Section 2 Zero-Sums

How the Supreme Court Misconstrued the Voting Rights Act, Limits Minority Representation, and Dangerously Pits Minorities Against Each Other

Salvador E. Pérez *


Man’s capacity for justice makes democracy possible; but man’s inclination to injustice makes democracy necessary.
—Reinhold Niebuhr 1 Reinhold Niebuhr, The Children of Light and the Children of Darkness: A Vindication of Democracy and a Critique of Its Traditional Defense, at xi (1944).

Fifty years after the passage of the Voting Rights Act (VRA), legal controversies surrounding voter identification, 2 See, e.g., Adam Liptak, Courts Strike Down Voter ID Laws in Wisconsin and Texas, N.Y. Times (Oct. 9, 2014),; Adam Liptak, Supreme Court Allows Texas to Use Strict Voter ID Law in Coming Election, N.Y. Times (Oct. 18, 2014),; Zachary Roth, Supreme Court Could Rule on Wisconsin Voter ID Law, MSNBC (Jan. 8, 2015, 11:25 AM), early voting, 3 See, e.g., Lyle Denniston, Early Voting in Ohio Blocked, SCOTUSblog (Sept. 29, 2014, 4:01 PM), voter registration, 4 See, e.g., Lyle Denniston, Court Allows North Carolina Voting Limits, SCOTUSblog (Oct. 8, 2014, 7:14 PM), campaign finance, 5 See, e.g., Robert Barnes, Supreme Court Strikes Down Limits on Federal Campaign Donations, Wash. Post (Apr. 2, 2014), and redistricting 6 See, e.g., Richard Hasen, Argument Preview: Racial Gerrymandering, Partisan Politics, and the Future of the Voting Rights Act, SCOTUSblog (Oct. 30, 2014, 3:02 PM), continue to raise fundamental questions that, when answered, inform a theory of democracy. What burdens on voters (importantly, the poor and minorities) are permissible? Are partisan considerations a legitimate basis for gerrymanders? Is political equality an acceptable justification for restricting campaign contributions and expenditures—or is that concept foreign to the First Amendment? Essentially, these controversies are concerned with the question of whose voice is given consideration. Who, in effect, gets to command the levers of power?

The central subject of this Essay is section 2 of the VRA 7 42 U.S.C. § 1973 (2013). Section 2 prohibits practices imposed or applied in a manner that results in a denial or abridgement of the right to vote based on race. Id. § 1973(a). Under the provision, plaintiffs must prove that (1) there exists a sufficiently large and geographically compact minority population to constitute a majority in a single-member district; (2) the minority group is politically cohesive; and (3) the white majority votes sufficiently as a bloc to usually defeat the minority’s preferred candidate. Thornburg v. Gingles, 478 U.S. 30, 50-51 (1986). Once these gatekeeping conditions have been met, section 2 has been violated if plaintiffs can show “based on the totality of circumstances” that members of a racial minority “have less opportunity than other members of the electorate to participate in the political process and to elect representatives of their choice.” 42 U.S.C. § 1973(b). In assessing section 2 claims under a totality of the circumstances, the proportionality inquiry—a comparison of the percentage of the total districts in which minority voters can elect their chosen candidate with the minority share of the citizen voting age population—is important. League of United Latin Am. Citizens v. Perry, 548 U.S. 399, 436 (2006). and the extent to which it should protect the voice of minority voters from being drowned out by that of the majority in the redistricting context. To be sure, this provision of the VRA—along with subsequent amendments and judicial interpretations—has led to the creation and proliferation of majority-minority districts and has no doubt succeeded in increasing minority representation in Congress. 8 Indeed, when evaluating its remedial value, one need only consider a counterfactual scenario in which the VRA does not exist and all congressional districts are demographic microcosms of the United States. One political scientist concluded that, in such a world, “the probability that a nationally representative district elects an African American to an open seat in Congress is 0.85%, a Latino, 0.47%.” Barry C. Edwards, Formulating Voting Rights Act Remedies to Address Current Conditions, 42 Am. Pol. Res. 376, 391 (2014). In other words, of 435 seats, African Americans would hold 4 and Latinos only 2. Id. One other related data point that highlights the Act’s remedial value is the exceedingly few majority-white constituencies represented by elected officials of color. See Janie Boschma, Why Don’t White Voters Elect Minority Reps to Congress?, Nat’l J. (Jan. 30, 2015), (“[M]inority lawmakers represent only 15 of the 318 districts where whites represent a majority of the population.”); Josh Kraushaar, Democrats’ Diversity Problem, Nat’l J. (Nov. 30, 2010), (“Of the 75 black, Hispanic, and Asian-American Democrats in Congress and governorships, only nine represent majority-white constituencies—and that declines to six in 2011.”). Despite the successes of section 2, however, it is no secret that women and people of color remain severely underrepresented in elected office at every level of government. Members of Congress, for example, are more than eighty percent male and about eighty percent white. 9 Aaron Blake, Yes, Politics Is Still Dominated by Old, White Men. Here’s Why., Wash. Post Fix (Sept. 3, 2014),

Nevertheless, in spite of this poverty of minority representation, the U.S. Supreme Court in 2009 misconstrued section 2 and removed a class of claims from its coverage, handicapping its effectiveness. In Bartlett v. Strickland, the Court ruled that the protections furnished to minority voters by remedial districts are only available in districts in which a minority makes up more than fifty percent of the electorate. 10 556 U.S. 1, 25-26 (2009) (plurality opinion). Chief Justice Roberts and Justice Alito joined Justice Kennedy’s opinion. Id. at 2. Justices Thomas and Scalia concurred only in the judgment, arguing that section 2 “does not authorize any vote dilution claim.” Id. at 26 (Thomas, J., concurring in the judgment) (emphasis added). In essence, the Court determined that in the redistricting context the statute guarantees minority voters the same “opportunity [as other voters] to participate in the political process and to elect representatives of their choice” 11 42 U.S.C. § 1973(b). unless, of course, those minority voters do not happen to live in a location characterized by especially large concentrations of voters of the same minority group.

The central argument here is not merely that Bartlett was wrongly decided. The thesis of this Essay is that by requiring that a minority group compose a numerical majority of the citizen voting age population (CVAP) for the creation of a section 2 district, the Court has limited already scarce opportunities for minority representation and—given shifting demographic realities—inflamed racial tensions between black and Latino communities. Redistricting in South Los Angeles is, in this regard, a glimpse into the future.

I. Latino Political Incorporation, Pursued—but at Black Representation’s Expense?

The size of the U.S. Latino electorate will likely double within a generation to compose nearly 16% of the CVAP. 12 See Paul Taylor et al., Pew Research Ctr., An Awakened Giant: The Hispanic Electorate Is Likely to Double by 2030, at 6 tbl.1 (2012), available at (showing that the Hispanic citizen voting eligible population is estimated to grow to 40 million by 2030, roughly sixteen percent of the anticipated total citizen voting eligible population of 256 million). A jurisdiction’s CVAP is its eligible voter population. By contrast, a jurisdiction’s voting age population (VAP) accounts for its entire adult population regardless of citizenship status. As such, CVAP and VAP can vary significantly within the same jurisdiction. This is especially true when it comes to Latino communities given their disproportionate noncitizen composition. As such, a Latino community’s VAP is very likely substantially larger than its CVAP. In California, that number is projected to reach 38% by 2040. 13 Mindy Romero, Cal. Civic Engagement Project, Policy Brief Issue 7, Is Demography Political Destiny?: Population Change and California’s Future Electorate 3 (2014), available at Each month for the next two decades, 50,000 U.S.-born Latinos will become eligible to vote. 14 Jonathan Capehart, 50,000 Shades of Dismay for the GOP, Wash. Post PostPartisan (Nov. 19, 2012, 2:33 PM ET),

These facts, unfortunately, obscure the historical exclusion of Latinos from the political process as well as their contemporary underrepresentation. Latinos comprise approximately 17% of the U.S. population, 15 State & County QuickFacts, U.S. Census Bureau, (last modified Feb. 5, 2015). To be sure, because of their disproportionate youth and noncitizen composition, Latinos make up only 10.8% of eligible voters (CVAP) nationwide. Mark Hugo Lopez & Ana Gonzalez-Barrera, Pew Research Ctr., Inside the 2012 Latino Electorate 5 (2013), available at
and yet only 29 U.S. Representatives (6.7%) and only 3 U.S. Senators (3%) are Latino. 16 Roque Planas, The Most Latino Congress Ever Is Coming in 2015, Huffington Post (Nov. 5, 2014, 9:59 PM EST),

Predictably, these conditions have led to pent-up demands for equal representation as well as increased Latino political cohesiveness and preferences for coethnic candidacies. Fervent desires for representation are expressed through an increasing willingness to engage the political process in ever more sophisticated ways, including political action committees to groom and support Latino candidates 17 See, e.g., Eliza Newlin Carney, Rules of the Game: Hispanic Caucus Leverages Latino Power, Roll Call (Mar. 24, 2013, 7:44 PM), (“Through its increasingly lucrative political action committee, known as BOLD PAC, the [Congressional Hispanic Caucus] helped elect nine more Latinos to the House in November . . . .”); Sheryl Gay Stolberg, In California Race, a Latina Democrat Carries Hopes of Her Party and People, N.Y. Times (May 4, 2014), (“On Monday, Cinco de Mayo, a new nonpartisan organization, the Latino Victory Project, announced an effort to promote Hispanic political engagement, in part by grooming Latino candidates . . . .”). (candidacies which have—not coincidentally—secured electoral rewards for party campaign committees 18 See Shaila Dewan, G.O.P. Gains by Tapping Democrats’ Base for State Candidates, N.Y. Times (Nov. 29, 2014), ). Additionally, campaigns to naturalize Latino permanent residents, register eligible Latino voters, and turn out registered Latino voters in a sustained fashion have become seemingly perennial. 19 See, e.g., Jacquellena Carrero, Latino Groups Target Voters Ahead of Midterms, NBC News (Apr. 17, 2014, 2:56 AM),; Miriam Jordan, Univision Gives Citizenship Drive an Unusual Lift, Wall St. J. (May 10, 2007, 12:01 AM ET), Some justifiably impatient advocates may go so far as to look to total raw population or voting age population (VAP) figures and say to themselves, “If our political power were reflective of our size, there would be 65 Latino House members, 20 See Lopez & Gonzalez-Barrera, supra note 15, app. at 11 tbl.1 (finding that Latinos comprise approximately 15.03% of the U.S. VAP). and about 18 of them would come from California. 21 California has 53 House members, see Directory of Representatives: California, U.S. House Representatives, (last visited Feb. 27, 2015), and Latinos constitute approximately 34% of California’s adult population, see Mark Baldassare et al., Pub. Policy Inst. of Cal., Latino Likely Voters in California 1 (2014), available at The state of our underrepresentation is such that we must pursue every avenue to political incorporation without compromise—even if it comes at the expense of other minority groups.”

This mindset is not without adherents. When black California Congresswoman Juanita Millender-McDonald suddenly died of cancer in 2007, the Congressional Hispanic Caucus targeted her seat, provoking the ire of the Congressional Black Caucus. As the Chairman of the Congressional Hispanic Caucus at the time, Representative Joe Baca, put it, “It’s time we have one of our own that speaks on our behalf.” 22 Steven Malanga, The Rainbow Coalition Evaporates, City J. (Winter 2008), Millender-McDonald’s district (which encompassed much of Long Beach, Watts, Compton, Signal Hill, and Carson) was 25% black and 43% Latino as of the 2000 census. 23 Harold Meyerson, Op-Ed., The Delicate Balance of Black and Brown, L.A. Times (June 24, 2007), Those numbers, however, did not translate into voting strength for the Latino community in light of handicaps related to low turnout and voting eligibility. 24 See Rachel Kapochunas, Ethnicity a Key Factor in Tuesday’s California House Special Primary, N.Y. Times (June 27, 2007), (noting that Latino voting strength is often reduced by low turnout and voting eligibility, and predicting that this would impact the Millender-McDonald runoff).

Candidates to replace Representative Millender-McDonald included two seasoned state legislators: State Senator Jenny Oropeza (a Latina) and Assemblywoman Laura Richardson (a black woman). 25 Id. Ethnic-based allegiances immediately flared, leading to support for the candidates from generally predictable quarters. Richardson drew the support of several prominent African American politicians and the California Legislative Black Caucus, 26 John L. Mitchell, Racial Issues Take a Back Seat in 37th, L.A. Times (July 3, 2007), while Oropeza garnered the support of the Congressional Hispanic Caucus and the California Latino Legislative Caucus. 27 Id. When it came to the electorate, a subsequent analysis of voter behavior found

very clear, and statistically significant evidence of racially polarized voting. Blacks voted almost unanimously for two African American candidates Laura Richardson and Valerie McDonald, and gave almost no votes at all to the Latino candidate Jenny Oropeza. In contrast, Latino voters in the district voted very heavily for Oropeza, and cast very few votes for the two major Black candidates in the contest. 28 Matt A. Barreto, A Summary of Voting Patterns in Los Angeles County 3-4 (2011), available at

The Millender-McDonald special election would be of only mild concern if it were anomalous. Central and southwestern Los Angeles County—like many other parts of the country—continues to experience dramatic demographic change. 29 Id. at 3. Over the course of the past twenty-five years, many cities and neighborhoods have transitioned from majority black to majority or plurality Latino. 30 See id. Additionally, numerous studies continue to find racial bloc voting, especially during primary contests. 31 Id.; Bernard L. Fraga, Race, Party, and Candidate Prospects Across the Multiple Stages of Congressional Elections 19 (Mar. 12, 2014) (unpublished manuscript), available at (finding that race is “the salient factor” in determining who runs for and wins primary elections).

Indeed, Latino population growth combined with relative black population decline and the persistence of racially polarized voting makes section 2 violations in this and similar geographies a growing possibility. Put simply, black-controlled congressional districts with less-than-majority-black electorates are vulnerable to Latino population growth because they are afforded no protection under current section 2 doctrine. And that vulnerability will, no doubt, lead to well-founded fears and recriminations. In California, for instance, during a particularly tense stage in the redistricting process, one black community leader exhorted the Citizens Redistricting Commission in the press, declaring, “The Voting Rights Act is being used to disadvantage Black people in Los Angeles.” 32 Jackie Dupont-Walker, Op-Ed., Black Community Leaders Urge Redistricting Commission to Stop Gutting Our Political Representation, L.A. Sentinel (July 15, 2011, 6:02 PM),

Prior to the 2010 redistricting cycle, three congressional districts in South Los Angeles were historically represented by African Americans, 33 See Jean Merl, Minority Representation a Challenge for Redistricting Commission, L.A. Times (July 25, 2011), (“That may mean trouble for three black Congress members—Democrats Karen Bass, Maxine Waters and Laura Richardson—in districts historically represented by African Americans.”). despite not having majority-black populations. At the time, these districts were three of the four congressional districts in California represented by black officials—the only other being Representative Barbara Lee’s, anchored by the city of Oakland. But because of relative declines in the black populations of both Los Angeles County 34 Id. (“L.A. County’s black population dropped from 9.5% to 8.3% between the census done in 2000 and the one completed [in 2010].”). and the State of California, 35 Dan Walters, California’s Black Population Shrinking Proportionately, Sacramento Bee Capitol Alert (Sep. 29, 2011), these districts were suddenly at risk in light of Latino population growth.

Presented with the strictures of section 2 and this demographic reality, the Commission considered a number of different options. One early option concentrated black voters into a single district, but some Commissioners criticized the plan for making adjacent districts less favorable than before for African American candidates. 36 See Raphael J. Sonenshein, League of Women Voters of Cal., When the People Draw the Lines: An Examination of the California Citizens Redistricting Commission 50 (2013), available at The Commission next gravitated toward a map that split black voters across three districts. 37 Id. This, in many ways, was the preferred outcome for the black community, as the three districts mirrored a profile of districts in which the black community had a history of successfully electing candidates of its choice. 38 Id.

The final map, however, reflected the following changes to the three districts: the 37th district (Culver City-Crenshaw) is now 34.09% black CVAP and 20.82% Latino CVAP; the 43rd district (Inglewood-Torrance) is now 32.54% black CVAP, with a nearly equal Latino CVAP of 28.72%; and the 44th district (Compton-Carson-San Pedro)—the Millender-McDonald district, more or less—is now a majority-Latino district created to comply with section 2. 39 See Meridian Pac., Inc., Plan Analysis—Demographics and Past Election Performance (2011), available at; Sandra Hernandez, From L.A.’s New Political Maps, an Interesting Congressional Race, L.A. Times (May 10, 2012), (“[T]he California Citizens Redistricting Commission established the 44th . . . as [a] majority Latino district[] . . . .”).

On balance, the black community retains control of two of the three districts, although it is unclear how long its grasp on the 37th and 43rd will hold. 40 In 2011, VRA counsel for the California Citizens Redistricting Commission discounted the viability of section 2 claims in this portion of Los Angeles County. Nevertheless, he qualified his view by referencing the record available at the time. Cal. Citizens Redistricting Comm’n, Business Meeting 25 (June 16, 2011) (transcript available at
Given the black population of the state, however, rough proportionality—a significant factor in the totality of the circumstances test for determining whether section 2 has been violated 41 See supra note 7. —would favor the existence of three black-controlled districts in California. 42 California’s black population is six percent of its total population. See Baldassare et al., supra note 21, at 1. Six percent of 53 (the size of California’s House delegation), Directory of Representatives: California, supra note 21, is roughly 3. This assertion, of course, assumes Gingles conditions are met. See supra note 7. Without a mandate to create section 2 districts with less-than-majority minority populations, however, this is a difficult proposition to maintain. Indeed, without the protection of such a mandate, black districts will be increasingly encroached upon and black voting strength will be diluted by Latino population growth on an expedited timetable.

II. Avoiding Political Famines and Political Feasts

The Supreme Court’s insistence in Bartlett that a proposed minority district be fifty percent of the CVAP limits the opportunities for minority representation to such a degree that current doctrine encourages explosive and racially charged political contests like the special election presented above. As in South Los Angeles, this limitation creates circumstances in which remedies are available for either blacks or Hispanics but not for both (and seemingly at each other’s expense). Ultimately, section 2—as presently misconstrued—will bring about a world in which black representation will be whittled down to a single congressional district in Los Angeles. To borrow language from Justice Souter, Latinos will enjoy a “political feast” while blacks suffer a “political famine.” 43 See Johnson v. De Grandy, 512 U.S. 997, 1017 (1994).

One way to avoid this world would be to abandon the formalism of Bartlett and embrace a functional understanding of section 2. A more functional approach would acknowledge that “a district may be a minority-opportunity district so long as a cohesive minority population is large enough to elect its chosen candidate when combined with a reliable number of crossover voters.” 44 Bartlett v. Strickland, 556 U.S. 1, 27 (2009) (Souter, J., dissenting) (emphasis added). (These de facto majority-minority districts have been termed “crossover districts” by many.) 45 See, e.g., Editorial, Narrowing the Voting Rights Act, N.Y. Times (Mar. 10, 2009), This seems, in fact, to be the approach the California Citizens Redistricting Commission at one point attempted to adopt with regard to black communities by allocating them across three congressional districts with black CVAPs in the 30-40% range. 46 See Sonenshein, supra note 36, at 49. But there is a vast difference in outcomes between a holding that permits redistricting bodies to draw these types of districts (as Bartlett allows) and one that commands them to do so. Effectively, instead of “foster[ing] our transformation to a society that is no longer fixated on race,” as Justice Kennedy has called for, 47 League of United Latin Am. Citizens v. Perry, 548 U.S. 399, 434 (2006) (quoting Georgia v. Ashcroft, 539 U.S. 461, 490 (2003)) (internal quotation mark omitted). his opinion in Bartlett promotes racial blocs by requiring states to pack minority voters into fifty-percent-plus-one majority-minority districts, “contracting the number of districts where racial minorities are having success in transcending racial divisions in securing their preferred representation.” 48 Bartlett, 556 U.S. at 27 (Souter, J., dissenting).

Equalizing the opportunity of minorities to participate in the political process and elect representatives of their choice indeed requires the Court to abandon its bright-line Bartlett rule. A recent notable empirical analysis that tested minority voter success in congressional elections under various possible VRA standards supports abandoning the Bartlett rule. 49 Edwards, supra note 8. It found, for example, that the “objective of the VRA is best accomplished by districts with 45% to 50% African American VAP and districts with 60% to 65% Latino VAP.” 50 Id. at 378. Edwards found that these standards best equalize the opportunity of minorities to elect representatives of their choice. His conclusion is based on an analysis of recent congressional elections results, 2010 population data, and voting simulations. Id. (It is important to recall here that VAP and CVAP are different figures.) 51 See supra note 12.

To be sure, creating crossover districts that protect minority voters against subordination necessitates flexibility and a willingness to apply different standards to different minority groups. Flexibility would undo inequitable dilution of minority voting power and increase opportunities for minority voters, thereby lowering the stakes and potential racial animosities between minority communities. Plainly, mandating the creation of crossover districts would mitigate interracial disputes by enlarging the pie of seats controlled by minority voters.

For example, if the Court were to endorse flexible standards and redistricting authorities were to adopt a 61.2% VAP standard for Latinos, “Latino voters [would] succeed in electing roughly 50 of their preferred candidates to Congress.” 52 Edwards, supra note 8, at 396. Fifty Latino-preferred House members would be a much-welcomed improvement on current representation (29). 53 See supra notes 16, 20-21 and accompanying text. The “current representation” figure of 29 (the number of Latinos in the U.S. House of Representatives) is admittedly a crude approximation considering that the Latino-preferred candidate may in fact not be Latino. Nevertheless, because of the prevalence of racially polarized voting and preferences for coethnic candidacies, equating the two is not inappropriate. To illustrate, prior to the 2014 midterm elections, only 5 of 23 congressional districts with majority-Latino electorates were represented by non-Latino members of Congress. Compare Mapping the Latino Electorate by Congressional District, Pew Res. Center (Oct. 16, 2014), (listing congressional districts by “Share Latino among Eligible Voters”), with Latino Members of Congress: 113th Session, NALEO Educ. Fund 2, _2012.pdf (last visited Feb. 27, 2015) (listing Latino members of the 113th Congress). Such flexibility would reap considerable rewards for black communities as well. Black representation could significantly increase in states such as Connecticut, Kentucky, Oklahoma, and Wisconsin “that currently could not create majority African American districts, but could potentially create districts with 30% to 50% African American voters.” 54 Edwards, supra note 8, at 397.

Instead of keeping faith with the VRA’s remedial purpose, the Supreme Court handicapped section 2 in favor of formalism and conserving an uncertain amount of judicial resources by striking from the statute’s coverage a class of meritorious claims. In its stodgy misconstruction, the Court distorted the law and perpetuated, as other actors have, “the perception that the Act is a blunt mandate to tally and bundle minority voters into districts pegged at talismanic target percentages,” thereby “treat[ing] the Act as a demographic imperative, deaf to local political conditions.” 55 Justin Levitt, Color by Numbers: The New Misreading of the Voting Rights Act 3 (Loyola Law Sch., L.A., Legal Studies Paper No. 2014-35, 2014), available at The Court, in the end, failed to see the Act for what it is—a sophisticated, tailored, and nuanced law that protects minority voters, allowing them to participate in the electoral process on an equal footing with other members of the electorate.


While the effectiveness of nonjudicial, political means of growing Latino representation remains to be seen, at present the political process merely serves as a safety valve for mounting frustrations within Latino communities over underrepresentation. The political thicket, too, is ripe with explosive potential for racially charged confrontations between minority communities. As the Millender-McDonald special election demonstrated, when opportunities for minority representation are scarce, blacks and Latinos eager to make their voices heard are often pitted against one another.

If the Supreme Court is to be faithful to the purposes of the VRA, it will have to reevaluate its section 2 doctrine, adopt a functional understanding of the commands of section 2, and mandate the creation of crossover districts. Such an approach would vindicate the preferences of minority voters, increase minority representation, disarm existing (and growing) racially charged electoral conflicts between minority communities, and reward successful efforts by racial minorities to transcend racial divisions in securing their preferred representatives.

Salvador E. Pérez received his J.D. from Stanford Law School and his M.P.A. from the John F. Kennedy School of Government at Harvard University in 2014. Mr. Perez thanks Matthew Forbes, Jacob Shelly, Jessica Witte, and the editors of Stanford Law Review Online for their insightful edits. He owes a debt of gratitude to Nathaniel Persily and Pamela S. Karlan for sparking and nurturing his interest in election law.