Marc Veasey v. Greg Abbott

STEPHEN A. HIGGINSON, Circuit Judge,

joined by GREGG COSTA, Circuit Judge, concurring:

As the Supreme Court has reminded, though great progress has been, made, *273“voting discrimination still exists; no one doubts that,” and Section 2 of the Voting Rights Act operates as a crucial “permanent, nationwide ban,” Shelby County v. Holder, — U.S. -, 133 S.Ct. 2612, 2619, 2631, 186 L.Ed.2d 651 (2013), on “even the most subtle forms of discrimination,” Chisom v. Roemer, 501 U.S. 380, 406, 111 S.Ct. 2354, 115 L.Ed.2d 348 (1991) (Scalia, J., dissenting). The courts, therefore, have a vital role in protecting the right “to participate equally in the political process.” Thornburg v. Gingles, 478 U.S. 30, 80, 106 S.Ct. 2752, 92 L.Ed.2d 25 (1986). Indeed, the Supreme Court has long cautioned that alleged discrimination against minorities calls for a “searching judicial inquiry,” United States v. Carolene Prods. Co., 304 U.S. 144, 152 n.4, 58 S.Ct. 778, 82 L.Ed. 1234 (1938), and with regard to Section 2, which mandates consideration of “the totality of circumstances,” 52 U.S.C. § 10301(b), Congress has made clear that, again in the Court’s words, “whether the political processes are ‘equally open’ depends upon a searching practical evaluation of the ‘past and present reality’ ” and “on a ‘functional’ view of the political process,” Gingles, 478 U.S. at 45, 106 S.Ct. 2752 (quoting S. Rep. No. 97-417, at 30 (1982)).

After a nine-day trial that saw the testimony of over forty witnesses, half of them experts, the district court concluded in a nearly one-hundred-and-fifty-page opinion that SB 14 — stricter than other voter ID laws that courts have upheld, including those after which Texas’s law was ostensibly modeled1 — violates Section 2 because it abridges minorities’ ability to participate equally in the political process. As the majority opinion shows, the district court’s Section 2 finding is a permissible view of a voluminous record informed by extensive testimony. I join the majority opinion and write separately to respond to arguments in the principal dissenting opinion because, contrary to those arguments, I perceive that the majority opinion’s Section 2 analysis fits comfortably with decisions of the Supreme Court, our court, and other circuits.

I.

As the majority opinion explains, our adoption of the Fourth and Sixth Circuits’ two-part test places Section 2’s totality-of-circumstances inquiry in a vote-denial framework that adheres to the text of Section 2, 52 U.S.C. § 10301, and the Supreme Court’s guidance in Gingles, 478 U.S. at 47, 106 S.Ct. 2752 (“The essence of a § 2 claim is that a certain electoral law, practice, or structure interacts with social and historical conditions to cause an inequality in the opportunities enjoyed by black and white voters to elect their preferred representatives.”). Use of the Gin-gles (or Senate) factors as nonexhaustive tools fleshing out this framework ensures the requisite causal linkage between past discrimination and a challenged voting practice’s disparate impact. Though some of the factors may have less relevance in vote-denial cases, others, particularly “Senate factors one, three, five, and nine,” aid in applying the Supreme Court’s admonition to discern the relevant social and historical effects of discrimination, and their interaction with a challenged law. Ohio State Conference of NAACP v. Hust-*274ed, 768 F.3d 524, 555 (6th Cir. 2014), vacated on other grounds, 2014 WL 10384647 (6th Cir. 2014); see also Daniel P. Tokaji, Applying Section 2 to the New Vote Denial, 50 Harv. C.R.-C.L. L. Rev. 439, 481-82 (2015) (concluding that most of the Senate factors are relevant to vote-denial claims, the fifth factor especially so). And case law belies the argument that the Senate factors have no place outside the vote-dilution context.

The Senate factors have roots' in this court, see Zimmer v. McKeithen, 485 F.2d 1297, 1305 (5th Cir. 1973) (en banc), and we have not limited them to vote-dilution cases. To the contrary, in a vote-denial case in which we affirmed a finding that Mississippi’s voter registration process violated Section 2’s results test, we noted that the trial court applied those “ ‘objective factors’ to aid the courts in evaluating a § 2 claim.” Miss. State Chapter, Operation Push, Inc. v. Mabus, 932 F.2d 400, 405 (5th Cir. 1991), aff'g Miss. State Chapter, Operation Push v. Allain, 674 F.Supp. 1245 (N.D. Miss. 1987) (applying all nine factors).2 And among our sister circuits, it is not just the Fourth and Sixth thát have found the Senate factors significant in vote-denial cases. See Gonzalez v. Arizona, 677 F.3d 383, 405-06 (9th Cir. 2012) (en banc) (explaining that “courts should .consider” the Senate factors in vote-denial cases); Johnson v. Governor of Fla., 405 F.3d 1214, 1227 n.26 (11th Cir. 2005) (en banc) (stating in dictum that the factors apply to denial cases); Smith v. Salt River Proj. Agr. Improvement & Power Dist., 109 F.3d 586, 596 n.8 (9th Cir. 1997) (rejecting the argument that the Senate factors “apply only to ‘vote dilution’ claims”). We do well to join this near-consensus of circuits in recognizing that the Senate Report and Gingles provide guidance in the vote-denial context.

II.

Today’s outcome is also not inconsistent with Crawford v. Marion County Election Board, 553 U.S. 181, 128 S.Ct. 1610, 170 L.Ed.2d 574 (2008). Crawford did not even discuss the Voting Rights Act, and held only that the lower courts “correctly concluded that the evidence in the record [was] not sufficient to support a facial attack on the validity of the entire statute” under the constitutional Anderson-Burdick framework. Id. at 189, 128 S.Ct. 1610. Furthermore, as Justice Stevens took care to note, the record there, unlike here, (1) did not quantify the voters without qualifying ID, (2) provided no “concrete evidence of the burden imposed on voters who currently lack photo identification,” and (3) said “virtually nothing about the difficulties faced by ... indigent voters.” Id. at 200-01, 128 S.Ct. 1610. To be sure, Crawford established that preventing voter fraud and safeguarding voter confidence are legitimate and important state interests. Id. at 194-97, 128 S.Ct. 1610. But it does not follow that assertion of those interests immunizes a voter ID law from all challenges, or that courts should be deterred from examining, as part of the Section 2 totality-of-circumstances inquiry, the tenuousness of the reasons given for the law. See League of Women Voters of *275N. Car. v. North Carolina, 769 F.3d 224, 246 (4th Cir. 2014); cf. Husted, 768 F.3d at 547 (“[Crawford] does not mean, however, that the State can, by merely asserting an interest in preventing voter fraud, establish that that interest outweighs a significant burden on voters.”).3

Nor does our decision contravene League of United Latin American Citizens v. Clements, 999 F.2d 831 (5th Cir. 1993) (en banc). Texas highlights our statements in that dilution case that socioeconomic disparities alone do not show “that minorities do not enjoy equal access to the political process,” and that the Senate Report “did not dispense with proof that participation in the political process is in fact depressed among minority citizens.” Id. at 866, 867. From this, Texas reasons that the district court erred by finding a Section 2 violation without “proof that the challenged law affects voting behavior.” The cited language in Clements, however, discussed Senate factors one and five: “the extent of any history of official discrimination” affecting political participation, and “the extent to which members of the minority group ... bear the effects of discrimination in such areas as education, employment and health, which hinder their ability to participate effectively in the political process.” Id. at 866 & n.30. In that context, we noted that the plaintiffs “offered no evidence of reduced levels of black voter registration, lower turnout among black voters, or any other factor tending to show that past discrimination” — -not the challenged law by itself— “ha[d] affected their ability to participate in the political process” as required by “these Zimmer [Senate] factors.” Id. at 867. Here, in sharp contrast, the district court heard such evidence and found, in its discussion of factor five, that the effects of past discrimination “have hindered the ability of African-Americans and Hispanics to effectively participate in the political process”; indeed, one expert “testified that these minorities register and turnout for elections at rates that lag far behind Anglo voters.” Veasey v. Perry, 71 F.Supp.3d 627, 697 (S.D. Tex. 2014).

The district court also heard from witnesses who were unable to vote because they lacked the required forms of ID, from some who struggled to obtain the required forms of ID or documents needed to obtain them, and from others who help disadvantaged individuals obtain photo IDs and attested to the difficulties those individuals face in doing so. See id. at 667-76. The court further credited expert testimony that SB 14 “would almost certainly decrease voter turnout, particularly among minorities,” by imposing burdens that fall more heavily on African-Americans and Hispanics. Id. at 655-56; see also id. at 664-65.4 The majority opinion rightly re*276jects Texas’s attempt to stretch Clements to require that plaintiffs wait until elections have occurred under the challenged law and then prove a direct impact on turnout. That would make impossible pre-enforcement Section 2 challenges, which the Supreme Court recently acknowledged. See Shelby County, 133 S.Ct. at 2619 (“[I]njunctive relief is available in appropriate [Section 2] cases to block voting laws from going into effect.”). And as multiple experts testified, given the myriad variables at play, it is extremely difficult to isolate the effect of a new law on voter turnout.5 Texas’s argument that the plaintiffs- were required to prove a direct impact on turnout is unsound.

It is also mistaken to suggest that the majority opinion conflicts with the Ninth Circuit’s decision in Gonzalez, 677 F.3d at 383. That court, after citing the Senate factors with approval and emphasizing the deference owed to a district court’s factual Section 2 determinations, affirmed a finding that the plaintiff had failed to establish the disparate impact of a voter ID law where, among other things, the district court rejected as unreliable the plaintiffs expert statistical analysis and the record included no evidence that Hispanics were even less likely to possess qualifying ID. Id. at 406-07 & n.33. The law at issue was also much less strict, requiring a voter to present at the polls either (A) one of a broader range of photo IDs or (B) two non-photo documents showing the voter’s name and address, such as a utility bill, bank statement, or voter registration card. Gonzalez v. Arizona, No. CV 06-1268-PHX, 2006 WL 3627297, at *1, *6 (D. Ariz. Sept. 11, 2006). Even so, two judges wrote separately to stress that the court’s holding was based on “the current record,” and that “[a] different record in a future case could produce a different outcome.” Gonzalez, 677 F.3d at 442 (Berzon, J., concurring).6

The Ninth Circuit of course said that “proof of ‘causal connection between the challenged voting practice and a prohibited discriminatory result’ is crucial” to a Sec-. tion 2 challenge. Id. at 405 (majority opinion) (quoting Smith, 109 F.3d at 595). But the district court here found that the rec*277ord established that causal connection. The trial judge found that SB 14 makes voting relatively more difficult for minorities, and without ameliorative measures will likely disproportionately suppress minority voting, by conditioning the right to vote on the possession of documents that, because of the effects of past discrimination, are harder for minority voters to obtain. Such interaction between present-day law and the effects of past discrimination is what Congress intended to combat. See Gingles, 478 U.S. at 69, 106 S.Ct. 2752 (“Congress intended that the Voting Rights Act eradicate inequalities in political opportunities that exist due to the vestigial effects of past purposeful discrimination.”); see also id. at 44, 106 S.Ct. 2752 n.9 (“[T]he purpose of the Voting Rights Act was ‘not only to correct an active history of discrimination, ... but also to deal with the accumulation of discrimination.’ ” (quoting S. Rep. No. 97-417, at 5)).

For this reason among others, we should not be guided by Frank v. Walker, 768 F.3d 744 (7th Cir. 2014). In that case, Judge Easterbrook — who did not mention the applicable clear-error standard of review — overlooked many of the district court’s factual findings. See Frank v. Walker, 773 F.3d 783, 792-93, 796-97 (7th Cir. 2014) (Posner, J., dissenting from denial of rehearing en banc); Tokaji, supra, at 460. Questioning other circuits’ approaches to vote-denial cases without offering a clear alternative, Judge Easter-brook went on to uphold a different voter ID law on the rationales that the law did not facially “draw any line by race,” and that the plaintiffs had not “show[n] a ‘denial’ of anything by Wisconsin, as § 2(a) requires” because the state had not directly caused minorities to be less likely or able to own qualifying IDs. Frank, 768 F.3d at 753. Ultimately, the Seventh Circuit read Section 2 as only “an equal-treatment requirement,” and rejected the plaintiffs’ challenge “because in Wisconsin everyone has the same opportunity to get a qualifying photo ID.” Id. at 754, 755.

This reasoning ignores that Section 2 prohibits voting procedures “imposed or applied ... in a manner which results in a denial or abridgement of the right ... to vote.” 52 U.S.C. § 10301(a) (emphasis added). Indeed, the opinion does not mention “abridgement” aside from a single quotation of the statute. Judge Easterbrook’s “equal-treatment” gloss — which he did not explain aside from saying that is “how [the statute] reads,” Frank, 768 F.3d at 754 — is puzzling because it is undisputed that, in response to a judicial ruling that Section 2 plaintiffs had to prove discriminatory intent, Congress revised the statute “to make clear that a violation could be proved by showing discriminatory effect alone,” Gingles, 478 U.S. at 35, 106 S.Ct. 2752; cf. Tex. Dep’t of Hous. & Cmty. Affairs v. Inclusive Cmtys. Project, Inc., — U.S. -, 135 S.Ct. 2507, 2513, 192 L.Ed.2d 514 (2015) (explaining that “disparate-treatment” plaintiffs must show a discriminatory intent or motive). And if Section 2 requires only equal treatment, or if a Section 2 burden is cognizable only if it is impossible for some minority voters to comply with the challenged law, Justice Scalia must have mistakenly stated that Section 2 would be violated if “a county permitted voter registration for only three hours one day a week, and that made it more difficult for blacks to register than whites.” Chisom, 501 U.S. at 408, 111 S.Ct. 2354 (Scalia J., dissenting). After all, ignoring disparities due to past discrimination, that law would give everyone the “same opportunity” to register.

Judge Easterbrook further seems to have reasoned that the only discrimination relevant to Section 2’s totality-of-the-circumstances inquiry is of the state-sponsored variety. See Frank, 768 F.3d at 753. *278The dissent agrees, adding that the official discrimination must be “contemporary” or at least “recent.” I have difficulty squaring that with Section 2’s directive to address the “totality of circumstances,” and with the Supreme Court’s admonitions to probe the interaction of the challenged practice “with social and historical conditions” as well as consider “the extent to which minority group members bear the effects of past discrimination in areas such as education, employment, and health, which hinder their ability to participate effectively in the political process.” Gingles, 478 U.S. at 47, 45, 106 S.Ct. 2752. As one of this court’s notable jurists .put it, “under the results standard of section 2, pervasive private discrimination should be considered, because such discrimination can contribute to the inability of [minorities] to assert their political influence and to participate equally in public life.” United States v. Marengo Cty. Comm’n, 731 F.2d 1546, 1567 n.36 (11th Cir. 1984) (Wisdom, J.); see also Gomez v. City of Watsonville, 863 F.2d 1407, 1418 (9th Cir. 1988) (rejecting the argument that only discrimination by the defendant is relevant to a Section 2 vote-dilution case); Solomon v. Liberty County, 899 F.2d 1012, 1032 (11th Cir. 1990) (en banc) (Tjoflat, J., concurring) (“Congress ... revised section 2 to prohibit election practices that accommodate or amplify the effect that private discrimination has in the voting process.” (ellipsis in original) (quoting David L. Eades, Recent Developments, Section 2 of the Voting Rights Act: An Approach to the Results Test, 39 Vand. L. Rev. 139, 172 (1986))).7

III.

Two related final points bear mentioning. First, Judge Easterbrook warned that the Frank plaintiffs’ interpretation of Section 2 could “sweep[ ] away almost all registration and voting rules.” Frank, 768 F.3d at 754. For example, he opined, “[m]otor-voter registration, which makes it simple for people to register by checking a box when they get drivers’ licenses, would be invalid, because black and Latino citizens are less likely to own cars and therefore less likely to get drivers’ licenses.” Id. The dissent advances a similar point, warning that voting regulations ranging from polling locations, early voting details, and registration times “can be challenged successfully under the majority’s rationale.” I agree that Section 2 challenges can be brought against a variety of election laws — but that is nothing new. See Holder v. Hall, 512 U.S. 874, 922, 114 S.Ct. 2581, 129 L.Ed.2d 687 (1994) (Thomas, J., concurring) (“The section thus covers all manner of registration requirements, the practices surrounding registration (including the selection of times and places where registration takes place and the selection of registrars), the locations of polling places, the times polls are open, the use of paper ballots as opposed to voting machines, and other similar aspects of the voting process that might be manipulated to deny any citizen the right to cast a ballot and have it properly counted.”); see also Allen v. State Bd. of Elections, 393 U.S. 544, 566-67, 89 S.Ct. 817, 22 L.Ed.2d *2791 (1969) (“Indicative of an intention to give the Act the broadest possible scope, Congress expanded the language in the final version of [Section] 2 to include any ‘voting qualifications or prerequisite to voting, or standard, practice, or procedure.’” (quoting then-42 U.S.C. § 1978)). Nor does it imperil our electoral system.

There is a difference between making voting harder in ways that interact with historical and social conditions to disproportionately burden minorities and making voting easier in ways that may not benefit all demographies equally (like motor-voter). The former can be characterized as “abridging” the right to vote; the latter cannot. Laws that neither “eliminate opportunities that racial minorities disproportionately use, [n]or impose a requirement that they disproportionately lack,”8 in other words, will not fail our test. Cf. 52 U.S.C. § 10301(b) (explaining that Section 2 does not “establishf ] a right to have members of a protected class elected in numbers equal to their proportion in population”). And as recent eases show, not all voter ID laws will, either. It is simplistic to lump all such laws together, overlooking that details — such as which forms of ID are accepted at the polls,9 what documentation is needed to get a free qualifying ID,10 and how the law is implemented 11 — matter. Especially significant are the accommodations made for those most affected by the ID requirement. In North Carolina, for instance, persons without qualifying ID can vote by swearing that they “subjectively believe a reasonable impediment prevented them from acquiring ID.” N. Car. State Conference of the NAACP v. McCrory, — F. Supp. 3d -, ---, 2016 WL 1650774, at *35-36 (M.D.N.C. Apr. 25, 2016). Only the impediment’s veracity may be challenged. Id. at ——, *120. South Carolina has a similar provision, which a three-judge court stressed in preclearing the state’s voter ID law. See South Carolina v. United States, 898 F.Supp.2d 30, 35-43 (D.D.C. 2012).

Second, we should not shy away from inquiring into such details, or from judging laws in their operative contexts, merely because it will require courts to draw fact-specific and even close distinctions. States have reacted to the Supreme Court’s decisions in Crawford and Shelby County by introducing a range of voting regulations *280that go beyond what had previously been upheld. It is healthy for these initiatives to be assessed against congressional mandates, and courts can and should distinguish between nondiscriminatory ones which safeguard voter integrity and those which, whatever' their intentions, interact with the effects of past discrimination to abridge minorities’- opportunities to participate in the political process. Such scrutiny should be seen not as heavy-handed judicial rejection of legislative priorities, but as part of a process of harmonizing those priorities with the fundamental right to vote — a topic with which over a quarter of our Constitution’s amendments have dealt in one way or another, and an individual right that cannot be compromised because an adverse impact falls on relatively few rather than many. See Operation Push, 932 F.2d at 404 (noting that the Mississippi legislature responded to a finding of a Section 2 violation by adopting ameliorative changes suggested by a district court); South Carolina, 898 F.Supp.2d at 35-36 (explaining that state officials adopted a broad interpretation of a voter ID law’s reasonable-impediment exception as litigation progressed); id. at 53 (Bates, J., concurring) (“An evolutionary process has produced a law that accomplishes South Carolina’s important objectives while protecting every individual’s right to vote and a law that addresses the significant concerns raised about [the law’s] potential impact on a group that all agree is disproportionately African-American.”); see also N. Car. State Conference of the NAACP v. McCrory, 156 F. Supp. 3d 683, 687-88, 698-700, 2016 WL 204481, at *2, *11 (M.D.N.C. Jan. 15, 2016) (noting that North Carolina adopted a reasonable-impediment exception “materially indistinguishable from South Carolina’s” during the course of litigation); Milwaukee Branch of the NAACP v. Walker, 357 Wis.2d 469, 851 N.W.2d 262, 278-79 (2014) (construing voter ID law so as to avoid constitutional infirmity).12

Cognizant that the Supreme Court may itself choose to refine Section 2 law in light of Gingles, Crawford, and Shelby County, or that Congress may revisit the topic as other affected groüps, such as young people, the working poor, and the elderly mobilize, I concur in the majority opinion, having offered these respectful responses to arguments made in dissent.

. See Texas v. Holder, 888 F.Supp.2d 113, 128 (D.D.C. 2012) ("SB 14 is far stricter than either Indiana's or Georgia’s voter ID laws.”), vacated on other grounds, - U.S. -, 133 S.Ct. 2886, 186 L.Ed.2d 930 (2013); see also N. Car. State Conference of the NAACP v. McCrory, - F. Supp. 3d -, -, 2016 WL 1650774, at *156 (M.D.N.C. Apr. 25, 2016) ("[North Carolina's voter ID law] is also less burdensome than the Texas ID requirement

. The Operation Push plaintiffs’ arguments mirror some in this case. Compare Operation Push, 932 F.2d at 403 (citing the argument that African-Americans' lack of access to transportation and less flexible work schedules made it more difficult for them to comply with the restrictive registration system), with Veasey v. Perry, 71 F.Supp.3d 627, 664-65 (S.D. Tex. 2014) (citing evidence that low-income Texans, who are disproportionately African-American or Hispanic, not only disproportionately lack qualifying ID, but also have more difficulty taking time off from work to secure ID and “live without vehicles for their own transportation to get to ID-issuing offices”).

. To the extent the dissent argues that League of United Latin American Citizens v. Clements forecloses consideration of the "tenuousness” factor, that case distinguished the “weight” of the state’s interest "from the conventional Zimmer [Senate] factor of tenuousness.” 999 F.2d 831, 871 (5th Cir. 1993) (en banc). And as the Supreme Court recently reminded, that a state interest is legitimate does not necessarily mean courts should ignore evidence of whether a specific law advances that interest or imposes needless burdens. See Whole Women’s Health v. Hellerstedt, - U.S. -, -, 136 S.Ct. 2292, 195 L.Ed.2d 665 (2016).

. The dissent cites law review articles for the propositions that studies collected therein show no effect from voter ID laws on turnout, or even show increased turnout. Those collected studies, which mostly date from 2009 and earlier, did not involve SB 14 and do not make the district court’s acceptance of expert testimony that Texas's law likely will depress turnout clearly erroneous. In any event, scholarship on the effects of voter ID laws is far from uniform. See, e.g., Zoltán Hajnal, et al., Voter Identification Laws and the Suppression of Minority Votes, at 15 (February 2016), http://pages.ucsd.edu/=zhajnal/page5/ *276documents/voterlDhajnaletal.pdf ("For Latinos, Blacks, and multi-racial Americans there are strong signs that strict photo identification laws decrease turnout.”); Bill Hobby, et ah, The Texas Voter ID Law and the' 2014 Election: A Study of Texas's 23rd Congressional District, at 13 (August 2015), https:// bakerinstitute.org/media/files/files/e0029eb8/ PoIitics-VoterID-Jones-080615.pdf (concluding that Hispanic non-voters in one Texas congressional district “were significantly more likely than Anglo non-voters to strongly agree or agree that a lack of photo ID was a reason that they did not cast a ballot in the 2014 general election”); Michael D. Gilbert, The Problem of Voter Fraud, 115 Colum. L. Rev. 739, 749 & n.55 (2015) ("Other studies suggest voter ID laws do depress votes.” (collecting studies)).

. Scholars have made the same point. See Tokaji, supra, at 475-76 ("Existing empirical methods are simply not up to the task of establishing the effect of a particular practice on turnout, let alone on turnout by particular subgroups, with any degree of precision.”); Samuel Issacharoff, Ballot Bedlam, 64 Duke L.J. 1363, 1383 (2015) (“There has not been enough time to test the observations against normal fluctuations in turnout ... and other confounding political factors.”); Michael J. Pitts, Empirically Measuring the Impact of Photo ID Over Time and Its Impact on Women, 48 Ind. L. Rev. 605, 606 (2015) ("[I]t can be difficult to determine the amount of actual disenfranchisement caused by photo identification laws.”); Gilbert, supra, at 750 ("Gathering relevant data and designing conclusive tests presents many challenges.”).

. The district judge, contrastingly to this case, noted that the record did not contain "adequate evidence on any of [the Senate] factors to enable an appropriate evaluation.” Gonzalez, 2006 WL 3627297, at *8.

. Because Texas selected the requisite voter qualifications and the manner of implementing them, which the trial court found interact with the effects of discrimination to cause racial disparities in opportunity to vote, considering the effects of private discrimination among other factors does not violate the Supreme Court’s warning against imposing disparate-impact liability when “the plaintiff cannot point to a defendant's policy or policies causing [a statistical] disparity.’’ Inclusive Cmties., 135 S.Ct. at 2523; cf. id. at 2524 (opining that a Fair Housing Act plaintiff might not be able to show “a causal connection between the Department’s policy and a disparate impact” if, "for instance ... federal law substantially limits the Department's discretion”).

. Tokaji, supra, at 475.

. See Veasey, 71 F.Supp.3d at 642 (chart showing that, in terms of types of ID accepted, SB 14 is the strictest voter ID law in the country).

. Compare Veasey, 71 F.Supp.3d at 668-69 (explaining that SB 14 requires a birth certificate or similar document to get a free qualifying ID), with McCrory, - F. Supp. 3d at -, 2016 WL 1650774, at *26 (explaining that a North Carolinian can secure free voter ID by supplying a Social Security number and two of approximately twenty supporting documents, including medical records, prison ID, and paycheck stubs), and Lee v. Va. State Bd. of Elections, - F. Supp. 3d -, -, 2016 WL 2946181, at *24 (E.D. Va. May 19, 2016) (“[E]ligible voters do not need to present any independent documentation to obtain a free voter form of identification under Virginia Code § 24.2-643 and its implementing regulations. The statute simply requires that a registrant provide her name, address, birth-date, and social security number and sign the registration form swearing that the information provided is true and correct.”).

.North Carolina’s ID requirement, for example, had a two-year "soft rollout,” and the state's more extensive educational efforts included mailings offering help to voters whom a study indicated might not have qualifying ID. See McCrory, - F. Supp. 3d at -- -, 2016 WL 1650774, at *19-26; see also Common Cause/Ga. v. Billups, 504 F.Supp.2d 1333, 1378-79 (N.D. Ga. 2007) (discussing Georgia’s "exceptional” educational efforts), vacated in part on other grounds, 554 F.3d 1340 (11th Cir. 2009).

. I also disagree with the opposite criticism that this interbranch engagement ameliorates too little, though that argument is contributory. See Richard L. Hasen, Softening Voter ID Laws Through Litigation: Is it Enough?, Wisc. L. Rev. Forward (forthcoming 2016), http:// papers.ssrn.com/sol3/papers.cfm?abstract_ id=2743946 .(with apologies to Professor Ha-sen for my citation of his draft version).