Amicus Curiae Brief — Husted v. A. Philip Randolph Inst., 138 S. Ct. 43 (2017) (No. 16-980)

aliens FILED

AND

BRIEFS No. 16-980

In the

Supreme Court of the Anited States

JON HUSTED, OHIO SECRETARY OF STATE,

Petitioner,

Vv.

A. PHILIP RANDOLPH INSTITUTE, ET AL.

Respondents.

ON WRIT OF CERTIORARI TO THE UNITED STATES

COURT OF APPEALS FOR THE SIXTH CIRCUIT

BRIEF FOR AMERICAN HISTORY

PROFESSORS AS AMICI CURIAE IN SUPPORT

OF RESPONDENTS

BRITTANY N. LOVEJOY: RICHARD P. BRESS

PATRICK K. O’BRIEN Counsel of Record

LATHAM & WATKINS LLP RYAN C. GROVER

505 Montgomery Street SAVANNAH K. BURGOYNE*

Suite 2000 LATHAM & WATKINS LLP

San Francisco, CA 94111 555 11th Street, NW

(415) 391-0600 Suite 1000

brittany.lovejoy@lw.com Washington, DC 20004

patrick.obrien@Iw.com (202) 637-2200

richard.bress@|w.com

ryan.gover@|w.com

savannah.burgoyne@lw.com

* Admitted in Virginia only;

all work supervised by a

member of the DC Bar.

Counsel for Amici Curiae

TABLE OF CONTENTS

Page

yr | © Us lig, | +. 5) Sees

INTERESTS OF AMICUS CURIAE w.0......0cccccecccseesseees 1

SUMMARY OF ARGUMENT................ccccccccsssssssesrseeseeees 3

I. ciceten 8 For + Non-Voting V Was Justified

As A Measure To Keep Accurate Voter

Rolls, But The Practice Raised Profound

II. Congress Acted To End The Practice Of

Purging Citizens From Registration

Lists For N wnlaeie When It Passed

The NVRA.... ARES ae OLE

A. abies Was Concerned Over

The Effect Of Non-Voter Purges

On Voter Participation................--.....0... 18

B. Congress Prohibited States From

Considering Non-Voting In A

Registrant Removal Process

Except To Confirm Independent

Evidence Of Voter Ineligibility .............. 23

Ill. The Department Of Justice Historically

Interpreted The NVRA As Prohibiting

Purge Practices Like Ohio’s

III in otcieienieiesctiieentnininniniameics 27

IV. HAVA Does Not Supersede The NVRA’s

Prohibition On Non-Voter Purge

ii

TABLE OF AUTHORITIES

Page(s)

CASES

Citizens’ Comnmittee for the Recall of Jack

Williams v. Marston,

IE SII siniceccncssiinntccainnteniiusnenssiinetiiiemnes 14

Duprey v. Anderson,

ee eT I BPD cccctcentnnsrmstnemseneveminuseninvnannaens 14

Gonzales v. Oregon,

RE ERIE EM eI nae ements 29

Hoffman v. Maryland,

928 F.2d 646 (4th Cir. 1991) .............ccccseccsseecrecesseresees 14

Michigan State VAW Community Action

Program Council (CAP) v. Austin,

198 N.W.2d 385 (Mich. 1972).........:cccce-rersessceseeeeeee 4,14

New York Trust Co. v. Eisner,

I enideieuisioneciienen 1

Reynolds v. Sims,

Ir I i esennienenais 3

Simms v. County Corrt,

ee es Os ID crcicncscsincntessrncensonsonnncsinnens 14

South Carolina v. Katzenbacii,

ER Eee ae ome mE 6

iii

TABLE OF AUTHORITIES—Continued

Page(s)

Toney v. White,

476 F.2d 203 (5th Cir.), vacated in part

on rel’g en banc, 488 F.2d 310 (5th Cir.

SUITE discecesichinsicecineihieasiceuiehibasendcitenieietindonapiipseiibtnincaiasiia 5, 15

Williams v. Osser,

350 F. Supp. 646 (E.D. Pa. 1972) ..........ccceeeeneees 15, 16

STATUTES

TE |

ee Pe Fa i ctnsinrsnesceerentetnnenscmmepsmntenicsnnemanninncienimesian 17

Se Se: ND seer casesesncecesusntinsssnntnvinsinmmapemnnsenttscasinanvain 17

Se er Oe INIT nchisicieniiioseigniviiacicdchnsecnipniuncianapciaiainiaiinenneninagaeg 17

Fe as Se INI apices encctentinincnsintcniineeneenteanainsniimnitita 24

a ee 6 Be Riccencccscsssnsnncisesvennecsteresien 18, 23, 30

Se es FR ee erntnensccnssenenncenenpreeneenincenesenssestamensant 25

Be ts I cecetrccnnivinssnecensinnenininouanicioemenatinewinees 25

| 6M | EIS ene DETTE TEE 26

Be Reds 1b CpG Psccrecerccncccenssnssenseesseeneveseneentininise 31

ee eas OF nN N ercttescccrnsemncerensensnecneamiveveavemantetenl 30

Pub. L. No. 89-110, 79 Stat. 437 (1965)..........ccseseseesseenes 6

Pub. L. No. 103-31, 107 Stat. 77 (1993)........scsecsesneeeesnenes 5

iv

TABLE OF AUTHORITIES—Continued

Page(s)

Pub. L. No. 107-252, 116 Stat. 1666 (2002)...................0.. 6

LEGLISLATIVE MATERIALS

8 8 en 17

Be, BRO DEO. BO CI vccceccccvecsccncscssescscensecnescees passim

H.R. Rep. No. 107-329, pt. 1 (2001)........00.cscecescesecessessees 30

H.R. Rep. No. 107-730 (2002) (Conf. Rep.)................ 6, 31

Bs Bs es Re CRs nccsencnccnenionncccssenesnmsentssecsees 16, 17

Be es Tr Oe riretetiaserectonsisimiccmsemsnemtnesenes passim

Voter Registration: Hearings Before tie

Subcomm. on Elections of the H. Comm.

on H. Admin., 100th Cong. (1988) ................ 17, 19, 21

Voter Registration: Hearings Before the

Subcomm. on Elections of the H. Comm.

on H. Admin., 101st Cong. (1989)................. 19, 20, 22

OTHER AUTHORITIES

Steve Barber et al., The Purging of

Empowerment: Voter Purge Laws and tie

Voting Rights Act, 23 Harv. C.R.-C.L. L.

I ee ee cee 12, 13, 16

Vv

TABLE OF AUTHORITIES—Continued

Page(s)

Committee for the Study of the American

Electorate, Creating the Opportunity:

How Voting Laws Affect Voter Turnout

Cs Fi d ccctevscniccnniniichitsasenisieanmmaeninegipitinloteneliiiie 16

Chandler Davidson, The Voting Rights Act:

A Brief History, in Controversies in

Minority Voting: The Voting Rights Act

in Perspective (Bernard Gofman &

Chandler Davidson ed5S., 1992) ..........ccsseeseeeserseeeees 7

Federal Election Commission, /mplementing

the National Voter Registration Act: A

Report to State and Local Election

Officials on Problems and Solutions

Discovered 1995-1996 (Mar. 1998)........0:cccccceeeeerees 29

Joseph P. Harris, A Model Registration

System: Report of the Committee on

Election Administration of the National

Municipal League, in Supplement to

National Municipal Review (Jan. 1927)................ 9,10

Joseph P. Harris, A Model Registration

System: Report of the Committee on

Election Administration of the National

Municipal League, Supplement to

National Municipal Review (1981)...............:0:0 8,12

Joseph P. Harris, Legislative Notes and

Reviews, 22 Am. Pol. Sci. Rev. 349 (1928)............8, 10

vi

TABLE OF AUTHORITIES—Continued

Page(s)

Joseph P. Harris, National Municipal League,

Model Voter Registration System (4th ed.

a 10

Joseph P. Harris, Registration of Voters in

the United States (1929) ............-sccsecesccorcsesseereeeeess 9, 11

Alexander Keyssar, The Right to Vote: The

Contested History of Democracy in the

8 7, 8,16

J. Morgan Kousser, Colorblind Injustice

J. Morgan Kousser, The Shaping of Southern

Politics: Suffrage Restriction and the

Establishment of One-Party South, 1880-

EL a ee 7

Stephen Loffredo, Poverty, Democracy and

Constitutional Law, 141 U. Pa. L. Rev.

SE 13

Arnold I. Menchel, Election Laws: The Purge

for Failure to Vote, 7 Conn. L. Rev. 372

National Commission on Federal Election

Reform, To Assure Pride and Confidence

in the Electoral Process (Aug. 2001),

http://web1.millercenter.org/

commissions/comm_2001.pdf ..............:cccceceseceeererensees 7

vii

TABLE OF AUTHORITIES—Continued

Page(s)

National Municipal League, A Model Election

BITTE ccasiiepdntitadtieneuintendadiaieliniatmionsenenddun

Report of the President’s Commission on

Registration and Voting Participation

GID ccnnsndcntnceensnescnnestetasenesstettubimmemmnssntameneoassnccines

John F.. Reynolds, Testing Democracy:

Electoral Behavior and Progressive

Reform in New Jersey, 1880-1920 (1988)........

Daniel P. Tokaji, Voter Registration and

Election Reform, 17 Wm. & Mary Bill Rts.

ie I cercecncdescennesninciaitbianeitinamneiatnasinaiotion

Q. Douglas Weeks, Permanent Registration

of Voters in the United States, 14 Temp.

SE isiinialei henner id tater ceca

Leonard D. White, Public Administration,

1927, 22 Am. Pol. Sci. Rev. 339 (1928).............

INTERESTS OF AMICUS CURIAE'

Amici curiae are professors who focus their

academic research and teachings on American history,

and in particular voting rights and elections. Amici

bring their objective expertise to this case and have no

personal stake in its outcome. As students and

teachers of history, amici believe that Petitioner’s

arguments are wrong in part because they are

profoundly ahistorical. This Court has aptly observed

on a number of occasions that “a page of history is

worth a volume of logic,” New York Trust Co. v.

Eisner, 256 U.S. 345, 349 (1921), and this case is no

exception. Amici submit this brief to describe the

history of state laws that purged the names of voters

from voter registration lists because those people failed

to vote—from their rise in the early twentieth century

to their ultimate and absolute rejection by Congress

when it passed the National Voter Registration Act of

1993.2

The scholars joining this brief include:

e Orville Vernon Burton, Professor of History,

Clemson University.

1 The parties have consented to the filing of this amicus

brief. No counsel for any party authored this brief in whole or in

part; and no such counsel or any party made a monetary

contribution intended to fund the preparation or submission of this

brief. No person or entity, other than amici and their counsel,

made a monetary contribution intended to fund the preparation or

submission of this brief.

2 All signatories speak for themselves only and not on

behalf of their respective institutions. Institutional affiliations are

listed for identification purposes.

2

F. Chandler Davidson, Radoslav A. Tsanoff

Professor Emeritus of Public Affairs and

Sociology, Rice University.

Steven Hahn, Professor of History, New York

University.

Pippa Holloway, Professor of History, Middle

Tennessee State University.

Alexander Keyssar, Matthew W. Stirling Jr.

Professor of History and Social Policy, Harvard

University.

J. Morgan Kousser, William R. Kenan, Jr.,

Professor of History and Social Science,

California Institute of Technology.

Allan J. Lichtman, Distinguished Professor of

History, American University.

Richard L. McCormick, President Emeritus and

Board of Governors Professor of History and

Education, Rutgers University.

John F’. Reynolds, Former Professor of History,

University of Texas at San Antonio.

J. Douglas Smith, Author and Historian, Chair

of Humanities, Colburn Conservatory of Music.

3

SUMMARY OF ARGUMENT

This case is about voting, a right “at the heart of

representative government.” Reynolds v. Sims, 377

U.S. 533, 555 (1964). The franchise has expanded

throughout our history via constitutional amendments

and federal legislation, as well as shifting norms and

attitudes. See id. at 555 & n.28. Even so, elections too

often are marred by low voter turnout, a problem

exacerbated by state practices that burden registration

and reduce access to the polls. In this case, Petitioner,

the Ohio Secretary of State, defends one such practice.

Ohio’s Supplemental Process automatically cancels the

registration of everyone who has not voted in two

years and then fails to respond to a single notice and

vote in the ensuing four years. Purging residents in

this way for exercising their right not to vote violates

federal law, regardless of whether the State provides

notice and an opportunity to challenge removal.

Respondents explain why that is so under the plain

text of the relevant statute. Amici agree and seek

here to place that text in its historical context. The

Court should reject Petitioner’s contrary revisionist

history and affirm the Sixth Circuit’s decision below.

In the early twentieth century, state and local

governments increasingly began purging the names of

citizens who failed to vote as an imprecise but

inexpensive way to keep voter rolls accurate.

Substantial “dead wood” on voter rolls, the theory

went, made for inaccurate registration lists that might

enable the likes of Tammany Hall to pad votes. The

working assumption behind these purging practices

was that people who did not vote in a jurisdiction failed

to do so because they had lost their voting eligibility in

that jurisdiction—perhaps they moved away, died, or

4

were convicted of a crime. And even if that assumption

proved incorrect in any particular case (7.e., the person

in fact remained eligible to vote in that jurisdiction),

proponents asserted that the affected person could

either prevent the purge from going into effect

(assuming that the person received advance notice) or

reregister (assuming that the person understood

reregistration was an option). None of that is to say

that these purges were innocent—historians like amici

have since found that in many cases registration

systems and purging practices had both the intent and

effect of disfranchising voters, particularly African-

American voters. Nevertheless, many States agreed

with the early proponents of these laws and adopted

purging for non-voting as a quick-and-easy method to

clean up voter registration lists. But over time even

once-proponents of purging for failure to vote came to

see the shortcomings of this practice, and Congress

ultimately banned the practice entirely.

Mounting experience with these purging practices

generated concerns about their efficacy and fairness.

Many who studied the causes of low voter turnout in

the United States concluded that cumbersome state

registration laws—including purging for non-voting—

were among the primary culprits. Critics also cited

evidence showing that these laws disproportionately

impacted poor and minority voters. Courts joined the

fray as well. The Michigan Supreme Court, for

example, questioned the logic of using failure to vote as

a proxy for voter ineligibility when the “absence of

baby-sitters” or the conscious exercise of the right not

to vote were equally plausible explanations for someone

sitting out an election. Michigan State UAW Cmty.

Action Program Council (CAP) v. Austin, 198 N.W.2d

5

385, 388 (Mich. 1972). And the Fifth Circuit exposed a

dark underbelly of the practice in a case where a local

election official violated federal law by applying

Louisiana’s purge process in a racially discriminatory

manner against African Americans. Toney v. White,

476 F.2d 203, 205-06, 208 (5th Cir.), vacated in part on

rel’g en banc, 488 F.2d 310 (5th Cir. 1973).

In the late 1980s, Congress addressed concerns

over state laws that purged registrants for non-voting

when it tackled comprehensive voting rights and

election law reform. Over the course of several years

of hearings, Congress heard from numerous groups

about the need to reform the complex maze of state and

local voter registration laws to increase citizen

participation in the voting process. At the same time,

state and local officials testified about their legitimate

need to maintain accurate voter registration lists.

The result was the National Voter Registration Act

of 1993, Pub. L. No. 103-31, 107 Stat. 77 (““NVRA”),

which struck a balance between these competing

interests. This balance came out against the continued

use of purge laws based on failure to vote—regardless

of whether the laws allowed for extended time periods

to measure non-voting and regardless of whether they

provided notice and an opportunity to challenge before

an individual was removed from a registration list.

Congress recognized that purging for non-voting was a

relatively common practice, and that States had a

legitimate interest in maintaining accurate voter

registration lists. Congress simply decided that a

person’s failure to vote should not be used to achieve

that goal. Congress directed States instead to use

more precise and less discriminatory methods for

identifying ineligible voters, including the National

6

Change of Address program available through the U.S.

Post Office.

Amici are not alone in this understanding of the

text and purpose of the NVRA. The United States

had—and over two decades consistently advocated—

precisely the same understanding, from the moment

Congress passed the NVRA until this late stage of this

case. That understanding of the NVRA was, and

remains today, correct, the Government’s recent about-

face notwithstanding.

That understanding, moreover, has not been

superseded legislatively. When Congress passed the

Help America Vote Act of 2002, Pub. L. No. 107-252,

116 Stat. 1666 (“HAVA”), it did not resurrect the

ability of States to purge voters who failed to vote. To

the contrary, Congress explicitly intended that HAVA

“leave[] NVRA intact, and ... not undermine it any

way.” H.R. Rep. No. 107-730, at 81 (2002) (Conf. Rep.)

(emphasis added).

ARGUMENT

I. Purging For Non-Voting Was Justified As A

Measure To Keep Accurate Voter Rolls, But

The Practice Raised Profound Concerns

State and local barriers to voting have taken many

forms over time. In innumerable well-documented

cases, state and local officials created flagrant obstacles

to voting designed to disenfranchise minority groups,

particularly African Americans. Congress admirably

fought against those obstacles through numerous

pieces of federal legislation, most notably the Voting

Rights Act of 1965, Pub. L. No. 89-110, 79 Stat. 437.

See, e.g., South Carolina v. Katzenbach, 383 U.S. 301,

308-15 (1966).

7

The practice of purging for non-voting was not

always driven by any overtly discriminatory animus.

As Petitioner recounts, these and other purge practices

emerged as a means to maintain accurate rolls when

States and municipalities shifted away from periodic

voter registration systems to permanent voter

registration systems in the early 1900s. See Pet’r Br. 3

(citing National Commission on Federal Election

Reform, To Assure Pride and Confidence in the

Electoral Process 28 (Aug. 2001),

http//web1.millercenter.org/commissions/co1am_2001.pdf).

But that is only part of the story. Petitioner

ignores the evidence that in many cases these voter

registration systems themselves also “served—and

often were intended to serve—as a means of keeping

African-American, working-class, immigrant, and poor

voters from the polls.” Alexander Keyssar, The Right

to Vote: The Contested History of Democracy in the

United States 312 (2000); see also Chandler Davidson,

The Voting Rights Aet: A Brief History, in

Controversies in Minority Voting: The Voting Rights

Act in Perspective 7, 10 (Bernard Gofman & Chandler

Davidson eds., 1992) (describing “statutory suffrage

restrictions” as a means to disfranchise African

Americans); J. Morgan Kousser, The Shaping of

Southern Politics: Suffrage Restrictionand tie

Establishment of One-Party South, 1880-19TO at 47-50

(1974) (describing registration laws, and particularly

the discretion afforded to registrars, as a means for

disfranchising African-American voters); Daniel P.

Tokaji, Voter Registration and Election Reform, 17

Wm. & Mary Bill Rts. J. 458, 457 (2008) (“There was

often an ulterior motive for these laws as well, namely

to impede the participation of groups that those

8

running elections wanted to exclude.”). “(The devil

was in the details,” Keyssar, swpra, at 312, and purging

often was driven by discriminatory animus and had

discriminatory effects. See J. Morgan Kousser,

Colorblind Injustice 34 (1999) (describing purging as a

practice used te “keep[] the black vote under control”

in the nineteenth century); John F. Reynolds, Testing

Democracy: Electoral Behavior and Progressive

Reform in New Jersey, 1880-1920 at 142, 146 (1988)

(discussing a 1911 New Jersey law that, among other

things, required voters who missed an election to

reregister and reportedly led to 800 of Trenton’s 1,000

black voters being removed from the registry).

In any event, the shift to permanent registration

systems certainly created a perceived need to clean

voter registration lists of the names of individuals no

longer eligible to vote in that jurisdiction. See Joseph

P. Harris, A Model Registration System: Report of the

Cominittee on Election Administration of the National

Municipal League, Supplement to Nationa] Municipal

Review 7 (1931) (“Nat’l Mun. League 1931 ed.”); see

also Joseph P. Harris, Legislative Notes and Reviews,

22 Am. Pol. Sci. Rev. 349, 349 (1928). Otherwise, any

benefits in efficiency and convenience resulting from

permanent registration systems could be substantially

offset by inaccurate voter rolls and a generalized fear

of voter fraud by political machines. Nat'l Mun.

League 1931 ed., at 37-38; Harris, Legislative Notes,

supra, at 349-51.

The question therefore became how to achieve most

effectively the goal of bona fide voter registration lists.

In 1927, the National Municipal League published A

Model Registration System, a report that provided

“best” practices already in use in some jurisdictions

9

that, at least according to the National Municipal

League, other States and cities should adopt. Joseph

P. Harris, A Model Registration System: Report of the

Committee on Election Administration of the National

Municipal League, in Supplement to National

Municipal Review 45-48, 77-83 (Jan. 1927) (“Nat’l Mun.

League 1927 ed.”). In Boston, for example, the local

police conducted a census of adult residents and

compared the results to registration records. /d. at 78,

80-81. Death reports were another “very practical and

effective means” of keeping voter registration lists

current. /d. at 78.

Among these “best” practices, the National

Municipal League recommended that a person’s failure

to vote in recent elections should be used as a basis for

removing ineligible individuals from voter registration

lists. Jd. at 78-79. Cities like Denver, Colorado and

Portland, Oregon, for instance, depended exclusively

on this method to the exclusion of all others because of

its certain effectiveness in identifying every voter who

died or moved away. /d. at 79; Joseph P. Harris,

Registration of Voters in the United States 225-26

(1929) (“Other methods fail to catch some electors who

have died or moved away, but all are caught by a

system of cancellation for failure to vote.”). Because of

their efficiency, the National Municipal League

advocated for the adoption of practices like these that

purged those who failed to vote in a one- or two-year

period, or in a general election, provided the voters

received notice and an opportunity to seek

reinstatement. Nat’l Mun. League 1927 ed., at 78-79.

Such measures, in the National Municipal League’s

view, “place[d] a mild penalty upon non-voting and

10

thereby stimulate|d] voting,” and would “not greatly

inconvenience[]” those who failed to vote. Jd. at 79.

Bills inspired by the National Municipal League’s

recommendations were introduced in seven States

immediately thereafter. See Leonard D. White, Public

Administration, 1927, 22 Am. Pol. Sci. Rev. 339, 344

(1928) (describing the National Municipal League’s

model registration system as the “most important

event in this field during 1927”); Harris, Legislative

Notes, supra, at 352. By 1939, twenty States had laws

on the books mirroring all or part of the Model

Registration System, see QO. Douglas Weeks,

Permanent Registration of Voters in the United States,

14 Temp. L.Q. 74, 75-76 (1939), and by the 1970s, fully

thirty-eight States plus the District of Columbia

purged citizens from voter registration lists based on

failure to vote. See Arnold I. Menchel, Election Laws:

The Purge for Failure to Vote, 7 Conn. L. Rev. 372,

373-78 (1975); see also Joseph P. Harris, National

Municipal League, Model Voter Registration System 44

(4th ed. 1954 & rev. 1957) (“Cancellation for failure to

vote is the principal means used in most permanent

registration jurisdictions to purge the lists.”).

Each of these laws permitted cancellation of

registrations for failure to vote, though they differed in

certain respects, most notably (1) the time period used

to track a person’s voting inactivity, and (2) the notice

(or lack thereof) provided to a person whose name had

been or would be purged. See Weeks, swpra, at 84-85;

Menchel, supra, at 373-78. Some laws considered a

failure to vote within two years; others four years or

more; some laws provided no notice; others provided

notice with a limited window to challenge removal or

seek reinstatement, and yet another provided multiple

11

notices before purging went into effect. See Weeks,

supra, at 84-85; Menchel, swpra, at 373-78. In all cases,

people could reregister if their names were purged

from the voter registration lists. See Harris,

Registration of Voters, supra, at 224-25; Menchel,

supra, at 376-77.

But as these laws took hold, questions emerged

about their wisdom. Voter participation rates in the

United States lagged well behind those in other

developed democracies. See Report of the President’s

Commission on Registration and Voting Participation

7-9(1963) (“1963 Report’). Groups that studied the

voter participation problem, like the President’s

Commission on Registration and Voting in 1963,

blamed “{rjestrictive legal and administrative

procedures in registration and _ voting [that]

disfranchise millions.” Jd. at 1. The Commission

highlighted the fundamental unfairness that results

when “[aJn unexpected business trip or a broken ankle

can deprive a citizen of his right to vote,” and when

that failure to vote two years ago means the person

“cannot vote now.” /d. at 11.

The National Municipal League voiced similar

concerns in 1973. National Municipal League, A Model

Election System 1 (1973) (“Nat’l Mun. League 1973

ed.”) (“[TJhe problem of non-voting in America is

directly related to the machinery states have created

for registering voters and administering elections.”).

“Frequent purges will result in more current and

accurate lists,” the National Municipal League

explained, “but they also increase the volume of

registrations that officials must process and may

seriously inconvenience voters. What the system

12

confers at one point, it often takes away later.” Jd. at

2A.

Both the President’s Commission on Registration

and Voting and the National Municipal League agreed

that States should never cancel a person’s registration

for failure to vote for any period less than four years, if

they did so at all. See 1963 Report 37; Nat’l Mun.

League 1973 ed., at 33. Indeed, the National Municipal

League went so far as to call laws with shorter non-

voting periods “discriminatory and undesirable”

because they disproportionately affected the large

number of people who voted only in presidential

elections. Nat’l Mun. League 1973 ed., at 33. The

National Municipal League more generally described

cancellation for non-voting as a flaw of permanent

registration systems that required citizens to

constantly reregister and thus called into question

their “permanent” label—a stark shift from its

previous longstanding support for aggressive purging.

Compare id, at 4, with Nat’] Mun. League 1931 ed., at

38-39. The National Municipal League now

recommended that States primarily use door-to-door

canvassing, not purging for non-voting, to keep

registration lists clean. Nat’l] Mun. League 1973 ed. at

24-34 (recommending voter purge laws based on

inactivity only for “States which do not implement a

door-to-door canvassing system”).

Critics also questioned the core assumption

underlying purging for non-voting—that a person’s

failure to vote was a reliable indicator of their

ineligibility to vote. See Menchel, supra, at 373, 393;

Steve Barber et al., The Purging of Empowerment:

Voter Purge Laws and the Voting Rights Act, 23 Harv.

C.R.-C.L. L. Rev. 483, 508 (1988). According to these

13

critics, that assumption had little validity in States that

purged the names of voters who merely failed to vote

for two years or in one general election. See Menchel,

supra, at 373. The assumption also lost persuasive

force in States that gave voters inadequate notice

before their registrations were cancelled. In Missouri,

for example, residents of St. Louis County who

received a notice were not told they had the right to

reregister if they failed to challenge the purge within

twenty days, but residents of Jackson County were.

Id. at 375; see also Barber, supra, at 501.

Critics further pointed out that the poor and

minority groups were disproportionately affected by

these purges both because they voted less frequently

and because they had greater difficulty navigating

reregistration once their registrations were purged.

See Barber, supra, at 491-92 (discussing a statistical

study of voters in Arizona showing that Mexican

Americans turned out less frequently in midterm

elections and concluding they would’ be

disproportionately impacted by a two-year purge for

failure to vote); Menchel, supra, at 392 (“The lower the

level of education, the more likely the voter will be, or

view himself as being, unable to register.”); see also

Stephen Loffredo, Poverty, Democracy and

Constitutional Law, 141 U. Pa. L. Rev. 1277, 1327-28

(1993) (“[P]Jeople who are literally struggling to find

enough to eat are highly unlikely to participate in the

political process. ... The politically quiescent attitude

of the poor, therefore, is less a matter of free choice,

than of the mutually reinforcing effects of low

resources, weak political incentives, and inadequate

skills that trap the poor in ... a cycle of defeat.”

(citations and internal quotation marks omitted)).

14

Litigants and judges hashed out these questions,

too. Challengers argued that these purge laws were

invalid under the U.S. and applicable _ state

constitutions. Courts often disagreed, holding that the

laws reasonably related to the legitimate state interest

of preventing voter fraud. See, e.g., Simms v. Cty. Ct.,

61 S.E.2d 849, 851-54 (W. Va. 1950); Citizens’ Comm.

for the Recall of Jack Williams v. Marston, 507 P.2d

113, 116-17 (Ariz. 1973); Duprey v. Anderson, 518 P 2d

807, 810-11 (Colo. 1974); see also Hoffman v. Maryland,

928 F.2d 646, 648-49 (4th Cir. 1991) (upholding

Maryland’s voter purge statute).

But in some cases, judges identified critical

weaknesses in these laws. In Michigan State UAW

Community Action Program Council (CAP) v.

Austin, the Michigan Supreme Court struck down that

State’s purge law with a two-year voting inactivity

trigger because the law affected the right to vote

protected by the Michigan Constitution and lacked a

compelling government interest. See 198 N.W.2d 385,

390 (Mich. 1972); see also Menchel, supra, at 381-82. In

doing so, the court questioned the validity of using a

person’s failure to vote as a proxy for identifying

ineligible voters because so many other explanations

existed for non-voting, “including illness, travel,

absence of baby-sitters, or a conscious protest against

all of the candidates in a particular election.” Michigan

State UAW, 198 N.W.2d at 388. That logical flaw had

profound real-world consequences given the sheer

number of registrations purged under the Michigan

law. From 1960 to 1970, over 600,000 registrations

were cancelled for non-voting in the city of Detroit

alone. /d.

15

The Fifth Circuit confronted a different problem in

Toney v. White. See 476 F.2d 203 (5th Cir. 1973),

vacated in part on reh’g en banc, 488 F.2d 310 (Sth Cir.

1973); see also Menchel, supra, at 386. There, a local

registrar had discriminated against African-American

voters by publishing a notice that 141 voters’ names

would be purged for non-voting without informing the

named individuals of their right under Louisiana law to

appear before the registrar and prove their right to

remain registered. Toney, 476 F.2d at 205-07. Of the

141 affected voters, 130 were African American. /d. at

206. The court concluded that this, along with other

discriminatory actions, violated the Fifteenth

Amendment and the Voting Rights Act. /d. at 208.

Even the divided decision of a three-judge district

court in Williams v. Osser that upheld purging for non-

voting exposed significant problems with these laws.

See 350 F. Supp. 646 (E.D. Pa. 1972); see also Menchel,

supra, at 383-85. In Williams, the challengers of

Pennsylvania’s purge law presented uncontested and

statistically valid evidence showing that 58.5% of

people removed from voter rolls for non-voting had not

actually moved, been convicted of a felony, or died and

were thus eligible to vote. 350 F. Supp. at 649-50.

Worse yet, fewer than half of those people took action

to reregister after they were purged. Jd. The court

nevertheless upheld the Pennsylvania law because it

concluded the law was rationally related to the state

interest of preventing voter fraud and, in the court’s

view, imposed only a “minimal” burden on voters. /d.

at 653. Viewing the statistical evidence, a dissenting

judge concluded that “[iJt is quite clear ... that non-

voting for two years is not a valid indicator of non-

residence” and would have found the law

16

unconstitutional under a more demanding level of

scrutiny. Jd. at 654-55 (Luongo, J. dissenting).

So matters stood in the late 1980s, when American

voter turnout hit historical lows. Voter participation in

the 1986 midterm elections fell to its lowest in over

four decades, with 112,000,000 eligible citizens failing to

vote. See Committee for the Study of the American.

Electorate, Creating the Opportunity: How Voting

Laws Affect Voter Turnout 3 (Oct. 1987). The

Committee for the Study of the American Electorate, a

nonprofit and nonpartisan group guided by public

officials and scholars, published a report raising the

alarm about the “appalling and embarrassing” voter

turnout figures that illustrated “a growing crisis for

American democracy.” See id. at 3; id. at 5 (“Simply

and bluntly, government of, for and by the people is in

danger of becoming government of, for and by the

few.”). The Committee did not mince words when it

came to purges based on voter inactivity. It found this

practice was “a major impediment to voting” and

eoncluded that ending the practice would substantially

enhance voter turnout. See id. at 77, 86-87.

By this point, forty States and the District of

Columbia had laws on the books that purged

individuals for non-voting, see Barber, supra, at 499,

and concerns about low voter participation, caused at

least in part by these purging laws, abounded. See

Creating the Opportunity, supra, at 77, 86-87; ef.

Keyssar, supra, at 312-13.

II. Congress Acted To End The Practice Of

Purging Citizens From Registration Lists For

Non-Voting When It Passed The NVRA

The dismal voter turnout to the 1988 presidential

elections spurred Congress into action. See S. Rep. No.

17

101-140, at 4 (1989). By the late 1980s, Congress had

considered an array of potential federal legislation

aimed at increasing the number of eligible citizens

registered to vote. See, e.g., H.R. Rep. No. 101-243

(1989); S. Rep. No. 101-140. Congress appreciated that

because the “failure to become registered is the

primary reason given by eligible citizens for not

voting,” making registration more accessible was

critical to increasing voter participation. See, e.g., H.R.

Rep. No. 108-9, at 3 (1993); see also S. Rep. No. 103-6,

at 2 (1993). At the same time, Congress recognized

that any new federal legislation should “minimiz/e]

potential new problems for State and local election

officials,” Voter Registration: Hearings Before the

Subcomm. on Elections of the H. Comm. on H.

Admin., 100th Cong. 32 (1988) (hereinafter 1988

Hearings) (statement of Rep. Swift), “protect the

integrity of the political process,” and “assure an

accurate and current voter registration roll,” S. Rep.

No. 103-6, at 1. Congress’s legislative efforts to

achieve these sometimes-competing goals culminated

in the NVRA.

Congress’s first stated goal in passing the NVRA

was to “increase the number of eligible citizens who

register to vote in elections for Federal office.” See 52

U.S.C. § 20501(b)(1). It aimed to accomplish this on the

front end by reforming state voter registration

processes to reduce government-imposed barriers to

registration. See e.g., H.R. Rep. No. 103-9, at 3; see

also 52 U.S.C. §§ 20504-20506 (requiring that States

allow citizens to register to vote simultaneously with a

driver’s license application, by mail, and in person). At

the same time, Congress recognized that, on the back

end, some of the methods States were using to

18

maintain voter rolls were stripping eligible voters from

registration lists, often in a discriminatory fashion. See

H.R. Rep. No. 103-9, at 5, 15; S. Rep. No. 103-6, at 3, 17-

18. In particular, testimony and evidence presented to

Congress confirmed that laws that purged individuals

based on their failure to vote were likely to remove

eligible voters from voter registration lists, directly

frustrating Congress’s commitment to increase voter

registration numbers.

To address that problem, Congress included in the

NVRA an express provision prohibiting state practices

that “result in the removal of the name of any person

from the official list of voters registered to vote in an

election for Federal office by reason of the person’s

failure to vote.” 52 U.S.C. §20507(b)(2). Having

recognized that a person’s failure to vote is not a fair or

accurate proxy for voter ineligibility, Congress

required a State to obtain some other, independent

evidence suggesting that a voter was ineligible prior to

taking steps to purge a particular voter. The Act’s

text, history, and purpose establish that Congress

intended to eliminate purge practices triggered by a

failure to vote.

A. Congress Was Concerned Over The

Effect Of Non-Voter Purges On Voter

Participation

The legislative history of the NVRA echoes the

very same concerns about fairness and efficacy that

critics of voter-inactivity purges had been raising for

decades. When considering the various bills that

eventually resulted in the NVRA, Congress carefully

weighed evidence that state purges of non-voters too

often resulted in the removal of eligible voters from

registration lists, thereby improperly reducing the

19

number of voters able to vote in a given election. See,

e.g., 1988 Hearings, at 285, 306, 379-80 (Creating the

Opportunity, swpra). Central to this issue was the

question of whether an individual’s failure to vote was

reliable evidence that the voter had moved, died, or

otherwise become ineligible to vote.

Evidence and testimony presented to Congress

established the important (if unremarkable) fact that

an individual’s failure to vote often was explained by a

number of reasons that had nothing at all to do with

their continued eligibility to vote. The League of

Women Voters, for example, provided the House

Subcommittee on Elections with a litany of reasons

why a voter might decline to vote in a given election,

which included “lack of interest in or confusion about a

particular election, disbelief that the issues presented

will adequately represent one’s concerns or that the

candidates are worthy of one’s support, inaccessibility

of the polling place, absence, emergency, [and] health.”

Voter Registration: Hearings Before the Subcomm. on

Elections of the H. Comm. on H. Admin., 101st Cong.

149 (1989) (hereinafter 1989 Hearings) (statement of

Nancy M. Neuman, President, League of Women

Voters of the United States); see also e.g., id. at 49

(statement of Rep. John Conyers, Jr.) (explaining that

“there are many reasons non-voters do not participate:

some are apathetic; some feel that existing choices do

not offer answers; others do not feel their vote will

make a difference”).

In addition, a report submitted to Congress by the

Citizens’ Commission on Civil Rights indicated that

“the pool of non-voters who make covsciors decisions

not to vote for a number of reasons ... has grown

dramatically since the 1960s.” 1988 Hearings, at 51

20

(“Barriers to Registration and Voting: An Agenda for

Reform”) (emphasis added). The Reverend Jesse

Jackson remarked upon the fundamentally

undemocratic nature of purging such conscious

objectors on the basis of their failure to vote, noting

that “[njo other rights guaranteed to citizens are bound

to the constant exercise of that right.” 1989 Hearings,

at 135 (testimony of Rev. Jesse Jackson, President,

National Rainbow Coalition). Mr. Jackson explained:

“We do not lose our right to free speech because we do

not speak out on every issue. By the same token, we

should not lose our right to vote because of our refusal

to vote in any particular election.” Jd. Congress

agreed:

(While voting is a right, people have an

equal right not to vote, for whatever

reason. However, many States continue

to penalize such non-voters by removing

their names from the voter registration

rolls merely because they have failed to

east a ballot in a recent election. Such

citizens may not have moved or died or

committed a felony. Their only “crime”

was not to have voted in a recent election.

S. Rep. No. 103-6, at 17.

The many possible alternative explanations for an

individual’s failure to vote thus called into question

both the wisdom and the suppressive effect of using an

individual’s non-voting status as evidence that the

voter may have moved, passed away, or otherwise

become ineligible to vote. The Committee for the

Study of the American Electorate presented evidence

to Congress demonstrating that the state practice of

using inactivity as a proxy for ineligibility “ha{d] a

21

serious negative effect on turnout” because it

eliminated as many as two million individuals from

voter rolls who would otherwise have been eligible to

vote. 1988 Hearings, at 306 (Creating the Opportunity,

supra). The Committee therefore recommended that

“In pursuing their legitimate responsibilities to provide

accurate voting lists, [States] should seek other

methods to cleanse the lists of those who have died or

moved.” /d.

The NVRA’s legislative history also reveals

concerns that non-voting purges might

disproportionately affect minority voters. Congress

understood that, historically, “selective purges” were

akin to the “poll tax/es], literacy tests, [end] residency

requirements” that state and local jurisdictions had

used to “discourage [the] participation” of minorities in

the electoral process throughout the nineteenth and

early twentieth century. H.R. Rep. No. 108-9, at 2-3;

see also S. Rep. No. 103-6, at 3. Though Congress

appreciated that contemporary state purge practices

were not likely fueled by the same discriminatory

animus as their historical predecessors, Congress

recognized that purges for non-voting could continue to

“disproportionately affect persons of low incomes, and

blacks and other minorities.” See S. Rep. No. 103-6, at

17-18.

Congressional testimony confirmed that state

registration laws were “not uniform,” “not

nondiscriminatory” and, in some cases, were

“interpreted in such a manner as to deny eligible

citizens their right to vote.” H.R. Rep. No. 103-9, at 4.

In particular, a number of interest groups voiced

specific concerns about the impact of state non-voter

purge laws on minority participation in the democratic

22

process. For example, the NAACP testified before

Congress that “there should be no non-voting purges”

because “[iJn many instances of non-voting purges

blacks are disproportionately purged because of their

infrequency of voting.” 1989 Hearings, at 153

(statement of Althea T.L. Simmons, Director,

Washington Bureau, NAACP); see also id. at 154

(testimony of Pamela Monroe Young, Legal Director,

NAACP).

The League of Women voters similarly expressed

“econcern|] about ‘purging’ procedures used in many

states, especially those that automatically remove

voters’ names from the list for not voting in a specified

number of elections.” Jd. at 149 (statement of Nancy

M. Neuman). The League noted that in New York City

350,000 names were removed from the lists for non-

voting before the April 1988 presidential primary,

“effectively negating the accomplishments of a city-

wide voter registration drive that had added 350,000

new names to the rolls.” Jd. And another 312,000

people were sent purge notices in February of 1989.

Id. The League strongly urged that “no voter’s name

should be removed from the list of registered voters for

not voting.” Id.

In sum, the legislative history of the NVRA reflects

serious and pervasive concerns—raised both by those

testifying before Congress and by Congress itself—

about the propriety of removing citizens from voter

rolls based on their failure to vote. Congress

understood that the continued use of that practice

contributed to the substantial voter turnout problem in

the United States, particularly among African

Americans and other minority groups.

23

B. Congress Prohibited States From

Considering Non-Voting In A Registrant

Removal Process Except To Confirm

Independent Evidence Of Voter

Ineligibility

In light of these concerns, Congress included in the

NVRA an express prohibition against state practices

that “resalt in the removal of the name of any person

from the official list of voters registered to vote in an

election for Federal office by reason of the person's

failure to vote.” 52 U.S.C. § 20507(b)(2) (emphasis

added) (“Failure-to-Vote Clause”). Ohio’s

Supplemental Process violates that prohibition. Under

the Supplemental Process, if an Ohio resident fails to

vote in a single election cycle, they are sent a notice

requesting their response. See Pet’r Br. 11. If the

voter does not respond to that solitary notice and does

not vote within the next four years, their voter

registration is canceled. Jd. Any such process that can

“result” in a voter being removed trom the from Ohio’s

voter rolls “by reason of [their] failure to vote,” is

impermissible under the NVRA. See 52 US.C.

§ 20507(b)(2).

Petitioner argues that the Supplemental Process

does not violate the NVRA because it only prohibits

States from “removing” registrants on the basis of

their failure to vote—i.e., that it does not prohibit

States from using the failure to vote as a “trigger[]” for

a notice-confirmation process. See Pet’r Br. 25-26. The

NVRA makes no such distinction. To the contrary, the

text broadly prohibits “/a/ny state pregram or

activity” that “result/s] in” purging “by reason of the

person’s failure to vote.” 52 U.S.C. § 20507(b)(2)

(emphasis added). The trigger for non-voting is itself a

24

“program or activity” that “results in” purging for non-

voting and thus falls under the NVRA’s prohibition.

The legislative history confirms that this statutory

prohibition applies to the entire voter-removal process,

from start to finish. In addition to subsection (b)(2)’s

prohibition against any “program or activity” used to

regulate voter registration rolls that results in purging

individuals for their failure to vote, Section &(b)

includes—at subsection (b)(1l)—an affirmative

requirement that such “program{s] or activit[ies]”

must “be uniform, nondiscriminatory, and in

compliance with the Voting Rights Act of 1965.” /d.

§ 20507(b)(1). The House Report explains that “the

intent of this section [is] to impose the uniform,

nondiscriminatory and conforming with the Voting

Rights Act standards on any activity that is used to

start, or has the effect of starting, a purge of the voter

rolls, without regard to how it is described or to

whether it also may have some other purpose.” H.R.

Rep. No. 103-9, at 15 (emphasis added); see also S. Rep.

No. 103-6, at 32 (same); id. at 19, 32 (explaining that use

of address change information from the National

Change of Address Program as a “trigger” is uniform

and nondiscriminatory). Subsection b(1) therefore

regulates every chronological step in a State’s voter

removal process, including the “activity that is used to

start, or has the effect of starting, a purge”—1.e., the

“trigger.” H.R. Rep. No. 103-9, at 15 (emphasis added).

Nothing in the NVRA or its legislative history

suggests that Congress intended the Failure-to-Vote

Clause in subsection (b)(2)—which applies to the very

same “program[s] and activitfies)’"—to have a

narrower scope.

25

Nor would excluding triggering rules from the

prohibition against purging for nonvoting be sensible:

Given the low voter participation rates in the United

States, voter inactivity would be a “trigger” pulled

with great regularity. Widespread use of voter

inactivity as a trigger by States to initiate notice-

confirmation processes would in turn result in broad

swaths of Americans being mailed purge notices. And

whether a significant proportion of the American

electorate remain on the voter rolls would then hinge

on whether or not registrants respond to those notices.

In addition, the process the NVRA authorizes for

removing voters who have moved out of jurisdiction

suggests that the only lawful “trigger” for a notice-

confirmation removal process is reliable evidence of

voter ineligibility. NVRA Sections 8(c) and &(d), taken

together, detail a safe-harbor method that States can

use to identify and subsequently purge voters who may

no longer live in an applicable jurisdiction. See 52

U.S.C. §20507(c)-(d). These provisions illustrate

Congress’s intent that States obtain independent

evidence of voter ineligibility—separate and apart

from a registrant’s failure to vote—prior to initiating a

voter purge notice-confirmation process.

Subsection (c) provides that a State seeking to

purge its voter registration lists of non-residents may

begin by consulting the U.S. Postal Services’ National

Change of Address (“NCOA”) system in order to

determine whether a voter has changed his or her

address to a location outside of the jurisdiction. See id.

§ 20507(c); H.R. Rep. No. 103-9 at 15. The House

Report explains that

[iJn order to provide some guidance to the

States, subsection (c) provides that a

26

State may meet the requirements of

conducting a general program that makes

a reasonable effort to keep voting lists

clean by establishing a program which

uses the [NCOA] program of the U.S.

Postal Service. Use of the NCOA

program by a State or any of its

registration jurisdictions could be deemed

to be in compliance with the requirements

that the program be _— uniform,

nondiscriminatory and in compliance with

the Voting Rights Act of 1965.

H.R. Rep. No. 103-9, at 15. Although use of the NCOA

is not mandatory, Congress envisioned that States

would use it, or some other “reasonable program,” to

obtain evidence of voter ineligibility prior to beginning

a voter removal process. See S. Rep. No. 103-6, at 19,

32. For example, States may opt instead to use

mailings that are returned as undeliverable as

objective evidence of voter ineligibility.

Once evidence that a registrant may have moved

has been obtained, the NVRA requires a State to use

the notice-confirmation procedures outlined in Section

8(d) to confirm that change of address. Section (d)(1)

provides that a State may purge an individual from a

voter registration list only (1) after a registrant

“confirms in writing” that he or she has changed

residence to a _ place outside the registrar’s

jurisdiction”; or (2) once the registrant fails to respond

to an address confirmation notice and fails to vote

during the next two Federal election cycles. See 52

U.S.C. §20507(d)(1). Congress’s use of the word

“confirm” further suggests that Congress intended

that States obtain evidence of a change in residence

27

prior to initiating a notice-confirmation process that

might result in the removal of the registrant.

The legislative history of the NVRA substantiates

this view. Congress was unambiguous in in its reports

that removal should be predicated on a _ voter’s

ineligibility. The Senate Report explains, for example,

that “[t)he Act allows the removal of a person’s name

from the official list by reason of a change of residence

outside the jurisdiction of the registrar,” and that

notice could be sent only “{ilf a State determines that a

registrant may have changed residence.” S. Rep. No.

103-6, at 19, 32-33. More broadly, the NVRA was

designed to “assure that voters’ names are maintained

on the rolls so long as they remain eligible to vote in

their current jurisdiction and to assure that voters are

not required to re-register except upon a change of

voting address to one outside their current registration

jurisdiction.” Jd. at 2. As discussed in Section II.A,

supra, Congress understood that, given the myriad

alternative reasons why an individual might fail to

vote, voter inactivity is not a dependable proxy for

voter ineligibility. Viewed in that context, the

legislative history strongly suggests that a voter

removal procedure can be lawfully triggered only by

particularized evidence of “a change of residence” (or

some other evidence of voter ineligibility)—a failure to

vote alone is not sufficient.

Ill. The Department Of Justice Historically

Interpreted The NVRA As Prohibiting Purge

Practices Like Ohio’s Supplemental Process

Immediately after Congress passed the NVRA, the

Department of Justice went to work enforcing its

provisions. And in the course of those enforcement

efforts, the Department of Justice interpreted the

28

NVRA as prohibiting all forms of purging for non-

voting, even in cases when voters received notice

before the purge went into effect. In 1994, for example,

Assistant Attorney General of the Department of

Justice Civil Rights Division Deval Patrick wrote to

the Senior Assistant Attorney General of Georgia

objecting to Georgia’s procedure of sending a

registration confirmation notice to registered voters

who failed to vote or otherwise contact election officials

for three years. See Br. for the United States (“United

States Sixth Cir. Br.”) Attachment 2 at 1-2, A. Philip

Randolph Inst. v. Husted, 838 F.3d 699 (6th Cir. 2016),

(No. 16-3746), 2016 WL 3923034. Including such people

in Georgia’s purge procedures “is directly contrary to

the language and purpose of the NVRA,” the

Department of Justice wrote, “and is likely to have a

disproportionate adverse effect on minority voters in

the state.” Jd. at 2 (emphasis added). Allowing such

purges to continue would “eliminate certain of the

gains to minority voters mandated by Congress in

enacting the NVRA and, accordingly, ‘would lead to a

retrogression in the position of racial minorities with

respect to their effective exercise of the electoral

franchise.” Jd. (citation omitted).

In the same vein, in 1996 the Department of Justice

challenged a Pennsylvania purge law that used a

person’s failure to vote within five years as a trigger to

send the person a notice, which resulted in the person’s

removal from registration lists if they failed to return

the notice or vote in the two subsequent general

elections. United States Sixth Cir. Br. Attachment 3,

at 14-18. The Department explained that “(t]he NVRA

was designed to render superfluous the need for large

scale purges and list cleaning systems such as

29

Pennsylvania’s purge for non-voting” through its

provision of alternative means of identifying voters

who changed residences. Jd. at 17. According to the

Department, Congress decided that “non-voting can

not be used as a proxy for ineligibility.” Jd. at 18 n.23.

And the Department correctly recognized that

Congress even meant to ban ali purges based on non-

voting, no matter how apparently reasonable, such as a

law that only purged voters who failed to vote for the

previous 100 years. Jd. As the Department explained,

“Congress embodied in the statute its determination

that any marginal benefit which might result from

permitting the use of non-voting was outweighed by

the burden on individuals’ rights not to vote and the

disparate systemic impact such a rule tends to have on

minorities and the poor.” Jd. at 19 n.23.

The list of cases in which the Department of Justice

advanced this understanding goes on and on. See

United States Sixth Cir. Br. at 2-3 (discussing other

instances when the Department of Justice challenged

provisions similar to Ohio’s Supplemental Process); see

also Federal Election Commission, /mplementing the

National Voter Registration Act: A Report to State and

Local Election Officials on Problems and Solutions

Discovered 1995-1996 at 5-22 (Mar. 1998) (describing

the Department of Justice’s objections to Georgia and

South Dakota measures that targeted individuals to

receive confirmation notices because they had failed to

vote in preceding elections).

None of this is to say that the Court owes deference

to the Department of Justice’s prior interpretations of

the NVRA; it does not. See Gonzales v. Oregon, 546

U.S. 248, 264 (2006). But the executive branch’s

consistent interpretation of the NVRA for more than

30

two decades and over three different presidential

administrations, from its inception up until this final

stage of this case, casts doubt about its sudden about-

face.

IV. HAVA Does Not Supersede The NVRA’s

Prohibition On Non-Voter Purge Practices

In the aftermath of the disputed 2000 presidential

election, Congress sought to “improve our country’s

election system.” H.R. Rep. No. 107-329, pt. 1, at 31

(2001). Nothing suggests tliat in so doing it intended to

affect in any way the scope or meaning of the NVRA’s

prohibition against non-voter purges. To the contrary,

Congress expressly provided that HAVA is not to “be

construed to authorize or require conduct prohibited

under ... or to supersede, restrict, or limit the

application of ... [the NVRA|).” 52 U.S.C.

§ 21145(a)(4).

Petitioner points to the provision HAVA added to

the Failure-to-Vote Clause stating that “nothing in [the

Failure-to-Vote Clause] may be construed to prohibit a

State from using the [notice-confirmation procedures

outlined in Section 8(d)]” to remove registrants from

voter rolls, and argues that this confirms the legality of

non-voter purge practices like Ohio’s, which engage in

a notice-confirmation procedure prior to removing a

registrant. See Pet’r Br. 36 (quoting 52 U.S.C.

§ 20507(b)(2)). It confirms no such thing. The far more

reasonable interpretation is that this merely clarifies

that use of the statutorily mandated procedure for

confirming independent indicators of ineligibility does

not constitute a practice that unlawfully purges for

non-voting.

Petitioner further contends that HAVA’s

requirement that States keep and maintain statewide

31

registration lists counsels in favor of Ohio’s

Supplemental Process. Pet’r Br. 38-39. HAVA

provides that “registrants who have not responded to a

notice and who have not voted in 2 consecutive general

elections for Federal office shall be removed from the

official list of eligible voters, except that no registrant

may be removed solely by reason of failure to vote.” 52

U.S.C. § 21083(a)(4)(A) (emphasis added). But under

this process, “[iJf individuals are to be removed from

the computerized list, they shall be removed in

accordance with the provisions of NVRA.” H.R. Rep.

No. 107-730, at 75 (2002) (Conf. Rep.). Again, Congress

provided explicitly that it did not intend for HAVA to

supersede the voter removal processes mandated by

the NVRA. And for the reasons articulated in Section

I1.B, supra, Ohio’s Supplemental Process is not “in

accordance with” the provisions of the NVRA.

From the outset, the Department of Justice shared

this post-HAVA understanding of the NVRA. In 2007,

for example, the Department agreed to a consent

judgment with local officials in New Mexico mandating

that the county “shall only place the name of any voter

on an inactive list based on objective information

indicating that the voter has become ineligible to vote

due to having moved, such as returned mail with no

forwarding address or National Change of Address

program data showing a move outside the County.”

United States Sixth Cir. Br. Attachment 7 at 9 4 13.

And in 2010—under yet another presidential

administration—the Department of Justice issued

guidance on the NVRA that, as the Department

explained to the Sixth Circuit below, “addresses the

precise issue presented in this case and articulates the

Department’s position that States must have reliable

32

evidence indicating a voter’s change of address before

they initiate the NVRA-prescribed process to cancel

the voter’s registration based on a change of

residence.” United States Sixth Cir. Br. 2; see also id.

at 15. The Department revised that guidance when it

switched its position in this case, in August 2017. See

Br. for the United States 14 n.4.

The real-world concerns that motivated the

Congresses that enacted the NVRA and HAVA to ban

the practice of purging for non-voting remain just as

relevant today. Amicus Professor Lichtman analyzed

voter registration files as part of his work as an expert

in a recent North Carolina voting rights case and

concluded that upwards of 37,000 North Carolina

citizens registered in 2002 but did not vote for the first

time until 2012. See Response Report of Allan

Lichtman, North Carolina State Conf. of the NAACP

v. McCrory, No. 1:13-cv-658 (M.D.N.C. submitted Mar.

24, 2015) (on file with authors); see also ECF No. 291-2

at 29, North Carolina State Conference of the NAACP

v. McCrory, No. 1:13-cv-658. These voters likely had

any number of explanations for their decade-long

dormancy, but whatever their reasons, they are

precisely the voters Congress sought to protect when

it ended the practice of purging for non-voting.

33

CONCLUSION

For the foregoing reasons, the Court should affirm

the decision of the Sixth Circuit.

Respectfully submitted,

BRITTANY N. LOVEJOY RICHARD P. BRESS

PATRICK K. O’BRIEN Counsel of Record

LATHAM & WATKINS LLP RYAN C. GROVER

505 Montgomery Street SAVANNAH K. BURGOYNE*

Suite 2000 LATHAM & WATKINS LLP

San Francisco, CA 94111 555 11th Street, NW

(415) 391-0600 Suite 1000

brittany.lovejoy@lw.com Washington, DC 20004

patrick.obrien@|w.com (202) 637-2200

richard. bress@]w.com

ryan.gover@Iw.com

savannah.burgoyne@Iw.com

* Admitted in Virginia only;

all work supervised by a

member of the DC Bar.

Counsel for Amici Curiae

September 22, 2017