting a map created using the "minimum disruption" factor alone is "fundamentally at odds with the multidimensional task" confronting the court.
¶ 21 As an example of the inherent flaw in the Hall Plaintiffs' approach of refusing to consider communities of interest, the court cited the lack of evidence of any community-of-interest rationale for the retention of existing boundary lines for Congressional District 2 (CD2), particularly in light of the
¶ 22 The Hall Aurora Map, which was proposed at closing arguments, made two significant changes to the Hall Minimum Disruption Map: it added Hispanic voters to Congressional District 1 (CD1) and united Aurora in Congressional District 7 (CD7). In addition to its failure to consider communities of interest similar to the Hall Minimum Disruption Map, the court found that the supplemental Hall Aurora Map placed incumbents from existing Congressional District 6 (CD6) and CD7 into the map's proposed CD7, something previous legislatures and courts have sought to avoid in congressional redistricting. The court also found that the map exacerbated the compactness problems of the existing CD7 by snaking from Jefferson County into Adams County via a frontage road, making the proposed district markedly non-compact with a truly odd shape. This map also included significant crop-producing areas in Arapahoe County in CD7 that the court reasoned should be linked with the agricultural lands of CD4. Additionally, the court held that the map's augmentation of CD1's Hispanic population by less than 3 percent would not have a "meaningful impact on minority voting strength." See Carstens v. Lamm, 543 F.Supp. 68, 86 (D.Colo.1982). For these reasons, the court rejected the Hall Aurora Map and held that it was "too flawed to be serviceable."
¶ 23 The court ruled that the Jefferson County Map suffered the same shortcomings as the Hall Minimum Disruption Map—a natural result because the Jefferson County Map was merely the Hall Minimum Disruption Map altered to keep Jefferson County whole. The court further found that no evidentiary basis was established as to why Jefferson County should be unified at the expense of the many other political subdivisions and communities of interest that the map divided. Accordingly, the court rejected this map as an unreasonable redistricting scheme.
¶ 24 The trial court found the consensus of the counties purportedly represented by Club 20 suspect. The evidence at trial showed that Club 20 is not an official organization of local governments but instead a lobbying group to which any person or entity can belong. For this reason, and because the decision-making process by which the map was endorsed by Club 20 was not substantiated, the court found that it could not conclude that the Club 20 Map represents any official consensus of any of these county groupings. The court noted that the board of Club 20 voted only on its policy to keep the Western Slope whole and not on the exact Club 20 Minimum Disruption Map submitted to the court. The court also found that Summit, Lake, Eagle, San Miguel, and Gunnison counties are not members of Club 20 and that the voting member of Club 20 from Summit County is a small business owner who represents herself individually and not Summit County. Hence, the court did not find the Club 20 Map to be an official representation of the views of the counties in the Western Slope. In addition, the court found no justification in the evidence presented at trial for the change in this map, also created by altering the Hall Minimum Disruption Map, to include Grand County but not Lake, Eagle, or Summit counties, in Congressional District 3 (CD3). Therefore, the
¶ 25 The trial court found that the CLF/CHBA maps sought to create Hispanic "influence districts" by aggregating various cities and counties so that populations of persons of Hispanic origin comprised 30 percent or more of the population in CD1, CD4, and CD7. The court acknowledged that Hispanics in Colorado have experienced discrimination and explicitly recognized the importance of the Latino community in the redistricting process. However, the court found that race was the predominant consideration in the drawing of the CLF/CHBA maps, creating a significant concern as to the constitutionality of the maps. See Hunt v. Cromartie, 526 U.S. 541, 546, 119 S.Ct. 1545, 143 L.Ed.2d 731 (1999) ("[A]ll laws that classify citizens on the basis of race, including racially gerrymandered districting schemes, are constitutionally suspect and must be strictly scrutinized.").
¶ 26 In addition to this potential constitutional infirmity, the court found that the CLF/CHBA maps actually lowered the Hispanic population in CD3, diluting the Hispanic influence in that district by separating the San Luis Valley from the growing Hispanic population on the Western Slope in Lake and Garfield counties.
¶ 27 The court found that the CLF/CHBA's proposal to move Pueblo and the San Luis Valley to CD4 unjustified. The court reasoned that there was no community of interest between the Hispanic populations in the San Luis Valley and Pueblo and the Eastern Plains counties of Lincoln, Elbert, Kit Carson, Cheyenne, and Kiowa. In addition, the court found that the CLF/CHBA's argument that the San Luis Valley was linked to the Eastern Plains due to water shortages in the Rio Grande River less than compelling because the San Luis Valley has recently enacted new groundwater management rules in an attempt to mitigate the impacts of its water scarcity issues. Hence, the court found that the shortage issues in the San Luis Valley were factually distinct from those faced in the South Platte and Arkansas basins of the Eastern Plains. The court also rejected the CLF/CHBA maps' attempt to link Pueblo County with the Eastern Plains based on agriculture, because agriculture has only a marginal influence in Pueblo County.
¶ 28 Further, the court found the CLF/CHBA's "nesting" approach problematic. Under the nesting approach, the maps were drawn by combining five state senate districts developed by the 2011 Colorado Reapportionment Commission. The court rejected this approach because the state reapportionment criteria and policies are different from the redistricting criteria and because the Commission's map had not been approved by this court at the time that the CLF/CHBA maps were submitted.
¶ 29 In addition, the court found that the maps failed to reflect changes in all parts of the state and instead only adjusted the three districts of greatest concern to the CLF/CHBA. Finally, the court found that the maps were not contiguous and split an inordinate number of political jurisdictions, such as Larimer County, which it split into three districts. For these reasons, the court rejected the CLF/CHBA maps.
¶ 30 While the trial court found the Thiebaut Maps' attempts to maximize competitiveness admirable, the court held that competitiveness could not serve as the sole basis for drawing district lines. Furthermore, the court found that the maps split too many municipalities and counties and that not all of the districts were sufficiently compact. Critically, the court also found that the Thiebaut Maps were the only maps submitted to the court that did not achieve numeric equality. Therefore, the court determined that it could not utilize these maps.
¶ 31 As discussed in detail below, the trial court found that there was an "abundant evidentiary basis" supporting the Moreno/South Map and that the Moreno/South Map best met the requirements for drawing congressional districts. The court held that
¶ 32 We granted certiorari to review the trial court's order and adoption of the Moreno/South Map at the request of both the Hall Plaintiffs and Douglas County. Before addressing the specifics of the adopted map, we must first set forth the applicable law that guides our present review.
¶ 33 At the outset of our analysis, we pause to explain the nature of our present inquiry. Because an apparent legislative impasse resulted in the General Assembly's failure to adopt a new map, the trial court was tasked with adopting a new map to remedy the unconstitutionality of the malapportioned holdover congressional boundaries. In order to determine a lawful remedy, the trial court must consider the evidence presented by the parties, hear and make credibility determinations with respect to the expert and lay testimony before it, consider all proposed plans, and ultimately adopt a legal redistricting scheme. See generally Beauprez, 42 P.3d 642. As chronicled above, in the present matter, the trial court completed this task through a "thorough, inclusive, and non-partisan" process. See id. at 647.
¶ 34 Unlike the trial court, it is not our role to draw a map or to pick a new map from the many that the parties submitted. Rather, at this stage, our sole task is to review the remedy selected by the trial court to determine whether it is lawful. Id. at 651-52.
¶ 35 Pursuant to section 2-1-102, C.R.S. (2011), as amended in 2010, the General Assembly has set forth a number of factors to guide the trial court in its adoption of a redistricting scheme. Thus, it is under this statute that we review the lawfulness of the adopted map. These factors can be separated into two distinct categories, each with its own standard of review: constitutional criteria and non-constitutional criteria.
¶ 36 Because the first two criteria in section 2-1-102 derive from the U.S. Constitution, they necessarily exist independent of our statute and take precedence over all other considerations. Nevertheless, the General Assembly has explicitly codified these two constitutional mandates to ensure that any adopted redistricting scheme will be constitutional. See §§ 2-1-102(1)(a)(I)-(II).
¶ 37 First, subsection 2-1-102(1)(a)(I) requires that an adopted plan evince "[a] good faith effort to achieve precise mathematical population equality between districts." Further, an adopted map must justify "each variance, no matter how small, as required by the constitution of the United States," and "[e]ach district shall consist of contiguous whole general election precincts" that do not overlap. This iron-clad standard derives from Article I, Section 2 of the U.S. Constitution, which the Supreme Court has interpreted as "our Constitution's plain objective of making equal representation for equal numbers of people the fundamental goal for the House of Representatives." Wesberry, 376 U.S. at 18, 84 S.Ct. 526.
¶ 38 Second, subsection 2-1-102(1)(a)(II) requires that a redistricting scheme be in "[compliance] with the federal `Voting Rights Act of 1965', in particular 42 U.S.C. sec. 1973...." In order to ensure that all citizens have equal and open access to participate in elections, section 1973 of the federal Voting Rights Act prohibits any and all race-based voting discrimination, including the adoption of districts that result in race-based voter dilution. This federal statute stems directly from the Fourteenth Amendment, which guarantees equal protection, and the Fifteenth Amendment, which forbids the denial of the right to vote on the basis of race. See Chisom v. Roemer, 501 U.S. 380,
¶ 39 In the present appeal, neither appellant claims that the Moreno/South Map adopted by the trial court violates either the equal population requirement or the prohibition against race-based voter dilution. Hence, we do not directly consider the constitutional criteria here.
¶ 40 Following the preliminary application of the constitutional criteria, the General Assembly has set forth a non-exhaustive list of four non-constitutional factors that may be considered by the trial court. § 2-1-102(1)(b); See Beauprez, 42 P.3d at 651 (holding that once the constitutional requirements "are satisfied, a court may consider non-constitutional criteria that have been articulated by the state as important state policy"); see also Upham v. Seamon, 456 U.S. 37, 41, 102 S.Ct. 1518, 71 L.Ed.2d 725 (1982) ("[W]henever adherence to state policy does not detract from the requirements of the Federal Constitution, we hold that a district court should ... honor state policies in the context of congressional reapportionment."). These factors are: (1) the preservation of political subdivisions; (2) the preservation of communities of interest; (3) district compactness; and (4) the minimal disruption of the existing district boundaries. §§ 2-1-102(1)(b)(I)-(IV). The statute is clear that the trial court is not limited to these factors, nor must it give weight to any of these enumerated factors over any others. Id. (stating that courts "[m]ay without weight to any factor, utilize factors including but not limited to [the enumerated non-constitutional factors]").
¶ 41 This non-exhaustive and unweighted list stands in stark contrast to the pre-amended version of the statute, which placed the constitutional and non-constitutional factors in the following hierarchy of consideration from most to least important: (1) equal population; (2) compliance with the Voting Rights Act; (3) preservation of political subdivisions; (4) preservation of communities of interest; (5) compactness; and (6) minimum disruption of prior district lines. See ch. 369, sec. 1, § 2-1-102, 2010 Colo. Sess. Laws 1735-36.
¶ 42 Although the revised version of the statute empowers the trial court with significant discretion, we note that the statute was not enacted in a vacuum. Rather, this statute is merely the newest chapter in our state's decades-long history of protracted redistricting battles. Accordingly, although we now interpret and apply this new statute for
¶ 43 Of paramount importance, we note the foundational goal of congressional redistricting under the United States Constitution: "fair and effective representation for all citizens." Reynolds, 377 U.S. at 565, 84 S.Ct. 1362. As with the constitutional factors discussed above, this principle stems directly from the Equal Protection Clause of the Fourteenth Amendment and "the democratic ideals of equality and majority rule." Id. at 566, 84 S.Ct. 1362. Accordingly, each of the General Assembly's enumerated non-constitutional factors must be interpreted in light of this overarching goal and applied in a holistic fashion.
¶ 44 With respect to the first non-constitutional factor, the preservation of county, city, and town boundaries, it is clear that citizens will best be represented by a redistricting scheme that unites their common local identities. Carstens, 543 F.Supp. at 88 ("These political subdivisions should remain undivided whenever possible because the sense of community derived from established governmental units tends to foster effective representation."). "Unnecessary fragmentation of these units not only `undermines the ability of constituencies to organize effectively but also ... increases the likelihood of voter confusion regarding other elections based on political subdivision geographics.'" Id. (quoting Amer. Bar Ass'n Special Comm. on Election Law and Voter Participation, Congressional Redistricting 12 (1981)).
¶ 45 Due to the paramount constitutional factors and the competing non-constitutional factors, however, fragmentation of municipal and county boundaries is sometimes unavoidable. Id. When such tough choices must be made, we agree with the Carstens court that it is generally more important to preserve county boundaries in rural areas, whereas municipal boundaries should often take precedence in the urbanized portions of the state. Id. ("As a general rule, county lines are more meaningful in sparsely populated areas because the residents rely on the county government to provide all necessary services. Municipal boundaries, on the other hand, take precedence in densely populated areas. These local units of government represent logical centers of community interest for urban residents who identify more closely with municipal rather than county services."). This localized application of the first factor is consistent with our ultimate goal of creating districts that maximize fair and effective representation.
¶ 46 Similarly, the second non-constitutional factor enumerated in our statute, the preservation of communities of interest, stems directly from the underlying purpose of maximizing fair and effective representation. See id. at 91 ("We are convinced that a plan which provides fair and effective representation for the people of Colorado must identify and respect the most important communities of interest within the state."). By grouping like-minded and similarly situated populations, this factor seeks to create cohesive districts that are organized around similar "ethnic, cultural, economic, trade area, geographic, and demographic factors." § 2-1-102(1)(b)(II). Indeed, "formulating a plan without any such consideration would constitute a wholly arbitrary and capricious exercise." Carstens, 543 F.Supp. at 91 (quoting, with approval, closing argument).
¶ 47 Before the trial court below, multiple current U.S. Representatives testified as to the importance of organizing the districts around issues that are uniquely faced in each area of the state due to regional differences in economy, geography, climate, and demographics. The testimony revealed that if an important issue is divided across multiple districts, it is likely to receive diffuse and unfocused attention from the multiple representatives it affects, as each is pulled in other directions by the many other issues confronting their districts. However, if a discrete and unique issue is placed in one district, that representative may familiarize herself with the complexities of the issue and the stakeholders it affects. Additionally, if the issue is of especially unique importance to her district, she may use a portion of her limited resources to designate a member of her staff to focus exclusively on the issue. Finally, the importance of this unique issue
¶ 48 Unlike the preservation of political subdivisions, which are relatively static, we acknowledge that the myriad ways to define communities of interest regularly evolve. This flexibility provides for the organization of congressional districts to address the most pressing issues of the day. A definition of communities of interest that focuses on current issues and communities is consistent with the 2010 amendment to section 2-1-102(1)(b)(II), which deleted a provision that had explicitly recognized as "[t]raditional communities of interest" the Western Slope and the Eastern Plains. Pragmatically, this allows for the dissolution of old communities of interest and the recognition of emerging communities of interest as the state's demographics continue to shift and change.
¶ 49 For example, in Beauprez, we explicitly approved of the trial court's decision to organize CD2 to include the coalition of municipalities that had formed to advocate for federal funds to assist in the cleanup of the radioactive contamination at the Rocky Flats Plant. 42 P.3d at 652. Now, a decade later, the cleanup of Rocky Flats has been completed and all parties before the trial court agreed that this community of interest no longer exists. This success, and the example that it provides, should guide the application of the communities of interest factor as courts strive to draw districts that maximize fair and effective representation to tackle the challenges of today and tomorrow—and not the challenges of yesterday.
¶ 50 In its arguments before this court, the Hall Plaintiffs imply that communities of interest should be legally defined by the existing congressional districts. We reject this argument because it would render the redistricting process nothing more than a mathematical recalibration of the boundaries of the existing districts. Although the primary charge of redistricting is to achieve equal population, the enumeration of non-constitutional factors in section 2-1-102(1)(b) clearly indicates that the legislature intended for the trial court to look beyond the raw data and to ensure that the present needs and demands of Coloradoans are met by representatives that are responsive and accountable. Under the Hall Plaintiffs' interpretation, courts would be forever beholden to the districts created decades earlier. As a separate, enumerated factor, the preservation of existing districts is nothing more than one of the many factors that may be considered. Thus, the preservation of existing districts should not be weighed so heavily as to subsume the General Assembly's recognition of the importance of unified communities of interest, other than those that have solidified around historic districts, to representative democracy. See §§ 2-1-102(a)(1)(II), (IV). Instead, the minimization of existing district boundaries should be carefully balanced with the many other competing factors.
¶ 51 The third enumerated non-constitutional factor in section 2-1-102, compactness, seeks to promote "fair and effective representation" by implicitly recognizing that the more densely located a representative's constituents, the easier it is to travel across and to physically engage with the district. Carstens, 543 F.Supp. at 87 ("Compact districts do, however, reduce electoral costs (in both time and money) and increase the opportunities for more effective representation by concentrating a congressperson's constituency in an easily accessible area."). Given the priority of equal population and the variable population distribution throughout Colorado, our districts will never be of comparable physical size. Id. at 88 ("The demographics of the State of Colorado are such that the Denver metropolitan districts will be extremely compact while the west slope and eastern plains districts will be quite large."). Instead, an adopted map should be reviewed for compactness as compared with previous district lines and competing map proposals to ensure that the map-drawing process has not been tainted by "partisan gerrymandering." Id. at 87.
¶ 52 Next, because section 2-1-102 is open-ended and because the 2010 amendment deleted the prohibition on the consideration of "non-neutral" political factors, we believe
¶ 53 Finally, we note that the important interests furthered by these non-constitutional factors will often abut and conflict with one another. Indeed, the individual factors, particularly the preservation of communities of interest, are not necessarily internally consistent and will often conflict. Additionally, every alteration that is made to one boundary for the sake of one factor will require alteration to another part of the map to balance population, which might then trigger even further alterations. In turn, this so-called "ripple effect" might split political subdivisions, divide communities of interest, or result in less compact districts. Accord In re Reapportionment of Colo. Gen. Assembly, 647 P.2d 191, 196 (Colo.1982) (noting, in the context of state legislative reapportionment, that "[e]ach detail of the reapportionment plan which we might disapprove would require the Commission to make changes which have a `ripple effect,' necessitating numerous other changes in the reapportionment scheme"). Again, such conflicts must always be resolved in light of the primary goal of this undertaking: "fair and effective representation for all citizens." Reynolds, 377 U.S. at 565, 84 S.Ct. 1362. Accordingly, "[i]f conflicts do, in fact, occur when these criteria are applied to redistricting plans, we are convinced that balancing the competing interests represents a realistic and practical means for resolving any potential problems." Carstens, 543 F.Supp. at 83.
¶ 54 In considering whether the trial court lawfully adopted the Moreno/South Map in light of the non-constitutional factors under section 2-1-102(1)(b), we apply an abuse of discretion standard. See Beauprez, 42 P.3d at 652. A court abuses its discretion only if its decision is "manifestly unreasonable, arbitrary, or unfair." Freedom Colo. Information, Inc. v. El Paso Cnty. Sheriff's Dep't, 196 P.3d 892, 899 (Colo.2008). "In assessing whether a trial court's decision is manifestly unreasonable, arbitrary, or unfair, we ask not whether we would have reached a different result but, rather, whether the trial court's decision fell within the range of reasonable options." E-470 Pub. Highway Auth. v. Revenig, 140 P.3d 227, 230-31 (Colo.
¶ 55 Given the broad discretion afforded by section 2-1-102 and the many competing interests and identifiable political, cultural, ethnic, economic, geographic, trade, and demographic groups within our diverse state, this standard merely directs us to review the trial court's adopted remedy to ensure that it was a reasonable balancing of these many factors in light of the testimony and evidence before it. See Beauprez, 42 P.3d at 652 (approving the election to split several municipalities in order to organize a congressional district around the community of interest related to the cleanup of Rocky Flats because "[t]he district court's decision to split these municipalities [was] supported by the record"); see also Carstens, 543 F.Supp. at 83-84 (summarizing Colorado's demographics and geography because such factors are "important to the development of a rational state policy for redistricting").
¶ 56 Thus, we are not presently tasked with determining whether the trial court adopted the perfect map or even the best of the proposed submissions. Instead, the narrow scope of our review is to determine whether the trial court's remedy reasonably balances the competing non-constitutional factors in a manner that will promote "fair and efficient representation for all citizens." Reynolds, 377 U.S. at 565, 84 S.Ct. 1362.
¶ 57 Under this standard, we hold that the map adopted by the trial court, the Moreno/South Map, reasonably balances the many non-constitutional factors that a court may consider and thus the adoption of this map did not constitute an abuse of discretion. As chronicled above, the trial court engaged in a "thorough, inclusive, and non-partisan" process. Beauprez, 42 P.3d at 647. The court accepted the input and contributions of all intervening parties and amici and actively participated in the ten-day trial before it. Further, the court's written order exhibits an appropriate respect for its role and explains its careful and thoughtful deliberations in weighing the testimony and evidence before it. While the Hall Plaintiffs highlight the fact that the adopted plan moves almost 1.4 million Coloradoans from their existing districts, it appears that the trial court adopted the proposed map that it determined best balanced the many competing interests at stake. We hold that this conclusion was reasonable and was supported by the evidence, and thus, was neither arbitrary nor capricious.
¶ 58 The Moreno/South Map makes only slight alterations to the boundaries of CD1 that have more or less remained constant since the 1970s. CD1 encompasses the entirety of the City and County of Denver, the state's largest population center and metropolitan hub. Because Denver has insufficient population to complete a district of the requisite population size, the adopted map adds several whole political subdivisions that are adjacent to the city's southern boundary and linked via the Santa Fe and Broadway transportation corridors, including Sheridan, Englewood, and Cherry Hills Village. Additionally, CD1 takes in unincorporated areas adjacent to Denver's southwest boundary, including unincorporated areas south of the Grant Ranch known as Ken Caryl and Columbine. The court justified the inclusion of these southern areas in CD1 because drawing CD1 to incorporate either the northern, eastern, or western municipalities surrounding Denver would have required the division of a major city across two congressional districts.
¶ 59 We hold that it was reasonable for the trial court to find that there is no genuine dispute that the City and County of Denver and the suburban areas to its immediate south constitute a community of interest given their proximity and similar economic and
¶ 60 In CD2, the Moreno/South Map includes the entirety of Summit, Grand, Clear Creek, Gilpin, Broomfield, and Larimer counties in addition to portions of Boulder, Jefferson, Park, and Eagle counties. The trial court found that CD2 is sufficiently compact and predominantly organized around five major communities of interest: (1) the I-70 corridor from just west of Denver at C-470 up through the mountain and resort towns of Summit, Grand, and Eagle counties to Avon; (2) Rocky Mountain National Park, which is located at the tri-county border of Boulder, Grand, and Larimer counties; (3) the beetle kill infestation and its acute devastation in Grand, Summit, Larimer, and Boulder counties; (4) the connection between the state's flagship public research universities, the University of Colorado Boulder (CU-Boulder) and Colorado State University Fort Collins (CSU-Fort Collins); and (5) the similar health and high-tech industries that have spun-off from these institutions and their important role in the economy of each university town.
¶ 61 First, the trial court recognized the regional importance of the I-70 transit corridor from the populous Front Range to the many recreational opportunities and resorts along the interstate to Avon. The trial court found that the federal government plays a critical role on this stretch of I-70 because of the importance of federal funds to deal with safety, maintenance, and congestion. Further, the trial court approved of splitting Eagle County to achieve equal population in a manner that places the area west of Avon in CD3 because the section of I-70 from Avon to the Utah border does not traverse the same rugged terrain and does not face the same congestion problems as the stretch from the Front Range to Avon. See Beauprez, 42 P.3d at 652 (recognizing the "ski corridor" through Summit, Eagle, and Grand counties).
¶ 62 Second, the trial court found a community of interest in the many municipalities surrounding Rocky Mountain National Park, which lies within Boulder, Grand, and Larimer counties. As a crown jewel of the National Park System, Rocky Mountain National Park attracts visitors from around the world and is a major economic engine for the gateway communities that surround the park. By moving Larimer County into CD2, the trial court reasoned that there would be one representative that could focus exclusively on Rocky Mountain National Park and advocate on its behalf for additional resources to deal with forest health, overpopulation of elk, transportation, and air and water quality issues.
¶ 63 Third, the court recognized an important community of interest in CD2 in the pine bark beetle infestation. At trial, there was testimony regarding the scourge of the bark beetle and the devastation that it has wrought throughout the state over the past decade. Although the court acknowledged that beetle kill is an issue throughout the state, it determined that the damage has been particularly devastating in Grand, Larimer, Boulder, Gilpin, Clear Creek, Summit, and Eagle counties. Thus, it found a community of interest in CD2 that needs federal awareness and support to help prevent further damage and also to mitigate the impacts of forest fires and falling trees.
¶ 64 Fourth, the trial court recognized a substantial community of interest between Boulder and Larimer counties related to the flagship public research institution in each county, CU-Boulder and CSU-Fort Collins, respectively. At trial, the court heard extensive testimony about the financial and demographic impact that each institution has on its surrounding community by attracting talented students and preeminent faculty from across the state and throughout the world.
¶ 66 The trial court rejected the Hall Plaintiffs' argument that the two institutions would be better served if they were represented by different congresspersons. The trial court found this argument "fundamentally at odds" with the concept of a community of interest. The court held that communities of interest should not be divided; rather, they should be unified so that their interests rise to concerns of the first order within their district.
¶ 67 Fifth, and finally, the court recognized an important community of interest in CD2 between Boulder and Larimer counties due to the high-tech and bio-tech industries that have spun off from research completed at CU-Boulder and CSU-Fort Collins. As a result of these universities, Boulder and Fort Collins share the same major employment industries: education, health care, and high-tech. These industries predominate both communities and are only expected to grow and become even more important in the future. Conversely, the trial court found that each county has experienced a similar decline in agriculture and manufacturing jobs; a trend that is also expected to continue. Both counties also share similar economic and ethnic demographics and typically have a lower unemployment rate than the rest of the state.
¶ 68 Although the court acknowledged that the previous boundaries of CD2 were altered by adding Larimer County and removing small sections of Jefferson, Adams, and Weld counties, the court reasoned that this changed was justified by the successful completion of the Rocky Flats cleanup. The court reasoned that the coalition of governments seeking to cleanup Rocky Flats served as a major community of interest for drawing CD2 after the 2000 census, see Beauprez, 42 P.3d at 647, but now that Rocky Flats is no longer an issue, the court needed to evaluate which issues are of greatest concern in the region at present.
¶ 69 We hold that the trial court was reasonable in placing its concern for present communities of interest above a mechanistic attempt to minimize the disruption of existing district boundaries. Further, we hold that the trial court did not abuse its discretion in organizing CD2 around the communities discussed above, which were amply supported by the record.
¶ 70 CD3 in the Moreno/South Map stretches over the majority of the western half of the state. It includes twenty-eight whole counties from Jackson County in the north and Pueblo County and the San Luis Valley counties in the south out to the western border with Utah. Thus, as it has for the last three decades, CD3 continues to combine the San Luis Valley with the mountainous
¶ 71 Eagle County is the only split county in CD3 and, as the trial court explained, this split appears to have been reasonable in order to keep the major Eagle County ski areas in CD2. The trial court found that west of Avon, where Eagle County is split, the I-70 corridor is flatter, less traveled, and allows for higher speeds out to the Utah border. Additionally, the trial court reasoned that this split served to keep Glenwood Springs (located in the western half of Eagle County) in the same congressional district as the other major population centers in the Roaring Fork basin, Aspen and Basalt.
¶ 72 The trial court also found that the borders of CD3 remained largely unchanged and thus preserved the previously recognized communities of interest within the district, including a rural populace, ranching, mining, tourism, public lands, energy production, and water. See Beauprez, 42 P.3d at 652. The trial court noted that 75 percent of the Western Slope is owned and managed by the federal government and that the district wholly contains three national parks: Sand Dunes National Park, Mesa Verde National Park, and Black Canyon of the Gunnison National Park.
¶ 73 As adopted, the boundaries of CD3 have changed slightly so that Las Animas and Otero counties are now wholly contained within CD4 and Lake County and the aforementioned portion of Eagle County are in CD3. The trial court concluded that this was reasonable because Lake County and the western portion of Eagle County share similar community interests with the Western Slope and because the change serves to make CD3 slightly more compact by eliminating its easternmost reach.
¶ 74 We conclude that this balancing of the non-constitutional criteria was reasonable. As adopted, CD3 minimizes disruption to its existing boundaries and is organized around well-established communities of interest that remain important today. Hence, we hold that the trial court did not abuse its discretion by adopting the boundaries for CD3 proposed by the Moreno Plaintiffs.
¶ 75 As adopted, CD4 contains eighteen whole Eastern Plains counties, including the new addition of Otero, Las Animas, and Elbert counties, as well as the city of Longmont in Boulder County and the new addition of the more rural portions of Douglas, Adams, and Arapahoe counties. Unlike in the past, CD4 no longer contains Larimer County.
¶ 76 Initially, the trial court determined that the inclusion of Las Animas and Otero counties in CD4 was a reasonable change. There was no debate at trial over the inclusion of these counties in CD4, and, as explained above, their addition makes both CD3 and CD4 more compact as these districts are now even more concentrated in the western and eastern portions of the state, respectively.
¶ 77 Next, as the trial court noted, all parties agree that Weld County and the Eastern Plains have a significant agricultural community of interest. The trial court concluded that the removal of Larimer County from the existing CD4 and placement in the new CD2 was appropriate, due to its waning ties with neighboring Weld County and the rest of the Eastern Plains. The trial court found that instead of agriculture as its primary focus, Larimer County now has a greater emphasis on a white collar economy in the education, health care, and technology sectors. The testimony at trial revealed that CSU-Fort Collins, once known almost exclusively as Colorado's agricultural school, has broadened substantially in the last decade to include significant research in other areas, such as technology and space. The trial court found that Weld County's emphasis remains on its role as the state's largest agricultural producer, as well as on oil and gas development in the Niobrara formation and the attendant regulatory and environmental issues associated with "fracking."
¶ 78 Demographically, the counties differ as well. Evidence at trial showed that Larimer County's unemployment rate parallels that found in Boulder County and is lower than almost anywhere else in the state,
¶ 79 Regarding transportation, evidence presented at trial showed Larimer County has a greater focus on access to mass transportation compared to Weld County's primary use of single-occupant vehicles. Open space and wildlife appears to be an issue of greater importance in Larimer County, where a sales tax to raise revenue for the purchase of open space was recently passed by popular vote. In contrast, Weld County has neither a parks and recreation department nor a dedicated revenue stream for purchasing open space or protecting wildlife. For these reasons, we conclude that it was reasonable for the trial court to determine that Larimer County was more appropriately placed in CD2 with Boulder County.
¶ 80 Because CD2 was reconfigured based on these existing and emerging communities of interest and because Larimer County was appropriately moved into CD2 to equalize population, the trial court had to consider what population center most appropriately matched the communities of interest found in CD4. The trial court considered the issue of alternative energy development, but found that while this was a growth industry in the Eastern Plains, there was no issue of federal import raised at trial in connection with alternative energy development as it is being implemented in CD4. In contrast, the trial court found that the growing use of fracking to expand oil and gas development presents significant issues at the federal level. Evidence at trial revealed that there are questions about the appropriate role for the federal government to play as overseer of regulation and disclosure of the composition and disposal of the fluids involved in the fracking process, as well as questions about the impact that the fracking industry will have on the local infrastructure, land, water, and air. The trial court found that fracking issues are very much in the public consciousness.
¶ 81 According to expert testimony at trial, Douglas, Elbert, and Arapahoe counties are the newest "frontier" of oil and gas development in the state. The trial court found that Weld County is already currently experiencing the economic development, infrastructure, environmental, and labor impacts from the oil and gas drilling in the Niobrara formation.
¶ 82 Douglas County argued that development of this industry in its borders is unlikely, but the trial court found that this argument belied the reality of what is happening in Douglas County, which has already begun to prepare for fracking. As evidence of Douglas County's concern regarding potential fracking, the trial court noted that the Douglas County Commissioners recently approved a $170,000 study to examine the transportation impacts of fracking on county residents and infrastructure. In addition, the trial court pointed to the fact that the commissioners are also investigating zoning changes, emergency management planning, temporary housing for drilling employees, transportation, and water quality issues. The trial court heard evidence that fracking sites are substantial and include tanks that hold hundreds of gallons of oil and drilling by-products, command centers, and hundred-foot tall drilling equipment. A Douglas County engineering representative visited fracking sites in Weld County on behalf of the county, and gave a "compelling" description at trial of a massive fracking operation just a short distance from a housing subdivision.
¶ 83 The evidence at trial also established that Douglas County residents have a high level of interest and concern on the issue of whether fracking facilities are safe in close proximity to population centers. There was not an empty chair at the Douglas County Fairground when over 300 people attended a county-sponsored public forum where fracking was discussed in April 2011. Testimony at trial revealed one of the speakers at this meeting was a planning manager for the city of Greeley with experience in managing the many issues that arise when fracking occurs near suburban development, where surface owners are comparatively less knowledgeable about their rights and the impact of drilling on or near their land.
¶ 85 In addition to Douglas County's shared concerns over the impacts of fracking, the trial court found that Douglas County also faces a similar risk of extreme drought as do the Eastern Plains counties. Douglas and Adams counties are among the most at-risk areas for severe and prolonged drought.
¶ 86 Like the trial court, we recognize that the Douglas County Commissioners opposed the inclusion of Douglas County in CD4. However, it is impossible to create a map where every interested party is pleased with the result, and it is not our role to attempt to do so. Rather, we simply review the map adopted by the trial court to ensure its legality. Here, there was ample evidence in the record supporting Douglas County's placement in CD4. The trial court found it unlikely that the representational interests of Douglas County would suffer adversely due to its placement. Furthermore, there was some level of compromise reached by the inclusion of Highlands Ranch in CD6 in the final Moreno/South Map. The evidence at trial showed that Highlands Ranch, although relatively small in physical area, comprises almost one-third of the population of Douglas County and is the area of Douglas County most closely connected to Denver. Thus, the trial court determined that Highlands Ranch was more closely related to the "exurban" bedroom communities in adopted CD6. Hence, we hold that it was not an abuse of discretion for the trial court to adopt a map with CD4 configured in this manner because its configuration was an appropriate balancing of the factors and was supported by the record.
¶ 87 CD5 remains largely unchanged from its previous boundaries and contains the entirety of El Paso, Teller, Fremont, and Chaffee counties, along with the bulk of Park County.
¶ 88 The only notable alteration to the existing boundaries of CD5 is the removal of Lake County. To achieve population equality following the 2010 census, CD5 needed to shed 7,445 people from its previous configuration. Lake County's population is approximately this size and was therefore moved out of CD5 and into CD3. Although the primary transportation route from Lake County to CD3 is closed during the winter months when Independence Pass is inaccessible, the trial court found that transportation issues between Lake County and the rest of CD3 was less of a concern due to advances in communications and internet connectivity within the region. Notably, it does not appear that Lake County objected to this change.
¶ 89 CD5 still contains the five military bases located in El Paso County, a community of interest that the trial court found to be of paramount significance. The court found that unified representation of these bases by a single congressperson provided an advantage to the five facilities, their employees, and the other businesses that are associated with and rely on these facilities. Although these five military facilities are a matter of statewide concern, the trial court heard evidence that having one congressperson as the point person on the military facilities in Colorado has benefited the state. The trial court found that a congressperson with that many military facilities in his or her district will have a "seat at the table" when matters involving military installations are at issue. In addition, the trial court found that combining El Paso County with Fremont, Chaffee, Teller, and most of Park
¶ 90 As adopted by the trial court, CD6 loosely follows the E-470 beltway along the eastern portion of the Denver Metro Region. It extends from northeast Thornton to Brighton around Denver International Airport and down south through Aurora and the communities immediately south of Denver, including Foxfield, Greenwood Village, Centennial, Littleton, and Highlands Ranch.
¶ 91 The trial court found that these areas are united as a community of interest because they can generally be characterized as "exurbs" of Denver. The trial court appears to have been particularly persuaded by the expert testimony of the Moreno Plaintiff's certified expert on Front Range community development and planning. Unlike the areas west of Denver that the expert witness characterized as "first ring suburbs," the communities to the east of Denver have generally developed much more recently. Accordingly, the municipalities and unincorporated areas within CD6 have generally newer infrastructure, which the trial court reasoned was important because these communities do not face the same maintenance and capital replacement needs of the older metro communities to the west of Denver. Additionally, the trial court found that many of these communities are still growing and will likely continue to grow in the coming decades. These areas also generally have higher median incomes than the rest of the state.
¶ 92 Although CD6 splits Adams, Arapahoe, and Douglas counties, the trial court found that this was necessary to balance the competing interest of keeping Aurora, which spans these three counties, in a single district. The trial court prioritized the unification of Aurora over the unification of the surrounding counties because, as the third largest city in the state, the city government serves a vital role as the sole provider of law enforcement, fire protection, and water facilities within city limits. Additionally, the trial court found that the city of Aurora would best be served by a single representative because it is uniquely dependent on federal appropriations. Three of the city's largest employers are the Buckley Air Force Base, the new Veterans Affairs hospital, and the University of Colorado Denver's Anschutz Medical Campus, which is heavily supported by federal research grants.
¶ 93 Finally, the court found that the adopted boundaries for CD6 create a competitive district because present party affiliation for Republican, Democratic, and Unaffiliated voters are all greater than 30 percent. Although competitiveness is not one of the non-constitutional factors enumerated in section 2-1-102, the trial court reasoned that this factor was properly considered because it "requires candidates running for office to work very hard, to listen to all views, and to reach out and engage as many people as possible."
¶ 94 As explained above, we hold that the trial court's consideration of competitiveness was a permissible and reasonable factor given the overarching goal of maximizing "fair and effective representation for all citizens." Reynolds, 377 U.S. at 565, 84 S.Ct. 1362. Further, it was reasonable for the trial court to place the important goal of making Aurora whole in a single district over the competing interest of minimizing county splits. See Carstens, 543 F.Supp. at 88 ("[M]unicipal boundaries ... take precedence [over county boundaries] in densely populated areas."). By organizing CD6 in this manner, the trial court appropriately moved the more rural areas of eastern Adams and Arapahoe counties and Elbert County into CD4, where they fit with that rural and agrarian community of interest. Conversely, the trial court's reasoning that the communities that are now included in CD6 all share similar interests as "exurb" communities was well supported by credible expert testimony. Hence, we hold that the trial court did not abuse its discretion by adopting these boundaries for CD6.
¶ 95 CD7 takes in the remainder of the metro suburbs to the north and west of Denver. This includes all of Lakewood, Golden, Wheat Ridge, Arvada, Westminster, Federal Heights, Northglenn, and Commerce City, a large portion of Thornton, and the surrounding unincorporated areas of Adams and Jefferson counties.
¶ 96 Again, the trial court relied heavily upon the testimony of the Moreno Plaintiffs' expert that these areas constitute "first ring suburbs" that were the first communities outside of Denver to urbanize in the 1950s and 1960s. As such, the trial court found that these communities share similar characteristics in that they are lower growth and have much older infrastructure than the communities in adopted CD6. Accordingly, there is a greater need for infrastructure improvement and replacement in CD7 as its schools, hospitals, roads, sewers, and water delivery systems reach the end of their useful lives. This problem is all the more magnified by the reality that the communities of CD7 generally have lower median incomes.
¶ 97 The trial court also found that CD7 has a higher Latino population than those of the "exurbs" in CD6. As a result, the trial court reasoned that it was especially important that CD6 be drawn so as to create a competitive district. The trial court found that competitive districts empower the Hispanic community by allowing its voting bloc to carry significant weight in elections. Again, the trial court reasoned that the more competitive a district, the more responsive a representative would be and the lesser chance that any voter bloc could be marginalized or ignored. As adopted, CD7 contains greater than 30 percent of each Unaffiliated, Democratic, and Republican voters.
¶ 98 Additionally, the court reasoned that CD7 was organized around a community of interest in renewable energy. CD7 is home to the National Renewable Energy Laboratory (NREL), which the trial court found to be a key employer within the district as well as a primary driver of spin-off businesses within the clean energy industry. The trial court also connected the federal presence of NREL (in Golden) within CD7 to the Federal Center in Lakewood, which contains regional offices for almost every agency within the federal government. Finally, the trial court also found a community of interest within CD7 related to the extension of FasTracks through the region in the coming decade, which is expected to ease traffic congestion on the 6th Avenue and I-70 corridors.
¶ 99 The trial court justified the alteration of the boundaries of adopted CD7 as compared to existing CD7 as necessary to improve compactness within the district. Similar to existing CD6, the boundaries of existing CD7 stretch from the west side of Denver in Lakewood, across the north of the city, and all the way out through the rural portions of Arapahoe and Adams counties. The trial court found that the boundaries of CD7 proposed by the Moreno/South Map increased compactness by concentrating CD7 entirely to the west and north of Denver. In contrast, the Hall Plaintiffs' maps proposed that CD7 connect Jefferson County to Adams County by a narrow frontage road and extend to far eastern, rural portions of Adams and Arapahoe counties. The trial court rejected this proposal on the grounds that it would create a CD7 that combined urbanized areas with rural communities that are "visually and functionally indistinguishable from adjacent lands in [the Eastern Plains counties]." Thus, the trial court found that it was appropriate to alter the boundaries of CD7 to increase compactness and the preservation of unique communities of interest.
¶ 100 On review, we hold that the trial court reasonably applied and balanced the competing interests of the non-constitutional factors by adopting the boundaries of CD7. It was reasonable to alter the existing boundaries of CD7 to increase compactness and to preserve the communities of interest surrounding clean energy and the need for infrastructure improvements within the "first ring suburbs." Such a conclusion was amply supported by the record. Hence, we hold that the court did not abuse its discretion with respect to CD7.
¶ 101 The trial court reasonably balanced the many competing non-constitutional factors in a manner that will maximize "fair and effective representation for all citizens." Reynolds, 377 U.S. at 565, 84 S.Ct. 1362. Its findings are supported by the record, which was compiled through a thorough, open, and fair process. Thus, we hold that the trial court properly applied the controlling law and adopted a lawful redistricting scheme.
¶ 102 For the reasons explained herein, we issued our order and mandate affirming the order of the trial court on December 5, 2011.
Justice COATS concurs in the judgment.
Justice EID dissents.
Justice COATS, concurring in the judgment.
¶ 103 Although for different reasons, I would also affirm the district court's order, redrawing as it does the congressional districts in this state in time for the 2012 general election. I write separately not only to briefly explain my own reasons for doing so but also to make clear that the only issue before the court today is the validity of the district court's remedial redistricting plan, protecting the plaintiffs' constitutional right to what has come to be known as one person/one vote. While it is not my intention either to criticize the district court's plan or to invite further legislative action, in light of our prior opinion in People ex rel Salazar v. Davidson, 79 P.3d 1221 (Colo.2003), cert. denied, 541 U.S. 1093, 124 S.Ct. 2228, 159 L.Ed.2d 260 (2004) (Rehnquist, C.J., joined by Scalia and Thomas, JJ., dissenting from denial of certiorari), I feel obliged to note that I do not consider this court foreclosed from reconsidering the question whether the general assembly is precluded from fulfilling its constitutional obligation to redistrict, notwithstanding the existence of a valid court-ordered plan following the last census.
¶ 104 I would affirm the district court's order in this case because redrawing the congressional map was indisputably made necessary by the population changes reflected in the 2010 census; the general assembly was unable to fulfill its obligation to redistrict in time for the general election of 2012; and the district court's redrawn congressional map does not violate federal law. In fact, other than complying with the Fourteenth Amendment's substantial equality requirement and the mandates of the Voting Rights Act, none of which are challenged here, the only limitation imposed on the courts in drawing congressional boundaries appears to be that they should, as much as possible, follow the policies and preferences of the state. See Upham v. Seamon, 456 U.S. 37, 41-42, 102 S.Ct. 1518, 71 L.Ed.2d 725 (1982). Unlike the majority, I do not believe our legislature's instructions for judicial redistricting, enacted following our Salazar decision, reflect such policies and preferences. In stark contrast to its elaborate provision for state senate and representative districts, see Colo. Const. art. V., §§ 46-48, the state constitution provides almost no guidance for or limitation on the general assembly's division of the state into congressional districts, see Colo. Const. art. V., § 44, other than requiring it do so. I do not believe the state constitution can be read to permit the delegation of that responsibility to the judicial branch of government. Because section 2-1-102, C.R.S. (2011), does not describe any redistricting policies or preferences that govern the legislature itself, the body with the constitutional duty to redistrict, but only purports to place limitations on the courts should they be forced to intercede, I do not believe the considerations it identifies amount to a redistricting policy of the state as contemplated by United States Supreme Court jurisprudence; and since the general assembly has for so long failed to redistrict itself, it would also be difficult to describe any existing or recently past congressional map as reflecting any state redistricting policies.
¶ 105 Until a half-century ago, any role for the judiciary in the inherently political process of redistricting not only went unrecognized; it was in fact forbidden. See Baker v. Carr, 369 U.S. 186, 208-37, 82 S.Ct. 691, 7 L.Ed.2d 663 (1962). In the intervening years, in combination with lifting its political-question ban, see id., the Supreme Court's determination that equal protection necessarily
¶ 106 In Salazar, this court held that the term "General Assembly" can be interpreted, for purposes of the redistricting mandate of the state constitution, so broadly as to include even an emergency map drawn by the courts in the absence of any action by the general assembly whatsoever. In light of the uniquely federal development of the judiciary's role in congressional redistricting, in particular, I continue to consider this construction untenable. See Salazar, 79 P.3d at 1243-53 (Kourlis, J., joined by Coats, J., dissenting). While I accord the principle of stare decisis the highest respect, see, e.g., Friedland v. Travelers Indem. Co., 105 P.3d 639, 651-53 (Colo.2005) (Coats, J., dissenting), it is clearly not to be understood as an immutable law. Whether the highest court of any jurisdiction will choose to follow or depart from its own prior decisions is ultimately a matter of discretion. See Burnet v. Coronado Oil & Gas Co., 285 U.S. 393, 405-13, 52 S.Ct. 443, 76 L.Ed. 815 (1932) (Brandeis, J., dissenting). Not least among the many reasons militating against a bar to the reconsideration of our holding in Salazar are the facts that it involves an interpretation of the state constitution, not subject to further refinement or modification by either the legislature or even the United States Supreme Court, see id., and that, due to its formal nature, affecting only the respective roles of the legislature and the courts, there can be no danger that a departure from the rule of Salazar might work a hardship or inequity on voters as the result of any reliance upon and ordering their behavior around it. See Planned Parenthood of Southeastern Pa. v. Casey, 505 U.S. 833, 854-55, 112 S.Ct. 2791, 120 L.Ed.2d 674 (1992).
¶ 107 I therefore concur in the judgment of the court to the extent that it affirms only the district court's plan for the 2012 general election. I would reject, however, any suggestion that the general assembly is, for that reason alone, precluded from yet fulfilling its constitutional obligation following the most recent census and congressional apportionment of representatives among the states.
Justice EID, dissenting.
¶ 108 The district court's plan in this case moves nearly one-third of Colorado's total population—almost 1.4 million people—to a different congressional district. This seismic shift is all the more astonishing given that Colorado did not gain or lose a congressional seat in the last census, and considering the
¶ 109 Section 2-1-102(1)(b) provides that a court "[m]ay, without weight to any factor, utilize factors including but not limited to:"
(Emphasis added). Section 2-1-102(1)(b) gives the district court broad discretion to consider the listed factors—and others—in drawing congressional boundaries. Maj. op. at ¶¶ 41, 55. But the district court does not have unbounded discretion to draw boundaries in any manner whatsoever. The text of section 2-1-102(1)(b) makes clear that the district court must, at the very least, consider and give some weight to all of the factors, including minimization of disruption of existing district lines, together with other factors it finds relevant. In my view, the district court abused its discretion by giving virtually no weight to the minimization factor in four of the seven districts.
¶ 110 The district court's lengthy order gives minimization more than passing consideration on only two occasions. First, the district court's order suggests that, because the boundaries of CD1, CD3, and CD5 remain essentially unchanged under the plan, the minimization factor had been given adequate weight. Dist. Ct. Order at 62 ("The changes to the 1st, 3rd, and 5th Congressional District thus reflect the Court's attempt to give effect the statutory criterion of minimizing disruption of current lines."). Yet at no point did the district court pause and consider the actual disruption that the plan caused overall—that is, the movement of almost one-third of Colorado's population to different districts. The court's "minimization in three districts is enough" rationale actually works against the adequacy of the plan, in that the huge shift of population to different districts was borne almost entirely by the four congressional districts that did change. Of course the addition (or loss) of a congressional seat may require significant shift of lines, see, e.g., Beauprez v. Avalos, 42 P.3d 642, 653 (Colo.2002) (affirming significant shift in lines after Colorado gained seventh congressional district), but that did not occur here; Colorado still has seven congressional districts.
¶ 111 In the only other significant discussion of the minimization factor in the order, the district court rejected the Hall Plaintiffs' map on the ground that the only interest it took into account was minimization. Dist. Ct. Order at 46. The district court, however, misperceived the argument that the Hall Plaintiffs make regarding minimization. The Hall Plaintiffs argue that current communities of interest oftentimes align with existing boundaries because those boundaries were drawn to reflect communities of interest at the time. See, e.g., Carstens v. Lamm, 543 F.Supp. 68, 72 (D.Colo.1982) (adopting redistricting plan after 1980 census); Beauprez, 42 P.3d at 653 (affirming district court's redistricting plan drawn after 2000 census). As the majority properly recognizes, communities of interest may change over time, and a district may no longer reflect communities of interest as they currently exist. See, e.g., maj. op. at ¶ 68 (noting that Rocky Flats no longer is an essential issue reflected in CD2). But existing districts provide the appropriate place to start.
¶ 112 Following the lead of the district court, the majority marginalizes the minimization factor throughout its opinion. It repeatedly dismisses the tremendous shift of population in the four changed districts as simply a necessary byproduct of the redistricting process. Maj. op. at ¶¶ 31, 57. In fact, according to the majority, the minimization factor "is nothing more than one of the many factors that may be considered," and plainly should not outweigh the communities
¶ 113 Importantly, the majority and the district court fail to recognize the significant interest that the minimization factor is designed to protect. Obviously, the interest is not the straw man that district lines should always remain the same. Nor is it only that existing district lines are likely to reflect communities of interest now because they did so in the past. It is that district lines, once drawn, reflect and encourage important relationships among constituents, community leaders, and the congressional representative surrounding particular issues—relationships that are lost when district lines change, or in this case, shift dramatically.
¶ 114 One striking example is the fact that Fort Collins in Larimer County has been a major population center for CD4 during the past three decades, and most of the congressional staff for the district is located there. Under the district court's plan, however, Fort Collins is no longer in CD4 but rather occupies the northernmost portion of CD2. It may be difficult to quantify the loss in constituent relationships caused by the move to CD2, but there is undoubtedly a loss.
¶ 115 Douglas County provides another case in point. Most Douglas County residents pay taxes to support the metro Scientific and Cultural Facilities District, the Stadium District, and the Regional Transportation District. Sixty percent of them commute to jobs in Denver. The county is part of the Denver Regional Council of Governments ("DRCOG"), a metropolitan planning organization ("MPO") mandated by federal law for urban areas with populations exceeding 50,000. See 49 U.S.C. § 5303 (2006). DRCOG, which includes Douglas County and eight other counties tied to the metro area, unites local and state officials to address regional transportation issues and to foster economic development, all with an eye toward obtaining federal funding. Under the district court's plan, most of Douglas County has been moved to CD4. Thus, most of the county will lose long-standing relationships with metro Denver communities fostered by its presence in CD6, and instead, Douglas County finds itself in the overwhelmingly agricultural CD4. The fact that Highlands Ranch was carved out from the move to CD4 at the last minute, maj. op. at ¶ 86, simply highlights, rather than solves, the problem.
¶ 116 It is true that, as the majority suggests, it is possible to identify interests that Douglas County shares with CD4, and that Larimer County shares with CD2. See maj. op. at ¶¶ 81-85 (noting that Douglas County and CD4 share an interest in issues arising from water use and energy development); id. at ¶ 60 (noting that the new CD2 includes Fort Collins and Boulder, both home to major state universities). Indeed, it is undoubtedly possible to draw similarities in interest between virtually any two geographic points in Colorado. The bottom line is that districts should be drawn in a manner that takes into account all of the factors listed in section 2-1-102(1)(b), including the minimization of disruption to existing districts. Because the district court's plan failed to do so, I respectfully dissent.
(1) [I]n adopting or enforcing any change to any [congressional] district, courts:
