NO. 16-1161

In the Supreme Court of the United States
BEVERLY R. GILL, et al.,
Appellants,

v.

WILLIAM WHITFORD, et al.,
Appellees.
On Appeal from the United States
District Court for the Western District of Wisconsin

BRIEF FOR AMICUS CURIAE
CENTER FOR MEDIA AND DEMOCRACY
IN SUPPORT OF APPELLEES
THEODORE R. BOEHM
HOOVER HULL TURNER

111 Monument Circle
Suite 4400
Indianapolis, IN 46204
(317) 822-4400
tboehm@hooverhullturner.com

JOSEPH H. YEAGER, JR.
Counsel of Record

HARMONY A. MAPPES
DULANY LUCETTA POPE
MATTHEW B. HARRIS
JEFFREY P. JUSTMAN
JULIANA YANEZ
FAEGRE BAKER DANIELS LLP

300 North Meridian Street
Suite 2700
Indianapolis, IN 46204
(317) 237-0300
jay.yeager@faegrebd.com
Counsel for Amicus Curiae
Becker Gallagher · Cincinnati, OH · Washington, D.C. · 800.890.5001

 i
TABLE OF CONTENTS
TABLE OF AUTHORITIES . . . . . . . . . . . . . . . . . . iii
STATEMENT OF INTEREST . . . . . . . . . . . . . . . . . 1
SUMMARY OF ARGUMENT . . . . . . . . . . . . . . . . . . 2
ARGUMENT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
I.

Wisconsin’s Pre-Act 43 Redistricting
Problems Sprang From Divided Government,
not Litigation, and the 2011 Process of
Enacting Act 43 Was Anything but a
Triumph of the Policy-Driven Process the
Legislature Describes . . . . . . . . . . . . . . . . . . . 7

II.

The Record Strongly Supports the District
Court’s Finding—and Act 43’s Guiding
Principle—that Past Party Preferences
Reliably Predict the Future Statewide
Behavior of Wisconsin’s Voters . . . . . . . . . . 14
A. Actions Speak Louder than Words: the
Legislature Has Changed its Tune Since
the 2011 Redistricting Process . . . . . . . . 15
B. The District Court Relied on Extensive
Data Proving that Voter Party Preference
Reliably Predicts Assembly Vote in the
Vast Majority of Districts . . . . . . . . . . . . 17
C. The Legislature’s Hand-Picked Examples
Prove Nothing . . . . . . . . . . . . . . . . . . . . . 21

III.

Voters Have Representational Interests and
Rights that Extend Far Beyond Constituent
Services . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24

 ii
IV.

The Authors of Act 43 Did not Assume Away
Party Volatility)They Planned for It . . . . . . 28

CONCLUSION . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32

 iii
TABLE OF AUTHORITIES
CASES
Ariz. Legislature v. Ariz. Indep. Redistricting
Comm’n, 135 S. Ct. 2652 (2015) . . . . . . . . . 10, 12
Arrington v. Elections Bd.,
173 F.Supp.2d 856 (E.D. Wis. 2001) . . . . . . . . . . 7
Baker v. Carr,
369 U.S. 186 (1962) . . . . . . . . . . . . . . . . . . . . . . 10
Baldus v. Members of Wis. Gov’t Accountability Bd.,
849 F.Supp.2d 840 (E.D. Wis. 2012) . . . . . . 10, 15
Baumgart v. Wendelberger,
No. 01-C-0121, 02-C-0366, 2002 WL 34127471
(E.D. Wis. May 30, 2002) (per curiam),
amended by 2002 WL 34127473
(E.D. Wis. July 11, 2002) . . . . . . . . . . . . . . . . . . . 7
Davis v. Bandemer,
478 U.S. 109 (1986) . . . . . . . . . . . . . . . . . . . 11, 27
Gaffney v. Cummings,
412 U.S. 735 (1973) . . . . . . . . . . . . . . . . . . . . . . 11
Peterson v. Borst,
786 N.E.2d 668 (Ind. 2003) . . . . . . . . . . . . . . . . . 8
Prosser v. Elections Bd.,
793 F.Supp. 859 (W.D. Wis. 1992) . . . . . . . . . . . 7
Reynolds v. Sims,
377 U.S. 533 (1964) . . . . . . . . . . . . . . . . . . . . 6, 10
Shaw v. Reno,
509 U.S. 630 (1993) . . . . . . . . . . . . . . . . . . . 25, 26

 iv
Vieth v. Jubelirer,
541 U.S. 267 (2004) . . . . . . . . . . . . . . . . 11, 27, 29
Wis. State AFL-CIO v. Elections Bd.,
543 F.Supp. 630 (E.D. Wis. 1982) . . . . . . . . . . . . 7
STATUTES
Wis. Stat. §§ 19.31-19.39 . . . . . . . . . . . . . . . . . . . . . . 9
Wis. Stat. §§ 19.81-19.98 . . . . . . . . . . . . . . . . . . . . . . 9
OTHER AUTHORITIES
Jowei Chen & Jonathon Rodden, Cutting Through
the Thicket: Redistricting Simulations and the
Detection of Partisan Gerrymanders, 14 Election
L.J. 331 (2015) . . . . . . . . . . . . . . . . . . . . . . . . . 8, 9
A. Scott Chinn, The Role of Indiana’s State and
Federal Courts in Legislative Redistricting, 37
Ind. L. Rev. 643 (2004) . . . . . . . . . . . . . . . . . . . . . 8
Tim Cullen & Dale Schultz, We led the Wisconsin
Senate. Now we’re fighting gerrymandering in
our state., Wash. Post, June 20, 2017 . . . . . . . . 25
Jeffmd, Daily Kos Elections’ statewide election
results by congressional and legislative districts,
Daily Kos, July 9, 2013, https://www.dailykos.
com/stories/2013/7/9/1220127/-Daily-Kos-Elec
tions-2012-election-results-by-congressionaland-legislative-districts . . . . . . . . . . . . . . . . . . . 31
Daniel R. Roy, How Political Turf Battle Morphed
Into Legal Contest Before State's Highest Court,
46 Res Gestae, June 2003 . . . . . . . . . . . . . . . . . . 8

 v
Boris Shor & Nolan McCarty, Ideological Mapping
of American Legislatures,105 Am. Pol. Sci. Rev.
530 (2011) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
Reference Manual on Scientific Evidence: Third
(Statistics) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
Wisconsin Election Results, Wisconsin Elections
Commission, http://elections.wi.gov/electionsvoting/results
(last visited Sept. 1, 2017) . . . . . . . . . . . 13, 19, 31
Wisconsin State Legislature Open GIS
Data, Legislative Technology Services
Bureau, http://data-ltsb.opendata.arcgis.com/dat
asets/2012-2020-wi-election-data-with-2011wards/data?selectedAttribute=PREREP12 (last
visited Sept. 1, 2017) . . . . . . . . . . . . . . . . . . . . . 13

 1
STATEMENT OF INTEREST1
The Center for Media and Democracy (the “Center”)
is a nonprofit, nonpartisan organization based in
Madison, Wisconsin. Founded in 1993, it has focused
on investigating and exposing abuses of power that
corrupt the proper workings of democratic government.
Its particular interest in this case is to expose incorrect
claims as to the legislative history of gerrymandering
in Wisconsin and distorted reports of Wisconsin
election results and to supply insight into the current
state of Wisconsin politics. The Center offers this brief
to respond specifically to points in Sections I.A and
II.B, C and D in the Brief for the Wisconsin State
Senate and Wisconsin State Assembly (the
“Legislature”) that are within the Center’s experience.
The Center also wishes to point out the damage a
gerrymander inflicts on others in addition to the
targeted minority party, in this case Wisconsin
Democrats. The credibility of the Center’s work is
established by its more than two decade history of
citations in the New York Times, Washington Post, the
Los Angeles Times, POLITICO, the Guardian, the New
Yorker, Bloomberg, WIRED, Vice and The Atlantic as
well as on CNN, MSNBC, NBC, CBS, PBS and NPR.

1

Counsel for all parties have consented to this filing. No counsel
for any party authored this brief in whole or in part, and no entity
or person, aside from Amicus Curiae, its members and counsel
have made any monetary contribution towards the preparation or
submission of this brief.

 2
SUMMARY OF ARGUMENT
The Legislature2 describes a fair, deliberative,
policy-driven legislative process that bears no relation
to the actual process by which Act 43 became the law
of Wisconsin. The short answer to many of the
Legislature’s claims is that the only policy driving Act
43 was preservation of one-party domination of the
Wisconsin Assembly and the incumbency of the
legislators who voted for it.3 That was the evidence,
and the Appellants have not contested the district
court’s finding of intent.
The Legislature argues that the power of
incumbency, differences among individual candidates,
and other factors drive election results. But actions
speak louder than words. In drafting Act 43, the
Legislature wholly ignored the factors it now applauds.
Instead it used reliable statistical analyses of historical
data to project the durability of the Republican
legislative majority under the Act 43 map against
various possible shifts in relative party strength.
Indeed, the evidence shows that party affiliation
explains well over 95% of a voter’s decision on
Assembly candidates. The Legislature now asks the
Court to ignore that core fact: in designing its
gerrymander, the Legislature embraced the very
principles and analysis it now argues against.
The legislators who supported Act 43 preserved
their own seats and positions in the legislative
2
3

The Wisconsin Legislature’s amicus brief is cited as “Br.”

The plaintiffs challenged only the Assembly map, but because
the Senate districts are nested within the Assembly districts, this
decision necessarily impacts the Senate map as well.

 3
hierarchy and uniquely benefited from it. This
gerrymander is thus the product of a strain of conflict
of interest that infects few other forms of legislation.
This gerrymander is immune from the checks and
balances that operate on normal legislation. In
Wisconsin, as in about half the states in the Union,
there is no voter initiative to correct this abuse of
power by enacting a redistricting procedure without the
Assembly’s approval. Without court intervention, this
gerrymander will survive despite its partisan origin
and anti-democratic impact.
Prior court intervention over the past three decades
in Wisconsin arose from the Legislature and the
Executive’s inability to agree on maps, not from any
inherent difficulty in drawing districts. The remaining
complexities in districting that the Legislature cites
arise from compliance with the Voting Rights Act and
the recognized constitutional prohibition on racial
gerrymandering. These will govern any redistricting
regardless of the outcome of this case. The Appellees
(“Wisconsin Voters”) proved and the Appellants have
not contested that computers are fully capable of
drawing maps without political data that still respect
traditional districting principles.
Modern technology offers a uniquely reliable
methodology for building a fortress around a transient
majority’s control through intentional maximization of
the number of districts deemed safe for one party. In
Wisconsin, as in most states, a map not dominated by
partisan animus will still produce some districts that
heavily favor one of the two major parties. But in Act
43 the State intentionally maximized the number of
safe districts, cutting meaningfully competitive

 4
districts in half in the process. The Legislature does
not address the basic data that show the durability of
a partisan majority riding the back of a well-drawn
gerrymander. And Wisconsin as a result has one of the
most partisan legislatures in the nation. Nor does the
Legislature offer an answer to the most obvious
question: If the authors of Act 43 did not believe in
analyzing past election results to predict future results
and maximize the likelihood of an impenetrable
Republican majority, then why did they spend nearly
half a million dollars of public money on sophisticated
computer analysis by outside consultants to do just
that?
In conferring this personal benefit on themselves,
the legislators who enacted Act 43 imposed a high cost
on the millions who must live under the Act 43 maps.
A gerrymander intentionally invades the
representative rights of millions of citizens.
A
legislature with few competitive races imposes multiple
barriers to the proper functioning of democratic
government. Candidates in both parties are obliged to
appeal not to the electorate as a whole, but only to a
majority of their party. The result is a legislative body
unrepresentative of the voters as a whole and resistant
to compromise. And the elimination of meaningful
races in the general election is damaging to the body
politic, producing large numbers of unopposed
candidates, fewer competitive races, increased voter
apathy, lower turnout, and suspicion of government.
There is of course no Constitutional right to
proportional representation, and nothing in this case
would change that. But individuals and political
parties have a constitutional right not to be targeted by

 5
the state for partisan reasons. That is what happened
to the Wisconsin Democratic party and to voters in
districts purposely manipulated to pack minority party
voters into supermajority districts or increase majority
party voters in other districts to levels deemed safe for
the majority.
The handful of races the Legislature offers as
examples of competitiveness are outliers among the
297 races in Wisconsin’s 99 Assembly districts in the
three election years since 2012. In fact only about
20)not the 75 the Legislature says)of those 297 races
were actually competitive. To be sure, an occasional
candidate will succeed in a competitive or even slightly
adverse district.
But the overwhelming and
uncontested data of record show these anomalies are
few and far between and extremely unlikely to
overcome the intended statewide partisan tilt of a
sophisticated computer-driven gerrymander such as
Act 43.
Wisconsin has seen litigation break out after each
census since 1980, but only because the government
was divided and could not agree on any map, forcing
courts to act. Establishment of an objective standard
for impermissible manipulation, rebuttable by a state’s
ability to justify its action as required by legitimate
districting considerations or otherwise, should over
time reduce the need for court intervention. And, of
course, as the district court noted, a state could
eliminate any presumption of impermissible motive
simply by employing a neutral process.
In addition to impairing the proper function of the
legislative branch, a gerrymander intentionally invades
the representative rights of millions of citizens. A

 6
gerrymander is intended to, and does, place the
election of the legislator for each district in the hands
of the voters in the primary of the party with a
majority in that district.
It thus purposefully
maximizes the number of voters who have no
meaningful voice in the selection of their legislators
and facilitates candidates from the far right and left.
Not only Democrats are deprived of a vote in a majority
of districts. In many states voting in a primary
requires a public declaration of affiliation with that
party. Independents and third party supporters who
are unwilling to do that have no vote in the only
contest that matters in the selection of their legislator.
They suffer the ultimate impairment of
representational rights. Thus, though Democrats are
the target of the Wisconsin gerrymander, Act 43 inflicts
collateral damage on many voters of both parties and
of no party.
ARGUMENT
The Wisconsin Legislature is systematically
depriving Wisconsin’s voters of the “fair and effective
representation for all citizens [which] is concededly the
basic aim of legislative apportionment.” Reynolds v.
Sims, 377 U.S. 533, 565-66 (1964). It has been the law
of the land for over 50 years that the Constitution
“guarantees the opportunity for equal participation by
all voters in the election of state legislators. Diluting
the weight of votes because of place of residence
impairs basic constitutional rights. . . .” Id. at 566. The
Wisconsin Voters proved below that the Legislature is
doing precisely that and worse: discriminating based
on voters’ own political views and associations. This

 7
tramples not only the Fourteenth Amendment but the
First Amendment as well.
The Center submits this brief because Wisconsin’s
voters need this Court’s help. Given the inherent
conflicts of interest and human nature, and given the
structure of Wisconsin government, this problem will
not fix itself.
I.

Wisconsin’s Pre-Act 43 Redistricting
P r o b l e m s Sp r a n g F r o m D i v i d e d
Government, not Litigation, and the 2011
Process of Enacting Act 43 Was Anything
but a Triumph of the Policy-Driven Process
the Legislature Describes.

The Legislature cites repeated litigation over
Wisconsin legislative maps since 1972 to warn that
partisan gerrymandering claims would invite
additional federal “intrusions into the legislative
process.” Br. 7-8.4 But in the three decades leading up
to Act 43, federal courts were not drawn into the
4

In the attempt following the 1980 census, the Republican
governor vetoed the Democrat-controlled legislature’s map, leading
to court intervention. See Wis. State AFL–CIO v. Elections Bd.,
543 F.Supp. 630, 631 (E.D. Wis. 1982). The same thing happened
following the 1990 census. Prosser v. Elections Bd., 793 F.Supp.
859, 861-62 (W.D. Wis.1992) (“Both houses of the Wisconsin
legislature have a Democratic majority, but not a large enough one
to override vetoes by the state’s Republican governor.”). And
following the 2000 census, the divided state legislature (Democrat
controlled Senate and Republican controlled Assembly) failed to
produce a map, again resulting in a court-drawn map. See
Arrington v. Elections Bd., 173 F.Supp.2d 856, 858-59 (E.D. Wis.
2001); Baumgart v. Wendelberger, No. 01-C-0121, 02-C-0366, 2002
WL 34127471, at *1 (E.D. Wis. May 30, 2002) (per curiam),
amended by 2002 WL 34127473 (E.D. Wis. July 11, 2002).

 8
mapmaking process by any tide of plaintiff-driven
litigation. Instead, divided state government failed to
agree on any map that met population standards
required by the most recent census. See id. Court
intervention was required then; and intervention will
be required in the future if the Legislature and
Governor cannot produce a proper map. None of this
implicates partisan gerrymandering claims.
The Legislature also complains that limiting
partisan gerrymanders will impose great burdens on
the Legislature. Yet with relative ease, a single
political scientist, expert witness Kenneth Mayer,
Professor of Political Science at the University of
Wisconsin, was able to draw a map adhering better to
traditional redistricting criteria than the Legislature’s
map and that did not discriminate against voters based
on political preference. Courts find the job similarly
manageable. See, e.g., Peterson v. Borst, 786 N.E.2d
668, 677-78 (Ind. 2003) (using software supplied by
parties without political data, Indiana Supreme Court
staff drew city-county council districts with minimal
population variation, regular shape, and respecting
internal lines within a few hours, resulting in a map
widely viewed as fair); A. Scott Chinn, The Role of
Indiana’s State and Federal Courts in Legislative
Redistricting, 1962-2003, 37 Ind. L. Rev. 643, 657-59
(2004); Daniel R. Roy, How Political Turf Battle
Morphed Into Legal Contest Before State’s Highest
Court, 46 Res Gestae, June 2003 at 20, 22. Indeed,
analysts can now create through computer algorithms
hundreds or even thousands of maps that comply with
traditional criteria. See, e.g., Jowei Chen & Jonathon
Rodden, Cutting Through the Thicket: Redistricting

 9
Simulations and the Detection of Partisan
Gerrymanders, 14 Election L.J. 331, 338 (2015).
Neither do partisan gerrymandering claims
threaten the Legislature’s discretion to make policy.
The Wisconsin process that produced Act 43 in
particular deserves little deference to the policymaking prerogatives normally accorded the legislative
branch. Remarkably, given the record before this
Court, the Legislature portrays that process as a
breakthrough of good government. Br. 8. The
Legislature describes “months of full-time work by
legislative aides meeting with caucus members and
drafting maps that complied with traditional criteria.”
Id.
But the legislative process that produced Act 43 was
not conducted in the usual sunshine of public
introduction of a bill, hearings, and amendments. The
press, public, and even most legislators first saw Act 43
on the eve of its passage. Individual Republicans, but
not Democrats, were permitted to view their new
districts at a secret location outside the Capitol
building. Even those Republicans were not given
access to the map as a whole and were sworn to secrecy
contrary at least to the spirit of Wisconsin’s strong
open meetings and records laws.
Jurisdictional
5
Statement Appendix 12a; Joint Appendix6 355; Wis.
Stat. §§ 19.31-19.39; 19.81-19.98. Due to the inherent
complexity of any written districting bill, which
identifies each district as a list of dozens of wards, it
was impossible for anyone newly introduced to Act 43
5

The Jurisdictional Statement Appendix is cited as “JSA.”

6

The Joint Appendix is cited as “JA.”

 10
to have a meaningful clue as to the boundaries of each
of these new districts before it was enacted. The entire
public process from introduction to approval by both
houses of the Legislature consumed only ten calendar
days. JSA 29a. See also Baldus v. Members of Wis.
Gov’t Accountability Bd., 849 F.Supp.2d 840, 845-46
(E.D. Wis. 2012) (describing secretive and hurried
process; “every effort was made to keep this work out
of the public eye and, most particularly, out of the eye
of the Democrats”).
In short, the Legislature’s claim that Act 43 is the
product of public-spirited deliberation and wise policy
choices is preposterous.
The Legislature’s claim that it exercised
“constitutionally conferred policy-making discretion”
assumes, wrongly, that Act 43 complies with the
Constitution of the United States.
Br. 6.
Gerrymandering is qualitatively different from other
legislation, and the key reasons for deference to the
policy calls of the legislative branch carry little weight
when it comes to redistricting, the primary motive of
which is partisan advantage.
First, the “policy” that drove Act 43 was indeed that
impermissible goal: partisan advantage. A majority of
the Court has already declared partisan gerrymanders
“incompatible with democratic principles.”
Ariz.
Legislature v. Ariz. Indep. Redistricting Comm’n, 135
S. Ct. 2652, 2658 (2015) (internal quotations omitted).
This Court held decades ago, and has never wavered
from the rule, that state discretion in districting is
constrained by the constitutional rights of the voters.
Baker v. Carr, 369 U.S. 186 (1962); Reynolds v. Sims,
377 U.S. 533 (1964). As the district court noted, the

 11
notion that some “partisan considerations” are lawful
in districting originates with Gaffney v. Cummings, 412
U.S. 735 (1973), but Gaffney held that partisan
considerations may be used to make districts more fair
and representative, not for the “invidious” purpose of
minimizing the voting strength of political groups. Id.
at 752-54; JSA 66a-68a.
The Gaffney line is entirely consistent with this
Court’s long-held consensus that partisan
gerrymandering presents, at the very least, a
significant constitutional problem.
In Vieth v.
Jubelirer, five Members of this Court explicitly
recognized that extreme partisan gerrymandering
violates the Constitution. See 541 U.S. 267, 307, 31216 (2004) (Kennedy, J., concurring in judgment); id. at
317–318 (Stevens, J., dissenting); id. at 343, 347–52
(Souter, J., joined by Ginsburg, J., dissenting); id. at
356–57, 366–67 (Breyer, J., dissenting). The other four
Justices in Vieth stated that they did not disagree with
that conclusion. See id. at 292 (plurality opinion) (a
dissent addresses “the incompatibility of severe
partisan gerrymanders with democratic principles. We
do not disagree with that judgment.”).7
7

Six Justices in Davis v. Bandemer, 478 U.S. 109 (1986), had
held the issue justiciable, but four of the six found proof of the loss
of a majority in the legislature insufficient to establish a lasting
impairment of voting rights. See id. at 118-27, 134-37. In Vieth
five Justices affirmed that the issue was justiciable, but Justice
Kennedy found no reliable measure of the impairment of voters’
First or Fourteenth Amendment rights. See 541 U.S. at 313-14.
Plaintiffs here offer proof of durability and a measure of an
impairment that shifts the burden of explanation to the State to
show that the impairment was not the product of partisan
gerrymander.

 12
Second, Act 43 is the product of a conflict of interest
different in kind from conflicts that affect any other
form of legislation. Each of the legislators who foists a
gerrymander upon the citizens of the state is among
the small group of individuals whose careers, positions
in the legislative hierarchy, and fundraising capacity
are cemented by their party’s majority status. Normal
checks and balances and deliberative policy processes
do not operate in this environment. See Ariz. Indep.
Redistricting Comm’n, 135 S. Ct. at 2672 (describing
the inherent conflict of interest “when legislators draw
district lines that they ultimately have to run in” and
the Founders’ purpose behind the Elections Clause “to
act as a safeguard against manipulation of electoral
rules by politicians and factions in the States to
entrench themselves or place their interests over those
of the electorate”) (internal quotations omitted).
Here, legislators used State power in their own selfinterest to disempower political opponents for a decade
or more. JSA 145a-146a. In the process, the
Legislature has maximized the number of safe majority
districts, rendering the election of legislators in the
general election a foregone conclusion in most of
Wisconsin, depriving independents of any meaningful
vote in most districts, and contributing to an
unrepresentative legislature. Throughout its defense
of gerrymandering the Legislature ignores entirely that
these consequences flow from legislation intentionally
designed for impermissible partisan reasons.
Finally, although the Legislature tells the Court
that partisan voting swings and dozens of competitive
races protected Wisconsin’s voters from the
gerrymander, the facts show just the opposite. The

 13
number of elections in competitive districts under
Act 43 does not approach the 75 claimed by the
Legislature. Br. 30. The Legislature does not disclose
its standard for a competitive election.
In his 2011 work for the Legislature, Professor
Keith Gaddie defined a district as “swing” if neither
major party had more than 52% (an advantage of up to
4%) of the total major party vote. SA 325. Act 43
reduced the number of swing districts by almost half,
from 19 to 10. Id. Even treating margins up to a 6%
advantage as competitive, half again Professor
Gaddie’s cutoff, the number of competitive elections
was 11 and 15 in 2008 and 2010, respectively, but then
dropped under Act 43 in 2012 to 12, then five, then
three in 2016. SA 325, Exs. 535 to 541; Wisconsin
Election Results.8 That adds up to 20, not 75, elections
in competitive districts out of 297 in the last three
cycles—less than 7% by a measure half again as
generous as that applied by the Legislature’s own
consultant.
The Legislature also overstates Republican
performance in close districts by choosing a skewed
group of “competitive” districts. Though the Brief
presumes that the 29 Act 43 districts at 55%-45% or
closer were competitive, it ignores that 14 of those 29
8

Citations herein to Wisconsin Elections Results reflect analyses
of official election data available at Wisconsin Elections
Commission, http://elections.wi.gov/elections-voting/results (last
visited Sept. 1, 2017). Election data is also available in
downloadable format at Wisconsin State Legislature Open GIS
Data, Legislative Technology Services Bureau, http://dataltsb.opendata.arcgis.com/datasets/2012-2020-wi-election-data-with2011-wards/data?selectedAttribute=PREREP12 (last visited Sept.
1, 2017).

 14
districts actually leaned Republican (i.e. were over 52%
Republican but did not reach the “safe” level of 55%).
SA 325. This is more than three times the four
districts that leaned Democratic. Compare Br. 29-31
with SA 325. Thus, the Legislature’s argument
proceeds not from a group of true swing districts but
from a larger group strongly tilted to favor the
Republican party.
II.

The Record Strongly Supports the District
Court’s Finding—and Act 43’s Guiding
Principle—that Past Party Preferences
Reliably Predict the Future Statewide
Behavior of Wisconsin’s Voters.

The Legislature does not deny that it deliberately
gerrymandered Wisconsin’s legislative districts in 2011
in a manner carefully calculated to give Republicans
durable and sizable majorities for a decade. Nor does
the Legislature contest the proof and the district
court’s finding that the Act 43 maps were not required
by traditional redistricting criteria. Instead, the
Legislature contends that the Wisconsin Voters’ claims
are based on “factual assumptions that are inconsistent
with how representative democracy works. . . .” Br. 17.
The Wisconsin Voters’ claims and the district court’s
findings are based not on “assumptions” but on hard
facts supported by voluminous evidence. Specifically,
the district court’s opinion, unlike the Legislature’s
argument, is based on evidence and careful analysis of
years of data from 99 Assembly districts, not cherrypicked examples from a handful of elections in the few
competitive districts that survived Act 43. And, of
course, the Legislature’s argument to this Court is
contradicted by the Legislature’s own research,

 15
analysis and conduct in successfully gerrymandering
the State of Wisconsin.
A. Actions Speak Louder than Words: the
Legislature Has Changed its Tune Since
the 2011 Redistricting Process.
The Legislature’s own conduct in 2011, proven in
Baldus, 849 F.Supp.2d at 845-46, and again in this
case, directly contradicts what it now tells this Court.
The district court, in a finding not contested here,
said:
In sum, from the outset of the redistricting
process, the drafters sought to understand the
partisan effects of the maps they were drawing.
They designed a measure of partisanship and
confirmed the accuracy of this measure with
Professor Gaddie. They used this measure to
evaluate regional and statewide maps that they
drew. They labeled their maps by reference to
their partisanship scores, they evaluated
partisan outcomes of the maps, and they
compared the partisanship scores and partisan
outcomes of the various maps. When they
completed a statewide map, they submitted it to
Professor Gaddie to assess the fortitude of the
partisan design in the wake of various electoral
outcomes.
The map that emerged from this process reduced
markedly the possibility that the Democrats
could regain control of the Assembly even with
a majority of the statewide vote. The map that
would become Act 43 had a pickup of 10

 16
Assembly seats compared to the Current Map.
As well, if their statewide vote fell below 48%,
the design of Act 43 ensured that the
Republicans would maintain a comfortable
majority.
JSA 138a-139a.
Were a party’s likely statewide success not largely
predictable based on the party’s showing in prior
elections, the Legislature would not have spent
$431,000 of state money and weeks of effort analyzing
past elections in various combinations of wards to test
the durability of ever more partisan majorities against
possible future swings in the relative public support for
the two major parties. JA 186. The Legislature claims
various other factors affect districting choices, but the
record reveals no evidence that in drawing its maps the
Legislature considered any nonpartisan policy or the
record or character of any individual candidate. Br. 20.
Indeed, the State here admits the Legislature’s vote
projections in 2011 “took no account of incumbency,
candidate strength, or other Assembly-district-specific
factors.” Appellants’ Brief 15.
In 2011 the Legislature hired consulting expert
Professor Keith Gaddie9 to evaluate the future vote
breakdown that could be expected under a series of
draft redistricting plans. SA 322; JSA 12a-13a.
Professor Gaddie calculated a “partisan score” based on
regression analysis to measure the party preference of
voters in existing and proposed districts. Id. He then
confirmed that his regression analysis tracked closely
9

The district court concluded: “We find [Gaddie’s] testimony
credible.” JSA 127a, n.181.

 17
with the Republican staff’s “composite score” of
partisanship and told his clients that their composite
score was “an almost perfect proxy for open seat vote,
the best proxy you’ll come up with.” JSA 17a-18a,
127a, Ex. 175 at 1. Professor Gaddie used methods
very similar to Professor Mayer’s to gauge
partisanship, and these experts for both sides agreed
that partisanship would in turn predict legislative
outcomes. JA 164-65; JSA 14a, 47a-48a, 54a, 127a128a, 131a. In fact, Professor Gaddie also noted to his
clients the “top-to-bottom party basis of the state
politics.” SA 322. Moreover, his projections regarding
the Republican advantages under the enacted plan
were borne out)in a good year for Republicans, the
Republicans did even a little better than Professors
Gaddie, and Mayer, predicted. JSA 148a, 153a, 158a
(“. . . we have the actual election results to confirm the
reliability of Professor Gaddie’s model . . . .”).
B. The District Court Relied on Extensive
Data Proving that Voter Party
Preference Reliably Predicts Assembly
Vote in the Vast Majority of Districts.
The Legislature claims voters “choose[] candidates
based on their records and positions, not just their
political parties.” Br. 20. But as the district court
found and the data demonstrate, voters’ choices are
overwhelmingly influenced by party rather than
individual candidate characteristics. Creating a robust
majority of safe districts produces a very high
probability of preserving one-party control throughout
the decade. The district court never suggested every
district would inevitably follow the party preference its
history suggested. But the underlying proven fact is

 18
that “safe” or “strongly leaning” districts rarely fall
outside the pattern.
The Wisconsin Voters’ expert Kenneth Mayer
testified at length and virtually without rebuttal on
these points. JSA 13a-14a, 147a-152a. Professor
Mayer’s testimony and the district court’s findings
considered statewide data—not a handful of selected
races—to demonstrate the high correlation between
party preference in presidential vote and Assembly
races. His regression analysis considered a variety of
independent factors that might influence elections and
proved presidential vote accounted for 93% to 94% of
the variation in Assembly vote. SA 47. Professors
Gaddie’s and Mayer’s projections are also closely
correlated, reinforcing their reliability as predictors of
Assembly vote. Professor Mayer actually compared his
estimate of the partisanship of each district against
Professor Gaddie’s own calculations. “The r-squared
for this regression is 0.96, indicating that the two
measures are almost perfectly related, and are both
capturing the same underlying partisanship.” SA 5556.
Professor Mayer noted that the presidential vote
share “correlates very strongly with other more
complex measures of partisan strength.” SA 39. He
showed graphically how closely Assembly vote tracks
presidential vote in Wisconsin:

 19

Further voting data from 2014 and 2016 shows the
same correlation.10

10

Wisconsin Election Results, supra. Graphs generated by
Microsoft Excel 2010.

 20

The ward-by-ward vote totals from the elections under
Act 43 show the high correlation between votes for
Republican candidates for governor and president and
votes for Republican Wisconsin State Assembly
candidates. The best-fit line for these data points, one
for each ward, shows a one-to-one ratio. They
demonstrate how the vote at the top of the ticket
matches closely the vote in each ward for the party’s
legislative candidates. In each of the three elections, in
numerical terms, the correlation exceeds 0.987.11
Because the real world is rarely perfect, however, the
relationship has just enough play to give the
Legislature the few outliers it discusses in its Brief.
In short, the district court’s finding of a high
correlation between a district’s party preference,
commonly measured by presidential vote, and its vote
11

“Multiple R,” Microsoft Excel 2010.

 21
for state legislators is not clearly erroneous.
correct and is confirmed by the experts.

It is

C. The
Legislature’s
Hand-Picked
Examples Prove Nothing.
Rather than address this extensive data and expert
testimony based on all races, the Legislature now offers
the Court anecdotes from five of the 792 Assembly
races held between 2002 and 2016. Anecdotes cannot
rebut the strong statistical correlation detailed in the
record. See, e.g., Reference Manual on Scientific
Evidence:
Third (Statistics), 217-18 (citing
shortcomings of anecdotes in establishing reliable
correlation between an alleged cause (here, individual
factors other than party affiliation) and effect (here,
statewide vote for a party)). But, these particular
anecdotes suffer from additional flaws.
(1)
The Legislature first points to four of
the 297 Assembly races in 2008, 2010 and 2012,
in which Republican candidates outperformed
the national ticket. Br. 20. These are not
random selections.
Under Act 43 the
Legislature’s composite score for District 96
narrowed from 53.6% to a 46.4% Democratic
advantage in 2012, and District 49 favored
Democrats by 50.41% to 49.59%, making it one
of the few truly competitive districts after
Act 43. SA 325. Moreover, the Legislature itself
identifies a major reason these two districts did
not follow form: the two Republicans who won
races in Democratic districts had demonstrated
a strong independent streak and opposed
Governor Walker on the 2011 Act 10 that

 22
sparked mass protests and launched a new era
of partisanship in Wisconsin. Br. 20.
(2)
The Legislature’s next set of districts
“that have changed hands within the same
districting cycle,”12 id., show in almost every
case the strong correlation between direction of
vote change in state Assembly races and top-ofticket races. Br. 20-22. Democrat Danou won in
2014 but was unseated by a Republican in
2016—a strong Republican year.
Br. 21.
Similarly, Republican Rivard lost to a Democrat
in 2012, a strong Democratic year—but then
Republican Quinn recovered the seat in 2014 (a
Republican year statewide and nationally) and
did better in 2016 along with the surge in
Republican support. Id. And Representative
Vruwink, a Democrat, won in the 2012
presidential election year but then lost in 2014
to a Republican, whose margin then predictably
increased in 2016 when President Trump won
Wisconsin. Id. None of these districts was
packed with Democrats: they ranged from
44.3% to 52.18% Republican post-Act 43.
SA 325.
(3)
The Legislature next looks to the 2002
to 2010 districting cycle to find two districts in
which party preference shifted. Br. 21. These
elections were conducted under court-drawn
maps that included at least twice as many
competitive districts as Act 43. SA 325. The
races the Legislature cites were in districts that
12

District 70, District 75 and District 92.

 23
would be expected to shift with the relative
political fortunes of the two major parties.
Specifically, District 28 swung from Republican
to Democrat in 2006 and back to Republican in
2010. Br. 21. District 68 swung from Democrat
to Republican in 2004, back to Democrat in 2008
and back to Republican in 2010. Id. These
changes reflected national trends as one would
expect in what the Legislature refers to as
“traditional ‘swing’ districts.” Id. But, in 2011,
Act 43 did away with many competitive districts,
and minimized this healthy effect that the
Legislature purports to value so highly. See pp.
12–14 above.
(4)
Last, the Legislature offers the 2010
example of an “embattled” Assembly speaker
who lost his seat to scandal in 2010. Br. 21. On
this the Legislature and the Center can agree )
a high-profile scandal can affect any race. But
the Legislature has not pointed to a single such
scandal that flipped a district under Act 43.
Even had it done so, it would have proven
nothing.
The statewide results were the
predictable, and predicted, result of the
Legislature’s calculated gerrymander. Partisan
gerrymanders work statewide, not district by
district.

 24
III.

Voters Have Representational Interests
and Rights that Extend Far Beyond
Constituent Services.

In Section II.C, the Legislature seems to claim that
those in the minority party suffer no harm from their
gerrymander because the minority party voters are
“[n]ot [d]eprived of [r]epresentation or [a]ccess to the
[p]olitical [p]rocess.” Br. 23. But, this truism—that all
voters are represented by the representative from their
district regardless of how they voted—ignores the realworld relationship between elected officials and their
constituents.
To be sure, each person residing within a district,
whether Republican, Democrat, Independent, or noneof-the-above, has a person in the state legislature
designated as his or her representative. Even voters
who did not support a legislator may be able to get help
with a pass to tour the White House or funding to
repair a local bridge or extend a bike trail. The
Legislature’s point is that the “name of the majority
party has no bearing on a great many of the services
that legislators provide their constituents on a day-today basis.” Br. 25. But, their representatives do not
represent or speak on behalf of the minority party
voters on the issues that matter most. As a practical
matter, an elected official in a gerrymandered safe
district can freely ignore the policy concerns of those
constituents who play no role in getting the
representative elected. And voters everywhere are
adversely affected by the distorted and
unrepresentative legislature Act 43 produces. As
former Wisconsin Senate Majority Leaders from both
parties recently wrote in the Washington Post, a

 25
“different kind of politics” takes hold when maps are
“rigged” and politicians no longer “ha[ve] to win votes
from the middle”:
With party control pre-decided, most legislators
care only about their primary elections.
Everyone knows who’s going to win the general
election. There’s much less incentive to reach
bipartisan compromise. Instead, we see policy
decided in partisan caucuses, leaving the
minority out in the cold.
Tim Cullen & Dale Schultz, We led the Wisconsin
Senate. Now we’re fighting gerrymandering in our
state, Wash. Post, June 20, 2017 (emphasis supplied).13
The Legislature misconstrues Shaw v. Reno, 509
U.S. 630, 648 (1993) for the proposition: “[P]resum[ing]
that voters who support[] losing candidates are
deprived of representation” is “antithetical to our
system of representative democracy.” Br. 23. In fact
Shaw says the opposite.
Describing a racial
gerrymander, the Court wrote:
The message that such districting sends to
elected representatives is equally pernicious.
When a district obviously is created solely to
effectuate the perceived common interests of one
racial group, elected officials are more likely to
believe that their primary obligation is to
represent only the members of that group,
13

https://www.washingtonpost.com/opinions/we-led-thewisconsin-senate-partisan-gerrymandering-in-our-state-andothers-goes-too-far/2017/06/20/8978544e-55d1-11e7-ba90f5875b7d1876_story.html?utm_term=.563d4d7fc4c0 (emphasis
added).

 26
rather than their constituency as a whole. That
is altogether antithetical to our system of
representative democracy.
Shaw, 509 U.S. at 648. The same holds true for a
partisan gerrymander. The Republican legislators
have every reason to cater to their core voters and
every incentive to ignore the interests of the minorityparty voters.
The Legislature contends that votes for the losing
candidate can force the winning candidate to “#adopt
more moderate centrist positions’ and to be more
responsive to independent and swing voters.” Br. 26
(quoting Opinion below (Griesbach, J., dissenting), JSA
287a). But that holds true only in competitive districts.
Act 43 intentionally minimized this salutary feature of
a republican form of government.
The Legislature claims “unexpectedly close”
elections serve as beneficial “wake-up call[s]” for
politicians. Br. 26. But in a gerrymandered legislature
the majority can sleep undisturbed. “Unexpectedly
close” elections are too rare to serve as any meaningful
check on the majority party’s gerrymander, or
realistically cause a candidate “to be more responsive
to the independent and swing voters.” Id.; see also pp.
12–14 above on frequency of competitive elections. The
Legislature also overlooks the data regarding
polarization of legislatures, which places Wisconsin
near the top of the list. Boris Shor & Nolan McCarty,
Ideological Mapping of American Legislatures, 105 Am.
Pol. Sci. Rev., 530, 537, 540, 546 (2011). The Wisconsin
Legislature’s documented high degree of political
polarization even before 2011 refutes the Legislature’s
claim that the ordinary processes of democratic

 27
elections will bring legislators to the center and cause
them to moderate their views. Br. 23-26.
This polarization has reinforced the dominance of a
caucus system in the legislature, demanding fealty to
the majority of the caucus and eliminating even the
possibility of dialog with minority legislators. See e.g.,
JSA 8a-9a (describing how caucus system blocks
minority-sourced legislation).
The relationship between elected officials and
constituents that the Legislature lauds in an audacious
effort to argue that gerrymanders do no harm is not
only counterfactual, it is not in keeping with this
Court’s previous statements on the subject. See p. 11
above; see also Vieth, 541 U.S. at 329–30 (Stevens, J.,
dissenting) (“Gerrymanders subvert that
representative norm because the winner of an election
in a gerrymandered district inevitably will infer that
her success is primarily attributable to the architect of
the district rather than to a constituency defined by
neutral principles.); Bandemer, 478 U.S. at 132
(“[U]nconstitutional discrimination occurs … when the
electoral system is arranged in a manner that will
consistently degrade a voter’s or a group of voters’
influence on the political process as a whole.”).

 28
IV.

The Authors of Act 43 Did not Assume
Away Party Volatility)They Planned for It.

The Legislature offers another surprising defense in
Section II.D of its Brief. Partisan gerrymandering
claims, it says, “ignore the reality” that how a voter
voted in past elections does not “necessarily dictate[]
what results that voter would prefer to see in the next
election.” Br. 27. This is an empirical claim that voter
movement from one party to another prevents
mapmakers from severely impeding the voting strength
of either party. Thus, lawmakers who spent some
$431,000 to build one of the most partisan districting
plans in U.S. history now contend that gerrymanders
do not work. JSA 246a-247a.
Gerrymandering claims do not rely on any
assumption that, at the individual level, “party
affiliation rarely changes” or “candidates do not
matter.”
Br. 29.
Indeed, anticipating vote
switching—and insulating partisan advantage against
it—is an essential part of any gerrymandering project,
and was here. These phenomena occur, but they do not
occur in sufficient numbers to overcome the advantage
that Act 43 built into a majority of districts.
The Legislature’s expert, Professor Gaddie,
conducted elaborate analyses to test how maps would
respond to changes in party preference. JSA 22a-23a,
27a, 41a-42a, 54a, 131a; SA 335-339 (Prof. Gaddie “S
Curves”). His predictions were borne out: the
gerrymandered map was robust enough in the face of
normal political ebbs and flows to give Republicans
what they paid for—a severe and durable partisan

 29
advantage.14 Professors Mayer and Jackman tested the
same thing: how would the gerrymander hold up in the
face of likely shifts in party preference? They correctly
predicted that the Act 43 map would reliably produce
Republican majorities even in years where the
Republican vote share sagged well below 50%.
JSA 148a.15 All the experts agreed on this general
method. JSA 149a at n.255.
Indeed, Professor Jackman projected almost exactly
how the efficiency gap would react in a year like 2016
when the Republicans did well. He predicted 10% and
the actual efficiency gap was around 11%. Br. of Eric
McGhee as Amicus Curiae in Support of Neither Party
at 17, n.6; SA 360. His sensitivity testing further
shows that even had the trend from 2012 to 2016 been
14

If accepted, this argument would also defeat claims of racial
vote dilution. While race is immutable, voting preferences are not.
Racial vote dilution claims therefore depend on a link between
voting preferences and race that is no less empirical than the link
between voting preferences and party.
15

The Legislature also points out that the gerrymander
challenged in Vieth and another judicial gerrymander cited by the
Vieth court in North Carolina failed to hold up. Br. 27. As with
the partisanship/vote correlation argument (see above at 14–21),
here the Legislature selects a few examples purporting to disprove
a general rule.
But, those picks again prove only the
unremarkable proposition that political behavior is not a perfect
science. This occasional failure of a gerrymander may reflect on
the skill of the gerrymanderers, the over-ambitious drawing of
maps or wave elections. The Legislature has not and cannot show
that any gerrymander meeting the standards set forth by the
district court in this case would likely fail. To the contrary, the
district court required proof that an intentionally discriminatory
map would likely endure throughout the period of the
gerrymander. JSA 148a-152a, 158a, 166a, 172a.

 30
Democratic, the gerrymander would still have produced
reliably pro-Republican seat shares. Id., JSA 148a,
152a at n.262, 165a.
Moreover experience confirms the accuracy of the
Legislature’s 2011 district-by-district partisanship
composite scores as predictors of 2012 and 2016
presidential election results.

 31
Graphs generated by Microsoft Excel 2010 from
composite scores at SA 325 and data compiled by
DailyKos.16
Statistics purporting to show “significant intradecade volatility in partisan balance” are less
persuasive still. Br. 28. As the Legislature would have
it, frequent changes in party control under the same
electoral map would suggest legislative power in
Wisconsin goes solely and simply to the party whose
“candidates do a better job than their counterparts at
earning votes.” Br. 29. In fact, Wisconsin’s actual
“intra-decade volatility” suggests the opposite. Under
the court-imposed non-gerrymandered map in force
from 2004-2010, both parties won and lost seats, with
Republicans gaining, losing, then regaining majorities
in both legislative chambers as one would expect
absent gerrymandered districts engineered to survive
common partisan shifts. By contrast, in the wake of
Act 43, starting in 2014 only four races changed party
out of 198 races held. Wisconsin Election Results,
supra. The “simple explanation” for this divergence is
not better candidates, but gerrymandered safe
majorities. And the record shows just how effective
that gerrymander has been. In 2012, 51.4% of the
statewide vote gave Democrats 39 Assembly seats, but
a “roughly equivalent vote share for Republicans (52%
in 2014) . . . translated into 63 seats”—yielding a 24seat disparity. JSA 154a. The Legislature does not
seek to explain this disparity.
16

See Jeffmd, Daily Kos Elections’ statewide election results by
congressional and legislative districts, Daily Kos, July 9, 2013,
https://www.dailykos.com/stories/2013/7/9/1220127/-Daily-KosElections-2012-election-results-by-congressional-and-legislativedistricts.

 32
CONCLUSION
The Legislature’s counter-factual and counterhistorical account only demonstrates why ordinary
political processes have not and will not solve this
problem. This Court should affirm the judgment of the
district court.
Respectfully submitted,
THEODORE R. BOEHM
HOOVER HULL TURNER
111 Monument Circle
Suite 4400
Indianapolis, IN 46204
(317) 822-4400
tboehm@hooverhullturner.com

JOSEPH H. YEAGER, JR.
Counsel of Record
HARMONY A. MAPPES
DULANY LUCETTA POPE
MATTHEW B. HARRIS
JEFFREY P. JUSTMAN
JULIANA YANEZ
FAEGRE BAKER DANIELS LLP
300 North Meridian Street
Suite 2700
Indianapolis, IN 46204
(317) 237-0300
jay.yeager@faegrebd.com

Counsel for Amicus Curiae