Case 1:17-cv-01427-TCB-WSD-BBM Document 137 Filed 03/26/18 Page 1 of 50

IN THE UNITED STATES DISTRICT COURT
FOR THE NORTHERN DISTRICT OF GEORGIA
ATLANTA DIVISION
NAACP, et al.,

*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*

Plaintiffs,
v.
BRIAN KEMP, in his official capacity
as Secretary of State for the State of
Georgia,
Defendant.
AUSTIN THOMPSON, et al.,
Plaintiffs,
v.
BRIAN KEMP, in his official capacity
as Secretary of State of the State of
Georgia,
Defendant.

Case No. 1:17-cv-01427TCB-WSD-BBM
CONSOLIDATED CASES

DEFENDANT BRIAN KEMP’S BRIEF IN OPPOSITION
TO PLAINTIFFS’ MOTION FOR A PRELIMINARY INUNCTION
COMES NOW BRIAN KEMP, Georgia Secretary of State (“Kemp”), by and
through his attorney of record, the Attorney General of the State of Georgia, and files
this Brief in Opposition to the NAACP Plaintiffs’ Motion for a Preliminary

1

 Case 1:17-cv-01427-TCB-WSD-BBM Document 137 Filed 03/26/18 Page 2 of 50

Injunction and Permanent Injunction, in which the Thompson Plaintiffs have joined.
ECF 103, 120.
INTRODUCTION
Plaintiffs challenge House Districts (“HDs”) 105 and 111 as racial
gerrymanders under the Fourteenth Amendment. ECF 1 at 24; ECF 84 at 39.
Plaintiffs premise their racial gerrymandering claims entirely on the correlation
between the race of voters and their partisanship. Despite direct evidence from the
State’s demographer that she used only partisan data to make the 2015 changes to HD
105 and HD 111, Plaintiffs argue that because state officials were aware of the
correlation, and because the partisan changes have a racial effect, the plan is
predominantly based on race. Under Plaintiffs’ rationale, the existence of a
correlation between race and party means that any time a district’s boundaries are
altered for a partisan reason, using partisan data, there will necessarily be a negative
racial impact and therefore racial intent. For the reasons that follow, Plaintiffs’ theory
is inconsistent with Supreme Court precedent and the evidence in this case.
Further, Kemp notes at the outset that any urgency is the product of Plaintiffs’
choices. H.B. 566 became effective in 2015 and was used in the 2016 elections, but
Plaintiffs did not file suit until April 24, 2017 and October 3, 2017. ECF 1 and ECF
84. Their Motion for Preliminary Injunction was filed shortly before candidates
2

 Case 1:17-cv-01427-TCB-WSD-BBM Document 137 Filed 03/26/18 Page 3 of 50

began qualifying to run for State House districts, including HDs 105 and 111, and
other offices. Candidate qualifying has now closed, and, as set out in greater detail
below, the ballot creation process is underway. In short, the entry of a preliminary
injunction as sought by Plaintiffs threatens to disrupt an election process that is
already underway and will confuse voters.
I.
A.

STATEMENT OF FACTS

H.B. 566 was an Effort to Consolidate Minor District Changes that
Representatives Started Requesting in 2014.
Representative Randy Nix, chair of the House Reapportionment Committee

from 2013 to 2016, testified that, during the 2014 legislative session, approximately
eight to ten members came forward requesting minor changes to their districts. Nix
Dep. 62:6-9, 66:9. 1 These changes included moving district lines to accommodate
incumbents who had moved or bought new property. Id. at 68:19-24. In response to
these requests, House Republican leadership then decided that “[i]f we do anything,
we’ll do it all at one time rather than trying to do things piecemeal.” Id. at 66:24 to
65:7. Mr. Nix insisted, however, that “any changes that were made had to be relatively
minor so that we didn’t jeopardize the good work that we had done when the maps
were drawn before [in 2011 and 2012].” Id. at 73:17-20.

1

The Nix Deposition appears at ECF 124.
3

 Case 1:17-cv-01427-TCB-WSD-BBM Document 137 Filed 03/26/18 Page 4 of 50

At the beginning of the 2015 legislative session, Rep. Nix announced that
redistricting would be considered “before the entire chamber . . . during the session.”
Nix Dep. 79:8-11. He made the same announcement before both the majority and
minority caucuses in the House and talked to the minority leader, Representative Stacey
Abrams, as well. Id. at 79:11-18. People on both sides of the aisle wanted to make
changes. Id. at 79:25 to 80:1.
B.

Chairman Nix Established Clear Guidelines for Changes that would be
Made in 2015, and Gina Wright was Tasked with Keeping all Changes
Within These Guidelines.
Rep. Nix outlined clear parameters for any redistricting that would take place in

2015. First, “it had to be done fairly, we wanted to offer it to everyone, not that
everyone could do it because of some of the changes some people wanted wouldn’t
work.” Nix Dep. 84:7-10. Second, Rep. Nix insisted that any changes to district lines
could not jeopardize the integrity of the map drawn in 2011 and modified in 2012. See
id. at 73:14-20, 74:14-16, 108:23-25 to 109:1. Rep. Nix was proud of the fact the
Obama Administration did not challenge the 2012 map and was insistent that any
changes in 2015 would not undermine the integrity of the map that led to federal
approval. Id. at 57:3-11, 94:24-25 to 95:1-7, 75:11-13. Third, Nix required that all
members who would be affected by a proposed change must consent to the change. Id.
at 72:19-25 to 73:1-7. Finally, Nix required all members who were requesting changes
4

 Case 1:17-cv-01427-TCB-WSD-BBM Document 137 Filed 03/26/18 Page 5 of 50

to do so through the Legislative and Congressional Reapportionment Office (the
“LCRO”). Id. at 79:13-18, 83:3-7. Rep. Nix tasked the LCRO with ensuring that any
changes fell into the broad requirements that he had established for the 2015
redistricting process––that all changes must (1) be fair and bipartisan, (2) maintain the
integrity of the 2012 map, and (3) be made with the consent of all affected members.
See id. at 80:15-20, 82:3-9, 141:23-25 to 142:1-3, 136:15-19 (describing how the
LCRO must inform Nix whether a proposed changes fits within the parameters he
established).
C.

The Process that the LCRO Used to Draw the Changes that Became H.B.
566.
Gina Wright is the Executive Director of the LCRO. She was the primary map

drawer for H.B. 566. Wright Dep., 15:25-16:20.2 Ms. Wright is the person in the LCRO
who is typically approached about a modification to a district. Deposition of Strangia
Dep., 30:24-31:1.3 Ms. Wright oversaw the redistricting of all seventeen districts
included in H.B. 566, although certain of her staff did provide technical assistance on
two areas that are not at issue in this litigation. Declaration of Gina Wright at ¶ 5,
attached hereto as Exhibit A. Ms. Wright was the only person in the LCRO who worked

2
3

The Deposition of Gina Wright appears at ECF 112.
The Deposition of Rob Strangia appears at ECF 110.
5

 Case 1:17-cv-01427-TCB-WSD-BBM Document 137 Filed 03/26/18 Page 6 of 50

on that portion of the H.B. 566 that touched any part of Gwinnett and Henry County,
including HD 105 and HD 111. Id.
During the 2015 redistricting process, Ms. Wright had access to political data
down to the census block level. The LCRO “allocate[s]” the political data that comes
in from the Secretary of State’s Office “to the block level . . . based on the percentage
and proportions of the population.” Wright Dep., 106:7-13. If “for example, you have
a block where, or several blocks, in a precinct where you have a lot of population right
here in these blocks and then nobody, you have zeroes out here, it’s going to obviously
allocate that proportion there of the voter data, not here because nobody lives here. So
it spreads out through the blocks.” Wright Dep., 106:17-23. The political data is
election results that come from statewide contested races. Exhibit A at ¶ 7.
Rob Strangia, a geographical information specialist in the LCRO, likewise
testified that political data the office uses goes all the way “down to the block level of
geography based on voting age population.” Strangia Dep., 25:7-9. The allocation is
done by Geographic Information System (GIS) software every two years. Exhibit A at
¶ 7. Strangia explained:
If you have a hundred registered voters in precincts ... and
there’s two blocks in the precinct, one block has 60 percent
of the voting age population, the other block has 40 percent
of the voting age population, it’s going to allocate the election
data based on those percentages to the block level geography.
It’s only an estimate.
6

 Case 1:17-cv-01427-TCB-WSD-BBM Document 137 Filed 03/26/18 Page 7 of 50

Strangia Dep., 25:20 to 26:3. For the same precinct in which the voting-age population
is split 60-40, and the vote is split on a 50-50 basis, Strangia testified that the software
would “put 60 percent of the votes for both candidates in the block with the larger
population and 40 percent in the one with the lower population.” Id. at 26:10-17.
The software presents a “typical and customary way of translating political data
down to the block level.” Id. at 26:19-22. It has been used not only by Georgia, but also
by Texas, Florida, and California. Id. at 26:23 to 27:23. The allocation is “a pretty
decent estimate, but it’s an estimate.” 4 Wright Dep., 111:23-24.
In his Expert Report and his Reply Report, Plaintiffs’ expert, Dr. Jowei Chen
said that he “found that the Legislature’s primary map-drawer had access only to racial
data, but not partisan data, at the sub-precinct level.” ECF 63-1 at 3; ECF 63-1 at 33;
ECF 94-1 at 16. Chen testified that he based that “finding” on his reading of the
depositions of Gina Wright and Rob Strangia. Chen Dep., 15:19 to 16:12.
Chen’s “finding” fails to account for the process that the LCRO actually uses
and the data that Ms. Wright actually had access to when drawing the maps for H.B.
566. This information was provided in both Ms. Wright’s and Mr. Strangia’s

4

While Plaintiffs’ expert, Dr. Peyton McCrary, believes such allocation to be less
predictable political data than racial population data, ECF 65-1 ¶ 47, it is nonetheless
what Wright used. Exhibit A at ¶¶ 7-10.
7

 Case 1:17-cv-01427-TCB-WSD-BBM Document 137 Filed 03/26/18 Page 8 of 50

depositions, both of whom testified that political data at the block level was available.
More to the point, Wright testified that she relied on political data, and not racial data,
in making the changes to HDs 105 and 111. Wright Dep., 30:1-31:10, 219:9-220:21;
Exhibit A at ¶¶ 7-10.
D.

After Mr. Nix Announced that Representatives Could Request Small
District Changes, Representatives Chandler and Strickland Decided to
Investigate Whether Small Changes Could Increase their Political
Chances.
HDs 105 and 111 were created in the redistricting that followed the 2010 Census.

Chandler Dep., 60:20-22; 5 Strickland Dep., 45:21 to 46:2. 6 To make changes to the
Districts in 2015, the Reapportionment Office focused on political data from statewide
races to draw to 2015 changes to HD 105 and HD 111. As stated above, Ms. Wright
performed the entire map drawing process for the 2015 changes to HD 105 and HD
111. Exhibit A ¶ 5. In redistricting HD 105 and HD 111, she understood the goal to be
improving the political performance of the two districts for the Republican incumbents.
Id. at ¶ 6. To draw the districts, she used a 2014 voting precinct data layer, which
included 2014 general election results, to measure the political effectiveness of the
districts. Id. at ¶ 8. The precinct layer included a total of support to the Republican
(“%TRepVots14”) and Democratic (“%TDemVots14”) candidates that represents the

5
6

The Deposition of Joyce Chandler appears at ECF 126.
The Deposition of Brian Strickland appears at ECF 127.
8

 Case 1:17-cv-01427-TCB-WSD-BBM Document 137 Filed 03/26/18 Page 9 of 50

average performance among all contested statewide general elections for 2014. Id. This
political data was available at the voting precinct level and was allocated to the precinct
boundaries on the map layer. Id. at ¶ 9. When a precinct is split, the redistricting
software allocates the votes from that election to the individual blocks. Id. When Ms.
Wright worked on the 2015 redistricting of HD 105 and HD 111, including all
surrounding affected districts, she did not keep data regarding the racial make-up of
the census blocks and precincts visible on the pending changes portion of the Maptitude
screen. Id. at ¶ 10. She was not working with that data because her goal was to improve
the Republican political performance of the districts, so she worked with the relevant
data for that goal, i.e., the total population and the political data (the %TRepVots14
and %TDemVots14 data). Id. at ¶ 10.
1.

The Election History of HD 105 Prior to 2015.

In 2012, Joyce Chandler defeated Renita Hamilton by a margin of 51.35% to
48.65%, a difference of 554 votes. Chandler Dep., Pls.’ Ex. 124. She outperformed
Republican Presidential candidate Mitt Romney, who lost in HD 105. Chandler Dep.,
Pls.’ Ex. 130 at 1.
In 2014, Rep. Chandler again defeated Renita Hamilton, this time by 789 votes.
Chandler Dep., Pls.’ Ex. 36 at 7. She won the Baycreek K, Baycreek C, Baycreek D,

9

 Case 1:17-cv-01427-TCB-WSD-BBM Document 137 Filed 03/26/18 Page 10 of 50

Baycreek F, Baycreek H, and Baycreek I precincts, while losing the Lawrenceville D,
Lawrenceville F, and Lawrenceville M precincts. Id.
Rep. Chandler does not look at racial information as part of her campaign
strategy. Chandler Dep. 60:2-4. She canvases all parts of her district. Chandler Dep.
66:20-22. In campaigning, she believes she just has “to work hard, period.” Chandler
Dep., 75:23-76:1.
2.

Plaintiffs Suggest that Chandler was Interested in Racial Data by Use
of Misleading Exhibit.

Rep. Chandler believed the demographics of HD 105 changed between 2012
and 2014 by becoming more Democratic than Republican, in that “the percentage of
Republican voters were less . . . than the percentage of Democrat voters.” Chandler
Dep. 126:16-127:3. After the 2014 election she wanted to redraw the lines of her
district because she “would like for [her] district to have been more Republican than
it was in the previous year.” Id. at 131:15-23. She hoped to improve her chances of
reelection in 2016 by tweaking her district. Id. at 178:5-7. Moving AfricanAmericans out of her district was never mentioned to her. Id. at 178:16-19. She did
not want to change the racial demographics of her district, remove African-American
voters from her district, dilute the voting strength of minorities in her district, nor add
white voters to her district. Id. at 187:17-188:3.

10

 Case 1:17-cv-01427-TCB-WSD-BBM Document 137 Filed 03/26/18 Page 11 of 50

Rep. Chandler was also not interested in the percentage of African-Americans
that would live in HD 104 and 105 after making a proposed change. Id. at 207:22208:1. She does not recall any discussions about increasing or decreasing the number
of black people in a district or getting below a certain percentage of black people as a
reason to change a district. Id. at 236:5-16. In fact, she does not remember any racial
considerations ever being discussed. Id. at 238:18-21.
Plaintiffs insinuate that Representative Chandler was interested in the racial
make-up of her district because of an email chain between Chandler and Dan
O’Connor where O’Connor appears to have sent Chandler both election data and
voter registration data by race. ECF 103-34. At her deposition, Representative
Chandler testified that she does not recall asking for any racial data for her district,
just a map, so she can campaign. Plaintiffs’ counsel then showed her the two pages
of voter registration data that are part of ECF 103-34, and represented that these
pages are part of the email from Dan O’Connor. Chandler Dep. 97-98. The data
attached to this exhibit was not an attachment to the Dan O’Connor email. See
Declaration of Howe Taing, attached hereto as Exhibit B. As is clear by looking at
the Bates stamp on the bottom right hand corner of documents, 7 the email is GA2-

7

Exhibit 128 to the Deposition is Exhibit 33 to Plaintiffs’ Motion for Preliminary
Injunction. Doc. 103-34.
11

 Case 1:17-cv-01427-TCB-WSD-BBM Document 137 Filed 03/26/18 Page 12 of 50

001195, the attachments are GA2-001197 and GA2-001198, which are attachments to
the email in GA2-001196. See Exhibit B at ¶¶ 5-6. Despite Representative
Chandler’s testimony that she did not recall requesting such data, the following
exchange occurs:
Q: Does Mr. O’Connor send you things that you don’t ask for?
A: I don’t know. I doubt it, but I don’t know.
Q: Why did you thank him for sending the voter registration data?
A: I don’t know. I would just courteous - - I would just usually respond thank
you to anybody.
Q: Let’s look at the spreadsheet on the second page.
Chandler Dep., 99:9-19. Counsel than repeatedly asks Representative Chandler if she
could find any election data or partisan data on the two pages. Id. at 99:21 – 101:2.
When Representative Chandler answered that she could only find “total voters,” she
is asked:
Q: So is that a no?
A: That’s a no, unless I’m just missing it.
Id. at 101:4-6. Chandler is then asked if it’s a “fair summary” that the two pages
(GA2-001197 and GA2-001198) provide voter registration by race but not by party.

12

 Case 1:17-cv-01427-TCB-WSD-BBM Document 137 Filed 03/26/18 Page 13 of 50

Id. at 101:7-22. Chandler is then, again, asked to explain why O’Connor would send
her this information, and not the election returns, if she had not asked for them.
Q: Do you have any reason to believe you did not request this information
from Mr. O’Connor?
A: I do not recall requesting anything like this.
Q: But that’s not my question. My question is: Do you have any reason to
believe that you did not request this information from Mr. O’Connor?
Id. at 101:23 – 102:6. The email from Dan O’Connor, 103-34 p. 1, clearly indicates
on its face that two excel files had been included. The email was a reply email from
Chandler, so the attachments were not included. From the face of this email, it
appears that O’Connor sent Representative Chandler both some registration data by
precinct and general election (GE) returns for HD 105. Plaintiffs however, insinuate
that only registration data, by race, was provided to Chandler, and because she
thanked O’Connor for the data, registration data must be all she wanted. ECF 103-1
at 16 (“O’Connor gave [Chandler] precinct-level racial data and Chandler thanked
him.”) (citing Ex. 33).
3.

The New District 105 Map

Once Chairman Nix opened up the redistricting process in 2015, HD 105 was
examined to determine whether any changes could make it a little more Republican.
13

 Case 1:17-cv-01427-TCB-WSD-BBM Document 137 Filed 03/26/18 Page 14 of 50

Chandler Dep. 138:5-9. To improve the Republican political performance of HD 105,
Representative Efstration’s HD 104 was a target district from which Ms. Wright could
move GOP voters because his district is “very Republican,” Id. at 112:13-113:3. Using
the political data layer in Maptitude, Ms. Wright began the redistricting of HD 105 with
the district at 50.98% (%TRepVots14) and 48.13% (%TDemVots14). HD 104, which
is adjacent to HD 105, was 69.68% (%TRepVotes14) and 28.87% (%TDemVots14).
Exhibit A ¶ 12. A detailed explanation of Wright’s process for redrawing these districts
is included below in Section II.A.1.a. See also Exhibit A ¶¶ 11-25.
4.

The Election History of HD 111 Prior to 2015.

In 2012, Brian Strickland defeated Bill Blackmon by a margin of 53% to 47%,
a difference of 1,477 votes. Strickland Dep., Pls.’ Ex. 288 at 2. He ran ahead of
Republican Presidential candidate Mitt Romney, who garnered only 50% of the vote
in HD 111. Strickland Dep. Pls.’ Ex. 290 at 1. Strickland carried the Lowes, North
Hampton, Stagecoach, Unity Grove, Pate’s Creek, Oakland, Dutchtown, and
McDonough Central precincts, losing in Mount Carmel, Wesley Lakes, and
Stockbridge West. Strickland Dep. Pls.’ Ex. 288 at 3.
Strickland explained that he did not look at “past voting information” in 2012
because, with a newly created district, “there was no past data to look at.” Strickland
Dep. 45:21 to 46:2. He did not look at racial information either. Id. at 46:8-11. Instead,
14

 Case 1:17-cv-01427-TCB-WSD-BBM Document 137 Filed 03/26/18 Page 15 of 50

Strickland “ma[de] sure to focus on everyone, and . . . campaigned in all communities.”
Id. at 46:21-23; see also id. at 48:4-5 (“[W]e worked the entire district.”). In his view,
he “outworked” Blackman, noting, “I knocked on doors. I was an unknown candidate
at that time. I had no record to run on. I was a young guy, too.” Id. at 119:16-20; see
also id. at 62:13-14 (“[I]f I work hard, then I can win any election.”).
Strickland explained his strategy: “Someone that has met me is more likely to
vote for me. In my experience, if I’m able to reach a voter, then I have a good chance
that they’ll vote for me.” Id. at 49:14-17. He testified, “In my experience, when I knock
on a door, whether they’re white, black, purple, green, orange, I get a very similar
response.” Id. at 50:9-11. Strickland does not know how individual people vote, but he
does believe that whether someone has “met [him] or been reached by [his] campaign”
was the key to electoral success. Id. at 50:23-24. He stated, “In my experience, no
matter who the person is, if I get a chance to reach them, I’m able to get votes whether
white or black.” Id. at 56:23-25.
Representative Andrew Welch observed that Strickland “has a very high
likability” in the polls. Welch Dep., 197:16-17. 8 People who Welch “know[s] are
typically Democratic leaning, they vote presidential Democratic, but they like him.”
Id. at 197:18-23. As a result, Strickland “gets crossover votes.” Id. at 197:3.
8

The Deposition of Andrew Welch appears at ECF 129.
15

 Case 1:17-cv-01427-TCB-WSD-BBM Document 137 Filed 03/26/18 Page 16 of 50

Strickland characterized the partisanship of HD 111 as “evenly split.” Id. at 58:6;
see also Jones Dep., 175:8-16 (“[p]retty even[ly]” split; “probably less Trump type”
Republicans); Welch Dep., 97:23 (HD 111 has “always been tight.”). Strickland
explained that, by looking at precinct-level results in benchmark races, one could see
“Republicans, Democrats switching around. I’d win places where a Democrat would
win state wide or the Democrat would win a presidential election too, so it was a good
mix.” Strickland Dep., at 58:16-20. He did not “consider racial information in
determining whether [he] thought [he] would win or lose.” Id. 46:8-11.
In 2014, Strickland defeated Jim Nichols by a margin of 53.1% to 46.9%, a
difference of some 1,124 votes. Strickland Dep. Ex 290 at 5. Again, Strickland carried
the Lowes, North Hampton, Stagecoach Unity Grove, Pate’s Creek, Oakland,
Dutchtown, and McDonough Central precincts, while losing Mount Carmel, Wesley
Lakes, and Stockbridge West. Id. And he ran ahead of Governor Deal, who garnered
only 48% of the vote in HD 111. Id. at 1.
5.

The New District 111 Map

Rep. Strickland heard that there would be a bill in 2015 dealing with changes to
the House districts. Strickland Dep., 123:23 to 124:10; id. at 148:25 to 149:8. He
believed that it was “traditional to have another look at revisiting or revising maps in
small ways in a midterm, if you will, in a ’15 year.” Id. at 124:11-14; see also Nix Dep.,
16

 Case 1:17-cv-01427-TCB-WSD-BBM Document 137 Filed 03/26/18 Page 17 of 50

70:12-13 (“[I]t was general knowledge that it had happened prior.”). Strickland
understood that the changes could not be “drastic or . . . really change the intent from
2010.” Id. at 150:4-6.
Rep. Welch testified that the 2015 changes to HD 111 came about because
Representative Dale Rutledge “was telling [Welch] he wanted to move somewhere
around Lawrenceville Street in McDonough.” Welch Dep., 77:18-20; see also
Strickland Dep., 128:19-22 (“I remember in particular Representative Rutledge
wanting to have a particular part of McDonough for personal reasons.”); Rutledge
Dep., 94:9-13. That move would put Rutledge into Welch’s district, “so we had to
redraw those lines.” Welch Dep., 77: 22-23. It would affect portions of the
McDonough Central and McDonough precincts, where Welch’s parents live. Id. at
80:24 to 81:2, 81:25. Welch told them that he “wasn’t happy because they were
talking about cutting my parents out. But I said, if that’s what you’re planning on
doing, we’ll work around it.” Id. at 84:10-15.
The possible changes were discussed at a meeting in early 2015 that may have
been attended by Strickland, Rutledge, Welch, Representative Yates, and Gina
Wright. Welch Dep., 77:6-9; Strickland Dep., 126:7-11. No data regarding the racial
or partisan effect of the change to accommodate Rutledge was presented at the
meeting. Welch Dep., 83:17 to 84:4. Strickland recalled a meeting at which the
17

 Case 1:17-cv-01427-TCB-WSD-BBM Document 137 Filed 03/26/18 Page 18 of 50

attendees looked at “benchmark political race[s].” Strickland Dep., at 155:17-24,
159:9-25.
Strickland “wasn’t interested in changing the lines at all” (id. at 162:22-23), but
when he “was told that changes were going to happen and that we were making
changes to maps,” he chose to participate. Id. at 163:4-6.
Welch knew that Strickland “was concerned about having to keep running in
such a tight district.” Welch Dep. 98:12-14. At the meeting, Welch told Strickland
“that they needed to do what they needed to do to adjust for Dale [Rutledge], and if
that swung any . . . of my district over to Brian’s, they could do that.” Id. at 98:19-23.
He told Gina Wright to “make it work within the parameters that they have. But if
they have to move my stuff around, then just don’t hurt Brian.” Id. at 101:6-9. None
of them wanted to make HD 111 “less safe.” Id. at 101:5.
In contrast to Strickland, Rutledge attributed the changes to Strickland’s desire
to “increase the Republican base in his district.” Rutledge Dep., 71:25 to 72:1. The
“plan” was to move Democrats from Strickland’s district into Rutledge’s and
Republicans into Strickland’s district. Id. at 103:9-21. There was no discussion of
moving people because of their race. Id. at 104:14-25. The changes to HD 111
“targeted Democrats and Republicans,” not African-Americans. Id. at 143:11 to
144:3.
18

 Case 1:17-cv-01427-TCB-WSD-BBM Document 137 Filed 03/26/18 Page 19 of 50

Rutledge knew that any change to Strickland’s district would mean changes to
his own. Id. at 74:19-20. And he knew that “if the districts are going to change, the
parties involved have to agree to the change.” Id. at 84:9-10.
Based on these concerns, Ms. Wright used the same process for redrawing HD
111 as she had for HD 105 “because the objective was to try and politically help the
incumbent for purposes of reelection.” Wright Dep., 224:2-5. She stated that the
legislators in the area of HD 111 did not suggest any particular changes to her, but left
it to her to suggest specific changes. Id. at 178:3-12.
Ms. Wright’s objective in making changes to the districts surrounding HD
111was to increase the political performance of Strickland in HD 111 while not
significantly lowering the political performance numbers for the surrounding
Republican incumbents. Exhibit A ¶ 27. Additionally, she was aware of the fact that
Representative Rutledge, the incumbent in HD 109, was considering purchasing a
house in an area that, under the 2012 plan, was in HD 111. Id.
Using the political data layer in Maptitude, she began the redistricting of HD 111
with the district at 50.14% (%TRepVots14) and 48.65% (%TDemVots14). Id. at ¶ 28.
HD 109 was 60.43% (%TRepVots14) and 38.44% (%TDemVots14). Id. HD 110 was
61.98% (%TRepvots14) and 36.97% (%TDemVots14). Id. A detailed explanation of

19

 Case 1:17-cv-01427-TCB-WSD-BBM Document 137 Filed 03/26/18 Page 20 of 50

Wright’s process for redrawing these districts is included below Section II.A.1.a. See
also Exhibit A ¶¶ 26-40.
E.

The Passage of HB 566
The Reapportionment Office’s work for the seventeen representatives who had

requested changes was bipartisan; “both parties, Republicans, Democrats, got together
and did tweaks to a map.” Strickland Dep., 124:17-19. The key to HB 566 was that
“both parties consented to it. It was completely bipartisan, and both parties got to make
tweaks.” Id. at 135:18-21. Strickland testified that “[i]t was communicated to me that
[Stacey Abrams, minority leader in the House] was in consent with the changes that
were being proposed by the Republicans, which is part of the reason it was bipartisan.”
Id. at 177:24 to 178:3, 178:11-13 (“But it was my understanding that she was also
involved and consented to the changes to 111.”).
Representative Jones testified that the changes made by H.B. 566 “certainly were
not controversial.” Jones Dep., 112:11. 9 She explained, “[W]e put it through the sieve
of our judgment of 180 people, the entire body. There were no dissents and ... if anyone
had objected at the time, it would have merited further scrutiny, and there was no
objection.” Id. at 112:17-22.
Representative Chuck Efstration concurred. He testified:
The purpose of House Bill 566, as I understood it to be, was to ...
9

The Deposition of Jan Jones appears at ECF 128.
20

 Case 1:17-cv-01427-TCB-WSD-BBM Document 137 Filed 03/26/18 Page 21 of 50

make district changes that were by the consent of or agreement of
all members affected, Republican and Democrat, as I recall.
And that was the reason that it passed unanimously in committee
and unanimously on the House floor.
Efstration Dep., 181:5-17. Efstration explained that he sits in the same row in the House
Chamber as the minority leader, Stacey Abrams, and that “there had been no partisan
objection whatsoever to the legislation at the time.” Id. at 184:14-18.
Rep. Efstration, a member of the House Reapportionment Committee, does not
recall any discussion of racial makeup at the committee meeting at which H.B. 566 was
considered. Id. at 124:16-125:11.
The vote in the House on H.B. 566 was 168-0.
Plaintiffs suggest that the summary of HB 566 which Dan O’Connor drafted is
misleading and “conveniently omitted” the true purpose of the legislation. ECF 1031 at 28-29. The omission is with Plaintiffs’ summary of the document. The
O’Connor summary, recites at the bottom of the second page:
District line changes can be made for a variety of reasons - - as some
examples, eliminating a split precinct (a precinct divided into multiple
districts), reuniting a neighborhood or community of interest, or
addressing technical concerns.
ECF 103-64 at 2 (emphasis added). By eliminating the beginning language, which
makes clear that the examples are not exclusive, or even necessarily related to the
purposes of the changes made in HB 566, Plaintiffs suggest that O’Connor’s memo is
21

 Case 1:17-cv-01427-TCB-WSD-BBM Document 137 Filed 03/26/18 Page 22 of 50

intentionally misleading. O’Connor testified he did not know anything about the
specific changes to HD 105 and HD 111 because he was not involved. O’Connor
Dep., 140:4 – 5 (“I wasn’t involved in the, you know, the map drawing of 111.”);
135:15 -16 (“And, again, I wasn’t involved in 111.”); 135:19 -20 (“I wasn’t involved
in [HD 105]”). Having not been involved in the redistricting, it is not surprising he
could not give specifics about the changes.
II.
A.

ARGUMENT AND CITATION OF AUTHORITY

Plaintiffs Have Not Met Their Burden of Showing That They are
Entitled to a Preliminary Injunction.
A preliminary injunction in advance of trial is an extraordinary measure.

United States v. Jefferson County, 720 F.2d 1511, 1519 (11th Cir. 1983); Univ. of
Texas v. Camenisch, 451 U.S. 390, 395 (1981). In order to prevail on a motion for
preliminary injunction, the movant must show: 1) a substantial likelihood of
prevailing on the merits; 2) that the plaintiff will suffer irreparable injury unless the
injunction issues; 3) that the threatened injury to the movant outweighs whatever
damages the proposed injunction may cause the opposing party; and 4) that if issued,
the injunction would not be adverse to the public interest.” Baker v. Buckeye
Cellulose Corp., 856 F.2d 167, 169 (11th Cir. 1988); Levi Strauss and Co. v. Sunrise
Int’l Trading Inc., 51 F.3d 982 (11th Cir. 1995). A preliminary injunction is a drastic

22

 Case 1:17-cv-01427-TCB-WSD-BBM Document 137 Filed 03/26/18 Page 23 of 50

remedy “which should not be granted unless the movant clearly carries the burden of
persuasion.” Canal Auth. of Fla. v. Callaway, 489 F.2d 567, 573 (11th Cir. 1974).
Plaintiffs have the burden of establishing their entitlement to a preliminary
injunction. Citizens for Police Accountability Political Comm. v. Browning, 572 F.3d
1213, 1217 (11th Cir. 2009). Plaintiffs’ motion for a preliminary injunction should
be denied because they have not shown any of the elements necessary to support their
request for this extraordinary remedy.
1.

Plaintiffs Have Not Shown a Substantial Likelihood of Success on
the Merits.

The most important factor in deciding whether to grant or withhold a
preliminary injunction is the consideration of a plaintiff’s likelihood of succeeding on
the merits, and a failure to meet this initial hurdle relieves a court from considering
the remaining factors. Church v. City of Huntsville, 30 F.3d 1332, 1341-45 (11th Cir.
1994) (citing Northeastern Fl. Chapter of the Ass’n of Gen. Contractors of Am. v.
City of Jacksonville, 896 F.2d 1283, 1285 (11th Cir. 1990) (“We need not address
every element because we conclude that no showing of irreparable injury was
made.”)). Here, Plaintiffs are neither likely to succeed on the merits of their claims
nor are they likely to satisfy the other requisites for relief.
a. The Redistricting of HD 105 and HD 111 Was Not
Predominantly Based on Race.
23

 Case 1:17-cv-01427-TCB-WSD-BBM Document 137 Filed 03/26/18 Page 24 of 50

Plaintiffs completely ignore both the direct testimony of the individual who
drafted the districts at issue in this litigation and the testimony of the legislators
representing HD 105 and HD 111, and ask this Court to infer discriminatory intent
because there is a correlation between a voter’s race and partisanship. The Supreme
Court has repeatedly condemned such an approach.
If district lines merely correlate with race because they are drawn on the
basis of political affiliation, which correlates with race, there is no racial
classification to justify.
Bush v. Vera, 517 U.S. 952, 967-968 (1996) (plurality). Three years later, the Court
again stressed that awareness of a relationship between race and partisanship does not
transform use of partisan data into racial intent.
Our prior decisions have made clear that a jurisdiction may engage in
constitutional political gerrymandering, even if it so happens that the
most loyal Democrats happen to be black Democrats and even if the
State were conscious of that fact.
Hunt v. Cromartie, 526 U.S. 541, 551 (1999) (emphasis in original).
Redistricting plans are subject to strict scrutiny only where “race was the
predominant factor motivating the legislature’s decision to place a significant number
of voters within or without a particular district.” Cooper v. Harris, 137 S. Ct. 1455,
1463 (quoting Miller v. Johnson, 515 U.S. 900, 916 (1995)). It is the Plaintiffs’
burden to show that race was the predominant factor in drawing these districts. Id.

24

 Case 1:17-cv-01427-TCB-WSD-BBM Document 137 Filed 03/26/18 Page 25 of 50

The Plaintiffs cannot meet this burden because the evidence in this case is that
partisanship considerations and data, not race, drove the redistricting of HD 105 and
HD 111.
Moreover, Gina Wright, Executive Director of the Legislative and
Congressional Reapportionment Office, who actually drafted the plans, testified that
she did not look at race as she was moving precincts and blocks in and out of these
districts. See Exhibit A at ¶¶ 8-10; Wright Dep., 219:9 to 220: 21. 10 Wright testified
that she looked at the racial demographics for the districts only at the end, after she
had finished making changes, and only, then, to see that she had not made a
significant change in the racial demographics of the districts. Id. Plaintiffs ignore
this testimony, and instead quote her testimony regarding the data available to her
when redistricting and, in the context of a discussion about the 2011 redistricting, her
practice of what data she would generally include in a pending changes box. 11 ECF
103-1 at 26 and 47. Prior to Shelby Co. v. Holder, 570 U.S. 529 (2013), it would
have been a practical necessity to consider race in order to comply with the non-

10

As Ms. Wright noted, neither district was majority minority. Wright Dep., 30:12 to
30:19.
11
Plaintiffs’ expert, Peyton McCrary, makes the same false assumption, taking
Wright’s deposition testimony from 2011 and attributing it to her work on the 2015
maps. ECF 65-1 ¶ 51. McCrary does this despite Wright’s clear testimony to the
contrary. Wright Dep.. 219:9 to 220:21.
25

 Case 1:17-cv-01427-TCB-WSD-BBM Document 137 Filed 03/26/18 Page 26 of 50

retrogression standard of Sec. 5 of the Voting Rights Act. Plaintiffs then misinterpret
this testimony as a concession that Ms. Wright used racial data while moving blocks
and precincts in and out of HD 105 and HD 111 in 2015.12
In another instance, Plaintiffs again state that “Wright testified that in finding []
balance, she looked at racial data,” referring the Court to page 29 of her deposition.
ECF 103-1 at 24. However, Wright’s testimony continues to the following page
where she explains that she looked at the racial make-up of the district “eventually
. . . to make sure that [she] did not do significant harm in that respect.” Wright Dep.
30:6-30:8. That is not a concession that she used race in making changes to the
districts.
Ms. Wright testified that she first made the changes to the districts and only
after completing those changes looked at resulting racial demographics.
[M]y objective was to see if I could find political improvement in the
Republican number for her district. So I’m going about this in a way of
knowing that, let’s see if we - - Republican area here, try and see if that
boosts the district.

12

Plaintiffs also recite in their brief that Wright ran reports providing detailed racial
breakdowns of the precincts, appearing to interpret this to mean that Wright was
interested in these demographics. ECF 103-1 at 31. Plaintiffs fail to include that
these reports were only run after the litigation was filed and at the request of counsel
from the Attorney General’s Office. Wright Dep. 295:19 – 296:19; 331:1-15;
Strangia Dep. 96:1-14 (testifying that report is not typical report prepared by the
LCRO). The same demographic report appears at Pls.’ Exhibits. 77, 146 and 167 in
the depositions. ECF 103-72.
26

 Case 1:17-cv-01427-TCB-WSD-BBM Document 137 Filed 03/26/18 Page 27 of 50

If that boosted the district, what my next objective would be is to
find a way to balance that population out, . . .
And, obviously, the places where I had split precincts would be
my first choice to go . . .
And then I added some . . . [population] to complete a balancing
of population. And it was at that point that I would then go back and
see, if I did these changes, what impact did that then have on the overall
percent total black, Hispanic, and whatnot.
...
I’m looking at what was my overall objective? Can I achieve that?
Then can I balance [the population]? And then what is the numerical
result of that after that is done?
Wright Dep. 219:9 – 220:21. See also Exhibit A at ¶¶ 8-10.
Plaintiffs contend that “[t]he high number and character of the split precincts in
Districts 105 and 111 indicate that Wright used racial data in creating them.” ECF
103-1 at 33. Again, Wright’s deposition testimony clearly refutes the suggestion.
Asked at her deposition about the precinct splits in HD 105, Wright testified that she
started by adding heavily Republican areas to the district and then balancing out the
population to get the district’s total population back to near the ideal district size.
Wright Dep. 213:14 – 214:2. Wright began by adding an area she knew to be “very,
very, Republican” to the district, Harbins C. See Exhibit A ¶¶ 13-15 and Exhibit 2
attached thereto; Wright Dep. 214:9 - 214:11. She then tried to take all of Harbins A,
but due to its size, settled on a road that cut through the entire precinct. Id.; Wright
27

 Case 1:17-cv-01427-TCB-WSD-BBM Document 137 Filed 03/26/18 Page 28 of 50

Dep., 214-216. Asked about the demographics of precinct Harbins A, and told that
the portion of Harbins A that was put in HD 105 has a lower African-American
percentage than that part of the precinct that was left in HD 104, Wright stated “I
would not know that. I didn’t look at that that way.” Wright Dep. 217:8-9.
Review of Exhibit 2 of Gina Wright’s declaration, Exhibit A attached hereto,
further supports Wright’s decision to split Harbins A using the boundary she chose.
The shape of the Harbins C precinct makes taking the area in Harbins A that is
adjacent to it a natural choice. Plaintiffs, and Plaintiffs’ expert Chen, criticize Wright
for choosing that part of Harbins A in grey on Exhibit 2 rather than the part just
above it in blue. ECF 103-1 at 47 and 63-1 at 24-25 and 34-35. 13 But the resulting
district shape between those two choices supports Wright’s decision. It is Plaintiffs,
not Wright, that want to make choices based on the racial composition of the blocks.
Wright chose not to look at race.
Wright testified further that she tried to balance the population total in HD 111
by putting precinct Lawrenceville M back together and settling on a boundary in

13

Chen further criticizes Wright’s split of the Baycreek H precinct, but fails to
mention that Wright made no change to the Baycreek H precinct that is split between
HD 105 and HD 114, a district that was not part of the 2015 redistricting. ECF 63-1
at 34.
28

 Case 1:17-cv-01427-TCB-WSD-BBM Document 137 Filed 03/26/18 Page 29 of 50

Harbins A. 14 Exhibit A ¶¶ 17-19 and Exhibit 3 attached thereto; Wright Dep., 219220. Needing additional population in HD 105 to balance the total population, she
naturally looked to a precinct that was already split. Exhibit A ¶ 30 and Exhibit 4
attached thereto; Wright Dep., 219:21 to 220:5. In fact, the two blocks in
Lawrenceville D that Wright added to HD 105 are majority African American.
Exhibit A ¶ 24.
Plaintiffs nonetheless report the Lawrenceville D precinct split as yet another
indication that Wright used race to lower the African American percentage in HD
105. Rather than focus on the boundary change to the Lawrenceville D precinct split
in 2015, i.e., the population of the two blocks that were moved, Plaintiffs instead
submerge the population of the two blocks that were moved with the population in
Lawrenceville D that was untouched and remained in HD 105, then they argue that
the Lawrenceville D precinct was split in a manner that disadvantaged minority
voters. Doc. 63-1 at 34. Plaintiffs’ characterization of this precinct split choice is
even more suspect when looking at a map. Review of Exhibits 2 and 4 of Wright’s
declaration show that had she chosen to swap the two parts of the Lawrenceville D
precinct between HD 104 and HD 105, HD 105 would not be contiguous.

14

Precinct Lawrenceville M is oddly shaped and a little difficult to see with the 2012
district boundaries overlay. The precinct extends to highway 316, just below the grey
area on the map with the district label 102. See Exhibit 3 to Wright Declaration.
29

 Case 1:17-cv-01427-TCB-WSD-BBM Document 137 Filed 03/26/18 Page 30 of 50

Chen’s criticism of the precinct splits in HD 111 is equally uninformative.
Again, Chen infers racial intent because in four (4) of the five (5) precinct splits in
HD 111 that portion of the precinct included in HD 111 has a lower AfricanAmerican concentration than the part of the precinct outside of HD 111. ECF 63-1 at
36-38. The first such precinct identified by Chen is Tussahaw, a precinct that
“performs pretty strongly Republican [and] was beneficial to [Strickland’s] T-rep
number” according to Wright. Wright Dep., 230:9-12. According to Chen, the
portion of the precinct that Wright included in HD 111 is 4.8% African-American
voting age population, and the part outside of HD 111 was 6.9% African-American
voting age population , numbers that are both nominal. ECF 63-1 at 36. More
importantly, Chen does not consider 1) that Wright took only roughly 25% of the
precinct’s population (Exhibit 10); 2) that Representative Welch (HD 110) lives in
Tussahaw and wanted to stay in his own district; or 3) that swapping those two parts
of Tussahaw would make HD 110 non-contiguous. See Exhibits 6 and 10 to Exhibit
A.
The next precinct split that Chen criticizes is Mount Carmel. According to
Chen, the African American voting age population included in HD 111 is 43.7%
while the part excluded from HD 111 is 45.2%, a minimal difference. ECF 63-1 at
36. Of course, since HD 111 began with a total African-American voting age
30

 Case 1:17-cv-01427-TCB-WSD-BBM Document 137 Filed 03/26/18 Page 31 of 50

population of 33.9%, ECF 63-1 at 25, adding either part of the Mount Carmel split
precinct necessarily increased the overall African-American voting age population in
this district. However, like the problems with the Tussahaw precinct, swapping the
split precinct parts of Mount Carmel between HD 73 and HD 111 would cause
contiguity problems. See Exhibits 6 and 14 to Exhibit A. Chen does not mention
these.
With respect to the Flippen precinct, Chen notes that the portion of the precinct
in HD 111 is 38.8% African-American voting age population while the part of the
precinct outside of HD 111 is 41.1% African-American voting age population. ECF
63-1 at 38. Like the Mount Carmel precinct, this precinct is over 33.9% AfricanAmerican voting age population, so addition of either part of the precinct increases
slightly the majority African-American percentage. However, with respect to the
choice of which part of the precinct to take, again, looking at the maps makes the
choice rather obvious. See Exhibits 6 and 8 to Exhibit A. Taking that part of the
Flippen precinct that is in HD 109 (orange) and swapping it with the part of Flippen
in HD 111 (purple) would make the Stagecoach and Stockbridge West precincts noncontiguous with the rest of HD 109.

31

 Case 1:17-cv-01427-TCB-WSD-BBM Document 137 Filed 03/26/18 Page 32 of 50

Chen’s analysis is not probative of any racial intent, rather, it merely reports
the racial demographic change in the districts and within precinct splits.15 Plaintiffs
themselves contend that there is a high correlation between race and partisanship in
Gwinnett and Henry counties. Yet when Wright’s deliberate attempts to increase
Republican political performance through the use of political data (election returns) is
successful, Plaintiffs argue that the resulting correlation to race must mean Wright
used race as a proxy for partisanship. The argument is circular.
Evidence that blacks constitute even a supermajority in one
congressional district while amounting to less than a plurality in a
neighboring district will not, by itself, suffice to prove that a jurisdiction
was motivated by race in drawing its district lines when the evidence
also shows a high correlation between race and party preference.
Hunt, 526 U.S. at 551-552 (emphasis added). The fact that the African-American
and/or Hispanic population percentage in HD 105 and HD 111 decreased slightly
with the 2015 redistricting is insufficient to establish that the districts were
predominantly based on race. The offense to the Equal Protection Clause is the use
of race to draw the districts, not a resulting minority percentage drop in the overall
district population.

15

Similarly, Francys Johnson, former President of the GA NAACP, does not point to
any evidence other than the fact that the percentage African-American in HD 105 and
111 decreased to support Plaintiffs’ contention that race predominated in the
redrawing of the districts. Johnson Dep. 15:21-25.
32

 Case 1:17-cv-01427-TCB-WSD-BBM Document 137 Filed 03/26/18 Page 33 of 50

[The Equal Protection Clause] prevents a State, in the absence of
“sufficient justification,” from “separating its citizens into different
voting districts on the basis of race.” Bethune-Hill v. Virginia State Bd.
of Elections, 580 U. S. ___, ___, 137 S. Ct. 788, 197 L. Ed. 2d 85
(2017) (internal quotation marks and alteration omitted).
Cooper v. Harris, 137 S. Ct. 1455, 1463 (2017). Wright testified, both in her
deposition and by declaration, that she did not look at the racial demographics until
after making all of the changes to these districts. Exhibit A ¶¶ 8-10; Wright Dep.,
29:19 to 30:9, 219:9 to 220:21.
It is hard to imagine a scenario where any change to district boundaries will
result in no change, up or down, to the racial demographics of a district. Therefore,
the only way Wright would have been able to redistrict HD 105 and HD 111 and not
change the racial demographics of the district, or if possible increase the racial
minority of the district, would have been by using race while drawing the districts.
Because the districts at issue are not majority African-American or majority Hispanic,
Wright had no reason to believe she was free to use race to ensure that the minority
percentage in the districts stayed unchanged.
It is only “[i]f a State has a good reason to think that all the ‘Gingles
preconditions’ are met, then so too it has good reason to believe that §2 requires
drawing a majority-minority district. But if not, then not.” Cooper, 137 S. Ct. 1455,
1470 (internal citation omitted). Here, there was no reason for Wright to pay
33

 Case 1:17-cv-01427-TCB-WSD-BBM Document 137 Filed 03/26/18 Page 34 of 50

attention to the racial demographics because the districts were not majority minority.
Furthermore, Wright had political data from contested statewide general elections to
measure the partisan advantage of each precinct and precinct split. Exhibit A ¶¶ 910; Strangia Dep. 24:21 to 26:22 (explaining how precinct election data is allocated
down to the block level for use in redistricting). Nor did Wright get specific direction
from the incumbent legislators about which particular areas to move in and out of
their districts. Exhibit A ¶ 42; Wright Dep. 178:3-12. Wright also did not discuss the
redistricting plan with Dan O’Connor. Exhibit A ¶ 43; Wright Dep., 193:2-5.
Finally, Dan O’Connor himself, while being questioned about a summary of the
redistricting plan that he drafted after the plan had been completed, repeatedly
testified that he had no role in the redistricting of HD 105 and HD 111. O’Connor
Dep. 140:4 – 5 (“I wasn’t involved in the, you know, the map drawing of 111.”);
135:15 -16 (“And, again, I wasn’t involved in 111.”); 135:19 -20 (“I wasn’t involved
in [HD 105]”). Plaintiffs try to paint O’Connor as the principal architect of the
redistricting plan, but both he and Wright testified unequivocally that O’Connor was
not involved in the creation of the redistricting plan. O’Connor, like Plaintiffs, does
believe that there is a correlation between race and partisanship. But that awareness,
particularly from someone that did not participate in drawing the district lines, is not

34

 Case 1:17-cv-01427-TCB-WSD-BBM Document 137 Filed 03/26/18 Page 35 of 50

evidence that race was a predominant factor in the redistricting. Hunt, 536 U.S. at
551; Vera, 517 U.S. at 967-68.
Plaintiffs’ expert, Peyton McCrary, also misinterprets Wright’s testimony
about the data she relied on in making changes to HD 105 and HD 111. McCrary
quotes Wright’s testimony in her deposition regarding the data she has available in
her redistricting software, but characterizes that testimony as pertaining to her
redistricting in 2015. ECF 65-1 ¶ 51. Wright makes it clear in her deposition that she
did not use the block level race data available to her in drawing HD 105 and HD 111.
Wright Dep. 219:9 to 220:21; Exhibit A ¶¶ 8-10. McCrary also concludes that
because, in his opinion, the disaggregation of election data below the precinct level is
not a particularly reliable indicator of how a precinct voted at the block level, Wright
must have used race instead. ECF 65-1 ¶¶ 47-48.
It is these fundamental flaws in McCrary’s report that then lead him to the
conclusion that HB 566 was the product of intentional racial discrimination.
Plaintiffs are simply using the correlation between race and partisanship to argue that
because legislators were intending a political result, and after the changes were made
African-Americans are a slightly smaller percentage of the population in these
districts, the map drawer must have used race to achieve the political end. The
Supreme Court has been very clear that:
35

 Case 1:17-cv-01427-TCB-WSD-BBM Document 137 Filed 03/26/18 Page 36 of 50

If district lines merely correlate with race because they are drawn on the
basis of political affiliation, which correlates with race, there is no racial
classification to justify.
Vera, 517 U.S. at 967-968. “The racial predominance inquiry concerns the actual
considerations that provided the essential basis for the lines drawn.” Bethune-Hill,
137 S. Ct. at 799. Plaintiffs ignore the direct testimony of Gina Wright that she relied
on the political data (election returns) to draw the districts, and instead infer intent
from the racial impact of the districts (i.e., a 2% decrease in African-American voting
age population), even as they acknowledge a correlation between race and
partisanship. The overwhelming evidence is that race was not the predominant factor
in drawing HD 105 and HD 111.
b. Plaintiffs’ Request for a Preliminary Injunction is Barred by the
Doctrine of Laches.
Laches applies to a request for equitable relief when (1) there was a delay in
asserting the claim; (2) the delay was not excusable; and (3) the delay caused the nonmoving party undue prejudice. United States v. Barfield, 396 F.3d 1144, 1150 (11th
Cir. 2005); Kason Indus. v. Component Hardware Group, 120 F.3d 1199, 1203 (11th
1997); see also Costello v. United States, 365 U.S. 265, 282 (1961).
Here, the legislation adopting the challenged redistricting plan was adopted in
2015. The NAACP Plaintiffs first waited until April 24, 2017 to file the complaint
challenging the plan, and then waited until February 20, 2018 to file their motion for
36

 Case 1:17-cv-01427-TCB-WSD-BBM Document 137 Filed 03/26/18 Page 37 of 50

preliminary injunction. The Thompson Plaintiffs did not file their Complaint until
October 3, 2017.
As outlined in the Declaration of Michael Barnes, Director of the Center for
Election Systems at the Office of the Secretary of State (“CES”), the Secretary of State
is responsible for creating each type of ballot that will be used in 158 of 159 Georgia
counties. Exhibit C at ¶ 4. To ensure that each ballot is accurate and complete, CES
had to begin this process for the May 22, 2018 primary, which includes elections HDs
105 and 111, on March 10, 2018. Id. at 7. Absentee ballots must be mailed to voters
by April 6, 2018, and counties could start receiving completed absentee ballots as early
as April 9, 2018. Id. at ¶ 5. Because of this timeline, the counties that encompass HDs
105 and 111 have to begin administering the May 22, 2018 election as early as April
7, 2018. See Exhibit E at ¶ 2; Exhibit F at ¶ 2.
Finally, candidates have already qualified for both the Republican Primary and
Democratic Primary in HDs 105 and 111. Exhibit D at ¶ 6. Five other districts were
redrawn in H.B. 566 to make the changes to HDs 105 and 111 (see Exhibit A at ¶ 5),
and candidates have qualified for the Republican and Democratic Primaries, as
applicable, in these districts as well. Exhibit D at ¶ 6. In HD 105, one of the four
candidates running for office lives in an area of HD 105 that was in HD 104 under the
2012 plan. Exhibit A at ¶ 50. This candidate has already paid a qualifying fee to run in
37

 Case 1:17-cv-01427-TCB-WSD-BBM Document 137 Filed 03/26/18 Page 38 of 50

HD 105 and is presumably already campaigning for office. The same is true for HD
111; two of the four candidates running for office live in an area of HD 111 that was
in HD 109 under the 2012 plan. Id. at ¶ 51. In addition, the voters in HD 105 and 111
will already be expecting to have the opportunity to vote for these candidates as their
representatives.
It is clear, then, that a preliminary injunction at this late stage in the election
process not only unduly prejudice the Secretary of State and cause harm to the counties
administering the election, but such an injunction would also harm the candidates and
voters who are already participating in the elections for the next representatives from
HDs 105 and 111.
2.

Plaintiffs Cannot Show That They Will Suffer Irreparable Harm
Absent the Granting of a Preliminary Injunction

Plaintiffs have not shown that they will suffer irreparable harm if a preliminary
injunction is not granted. “A showing of irreparable harm is the ‘sine qua non’ of
injunctive relief.” Siegel v. LePore, 234 F.3d 1163, 1176 (11th Cir. 2000) (citations
omitted). When a plaintiff has not shown a likelihood of success on the merits,
claims for irreparable injury based on an alleged constitutional injury have no merit.
Overstreet v. Lexington-Fayette Urban County Gov’t, 305 F.3d 566, 578 (6th Cir.
2002). Here, Plaintiffs have failed to show irreparable harm.

38

 Case 1:17-cv-01427-TCB-WSD-BBM Document 137 Filed 03/26/18 Page 39 of 50

3.

The Damage to the Defendants Outweighs Any Alleged Injury to
Plaintiff.

On a motion for preliminary injunction, the plaintiff bears the burden of
showing that the perceived injury outweighs the damages that the preliminary
injunction might cause to the defendants. Baker v. Buckeye Cellulose Corp., 856
F.2d 167, 169 (11th Cir. 1988). “Only in rare instances is the issuance of a
mandatory preliminary injunction proper.” Harris v. Wilters, 596 F.2d 678, 680 (5th
Cir. 1979). Additionally, in election cases courts should give consideration to the
proximity of the election and the potential for any voter confusion that a last minute
change to the State’s processes may lead to. Purcell v. Gonzalez, 549 U.S. 1, 4-5
(2006).
As explained above, here, candidates have already qualified for office. And
some of the candidates that qualified for HD 105 and HD 111 will not live in those
districts if they revert to the 2012 boundaries. Exhibit A ¶¶ 49-51 and exhibits
attached thereto.
If this Court ordered that the election for the seven affected House Districts
must be conducted using the 2012 boundaries, county registrars for all of the affected
counties would have to reassign all of the affected voters to new House Districts. See
Harvey Decl. ¶ 6. The process is manual, and requiring that county registrars engage
in this process, while in the midst of conducting an election has never been attempted.
39

 Case 1:17-cv-01427-TCB-WSD-BBM Document 137 Filed 03/26/18 Page 40 of 50

Id. at ¶ 7. Nor was mid-election redistricting contemplated when Georgia’s voter
registration system was designed. Id. In addition, mid-election redistricting
combined with a new special election would lead to voter confusion and likely lower
turnout. Id. at ¶ 5; Exhibit F at ¶ 10.
4.

The Preliminary Injunction Will Not Serve the Public Interest

A plaintiff also bears the burden of showing that the preliminary injunction
would serve the public interest. Baker, 856 F.2d at 169. Contrary to the Plaintiffs’
contention, reversing the changes made to HDs 105 and 111 for the purpose of
conducting a special election either parallel to the regularly scheduled General Primary
/ Nonpartisan General Election or on its own would not be in the public interest. Such
an order would confuse candidates and voters, disrupt an ongoing process, and tax the
public with costs.
In North Carolina v. Covington, 137 S. Ct. 1624 (2017), the United States
Supreme Court reversed a district court’s order directing the conduct of special
elections for a truncated one-year term. Significantly, the district court ordered those
special elections stating, “While special elections have costs, those costs pale in
comparison to the injury caused by allowing citizens to continue to be represented by
legislators elected pursuant to a racial gerrymander.” Id. at 1625 (quoting App. to Juris.

40

 Case 1:17-cv-01427-TCB-WSD-BBM Document 137 Filed 03/26/18 Page 41 of 50

Statement 200). Put simply, the Supreme Court’s per curiam reversal of that order
shows that racial gerrymandering claims like those of Plaintiffs are not talismanic.
Rather, this Court must weigh the equities, not simply looking for a “fitting
remedy” for any legal violations, but also “taking account of ‘what is necessary, what
is fair, and what is workable.’” Id. (quoting New York v. Cathedral Academy, 434 U.S.
125, 129 (1977)). The Supreme Court identified several “obvious considerations” to
be part of the balancing process: “the severity and nature of the particular constitutional
violation, the extent of the likely disruption to the ordinary processes of governance if
early elections are imposed, and the need to act with proper judicial restraint when
intruding on state sovereignty.” Id. at 1626. Those factors are not exclusive, “but they
are among the matters a court would generally be expected to consider in its ‘balancing
of the individual and collective interests’ at stake.” Id. (quoting Swann v. CharlotteMecklenburg Bd. of Ed., 402 U.S. 1, 16 (1971)).
The Seventh Circuit noted in Fulani v. Hogsett, 917 F. 2d 1028 (7th Cir. 1990),
that claims “against a state electoral process must be expressed expeditiously.” Id. at
1031. It noted, “As time passes, the state’s interest in proceeding with the election
increases as resources are committed and irrevocable decisions are made.” Id.
As of the filing of this Response, candidates have qualified, and the ballot
preparation process is well underway. The Secretary of State is required by federal law
41

 Case 1:17-cv-01427-TCB-WSD-BBM Document 137 Filed 03/26/18 Page 42 of 50

to transmit all absentee ballots to military and civilian overseas voters (“UOCAVA
voters”) at least 45 days prior to the date of any election for federal office. United
States v. Georgia, 778 F.3d 1202, 1203 (11th Cir. 2015); United States v. Alabama,
778 F.3d 926, 934-35 (11th Cir. 2015). See also 52 U.S.C. § 20302. 16 Thus, the
Secretary of State is legally required to transmit all UOCAVA ballots to voters by no
later than April 6, 2018.17 Exhibit C ¶ 5. In addition, advance voting for the May 22,
2018 Primary Election begins on April 30, 2018. Exhibit D, Harvey Decl. Exhibit 1 at
2. As Michael Barnes explains, CES prepares all election ballots for every county in
Georgia except Pike County. Exhibit C ¶ 4. For the May 22, 2018 General Primary
Election, CES began preparing ballots on March 10, 2018, the day after the closing of
candidate qualification and more than 80 days before the election. Id. ¶ 7. CES is
transmitting ballot images, including images for printed absentee ballots, to the
counties designated printer between March 24, 2018, and April 1, 2018. Id. ¶ 6. Barnes
notes, “[T]hese ballots must be transmitted by no later than 50 days prior to the date of
the election” so that they are in the hands of the local election officials in time for them

16

Formerly 42 U.S.C. § 1973ff(a)(8)(A).
Georgia law provides that absentee ballots may be transmitted to UOCAVA voters
up to 49 days prior to the date of the election. O.C.G.A. § 21-2-384(a)(2).
42

17

 Case 1:17-cv-01427-TCB-WSD-BBM Document 137 Filed 03/26/18 Page 43 of 50

to be sent to the voters covered by the Uniformed and Overseas Citizens Absentee
Voting Act (“UOCAVA”) and state law not less than 45 days before the election. 18 Id.
In addition, notices will have to be sent to voters telling them which precincts
they will vote in and which offices they will vote for. As Tina Lunsford notes, those
notices should go out no less than 60 days before the election so that voters covered by
the Uniformed and Overseas Citizens Absentee Voting Act, 52 U.S.C. §§ 2030120311, are reached in timely fashion. Exhibit E, Lunsford Decl. ¶ 8.
The declarations of Chris Harvey, Lynn Ledford, and Tina Lunsford show that
in order to use the 2012 district boundaries in a special election conducted in
conjunction with the regularly scheduled 2018 elections would result in the imposition
of both monetary and nonmonetary costs. Those declarations also detail the magnitude
of the tasks that public officials must perform in order to redistrict and put on a special
election.
Lunsford and Ledford each outline the steps they must take and the monetary
costs that are associated with each step. Lunsford states, “The estimated cost to hold a
court-ordered Special Election with the General Primary/General Election would be
$66, 995.” Exhibit E ¶ 7. That estimated cost would triple if the elections were not all
on the same ballot. Id. Ledford estimates her costs for HDs 104 and 105 as $168,000,

18

O.C.G.A. § 21-2-384(a)(2) implements the requirements of UOCAVA.
43

 Case 1:17-cv-01427-TCB-WSD-BBM Document 137 Filed 03/26/18 Page 44 of 50

exclusive of the cost of advance in-person satellite locations, runoffs, and other
redistricting changes. Exhibit F ¶ 8.
As for nonmonetary costs, changing the district lines and ordering a special
election will disqualify some candidates, cause others to change their strategy in
midstream, and confuse voters. As Gina Wright notes, one of the four candidates who
has qualified to run in HD 105 lives in a portion of HD 105 that was in HD 104 in 2012.
Exhibit A ¶51. She also points out that two of the four candidates who have qualified
to run in the 2015 version of HD 111 live in portions of that district that were in HD
109 under the 2012 plan. Id. Accordingly, absent some other changes to the ballots, the
Plaintiffs contemplate disqualifying three candidates.
Putting the districts back as they were in 2012 would alter their political
landscapes, given that it would undo Wright’s effort to add Republican voters to HDs
105 and 111. The candidates who remain after the disqualifications will have to reach
voters in precincts other than the ones they presently have in mind. Democrats in HD
111, for example, would likely target precincts like McDonough Central, and
Republicans could no longer reach their partisans in Tussahaw.
The affected voters would be confused by the change in their representatives.
Ledford explains that running elections in parallel causes voter confusion. Exhibit F ¶
10. She notes further that, if there are stand-alone elections for HDs 104 and 105 in
44

 Case 1:17-cv-01427-TCB-WSD-BBM Document 137 Filed 03/26/18 Page 45 of 50

Gwinnett County, voters may think that everyone is entitled to vote in those elections.
Anyone who shows up to find his or her polling place closed is unlikely to appreciate
the County’s efforts.
As Ledford also notes, there is no guarantee that the polling places that are used
for regularly scheduled elections will be available if a special election is ordered.
Ledford also points out the effect of the Census Bureau’s designation of Gwinnett
County as a covered § 203 jurisdiction for Spanish: all sorts of election materials and
the ballots will have to be translated. See 52 U.S.C. § 10503.
All of this effort to hold a special election must be balanced against the return.
As Lynn Ledford observes, “[S]pecial elections historically do not garner the same
level of attention and voter turnout as regularly scheduled elections do.” Exhibit F ¶
10.
Ledford’s experience is reflected in the two stand-alone special elections
conducted most recently in Georgia. A special election for HD 175 was held on
February 13, 2018, and voter turnout was only 9.93% of the registered voters in the
district. See http://results.enr.clarityelections.com/GA/72657/Web02-state.192841/#/.
In the special election for Senate District 17 and HD 111 held on January 16, 2018,
turnout

was

only

6.99%

of

the

registered

voters

http://results.enr.clarityelections.com/GA/72405/Web02-state/#/.
45

involved.

See

 Case 1:17-cv-01427-TCB-WSD-BBM Document 137 Filed 03/26/18 Page 46 of 50

As noted above, Defendant believes that Plaintiffs are unlikely to prevail in this
matter and therefore the motion for preliminary injunction should be denied. However,
even if the Court believed that Plaintiffs were likely to prevail, the public interest is not
served by scheduling a special election.
III.

CONCLUSION

As Plaintiffs cannot satisfy the standards for an injunction, their request for a
preliminary injunction should be denied.
Respectfully submitted this 26th day of March, 2018.
s/ Frank B. Strickland
Frank B. Strickland
Special Assistant Attorney General
Georgia Bar No. 687600
fbs@sbllaw.net
John J. Park, Jr.
Georgia Bar No. 547812
jjp@sbllaw.net
Barclay S. Hendrix
Georgia Bar No. 917852
Barclay.hendrix@sbllaw.com
STRICKLAND BROCKINGTON
LEWIS LLP
Midtown Proscenium Suite 2200
1170 Peachtree Street NE
Atlanta, Georgia 30309
678-347-2200 (telephone)
678-347-2210 (facsimile)
CHRISTOPHER M. CARR
Attorney General 112505
46

 Case 1:17-cv-01427-TCB-WSD-BBM Document 137 Filed 03/26/18 Page 47 of 50

ANNETTE M. COWART 191199
Deputy Attorney General
RUSSELL D. WILLARD 760280
Senior Assistant Attorney General
CRISTINA CORREIA 188620
Assistant Attorney General
40 Capitol Square SW
Atlanta, GA 30334
ccorreia@law.ga.gov
404-656-7063
404-651-9325
Attorneys for Defendant

47

 Case 1:17-cv-01427-TCB-WSD-BBM Document 137 Filed 03/26/18 Page 48 of 50

CERTIFICATE OF COMPLIANCE
Pursuant to L.R. 7.1(D), the undersigned hereby certifies that the foregoing
DEFENDANT BRIAN KEMP’S BRIEF IN OPPOSITION TO PLAINTIFFS’
MOTION FOR A PRELIMINARY INUNCTION has been prepared in Times New
Roman 14, a font and type selection approved by the Court in L.R. 5.1(C).
s/ Frank B. Strickland
Frank B. Strickland
Special Assistant Attorney General
Georgia Bar No. 687600

48

 Case 1:17-cv-01427-TCB-WSD-BBM Document 137 Filed 03/26/18 Page 49 of 50

CERTIFICATE OF SERVICE
I hereby certify that on March 26, 2018, I served the within and foregoing
DEFENDANT BRIAN KEMP’S BRIEF IN OPPOSITION TO PLAINTIFFS’
MOTION FOR A PRELIMINARY INUNCTION with the Clerk of Court using the
CM/ECF system, which will send notification of such filing to all parties to this matter
via electronic notification or otherwise:
Jon Greenbuam
jgreenbaum@lawyerscommittee.org
Julie Houk
houk@lawyerscommittee.org
John Powers
jpowers@lawyerscommittee.org
Ezra Rosenberg
erosenberg@lawyerscommittee.org
Lawyers’ Committee for Civil Rights
Under Law
1401 New York Avenue, Suite 400
Washington, DC 20005
William Vance Custer, IV
bill.custer@bryancave.com
Jennifer Burch Dempsey
Jennifer.dempsey@bryan cave.com
Julia Fenwick Ost
Julie.ost@bryancave.com
Bryan Cave, LLP-ATL
One Atlantic Center
14th Floor
1201 West Peachtree St, NW
Atlanta, GA 30309-3488

49

 Case 1:17-cv-01427-TCB-WSD-BBM Document 137 Filed 03/26/18 Page 50 of 50

Bradley S. Phillips
Brad.phillips@mto.com
Gregory D. Phillips
Gregory.phillips@mto.com
John F. Muller
Thomas P. Clancy
Thomas.clancy@mto.com
Munger, Tolles & Olson, LA-CA
50th Floor
350 South Grand Avenue
Los Angeles, CA 90071-1560
Quinton Washington
quinton@bellwashington.com
Bell & Washington LLP
196 Peachtree Street SW, Suite 310
Atlanta, GA 30303
Marc Erik Elias
melias@perkinscoie.com
Aria C. Branch
abranch@perkinscoie.com
Perkins Coie LLP
700 13th Street, N.W., Suite 600
Washington, D.C. 20005
Abha Khanna
Akhanna@perkinscoie.com
Perkins Coie LLP
1201 Third Avenue, Ste. 4900
Seattle, WA 98101
This 26th day of March, 2018.
s/ Frank B. Strickland
Frank B. Strickland
Georgia Bar No. 687600
50