Democratic National Committee, DSCC v. Reagan

          AMENDED FINDINGS OF FACT AND CONCLUSIONS OF LAW ^[1]

          Douglas L. Rayes, United States District Judge

         Plaintiffs challenge two aspects of Arizona's election system: (1) Arizona's policy to not count provisional ballots cast in the wrong precinct, which derives from the collective effect of A.R.S. §§ 16-122, -135, -584, and related rules in the Arizona Election Procedures Manual; and (2) Arizona House Bill 2023 (“H.B. 2023”), codified at A.R.S. § 16-1005(H)-(I), which makes it a felony for anyone other than the voter to possess that voter's early mail ballot, unless the possessor falls within a statutorily enumerated exception. Plaintiffs allege that the challenged laws violate § 2 of the Voting Rights Act of 1965 (“VRA”) by adversely and disparately impacting the electoral opportunities of Hispanic, African American, and Native American Arizonans, who Plaintiffs claim are among their core constituencies. Plaintiffs also contend that these provisions violate the First and Fourteenth Amendments to the United States Constitution by severely and unjustifiably burdening voting and associational rights. Lastly, Plaintiffs claim that H.B. 2023 violates § 2 of the VRA and the Fifteenth Amendment to the United States Constitution because it was enacted with the intent to suppress voting by Hispanic and Native American voters. (Doc. 360 at 4-7.)^[2] Plaintiffs seek (1) a declaration that the challenged election practices are unlawful and (2) a permanent injunction requiring Defendants to partially count out-of-precinct (“OOP”) provisional ballots for races for which the voter otherwise was eligible to cast a vote and enjoining Defendants from implementing, enforcing, or giving any effect to H.B. 2023. (Doc. 233 at 41-42.)

         The Court presided over a ten-day bench trial beginning October 3, 2017 and ending October 18, 2017. Pursuant to Federal Rule of Civil Procedure 52, and for the following reasons, the Court finds against Plaintiffs and in favor of Defendants on all claims.^[3]

         I. PARTIES

         Plaintiffs are the Democratic National Committee (“DNC”), the Democratic Senatorial Campaign Committee (“DSCC”), and the Arizona Democratic Party (“ADP”). The DNC is a national committee dedicated to electing local, state, and national candidates of the Democratic Party to public office. The DSCC is a Democratic political committee dedicated to encouraging the election of Democratic Senate candidates to office and is comprised of sitting Democratic members of the United States Senate. The ADP is a state committee dedicated to electing candidates of the Democratic Party to public office throughout Arizona.

         Defendants are Arizona Secretary of State Michele Reagan and Arizona Attorney General Mark Brnovich. Secretary Reagan is Arizona's chief elections officer. Attorney General Brnovich is Arizona's chief legal officer, charged with enforcing state criminal statutes, including H.B. 2023 and other election-related offenses. Secretary Reagan drafts, and Attorney General Brnovich (in conjunction with the Governor of Arizona) approves, the Election Procedures Manual. A.R.S. §§ 41-191 et seq, 16-1021, -452.

         The Court also permitted the following parties to intervene as defendants: (1) the Arizona Republican Party (“ARP”), a state committee dedicated to electing candidates of the Republican Party to public office; (2) Debbie Lesko, who at the time of intervention was an Arizona State Senator representing Arizona's 21st legislative district and Precinct Committeewoman for Arizona's 21st legislative district, and who recently was elected to represent Arizona's 8th congressional district in the United States House of Representatives; (3) Tony Rivero, a member of the Arizona House of Representatives representing Arizona's 21st legislative district; (3) Bill Gates, who at the time of intervention served as a City of Phoenix Councilman and Precinct Committeeman for Arizona's 28th legislative district, and who now serves as a member of the Maricopa County Board of Supervisors representing district 3; and (4) Suzanne Klapp, a City of Scottsdale Councilwoman and Precinct Committeewoman for Arizona's 23rd legislative district. (Docs. 39, 44, 56, 126.)

         II. OVERVIEW OF TRIAL TESTIMONY

         A. Plaintiffs' Expert Witnesses

         1. Dr. Allan Lichtman

         Dr. Allan Lichtman is a Distinguished Professor of History at American University in Washington, D.C., where he has been employed for 42 years. Dr. Lichtman formerly served as Chair of the History Department and Associate Dean of the College of Arts and Sciences at American University. He received his B.A. in History from Brandeis University in 1967 and his Ph.D. in History from Harvard University in 1973, with a specialty in the mathematical analysis of historical data. Dr. Lichtman's areas of expertise include political history, electoral analysis, and historical and quantitative methodology. (Ex. 91 at 3-4.)

         Dr. Lichtman has worked as a consultant or expert witness for plaintiffs and defendants in more than 80 voting and civil rights cases, including League of United Latin American Citizens (LULAC) v. Perry, 548 U.S. 399 (2006), in which Justice Kennedy's majority opinion authoritatively cited Dr. Lichtman's statistical work. Dr. Lichtman also has testified several times for plaintiffs and defendants on issues of intentional discrimination and application of Section 2 in VRA cases. (Ex. 91 at 4.)

         Dr. Lichtman opined, generally, that under the totality of the circumstances H.B. 2023 causes minority voters to have less opportunity to participate in the political process than non-minority voters, and that the law was passed with the intent to suppress minority voters.^[4] He supported his opinions with the standard sources used in political and historical analysis, including scholarly books, articles, reports, newspapers, voter registration and turnout data, and scientific surveys.

         Dr. Lichtman's underlying sources, research, and statistical information are useful. The surveys and data he supplied reveal significant socioeconomic disparities between non-minorities and minorities, including in areas of poverty, unemployment, education, transportation, and health. (Ex. 91 at 3-4.) His report also contains evidence that Arizona exhibits racially polarized voting and has a history of racial appeals in political campaigns that continue to this day. (Ex. 91 at 30, 44-45.) Dr. Lichtman opined that the strong ties between race and partisanship in Arizona make targeting minorities the most effective and efficient way for Republicans to advance their political prospects. (Ex. 93 at 4-5.)

         Although the Court finds Dr. Lichtman's curation of material facts surrounding the legislative history and his underlying research to be helpful and reliable, the Court did not find Dr. Lichtman's ultimate opinions useful. Dr. Lichtman applied the law as he interpreted it to the data he assembled. In this respect, his opinions presented more like an attorney's closing argument than an objective analysis of data, and the credibility of his trial testimony was undermined by his seeming effort to advocate a position rather than answer a question. Moreover, applying law to facts is this Court's duty, and it is one the Court can do without the assistance of an expert opining on how he interprets the law and thinks it should be applied. The Court also has not considered Dr. Lichtman's opinions on the ultimate issue of legislative intent, both because this issue is not the proper subject of expert testimony and because it invades the province of the Court.

         2. Dr. David Berman

         Dr. David Berman is a Professor Emeritus of Political Science and a Senior Research Fellow at the Morrison Institute for Public Policy at Arizona State University. As a political science professor, he has taught undergraduate survey courses in American government and politics, state and local politics, and Arizona government and politics, as well as more specialized courses, including undergraduate seminars on Arizona politics during which students interacted with state and local office holders and political participants. He has also taught advanced graduate courses focusing on research methods in these areas. (Ex. 89 at 3.)

         As a Senior Research fellow with the Morrison Institute, Dr. Berman specializes in research and writing on governance and election issues in Arizona, including redistricting, direct democracy, and campaign finance. He has been a professor at Arizona State University since 1966, and his previous work experience was as a Research Associate at the National League of Cities in Washington, D.C. from 1964 to 1966. (Ex. 89 at 3-4.)

         Dr. Berman opined that Arizona has a long history of discrimination against the voting rights of Native Americans, Hispanics, and African Americans, and that this discrimination is part of a more general pattern of political, social, and economic discrimination against minority groups in areas such as school segregation, educational funding and programming, equal pay and the right to work, and immigration.

         The Court finds Dr. Berman credible. His opinions were well-researched and rendered using standard sources and methodologies in his field of expertise, and his sources were well-identified. Dr. Berman has authored ten books and over 70 published papers, book chapters, or refereed articles dealing with state and local government, politics, and public policy, and his opinions were based substantially on these prior works. In particular, Dr. Berman drew heavily upon his book Arizona Politics and Government: The Quest for Autonomy, Democracy, and Development (University of Nebraska Press, 1998) and his review of archival papers and collections. (Ex. 89 at 3-4.) The Court affords great weight to Dr. Berman's opinions.

         3. Dr. Jonathan Rodden

         Dr. Jonathan Rodden is a tenured Professor of Political Science at Stanford University and the founder and director of the Stanford Spatial Social Science Lab, a center for research and teaching with a focus on the analysis of geo-spatial data in the social sciences. Students and faculty members affiliated with the lab are engaged in a variety of research projects involving large, fine-grained, geo-spatial data sets, including individual records of registered voters, Census data, survey responses, and election results at the level of polling. Prior to joining the Stanford faculty, Dr. Rodden was the Ford Professor of Political Science at the Massachusetts Institute of Technology. He received his Ph.D. from Yale University and his B.A. from the University of Michigan, Ann Arbor, both in political science. (Ex. 95 at 5-6.)

         Dr. Rodden has expertise in the use of large data sets and geographic information systems to analyze aspects of political representation. He has developed a national data set of geo-coded precinct-level election results that has been used extensively in policy-oriented research related to redistricting and representation. He also has worked extensively with Census data from the United States and other countries.

         Dr. Rodden has published papers on political geography and representation in a variety of academic journals and has been featured in popular publications like the Wall Street Journal, the New York Times, and Boston Review. Dr. Rodden has testified as an expert witness in three recent election law cases. (Ex. 95 at 6.)

         Here, Dr. Rodden analyzed the rates and causes of OOP voting in Arizona during the 2012, 2014, and 2016 general elections. The Court finds his use of a combination of individual-level and aggregate data analyses, both of which have been accepted in previous cases analyzing questions under the VRA, to be valid and generally trustworthy, and affords them great weight. (Ex. 97 at 7-9.)

         Dr. Rodden found that Hispanic, Native American, and African American voters cast OOP ballots at statistically higher rates than their non-minority counterparts. (Ex. 95 at 3-4; Ex. 97 at 2-4.) Focusing on Maricopa County in the 2012 election, Dr. Rodden found that the rate of OOP voting was “131 percent higher for Hispanics, 74 percent higher for African Americans, and 39 percent higher for Native Americans than whites.” (Ex. 95 at 3-4.)

         Further, Dr. Rodden found that OOP voters are substantially more likely to be young and to live in neighborhoods characterized by large numbers of renters and with high rates of transience, and that the rate of OOP voting was 65 percent higher for Democratic voters than for Republican voters in Maricopa County, and 56 percent higher in Pima County. Dr. Rodden found that “changes in polling place locations are associated with higher rates of out-of-precinct voting, ” and that “African Americans and Hispanics are substantially more affected by this than whites. In particular, the impact of precinct consolidation, while statistically significant for all groups, is more than twice as large for Hispanics and African Americans as for non-Hispanic whites.” (Ex. 95 at 3-4.) When analyzing Arizona's non-metropolitan counties, Dr. Rodden found that OOP voting is “negligible in majority-white precincts, but increases dramatically in precincts where Hispanics and Native Americans make up majorities.” (Ex. 96 at 58.)

         In addition to his analysis of OOP voting, Dr. Rodden employed standard and accepted methods in his field to analyze the “mailability” of Arizona's non-metropolitan counties in order to estimate the populations that likely would be most affected by H.B. 2023's ballot collection restrictions. Though somewhat imprecise, the Court finds his method of analysis to be creative given the lack of direct data available on the subject, generally reliable, and based on sufficient data given the circumstances. Dr. Rodden found that “[o]utside of Maricopa and Pima counties” “around 86 percent of non-Hispanic whites have home mail service, ” but “only 80 percent of Hispanics do, and only 18 percent of Native Americans have such access.” (Ex. 97 at 4.)

         Dr. Rodden's error rate is unknown, however, due to the lack of direct data. Also, his analysis did not include Arizona's metropolitan counties and therefore does not reveal whether, on a statewide basis, minorities have disparate access to home mail service as compared to non-minorities. Further, mail access is an imprecise proxy for determining the number and demographics of voters who use or rely on ballot collection services. Simply because a voter lacks home mail access does not necessarily mean that she uses or relies on a ballot collector to vote, let alone a ballot collector who does not fall into one of H.B. 2023's exceptions. Accordingly, although Dr. Rodden's analysis provided useful insight into home mail access in non-metropolitan counties, the Court is mindful of its limitations and affords these opinions moderate weight.

         B. Plaintiffs' Lay Witnesses

         Plaintiffs called the following lay witnesses to testify at trial: Carmen Arias, Michael Blair, Delilah Correa, Charlene Fernandez, LeNora Fulton, Steve Gallardo, Kate Gallego, Kathleen Giebelhausen, Marva Gilbreath, Leah Gillespie, Carolyn Glover, Leonard Gorman, Shari Kelso, Scott Konopasek, Joeseph Larios, Daniel Magos, Lori Noonan, Patrick O'Connor, Martin Quezada, Nellie Ruiz, Spencer Scharff, Sam Shaprio, Ken Clark, and John Powers. These witnesses include individual voters, representatives from state, county, and municipal governments, community advocates who have collected ballots as part of get-out-the-vote (“GOTV”) efforts, community advocates focusing of Native American issues, Democratic Party operatives, a California state elections official, and a former United States Department of Justice (“DOJ”) official.

         C. Defendants' Expert Witnesses

         1. Dr. Donald Critchlow

         Dr. Donald Critchlow works at Arizona State University as the Director of the Center for Political Thought and Leadership, an organization funded by a grant from the Charles Koch Foundation. (Tr. 1533-37.) He opined on the relationship between racial discrimination and voting in Arizona. Dr. Critchlow made credible observations that discrimination in Arizona has not been linear and that Arizona has taken effective action to combat discrimination and encourage participation in voting.

         With that said, Dr. Critchlow has never published a book or article focused specifically on Arizona history, nor has he taught courses in Arizona history or politics. (Tr. 1531-32.) Further, in many respects he offered one-sided opinions of Arizona's history, ignored incidents of discrimination, and failed to address the key political shift between the Democratic and Republican parties during the Civil Rights Movement. For example, he either was unfamiliar with or totally discounted the Republican strategy of confrontation of minority voters at the polls during “Operation Eagle Eye” in the 1960s. (Ex. 89 at 16; Tr. 1549.) Additionally, although Dr. Critchlow acknowledged that Arizona has a history of discrimination, his report appears to attribute past racial discrimination in Arizona only to the Democratic Party and claims that discrimination has not existed since the 1960s (in the Republican era). (Ex. 521 at 4.) For these reasons, the Court affords little weight to Dr. Critchlow's opinions

         2. Sean Trende

         Sean Trende critiqued Dr. Lichtman's analysis of Arizona's voting patterns and history of racial discrimination, but offered no new information or analysis. Though the Court found some of his criticisms worth considering, overall they were insignificant. For example, although Trende generally agreed with Dr. Lichtman that Arizona experiences racially polarized voting, he made much of the irrelevant fact that Arizona voting is not as racially polarized as voting in Alabama. (Tr. 1837.) Additionally, Trende's opinions on the weight to give certain evidence and on the proper interpretation and application of the law and evidence-like those of Dr. Lichtman's-were not helpful and invade the province of the Court. Moreover, Trende does not have a Ph.D and has never written a peer-reviewed article. He has spent most of his professional career working as a lawyer or political commentator. He is not a historian and says nothing about the historical methods Dr. Lichtman utilized. (Tr. 1861-62.) For these reasons, the Court affords Trende's opinions little weight.

         3. Dr. M.V. Hood

         Dr. M.V. Hood is a Professor of Political Science at the University of Georgia. Dr. Hood responded to the reports of Drs. Lichtman, Rodden, and Berman. (Ex. 522 at 2-3.) For a number of reasons, the Court affords little weight to Dr. Hood's opinions.

         Dr. Hood criticizes Dr. Berman's use of older historical information. Yet Dr. Critchlow, another expert retained by Defendants, agrees with Dr. Berman that older historical information is relevant to understanding patterns. (Ex. 521 at 8-10; Ex. 522 at 11.) Moreover, Dr. Hood admitted at trial that he examines historical information going back 50 to 200 years. (Tr. 2122-23.)

         Dr. Hood opined that H.B. 2023 does not hinder Native American voting because the rates of early voting on the Navajo Nation increased from 2012 to 2016. He based that opinion on early votes cast in three counties. This opinion is not reliable. Dr. Hood's analysis did not include an assessment of racial disparities and turnout. He also conceded that myriad factors could affect turnout. (Tr. 2111-14.)

         Dr. Hood prepared a cross-state comparative analysis of ballot collection laws and policies related to counting OOP ballots. Although his analysis offered some insight, it overall was not useful because he did not address statutory differences and nuances, and his analysis reflected an incomplete understanding of the laws he categorized. For example, some of the states he labeled as prohibiting ballot collection do not have laws comparable to H.B. 2023 because they prohibit only the delivery of the ballot, not the collection and mailing of the ballot on someone else's behalf. (Ex. 92 at 52-53.)

         The Court also notes that Dr. Hood's testimony either has been rejected or given little weight in numerous other cases due to concerns over its reliability. See Ne. Ohio Coal. for the Homeless v. Husted, No. 2:06-CV-896, 2016 WL 3166251, at *24 (S.D. Ohio June 7, 2016), aff'd in part, rev'd in part, 837 F.3d 612 (6th Cir. 2016); Veasey v. Perry, 71 F.Supp.3d 627, 663 (S.D. Tex. 2014); Frank v. Walker, 17 F.Supp.3d 837, 881-84 (E.D. Wis. 2014), rev'd on other grounds, 768 F.3d 744 (7th Cir. 2014); Fla. v. United States, 885 F.Supp.2d 299, 324 (D.D.C. 2012); Common Cause/Ga. v. Billups, No. 4:05-cv-0201, 2007 WL 7600409, at *14 (N.D.Ga. Sept. 6, 2007). Additionally, most of Dr. Hood's work has been as an expert on behalf of states defending against allegations that their laws violated the Constitution or the VRA. (Tr. 2123-25.)

         4. Dr. Janet Thornton

         Dr. Janet Thornton is a Managing Director at Berkeley Research Group. Dr. Thornton did not conduct her own analysis, but instead offered her opinion that Dr. Rodden's statistical work is flawed. (Ex. 525 at 1.) For example, she challenged Dr. Rodden's approaches to measuring racial disparities in OOP voting. One approach uses individual surname data and geographic coordinates to infer race. Among Dr. Thornton's critiques was the presence of measurement error, which is well-taken. Indeed, even Dr. Rodden concedes measurement error exists, especially as it pertains to African American probabilities. Dr. Thornton did not critique the Hispanic probabilities assessed by Dr. Rodden, however, and Dr. Rodden credibly explained that the measurement error for Hispanic probabilities leads only to the under-estimation of racial disparities.

         The second approach that Dr. Rodden employed relied on data collected by the Census Department on race and ethnicity at the lowest possible level of geographic aggregation. Dr. Thornton's challenge to the aggregate approach was neither about the data nor the presence of racial disparities in OOP voting, but rather the statistical model employed by Dr. Rodden. Dr. Rodden, however, credibly showed that results similar to those reported by his analysis are obtained using the alternative model specification or measurement strategies recommended by Dr. Thornton.

         Dr. Thornton's opinion that there should have been a systematic decline in the number of ballots cast in Arizona's 13 non-metropolitan counties during 2016 if the limits on ballot collection impacted the ability of rural and minority persons to vote is simplistic and not credible. The statistical evidence suggests that increased turnout in rural counties for the 2016 election was driven by non-minority voters, not Native American and Hispanic voters. (Ex. 98 at 21-26.) Moreover, many factors impact voter turnout, including controversial candidates and partisan mobilization efforts, all of which might drown out the potentially deleterious effects of H.B. 2023. Overall, the Court finds that Dr. Thornton's critiques do not significantly undermine Dr. Rodden's opinions and therefore affords them less weight.

         D. Defendants' Lay Witnesses

         Defendants called the following lay witnesses to testify at trial: Brad Nelson, Eric Spencer, Helen Purcell, James Drake, Michael Johnson, Michelle Ugenti-Rita, Amy Chan (formerly Amy Bjelland), Tony Rivero, and Scott Freeman. These witnesses include current and former lawmakers, elections officials, and law enforcement officials.

         E. Witnesses Testifying By Deposition

         In addition to the live testimony, the following witnesses testified by deposition: Sheila Healy, Randy Parraz, Samantha Pstross, Secretary Reagan, Spencer Scharff, Donald Shooter, Eric Spencer, Robyn Stallworth-Pouquette, Alexis Tameron, Victor Vasquez, and Dr. Muer Yang. The parties each raised admissibility objections to certain of these deposition designations. The Court addresses these objections, along with other outstanding evidentiary matters, in a separate order.

         III. OVERVIEW OF CHALLENGED ELECTIONS PRACTICES

         A. H.B. 2023

         Voting in Arizona involves a flexible mixture of early in-person voting, early voting by mail, and traditional, in-person voting at polling places on Election Day. Arizona voters do not need an excuse to vote early and Arizona permits early voting both in person and by mail during the 27 days before an election. A.R.S. § 16-541. For those voters who prefer to vote early and in-person, all Arizona counties operate at least one in-person early voting location. Some of these locations are open on Saturdays. (Doc. 361 ¶ 59.)

         Arizona has allowed early voting by mail for over 25 years, and it has since become the most popular method of voting, accounting for approximately 80 percent of all ballots cast in the 2016 election. In 2007, Arizona implemented permanent no-excuse early voting by mail, known as the Permanent Early Voter List (“PEVL”). Arizonans now may vote early by mail either by requesting an early ballot on an election-by-election basis, or by joining the PEVL, in which case they will be sent an early ballot as a matter of course no later than the first day of the 27-day early voting period. A.R.S. §§ 16-542, -544. In 2002, Arizona also became the first state to make available an online voter registration option, allowing voters to register online through Arizona's Motor Vehicle Division (“MVD”) website, www.servicearizona.com. When registering online through the MVD, voters can enroll in the PEVL by clicking a box. (Doc. 361 ¶ 56.)

         To be counted, an early ballot must be received by the county recorder by 7:00 p.m. on Election Day. A.R.S. § 16-548(A). Early ballots contain instructions that inform voters of the 7:00 p.m. deadline. Voters may return their early ballots by mail postage-free, but they must mail them early enough to ensure that they are received by this deadline. Additionally, some Arizona counties provide special drop boxes for early ballots, and voters in all counties may return their early ballots in person at any polling place, vote center, or authorized election official's office without waiting in line. (Doc. 361 ¶¶ 57, 61.)

         Since 1997, it has been the law in Arizona that “[o]nly the elector may be in possession of that elector's unvoted early ballot.” A.R.S. § 16-542(D). In 2016, Arizona amended A.R.S. § 16-1005 by enacting H.B. 2023, which limits who may collect a voter's voted or unvoted early ballot:

H. A person who knowingly collects voted or unvoted early ballots from another person is guilty of a class 6 felony. An election official, a United States postal service worker or any other person who is allowed by law to transmit United States mail is deemed not to have collected an early ballot if the official, worker or other person is engaged in official duties.

I. Subsection H of this section does not apply to:

1. An election held by a special taxing district formed pursuant to title 48 for the purpose of protecting or providing services to agricultural lands or crops and that is authorized to conduct elections pursuant to title 48.

2. A family member, household member or caregiver of the voter. For the purposes of this paragraph:

(a) “Caregiver” means a person who provides medical or health care assistance to the voter in a residence, nursing care institution, hospice facility, assisted living center, assisted living facility, assisted living home, residential care institution, adult day health care facility or adult foster care home.

(b) “Collects” means to gain possession or control of an early ballot.

(c) “Family member” means a person who is related to the voter by blood, marriage, adoption or legal guardianship.

(d) “Household member” means a person who resides at the same residence as the voter.

A.R.S. § 16-1005(H)-(I). Voters therefore may entrust a caregiver, family member, household member, mail carrier, or elections official to return their early ballots, but may not entrust other, unauthorized third parties to do so.

         B. Rejection of OOP Ballots

         Since at least 1970, Arizona has required voters who choose to vote in person on Election Day to cast their ballots in their assigned precinct and has enforced this system by counting only those ballots cast in the correct precinct. (Doc. 361 ¶ 46.) Because elections involve many different overlapping jurisdictions, the precinct-based system ensures that each voter receives a ballot reflecting only the races for which that person is entitled to vote. (Ex. 95 at 10.) If a voter arrives at a precinct but does not appear on the precinct register, Arizona allows the voter to cast a provisional ballot. A.R.S. §§ 16-122, -135, -584. After Election Day, county elections officials review all provisional ballots. If a voter's address is determined to be within the precinct, the provisional ballot is counted. Arizona does not count any portion of a provisional ballot cast outside of a voter's correct precinct. A majority of states do not count OOP ballots, putting Arizona well within the mainstream on this issue.^[5] Indeed, at no point has the DOJ objected to this practice, and Plaintiffs object to it for the first time in this case.

         In 2011, Arizona amended its elections code to allow counties to choose whether to conduct elections under the traditional precinct model or to use a “vote center” system. 2011 Ariz. Legis. Serv. Ch. 331 (H.B. 2303) (April 29, 2011) (amending A.R.S. § 16-411). Unlike the precinct-based system, the vote center model requires each vote center to be equipped to print a specific ballot, depending on each voter's particular district, that includes all races for which that voter is eligible to vote. Thus, under a vote center system, voters may cast their ballots at any vote center in the county in which they reside and receive the appropriate ballot. A.R.S. § 16-411(B)(4). Graham, Greenlee, Cochise, Navajo, Yavapai, and Yuma counties have adopted the vote center model. These counties are mostly rural and sparsely populated. Precinct-based voting requirements, such as Arizona's policy to not count OOP ballots, have no impact on voters in these counties. By comparison, the most populous counties in Arizona, such as Maricopa, Pima, and Pinal, currently adhere to the traditional precinct-based model.

         IV. PRELIMINARY ISSUES

         A. Standing

         Article III of the United States Constitution limits federal courts to resolving “Cases” and “Controversies, ” one element of which is standing. To have standing to litigate in federal court, a plaintiff “must have suffered or be imminently threatened with a concrete and particularized ‘injury in fact' that is fairly traceable to the challenged action of the defendant and likely to be redressed by a favorable judicial decision.” Lexmark Int'l, Inc. v. Static Control Components, Inc., -- U.S. --, 134 S.Ct. 1377, 1386 (2014) (citing Lujan v. Defenders of Wildlife, 504 U.S. 555, 560-61 (1992)). Only one plaintiff needs to have standing when only injunctive relief is sought. Crawford v. Marion Cty. Election Bd., 472 F.3d 949, 951 (7th Cir. 2007), aff'd, 553 U.S. 181, 189 n.7 (2008).

         Plaintiffs have organizational standing to challenge the election regulations at issue. Ballot collection was a GOTV strategy used primarily by the Democratic Party to increase electoral participation by otherwise low-efficacy voters. (Tr. 416-26, 632-33, 659, 902, 930; Healy Dep. 28:15-29:13.) H.B. 2023's limitations will require Democratic organizations, such as the ADP, to retool their GOTV strategies and divert more resources to ensure that low-efficacy voters are returning their early mail ballots. Additionally, credible expert testimony shows that minority voters, who tend to vote disproportionately for Democratic candidates, vote OOP at higher rates than non-minority voters. Thus, Arizona's policy to not count OOP ballots places a greater imperative on organizations like the ADP to educate their voters. These are sufficiently concrete and particularized injuries that are fairly traceable to the challenged provisions. See Crawford, 472 F.3d at 951 (“Thus the new law injures the Democratic Party by compelling the party to devote resources to getting to the polls those of its supporters who would otherwise be discouraged by the new law from bothering to vote.”); One Wis. Inst., Inc. v. Nichol, 186 F.Supp.3d 958, 967 (W.D. Wis. 2016) (finding expenditure of resources for educating voters about how to comply with new state voter registration requirements sufficient to establish standing).

         Plaintiffs also have associational standing to challenge these provisions on behalf of their members.

[A]n association has standing to bring suit on behalf of its members when: (a) its members would otherwise have standing to sue in their own right; (b) the interests it seeks to protect are germane to the organization's purpose; and (c) neither the claim asserted nor the relief requested requires the participation of individual members in the lawsuit.

Hunt v. Wash. State Apple Adver. Comm'n, 432 U.S. 333, 343 (1977). A number of self-identified Democratic voters testified either that they have used ballot collection services in the past, or that they have voted OOP. The voting rights of such individuals are germane to Plaintiffs' goal of electing Democratic candidates to local, state, and federal offices. Further, neither the claims asserted nor the relief requested requires individual members to participate in this lawsuit.

         Finally, Plaintiffs' asserted injuries can be redressed by a favorable decision of this Court. “[W]hen a plaintiff challenges the constitutionality of a rule of law, it is the state official designated to enforce that rule who is the proper defendant[.]” Am. Civil Liberties Union v. The Fla. Bar, 999 F.2d 1486, 1490 (11th Cir. 1993). Here, county officials are responsible for counting ballots and verifying proper voter registration, see A.R.S. §§ 16-621(A), -584(E), but Secretary Reagan and Attorney General Brnovich also play a role in determining how OOP ballots are counted. Arizona law requires Secretary Reagan, after consulting with county officials, to “prescribe rules to achieve and maintain the maximum degree of correctness, impartiality, uniformity and efficiency on the procedures for early voting and voting, and of producing, distributing, collecting, counting, tabulating and storing ballots.” A.R.S. § 16-452(A). These rules are prescribed in the Election Procedures Manual and have the force of law. A.R.S. § 16-452(B)-(C). “Any person who does not abide by the Secretary of State's rules is subject to criminal penalties, ” Ariz. Libertarian Party, Inc. v. Bayless, 351 F.3d 1277, 1280 (9th Cir. 2003) (citing A.R.S. § 16-452(C)), and Attorney General Brnovich is authorized to prosecute such violations, A.R.S. § 16-1021. Although county officials are responsible for physically counting ballots, they are not empowered to count or reject ballots at their discretion. Rather, “[a]ll proceedings at the counting center shall be under the direction of the board of supervisors or other officer in charge of elections and shall be conducted in accordance with the approved instructions and procedures manual[.]” A.R.S. § 16-621(A).

         Though the Court cannot require Secretary Reagan and Attorney General Brnovich to physically count OOP ballots for races for which the voter was otherwise eligible to cast a vote, it can require them to prescribe such a procedure in the Election Procedures Manual, which county election officials then would be bound by law to follow. Further, Attorney General Brnovich can ensure compliance with such a directive because he is authorized to prosecute county officials who violate it.

         Likewise, Attorney General Brnovich is empowered to enforce state election laws like H.B. 2023. He is not the only official with such authority; Attorney General Brnovich is authorized to enforce Arizona's election laws “[i]n any election for state office, members of the legislature, justices of the supreme court, judges of the court of appeals or statewide initiative or referendum, ” but in elections for “county, city or town office, community college district governing board, judge or a county, city, or town initiative or referendum, ” that authority resides with “the appropriate county, city or town attorney[.]” A.R.S. § 16-1021. But most elections will include statewide races and therefore Attorney General Brnovich likely will share enforcement authority in most circumstances. Moreover, although Attorney General Brnovich might lack authority to direct the enforcement activities of county and municipal prosecutors, there is no reason to believe that these local law enforcement officials will attempt to enforce H.B. 2023 should the Court declare it unconstitutional or unlawful under the VRA.

         Lastly, although there is no evidence that Secretary Reagan or other state or local elections officials play a direct role in the enforcement of H.B. 2023, Secretary Reagan has some indirect involvement in the law's implementation by virtue of her responsibility for drafting the Election Procedures Manual. If the Court were to enjoin H.B. 2023's implementation and enforcement, the Election Procedures Manual would need to reflect as much.

         B. Effect of Preliminary Appellate Proceedings

         On September 23, 2016, the Court denied Plaintiffs' motion to preliminarily enjoin enforcement of H.B. 2023. (Doc. 204.) On October 4, 2016, the Court also denied Plaintiffs' motion to preliminary enjoin enforcement of H.B. 2023 pending Plaintiffs' appeal of the Court's September 23 order. (Doc. 213.) Plaintiffs thereafter moved the Ninth Circuit Court of Appeals for an injunction pending appeal, which was denied by a three-judge motions panel. Later, on October 28, 2016, a divided three-judge merits panel affirmed the Court's order denying Plaintiffs' preliminary injunction motion. Chief Judge Thomas dissented.

         On November 2, 2016, a majority the Ninth Circuit's non-recused active judges voted to rehear the case en banc. Two days later, a majority of the en banc panel voted to preliminarily enjoin enforcement of H.B. 2023 pending the panel's rehearing, essentially for the reasons provided in Chief Judge Thomas' dissent.^[6] This preliminary injunction was short-lived, however, as the United States Supreme Court stayed the order on November 5, 2016, pending the Ninth Circuit's final disposition of the appeal.

         In light of this history, the parties disagree over the effect that Chief Judge Thomas' dissent should have on the Court's post-trial analysis. As explained during the final pretrial conference, although the Court has considered Chief Judge Thomas' dissent, the Court is not bound by its factual analysis. To date, all appellate proceedings have occurred at the preliminary injunction stage on a less developed factual record. Findings and conclusions rendered at the preliminary injunction stage are just that-preliminary. They do not necessarily preclude the Court from making different findings or conclusions after thorough factual development and a full trial on the merits. Accordingly, although the Court is mindful of Chief Judge Thomas' critiques and their preliminary adoption by a majority of the en banc panel, the Court is not bound to make identical findings and conclusions as those made at a preliminary phase of the litigation.

         And with that, the Court proceeds to the merits.

         V. FIRST AND FOURTEENTH AMENDMENTS^[7]

         “[T]he Constitution of the United States protects the right of all qualified citizens to vote, in state as well as in federal elections.” Reynolds v. Sims, 377 U.S. 533, 554 (1964). Relatedly, the First and Fourteenth Amendments protect the right of the people to associate for political purposes. Tashjian v. Republican Party of Conn., 479 U.S. 208, 214 (1986). “It does not follow, however, that the right to vote in any manner and the right to associate for political purposes . . . are absolute.” Burdick v. Takushi, 504 U.S. 428, 433 (1992). Rather, the Constitution empowers states to regulate the “Times, Places and Manner of holding Elections for Senators and Representatives, ” U.S. Const. art. I, § 4, cl. 1, and states retain “control over the election process for state offices, ” Tashjian, 479 U.S. at 217. “Common sense, as well as constitutional law, compels the conclusion that government must play an active role in structuring elections.” Burdick, 504 U.S. at 433. “[A]s a practical matter, there must be a substantial regulation of elections if they are to be fair and honest and if some sort of order, rather than chaos, is to accompany the democratic processes.” Storer v. Brown, 415 U.S. 724, 730 (1974).

         Like an individual's voting and associational rights, however, a state's power to regulate elections is not absolute; it is “subject to the limitation that [it] may not be exercised in a way that violates other . . . provisions of the Constitution.” Williams v. Rhodes, 393 U.S. 23, 29 (1968); see Wash. State Grange v. Wash. State Republican Party, 552 U.S. 442, 451 (2008). But because all election regulations “invariably impose some burden upon individual voters, ” Burdick, 504 U.S. at 433, “not every voting regulation is subject to strict scrutiny, ” Pub. Integrity Alliance, Inc. v. City of Tucson, 836 F.3d 1019, 1024 (9th Cir. 2016).

Instead, . . . a more flexible standard applies. A court considering a challenge to a state election law must weigh “the character and magnitude of the asserted injury to the rights protected by the First and Fourteenth Amendments that the plaintiff seeks to vindicate” against “the precise interests put forward by the State as justifications for the burden imposed by its rule, ” taking into consideration “the extent to which those interests make it necessary to burden the plaintiff's rights.”

Burdick, 504 U.S. at 434 (quoting Anderson v. Celebrezze, 460 U.S. 780, 789 (1983)). This framework commonly is referred to as the Anderson/Burdick test, after the two Supreme Court decisions from which it derives.

         Under this framework, the degree to which the Court scrutinizes “the propriety of a state election law depends upon the extent to which a challenged regulation burdens First and Fourteenth Amendment rights.” Id. A law that imposes severe burdens is subject to strict scrutiny, meaning it must be narrowly tailored to serve a compelling state interest. Id. “Regulations imposing . . . [l]esser burdens, however, trigger less exacting review, and a State's ‘important regulatory interests' will usually be enough to justify ‘reasonable, nondiscriminatory restrictions.'” Timmons v. Twin Cities Area New Party, 520 U.S. 351, 358 (1997) (quoting Burdick, 504 U.S. at 434); see also Pub. Integrity Alliance, 836 F.3d at 1024 (“Applying these precepts, ‘[w]e have repeatedly upheld as ‘not severe' restrictions that are generally applicable, evenhanded, politically neutral, and protect the reliability and integrity of the election process.'” (quoting Dudum v. Arntz, 640 F.3d 1098, 1106 (9th Cir. 2011)). Additionally, when applying Anderson/Burdick, the Court considers the state's election regime as a whole, including aspects that mitigate the hardships that might be imposed by the challenged provisions. See Ohio Democratic Party v. Husted, 834 F.3d 620, 627 (6th Cir. 2016); see also Crawford v. Marion Cty. Election Bd., 553 U.S. 181, 199 (2008) (considering mitigating aspects of Indiana's election laws).

         A. Application to H.B. 2023

         1. Burden on Voting Rights

         At most, H.B. 2023 minimally burdens Arizona voters as a whole. In fact, the vast majority of Arizona voters are unaffected by the law. Although voting by early mail ballot has steadily increased in Arizona, in any given election there remains a subset of voters who choose to vote in person, either early at a designated early voting site or on Election Day. In-person voters are not impacted by limitations on who may collect early mail ballots. For example, 2, 323, 579 registered voters cast ballots during the 2012 general election. (Ex. 543 at 2.) Of these, 1, 542, 855 submitted early mail ballots, over 99 percent of which were counted. (Ex. 95 at 17.) Thus, roughly a third of all Arizonans voted in person during the 2012 general election. Similarly, approximately 80 percent of ...