Opinion
1:21-MI-55555-JPB
10-11-2023
IN RE: GEORGIA SENATE BILL 202
ORDER
J. P. BOULEE, UNITED STATES DISTRICT JUDGE
This matter is before the Court on Plaintiffs' Joint Motion for Preliminary Injunction [Doc. 566]. This Court finds as follows:
Plaintiffs include the United States of America in Civil Action No. 1:21-CV-2575, and the named private plaintiffs in the following four cases: Sixth Dist. of the Afr. Methodist Episcopal Church v. Kemp, 1:21-CV-1284; The New Ga. Project v. Raffensperger, 1:21-CV-1229; Ga. State Conf, of the NAACP v. Raffensperger, 1:21-CV-1259; and The Concerned Black Clergy of Metro. Atlanta Inc. v. Raffensperger, 1:21-CV-1728.
BACKGROUND
Georgia Senate Bill 202 (“S.B. 202”) governs election-related processes and was signed into law by Governor Brian Kemp on March 25, 2021. Shortly after S.B. 202's passage, Plaintiffs filed complaints against Georgia state officials and counties (collectively, “Defendants”) challenging various provisions contained within the law. In the instant motion, Plaintiffs ask the Court to enjoin the enforcement of five provisions of S.B. 202 because, in Plaintiffs' view, the provisions violate the Fourteenth Amendment, the Fifteenth Amendment and Section 2 of the Voting Rights Act (“VRA”). As to all three claims, Plaintiffs assert that the provisions intentionally discriminate against black voters.
This list includes Brian Kemp, Governor of the State of Georgia, in his official capacity; Brad Raffensperger, Secretary of State of Georgia, in his official capacity; and individual members of the Georgia State Elections Board, in their official capacities.
The complaints name as defendants the boards of elections and registration (as well as members of those boards) from the following counties: Bibb, Chatham, Clarke, Clayton. Cobb, Columbia, DeKalb, Fulton, Gwinnett, Hall and Richmond. The master docket contains a complete list of County Defendants.
While the private plaintiffs assert violations of the Fourteenth and Fifteenth Amendments and Section 2, the United States only alleges a violation of Section 2.
As stated above, Plaintiffs seek to enjoin the enforcement of five different provisions of S.B. 202. The first provision that Plaintiffs seek to enjoin involves the use of drop boxes (the “Drop Box Provision”). The Drop Box Provision governs, among other things, the number of drop boxes allowable in each county, the location of the drop boxes and the hours of drop box accessibility. The second provision prohibits the distribution of food, drinks and other gifts to voters waiting in line at polling places (the “Food, Drink and Gift Ban”). Plaintiffs also seek to enjoin a provision that pertains to the deadline for submitting applications to vote absentee (the “Application Deadline Provision”). The fourth provision prohibits the counting of out-of-precinct provisional ballots cast before 5:00 P.M. (the “Provisional Ballot Provision”). The fifth and final provision at issue involves the requirement that voters applying for an absentee ballot provide their driver's license or state identification card number (the “Identification Provision”).
Although Plaintiffs challenged other provisions of S.B. 202, those provisions are not at issue here.
RELEVANT FACTS
In summarizing the facts of this case, the Court focuses on recent history and the sequence of events leading to the passage of S.B. 202, the 2021 Legislative Session and the challenged provisions.
1. Recent History and the Sequence of Events Leading to the Passage of S.B. 202
In considering the recent history and the sequence of events leading to the passage of S.B. 202, the Court discusses both the 2018 and 2020 elections.
A. The 2018 Election
Leading up to the 2018 election, black Georgians, who Plaintiffs contend overwhelmingly prefer Democratic candidates, increased and expanded their voter engagement efforts. [Doc. 566-3, p. 4]. These efforts included, among other things, offering “food, water, and music at the polling places” to encourage black voters to stay in line. Id. Plaintiffs contend that because of these voter engagement efforts, black voter turnout increased from 35% in the 2014 midterm election to over 49% in the November 2018 midterm election. [Doc. 566-1, p. 14]. Notably, in the 2018 election cycle, black Georgians began using absentee voting in greater numbers than they had in previous election cycles. [Doc. 566-42, p. 11].
In 2018, Democratic candidate Stacey Abrams lost Georgia's gubernatorial race. This race, which garnered statewide and nationwide attention, was close. Indeed, Governor Kemp, Abrams's opponent, won the election by a margin of 1.39%-only 8,744 votes. [Doc. 601-3, p. 3]. After Abrams's loss, her supporters voiced numerous criticisms about the election and lodged complaints about how provisional and absentee ballots were counted and how the State conducted list maintenance. Id. The complaints even included allegations that voting machines “flipped votes” from Abrams to Governor Kemp. Id. These criticisms eventually led to lawsuits. Id.
Following the 2018 election, the Georgia Legislature (“Legislature”) acted on these complaints and considered at least forty election bills. Id. at 5. Ultimately, during the 2019 Legislative Session, the Legislature was able to pass House Bill 316 (“H.B. 316”)-an omnibus election bill that included changes to the process for absentee ballots, provisional ballots, voter-list maintenance and election equipment. Id. In passing this bill, the relevant legislative committees held at least ten hearings. Id.
According to Ryan Germany, who was formerly employed as general counsel for the Georgia Secretary of State, “[m]any, if not most, of the complaints about Georgia's elections processes raised in the lead-up and following the 2018 election were based on inaccurate and misleading descriptions. However, those complaints received substantial media attention and many people seemed to believe the inaccurate and misleading descriptions.” [Doc. 601-3, p. 5]. Therefore, in Germany's view, the Legislature “needed to consider not just the actual issues but also voter perception about the election when considering whether any legislative response was necessary to address those concerns.” Id.
B. The 2020 Election
The 2020 election was held during the COVID-19 pandemic, which required state and county officials to consider and implement a host of measures to ensure that the public could vote safely. Id. at 6. One such measure was the implementation of drop boxes, which were not expressly authorized before in Georgia. Id. Another such measure was the State's decision to send unsolicited absentee ballot applications to all active voters for the June 2020 primary election. Id. In an interview addressing this measure, the Speaker of the Georgia House of Representatives, David Ralston, a Republican, warned that sending these unsolicited applications would “drive up turnout” in 2020 and lead to electoral outcomes that he did not favor. Indeed, Speaker Ralston clarified in the interview that sending applications to all voters “will be extremely devastating to Republicans and conservatives in Georgia.” [Doc. 601, p. 42].
Similar to 2018, organizations continued their various voter engagement efforts to mobilize black voters for the 2020 election. For the 2020 election, black voter turnout increased from 52% in the 2016 presidential election to over 57% in the November 2020 presidential election. [Doc. 566-42, p. 14]. In 2020, a greater percentage of black voters voted by absentee ballot than white voters. Id. at 12.
In 2020, Georgia was not able to declare a winner for president on the day of the election. Before a winner was determined, Georgia State Representative Barry Fleming wrote an op-ed titled “Republican Party wins on Election Day, and future is bright.” [Doc. 566-114, p. 3]. In the op-ed, Representative Fleming asserted that Republicans made gains in the United States House of Representatives and maintained control in state houses and governorships across the nation. Id. at 4. As to the presidential election, Representative Fleming accused “the Democrats who run Pennsylvania” of ignoring state requirements on balloting. Id. at 5. In Georgia and other closely contested states, Representative Fleming opined that Democrats “are relying on the always-suspect absentee balloting process to inch ahead.” Id. Representative Fleming stated that “[i]f elections were like coastal cities, absentee balloting would be the shady part of town down near the docks you do not want to wander into because the chance of being shanghaied is significant.” Id. Representative Fleming ended the op-ed by explaining that the Legislature would address the issue in the next session in January. Id.
It was eventually determined that Donald Trump, the Republican presidential candidate, lost the Georgia election by approximately 12,000 votes. Similar to the aftermath of the 2018 election, complaints and lawsuits pertaining to the 2020 election arose almost immediately. [Doc. 601-3, p. 8]. These complaints and lawsuits involved claims of improper counting of absentee ballots and the hacking of voting machines. Id. Widespread voter fraud, especially in areas with a higher black population, was also alleged. Id.
Following the 2020 election but before the 2021 Legislative Session began, some Republican members of the Legislature convened a series of legislative committee hearings relating to election impropriety in Georgia. [Doc. 566-8, p. 5]. These hearings were held in response to claims that the election was compromised by widespread voter fraud. Id. at 6.
Rudy Giuliani attended on Trump's behalf and, at one of the hearings, described black election officials as looking like drug dealers. [Doc. 566-1, p. 21]; [Doc. 566-9, p. 10].
2. 2021 Legislative Session and the Passage of S.B. 202
Although many of the complaints following the 2020 election were based on “inaccurate and misleading information,” the complaints nevertheless received substantial media coverage and influenced voter perception. [Doc. 601-3, p. 9]. As a result, in the legislative session following the 2020 election, the Legislature considered various bills relating to elections.
In Georgia, a regular legislative session lasts forty legislative days. [Doc. 566-3, p. 19]. In the Georgia House of Representatives (“House”), legislation pertaining to election matters is typically handled by the Governmental Affairs Committee. Id. at 9. In the Georgia Senate (“Senate”), election-related matters are addressed by the Ethics Committee. [Doc 566-7, p. 3].
For the 2021 Legislative Session, the Governmental Affairs Committee did not handle election matters. Instead, a special committee-the Special Election Integrity Committee (the “Election Committee”)-was formed to address election security, accessibility and confidence. [Doc. 566-3, p. 19]. The Election Committee was chaired by Representative Fleming and had fourteen members- ten Republicans and four Democrats. Id. at 10.
The Election Committee was one of three special committees used in the House during the 2021 Legislative Session. [Doc. 601-3, p. 10].
In the 2021 Legislative Session, the Election Committee considered almost fifty bills, which was an increase in the amount of election-related bills normally introduced in a single legislative session. Id. In total, over a six-week period, the Election Committee held twelve full committee hearings and four subcommittee meetings concerning election-related bills. [Doc. 601, p. 34].
One of the more significant bills considered by the Election Committee was House Bill 531 (“H.B. 531”). [Doc. 566-3, p. 12]. Among other things, H.B. 531 instituted new identification requirements for mail-in voting and prevented county election offices from receiving private grants. Id. H.B. 531 also required that drop boxes be located indoors at early voting sites and be used only during voting hours. Id. Lastly, H.B. 531 eliminated one Sunday of early voting and eliminated out-ofprecinct voting. Id. at 12-13. This bill was introduced on February 18, 2021, and the Election Committee met five different times to discuss the bill. Id. at 12, 15. Ultimately, the House passed H.B. 531 on March 1, 2021.
The Senate also considered its own election bills. Similar to the House Election Committee, the Senate Ethics Committee held thirteen hearings over a six-week period. [Doc. 601, p. 34]. One bill that the Ethics Committee considered was Senate Bill 241 (“S.B. 241”), a large elections bill that eliminated no-excuse absentee voting, required identification for absentee voting, limited the use of mobile voting units and allowed the State Election Board (“SEB”) to recommend to the Legislature that election superintendents be removed on certain bases. [Doc. 566-7, p. 7]. As to S.B. 241 specifically, the Ethics Committee held three hearings. Id. at 8.
S.B. 202 was originally introduced in the Senate as a three-page election bill that addressed which types of voters could be sent absentee ballot applications by the Secretary of State's office. [Doc. 566-6, p. 8]. The Senate passed S.B. 202 on March 8, 2021, and sent the bill to the House. [Doc. 601-3, p. 16]. On March 17, 2021, with five legislative days remaining in session, the House introduced a substitute S.B. 202-an omnibus bill spanning over ninety pages. [Doc. 566-3, p. 17]. Significantly, the revised version of S.B. 202 contained many provisions from H.B. 531 and S.B. 241. Id. at 18. Defendants presented evidence that “[s]uch incorporation of various legislative proposals into a single omnibus piece of legislation is commonplace.” [Doc. 601-3, p. 17]. Three hearings were held on S.B. 202 before the bill was passed out of the Election Committee on March 22, 2021. [Doc. 566-2, p. 20]. On March 25, 2021, the House passed S.B. 202 after it was debated on the floor for two hours. Id. S.B. 202 was then returned to the Senate for an “agree” or “disagree” vote. [Doc. 566-8, p. 16]. That same day, the Senate passed S.B. 202, and Governor Kemp signed it into law. Id.
S.B. 202 had been “engrossed,” which meant that amendments to the bill were not permitted.
Plaintiffs assert that the legislative process surrounding S.B. 202 lacked transparency. [Doc. 566-6, p. 4]. According to Plaintiffs, transparency was lacking due to the time of the committee meetings and the lack of advance notice of substantive changes in bills. Specifically, as to the time of the committee meetings, Plaintiffs provided evidence that these meetings were sometimes held as early as 7:30 A.M. and as late as 5:00 P.M., making it more difficult for the public to attend. Id. Plaintiffs also alleged that some of the subcommittee meetings were not livestreamed or recorded. Id. As to the lack of advance notice, Plaintiffs stated that amendments or substitutes to the election bills were sometimes not made publicly available until the day of the hearing. [Doc. 566-3, p. 13].
3. The Challenged Provisions
Plaintiffs challenge five provisions of S.B. 202. In the paragraphs that follow, the Court discusses how the provisions at issue changed existing law, the alleged justifications for the change in law and what effect, if any, the changes have had on black voters.
A. Drop Box Provision
In April 2020, as a result of the COVID-19 pandemic, the SEB enacted an emergency rule (“Emergency Rule”) authorizing county registrars to establish drop box locations as a means for absentee voters to return their ballots. See [Doc. 566- 102, pp. 2-3]. The SEB thereafter extended the Emergency Rule through the January 2021 runoff election. [Doc. 566-105, pp. 2-3]. While the Emergency Rule authorized counties to “establish one or more drop box locations,” it neither required counties to have drop boxes nor regulated how many drop box locations a county could establish. See [Doc. 566-102, pp. 2-3]. The Emergency Rule also did not regulate the placement of drop boxes other than requiring that they be located “on county or municipal government property generally accessible to the public.” Id. The Emergency Rule further provided that drop box locations may be open beginning forty-nine days before Election Day until 7:00 P.M. on Election Day. Id. Finally, the Emergency Rule established requirements for the surveillance of drop boxes, including adequate lighting and continuous monitoring by video recording devices. Id.
According to Plaintiffs, drop boxes were extremely popular with voters in the 2020 election cycle. [Doc. 566-1, p. 29]. Plaintiffs explain that during the time when the Emergency Rule was in place, most drop boxes were located outside and available to voters 24-hours a day and seven days a week. Id. Even though the drop boxes were popular with some voters, they were also the subject of various complaints. Indeed, voters complained that drop boxes were not secure and were not properly monitored. [Doc. 601-7, pp. 3-4]; [Doc. 601-3, p. 24]. Voters also voiced concerns relating to drop box misconduct and unlawful ballot harvesting. [Doc. 601-3, pp. 24-25]; [Doc. 601-2, p. 7]. According to Defendants, even though there was no evidence of widespread voter fraud in Georgia involving drop boxes, these types of complaints required the State to devote substantial resources to investigate and address the issues. [Doc. 601-3, p. 25]; [Doc. 601-5, p. 10].
The Drop Box Provision changed the Emergency Rule in several ways. Specifically, under the Drop Box Provision, drop boxes must be located indoors. Moreover, drop boxes can only be available during early voting hours and must be under constant surveillance by a person. The Drop Box Provision also requires that all counties establish at least one drop box, with an additional drop box to be established for every 100,000 registered voters.
Plaintiffs contend that the Drop Box Provision disparately impacts black voters because (1) voters in counties with a more significant black population reported the greatest decreases in the number of drop boxes and (2) more black voters returned their absentee ballots in the final four days of an election cycle, after early voting hours concluded. With respect to drop box decreases, Plaintiffs' expert asserts that about 75% of black voters, compared to 54% of white voters, reported a decrease in the number of drop boxes in their county for November 2022 relative to November 2020. [Doc. 566-46, p. 66]. Moreover, the largest decreases in drop boxes occurred in the three Georgia counties with the highest black populations. Specifically, Plaintiffs' expert asserts that Fulton County, which accounts for 13.8% of the state's black population, had twenty-nine fewer drop boxes; DeKalb County, which has 12.1% of the state's black population, had twenty-eight fewer drop boxes; and Gwinnett County, which has 7.8% of the state's black population, had eighteen fewer drop boxes. [Doc. 566-42, pp. 2930]. Notably, the expert report also shows that in 133 other Georgia counties, which comprise 34.7% of the state's black population, there was either no change or an increase in the number of drop boxes. Id. at 29.
As to Plaintiffs' contention that more black voters return their absentee ballots in the final four days of an election cycle, Plaintiffs' expert provided several graphs showing that in the 2020 and 2021 elections, drop box usage significantly increased in the final four days of the election cycle. Id. at 32-34. On a statewide level, Plaintiffs' expert did not have the racial information concerning who was using these drop boxes in the final four days. Plaintiffs' expert did, however, have racial data from one county in Georgia which showed that black voters were more likely than white voters to use drop boxes in the final four days before an election. Id. at 50-51.
B. Food, Drink and Gift Ban
Prior to the enactment of S.B. 202, Georgia law prohibited improper campaigning or solicitation of votes at polling locations. See O.C.G.A. § 21-2414. Georgia's pre-S.B. 202 voting laws also provided that any participation in the giving or receiving of money or gifts to solicit votes was punishable as a felony. See Id. § 21-2-570. According to Defendants, in recent elections, a growing number of organizations began approaching voters in line and offering items of value to voters. [Doc. 601, p. 13]. These items were not limited to just food and drink-they also included other items of value such as phone chargers. [Doc. 6012, p. 17]. Defendants also provide testimony describing situations where clearly identifiable partisans were delivering food to voters in line and where a political candidate was observed giving pizza to voters in line. Id. at 17-18. Defendants explain that these activities led to confusion among county and election officials as to the lawfulness of such activities. [Doc. 601-3, pp. 29-30]. The activities also generated complaints from voters who believed that such actions were unlawful attempts to influence voters. Id. To address these complaints and concerns, the Legislature enacted the Food, Drink and Gift Ban. [Doc. 601, p. 14].
The Food, Drink and Gift Ban prohibits the distribution of food, drinks and other gifts to voters waiting in line at polling places. Specifically, the Food, Drink and Gift Ban provides that
[n]o person shall . . . give, offer to give, or participate in the giving of any money or gifts, including, but not limited to, food and drink, to an elector . . . [or] establish or set up any tables or booths on any day in which ballots are being cast.
Plaintiffs contend that the Food, Drink and Gift Ban disparately impacts black voters because black voters are more likely to face long lines at polling places. [Doc. 566-1, p. 31]. To support this contention, Plaintiffs first provide evidence showing that Georgia voters face longer lines than voters in other states. Plaintiffs present statistics showing that in presidential elections between 2008 and 2020, 22% of Georgia voters waited more than thirty minutes to vote compared to 12.9% of voters from other states. [Doc. 566-48, p. 21].
Plaintiffs then provide evidence showing that in previous elections, black voters in particular experienced longer wait times. For instance, Plaintiffs provide statistics that in the November 2020 election, black voters waited on average over ten minutes longer to vote than white voters. Id. at 33. Plaintiffs also point to statistics from previous elections showing that black voters were more likely to face wait times of over sixty minutes to vote compared to white voters. [Doc. 566- 42, p. 22]. Specifically, as to voters waiting over sixty minutes, Plaintiffs presented evidence that: 0.5% of black voters compared to 0.4% of white voters in 2014 waited for more than sixty minutes; 3% of black voters compared to 5% of white voters in 2016 waited for more than sixty minutes; 7% of black voters compared to 5% of white voters in 2018 waited for more than sixty minutes; and 12% of black voters compared to 8% of white voters in 2020 waited for more than sixty minutes. Id. Importantly, however, Plaintiffs' expert conceded that too few respondents reported waits of over sixty minutes to make racial comparisons for those wait times reliable. Id.
According to Plaintiffs, the line warming efforts in elections prior to S.B. 202 facilitated the voting process for black voters by alleviating some of the burdens of waiting in long lines. [Doc. 566-1, p. 48]. Plaintiffs provide declarations from constituents that explain how receiving food and water encourages them to stay in line to vote. For example, Rhonda Briggins, a social action chair for Delta Sigma Theta Sorority, Inc. (“Delta”), shares the belief that line warming efforts help reaffirm the dignity of black voters. [Doc. 566-10, p. 11]. Briggins maintains that without ready access to necessities like food and water, black voters would be forced to leave the line. Id. In another example, Tamara Scott, a member of the Stone Mountain chapter of Delta, shares her experience voting in the November 2020 general election. [Doc. 566-20, p. 4]. Scott recalls a group of citizens handing out food and water to voters in line, as she waited for approximately four hours to cast her ballot. Id. Scott explained that although she contemplated leaving the line, receiving food and water was one of the reasons that she decided to stay. Id. at 5.
However, it is noteworthy that other provisions of S.B. 202 expressly address long wait times at the polls. For instance, under S.B. 202, large precincts that experience lines on election day of one hour or more are required to reduce the size of the precinct, increase the number of poll workers or obtain more equipment in subsequent elections. O.C.G.A. § 21-2-263. According to Defendants, these provisions have resulted in shorter wait times on election day. Indeed, Ryan Germany, former general counsel for the Georgia Secretary of State, noted in a declaration that average wait times were two minutes and nineteen seconds for the November 2022 general election and one minute and forty-five seconds in the December 2022 runoff election. [Doc. 601-19, p. 53]. Germany also points to a recent study showing that fewer than 10% of Georgia voters reported waiting in a voting line for more than thirty minutes in 2022. Id.
C. Application Deadline Provision
In Georgia, voters are allowed to vote absentee so long as they apply to do so. Prior to the enactment of S.B. 202, Georgia voters could apply for an absentee ballot up to four days before a primary, general or runoff election-the Friday before a Tuesday election. S.B. 202 now requires voters to apply for an absentee ballot no later than eleven days prior to an election, pushing up the last day to apply for an absentee ballot by one week.
In support of S.B. 202, Defendants assert that the four-day rule led to inefficiencies in the early voting process, as well as complaints from county officials. [Doc. 601, pp. 16-17]. According to Defendants, absentee ballots requested toward the end of the absentee ballot request period often went uncast because there was not enough time for the application to be processed and the corresponding absentee ballot to be mailed, received, completed and returned. Id. at 15-16. As such, Defendants aver that the Legislature enacted the Application Deadline Provision for the following reasons: to ensure that more ballots were properly cast and counted; to improve efficiencies and voter confidence in the absentee-voting system; and to free up county officials' time to focus on early voting activities in the week leading up to an election. Id. at 60-62.
Plaintiffs argue, however, that increased drop box availability-such as what existed under the Emergency Rule-mitigates this risk. [Doc. 556-1, p. 34].
Plaintiffs generally assert that black voters are disparately impacted by the Application Deadline Provision because black voters use absentee voting more than white voters. [Doc. 566-1, p.46]. Plaintiffs contend more specifically that narrowing the absentee ballot request window disproportionately impacts black voters because black voters have been more likely than white voters to submit applications late in the election cycle. Id. To support this assertion, Plaintiffs point to evidence from the report prepared by Dr. Bernard L. Fraga, one of Plaintiffs' experts, showing that during the November 2020 and January 2021 elections, a higher percentage of black voters who requested absentee ballots did so during the timeframe ultimately eliminated by S.B. 202: 2.2% of black voters' applications as compared to 1.4% of white voters' applications (November 2020) and 1.8% of black voters' applications as compared to 1.2% of white voters' applications (January 2021). [Doc. 566-46, pp. 49-50]. Additionally, Plaintiffs brought evidence of the overall number of absentee ballot requests that would have been rejected had S.B. 202 been in effect in November 2020-over 30,000- noting that more than 12,000 of those requests came from black voters. Id. at 50-51; [Doc. 566-1, p. 34]. Specifically, Dr. Fraga's report shows that in November 2020, 12,664 black voter absentee applications and 12,697 white voter absentee applications would have been rejected. [Doc. 566-46, pp. 50-51].
Dr. Fraga's report shows the inverse occurred in November 2018: 8.6% of white voters' absentee requests would have been rejected as untimely had S.B. 202's deadline been in effect, as compared to 7.8% of black voters' requests. [Doc. 566-46, pp. 49-50].
Plaintiffs also point out that the 12,664 black voter applications that would have been rejected represent 40% of these requests despite the fact that black registrants only made up 30% of all voter registrants for the November 2020 election. See [Doc. 566-42, p. 10].
Plaintiffs also cite to other statistics in Dr. Fraga's Report which show that of those applications that were rejected, 11.0% of black voters' rejected applications were rejected for arriving too late-as compared to 7.8% of white voters' rejected applications-in November 2020, and 28.5% of black voters' rejected applications were rejected for arriving too late-as compared to 24.2% of white voters' rejected applications-in November 2022 (after S.B. 202 was in effect). Id. at 45-46.
Applications can be rejected for a number of reasons.
Dr. Fraga's report shows the opposite result in another election: in January 2021, white voters' rejected applications were more frequently rejected for being too late (14.5%) as compared to black voters' (6.0%). [Doc. 566-46, pp. 45-46].
To counter this evidence, Defendants rely on statistics from their expert, Dr. Justin Grimmer, which show that in November 2022, a very small percentage of overall absentee ballot requests were rejected for being late. [Doc. 601-17, pp. 6971]. According to Dr. Grimmer's report, only 0.25% of all mail-in absentee ballot applications in the 2022 general election were rejected for arriving outside the deadline imposed by S.B. 202 (706 total rejections). Id. at 70-71. Broken down by race, the data shows that 0.27% of black voters' absentee ballot applications arrived after the deadline, compared to 0.22% of white voters' requests. Id. at 70.
D. Provisional Ballot Provision
As a general rule, to vote on Election Day, a voter must vote at his or her designated polling place or precinct. For almost twenty years, Georgia voters who failed to comply with this rule by appearing to vote somewhere other than their designated polling place have been permitted to cast provisional ballots. Before the enactment of S.B. 202, these provisional ballots were counted so long as the voter was eligible to vote, even if that voter appeared to vote at the wrong polling place.
The Provisional Ballot Provision changes the previous procedure and provides, in pertinent part, the following:
If the person presents himself or herself at a polling place in the county in which he or she is registered to vote, but not at the
precinct at which he or she is registered to vote, the poll officials shall inform the person of the polling location for the precinct where such person is registered to vote. The poll officials shall also inform such person that any votes cast by a provisional ballot in the wrong precinct will not be counted unless it is cast after 5:00 P.M. and before the regular time for the closing of the polls on the day of the primary, election, or runoff and unless the person executes a sworn statement, witnessed by the poll official, stating that he or she is unable to vote at his or her correct polling place prior to the closing of the polls and giving the reason therefor.
...
If the person is voting a provisional ballot in the county in which he or she is registered to vote but not at the precinct in which he or she is registered to vote during the period from 5:00 P.M. to the regular time for the closing of the polls on the day of the primary, election, or runoff, the person shall execute a sworn statement, witnessed by the poll official, stating that he or she is unable to vote at his or her correct polling place prior to the closing of the polls and giving the reason therefor.
As made clear by the text of the rule above, the Provisional Ballot Provision does not completely eliminate out-of-precinct voting. Instead, the Provisional Ballot Provision merely changed the rule regarding counting these ballots. Under the Provisional Ballot Provision, these provisional ballots are not counted if they are cast before 5:00 P.M. on Election Day.
Defendants offered several justifications for the Provisional Ballot Provision. According to Defendants, the increased use of out-of-precinct provisional ballots in recent elections led to inefficiencies on Election Day and decreased voter confidence. As to the inefficiencies on Election Day, Defendants presented evidence that processing a single out-of-precinct voter is a timeconsuming process, often taking fifteen to twenty minutes for each ballot. [Doc. 601-16, p. 6]. Once cast, election officials must manually duplicate out-of-precinct provisional ballots so that they can be counted. [Doc. 601-13, pp. 21-22]. This process, in addition to taking time (and potentially delaying election results), can lead to allegations of fraud and decreased voter confidence. Id. The ballot duplication process, though proper, generates numerous complaints because voters see election officials filling out ballots and suspect fraudulent votes are being cast. See id; [Doc. 601-9, pp. 17-18]. Moreover, if voters can simply arrive at any polling place on Election Day and vote, election officials cannot predict how many voters will appear at any given precinct. [Doc. 601, p. 65]. This can lead to longer lines or an inadequate number of poll workers. Id.
Plaintiffs assert that the Provisional Ballot Provision disproportionately harms black voters because black voters are more likely than white voters to cast such ballots. To support this claim, Plaintiffs presented limited data. According to Plaintiffs, in 2018, 12.1% more black voters than white voters cast provisional ballots, and, in 2021, 18.3% more black voters than white voters cast provisional ballots. [Doc. 566-42, p. 55]. In 2020, however, white voters were 2.5% more likely than black voters to cast provisional ballots. Id. In response, Defendants brought statistics showing the small percentage of provisional ballots cast overall, noting that well under 1% of all ballots cast in the last five elections were provisional ballots-for both black and white voters. [Doc. 601, p. 73]; [Doc. 60117, p. 54]. Indeed, the 2018 election resulted in the highest share of provisional ballots from the past five years, and even in that year, only 0.22% of all votes were cast as provisional votes. [Doc 601-17, pp. 53-54].
E. Identification Provision
All Georgia voters are permitted to vote absentee. To vote absentee, voters have always had to comply with certain identification requirements. [Doc. 566-31, p. 4]. Before the enactment of S.B. 202, a voter had to comply with two identification requirements. Specifically, the voter was required to provide their date of birth and their signature, which had to match the voter's signature in the voter file. Id. at 3-4.
In response to numerous lawsuits about the signature-match requirement, S.B. 202 changed the identification requirements. In lieu of a signature, the Identification Provision requires that a voter provide the number of his or her Georgia driver's license or identification card (hereinafter, “DDS ID”). The DDS ID number must be provided at two stages of the absentee voting process. First, the DDS ID number must be supplied on the absentee ballot application. In the event a voter does not have a DDS ID, the voter may provide a copy of another form of identification. These other forms of identification include, but are not limited to, “a copy of a current utility bill, bank statement, government check, paycheck, or other government document that shows the name and address of such elector.” O.C.G.A. § 21-2-417(c). Significantly, in the event an individual does not possess a valid DDS ID, individuals may obtain one free of charge.
As stated above, a DDS ID number must be supplied twice during the absentee voting process. In addition to providing the number when submitting an application, a voter must also provide his DDS ID number on the absentee ballot envelope. If a voter lacks a DDS ID at this stage, the voter may simply write the last four digits of his Social Security number to confirm his identity.
Defendants offered numerous justifications for the Identification Provision, including the State's interests in orderly and efficient election administration, preventing fraud and increasing voter confidence. [Doc. 601, p. 67]. Specifically, Defendants contend that the Identification Provision makes elections more efficient because the Identification Provision is an objective requirement. Indeed, before S.B. 202, the signature-match process took election officials three to four minutes per voter, while the objective verification takes less than one minute. Id.
In Georgia, 97% of registered voters have a DDS ID associated with their voting record, and the number of registered voters without a DDS ID has been declining. For instance, in April 2021, approximately 272,700 registered voters lacked a DDS ID, while in November 2022, roughly 243,000 registered voters lacked a DDS Id. Plaintiffs contend that the Identification Provision disproportionately impacts black voters because 56.5% of the 272,700 registered voters lacking a DDS ID and 53% of the 243,000 registered voters lacking a DDS ID are black. Plaintiffs also assert that the Identification Provision disproportionately impacts black voters because 5.6% of all black registered voters lack a DDS ID while only 2% of all white registered voters lack a DDS ID.
ANALYSIS
Plaintiffs seek a preliminary injunction in this case. A plaintiff seeking preliminary injunctive relief must show (1) a substantial likelihood of success on the merits; (2) irreparable injury absent an injunction; (3) that the balance of equities is in his favor; and (4) that an injunction would not be adverse to the public interest. Sofarelli v. Pinellas County, 931 F.2d 718, 723-24 (11th Cir. 1991). Because a preliminary injunction “is an extraordinary and drastic remedy,” the Court may not issue such relief “unless the movant clearly establish[es] the burden of persuasion as to each of the four prerequisites.” Siegel v. LePore, 234 F.3d 1163, 1176 (11th Cir. 2000) (internal punctuation omitted) (quoting McDonald's Corp. v. Robertson, 147 F.3d 1301, 1306 (11th Cir. 1998)). Granting a preliminary injunction is thus the exception rather than the rule. See id.
1. Likelihood of Success on the Merits
To obtain a preliminary injunction, the moving party must show a substantial likelihood that he will ultimately prevail on the merits of his claim. Sofarelli, 931 F.2d at 723. This factor is generally considered the most important of the four factors, see Garcia-Mir v. Meese, 781 F.2d 1450, 1453 (11th Cir. 1986), and failure to satisfy this burden-as with any of the other prerequisites-is fatal to the claim, see Siegel, 234 F.3d at 1176. In this case, Plaintiffs contend that the provisions violate the Fourteenth Amendment, the Fifteenth Amendment and Section 2 of the VRA. In the analysis that follows, the Court will analyze whether Plaintiffs have shown that they have a substantial likelihood of success on any of their claims.
A. Violation of the Fourteenth & Fifteenth Amendments
Plaintiffs assert that the provisions violate both the Fourteenth and Fifteenth Amendments. The Fourteenth Amendment provides, in pertinent part, that “[n]o State shall . . . deny to any person within its jurisdiction the equal protection of the laws.” U.S. Const. amend. XIV. In a case such as this, a successful Fourteenth Amendment claim “requires proof of both an intent to discriminate and actual discriminatory effect.” Greater Birmingham Ministries v. Sec'y of State for State of Ala., 992 F.3d 1299, 1321 (11th Cir. 2021). The Fifteenth Amendment states that “[t]he right of citizens of the United States to vote shall not be denied or abridged by the United States or by any State on account of race, color, or previous condition of servitude.” U.S. Const. amend. XV. Similar to the Fourteenth Amendment, successful Fifteenth Amendment claims require a showing of “racially discriminatory motivation.” Greater Birmingham Ministries, 992 F.3d at 1321.
Claims brought pursuant to the Fourteenth and Fifteenth Amendments are subject to a two-prong burden-shifting test. Id. Under the first prong, a plaintiff “must prove both that the law will have a discriminatory impact and that it was adopted with discriminatory intent.” League of Women Voters of Fla. Inc. v. Fla. Sec'y of State, 66 F.4th 905, 922 (11th Cir. 2023). “If Plaintiffs are unable to establish both intent and effect, their constitutional claims fail.” Greater Birmingham Ministries, 992 F.3d at 1321. Under the second prong, “‘the burden shifts to the law's defenders to demonstrate that the law would have been enacted without this [racial discrimination] factor.'” Id. (quoting Hunter v. Underwood, 471 U.S. 222, 228 (1985)).
As stated immediately above, the first prong requires courts to determine whether the law has both a discriminatory intent and effect. The Eleventh Circuit Court of Appeals has held that the “concept of discriminatory intent is sometimes subtle and difficult to apply.” Dowdell v. City of Apopka, 698 F.2d 1181, 1185 (11th Cir. 1983). Consequently, courts must “evaluate all available direct and circumstantial evidence of intent in determining whether a discriminatory purpose was a motivating factor in a particular decision.” Burton v. City of Belle Glade, 178 F.3d 1175, 1189 (11th Cir. 1999). The Supreme Court of the United States, in Village of Arlington Heights v. Metropolitan Housing Development Corp., 429 U.S. 252, 265 (1977), identified the following non-exhaustive list of factors that a court should consider when assessing discriminatory intent: (1) the impact of the challenged law; (2) the historical background; (3) the specific sequence of events leading up to its passage; (4) the procedural and substantive departures; and (5) the contemporary statements and actions of key legislators. The Eleventh Circuit has supplemented the Arlington Heights list to include the following additional factors: (6) “the foreseeability of the disparate impact;” (7) “the knowledge of that impact;” and (8) “the availability of less discriminatory alternatives.” Greater Birmingham Ministries, 992 F.3d at 1321. In the analysis that follows, the Court will consider each of these factors in determining whether Plaintiffs have met their burden of showing that the Legislature acted with a discriminatory intent.
i. Impact of the Challenged Law
In determining whether Plaintiffs are able to show that the Legislature acted with a discriminatory intent, the first factor that the Court should consider is the impact of the challenged law. This requires the Court to analyze whether the law bears more heavily on one race than another. Id. “When considering disparate effect the focus should not be on absolute numbers but rather on whether the challenged requirements ‘operate to disqualify [minority voters] at a substantially higher rate.'” Williams v. City of Dothan, 818 F.2d 755, 764 (11th Cir. 1987).
Before proceeding to analyzing whether each individual provision disparately impacts black voters, the Court notes that Defendants presented evidence showing that in a survey conducted by the University of Georgia's School of Public and International Affairs concerning the November 2022 election-after S.B. 202 was implemented-voters reported high levels of satisfaction. [Doc. 677, p. 252]. Indeed, 99.5% of black voters reported no problem when voting, and no black voters reported poor experiences when voting. Id. at 116, 252.
(a) Drop Box Provision
Plaintiffs argue that the Drop Box Provision disparately impacts black voters because black voters now have access to fewer drop boxes than they did in 2020 and black voters were more likely than white voters to cast absentee ballots in the final four days of the election cycle. As previously noted, Plaintiffs provided evidence showing that (1) the counties with the highest decreases in drop boxes had the highest percentage share of the state's black population and (2) voters used drop boxes at higher rates in the last four days of an election cycle. The Court, however, finds several problems with Plaintiffs' proffered evidence.
First, Plaintiffs submitted evidence showing that 75% of black voters and 54% of white voters reported a decrease in the number of drop boxes in their county for November 2022 relative to November 2020. Plaintiffs also provided evidence showing that the largest decreases in drop boxes occurred in three Georgia counties with the highest black populations, which comprise a total of 33.7% of the state's black population. Plaintiffs contend that these statistics demonstrate that black voters are disparately impacted by the Drop Box Provision's restriction on the number of additional drop boxes that counties can establish. The Court disagrees. Even assuming that Plaintiffs' evidence from only three Georgia counties is sufficient to support a finding of disparate impact, the Court cannot overlook that additional evidence was presented that showed that in 133 other Georgia counties, which comprise 34.7% of the state's black population, there was no change or an increase in the number of drop boxes in those counties. Therefore, Plaintiffs have not demonstrated, at least at this stage of the proceedings, a disparate impact with respect to decreases in drop box locations. Moreover, evidence related to the removal of drop boxes-“absent information about the race of the voters” who used the drop boxes in those counties-“does not prove that the [Drop Box Provision] will have a disparate impact on black voters.” League of Women Voters, 66 F.4th at 935-36. Indeed, “testimony that some counties will offer fewer drop boxes in the future cannot establish disparate impact without reliable evidence of voting patterns as to race.” Id. at 936.
Plaintiffs further contend that eliminating drop box access after the early voting period will disproportionately impact black voters because black voters are more likely to submit their absentee ballots in the four days leading up to Election Day. This argument fails, at least at this stage of the proceedings, for two reasons. First, although Plaintiffs have brought statewide evidence showing an overall increase in drop box usage for all voters in the four days leading up to an election, this evidence does not necessarily demonstrate a disparity in drop box usage during this period when comparing black voters to white voters. Indeed, Plaintiffs provided no evidence concerning the race of the voters depositing their absentee ballots in the drop boxes. Second, the evidence that Plaintiffs provide that does demonstrate racial disparity in drop box usage during this time period only pertains to one county in Georgia. The Court is not convinced that a statewide statistical relationship can be inferred from data from a single county. See League of Women Voters, 66 F.4th at 933-34 (declining to reach a statistical conclusion about a state from an analysis driven predominantly by data from a single county).
To summarize, Plaintiffs have not demonstrated that they are substantially likely to succeed in showing that the Drop Box Provision has a disparate impact on black voters.
(b) Food, Drink & Gift Ban
Plaintiffs argue that the Food, Drink and Gift Ban disproportionally impacts black voters because black voters are more likely to wait in long lines at the polls. At least on this record, the Court is not persuaded. Plaintiffs presented evidence showing that, for voters waiting in line for more than sixty minutes, there were statistical differences of 0.1%, 2%, 2% and 4% for the elections held between 2014 and 2020. However, in League of Women Voters, the Eleventh Circuit opined that percentage point differences this small were not significant. See 66 F.4th at 935 (noting that “[a] difference of only 1.3 percentage points is not substantial,” and that statistical differences of just over two percentage points are “so small . . . that they cannot support a finding of disparate impact.”). Here, at least for 2014, 2016 and 2018, the statistics underlying Plaintiffs' arguments show even smaller differences than those reviewed and rejected by the Eleventh Circuit in League of Women Voters. As such, this Court similarly finds that the differences presented by Plaintiffs in this case are not statistically significant enough to demonstrate that black voters wait in longer lines at a meaningfully higher rate than white voters. Nonetheless, even assuming these percentage point differences were statistically significant, Plaintiffs' expert concedes in his report that too few respondents reported wait times of longer than sixty minutes to make racial comparisons for those wait times reliable.
Even if this Court were to assume that black voters wait in longer lines than white voters, “that finding would not alone support a conclusion that [the Food, Drink and Gift Ban] has a disparate impact on black voters.” Id. Here, Plaintiffs have failed to provide sufficient evidence to show that “by restricting the ability of third parties to hand out water bottles and snacks, the [Food, Drink and Gift Ban] makes it harder for voters in line to cast their ballots.” Id. Ultimately, especially at the preliminary injunction phase where Plaintiffs have to show a substantial likelihood of success, the Court is not convinced that voters would leave the line if not provided food, drinks and other gifts. Id.
(c) Application Deadline Provision
Plaintiffs contend that the Application Deadline Provision disproportionately affects black voters because black voters (1) use absentee voting more than white voters and (2) are more likely to submit absentee ballot requests that will be rejected for failure to comply with S.B. 202's new eleven-day deadline. [Doc. 566-1, p. 46]. As an initial matter, the Court is not convinced that data showing that black voters vote absentee more often than white voters proves that the Application Deadline Provision of S.B. 202 disproportionately impacts black voters. Put differently, even if black voters use absentee voting more frequently, the Court is not persuaded that this necessarily means that black voters will be disparately impacted by a provision governing when absentee ballot applications should be returned. Without more, generalized evidence regarding the use of absentee voting is not sufficient to show that this particular provision, pertaining to one aspect of absentee voting, is discriminatory.
Moreover, the Court is not convinced that Plaintiffs have shown reliable and statistically significant evidence demonstrating a disparate impact on this basis in any case. Plaintiffs point to the 4.9% difference between black voters and white voters who used absentee voting in 2020. [Doc. 566-42, p. 12]. However, the evidence also shows that in 2018, the difference was only 2.37% (and, in 2022, 1.84%). Id. As discussed previously, the Eleventh Circuit has found that differences of just over two percentage points did not support a finding of disparate impact. See League of Women Voters, 66 F.4th at 935. As such, even assuming the percentage point difference in 2020 was statistically significant enough to support a finding of disparate impact, the Court cannot ignore evidence from other years which support the opposite conclusion.
The Court next considers the evidence brought by Plaintiffs which purportedly shows that the Application Deadline Provision disproportionately impacts black voters because black voters are more likely to submit absentee ballot requests late in the election cycle (closer to the previous four-day deadline). First, the Court addresses Plaintiffs' counterfactual data-that is, the statistics indicating that if S.B. 202 had been in effect at the time of the November 2020 and January 2021 elections, a disproportionate number of black voter absentee applications would have been rejected due to untimeliness (falling outside of S.B. 202's elevenday deadline). Dr. Fraga's report shows that in November 2020, had S.B. 202's deadline been in effect, 2.2% of black voters' applications (12,664 total applications) would have been rejected as untimely as compared to 1.4% of white voters' applications (12,697 total applications)-a difference of 0.8%. [Doc. 56646, pp. 49-51]. And in January 2021, 1.8% of black voters' applications (8,176 total applications) as compared to 1.2% of white voters' applications (8,349 total applications) would have been rejected-a difference of 0.6%. Id. These statistics reveal even smaller differences among white and black absentee voters than those reviewed and dismissed by the Eleventh Circuit in League of Women Voters. See 66 F.4th at 934-35. Given the Eleventh Circuit's precedent in that case, and for the same reasons previously discussed, this Court cannot find that such small percentage point differences are statistically significant enough to demonstrate a disparate impact.
Dr. Fraga's report shows a difference of 0.8% in the opposite direction for the November 2018 election: 8.6% of white voter applications would have been rejected as compared to 7.8% of black voter applications. [Doc. 566-46, p. 49].
Moreover, as noted above, these statistics counterfactually impose a deadline that did not exist at the time. However, the Court need not rely only on counterfactual data. Plaintiffs and Defendants also bring statistical evidence of the new deadline's effect on absentee ballot applications in the 2022 general election, after S.B. 202 was enacted. Plaintiffs and Defendants present these statistics differently.
Plaintiffs point to data showing that of those absentee ballot requests that were rejected in November 2022, 28.5% of black voters' rejected applications were rejected for arriving too late-as compared to 24.2% of white voters' rejected applications. [Doc. 566-46, pp. 45-46]. Defendants, on the other hand, rely on statistics showing that of all Georgia voters requesting absentee ballots, only 0.25% of total applications were rejected for falling outside of S.B. 202's elevenday deadline (0.27% of black voters' applications and 0.22% of white voters' applications). [Doc. 601-17, pp. 70-71]. In short, by using differing denominators (absentee ballot applications that were rejected, as in Dr. Fraga's Table 7, versus absentee ballots applications that were requested, as in Dr. Grimmer's Table 12), but the same numerator (absentee ballot applications rejected for missing the deadline), Plaintiffs and Defendants offer starkly different percentages for the Court to consider (differences of 4.3% and 0.05%, respectively).
Plaintiffs also reference the same statistics from the November 2020 election, which show that of those absentee-by-mail ballots that were rejected, 11.0% of black voters' rejected applications were rejected as untimely, as compared to 7.8% of white voters' rejected applications. [Doc. 566-46, pp. 45-46].
In the Court's view, in considering whether black voters have been disparately impacted by the Application Deadline Provision, the relevant question is how many black voters attempted to vote absentee-by-mail and were unable to do so because their application missed S.B. 202's new deadline (compared to white voters whose applications were rejected for the same reason). And, for the November 2022 election, the answer to this question reveals a small disparity between black and white voters: 0.05% (0.27% of black voters' applications rejected for untimeliness as compared to 0.22% of white voters' applications). Again, given the Eleventh Circuit's precedent on the statistical significance of similar evidence in League of Women Voters, the Court cannot find that Plaintiffs have provided sufficient evidence to show that the Application Deadline Provision has a disparate impact on black voters. See 66 F.4th at 934-35.
Using Plaintiffs' methodology-the percentage of rejections that were related to untimeliness-it is difficult to understand the overall population of black voters whose absentee ballot applications were rejected for failure to comply with the new deadline. Cf. Brnovich v. Democratic Nat'l Comm., 141 S.Ct. 2321, 2345 (2021) (explaining how statistics can be used to make a small difference in outcomes appear larger).
(d) Provisional Ballot Provision
Plaintiffs assert that the Provisional Ballot Provision has a disparate impact on black voters because black voters are more likely to cast provisional ballots in the wrong precinct. To support this claim, Plaintiffs presented limited data from a select number of counties which showed that in 2018, 12.1% more black voters cast provisional ballots, and in 2021, 18.3% more black voters cast provisional ballots. [Doc. 566-42, p. 55]. In 2020, however, white voters were 2.5% more likely to cast provisional ballots. Id. Defendants, on the other hand, pointed to the small percentage of provisional ballots cast overall, noting that less than 1% of ballots cast in the last five elections were provisional ballots-for both black and white voters. [Doc. 601, p. 73]; [Doc. 601-17, p. 54].
As an initial matter, the Court is concerned that the statistical evidence presented by Plaintiffs may be misleading given that such a small percentage of voters voted via provisional ballot. Moreover, in claiming that black voters are more likely to cast provisional ballots than white voters, Plaintiffs relied on evidence from only thirteen Georgia counties. As such, the Court is not completely convinced that black voters are more likely to cast provisional ballots. Although the Court is concerned about the relatively small sample size, in deciding this motion, the Court will nevertheless accept as true the fact that black voters are more likely to cast provisional ballots than white voters.
Indeed, in 2020, only 0.19% of all votes were cast by provisional ballots. And, in 2018-the election with the highest share of provisional ballots in the last five years- only 0.22% of ballots were provisional. See [Doc. 601-17, pp. 53-54].
A finding that black voters are more likely to cast provisional ballots “is not evidence that [the Provisional Ballot Provision] will have a disparate impact on black voters in [Georgia].” League of Women Voters, 66 F.4th at 935. This Court finds that to show a disparate impact, Plaintiffs should have offered more evidence. Indeed, to truly know whether the Provisional Ballot Provision disparately impacts black voters, it would have been helpful for the Court to know how many of those provisional ballots cast in previous election years were cast after 5:00 P.M. as opposed to before 5:00 P.M. As to the ballots cast before 5:00 P.M., which would not be counted under the Provisional Ballot Provision, the Court would be curious to know the race of the voters casting those ballots and whether any type of racial disparity existed as to those ballots specifically. In sum, because there is no reliable evidence from which to infer that black voters are more likely than white voters to cast provisional ballots in the wrong precinct before 5:00 P.M. on Election Day, the Court cannot find that the Provisional Ballot Provision disparately impacts black voters. See id. at 936 (explaining that without evidence concerning the race of the voter using drop boxes at particular times, the evidence was insufficient to support a finding of disparate impact).
To underscore this point, at the hearing, Plaintiffs presented testimony from Mark Burns, who cast an out-of-precinct ballot in 2022. Burns testified that in 2022, he elected to vote in the incorrect precinct because traveling to his precinct required a 1 or 1.5 mile walk uphill. [Doc. 677, p. 146]. Ultimately, even though Burns voted in the wrong precinct, Burns's vote was counted because he cast his provisional ballot around 6:00 P.M. Id. at 150.
(e) Identification Provision
Plaintiffs assert that the Identification Provision disproportionately impacts black voters because black voters are less likely than white voters to have a DDS Id. To support this claim, Plaintiffs presented evidence that 5.6% of all black registered voters lack a DDS ID while only 2% of all white registered voters lack a DDS Id. Defendants do not dispute the statistical evidence submitted by Plaintiffs. Instead, Defendants assert that the Identification Provision does not have a disparate impact on black voters because voters without an appropriate DDS ID may obtain one for free, and voters may use a host of other forms of identification to vote by absentee ballot.
The Court recognizes that Plaintiffs presented evidence that showed a statistical difference of 3.6% in black registered voters that lack a DDS ID compared with white registered voters. However, Plaintiffs presented no evidence that black registered voters fail to have the other acceptable forms of identification allowed by the statute at a statistically higher rate than white voters. These other forms of identification include utility bills, bank statements, paychecks and other government documents that include a name and an address. O.C.G.A. § 21-2-417(c). Ultimately, without this additional evidence, the Court cannot find at this time that the Identification Provision has a disparate impact on black voters.
***
As explained above, Plaintiffs have not shown, at least at this stage of the proceedings, that any of the provisions have a disparate impact on black voters. The Court thus weighs this factor in favor of Defendants and against a discriminatory purpose finding.
ii. Historical Background and Specific Sequence of Events Leading up to the Passage of S.B. 202
Courts are required to consider the historical background as well as the specific sequence of events in analyzing the intent of the Legislature. In their argument, Plaintiffs discuss the historical background in combination with the specific sequence of events leading to the passage of the legislation. As such, the Court considers these factors together.
The Court has reviewed Plaintiffs' expert report providing a detailed history of the disenfranchisement of black voters in the United States and in Georgia specifically. See [Doc. 566-49]. Although Plaintiffs' report provides useful context pertaining to Plaintiffs' challenges to S.B. 202, the Eleventh Circuit has instructed that courts should give little weight to distant history and focus on the “precise circumstances surrounding the passing of the law in question.” League of Women Voters, 66 F.4th at 923. As such, the Court focuses on the evidence Plaintiffs provide regarding recent history, as well as the specific sequence of events leading to the passing of the legislation.
In their motion, Plaintiffs argue that these factors weigh in their favor because: (1) S.B. 202 was passed alongside increasingly successful, black-led efforts in Georgia to mobilize voters of color, which included the promotion of absentee voting; (2) S.B. 202 was passed after historic wins by candidates preferred by black voters, in a state where voting remains highly racially polarized; and (3) S.B. 202 grew out of a tenuous and racialized voter fraud context.
In considering the sequence of events leading to the passage of S.B. 202, the Court recognizes that black voters used absentee voting at a rate higher than white voters in 2020 (although not by an overwhelming margin) and that candidates preferred by black voters, and the Democratic Party, were successful in 2020. While these two findings could potentially suggest that the Legislature acted with a discriminatory intent, the Court cannot ignore what appears to be the Legislature's legitimate justifications for passing the law. These justifications include, but are not limited to, the State's interest in orderly and efficient election administration, preventing fraud or the appearance of fraud and increasing voter confidence. Indeed, the Legislature was consistent in its messaging that S.B. 202 was “designed to address the lack of elector confidence in the election system on all sides of the political spectrum, to reduce the burden on election officials, and to streamline the process of conducting elections in Georgia by promoting uniformity in voting.” [Doc. 566-59, p. 5].
The Eleventh Circuit has held that partisan discrimination must not be conflated with racial discrimination. League of Women Voters, 66 F.4th at 925.
Plaintiffs argue that the State's proffered justification of preventing fraud or the appearance of fraud is pretextual because many of the allegations of voter fraud were unfounded. This argument is unpersuasive because “[u]nder settled precedent, evidence of existing voter fraud is unnecessary for legislation that aims to prevent future fraud.” League of Women Voters of Fla., Inc. v. Fla. Sec'y of State, No. 22-11143, 2023 WL 6157350, at *3 (11th Cir. Sept. 21, 2023). Indeed, the Eleventh Circuit has recognized that “even if voter fraud had never occurred . . ., the legislators' stated justification of preventing future fraud would be presumptively lawful.” Id.
Plaintiffs also seem to argue that the comments Giuliani made during the December 2020 hearings are circumstantial evidence of the Legislature's intent. Because statements of individual legislators have limited probative value, see League of Women Voters, 66 F.4th at 932, this isolated statement by a non-legislator has even less probative value.
Here, in light of the extensive complaints surrounding the 2020 election, it makes sense that the Legislature would “consider not just the actual issues but also voter perception about the election when considering whether any legislative response was necessary.” [Doc. 601-3, p. 5]. The Legislature's reaction to both actual issues and voter perceptions pertaining to the 2020 election (after Trump, a white Republican, lost) is even less suspect given that the Legislature reacted similarly and amended voting laws after the disputed 2018 election (when Abrams, a black Democrat, lost). And, responding to constituent concerns is, of course, exactly what we expect state legislators to do in a representative democracy. It is not the role of the federal judiciary to second guess their responses and strike down legislation when the concerns that prompted it may not be completely valid. That is, instead, fodder for debate in the next legislative session.
Perhaps the most straightforward example of responding to voter perceptions involves provisional ballots. To process a provisional ballot, election officials must manually duplicate the ballot so that it can be counted. [Doc. 601-13, pp. 21-22]. Although this process is not in any way fraudulent, voters perceived it as such-raising concerns when they witnessed election officials filling out ballots. See id.; [Doc. 601-9, pp. 17-18]. Through the Provisional Ballot Provision, legislators understandably sought to address these misperceptions by aiming to reduce the number of provisional ballots cast (and thus minimize the number of voters confused by the process).
To summarize, given the State's justifications for S.B. 202, the Court finds that this factor weighs in favor of Defendants and against a finding of discriminatory intent.
iii. Procedural and Substantive Departures
The Court next considers the procedural process used by the Legislature in passing S.B. 202 and analyzes whether there were significant procedural or substantive departures from the normal procedure which would suggest intentional discrimination. Plaintiffs contend that the Legislature departed from its normal legislative practice when it passed S.B. 202. As to this factor, Plaintiffs assert that S.B. 202's legislative record was “marked by a lack of transparency and a rushed process that excluded Black participation from legislators and the public alike.” [Doc. 566-1, p. 55]. Concerning the rushed process specifically, Plaintiffs aver that legislators received drafts of bills either just before or at committee hearings and that S.B. 202 morphed from three pages to over ninety pages in a short amount of time. Plaintiffs also argue that there was very little debate during the three committee hearings on the law before it eventually passed both the House and Senate pursuant to a party-line vote. As to the exclusion of black participation, Plaintiffs contend that the Election Committee members of the minority party, who were all black, were excluded from meetings to discuss election bill provisions. Plaintiffs also assert that election officials were largely ignored by the Legislature.
After reviewing Plaintiffs evidence, the Court was able to discern the following departures from the ordinary legislative procedure: (1) the higher number of election bills considered in the 2021 Legislative Session; (2) the formation of the Election Committee; (3) the lack of a fiscal analysis; (4) the timeliness of the receipt of amendments; and (5) the times of the hearings. Plaintiffs, however, failed to provide sufficient evidence which would show that these departures were significant. For instance, Plaintiffs brought evidence showing that the times of the hearings (7:30 A.M. and 5:00 P.M.) were unusual. Plaintiffs did not, however, offer any additional evidence concerning when hearings are normally held. Without this context, the Court is unable to determine whether this specific variation from procedure is significant or not. Defendants, on the other hand, did present evidence as to how the Legislature has worked in the past. In fact, Defendants offered evidence that the process for considering and passing H.B. 316 in 2019 was very similar to the process for considering and passing S.B. 202 in 2021. Defendants also brought evidence that the Election Committee was one of three special committees utilized during the legislative session. The Court is thus not convinced that Plaintiffs have shown significant departures from the normal procedure.
Plaintiffs also argue that this factor weighs in their favor because S.B. 202 was passed after three hearings and a short debate. Again, Plaintiffs never provided any evidence as to the normal number of hearings on a bill or the typical length of a debate or even claim that this procedure was abnormal. Defendants, on the other hand, provided evidence that four hearings for a single bill is a significant amount of consideration by a legislative committee. [Doc. 601-3, p. 15]. Moreover, even though S.B. 202 technically only had three hearings, the Court cannot ignore that in the six-week period leading up to the passage of S.B. 202, the House and Senate considered other election bills, many of which had provisions incorporated into S.B. 202. In total, twenty-five hearings were held between the House and the Senate. Based on this record, this Court cannot find at this stage of the proceedings that the legislative process surrounding S.B. 202 departed from the ordinary procedure in any significant procedural or substantive way. Accordingly, this factor weighs in favor of Defendants and against a discriminatory purpose finding.
iv. Contemporaneous Statements and Actions of Key Legislators
Courts are also required to consider the contemporary statements and actions of key legislators. In arguing that this factor supports a finding of discriminatory intent, Plaintiffs rely on the statements of two legislators: Representative Fleming and Speaker Ralston.
As stated above, Plaintiffs contend that a statement by Representative Fleming shows that the Legislature acted with discriminatory intent. In the statement at issue, Representative Fleming compared absentee voting to the shady part of town down near the docks. Specifically, Representative Fleming stated that “[i]f elections were like coastal cities, absentee balloting would be the shady part of town down near the docks you do not want to wander into because the chance of being shanghaied is significant.” [Doc. 566-114, p. 5].
While this analogy is not one that the Court would use, the Court is not convinced that this statement shows racial animus towards black voters specifically. Moreover, this isolated statement does not inform the Court about the Legislature's motivations as a whole. Indeed, what motivates one legislator to act “is not necessarily what motivates scores of others to enact [legislation].” League of Women Voters, 66 F.4th at 932. Here, many legislators besides Representative Fleming favored S.B. 202 as demonstrated by the fact that S.B. 202 passed by a margin of 100 to 74 in the House and 34 to 20 in the Senate. Thus, Representative Fleming's vote was only one vote of the 134 votes supporting S.B. 202.
Plaintiffs also point to a statement by Speaker Ralston. In an April 2020 interview, Speaker Ralston spoke out against sending these unsolicited absentee ballot applications to every voter. Speaker Ralston opined that sending out unsolicited applications would be extremely devastating to Republicans and conservatives in Georgia and would drive up turnout in 2020. Similar to Representative Fleming's statement, the Court is not persuaded that this statement contains an animus towards black voters. Rather, the statement suggests that Speaker Ralston was motivated by a partisan purpose since he said that increased turnout would be negative to Republicans. “A connection between race and partisan voting patterns is not enough to transform evidence of partisan purpose into evidence of racially discriminatory intent.” Id. at 931.
To summarize, although Speaker Ralston's statement indicates that it was made for a partisan purpose, neither his statement nor Representative Fleming's showed racial animus towards black voters. Accordingly, these two statements are not sufficient to show that the Legislature was motivated by a discriminatory intent. This factor thus weighs against a finding of discriminatory intent.
v. Foreseeability and Knowledge of the Disparate Impact
The Court will next consider the foreseeability and knowledge of the disparate impact. Before addressing Plaintiffs' arguments, it is important to note that Plaintiffs have not shown, at least at this stage of the proceedings, that any of the provisions have a disparate impact on black voters. See supra Part l.A.i. Without this showing, the Court questions whether the Legislature could have foreseen or known about a disparate impact. See League of Women Voters, 66 F.4th at 938. The Court will nevertheless address this factor as if Plaintiffs demonstrated a disparate impact as to each of the challenged provisions.
Plaintiffs contend that the disparate impact was foreseeable because of the testimony of certain citizen witnesses and legislators at the legislative hearings. Beginning with Plaintiffs' first argument, several citizen witnesses testified at the hearings concerning the negative impacts of S.B. 202 on black voters. At one of the hearings, Cindy Battles, who is the Policy and Engagement Director for the Georgia Coalition for the People's Agenda, testified that the provisions would make it harder for voters to vote. [Doc. 566-3, p. l6]. As to black individuals specifically, Battles stated that the prohibition or restriction of Sunday voting seemed “like a direct attack on the Black community.” Id. Aunna Dennis, the Executive Director of Common Cause Georgia, also testified during a legislative hearing on S.B. 202. Dennis opined that voters of color and low-income voters would be disparately impacted by the Provisional Ballot Provision because those voters are more likely to move within their county, thus increasing the likelihood that they will vote in the incorrect precinct. Id. James Woodall, the state president for the Georgia NAACP, testified in a similar manner. He explained during his testimony that an identification requirement would disproportionately affect black voters (especially those in pretrial detention) and that black voters wait in longer lines on average. [Doc. 566-83, pp. 5-9].
S.B. 202 did not eliminate Sunday voting.
As an initial matter, the Court notes that the testimony from these citizens concerning any potential disparate impact on black voters was general in nature, rather than specific. For instance, while the witnesses opined that they thought the law would disparately impact black voters, the testimony was not supported by evidence or statistics. In the Court's view, without corroborating evidence to support a witness's generalized assertion of disparate impact, this is weak evidence of foreseeability. Indeed, the Supreme Court has held that remarks “made in the course of legislative debate or hearings other than by persons responsible for the preparation or the drafting of a bill, are entitled to little weight.” Ernst & Ernst v. Hochfelder, 425 U.S. 185, 203 n.24 (1976) (determining that witness testimony from various industry representatives in legislative hearings was entitled to little weight).
In addition to arguing that the citizen testimony is evidence of foreseeability, Plaintiffs also contend that the disparate impact on black voters was foreseeable because of statements made by other legislators. For example, Georgia State Representative Burnough testified that in her county, the number of drop boxes would decrease from nine to one, and this change would “make voting worse than what it has been and the access will be limited.” [Doc. 566-1, pp. 21-23]. Representative Burnough also argued that many Georgians, disproportionately black, cast out of precinct provisional ballots on Election Day and that “communities of color are disproportionately affected by bills like these.” [Doc. 566-81, pp. 67-68].
As a general rule, “warnings of the potentially vast impact of a bill by ‘legislative opponents[-]who [i]n their zeal to defeat a bill . . . understandably tend to overstate its reach'-should be ‘entitled to little weight.'” League of Women Voters, 66 F.4th at 941 (quoting Ernst & Ernst, 425 U.S. at 203). Consequently, the Eleventh Circuit has held that “the concerns expressed by political opponents during the legislative process are not reliable evidence of legislative intent.” Id. In short, this Court will afford “little weight” to the warnings of political opponents.
In a footnote, Plaintiffs also point to text messages exchanged between Representative Fleming and others (non-legislator proponents of the bill); the text messages discussed the testimony of legislative opponents and some of the witnesses. [Doc. 566-1, p. 60]. According to Plaintiffs, these text messages show that Representative Fleming dismissed the harms to black voters. Id. But, as already stated above, what other legislative opponents say about a bill is unreliable evidence of foreseeability. The Court has also held that the testimony of these citizen witnesses is weak evidence of foreseeability because the testimony was general in nature. Therefore, the Court is not convinced that Representative Fleming's alleged dismissal of the testimony shows foreseeability. Even if it did, as previously discussed, isolated statements like Representative Fleming's text exchange do not inform the Court about the Legislature's motivations as a whole. Certainly, there is no evidence before the Court that other legislators were dismissive of the testimony. Indeed, what motivates one legislator to act “is not necessarily what motivates scores of others to enact [legislation].” League of Women Voters, 66 F.4th at 932.
To summarize, Plaintiffs presented limited evidence in the form of testimony of legislative opponents and others from legislative hearings to support their assertion that the legislators knew or foresaw the disparate impact that S.B. 202 would have on black voters. Especially here, where Plaintiffs seek the drastic remedy of injunctive relief, this Court cannot find that Plaintiffs have presented enough evidence to show that the Legislature foresaw or knew that S.B. 202 would have a disparate impact on minority voters. Thus, for purposes of this motion only, this factor weighs against a finding of discriminatory intent.
vi. Availability of Less Discriminatory Alternatives
The final factor that courts consider is the availability of less discriminatory alternatives. Plaintiffs argue that less discriminatory alternatives exist as to some of the challenged provisions. As to the Application Deadline Provision specifically, Plaintiffs argue that a seven or eight-day deadline instead of the eleven-day deadline would have been a less discriminatory alternative. As to the Identification Provision, Plaintiffs assert that a less discriminatory alternative would have been to require a social security number and birthdate rather than an identification number. Finally, as to the Drop Box Provision, Plaintiffs suggest that the Legislature could have continued the Emergency Rule from 2020.
As an initial matter, the Court acknowledges that the Legislature did not include the alternative options that Plaintiffs would have preferred. Importantly, the Eleventh Circuit has held that the failure to “‘include the alternative option[s] that Plaintiffs would have preferred' is not evidence of discriminatory intent.” League of Women Voters, 66 F.4th at 942 (quoting Greater Birmingham Ministries, 992 F.3d at 1327).
The Court is not convinced that Plaintiffs have shown that the Legislature failed to consider alternatives that would lessen any potentially discriminatory impact. Indeed, the evidence before the Court suggests, at least in regard to other portions of S.B. 202, that alternatives were considered and even implemented. For instance, in one of the precursor bills to S.B. 202, no-excuse absentee voting for people under the age of sixty-five was eliminated. [Doc. 601-20, p. 10]. Weekend voting was also eliminated. Id. at 7. Notably, S.B. 202 maintains both no-excuse absentee voting and weekend voting.
S.B. 202 also includes provisions and ideas that were typically supported by Democrats, including provisions that required more staff, equipment and polling places in large precincts with long lines. [Doc. 566-3, p. 16]. Moreover, in response to concerns about pretrial detainees lacking a DDS ID, S.B. 202 was amended to ensure that pretrial detainees in jails could access their driver's licenses for purposes of voting. [Doc. 601-20, p. 8]. These changes to S.B. 202 “loosened the restrictions that it would have otherwise imposed” and demonstrates that S.B. 202's proponents were receptive to input during the legislative process. League of Women Voters, 66 F.4th at 941. Simply put, the Legislature was not required to adopt all suggested changes, and the record supports a finding that the Legislature considered alternative measures. See id. (“Plainly, the Legislature adopted some alternatives that were more palatable to the bill's opponents. It did not accept all of them, nor was it required to do so.”).
In sum, the record at this stage of the proceedings does not convincingly support Plaintiffs' claim that the Legislature intended to discriminate by failing to include alternative options that were less discriminatory. Thus, for purposes of this motion only, this factor weighs against a finding of discriminatory intent.
***
The Court has carefully considered the factors above in considering whether Plaintiffs have shown that the Legislature intended to discriminate against black voters when it passed S.B. 202. After weighing the factors, the Court cannot find that Plaintiffs are substantially likely to succeed on their claims pursuant to the Fourteen and Fifteenth Amendments. To the extent that Plaintiffs seek preliminary injunctive relief on these claims, the motion is DENIED.
B. Violation of Section 2 of the VRA
Plaintiffs also allege that the provisions violate Section 2 of the VRA. Congress enacted the VRA in an effort to end the “denial of the right to vote based on race.” Brnovich v. Democratic Nat'l Comm., 141 S.Ct. 2321, 2330 (2021). Section 2 of the VRA states that “[n]o voting qualification or prerequisite to voting, or standard, practice, or procedure shall be imposed or applied by any State or political subdivision to deny or abridge the right of any citizen of the United States to vote because he is a member of a language minority group.” 52 U.S.C. § 10301(b)(2).
Plaintiffs argue that they are entitled to preliminary injunctive relief on their Section 2 claim because they have shown discriminatory intent that is sufficient to constitute a violation of the Fourteenth Amendment. Specifically, Plaintiffs aver that “[a] showing of discriminatory intent that is ‘sufficient to constitute a violation of the [F]ourteenth [A]mendment' is also ‘sufficient to constitute a violation of [S]ection 2.'” [Doc. 566-1, p. 36]. In short, Plaintiffs assert that the Arlington Heights analysis applies to their Section 2 claim.
In the previous section, this Court applied the Arlington Heights factors and found that Plaintiffs were not substantially likely to succeed on their constitutional claims. Because Plaintiffs are not substantially likely to succeed on their constitutional claims, they are also not substantially likely to succeed on their Section 2 claim. For the reasons explained in the preceding section, Plaintiffs failed to meet their burden to show that the law was passed with a discriminatory intent. Thus, to the extent that Plaintiffs seek preliminary injunctive relief on their Section 2 claim, the motion is DENIED.
Defendants argue that a Section 2 discriminatory intent claim is not a cognizable cause of action. [Doc. 601, pp. 30-31]. In Defendants' view, the only valid cause of action under Section 2 is a results-based claim. Because Plaintiffs failed to meet their burden to show discriminatory intent, the Court need not resolve this issue to decide the current motion. An order on Defendants' Motion for Judgment on the Pleadings [Doc. 549] concerning this issue is forthcoming.
2. Other Preliminary Injunction Prerequisites
As explained in detail above, Plaintiffs failed to show a substantial likelihood of success on the merits as to their claims that the provisions intentionally discriminate against black voters in violation of the Fourteenth Amendment, Fifteenth Amendment and Section 2 of the VRA. Therefore, this Court will not address whether Plaintiffs will be irreparably harmed should an injunction not issue or whether the balance of equities favors injunctive relief. See Church v. City of Huntsville, 30 F.3d 1332, 1342 (11th Cir. 1994) (recognizing that when plaintiffs fail to establish a substantial likelihood of success on the merits, the other prerequisites need not be addressed).
In a case like this, where the government is the party opposing the preliminary injunction, the final two preliminary injunction factors (the balance of the equities and the public interest) merge. Swain v. Junior, 958 F.3d 1081, 1091 (11th Cir. 2020). In the context of an election, the balance of the equities and the public interest factors are considered in tandem because “the real question posed . . . is how injunctive relief . . . would impact the public interest in an orderly and fair election, with the fullest voter participation possible and an accurate count of the ballots cast.” Curling v. Kemp, 334 F.Supp.3d 1303, 1326 (N.D.Ga. 2018).
CONCLUSION
For the reasons explained above, Plaintiffs' Joint Motion for Preliminary Injunction [Doc. 566] is DENIED.
SO ORDERED.