Ohio Organizing Collaborative v. Husted

Full title: The Ohio Organizing Collaborative, et al., Plaintiffs, v. Jon Husted, et…

Court: United States District Court, S.D. Ohio, Eastern Division.

Date published: May 24, 2016

189 F. Supp. 3d 708 (S.D. Ohio 2016)

Opinion

Case No. 2:15-cv-1802

05-24-2016

The Ohio Organizing Collaborative, et al., Plaintiffs, v. Jon Husted, et al., Defendants.

Bruce V. Spiva, Ceridwen B. Cherry, Elisabeth Carmel Frost, Koushik Pal, Marc E. Elias, Rhett P. Martin, Perkins Coie LLP, Washington, DC, Amanda R. Callais, Perkins Coie LLP, Pro Hac, Vice, Charles Grant Curtis, Jr., Perkin, Joshua L. Kaul, Perkins Coie LLP, Madison, WI, Donald Joseph McTigue, John Corey Colombo, McTigue & McGinnis LLC, Columbus, OH, for Plaintiffs. Steven T. Voigt, Bridget C. Coontz, Brodi J. Conover, Ryan L. Richardson, Sarah Elaine Pierce, Tiffany L. Carwile, Zachery Paul Keller, Ohio Attorney General's Office, Columbus, OH, for Defendants.

MICHAEL H. WATSON, JUDGE, UNITED STATES DISTRICT COURT

Bruce V. Spiva, Ceridwen B. Cherry, Elisabeth Carmel Frost, Koushik Pal, Marc E. Elias, Rhett P. Martin, Perkins Coie LLP, Washington, DC, Amanda R. Callais, Perkins Coie LLP, Pro Hac, Vice, Charles Grant Curtis, Jr., Perkin, Joshua L. Kaul, Perkins Coie LLP, Madison, WI, Donald Joseph McTigue, John Corey Colombo, McTigue & McGinnis LLC, Columbus, OH, for Plaintiffs.

Steven T. Voigt, Bridget C. Coontz, Brodi J. Conover, Ryan L. Richardson, Sarah Elaine Pierce, Tiffany L. Carwile, Zachery Paul Keller, Ohio Attorney General's Office, Columbus, OH, for Defendants.

FINDINGS OF FACT AND CONCLUSIONS OF LAW

MICHAEL H. WATSON, JUDGE, UNITED STATES DISTRICT COURT

On May 8, 2015, The Ohio Organizing Collaborative ("OOC"), Jordan Isern, Carol Biehle, and Bruce Butcher filed suit against Ohio Secretary of State Jon Husted ("the Secretary") and Ohio Attorney General Mike DeWine in their official capacities, challenging various Ohio election laws, directives, and policies. Compl., ECF No. 1. Thereafter, Plaintiffs amended their complaint to substitute OOC with the Ohio Democratic Party ("ODP"), the Democratic Party of Cuyahoga County ("DPCC"), and the Montgomery County Democratic Party ("MCDP"). Am. Compl., ECF No. 41.

The case proceeded directly to trial on an expedited schedule, including expedited discovery, and the Court held a ten-day bench trial. As explained in greater detail herein, the Court finds that Senate Bill 238's ("S.B. 238") amendments to Ohio Revised Code § 3509.01 reducing the early in-person ("EIP") voting period violate the Fourteenth Amendment to the United States Constitution and Section 2 of the Voting Rights Act of 1965 ("VRA") and enjoins Defendants from enforcing or giving any effect to that provision. The Court finds in favor of Defendants on the remainder of Plaintiffs' claims.

I. Introduction

Plaintiffs challenge the following election laws, directives, and policies: • S.B. 238's reduction of the number of days in the EIP voting period and elimination of same-day registration

• Ohio Revised Code § 3501.10(C)'s limitation of one EIP voting location per county

• Senate Bill 200's ("S.B. 200") change to the law regarding the minimum number of direct recording electronic voting machines ("DRE machines") that counties are required to maintain if they use DREs as their primary voting device and Directive 2014-26's policy regarding the minimum number of DRE machines counties are required to deploy on Election Day

• Senate Bill 205's ("S.B. 205") restrictions on unsolicited absentee ballot mailings (including the prohibition on boards of elections ("BOE") conducting such mailings, the prohibition on including prepaid postage in those mailings, and the limitations on the Secretary's mailings) and the Secretary's policy of excluding certain voters from unsolicited absentee ballot application mailings as outlined in Directive 2014-15

• S.B. 205's and Senate Bill 216's ("S.B. 216") addition of categories of information required to be provided on absentee ballot envelopes and provisional ballot affirmation forms

• S.B. 216's reduction in the cure period for provisional ballots cast due to a lack of identification and its prohibition on elections officials completing on a voter's behalf a provisional ballot affirmation form

• S.B. 216's failure to require BOEs to consolidate multi-precinct poll books

Plaintiffs claim that these laws, directives, and policies, hereinafter referred to as "the challenged provisions," disproportionately burden the right to vote of African Americans, Hispanics, and young people and are therefore unconstitutional and violate the VRA. They further allege that the Ohio General Assembly passed the challenged provisions at least in part with the intent to discriminate against those groups of voters, that S.B. 205 and S.B. 216 violate the Civil Rights Act of 1964 ("CRA"), that portions of S.B. 216 violate procedural due process and equal protection, and that all of the challenged provisions unconstitutionally "fence out" Democratic voters. They seek declaratory and injunctive relief, as well as orders mandating Defendants take certain actions.¹

1 Plaintiffs also seek declaratory and injunctive relief with respect to Directives 2014-14 (instructing BOEs on conducting the general voter records maintenance program) and 2014-18 (instructing BOEs on absentee voting requirements). Am. Compl. 62, ECF No. 41. Plaintiffs have not, however, developed an argument with respect to those specific directives.

As a preliminary matter, while Plaintiffs' claims are based on the alleged effects of the challenged provisions on African Americans, Hispanics, and young people, the evidence adduced relates almost entirely to African Americans. Plaintiffs point to some evidence of the impact of S.B. 238 on young people, but they did not develop an argument based on that evidence. Rather, Plaintiffs focused their arguments on the impact of the challenged provisions on African Americans. The Court therefore finds that Plaintiffs' claims fail with respect to Hispanics and young people and addresses their claims with respect to African Americans only.² The Court also notes that the State of Ohio does not collect, and therefore does not have, information on the racial identity of its voters.³ As explained in greater detail below, the inability to confirm voter race is a key deficiency in the experts' analyses of the impact of the challenged provisions on different racial groups. Although the experts employed various methods to address this issue, the Court is nevertheless forced to evaluate the challenged provisions' burdens on the fundamental right to vote based in part on somewhat speculative expert evidence.

2 In so finding, the Court does not opine on the impact of the challenged provisions on these groups but finds only that Plaintiffs have failed to present sufficient evidence of any such impact.

3 This is presumably due in part to the fact that Ohio has never been subject to the Section Five preclearance requirements of the VRA.

With this in mind, the Court, pursuant to Federal Rule of Civil Procedure 52, makes the following findings of fact and conclusions of law. The Court will begin by briefly describing the parties. It will then identify the witnesses who testified at trial and address the weight afforded to the opinions of the expert witnesses. Next, the Court will briefly describe the relevant history of Ohio's election administration and the voting opportunities currently available to Ohio voters. After addressing some preliminary legal issues, the Court will finally address the merits of Plaintiffs' various claims, addressing each legal theory and challenged provision in turn.

II. Parties

A. Plaintiffs

The Plaintiffs in this case are ODP, DPCC, MCDP (collectively "Democratic Party Plaintiffs"), Jordan Isern, Carol Biehle, and Bruce Butcher ("Reverend Butcher").

The Democratic Party Plaintiffs claim African Americans as one of their key constituencies. Tr. Trans. 31-32, ECF No. 97 (Martin); Tr. Trans. 87, ECF No. 101 (Owens); Id. at 113-14 (Beswick). They engage in get-out-the-vote ("GOTV") efforts, primarily with their constituencies. Tr. Trans. 19-20, 32-34, ECF No. 97 (Martin); Tr. Trans. 90-91, 94-95, ECF No. 101 (Owens); Id. at 117-18, 127-28 (Beswick).

Reverend Butcher is a pastor in Summit County, Ohio, who has been and will continue to be involved in GOTV efforts. Tr. Trans. 102-17, ECF No. 96 (Butcher). Carol Biehle is a voter who votes in a multi-precinct voting location. Id. at 172 (Biehle). Jordan Isern did not testify at trial, nor did Plaintiffs adduce any evidence concerning his role as a plaintiff in this lawsuit.

B. Defendants

The Defendants in this case are Jon Husted and Mike DeWine, both sued in their official capacities. Jon Husted is the Secretary of State of Ohio and, as such, is Ohio's chief election officer. Ohio Rev. Code § 3501.04. Mike DeWine is the Attorney General of Ohio and, as such, represents the State in all legal matters.

III. Plaintiffs' Witnesses

A. Expert Witnesses

1. Dr. Jeffrey Timberlake

Dr. Jeffrey Timberlake is a tenured Associate Professor of Sociology at the University of Cincinnati whose research focuses on, inter alia , racial and ethnic inequalities. PX 109 at 5 (Timberlake Rep.). At trial, the Court admitted Dr. Timberlake as an expert in sociology. Tr. Trans. 197, ECF No. 97. He submitted an initial report evaluating the Senate Factors relevant to establishing a violation of Section 2 of the VRA and analyzing the likely impacts of the challenged provisions on the enfranchisement of minority voters. PX 109 at 1-2 (Timberlake Rep.). Using public data sources and various statistical methods, Dr. Timberlake concluded that African Americans in Ohio are subject to various types of discrimination and inequalities that negatively affect their ability to participate in the political process and that the challenged provisions will disproportionately and negatively impact minorities. PX 109 at 61-62 (Timberlake Rep.). He also submitted a rebuttal report evaluating the expert testimony of Defendants' experts Dr. Trey Hood and Sean Trende. PX 110 (Timberlake Reb.).

Much of Dr. Timberlake's reports examine the impact of the challenged provisions on African American versus white voters. In doing so, Dr. Timberlake employed two different methods to address the lack of information regarding the racial identity of Ohio's voters. See Tr. Trans. 106, ECF No. 97 (Timberlake).

In his initial report, Dr. Timberlake employed a "binning" method to compare, inter alia , the usage rates of various voting methods among minority and non-minority voters. PX 109 at 51-60 (Timberlake Rep.). He classified Ohio's counties by minority and poverty level, placing each of the eighty-eight counties into one of three county types: high minority, low minority/high poverty, and low minority/low poverty. Id. at 7.⁴ Using these groupings, Dr. Timberlake presented data comparing voting behaviors among the county types. Id. at 51-56.⁵

4 "[H]igh minority" counties are the seven counties that have at least 18% minority residents of voting age. Id. These counties feature at least six of Ohio's seven largest cities, are on average 25.9% minority, have an average voting age population ("VAP") of 550,000, and include nearly three-quarters of all of Ohio's minorities. Id. "[L]ow minority/high poverty" counties are the seventeen counties with between 2% and 15% minority VAP and poverty rates of at least 15%. Id. "[L]ow minority/low poverty" counties are the sixty-four counties with between 1.7% and 16.8% minority VAP and poverty rates below 15%. Id.

5 Dr. Timberlake created two errata to address some of the errors in his initial table organizing these groupings. See PX 111 (First Erratum); PX 112 (Second Erratum); Tr. Trans. 111-17, ECF No. 97 (Timberlake); Tr. Trans. 111-16, ECF No. 100 (Timberlake). The second erratum divided the counties using a 15% cut-off and placed two additional counties in the high minority category. Tr. Trans. 111-16, ECF No. 100 (Timberlake). Dr. Timberlake testified that the changes made in the errata were minor and did not alter any of his substantive conclusions. Id. at 116-17.

This binning analysis is subject to the ecological inference problem, or making false inferences about individuals using aggregate-level data. DX 18 at 10 (Hood Reb.); DX 20 at 15 (McCarty Reb.); PX 110 at 2 (Timberlake Reb.). The problem is best illustrated with respect to Dr. Timberlake's evaluation of Golden Week usage:

Although [high minority counties] have a high percentage of minorities relative to other counties in Ohio, non-Hispanic whites still comprise more than 70% of the population even in the high minority county grouping. While one may observe a higher incidence of votes cast during Golden Week compared to the two other county groups, there is absolutely no way in which a researcher can confidently infer that minorities are the voters utilizing early in-person voting at higher rates than whites. In racial terms, we simply do not know who it is that is voting early in-person.

DX 18 at 10 (Hood Reb.); see also Tr. Trans. 100-01, ECF No. 99 (Hood); DX 20 at 15 (McCarty Reb.) ("[W]e cannot assume that because there is heavy usage of Golden Week in counties with high black populations that it must be that blacks use Golden Week more often. It is possible that whites in counties with large black populations use Golden Week more often than whites in other counties."); Tr. Trans. 36, ECF No. 98 (McCarty). The Court finds that the ecological inference problem significantly diminishes the weight of Dr. Timberlake's binning analysis and that such analysis, standing alone, provides little support for the conclusions drawn therefrom. Accordingly, Dr. Timberlake's conclusions with respect to restrictions on absentee ballot mailings and provisional balloting, for which this binning analysis is the primary support, see PX 109 at 55-58 (Timberlake Rep.), are entitled to little weight.

In his rebuttal report, however, Dr. Timberlake addressed this ecological inference problem by employing a different methodology to assess the likely racially disparate impact of the challenged provisions. PX 110 at 2 (Timberlake Reb.). First, by looking to information regarding a voter's address, the date he or she voted, and the method by which he or she voted, Dr. Timberlake determined whether a voter voted during Golden Week and whether it was in-person or by mail. Tr. Trans. 125, ECF No. 97 (Timberlake). He then matched the voter's address to a census block, coded the block, and merged it with census data to get a sense of the racial composition of the block. Id. A census block is the smallest grouping of geography used by the census and tends to have the most homogenous racial and ethnic composition. Id. at 128. He then "divided the number of voters [in the voter files] by the total population in each block to derive the population-adjusted rate of use of Golden Week and [early-in-person] voting for each block." PX 110 at 2 (Timberlake Reb.). Finally, he "generated variables to measure the percent African American in each block, and whether each block was ‘homogenous black’ or ‘homogeneous white’ (100% black or white population, respectively), or ‘nearly homogeneous black’ or ‘nearly homogenous white’ (at least 90% black or white, respectively)." Id. The inference drawn from this method is that if a voter lives in a block that is 100% African American, then that voter must be African American. Id. ; Tr. Trans. 128, ECF No. 97 (Timberlake).

He applied this method to evaluate Golden Week EIP usage in Cuyahoga, Hamilton, and Mahoning Counties in the 2008, 2010, and 2012 elections. PX 110 at 5-6 (Timberlake Reb.). He first examined the linear relationship between the percentage of African Americans in a census block to the percentage of EIP and Golden Week votes cast in that block, finding that in 2008, 2010, and 2012, the voting rate increases as the percentage of African Americans in the block increases. Id. at 5-6, 9-10. Next, he compared Golden Week voting rates in 100% homogenous black and white blocks, finding that the Golden Week voting rate in 100% homogenous black blocks in 2008 and 2012 was greater than in 100% homogenous white blocks. Id. at 5-6. He found the same when comparing nearly homogenous black and white blocks. Id. He also concluded that EIP voting in general in those blocks was higher among blacks than among whites. Id. at 9-10. These findings, he opines, strongly suggest that African Americans voted EIP, and specifically during Golden Week, at a much higher rate than did whites. Id. at 7, 10.

The Court finds this census block analysis credible. Although the analysis relied on data from only three counties, the Court finds it probative, as those counties are three of the largest counties in Ohio and contain nearly two-fifths of Ohio's minority population. See PX 109 at 8 (Timberlake Rep.); PX 110 at 1 (Timberlake Reb.); see also N.A.A.C.P. v. Husted , 768 F.3d 524, 535 (6th Cir.2014), vacated as moot by 2014 WL 10384647⁶ (district court did not clearly err in relying on findings based on an analysis of only five counties because those counties made up one-third of Ohio's population and nearly seventy-three percent of all African Americans living in Ohio).

6 For ease of reading, all additional citations to N.A.A.C.P. v. Husted , 768 F.3d 524, 535 (6th Cir.2014) herein will omit the citation indicating the decision was vacated as moot.

The Court further finds that the census block analysis with respect to Golden Week and EIP usage coupled with the binning analysis of the same provide sufficient support for Dr. Timberlake's conclusions regarding the disproportionate use of Golden Week and EIP voting by African Americans in Ohio. See N.A.A.C.P. v. Husted , 768 F.3d at 534–37 (finding that the district court did not err in relying on a similar census block analysis conducted by Dr. Daniel Smith).

In so finding, the Court notes three things. First, the Court disregards Dr. Timberlake's discussion of Dr. Daniel Smith's expert report in N.A.A.C.P. v. Husted , 43 F.Supp.3d 808 (S.D.Ohio 2014). While Federal Rule of Evidence 703 sometimes permits experts to rely on inadmissible evidence when forming their opinions, it does not permit an expert to bolster his own opinion by testifying that another, non-testifying expert's conclusions are the same as his own. Mike's Train House, Inc. v. Lionel, L.L.C. , 472 F.3d 398, 409 (6th Cir.2006). Nevertheless, the Court independently notes that another judge of this Court and the Sixth Circuit credited Dr. Smith's similar analysis in N.A.A.C.P. v. Husted . See 43 F.Supp.3d at 829–30 ; 768 F.3d at 534–37 (district court did not clearly err in crediting Dr. Smith's analysis). Second, Dr. Timberlake's use of Dr. Roscigno's report does not impact the weight afforded to Dr. Timberlake's opinions. See Tr. Trans. 132,134-36, ECF No. 100 (Timberlake); Tr. Trans. 182-84, ECF No. 97 (Timberlake). Third, Dr. Timberlake's decision to evaluate the 2008 and 2012 elections as opposed to the 2010 and 2014 elections does not diminish the weight afforded to his opinions. Cf. N.A.A.C.P. v. Husted , 768 F.3d at 535–36 (it is incorrect to assume that future elections in Ohio will not contain the same campaign effects as did the 2008 and 2012 elections).

The Court also accepts Dr. Timberlake's findings regarding the socioeconomic disparities between African Americans and whites, racially polarized voting, and statistics regarding the race of elected officials in Ohio. See PX 109 at 12-42, 45-50 (Timberlake Rep.). However, the Court considers his opinions based on these findings in light of the related evidence offered by Dr. Hood. See DX 16 at 2-9 (Hood Reb.).

2. Dr. Muer Yang

Dr. Muer Yang is an assistant professor in the Department of Operations and Supply Chain Management at the Opus College of Business, University of St. Thomas and has been studying voting lines since 2007. PX 113 at 3 (Yang Rep.); Tr. Trans. 162-63, ECF No. 98 (Yang). At trial, the Court admitted Dr. Yang as an expert in operations management. Id. at 165.

Dr. Yang authored an initial report in which he used discrete event simulation models to demonstrate the likely effect of S.B. 200, S.B. 205, and Directive 2014-26 on wait times at voting locations. PX 113. He also authored a rebuttal report criticizing Dr. Allen's use of queuing theory. PX 114.

Dr. Yang concluded that S.B. 200 and Directive 2014-26 will reduce the number of DRE machines available for use, which will, in turn, increase voter wait times on Election Day. PX 113 at 20 (Yang Rep.). The Court credits Dr. Yang's analysis but considers his findings in light of his assumption that counties will adhere to the minimum number of DRE machines required by statute and his failure to account for back-up paper ballot and optical scanning methods in DRE counties. See, e.g. , Tr. Trans. 199, 203, 209, ECF No. 98 (Yang); DX 19 at 12 (Allen Reb.).

3. Dr. Lorraine Minnite

Dr. Lorraine Minnite is an Associate Professor in the Department of Public Policy and Administration at the University of Rutgers, Camden, and her area of expertise includes the incidence and effect of voter fraud in American elections. PX 107 at 1-2 (Minnite Rep.); Tr. Trans. 16-17, ECF No. 102 (Minnite). The Court admitted Dr. Minnite as an expert at trial, but the parties disputed whether she should be admitted as an expert on the incidence and effect of voter fraud in American politics or American elections. Tr. Trans. 18-19, ECF No. 102. Having heard Dr. Minnite's testimony, the Court admits her as an expert on the incidence and effect of voter fraud in American elections.

Dr. Minnite concluded that voter fraud is extremely rare. PX 107 at 1 (Minnite Rep.). The Court finds Dr. Minnite credible and affords her opinions significant weight. In so doing, the Court finds that Defendants' concerns with her report and testimony, specifically her definition of voter fraud, her reliance on only reported cases of voter fraud, and her admission that a state has a rational reason to guard against fraud, do not diminish the credibility of her opinions.

4. Dr. David Canon

Dr. David Canon is a professor and chair of the political science department at the University of Wisconsin, Madison. PX 106 at 2 (Canon Rep.); Tr. Trans. 4-5, ECF No. 101 (Canon). At trial, the Court admitted Dr. Canon as an expert in political science and the impact of election laws on turnout. Tr. Trans. 8, ECF No. 101. Most of Dr. Canon's report is a critique of Defendants' experts' citations to an article that Dr. Canon coauthored.

B. Lay Witnesses

In addition to individual Plaintiffs Reverend Butcher and Carol Biehle, Plaintiffs called a number of Democratic political employees and volunteers: Gregory Beswick, Executive Director of the ODP; Nick Martin, Executive Director of the CCDP; Mark Owens, Chair of the MCDP; Joseph Longley, a former Democratic campaign worker; Terri Taylor, a volunteer for the Democratic Party; Rachel Bowman, a field organizer for President Barack Obama's reelection campaign; Matthew Caffrey, 2009-2011 President of the College Democrats of The Ohio State University and paid worker for President Obama's campaign in 2012; and Andrew Kohn, a paid worker for the Obama for America 2008 campaign. Tr. Trans. 85, 108, ECF No. 101 (Owens and Beswick); Tr. Trans. 5, 69-70, 200, ECF No. 97 (Martin, Longley, Taylor); Tr. Trans. 125, 150, 157-58, 166, ECF No. 96 (Caffrey and Kohn); Tr. Trans. 10, ECF No. 98 (Bowman).

Plaintiffs called two Democratic politicians: former Ohio Senator Nina Turner and Cleveland City Council member Phyllis Cleveland. Tr. Trans. 43, ECF No. 96 (Turner); Tr. Trans. 76, ECF No. 102 (Cleveland).

Plaintiffs also called five Democratic board officials: Brad Cromes, former Deputy Director of the Portage County BOE; Timothy Burke, Chair of the Hamilton County BOE and Democratic Party; William Anthony, Director of the Franklin County BOE; Eben McNair, who is a member of the Cuyahoga County BOE and holds various positions in the CCDP; and Anthony Perlatti, Deputy Director of the Cuyahoga County BOE. Tr. Trans. 4-5, ECF No. 103 (Cromes); Tr. Trans. 152, ECF No. 100 (Burke); Tr. Trans. 196, ECF No. 96 (Anthony); Tr. Trans. 4-5, ECF No. 100 (McNair); Tr. Trans. 217, ECF No. 97 (Perlatti).

IV. Defendants' Witnesses

A. Expert Witnesses

1. Dr. M.V. (Trey) Hood

Dr. Hood is a tenured professor of Political Science and the Director of Graduate Studies of Political Science at the University of Georgia. Tr. Trans. 5, ECF No. 99 (Hood). His research is concentrated in American politics and policy, specifically election administration, southern politics, voting behavior, and racial and ethnic politics. Id. In the past four years, Dr. Hood has offered expert testimony in nine cases, at least eight of which were on behalf of state defendants. DX 15 at 1; Tr. Trans. 170-71, ECF No. 99 (Hood).

At trial, the Court admitted Dr. Hood as an expert in political science, public policy related to election laws, election administration, voter fraud, and voter behavior. Tr. Trans. 12, ECF No. 99. He authored an initial report in which he discusses changes to Ohio's EIP voting procedures, the provision of DRE machines, the administration of multi-precinct voting locations, and the changes to administrative procedures regarding provisional and absentee ballots. See DX 15 at 2. He also authored a rebuttal report in response to Dr. Timberlake's initial report. DX 18.

Dr. Hood concluded that the challenged provisions would not deny African Americans equal opportunity to participate in the political process and/or were supported by legitimate administrative concerns. DX 15 at 40-42 (Hood Rep.); DX 18 at 17 (Hood Reb.). Much of Dr. Hood's report is a summary or restatement of the declarations of elections officials and his opinions based on those declarations. The Court affords those opinions little weight for the following reasons: the declarants were selected by defense counsel; Dr. Hood never personally questioned the declarants or developed the questions asked of them; he did not conduct any follow-up interviews with the declarants or confirm the declarations with hard data; and while the declarations are sworn statements, they were prepared for purposes of litigation. See Tr. Trans. 118-22, ECF No. 99 (Hood).

2. Dr. Nolan McCarty

Dr. Nolan McCarty is a professor of politics and public affairs at Princeton University, and his focus area is "American politics doing quantitative work on legislative and electoral behavior." Tr. Trans. 27-28, ECF No. 98 (McCarty). He was an expert witness on behalf of defendants in N.A.A.C.P. v. Husted , where he examined whether there was a change in voting behavior that could be attributed to the 2012 change in Ohio election laws. Id. at 32. At trial in this case, the Court admitted Dr. McCarty as an expert. Id. at 31.

Dr. McCarty authored a rebuttal report primarily criticizing Dr. Timberlake's binning methodology and evaluation of the 2008 and 2012 elections and offering his own analysis of the effect of the elimination of Golden Week. DX 20.

To evaluate the actual impact of the elimination of Golden Week, Dr. McCarty compared the 2010 midterm election when the challenged provisions were not in place to the 2014 midterm general election when they were in place. DX 20 at 3 (McCarty Reb.); Tr. Trans. 113-17, ECF No. 98 (McCarty). He concluded that the challenged provisions did not impact participation overall between 2010 and 2014, opining that African American EIP voters are highly engaged voters that would not be adversely affected by the elimination of Golden Week. Tr. Trans. 60, 65, 71-73, ECF No. 98 (McCarty); DX 20 at 9-14 (McCarty Reb.).

To address the lack of data on the race of Ohio's voters, Dr. McCarty surmised each voter's race based on his or her surname and the racial composition of the census tract in which he or she resides. DX 20 at 6-8 (McCarty Reb). He used 2000 census data on the distribution of surnames by race to compute the probability that an individual with a particular surname identified as African American in the census, used the 2010 census tract files to estimate the probability of living in each of Ohio's census tracts conditional on identifying as African American, and then computed the probability that a respondent with a particular surname who lives in a particular tract would identify as African American. Id. at 6. Dr. McCarty was provided with a geocoded voter file that had already matched the voter's location to a census tract, and he engaged in the racial-imputation methodology. Tr. Trans. 56, ECF No. 98 (McCarty). According to Dr. McCarty, this methodology is generally accepted in the field of political science and has been used substantially over the past ten years. Tr. Trans. 57, ECF No. 98 (McCarty); see also id. at 93-94 (McCarty re-ran the analysis using different probability thresholds and found no significant change in the results).

Dr. McCarty chose to use census tracts in part because of his concerns about the accuracy of the census block assignments, as they were five years old, and, given that they are so small, the populations of census blocks can change fairly dramatically over that period of time. Tr. Trans. 104-08, ECF No. 97 (McCarty). He felt "very confident" that had he conducted the analysis using census blocks, the thrust of his conclusions would not change. Tr. Trans. 157-58, ECF No. 98 (McCarty). Dr. McCarty's use of census tracts instead of census blocks as part of this method nevertheless calls into question the degree of certainty in determining a voter's race, as census blocks are smaller and much more racially homogenous than census tracts. See id. at 102-12 (McCarty). Additionally, Dr. McCarty conceded that the number of African Americans that he coded was smaller than the percentage of African Americans in the VAP because he was intentionally conservative in his assessment of who is an African American. Id. at 97.

The Court finds that Dr. McCarty's surname methodology and use of census tracts rather than census blocks diminishes somewhat the weight of his analysis of 2010 and 2014 voter turnout. The import of his individual level analyses is further diminished by the fact that they examine midterm elections, which generally produce significantly lower turn out than presidential elections. Furthermore, his comparisons of aggregate rates of early voting across 2010 and 2014 are affected by his failure to account for individuals who registered between 2010 and 2014. See id. at 88, 151 (in his analysis of 2010, unregistered voters were treated as registered voters who did not vote, thereby driving down the 2010 turnout rate). Additionally, Dr. McCarty's failure to compare the ratios of usage of Golden Week between African Americans and whites limits the import of his opinions in determining whether African Americans will be disproportionately affected by the elimination of Golden Week.

3. Sean Trende

Sean Trende holds a B.A. in history and political science, an M.A. in political science, and a J.D., and he has been studying and following United States elections for almost twenty years. Trende Decl. ¶¶ 6-9, ECF No. 76-3. He has been a Senior Elections analyst at RealClearPolitics, a political website, since 2009 and a Senior Columnist for Crystal Ball since January 2014. Id. ¶¶ 10, 13; Tr. Trans. 42, ECF No. 103 (Trende). After hearing Trende's qualifications at trial, the Court orally admitted him as an expert, over Plaintiffs' objection, in the fields of campaigns and elections, voter behavior, voter turnout, demographic trends, and political history. Tr. Trans. 49, ECF No. 103 (Trende).

Trende authored an initial report in which he places the challenged provisions in a national context and examines the existence of a causal link between the challenged provisions and any anticipated changes in African American, as compared to white, voter participation. DX 14; Tr. Trans. 49-50, ECF No. 103 (Trende). He also authored a rebuttal report responding to Dr. Timberlake's initial report. DX 17. As explained in greater detail in the Court's Order on Plaintiffs' motion to exclude Trende as an expert, ECF No. 115, much of Trende's report is irrelevant to the Court's analysis.

4. Dr. Theodore Allen

Dr. Theodore Allen is a tenured associate professor at The Ohio State University, teaching in the area of integrated systems engineering. Tr. Trans. 114-15, ECF No. 102 (Allen). At trial, the Court admitted Dr. Allen as an expert in the field of integrated systems engineering and waiting-line analysis, specifically in elections. Id. at 118. Dr. Allen authored an initial report in which he used queuing theory to evaluate the impact of opening multiple EIP voting locations per county and the merits of S.B. 200's new DRE machine formula. DX 16 (Allen Rep.). He also authored a rebuttal report responding to Dr. Yang's initial report. DX 19.

Dr. Allen opined, inter alia , that splitting early voting resources among multiple EIP voting locations would have caused waiting lines to significantly increase in 2012, and therefore, using multiple EIP voting locations would not, in practice, provide a better voting experience to Ohioans. DX 16 at ¶ 29 (Allen Rep.). The Court finds that Dr. Allen's opinions regarding waiting lines resulting from opening multiple voting locations are entitled to little to moderate weight because his use of queuing analysis, specifically the "M/M/c" formula⁷ , is questionable. See DX 114 at 1-3 (Yang Reb.) (identifying three assumptions that the "M/M/c" formula relies on that do not apply to voting queues); Tr. Trans. 168-69, ECF No. 102 (Allen) (confirming that discrete event simulation can accommodate a much broader set of assumptions than the queuing model he used and conceding that if he had a lot of time and someone to check his work, discrete event simulation would be preferable to queuing theory in conducting the type of analysis in his report); see also id. at 172, 174, 178.

7 The M/M/c formula is a square root staffing formula based on queuing theory that "gives you a sense of how many resources you need as a function of arrival service processes." Tr. Trans. 127, ECF No. 102 (Allen). In this context, the formula considers the number of arrivals at a polling place, the service processes (the number of machines) and the service time (the time it takes to cast a ballot) to estimate wait times. DX 16 at ¶¶ 15, 19 (Allen Rep.).

B. Lay Witnesses

Defendants called four lay witnesses: Matthew Damschroder, Assistant Secretary of State and Chief of Staff to the Secretary; Sherry Poland, Director of the Hamilton County BOE; Timothy Ward, Director of the Madison County BOE; and Mark Munroe, Chair of the Mahoning County BOE. Tr. Trans. 48, ECF No. 104 (Damschroder); Id. at 9 (Poland); Tr. Trans. 223, ECF No. 103 (Ward); Tr. Trans. 94, ECF No. 102 (Munroe).

V. Ohio's Relevant History of Election Administration and Reform

The 2004 general election in Ohio saw several election administration problems, including extremely long lines at the polls. See, e.g. , Tr. Trans. 199, ECF No. 96 (Anthony); N.A.A.C.P. v. Husted , 768 F.3d at 530–31. In response, Ohio created no-excuse absentee voting and provided an early voting period during which voters could vote by mail or EIP. Ohio Rev. Code § 3509.02 ; DX 14E (Directive 2014-18); N.A.A.C.P. v. Husted , 768 F.3d at 531 (discussing Substitute House Bill 234).

After the 2005 reforms, BOEs in Ohio's three largest counties encouraged absentee voting by mail by mailing unsolicited absentee ballot applications and, in some cases, prepaying postage for return mailings. See Tr. Trans. 225-27, ECF No. 96 (Anthony); Tr. Trans. 233, ECF No. 97 (Perlatti); Tr. Trans. 172-73, ECF No. 100 (Burke). Thereafter, a large number of voters took advantage of early voting. PX 109 at 56, Table 6 (Timberlake Rep.).

Nevertheless, voters in Ohio's largest counties still waited in significantly long lines to vote early and on Election Day in 2008 and 2012. Tr. Trans. 205-09, 227-28, ECF No. 96 (Anthony); id. at 57-59 (Turner); Tr. Trans. 143-44, ECF No. 101 (Beswick); Tr. Trans. 229-30, ECF No. 97 (Perlatti); id. at 45-46 (Martin); id. at 207-09 (Taylor); id. at 79 (Longley); Tr. Trans. 16-17, 20, ECF No. 100 (McNair); id. at 100 (Burke); Tr. Trans. 83, ECF No. 102 (Cleveland); Tr. Trans. 23-24, ECF No. 98 (Bowman).

Following the 2012 election, the General Assembly enacted S.B. 238, S.B. 200, S.B. 205, and S.B. 216, all of which are challenged in the instant suit.

In 2014, Ohio organizations challenged S.B. 238's elimination of Golden Week, as well as certain early voting directives, in Ohio State Conference of N.A.A.C.P. v. Husted, 43 F.Supp.3d 808 (S.D.Ohio 2014). The parties ultimately reached a settlement in which the State agreed, inter alia , to provide EIP voting on the final two Saturdays and Sundays before presidential general elections and evening EIP voting hours until 7 p.m. during the final week before those elections. DX 14K at 4. Golden Week was not reinstated, however.

VI. Voting Opportunities Currently Available to Ohio Voters

Ohio voters have several methods by which to vote. Voters may vote in person on Election Day between 6:30 a.m. and 7:30 p.m. Ohio Rev. Code § 3501.32(A). Ohio also allows no-excuse absentee in-person voting before Election Day. For the 2016 general election, Ohio currently will have twenty-three days of EIP voting spread out over four weeks, which includes two Saturdays, two Sundays, and ten days where voting is permitted until either 6:00 p.m. or 7:00 p.m. DX 14J (2016 voting calendar). The period totals 207 hours of EIP voting. DX 14 at ¶ 92 (Trende Rep.). Additionally, Ohio allows no-excuse absentee voting by mail. Voters may request an absentee ballot application and mail it to the BOE, return it in person to the BOE, or have a family member return it to the BOE. Ohio Rev. Code § 3509.05(A).

VII. Preliminary Legal Issues

Before turning to the merits of Plaintiffs' claims, the Court first addresses several legal issues raised by the parties.

A. Standing

"Article III of the Constitution limits federal courts' jurisdiction to certain ‘Cases' and ‘Controversies.’ " Clapper v. Amnesty Int'l USA , –––U.S. ––––, 133 S.Ct. 1138, 1146, 185 L.Ed.2d 264 (2013). "One element of the case-or-controversy requirement is that plaintiffs must establish that they have standing to sue." Id. (internal quotation marks and citations omitted).

Article III standing requires: (1) an injury in fact, (2) fairly traceable to the defendant's conduct, (3) that is likely redressable by a favorable ruling. Sandusky Cnty. Democratic Party v. Blackwell , 387 F.3d 565, 573 (6th Cir.2004) (citations omitted). Each element must be proven with the requisite "degree of evidence required at the successive stage of the litigation." Lujan v. Defenders of Wildlife , 504 U.S. 555, 561, 112 S.Ct. 2130, 119 L.Ed.2d 351 (1992). "The party invoking federal jurisdiction bears the burden of establishing these [standing] elements." Id.

Under certain circumstances, an organization may have standing to sue on behalf of its members (representational standing). Sandusky Cnty. Democratic Party , 387 F.3d at 574. An organization may also have standing to assert an injury to itself where it satisfies the Article III standing requirements (organizational standing). See Am. Canoe Ass'n, Inc. v. City of Louisa Water & Sewer Comm'n , 389 F.3d 536, 544 (6th Cir.2004).

Here, ODP has established representational standing to bring suit on behalf of its members.

An association has standing to bring suit on behalf of its members when its members would otherwise have standing to sue in their own right, the interests at stake are germane to the organization's purpose, and neither the claim asserted nor the relief requested requires the participation of individual members in the lawsuit. The individual participation of an organization's members is not normally necessary when an association seeks prospective or injunctive relief for its members.

Sandusky Cnty. Democratic Party , 387 F.3d at 574 (internal quotation marks and citations omitted).

ODP's claimed key constituencies include African Americans. See Tr. Trans. 31-32, ECF No. 97 (Martin); Tr. Trans. 87, ECF No. 101 (Owens); Id. at 113-14 (Beswick); see also PX 109 at 39-41 (Timberlake Rep.) (opining African Americans in Ohio consistently vote overwhelmingly for Democrats). Although ODP did not adduce evidence of the racial breakdown of its membership, the Court cannot ignore the likelihood that at least some of ODP's members are African American. These members would have standing to sue in their own right. First, Plaintiffs have presented evidence of an injury in fact (a real and imminent burden on the right to vote) traceable to the challenged laws⁸ that would be redressable by a finding that the laws violate the United States Constitution and/or the VRA and the granting of associated injunctive relief. See Sandusky Cnty. Democratic Party , 387 F.3d at 574. Second, the interest in the right to vote is germane to the purpose of the ODP, who engages in GOTV efforts targeted towards its members and constituents in an effort to successfully elect Democratic candidates. See Tr. Trans. 117-18, 127-28, ECF No. 101 (Beswick); id. at 136 (In 2016, ODP intends to engage in GOTV voter-education efforts with respect to the election law changes that occurred in 2014.). Last, neither the claims asserted nor the injunctive relief requested requires the participation of individual members in the lawsuit.⁹

8 Except, as discussed infra , Plaintiffs have not established a burden on the right to vote resulting from S.B. 200 or Ohio Revised Code § 3501.10(C).

9 Defendants claim that ODP cannot claim to represent all of Ohio's African Americans. The Court need not decide whether ODP can claim to represent all of the African Americans in Ohio, however, as it is sufficient that ODP has standing to assert the rights of its African American members.

To the extent that ODP cannot establish representational standing for failure to either identify at least one specific member who will be injured by each of the challenged provisions or provide evidence that all of its members would be injured, see Summers v. Earth Island Inst. , 55 U.S. (14 How.) 488, 497–98, 14 L.Ed. 510 (2009), the Court finds that ODP nevertheless has organizational standing.

ODP has established an injury in fact, as the record reflects that: (1) the challenged provisions will make it more difficult for its members and constituents to vote, which hinders ODP's mission of electing its candidates, and (2) the challenged provisions will force ODP to divert resources from ensuring their members and constituents vote to counteracting the negative effects of the challenged provisions. See, e.g. , Havens Realty Corp. v. Coleman , 455 U.S. 363, 378, 102 S.Ct. 1114, 71 L.Ed.2d 214 (1982) (a drain on an organization's resources constitutes a concrete and demonstrable injury); Miami Valley Fair Hous. Ctr., Inc. v. Connor Grp. , 725 F.3d 571, 576–77 (6th Cir.2013) (same); Common Cause/Georgia v. Billups , 554 F.3d 1340, 1350–51 (11th Cir.2009) (N.A.A.C.P. established injury where evidence showed it would divert resources from its regular activities to educate and assist voters in complying with a new photo identification voting requirement); Florida State Conference of N.A.A.C.P. v. Browning , 522 F.3d 1153, 1160–66 (11th Cir.2008) (organization established actual and imminent injury); Crawford v. Marion Cnty. Election Bd. , 472 F.3d 949, 951 (7th Cir.2007) (same), aff'd, Crawford v. Marion Cnty. Election Bd. , 553 U.S. 181, 189 n. 7, 128 S.Ct. 1610, 170 L.Ed.2d 574 (2008) ; Tr. Trans. 117-18, 127-30, 136-37, 142, 145-48, ECF No. 101 (Beswick).¹⁰ This injury is caused by the challenged provisions and would be redressable by an injunction enjoining the provisions' enforcement.¹¹

10 The Court is cognizant of the Sixth Circuit's opinion in Fair Elections Ohio v. Husted , in which the majority disagreed with the dissent's reliance on Havens and Miami Fair Housing Center for the proposition that the drain or diversion of an organization's resources constitutes an injury for standing purposes. 770 F.3d 456, 460 n. 1 (6th Cir.2014). This does not change the Court's analysis, however. The Sixth Circuit did not hold that the diversion of resources cannot constitute an injury but rather found that the plaintiff failed to establish such an injury with specific evidence. Id. at 460–61.

11 Defendants assert that an organization cannot have organizational standing to bring a VRA or intentional racial discrimination claim, but Defendants have not cited any Sixth Circuit or United States Supreme Court cases explicitly finding the same. Absent such binding authority, the Court declines to find that the Democratic Party Plaintiffs lack organizational standing to bring those claims.

Having found that ODP has representational and, alternatively, organizational standing, the Court need not address whether the remaining plaintiffs have standing. See Am. Civil Liberties Union of Kentucky v. Grayson Cnty., Ky. , 591 F.3d 837, 845 (6th Cir.2010) (citing cases) ("The presence of one party with standing is sufficient."); Crawford , 472 F.3d at 951 (only one plaintiff with standing is required when only injunctive relief is sought), aff'd , Crawford , 553 U.S. at 189 n. 7, 128 S.Ct. 1610. The Court also need not determine whether Plaintiff Carole Biehle has standing to assert her claim regarding the consolidation of poll books because, as explained in greater detail herein, that claim is moot.

B. Federalism, Anti-Commandeering, Constitutional Avoidance, Judicial Restraint, and Laches

Having reviewed the briefing and relevant authority, the Court rejects Defendants' argument that Plaintiffs' claims are precluded by the doctrines of federalism, anti-commandeering, constitutional avoidance, judicial restraint, and/or laches.

C. Effect of N.A.A.C.P. v. Husted

As mentioned above, this is not the first time that S.B. 238 has been challenged. S.B. 238's elimination of Golden Week, as well as certain early voting directives, were litigated in 2014 as part of Ohio State Conference of N.A.A.C.P. v. Husted , 43 F.Supp.3d 808 (S.D.Ohio 2014) (" N.A.A.C.P. v. Husted I "). In that case, another judge of this Court declared S.B. 238 unconstitutional and in violation of § 2 of the VRA and granted a preliminary injunction enjoining its enforcement.

The United States Court of Appeals for the Sixth Circuit affirmed the district court's decision in Ohio State Conference of N.A.A.C.P. v. Husted , 768 F.3d 524 (6th Cir.2014) (" N.A.A.C.P. v. Husted II") The Sixth Circuit later vacated its decision, however, after the United States Supreme Court stayed the decision pending a petition for writ of certiorari. N.A.A.C.P. v. Husted , No. 14–3877, 2014 WL 10384647, at *1 (6th Cir. Oct. 1, 2014). In so doing, the Sixth Circuit noted that the preliminary injunction that was the subject of the appeal was limited to the 2014 election, which no longer had any effect due to the Supreme Court's stay. Id. The parties settled.

Plaintiffs argue the Court should give persuasive, if not preclusive, effect to the district court's decision and the Sixth Circuit's affirmance thereof. Defendants argue that, because the Sixth Circuit's opinion was later vacated, it is not entitled to preclusive effect.

Although the Court finds that the Sixth Circuit's vacated opinion in N.A.A.C.P. v. Husted II is not binding, see, e.g. , Dodrill v. Ludt , 764 F.2d 442 (6th Cir.1985), Fleet Aerospace Corp. v. Holderman , 848 F.2d 720, 722 (6th Cir.1988), the Court is free to find the reasoning therein persuasive. See Matte i v. Matte i , 126 F.3d 794, 801 n. 6 (6th Cir.1997) (finding the reasoning in a vacated and unpublished decision nonetheless persuasive); Barrett v. Harrington , 130 F.3d 246, 258 n. 18 (6th Cir.1997) ("Because this Court only looks to the case as persuasive authority, it is irrelevant that the case has been vacated ...."). Accordingly, the Court will give persuasive effect to the Sixth Circuit's vacated opinion as it sees fit, keeping in mind that the Sixth Circuit's opinion was vacated for reasons unrelated to the merits.

Merits

The Court now turns to the merits of Plaintiffs' claims. Plaintiffs argue that the challenged provisions violate: (1) their rights under the First and Fourteenth Amendments to the United States Constitution pursuant to Anderson /Burdick , (2) Section 2 of the VRA, (3) the Fourth and Fifteenth Amendments to the United States Constitution under an intentional discrimination theory, (4) their rights under the First and Fourteenth Amendments to the United States Constitution under a partisan fencing theory; (5) Section 1971 of the Civil Rights Act of 1964, (6) their procedural due process rights under the Fourteenth Amendment to the United States Constitution, and (7) their equal protection rights under the Fourteenth Amendment to the United States Constitution pursuant to Bush v. Gore . Plaintiffs seek, inter alia , a declaration that the challenged provisions violate the First, Fourteenth, and Fifteenth Amendments to the United States Constitution, the VRA, and the CRA and an injunction enjoining Defendants from enforcing or giving any effect to the challenged provisions.

VIII. Equal Protection (Anderson/Burdick )

Plaintiffs claim the challenged provisions violate the Equal Protection Clause of the Fourteenth Amendment to the United States Constitution by imposing burdens on the voting rights of African Americans that outweigh the interests furthered by the provisions. They bring their claim under 42 U.S.C. § 1983. Voting is a fundamental right, and "[o]ther rights, even the most basic, are illusory if the right to vote is undermined." Wesberry v. Sanders , 376 U.S. 1, 17, 84 S.Ct. 526, 11 L.Ed.2d 481 (1964). "The right to vote is protected in more than the initial allocation of the franchise. Equal protection applies as well to the manner of its exercise.' " League of Women Voters v. Brunner , 548 F.3d 463, 477 (6th Cir.2008) (quoting Bush v. Gore , 531 U.S. 98, 104, 121 S.Ct. 525, 148 L.Ed.2d 388 (2000) ). " ‘[H]aving once granted the right to vote on equal terms'—such as expanding early voting opportunities—‘the State may not, by later arbitrary and disparate treatment, value one person's vote over that of another’—for example, by making it substantially harder for certain groups to vote than others." N.A.A.C.P. v. Husted II , 768 F.3d at 542 n. 4 (quoting Bush v. Gore , 531 U.S. at 104–05, 121 S.Ct. 525 ).

"The Equal Protection Clause applies when a state either classifies voters in disparate ways, or places restrictions on the right to vote." Obama for Am. v. Husted , 697 F.3d 423, 428 (6th Cir.2012) (internal citations omitted). "At the same time, the Constitution vests states with the authority to prescribe " ‘[t]he Times, Places and Manner of holding Elections for Senators and Representatives.’ " NE OH Coalition for the Homeless v. Husted , 696 F.3d 580, 592 (6th Cir.2012) (quoting U.S. Const. Art. I, § 4, cl. 1 ). In the face of an equal protection challenge, the Court balances these competing interests, considering the character of the state action and the nature of the burden it places on voters in setting the standard of review to apply. "While a rational basis standard applies to state regulations that do not burden the fundamental right to vote, strict scrutiny applies when a state's restriction imposes ‘severe’ burdens." Id. (citing Obama for Am. , 697 F.3d at 428 ). The Court applies the flexible Anderson /Burdick balancing test to most cases falling in-between. Id. (citations omitted). "Although Anderson and Burdick were both ballot-access cases, the Supreme Court has confirmed their vitality in a much broader range of voting rights contexts." Obama for Am. , 697 F.3d at 429 (internal citation omitted).

Under the Anderson /Burdick test,

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 plaintiffs' rights."

Burdick v. Takushi , 504 U.S. 428, 434, 112 S.Ct. 2059, 119 L.Ed.2d 245 (1992) (quoting Anderson v. Calebrezze , 460 U.S. 780, 789, 103 S.Ct. 1564, 75 L.Ed.2d 547 (1983) ). "There is no ‘litmus test’ to separate valid from invalid voting regulations; courts must weigh the burden on voters against the state's asserted justifications and ‘make the ‘hard judgment’ that our adversary system demands.' " Obama for Am. , 697 F.3d at 429 (quoting Crawford , 553 U.S. at 190, 128 S.Ct. 1610 (Stevens, J., announcing the judgment of the Court)). "However slight that burden may appear.... it must be justified by relevant and legitimate state interests ‘sufficiently weighty to justify the limitation.’ " Crawford , 553 U.S. at 191, 128 S.Ct. 1610 (citation omitted); Obama for Am. , 697 F.3d at 433.

Notably, "how Ohio's early-voting system compares to that of other states is not relevant under the Anderson – Burdick balancing test." N.A.A.C.P. v. Husted II , 768 F.3d at 546.

In addressing Plaintiffs' equal protection claim with respect to each challenged provision, the Court will describe the relevant provision, evaluate the extent of the burden it imposes, if any, and apply the appropriate standard to determine whether the provision satisfies equal protection despite any imposed burden.

A. S.B. 238: Elimination of Golden Week

Plaintiffs first challenge S.B. 238's elimination of Golden Week.

In 2005, the Ohio General Assembly passed Substitute House Bill 234 ("H.B. 234") in an effort to remedy the manifold problems experienced during the 2004 election. H.B. 234 instituted "no fault" early voting, and BOEs were required to make absentee ballots available no later than thirty-five days before the election. Ohio Rev. Code § 3509.01(B)(2) (2014) (as amended Feb. 25, 2014). As Ohio law requires voters to be registered at least thirty days prior to an election, voters in Ohio had a period during which they could both register and vote on the same day. See Ohio Rev. Code § 3503.01(A) ; DX 15 at 3, 15 (Hood Rep.). The opportunity to register and vote simultaneously is referred to as "same-day registration" ("SDR"), and the period during which voters were allowed to do so is referred to as "Golden Week."

S.B. 238, effective in 2014, amended Ohio Revised Code § 3509.01(B) to make the first day of early voting, by mail and EIP, the day after the close of voter registration. Ohio Rev. Code § 3509.01(B)(2)-(3). In doing so, S.B. 238 eliminated Golden Week and reduced the number of available EIP voting days. See DX 15 at 3 (Hood Rep.). There will currently be three fewer days for EIP voting in the 2016 general election than there were in the 2012 general election, which included the Golden Week period. Compare DX 14H (2012 voting calendar: twenty-six days) with DX 14J (2016 voting calendar: twenty-three days).

For the following reasons, the Court finds that the elimination of Golden Week imposes a modest burden—which the Court defines as a more than minimal but less than significant burden—on the right to vote of African Americans and that the State's justifications for enacting S.B. 238 fail to outweigh that burden. The Court's analysis largely mirrors that of N.A.A.C.P. v. Husted I and N.A.A.C.P. v. Husted II. Recognizing that N.A.A.C.P. v. Husted II was decided in the context of a preliminary injunction, the Court specifically treats it as highly persuasive authority given that the decision was vacated for reasons unrelated to the merits.¹² 1. Burden

12 At trial, Defendants repeatedly asserted that they were successful before the Supreme Court in obtaining a stay of the Sixth Circuit's decision. While the Supreme Court's stay led to the Sixth Circuit vacating its decision, two considerations are worth mentioning. First, in issuing the stay, the Supreme Court did not comment on the merits of the Sixth Circuit's decision. Second, the Supreme Court issued its stay on September 29, 2014, about five weeks before the 2014 general election. On October 8 and October 9, 2014, the Supreme Court issued two orders effectively preventing two other Court of Appeals decisions from altering state election procedures. See Veasey v. Perry , 769 F.3d 890, 894 (5th Cir.2014) (discussing Frank v. Walker , ––– U.S. ––––, 135 S.Ct. 7, 190 L.Ed.2d 245 (2014) and North Carolina v. League of Women Voters of N.C. , ––– U.S. ––––, 135 S.Ct. 6, 190 L.Ed.2d 243 (2014) ). The Supreme Court's actions suggest that the timing of the Sixth Circuit's opinion, which would have required Ohio to change its election procedures so close to the election, may have been a factor in the stay. See Veasey , 769 F.3d at 895 ("While the Supreme Court has not explained its reasons for issuing these stays, the common thread is clearly that the decision of the Court of Appeals would change the rules of the election too soon before the election date.").

The elimination of Golden Week imposes a modest burden on the right to vote of African Americans in two ways: (1) by reducing the overall EIP voting period, and (2) by eliminating the opportunity for SDR.

a. Reduction in the EIP Voting Period

The elimination of Golden Week burdens voters by reducing the overall period for EIP voting. The record reflects that over 60,000 people voted during Golden Week in 2008, and over 80,000 people voted during Golden Week in 2012. PX 109 at 51 (Timberlake Rep.); Tr. Trans. 108, ECF No. 97 (Timberlake). Individuals who would have voted during Golden Week in future elections must now vote on other days during the early voting period, vote absentee by mail, vote on Election Day, or not vote at all. See Tr. Trans. 180, ECF No. 98 (Yang); PX 113 at 12, 16 (Yang Rep.); Tr. Trans. 192, ECF No. 102 (Allen).

The elimination of the extra days for EIP voting provided by Golden Week will disproportionately burden African Americans, as expert and anecdotal evidence reflects that African Americans vote EIP, and specifically EIP during Golden Week, at a significantly higher rate than other voters. For example, individual level data for three of the largest counties in Ohio containing nearly two-fifths of Ohio's minority population—Cuyahoga, Hamilton, and Mahoning—show that in 2008, 2010, 2012, and 2014, "the rate of EIP voting increases with respect to the percentage of African Americans in the census block." PX 110 at 10 (Timberlake Reb.) (emphasis added). It also shows that in 2008 and 2012, "the voting rate [during Golden Week ] gets bigger as the percentage of blacks in the block gets bigger, suggesting that blacks are more likely to use Golden Week EIP voting in those years." Id. at 5-6 (emphasis added). To be sure, it is unknown whether the voters in those blocks are African American or white.

However, looking at homogenous and nearly homogenous black and white blocks, the same individual level data shows that usage rates of EIP voting were far higher among African Americans than among whites in 2008, 2010, 2012, and 2014. See PX 110 at 9-10 (Timberlake Reb.) (EIP voting rate in homogenous black blocks was 4.316 times higher than homogenous white blocks in 2008, 1.390 times higher in 2010, 4.4476 times higher in 2012, and 2.330 times higher in 2014, and similar rates were found for nearly homogenous black blocks versus nearly homogenous white blocks). Defendants' expert admitted that the rate at which African Americans used EIP voting in 2010 and 2014 was slightly higher than the white rate. Tr. Trans. 76-77, ECF No. 98 (McCarty). The data also shows that the usage rates of Golden Week specifically were far higher among African Americans than among whites in both 2008 and 2012. PX 110 at 6 (Timberlake Reb.) (3.514 times higher in 2008 and 5.186 times higher in 2012 for homogenous black blocks; 4.100 times higher in 2008 and 5.584 times higher in 2012 for nearly homogenous black blocks). In other words, in 2008, for example, "the rate of voting early in person during golden week is three and a half times greater in homogeneous black blocks than homogeneous white blocks." Tr. Trans. 128-29, ECF No. 97 (Timberlake).

Additionally, data from the Current Population Survey ("CPS") indicates that "in 2008, 19.9% of Ohio's African American voters made use of EIP voting compared to only 6.2% of whites[,and that] [i]n 2012, 19.6% of blacks used EIP voting versus 8.9% of whites." PX 109 at 54-55 (Timberlake Rep.) (citing CPS Voting and Registration Supplement and Daniel A. Smith (2014), p. 31); Tr. Trans. 133, ECF No. 97 (Timberlake) (same).

These findings are corroborated by anecdotal evidence reflecting that African Americans are significantly more likely than whites in Ohio to utilize EIP voting, and specifically during Golden Week. See Tr. Trans. 5, 23-24, ECF No. 97 (Martin) (based on analysis conducted by either the ODP or CCDP, more than half of the ballots cast EIP in Cuyahoga county in 2008 were cast by African American or Hispanic voters; only one in five ballots were cast by African American or Hispanic voters by mail or on Election Day, and the numbers were very similar in 2012); Id. at 207-08, 210 (Taylor) (on three different occasions in 2012, Taylor observed that almost everybody in line at the Cuyahoga County early vote location was African American); Tr. Trans. 17, ECF No. 100 (McNair) (observed that EIP voting in Cuyahoga County was primarily used by African Americans); Tr. Trans. 222, 230, ECF No. 97 (Perlatti) (based on his observations, the majority of individuals in Cuyahoga County utilizing EIP voting, and Golden Week voting in 2012, are/were African American); Tr. Trans. 155-56, ECF No. 100 (Burke) (based on her observations, the racial makeup of voters using Golden Week in Hamilton County was overwhelmingly African American); Tr. Trans. 91-92, ECF No. 101 (Owens) (observed primarily African Americans using Golden Week in Montgomery County).

Moreover, to the extent the voters who would have voted during Golden Week choose to vote on other early voting days or on Election Day, that will likely result in longer lines at the polls, thereby increasing the burdens for those who must wait in those lines and deterring voting. See PX 113 at 12-16 (Yang Rep.) (explaining simulation model indicating that if 5%, 16%, or 45% of the voters who would have voted early now vote on Election Day instead, wait times will increase even if BOEs do not reduce the number of DRE machines used); see also Tr. Trans. 192, ECF No. 102 (Allen) ("[I]n general, more people means expected longer lines."). African Americans will disproportionately bear this burden, because, as explained infra , they have greater time and resource limitations that may prevent them from waiting in line on Election Day and are less likely to vote absentee.

Based on this evidence, it is reasonable to conclude that the reduction in overall time to vote will burden the right to vote of African Americans, who use EIP voting significantly more than other voters. See N.A.A.C.P. I , 43 F.Supp.3d at 841 (making a similar finding); N.A.A.C.P. II , 768 F.3d at 539–42 (affirming the district court's finding).

b. SDR

The elimination of Golden Week also eliminated the opportunity for SDR, a mechanism which the record reflects over 10,000 people used in each of the last two presidential elections. In the 2008 general election, approximately 12,842 individuals used Golden Week to both register (including both new and updated registrations) and vote on the same day. PX 109 at 51 (Timberlake Rep.); Tr. Trans. 108, ECF No. 97 (Timberlake). In the 2012 general election, approximately 14,253 individuals used Golden Week to both register and vote on the same day. PX 109 at 51 (Timberlake Rep.). Of these, 5,844 were new registrations, and 8,409 were updates to existing registrations. DX 15 at 7 (Hood Rep.). Of the 5,844 new registrations, 1,789 were in Ohio's three largest urban counties. Id. at 8.

The opportunity for SDR during Golden Week alleviated the costs to voters of having to register and vote at separate times. Indeed, it may be more difficult for voters with time, resource, transportation, and childcare restraints to make two separate trips to register and vote, and Golden Week allowed individuals to do both at once. See PX 109 at 50 (Timberlake Rep.); Tr. Trans. 38-39, ECF No. 104 (Poland); Tr. Trans. 108-09, 112, ECF No. 96 (Butcher); Id. at 161-62 (Kohn); Tr. Trans. 33-36, ECF No. 97 (Martin); Tr. Trans. 120-21, ECF No. 101 (Beswick). The elimination of SDR means that voters must now register and vote at separate times, which increases the "cost of voting," especially for socio-economically disadvantaged groups. See PX 109 at 50 (Timberlake Rep.); Tr. Trans. 195-96, ECF No. 97 (Timberlake).

Expert and anecdotal evidence in the record indicate that, unfortunately, African Americans in particular are more likely to be subject to economic, transportation, time, and childcare constraints that increase the cost of voting. Specifically, relative to whites, African Americans in Ohio are less likely to work in professional and managerial jobs; are more likely to work in service and sales jobs, including hourly wage jobs; have lower incomes; are nearly three times more likely to live in poverty; and are more than two and a half times more likely to live in a neighborhood in which more than 20% of the residents are in poverty. PX 109 at 19, 22-24, 31-32 (Timberlake Rep.); see also Tr. Trans. 46, ECF No. 96 (Turner) (describing the majority African-American Hough community as one of the poorest communities in Cleveland); Tr. Trans. 78, ECF No. 102 (Cleveland) (describing two neighborhoods in Cleveland with more than 95% African Americans as historically poor).

Further, African Americans in Ohio are generally more transient and rely more heavily on public transportation than whites. See PX 109 at 17-18 (Timberlake Rep.) (for the period between 2008 and 2012, white households are almost twice as likely as African American households to be homeowners, and African Americans are more likely than whites to have moved residences); Tr. Trans. 176-77, ECF No. 97 (same); PX 109 at 23 (Timberlake Rep.) (on average, African Americans have one fewer vehicle per household than whites in Ohio; over 20% of African Americans have no automobile available in the household compared to 6% for whites; and about 12% of African Americans rely on walking or public transportation to get to work compared to about 3% of whites); Tr. Trans. 80, ECF No. 102 (Cleveland) (less than fifty percent of her majority African-American district in Cleveland own a vehicle, particularly those who live in the public housing areas); Tr. Trans. 159, ECF No. 96 (Kohn) (many individuals in a predominantly African-American area of Columbus used public transportation).

These disparities suggest that the cost of voting is therefore generally higher for African Americans, as they are less likely to be able to take time off of work, find childcare, and secure reliable transportation to the polls. Moreover, greater levels of transience may result in more frequent changes of address, which in turn requires individuals to update their registration more frequently. SDR provided an opportunity to do so and vote at the same time. As such, African Americans disproportionately make up the group that benefits the most from SDR, and the elimination of that opportunity burdens their right to vote.¹³ C.f. N.A.A.C.P. v. Husted I , 43 F.Supp.3d at 850 (discussing plaintiffs VRA claim and finding that the elimination of SDR will impact African Americans, as the record indicated that "they tend to disproportionately make up the groups that benefit the most from same-day registration: the poor and the homeless."); N.A.A.C.P. v. Husted II , 768 F.3d at 551, 555 (affirming the district court's finding of a VRA violation).

13 There is also expert evidence, although entitled to little weight, that the elimination of SDR disproportionately burdens the voting rights of African Americans. See Tr. Trans. 37-39, 55-56, ECF No. 101 (Canon); PX 109 at 51 (Timberlake Rep.) (usage rate of Golden Week was greater in high minority counties than in other counties in 2008 and 2012).

c. Nature of the Burden

Together, this evidence of the effects of the reduction in EIP voting days and the elimination of SDR demonstrates that S.B. 238 imposes a modest, as well as a disproportionate, burden on African Americans' right to vote.

In so finding, the Court, as in N.A.A.C.P. v. Husted , is mindful of the numerous opportunities to cast a ballot in Ohio, including vote by mail, in person on Election Day, and on other EIP voting days. Similar to N.A.A.C.P. v. Husted , however, the Court finds that while these opportunities mitigate some of the burdens imposed by S.B. 238, the record reflects that they do not eliminate or significantly decrease those burdens.

Specifically, the anecdotal evidence suggests that voting by mail is not a viable alternative to EIP voting for many African Americans. First, the record reflects that African Americans are distrustful of voting by mail. See Tr. Trans. 59-60, ECF No. 97 (Martin) ("Many voter[s] that we talked to, and I experienced personally, I've had a vote-by-mail application alongside the registration and they said, I wouldn't prefer that. I don't trust the vote by mail. I'd like to cast my ballot in person. So not every voter—vote by mail is not for every voter."); Tr. Trans. 226, ECF No. 96 (Anthony) ("I will say to them, you know, you can vote by mail and just mail it in .... And they would tell me, well, I don't trust the mail. I don't know what it's going to cost me and I don't know if you guys are going to count it or not.... A lot of folks think their absentee ballot would not be counted ...."); Id. at 62-63 (Turner) (African Americans are very suspicious of the vote-by-mail process and tend to vote in person); McDonald Dep. 66-67, ECF No. 74-10 ("I've noticed that the African/American community is more likely not to utilize vote by mail because they have the fear factor of it being delivered to us ...")¹⁴ ; Tr. Trans. 106, ECF No. 96 (Butcher) (in his GOTV efforts with African American voters, he had to overcome the notion that the vote did not matter and the historical suspicion of the voting process); Id. at 158-59 (Kohn) ( "encountered a lot of skepticism" in immigrant Somali and predominantly African American communities on the east side of Columbus "about whether their vote would count, how to vote and just about the process in general."); see also N.A.A.C.P. v. Husted II , 768 F.3d at 542 (affirming the district court's finding that voting by mail is not a viable alternative means of access to the ballot for African Americans, lower-income individuals, and the homeless because, inter alia , they are distrustful of the mail and/or voting by mail).

14 Defendants object to this portion of McDonald's deposition on the grounds that it calls for an expert opinion and for speculation. Defendants' objection is overruled. McDonald did not testify to his opinion (expert or otherwise); he testified to his observation, as evidenced by the words, "I've noticed ...." McDonald Dep. 66-67, ECF No. 74-10.

Second, as explained in greater detail herein, voting by mail entails correctly filling out an absentee ballot application, paying postage to return that application by mail or making arrangements to return the application in person, correctly filling out an absentee ballot, and paying postage (sometimes in irregular amounts) to return the ballot by mail or making arrangements to return it in person. These costs and relatively complex requirements coupled with evidence described above that African Americans tend to have more limited financial, transportation, and childcare resources and lower levels of educational attainment further suggest that voting by mail may not be a suitable alternative for many African American voters. See PX 109 at 19, 22-24, 31-32 (Timberlake Rep.); see also Tr. Trans. 46, ECF No. 96 (Turner); Tr. Trans. 78, ECF No. 102 (Cleveland); PX 109 at 4-5 (Timberlake Rep.) (discussing cost of voting analysis); N.A.A.C.P. v. Husted I , 43 F.Supp.3d at 843 (finding the same); N.A.A.C.P. v. Husted II , 768 F.3d at 542 (same). Moreover, as noted in N.A.A.C.P. v. Husted I , S.B. 238 reduces "the overall period for voting by mail by the same amount as the time for EIP voting," thereby providing less time to vote by mail than before S.B. 238. N.A.A.C.P. v. Husted I , 43 F.Supp.3d at 843 (citing Ohio Rev. Code § 3509.01(B)(2) ).

Additionally, the costs of registering and voting at separate times, the cost of voting in general, and evidence that African Americans fare worse in various socio-economic measures also reduces the viability of Election Day voting as an alternative.

Accordingly, while Ohio's election system provides multiple ways to vote, the record suggests that those options do not eliminate or significantly decrease the burden imposed on the right to vote of African Americans as a result of the elimination of Golden Week.

Notably, the burden imposed by S.B. 238 in the context of Ohio's current voting scheme is less than the burden it imposed when it was first challenged in N.A.A.C.P. v. Husted in 2014. At the time the plaintiffs in N.A.A.C.P. v. Husted challenged S.B. 238, the Secretary had also issued a directive setting uniform EIP voting hours for the entire state that did not include voting hours for several weekends or evening voting hours on any day. See N.A.A.C.P. v. Husted I , 43 F.Supp.3d at 812–13. The district court found, and the Sixth Circuit affirmed, that the combined effects of S.B. 238 and the directive created a significant burden on the right to vote. Id. at 842–44 ; N.A.A.C.P. v. Husted II , 768 F.3d at 546. Specifically, the Court found that the lack of evening voting hours and the availability of only one Sunday for EIP voting burdened the voting rights of African Americans and lower-income voters because of the importance of Sunday "Souls to the Polls" efforts in African American communities and the inability of lower-income voters to find the time to vote during daytime hours. N.A.A.C.P. v. Husted I , 43 F.Supp.3d at 842. As a result of the settlement in that case, Ohio now provides, for presidential general elections, EIP voting hours on two Saturdays and two Sundays and provides for ten days where voting is permitted until either 6:00 p.m. or 7:00 p.m. The addition of those EIP voting hours somewhat reduces the burden imposed by S.B. 238 today as compared to the burden it imposed in 2014 when Ohio did not provide those hours. For the reasons articulated above, however, those hours do not eliminate the burden currently imposed by S.B. 238, nor do the hours ameliorate the burden such that it is outweighed by the State's tenuous justifications, as discussed infra.

The Court is also mindful that unlike in N.A.A.C.P v. Husted , the instant record includes evidence of an election that occurred while S.B. 238 was in effect. Defendants rely on this analysis for the proposition that S.B. 238 did not negatively affect voters. Specifically, Defense expert Dr. McCarty compared voter turnout in Golden Week and non-Golden Week elections (2010 and 2014). He found that even with S.B. 238 in effect, African Americans voted EIP at a slightly higher rate in 2014 than 2010 and that African Americans who use early voting tend to be highly engaged voters who will not be substantially impacted by the elimination of Golden Week. See DX 20 at 8 (McCarty Reb.) (finding that in 2010, when Golden Week existed, 6.9% of the votes cast by African Americans were EIP, but in 2014, when Golden Week did not exist, the rate was 7.1%); see id. at 12-13; Tr. Trans. 60, 71-72, ECF No. 98 (McCarty).

Dr. McCarty's analysis does not change the Court's finding. First, the probative value of his results is diminished by the fact that they are based on data from midterm as opposed to presidential elections. Midterm elections, which generally produce much lower turnout than presidential elections, see, e.g. , Tr. Trans. 43-44, ECF No. 97, may not paint an accurate picture of voter behavior. Indeed, those who vote in midterm elections are arguably already higher-propensity voters. Second, as discussed supra , the Court is skeptical of the accuracy of Dr. McCarty's methodology in surmising the race of particular voters based on surnames and addresses. Moreover, his data shows that in 2010, African Americans still voted EIP during Golden Week at a higher rate than whites, and African Americans still voted EIP at a higher rate than whites in 2010 and 2014. Tr. Trans. 153-58, 163–64, ECF No. 97 (Timberlake). Last, Dr. McCarty's analysis, which does not meaningfully address the impacts of the elimination of SDR, does not change the Court's finding of the burden imposed by SDR's elimination. Accordingly, the Court finds that the evidence regarding the 2014 midterm election does not persuasively indicate that S.B. 238 is not substantially likely to burden the right to vote of African Americans in the 2016 general presidential election.¹⁵

15 Nor does the Court find persuasive the evidence regarding the effects of similar laws in North Carolina and Georgia given the record evidence regarding the marked differences between the examined elections in those states and the elections in Ohio.

In sum, the record includes statistical and anecdotal evidence reflecting that: over 10,000 voters have used EIP voting in the past; African Americans use EIP voting at significantly higher rates than whites; many thousands of individuals have used Golden Week in the past; African Americans used Golden Week at a significantly higher rate than whites in previous elections; the elimination of the early voting days will therefore burden African Americans; the elimination of SDR will increase the cost of voting for African Americans; and other voting mechanisms do not suffice to reduce these burdens. Thus, although the Court cannot predict how many African Americans will turn out in future elections, it is reasonable to conclude from this evidence that their right to vote will be modestly burdened by S.B. 238's reduction in the EIP voting period and elimination of SDR.

2. Whether the State's Justifications Outweigh the Burden

Having found that S.B. 238 imposes a modest burden on the right to vote of African Americans in Ohio, the Court must apply the Anderson /Burdick standard to weigh that burden against the precise interests offered by Defendants as justifications for that burden. "Put differently, [Defendants] must articulate specific, rather than abstract state interests, and explain why the particular restriction imposed is actually necessary, meaning it actually addresses, the interest put forth." N.A.A.C.P. v. Husted II , 768 F.3d at 545. Defendants articulate four primary justifications for S.B. 238's elimination of Golden Week: (1) preventing voter fraud; (2) reducing costs; (3) reducing administrative burdens; and (4) increasing voter confidence and preventing voter confusion. For the following reasons, the Court finds that these justifications do not outweigh the burden imposed. a. Preventing Voter Fraud

First, while the record includes general opinion evidence that Golden Week increases the opportunity for voter fraud, see Tr. Trans. 112, ECF No. 104 (Damschroder); Tr. Trans. 253, ECF No. 103 (Ward); DX 15 at 15 (Hood Rep.), actual instances of voter fraud during Golden Week are extremely rare. See Tr. Trans. 134, ECF No. 104 (Damschroder) (confirming that is the Secretary's position); Tr. Trans. 225, ECF No. 97 (Perlatti unaware of any instances of fraud due to Golden Week in Cuyahoga County); Tr. Trans. 24, ECF No. 100 (McNair testifying to the same); Id. at 161-62, 168-70, 191 (Burke testifying that in the few instances where fraud was discovered in Hamilton County, the votes were not counted); Tr. Trans. 110, ECF No. 102 (Munroe confirming that he believes voter fraud is very rare); Tr. Trans. 10, ECF No. 103 (Cromes did not observe any instances of fraud during Golden Week in Portage County); Id. at 252 (Ward confirming he is unaware of any instances of fraud during Golden Week in Madison County); see also PX 107 at 10-11 (Minnite Rep.) (in-person voter fraud is particularly uncommon).

There were two discovered instances of voter fraud in Hamilton County during Golden Week in 2012, but the ballots were not ultimately counted in either instance. Tr. Trans. 17, 38, ECF No. 104 (Poland). Damschroder testified about two types of incidents that occurred in Franklin County during Golden Week in 2008. First, six out-of-town students voted during Golden Week who were not registered to vote in Ohio, but they self-reported, and five of the six ballots were successfully withdrawn. Tr. Trans. 106, 138, ECF No. 104 (Damschroder). Second, there were incidents where registrants' acknowledgment cards were returned as undeliverable and the residences were found to be empty, but those ballots were not counted, and no indictments issued. Id. at 107, 138.

This very limited evidence of voter fraud is insufficient to justify the modest burden imposed by S.B. 238. See N.A.A.C.P. v. Husted II , 768 F.3d at 547 (recognizing that preventing voter fraud is a legitimate interest but concluding that only a handful of actual examples of voter fraud is insufficient to outweigh the burden imposed by S.B. 238).

Defendants argue that elections officials may not have enough time before Election Day to verify the registration of a voter who registered during Golden Week. This concern is insufficient to outweigh the burden imposed by S.B. 238. Notably, under the current scheme, a voter can register on the last day of the registration period and cast an in-person ballot the very next day before the mail-verification process has been completed.¹⁶ Tr. Trans. 135-36, ECF No. 104 (Damschroder); see also Tr. Trans. 273-74, ECF No. 97 (Perlatti) (voter could register thirty days before election, cast a ballot during early voting, and the acknowledgment card could be returned after the election); Tr. Trans. 156, ECF No. 99 (Hood); Tr. Trans. 171, ECF No. 100 (Burke).¹⁷ As another judge of this Court and the Sixth Circuit recognized, that concern "does not withstand logical scrutiny, but instead is perhaps better characterized as a justification for limiting the registration deadline itself to beyond 30 days before the election." N.A.A.C.P. v. Husted I , 43 F.Supp.3d at 844 ; N.A.A.C.P. v. Husted II , 768 F.3d at 547.

16 BOEs verify new or updated Golden Week registrations through a mail verification process. The BOE sends the voter a nonforwardable acknowledgment card to the voter's address. If the card is not returned, the voter is assumed to reside at the address. If the card is returned as undeliverable, the BOE will send a forwardable confirmation card to the voter. If the voter returns the card with the correct information, the ballot is counted. If the card is not returned, is returned as undeliverable, or it contains incorrect information, it is not counted. Tr. Trans. 223, ECF No. 97 (Perlatti); Tr. Trans. 23, ECF No. 100 (McNair).

17 The registration period for the 2016 general election ends on the 28th day before the election, as the 29th day is a holiday. DX 14J (2016 voting calendar); Tr. Trans. 136, ECF No. 104 (Damschroder).

This point is illustrated by the simple hypothetical of a voter (under the status quo imposed by SB 238) registering to vote 30 days before the election and then returning to his local Board to cast an EIP ballot on the 29th day before the election. The potential that that voter's registration could not be verified in time is nearly exactly the same as a voter who could previously register and vote on the 30th day before the election. In fact, if the potential for fraud is measured in relation to time before the election, that potential would increase as the election grew closer because less time remains to verify absentee voter registrations. In sum, the potential for fraud [on this ground] exists whether voters are allowed to register and vote on the same day or not, and is best combatted by election officials following the law and applicable procedures and not counting absentee votes prior to the proper verification of registration.

N.A.A.C.P. v. Husted I , 43 F.Supp.3d at 844. Thus, because ballots cast by same-day registrants were segregated from other ballots in order to verify registration, Tr. Trans. 135, ECF No. 104 (Damschroder testifying that was the case in at least 2012); Tr. Trans. 223-24, ECF No. 97 (Perlatti); Tr. Trans. 23, ECF No. 100 (McNair); Id. at 162 (Burke); Tr. Trans. 10, ECF No. 103, it would have been no more difficult to verify those ballots than to verify the ballots cast by someone who registered on the last day of the registration period and voted the next day or by someone who voted very close to Election Day. See N.A.A.C.P. v. Husted II , 768 F.3d at 547 (recognizing the same); compare with North Carolina State Conference of the N.A.A.C.P. v. McCrory , 182 F.Supp.3d 320, 478–80, 2016 WL 1650774, at *127 (M.D.N.C. April 25, 2016) (finding that unlike in N.A.A.C.P. v. Husted , the record showed that SDR's proximity to Election Day did not provide BOEs a sufficient number of days to verify SDR voters).

Accordingly, while the interest in preventing voter fraud is a valid one, absent more than very limited evidence of actual occurrences of voter fraud, it is insufficient to justify the burden imposed by S.B. 238.

b. Reducing Costs

Defendants' cost justification also fails to outweigh the burden imposed by S.B. 238. First, cost savings from the elimination of Golden Week are minimal. Tr. Trans. 220-21, ECF No. 97 (approximately $41,000 cost to Cuyahoga County BOE to administer Golden Week in 2012 out of overall budget of over $13 million); Id. at 221 (Cuyahoga County BOE's budget request for 2016 will be between $16 million and $17 million); Tr. Trans. 157-58, ECF No. 100 (Burke) (as a member of Hamilton County BOE, Burke was never made aware of any concerns about the cost of Golden Week in any election, as it would have added a very small percentage to the BOE's budget); Tr. Trans. 35, ECF No. 104 (Poland) ($8,000 to $12,000 in staffing costs alone to Hamilton County BOE to administer Golden Week in 2014 out of an approximately $8 million budget; approximately $16,000 to $24,000 in costs to administer Golden Week in Hamilton County in 2012); Tr. Trans. 96, 109-10, ECF No. 102 (Munroe) (Mahoning County BOE incurred "a few thousand dollars" in costs for the administration of Golden Week in 2012 out of approximately $2.5 million budget for that year, and sixty to seventy voters used Golden Week to both register and vote); Tr. Trans. 250-51, ECF No. 103 (Martin) (estimated cost to Madison County had it administered Golden Week in 2015 would have been around $1,000 out of approximately $400,000 budget). Moreover, if BOEs have additional budgetary needs, they may request additional funds from the County Commissioners or apply to the Court of Common Pleas for funding that is absolutely necessary to conduct an election. Tr. Trans. 104, ECF No. 102 (Munroe).

Second, Ohio law requires BOEs to keep their offices open until 9 p.m. on the final day of registration and to remain open for a period of time necessary for the performance of its duties at all other times during each week. Ohio Rev. Code § 3501.10(B). Golden Week would therefore presumably take place during times when BOEs are already open for business. See Tr. Trans. 202, ECF No. 100 (Burke) (Hamilton County BOE was open for business during the week that would have been Golden Week in 2014). Therefore, any BOEs that conduct EIP voting at their offices are unlikely to incur substantial additional overhead costs. See N.A.A.C.P. v. Husted II , 768 F.3d at 549 (affirming a similar finding in N.A.A.C.P. v. Husted I ).

To be sure, reducing costs is a legitimate interest. But "it is not enough merely to assert that a restriction on voting saves costs .... Rather, where more than minimal burdens on voters are established, the State must demonstrate that such costs would actually be burdensome."). N.A.A.C.P. v. Husted II , 768 F.3d at 548 (discussing Obama for Am. , 697 F.3d at 433-34 ). Defendants have not offered any evidence that counties were unable to manage the minimal costs of maintaining Golden Week in the past or would be unable to do so in the future. The interest in managing costs is therefore insufficient to justify the modest burden imposed by eliminating Golden Week.

c. Reducing Administrative Burdens

Defendants' administrative concerns are also unavailing. The record does reflect that the five-week period before an election is an extremely busy time for BOEs, and thus, Golden Week and the extra responsibility associated with it would increase administrative burdens. See Tr. Trans. 227-31, ECF No. 103 (Ward) (discussing ballot preparation, logic and accuracy tests, and processing of voter registrations and absentee ballot applications that occur during that period and testifying that staff stays until 9:00 p.m. or 10:00 p.m. to complete all of those tasks); Tr. Trans. 14-15, ECF No. 104 (Poland) (testifying that around thirty-five days before the election, the BOE is editing registrations, preparing absentee ballots by mail, testing voting equipment, printing ballots, and dealing with campaign finance deadlines).

Similar to their cost concerns, however, Defendants fail to present sufficient evidence that BOEs were unable to manage these administrative burdens when Golden Week was in place or that they will be unable to do so should it be reinstated.¹⁸ See McCrory , 182 F.Supp.3d at 478–80, 2016 WL 1650774, at *127 (finding that unlike in N.A.A.C.P. v. Husted , the record showed that SDR placed additional burdens on BOE staff that prevented them from timely processing SDR registrants as required by statute). Absent such evidence, the Court finds the administrative burdens imposed by Golden Week insufficient to justify the burden imposed by its elimination. Cf. Obama for Am. , 697 F.3d at 433–44.

18 The only evidence in support of that notion is that in 2010, the Ohio Association of Election Officials task force, aware of these administrative concerns, recommended that early voting begin twenty-one days before Election Day. See Tr. Trans. 232, ECF No. 103. In fact, the record also includes evidence that Golden Week aids in election administration in that it (1) provided boards more time to mail out and process absentee ballots, Tr. Trans. 225-26, ECF No. 97 (Perlatti); Tr. Trans. 21-22, ECF No. 100 (McNair), and (2) relieved pressure on the polls on Election Day, Tr. Trans. 225-26, ECF No. 97 (Perlatti); Tr. Trans. 156, ECF No. 100 (Burke); Tr. Trans. 7, ECF No. 103 (Cromes).

d. Increasing Voter Confidence and Preventing Voter Confusion

Finally, Defendants have adduced insufficient evidence in support of their final justification for S.B. 238—increasing voter confidence and preventing voter confusion—citing only two elections officials' concerns that voters could become confused about deadlines for registration as a result of Golden Week. See, e.g. , Tr. Trans. 98, ECF No. 102 (Munroe). Defendants adduced insufficient evidence of actual voter confusion to substantiate those concerns.

In sum, Plaintiffs have adduced sufficient evidence demonstrating: that a significant number of individuals used EIP voting and Golden Week in past elections; that African Americans did so at a higher rate than whites; that alternative methods of voting may not be viable alternatives for African Americans because of their distrust of voting by mail, the cost of voting on Election Day, and the costs of registering and voting at separate times; and that African Americans are among those that will be most affected by the elimination of SDR. Additionally, Defendants' justifications for S.B. 238, while they may be legitimate, are minimal, unsupported, or not accomplished by S.B. 238. In other words, Defendants have failed to establish that the justifications are "actually necessary" to burden the right to vote of African Americans. See N.A.A.C.P. v. Husted II , 768 F.3d at 545. Indeed, although the availability of weekend and evening EIP voting hours renders the burden in this case less severe than that in N.A.A.C.P. v. Husted , the burden here remains more than minimal, and Defendants' justifications do not, either individually or together, outweigh that burden. Accordingly, Plaintiffs succeed on their Anderson /Burdick claim based on S.B. 238.

B. Ohio Revised Code § 3501.10(C) —One EIP Voting Location per County

Plaintiffs next challenge Ohio Revised Code § 3501.10(C), which limits counties to one EIP voting location per county, whether it be the offices of the county's BOE or elsewhere. Id.

1. Burden

Plaintiffs have failed to prove that the limitation to one EIP voting center per county burdens African Americans. Plaintiffs first argue that a single EIP voting location requires large numbers of voters to travel greater distances to vote early than they would have had to travel if additional early voting locations were open. It does not follow, however, that because there exists the possibility of a more convenient alternative (as Plaintiffs ask for nothing more than discretion to open additional EIP voting centers), the status quo amounts to a burden. In other words, it is quite possible that the status quo is not burdensome, even though an alternative would be even more convenient. Indeed, Plaintiffs have offered no evidence that the location of existing EIP voting centers are so far away from African Americans that they create a burden on African Americans. See Tr. Trans. 237-38, ECF No. 96 (Anthony) (Franklin County's early voting site is intentionally located on a public transportation route and in the middle of a population that largely walks; approximately 60% of eligible voters in Franklin County live in the area of the early voting site); Tr. Trans. 20, ECF No. 104 (Poland) (Hamilton County's early voting site, the BOE in downtown Cincinnati, is on a public transport line, a few blocks away from the central hub for public transport, and minutes away from major interstates); Tr. Trans. 252, ECF No. 97 (Perlatti) (Cuyahoga County's early voting site, the BOE office in downtown Cleveland, is accessible by public transport and has good freeway access).

Plaintiffs argue that Ohio has traditionally had long lines for early voting. See Tr. Trans. 57-59, ECF No. 96 (Turner); id. at 205-09 (Anthony); Tr. Trans. 143, ECF No. 101 (Beswick); Tr. Trans. 229-30, ECF No. 97 (Perlatti); id. at 45-46, 57 (Martin); id. at 207-09 (Taylor); Tr. Trans. 16-17, 20, ECF No. 100 (McNair); id. at 195 (Burke); Tr. Trans. 12, ECF No. 103 (Cromes); Tr. Trans. 115, ECF No. 96 (Butcher); id. at 142 (Caffrey); Tr. Trans. 22-23, ECF No. 98 (Bowman); Tr. Trans. 92, ECF No. 101 (Owens); PX 19 at 3-4 (photos). There is no persuasive evidence, however, that the long lines are caused by the limitation of having only one early voting location.¹⁹ Accordingly, Plaintiffs have failed to prove that Ohio Revised Code § 3501.10(C) imposes a burden on African Americans.

19 Dr. Allen argues the contrary—that dividing a set amount of resources among multiple EIP voting locations would actually increase wait times. See DX 16 at ¶ 25; Tr. Trans. 126, ECF No. 102. The Court gives this evidence little weight for the reasons articulated in Dr. Yang's rebuttal report. PX 114 at 4-5 (Yang Reb.). Nevertheless, although Plaintiffs attack Dr. Allen's queuing theory analysis, Plaintiffs bear the burden of showing that it is the limitation to one EIP voting location that causes increased wait times (assuming increased wait times amounted to a burden on the right to vote). Discrediting Dr. Allen's evidence to the contrary does not satisfy that burden, and Plaintiffs have failed to offer persuasive evidence of such causation. For example, Dr. Yang's analysis assumes: (1) the invalidation of the one EIP center provision will result in more EIP centers actually being opened, (2) the additional EIP centers will decrease the travel time for certain constituents, (3) the benefits from the increase in EIP voting centers will not be offset by inconveniences of those additional centers (lack of parking, limited facility space, difficulty of finding the polling station, etc.), (4) the DRE resources will not remain constant, and (5) there will not be a corresponding increase in EIP voter turnout which adds to longer lines. In sum, even if the Court presumed that long wait lines was a burden on African Americans, Plaintiffs have failed to show that it is the limitation on having only one EIP voting center that causes that burden.

The failure to do so implicates Plaintiffs' standing to challenge this provision. With respect to representational standing, Plaintiffs must establish an injury-in-fact suffered by those whom they purport to represent. The Court has found that no burden on the right to vote results from this provision, and Plaintiffs have not articulated any other injury caused by the provision. Plaintiffs therefore have not established representational standing to challenge Ohio Revised Code § 3501.10(C). Moreover, the existence of organizational standing is questionable. While ODP intends to engage in GOTV voter-education efforts with respect to the election law changes that occurred in 2014, it seems unlikely that it would educate voters on the EIP voting location limitation, as that does not directly impact a voter's obligations. Moreover, the limitation of one EIP voting location per county existed before the 2014 changes.

2. Whether the Law is Rationally Related to a Legitimate State Interest

Even assuming standing exists, the Court finds that Ohio Revised Code § 3501.10(C) is rationally related to a legitimate state interest. See NE OH Coalition for the Homeless , 696 F.3d at 592 (rational basis standard applies to laws that do not burden the fundamental right to vote).

Denying BOEs the discretion to open more than one EIP voting site promotes uniformity, which in turn promotes the fair administration of elections. Indeed, granting counties the discretion to open more than one early voting site could result in different voting opportunities within a single district. Tr. Trans. 101, ECF No. 104 (Damschroder) ("I think like my example from earlier, about the Delaware County voter in the same congressional district as the Franklin County congressional district voter. Delaware might not choose to have multiple locations or Delaware might choose to have multiple locations and Franklin County might not. And so then voters wouldn't have, even within the same district or political subdivision, have the same opportunities to request or vote an absentee ballot in person."). Granting counties the discretion to open more than one early voting center could also "open[ ] the door for controversy because ... you could potentially have a scenario where boards were tying [on whether to open another EIP voting location] and then a Secretary of State from one party or the other was breaking the tie. Those tie votes could change from election to election ...." Id.

In short, limiting counties to one EIP voting location helps ensure everyone has the same opportunity to vote EIP, which is a rational basis upon which to justify the limitation.²⁰ Accordingly, Plaintiffs' Anderson / Burdick claim based on Ohio Revised Code § 3501.10(C) fails.

20 Notably, Plaintiffs did not adduce evidence that any BOE would like to open an additional EIP voting location; to the contrary, evidence indicates it would be extremely difficult for BOEs to do so. Tr. Trans. 258, ECF No. 97 (Perlatti) (it is possible but would be very difficult); Tr. Trans. 22, ECF No. 104 (Poland) (under the current 2016 budget, Hamilton County would not be able to accommodate an additional location in 2016); Tr. Trans. 103, ECF No. 102 (Munroe) (an additional location would not fit within the Mahoning County BOE's 2016 budget); Tr. Trans. 20-22, ECF No. 104 (Poland) (discussing logistical considerations in opening an additional voting site).

C. S.B. 200 and Directive 2014-26—Reduction in DREs

Plaintiffs next challenge the formulas provided in S.B. 200 and Directive 2014-26 to be used to calculate the minimum number of DRE machines that counties which use DRE machines as their primary voting device are required to maintain in their inventories and deploy on Election Day.

Counties in Ohio are permitted to choose the type of voting equipment that best fits their needs. Ohio Rev. Code § 3506.02 ; Tr. Trans. 83, ECF No. 104 (Damschroder). Counties that choose to use DRE machines are required by statute to maintain a certain number in their inventory. Ohio Rev. Code § 3506.22. Before S.B. 200, the ratio was 1 machine for every 175 registered voters within a county.

S.B. 200 modified the ratio. Under the new formula, the ratio remains 1 to 175, but the denominator is now calculated by subtracting the total number of absentee ballots counted in the last presidential election from the greater of either (i) the total number of registered voters in the county at the voter registration deadline in the last presidential election, or (ii) the average of the total number of registered voters in the county as of the voter registration deadlines for the last two presidential elections. Ohio Rev. Code § 3506.22(B).

Directive 2014-26, which addressed requirements for the November 4, 2014 general election, required BOEs using DRE machines as their primary voting system on Election Day to "deploy at least one DRE voting machine for every 175 registered voters in a precinct or voting location ..." DX 62 at 2. It provides that "[i]n determining the number of registered voters, a Board does not have to count electors who have failed to respond within 30 days to any confirmation notice ... [and/or] any registered voter who has requested an absent voter ballot (by mail or in person) as of the date the allocation decision is made by the Board." Id. at 2-3. Directive 2015-15 imposed the same requirement for the November 3, 2015 general election. DX 27 at 2-3.

Thus, while S.B. 200 addresses the ratio requirement concerning a county's inventory of DRE machines, Directives 2014-26 and 2015-15 address the ratio requirement for DRE machines deployed on Election Day. See Tr. Trans. 89, ECF No. 104 (Damschroder) ("[T]he new ratio in the law is that the board can exclude voters who cast.... an absentee ballot in the last presidential election from the denominator when making an inventory purchase decision in the future for DREs, which is different than [Directive 2015-15] which [is] specific instructions for allocation for a specific election.").

1. Burden

The Court finds Plaintiffs have failed to show that S.B. 200 or Directive 2014-26 will burden the right to vote of African Americans.

The evidence indicates that S.B. 200's new formula will reduce the minimum number of DRE machines that counties are required to have in their inventory and that to the extent a county's DRE machine supply falls below the number required under the law previously in place , the result will be longer wait times to vote. PX 113 at 8-11, 16-18 (Yang Rep.)²¹ ; Tr. Trans. 193, ECF No. 98 (Yang) ("So if we ... keep 175 unchanged, just make the numerator smaller because we are removing the number of absentee voters in that formula ... it will result in smaller number and that will be smaller number of voting machines."); id. at 175 (Yang) ("So if what they actually used on [E]lection [D]ay is more than the new minimum, less than what the old minimum [sic], the expected wait time will be longer than what happened before but will be shorter than what I reported in my analysis."); Tr. Trans. 179-80, ECF No. 102 (Allen).²²

21 In using the minimum number of DRE machines required under S.B. 200 to calculate wait times, Dr. Yang's analysis appears to assume that counties will deploy that minimum number of machines on Election Day.

22 These burdens could be exacerbated by DRE failures, see Tr. Trans. 202, ECF No. 98. Indeed, DRE machines have failed or had technical issues in the past. See Tr. Trans. 12-15, ECF No. 100 (McNair); Tr. Trans. 19, ECF No. 103 (Cromes); see also Tr. Trans. 219, ECF No. 96 (Anthony), but only to the extent not mitigated by the use of backup paper ballots, see DX 63 (Directive 2014-23) (for the November 4, 2014 election, BOEs who used DRE machines on Election Day were required to supply all precincts with backup paper ballots).

The existence of a burden on the right to vote resulting from S.B. 200 is therefore conditional upon a county's DRE supply falling below the minimum number of DREs required under the old formula as a result of the county's adherence to the new formula.

Plaintiffs have not adduced evidence of a single county's inventory falling below the minimum number of DRE machines required before S.B. 200. This is unsurprising, as the Secretary instructed BOEs not to divest themselves of DRE machines even if their stock exceeds the new statutory floor. DX 14BB; Tr. Trans. 91, ECF No. 104 (Damschroder).²³ In fact, Franklin County and Montgomery County plan to exceed the minimum required under S.B. 200 for the 2016 election. DX 19 at ¶ 16 (Allen Reb.). Plaintiffs have therefore failed to demonstrate a burden caused by S.B. 200.

23 Plaintiffs assert that to the extent that counties exceed the old minimum, the change in the DRE formula serves no purpose and argue that the Court should therefore focus its analysis on the counties where S.B. 200 will result in a reduction in DRE inventory below the old minimum. But the Court is charged with determining the nature and severity of the burden on African American voters statewide. The fact that certain counties will not be affected by S.B. 200 is relevant to determining the extent of the burden.

The Court is mindful that because BOEs are not required to replace inoperable DRE machines as long as the counties' inventories remain above the new minimum, inventories may drop below the old statutory minimum overtime. As Dr. Yang explained, fewer DRE machines could cause longer lines at the polls. Absent evidence indicating the likelihood of that occurring, however, the mere potential for a burden to arise at some point in the future is insufficient to establish a current burden on the right to vote of African Americans, or at least for the 2016 general election.

The same analysis applies to Directive 2014-26. Although Plaintiffs do not offer any statistical evidence about the specific burden imposed if counties deploy the minimum number of DRE machines pursuant to the Directive, their evidence with respect to S.B. 200 shows that the fewer machines deployed, the longer the wait times. This burden, however, only arises if counties deploy the minimum number of DRE machines pursuant to the formula outlined in Directive 2014-26. Plaintiffs have not presented any evidence that any DRE counties intend to adhere to the minimum numbers permitted by the Directive. In fact, BOEs can exceed the minimum number of DRE machines required to be deployed on Election Day, and the Secretary instructs BOEs to consider the nature of local contests in determining whether to exceed the minimum numbers required by law. Tr. Trans. 96-97, ECF No. 104 (Damschroder); DX 27 at 1 (Directive 2015-15).

Moreover, in contrast to S.B. 200, Directive 2014-26 does not require BOEs to account for individuals who have requested absentee ballots in calculating the minimum number of DRE machines to be deployed. DX 62 at 2-3 ("In determining the number of registered voters, a Board does not have to count electors who have failed to respond within 30 days to any confirmation notice ... [and/or] any registered voter who has requested an absent voter ballot ....") (emphasis added). Accordingly, there is no evidence of an imminent burden.

In sum, Plaintiffs bear the burden of establishing that S.B. 200 and Directive 2014-26 impose a burden on the right of African Americans to vote. While those provisions create the potential for the imposition of a burden, Plaintiffs have failed to establish the existence of the circumstances necessary for that potential to become a reality for the 2016 general election.

The failure to do so implicates Plaintiffs' standing to challenge these provisions. The Court has found that no burden on the right to vote results from these provisions, and Plaintiffs have not articulated any other injury caused by the provisions. Plaintiffs therefore have not established representational standing to challenge S.B. 200 and Directive 2014-26. Moreover, the existence of organizational standing is questionable, as it seems unlikely that ODP's intended GOTV voter-education efforts would include changes in the DRE formula, as that does not directly impact a voter's obligations.

2. Whether the Provisions are Rationally Related to a Legitimate State Interest

Even assuming standing exists based on an alternative injury, however, the Court finds that S.B. 200 and Directive 2014-26 are rationally related to the legitimate government interest of promoting efficient election administration. By allowing counties to account for voters who are unlikely to vote in person when calculating the need for DRE machines, the new formula imposed by S.B. 200 promotes election administration by helping ensure that BOEs do not use funds to purchase more DRE machines than necessary. See Tr. Trans. 129, ECF No. 102 (Allen) (the new formula is "a conceptual step towards a formula that is supported by applied waiting-line analysis, or a queuing theory."); Tr. Trans. 82, ECF No. 104 (Damschroder) (formula change "seem[s] reasonable" because "the machine allocation is primarily for [E]lection-[D]ay machine inventory. And it seems reasonable for a board to be allowed to exclude from any future purchases of voting machines that happen to be DREs ... voters who have ... demonstrated a willingness to vote absentee before [E]ection [D]ay and not use—not need to use the service of a DRE on [E]ection [D]ay.").

Additionally, Directive 2014-26 promotes flexibility in election administration. Specifically, with inventories unchanged, the new formula gives BOEs greater flexibility to apportion DRE machines to the precincts where the machines are most needed. Id. at 95.

In sum, S.B. 200 and Directive 2014-26 do not impose a burden on the right to vote of African Americans, and, assuming standing nevertheless exists to challenge the provisions, the provisions are rationally related to a legitimate state interest. Accordingly, Plaintiffs' claim fails.

D. S.B. 205 and Directive 2014-15—Absentee Ballot Mailings

Plaintiffs next challenge S.B. 205's restrictions on unsolicited absentee ballot application mailings and the Secretary's policy, as outlined in Directive 2014-15, regarding the universe of voters to whom he sends his unsolicited absentee ballot application mailings.

S.B. 205 prohibits any public office or public official from mailing unsolicited absentee ballot applications; prohibits BOEs from pre-paying postage for returning absentee ballot applications or absentee ballots; prohibits the Secretary from mailing unsolicited absentee ballot applications absent specific authorization from the General Assembly; and restricts the Secretary's authorized mailings to general elections.²⁴ Directive 2014-15 identified the voters to whom the Secretary would send its authorized mailing for the November 2014 general election: (i) every registered Ohio voter in "active" status as of August 1, 2014, and (ii) every registered Ohio voter who voted in the 2010 general election or the 2012 presidential election, regardless of voter status. DX 14F at 2 (Directive 2014-15). The directive also provided that the mailing would be followed by a supplemental mailing to individuals who newly registered or changed their registration between certain dates. DX 14F at 3. Plaintiffs challenge the policy formulated in this directive.

24 Plaintiffs assert that Ohio Revised Code § 3509.03(I) prohibits the Secretary from including prepaid postage in its mailings, but that provision prohibits only BOEs from including prepaid postage.

Before S.B. 205, a few counties mailed unsolicited absentee ballot applications to voters. Id. at 55. Franklin County was the only county that did so in 2006, approximately thirteen counties did so in the 2008 general election, and "the number edged up a little bit more in 2010." Id. at 61. These counties differed as to whether they provided return envelopes and postage in their mailings and as to the universe of voters to whom they sent the mailings. Id. at 60-62. In 2012 and 2014, the Secretary mailed unsolicited absentee ballot applications statewide to the voters identified in Directive 2014-15 to ensure fairness and uniformity across counties and to help reduce the potential for longer lines on Election Day. DX 14F at 1 (Directive 2014-15); Tr. Trans. 54-56, ECF No. 104 (Damschroder). The Secretary will again mail applications statewide for the 2016 general election. Tr. Trans. 54-55, ECF No. 104 (Damschroder). The Secretary cannot, however, conduct such mailings in odd-year elections. Tr. Trans. 235, ECF No. 97 (Perlatti); Tr. Trans. 174, ECF No. 100 (Burke). Notably, before S.B. 205, BOEs were not subject to that same limitation. See Tr. Trans. 173, ECF No. 100 (Burke) (Hamilton County sent its mailings in odd and even numbered general elections and primaries).

1. Burden

The Court finds S.B. 205's restrictions regarding unsolicited absentee ballot application mailings and Directive 2014-15's policy regarding the voters to whom the Secretary directs its mailings impose a minimal burden on the right to vote of African Americans.

First, the counties' unsolicited absentee ballot application mailings created a consistency that was valuable to voters in those counties and encouraged absentee voting. See Tr. Trans. 233, ECF No. 97 (Perlatti) (discussing habitual nature of voters and explaining that Cuyahoga County's consistent mail program made voters more comfortable with the vote by mail process); Tr. Trans. 173-74, ECF No. 100 (Burke) (Hamilton County previously sent absentee ballot application mailings in odd numbered years, even numbered general elections, and presidential primaries, and this past election, the Hamilton County BOE had a number of complaints from voters who at the last minute realized they had not received an application for an absentee ballot); id. at 10-11, 16 (McNair) (Cuyahoga County mailed absentee ballot applications for every election because consistency in election administration is important for voters; they also encouraged absentee voting); Tr. Trans. 60, ECF No. 96 (Turner) (Senator Turner's office received calls about whether voters would receive absentee ballot applications); Tr. Trans. 16, ECF No. 97 (Martin) (after the prohibition on county mailings, CCDP experienced increased questions from and confusion on behalf of Democratic voters). These mailings alleviated pressure on the polls. See Tr. Trans. 233, ECF No. 97 (Perlatti); Tr. Trans. 173-74, ECF No. 100 (Burke); id. at 10-11, 16 (McNair).

Second, the prepaid postage included in certain BOEs' unsolicited absentee ballot application mailings (both for returning the application and the ballot itself) helped ensure that indigency was not a barrier to voting by mail and that confusion about how much postage was owed did not result in ballots not being delivered to BOEs. See Tr. Trans. 233, ECF No. 97 (Perlatti) (Cuyahoga County conducted unsolicited absentee ballot application mailings that included prepaid postage between 2008 and 2011); Tr. Trans. 225-27, ECF No. 96 (Anthony) (Franklin County BOE included prepaid postage to eliminate voter confusion about cost of postage, as multiple-page ballots required a larger envelope and irregular postage costs); Tr. Trans. 212, ECF No. 97 (Taylor) (received complaints from some elderly women who depended on postage); Tr. Trans. 174-75, ECF No. 100 (Burke) (aware of incidents in Hamilton County where voters did not provide sufficient amount of postage on their absentee ballots); Tr. Trans. 123, ECF No. 101 (Beswick) (describing voter confusion over postage amount); see also Tr. Trans. 233-34, ECF No. 97 (Perlatti) (amount of postage on absentee ballots varied); Tr. Trans. 10, ECF No. 100 (McNair) (same).

Third, the prohibition on the Secretary conducting unsolicited absentee ballot application mailings absent authorization and the restriction of such mailings to general elections limits the availability of an alternative means of facilitating absentee voting in some elections and eliminates it for other elections.

S.B. 205's restrictions therefore make it more difficult for voters to cast an absentee ballot by mail. This burden disproportionately affects lower-income voters, who are disproportionately African American. Indeed, Ohio's three largest counties, which include a disproportionate number of Ohio's African American population, used unsolicited absentee ballot mailings and the inclusion of prepaid postage (with the exception of Hamilton County) before S.B. 205. See Tr. Trans. 233, ECF No. 97 (Perlatti); Tr. Trans. 10, ECF No. 100 (McNair); id. at 172-73 (Burke) (Hamilton County did not include prepaid postage); Tr. Trans. 225, ECF No. 96 (Anthony). The prohibition on counties sending unsolicited absentee ballot application mailings and including prepaid postage thus burdens African American voters. Further, the Secretary's exclusion of many inactive voters from his mailings disproportionately impacts African Americans, who have higher rates of residential instability and are thus more likely than others to have out-of-date voter registrations and to be in an inactive status. Moreover, to the extent these provisions make it more difficult for voters to cast absentee ballots, they will in turn cause some voters to vote in person, which will result in increased wait times to vote. See PX 113 at 21 (Yang Rep.).

These burdens are, however, mitigated by several considerations. First, the Secretary's statewide mailing in 2016 will offset the burdens imposed by S.B. 205 for that election by providing unsolicited applications to certain voters. See Tr. Trans. 54-55, ECF No. 104 (Damschroder) (the Secretary will again mail applications statewide for the 2016 general election). Indeed, not all counties conducted mailings before S.B. 205 and not all included prepaid postage or directed the mailings to the same universe of voters. Id. at 60-62. Therefore, the Secretary's mailing, which is directed to all counties, reaches more voters statewide. See Tr. Trans. 199-200, ECF No. 100 (Burke). The Secretary also plans to conduct a supplemental mailing to individuals who newly registered or changed their registrations. The Secretary's mailing does not entirely eliminate the burdens imposed, however. The Secretary cannot conduct the mailings in odd-year elections. In at least one instance, his exclusion of a class of voters from his mailings resulted in them reaching less people in a particular county than would have that county's mailing. See id. at 199-200 (the Secretary's mailings reached fewer voters in Hamilton County than did the BOE's mailings because the BOE sent its mailings to a greater universe of voters). The Secretary's mailings also do not account for counties' individual needs. Tr. Trans. 235-37, ECF No. 97 (Perlatti); Tr. Trans. 18-19, ECF No. 100 (McNair) (opining that the Secretary's mailing was sent too early in 2012 because people do not focus on the election that early). Moreover, the record reflects that the mitigation of the burdens accomplished by the Secretary's mailing is guaranteed only for the 2016 election, as there is no evidence that the General Assembly will authorize the Secretary to conduct mailings after the 2016 general election. Nevertheless, the Secretary's statewide mailing for the 2016 general election will somewhat mitigate the burdens imposed by S.B. 205's restriction on BOE mailings.

Second, the burdens imposed by S.B. 205 are mitigated by the fact that absentee ballot applications are widely available. Candidates and political parties mail applications to voters, applications are available at many libraries, and an application is available as a downloadable form on the Secretary's and other websites. Tr. Trans. 54, ECF No. 104 (Damschroder). Moreover, BOEs may place applications on their websites or in various public places, and an individual can request an application verbally or in writing. DX 14E (Directive 2014-18). Notably, private groups, including the Democratic Party Plaintiffs, may conduct their own mailings.

Third, the burden imposed by the prohibition on including prepaid postage in mailings is somewhat diminished by the fact that one of Ohio's largest counties, Hamilton County, did not include prepaid postage in its mailings before S.B. 205 and has an understanding with its Post Office that the Post Office will still deliver to the BOE any envelopes that have insufficient postage. See Tr. Trans. 172-73, 175, ECF No. 100 (Burke). Voters also need not mail back their ballot; the voter or a family member can return it in person. While transportation, time, and childcare constraints reduce the efficacy of the option to return the ballot in person for African Americans, it cannot be entirely eliminated as an alternative to obtaining postage to mailing the ballot.

Fourth, the severity of the burden imposed on the right to vote by S.B. 205 and Directive 2014-15 must be considered in light of available alternative voting methods. One could argue that given the Court's finding that the complexities of the vote by mail process likely do not make it a viable alternative to voting in person for African Americans, any additional burdens on the ability to vote by mail would more than minimally burden the right to vote. The Court finds more persuasive, however, that voting EIP or on Election Day, which evidence indicates African Americans prefer over voting by mail, are viable alternatives to voting by mail that significantly reduce the overall burden on the right to vote imposed by S.B. 205.

In sum, while S.B. 205's restrictions on unsolicited absentee ballot application mailings and the Secretary's policy as to the recipients of his authorized mailings impose a burden on the right to vote of African Americans, the Secretary's mailings, the other means of obtaining an absentee ballot application, and the opportunity to vote EIP render that burden minimal.

2. Whether the Justifications Outweigh the Burden Imposed

Having so found, the Court employs a standard akin to that of rational basis to determine whether the justifications for the restrictions outweigh that burden. First, S.B. 205's restriction on unsolicited absentee ballot mailings by any public office other than the Secretary and prohibition on prepaid postage furthers the legitimate state interest of helping to ensure that voters across Ohio have the same opportunity to vote absentee by mail regardless of the county in which they live. See Tr. Trans. 55-65, ECF No. 104 (Damschroder); Timmons v. Twin Cities Area New Party , 520 U.S. 351, 364, 117 S.Ct. 1364, 137 L.Ed.2d 589 (1997) (states have an interest in protecting the fairness of election processes). The fact that the exclusion of certain voters from the Secretary's mailings resulted in fewer people receiving the mailings in Hamilton County than if Hamilton County conducted its own mailings does not significantly diminish the value of reaching a greater number of voters statewide. Indeed, S.B. 205 may not accomplish complete uniformity, but it certainly promotes it, and there are relatively easy alternative means for those excluded from the Secretary's mailings to otherwise obtain an absentee ballot. Second, Directive 2014-15's limitation of the Secretary's mailings to active voters promotes the State's legitimate interest in efficient election administration. See Timmons , 520 U.S. at 364, 117 S.Ct. 1364 (states have an interest in protecting efficiency of election processes). Specifically, because the Secretary sends each absentee ballot application to a specific address by nonforwardable mail, inactive voters who failed to confirm their address would be less likely to receive the mailing than active voters. Tr. Trans. 57-58, ECF No. 104 (Damschroder); DX 15 at 36 (Hood Rep.). Even if the application was forwarded to the voter's correct address, the voter would not be permitted to use the application, as the absentee ballot applications sent by the Secretary are pre-populated with the voter's name and current address. DX 14F at 2 (Directive 2014-15); Tr. Trans. 57-58, ECF No. 104 (Damschroder). This means that sending an application to an inactive voter who has moved would provide the inactive voter with an application that he or she would not be able to use. Id. at 57-58.²⁵ Directive 2014-15 prevents this waste of resources, which promotes efficient election administration.

25 Because of this, the Secretary's office mailed inactive voters a postcard in 2012 and 2014 reminding them to update their registration, and it plans to do so again in 2016. Id. at 58-59.

The Court finds that these justifications outweigh the minimal burden imposed by S.B. 205's restrictions and the policy outlined in Directive 2014. Indeed, in light of the evidence that African Americans distrust voting by mail and that viable alternative methods of voting are available, the interests in fairness and election administration are sufficiently weighty to justify imposing the minimal burdens caused by S.B. 205's restrictions. Plaintiffs' equal protection claim based on these restrictions accordingly fails.

E. S.B. 205 and S.B. 216—Informational Requirements

Plaintiffs next challenge S.B. 205's and S.B. 216's addition of date of birth and address to the identification fields that voters are required to fill out on absentee ballot envelopes and provisional ballot affirmation forms.

S.B. 205 concerns absentee ballots. To cast an absentee ballot, a voter is required to fill out and return an application to his or her BOE. Ohio Rev. Code § 3509.03 ; DX 14E (Directive 2014-18). The application requires the voter's (i) name; (ii) address; (iii) date of birth; (iv) signature; and (v) either a driver's license number, the last four digits of the voter's social security number, or a copy of the voter's current and valid photo identification or other various documents showing the voter's name and address. Id.

After verifying the voter's eligibility, the BOE provides the voter with a ballot and ballot envelope. DX 14E at 9. The voter can return the ballot to the BOE by mail or in person, and a voter's family member may also deliver the completed ballot. Ohio Rev. Code § 3509.05(A). Some BOEs maintain a drop box, allowing voters to drop off their ballot at any time. Tr. Trans. 235-36, ECF No. 96 (Anthony); Tr. Trans. 20, ECF No. 104 (Poland); id. at 64 (Damschroder).

The five required identifiers on the absentee ballot envelope mirror the information required on absentee ballot applications, Ohio Revised Code § 3509.06(D)(3)(a) ; on provisional ballot identification envelopes, Tr. Trans. 25, ECF No. 104 (Poland); and for voter registration, id. at 119. S.B. 205 added the date of birth and address requirements. BOEs preprint the address on the envelope. DX 14C (Directive 2014-27); Tr. Trans. 114-15, ECF No. 104 (Damschroder). Accordingly, the only new information that a voter is required to fill out on an absentee-ballot envelope as a result of S.B. 205 is the voter's date of birth. Tr. Trans. 115, ECF No. 104 (Damschroder).

The year of the date of birth need not be accurate for the ballot to be processed. Ohio Rev. Code § 3509.06(D)(3)(a)(iii)(I) ; Tr. Trans. 116-17, ECF No. 104 (Damschroder). Even if a voter provides a date of birth on the envelope that differs from the date of birth contained in the statewide registration database, the vote may still be counted if, by a vote of at least three BOE members, the BOE finds that the voter has provided all other required information. Ohio Rev. Code § 3509.06(D)(3)(a)(iii)(III).

If an absentee ballot envelope is missing required information, or if the information does not conform to that contained in the registration database, the BOE must send written notice to the voter detailing the defect with a pre-addressed reply envelope, and the voter has until seven days after the election to cure the defect. Ohio Rev. Code § 3509.06(D)(3)(b) ; DX 14C at 2. The BOE may identify for the voter the specific information that is missing or in error. DX 14C at 2. The failure to timely cure the deficiencies results in the ballot not being counted.

S.B. 216 concerns provisional ballots. Provisional ballots provide voters with the opportunity to vote on Election Day even when circumstances raise questions about their eligibility. See Tr. Trans. 121, 132, ECF No. 104 (Damschroder). Scenarios that raise eligibility questions include: (i) the individual does not appear on the list of eligible voters for the precinct; (ii) the individual is unable to provide acceptable identification; (iii) the individual has already requested an absentee ballot; and (iv) the individual's signature does not appear to match the signature on his registration form. Ohio Rev. Code § 3505.181(A).

An individual voting provisionally must complete a ballot affirmation form containing his (i) name; (ii) date of birth; (iii) current address; (iv) signature; and (v) include a method of identification. Ohio Rev. Code § 3505.182 ; DX 66 (Form 12-B). S.B. 216 imposed the address and date of birth requirements.

The date of birth field requires only a correct month and day of birth. Ohio Rev. Code § 3505.183(B)(3)(e). Moreover, providing a date of birth that does not match that in the statewide registration database does not require invalidation of the ballot if a BOE finds by a vote of at least three of its members that the voter satisfies all other requirements. Ohio Rev. Code § 3505.183(B)(3)(e)(ii) ; DX 14D (Directive 2014-20).

If a provisional voter's current address does not match the information in the voter registration system, and if the voter indicates he or she has moved, then the individual's provisional ballot will be counted so long as the voter has not voted elsewhere. Ohio Rev. Code § 3505.183(B)(3)(f). The complete failure to provide a date of birth, however, prevents the ballot from being counted. DX 14D (Directive 2014-20).²⁶

26 Plaintiffs assert that the complete failure to provide one's address on the provisional ballot affirmation form also results in automatic disenfranchisement, and Defendants do not deny that assertion, but the parties do not cite to any authority in support.

BOEs may begin reviewing provisional ballot envelopes the day after an election. Ohio Rev. Code § 3505.183(G)(1) ; DX 14D at 2 (Directive 2014-20). BOEs may not, however, start counting any provisional ballot until after the board members have determined, by majority vote at a public meeting, the validity of all provisional ballots within the county. Ohio Rev. Code § 3505.183(F) ; DX 14D at 2 (Directive 2014-20). BOEs must complete the validity determination and the count of provisional ballots no later than twenty-one days after an election. Ohio Rev. Code § 3505.32(A).

1. Burden

The Court finds the burden imposed by S.B. 205's and S.B. 216's additional identifying requirements for absentee ballot envelopes and provisional ballot affirmation forms is minimal.

A significant number of Ohioans voted absentee by mail in past elections. In 2012, approximately one-third of the ballots were cast either EIP or by mail. Tr. Trans. 76, ECF No. 104 (Damschroder). In 2010, 21.5% of the votes in Ohio were cast absentee by mail, compared to 22.1% in 2012 and 22.5% in 2014. DX 15 at 6 (Hood Rep.). Although the numbers are significantly less, a number of Ohioans were required to vote provisionally in past elections. See id. at 34 (208,084 provisional ballots cast in 2012 (3.69% of total votes cast) and 49,262 provisional ballots cast in 2014 (1.56% of total votes cast)); PX 109 at 58 (Timberlake Rep.) (identifying the same numbers).

Imposing additional requirements increases the amount of information that a voter must provide to vote absentee by mail or provisionally, which inevitably increases errors. The increase in errors in turn increases the opportunity for disenfranchisement in that the failure to cure those errors or omissions will result in an absentee ballot not being counted, and the omission of the date of birth on a provisional ballot affirmation form will result in disenfranchisement. The new requirements will especially burden voters with low levels of English proficiency or low literacy, as those individuals may find it more difficult to fill out their date of birth on the absentee ballot envelope or provisional ballot affirmation form.

This burden is minimal, however. With respect to absentee ballots, the only burden imposed by S.B. 205 is filling out one's birth date, and the law provides corrective mechanisms for deficiencies and allows for the vote to be counted even if the voter lists an inaccurate or non-matching birth date. With respect to provisional ballot affirmation forms, the only "burden" imposed by S.B. 216 is filling out one's birth date and address, and the law provides corrective mechanisms and allows for the vote to be counted even if the voter lists an incomplete or non-matching birth date or a nonmatching address. Specifically, a voter is required to list only a correct month and day of birth; a ballot with a non-matching birth date can still be counted by a vote of at least three BOE members; and a ballot with a non-matching address can still be counted if the voter has moved and not voted elsewhere.

Plaintiffs argue that African Americans have lower levels of educational attainment and thus will be among those who are more burdened by having to add their date of birth and address. The evidence to which Plaintiffs refer, however, is evidence that, relative to whites, African Americans are less likely to score proficient or better on standardized testing, have higher high-school dropout rates, and are less likely to have obtained a baccalaureate degree or higher. PX 109 at 26-27 (Timberlake Rep.); Tr. Trans. 186-87, ECF No. 97 (Timberlake). This evidence does not equate to evidence of literacy or English proficiency rates, and the Court is unwilling to find that lower standardized test scores or less advanced degrees indicate that African Americans who choose to vote by mail or who vote by provisional ballot will be unable or find it significantly difficult to fill out their date of birth and address.

Moreover, the expert evidence offered to establish that African Americans use absentee ballots and have those ballots rejected at higher rates than whites, see PX 109 at 55-56 (Timberlake Rep.), is entitled to little weight because it is subject to the ecological inference problem and does not distinguish between absentee by mail and EIP absentee ballots or identify the different reasons for rejection. See DX 18 at 13 (Hood Reb.). Notably, in 2014, very few absentee ballots were rejected for errors or omissions of address or date of birth. Specifically, the absentee ballot rejection rate in Ohio in 2014 was 1.5%, and two-thirds of rejected voters missed the filing deadline, failed to sign the envelope, had a non-matching signature, had already voted, or were deceased. DX 17 at ¶¶ 70-71 (Trende Reb.). Fifty-eight ballots, or .05%, were rejected due to lack of address, and six ballots, .006%, were rejected for failing to provide adequate identification. Id. at ¶ 71. Similarly, the evidence offered to establish that African Americans use provisional ballots and have those ballots rejected at a higher rate than whites, PX 109 at 57-58 (Timberlake Rep.), is entitled to little weight because it too is subject to the ecological inference problem, and it does not identify the different reasons for rejection. It is also worth noting that the provisional ballot system is itself a way to ensure that an otherwise deficient ballot is counted.

As a whole, this evidence demonstrates only a minimal burden on the right of African Americans to vote as a result of the additional information requirements.

2. Whether the Justifications Outweigh the Burden Imposed

Having found a minimal burden, the Court employs a standard akin to rational basis to determine whether the justifications for the additional requirements outweigh the burden imposed. The Court finds they do.

First, by requiring the additional identifiers, S.B. 205 and S.B. 216 make the five fields necessary for requesting an absentee ballot, casting an absentee ballot, casting a provisional ballot, and for voter registration the same. Compare Ohio Rev. Code § 3509.03 (absentee ballot application); Ohio Rev. Code § 3509.06 (absentee ballot envelope); Ohio Rev. Code § 3505.182 (provisional ballot affirmation form); Ohio Rev. Code § 3503.14(A) (voter registration). This serves the legitimate state interest of preventing confusion among election administrators. Cf. Timmons , 520 U.S. at 364, 117 S.Ct. 1364 ; see Tr. Trans. 119-20, ECF No. 104 (Damschroder) (describing prior confusion among the Secretary and BOEs); DX 15 at 37 (Hood) (conforming the information required on the absentee-ballot envelope to that required to request an absentee ballot or cast a provisional ballot "help[s] to provide standardization across these processes.") (emphasis in original).

The additional requirements also assist election administration by helping verify the identity of the voter, as identifying the voter is essential to determining voter eligibility. See DX 15 at 37 (Hood) (The newly required information "ensures that the individual casting the ballot is, indeed, the individual who was sent the absentee ballot."); Tr. Trans. 24-27, ECF No. 104 (Poland) (date of birth helps identify voter and address helps ensure ballot was cast in correct precinct); id. at 120-21 (Damschroder) (there exist voters in the database who share first and last names and the last four digits of their social security numbers); Tr. Trans. 238, 259, ECF No. 103 (Ward) (birth date is important to help identify voters with other matching identifying information); id. at 21-22 (Cromes) (address helps determine individual's precinct); Tr. Trans. 123-24, ECF No. 104 (Damschroder) ( "And, again, with the provisional, because the ballot[']s being cast because the voter's eligibility is in question, having additional data elements to search, first, the county voter file and then the statewide voter file to ... confirm the voter's eligibility so that the board can count the ballot."); id. at 124 (address helps BOEs determine that the voter cast the correct ballot style in the correct precinct). "There is no question about the legitimacy or importance of the State's interest in counting only the votes of eligible voters." Crawford , 553 U.S. at 196, 128 S.Ct. 1610. While there is evidence that some BOEs were able to identify absentee voters without notable difficulty before S.B. 205, see Tr. Trans. 243, ECF No. 97 (Perlatti); Tr. Trans. 26, ECF No. 100 (McNair); id. at 176-77, 198 (Burke); Tr. Trans. 21, ECF No. 103 (Cromes); Tr. Trans. 35-36, ECF No. 105. (Damschroder), the Court finds that to the extent other counties did not have the same experience, the importance of properly identifying voters outweighs the minimal burdens imposed by the additional requirements.

Moreover, the additional identifiers allow the provisional ballot affirmation to automatically serve as voter registration for future elections, which promotes the State's legitimate interest in efficient election administration. See Tr. Trans. 125, ECF No. 104 (Damschroder).

The Court finds that together, these justifications for S.B. 205 and S.B. 216 outweigh the minimal burdens those provisions impose, and thus, Plaintiffs' equal protection claims on these grounds fail.

F. S.B. 216—Other Provisional Ballot Restrictions

Plaintiffs also challenge two additional aspects of S.B. 216: (1) its reduction in the period to cure a provisional ballot for failure to provide adequate identification, and (2) its prohibition on elections officials filling out provisional ballot affirmation forms on a voter's behalf.

One circumstance under which a voter may be forced to vote provisionally is if he fails to provide valid identification. When an individual casts a provisional ballot, he receives notice (i) with a telephone hotline number for learning about the status of his ballot; (ii) explaining the reasons why additional information may be needed for his ballot to count; and (iii) giving instructions if he did not provide identification. DX 14D at 1 (Directive 2014-20); DX 67 (Form 12-H).

A provisional voter who does not provide identification at the time of voting has a period of time after Election Day during which to do so at his county's BOE. Ohio Rev. Code § 3505.182(D)(4). S.B. 216 reduced that period from ten days to seven days.

S.B. 216 also prevents elections officials, with limited exceptions, from completing the information required on the provisional ballot affirmation form on a voter's behalf. See Ohio Rev. Code §§ 3505.181(B)(2), (F) (permitting assistance to blind, disabled, or illiterate voters).

1. Burden

These restrictions impose a minimal burden on the right to vote of African Americans.

a. Reduction in Cure Period

The reduction in the cure period burdens the right to vote of those for whom the eighth, ninth, or tenth day after Election Day is the most convenient or only time for them to cure their ballots. In Cuyahoga County, a very small number of ballots needed correcting during the eliminated three-day period. Tr. Trans. 30, ECF No. 100 (McNair). There is some expert evidence, although entitled to little weight, that African Americans make up a disproportionate share of provisional voters and therefore will be more significantly burdened by the reduction in the cure period. See PX 109 at 58 (Timberlake Rep.) (In 2008 and 2012, voters in high-minority counties were between approximately 30% and 50% more likely than voters in other Ohio counties to have to vote by provisional ballot.).

This burden is minimal. The cure period is limited to curing the failure to provide proof of identity on Election Day.²⁷ While Plaintiffs have adduced evidence of African American provisional ballot use and rejection rates, they have not shown that those voters voted provisionally due to their failure to show proof of identity. As such, they have not shown that the reduction in the period during which to cure such deficiencies would likely impose more than just a minimal burden. Moreover, the reduction in the cure period still leaves voters with seven days to cure their ballot deficiencies, which Plaintiffs have not proven to be insufficient. The Court therefore finds the reduction in the cure period imposes only a minimal burden.

27 Voters who voted provisionally due to a voter challenge also have seven days to cure any deficiencies that resulted in the challenge, Ohio Revised Code § 3505.182(E), but Plaintiffs make no argument as to the length of the period to cure such deficiencies.

b. Prohibition on Assistance

The bar on elections officials from completing a provisional ballot affirmation form on a voter's behalf—with the exception of blind, disabled, or illiterate voters—limits the assistance that provisional voters with low literacy or limited English proficiency can receive from a poll worker, which could result in increased voter error. Plaintiffs again cite evidence that, relative to whites, African Americans are less likely to score proficient or better on standardized testing, have higher high-school dropout rates, and are less likely to have obtained a baccalaureate degree or higher. Again, however, such evidence does not establish that African Americans have lower literacy or English proficiency rates such that they would be unable or find it significantly difficult to fill out a provisional ballot affirmation form. Indeed, the prohibition on assistance does not prohibit elections officials from providing instructions on how to complete the provisional ballot form. Accordingly, the only burden is having to fill out the ballot form oneself. The Court is unwilling to find that lower standardized test scores or levels of educational attainment equate to a voter's inability to complete a provisional ballot form with instruction. The Court therefore characterizes the burden imposed by the prohibition as minimal.

2. Whether the Justifications Outweigh the Burdens

The Court must next determine whether the justifications for these provisions outweigh the burdens imposed under a standard akin to rational basis. The Court finds that they do.

S.B. 216's provisions further the legitimate interest in improved election administration by affording BOEs more time to determine the validity of provisional ballots. Cf. Obama for Am. , 697 F.3d at 442 (state has legitimate regulatory interest in relieving election workers of burdens during busy periods). Although there is evidence that one BOE did not need the extra time, see Tr. Trans. 30, ECF No. 100 (McNair), the Court finds that the importance of efficient election administration outweighs the minimal burden imposed by the reduction in the cure period.

Additionally, preventing poll workers from incorrectly filling out a voter's information is an important regulatory interest, as it helps promote accuracy and prevent ballot rejection. Accordingly, the Court finds that the justifications for these portions of S.B. 216 outweigh the minimal burdens imposed, and Plaintiffs' Anderson /Burdick claim on these grounds therefore fail.

G. Cumulative Impact

In addition to arguing that each challenged provision, alone, violates the Equal Protection Clause, Plaintiffs argue that the combined effects of the challenged provisions amount to a violation. The Court therefore considers the cumulative impact of the remaining challenged provisions.²⁸

28 The Court excludes from this analysis S.B. 238's elimination of Golden Week as inclusion of that provision would necessarily result in a finding that the cumulative impact violates Equal Protection. The Court also excludes consideration of Ohio Revised Code § 3501.10(C) (one EIP voting location) and S.B. 200 and Directive 2014-26 (reduction in DREs) as the Court found those provisions impose no burden on African Americans. Accordingly, they, too, should be excluded from any cumulative impact analysis lest they potentially be struck down due solely to the cumulative impact of other provisions whose minimal burdens together constitute a more significant burden.

The Court first considers whether the separate burdens on absentee voting combine to create a more than minimal burden on African Americans. Even considering the cumulative impact of S.B. 205's and Directive 2014-15's restrictions on absentee ballot mailings (no unsolicited mailings by BOEs, limited mailing by Secretary, no prepaid postage) and S.B. 205's informational requirements on absentee ballots (needing to add date of birth and address on the ballot envelope), the Court finds only a minimal burden on the voting rights of African Americans.

As a result of the absentee ballot regulations, it will be more difficult for voters to vote absentee because (except for when the Secretary mails them in a general election upon approval of the General Assembly) they can no longer receive an absentee ballot application without first asking for one, they will likely not be provided prepaid postage to return any ballot they do receive, and the voter will be required to fill out his or her date of birth and address on the ballot envelope. The Court finds that together, these burdens on the right to vote of African Americans are minimal. The burdens fall only on those who choose to vote by mail. As noted above, there is evidence that African Americans distrust voting by mail and prefer voting either EIP or on Election Day. The Court has reinstated the full EIP voting period, arguably reducing the impact of these provisions. For those African Americans who choose to vote by mail, many of the burdens are mitigated by the fact that absentee ballots are still accessible upon request, private organizations can send unsolicited applications, applications are available online and in public places, and there are provisions allowing a ballot to be counted even with an inaccurate date of birth. Moreover, as the Court also noted, these provisions impact the indigent and those of low literacy. Although the evidence shows that African Americans, on average, are poorer and have lower levels of educational attainment than white voters, such evidence does not establish that African American voters who choose to fill out an absentee ballot will be unable to fill out their date of birth or address.

In sum, the Court finds that even considered together, the burdens on absentee voting are minimal, and the justifications that outweigh the individual burdens suffice as well to outweigh the cumulative burdens.

With respect to the burdens on provisional ballots, the challenged regulations together require voters to add their birth date and address to a provisional ballot, prohibit poll workers from filling out such information on their behalf, and reduce the time period to cure certain defects in provisional ballots by three days. Together, these provisions still impose only a minimal burden on African Americans.

As noted above, there exist mechanisms by which provisional ballots can be counted even with incomplete or nonmatching birth dates or nonmatching addresses. Moreover, the Court has generally characterized the burden of inputting a date of birth or address as minimal, and the inability of a poll worker to fill out that information on a voter's behalf does not make that burden any more significant. Moreover, the reduction in cure period applies only to voters who failed to provide proof of identity. To the extent it "combines" with the other regulations regarding provisional ballots by requiring such a voter to not only input his or her address and date of birth but to also return within seven days instead of ten to show adequate proof of identity, the Court finds these challenged provisions do not relate to one another in a way that increases the burdens they impose from the burdens the Court has already found they impose separately.

Accordingly, considering the combined challenged provisions on provisional voting, the Court finds they impose no more than a minimal burden on the right to vote of African Americans and that the burden is outweighed by the justifications addressed in the above analysis.

Finally, Plaintiffs presumably argue that the regulations on absentee voting and provisional voting themselves combine to create an unjustified burden on the right of African Americans to vote. A voter cannot validly vote both absentee and provisionally. Accordingly, to the extent these regulations combine to create a burden on a voter, it must arise in the situation where the absentee ballot regulations make it so difficult for a voter to vote absentee that he is unable to do so and is, instead, forced to vote in person. Then, upon appearing to vote in person, the voter must be forced to vote provisionally and be burdened by the restrictions on provisional ballots. There has been no evidence presented as to the likelihood of these provisions interacting in a way that seriously burdens the right to vote of any African American. This is especially so given the Court's findings on the wide availability of absentee ballots, the fact that not all counties included prepaid postage before S.B. 205 took effect, the relative easiness of adding one's birth date and address to a form, the ability to count a ballot notwithstanding an error as to one's birth date or address, and the fact that, in 2014, very few absentee ballots were rejected for errors or omissions of address or date of birth. For the reasons addressed above, the State's justifications of uniformity, voter verification, and efficiency of administration overcome the minimal burden of even the combined effects of these provisions. Plaintiffs' Anderson /Burdick claim based on the cumulative impact on the remaining challenged provisions therefore fails.

In sum, Plaintiffs succeed on their Anderson /Burdick claim based only on S.B. 238, as the Court finds that the modest burden imposed by that provision outweighs the justifications set forth by Defendants.

IX. Voting Rights Act

Plaintiffs next argue that the challenged provisions interact with the ongoing effects of historical racial discrimination and impose disproportionate burdens on African American voters in violation of § 2 of the VRA.

Section 2 of the VRA, 52 U.S.C. § 10301, states:

No voting qualification or prerequisite to voting or standard, practice, or procedure shall be imposed or applied by any State or political subdivision in a manner which results in a denial or abridgement of the right of any citizen of the United States to vote on account of race or color, or in contravention of the guarantees set forth in section 10303(f)(2) of this title ....

52 U.S.C. § 10301.

Section 2 is violated:

if, based on the totality of circumstances, it is shown that the political processes leading to nomination or election in the State or political subdivision are not equally open to participation by members of a class of citizens protected by subsection (a) in that its members have less opportunity than other members of the electorate to participate in the political process and to elect representatives of their choice.

Id.

Section 2 "does not require proof of discriminatory intent. Instead, a plaintiff need show only that the challenged action or requirement has a discriminatory effect on members of a protected group ..." Moore v. Detroit Sch. Reform Bd. , 293 F.3d 352, 363 (6th Cir.2002). "[T]he critical question in a § 2 claim is whether the use of a contested electoral practice or structure results in members of a protected group having less opportunity than other members of the electorate to participate in the political process and to elect representatives of their choice." Thornburg v. Gingles , 478 U.S. 30, 62, 106 S.Ct. 2752, 92 L.Ed.2d 25 (1986) (citing S. REP. at 2, 27, 28, 29, n.118, 36). Additionally, courts should interpret the VRA "in a manner that provides the broadest possible scope in combating racial discrimination." Chisom v. Roemer , 501 U.S. 380, 403, 111 S.Ct. 2354, 115 L.Ed.2d 348 (1991) (internal quotation marks and citation omitted).

The Sixth Circuit has held that, in a vote denial case, such as this one, there are two elements to proving a § 2 claim:

First, the challenged "standard, practice, or procedure" must impose a discriminatory burden on members of a protected class, meaning that members of the protected class "have less opportunity than other members of the electorate to participate in the political process and to elect representatives of their choice." 42 U.S.C. § 1973(a)-(b). Second, the Supreme Court has indicated that that burden must in part be caused by or linked to "social and historical conditions" that have or currently produce discrimination against members of the protected class. Gingles , 478 U.S. at 47, 106 S.Ct. 2752 ("The essence of a § 2 claim is that a certain electoral law, practice, or structure interacts with social and historical conditions to cause an inequality in the opportunities enjoyed by black and white voters to elect their preferred representatives."). In assessing both elements, courts should consider "the totality of the circumstances." 42 U.S.C. § 1973(b).

N.A.A.C.P. v. Husted II , 768 F.3d at 554 ; see also League of Women Voters of N.C. v. North Carolina , 769 F.3d 224, 240 (4th Cir.2014) (adopting the Sixth Circuit's two factor test); Veasey v. Abbott , 796 F.3d 487, 504 (5th Cir.2015), reh'g en banc granted , 815 F.3d 958 (5th Cir.2016) (same).

There are seven factors listed in the Senate Judiciary Committee report for the 1982 amendments to Section 2 ("Senate Factors") that the Court uses to assess the "totality of the circumstances":

1. The extent of any history of official discrimination in the state or political subdivision that touched the right of the members of the minority group to register,

to vote, or otherwise to participate in the democratic process;

2. The extent to which voting in the elections of the state or political subdivision is racially polarized;

3. The extent to which the state or political subdivision has used unusually large election districts, majority vote requirements, anti-single shot provisions, or other voting practices or procedures that may enhance the opportunity for discrimination against the minority group;

4. If there is a candidate slating process, whether the members of the minority group have been denied access to that process;

5. The extent to which members of the minority group in the state or political subdivision bear the effects of discrimination in such areas as education, employment and health, which hinder their ability to participate effectively in the political process;

6. Whether political campaigns have been characterized by overt or subtle racial appeals;

7. The extent to which members of the minority group have been elected to public office in the jurisdiction.

Additional factors that in some cases have had probative value as part of plaintiffs' evidence to establish a violation are:

whether there is a significant lack of responsiveness on the part of elected officials to the particularized needs of the members of the minority group.

whether the policy underlying the state or political subdivision's use of such voting qualification, prerequisite to voting, or standard, practice or procedure is tenuous.

Gingles , 478 U.S. at 36–37, 106 S.Ct. 2752 (quoting S. REP. NO. 97-417 at 28-29). The Senate Factors are particularly relevant with respect to the second element of a § 2 claim. N.A.A.C.P. v. Husted II , 768 F.3d at 554. However, "this list of typical factors is neither comprehensive nor exclusive," and "there is no requirement that any particular number of factors be proved, or that a majority of them point one way or the other." Gingles , 478 U.S. at 45, 106 S.Ct. 2752 (quoting S. REP. at 29). Senate Factors one, three, five, and nine are particularly relevant in vote denial claims "in that they specifically focus on how historical or current patterns of discrimination ‘hinder [minorities'] ability to participate effectively in the political process." N.A.A.C.P. v. Husted II , 768 F.3d at 555.

Despite Defendants' argument, the Court need not identify an objective benchmark against which to assess the burdens imposed by the challenged provisions when assessing a § 2 claim of this sort. Id. at 556. Rather, the relevant benchmark is inherently built into § 2 claims and is whether members of the minority have less opportunity than other members of the electorate to participate in the political process and elect representatives of their choice. Id. Thus, the Court considers the challenged provisions as they are now, wholly within the State of Ohio (rather than comparing Ohio across other states) and determines whether those provisions result in minorities having less opportunity than other members of the electorate to participate in the political process and elect representatives of their choice. Id. at 559 ("The focus [in a § 2 claim] is on the internal processes of a single State or political subdivision and the opportunities enjoyed by that particular electorate.").

At the same time, a court does not improperly graft a retrogression analysis on a § 2 claim or convert a § 2 claim into a § 5 claim by comparing the opportunities of minorities to vote under prior law compared with current law when analyzing the "totality of the circumstances." Id. at 557–59. To the contrary, an analysis of whether a change in law results in a decreased opportunity of minorities to vote as compared to other voters is exactly the type of analysis required by § 2 claims. Id. at 558 ("The fact that a practice or law eliminates voting opportunities that used to exist under prior law that African Americans disproportionately used is therefore relevant to an assessment of whether, under the current system, African Americans have an equal opportunity to participate in the political process as compared to other voters."); see also League of Women Voters of N.C. , 769 F.3d at 242 (agreeing with N.A.A.C.P. v. Husted II ).

A. Discriminatory Burden

1. S.B. 238

The Court has already found, supra , that S.B. 238 imposes a burden on the rights of African Americans to vote. Based on the evidence discussed in the above Anderson /Burdick section, the Court finds that S.B. 238 results in less opportunity for African Americans to participate in the political process than other voters. The Court thus finds that the first element of a VRA claim has been satisfied with respect to that provision.

2. Ohio Revised Code § 3501.10(C), S.B. 200, and Directive 2014-26

Plaintiffs' failure to prove that Ohio Revised Code § 3501.10(C) (one EIP voting location) and S.B. 200 and Directive 2014-26 (DRE machine formulas) impose any burden on African Americans for Anderson / Burdick purposes precludes a finding that they impose a discriminatory burden under the VRA.

3. S.B. 205—Unsolicited Absentee Ballot Mailings

The Court has already found S.B. 205's prohibition on mailing unsolicited absentee ballot applications and on including prepaid postage in those mailings imposes a minimal burden on the rights of African Americans to vote and that the burden disproportionately affects African Americans. However, the Court does not find, for § 2 purposes, that there is a discriminatory burden that amounts to African Americans having less of an opportunity to participate in the political process or to elect representatives of their choice. African Americans can still request absentee ballots, even if they will not be sent unsolicited. Likewise, absentee ballot applications are available in many public places, private organizations can send unsolicited absentee ballot applications, and such applications are available online. Moreover, the evidence shows African Americans distrust voting by mail. Those who would choose to vote by mail but can no longer do so due to the failure to receive an unsolicited ballot or prepaid return postage may still vote absentee in person or in person on Election Day. Accordingly, the mere fact that public officials are prohibited from sending unsolicited absentee ballot applications, even if that prohibition imposes a disproportionate burden on African Americans, does not result in African Americans having less opportunity to participate in the political process or to elect representatives of their choice. For this reason, the Court finds Plaintiffs have not established a § 2 claim with respect to S.B. 205's restrictions on unsolicited absentee ballot application mailings.

4. S.B. 205 and S.B. 216—Additional Identifying Requirements

With respect to the additional identifying requirements imposed by S.B. 205 and S.B. 216, the Court finds that there is little to no evidence of a discriminatory burden on African Americans. The burden of having to list one's address or birth date is imposed on all voters and may be relatively more severe on voters with low literacy or English proficiency. While there is some evidence that African Americans in Ohio generally fare worse than whites in Ohio in standardized testing scores, high school drop-out rates, and higher education attainment, the Court is unwilling to find that lower standardized test scores or education levels equate to an adult voter's inability to provide one's birth date or address. Moreover, as also noted above, the evidence regarding absentee and provisional ballot rejection rates by race is entitled to little weight. Thus, Plaintiffs have not shown that these provisions result in African Americans having less of an opportunity than other members of the electorate to participate in the political process and to elect representatives of their choice. Accordingly, they have failed to show a discriminatory burden.

5. S.B. 216—Other Provisional Ballot Requirements

Plaintiffs' claim based on S.B. 216 (other provisional ballot requirements) also fails. Although the Court found S.B. 216 imposes a minimal burden on African Americans, the Court found that the burden is on those for whom the eighth, ninth, or tenth day after Election Day is the most convenient or only time for them to cure their ballots. As noted above, the evidence that African Americans make up a disproportionate share of provisional voters and will thus be more significantly burdened than other members of the electorate by this reduction is entitled to little weight. Likewise, there is little to no evidence that African Americans will be disproportionately impacted by the prohibition on poll worker assistance, as the Court declined to find that evidence of lower education attainment equates to evidence of lower literacy and English proficiency that would lead to a burden caused by having to fill out a provisional ballot affirmation form oneself.

Further, even if African Americans are disproportionately affected by these provisions, there are still seven days within which voters can cure their failure to provide identification, and they can still fill out the provisional ballot affirmation form without poll worker assistance. Therefore, the Court simply cannot find that the reduction in the cure period or prohibition on poll worker assistance will result in African Americans having less of an opportunity to participate in the political process and to elect representatives of their choice. As such, the Court finds S.B. 216 (other provisional ballot requirements) does not create a discriminatory burden.

In sum, the Court finds that of the challenged provisions, only S.B. 238 imposes a discriminatory burden on African Americans for § 2 purposes.

B. Causal Link

The Court next considers whether the discriminatory burden imposed by S.B. 238 is in part caused by or linked to social and historical conditions that have or currently produce discrimination against African Americans. In doing so, the Court looks to the totality of the circumstances and considers whether the Senate Factors have been met in this case. The Court concludes the following with respect to the Senate Factors:

1. The extent of any history of official discrimination in the state or political subdivision that touched the right of the members of the minority group to register, to vote, or otherwise to participate in the democratic process.

Plaintiffs have provided evidence that Ohio had facially discriminatory voting laws between 1802 and 1923. PX 109 at 34-35 (Timberlake Rep.). 2. The extent to which voting in the elections of the state or political subdivision is racially polarized.

Plaintiffs have provided some evidence that Ohio elections are racially polarized. As the Court has found, African Americans in Ohio consistently vote overwhelmingly for Democrats. Id. at 39-41.

3. The extent to which the state or political subdivision has used unusually large election districts, majority vote requirements, anti-single shot provisions, or other voting practices or procedures that may enhance the opportunity for discrimination against the minority group.

Plaintiffs have provided evidence that the state of Ohio has used voting practices or procedures that enhance the opportunity for discrimination against African Americans. Ohio's implementation of permissive voter challenges enhances the opportunity for discrimination against minorities. The evidence demonstrates that various groups have used Ohio's voter-challenge law to target minorities. For example, in 2004, Ohio Republicans had a plan to put 3,600 challengers inside precincts serving mostly African American voters and to object to the voter eligibility of about 23,000 legally registered Ohio voters. PX 107 at 17 (Minnite Rep.) In 2012, groups affiliated with the Tea Party's True the Vote , which was banned from Columbus and other locations, were granted permission to serve as poll watchers in various precincts in Ohio. PX 109 at 36 (Timberlake Rep.). The majority of those poll watchers appeared to disparately target African Americans. Id. ²⁹

29 The Court is generally aware that both major parties and others historically have used poll watchers.

The Court credits Dr. Timberlake's opinion that the use of poll watchers to target African Americans, in conjunction with Ohio's history of official voting exclusion and the other discriminatory burdens African Americans face, enhances the opportunity for confusion and intimidation among minority voters. Id. The Court's conclusion that Ohio's implementation of poll watchers satisfies this factor is buttressed by the identical findings of another judge of this Court and the Sixth Circuit. See N.A.A.C.P. v. Husted II , 768 F.3d at 557 ("More recently, Ohio has implemented ‘voting practices that suppress minority political participation,’ such as poll watchers and voter ID laws-practices that fall within the ambit of Senate factor three."); N.A.A.C.P. v. Husted I , 43 F.Supp.3d at 833 (citing the use of poll watchers in 2012 as evidence of the third Senate Factor).

4. If there is a candidate slating process, whether the members of the minority group have been denied access to that process.

Plaintiffs have offered no evidence with respect to this factor.

5. The extent to which members of the minority group in the state or political subdivision bear the effects of discrimination in such areas as education, employment and health, which hinder their ability to participate effective in the political process.

Plaintiffs have adduced evidence indicating that African Americans in the state of Ohio bear the effects of discrimination in areas such as employment and education. See supra pp. 39-41. In addition, they are more likely to be transient than whites and are more likely than whites to rely on public transportation. The Court finds this discrimination hinders African Americans' ability to participate effectively in the political process. See PX 109 at 31-33 (Timberlake Rep.).

6. Whether political campaigns have been characterized by overt or subtle racial appeals.

There is some evidence of racial appeals in state and national political campaigns. Dr. Timberlake opines that recent presidential elections have been characterized by overt and subtle racial appeals. See PX 109 at 42 (Timberlake Rep.). Defendants' expert, Dr. Hood, minimizes the facts that Dr. Timberlake relied upon in forming that opinion. Dr. Hood argues those facts do not show that Ohio's political campaigns have been characterized by overt or subtle racial appeals because Dr. Timberlake relies on only five instances which are not strongly connected to a particular candidate or campaign. In order to evaluate his opinion, the Court has reviewed the facts Dr. Timberlake relied upon.

The Court concludes that the facts show that Ohio's two most recent presidential elections—the first two in which an African American was a major-party candidate for president—were characterized by racial appeals. However, Dr. Timberlake has not shown that Ohio has a history of political campaigns characterized by racial appeals or that such racial appeals have become the norm. Accordingly, the Court finds there is some evidence of racial appeals in political campaigns in Ohio, but this factor does not strongly support Plaintiffs' § 2 claim.

7. The extent to which members of minority groups have been elected to public office in the jurisdiction.

This factor is neutral. There has never been an African American Governor, Attorney General, or State Auditor in Ohio, none of the current Ohio Supreme Court justices are African American, and only one of the eleven elected State Board of Education members is African American. PX 109 at 46 (Timberlake Rep.). On the other hand, the evidence shows that Ohio has made strides with respect to minority representation in public office, and African American representation in the United States House of Representatives, the Ohio General Assembly, and Columbus, Cleveland, and Cincinnati city councils is comparable to the African American share of the VAP. See DX 18 at 7-8 (Hood Reb.). Accordingly, this factor is neutral.

8. Whether there is a significant lack of responsiveness on the part of elected officials to the particularized needs of the members of the minority group.

Plaintiffs have not shown a significant lack of responsiveness on the part of elected officials to the particularized needs of African Americans. Plaintiffs argue that the process surrounding the enactment of S.B. 238 shows that Ohio elected officials have been unresponsive to the needs of minorities. Essentially, Plaintiffs reiterate their argument that the legislature was aware that repealing Golden Week would disproportionately harm African Americans and enacted S.B. 238 anyway, without even debating an amendment that would have required the Secretary to assess the racial impact it would have. Such evidence, while suggesting indifference by members of the legislature to the concerns of their colleagues, does not establish that, generally, elected officials in the State of Ohio are significantly unresponsive to the particularized needs of the members of the minority group.

9. Whether the policy underlying the state or political subdivision's use of such voting qualification, prerequisite to voting, or standard, practice or procedure is tenuous.

The Court has found that the justifications offered in support of the elimination of Golden Week were either not supported by evidence or did not withstand logical scrutiny.

Thus, the Court has found that Plaintiffs established five of the nine Senate Factors: Senate Factors one, two, three, five, and nine.³⁰ Particularly relevant to this analysis are factors five and nine. The Court now turns to an analysis of the second element of a VRA § 2 claim with respect to S.B. 238.

30 There is also some evidence with respect to factors six and eight but not enough to make those factors weigh strongly in Plaintiffs' favor.

These factors show that the burdens imposed by S.B. 238 are at least in part caused by the social and historical conditions that have produced discrimination against African Americans. Factor five shows that discrimination in employment and education already hinders African Americans' ability to participate effectively in the political process. The burdens to voting that are imposed disproportionately on African Americans by the elimination of Golden Week are caused in part by this existing discrimination in employment and education. As Dr. Timberlake explains, African Americans are less likely to own a vehicle and therefore must rely on public transportation to vote in person. PX 109 at 31 (Timberlake Rep.). They are less likely to have the flexibility to take time off of work to vote in person. Id. at 32. They are less likely to be able to arrange for childcare in order to vote. Id. Voting in person costs the voter the money and time to get to and from the polling place, and it therefore is more difficult for African Americans than whites to have the resources and ability to take leave from work in order to make two separate trips to register and vote if Golden Week is eliminated or to find a day on which they can vote if the period for EIP is shortened. Id. ; see also supra , p. 39-41.

Further, factor nine shows that S.B. 238 was passed based upon tenuous justifications.

Having considered all of the Senate Factors and the totality of the circumstances, the Court agrees with the reasoning in N.A.A.C.P. v. Husted I and N.A.A.C.P. v. Husted II and concludes that S.B. 238 interacts with the historical and social conditions facing African Americans in Ohio to reduce their opportunity to participate in Ohio's political process relative to other groups of voters and that Plaintiffs have succeeded on a § 2 claim. N.A.A.C.P. v. Husted II , 768 F.3d at 556–57 ("African Americans' lower-socioeconomic status in turn plays a key role in explaining why the disproportionate impact of SB 238 and Directive 2014-17 burdens African Americans' voting opportunities."); N.A.A.C.P. v. Husted I , 43 F.Supp.3d at 849–50 (finding the plaintiffs showed a strong likelihood of success on their § 2 claim).³¹

31 For the reasons explained supra , the record evidence showing that African American EIP voting did not decrease between 2010 and 2014 does not change this finding.

As the Court finds that the remaining challenged provisions do not impose a discriminatory burden on African Americans, the Court need not consider whether they cumulatively violate the VRA.

X. Fourteenth and Fifteenth Amendment: Invidious Discrimination

Plaintiffs next argue the General Assembly passed the challenged provisions, at least in part, with the intent to discriminate against African Americans, in violation of the Fourteenth and Fifteenth Amendments to the United States Constitution.

"Under the Equal Protection Clause of the Fourteenth Amendment, courts apply strict scrutiny to statutes that involve suspect classifications or infringe upon fundamental rights." Moore , 293 F.3d at 368.

Here, the challenged provisions are facially neutral with respect to race. "Where facially neutral legislation is challenged on the grounds that it discriminates on the basis of race, the enactment will be required to withstand strict scrutiny only if the plaintiff can prove that it ‘was motivated by a racial purpose or object’ or ‘is unexplainable on grounds other than race.’ " Moore , 293 F.3d at 369 (quoting Hunt v. Cromartie , 526 U.S. 541, 546, 119 S.Ct. 1545, 143 L.Ed.2d 731 (1999) ).³² "Proving that a law has a racially disparate impact, without more, is therefore insufficient to establish a violation of either the Fourteenth or the Fifteenth Amendment." Id. (citing Village of Arlington Heights v. Metro. Hous. Dev. Corp. , 429 U.S. 252, 264–65, 97 S.Ct. 555, 50 L.Ed.2d 450 (1977) ). A plaintiff need not prove that "the challenged action rested solely on racially discriminatory purposes ... [;]" it need only be a motivating factor in the decision. Arlington Heights , 429 U.S. at 265–66, 97 S.Ct. 555.

32 Claims brought under the Fifteenth Amendment also require proof of discriminatory intent. Reno v. Bossier Parish Sch. Bd. , 520 U.S. 471, 481, 117 S.Ct. 1491, 137 L.Ed.2d 730 (1997) (citations omitted).

"Determining whether invidious discriminatory purpose was a motivating factor demands a sensitive inquiry into such circumstantial and direct evidence of intent as may be available." Arlington Heights , 429 U.S. at 266, 97 S.Ct. 555. The Supreme Court has identified

five factors that are relevant for determining whether facially neutral state action was motivated by a racially discriminatory purpose: (1) the impact of the official action on particular racial groups, (2) the historical background of the challenged decision, especially if it reveals numerous actions being taken for discriminatory purposes, (3) the sequence of events that preceded the state action, (4) procedural or substantive departures from the government's normal procedural process, and (5) the legislative or administrative history.

Moore , 293 F.3d at 369 (citing Arlington Heights , 429 U.S. at 266–68, 97 S.Ct. 555 ). "[A]n invidious discriminatory purpose may often be inferred from the totality of the relevant facts." Washington v. Davis , 426 U.S. 229, 242, 96 S.Ct. 2040, 48 L.Ed.2d 597 (1976).³³ Here, the first factor weighs in favor of finding a discriminatory intent with respect to S.B. 238 because that provision has a disparate impact on African Americans. See supra , pp. 34-48. However, as noted above, a disparate impact is not alone sufficient to invoke strict scrutiny. Thus, the Court turns to a consideration of the final four factors to determine whether Plaintiffs have shown the challenged provisions were motivated in part by discrimination.³⁴

33 Defendants dispute the propriety of considering certain legislative history materials under this five-factor inquiry, arguing that the legislature speaks through its journal. They specifically object to the admission of videos and transcripts of legislative session hearings on the challenged provisions. As Plaintiffs correctly contend, however, such evidence is properly considered as evidence of discriminatory purpose under Arlington Heights , which explicitly states that legislative history may be highly relevant to the inquiry, "especially where there are contemporary statements by members of the decisionmaking body, minutes of its meetings, or reports." 429 U.S. at 268, 97 S.Ct. 555. Nevertheless, the extent to which such evidence demonstrates the intent of the entire legislature in enacting the laws is highly questionable. Indeed,

Inquiries into congressional motives or purposes are a hazardous matter. When the issue is simply the interpretation of legislation, the Court will look to statements by legislators for guidance as to the purpose of the legislature, because the benefit to sound decisionmaking in this circumstance is thought sufficient to risk the possibility of misreading Congress' purpose. It is entirely a different matter when [courts] are asked to void a statute that is, under well-settled criteria, constitutional on its face, on the basis of what fewer than a handful of Congressmen said about it. What motivates one legislator to make a speech about a statute is not necessarily what motivates scores of others to enact it, and the stakes are sufficiently high for [courts] to eschew guesswork.

Veasey v. Abbott , 796 F.3d 487, 501 (5th Cir.2015) (quoting United States v. O'Brien , 391 U.S. 367, 383–84, 88 S.Ct. 1673, 20 L.Ed.2d 672 (1968) ), reh'g en banc granted , 815 F.3d 958 (5th Cir.2016) (evaluating claim of discriminatory purpose under the Arlington Heights standard).

34 Although Plaintiffs assert in their briefing that "the challenged provisions" were enacted with the intent to discriminate on the basis of race, the evidence offered in support relates primarily to S.B. 238.

It requires too much speculation to find that the historical background of the challenged decisions or the sequence of events preceding the state action show a racial motivation. Indeed, Plaintiffs seem to argue more that it was Democratic successes in 2008, rather than racial bias , that prompted the General Assembly and Secretary Husted to take the challenged actions. Plaintiffs emphasize that President Obama won the African American vote in Ohio by a margin of 97-2, Tr. Trans. 200, ECF No. 103 (Trende), and that one expert has opined that Republicans have a general incentive to disenfranchise non-white voters, id. at 194. Those facts are insufficient, however, to justify a leap to the conclusion that the General Assembly passed the challenged provisions or the Secretary issued the challenged directives partly in order to disenfranchise minority voters.

Plaintiffs have also failed to show that procedural or substantive departures from the General Assembly's normal procedural process indicate a racially discriminatory purpose. First, although Plaintiffs offer evidence of the procedure in which S.B. 238 was passed, Plaintiffs offer no such evidence with respect to the other challenged provisions. See PX 10 at 9 (Sen. Gentile) (S.B. 238 was introduced only eight days before it passed the Senate, and only two hearings were held the day before the vote); PX 14 at 59 (Rep. Heard) (discussion on the bill was cut off in the House on at least one occasion). Moreover, Plaintiffs offered no evidence of the typical procedural process by which bills are passed such that the Court is unable to determine whether or how vastly the passage of S.B. 238 differed from the normal process. Finally, Plaintiffs offer no evidence linking any procedural or substantive departure to a racial motivation. "Allegations that the Legislature acted with haste and did not engage in extensive fact-finding might be a legitimate and even a valid critique of its behavior, but it does not lead to an inference of racial discrimination. Similarly, the Legislature's rejection of amendments ... is simply the operation of the democratic process." Moore , 293 F.3d at 370. There is simply no evidence to show that S.B. 238—or any of the challenged procedures—were passed in an unusual manner in order to further a racially discriminatory purpose.

The legislative or administrative history likewise fails to show the challenged provisions were passed by the General Assembly, as a whole, in part because of the impact it would have on minorities. There is some evidence the General Assembly knew some of the provisions would have a disparate impact on minorities. During floor discussions, several General Assembly members commented that the challenged statutes would make voting more difficult for voters generally, and some stated it would make voting more difficult for minorities specifically. See, e.g. , PX 12 at 40-41 (Rep. Reece) (reading commentaries about the Republican agenda to decrease access to the polls); id. at 53 (Rep. Ramos) (In opposition to S.B. 205: "African Americans ... live in cities. So when you are disenfranchising city voters, you are in fact disenfranchising African American ... voters."); id. at 60 (stating S.B. 205 "disenfranchises those amongst us who have historically been disenfranchised the most"); id. at 15 (Sen. Smith) (citing errors for which applications were denied classifying the errors as "minor paperwork issues" and claiming that S.B. 205 "will make the problem worse"); PX 18 at 9-10 (Sen. Turner) ("This bill [S.B. 205] does nothing to advance the cause of expanding and protecting the right to vote in the great state of Ohio. It does everything, everything, to put up more road blocks and barriers to protecting the right to vote."); id. at 18-21 (stating the same, emphasizing the disenfranchising effect on African American voters and citing several studies); PX 8 at 26-27 (Sen. Tavares) (expressing discontent with S.B. 216, citing the change will confuse voters, put the burden on voters, and disenfranchise "people of color."); PX 14 at 11-13 (Rep. Clyde) ("[S.B. 216] makes it harder to vote by requiring voters to provide more unnecessary information ... [and] if they leave something off or make a tiny insignificant error, their vote will be thrown out. They will be disenfranchised."); PX 18 at 19 (Sen. Turner) (explaining S.B. 238 will have a disproportionate impact on the African American community).

This evidence does not, however, carry the day. "A disproportionate effect does not violate the Equal Protection Clause, even if it was foreseen. The Supreme Court has instructed that the ‘intent’ in ‘intentional discrimination’ means more than ‘intent as awareness of consequences.’ " United States v. Blewett , 746 F.3d 647, 659 (6th Cir.2013) (quoting Pers. Adm'r of Mass. v. Feeney , 442 U.S. 256, 279, 99 S.Ct. 2282, 60 L.Ed.2d 870 (1979) ). "A violation arises only when a public official takes an action ‘because of,’ not merely ‘in spite of,’ its adverse effects upon an identifiable group." Id. Thus, even though Plaintiffs have introduced evidence that at least some in the General Assembly were aware the challenged provisions would disproportionately affect minorities, the Court does not find from the totality of the circumstances that Plaintiffs have established the challenged provisions were passed, at least in part, because of the disproportionate harm the provisions would cause minorities.

Additionally, Plaintiffs have failed to point to any statements by the General Assembly suggesting a racial motive. Specifically, the Court does not find the statements by Representatives Dovilla and Brenner indicate a racially discriminatory purpose. Rather, they discuss a desire to treat urban, suburban, and rural voters equally. See PX 12 at 5 (Rep. Dovilla) ("Treating voters consistently throughout the state, whether they are from a rural, suburban or urban area, is an essential element of protecting each citizen's equal right to the ballot box.... [T]he bill ... is consistent with the bipartisan deal struck in 2012 ... whereby every registered voter in Ohio, not just those residing in one urban county, received an application."); PX 12 at 70-72 (Rep. Brenner) ("Every single voter should be treated equally.... "[I]f you are allowing one local board of elections and it happens to be in an urban county to be able to send out absentee ballot request forms with prepaid postage when most of the rural counties can't afford it, you are definitely disenfranchising the voters in the rural counties because that isn't equal."). Thus, the Court concludes, based on a consideration of the Arlington Heights factors, and after considering the totality of the relevant facts, that Plaintiffs have failed to show a discriminatory purpose.

Without evidence of a discriminatory purpose (and because Plaintiffs do not argue the challenged provisions are unexplainable on grounds other than race), Plaintiffs fail to show that the challenged provisions are racially invidious laws subject to strict scrutiny under the Fourteenth Amendment, and their Fifteenth Amendment claim also fails.

XI. Partisan Fencing: First and Fourteenth Amendments

Plaintiffs argue that the challenged provisions were intended to suppress Democrats' votes, in violation of the First and Fourteenth Amendments. In support, they argue that those Amendments prohibit discrimination on the basis of partisan affiliation or viewpoint, citing Carrington v. Rash , 380 U.S. 89, 85 S.Ct. 775, 13 L.Ed.2d 675 (1965) and Justice Kennedy's concurrence in Vi e th v. Jubelirer , 541 U.S. 267, 124 S.Ct. 1769, 158 L.Ed.2d 546 (2004).

In Carrington , the Supreme Court found that a Texas constitutional provision prohibiting any member of the armed forces who moved his home to Texas during the course of his military duty from voting in the state violated equal protection. Carrington , 380 U.S. at 95–96, 85 S.Ct. 775. In evaluating whether the classifications drawn in the law were "reasonable in light of its purpose," the Court rejected the state's argument that it had a legitimate interest in preventing a concentration of military personnel from overwhelming the civilian community and voting in a manner contrary to that community's interests. Carrington , 380 U.S. at 93–94, 85 S.Ct. 775. In so doing, the Court noted that the " ‘[f]encing out’ from the franchise a sector of the population because of the way they may vote is constitutionally impermissible." Id. at 94, 85 S.Ct. 775.

Courts have since applied Carrington in determining whether state laws that explicitly disenfranchise a specific group of voters violate Equal Protection. Specifically, courts have relied on Carrington 's"fencing out" reasoning in finding that a state's interest in having a more committed, knowledgeable electorate is not a valid justification for such laws. See, e.g. , Cipriano v. City of Houma , 395 U.S. 701, 704–06, 89 S.Ct. 1897, 23 L.Ed.2d 647 (1969) (finding unconstitutional a state law that permits only property taxpayers to vote in certain elections and rejecting the argument that the law is justified by the interest in limiting the franchise to those who have a special pecuniary interest in or will be directly affected by the election); Evans v. Cornman , 398 U.S. 419, 423–25, 90 S.Ct. 1752, 26 L.Ed.2d 370 (1970) (applying heightened scrutiny and finding an equal protection violation where Maryland prohibited individuals residing on a federal reservation or enclave within the state from registering to vote and rejecting the argument that the law is justified by the state's interest in ensuring that voters are substantially interested in or affected by electoral decisions); Dunn v. Blumstein , 405 U.S. 330, 355, 92 S.Ct. 995, 31 L.Ed.2d 274 (1972) (finding residential duration requirements for voter registration did not pass heightened scrutiny and rejecting the argument that the requirements are justified by the interest in ensuring that voters share a common interest in matters pertaining to the community).

Carrington and its progeny do not, however, appear to create a separate equal protection cause of action to challenge a facially neutral law that was allegedly passed with the purpose of fencing out voters of a particular political affiliation. Moreover, Carrington was decided before Anderson and Burdick , which appears to be the proper standard under which to evaluate an equal protection challenge to laws that allegedly burden the right to vote of certain groups of voters.

Furthermore, Plaintiffs have not made out a First Amendment claim. Plaintiffs argue nothing about the challenged provisions burdening their rights to free expression or association under the traditional First Amendment framework and instead state that there is evidence Defendants intended to suppress the votes of Democrats. See, e.g. , Pls.' Tr. Br. 69, ECF No. 79. They seem to argue that the evidence of intentional discrimination against Democrats is sufficient to state a claim for "partisan fencing" under the First Amendment. In support, they cite a quotation from Justice Kennedy's concurrence in Vieth v. Jubelirer , 541 U.S. 267, 314, 124 S.Ct. 1769, 158 L.Ed.2d 546 (2004). The plurality holding in that case is that partisan gerrymandering claims are justiciable under the Fourteenth Amendment, but there is no acceptable standard by which to evaluate such claims. Id. at 309–13, 124 S.Ct. 1769 (Kennedy, J., concurring). In the course of his concurrence, Justice Kennedy stated that future partisan gerrymandering cases may be brought more appropriately under the First Amendment. Id. at 314, 124 S.Ct. 1769.

Plaintiffs have cited no case law that has extended Justice Kennedy's statements to the recognition of a new, general First Amendment "partisan fencing" cause of action (either within or outside of the context of partisan gerrymandering) that is proven solely by evidence of intent to discriminate based on political viewpoint. Moreover, interpreting Carrington , above, one court has noted, "[a] careful reading of the Carrington opinion would seem to reveal that the term ‘partisan fencing’ does not create an independent cause of action aside from a typical First Amendment and Equal Protection Clause claim." Lee v. Va. State Bd. of Elections , No. 3:15CV357, 155 F.Supp.3d 572, 583–84, 2015 WL 9274922, at *9 (E.D.Vir. Dec. 18, 2015). As there is no recognized separate First Amendment claim for "partisan fencing," and as Plaintiffs have failed to develop an argument under a traditional First Amendment analysis, their First Amendment claim fails.

XII. Civil Rights Act of 1964

Plaintiffs aver that S.B. 205's and S.B. 216's requirements for casting absentee and provisional ballots violates § 10101(a)(2)(B) of the CRA, formerly enumerated as § 1971(a)(2)(B). Plaintiffs concede, however, that the Court is bound by the Sixth Circuit's holding in McKay v. Thompson , 226 F.3d 752, 756 (6th Cir.2000), that there is no private right of action under § 10101(a)(2)(B), and they raise the claim only to preserve it for appeal. Accordingly, Plaintiffs' CRA claim fails.

XIII. Procedural Due Process

Plaintiffs allege that "Ohio's failure to provide notice and an opportunity to cure defects on the provisional ballot affirmation form violates voters' rights to procedural due process." Am. Compl. ¶ 206, ECF No. 41. Out of the 140 pages of briefing, Plaintiffs' argument in support of this allegation is contained in a footnote in their trial brief, is largely conclusory, and does not contain citations to evidence. Indeed, Plaintiffs did not address their procedural due process claim in their proposed findings of fact and conclusions of law. The Court therefore finds that Plaintiffs have waived their procedural due process claim. See Cook v. Donahoe , No. 3:11–cv–132, 2013 WL 93663, at *4 (S.D.Ohio Jan. 8, 2013) ("[I]ssues adverted to in a perfunctory manner, unaccompanied by some effort at developed argumentation, are deemed waived. It is not sufficient for a party to mention a possible argument in the most skeletal way, leaving the court to ... put flesh on its bones.") (citation omitted).

XIV. Equal Protection: No Uniform Standard

Plaintiffs aver that S.B. 216 violates equal protection by granting BOEs the discretion to decide whether to consolidate poll books in multi-precinct voting locations.

Since the filing of Plaintiffs' amended complaint, the Secretary has issued a permanent directive requiring BOEs to consolidate poll books in multi-precinct voting locations. See Secretary of State Directive 2015-24, available at http://www.sos.state.oh.us/elections/electionsofficials/Rules.aspx. Plaintiffs' claim is therefore moot.

CONCLUSION

The Court finds that Plaintiffs have succeeded on their equal protection Anderson /Burdick and VRA claims based on S.B. 238 (Counts I & IV). Plaintiffs waived their procedural due process claim (Count VI), which is dismissed with prejudice. Plaintiffs' equal protection claim regarding the consolidation of poll books (Count VII) is dismissed as moot. Defendants are entitled to judgment on the remainder of Plaintiffs' claims.

Accordingly, the Court:

A. DECLARES that S.B. 238's amendments to Ohio Revised Code § 3509.01 reducing the early in-person voting period from thirty-five days before an election to the period beginning the day following the close of voter registration are unconstitutional and in violation of Section 2 of the Voting Rights Act of 1965 and are accordingly unenforceable.

B. PERMANENTLY ENJOINS Defendants from enforcing and implementing S.B. 238's amendments to Ohio Revised Code § 3509.01 reducing the early in-person voting period from thirty-five days before an election to the period beginning the day following the close of voter registration.

IT IS SO ORDERED.