Carter, C., Pets. v. Chapman, L.

Opinion

[J-20-2022] IN THE SUPREME COURT OF PENNSYLVANIA MIDDLE DISTRICT

BAER, C.J., TODD, DONOHUE, DOUGHERTY, WECHT, MUNDY, BROBSON, JJ.

CAROL ANN CARTER, MONICA : No. 7 MM 2022 PARRILLA, REBECCA POYOUROW, : WILLIAM TUNG, ROSEANNE MILAZZO, : BURT SIEGEL, SUSAN CASSANELLI, LEE : ARGUED: February 18, 2022 CASSANELLI, LYNN WACHMAN, : MICHAEL GUTTMAN, MAYA FONKEU, : BRADY HILL, MARY ELLEN BALCHUNIS, : TOM DEWALL, STEPHANIE MCNULTY : AND JANET TEMIN, : : Petitioners : : : v. : : : LEIGH M. CHAPMAN, IN HER OFFICIAL : CAPACITY AS THE ACTING SECRETARY : OF THE COMMONWEALTH OF : PENNSYLVANIA; JESSICA MATHIS, IN : HER OFFICIAL CAPACITY AS DIRECTOR : FOR THE PENNSYLVANIA BUREAU OF : ELECTION SERVICES AND NOTARIES, : : Respondents : : ----------------------------------------------------------- : PHILIP T. GRESSMAN; RON Y. DONAGI; : KRISTOPHER R. TAPP; PAMELA GORKIN; : DAVID P. MARSH; JAMES L. : ROSENBERGER; AMY MYERS; EUGENE : BOMAN; GARY GORDON; LIZ MCMAHON; : TIMOTHY G. FEEMAN; AND GARTH : ISAAK, : : Petitioners : : : v. : : : LEIGH M. CHAPMAN, IN HER OFFICIAL : CAPACITY AS THE ACTING SECRETARY : OF THE COMMONWEALTH OF : PENNSYLVANIA; JESSICA MATHIS, IN : HER OFFICIAL CAPACITY AS DIRECTOR : FOR THE PENNSYLVANIA BUREAU OF : ELECTION SERVICES AND NOTARIES, : : Respondents :

OPINION

OPINION FILED: March 9, 2022 CHIEF JUSTICE BAER DECIDED: February 23, 2022 I. Introduction

Pennsylvania’s current congressional districting plan is irrefutably unconstitutional

based upon the reapportionment of the House of Representatives following the 2020

Decennial Census conducted pursuant to Article I, Section 2 of the United States

Constitution. Due to this Commonwealth’s loss of population relative to the nation as a

whole, Pennsylvania’s allotted number of congressional representatives declined from

eighteen to seventeen. As a result, Pennsylvania now requires a new congressional

districting plan drawn with only seventeen districts for the upcoming May 17, 2022,

Primary Election.

Because the General Assembly and the Governor failed to agree upon a

congressional redistricting plan, this Court was tasked with that “unwelcome obligation.”

League of Women Voters of Pennsylvania v. Commonwealth, 178 A.3d 737, 823 (Pa.

2018) (“LWV II”). This is not uncharted territory, as a similar scenario unfolded following

the inability of the political branches to enact a plan in the wake of the 1990 Decennial

Census. In Mellow v. Mitchell, 607 A.2d 204 (Pa. 1992), this Court assumed plenary

jurisdiction of an action originating in the Commonwealth Court and designated a

[J-20-2022] - 2 Commonwealth Court judge as master to conduct hearings, make findings of fact, and

render conclusions of law before the Court decided on an appropriate redistricting plan.

Mellow, 607 A.2d at 206. The same procedure was adhered to in this case.

Our Special Master expended tremendous effort by expeditiously conducting

hearings, making extensive findings of fact, providing a comprehensive report to this

Court analyzing the merits of the various congressional redistricting plans submitted

before it, and ultimately recommending the adoption of the plan created by the

Pennsylvania Legislature in House Bill 2146 (“H.B. 2146”), which Governor Tom Wolf

vetoed on January 26, 2022. We acknowledge and thank her for her effort.

After deliberating and affording due consideration to our Special Master’s findings

and recommendation and reviewing de novo the relative merit of the submitted

congressional plans, the Court respectfully declined to adopt the Special Master’s

analysis and ultimate plan selection. Rather, on February 23, 2022, we entered a per

curiam order, directing that the Pennsylvania primary and general elections for seats in

the United States House of Representatives commencing in 2022 shall be conducted in

accordance with the plan submitted to the Special Master by the Carter Petitioners, who

we name herein below (“Carter Plan”).1 Our order indicated that an opinion would follow,

and this opinion is filed in accordance therewith.

In full cognizance that the redistricting of congressional districts falls squarely

within the purview of the General Assembly, U.S. CONST., art. I, § 4, cl. 1, we have fulfilled

our obligation to select a redistricting plan only because the Legislature was unable to do

so.2 In making our selection, we were guided by our decision in LWV II, where we applied

1 Justices Todd, Mundy, and Brobson dissented as to the selection of the Carter Plan as the congressional redistricting plan. 2The Elections Clause of the United States Constitution provides that the “Times, Places and Manner of holding Elections for Senators and Representatives, shall be prescribed

[J-20-2022] - 3 the traditional core districting criteria requiring that congressional districts be compact,

contiguous, as nearly equal in population as practicable, and which minimize divisions of

political subdivisions, while taking into consideration the subordinate historical

considerations, such as communities of interests, the preservation of prior district lines,

and the protection of incumbents. LWV II, 178 A.3d at 816-17. Finally, we have ensured

that the congressional districting plan that we adopted does not violate Pennsylvania’s

Free and Equal Elections Clause by “dilut[ing] the potency of an individual’s ability to

select the congressional representative of his or her choice,” id. at 816, and complies with

the Voting Rights Act, 52 U.S.C. § 10301.3

This Court acknowledges that there is no perfect redistricting plan. Each map

involves trade-offs between the requisite traditional core redistricting criteria, as well as

the subordinate historical redistricting considerations. The task of balancing these criteria

and considerations is better suited to the Commonwealth’s political branches, rather than

the judiciary. Nevertheless, given our unwelcomed circumstance, we have endeavored

to adopt a plan that, as phrased in League of Women Voters of Pennsylvania v.

Commonwealth, 181 A.3d 1083, 1087 (Pa. 2018) (“LWV III”), is “superior or comparable”

to all of the plans submitted on the designated criteria.

As evidenced by the views expressed by our esteemed colleagues and the Special

Master, reasonable minds can disagree in good faith as to which submitted plan best

in each State by the Legislature thereof; but the Congress may at any time by Law make or alter such Regulations, except as to the Places of chusing Senators.” U.S. CONST., art. I, § 4, cl. 1. Congress passed 2 U.S.C. § 2a, pursuant to the Elections Clause, which provides that, following the decennial census and reapportionment, the Clerk of the House of Representatives shall “send to the executive of each State a certificate of the number of Representatives to which such State is entitled” and the state shall be redistricted “in the manner provided by the law thereof.” 3 The Free and Equal Elections Clause provides that “Elections shall be free and equal; and no power, civil or military, shall at any time interfere to prevent the free exercise of the right of suffrage.” PA. CONST. art. I, § 5.

[J-20-2022] - 4 balances the requisite criteria and considerations. Nevertheless, for the reasons set forth

below, we adopt the plan submitted to the Special Master by the Carter Petitioners as the

2022 Congressional Redistricting Plan.

II. Procedural History

This matter commenced on December 17, 2021, when two separate petitions for

review were filed in the Commonwealth Court’s original jurisdiction. At Commonwealth

Court docket number 464 M.D. 2021, Carol Ann Carter et al. (collectively referred to as

“Carter Petitioners”) presented a petition for review.4 The Carter Petitioners identified

themselves as citizens of the United States who are registered to vote in Pennsylvania.

They named as respondents to their petition Veronica Degraffenreid, in her capacity as

then-Acting Secretary of the Commonwealth of Pennsylvania,5 and Jessica Mathis, in her

capacity as Director for the Pennsylvania Bureau of Election Services and Notaries

(collectively referred to as “Respondents”). At Commonwealth Court docket number 465

M.D. 2021, Philip T. Gressman, et al. (collectively referred to as “Gressman Petitioners”)

filed a petition for review.6 7 The Gressman Petitioners identified themselves as United

States citizens who are registered to vote in Pennsylvania. They further described

themselves as “leading professors of mathematics and science[.]” Gressman Petitioners’

4Additional Carter Petitioners included: Monica Parrilla, Rebecca Poyourow, William Tung, Roseanne Milazzo, Burt Siegel, Susan Cassanelli, Lee Cassanelli, Lynn Wachman, Michael Guttman, Maya Fonkeu, Brady Hill, Mary Ellen Balchunis, Tom DeWall, Stephanie McNulty, and Janet Temin. 5 Leigh Chapman later became the Acting Secretary of the Commonwealth of Pennsylvania and was substituted for Acting Secretary Degraffenreid. 6 Additional Gressman Petitioners were Ron Y. Donagi, Kristopher R. Tapp, Pamela

Gorkin, David P. Marsh, James L. Rosenberger, Amy Myers, Eugene Boman, Gary Gordon, Liz McMahon, Timothy G. Feeman, and Garth Isaak. 7 We will refer to the Carter Petitioners and the Gressman Petitioners collectively as “Petitioners.”

[J-20-2022] - 5 Petition for Review, 12/17/2021, at ¶10. The Gressman Petitioners also designated

Respondents as the opposing parties.

The petitions for review were substantially similar in their alleged facts, claims

presented, and relief requested. Factually, Petitioners asserted that this Court in LWV III,

utilized data from the 2010 Census when we adopted the 2018 congressional district plan

(“2018 Plan”), which appropriately divided the Commonwealth into eighteen districts.

Petitioners, however, explained that the 2020 Census reflected a population shift that

resulted in the Commonwealth losing one of its congressional districts, rendering the 2018

Plan unconstitutionally malapportioned.

Stated broadly, Petitioners claimed that the 2018 Plan violated their state and

federal rights to cast undiluted votes. In terms of relief, Petitioners asked the

Commonwealth Court to: (1) deem the 2018 Plan unconstitutional; (2) enjoin

Respondents and related parties from implementing, enforcing, or giving effect to that

plan; and (3) adopt a constitutionally acceptable congressional district plan in time for the

impending 2022 election cycle.

On December 20, 2021, the Commonwealth Court consolidated the petitions for

review and, in a separate order, established a process, in compliance with this Court’s

prior decision in Mellow, supra, by, inter alia, setting deadlines for: (1) the filing of

applications to intervene; (2) submitting proposed seventeen-district congressional

reapportionment plans consistent with constitutional principles and the 2020 Census; and

(3) conducting hearings in the event that the court would be required to choose a new

map due to political gridlock.

The following day, December 21, 2021, Petitioners filed in this Court Applications

for Extraordinary Relief. In those applications, Petitioners asked this Court, inter alia, to

exercise its extraordinary jurisdiction pursuant to 42 Pa.C.S. § 726 and Pa.R.A.P. 3309

[J-20-2022] - 6 to address expeditiously the merits of the claims that they presented in their petitions for

review.8 This Court eventually denied those applications without prejudice to reapply for

similar relief, as future developments might dictate.

While these applications were pending in this Court, the Commonwealth Court held

a hearing on the ten applications to intervene that had been filed in that court. By order

dated January 14, 2022, the court set new deadlines regarding the judicial process that

would address the petitions for review, and it granted intervenor status to the following

applicants: (1) the Speaker and Majority Leader of the Pennsylvania House of

Representatives; (2) the President Pro Tempore and Majority Leader of the Pennsylvania

Senate; (3) Pennsylvania State Senators Maria Collett, Katie J. Muth, Sharif Street, and

Anthony H. Williams; (4) Tom Wolf, Governor of the Commonwealth of Pennsylvania; (5)

Senator Jay Costa and members of the Democratic Caucus of the Senate of

Pennsylvania; (6) Representative Joanna E. McClinton, Leader of the Democratic Caucus

of the Pennsylvania House of Representatives; and (7) Congressman Guy

Reschenthaler, Swatara Township Commissioner Jeffrey Varner, Tom Marino, Ryan

Costello, and Bud Shuster.

The Commonwealth Court directed that these intervenors would participate in the

litigation as parties. The court directed all parties to submit at least one but no more than

8 Section 726 of the Pennsylvania Judicial Code provides as follows:

Notwithstanding any other provision of law, the Supreme Court may, on its own motion or upon petition of any party, in any matter pending before any court or magisterial district judge of this Commonwealth involving an issue of immediate public importance, assume plenary jurisdiction of such matter at any stage thereof and enter a final order or otherwise cause right and justice to be done.

42 Pa.C.S. § 726. Pennsylvania Rule of Appellate Procedure 3309 explains the process for applying for relief under 42 Pa.C.S. § 726.

[J-20-2022] - 7 two proposed congressional redistricting plans, along with a supporting brief and/or an

expert report by January 24, 2022. The court also required each party to file a responsive

brief and/or expert report by January 26, 2022. In addition, the court directed these

parties to submit a joint stipulation of facts, and the court set January 27th and 28th of 2022

as the dates of the evidentiary hearings on this matter. Concerning those hearings, the

court explained that each of the parties would be permitted to present one witness and to

cross-examine the other parties’ witnesses.

In the same order, the Commonwealth Court granted amicus status to the following

applicants: (1) Voters of the Commonwealth of Pennsylvania; (2) Citizen-Voters; (3) Draw

the Lines-PA; and (4) Khalif Ali et al. The court limited the amicus participants’ litigation

contribution to the submission of one proposed congressional redistricting plan and a

supporting brief and/or expert report.

Subsequently, the parties and amici submitted congressional redistricting maps,

expert reports, and briefs in support thereof. The Commonwealth Court held hearings on

January 27th and 28th of 2022, at which numerous experts testified.

On January 29, 2022, the Carter Petitioners filed in this Court another Application

for Extraordinary Relief, requesting that this Court immediately assume jurisdiction over

the redistricting litigation. By order dated February 2, 2022, this Court granted the Carter

Petitioners’ Application for Extraordinary Relief, obtaining original jurisdiction over the

matter.

In conformance with this Court’s decision in Mellow, supra, we: (1) designated as

a Special Master the Honorable Patricia A. McCullough, the Commonwealth Court judge

who was presiding over the matter when we assumed plenary jurisdiction; (2) explained

that the proceedings that already had occurred in the Commonwealth Court shall be

considered part of the Special Master’s record; (3) directed the Special Master to file in

[J-20-2022] - 8 this Court on or before February 7, 2022, a report containing proposed findings of fact

and conclusions of law supporting her recommendation of a redistricting plan; and (4) set

a schedule for the parties and amicus participants to file exceptions and briefs in this

Court.

On February 7, 2022, the Special Master submitted her comprehensive report.

While we do not provide a detailed summary of that report, we highlight that the report

deemed the 2018 Plan constitutionally deficient because, inter alia, it created boundaries

for eighteen congressional seats based upon the 2010 Census but the 2020 Census

resulted in Pennsylvania being limited to seventeen congressional seats. The report

further observed that the General Assembly and Governor were unable to agree upon a

congressional redistricting plan to replace the 2018 Plan, thus, thrusting upon the

Pennsylvania judiciary the task of selecting such a plan.

The Special Master ultimately received thirteen congressional redistricting plans

to study. Although the Special Master used several metrics to choose the most desirable

plan, she eliminated multiple plans from consideration due to the following alleged

shortcomings: (1) the splitting of the City of Pittsburgh into separate districts; (2) the

yielding of a partisan advantage contrary to Pennsylvania’s political geography; and (3)

the failure to achieve a maximum population deviation of one person.

Regarding the remaining plans, the Special Master ultimately chose H.B. 2146 to

replace the 2018 Plan. As will be discussed in more detail infra, the Special Master

appears to have given H.B. 2146 preferential treatment because “it is the General

Assembly’s prerogative, rather its constitutional mandate, to redraw the state’s

congressional districts under Article I, section 4 of the United States Constitution and its

related provisions in the Pennsylvania Constitution and state statutes.” Report at 208,

¶62; id. at 214, ¶94 (“The Court believes that, in the context of this case, where it must

[J-20-2022] - 9 recommend one map of many, as a matter of necessity, the interests of the

Commonwealth as a sovereign state and political entity in its own right, would best be

served by factoring in and considering that H.B. 2146 is functionally tantamount to the

voice and will of the People[.]”).

While the Special Master provided her recommendation of a congressional district

plan, we are mindful that this Court obtained original jurisdiction over this litigation when

we granted the Carter Petitioners’ Application for Extraordinary Relief; accordingly, our

scope of review of the matter is de novo. LWV II, 178 A.3d at 801 n.62. While Judge

McCullough’s findings of fact are not binding on this Court, they are afforded due

consideration, as she presided over the evidentiary hearing. Id.

In accordance with this Court’s order of February 2, 2022, the following parties and

amicus participants have filed exceptions in this Court: (1) Carter Petitioners; (2)

Gressman Petitioners; (3) Respondents; (4) Congressman Guy Reschenthaler, Swatara

Township Commissioner Jeffrey Varner, Tom Marino, Ryan Costello, and Bud Shuster;

(5) Senator Jay Costa and members of the Democratic Caucus of the Senate of

Pennsylvania; (6) Tom Wolf, Governor of the Commonwealth of Pennsylvania; (7)

Representative Joanna E. McClinton, Leader of the Democratic Caucus of the

Pennsylvania House of Representatives; (8) Khalif Ali et al.; (9) Citizen-Voters; and (10)

Draw the Lines-PA.

In relevant part, the exceptions challenge the way that the Special Master

eliminated plans and the criteria that she utilized in choosing H.B. 2146. For example,

several of the parties and amici are of the view that it was error for the Special Master to

reject plans because they split the City of Pittsburgh, attempted to accomplish partisan

fairness, or failed to achieve a maximum population deviation of one person. Some also

[J-20-2022] - 10 insist, inter alia, that the Special Master erroneously favored H.B. 2146 simply because it

was produced by the Legislature.

The following parties have filed briefs in support of the Special Master’s Report:

(1) Voters of the Commonwealth of Pennsylvania;9 (2) the Speaker and Majority Leader

of the Pennsylvania House of Representatives; and (3) the President Pro Tempore and

Majority Leader of the Pennsylvania Senate. Lastly, the following parties filed amicus

briefs in the Court: (1) Philadelphia County Board of Elections; (2) Washington County

Public Officials; (3) Concerned Citizens for Democracy; and (4) Williamsport/Lycoming

Chamber of Commerce and Greater Susquehanna Valley Chamber of Commerce.

On February 18, 2022, this Court heard argument on the parties’ exceptions to the

Special Master’s Report. We would like to extend our gratitude to the parties and their

counsel who participated in that hearing. Their submissions and advocacy have greatly

aided this Court in completing the task of selecting an appropriate redistricting plan.

III. Case Law

In Mellow, supra, we explained that Pennsylvania lost two congressional districts

following the 1990 census, and the General Assembly failed to enact a timely remedial

reapportionment plan. State senators subsequently filed an action in the Commonwealth

Court seeking: (1) a declaration that the existing congressional apportionment law was

unconstitutional; (2) an injunction to enjoin the implementation of the congressional

election until a valid plan could be adopted; and (3) the adoption of a valid plan in the

event the Legislature was unable to do so. Mellow, 607 A.2d at 205. Upon the senators’

request, this Court assumed plenary jurisdiction over the matter and designated a

Commonwealth Court judge as special master to conduct hearings, to make findings of

fact, and to render conclusions of law. Id. at 206.

9The Voters of the Commonwealth of Pennsylvania additionally advocated in favor of the map they submitted.

[J-20-2022] - 11 In Mellow, this Court adopted the master’s factual findings, as well as his

recommended decision regarding the selection of one of the six congressional

redistricting plans submitted. Id. Initially, the Court examined the master’s reasons for

recommending the plan, i.e., the plan had a low maximum population deviation, contained

minimal splits of municipalities, achieved an enlarged number of congressional districts

with a majority African American population, and came closest to implementing factors

relating to communities of interest. Id. at 206. The Court proceeded to resolve numerous

exceptions to the master’s report filed by the parties, ultimately concluding that the

master’s conclusions of law were sound. Notably, the Court then addressed what it

termed as “Additional Criteria,” which included an examination of the political fairness of

the plan, finding that the plan “results in a politically fair balance in the Pennsylvania

delegation between Democrats and Republicans,” considering that it divided the two-seat

congressional loss equally between both parties. Id. at 210.

Following Mellow, which was decided in 1992, this Court, once again, was faced

with having to adopt a congressional redistricting map under the circumstances presented

in our seminal 2018 decision in LWV II. Unlike the instant case, where the General

Assembly and the Governor failed to enact a redistricting map after a change in

Pennsylvania’s population resulted in the loss of a congressional district, voters in LWV II

commenced an action in the Commonwealth Court challenging an existing congressional

redistricting plan enacted in 2011 (“2011 Plan”). The petitioners alleged, inter alia, that

the 2011 Plan violated the Free and Equal Elections Clause of Article I, Section 5 of the

Pennsylvania Constitution by intentionally discriminating against the petitioners and other

Democratic voters by using redistricting to maximize Republican congressional seats and

entrench Republican power. LWV II, 178 A.3d at 766. They contended that the 2011

Plan had the actual discriminatory effect of disadvantaging Democratic voters and

[J-20-2022] - 12 burdening severely their representational rights. Petitioners thereafter filed an application

for extraordinary relief in this Court.

We granted the application, assumed plenary jurisdiction, and remanded the

matter to the Commonwealth Court for the creation of an evidentiary record. Upon review

of the findings of fact and conclusions of law submitted by then-Judge, now-Justice,

Brobson, this Court, on January 22, 2018, entered a per curiam order: (1) declaring that

the 2011 Plan clearly, plainly, and palpably violated the Pennsylvania Constitution; (2)

striking the 2011 Plan as unconstitutional; and (3) enjoining its use at the May 2018

primary election. League of Women Voters of Pennsylvania, 175 A.3d 282, 289 (Pa.

2018) (“LWV I”). Our per curiam order further afforded the General Assembly the

opportunity to submit a congressional districting plan that comported with our state

charter, if approved by the Governor. Absent such submission, the Court declared that it

would proceed expeditiously to adopt a plan based on the evidentiary record developed

in the Commonwealth Court.10 Id. at 290. No such plan was ever adopted by the

Legislature.

In our subsequent opinion in support of our per curiam order, the Court explained

that the “Free and Equal Elections Clause was specifically intended to equalize the power

of voters in our Commonwealth’s election process, and it explicitly confers this

guarantee[.]” LWV II, 178 A.3d at 812. In determining how to assess a claim alleging

congressional vote dilution under the Free and Equal Elections Clause of the state

charter, the Court turned to the neutral criteria that traditionally governed the formation of

the Commonwealth’s state legislative districts, as set forth in Article 2, Section 16 of the

Pennsylvania Constitution. Id. at 815-16. These criteria require an examination of

whether the congressional districts created under the redistricting plan: (1) are composed

10This author filed a concurring and dissenting statement, and then-Chief Justice Saylor and Justice Mundy filed dissenting statements.

[J-20-2022] - 13 of compact territory; (2) are comprised of contiguous territory; (3) are as nearly equal in

population as practicable; and (4) do not divide any county, city, incorporated town,

borough, township, or ward, except where necessary to ensure equality of population

(collectively, “traditional core criteria”). Id. at 816-17. We explained that these criteria

emphasize greatly the creation of representational districts that “maintain the

geographical and social cohesion of the communities in which people live and conduct

the majority of their day-to-day affairs,” and “accord equal weight to the votes of residents

in each of the various districts.” Id. at 814.

Finding these traditional core criteria to be “deeply rooted in the organic law of our

Commonwealth,” and the “foundational requirements which state legislative districts must

meet under the Pennsylvania Constitution,” the Court adopted them as a measure to

assess whether a congressional districting plan dilutes the potency of a voter’s ability to

select his or her preferred congressional representative in violation of the Free and Equal

Elections Clause. Id. at 816. We explained that these traditional core criteria provide a

“floor” of protection against the dilution of one’s vote and that the subordination of these

criteria to extraneous considerations, such as partisan gerrymandering, is

unconstitutional. Id. at 817. Additionally, we observed that congressional districting

maps must also comply with federal law, specifically, the Voting Rights Act, 52 U.S.C.

§ 10301. Id. at 817 n.72.

The Court in LWV II further recognized additional factors that have historically

played a role in the creation of legislative districts, such as “the preservation of prior

district lines, the protection of incumbents, and the maintenance of the political balance

which existed after the prior reapportionment.” Id. at 817. Additionally recognized as a

subordinate historical factor was the preservation of communities of interest because

“[w]hen an individual is grouped with other members of his or her community in a

[J-20-2022] - 14 congressional district for purposes of voting, the commonality of the interests shared with

the other voters in the community increases the ability of the individual to elect a

congressional representative for the district who reflects his or her personal preferences.”

Id. at 816.

We clarified that these historical factors are wholly subordinate to the traditional

core criteria requiring compact and contiguous districts, minimization of the division of

political subdivisions, and maintenance of population equality among congressional

districts. Id. at 817. We will refer to these factors as “subordinate historical

considerations.”

Relevant here, we recognized that “there exists the possibility that advances in

map drawing technology and analytical software can potentially allow mapmakers, in the

future, to engineer congressional districting maps, which, although minimally comporting

with these [traditional core] criteria, nevertheless operate to unfairly dilute the power of a

particular group’s votes for a congressional representative.” Id. (referencing trial

testimony discussing the concept of an efficiency gap metric used to determine partisan

fairness based upon the number of “wasted” votes for the minority political party under a

particular redistricting plan). Because this Court was resolving LWV II based exclusively

on the degree to which the traditional core criteria were subordinated to pursue partisan

advantage, we did not discuss a means by which to differentiate among myriad

redistricting plans that, on their face, satisfy the traditional core criteria. Id.

Applying this jurisprudence to the 2011 Plan, the Court in LWV II concluded that it

clearly violated the traditional core criteria, thereby depriving the petitioners of their state

constitutional right to free and equal elections. Id. at 818. The Court found that the 2011

Plan revealed “tortuously drawn districts that cause plainly unnecessary political-

subdivision splits,” and “oddly shaped, sprawling districts which wander seemingly

[J-20-2022] - 15 arbitrarily across Pennsylvania, leaving 28 counties, 68 political subdivisions, and

numerous wards, divided among as many as five congressional districts, in their wakes.”

Id. at 819. We emphasized that the congressional districts “often rend municipalities from

their surrounding metropolitan areas and quizzically divide small municipalities which

could easily be incorporated into single districts without detriment to the traditional

redistricting criteria.” Id. Accordingly, we concluded that the 2011 Plan did not comply

with traditional core redistricting criteria and, thus, violated the Free and Equal Elections

Clause. Id. at 820.

As to the appropriate remedy in LWV II, the Court acknowledged that while the

primary responsibility for apportioning congressional districts rests with the General

Assembly, it becomes the judiciary’s task to determine the appropriate redistricting plan

when the Legislature is unable or chooses not to act. Id. at 821-22. Accordingly, based

upon both state and federal case law, we found sufficient authority for this Court to

formulate a valid redistricting plan.11 Id. at 824.

The Court thereafter prepared a constitutionally sound plan, i.e., the 2018 Plan,

which was implemented for the May 2018 primary election. LWV III, supra. The 2018

Plan was based upon the record developed in the Commonwealth Court and relied

significantly upon the submissions provided by the parties, intervenors, and amici.

LWV III, 181 A.3d at 1087. In LWV III, this Court found that the 2018 Plan satisfied the

traditional core criteria as it split only 13 counties, four of which are split into three districts

and nine of which are split into two districts. Id. The Court opined that the 2018 Plan was

“superior or comparable” to all plans submitted in compactness, by whichever calculation

methodology was employed. Id. Finally, the Court observed that the 2018 Plan achieves

the constitutional guarantee of one person, one vote. Id.

11This author filed a concurring and dissenting opinion, and then-Chief Justice Saylor and Justice Mundy filed dissenting opinions.

[J-20-2022] - 16 IV. Special Master Recommendation and Exceptions to Special Master’s Report

Based upon the processes and guidelines set forth in Mellow and the LWV

decisions, we turn to our review of the Special Master’s Report and recommendation and

the numerous exceptions and responses filed by the parties and amici. For the reasons

set forth below, we respectfully declined to adopt the Special Master’s recommendation

to select H.B. 2146. Below, we focus upon the following three aspects of the Special

Master’s analysis: (1) the Special Master’s conclusion that certain plans improperly

yielded a partisan advantage to the Democratic Party contrary to Pennsylvania’s political

geography; (2) the Special Master’s finding that certain plans failed to achieve a maximum

population deviation of one person; and (3) the Special Master’s preferential treatment of

H.B. 2146. As discussed below, we respectfully disagree with the reasons provided for

narrowing the plans on these bases. Thus, the exceptions filed by the parties and amici

to the Special Master’s Report are sustained in part, consistent with the following analysis.

1. Partisan Advantage

Several of the exceptions challenge the Special Master’s discrediting of six of the

thirteen maps for “yield[ing] a partisan advantage to the Democratic Party” based upon

either their mean-median scores or their efficiency gap scores, which, as discussed infra,

are generally accepted metrics for evaluating the partisan fairness of a redistricting plan.12

Report at 197, ¶ 41-42. The Report viewed this asserted partisan advantage as contrary

to the “natural and undisputed Republican tilt” in the Commonwealth resulting from the

clustering of Democratic voters in the urban areas. Id. at ¶ 40 The Special Master

deemed the drawing of district lines to negate this tilt to be “a subspecies of unfair

12 Specifically, the Report gave “less weight” to the Gressman Plan, the House Democratic Caucus, the Carter Plan, the Governor’s Plan, the Senate Democratic Caucus 2 Plan, the House Democratic Caucus Plan, and the Draw the Lines Plan because these plans provide a “partisan advantage to the Democratic Party.” Report at 197, ¶ 41-42.

[J-20-2022] - 17 gerrymandering.” Id. She explained, “[A]ny map that prioritizes proportional election

outcomes, for example, by negating the natural geographic disadvantage, to achieve

proportionality at the expense of traditional redistricting criteria, violates” the Free and

Equal Elections Clause. Id. at 198, 44. Nevertheless, while discounting these six maps

due to the absence of a sufficient “Republican tilt,” the Special Master credited H.B. 2146

for the same attribute, observing that the Republican majority in the General Assembly

“developed and proposed a plan, H.B. 2146, that favors Democrats, which ultimately

underscores the partisan fairness of the plan.” Report at 211, ¶ 79; 216. ¶ 97.

Respectfully, we reject this contradictory logic, which uses partisan advantage to

discredit some but not all plans. Moreover, the record does not support the conclusion

that all of the enumerated maps in fact “prioritized proportional election outcomes” at the

expense of the traditional core criteria, given the various maps’ exceptional performances

on these criteria. Instead, it appears that the mapmakers were cognizant of this Court’s

expressed concern that maps could be engineered in the future to meet the requisite

traditional core criteria while operating to dilute votes. LWV II, 178 A.3d at 817. Indeed,

we conclude that consideration of partisan fairness, when selecting a plan among several

that meet the traditional core criteria, is necessary to ensure that a congressional plan is

reflective of and responsive to the partisan preferences of the Commonwealth’s voters.

Thus, for purposes of our review, we return these six plans to the same status as the

other submitted plans.13

2. Population Deviation

The Special Master further discounted the two plans that failed to reach a

maximum population deviation of one person, despite finding that all the proposed plans

13We additionally credit Dr. Jonathan Rodden’s observation that “it is not the case that the human geography in Pennsylvania somehow requires that we draw unfair districts.” Transcript of Jan. 27, 2022 (“Tr.”) at 192-93.

[J-20-2022] - 18 satisfied the constitutional requirement that congressional districts be created “as nearly

equal in population as practicable.”14 Report, at 138, CL 1; see PA. CONST. art. II, § 16;

U.S. CONST. art. I, § 2. In other words, while the districts in most of the plans deviated by

only one person, the two discounted plans deviated from the ideal district population of

764,865 by plus one person or minus one person.15

While we acknowledge that the Special Master is justified in flagging these plans

due to their slightly greater population deviation, we respectfully disagree that a

population deviation of an additional person serves as an indelible mark against these

plans. Rather, under the relevant case law discussed infra, a failure to achieve the lowest

population deviation requires further investigation into the justification for the population

deviation. See Karcher v. Daggett, 462 U.S. 725, 740 (1983). The Special Master,

however, did not engage in any such analysis. Accordingly, we conclude that it was

improper to discredit these two plans without considering the reasons for the minor

population deviation. Indeed, as set forth in detail infra, we ultimately conclude that the

Carter Petitioners sufficiently justified the deviation present in their plan of plus or minus

one person.

3. Preferential Treatment of H.B. 2146

After rejecting the majority of the plans based, inter alia, upon their alleged

“Democratic partisan advantage” or the two-person population deviation, the Special

Master was left with four plans to consider, Voters of the Commonwealth of Pennsylvania,

Reschenthaler 1, Reschenthaler 2, and H.B. 2146. According to the Special Master,

Republican Legislative Intervenors requested that some degree of deference be given to

14The two plans discounted under this rationale were the Carter Plan and the House Democratic Plan. 15The ideal district population is determined by dividing the Commonwealth’s population as determined by the 2020 Census, which is 13,002,700, by the seventeen allotted districts, which results in a population of 764,864.7. Report at 3, n.6.

[J-20-2022] - 19 H.B. 2146 because it had gone through the legislative process and was passed by the

Legislature. The Special Master initially indicated that she would not afford H.B. 2146

any special deference and instead would assess the plan the same as the other parties

and amici and their respective maps.

Nevertheless, the Special Master ultimately recommended the adoption of H.B.

2146, emphasizing that “the decisions and policy choices expressed by the legislative

branch are presumptively reasonable and legitimate, absent a showing of an

unconstitutional defect or deficiency.” Report. at 213, ¶ 90 (citing Upham v. Seamon, 456

U.S. 37, 41-42 (1982)). The Special Master reasoned that “H.B. 2146 represents [t]he

policies and preference of the state, and constitutes a profound depiction of what the

voters in the Commonwealth of Pennsylvania desire, through the representative model of

our republic and democratic form of government, when compared to the Governor or any

other of the parties or their amici.” Id. at 214, ¶ 93 (internal quotation marks and citations

omitted).

To the extent that the Special Master’s recommendation was premised upon

bestowing H.B. 2146 preferential treatment simply because it had made it partway

through the legislative process, we reject her endorsement of this plan on this basis alone.

Upham, relied upon by the Special Master in affording H.B. 2146 special consideration,

is readily distinguishable from the present matter. There, the United States Supreme

Court was tasked with reviewing a district court’s decision to reject a congressional

reapportionment plan in favor of its own drafted plan. Importantly, the at-issue plan had

already been duly enacted and was awaiting preclearance from the United States

Attorney General when a suit was filed in the federal district court, challenging the

constitutionality of the reapportionment plan and its validity under the Voting Rights Act.

Thus, Upham, unlike this case, involved a fully-enacted plan that was not vetoed by the

[J-20-2022] - 20 Governor.16 Moreover, by relying upon Upham, the Special Master ignored a separate

line of cases where courts have, in similar circumstances, declined to afford deference to

vetoed plans.17

In our view, declining to afford preferential treatment to a plan passed by the

Legislature but vetoed by the Governor is not only logical, see Cartsen, 543 F. Supp. at

79 (observing that if it were to accept the argument that a vetoed redistricting plan should

receive priority during deliberations, “a partisan state legislature could simply pass any

bill it wanted, wait for a gubernatorial veto, file suit on the issue and have the Court defer

to their proposal”), but also comports with this Commonwealth’s constitutional precepts.18

16 A second case cited by the Special Master, Perry v. Perez, 565 U.S. 388 (2012), is likewise distinguishable, as that case also involved a challenge to new electoral plans that had already been duly enacted. See Perry, 565 U.S. at 391-92 (reviewing the implementation of interim maps that were allegedly inconsistent with the State of Texas’ enacted plans). 17 See, e.g., Johnson v. Wisconsin Elections Commn., 967 N.W.2d 469, 490 n.8 (Wis. 2021) (“The legislature asks us to use the maps it passed during this redistricting cycle as a starting point, characterizing them as an expression of ‘the policies and preferences of the State[.]’ The legislature's argument fails because the recent legislation did not survive the political process.”) (internal citations omitted); Carstens v. Lamm, 543 F.Supp. 68, 79 (D. Colo. 1982) (affording no deference to vetoed redistricting plan and instead, regarding “the plans submitted by both the Legislature and the Governor as ‘proffered current policy’ rather than clear expressions of state policy”) (footnote omitted); and Hippert v. Ritchie, 813 N.W.2d 374, 379, n. 6 (Minn. 2012) (acknowledging that in Perry, supra, the United States Supreme Court held that a federal district court, when creating an interim congressional redistricting plan, should defer to the duly enacted redistricting plan, but finding that in this case, the legislature’s redistricting plan was not entitled to such deference because it “was never enacted into law”). 18 As this Court explained in Scarnati v. Wolf, 173 A.3d 1110, 1120 (Pa. 2017), by “conferring upon the Governor the authority to nullify legislation that has passed both legislative houses, [Pa. Const. art. IV,] Section 15 entrusts him with the obligation both to examine the provisions of the legislation within the ten days allotted by Section 15 and to either approve it or return it, disapproved, for legislative reconsideration.”). Consequently, the Governor is “an integral part of the lawmaking power of the state.” Id. (internal quotation marks omitted). See also id. (observing that “[n]o bill may become law without first being submitted to the Governor for approval or disapproval”).

[J-20-2022] - 21 Finally, by disregarding the Governor’s veto and affording H.B. 2146 preference

because it purportedly represented “the will of the people,” the Special Master improperly

elevated the General Assembly’s role in passing legislation over that of the Executive

Branch, which is an inappropriate departure from basic constitutional principles of checks

and balances, see, e.g., Carstens, 543 F. Supp. at 79 (finding that the legislature’s vetoed

plan, while certainly entitled to careful consideration, could not “represent current state

policy any more than the Governor’s proposal” because “[b]oth the Governor and the

General Assembly are integral and indispensable parts of the legislative process”), and

offensive to the separation-of-powers doctrine.

V. Standard for Choosing New Redistricting Plan

Having rejected the Special Master’s process of winnowing the maps, we review

these maps de novo under this Court’s precedent in Mellow and LWV II. In selecting one

of the various congressional districting plans submitted by the parties and amici, we find

ourselves bound by the same commands that the Legislature must satisfy when

performing such task. First and foremost, we begin, with the traditional core criteria of

ensuring that the districts are compact, contiguous, are as nearly equal in population as

practicable, and do not divide any county, city, incorporated town, borough, township, or

ward, except where necessary. LWV II, at 178 A.3d at 816- 17. As noted, these

traditional core criteria provide a “‘floor’ of protection for an individual against the dilution

of his or her vote in the creation of [congressional] districts.” Id. at 817.

Second, we may also examine the subordinate historical considerations, including,

inter alia, communities of interests, the preservation of prior district lines, and the

protection of incumbents. Id. As noted, we must keep in mind that these factors are

wholly subordinate to the traditional core criteria. Id.

[J-20-2022] - 22 Third, we ensure that the congressional districting plan does not violate

Pennsylvania’s Free and Equal Elections Clause by “diluting the potency of an individual’s

ability to select the congressional representative of his or her choice.” LWV II, 178 A.3d

at 816. While the traditional core criteria protect against the creation of obviously

gerrymandered districts, such as those present in the 2011 Plan, they do not necessarily

prevent all forms of vote dilution. As noted supra, this Court observed in LWV II that

“advances in map drawing technology and analytical software can potentially allow

mapmakers, in the future, to engineer congressional districting maps, which, although

minimally comporting with these [traditional core] criteria, nevertheless operate to unfairly

dilute the power of a particular group’s vote for a congressional representative.” Id.

Partisan fairness metrics provide tools for objective evaluation of proposed congressional

districting plans to determine their political fairness and avoid vote dilution based on

political affiliation.

Fourth, and finally, in adopting a congressional redistricting plan, we guarantee

that the dictates of the Voting Rights Act, 52 U.S.C. § 10301, have been respected.

As mentioned throughout, many of the plans submitted by the parties and the amici

curiae satisfy these rigorous standards set forth in LWV II. Moreover, as demonstrated

in our respected colleagues’ responsive opinions, reasonable minds may disagree as to

which of these plans best balances the designated criteria and considerations.

Nevertheless, having been thrust into the position of choosing a redistricting plan due to

the political stalemate between the Legislature and the Governor, we applied the

aforementioned designated criteria and considerations and selected the Carter Plan as

the 2022 Congressional Redistricting Plan. Our reasons for doing so follow.

VI. Adoption of Carter Plan

[J-20-2022] - 23 We initially observe that the parties and their experts generally agree on the

metrics to be used in judging a plan’s performance on the traditional core criteria, the

subordinate historical considerations, and the evaluations of partisan fairness. However,

through no fault of the experts, the results of these metrics vary based on differences in

their application of the metrics and divergences in the data sets. For example, the

seemingly simple task of counting how many counties are split by a plan varies between

experts based on their assessment of a naturally noncontiguous piece of Chester

County.19 Additionally, some of the standards used to evaluate partisan fairness vary

based upon how many past elections are included in the relevant dataset. Given these

variations, we rely upon the analyses performed by Dr. Daryl DeFord, which evaluate all

of the submitted plans using the same methods and data sets.20 See, inter alia, Exh. 1

of Post-Trial Submission of Gressman Petitioners. We appreciate Dr. DeFord’s efforts in

this regard as it allows the Court to engage in an apples-to-apples comparison of the

plans on each metric.

A. Description of the Carter Plan

The Carter Plan was created by Dr. Jonathan Rodden, 21 who submitted an expert

report and testified as to his decision-making process at the hearing in this case. Dr.

Rodden explained that he used the 2018 Plan “as a guide” with the goal of “preserving

19 As described by one of the experts, a small portion of Chester County is rendered “technically non-contiguous” if the boundary between Chester County and Delaware County is used as a district boundary. In such case, that six-person portion of Chester County is “marooned in Delaware County due to a bend in the Brandywine Creek at the intersection with the [s]outhern state boundary.” Expert Report of Jonathan Rodden (“Rodden Report”) at 21. While some experts included this in the count of county splits, others did not. 20 Dr. DeFord is an assistant professor of data analytics at Washington State University. 21Dr. Rodden is a professor of political science at Stanford University and director of the Stanford Spatial Social Science Lab.

[J-20-2022] - 24 the cores and boundaries of districts where feasible given equal population requirements

and meeting or surpassing [the 2018 Plan’s] adherence to traditional districting criteria[.]”

Rodden Report at 1.

He opined that the 2018 Plan was a “reasonable starting point” because it

“performed very well according to traditional redistricting criteria,” observing that it “was a

compact plan” that involved “relatively few county splits and other jurisdictional splits.”

Rodden Report at 6; Tr. at 88. He additionally recognized that the 2018 Plan “was broadly

recognized” as a fair plan by those who study redistricting, following its use in the 2018

and 2020 elections. Id. at 89. He observed that it “produce[d] relatively competitive

elections” with “outcomes that are roughly in line with overall partisan preferences of

Pennsylvania voters.”22 Rodden Report at 6.

Dr. Rodden provided a detailed district-by-district assessment of the adjustments

needed to achieve population equality, given the different rates of population growth.

Rodden Report at 8-9, 12-20. He additionally explained the rationale behind each

decision to alter district boundaries, with due consideration paid to the give and take

between traditional core criteria which require maximizing compactness and minimizing

county splits. Id.

In adjusting the 2018 Plan to the population changes of the 2020 Census, Dr.

Rodden observed that Pennsylvania’s urban areas, especially in Southeastern

Pennsylvania, “have experienced population growth on par with the United States as a

whole” in the years since the 2010 Census. Rodden Report at 1. As a result, only minimal

adjustments in the 2018 Plan boundaries were needed for the urban districts in

Southeastern and Southwestern Pennsylvania to achieve the population targets under

22In those elections, the average Democratic vote share was 52.7 percent, and the Pennsylvania congressional delegation was split evenly between Republicans and Democrats, with several competitive districts. Rodden Report at 4.

[J-20-2022] - 25 the 2020 Census. However, the “precipitous decline in population” in the rural areas of

Central Pennsylvania required more substantial changes in those districts to achieve the

necessary equal population, resulting in the absorption of former-District 12 of the 2018

Plan into the surrounding districts, Districts 9, 15, and 13. Id. at 1, 20.

Dr. Rodden expressly stated that he “did not consider racial data [when] drawing

districts or making adjustments for population changes in the map.” Rodden Report at

23. Likewise, he explained that he “did not consider partisan performance” when drawing

the map. Id. However, after completing the map, he “was asked to evaluate the districts’

partisan performance,” which he deemed to be “consistent with and responsive to

Pennsylvania voters’ partisan preferences.” Id. at 1. As incorporated into the discussion

below, Dr. Rodden also addressed the plan’s performance on the requisite traditional core

criteria as well as the subordinate historical considerations.

B. Special Master’s Rejection of the Carter Plan

The Special Master rejected the Carter Plan, reasoning that in using the 2018

Plan, the Carter Plan erroneously elevated the subordinate historical considerations of

preservation of prior district lines above the traditional core criteria, in violation of this

Court’s decision in LWV II, which held that the historical considerations are “wholly

subordinate” to the traditional core criteria. Report at 183, CL 2 (quoting LWV II, 178 A.3d

at 817); 187, FF10. Specifically, she faulted the Carter Plan for “opting to draw less

compact districts instead of disrupting” the district boundaries of the 2018 Plan. Id. at

186, FF9. Additionally, while acknowledging that the so-called “least-change” approach

may be appropriate when applied to a legislatively enacted plan, the Special Master

concluded that “choosing a plan based upon its similarity to a previously court-drawn

redistricting plan is not constitutionally sound.” Id. at CL 5. She theorized that use of the

least-change approach for a court map could allow a court to adopt continuously “features

[J-20-2022] - 26 of its prior plan, effectively rendering impossible any future challenge to the plan.” Id. at

188, FF 11.

Respectfully, this Court does not view the Carter Plan’s utilization of the 2018 Plan

as a starting point to be either a prerequisite or a disqualifying attribute. Instead, we deem

it to be one of several reasonable starting points. Such method is particularly useful here,

considering that the 2018 Plan was adopted only four years ago and in strict conformity

with the traditional core criteria explicated in LWV II. LWV III, 181 A.3d at 1086-87. Thus,

the 2018 Plan provided a reasonable starting point of contiguous and compact districts

that minimized divisions of political subdivisions, even if it no longer provided districts of

equal population.

Our decision to adopt the Carter Plan, however, is not based upon its starting point

but rather its end point. Stated another way, we do not select the Carter Plan because it

utilized the least change approach but because the least change approach worked in this

case to produce a map that satisfies the requisite traditional core criteria while balancing

the subordinate historical considerations and resulted in a plan that is reflective of and

responsive to the partisan preferences of the Commonwealth’s voters, as set forth below.

C. Traditional Core Criteria

1. Contiguity

Starting with the simplest and least contentious of the traditional core criteria, the

seventeen districts in the Carter Plan, like every map submitted, are all contiguous.

2. Compactness

Turning to compactness, we find that all of the submitted plans are on a higher

plane of compactness than the unconstitutional 2011 Plan with its “oddly shaped,

sprawling districts which wander seemingly arbitrarily across Pennsylvania.” LWV II, 178

A.3d at 819. Moreover, utilizing the various accepted metrics, the submitted maps are all

[J-20-2022] - 27 within a relatively narrow range comparable to the 2018 Plan, which this Court deemed

constitutionally sufficient.23

While well within the range of the submitted plans, we acknowledge that the Carter

Plan is slightly less compact than some of the other maps. We discount, however, the

Special Master’s suggestion that any reduction in compactness resulted from adherence

to the 2018 Plan lines. Instead, minor reductions resulted from a trade-off acknowledged

by numerous experts between two of the traditional core criteria: compactness and

minimization of political subdivision splits. It is easily comprehended that adherence to

county and city lines will decrease compactness because many of the boundaries follow

geographic features such as rivers, which meander across our Commonwealth. A

mapmaker must, therefore, balance more compact districts with respect for the integrity

of political subdivisions.

In our view, Dr. Rodden’s Report sufficiently justifies the slightly less-compact

aspect of the Carter Plan by explaining various decision points where he sacrificed

compactness in favor of unifying counties or other political subdivisions. Rodden Report

at 8-9, 12-20, 22-23, Tr. at 105-06. Additionally, we recognize that the Carter Plan is less

compact in part due to the decision to keep Pittsburgh within a single district. Rather than

utilizing a relatively smooth dividing line, the Carter Plan traces Pittsburgh’s jagged city

23 Several metrics are used to evaluate compactness, each testing a slightly different aspect of that concept. We need not delve into the details of the computations of these accepted metrics, which are not contested, but rather look broadly to the results across the metrics. Specifically, using the Mean Polsby-Popper metric in which larger scores indicate greater compactness, the submitted maps range from 0.27 to 0.38, with the Carter Plan scoring 0.31 and the 2018 Plan scoring 0.32. On the Mean Reock score, under which higher scores again indicate greater compactness, the submitted maps range from 0.38 to 0.44, with the Carter Plan at 0.41 and the 2018 Plan at 0.43. The Carter Plan again is within the midrange of the Mean Convex Hull metric where larger scores indicate more compact districts, with the maps ranging from 0.75 to 0.81, the Carter Plan at 0.78 and the 2018 Plan at 0.79. Finally, addressing the Cut Edges metric, for which a lower score demonstrates more compact districts, the Carter Plan at 5896 falls within the range of maps from 5,061 to 6821, where the 2018 Plan is at 5,789.

[J-20-2022] - 28 line. Given the thorough explanation for the choices made and the realities of existing

but irregular county and municipality boundaries, we deem the Carter Plan to be

sufficiently compact in comparison to the other submitted plans.

3. Equal Population

The Carter Plan included four districts with a population of 764,865, four districts

with one additional person at 764,866, and nine districts with one less person at 764,864.

Rodden Report at 21. As stated supra, the Special Master found that each proposed plan

satisfied the constitutional requirement that congressional districts be as nearly equal in

population as practicable. Report at 138, CL 1. Nevertheless, she gave less weight to

the Carter Plan because districts in the plan had a maximum deviation of two persons,

whereas some plans achieved a maximum deviation of only one person. As noted above,

we respectfully rejected the Special Master’s discounting based upon its maximum

population deviation, without considering whether the slight difference between the one-

and two-person population deviation was justified.24

Although a challenge under the equal population requirement is not presently

before this Court, the case law is nonetheless instructive in reviewing whether the Carter

Plan sufficiently met the traditional core criterion of equal population. While the criterion

of equal population is exacting and enforced strictly, the United States Supreme Court

has conceded that “precise mathematical equality … may be impossible to achieve in an

imperfect world,” and consequently, the United States Constitution’s equal population

standard requires only that districts be apportioned to achieve population equality “as

nearly as is practicable.” Karcher v. Daggett, 462 U.S. 725, 730 (1983).

24While the Special Master merely gave the Carter Plan less weight, some parties and amici argued that the Carter Plan failed to meet the equal population requirement because nine plans achieved a deviation of one person.

[J-20-2022] - 29 Under the relevant caselaw, a challenge to population equality requires the parties

challenging the proposed plan to show that the population deviation “could have been

reduced or eliminated altogether by a good-faith effort to draw districts of equal

population.” Id. at 730. This burden may be satisfied by the presentation of a plan with

a lower population deviation, particularly where the party being challenged presents an

alternative plan that achieves a lower population deviation. See Vieth v. Pennsylvania,

195 F. Supp. 2d 672, 675–76 (M.D. Pa. 2002) (where defendants themselves presented

a plan with a lower population deviation).

We will assume arguendo that this step is met as the Carter Petitioners appended

to their exceptions filed in this Court a slightly revised plan containing only a one-person

deviation. The reduced deviation was achieved at the expense of an additional split in a

Vote Tabulation District.25 Notably, however, the ability to achieve a lower maximum

population deviation, by itself, does not establish the unconstitutionality of a plan with a

larger deviation. Karcher, 462 U.S. at 740. Rather, the burden merely shifts to the

proponent of the plan to prove “with some specificity” that the deviations in its proposed

plan were necessary to achieve a legitimate state objective. Id. at 740-41.

The specificity required for demonstrating that the deviation was necessary is

flexible and requires a case-by-case consideration of the following factors: the size of the

deviation, the importance of the legitimate state interest necessitating the deviation, the

consistency with which the plan reflects those interests, and whether alternatives might

substantially vindicate those interests yet achieve a smaller deviation. Id. at 741.

25 Dr. Rodden explained that a “vote tabulation district” is the term for the level at which ballots differ between local races. He attempted to minimize these splits because district divisions at this level create difficulties and potential errors for local election officials as they determine which ballot a voter should complete. Tr. at 95-96. He observed that these errors relating to vote tabulation districts can result in voters being provided incorrect ballots, which can have significant consequences in close elections. Tr. at 97.

[J-20-2022] - 30 Accordingly, “the greater the deviation, the more compelling the government's justification

must be.” Vieth, 195 F. Supp. 2d at 677.

While “there are no de minimis population variations,” Karcher, 462 U.S. at 734,

the size of the deviation between a one-person and two-person deviation is as small a

population deviation as is possible and thus results in a low burden of justification. The

Karcher court provided a non-exhaustive list of legislative policies that might justify a slight

population variance, including respecting municipal boundaries and preserving prior

districts. Id. at 740. Since Karcher, federal courts have also recognized a legitimate state

interest in avoiding splitting of election precincts and not unduly departing from “the useful

familiarity of existing districts.” Mellow, 607 A.2d at 206 (Pa. 1992) (collecting cases).

In the brief filed in support of their exceptions in this Court, the Carter Petitioners

explained that their attempts to reach zero deviation required not only the manipulation

of several census blocks, but also the additional split of a Vote Tabulation District at the

intersection of Districts 3 and 5 in South Philadelphia, which the original plan was able to

keep intact. Carter Exceptions Brief at Exhibit A, 2–3. We addressed a similar justification

in Mellow, where the proposed plan fell below others regarding population deviation

precisely because the cost of maximum mathematical equality “require[d] manipulation of

the smallest census unit, the census block.” Mellow, 607 A.2d at 218. In Mellow, we

found that the election administration problems arising from requiring voters in a single

precinct to look to two different sets of congressional candidates “is not a minor one.” Id.

In doing so, we accepted the justification and ultimately adopted a proposed plan with a

larger, but still slight, population deviation than other plans submitted.26 Id.

26In Mellow, the Court adopted a plan with a 63-person maximum population deviation, despite the submission of a plan with only a one-person deviation. In the adopted plan, the smallest district included 565,754 persons, while the largest district had 565,817 persons. Mellow, 607 A.2d at 226 (Appendix A to Opinion of President Judge Craig).

[J-20-2022] - 31 In the present case, the Carter Petitioners have satisfied their burden by stating,

with specificity, that the two-person deviation was required to prevent the additional split

of a Vote Tabulation District. This is a recognized legitimate state interest, and there has

been no evidence nor allegations of bad faith on the part of the Carter Petitioners. The

Carter Plan represents a good-faith effort to draw districts of equal population, and the

two-person deviation was the byproduct of legitimate efforts to limit the number of splits.

Accordingly, the Carter Plan satisfies the equal population requirement and is

comparable, given the very minimal deviation, to the other submitted plans.

4. Splits of Political Subdivisions

While the traditional core criterion of contiguity is very straight forward, it is less

clear how to assess whether a plan has satisfied the requirement that it “not divide any

county, city, incorporated town, borough, township, or ward, except where necessary to

ensure equality of population.” LWV III, 181 A.3d at 1085. In practical terms, there are

only a few political subdivisions which are necessary to split to comply with the maximum

population of a district.

Following the 2022 Census, the ideal population for Pennsylvania’s congressional

districts is 764,865. Thus, only Philadelphia, Allegheny, and Montgomery Counties

exceed this population, and the only city with an excess population is Philadelphia.

Beyond these required divisions, mapmakers must divide the Commonwealth by grouping

together whole political subdivisions with parts of others to achieve the necessary equal

population. Inevitably, there are tradeoffs inherent in this process. A plan that prioritizes

minimizing the number of county splits may well incur more municipality and ward splits

to achieve the critical equal population of the district as a whole. To complicate matters

further, some boroughs span a county line, requiring a mapmaker to choose potentially

between splitting the county or the borough. Additionally, reasonable minds can differ as

[J-20-2022] - 32 to whether it is preferable to split fewer total political subdivisions but to split some in

multiple pieces. For example, Philadelphia’s population requires it to be split in at least

three pieces, but some proposed plans split it into four pieces.

Neither our constitution nor our caselaw provides guidance as to whether the unity

of one type of political subdivision should be prioritized over that of another. Instead, we

observe that these determinations are best left to the political branches, and thus, we do

not rank the order of the importance of the splits. Instead, for the purpose of choosing a

plan, we look wholistically across the plans for a minimization of the splits and for a

justification for the splits to ensure that the decisions were based on valid redistricting

criteria and not for vote dilution purposes.

Turning back to the submitted plans, we emphasize that all of the plans are a far

cry from the unconstitutional 2011 Plan which splintered the Commonwealth, including

the division of twenty-eight counties. LWV II, 178 A.3d at 819. In contrast, the Carter

Plan splits half as many counties. Indeed, Dr. Rodden testified that he prioritized

maintaining the counties as whole entities and, when counties are split, avoiding splitting

them multiple times. While we do not opine as to which division is preferable, we merely

observe that the Carter Plan is one of the best in terms of keeping counties whole and

falls within all other ranges of the plans submitted.27 Accordingly, we conclude that the

Carter Plan is superior or comparable to all the other submitted plans on this criterion.

27 As discussed supra, experts disagree as to how to count the separation of a non- contiguous portion of Chester County from the rest of the county when a plan uses the border between Chester and Delaware Counties as a district boundary. Dr. DeFord’s comparison of the plans’ splits indicates that this split was included in the Carter Plan’s total of 14 divided counties, such that an argument could be made that the Carter Plan should actually be attributed with 13 splits, which would tie for the least split counties. In comparison, the other maps range from 13-17 counties, while the 2018 Plan divided 14.

In terms of city splits, the Carter Plan splits Philadelphia into three pieces as is required by its population but does not fragment it into 4 pieces as do some maps. While

[J-20-2022] - 33 D. Subordinate Historical Considerations

Having determined that the Carter Plan meets or exceeds the other submitted

plans in terms of its adherence to the traditional core criteria, we next consider the

subordinate historical considerations which this Court and other courts have recognized

as relevant considerations in designing a congressional districting plan.

1. Communities of Interest

As discussed above, respect for communities of interest increases an individual’s

ability to elect a congressional representative who reflects his or her personal preferences

based upon “the commonality of the interests shared with the other voters in the

community.” Id. at 816. We observe that the Special Master found that Dr. Rodden “did

not explicitly examine or appear to have considered the specific considerations that need

to be taken into account when establishing that splits maintain the surrounding

communities of interest.” Report at 156, FF12.

Respectfully, we do not read the record to support that finding, given that Dr.

Rodden elucidated several choices that he faced relating to communities of interest. For

example, in forming District 7, he drew the boundaries to “unify Carbon County with the

rest of the Lehigh Valley” and to keep together the Allentown-Bethlehem-Easton area,

which the United States Census Department recognizes as “a metropolitan statistical

area.” Rodden Report at 14, 17. Similarly, the Carter Plan centers District 10 around

Harrisburg, keeping the greater Capital Region intact rather than dividing the area into

the Carter Plan retains Pittsburgh in a single district, it nevertheless splits Williamsport, which is in the range of the other maps which either split one or two cities. While the submitted maps split between 19 and 25 municipalities, the Carter Plan divides 23, and the 2018 Plan separates 29. In terms of wards, the Carter Plan divides 21, which is in the midrange of the submitted maps that divide from 14 to 41; the 2018 Plan split 29. In total, the Carter Plan divides 58 political subdivisions, whereas all the maps range from 49 to 79 total splits. The 2018 Plan had 72 total splits.

[J-20-2022] - 34 multiple district as do some of the plans. The Plan addresses complaints raised regarding

the 2018 Plan, which separated State College from its surrounding area, by placing the

entirety of Centre County in District 15. In addition, unlike several of the plans, the Carter

Plan does not split the City of Pittsburgh, which many, including the Special Master, have

argued results in the division of a community of interest.28

Given the choices made to protect communities of interest, we conclude that the

Carter Plan sufficiently considered this historical redistricting consideration.

2. Preservation of Prior Districts

As has been repeatedly observed by this Court and the United States Supreme

Court, the preservation of prior districts is a legitimate redistricting objective, but one that

is subordinate to the traditional redistricting criteria. See Karcher v. Daggett, 462 U.S.

725, 740 (1983). As discussed supra, the Carter Plan used the 2018 Plan as a starting

point with the intent of preserving district cores and boundaries as much as possible,

given the population changes. The Carter Petitioners argue that the preservation of

districts is beneficial in part because it “create[s] continuity for the overwhelming majority

of Pennsylvania residents.” Carter Plan Brief at 6. The data presented at the hearing

demonstrates that the Carter Plan “laps the field” by ensuring that 86.6 percent of the

population falls in the same district as under the 2018 Plan, while the next highest plan

included only 82.4 percent. Tr. at 407-408 (Dr. Moon Duchin); Rodden Response Report

at 22.

3. Incumbents

28 While we do not view the splitting of Pittsburgh as a disqualifying feature as did the Special Master, we recognize that it is relevant to a plan’s consideration of communities of interest. Moreover, given the history of the recent congressional districting plans, we deem it preferable to retain it within a single district.

[J-20-2022] - 35 A plan’s treatment of incumbents is a relevant consideration because it can reveal

partisan bias where a map protects one party’s incumbents but pairs the other party’s

incumbents against each other, absent other justification.

In this case, the Special Master observed that the Carter Plan pairs two incumbent

Republican representatives, opining that it does so “without any explicit or apparent

justification.” Report at 204. Our review of the record does not support this conclusion.

To the contrary, Dr. Rodden stated that he intentionally considered incumbent addresses

when drawing the Plan to avoid “inadvertently double-bunking sitting congressional

representatives in the same district.” Rodden Report at 23. Moreover, he explained that

the two incumbents paired in District 15 of the Carter Plan resulted from the absorption

of the former-District 12 into District 15 and surrounding districts, which was necessitated

by the significant population loss in Central Pennsylvania since the 2010 Census. We

find this pairing to be justified by the loss of population in this area and not suggestive of

partisan bias, and we further conclude that the Carter Plan pays due consideration to

incumbents.

E. Partisan Fairness

We reiterate this Court’s concern that advances in mapmaking have the potential

to create a plan that will “dilute the power of a particular group’s vote” despite meeting the

traditional core criteria. LWV II, 178 A.3d at 817. Accordingly, we deem it appropriate to

evaluate proposed plans through the use of partisan fairness metrics to ensure that all

voters have “an equal opportunity to translate their votes into representation.” Id. at 814.

In recent years, numerous metrics have been developed to allow for objective

evaluation of proposed districting plans to determine their partisan fairness. For example,

some of the metrics attempt to ascertain a map’s responsiveness to voters, evaluating

whether a party with a majority of votes is likely to win a majority of seats, or whether it is

[J-20-2022] - 36 likely to produce “anti-majoritarian” results, without focus on exact proportionality of

representation. Others attempt to measure whether and to what extent a map favors one

party. In utilizing these tools, we do not prioritize one metric over another, but rather look

wholistically to a plan’s performance across the assessments.

Turning to the Carter Plan specifically, we initially observe that Dr. Rodden

expressly stated that he “did not consider partisan performance” when drawing the map

but instead considered the relevant metrics after it was completed. Rodden Report at 23.

In so doing, he provided detailed assessments of several of the districts. In sum, he views

the Carter Plan as producing “8 districts where Democrats are expected to win, one of

which (District 8) is potentially competitive; 8 districts where Republicans are quite likely

to win, two of which are at least potentially competitive (1 and 10); and one district (District

7) that is a toss-up with a very slight Democratic lean.” Id. at 25.29 Moreover, Dr. Rodden

viewed the Carter Plan as “similar to that of the [2018 Plan], reflective of Pennsylvania’s

statewide partisan preferences, and consistent with changes in population as they relate

to partisanship.” Id. He additionally opined that based on the competitiveness of several

of the districts, the Carter Plan would be responsive to changes in Pennsylvania voters’

partisan preferences. Id.

Dr. Rodden’s assessment is supported by the plan’s performance on the various

metrics. In contrast to some of the submitted plans, the Carter Plan consistently scores

better than average on the measures and equals or surpasses the standards set by the

29 Some of the other parties and amici have oversimplified Dr. Rodden’s assessment as describing a split of ten Democratic seats and seven Republican seats; we reject that view based on Dr. Rodden’s description of the plan, which is further supported by the Carter Plan’s performance on the metric’s discussed below.

[J-20-2022] - 37 2018 Plan.30 Thus, we conclude the Carter Plan is superior or comparable to the other

maps in regard to partisan fairness.

F. Voting Rights Act

While formal Voting Rights Act assessments were not performed in relation to the

submitted plans, the Carter Plan, like all the submitted plans but one, retains the two

majority-minority districts present in the 2018 Plan according to Dr. DeFord’s

assessment.31 Indeed, unlike some of the plans, the Carter Plan’s majority-minority

districts hew closely to the same Philadelphia area districts included in the 2018 Plan,

which to our knowledge has never been challenged as violative of the VRA. As explained

by Dr. Rodden, the boundaries of the Philadelphia area district remained largely

unchanged because the population of this area grew at a similar rate to the United States

as a whole. Rodden Report at 12. Additionally, Dr. Rodden expressly indicated that he

“did not consider racial data [when] drawing districts or making adjustments for population

changes in the map.” Rodden Report at 23. Moreover, no party or amici have raised any

concerns regarding the Carter Plan’s compliance with the VRA.

30 We set forth a few of the partisan fairness metrics. The Carter Plan was one of the best performers on the Majority Responsiveness Metric, where a responsive map is confirmed by a low number of anti-majoritarian elections, which are balanced between the political parties. The Carter Map had only 3 anti-majoritarian elections, with one favoring Democrats and two favoring Republicans. In contrast, H.B. 2146 had one of the highest anti-majoritarian results, with all five favoring Republicans. The Carter Plan had the least biased score (-0.4%) on the average efficiency gap metric, on which negative numbers favor Republicans and positive numbers favor Democrats. The submitted plans ranged from -7.8% to +3.3%, including H.B. 2146 which had one of the highest efficiency gaps favoring Republicans at 6.3%. The 2018 Plan had an average efficiency gap of - 2.6%. In regard to the mean-median metric, upon which numbers closer to zero demonstrate a more balanced plan, the Carter Plan scored -1.6%, which demonstrated a slight Republican tilt, where other plans ranged from -2.9% to -0.3%, with H.B. 2146 having the most significant skew in favor or Republicans at -2.9%. The 2018 Plan had an average mean-median score of -1.9%. 31All other plans submitted also included two majority-minority districts, other than the Gressman Plan which was drawn in part to add an additional majority-minority district.

[J-20-2022] - 38 VII. Conclusion

We reiterate that this Court has been forced into an unusual but not unprecedented

role of selecting a congressional redistricting plan for the impending May 17, 2022,

Primary Election. There is no perfect plan, nor can there be, as many of the criteria work

at cross-purposes to each other and require mapmakers to balance opposing criteria.

Our task is to discern which plan, in our view, best abides by the traditional core criteria

with attention paid to the subordinate historical considerations and awareness of partisan

fairness. As noted, several of the maps submitted would be reasonable choices to be

made by a legislature. After careful consideration and for the reasons set forth above,

we adopt the Carter Plan for the Pennsylvania primary and general elections for seats in

the United States House of Representatives commencing in 2022. We grant, in part, the

exceptions to the extent they are consistent with this opinion and dismiss as moot the

exceptions in all other respects.

Justices Donohue, Dougherty, and Wecht join the opinion.

Justices Donohue, Dougherty, and Wecht file concurring opinions.

Justices Todd, Mundy, and Brobson file dissenting opinions.

[J-20-2022] - 39