Farrakhan v. Gregoire

Opinion

FOR PUBLICATION UNITED STATES COURT OF APPEALS FOR THE NINTH CIRCUIT

MUHAMMAD SHABAZZ FARRAKHAN,  aka Ernest S. Walker; AL-KAREEM SHADEED; MARCUS X. PRICE; RAMON BARRIENTES; TIMOTHY No. 06-35669 SCHAAF; CLIFTON BRICENO, Plaintiffs-Appellants,  D.C. No. CV-96-0076-RHW v. OPINION CHRISTINE O. GREGOIRE; SAM REED; HAROLD W. CLARKE; STATE OF WASHINGTON, Defendants-Appellees.  Appeal from the United States District Court for the Eastern District of Washington Robert H. Whaley, District Judge, Presiding

Argued and Submitted April 8, 2008—Seattle, Washington

Filed January 5, 2010

Before: Stephen Reinhardt, A. Wallace Tashima, and M. Margaret McKeown, Circuit Judges.

Opinion by Judge Tashima Dissent by Judge McKeown

107 FARRAKHAN v. GREGOIRE 113 COUNSEL

Lawrence A. Weiser, Esq., University Legal Assistance at Gonzaga Law School, Spokane, Washington, for the plaintiffs-appellants.

Ryan P. Haygood, NAACP Legal Defense & Educational Fund, Inc., New York, New York, for the plaintiffs- appellants.

Carol Murphy, Deputy Solicitor General, Olympia, Washing- ton, for the defendants-appellees.

Peter A. Danelo, Heller Ehrman, Seattle, Washington, for the amici curiae American Civil Liberties Union and ACLU of Washington.

Juan Cartagena, New York, New York, for the amicus curiae Community Service Society.

Sam Hirsch, Jenner & Block, Washington, DC, for the amici curiae National Black Police Association, National Latino Officers Association of America, and Zachary W. Carter, et al., eight prominent former state and federal law-enforcement officials.

Derek S. Tarson, Debevoise & Plimpton, New York, New York, for the amici curiae Alfred Blumstein, et al., 23 leading criminologists.

OPINION

TASHIMA, Circuit Judge:

Plaintiffs, minority citizens of Washington state who have lost their right to vote pursuant to the state’s felon disenfran- 114 FARRAKHAN v. GREGOIRE chisement provision, filed this action in 1996 challenging that provision on the ground that, due to racial discrimination in the state’s criminal justice system, the automatic disenfran- chisement of felons results in the denial of the right to vote on account of race, in violation of § 2 of the Voting Rights Act (“VRA”), 42 U.S.C. § 1973. We earlier reversed the dis- trict court’s grant of summary judgment to Defendants. See Farrakhan v. Washington, 338 F.3d 1009 (9th Cir. 2003), cert. denied, 543 U.S. 984 (2004) (“Farrakhan I”). On remand, the district court again granted summary judgment to Defendants. Plaintiffs timely appeal. We reverse and grant summary judgment to Plaintiffs.

I. BACKGROUND

Plaintiffs Muhammad Shabazz Farrakhan, Al-Kareem Shadeed, Marcus Price, Ramon Barrientes, Timothy Schaaf, and Clifton Briceno (collectively, “Plaintiffs”) are minority citi- zens1 who were convicted of felonies in Washington. Id. at 1012. As a result of their felony convictions, Plaintiffs lost their right to vote pursuant to Washington’s felon disenfran- chisement law as set forth in Article VI, § 3 of the Washing- ton Constitution.2

Plaintiffs alleged that “minorities are disproportionately prosecuted and sentenced, resulting in their disproportionate representation among the persons disenfranchised under the 1 Farrakhan, Shadeed, Price, and Schaaf are African American; Barri- entes is Latino; and Briceno is Native American. 2 Article VI, § 3 provides: “All persons convicted of infamous crime unless restored to their civil rights . . . are excluded from the elective fran- chise.” An “infamous crime” is defined as one that is “punishable by death in the state penitentiary or imprisonment in a state correctional facility.” Wash. Rev. Code § 29A.04.079. Plaintiffs’ suit included a challenge to the state’s civil rights restoration procedure, see Wash. Rev. Code § 9.94A.637, but that challenge was dismissed by this Court for lack of standing, see Farrakhan I, 338 F.3d at 1021-23, and is not at issue on this appeal. FARRAKHAN v. GREGOIRE 115 Washington Constitution”; consequently, that the Washington felon disenfranchisement law “causes vote denial and vote dilution on the basis of race, in violation of the VRA . . . .” Farrakhan v. Locke, 987 F. Supp. 1304, 1307 (E.D. Wash. 1997). The district court granted Defendants’3 motion to dis- miss as to Plaintiffs’ vote dilution claim, but permitted Plain- tiffs’ vote denial claim to proceed.4 Id. at 1315.

On subsequent cross-motions for summary judgment, the district court granted Defendants’ motion and denied Plain- tiffs’ motion. Farrakhan v. Locke, No. CS-96-76-RHW, 2000 U.S. Dist. LEXIS 22212 (E.D. Wash. Dec. 1, 2000). The court found that “Plaintiffs’ evidence of discrimination in the criminal justice system, and the resulting disproportionate impact on minority voting power, is compelling.” Id. at *14. Nevertheless, it concluded that such evidence was “legally insufficient to establish causation under the VRA,” id. at *17, because “it is discrimination in the criminal justice system, not the disenfranchisement provision itself, that causes any vote denial,” id. at *15.

On appeal, we reversed the district court’s 2000 order and remanded for further proceedings. Farrakhan I, 338 F.3d at 1012, 1023. We first held that Plaintiffs’ challenge to Wash- ington’s disenfranchisement law “is cognizable under Section 2 of the VRA.” Id. at 1016. We then held that the district court “erred in failing to consider evidence of racial bias in Washington’s criminal justice system” and that it “miscon- strued the causation requirement of a Section 2 analysis.” Id. We explained that “a Section 2 ‘totality of the circumstances’ inquiry requires courts to consider how a challenged voting 3 Defendants are the State of Washington, the Governor, the Secretary of the Department of Corrections, and the Secretary of State (collectively, “Defendants” or the “State”). 4 Plaintiffs also asserted a number of constitutional claims, all of which were dismissed pursuant to Rule 12(b)(6), see Farrakhan v. Locke, 987 F. Supp. at 1314, and are not at issue on this appeal. 116 FARRAKHAN v. GREGOIRE practice interacts with external factors such as ‘social and his- torical conditions’ to result in denial of the right to vote on account of race or color.” Id. at 1012 (quoting Thornburg v. Gingles, 478 U.S. 30, 47 (1986)). Consequently, “evidence of discrimination can be relevant to a Section 2 analysis.” Id.

Following remand, the parties conducted additional discov- ery and ultimately filed new cross-motions for summary judg- ment. In their motion, Plaintiffs relied heavily on the reports of two expert witnesses: Dr. Robert Crutchfield, a Professor of Sociology at the University of Washington, who has “con- ducted extensive research on racial disparity in the Washing- ton State criminal justice system,” Crutchfield Report at 9, and Dr. Katherine Beckett, an Associate Professor of Sociol- ogy at the University of Washington, who “conducted a 2004 study entitled Race and Drug Law Enforcement in Seattle,” Beckett Report at 16.

Dr. Crutchfield’s expert report consisted of an extensive lit- erature review of the empirical research that has been con- ducted on racial disparities in the various levels of Washington’s criminal justice system (policing and investiga- tion, prosecution, and sentencing). He described studies show- ing, inter alia, that the racial disparities in the state’s criminal justice system cannot be explained by “legitimate” factors, such as racial minorities’ higher level of involvement in crim- inal activity,5 Crutchfield Report at 4-9; evidence of “unwar- ranted” racial disparities in the rates of vehicle searches, id. at 18, 21; and “observable racial differences” in the process- ing of criminal cases (e.g., charging and bail recommenda- 5 For example, whereas national studies have shown that 80% of the racial disparity in imprisonment can be explained by differential rates of crime commission (while 20% of the disparity cannot be accounted for on this basis), studies focusing on Washington have shown that “substantially more than one half of Washington State’s racial disproportionality cannot be explained by higher levels of criminal involvement.” Crutchfield Report at 9. FARRAKHAN v. GREGOIRE 117 tions, lengths of confinement, and alternative sentencing), id. at 26-30.

Dr. Beckett’s report described the findings of her study “analyzing the extent and causes of racial disparity in Seattle drug [possession and] delivery arrests.” Beckett Report at 1. Her research found that “blacks and Latinos are over- represented, and whites under-represented, among Seattle’s drug arrestees,” and that “the organizational practices that produce these disparities” — specifically, the police’s focus on crack cocaine, on outdoor drug activity, and on the down- town area — “are not explicable in race neutral terms.” Id. at 3.

The district court again granted the State’s motion for sum- mary judgment and denied Plaintiffs’ motion. Farrakhan v. Gregoire, No. CV-96-076-RHW, 2006 WL 1889273, at *1 (E.D. Wash. July 7, 2006). Reviewing the reports of Plain- tiffs’ expert witnesses, the district court found that Plaintiffs had presented “compelling evidence of racial discrimination and bias in Washington’s criminal justice system.” Id. at *6. Moreover, “[c]ontrary to Defendants’ assertion that these reports are based solely on statistics and are thus insufficient evidence for a VRA claim,” the district court found that “these experts’ conclusions, drawn from the available statisti- cal data, are admissible, relevant, and persuasive.” Id. The district court also found it significant that Defendants had not “present[ed] any evidence to refute Plaintiffs’ experts’ con- clusions.” Id. Thus, the district court concluded that it was “compelled to find that there is discrimination in Washing- ton’s criminal justice system on account of race,” id., and that such discrimination “clearly hinder[s] the ability of racial minorities to participate effectively in the political process, as disenfranchisement is automatic,” id. (quoting Farrakhan I, 338 F.3d at 1220) (internal quotation marks omitted) (alter- ation in original). 118 FARRAKHAN v. GREGOIRE Nevertheless, the district court went on to hold that “the totality of the circumstances does not support a finding that Washington’s felon disenfranchisement law results in dis- crimination . . . on account of race.” Id. at *9. Explaining that discrimination in the criminal justice system is simply one factor to consider in the totality of the circumstances analysis (falling within the scope of Senate Factor 5), the district court concluded that the remaining Senate Factors6 weigh in Defen- dants’ favor. Id. First, the district court determined that “the first Senate factor strongly favors” Defendants’ position because Plaintiffs had not shown any history of official dis- crimination in Washington. Id. at *7. Next, the district court concluded that Plaintiffs “failed to present any substantial evi- dence regarding” Senate Factors 2, 3, 4, 6, 7, and 8. Id. at *8. The court acknowledged that “several of these factors are not relevant in a VRA vote denial claim,” but found that Factors 7 and 8 — “the extent to which minority group members have been elected to political office in Washington” and the “level of responsiveness elected officials have to the particularized needs of” minorities — are “certainly relevant to Plaintiffs’ VRA claim.” Id. Finally, the court concluded that Senate Fac- tor 9 — whether the state’s policy justifications are “tenuous” — “also favors Defendants’ position.” Id. Although Defen- dants did “not explain why disenfranchisement of felons is ‘necessary’ to vindicate any identified state interest,” id., the district court concluded that, in light of the Constitution’s explicit recognition of the states’ power to disenfranchise fel- ons,7 its “ability to examine the tenuousness of Washington’s felon disenfranchisement law is extremely limited,” id. Thus, the district court concluded that “[a]lthough the evidence of 6 The Senate Factors are described and discussed in Part III.A, infra, at 13-14. 7 Section 2 of the 14th Amendment acknowledges the practice of felon disenfranchisement by providing that disenfranchisement “for participa- tion in rebellion, or other crime” will not result in the reduction of repre- sentatives to Congress that otherwise would occur when a state denies the right to vote to any male citizens over the age of 21. U.S. Const. amend. XIV, § 2. FARRAKHAN v. GREGOIRE 119 racial bias in Washington’s criminal justice system is compel- ling,” under the totality of the circumstances test, Plaintiffs had failed to establish a violation of VRA § 2. Id. at *9.

Subsequent to oral argument, and well after this case had been submitted for decision, Washington law regarding the voting rights of felons was amended. Washington law now provides that the voting rights of felons will be “provisionally restored,” at such time as those convicted under Washington state law are no longer under the authority of the Washington Department of Corrections, and, as to those convicted under federal law or in any other state, they are not in custody. See Wash. Laws of 2009, ch. 325, HB 1517. We requested sup- plemental briefing on what effect, if any, this new law might have on this case. Following our review of the parties’ briefs, we conclude that the new law does not affect our analysis or resolution of any of the issues on this appeal, with one narrow exception: the claim of one of the Plaintiffs has been mooted because he is no longer under the authority of the Department of Corrections.8

The dissent characterizes the amendment as a “significant legislative change” and would remand the case to the district court to allow it the opportunity to determine whether there are “meaningful analytical differences” between incarcerated and non-incarcerated felons. Diss. at 157. Neither party, how- ever, has ever suggested to this court — including in the sup- plemental briefing — that there are any material differences between incarcerated and nonincarcerated felons that are rele- vant to the outcome of this case.9 In the absence of any con- 8 Defendants argue that the amendment moots the case. However, with five of the original six Plaintiffs facing the same circumstance of disen- franchisement that they faced before the passage of the amendment, the case is not moot. 9 For example, the state has never argued that there are administrative difficulties in permitting incarcerated felons to vote that would justify applying a different rule to them than to non-incarcerated felons. 120 FARRAKHAN v. GREGOIRE tention, especially by the State, that such differences exist, there is neither reason nor need to remand to the district court for the purposes urged by the dissent.

Thus, we are not, contrary to the dissent’s assertion, the first court to be “presented with the question whether [incar- cerated and nonincarcerated felons] present a meaningful dis- tinction under the VRA’s totality of the circumstances inquiry.” Diss. at 158 (footnote omitted). In fact, we are not presented with that question at all. Rather, what the State con- tends regarding the amended law is that the provisions modi- fying the period during which felons are deprived of the right to vote are sufficient, when taken in concert with the other rel- evant considerations, to require us to uphold the grant of sum- mary judgment under the totality of the circumstances test that we ordinarily apply in voting rights cases. We consider that argument below, in Section III.E.

II. STANDARD OF REVIEW

We review de novo the district court’s conclusions of law regarding the application of § 2 of the VRA. Smith v. Salt River Project Improvement & Power Dist., 109 F.3d 586, 591 (9th Cir. 1997) (“Salt River”); see also Thornburg, 478 U.S. at 79 (stating that an appellate court reviewing a § 2 claim can “correct errors of law, including those that may infect a so- called mixed finding of law and fact, or a finding of fact that is predicated on a misunderstanding of the governing rule of law”) (internal citation and quotation marks omitted); Gomez v. City of Watsonville, 863 F.2d 1407, 1411 (9th Cir. 1988) (stating in a § 2 case that “the district court’s findings will be set aside to the extent that they rest upon an erroneous view of the law”). Except to note that we also review a district court’s ruling on summary judgment de novo, Fin. Mgmt. Advisors, LLC v. Am. Int’l Specialty Lines Ins. Co., 506 F.3d 922, 925 (9th Cir. 2007), we defer a fuller discussion of the standard that governs our review of the district court’s sum- mary judgment rulings to Part III.D.1, infra. FARRAKHAN v. GREGOIRE 121 III. ANALYSIS

A. Statutory Background

[1] Congress enacted the VRA of 1965, pursuant to its enforcement power under § 2 of the Fifteenth Amendment, for the “broad remedial purpose of ‘rid[ding] the country of racial discrimination in voting.’ ” Farrakhan I, 338 F.3d at 1014 (quoting South Carolina v. Katzenbach, 383 U.S. 301, 315 (1966)). As originally enacted, the VRA focused in large part on certain “covered” jurisdictions with a history of voting discrimination.10 The VRA required such jurisdictions to pre- clear any change in voting procedures with the Department of Justice; it also banned literacy tests11 and permitted the federal government to monitor elections in those jurisdictions. Voting Rights Act of 1965, Pub. L. No. 89-110, tit. I, §§ 4, 5, 6(b), 7, 9, & 13(a), 79 Stat. 437 (1965), codified at 42 U.S.C. § 1973b et seq. (1965). Section 2 of the 1965 VRA, in con- trast, was not restricted to “covered” jurisdictions. Mirroring the language of the Fifteenth Amendment, § 2 originally pro- vided that “[n]o voting qualification or prerequisite to voting, or standard, practice, or procedure shall be imposed or applied by any State or political subdivision to deny or abridge the right of any citizen of the United States to vote on account of race or color.” 42 U.S.C. § 1973 (1965).

[2] In 1980, a plurality of the Supreme Court concluded, in City of Mobile v. Bolden, 446 U.S. 55, 58 (1980), that § 2 10 “A jurisdiction was ‘covered’ for purposes of section 5 if it used a lit- eracy or other test for registering or voting and if less than half of its vot- ing age population voted in the 1964 election. The original covered jurisdictions were Alabama, Georgia, Louisiana, Mississippi, South Caro- lina, Virginia, and large parts of North Carolina.” United States v. Blaine County, Mont., 363 F.3d 897, 901 n.4 (9th Cir. 2004). 11 Congress amended the VRA in 1970 to make the ban on literacy tests nationwide for a five-year period. See Oregon v. Mitchell, 400 U.S. 112, 117 (1970). In 1975, Congress made the nationwide literacy test ban per- manent. See Blaine County, 363 F.3d at 901. 122 FARRAKHAN v. GREGOIRE “was intended to have an effect no different from that of the Fifteenth Amendment itself,” id. at 61; consequently, that plaintiffs raising claims under VRA § 2 were required to show direct evidence of discriminatory intent, as is required for Fif- teenth Amendment claims, id. at 62-63. In direct response to Bolden, Congress amended § 2 in 1982 “to make clear that proof of discriminatory intent is not required to establish a violation of Section 2.” S. Rep. No. 97-417, at 2 (1982), reprinted in 1982 U.S.C.C.A.N. 177, 179 (“Senate Report”). Section 2(a) now provides:

No voting qualification or prerequisite to voting or standard, practice, or procedure shall be imposed or applied by any State or political subdivision in a manner which results in a denial or abridgement of the right of any citizen of the United States to vote on account of race or color . . . .

42 U.S.C. § 1973(a) (emphasis added). Section 2(b) further explains that

A violation of subsection (a) . . . is established if, based on the totality of the circumstances, it is shown that the political processes leading to nomina- tion or election in the State or political subdivision are not equally open to participation by members of a class or citizens protected by subsection (a) . . . in that its members have less opportunity than other members of the electorate to participate in the politi- cal process and to elect representatives of their choice.

42 U.S.C. § 1973(b). Although the debate surrounding this amendment focused almost exclusively on vote dilution claims,12 the language of the amendment makes clear that the 12 The Senate hearings “focused on whether replacing the Bolden test with a results test would effectively mandate proportional representation— FARRAKHAN v. GREGOIRE 123 new “results test” applies both to vote dilution and vote denial claims.13 See Chisom v. Roemer, 501 U.S. 380, 394 (1991) (“[P]laintiffs can prevail under § 2 by demonstrating that a challenged election practice has resulted in the denial or abridgment of the right to vote based on color or race.”); Salt River, 109 F.3d at 594-95 (applying § 2 results test to a vote denial claim).

The Senate Report on the 1982 amendments listed “typical factors” that courts might consider in determining whether, under the totality of the circumstances, a challenged voting practice “results in” the denial or abridgement of the right to vote on account of race. These are:

(1) the extent of any history of official discrimina- tion in the state or political subdivision that touched the right of the members of the minority group to register, to vote, or otherwise to participate in the democratic process;

(2) the extent to which voting in the elections of the state or political subdivision is racially polarized;

that is, the election of racial minorities in numbers proportionate to their population.” Daniel P. Tokaji, The New Vote Denial: Where Election Reform Meets the Voting Rights Act, 57 S.C. L. Rev. 689, 705 (2006). Congress ultimately included a provision in the statute clarifying that “nothing in this section establishes a right to have members of a protected class elected in numbers equal to their proportion in the population.” 42 U.S.C. § 1973(b). 13 “Vote denial” refers to practices that prevent people from voting or having their votes counted. Historically, examples of practices resulting in vote denial include literacy tests, poll taxes, all-white primaries, and English-only ballots. “Vote dilution,” on the other hand, refers to practices that diminish minorities’ political influ- ence in places where they are allowed to vote. Chief examples of vote-dilution practices include at-large elections and redistricting plans that keep minorities’ voting strength weak. Tokaji, supra, at 691. 124 FARRAKHAN v. GREGOIRE (3) the extent to which the state or political subdivi- sion has used unusually large election districts, majority vote requirements, anti-single shot provi- sions, or other voting practices or procedures that may enhance the opportunity for discrimination against the minority group;

(4) if there is a candidate slating process, whether the members of the minority group have been denied access to that process;

(5) the extent to which members of the minority group in the state or political subdivision bear the effects of discrimination in such areas as education, employment and health, which hinder their ability to participate effectively in the political process;

(6) whether political campaigns have been character- ized by overt or subtle racial appeals;

(7) the extent to which members of the minority group have been elected to public office in the juris- diction;

(8) whether there is a significant lack of responsive- ness on the part of elected officials to the particular- ized needs of the members of the minority group;

(9) whether the policy underlying the state or politi- cal subdivision’s use of such voting qualification, prerequisite to voting, or standard, practice or proce- dure is tenuous.

S. Rep. No. 97-417, at 28-29.14 The Senate Report empha- 14 Hereinafter, the factors listed in the Senate Report will be referred to as the “Senate Factors.” Senate Factors 8 and 9 were not numbered in the Senate Report, but were provided as “additional factors that in some cases have had probative value.” S. Rep. No. 97-417, at 29. Farrakhan I, how- ever, numbered these as Factors 8 and 9. We follow that practice. FARRAKHAN v. GREGOIRE 125 sized, however, that “there is no requirement that any particu- lar number of factors be proved, or that a majority of them point one way or the other,” and that, “[w]hile these enumer- ated factors will often be the most relevant ones, in some cases other factors will be indicative of the alleged dilution.” Id. at 29.

B. Law of the Case

[3] As a preliminary matter, Defendants argue that the VRA does not apply to state felon disenfranchisement laws and that the district court’s grant of summary judgment should be affirmed on that basis alone. In Farrakhan I, how- ever, we clearly held that vote denial claims challenging felon disenfranchisement laws are cognizable under § 2 of the VRA. Farrakhan I, 338 F.3d at 1016. Defendants acknowl- edge that Farrakhan I is the law of the case, but argue that the exceptions to the law of the case doctrine permit this panel to “reexamine” Farrakhan I. We disagree because, as discussed below, none of the exceptions to the law of the case doctrine applies. Therefore, Farrakhan I remains binding on this panel.

[4] “The law of the case doctrine states that the decision of an appellate court on a legal issue must be followed in all sub- sequent proceedings in the same case.” Jeffries v. Wood, 114 F.3d 1484, 1489 (9th Cir.) (en banc) (internal quotation marks omitted) (quoting Caldwell v. Unified Capital Corp. (In re Rainbow Magazine, Inc.), 77 F.3d 278, 281 (9th Cir. 1996)), overruled on other grounds by Lindh v. Murphy, 521 U.S. 320 (1997), cert. denied, 522 U.S. 1008 (1997). Nevertheless, “a panel of this court has discretion to depart from the law of the case . . . where: ‘(1) the decision is clearly erroneous and its enforcement would work a manifest injustice, (2) intervening controlling authority makes reconsideration appropriate, or (3) substantially different evidence was adduced at a subse- quent trial.’ ” Tahoe-Sierra Preservation Council, Inc. v. 126 FARRAKHAN v. GREGOIRE Tahoe Reg’l Planning Agency, 216 F.3d 764, 787 (9th Cir. 2000) (quoting Jeffries, 114 F.3d at 1489).

[5] Defendants appear to invoke the first and second excep- tions, arguing that “[t]he subsequent intervening authority of sister circuits reveals that this Court’s conclusion was clearly erroneous and works a manifest injustice.” They rely on post- Farrakhan I cases from the Second and Eleventh Circuits, which held that the VRA does not apply to felon disenfran- chisement laws. See Hayden v. Pataki, 449 F.3d 305 (2d Cir. 2006) (en banc); Johnson v. Governor of the State of Fla., 405 F.3d 1214 (11th Cir.) (en banc), cert. denied, 546 U.S. 1015 (2005)).15 To the extent Defendants suggest that these cases constitute “intervening controlling authority” that would make reconsideration appropriate, such argument is clearly incor- rect. Out-of-circuit cases are not binding on this Court and therefore do not constitute “controlling authority.” Defendants have cited no case to the contrary.

[6] Moreover, although Hayden, Johnson, and Simmons created a circuit split with our decision in Farrakhan I, we do not agree that those decisions demonstrate that Farrakhan I was “clearly erroneous.” First, both Hayden and Johnson were rendered over vigorous dissents. See Hayden, 449 F.3d at 343-62 (Parker, J., dissenting, joined by Calabresi, Pooler, and Sotomayor, JJ.); id. at 362-67 (Calabresi, J., dissenting); id. at 367-68 (Sotomayor, J., dissenting); id. at 368-69 (Katz- mann, J., dissenting); Johnson, 405 F.3d at 1239-44 (Wilson, J., dissenting in relevant part); id. at 1247-51 (Barkett, J., dis- senting).16 Thus, even if we assume that Farrakhan I was erroneous, such error was hardly “clear,” given the vigorous dissenting opinions in the First, Second, and Eleventh Cir- 15 Since this case was argued and submitted for decision, the First Cir- cuit has also held that the VRA does not apply to felon disenfranchisement laws. See Simmons v. Galvin, 575 F.3d 24 (1st Cir. 2009). 16 Simmons, too, was filed over a vigorous dissent. See 575 F.3d at 45 (Torruella, J., dissenting). FARRAKHAN v. GREGOIRE 127 cuits. Second, Farrakhan I was called en banc but failed to attract a majority vote of the nonrecused active judges in favor of en banc rehearing. Farrakhan v. Washington, 359 F.3d 1116 (9th Cir. 2004) (denying petition for rehearing en banc). That a majority of this Court’s active judges did not consider Farrakhan I worthy of en banc rehearing also sup- ports a conclusion that the decision was not “clearly errone- ous.” Cf. Jeffries, 114 F.3d at 1493 (holding that a three-judge panel “should not have exercised its discretion to depart from its prior decision” in part because “further appellate review of [that decision] was sought and denied prior to the panel’s change of heart”).17 Finally, although it did not directly address the question whether challenges to felon disenfran- chisement laws are cognizable under VRA § 2, the Sixth Cir- cuit treated them as such when it decided a § 2 vote dilution challenge to Tennessee’s felon disenfranchisement law. See Wesley v. Collins, 791 F.2d 1255, 1259-62 (6th Cir. 1986). Taking Wesley into account, there is a close split among the circuits that have faced VRA challenges to felon disenfran- chisement laws on whether such challenges are cognizable, lending further support to the conclusion that Farrakhan I cannot be considered “clearly erroneous” for the purpose of departing from the law of the case.

[7] We thus conclude that Farrakhan I remains binding Circuit law.

C. Standing

Defendants next argue that Plaintiffs “lack standing to claim that they were denied the right to vote on account of race” because they have not shown that their own felony con- victions were the result of racial discrimination. Defendants misconstrue the requirements for Article III standing. 17 Like the panel decision at issue in Jeffries, see 114 F.3d at 1493, Far- rakhan I was denied both en banc rehearing, Farrakhan, 359 F.3d 1116, and a writ of certiorari from the Supreme Court, 543 U.S. 984. 128 FARRAKHAN v. GREGOIRE [8] To establish Article III standing, Plaintiffs must demon- strate: (1) that they have suffered an injury in fact that is both “concrete and particularized” and “actual and imminent,” (2) that the injury is fairly traceable to the challenged action, and (3) that a decision in Plaintiffs’ favor would likely redress the injury. Lujan v. Defenders of Wildlife, 504 U.S. 555, 560-61 (1992). That test is easily satisfied here. Plaintiffs have suf- fered an injury in fact that is concrete, particularized, and actual: they have been denied the right to vote. That injury is directly traceable to the challenged action: Washington’s felon disenfranchisement law. And a decision invalidating Washington’s felon disenfranchisement provision would redress Plaintiffs’ injury: it would restore their right to vote.

The State attempts to import a merits question — that is, a question regarding whether plaintiffs can prove a violation — into the standing inquiry. This is incorrect. See Warth v. Seldin, 422 U.S. 490, 500 (1975) (“[S]tanding in no way depends on the merits of the plaintiff’s contention that partic- ular conduct is illegal.”). Standing is a threshold question, the purpose of which is to ensure that there is an actual “case or controversy” and that the plaintiff is the correct party to bring suit. See id. at 498-99. Whether Plaintiffs can succeed on their VRA claim is irrelevant to the question whether they are enti- tled to bring that claim in the first place.18 18 Moreover, as Plaintiffs correctly point out, they need not show that their own convictions were the result of racial discrimination to succeed on a § 2 vote denial claim. First, a § 2 claim focuses on the effect of the challenged practice on minority voters as a class, rather than on the dis- crimination faced by the plaintiff in a given case. See 42 U.S.C. § 1973(b) (“A violation of subsection (a) of this section is established if . . . it is shown that the political process leading to nomination or election in the State or political subdivision are not equally open to participation by mem- bers of a class of citizens protected by subdivision (a) . . . .” (emphasis added)). Second, in amending § 2, Congress expressly eliminated the requirement that plaintiffs raising § 2 claims prove intentional discrimina- tion. See S. Rep. No. 97-417, at 16 (“[P]roof of a discriminatory purpose should not be a prerequisite to establishing a violation of Section 2 of the Voting Rights Act.”); id. at 28 (“[T]he specific intent of this amendment is that the plaintiffs may choose to establish discriminatory results without proving any kind of discriminatory purpose.”). FARRAKHAN v. GREGOIRE 129 [9] In any event, neither this Court nor the other circuits that have considered vote denial claims under § 2 have ever held that a plaintiff lacked standing because he or she did not allege that he/she had been personally discriminated against. See Hayden, 499 F.3d 305; Johnson, 405 F.3d 1214; Farrak- han I, 338 F.3d 1009; Salt River, 109 F.3d 586. Because Plaintiffs have alleged an injury in fact that is traceable to the Washington law and can be redressed by a favorable ruling, we reject Defendants’ argument that Plaintiffs lack standing.

D. The cross-motions for summary judgment

1. The summary judgment standard

[10] “Summary judgment is appropriate where no genuine issue of material fact exists and a party is entitled to prevail in the case as a matter of law.” Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 250 (1986); see also Fed. R. Civ. P. 56(c). Here, the parties agree that there are no disputed material facts. Indeed, “[a]lthough summary judgment rules provided [Defendants] with an opportunity to respond to [Plaintiffs’] materials, [Defendants] did not offer any fact-based or expert- based refutation in the manner the rules provide.” See Beard v. Banks, 548 U.S. 521, 534 (2006) (plurality opinion) (citing Fed. R. Civ. P. 56(e)). Federal Civil Rule 56(e)(2) provides that “[w]hen a motion for summary judgment is properly made and supported, an opposing party may not rely merely on allegations or denials in its own pleading; rather, its response must — by affidavits or as otherwise provided in this rule — set out specific facts showing a genuine issue for trial. If the opposing party does not so respond, summary judgment should, if appropriate, be entered against that party.” Likewise, Rule 56.1(b) of the Local Rules of the East- ern District of Washington (“Local Rule”) provides that “[a]ny party opposing a motion for summary judgment must file with its responsive memorandum a statement . . . setting forth the specific facts which the opposing party asserts estab- lishes a genuine issue of material fact precluding summary 130 FARRAKHAN v. GREGOIRE judgment. Each fact must explicitly identify any fact(s) asserted by the moving party which the opposing party dis- putes or clarifies.” If the moving party’s statement of facts are not controverted in this manner, “the Court may assume that the facts as claimed by the moving party are admitted to exist without controversy.” Local Rule 56.1(d); see also Beard, 548 U.S. at 527.

[11] Here, Defendants failed specifically to challenge the facts identified in Plaintiffs’ statement of undisputed facts as required by the rules. Defendants did, in their Supplemental Statement of Material Facts, raise some questions about Plain- tiffs’ expert’s reports, but those questions were not supported by affidavit or counter-experts. Moreover, none of the ques- tions raised by Defendants in their Supplemental Statement contradicts, or even suggests that there is some dispute about the ultimate conclusions of Plaintiffs’ experts’ reports. There- fore, “by failing specifically to challenge the facts identified in [Plaintiffs’] statement of undisputed facts, [Defendants are] deemed to have admitted the validity of the facts contained in the [Plaintiffs’] statement.” Beard, 548 U.S. at 527.

In addition to failing to challenge any of Plaintiffs’ facts in the manner required by Rule 56(e) and Local Rule 56.1, Defendants insisted before the district court that “[N]o ques- tion of material fact remains in this case; and therefore, this case is ready to be ruled upon at summary judgment.” Just as Defendants’ counsel insisted before the district court that there were no disputes of material facts, Defendants’ counsel at oral argument before this Court repeatedly insisted that there were no disputes of material fact. Therefore, Defendants do not, and have not, disputed any of Plaintiffs’ factual asser- tions, including the assertions put forth by Plaintiffs’ experts, either in their briefing before this Court or during oral argu- ment. Instead, Defendants have stated repeatedly both before the district court and this Court that no question of material FARRAKHAN v. GREGOIRE 131 fact exists and that this case is ready to be ruled upon at sum- mary judgment.19

Defendants do argue that the district court erred in conclud- ing there is discrimination in Washington’s criminal justice system on account of race because, according to Defendants, Plaintiffs’ evidence of racial bias in Washington’s criminal justice system “is very limited,” and is inadequate to demon- strate that even Senate Factor 5 favors Plaintiffs’ claims as a matter of law. In other words, Defendants argue that they are entitled to summary judgment because even accepting Plain- tiffs’ evidence as uncontroverted, it fails as a matter of law to demonstrate that the felon disenfranchisement law violates § 2.

When a moving party without the ultimate burden of per- suasion at trial demonstrates that it is entitled to prevail as a matter of law by showing that the nonmoving party has not adduced sufficient evidence of an essential element to carry its ultimate burden of persuasion at trial, the moving party is entitled to summary judgment. See Nissan Fire & Marine Ins. Co. v. Fritz Cos., 210 F.3d 1099, 1102 (9th Cir. 2000). Put differently, when the nonmoving party has the burden of proof at trial, as Plaintiffs do here, the party moving for sum- mary judgment, in this case the State, need only point out that there is an absence of evidence to support the nonmoving party’s case. See Devereaux v. Abbey, 263 F.3d 1070, 1076 (9th Cir. 2001) (en banc). If, on the other hand, the State fails to demonstrate that there is an absence of evidence to support 19 For example, at oral argument, Defendants “disagree[d]” with the panel’s suggestion that there may be a dispute about a material fact; insisted that there is no issue of material fact for trial because Defendants do not dispute the conclusions of Plaintiffs’ expert reports, but instead challenge their legal adequacy; stated that either Defendants win on sum- mary judgment or Plaintiffs win on summary judgment because Plaintiffs’ evidence is “inadequate” to meet Plaintiffs’ burden; and that “[t]here are no factual issues.” 132 FARRAKHAN v. GREGOIRE Plaintiffs’ case, then the State’s summary judgment motion must be denied.

As the Supreme Court has noted, Defendants’ litigation strategy is a perilous one. “It has always been perilous for the opposing party neither to proffer any countering evidentiary material nor file a 56(f) affidavit. And the peril rightly contin- ues after the amendment to Rule 56(e).” See Adickes v. Kress & Co., 398 U.S. 144, 161 (1970) (internal citation, quotation marks, and brackets omitted). Declining to “offer any evi- dence opposing summary judgment . . . is not the recom- mended approach when the opposing party feels that the movant has not met his burden. . . . [I]n most cases the better response to a summary-judgment motion is not simply to test the sufficiency of the movant’s case by challenging the legal sufficiency of the evidence presented on the motion, but to introduce contradictory evidence to establish that a question of material fact remains in dispute.” 10A Charles Alan Wright et al., Federal Practice and Procedure Civil 3d § 2727, at 516 (1998). Nevertheless, that is the approach the Defendants have pursued, and we are charged with deciding this case in the procedural posture and on the record evidence as it is brought before us.

[12] Plaintiffs, on the other hand, would be entitled to sum- mary judgment based upon their motion if they make out a prima facie case that would entitle them to judgment as a mat- ter of law if uncontroverted at trial. See UA Local 134 United Ass’n of Journeymen & Apprentices of Plumbing & Pipefit- ting Indus. of U.S. & Canada, AFL-CIO, 48 F.3d at 1471; see also 10A Wright et al., supra, § 2727, at 486. Given that Defendants have not adduced any evidence to show that there is a genuine issue for trial, Plaintiffs need only demonstrate that their uncontroverted evidence entitles them to judgment as a matter of law. See Ritchie v. United States, 451 F.3d 1019, 1023 (9th Cir.2006) (“[J]udgment as a matter of law is appropriate only if no reasonable jury could find in favor of the non-moving party.”). FARRAKHAN v. GREGOIRE 133 [13] Thus, because the parties agree that the facts are uncontroverted and agree further that the only question left for the court is to determine the legal significance of those facts and the reasonable inferences to be drawn from them, we conclude that summary judgment in this case is appropri- ate:

The fact that difficult questions of law exist or that parties differ on the legal conclusion to be drawn from the facts is not in and of itself a ground for denying summary judgment inasmuch as refusing to grant the motion does not obviate the court’s obliga- tion to make a difficult decision; a denial merely postpones coming to grips with the problem at the cost of engaging in a full-dress trial that is unneces- sary for a just adjudication of the dispute. Therefore, when the only question is what legal conclusions are to be drawn from an established set of facts, the entry of summary judgment usually should be directed.

10A Wright et al., supra § 2725, at 411-12; see also Farrak- han, 359 F.3d at 1117 (Kozinski, J., dissenting from denial of rehearing en banc) (“[T]he record is settled. . . . No triable issues of fact remain.”); Smith v. Califano, 597 F.2d 152, 155 n.4 (9th Cir. 1979) (“The parties here have agreed on the material facts, the dispute involving the proper interpretation of relevant statutes and regulations. Because the case could thus be resolved as a matter of law, summary judgment was the proper procedural device.”); cf. Aramark Facility Servs. v. Serv. Employees Int’l Union, Local 1877, AFL-CIO, 530 F.3d 817, 822 (9th Cir. 2008) (“Here, the district court resolved the matter on the parties’ cross-motions for summary judgment, which necessarily present questions of law.”). Thus, the legal question presented to us, although difficult, is a straightfor- ward one: Have Plaintiffs demonstrated a prima facie case that the felon disenfranchisement law violates § 2 of the VRA, i.e., that: (1) there are significant statistical racial dis- 134 FARRAKHAN v. GREGOIRE parities in the operation of the criminal justice system; (2) those disparities cannot be explained in race-neutral ways; and (3) those non-race-neutral disparities in the criminal jus- tice system lead to significant racial disparities in the qualifi- cation to vote, such that Plaintiffs would be entitled to judgment as a matter of law based upon the uncontroverted evidence?

2. The merits of the cross-motions

Defendants’ summary judgment motion is premised on Plaintiffs’ having failed to produce sufficient evidence to establish a § 2 vote denial claim. The nub of Defendant’s argument is that all of the Senate Factors are relevant to Plain- tiffs’ vote denial claim; that the district court was correct to consider them in its totality-of-the-circumstances analysis; and that, because Plaintiffs failed to produce probative evi- dence relating to Senate Factors other than 5, Plaintiffs failed to produce sufficient evidence to make out a § 2 vote denial claim. Thus, Defendants contend, they were entitled to sum- mary judgment.

Plaintiffs, on the other hand, contend that, having con- cluded that racial discrimination exists in the Washington criminal justice system (Factor 5), the district erred in then requiring Plaintiffs to produce evidence regarding other Sen- ate Factors not relevant to their vote denial claim. While those factors may be pertinent to a vote dilution claim, Plaintiffs contend that they “cast no light on Plaintiffs’ vote denial chal- lenge.”

We agree with Plaintiffs for the reason that, given the strength of their Factor 5 showing, the district court erred in requiring them to prove Factors that had little if any relevance to their particular vote denial claim. Although the district court was required to consider the “totality of the circum- stances,” not all of the Senate Factors were equally relevant, or even necessary, to that analysis in this case. Some Senate FARRAKHAN v. GREGOIRE 135 Factors may be relevant as circumstantial evidence with respect to certain vote denial claims, but proof of those Fac- tors was not required where, under Factor 5, Plaintiffs pro- vided strong, indeed “compelling,” direct evidence of the alleged violation. There is indeed, as the Senate Report stressed, no requirement that any particular number of Factors support a particular claim. S. Rep. No. 97-417, at 29. Even one may be enough in some instances.20

We first address the district court’s treatment of the various Senate Factors to explain why that treatment was erroneous. We then consider whether the evidence produced by Plaintiffs was sufficient to preclude a grant of summary judgment to Defendants. Finally, we consider whether Plaintiffs were enti- tled to summary judgment.

a. The district court’s treatment of the Senate Factors

i. Senate Factors 7 and 8

[14] In its listing of the Factors that typically may be rele- vant to a § 2 claim, the Senate Report made clear that “there is no requirement that any particular number of Factors be proved or that a majority of them point one way or the other.” S. Rep. No. 97-417, at 29; see id. at 29 n.118 (stating that the Factors were not intended “to be used[ ] as a mechanical ‘point-counting’ device”); see also Gomez, 863 F.2d at 1412 (noting the Senate Report’s emphasis that the “list of factors 20 Contrary to what the dissent contends, we do not “dictat[e] that a dis- trict court should not consider certain factors . . . in vote denial cases.” See Diss. at 160. We hold only that different factors will be of relevance in dif- ferent cases, depending on the circumstances of those cases; that courts should consider each factor in light of the circumstances of the case before them; and that where, as here, plaintiffs provide compelling evidence of a law or system of laws that, as implemented, necessarily results in the discriminatory deprivation of racial minorities’ right to vote, that depriva- tion is sufficient, and the plaintiffs need not present additional evidence regarding other factors that are of less relevance to the plaintiffs’ claim. 136 FARRAKHAN v. GREGOIRE was not a mandatory seven-pronged test” but “only meant as a guide to illustrate some of the variables that should be con- sidered by the court”). Thus, “while the basic ‘totality of the circumstances’ test remains the same, the range of factors that [are] relevant in any given case will vary depending upon the nature of the claim and the facts of the case.” Gomez, 863 F.3d at 1412. Where the evidence of one central Factor in a particular case is compelling, that Factor may be sufficient. Moreover, as the Supreme Court has recognized, the enumer- ated Factors are “particularly [pertinent] to vote dilution claims,” Thornburg, 478 U.S. at 45, and, it follows, not as pertinent, generally, in vote denial cases. Thus, in vote denial cases, there is even more flexibility in determining whether, under the totality of the circumstances test, a single factor is controlling and whether any weight may or should be given to the presence or absence of others.

The district court acknowledged that it was “not bound by the list of Senate factors,” but found that several of the Fac- tors were relevant to Plaintiffs’ vote denial challenge. Farrak- han, 2006 WL 1889273, at *7. Specifically, the district court found that Factors 7 and 8 — the extent of minority represen- tation among elected officials, and the level of responsiveness of elected officials to minorities’ needs — were “certainly rel- evant to Plaintiffs’ VRA claim.” Id. at *8.21 We conclude, however, that, in light of its finding of “compelling evidence of racial discrimination and bias in Washington’s criminal justice system,” the district court erred in according any weight to Plaintiffs’ failure to introduce evidence regarding 21 The district court also noted that Plaintiffs had “failed to present any substantial evidence regarding” Senate Factors 2, 3, 4, and 6. Id. However, it then “admitted[ ]” that “several of these factors are not relevant in a VRA vote denial claim.” Id. Because we interpret this to mean that the district court did not rely on these factors in its totality of the circum- stances analysis, we do not address these factors. In any event, we agree with the district court that these factors are not relevant to Plaintiffs’ vote denial claim. Therefore, to the extent the district court did weigh these fac- tors in its analysis, we conclude that it erred. FARRAKHAN v. GREGOIRE 137 Factors 7 and 8. These factors are not essential to a § 2 vote denial claim and in this case, while their presence might be of some relevance, their absence is insufficient cause to justify in any respect the denial of Plaintiffs’ claim.

To understand which Senate Factors might be relevant to deciding a vote denial claim, it is important to recognize the analytical distinction between vote denial and vote dilution theories. A vote dilution claim does not allege that minority voters are denied access to the polls; rather, the claim is that, although minority voters have the formal right to vote, the challenged voting scheme “operates to minimize or cancel out [the minority voters’] ability to elect their preferred candi- dates.”22 Thornburg, 478 U.S. at 48. In other words, the focus of a vote dilution challenge is on the effectiveness of the minority plaintiffs’ votes. Naturally then, the Factors most rel- evant to a vote dilution claim are those that examine whether minorities have the capacity to be politically influential as a group, and, if so, whether their political influence has been weakened — for example, whether the minority group is politically cohesive, whether the white majority votes in a bloc, whether voting is racially polarized, whether minorities have succeeded in being elected to public office, and whether elected officials have been responsive to the particularized needs of the minority group. See Thornburg, 478 U.S. at 48 & n.15.

Vote denial claims, in contrast, challenge laws, as amici point out, “that directly exclude otherwise qualified voters from participating.” Whereas vote dilution claims “implicate the value of aggregation,” vote denial claims “implicate the value of participation.” Tokaji, supra, at 718 (emphasis added). Thus, the primary question in such cases is not whether a “denial or abridgement” occurs, but whether such 22 “Chief examples of vote-dilution practices include at-large elections and redistricting plans to keep minorities’ voting strength weak.” Tokaji, supra, at 691. 138 FARRAKHAN v. GREGOIRE denial is “on account of race.” In vote denial claims brought under the “results test,” the “on account of” element is proved by showing that a “discriminatory impact . . . is attributable to racial discrimination in the surrounding social and histori- cal circumstances.”23 Farrakhan I, 338 F.3d at 1019. Conse- quently, factors that examine the political strength of minority voters in the jurisdiction are of lesser relevance.

Given the analytical distinction between vote dilution and vote denial, it is clear that Senate Factors 7 and 8, while rele- vant to the former,24 are of lesser relevance to a vote denial claim. The “extent to which members of the minority group have been elected to public office in the jurisdiction” (Senate Factor 7) simply has no bearing on the question whether minorities are being denied the right to vote “on account of race.” Even if a majority of the elected officials in the juris- diction were members of the minority group, it would still violate § 2 to deny minority citizens the right to vote on dis- criminatory grounds. The fact that minority candidates have had success in the state does not cure the discriminatory denial of the franchise to minority voters.25 Likewise, whether elected officials have been responsive to “the particularized 23 In the challenge under review, to felon disenfranchisement laws, the “social circumstance” is the operation of the criminal justice system. See Farrakhan I, 338 F.3d at 1012, 1019-20. In Salt River, the social circum- stance at issue was land ownership. See 109 F.3d at 589. 24 The Supreme Court has, in fact, stated that Senate Factor 7 is an essential factor in a vote dilution challenge — that is, a factor that must be proved for the plaintiffs to succeed. See Thornburg, 478 U.S. at 48 n.15. As for Senate Factor 8, the Supreme Court explained that, while proving that factor “might be supportive of a [vote dilution] challenge,” it is “ ‘not essential to’ such a claim.” Gomez, 863 F.3d at 1413 (quoting Thornburg, 478 U.S. at 48 n.15) (emphasis in original). 25 The Senate Report strongly indicates that the Senate Committee included Senate Factor 7 to help § 2 plaintiffs prove vote dilution claims. See S. Rep. No. 94-417, at 29 n.115. This lends further support to the con- clusion that although it may help them do so, the plaintiffs in a vote denial case are not required to produce evidence supporting Factor 7, and may not be penalized for failing to do so. FARRAKHAN v. GREGOIRE 139 needs of the members of the minority group” (Factor 8) may be probative of the minorities’ ability to influence the political process, but generally does not indicate whether minorities are being denied access to the polls on account of their race. If minorities are disproportionately deprived of their right to vote, and if that disparity is caused by racial discrimination, then whether the elected officials have been responsive to minority issues is simply of little relevance.26

[15] Accordingly, the district court erred in concluding that Plaintiffs’ “failure to produce any evidence” as to Factors 7 and 8 provided any support for its grant of summary judgment to Defendants. Farrakhan, 2006 WL 1889273, at *8. Plain- tiffs’ failure to produce evidence regarding those factors is without legal significance because proof relating to them is not necessary to establish a vote denial claim. This is espe- cially so in a case in which a “compelling” showing of dis- crimination has been made. Defendants, while contending that the district court was correct to rely on the absence of evi- dence regarding Factor 7 and Factor 8, do not even attempt to explain why such evidence is relevant to Plaintiffs’ vote denial claim. Their unsupported assertion that all of the Sen- ate Factors are “relevant” does not make them so.27 26 Moreover, as with Senate Factor 7, Congress made clear that proving Factor 8 “is not an essential part of plaintiffs’ case.” S. Rep. No. 94-417, at 29 n.116. Even “Defendants’ proof of some responsiveness would not negate Plaintiffs’ showing by other, more objective factors enumerated here that minority voters nevertheless were shut out of equal access to the political process.” Id. Thus, the Senate Report makes clear that Factor 8 cannot negate Plaintiffs’ showing that the disproportionate disenfranchise- ment of minority voters in Washington is caused by racial discrimination in the state’s criminal justice system. 27 Because “the ingenuity of such schemes” to deny minorities the right to vote “seems endless,” S. Rep. No. 97-417, at 6, we do not imply that Senate Factors 7 and 8 are never relevant to establishing vote denial claims — only that the absence of such evidence may not serve as a justi- fication for denying them. As we have noted, in cases in which the evi- dence of discrimination in the surrounding social and historical 140 FARRAKHAN v. GREGOIRE ii. Senate Factor 1

Plaintiffs also argue that the district court erred in placing “near-dispositive weight” on Senate Factor 1 (“extent of any history of official discrimination in the state” in the area of voting). We agree.

[16] The district court misperceived the relationship between Factor 1 and § 2 vote denial claims. Although Factor 1 may be supportive of a § 2 vote denial claim28 — especially where the plaintiff alleges that the voting qualification itself is discriminatory — proving Factor 1 is not necessary to suc- ceed on such a challenge. Cf. Thornburg, 478 U.S. at 48 n.15 (distinguishing between factors that are “essential” to proving a vote dilution claim and factors that are “supportive of, but not essential to,” such a claim) (emphasis in original). Show- ing that a state has a history of discriminating against minority voters can strongly support an argument that the state voting qualification being challenged was enacted with a discrimina- tory purpose. The failure to show that a state has a history of discriminatory voting practices, however, does not negate a showing that the current voting practice at issue is discrimina- tory.

[17] Plaintiffs do not contend that Washington’s felon dis- enfranchisement law was enacted with a discriminatory pur- pose; their claim, rather, is that the provision interacts with a racially discriminatory criminal justice system and, as a result,

circumstances is less “compelling,” these factors may provide circumstan- tial evidence that the disparate impact of a particular practice on minority voters is attributable to such discrimination. Here, we merely hold that, where plaintiffs provide direct evidence of racial discrimination under Factor 5, the absence of evidence regarding Factors 7 and 8 is irrelevant to the district court’s totality of the circumstances analysis. 28 In this way, Factor 1 differs from Factors 7 and 8, which are primarily relevant to vote dilution claims and generally do not lend as high a level of support to a § 2 vote denial claim. FARRAKHAN v. GREGOIRE 141 racial minorities are disproportionately denied the right to vote. If Plaintiffs adduce evidence that the disproportionate disenfranchisement in Washington is attributable to discrimi- nation in the criminal justice system, they may show a viola- tion of § 2 under the “results test” that was sufficient to survive summary judgment. Here, in fact, the district court found the evidence “compelling.” That Washington has not historically discriminated against minorities in voting does not negate a showing that this voting law has a discriminatory result. See Tokaji, supra, at 721 (“A court does not need to rely on . . . circumstantial evidence . . . when there is direct evidence that an electoral process has the result of dispropor- tionately denying minority votes.” (emphasis in original)).

This conclusion draws support from our precedent. In Gomez, the district court denied a § 2 vote dilution challenge to the city’s at-large election scheme. After determining that the district court had erred in its application of the factors “es- sential” to such a claim, we assessed the district court’s treat- ment of the “other factors,” including Factor 1. 863 F.2d at 1417-19. Although we were “troubled” by the district court’s conclusion that there had been no official discrimination against Hispanics, we did not believe “that the district court had committed clear error” in so concluding. Id. at 1418. Nev- ertheless, we concluded that “even without such a showing, plaintiffs have clearly established a violation of Section 2.” Id. at 1419 (emphasis added). Once the factors “essential” to plaintiffs’ vote dilution claim had been satisfied, it made no difference that they had not proved a history of official dis- crimination in voting.

[18] As in Gomez, Plaintiffs here established a violation of § 2 by adducing evidence sufficient to establish a vote denial claim — that “there is discrimination in Washington’s crimi- nal justice system on account of race,” Farrakhan, 2006 WL 1889273, at *6, and that such discrimination “clearly hinder[s] the ability of racial minorities to participate effec- tively in the political process,” id. (quoting Farrakhan I, 338 142 FARRAKHAN v. GREGOIRE F.3d at 1220 (internal quotation mark omitted) (alteration in original)). Plaintiffs’ evidence of racial discrimination in the Washington justice system was, the district court states, “compelling.” Accordingly, Plaintiffs were not required to produce further circumstantial evidence, and the district court’s conclusion that Plaintiffs’ failure to prove Senate Fac- tor 1 “strongly favors a finding that Washington’s felon disen- franchisement law does not violate § 2 of the VRA” was erroneous.

iii. Senate Factor 9

Finally, Plaintiffs contend that the district court’s conclu- sion that Senate Factor 9 favors Defendants was erroneous. Because, under the totality of the circumstances test, Plaintiffs established a § 2 violation based on the district court’s finding of racial discrimination in Washington’s criminal justice sys- tem, it does not matter whether, as Plaintiffs claim, the state’s policy justification for felon disenfranchisement is tenuous.

[19] Like Factor 1, Factor 9 is a factor that could support Plaintiffs’ vote denial claim circumstantially but is not neces- sary to proving it. This conclusion draws direct support from the Senate Report’s discussion of Factor 9, which explains that “even a consistently applied practice premised on a racially neutral policy would not negate a plaintiff’s showing through other factors that the challenged practice denies minorities fair access to the process.” S. Rep. No. 97-417, at 29 n.117. It is also in line with Congress’ express objective in amending § 2 of “broaden[ing] the protection afforded by the Voting Rights Act.” Chisom, 501 U.S. at 404. Under this approach, the district court’s finding that Factor 9 “favors the defendants’ position” is erroneous. If Plaintiffs can prove that the denial of their right to vote was “on account of” race, it did not matter whether the state’s policy reasons were tenuous — a § 2 violation had been established. Accordingly, we hold that the district court erred in concluding that Plaintiffs’ fail- ure to demonstrate the tenuousness of the state’s felon disen- FARRAKHAN v. GREGOIRE 143 franchisement policy weighed against finding a § 2 violation; to the contrary, in this case Factor 9 was simply neutral.

b. Plaintiffs’ evidence that vote denial is “on account of race”

[20] Ultimately then, the plaintiff’s burden in any § 2 case is to prove that the challenged voting qualification “results in a denial or abridgement of the right of any citizen of the United States to vote on account of race or color.” 42 U.S.C. § 1973(a). In the case of automatic felon disenfranchisement, there is no question that the challenged provision constitutes a denial of the right to vote. Consequently, the sole remaining issue is causation — whether the denial of the right to vote is “on account of race or color.” As we explained in Farrakhan I, the “on account of” requirement may be met “where the dis- criminatory impact of a challenged voting practice is attribut- able to racial discrimination in the surrounding social and historical circumstances,” which include the state’s criminal justice system. Farrakhan I, 338 F.3d at 1019-20.

Here, the district court repeatedly declared that Plaintiffs have presented “compelling” evidence of racial discrimination in Washington’s criminal justice system. Indeed, after consid- ering Plaintiffs’ evidence, the district court concluded that it “has no doubt that members of racial minorities have experi- enced racial discrimination in Washington’s criminal justice system.” Farrakhan, 2006 WL 1889273, at *9.

[21] Based on the uncontroverted facts, we reach the same conclusion as the district court. The expert reports, which were not refuted by the State, provide compelling circumstan- tial evidence of discrimination in Washington’s criminal jus- tice system. Dr. Crutchfield’s report states that criminal justice practices disproportionately affect minorities beyond what can be explained by non-racial means. For example, African Americans in Washington State were over nine times more likely to be in prison than Whites, even though the ratio 144 FARRAKHAN v. GREGOIRE of Black to White arrest for violent offenses was only 3.72:1, suggesting that substantially more than one half of Washing- ton State’s racial disproportionality in its criminal justice sys- tem cannot be explained by higher levels of criminal involvement as measured by violent crime arrest statistics. A study of the Washington State Patrol shows that Native Amer- icans were more than twice as likely to be searched as Whites; African Americans were more than 70 percent more likely to be searched than Whites; and Latinos were more than 50 per- cent more likely to be searched. A study of the Vancouver, Washington Police Department (“VPD”) indicated that of those stopped for traffic violations by the VPD, African Americans are nearly twice as likely to be searched as Whites, and Latino were three times more likely to be searched. This, despite the fact that searches of Whites more frequently resulted in the seizure of contraband than searches of African Americans and Latinos. According to Dr. Crutchfield, these findings suggest that African Americans and Latinos are at greater risk for searches that could lead to felony charges, but because those searches are less fruitful then searches against Whites, it is likely that minorities are being placed at greater risk for no legitimate purpose.

Dr. Crutchfield also indicated that the significant racial dis- parities in arrest rates are not fully warranted by race or ethnic differences in illegal behavior. The Seattle Police Department (“SPD”) arrested African Americans and Latinos for drug possession at rates much higher than their proportion among users. Whites, on the other hand, were arrested for drug pos- session at rates much lower than their proportion among users. The most significant cause of the racial disparity in Seattle drug arrests resulted from the SPD’s focus on crack cocaine, a focus that, according to Dr. Crutchfield, cannot be justified by drug use or distribution patterns.

Dr. Crutchfield also reported that charging and bail prac- tices are infected with racial disparities. Whites are less likely to have charges filed than minorities, a significant disparity FARRAKHAN v. GREGOIRE 145 that persists even after a number of legally relevant character- istics, such as offense seriousness, offenders’ criminal histo- ries, and weapons charges, are taken into account. Minority defendants were less likely to be released on their own recog- nizance than others, even after adjusting for differences among defendants in the severity of their crimes, prior crimi- nal records, ties to the community, and the prosecuting attor- ney’s recommendation. Whether defendants are released on their own recognizance, as opposed to being required to post bail, is important because defendants released on their own recognizance are likely to receive more lenient treatment in both charging and sentencing. Thus, to the extent that minori- ties are disadvantaged in pre-trial release, this has real poten- tial for contributing to disparities in felony conviction rates.

Likewise, Dr. Beckett reported that the disparity between whites and minorities (specifically, blacks and Latinos) in drug possession and delivery arrests is largely the result of three organizational practices — the police’s focus on crack cocaine, on outdoor drug venues, and on the downtown area — that are not “explicable in race-neutral terms.” Beckett Report at 2. Dr. Beckett stated that the focus on crack cannot be explained by the frequency of its exchange, by the level of violence in the crack market, or by the health problems asso- ciated with crack as opposed to other serious drugs, such as cocaine. Id. at 10-12. She also reported that the focus on out- door drug activity cannot be explained by either greater citi- zen complaints or greater yield from such arrests, concluding that the outdoor focus is an “(ineffecient) policy choice” rather than “an organizational or legal necessity.” Id. at 13. Finally, Dr. Beckett explained that the concentration of law enforcement resources downtown is out of proportion to the level of drug crime there and is also not explainable vis-a-vis citizen complaints. Id. at 21-23.

[22] On this uncontroverted record, the district court found that “there is discrimination in Washington’s criminal justice system on account of race,” Farrakhan, 2006 WL 1889273, 146 FARRAKHAN v. GREGOIRE at *6, and that such discrimination “clearly hinder[s] the abil- ity of racial minorities to participate effectively in the political process, as disenfranchisement is automatic,” id. (quoting Farrakhan I, 338 F.3d at 1020). Having so found, the district court should not have required Plaintiffs to produce additional circumstantial evidence; they had presented evidence that, if accepted by a finder of fact, would establish a § 2 violation under the totality of the circumstances. Thus, the district court erred in granting Defendants summary judgment.

c. Defendants’ challenges to the district court’s legal conclusions with respect to Plaintiffs’ evidence

Defendants contend, however, that the district court erred in the conclusions it drew from the evidence adduced by Plaintiffs with respect to Factor 5. For that reason, they say, its result was correct, although its analysis was wrong. Defen- dants contend that the district court erred in finding that the Washington criminal justice system was racially discrimina- tory, and that, in the absence of such a determination, no basis whatsoever exists for challenging the felon disenfranchise- ment law. We agree that Plaintiffs’ challenge is founded on the premise that Washington’s criminal justice system is racially discriminatory and that, in the absence of evidence supporting that claim, Plaintiffs’ § 2 challenge would fail. We disagree, however, with Defendants’ contention that the dis- trict court erred in its conclusion that Plaintiffs introduced “compelling evidence of racial discrimination and bias in Washington’s criminal justice system.”

Specifically, Defendants contend that the district court committed three distinct legal errors in analyzing Plaintiffs’ evidence of racial discrimination. None of these arguments has merit.

[23] First, Defendants argue that the district court erred as a matter of law in extrapolating Dr. Beckett’s Seattle-specific findings to the whole of Washington state. However, it was FARRAKHAN v. GREGOIRE 147 not unreasonable to draw inferences from Dr. Beckett’s Seattle-specific findings. Dr. Crutchfield reported that “a large proportion of the minority population of Washington State resides in the City of Seattle or in the surrounding county, King County.” Crutchfield Report at 15; see also id. at 27 (stating that “King County has the largest minority pop- ulation in the state and contains the state’s most diverse city, Seattle, so it is an opportune location in which to complete a study of racial and ethnic disparities in the prosecution of criminal cases”). Given that much of the state’s minority pop- ulation resides in Seattle, it was reasonable for the district court to look to a Seattle-focused study in assessing racial dis- crimination in the state as a whole. Indeed, as Dr. Crutchfield reported, counties “with smaller minority populations were likely to produce larger racial disparities” in imprisonment, which suggests that the district court’s extrapolation from a Seattle-based study actually underestimated the racial dis- crimination in the state as a whole. And, as we have noted, Defendants presented no evidence to counter either Dr. Crutchfield’s or Dr. Beckett’s findings. Thus, the district court did not err in extrapolating the Seattle findings to the state as a whole.

Second, Defendants contend that the district court erred in relying on statistical disparity alone, in contravention of Salt River. This is plainly incorrect. To be sure, Salt River made clear that “a bare statistical showing of disproportionate impact on a racial minority does not satisfy the § 2 ‘results’ inquiry” because causation cannot be inferred from impact alone. 109 F.3d at 595. In Salt River, the plaintiffs challenged a voting qualification which required voters to own property in order to be eligible to vote. The Salt River plaintiffs, how- ever, demonstrated only that “proportionately fewer African- Americans than non-Hispanic whites residing in the [voting] District live in owner-occupied homes.” 109 F.3d at 590. The plaintiffs “stipulated to the nonexistence of virtually every cir- cumstance which might indicate that landowner-only voting results in racial discrimination,” id. at 595, and the district 148 FARRAKHAN v. GREGOIRE court concluded (and this Court agreed) that “the observed difference in rates of home ownership between non-Hispanic whites and African-Americans is not substantially explained by race but is better explained by other factors independent of race,” id. at 591. Thus, the Salt River plaintiffs’ evidence of statistical disparity alone was insufficient to prove that the racial disparity in voting was “on account of race.” Id. at 591, 595-96.

In this case, by contrast, Plaintiffs have introduced expert testimony demonstrating that the statistical disparity and dis- proportionality evident in Washington’s criminal justice sys- tem arises from discrimination, and the State has failed to refute that showing. See Farrakhan, 2006 WL 1889273, at *6 n.7. If Plaintiffs in this case demonstrated only that African Americans, Latinos, and Native Americans are disproportion- ately affected by Washington’s disenfranchisement law, that clearly would not be enough under Salt River. Unlike in Salt River, however, Plaintiffs have produced evidence that Wash- ington’s criminal justice system is infected with racial bias. The experts’ conclusions are not “statistical disparity alone,” but rather speak to a durable, sustained difference in treatment faced by minorities in Washington’s criminal justice system — systemic disparities which cannot be explained by “factors independent of race.”29

Plaintiffs here have introduced evidence demonstrating what the Salt River plaintiffs could not. Plaintiffs have dem- onstrated that police practices, searches, arrests, detention 29 This, of course, stands in stark contrast to Salt River. In that case, the defendants presented an expert who analyzed the statistical disparity in home ownership using a multivariable analysis. Salt River, 109 F.3d at 590. The defendants’ expert in that case “testified that multiple regression analysis did not indicate a strong correlation between race and home own- ership” and posited that the strongest indicator of home ownership was “persons per dwelling unit.” Id. The district court relied heavily on this expert’s testimony in concluding that the racial disparity in home owner- ship was “not substantially explained by race.” Id. at 591. FARRAKHAN v. GREGOIRE 149 practices, and plea bargaining practices lead to a greater bur- den on minorities that cannot be explained in race-neutral ways. The emphasis on crack cocaine and street drug traffick- ing is not proportional to its harm to the community or its share of the drug trade. The proportion of African Americans and Latinos arrested for drug possession bears no correlation the proportion of users among the races. Searching African Americans and Latinos at higher rates than Whites even though searches of African Americans and Latinos yield less seizures makes little sense in non-racial terms. Detaining minority defendants in disproportionate numbers to Whites even after accounting for differences among defendants in the severity of their crimes, prior criminal records, ties to the community, and the prosecuting attorney’s recommendation, cannot be understood as race neutral.

[24] Plaintiffs’ evidence suggests not only that Washing- ton’s criminal justice system adversely affects minorities to a greater extent than non-minorities, but also that this differen- tial effect cannot be explained by factors other than racial dis- crimination. This method of proving racial discrimination is familiar in our antidiscrimination jurisprudence: The three- step analysis required by Batson v. Kentucky, 476 U.S. 79 (1989), proves discriminatory intent through the same circum- stantial inference from a lack of race-neutral explanations. See, e.g., Green v. Lamarque, 532 F.3d 1028, 1029-30 (9th Cir. 2008). Nothing in Salt River undermines the use of such circumstantial evidence of racial discrimination.

Defendants also contend that Plaintiffs did not produce any evidence connecting asserted bias in the criminal justice sys- tem to the ability of protected minorities to participate effec- tively in the political process. According to Defendants, the district court’s finding of such a connection lowered Plain- tiffs’ burden. The district court, however, relied directly on Farrakhan I’s explanation that a finding of discrimination in the criminal justice system would establish the requisite con- 150 FARRAKHAN v. GREGOIRE nection, because, under Washington law, “disenfranchisement is automatic.” Farrakhan I, 338 F.3d at 1020.

[25] Finally, Defendants argue that, apart from these three asserted errors, the district court erred in concluding that Plaintiffs’ evidence demonstrates more than statistical dispar- ity because, according to Defendants, “the evidence offered by Plaintiffs actually falls far short of any such showing.” As noted above, Plaintiffs’ experts concluded that many of the racial disparities in Washington’s criminal justice system can- not be accounted for by race-neutral explanations and Defen- dants did not refute those conclusions with contrary evidence. Although Defendants criticized the experts’ studies and the conclusions, the reports, when objectively viewed, support a finding of racial discrimination in Washington’s criminal jus- tice system, and the district court did not err in so concluding.

d. Defendants’ arguments that even if Washington’s criminal justice system is infected with racial bias, there is no § 2 violation

Defendants also argue that even if Plaintiffs have demon- strated that Washington’s criminal justice system is infected with racial bias, Defendants are still entitled to summary judg- ment because Plaintiffs have failed to show a discriminatory intent or discriminatory motive. This, they plainly do not have to show under § 2, as amended. See S. Rep. No. 97-417, at 2 (“[P]roof of discriminatory intent is not required to establish a violation of Section 2.”).

Defendants next argue that even if Washington’s criminal justice system is infected with racial bias and that such infec- tion spreads to voting qualifications, Plaintiffs still have failed to show a § 2 violation because “[t]he simple fact is that the voter fully controls whether he or she will forfeit the right to vote under Washington’s felon disenfranchisement law. The voter need only refrain from committing a felony to retain his or her right to participate fully in the electoral process.” How- FARRAKHAN v. GREGOIRE 151 ever, Farrakhan I directly addressed and rejected this claim, holding that, “when felon disenfranchisement results in denial of the right to vote . . . on account of race or color, Section 2 affords disenfranchised felons the means to seek redress.” 338 F.3d at 1016 (emphasis added). Cf. Hunter v. Underwood, 471 U.S. 222, 223-24 (1985) (holding, in a case brought by two individuals convicted of presenting worthless checks, that a provision of the Alabama Constitution disenfranchising those convicted of crimes of moral turpitude violates the Equal Protection Clause because its enactment was motivated by racial bias).

If Farrakhan I and Hunter, inferentially, had not already decided this question, we would nonetheless reject Defen- dants’ argument. Plaintiffs’ evidence demonstrates that in the total population of potential “felons,” i.e., those who have committed crimes, minorities are more likely than Whites to be searched, arrested, detained, and ultimately prosecuted. And they have introduced evidence showing that these dispar- ities cannot be explained in race-neutral ways. Plaintiffs have demonstrated that in a total mass of potential “felons,” i.e., those that have committed crimes, minorities are more likely than Whites to be searched, arrested, detained, and ultimately prosecuted. And they have shown that these disparities cannot be explained away in race-neutral ways.

To be sure, one of the early (if not the first) decision points in the process of becoming a felon is the decision by the per- son to commit a crime. Plaintiffs have not attempted to dem- onstrate that that decision point is infected by racial bias. Before one who commits a criminal act becomes a felon, however, numerous other decisions must be made by State actors. Police departments decide where to spend resources, officers decide which individuals to search and arrest, prose- cutors decide which individuals to charge (including whether to charge a felony or a misdemeanor), detain, and prosecute. If those decision points are infected with racial bias, resulting in some people becoming felons not just because they have 152 FARRAKHAN v. GREGOIRE committed a crime, but because of their race, then that felon status cannot, under § 2 of the VRA, disqualify felons from voting.

3. Plaintiffs are entitled to summary judgment

Plaintiffs also contend that the district court erred in deny- ing their motion for summary judgment.30 We agree. And, contrary to the dissent’s suggestion, Diss. at 160-61, we are mindful of our obligation, when considering a motion for summary judgment, that we view the evidence in the light most favorable to the nonmoving party, and draw “all justifi- able inferences” in that party’s favor. See Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 255 (1986). However, “[o]n a motion for summary judgment, ‘facts must be viewed in the light most favorable to the nonmoving party only if there is a “genuine” dispute as to those facts.’ ” Ricci v. DeStefano, 129 S. Ct. 2658, 2677 (2009) (quoting Scott v. Harris, 550 U.S. 372, 380 (2007)); see also Beard v. Banks, 548 U.S. 521, 529-30 (2006) (noting that “we must distinguish between evi- dence of disputed facts and disputed matters of professional judgment”). Here, there is no such “genuine” dispute. Plain- tiffs carried their burden of producing evidence of discrimina- tion; defendants were then required to “do more than simply show that there is some metaphysical doubt as to the material facts.” Matsushita Elec. Indus. Co. v. Zenith Radio Corp., 475 U.S. 574, 586-87 (1986). Based on the substantial showing by Plaintiffs, Defendants had the burden “to set forth any specific facts showing that there is a genuine issue . . . for trial.” Dep’t of Commerce v. U.S. House of Representatives, 525 U.S. 316, 330 (1999). This they did not do. The record is uncontro- verted. 30 “Ordinarily, a denial of a motion for summary judgment is not a final order and thus not appealable. 28 U.S.C. § 1291. However, the district court’s grant of summary judgment [to Defendants] was a final decision giving us jurisdiction to review its denial of plaintiff’s motion for sum- mary judgment.” Abend v. MCA, Inc., 863 F.2d 1465, 1482 n.20 (9th Cir. 1988). FARRAKHAN v. GREGOIRE 153 In its procedural posture, this case is also, in many respects, similar to the recent case of Ricci v. DeStefano, which also involved the resolution of cross-motions for summary judg- ment, and in which the Court noted: “As the District Court noted, although ‘the parties strenuously dispute the relevance and legal import of, and inferences to be drawn from, many aspects of this case, the underlying facts are largely undisput- ed.’ ” 129 S. Ct. at 2665 (quoting Ricci v. DeStefano, 554 F. Supp. 2d 142, 145 (D. Conn. 2006)). The Court granted sum- mary judgment to petitioners, because “there [was] no evi- dence — let alone the required strong basis in evidence” to support the respondent’s position. Id. at 2681.

[26] In any case, even viewing the evidence in the light most favorable to Defendants, Plaintiffs have demonstrated that racial minorities are overrepresented in the felon popula- tion based upon factors that cannot be explained by non-racial reasons. Given that uncontroverted showing, in the words of the district court, there can be “no doubt that members of racial minorities have experienced discrimination in Washing- ton’s criminal justice system.” In the face of this showing, all Defendants did was question the credibility of Plaintiffs’ experts without “rais[ing] a genuine issue of material fact regarding” the ultimate effect of Washington’s felon disen- franchisement law. Dep’t of Commerce, 525 U.S. at 331. They have “not offer[ed] any fact-based or expert-based refu- tation,” Beard, 548 U.S. at 534, that challenges the conclu- sions reached by Plaintiffs’ experts. Section 2 of the VRA demands that such racial discrimination not spread to the bal- lot box. Thus, based on the uncontroverted record, Plaintiffs are entitled to summary judgment.

E. Washington’s Amended Felon Disenfranchisement Law

Defendants argue that Washington’s recent amendment to its felon disenfranchisement law alters the totality of the cir- cumstances analysis required by § 2 of the VRA. The amend- 154 FARRAKHAN v. GREGOIRE ment modified Washington’s felon disenfranchisement scheme by providing for the provisional restoration of the vot- ing rights of felons upon their release from prison or from community custody (a Washington program through which offenders live in the community, but are subject to restraints imposed by the Department of Corrections). See Wash. Laws of 2009, ch. 325, HB 1517; Wash. Rev. Code § 9.94A.030(5). A released felon’s provisionally restored right to vote may be revoked for willful failure to fulfill all financial obligations imposed as part of his sentence. Id. at 2(a). Under the previ- ous law, a felon was barred from voting until he had com- pleted all the requirements of his criminal sentence, including any financial obligations imposed as part of that sentence and had obtained certificates of discharge restoring his civil rights. See Farrakhan I, 338 F. 3d at 1012.

Defendants first contend that if felon disenfranchisement laws disproportionately affect minorities, then the amend- ment, which reduces the number of felons disenfranchised, disproportionately benefits minorities, and in so doing pro- vides evidence of positive action by the state with regard to minority voting rights that is relevant to Senate Factors 1, 3 and 8. However, a mere decrease in the length of time for which the State’s discriminatory criminal justice system deprives minorities of the right to vote does not change our determination that those Factors have little relevance to this case.

[27] In terms of Factor 5, the gravamen of defendants’ argument is that the amendment decreases the total number of minorities who are without voting rights at any given time, and so diminishes the extent of the discriminatory effects of the State’s felon disenfranchisement system.31 We hope that 31 Whether this is true remains to be seen. The amended Washington law only “provisionally” restores the voting rights of felons upon their release from custody. HB 1517 § 1.1. Permanent restoration of the voting rights requires additional action by the sentencing court, the indeterminate sen- tence review board or the governor. Id. at 5(f). FARRAKHAN v. GREGOIRE 155 defendants are correct about the positive effects of the amend- ment: it appears that under the old law almost a quarter of oth- erwise qualified African American men in Washington were disenfranchised. See Jamie Fellner & Marc Mauer, Losing the Vote: The Impact of Felony Disenfranchisement Laws in the United States 9-10 (1998); see also Jeff Manza & Christopher Uggen, Locked Out: Felon Disenfranchisement and American Democracy 250 (2006) (over 17% of the entire adult black population of Washington disenfranchised). However, no matter how well the amended law functions to restore at an earlier time the voting rights of felons who have emerged from incarceration, it does not protect minorities from being denied the right to vote upon conviction by a criminal justice system that Plaintiffs have demonstrated is materially tainted by discrimination and bias. Accordingly, it does not alter our analysis as to Senate Factor 5 or as to the totality of the cir- cumstances.

IV. CONCLUSION

We are bound by Farrakhan I’s holding that § 2 of the VRA applies to Washington’s felon disenfranchisement law. Plaintiffs have demonstrated that the discriminatory impact of Washington’s felon disenfranchisement is attributable to racial discrimination in Washington’s criminal justice system; thus, that Washington’s felon disenfranchisement law violates § 2 of the VRA. The judgment of the district court granting Defendants’ motion for summary judgment and denying Plaintiffs’ motion for summary judgment is REVERSED, and the case is REMANDED with instructions to GRANT summary judgment to Plaintiffs.

McKEOWN, Circuit Judge, dissenting:

In granting summary judgment to plaintiffs, the majority has charted territory that none of our sister circuits has dared 156 FARRAKHAN v. GREGOIRE to explore. The First, Second, and Eleventh Circuits have determined that vote denial challenges to felon disenfran- chisement laws are not cognizable under the Voting Rights Act. See Simmons v. Galvin, 575 F.3d 24 (1st Cir. 2009); Hayden v. Pataki, 449 F.3d 305 (2d Cir. 2006) (en banc); Johnson v. Governor of State of Fla., 405 F.3d 1214 (11th Cir. 2005) (en banc). That preliminary question was settled by our circuit in Farrakhan v. Washington, 338 F.3d 1009 (9th Cir. 2003) (“Farrakhan I”). While I believe that the felon dis- enfranchisement challenge is not a comfortable fit within the Voting Rights Act, I do not dispute the continuing validity of Farrakhan I. The wisdom of Farrakhan I is not within the purview of the panel to reconsider here. See Miller v. Gam- mie, 335 F.3d 889, 893 (9th Cir. 2003) (en banc) (holding that prior circuit authority is binding on three-judge panels unless “clearly irreconcilable with the reasoning or theory of inter- vening higher authority”). However, in part because the hold- ing of Farrakhan I places us in a crowd of one amongst the circuits, I believe we should be particularly mindful before reversing the district court and invalidating felon disenfran- chisement in the State of Washington. The majority has failed to act with appropriate caution. I respectfully dissent.

I first note that the landscape of this case has changed from the time the district court dismissed the case and even since we heard oral argument. As of July 26, 2009, Washington law now provides that the State will provisionally restore voting rights to felons convicted in Washington state courts so long as the individual is no longer under the authority of the Department of Corrections, and to those convicted of federal felonies or felonies in other states as long as the person is no longer incarcerated. Washington Laws of 2009, chapter 325, HB 1517.

Following this significant legislative change, we are left to consider the Voting Rights Act challenge of only those felons still serving their prison terms. Interestingly, the case up to this point has never contemplated the two distinct sets of fel- FARRAKHAN v. GREGOIRE 157 ons affected by the prior Washington law—those still incar- cerated and those already released. Both the parties and the courts have seemingly considered felons generally, as a single group; the bifurcation of classes of felons came about as a consequence of this new legislation. Thus, within this litiga- tion, no court has addressed whether these two sets of individ- uals present meaningful analytical differences. This posture is not surprising because the statute did not make such a distinc- tion before it was amended and, as the State notes in supple- mental briefing, the case now presents “a substantially different controversy.” Although the majority concludes that the new law has limited effect on the case, see Maj. Op. at 119, the supplemental briefing suggests otherwise. Indeed, the State of Washington claims the entire case is moot because the statute at issue has substantially changed, that a significant part of the case involved the “continuing disenfranchisement of felons upon release from incarceration,”—a point that is no longer at issue—and that the new law “necessarily alters the totality of the circumstances” analysis. The State views the new law as a game changer supporting affirmance of the dis- trict court.

It bears noting that none of the three recent felon disenfran- chisement cases to percolate through the circuit courts has encompassed both classes of felons. In Simmons v. Galvin, the First Circuit considered a challenge brought solely by cur- rently incarcerated felons to the Massachusetts law prohibit- ing incarcerated felons from voting. 575 F.3d at 26. Similarly, in Hayden v. Pataki, 449 F.3d 305 (2d Cir. 2006) (en banc), the Second Circuit considered N.Y. Election Law § 5-106, which “disenfranchises only currently incarcerated prisoners and parolees.” Id. at 314. That court remarked that “the statute may not raise the same issues that are implicated by provi- sions disenfranchising for life those convicted of felonies, such as the . . . provision of the . . . Washington Constitution addressed in Farrakhan.” Id. (emphasis added). The court did not elaborate on the contours of any distinctions. The Elev- enth Circuit, in Johnson v. Governor of State of Fla., 405 F.3d 158 FARRAKHAN v. GREGOIRE 1214 (11th Cir. 2005) (en banc), considered the converse class of individuals—“Florida citizens who have been con- victed of a felony and have completed all terms of their incar- ceration, probation, and parole but who are barred from voting under the state’s felon disenfranchisement law.” Id. at 1216-17. Thus, the Eleventh Circuit, too, did not contemplate a bifurcated group of felons.

In an earlier case, Wesley v. Collins, 791 F.2d 1255 (6th Cir. 1986), the Sixth Circuit considered a Voting Rights Act challenge to Tennessee’s felon disenfranchisement law brought by a public interest group and an African-American man convicted of a felony. Id. at 1257. Though Tennessee’s law appears to have affected both currently incarcerated pris- oners and felons already released, see id., the court did not focus on this distinction when it dismissed the plaintiffs’ chal- lenge. See id. at 1260-62. Thus, even considering Wesley, because of the recent statutory change, ours is the only court clearly presented with the question whether the different groups of felons present a meaningful distinction under the VRA’s totality of the circumstances inquiry.1

Thus, the enactment of HB 1517 is the first reason I would remand this case to the district court. It is not our job to con- sider, in the first instance, the effect this new law has on plaintiffs’ case and whether the totality of the circumstances analysis under § 2 of the Voting Rights Act should be differ- ent now that plaintiffs’ case remains viable only as to cur- rently incarcerated felons. 1 In addition, the Fourth Circuit has considered a challenge to a felon disenfranchisement law in the Commonwealth of Virginia. In an unpub- lished decision, Howard v. Gilmore, No. 99-2285, 2000 WL 203984 (4th Cir. Feb. 23, 2000), the court determined that the plaintiff failed “to plead any nexus between the disenfranchisement of felons and race.” Id. at *1. It is unclear from the decision whether the plaintiff was currently incarcer- ated or already released at the time of his suit. Regardless, the Fourth Cir- cuit’s decision makes no mention of any distinction between classes of felons. FARRAKHAN v. GREGOIRE 159 Next, I take issue with the majority’s conclusion that plain- tiffs prevail by offering evidence regarding Senate Factor 5 alone. As detailed in the majority opinion, Maj. Op. at 123-24, the Senate Report on the 1982 amendments to the Voting Rights Act listed “typical factors” that courts might consider in determining whether, under the totality of the cir- cumstances, a challenged voting practice “results in” the denial or abridgement of the right to vote on account of race. Evidence of racial discrimination in the Washington criminal justice system falls primarily under Senate Factor 5—”the extent to which members of the minority group in the state or political subdivision bear the effects of discrimination in such areas as education, employment and health, which hinder their ability to participate effectively in the political process.” S. Rep. No. 97-417, at 29. The majority concludes that having found discrimination in the Washington criminal justice sys- tem, “the district court should not have required Plaintiffs to produce additional circumstantial evidence” because the evi- dence of discrimination in the criminal justice system alone “would establish a § 2 violation under the totality of the cir- cumstances.” Maj. Op. at 145-46. As a result, the majority itself considers only evidence of Factor 5 in granting sum- mary judgment to plaintiffs.

I take issue with the majority’s limitation. My view is largely driven by my disagreement with the majority’s con- clusion that there is a per se “analytical distinction” between vote denial and vote dilution cases in the circumstance pre- sented here. See Maj. Op. at 137. To be sure, there are differ- ences between the two types of cases, but those differences should not force an absolute dichotomy in our analysis. As I have already noted, the felon disenfranchisement challenge is not a comfortable fit within the Voting Rights Act. That said, there is arguably a continuum of conduct that constitutes a denial or abridgement of the right to vote within the context of the Voting Rights Act, and this case need not be shoe- horned into a single category. 160 FARRAKHAN v. GREGOIRE Indeed, academic literature suggests that one of the driving concerns surrounding felon disenfranchisement laws— advanced in this litigation through a vote denial claim—is the effect the regulations have on the voting power of minority blocs, which is the thrust of a vote dilution inquiry. See e.g. Pamela S. Karlan, Convictions and Doubts: Retribution, Rep- resentation, and the Debate Over Felon Disenfranchisement, 56 Stan. L. Rev. 1147, 1155-64 (2004). “Virtually every con- temporary discussion of criminal disenfranchisement in the United States begins by noting the sheer magnitude of the exclusion, and its racial salience.” Id. at 1156. This observa- tion is not surprising, as “groups of voters elect representa- tives, individual voters do not.” Id. (quoting Davis v. Bandemer, 478 U.S. 109, 167 (1986) (Powell, J., concurring in part and dissenting in part)).

Thus, taking away the right to vote of minority felons may very well have a significant effect on the voting power of minorities as a whole in any given jurisdiction. As a result, those urging the repeal of felon disenfranchisement laws are often driven not only by their concern for the rights of the individual felons, but also by their worries about the effect that such laws have on the voting power of minority voting blocs. Indeed, the concern for the effect on the voting power of minorities is evidenced by Wesley, in which the Sixth Cir- cuit considered the plaintiffs’ challenge to the Tennessee felon disenfranchisement law as a vote dilution claim. 791 F.2d at 1260-62. Based on the interwoven concerns in vote denial and vote dilution cases, I am not comfortable dictating that a district court should not consider certain factors— Senate Factors or otherwise—in vote denial cases, nor do I agree with the majority’s conclusion that the plaintiffs prevail solely by establishing evidence that falls within Senate Factor 5.

As to Senate Factor 5 itself, significant factual issues remain. The existence of these unresolved issues is another reason why I part company with the majority. The majority FARRAKHAN v. GREGOIRE 161 makes much of the district court’s conclusion that plaintiffs have presented “compelling” evidence of racial discrimination in the criminal justice system. Maj. Op. at 143. The district court made this conclusion, of course, while considering the State’s motion for summary judgment, thereby viewing the evidence in the light most favorable to the plaintiffs. In deter- mining whether to now grant summary judgment to plaintiffs, the majority should view the evidence in the light most favor- able to the defendants. Thus, the district court’s conclusion that the evidence is “compelling” is of little use at this stage; the majority seriously errs in failing to acknowledge that conundrum.

A review of the evidence reveals the risk the majority takes in viewing the evidence in the light most favorable to the plaintiffs while granting summary judgment to plaintiffs—a complete reversal of the normal procedure on summary judg- ment. For example, in reviewing Professor Beckett’s report, the district court “extrapolate[d]” her “drug-arrest-in-Seattle- specific findings to Washington felony arrests and convictions in general.” When put to the test, it is unclear whether the extrapolation would hold up, as Beckett’s study does not con- sider non-drug related arrests in Seattle or any arrests outside of Seattle. While Dr. Crutchfield’s report does encompass the entire State, the studies he details do not paint a definitive pic- ture of racial discrimination in the Washington criminal jus- tice system. In making this observation, I do not deny the existence of discrimination, my point rests on the evidence presented. For example, Crutchfield discusses the Klement and Siggins (2001) study of drug enforcement patterns in Seattle. Crutchfield notes that within the drug-crime category, the police department focuses most heavily on “observable street level drug markets,” which have much more of a “mi- nority flavor” than the general population. Crutchfield goes on to explain that “[b]usiness owners and residents call the police when visible drug activity threatens their interests” and that drug sales in the “street markets” are more likely to affect those interests than other sorts of drug crimes. A reasonable 162 FARRAKHAN v. GREGOIRE factfinder may very well conclude that the police focus on street markets has little to do with racial discrimination, but instead relates much more strongly to the police department’s desire to target crimes likely to affect the well-being of the greatest majority of businesses and individuals. Alternatively, a factfinder may determine that the focus results from the fact that police are tasked with responding to citizens’ calls; if people are more likely to call the authorities only when they can actually view a drug crime occurring out in the open, i.e. on the street, then of course it is more likely that police arrests will over represent street market drug sales, as compared to other types of drug crimes. Or finally, it may be that this approach to policing is race-based. The point is that there are material factual questions as to cause and effect.

I stress these examples to emphasize my view that the majority errs in granting summary judgment to plaintiffs. The proper course at this stage is to remand to the district court for consideration of the plaintiffs’ motion for summary judgment. On remand, a factfinder should be able to weigh the evidence concerning whether there is racial discrimination in the Wash- ington criminal justice system, along with other factors (the Senate Factors and perhaps additional relevant considerations) to determine if plaintiffs have demonstrated a violation of the Voting Rights Act. This court overreaches when it bypasses that crucial exercise. And, considering the potential holes in the evidence, the majority is remiss in doing so.