Leslie Feldman, et al v. Arizona Sec'y of State's Ofc., et al

Filing

FILED OPINION (SIDNEY R. THOMAS, CARLOS T. BEA and SANDRA S. IKUTA) AFFIRMED. Judge: SRT Dissenting, Judge: CTB Authoring, FILED AND ENTERED JUDGMENT. [10184244]

Case: 16-16865, 11/02/2016, ID: 10184244, DktEntry: 33-2, Page 1 of 33

FILED
NOV 02 2016

Feldman v. Sec. of State of Arizona, No. 16-16865

MOLLY C. DWYER, CLERK
U.S. COURT OF APPEALS

THOMAS, Chief Judge, dissenting:
Voting should be easy in America. In Arizona, it is not, and the burden falls
heaviest on minority voters. At issue in this appeal is Arizona’s refusal to count
ballots cast out-of-precinct, even for races in which the citizen is entitled, qualified,
and eligible to vote. Statistically significant evidence shows that this practice
disproportionately and adversely impacts minority voters. Because the district
court erred in denying the motion for a preliminary injunction directing that these
legitimate votes be counted, I respectfully dissent.
I
Under federal and state law, a voter who appears to vote at the wrong
precinct is entitled to cast a provisional ballot. 52 U.S.C. § 21082; A.R.S. § 16584. In Arizona, these provisional ballots are placed in a tub and taken to a voting
center. There, the voters’ addresses are compared with precinct geography. If a
ballot was cast by a person who lives in the precinct where he or she voted, then it
is counted. If the voter’s address was outside the precinct, the ballot is rejected in
its entirety, even as to races and ballot measures for which the voter was entitled to

-1-

Case: 16-16865, 11/02/2016, ID: 10184244, DktEntry: 33-2, Page 2 of 33

cast a ballot. These races include Presidential, Senatorial, and statewide elections,
and Congressional elections if the voter lives in the Congressional district.
Arizona is among the nation’s leaders in the number of ballots deemed
provisional. It leads the nation, by a wide margin, in the number of provisional
ballots rejected, and therefore not counted. The primary reason for rejecting
provisional ballots in Arizona is that they were cast out-of-precinct. The following
graph illustrates Arizona’s experience in comparison with that of other states:

Since 2006, Arizona has rejected over 121,000 provisional ballots cast out-

-2-

Case: 16-16865, 11/02/2016, ID: 10184244, DktEntry: 33-2, Page 3 of 33

of-precinct. Of the voters visiting a polling place in Arizona in the 2012 general
election, 22% were asked to cast a provisional ballot, and over 33,000 of
these—more than 5% percent of all in-person ballots cast—were rejected. The
provisional voting rate was 18% in 2014. No other state rejects a larger share of its
in-person ballots.
Rejected out-of-precinct provisional ballots are most prevalent in the
relatively urban counties, especially Maricopa and Pima. The vast majority of
Arizonans live in Maricopa and Pima Counties. Indeed, Maricopa County
accounts for 61% of Arizona’s population and almost 70% of all out-of-precinct
ballots cast.
Why is there such a high rate of out-of-precinct voting in Phoenix and
Tucson? The answer largely is that are relatively few polling places in those cities,
and polling sites change with great frequency. As the plaintiffs’ expert put it:
“Voters must invest significant effort in order to negotiate a dizzying array of
precinct and polling place schemes that change from one month to the next.” As
one State Senator observed, it is not uncommon for a voter’s assigned polling
location to change nearly every election. And, significantly for our consideration,
changes in polling place locations are statistically associated with higher rates of
out-of-precinct voting.

-3-

Case: 16-16865, 11/02/2016, ID: 10184244, DktEntry: 33-2, Page 4 of 33

The 2012 election cycle in Maricopa County provides an example. In the
general mid-term election in November 2010, there were 1,143 polling places. For
the Presidential Preference Primary in February 2012, there were 211. For the
general 2012 election, there were 724. In 2008, 2012, and 2016, Maricopa County
used a completely different precinct system for the Presidential Preference Primary
than for the general election. In returning to a precinct model, the County places
one polling place in each precinct. However, the number of registered voters
varies widely from precinct to precinct. For example, one precinct had
approximately 100 registered voters; another in the same geographic area had
9,000.
For the Presidential Preference Primary in 2016, Maricopa County used a
“vote center model,” in which voters can vote in any polling place in the county,
with the appropriate precinct ballot being generated for each voter at the vote
center. However, for the 2016 general election, Maricopa County switched back to
the precinct-model system, assigning voters back to hundreds of precincts.
Geography also plays a role. Many polling places are located directly on
precinct boundaries. Multiple polling places are often clustered together,
sometimes even in the same building. Some of these voting places are outside the
boundaries of the voter’s actual precinct. Many voters cast their ballots in

-4-

Case: 16-16865, 11/02/2016, ID: 10184244, DktEntry: 33-2, Page 5 of 33

incorrect precincts simply because they stood in the wrong line at a multi-precinct
location.
An assigned polling place is not necessarily the poll closest to the voter’s
residence. In fact, in Maricopa County, one quarter of out-of-precinct voters cast
ballots in an incorrect polling place that was actually closer to their home than their
assigned polling place. Indeed, most out-of-precinct votes in Maricopa County are
cast very close to the assigned polling place.
Causing additional confusion is the fact that the City of Phoenix conducts
elections at the same time at completely different polling places. Thus, a citizen
who wished to vote in person in both city and state elections would have to travel
to two entirely different voting places on election day.
The reduction of the number of polling places in Phoenix has also had an
impact on voting. During the last Presidential election, voters in some precincts
waited four to six hours to cast their ballots. One Congressman testified that
voters did not complete voting in his district until well after midnight. A State
Senator testified that there was only one polling place in his district of 70,000
people. In that district, it would take a voter using public transportation 50 minutes
to get to the voting booth. He testified that in West Phoenix, an area consisting of

-5-

Case: 16-16865, 11/02/2016, ID: 10184244, DktEntry: 33-2, Page 6 of 33

over 200,000 people in predominantly Hispanic neighborhoods, there were only
two voting centers. In his district, voters waited up to 5 hours to vote.
In addition, there were voters who were not told where their correct precinct
was located when their ballots were categorized as provisional, thereby preventing
them from voting at the correct precinct. One voter left the hospital to vote after
undergoing heart by-pass surgery. Two polling places where he had voted
previously were closed. He found a pamphlet sent to him by the county listing his
polling place, went to the place indicated, and voted a provisional ballot. The
election workers did not tell him that his vote would not be counted, nor did they
identify his correct voting place. He returned to the hospital after voting. His vote
was rejected in a race that ended up being decided by a handful of votes.
How does this complicated, kaleidoscopic method of designating polling
places affect minority voting? The record is undisputed: it has a statistically
significant adverse affect on minority voters.
The numbers are startling. The rate at which in-person ballots were rejected
and not counted because the votes were cast out-of-precinct was 131% higher for
Hispanics, 74% higher for African Americans, and 39% higher for Native
Americans than for white voters.

-6-

Case: 16-16865, 11/02/2016, ID: 10184244, DktEntry: 33-2, Page 7 of 33

According to the data collected by Maricopa County, many voters whose
ballots were classified as having been cast in the wrong precinct did not make a
mistake at all. Their ballots were marked as “out-of-precinct” and discarded even
though their registration precinct matched residency records for the precinct. In
other words, the citizen voted in the right place, but the voter’s properly cast ballot
was improperly rejected as being cast out-of-precinct.
The rate at which these ballots were rejected along ethnic and racial lines
was also significant: 80% higher for Hispanics, 34% higher for African
Americans, and 26% higher for Native Americans in comparison with white
voters. This problem is not trivial: fully 35% of the ballots rejected as being outof-precinct were discarded in error. And the disparity between white and nonwhite voters has proven consistent over time.1

1

The majority and the district court seem to discount this effect because the
plaintiffs are not challenging the precinct system per se. However, that is a red
herring. Plaintiffs are challenging the effect of the system, which can easily be
remedied by counting the ballots cast, rather than changing the entire precinct
system.
-7-

Case: 16-16865, 11/02/2016, ID: 10184244, DktEntry: 33-2, Page 8 of 33

The following graph illustrates the ethnic and racial disparity of outof-precinct ballots cast in Maricopa County from 2010-2014:

Overall, as the district court noted, white voters accounted for 56% of the
out-of-precinct ballots, despite casting 70% of all in-person votes. In contrast,
Hispanic voters made up 15% of in-person voters, but accounted for 26% of outof-precinct votes. African Americans accounted for 10% of in-person voters, but
13% of out-of-precinct votes.
Similar results occurred in Pima County for the general elections of 2010
and 2012. The rates at which African Americans and Hispanics cast

-8-

Case: 16-16865, 11/02/2016, ID: 10184244, DktEntry: 33-2, Page 9 of 33

out-of-precinct ballots in Pima County were significantly higher than the rate for
whites in both years. In 2012, the rate at which African American voters cast outof-precinct ballots was 37% higher than for white voters. For Hispanics, the rate
was 123% higher than for white voters. The rate was also higher in 2012 for
Native American voters by 47% in comparison to white voters. Each of these
differences is statistically significant. These racial and ethnic differences were also
statistically significant in the 2010 mid-term general election.
This disparity also exists in rural areas. In non-metropolitan counties, outof-precinct voting is negligible in majority-white precincts, but increases
dramatically in precincts where Hispanics and Native Americans make up
majorities.
Based on these facts, the plaintiffs filed suit under § 2 of the Voting Rights
Act and the First and Fourteenth Amendments, alleging that Arizona’s practice of
discarding out-of-precinct ballots disparately burdens minorities and leaves them
with less opportunity than other members of the electorate to participate in the
political process. The district court denied their motion for a preliminary
injunction, which brings us to the instant appeal.

-9-

Case: 16-16865, 11/02/2016, ID: 10184244, DktEntry: 33-2, Page 10 of 33

II
The district court erred in denying the motion for a preliminary injunction
founded on Voting Rights Act violations. As I noted in the companion appeal, the
Voting Rights Act of 1965 “was designed by Congress to banish the blight of
racial discrimination in voting, which has infected the electoral process in parts of
our country for nearly a century.” State of S.C. v. Katzenbach, 383 U.S. 301, 308
(1966) abrogated by Shelby Cty., Ala. v. Holder, __ U.S. __, 133 S. Ct. 2612
(2013). The Act “implemented Congress’ firm intention to rid the country of racial
discrimination in voting. It provided stringent new remedies against those practices
which have most frequently denied citizens the right to vote on the basis of their
race.” Allen v. State Bd. of Elections, 393 U.S. 544, 548 (1969).
The central purpose of the Act was “[t]o enforce the fifteenth amendment to
the Constitution of the United States.” Chisom v. Roemer, 501 U.S. 380, 383
(1991) (quoting Pub.L. 89–110, 79 Stat. 437, 42 U.S.C. § 1973 et seq.). The
Fifteenth Amendment provides 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, § 1.

-10-

Case: 16-16865, 11/02/2016, ID: 10184244, DktEntry: 33-2, Page 11 of 33

At issue in this preliminary injunction appeal, as well as in the companion
appeal, is § 2 of the Act, which is “a restatement of the Fifteenth Amendment.”
Chisom, 501 U.S. at 392. Section 2 provides, without limitation, that any voting
qualification that denies citizens the right to vote in a discriminatory manner
violates the Voting Rights Act. 52 U.S.C. § 10301; see also Allen, 393 U.S. at
566–67 (noting that Congress intentionally chose the expansive language “voting
qualifications or prerequisite to voting, or standard, practice, or procedure” for § 2
so as to be “all-inclusive of any kind of practice” that might be used by states to
deny citizens the right to vote (internal quotation marks omitted)). As amended in
1982, § 2 makes “clear that certain practices and procedures that result in the
denial or abridgment of the right to vote are forbidden even though the absence of
proof of discriminatory intent protects them from constitutional challenge.”
Chisom, 501 U.S. at 383–84.
To succeed on a § 2 claim, a plaintiff must show (1) that “the challenged
standard, practice, or procedure must impose a discriminatory burden on members
of a protected class, meaning that members of the protected class have less
opportunity than other members of the electorate to participate in the political
process and to elect representatives of their choice” and (2) “that burden must in
part be caused by or linked to social and historical conditions that have or currently

-11-

Case: 16-16865, 11/02/2016, ID: 10184244, DktEntry: 33-2, Page 12 of 33

produce discrimination against members of the protected class.” League of Women
Voters of N.C. v. North Carolina, 769 F.3d 224, 240 (4th Cir. 2014) (internal
quotations omitted); see also Veasey v. Abbott, 830 F.3d 216, 244 (5th Cir. 2016).
A
The record of this case demonstrates that the challenged practice of refusing
to count votes cast out-of-precinct for races in which the voters were eligible to
participate caused minority voters “less opportunity than other members of the
electorate to participate in the political process and to elect representatives of their
choice.” League of Women Voters, 769 F.3d at 240.2 Unlike the companion
appeal, which involved a situation in which statistical evidence was not possible to
obtain, this case involves hard, statistically significant proof of discriminatory
effect.
The district court acknowledged this proof, and credited it for the purposes
of its analysis. However, it deemed the evidence insufficient at stage one of the
Voting Rights Act analysis because, in the district court’s view, the rejected out-of-

2

As the majority properly notes, the district court assumed, for the purposes
of the motion, that expert evidence tendered by the plaintiffs was sufficient to show
a cognizable disparate burden under the Voting Rights Act. The majority stated it
had “grave doubts” as to this conclusion, but accepted, without deciding, that the
plaintiffs carried their burden of proof. Because the first part of the § 2 analysis
greatly informs the second part of the examination, it is necessary for me to discuss
the proof and the district court’s opinion in some detail.
-12-

Case: 16-16865, 11/02/2016, ID: 10184244, DktEntry: 33-2, Page 13 of 33

precinct votes did not constitute a significant portion of the total votes cast. In the
district court’s view, the plaintiffs were required to prove that the rejection of outof-precinct ballots cast by minorities “meaningfully reduces the likelihood that
minority as compared to white voters will cast ballots that are ultimately counted.”
No other court in the nation has imposed such a requirement. There is no
support for the district’s court new requirement either in the text of the Voting
Rights Act or in any case construing it. The standard at stage one is simply that the
“the challenged standard, practice, or procedure must impose a discriminatory
burden on members of a protected class, meaning that members of the protected
class have less opportunity than other members of the electorate to participate in
the political process and to elect representatives of their choice.” League of
Women Voters, 769 F.3d at 240. The key phrases here are “burden” and “less
opportunity.” The standard does not include “meaningful” or significant overall
electoral impact.
A Voting Rights Act plaintiff need not show that the challenged voting
practice caused a disparate impact by itself. Farrakhan v. Washington, 338 F.3d
1009, 1018–19 (9th Cir. 2003). Nor must the challenged practice make voting
impossible or cause significant electoral impact. It suffices for a violation of the
Voting Rights Act that the practice simply makes voting more burdensome.

-13-

Case: 16-16865, 11/02/2016, ID: 10184244, DktEntry: 33-2, Page 14 of 33

Thornburg v. Gingles, 478 U.S. 30, 35–36, 44, 47 (1986). And, as the Fourth
Circuit succinctly put it, “what matters for purposes of Section 2 is not how many
minority voters are being denied equal electoral opportunities, but simply that
‘any’ minority is being denied equal electoral opportunities.” League of Women
Voters, 769 F.3d at 244. As Justice Scalia put it:
If, for example, a county permitted voter registration for only three
hours one day a week, and that made it more difficult for blacks to
register than whites, blacks would have less opportunity “to
participate in the political process” than whites, and § 2 would
therefore be violated—even if the number of potential black voters
was so small that they would on no hypothesis be able to elect their
own candidate,
Chisom, 501 U.S. at 408 (Scalia, J., dissenting) (emphasis in original); see also id.
at 397 (Maj. Op.).
Put another way, the district court and the State are arguing that the minority
voters’ claims must fail because their votes really didn’t matter to the electoral
outcome. That proposition is contrary to the entire theory of the Voting Rights
Act. As the Supreme Court has observed: “No right is more precious in a free
country than that of having a voice in the election of those who make the laws
under which, as good citizens, we must live.” Wesberry v. Sanders, 376 U.S. 1, 17
(1964).

-14-

Case: 16-16865, 11/02/2016, ID: 10184244, DktEntry: 33-2, Page 15 of 33

Black votes matter. Hispanic votes matter. Native American votes matter.
White votes matter. All votes matter.
And the district court and the State are wrong to assume that, because the
improperly rejected votes constituted a relatively small portion of the total votes
cast, those votes really didn’t matter. Arizona has had a long history of very close
elections. Indeed, earlier this year, in the Fifth Congressional District Republican
primary, Andy Briggs defeated Christine Jones by 27 votes.
And the list goes on. In the 2014 Second Congressional District, Martha
McSally defeated Ron Barber by 167 votes. The 2012 Democratic primary in the
Fourth Congressional District between Johnnie Robinson and Mikel Weisser was
decided by 19 votes. Proposition 112, a 2010 statewide ballot measure seeking to
shorten the filing deadline for initiative petitions, lost by 194 votes. The 2002
Arizona Gubernatorial election was decided by slightly over 10,000 votes. In the
1994 Democratic primary for U.S. Senate, then-Congressman Sam Coppersmith
edged then-Arizona Secretary of State Dick Mahoney by 59 votes. A 1992
Republican legislative primary election between Richard Kyle and John Gaylord
resulted in a tie, with the winner being decided in a hand of poker, which Kyle won
by drawing a pair of sevens over Gaylord’s failed heart flush.
Indeed, the very first Arizona gubernatorial election in 1912 was very close,

-15-

Case: 16-16865, 11/02/2016, ID: 10184244, DktEntry: 33-2, Page 16 of 33

and the second race for Governor in 1916 was decided by 30 votes. In sum, close
elections have long been part of the fabric of Arizona politics. So votes, in fact, do
matter in Arizona, and disenfranchisement of any segment of voters can have an
effect on the outcome of an Arizona election.
The district court’s imposition of a “meaningful electoral effect”
requirement constituted legal error. For the reasons I have previously discussed,
the total number of votes affected is not the relevant inquiry; the proper test is
whether minority votes are burdened. If the right to vote in-person by minority
voters is burdened, it is not relevant that minorities may vote by absentee ballot. In
addition, it is, to say the least, ironic that the State and the district court would rely
on early absentee voting in this appeal, when in the companion case, both
dismissed absentee voting as a mere “convenience” because, in their view, inperson voting was the intended primary means for voters to cast their ballots.
Here, the plaintiffs established through statistically significant evidence that
the practice of not counting out-of-precinct ballots for races in which the voter was
qualified to vote afforded minority voters “less opportunity than other members of
the electorate to participate in the political process and to elect representatives of
their choice.” League of Women Voters, 769 F.3d at 240. Plaintiffs more than
satisfied their burden as to the first part of their § 2 Voting Rights Act claim.

-16-

Case: 16-16865, 11/02/2016, ID: 10184244, DktEntry: 33-2, Page 17 of 33

B
The second part of a § 2 claim requires a plaintiff to show that the burden on
voting imposed by the challenged practice is “in part [ ]caused by or linked to
social and historical conditions that have or currently produce discrimination
against members of the protected class.” League of Women Voters, 769 F.3d at
240; Veasey, 830 F.3d at 244.
As I discussed in my dissent to the companion appeal, the Supreme Court
has identified several factors to be taken into consideration, consistent with the
legislative history of the Voting Rights Act, namely:
(1) the extent of any history of official discrimination in the state or political
subdivision that touched the right of the members of the minority group to
register, to vote, or otherwise to participate in the democratic process;
(2) the extent to which voting in the elections of the state or political
subdivision is racially polarized;
(3) the extent to which the state or political subdivision has used unusually
large election districts, majority vote requirements, anti-single shot
provisions, or other voting practices or procedures that may enhance the
opportunity for discrimination against the minority group;
(4) if there is a candidate slating process, whether the members of the
minority group have been denied access to that process;
(5) the extent to which members of the minority group in the state or
political subdivision bear the effects of discrimination in such areas as
education, employment and health, which hinder their ability to participate
effectively in the political process;

-17-

Case: 16-16865, 11/02/2016, ID: 10184244, DktEntry: 33-2, Page 18 of 33

(6) whether political campaigns have been characterized by overt or subtle
racial appeals; and
(7) the extent to which members of the minority group have been
elected to public office in the jurisdiction.
Gingles, 478 U.S. at 37. In addition, the Court added that in some cases, there was
probative value in inquiring “whether there is a significant lack of responsiveness
on the part of elected officials to the particularized needs of the members of the
minority group” and “whether the policy underlying the state or political
subdivision’s use of such voting qualification, prerequisite to voting, or standard,
practice or procedure is tenuous.” Id. (citing S. Rep., at 28–29, U.S.Code Cong. &
Admin. News 1982, pp. 206–207).
Without repeating in detail the evidence tendered by the plaintiffs as to the
Gingles factors that I discussed in my prior dissent, Feldman v. Arizona Sec’y of
State, 2016 WL 6427146, at *29–31 (9th Cir. 2016), it is clear that they satisfied
their burden.
As to the first factor, the extent of any history of official discrimination in
the state or political subdivision that touched the right of the members of the
minority group to register, to vote, or otherwise to participate in the democratic
process, the plaintiffs produced expert testimony through Dr. David R. Berman of
Arizona State University. He detailed Arizona’s long history of imposing burdens

-18-

Case: 16-16865, 11/02/2016, ID: 10184244, DktEntry: 33-2, Page 19 of 33

on minority voters. In 1912, shortly after gaining statehood, Arizona imposed a
literacy test for voting. In Cochise and Pima Counties, the denial of the right to
vote meant that nearly half the precincts lacked enough voters to justify holding
primary elections in 1912. From 1912 to the early 1960s, election registrars
applied the literacy test to reduce the ability of African Americans, Native
Americans, and Hispanics to register to vote. In an action filed against Arizona to
enforce the Voting Rights Act, the United States Justice Department estimated that
73,000 people could not vote because of the existence of the literacy test.
The passage of the Voting Rights Act in 1965 caused the suspension of the
literacy test in Arizona, but the statute remained in effect until it was repealed in
1972, after Congress banned its use in 1970 through an amendment to the Voting
Rights Act. Arizona subsequently unsuccessfully challenged the Congressional
ban on literacy tests. Oregon v. Mitchell, 400 U.S. 112, 118 (1970). In Mitchell,
the Court noted that, in Arizona, only two counties out of eight with Hispanic
populations in excess of 15% showed voter registration equal to the state-wide
average. Id. at 132. In the 1960s, there were a number of initiatives to discourage
minority voting in Arizona, such as “Operation Eagle Eye.” Under Operation
Eagle Eye, minority voters were challenged at the pools on a variety of pretexts,

-19-

Case: 16-16865, 11/02/2016, ID: 10184244, DktEntry: 33-2, Page 20 of 33

with the goal of preventing minority voting or slowing down the process to create
long lines at the polls and discourage voting.
Native Americans in Arizona especially suffered from voting restrictions.
Although Native Americans were U.S. citizens, the Arizona Supreme Court held in
1928 that they could not vote because they were under federal guardianship.
Porter v. Hall, 271 P. 411, 419 (Ariz. 1928). Even after that ban was overruled in
1948 in Harrison v. Laveen, 196 P.2d 456 (Ariz. 1948), Native Americans faced
significant obstacles to voting. See generally, Patty Ferguson-Bohnee, The History
of Indian Voting Rights in Arizona: Overcoming Decades of Voter Suppression, 47
Ariz. St. L. J. 1099, 1112 (2015).
Because of its long history of imposing burdens on minority voting, Arizona
became one of nine states subject to the pre-clearance requirements of the Voting
Rights Act after it was amended in 1975 to protect language minorities. 40 Fed.
Reg. 43746. Under the pre-clearance provision, Arizona was required to obtain the
approval of the United States Department of Justice before implementing any law
affecting the voting rights and representations of minorities. Since 1982, the
Department of Justice has vetoed four statewide redistricting plans proposed by
Arizona that appeared to discriminate against minorities. As Dr. Berman testified:
“Arizona has a long history of discrimination against Native Americans, Hispanics

-20-

Case: 16-16865, 11/02/2016, ID: 10184244, DktEntry: 33-2, Page 21 of 33

and African Americans when it comes to their voting rights. This discrimination
has been reflected in legislation relating to voter requirements, election law and the
manner in which elections have been administered, efforts to intimidate voters, and
instances of racial appeals, both subtle and not so subtle during campaigns.” He
testified that “[l]ooking at the history of abuse and neglect, there is no reason to
assume that discrimination in regard to voting and election practices is a relic of
the past and that the protections provided by preclearance are not needed in
Arizona.”
As to the second factor, the extent to which voting in the elections of the
state or political subdivision is racially polarized, Dr. Allan Lichtman of American
University provided expert testimony detailing the history of polarized voting in
Arizona. Statistical analysis showed the sharp polarization between white and nonwhite voters. Indeed, the data showed that for every election studied, the
preferences of white and non-white voters diverged significantly.
For the reasons described in both the discussion of the first Gingles factor
and in stage one of the Voting Rights Act analysis, plaintiffs demonstrated a
likelihood of success as to Gingles factor three: Arizona has used voting practices
or procedures that may enhance the opportunity for discrimination against the
minority group.

-21-

Case: 16-16865, 11/02/2016, ID: 10184244, DktEntry: 33-2, Page 22 of 33

Because the voting access issues affect the right to vote for a candidate, the
fourth factor concerning the candidate slating process is not relevant, and the
plaintiffs’ expert conceded that there did not seem to be candidate slating by
political parties in recent Arizona history.
The fifth factor, which I shall discuss in more detail later, the extent to
which members of the minority group in the state or political subdivision bear the
effects of discrimination in such areas as education, employment and health, which
hinder their ability to participate effectively in the political process, falls decisively
in favor of the plaintiffs. The plaintiffs’ expert opined that “[t]he persistent effects
of discrimination are substantially demonstrated in the deficient
socio-economic position of Hispanic, Native American, and African American
people in Arizona.” The plaintiffs tendered significant evidence showing that
Arizona minorities suffered in education and employment opportunities, with
disparate poverty rates, depressed wages, higher levels of unemployment, lower
educational attainment, less access to transportation, residential transiency, and
poorer health.
The plaintiffs also provided substantial evidence as to the sixth factor,
namely, whether political campaigns have been characterized by overt or subtle
racial appeals.

-22-

Case: 16-16865, 11/02/2016, ID: 10184244, DktEntry: 33-2, Page 23 of 33

Finally, the plaintiffs provided evidence supporting the seventh Gingles
factor, namely, the extent to which members of the minority group have been
elected to public office in the jurisdiction. As of January 2016, Hispanics
constituted over 30% of the population, but held only 19% of the seats in the
Arizona legislature. African-Americans made up 4.7% of the population, but held
1% of the legislative seats. Native Americans fared slightly better, constituting
5.3% of the population and holding 4.4% of the legislative seats.
In sum, the plaintiffs tendered significant, and mostly uncontradicted
evidence, satisfying the Gingles factors at stage two. But, again, the Gingles
factors are not the end of the story. We are obligated to look to the “totality of the
circumstances.” 52 U.S.C. § 10301(b). When we do so, we can easily see how the
effects of discrimination hinder minority voters’ ability to cast ballots in person.
And in assessing the totality of the circumstances, we also must be mindful that the
Voting Rights Act does not require proof of intentional discrimination; indeed,
Congress specifically amended § 2 of the Voting Rights Act in 1982 to relieve
plaintiffs of the burden of proving discriminatory intent. Chisom, 501 U.S. at 394;
see also Ruiz v. City of Santa Maria, 160 F.3d 543, 557 (9th Cir. 1998) (noting
Congress’s statement that the “intent test” was “unnecessarily divisive in that it
involved charges of racism on the part of individual officials or entire communities

-23-

Case: 16-16865, 11/02/2016, ID: 10184244, DktEntry: 33-2, Page 24 of 33

[and] placed an inordinately difficult burden of proof on plaintiffs and [ ] asked the
wrong question” (internal alterations and quotation marks omitted)). Rather,
courts must consider how the challenged practice “interacts with social and
historical conditions to cause an inequality in the opportunities enjoyed by black
and white voters to elect their preferred representatives.” Gingles, 478 U.S. at 47.
And proof of direct causation is not required; it suffices that the challenged
practice be “linked” “in part” to social and historic conditions. League of Women
Voters, 769 F.3d at 240; Veasey, 830 F.3d at 244.
As to the issues raised in this appeal, the fifth Gingles factor is especially
relevant, namely “the extent to which members of the minority group in the state or
political subdivision bear the effects of discrimination in such areas as education,
employment and health, which hinder their ability to participate effectively in the
political process.” 478 U.S. at 37. The district court recognized that the plaintiffs
had established this factor, stating that it “did not discount Arizona’s history of
racial discrimination or the lingering effects on the socio-economic status of
minorities.”
The majority seems to require, and the district court implied, that the fifth
Gingles factor requires proof that intentional discrimination caused minorities
adverse socioeconomic effects. However, that is not the proof required. Our

-24-

Case: 16-16865, 11/02/2016, ID: 10184244, DktEntry: 33-2, Page 25 of 33

Circuit has considered the factor satisfied when there was a history of
discrimination and lowered socioeconomic status. See, e.g., United States v.
Blaine County, Montana, 363 F.3d 897, 914 (9th Circ. 2004) (holding factor
satisfied when there was evidence of lowered minority socioeconomic status and
historical evidence of discrimination); Old Person v. Cooney, 230 F.3d 1113, 1129
(9th Cir. 2000) (factor satisfied with showing that American Indians have a lower
socio-economic status than whites in Montana and that social and economic factors
hinder the ability of American Indians in Montana to participate fully in the
political process); see also League of Women Voters, 760 F.3d at 235 (employing a
similar analysis). Thus, any contrary conclusion is legally erroneous.
The plaintiffs tendered significant evidence showing that Arizona minorities
suffered in education and employment opportunities, with disparate poverty rates,
depressed wages, higher levels of unemployment, lower educational attainment,
less access to transportation, more residential transiency, and poorer health. The
district court seemed to doubt that these factors were related to lower out-ofprecinct voting by minorities, but the record speaks otherwise. These factors
directly contribute to the statistically significant disparity in out-of-precinct voting
by minorities as compared to whites. Indeed, these considerations go to the heart

-25-

Case: 16-16865, 11/02/2016, ID: 10184244, DktEntry: 33-2, Page 26 of 33

of why Arizona’s refusal to count legitimate out-of-precinct votes most severely
affects Arizona’s minority voters.
For instance, minority voters often cannot afford home ownership and they
have higher rates of residential mobility than white voters. Because of this, and
given the “dizzying array” of changes in polling location, minority voters are more
likely to vote in the wrong precinct. Indeed, African American and Hispanic
voters are substantially more affected by polling place changes than white voters.
In particular, the impact of precinct consolidation, while statistically significant for
all groups, is more than twice as large for Hispanics and African Americans as for
non-Hispanic whites.
Data also indicates that significant numbers of Hispanic and African
American voters in Phoenix do not have access to an automobile. Reliance on
public transportation disparately burdens minority voters in several ways. Among
voters who are transported to the incorrect polling place, minority voters have less
opportunity to travel to the correct polling place. Travel distances also vary
significantly between white and minority voters. The data shows that minority
voters have to travel much farther than white voters to get to assigned polling
places. Hispanics and Native Americans are more likely than whites to live further
from their assigned polling places, and Hispanics are more likely to live in

-26-

Case: 16-16865, 11/02/2016, ID: 10184244, DktEntry: 33-2, Page 27 of 33

proximity to multiple polling places to which they are not assigned. One
Congressman serving a district that was 65% Hispanic testified that it took an hour
and a half for his constituents to get to the polls via public transportation. And
many voting centers are not located near public transportation lines. These are
factors considered significant by the Fourth Circuit, in its consideration of out-ofprecinct voting in League of Women Voters, 769 F.3d at 233–34.
In addition, in Maricopa County, there are significantly fewer polling places
in Hispanic areas, using population density as a metric, than in predominantly
white neighborhoods. Election day issues in these consolidated polling places,
such as long delays in access to voting, disproportionately affect low-income
minority voters, who typically have very little flexibility in their work day, and
must vote during a narrow window of time in the morning or evening. In his study
of the 2008 and 2012 elections, Dr. Lichtman concluded that minorities were 61%
more likely than whites to experience waiting times of 31 minutes or more. The
difference was statistically significant.
Language barriers also pose significant hindrances to minority voters who
are not fluent in English. The Plaintiffs tendered evidence that voters in Spanishspeaking areas in Maricopa County received mistranslated or incorrect information
from election offices, creating confusion for voters who are not fluent in English.

-27-

Case: 16-16865, 11/02/2016, ID: 10184244, DktEntry: 33-2, Page 28 of 33

In the 2012 election, Maricopa County sent Spanish-speaking documents with the
wrong election date to Hispanic voters. The English version contained the correct
information. In a special election this year, over 1.3 million Spanish-speaking
households received a ballot with erroneous descriptions of ballot initiatives.
There is an additional consequence of all these hindrances to voting:
suppression of voter turnout. The plaintiffs’ expert used the term “calculus of
voting,” to describe the overall effect on voter turnout when the barriers to voting
exceed the benefits. He noted that “recent research has demonstrated that changes
in polling locations associated with precinct consolidation have a substantial effect
on turnout.” Data showed that in Maricopa and Pima Counties, such changes were
far more likely to affect minority voters as compared with white voters. Arizona
has the second worst turnout of African American voters in the nation. Turnout
among African Americans in Arizona’s 2012 Presidential election was 46%; the
national average was 66%. The turnout of Hispanic voters in Arizona was 39%,
compared with a 62% turnout of white voters.
In summary, these historic and socio-economic Gingle factors are
significantly associated with the statistically significant difference in white versus
minority out-of-precinct voting. At stage two, the plaintiffs were required only to
show that, under the totality of the circumstances, the discriminatory impact of the

-28-

Case: 16-16865, 11/02/2016, ID: 10184244, DktEntry: 33-2, Page 29 of 33

challenged practice was linked to historic and social factors. They satisfied this
burden.
C
Given the plaintiffs’ uncontested proof of the undue burden imposed on outof-precinct minority voters who were eligible to vote in some races, and the proof
of association between this burden and discriminatory historical and
socioeconomic factors, the district court eIIed in denying a preliminary injunction.
I rr
The district court also erred in its analysis of the plaintiffs’ Fourteenth
Amendment claims. Under Burdick v. Takushi, 504 U.S. 428 (1992) and
Anderson v. Celebrezze, 460 U.S. 780 (1983), we must weigh the nature and
magnitude of the burden imposed by the law against the state’s interest and
justification for it. Nader v. Brewer, 531 F.3d 1028, 1034 (9th Cir. 2008).
The practices challenged under Anderson-Burdick are evaluated from the
vantage point of the burdened voters. See Crawford v. Marion Cty. Election Bd.,
553 U.S. 181, 186 (2008). The burden in this case is disenfranchisement. It is
uncontested that Arizona has disenfranchised a significant number of minority
voters, some erroneously, by virtue of its prohibition on out-of-precinct voting,
even when legitimate ballots were cast. Their legitimate votes were not counted.

-29-

Case: 16-16865, 11/02/2016, ID: 10184244, DktEntry: 33-2, Page 30 of 33

Restrictions on fundamental voting rights cause irreparable injury. League
of Women Voters, 769 F.3d at 247. “[O]nce the election occurs, there can be no
do-over and no redress.” Id. Thus, “the injury to these voters is real and
completely irreparable.” Id.
The State’s justification for the challenged practice is not specific to the
challenged practice. The State cites “significant and numerous” advantages
attending a precinct-based voting system. The purported advantages of such a
system—such as capping the number of voters attempting to vote in one place on
election day and putting polling places closer to voter residences—are belied by
the voters’ experiences. But more importantly, these justifications speak only to
Arizona’s choice to use a precinct-based system; they do not justify Arizona’s
choice to discount ballots cast out of precinct, even when the ballots contain votes
the voters were eligible to cast.
Under the proper “balancing and means-end fit framework,” we must“tak[e]
into consideration the extent to which [the state’s] interests make it necessary to
burden the plaintiff’s rights.” Pub. Integrity All., Inc. v. City of Tucson, 2016 WL
4578366, at *3 (9th Cir. 2016) (internal quotations omitted). The State’s
articulated interest here is administrative efficiency. The State argues that it will
take up to 15 minutes to process legitimate out-of-precinct votes. But in Maricopa

-30-

Case: 16-16865, 11/02/2016, ID: 10184244, DktEntry: 33-2, Page 31 of 33

County, voters were waiting for between 4 to 6 hours to cast their ballots. The
evidence showed that many voters who ended up voting in the wrong precinct
traveled there using public transportation and may have had to take time off work.
Spending a few minutes of administrative time to permit these citizens’ votes to be
counted pales in comparison with the sacrifice made by these voters in pursuit of
the exercise of their franchise.
Perhaps more importantly, the requested relief does not involve altering preelection or election day procedures. Voters still must vote in their precincts if their
votes are to be counted as to precinct-specific contests. If a voter casts a ballot in
the wrong precinct, it would still be treated as a provisional ballot. The only
difference would be that the out-of-precinct vote would be counted as to those
elections in which the voter was eligible to vote regardless of precinct.
In addition, when one analyzes how provisional ballots are treated, the
burdens are relatively low for the State. It already manually examines the
provisional ballots and manually compares addresses. If the ballot is cast in the
correct precinct, it is counted. If not, the only additional burden that would be
imposed would be to count the votes for the race for which the voter is qualified
and eligible to vote. Arizona law provides the State ten days to count provisional
ballots. The State is already using manual procedures as to write-in and damaged

-31-

Case: 16-16865, 11/02/2016, ID: 10184244, DktEntry: 33-2, Page 32 of 33

ballots. These administrative burdens should not be discounted, but in comparison
to the hardships faced by minority voters on election day, the scales weigh in favor
of the voters.
The State’s interest in administrative efficiency simply does not justify the
means employed: disenfranchisement of out of precinct voters. The plaintiffs were
entitled to a preliminary injunction on their Fourteenth Amendment claim.
To be sure, courts must exercise great caution in deciding election
challenges close to election day. Purcell v. Gonzalez, 549 U.S. 1, 4-5 (2006). But
the relief sought by the plaintiffs does not affect polling procedure; the plaintiffs
simply seek to have legitimately cast ballots, which have already been properly
collected, counted. And the lynchpin of the district court’s analysis, and relied
upon by the majority, is that there are not a substantial number of ballots at issue
relative to the entire number of votes cast.
IV
The district court should have granted a preliminary injunction. The
plaintiffs showed a statistically significant relationship between Arizona’s practice
of declining to count legitimate out-of-precinct votes and a disparate burden on the
franchise of minority voters. The district court erred as a matter of law in requiring
that the practice have a “meaningful” impact on election results. The plaintiffs

-32-

Case: 16-16865, 11/02/2016, ID: 10184244, DktEntry: 33-2, Page 33 of 33

established a likelihood of success on both their Voting Rights Act and
Constitutional claims. Their ballots should be counted in all races in which they
are eligible to vote.
For these reasons, I respectfully dissent.

-33-