angela m. spiveyiframe.dri.org/dri/course-materials/2017-classactions/pdfs/07_spivey.pdf301 f.r.d....
TRANSCRIPT
Ascertainability
Angela M. Spivey
McGuireWoods LLP
1230 Peachtree Street NE Atlanta, GA 30309 (404) 443-5720
Angela M. Spivey is managing partner of the McGuireWoods LLP Atlanta office. She defends and counsels food companies on a host of issues, including implementation of recalls, defense of widespread international outbreaks and resulting civil litigation, defense of corporations and individuals in OCI criminal investigations, and regulatory oversight and compliance. Ms. Spivey is a national speaker at various food industry group conferences and has first-chair trial experience before state and federal courts on matters ranging from product liability, personal injury, contract disputes, and business torts.
Ascertainability ■ Spivey ■ 3
Ascertainability
I. Introduction ...................................................................................................................................................5 II. Briseño v. Conagra Foods, Inc., U.S. Court of Appeals for the Ninth Circuit. Appellant’s
Opening Brief .................................................................................................................................................6 III. Briseño v. Conagra Foods, Inc., U.S. Court of Appeals for the Ninth Circuit. Answering
Brief of Plaintiffs-Appellees .........................................................................................................................81 IV. Briseño v. Conagra Foods, Inc., U.S. Court of Appeals for the Ninth Circuit. Appellant’s
Reply Brief ..................................................................................................................................................171 V. Briseño v. Conagra Foods, Inc., U.S. Court of Appeals for the Ninth Circuit. Opinion ..........................209
Table of Contents
Ascertainability ■ Spivey ■ 5
Ascertainability
I. IntroductionThe federal courts of appeals have adopted different approaches regarding whether and to what
extent a class must be ascertainable to be certified. The following materials provide context for an overview of the current state of the law on ascertainability and practical suggestions for defeating class certification under the various standards.
6 ■ Class Actions ■ July 2017
II. Briseño v. Conagra Foods, Inc., U.S. Court of Appeals for the Ninth Circuit. Appellant’s Opening BriefNo. 15-55727
In the
United States Court of Appeals For the Ninth Circuit
__________________
ROBERT BRISEÑO, Plaintiff-Appellee,
v.
CONAGRA FOODS, INC., Defendant-Appellant. _________________
ON APPEAL FROM THE UNITED STATES DISTRICT COURT FOR THE CENTRAL DISTRICT OF CALIFORNIA
(The Honorable Margaret M. Morrow) Case No. 2:11-cv-05379 __________________
APPELLANT’S OPENING BRIEF__________________
A. Brooks GreshamLaura E. Coombe MCGUIREWOODS LLP1800 Century Park East, 8th Floor Los Angeles, CA 90067 Telephone: (310) 315-8291
Angela M. Spivey MCGUIREWOODS LLP1230 Peachtree St, NE, Suite 2100 Atlanta, GA 30309 Telephone: (404) 443-5720
R. Trent Taylor MCGUIREWOODS LLPGateway Plaza 800 E. Canal St. Richmond, VA 23219 Telephone: (804) 775-1182
E. Rebecca Gantt MCGUIREWOODS LLPWorld Trade Center 101 W. Main St., Suite 9000 Norfolk, VA 23510 Telephone: (757) 640-3731
Counsel for Defendant-Appellant ConAgra Foods, Inc.
Case: 15-55727, 09/21/2015, ID: 9691076, DktEntry: 13, Page 1 of 75
Ascertainability ■ Spivey ■ 7 -i-
CORPORATE DISCLOSURE STATEMENT
Pursuant to Federal Rule of Appellate Procedure 26.1, Defendant-Appellant
ConAgra Foods, Inc., hereby states, by and through counsel, that it does not have a
parent corporation and that no publicly held corporation owns 10% or more of its
stock.
Dated: September 21, 2015 Respectfully submitted,
MCGUIREWOODS LLP
/s/ A. Brooks Gresham A. Brooks Gresham
Counsel for Defendant-Appellant ConAgra Foods, Inc.
Case: 15-55727, 09/21/2015, ID: 9691076, DktEntry: 13, Page 2 of 75
8 ■ Class Actions ■ July 2017 -ii-
TABLE OF CONTENTS
CORPORATE DISCLOSURE STATEMENT ......................................................... i STATEMENT WITH RESPECT TO ORAL ARGUMENT ................................ xii INTRODUCTION .................................................................................................... 1 JURISDICTIONAL STATEMENT ......................................................................... 4 STATEMENT OF THE ISSUES.............................................................................. 5 STATEMENT OF THE CASE ................................................................................. 5 SUMMARY OF THE ARGUMENT ....................................................................... 8 ARGUMENT .......................................................................................................... 10 I. STANDARD OF REVIEW .......................................................................... 11 II. THE DISTRICT COURT ERRONEOUSLY HELD THAT THE
PUTATIVE CLASSES ARE ASCERTAINABLE ..................................... 12 A. Ascertainability is a threshold requirement for certification ............. 13 B. A class is only ascertainable where its members can be
determined in an objectively verifiable and administratively feasible manner ................................................................................... 14
C. The district court abused its discretion by relying solely on the proffered class definition to hold that the classes’ members were ascertainable .............................................................................. 18
III. THE NAMED REPRESENTATIVES DO NOT HAVE CLAIMS AND DEFENSES THAT ARE TYPICAL OF THOSE OF THE PUTATIVE CLASS MEMBERS ................................................................. 23
IV. THE DISTRICT COURT ERRONEOUSLY HELD THAT COMMON QUESTIONS PREDOMINATE OVER INDIVIDUAL ONES ............................................................................................................ 27 A. Individual issues predominate with respect to materiality
because there is no reliable classwide method of proof ..................... 28 1. Plaintiffs must show they can prove materiality to avoid
individualized issues of reliance and causation ....................... 29 2. The district court erroneously relied on surveys that do
not support a finding of classwide materiality ......................... 31
Case: 15-55727, 09/21/2015, ID: 9691076, DktEntry: 13, Page 3 of 75
Ascertainability ■ Spivey ■ 9
TABLE OF CONTENTS(continued)
Page
-iii-
a. Plaintiffs’ proffered method of proving materiality was not tailored to their theory that consumers factor a belief that the Wesson Oil natural label means the absence of GMOs into their purchasing decision .......................................................................... 32
b. The district court erred in finding it appropriate to apply a classwide presumption of materiality when at least one-third of a proposed class does not find the label claim material .................................................. 38
3. The district court did not consider ConAgra’s own superior rebuttal evidence which shows that the term ‘natural’ holds so many different meanings for different consumers that materiality cannot be shown ........................... 41
B. Individual issues predominate with respect to damages .................... 45 1. Comcast requires a damages model linked to Plaintiffs’
theory of liability ...................................................................... 46 2. The district court twice rejected one of the models that
formed the basis of the hybrid model ...................................... 47 3. The district court erroneously accepted a “hybrid” model
that was incapable of reliably or persuasively calculating classwide damages ................................................................... 48
V. A CLASS ACTION IS NOT A SUPERIOR VEHICLE FOR ADJUDICATING PLAINTIFFS’ CLAIMS BECAUSE IT IS UNMANAGEABLE..................................................................................... 57
CONCLUSION ....................................................................................................... 59 STATEMENT OF RELATED CASES .................................................................. 59 CERTIFICATE OF COMPLIANCE ...................................................................... 61 CERTIFICATE OF SERVICE ............................................................................... 62
Case: 15-55727, 09/21/2015, ID: 9691076, DktEntry: 13, Page 4 of 75
10 ■ Class Actions ■ July 2017 -iv-
TABLE OF AUTHORITIES
Page(s)
Cases
Algarin v. Maybelline, LLC,300 F.R.D. 444 (S.D. Cal. 2014) .................................................................. 16, 42
Allen v. Hyland’s, Inc.,300 F.R.D. 643 (C.D. Cal. 2014) ........................................................................ 44
Amchem Products, Inc. v. Windsor,521 U.S. 591 (1997) ................................................................................ 27, 28, 32
Astiana v. Ben & Jerry’s Homemade, Inc.,No. C 10-4387-PJH, 2014 WL 60097 (N.D. Cal. Jan. 7, 2014) ................... 13, 16
Astiana v. Kashi Co.,291 F.R.D. 493 (S.D. Cal. 2013) .................................................................. 42, 44
Bakalar v. Vavra,237 F.R.D. 59 (S.D.N.Y. 2006) .......................................................................... 20
Beck v. Maximus, Inc.,457 F.3d 291 (3d Cir. 2006) ............................................................................... 26
Berger v. Home Depot USA, Inc.,741 F.3d 1061 (9th Cir. 2014) ............................................................................ 13
Brazil v. Dole Packaged Foods, LLC,No. 12-CV-01831-LHK, 2014 WL 5794873(N.D. Cal. Nov. 6, 2014) ..................................................................................... 48
Bruton v. Gerber Products Co.,No. 12-CV-02412-LHK, 2014 WL 2860995(N.D. Cal. June 23, 2014) ............................................................................. 16, 21
Byrd v. Aaron’s Inc.,784 F.3d 154 (3d Cir. 2015) ............................................................. 13, 14, 15, 17
Caro v. Procter & Gamble Co.,18 Cal. App. 4th 644 (1993) ......................................................................... 30, 41
Case: 15-55727, 09/21/2015, ID: 9691076, DktEntry: 13, Page 5 of 75
Ascertainability ■ Spivey ■ 11 -v-
Carrera v. Bayer Corp.,727 F.3d 300 (3d Cir. 2013) ......................................................................... 20, 21
Clausen v. M/V New Carissa,339 F.3d 1049 (9th Cir. 2003) ............................................................................ 56
Comcast v. Behrend,133 S. Ct. 1426 (2013) .................................................................................passim
Daubert v. Merrell Dow Pharmaceuticals, Inc.,509 U.S. 579 (1993) ...................................................................................... 50, 51
Ellis v. Costco Wholesale Corp.,657 F.3d 970 (9th Cir. 2011) ............................................................ 11, 12, 19, 51
EQT Production Co. v. Adair, 764 F.3d 347, 359 (4th Cir. 2014) ............. 15, 20, 32
Exxon Corp. v. Texas Motor Exchange of Houston, Inc.,628 F.2d 500 (5th Cir. 1980) .............................................................................. 34
Fairbanks v. Farmers New World Life Ins. Co.,197 Cal. App. 4th 544 (2011) ............................................................................. 40
Fancaster, Inc. v. Comcast Corp.,832 F. Supp. 2d 380 (D.N.J. 2011) ..................................................................... 35
Forrand v. Federal Express Corp.,No. CV 08-1360-DSF, 2013 WL 1793951(C.D. Cal. Apr. 25, 2013) ................................................................................... 32
Fractus, S.A. v. Samsung,Civ. No. 6:09-cv-203-LED-JDL, 2011 WL 7563820(E.D. Tex. Apr. 29, 2011) ................................................................................... 35
Garcia v. Medved Chevrolet, Inc.,263 P.3d 92 (Colo. 2011) .................................................................................... 30
Gartin v. S&M NuTec LLC,245 F.R.D. 429 (C.D. Cal. 2007) ........................................................................ 57
General Telephone Co. of Southwest v. Falcon,457 U.S. 147 (1982) ............................................................................................ 17
Case: 15-55727, 09/21/2015, ID: 9691076, DktEntry: 13, Page 6 of 75
12 ■ Class Actions ■ July 2017 -vi-
Gianino v. Alacer Corp.,846 F. Supp. 2d 1096 (C.D. Cal. 2012) .............................................................. 58
Guido v. L’Oreal USA, Inc.,Nos. 2:11-CV-01067 & -05465-CAS, 2014 WL 6603730(C.D. Cal. Jul. 24, 2014) ..................................................................................... 54
Gunnells v. Healthplan Services, Inc.,348 F.3d 417 (4th Cir. 2003) .............................................................................. 58
Hanon v. Dataproducts Corp.,976 F.2d 497 (9th Cir.1992) ............................................................................... 24
Harolds Stores, Inc. v. Dillard Dep’t Stores, Inc.,82 F.3d 1533 (10th Cir. 1996) ............................................................................ 34
Henry Schein, Inc. v. Stromboe,102 S.W.3d 675 (Tex. 2003) .............................................................................. 40
In re Blood Reagents Antitrust Litig.,783 F.3d 183 (3d Cir. 2015) ............................................................................... 51
In re Celexa & Lexapro Marketing & Sales Practices Litigation,291 F.R.D. 13 (D. Mass. 2013) ........................................................................... 39
In re Clorox Consumer Litigation,301 F.R.D. 436 (N.D. Cal. 2014) ........................................................................ 16
In re Countrywide Financial Corp. Mortgage Marketing & Sales Practices Litigation, Nos. 08md1988-DMS, 2011 WL 6325877 (S.D. Cal. Dec. 16, 2011) .............................................................................. 40, 41
In re NJOY, Inc. Consumer Class Action Litigation,No. CV-14-00428-MMM, 2015 WL 4881091(C.D. Cal. Aug. 14, 2015) ................................................................................... 54
In re POM Wonderful LLC Marketing & Sales Practices Litigation,ML 10-02199-DDP, 2014 WL 1225184(C.D. Cal. Mar. 25, 2014) ............................................................................. 16, 21
In re Tobacco II Cases,207 P.3d 20 (Cal. 2009) ...................................................................................... 29
Case: 15-55727, 09/21/2015, ID: 9691076, DktEntry: 13, Page 7 of 75
Ascertainability ■ Spivey ■ 13 -vii-
In re Vioxx Class Cases,180 Cal. App. 4th 116 (2009) ....................................................................... 39, 40
In re WellPoint, Inc. Out-of-Network UCR Rates Litigation,No. MDL 09-2074-PSG, 2014 WL 6888549(C.D. Cal. Sept. 3, 2014) ..................................................................................... 16
Ivie v. Kraft Foods Global, Inc.,No. C-12-02554-RMW, 2013 WL 685372(N.D. Cal. Feb. 25, 2013) ................................................................................... 43
Jimenez v. Allstate Insurance Co.,765 F.3d 1161 (9th Cir. 2014) ............................................................................ 11
Johnson v. Harley-Davidson Motor Co. Group, LLC,285 F.R.D. 573 (E.D. Cal. 2012) ........................................................................ 41
Jones v. ConAgra Foods, Inc.,No. C 12-01633-CRB, 2014 WL 2702726(N.D. Cal. June 13, 2014) ................................................................ 16, 21, 43, 44
Karhu v. Vital Pharmaceuticals, Inc.,--- F. App’x ---, 2015 WL 3560722 (11th Cir. June 9, 2015) ............................ 16
Khoday v. Symantec Corp.,Civil No. 11-180, 2014 WL 1281600 (D. Minn. Mar. 13, 2014) ....................... 54
Koos v. First Nat’l Bank,496 F.2d 1162 (7th Cir. 1974) ............................................................................ 25
Kosta v. Del Monte Foods, Inc.,No. 12-CV-1722 YGR, 2015 WL 4593175(N.D. Cal. July 30, 2015) .............................................................................. 32, 33
Leyva v. Medline Industries Inc.,716 F.3d 510 (9th Cir. 2013) .............................................................................. 45
Ludlow v. BP, P.L.C.,--- F.3d ---, 2015 WL 5235010 (5th Cir. Sept. 8, 2015) ..................................... 51
Lust v. Merrell Dow Pharmaceuticals, Inc.,89 F.3d 594 (9th Cir. 1996) ................................................................................ 52
Case: 15-55727, 09/21/2015, ID: 9691076, DktEntry: 13, Page 8 of 75
14 ■ Class Actions ■ July 2017 -viii-
Marcus v. BMW of North America, LLC,687 F.3d 583 (3d Cir. 2012) ......................................................................... 14, 22
Martin v. Pacific Parking Systems Inc., 583 F. App’x 803(9th Cir. 2014) ..............................................................................................passim
Massachusetts Mutual Life Insurance Co. v. Superior Court,97 Cal. App. 4th 1282 (2002) ............................................................................. 29
Mevorah v. Wells Fargo Home Mortgage(In re Wells Fargo Home Mortgage), 571 F.3d 953 (9th Cir. 2009) ................. 28
Mullins v. Direct Digital,795 F.3d 654 (7th Cir. 2015) .................................................................. 15, 20, 58
Oracle America, Inc. v. Google Inc.,No. 10-CV-03561, 2012 WL 850705 (N.D. Cal. Mar. 13, 2012) ...................... 54
Oshana v. Coca-Cola Co.,No. 04 C 3596, 2005 WL 1661999 (N.D. Ill. July 13, 2005) ............................. 39
Parra v. Bashas’, Inc.,536 F.3d 975 (9th Cir. 2008) .............................................................................. 12
Pizza Hut, Inc. v. Papa John’s Int’l,227 F.3d 489 (5th Cir. 2000) .............................................................................. 36
Prism Technologies LLC v. AT&T Mobility, LLC,Case Nos. 8:12CV122-126, LLC, 2014 U.S. Dist. LEXIS 132619 (D. Neb. Sept. 22, 2014) ..................................................................................... 53
Randolph v. J.M. Smucker Co., 303 F.R.D. 679(S.D. Fla. 2014). ............................................................................................ 44, 56
Red v. Kraft Foods, Inc.,No. CV-10-1028-GW, 2012 WL 8019257 (C.D. Cal. Apr. 12, 2012) ................................................................................... 23
Rikos v. Procter & Gamble Co.,--- F.3d ---, 2015 WL 4978712 (6th Cir. Aug. 20, 2015) ................................... 13
Case: 15-55727, 09/21/2015, ID: 9691076, DktEntry: 13, Page 9 of 75
Ascertainability ■ Spivey ■ 15 -ix-
Rojas v. General Mills, Inc.,No. 12-cv-05099-WHO, 2013 WL 5568389 (N.D. Cal. Oct. 9, 2013) ...................................................................................... 43
Rubio v. Capital One Bank,613 F.3d 1195 (9th Cir. 2010) ............................................................................ 43
Sethavanish v. ZonePerfect Nutrition Co.,No. 12-2907-SC, 2014 WL 580696(N.D. Cal. Feb. 13, 2014) ................................................................. 16, 20, 21, 23
Stearns v. Ticketmaster Corp.,655 F.3d 1013 (9th Cir. 2011) .....................................................................passim
Steroid Hormone Product Cases, 104 Cal. Rptr. 3d 329(Cal. Ct. App. 2010) ............................................................................................ 30
Strawn v. Farmers Ins. Co. of Oregon,258 P.3d 1199 (Or. 2011) ................................................................................... 40
THOIP v. Walt Disney Co.,690 F. Supp. 2d 218 (S.D.N.Y. 2010) ................................................................ 35
Thurston v. Bear Naked, Inc.,No. 3:11-cv-02890-H (BGS), 2013 WL 5664985(S.D. Cal. July 30, 2013) .................................................................................... 44
Universal City Studios, Inc. v. Nintendo Co.,746 F.2d 112 (2d Cir. 1984) ............................................................................... 34
Vinole v. Countrywide Home Loans, Inc.,571 F.3d 935 (9th Cir. 2009) .............................................................................. 28
Wal-Mart Stores, Inc. v. Dukes,131 S. Ct. 2541 (2011) .................................................................................passim
Water Pik, Inc. v. Med-Systems, Inc.,726 F.3d 1136 (10th Cir. 2013) .......................................................................... 35
Webb v. Carter’s, Inc.,272 F.R.D. 489 (C.D. Cal. 2011) ........................................................................ 30
Case: 15-55727, 09/21/2015, ID: 9691076, DktEntry: 13, Page 10 of 75
16 ■ Class Actions ■ July 2017 -x-
Wells Fargo & Co. v. WhenU.com, Inc.,293 F. Supp. 2d 734 (E.D. Mich. 2003) ............................................................. 35
Wiener v. Dannon Co.,255 F.R.D. 658 (C.D. Cal. 2009) ........................................................................ 25
Xavier v. Philip Morris USA, Inc.,787 F. Supp. 2d 1075 (N.D. Cal. 2011) ........................................................ 14, 18
Yokoyama v. Midland National Life Insurance Co.,594 F.3d 1087 (9th Cir. 2010) ............................................................................ 11
Zinser v. Accufix Research Institute, Inc.,253 F.3d 1180 (9th Cir. 2001) ............................................................................ 57
Statutes
28 U.S.C. § 1292(e) ................................................................................................... 4
28 U.S.C. § 1332(d) ................................................................................................... 4
Rules
Ninth Cir. R. 28-2.6 ................................................................................................. 59
Fed. R. App. P. 5 ........................................................................................................ 4
Fed. R Civ. P. 23 ...............................................................................................passim
Fed. R Civ. P. 23(a) .......................................................................................... 7, 8, 10
Fed. R Civ. P. 23(a)(3) ................................................................................. 23, 25, 26
Fed. R Civ. P. 23(b)(2) ....................................................................................... 6, 7, 8
Fed. R Civ. P. 23(b)(3) ......................................................................................passim
Fed. R. Civ. P. 23(b)(3)(D) ...................................................................................... 57
Fed. R. Civ. P 23(f) ............................................................................................ 1, 4, 7
Fed. R. Evid. 702 ............................................................................................... 50, 51
Case: 15-55727, 09/21/2015, ID: 9691076, DktEntry: 13, Page 11 of 75
Ascertainability ■ Spivey ■ 17 -xi-
Other Authorities
7A Charles A. Wright et al., Federal Practice & Procedure § 1760 (3d ed. 2005 & Supp. 2015) .......................................................................... 13, 17
FDA, DRAFT Guidance for Industry: Voluntary Labeling Indicating Whether Foods Have or Have Not Been Developed Using Bioengineering (Jan. 2001), available at http://www.fda.gov/Food/GuidanceRegulation/GuidanceDocumentsRegulatoryInformation/LabelingNutrition/ucm059098.htm (lastvisited Sept. 18, 2015) ........................................................................................ 43
Federal Judicial Center, Reference Manual on Scientific Evidence(3d ed. 2011) ....................................................................................................... 35
1 Joseph M. McLaughlin, McLaughlin on Class Actions § 4.2(11th ed. 2014) .................................................................................................... 16
Random House Webster’s College Dictionary 245 (2000) ..................................... 13
1 William B. Rubenstein, Newberg on Class Actions § 3.3(5th ed. 2015) ...................................................................................................... 17
Case: 15-55727, 09/21/2015, ID: 9691076, DktEntry: 13, Page 12 of 75
18 ■ Class Actions ■ July 2017 -xii-
STATEMENT WITH RESPECT TO ORAL ARGUMENT
Pursuant to Federal Rule of Appellate Procedure 34(a), Defendant-Appellant
states that oral argument should be heard in this case. Argument would
significantly aid this Court’s decisional process because of the complex and
important issues this case presents.
Case: 15-55727, 09/21/2015, ID: 9691076, DktEntry: 13, Page 13 of 75
Ascertainability ■ Spivey ■ 191
INTRODUCTION
The Plaintiffs in this case are pursuing numerous state law claims on behalf
of classes of purchasers of certain varieties of Wesson brand cooking oils
manufactured by Defendant ConAgra Foods, Inc. (“ConAgra”). These cooking
oils contain the label statement “100% Natural,” and Plaintiffs contend this label is
misleading for one reason: the products are allegedly derived from genetically
modified organisms (“GMOs”). After initially denying certification, the district
court below certified damages classes of consumers in eleven different states. This
Court granted ConAgra’s petition for immediate review of this decision pursuant to
Federal Rule of Civil Procedure 23(f).
As ConAgra highlighted in that petition, the district court’s incorrect
application of the law governing class certification implicates several important
issues that continue to trouble district courts adjudicating the flood of food labeling
class actions in this Circuit.1 In many of these cases, the plaintiffs have been
unable to certify the proposed classes because their sweeping claims are inherently
incompatible with the strictures of Rule 23. This case, for example, involves all
purchasers of Wesson Oil since 2007 in eleven states, and attempts to prove that a
1 At least three other actions currently pending in this Court raise similar issues. See Jones v. ConAgra Foods, Inc. (No. 14-16327); Brazil v. Dole Packaged Foods, LLC (No. 14-17480); Bruton v. Gerber Products Co. (No. 15-15174).
Case: 15-55727, 09/21/2015, ID: 9691076, DktEntry: 13, Page 14 of 75
20 ■ Class Actions ■ July 20172
word susceptible to a host of different meanings (“natural”) is misleading based on
one highly specific interpretation of that word (“no GMOs”).
Plaintiffs in these cases struggle to show that the classes they propose are
sufficiently cohesive to be good candidates for litigation as a class action. This
case presents claims that are uniquely poor candidates for certification. The
“100% Natural” label is very different from other labels whose meaning is facially
clear (like “no sugar added”) or well-defined by regulatory authorities (like
“organic”). Not only do Plaintiffs challenge a label whose variable meaning is
largely in the eye of the beholder, but they seek to attribute a precise meaning to
the word “natural.” Their claims depend on class members interpreting that word
to mean the absence of GMOs, an issue itself subject to debate and variation over
the course of the eight or more years covered by these class actions.
Thus unsurprisingly, Plaintiffs have failed twice to muster sufficient proof
that they could show the natural label was material or that damages could be
computed on a classwide basis, so that common issues would predominate over
individual ones. The district court rightly rejected their efforts the first time, but
wrongly allowed certification the second time even though Plaintiffs had added
very little to their first set of arguments and evidence. As to the materiality of the
natural label, Plaintiffs relied largely on third party generalized surveys that were
not tied to consumers of Wesson Oil. They also failed to present a single viable
Case: 15-55727, 09/21/2015, ID: 9691076, DktEntry: 13, Page 15 of 75
Ascertainability ■ Spivey ■ 213
method for demonstrating that class members factor a belief that the Wesson Oil
natural label means the absence of GMOs into their purchasing decisions. As to
damages, Plaintiffs first attempted to ignore the requirement of the Supreme
Court’s recent decision in Comcast v. Behrend, 133 S. Ct. 1426 (2013), that a
damages model be tied to their specific theory of liability. When that was
unsuccessful, they hastily crafted an unreliable and untested “hybrid damages
model” that attempted to cure the deficiencies of two other models by simply
multiplying together their outputs.
Although Plaintiffs’ claims are uniquely problematic among food labeling
class actions, this case also raises the broadly applicable issue of ascertainability –
whether class members can be easily and reliably identified. Many courts have
required plaintiffs to propose both a clear definition and to show the definition can
be implemented in an administratively feasible manner, but the district court here
relied only on the class definition. Yet administrative feasibility is an important
procedural safeguard that protects the viability of the class action mechanism and
the due process rights of absent class members and defendants. While this Court
has already adopted this requirement in an unpublished decision, it should,
consistent with a number of its sister Circuits, hold that Plaintiffs must propose a
workable method of identification.
Case: 15-55727, 09/21/2015, ID: 9691076, DktEntry: 13, Page 16 of 75
22 ■ Class Actions ■ July 20174
On these and other fronts, Plaintiffs have been entirely unable to propose a
suitable way to proceed as a class. The district court should have denied Plaintiffs’
amended motion for class certification, just as it did their original motion. Instead,
it allowed them to proceed on flimsy and generalized evidence, a result that if
affirmed, would effectively lower the bar for certification under Rule 23 in this
Circuit.
JURISDICTIONAL STATEMENT
The district court had subject matter jurisdiction under 28 U.S.C. § 1332(d)
because this is a consolidated class action in which (1) the matter in controversy
exceeds the sum or value of $5,000,000, (2) there are over 100 class members, and
(3) at least one member of the putative classes is a citizen of a state different from
the defendant. ER05868-69, ¶ 7. Defendant ConAgra is incorporated in Delaware
and headquartered in Nebraska. ER05883, ¶ 32. Plaintiffs are residents of eleven
different states: California, Colorado, Florida, Illinois, Indiana, Nebraska, New
York, Ohio, Oregon, South Dakota, and Texas. ER00139-140.
This Court has appellate jurisdiction under 28 U.S.C. § 1292(e). The district
court entered an order certifying the class in part on February 23, 2015. ER00136-
275. Pursuant to Federal Rule of Civil Procedure 23(f) and Federal Rule of
Appellate Procedure 5, ConAgra timely filed a petition for permission to appeal on
Case: 15-55727, 09/21/2015, ID: 9691076, DktEntry: 13, Page 17 of 75
Ascertainability ■ Spivey ■ 235
March 9, 2015. No. 15-80040, ECF No. 1 (9th Cir.). On May 13, 2015, this Court
granted the petition. Id. ECF No. 3.
STATEMENT OF THE ISSUES
1. Whether a class is ascertainable when there is no evidence that members’ identification would be administratively feasible.
2. Whether class representatives are typical when they and class members are each subject to unique defenses.
3. Whether the district court erred by failing to require persuasive evidence regarding materiality that specifically linked the class and its claims, when materiality must be shown to predominate on a classwide basis.
4. Whether the district court erred by relying on an untested “hybrid” damages method that failed to satisfy Comcast Corp. v. Behrend.
5. Whether a class action is superior where it would require eleven separate trials and where Plaintiffs have not shown how they will identify class members.
STATEMENT OF THE CASE
This case is a consolidation of putative class actions that consumers in
eleven different states brought against ConAgra. ConAgra manufactures and
markets numerous familiar and trusted food product brands sold across the United
States, such as Healthy Choice, Orville Redenbacher’s, Swiss Miss, and the subject
of this case, Wesson brand cooking oils.
Plaintiffs allege that they purchased Wesson Vegetable Oil, Canola Oil,
Corn Oil, and Best Blend beginning on June 27, 2007, and that such products were
Case: 15-55727, 09/21/2015, ID: 9691076, DktEntry: 13, Page 18 of 75
24 ■ Class Actions ■ July 20176
labeled “100% Natural.” ER05866-883. Plaintiffs contend that this label is false
and misleading under the relevant state laws because the cooking oils contain
GMOs. Id. Plaintiff Robert Briseño first filed his class action complaint on behalf
of a nationwide class of consumers over four years ago. ER08343. Over the
following several months, the district court consolidated Briseño’s action with
seven other actions filed in federal district courts. ER00136 at n.2. Pursuant to
those consolidation orders, the district court directed Plaintiffs to file a
consolidated complaint, ER08348, which Plaintiffs did on January 12, 2012,
pursuing a nationwide class of consumers and fifteen state subclasses. Id.
After the district court granted in part ConAgra’s motion to dismiss,
ER08353, Plaintiffs filed their Second Consolidated Amended Complaint on
December 19, 2012, which remains the operative complaint in this action.
ER05862-5913. Plaintiffs sought certification of fifteen state classes and alleged
violations of each state’s consumer protection and other laws, including express
and implied warranty and unjust enrichment claims. ER05866. The district court
later allowed the dismissal of the named plaintiffs from four states, leaving the
eleven state classes now at issue. ER00137-138 & nn. 8, 12.
Plaintiffs filed their first class certification motion on May 5, 2014, seeking
certification of state classes pursuing both monetary damages pursuant to Rule
23(b)(3) and injunctive relief pursuant to Rule 23(b)(2). ER00276-316. On
Case: 15-55727, 09/21/2015, ID: 9691076, DktEntry: 13, Page 19 of 75
Ascertainability ■ Spivey ■ 257
August 1, 2014, the district court denied Plaintiffs’ motion without prejudice.
ER00040-105. It concluded that all Rule 23(a) requirements were met. ER00090.
However, it denied certification of the Rule 23(b)(2) injunctive relief classes
because it concluded that the named plaintiffs lacked standing as they had not
indicated that they would buy Wesson Oil in the future. ER00095. The district
court also denied certification of the Rule 23(b)(3) damages classes because
Plaintiffs did not show that common issues predominated over individual ones,
either as to reliance and causation or as to damages. ER00098, ER00101.
The district court allowed Plaintiffs a second opportunity to seek class
certification. ER00105. Plaintiffs filed an amended certification motion on
September 8, 2014. ER05228-5323. ConAgra opposed this motion and filed a
motion to strike various expert declarations that accompanied Plaintiffs’ amended
motion. ER05324-5407; ER08035-71. After holding a hearing, on February 23,
2015, the district court entered an order denying certification of the injunctive
relief classes but granting certification of the damages classes in eleven states
(although it denied certification of some individual state law claims). ER00136-
275. It also struck several of Plaintiffs’ expert opinions. Id.
ConAgra filed a petition for permission to appeal pursuant to Rule 23(f),
which this Court granted, and it now appeals the certification of the damages
Case: 15-55727, 09/21/2015, ID: 9691076, DktEntry: 13, Page 20 of 75
26 ■ Class Actions ■ July 20178
classes. Plaintiffs did not appeal the district court’s refusal to certify classes
seeking injunctive relief pursuant to Rule 23(b)(2).
SUMMARY OF THE ARGUMENT
First, the district court required no proof that class members could be
ascertained in an administratively feasible manner. Though the district court
acknowledged that Plaintiffs had the burden to prove the administrative feasibility
of identifying class members according to their proposed class definition, Plaintiffs
offered no explanation for how they would identify class members. The district
court nonetheless held that because Plaintiffs’ proposed class definition was
“objective,” the class was per se ascertainable.
Second, although both the class representatives and class members are
subject to unique defenses, the district court held that the representatives satisfied
Rule 23(a)’s typicality requirement. But this decision was in the face of evidence
that (1) a substantial portion of class members likely did not understand the natural
label to mean the absence of GMOs and (2) class representatives’ claims of being
misled were inherently suspect in light of their own continued willingness to
purchase products containing GMOs.
Third, Plaintiffs failed to show that the natural label was a material factor in
class members’ purchases of Wesson Oils because they believed the label indicated
the absence of GMOs. Although the district court allowed Plaintiffs to rely upon a
Case: 15-55727, 09/21/2015, ID: 9691076, DktEntry: 13, Page 21 of 75
Ascertainability ■ Spivey ■ 279
“reasonable consumer” standard to show that common issues predominated,
Plaintiffs neglected to present evidence or methods to satisfy this standard. Even
after the district court gave them a second opportunity to present adequate
evidence, Plaintiffs relied heavily on generalized third-party surveys, none of
which specifically addressed Wesson Oils, and not one of which linked the
elements of Plaintiffs’ theory of liability together. Instead, these surveys reflect
the general disagreement among consumers as to the meaning of natural claims.
Fourth, the district court allowed similarly threadbare proof in the context of
Plaintiffs’ damages model, which the Supreme Court held in Comcast must be
closely linked to Plaintiffs’ theory of liability. At the last minute, one of
Plaintiffs’ experts opined that she could cobble together two divergent models to
satisfy the requirement that Plaintiffs isolate damages to their belief in the absence
of GMOs. But it is undisputed that the expert did not show that she had ever used
this “hybrid damages model” before, or that any other expert in her field had. And
with good reason – this hybrid damages model does not comply with Comcast
because it is unable to isolate the price premium associated with the believed
absence of GMOs.
Fifth, Plaintiffs did not demonstrate that the class action is a superior method
for adjudicating their claims. Because of differences in state law, the district court
would likely have to conduct eleven separate trials. Further, Plaintiffs have not
Case: 15-55727, 09/21/2015, ID: 9691076, DktEntry: 13, Page 22 of 75
28 ■ Class Actions ■ July 201710
shown how or if they can identify class members, posing significant problems for
notice and damages allocation.
It is Plaintiffs’ burden to prove that all of the criteria for class certification
are met, and it is the district court’s duty to engage in a rigorous analysis of
Plaintiffs’ evidence and methods. Neither occurred here. Accordingly, the
decision to certify the damages classes was an abuse of discretion.
ARGUMENT
Because “Rule 23 does not set forth a mere pleading standard,” Plaintiffs
bear the burden of affirmatively demonstrating their compliance with Rule 23.
Wal-Mart Stores, Inc. v. Dukes, 131 S. Ct. 2541, 2551 (2011). In order to maintain
a class action under Rule 23(b)(3), Plaintiffs must first satisfy the prerequisites set
forth in Rule 23(a) of numerosity, commonality, typicality, and adequacy. As
discussed below, Plaintiffs must also show that the putative class is ascertainable.
Finally, Plaintiffs must demonstrate that Rule 23(b)(3) itself is satisfied, which
imposes two additional criteria. The court must find that “questions of law or fact
common to class members predominate over any questions affecting only
individual members, and that a class action is superior to other available methods
for fairly and efficiently adjudicating the controversy.” These two conditions are
known as predominance and superiority. In this appeal, ConAgra challenges the
Case: 15-55727, 09/21/2015, ID: 9691076, DktEntry: 13, Page 23 of 75
Ascertainability ■ Spivey ■ 2911
district court’s findings regarding ascertainability, typicality, predominance, and
superiority.
I. Standard of review
This Court reviews a district court’s grant of class certification for abuse of
discretion. Yokoyama v. Midland Nat’l Life Ins. Co., 594 F.3d 1087, 1090 (9th Cir.
2010). To the extent a ruling is premised on a finding of fact, that finding is
reviewed for clear error. Stearns v. Ticketmaster Corp., 655 F.3d 1013, 1018 (9th
Cir. 2011). However, no deference is accorded to district courts when reviewing
questions of law which are subject to de novo review. Id. “‘The district court
abuses its discretion if its certification order is premised on impermissible legal
criteria.’” Ellis v. Costco Wholesale Corp., 657 F.3d 970, 980 (9th Cir. 2011)
(citations omitted). Additionally, “[a] class certification order is an abuse of
discretion . . . if [the district court’s] application of the correct legal rule was based
on a ‘factual finding that was illogical, implausible, or without support in
inferences that may be drawn from facts in the record.’” Jimenez v. Allstate Ins.
Co., 765 F.3d 1161, 1164 (9th Cir. 2014) (citations omitted).
Case: 15-55727, 09/21/2015, ID: 9691076, DktEntry: 13, Page 24 of 75
30 ■ Class Actions ■ July 201712
The district court also abuses its discretion where it “‘relies upon an
improper factor, omits consideration of a factor entitled to substantial weight, or
mulls the correct mix of factors but makes a clear error of judgment in assaying
them.’” Parra v. Bashas’, Inc., 536 F.3d 975, 977-78 (9th Cir. 2008) (citation
omitted). In conducting its “rigorous analysis” of whether the criteria for
certification are present, the district court may need to “probe behind the
pleadings” and may also be required to engage in inquiries that overlap with the
factual and legal merits of the plaintiffs’ underlying claims. Dukes, 131 S. Ct. at
2551-52. Reversible error occurs where the district court does not follow these
mandates. Ellis, 657 F.3d at 980.
II. The district court erroneously held that the putative classes are ascertainable.
Plaintiffs failed to meet their burden to demonstrate that the proposed class
was ascertainable, because they did not show that identifying class members was
administratively feasible. Although the district court correctly recognized that the
ability to accurately identify the members of a putative class in a feasible and
verifiable manner is a fundamental requirement, it failed to apply this threshold test
of ascertainability, holding that Plaintiffs had satisfied their burden simply by
offering an objective class definition. Id. Because Plaintiffs failed to put forward
any method of identifying class members, let alone propose a feasible and
Case: 15-55727, 09/21/2015, ID: 9691076, DktEntry: 13, Page 25 of 75
Ascertainability ■ Spivey ■ 3113
verifiable method, the district court abused its discretion in concluding that the
putative classes are ascertainable.
A. Ascertainability is a threshold requirement for certification.
Although ascertainability is not explicitly mentioned in Rule 23, this Court
has referred to ascertainability as a “threshold” test for certification. Berger v.
Home Depot USA, Inc., 741 F.3d 1061, 1071 n.3 (9th Cir. 2014). See also 7A
Charles A. Wright et al., Federal Practice & Procedure § 1760 (3d ed. 2005 &
Supp. 2015) (noting that ascertainability is an “essential prerequisite” for
certification). The need to identify class members flows from the very concept of a
“class,” defined as “a number of persons or things regarded as belonging together
because of common attributes, qualities, or traits.” Random House Webster’s
College Dictionary 245 (2000).
Every Court of Appeals to consider the question has agreed that
ascertainability is a prerequisite for certification. See Byrd v. Aaron’s Inc., 784
F.3d 154, 161 n.4 (3d Cir. 2015) (collecting cases); Rikos v. Procter & Gamble
Co., --- F.3d ---, 2015 WL 4978712, at *22 (6th Cir. Aug. 20, 2015). Like the
court below, district courts throughout this circuit similarly require class action
plaintiffs to demonstrate ascertainability. See, e.g., Astiana v. Ben & Jerry’s
Homemade, Inc., No. C 10-4387-PJH, 2014 WL 60097, at *3 (N.D. Cal. Jan. 7,
2014).
Case: 15-55727, 09/21/2015, ID: 9691076, DktEntry: 13, Page 26 of 75
32 ■ Class Actions ■ July 201714
In light of these authorities, it cannot be seriously disputed that
ascertainability is a prerequisite to certification, and Plaintiffs did not argue to the
contrary. It is simply common sense that to bring a class action, Plaintiffs must be
able to identify who the members of the class actually are. See Byrd, 784 F.3d at
162 (describing ascertainability as “grounded in the nature of the class-action
device itself”). As the Third Circuit has highlighted, “[t]he ascertainability
requirement serves several important objectives,” such as protecting both absent
class members and defendants. Marcus v. BMW of N. Am., LLC, 687 F.3d 583,
593 (3d Cir. 2012). In sum, “[w]ithout an objective, reliable way to ascertain class
membership, the class quickly would become unmanageable, and the preclusive
effect of final judgment would be easy to evade.” Xavier v. Philip Morris USA,
Inc., 787 F. Supp. 2d 1075, 1088-90 (N.D. Cal. 2011).
B. A class is only ascertainable where its members can be determined in an objectively verifiable and administratively feasible manner.
Despite this virtually uniform agreement that an ascertainability requirement
inheres in Rule 23, some disagreement has developed as to the precise meaning of
that requirement. Specifically, some Courts of Appeals differ as to whether
plaintiffs can satisfy their burden merely by coming up with a class definition that
is objective and clear on its face, or whether they must additionally demonstrate
that this method would be verifiable and administratively feasible in practice. See,
Case: 15-55727, 09/21/2015, ID: 9691076, DktEntry: 13, Page 27 of 75
Ascertainability ■ Spivey ■ 3315
e.g., Mullins v. Direct Digital, 795 F.3d 654, 661-62 (7th Cir. 2015) (recognizing
disagreement and finding that a class was ascertainable because plaintiffs’ class
definition was neither vague nor based on subjective criteria).
This Court should follow the well-reasoned decisions of the Courts of
Appeals that require plaintiffs to show administrative feasibility. For example, the
Third Circuit recently stated that “[t]he ascertainability inquiry is two-fold,
requiring a plaintiff to show that (1) the class is ‘defined with reference to
objective criteria’; and (2) there is ‘a reliable and administratively feasible
mechanism for determining whether putative class members fall within the class
definition.’” Byrd, 784 F.3d at 163 (citations omitted). In EQT Production Co. v.
Adair, the Fourth Circuit considered a class that was objectively defined but
remanded to the district court to reconsider ascertainability because the method of
identification—looking to land records—would likely be “a complicated and
individualized process.” 764 F.3d 347, 359 (4th Cir. 2014).
In fact, this Court has also required administrative feasibility, albeit in an
unpublished opinion. In Martin v. Pacific Parking Systems Inc., this Court held
that there was no ascertainability where membership turned on whether a putative
member had used a personal credit card for a parking transaction. 583 F. App’x
803 (9th Cir. 2014). It agreed with the district court that “there was no reasonably
efficient way to determine which of the hundreds of thousands of individuals who
Case: 15-55727, 09/21/2015, ID: 9691076, DktEntry: 13, Page 28 of 75
34 ■ Class Actions ■ July 201716
used the parking lots” had used a personal card and that the plaintiff had “not
demonstrated that it would be administratively feasible” to make such a
determination. Id. at 804. See also Karhu v. Vital Pharm., Inc., --- F. App’x ---,
2015 WL 3560722 at *3 (11th Cir. June 9, 2015).
District courts both within and outside of this circuit have similarly held that
for a class to be ascertainable, the proposed class definition must be clear and
objective on its face and plaintiffs must have proposed a feasible and verifiable
method for applying that definition. Appeals from two such decisions are
currently pending in this Court. Bruton v. Gerber Prods. Co., No. 12-CV-02412-
LHK, 2014 WL 2860995, at *4 (N.D. Cal. June 23, 2014); Jones v. ConAgra
Foods, Inc., No. C 12-01633-CRB, 2014 WL 2702726, at *8-11 (N.D. Cal. June
13, 2014). 2
Treatises also endorse the administrative feasibility requirement. See 1
Joseph M. McLaughlin, McLaughlin on Class Actions § 4.2 (11th ed. 2014) (“[I]t
must be administratively feasible for the court to determine whether a given person
2 See also, e.g., In re WellPoint, Inc. Out-of-Network UCR Rates Litig., No. MDL 09-2074-PSG, 2014 WL 6888549, at *15 (C.D. Cal. Sept. 3, 2014); In re Clorox Consumer Litig., 301 F.R.D. 436, 441-42 (N.D. Cal. 2014); Algarin v. Maybelline, LLC, 300 F.R.D. 444, 455 (S.D. Cal. 2014); In re POM Wonderful LLC Mktg. & Sales Practices Litig., ML 10-02199-DDP, 2014 WL 1225184, at *6 (C.D. Cal. Mar. 25, 2014); Sethavanish v. ZonePerfect Nutrition Co., No. 12-2907-SC, 2014 WL 580696, at *4 (N.D. Cal. Feb. 13, 2014); Astiana, 2014 WL 60097, at *3, 2014).
Case: 15-55727, 09/21/2015, ID: 9691076, DktEntry: 13, Page 29 of 75
Ascertainability ■ Spivey ■ 3517
fits within the class definition without effectively conducting a mini-trial of each
person’s claim.”); 7A Federal Practice & Procedure § 1760 (“[Ascertainability]
will not be deemed satisfied unless the class description is sufficiently definite so
that it is administratively feasible for the court to determine whether a particular
individual is a member.”); 1 William B. Rubenstein, Newberg on Class Actions §
3.3 (5th ed. 2015).
All of this authority is supported by sound reasoning. Most fundamentally,
it is difficult to see the value of a definition that is clear on paper but
inadministrable in practice. Such an approach contravenes the duty of a district
court at certification to “probe behind the pleadings” to determine if what plaintiffs
plead is actually so. Dukes, 131 S. Ct. at 2551-52. “[A] party cannot merely
provide assurances to the district court that it will later meet Rule 23’s
requirements.” Byrd, 784 F.3d at 164. Additionally, the “‘principal purpose’” of
the class action mechanism is “‘efficiency and economy of litigation.’” Gen. Tel.
Co. of Sw. v. Falcon, 457 U.S. 147, 159 (1982) (citation omitted). Administrative
feasibility serves this core function, and a class definition that is guaranteed to
swamp potential efficiencies should be rejected. Reliable identification also serves
res judicata, for if the “class wins, any relief must be reasonably limited to those
who are entitled to it, and if a plaintiff class loses, the preclusive effect of final
Case: 15-55727, 09/21/2015, ID: 9691076, DktEntry: 13, Page 30 of 75
36 ■ Class Actions ■ July 201718
judgment must be enforced against all class members.” Xavier, 787 F. Supp. 2d at
1091 (emphasis omitted).
For these and numerous other reasons discussed by the relevant authorities,
this Court should affirm that in addition to proposing a suitable class definition,
Plaintiffs must propose an administratively feasible method of applying that
definition to identify the members of a class.
C. The district court abused its discretion by relying solely on the proffered class definition to hold that the classes’ members were ascertainable.
In this case, although the district court explicitly acknowledged that
administrative feasibility is required, it ultimately allowed Plaintiffs to rest solely
on their definition of the putative classes: “whether an individual purchased
Wesson Oils during the class period.” ER00182. It found that the class was
ascertainable despite Plaintiffs’ “inability to identify the specific members of a
putative class.” ER00183. Plaintiffs did not, in fact, propose any concrete method
for identifying class members.3 Instead, as the district court’s opinion reflects,
3 The closest they came was in their Reply in support of the original motion for class certification, where Plaintiffs briefly suggested a number of potential methods that would vary from consumer to consumer. ER6297-98. However, they did not propose any specific methods in connection with their amended class certification motion, and the district court neither relied on nor evaluated any such methods in either certification opinions. This court in Martin affirmed a denial of class certification on this exact basis – because the plaintiff “proposed no plan to
Case: 15-55727, 09/21/2015, ID: 9691076, DktEntry: 13, Page 31 of 75
Ascertainability ■ Spivey ■ 3719
Plaintiffs’ amended motion for class certification rested solely on their “objective”
class definition. ER05256-57; ER05618-20. Because the district court did not
conduct any inquiry as to the feasibility of Plaintiffs’ class definition, it abused its
discretion by premising certification “‘on impermissible legal criteria.’” Ellis, 657
F.3d at 980 (citations omitted).
Although the district court did not rely upon any method for class member
identification, many of the authorities it cited discussed self-identification –
whether it is sufficient for putative class members to simply state that they
purchased the relevant product during the relevant timeframe, without
corroboration. ER00183-184. As these citations reflect, the underlying
disagreement regarding administrative feasibility often arises in cases involving
low-cost consumer goods. In such cases, class members would not have kept
receipts or other records of their purchases and defendants may not have records of
retail purchasers. The plaintiffs may propose that putative members can simply
state that they are members of a class, and the question of administrative feasibility
becomes conflated with whether a court is willing to accept putative members’
uncorroborated statements of what they remember as reliable proof of class
the district court for manageably determining which individuals are members.” 583 F. App’x at 804.
Case: 15-55727, 09/21/2015, ID: 9691076, DktEntry: 13, Page 32 of 75
38 ■ Class Actions ■ July 201720
membership. Compare, e.g., Carrera v. Bayer Corp., 727 F.3d 300, 308-12 (3d
Cir. 2013), with Mullins, 795 F.3d at 662.
This Court need not engage in this debate here, because the district court did
not require any method of identification at all. Because the question of whether
Plaintiffs must show that member identification is feasible and verifiable is distinct
from which methods are in fact feasible and verifiable, this Court should reject the
misstep of the district court. It apparently failed to require administrative
feasibility because doing so would “effectively prohibit class actions involving low
priced consumer goods.” ER00184. Excusing Plaintiffs from providing an
acceptable identification method because a subsector of a portion of class actions
(consumer class actions involving low-priced goods) may be impacted throws the
baby out with the bathwater because many class actions do not involve such goods.
E.g., EQT Production, 764 F.3d at 358-60 (gas rights holders); see also Bakalar v.
Vavra, 237 F.R.D. 59, 64-66 (S.D.N.Y. 2006) (rights to valuable artwork).
Because administrative feasibility serves important purposes, it should be
addressed in every case.
Even within the subsector of low-cost consumer class actions, each case is
different and not all will be barred by the application of a proper ascertainability
test. Sethavanish v. ZonePerfect Nutrition Co., No. 12-2907-C, 2014 WL 580696,
at *5 (N.D. Cal. Feb. 13, 2014) (noting that cases like Carrera “do not bar
Case: 15-55727, 09/21/2015, ID: 9691076, DktEntry: 13, Page 33 of 75
Ascertainability ■ Spivey ■ 3921
certification in consumer class actions altogether”). “Class actions, and consumer
class actions in particular, each fall on a continuum of ascertainability dependent
upon the facts of the particular case or product. . . . [R]elevant considerations
include the price of the product, the range of potential or intended uses of a product
and the availability of purchase records.” In re POM Wonderful LLC Mktg. &
Sales Practices Litig., ML 10-02199-DDP, 2014 WL 1225184, at *6 (C.D. Cal.
Mar. 25, 2014). In some instances, reliable records may exist, such as receipts,
consumers’ use of loyalty cards, Carrera, 727 F.3d at 308, or “retailer or banking
records [that] make it economically and administratively feasible to determine who
is in (and who is out) of a putative class.” Sethavanish, 2014 WL 580696, at *5.4
Plaintiffs here apparently made no attempt to obtain data from retailers. See
generally ER00276-353; ER06260-6312; ER05228-5323; ER05579-5629.
Even if this Court does address, in the first instance, whether consumer self-
identification without any form of corroborating information is administratively
feasible and objectively verifiable in this case, it should hold that it is not, as it has
previously recognized in an unpublished decision. Martin, 583 F. App’x at 804
(noting that the plaintiff had proposed no method of identification “beyond
4 Even when the only proof available is in fact consumers’ say-so, courts may draw nuanced distinctions based, for example, on how inherently reliable those memories are likely to be. Bruton, 2014 WL 2860995, at *5-10; Jones, 2014 WL 2702726, at *10.
Case: 15-55727, 09/21/2015, ID: 9691076, DktEntry: 13, Page 34 of 75
40 ■ Class Actions ■ July 201722
suggesting that individuals self-identify themselves.”). Plaintiffs offered no reason
why putative members would be particularly likely to reliably remember whether
they bought a specific variety of Wesson Oil as opposed to one of many competing
brands of cooking oil as long as eight years ago.
Self-identifications also pose a fundamental constitutional problem: there is
no apparent way to protect ConAgra’s due process right to verify and challenge
such self-identifications in any administratively feasible manner. See, e.g.,
Marcus, 687 F.3d at 593. As the Supreme Court has recognized, “a class cannot be
certified on the premise that [the defendant] will not be entitled to litigate its . . .
defenses to individual claims.” Dukes, 131 S. Ct. at 2561. A defendant’s due
process rights cannot be sacrificed simply to provide class action plaintiffs a
damages remedy, particularly where injunctive or regulatory remedies may be
available to prevent mislabelings, and where the class action mechanism is already
an “‘exception to the usual rule’” that litigation is to be conducted only by
individual named parties. Dukes, 131 S. Ct. at 2550 (citation omitted). Rule
23(b)(3) is an “‘adventuresome innovation’” that carries special procedural
protections, id. at 2558 (citation omitted), and such protections should not be set
aside in order to provide plaintiffs one specific variety of available remedies.
Put simply, Plaintiffs’ prior emphasis on arguments that applying the
ascertainability requirement here would be the end of consumer class actions is
Case: 15-55727, 09/21/2015, ID: 9691076, DktEntry: 13, Page 35 of 75
Ascertainability ■ Spivey ■ 4123
exaggerated hyperbole. Plaintiffs make this argument to be relieved of their
obligation to demonstrate administrative feasibility – an obligation that the district
court recognized but failed to apply. There are paths forward for plaintiffs in low-
cost class actions to meet this requirement, as district courts have found. See, e.g.,
Sethavanish, 2014 WL 580696, at *5; Red v. Kraft Foods, Inc., No. CV-10-1028-
GW, 2012 WL 8019257, at *5 (C.D. Cal. Apr. 12, 2012). Ultimately, there must
be a balancing of interests: while application of this requirement should not be so
strict as to eliminate all low-cost class actions, courts should not make the mistake
of completely ignoring the requirement or applying it half-heartedly at the expense
of defendants’ due process rights and Rule 23’s procedural protections.
The bottom line is this: the district court recognized correctly that
administrative feasibility is a requirement as part of the ascertainability inquiry.
Plaintiffs did not even attempt to meet this requirement. Their failure to do so
should have ended this inquiry, but the district court chose not to apply this
requirement. This is error, and this alone mandates reversal of the district court’s
grant of class certification.
III. The named representatives do not have claims and defenses that are typical of those of the putative class members.
Under Rule 23(a)(3), Plaintiffs must show that “the claims or defenses of the
representative parties are typical of the claims or defenses of the class.” “The
purpose of the typicality requirement is to assure that the interest of the named
Case: 15-55727, 09/21/2015, ID: 9691076, DktEntry: 13, Page 36 of 75
42 ■ Class Actions ■ July 201724
representative aligns with the interests of the class.” Hanon v. Dataproducts Corp.,
976 F.2d 497, 508 (9th Cir.1992). “The test of typicality ‘is whether other
members have the same or similar injury, whether the action is based on conduct
which is not unique to the named plaintiffs, and whether other class members have
been injured by the same course of conduct.’” Id. (citation omitted).
The claims and defenses of the class representatives and class members
diverge from each other in two critical respects. First, the representatives have
alleged that they were deceived by the “100% Natural” label because they thought
that the label meant that the product was GMO-free. ER07873-75. But ConAgra
presented substantial evidence that the same does not hold true for the majority of
class members. Specifically, ConAgra’s expert Dr. Dominique Hanssens
conducted a survey of California cooking oil consumers who had purchased or
would consider purchasing Wesson Oil that showed:
The term “100% Natural” has no statistically measurable impact on consumers purchasing Wesson Oil.
Forty percent of consumers either did not know or were not sure what the “100% Natural” label meant.
Only 2.6% of consumers defined the “100% Natural” label as meaning GMO-free.
Sixty percent of consumers either did not believe that, or did not know whether, Wesson Oil was GMO-free.
Case: 15-55727, 09/21/2015, ID: 9691076, DktEntry: 13, Page 37 of 75
Ascertainability ■ Spivey ■ 4325
ER05355-56 (citing ER00797-800, ¶¶ 52-53, 58, 61; ER00887). Even Plaintiffs’
evidence tends to show the same disparity. Although their surveys were highly
generalized and, unlike Dr. Hanssens’, not targeted to actual or potential Wesson
Oil consumers, they found that around sixty percent of consumers prefer natural
labels and that between under one-half and slightly more than two-thirds of
consumers generally associate natural labels with an absence of GMOs. See infra
Part IV.A.2.b. This evidence suggests that at least one-third of putative class
members will not share the class representatives’ experiences. Id.
The district court effectively declined to consider this evidence because it
held that all that matters is whether class representatives, not class members, are
subject to unique defenses. ER00189-190. It did not provide a citation for this
conclusion, however, and the text of Rule 23(a)(3) certainly does not contemplate
this lopsided requirement, but asks simply whether the representatives and class
members mirror each other. See Koos v. First Nat’l Bank, 496 F.2d 1162, 1164-65
(7th Cir. 1974) (“Where it is predictable that a major focus of the litigation will be
on an arguable defense unique to the named plaintiff or a small subclass, then the
named plaintiff is not a proper class representative.” (emphasis added)). Where
many class members may be subject to a rebuttal of an inference of materiality
based on the evidence discussed above, e.g., Wiener v. Dannon Co., 255 F.R.D.
658, 669 (C.D. Cal. 2009), it is likely that the representatives will not share
Case: 15-55727, 09/21/2015, ID: 9691076, DktEntry: 13, Page 38 of 75
44 ■ Class Actions ■ July 201726
defenses unique to many class members and will accordingly not be “focused on
common concerns of the class.” Beck v. Maximus, Inc., 457 F.3d 291, 301 (3d Cir.
2006).
Second, the named representatives are in fact subject to unique defenses.
Attempting to prove standing for the purpose of certifying the injunctive classes,
ER00193-194, they inconsistently alleged (1) that the natural label made a
difference to them because they do not like GMOs and (2) that they would
consider buying Wesson Oils in the future even if they still contained GMOs. See,
e.g., ER07874-75; ER00189, n.174; ER00179-180. Further, after commencing this
litigation the named representatives continued to buy other food products labeled
natural but containing GMOs. ER05359. As the district court itself noted, “these
statements could support an inference that plaintiffs’ belief [that] Wesson Oils did
not contain GMOs was not material to their purchasing decision.” ER00253.
Even under the district court’s definition of typicality, this poses a problem for the
putative classes. See, e.g., Stearns, 655 F.3d at 1019 (representative atypical where
he “was not really deceived into joining the Entertainment Rewards program”).
Because both a substantial portion of class members and the class
representatives will each be subject to defenses that the other does not share, the
representatives do not satisfy Rule 23(a)(3), and the district court erred in finding
to the contrary.
Case: 15-55727, 09/21/2015, ID: 9691076, DktEntry: 13, Page 39 of 75
Ascertainability ■ Spivey ■ 4527
IV. The district court erroneously held that common questions predominate over individual ones.
The district court further erred in concluding that common issues
predominated as to materiality and damages. As to both, Plaintiffs failed to satisfy
their demanding burden to demonstrate that these issues are susceptible to class
adjudication. Because they did not actually develop concrete methodologies tied
to their underlying theory of injury, Plaintiffs’ approach was to (1) patch together
questionable pieces of evidence as to materiality that failed to show a cohesive
view of the natural label, and (2) proffer untested expert opinions as to damages.
By not subjecting this deficient approach to sufficient scrutiny, the district court
neglected to apply a rigorous analysis.
Rule 23(b)(3) provides that in order for a class to be certified, the court must
find, among other criteria, that “the questions of law or fact common to class
members predominate over any questions affecting only individual members.”
Predominance is a “demanding” standard, Amchem Prods., Inc. v. Windsor, 521
U.S. 591, 624 (1997), and courts have a “duty to take a ‘close look’ at whether
common issues predominate over individual ones.” Comcast, 133 S. Ct. at 1432
(citation omitted). As with all aspects of a certification decision, courts must
“‘probe behind the pleadings’” and engage in a “‘rigorous analysis’” before
certifying a class. Id. (citations omitted).
Case: 15-55727, 09/21/2015, ID: 9691076, DktEntry: 13, Page 40 of 75
46 ■ Class Actions ■ July 201728
The “predominance inquiry tests whether proposed classes are sufficiently
cohesive to warrant adjudication by representation.” Amchem, 521 U.S. at 623.
When class members’ claims “require a fact-intensive, individual analysis,” then
certification is likely inappropriate. Vinole v. Countrywide Home Loans, Inc., 571
F.3d 935, 947 (9th Cir. 2009). The key concern in a predominance inquiry is “the
balance between individual and common issues.” Mevorah v. Wells Fargo Home
Mortgage (In re Wells Fargo Home Mortg.), 571 F.3d 953, 959 (9th Cir. 2009).
A. Individual issues predominate with respect to materiality because there is no reliable classwide method of proof.
Plaintiffs’ claims are inherently individualized because the putative class
members contend that they each were misled by the natural label in their individual
purchasing decisions. While the district court eased Plaintiffs’ burden in this
regard by concluding they were allowed to present classwide proof of materiality,
it failed to take the necessary next step of ensuring that such proof was actually
available. Plaintiffs submitted, and the district court relied upon, a hodgepodge of
various third-party generalized surveys and marketing data, most of which had
nothing to do with Wesson Oil consumers. Most importantly, there was no single
survey or method that addressed whether consumer beliefs regarding the absence
of GMOs were material to their purchases of Wesson cooking oils. Revealingly,
Plaintiffs did attempt to present such a survey, but the district court deemed it to be
inadmissible and did not consider it. That should have ended the matter.
Case: 15-55727, 09/21/2015, ID: 9691076, DktEntry: 13, Page 41 of 75
Ascertainability ■ Spivey ■ 4729
1. Plaintiffs must show they can prove materiality to avoid individualized issues of reliance and causation.
Many of Plaintiffs’ causes of action require them to show reliance and
causation by proving that the alleged misrepresentation caused class members to
purchase the product. See, e.g., In re Tobacco II Cases, 207 P.3d 20, 39 (Cal.
2009). As the district court recognized, such issues are individualized and would
normally predominate over common ones. ER00096-97. Plaintiffs sought to cure
this problem by pointing to various state laws’ allowance for classwide proof of
materiality as a way to avoid proving individual reliance and causation at
certification. Accordingly, the district court considered whether Plaintiffs had
“demonstrate[ed] that the materiality of [the natural label] can be established by
common survey proof” on a classwide basis. ER00250.5 As an example of the
law of one of the state classes, California courts have held that “‘plaintiffs satisfy
their burden of showing causation as to each by showing materiality as to all.’”
Mass. Mut. Life Ins. Co. v. Superior Ct., 97 Cal. App. 4th 1282, 1292 (2002)
(citation omitted). Materiality, in turn, “is judged by a ‘reasonable man’ standard,”
5 It held that claims of nine of the eleven state classes allowed classwide proof of materiality. ER00201-250. Although the district court also certified Indiana and Nebraska classes, the claims of those classes do not involve materiality and therefore those two classes are not at issue in this claim of error. Because the district court subsequently treated materiality as a uniform issue for all nine classes without addressing any state law differences, ConAgra will do the same here using California law as a representative example, without conceding that there are no such differences.
Case: 15-55727, 09/21/2015, ID: 9691076, DktEntry: 13, Page 42 of 75
48 ■ Class Actions ■ July 201730
such that “a misrepresentation is deemed material if a reasonable man would attach
importance to its existence or nonexistence in determining his choice of action in
the transaction in question.” Steroid Hormone Prod. Cases, 104 Cal. Rptr. 3d 329,
338 (Cal. Ct. App. 2010) (quotation marks and citations omitted).
In Stearns, this Court required plaintiffs seeking to establish predominance
through classwide methods to show that the alleged misrepresentation is “material
to all class members,” for if it is not, “the issue of reliance would ‘vary from
consumer to consumer’ and the class should not be certified.” 655 F.3d at 1022-23
(emphasis added) (citation omitted). Stearns emphasized that while the
“reasonable consumer” test does lessen this burden of proof, it does not mean that
“predominance would be shown in every . . . case.” Id. at 1020. Therefore,
Plaintiffs must first present evidence showing that, with respect to the label claim
at issue, materiality itself is a common issue among consumers. E.g., Webb v.
Carter’s, Inc., 272 F.R.D. 489, 502-03 (C.D. Cal. 2011). And even if Plaintiffs
establish a presumption of reliance and causation, ConAgra is entitled to offer
evidence to rebut it. E.g., Caro v. Procter & Gamble Co., 18 Cal. App. 4th 644,
667-68 & n.20 (1993); Garcia v. Medved Chevrolet, Inc., 263 P.3d 92, 99 (Colo.
2011).
Case: 15-55727, 09/21/2015, ID: 9691076, DktEntry: 13, Page 43 of 75
Ascertainability ■ Spivey ■ 4931
2. The district court erroneously relied on surveys that do not support a finding of classwide materiality.
The district court in its original decision denying class certification held that
“the evidence regarding the materiality of ‘100% Natural’ is in conflict” and “weak
at best.” ER00097. It found that “Plaintiffs adduce[d] no survey evidence
concerning the actual reaction of consumers to the ‘100% Natural’ label on
Wesson Oils specifically or the presence of such a label on cooking oils generally.”
Id. It also emphasized that Plaintiffs had not presented evidence directly related
“to the specific issue raised in this case—i.e., whether consumers believe the label
means the product contains no genetically modified organisms.” Id.
However, in its second order granting class certification, the district court
inexplicably reversed course and held that Plaintiffs were entitled to the classwide
inference of materiality, despite Plaintiffs’ failure to correct the deficiencies
initially identified. The new evidence Plaintiffs proffered in their amended
certification motion suffered from the same critical flaws as that in support of their
first motion. Plaintiffs still have not shown how they will prove that Wesson Oil
consumers believe that the 100% Natural label means that the product does not
contain GMOs and attach weight to this belief in their purchasing decisions. The
district court was right the first time, and it should not have accepted Plaintiffs’
still “weak” evidence as establishing materiality. The district court’s ruling was
therefore in error for at least two reasons.
Case: 15-55727, 09/21/2015, ID: 9691076, DktEntry: 13, Page 44 of 75
50 ■ Class Actions ■ July 201732
a. Plaintiffs’ proffered method of proving materiality was not tailored to their theory that consumers factor a belief that the Wesson Oil natural label means the absence of GMOs into their purchasing decision.
Plaintiffs must proffer a method of proving materiality that targets their
theory of liability – that consumers believed the natural label meant the Wesson
Oils did not contain GMOs and factored this into their purchasing decision.
Amchem, 521 U.S. at 623 (noting that the predominance inquiry must “train[] on
the legal or factual questions that qualify each class member’s case as a genuine
controversy”). It is axiomatic that the common issues that must predominate are
the issues that actually matter in the case at hand. See EQT Production, 764 F.3d
at 366 (reversing on predominance grounds where the district court failed to
consider whether the plaintiffs’ evidence was “relevant to assessing the
defendants’ ultimate liability”); Forrand v. Federal Express Corp., No. CV 08-
1360-DSF, 2013 WL 1793951, at *3 (C.D. Cal. Apr. 25, 2013) (“[For the]
predominance requirement to be satisfied, a plaintiff must bring forth a . . . method
that can be applied classwide and that ties plaintiff’s legal theory to . . . the
defendant’s allegedly illegal conduct.”); Kosta v. Del Monte Foods, Inc., No. 12-
CV-1722 YGR, 2015 WL 4593175, at *12 (N.D. Cal. July 30, 2015) (“A fill-in-
the-blanks declaration from an expert, without any real consideration of the
Case: 15-55727, 09/21/2015, ID: 9691076, DktEntry: 13, Page 45 of 75
Ascertainability ■ Spivey ■ 5133
specific product attributes at issue here, is not sufficient to establish that the
materiality of the label statements here is a common question.” (emphasis added)).
Plaintiffs’ evidence failed to comprehensively link (1) Wesson Oils, (2) a
natural claim, (3) the absence of GMOs and (4) consumers’ purchasing decisions.
The district court’s decision first cited highly generalized national surveys that
some consumers may prefer “natural” products when shopping for food generally
and that they might pay more for “some” unspecified types of natural products.
ER00250 (citing ER03680-3703; ER03726; ER03755). Then, it referenced data
showing that consumers find the natural label on Wesson products generally
favorable. ER00250-251. As the district court acknowledged, however, none of
this information “directly links consumers’ understanding of the ‘100% Natural’
label to the specific issue raised in this case – i.e., whether customers believe the
label means the product contains no . . . GMO ingredients.” ER00251.
Next, the district court turned to four generalized, third-party (and mostly
online) surveys showing “that consumers believe ‘natural’ means, among other
things, no GMOs.” Id. (citing ER03680-3703; ER03819-3928; ER03929-3962;
ER04931-4967). These surveys were the primary new evidence that Plaintiffs
presented in their amended motion, but they still suffered from the same flaw the
district court identified initially: they did not involve Wesson Oil products or
marketing, nor did they purport to describe consumers of Wesson Oil. See Kosta,
Case: 15-55727, 09/21/2015, ID: 9691076, DktEntry: 13, Page 46 of 75
52 ■ Class Actions ■ July 201734
2015 WL 4593175, at *11 (rejecting “broad statements about why labels generally
matter to consumers of food products”). Instead, these surveys were conducted of
consumers of food generally, necessarily referring to an immense variety of
products with different labels, different ingredients, and sets of consumers. E.g.,
ER03688 (referring to packaged and processed foods); ER03862 (“[p]roduction
and [f]arming”); ER03947 (“food or beverage”). In this way, the surveys simply
failed to demonstrate that reasonable consumers shared a uniform understanding of
whether Wesson Oil contained GMOs.6
This failure to focus on the products at issue and accordingly, on the class
members who purchased them, is a fatal flaw in the surveys Plaintiffs presented.
The cases are legion that hold that “‘to be probative and meaningful surveys must
rely upon responses by potential consumers of the products in question.’”
Universal City Studios, Inc. v. Nintendo Co., 746 F.2d 112, 118 (2d Cir. 1984)
(punctuation and citation omitted) (emphasis added); see also Harolds Stores, Inc.
v. Dillard Dep’t Stores, Inc., 82 F.3d 1533, 1544-46 (10th Cir. 1996); Exxon Corp.
v. Texas Motor Exch. of Houston, Inc., 628 F.2d 500, 506-07 (5th Cir. 1980);
6 The district court also referenced ConAgra’s receipt of complaints from individuals after they learned that the products contained GMOs. ER00251. This last piece of largely subjective and piecemeal evidence has little relevance for the reasonable consumer inquiry, little to say about consumers’ state of mind at the time of a purchase, and can hardly be a sufficient method of proof for millions of putative class members.
Case: 15-55727, 09/21/2015, ID: 9691076, DktEntry: 13, Page 47 of 75
Ascertainability ■ Spivey ■ 5335
Fancaster, Inc. v. Comcast Corp., 832 F. Supp. 2d 380, 403 (D.N.J. 2011); THOIP
v. Walt Disney Co., 690 F. Supp. 2d 218, 236-37 (S.D.N.Y. 2010).
For instance, the Tenth Circuit upheld a district court’s finding that a survey
was “devoid of any probative value and therefore irrelevant” in part because it was
overinclusive. The respondents to the survey were potential purchasers of all sinus
remedies rather than of sinus-irrigation products in particular (which were at issue
in the suit and which formed a small percentage of sinus remedy sales). Water Pik,
Inc. v. Med-Systems, Inc., 726 F.3d 1136, 1145 (10th Cir. 2013). See also Fractus,
S.A. v. Samsung, Civ. No. 6:09-cv-203-LED-JDL, 2011 WL 7563820, at *1 (E.D.
Tex. Apr. 29, 2011) (excluding survey evidence that was not sufficiently tied to the
patent at issue but addressed the market as a whole). Here, Plaintiffs’ surveys
required extrapolation from consumers of all foods or all packaged foods to
Wesson Oil consumers, supra, p.34; yet courts typically do not permit the results
of a survey of one group of people to be extrapolated as evidence for another group
of people. See, e.g., Wells Fargo & Co. v. WhenU.com, Inc., 293 F. Supp. 2d 734,
752-53 (E.D. Mich. 2003); Federal Judicial Center, Reference Manual on Scientific
Evidence 377-78 (3d ed. 2011) (“A survey that provides information about a
wholly irrelevant population is itself irrelevant. Courts are likely to exclude the
survey or accord it little weight.” (footnotes omitted)).
Case: 15-55727, 09/21/2015, ID: 9691076, DktEntry: 13, Page 48 of 75
54 ■ Class Actions ■ July 201736
The district court was apparently aware of this body of law given its initial
critique of Plaintiffs’ original survey evidence—indeed, it stated in the first hearing
on class certification that “offering an opinion that across the surveys this is what
consumers think is just not appropriate.” ER00012. Yet the district court
abandoned this position to grant class certification even though it still lacked a
single method of proof tailored to Plaintiffs’ theory of materiality. Tellingly,
Plaintiffs attempted to present precisely such a study, but the district court
disregarded the study as inadmissible. ER00251, n.245 (referencing the Kozup
survey). That exclusion should have been the end of the matter, but the court
inexplicably combined inadequate surveys to make an unwarranted finding
regarding consumers’ understanding of Wesson Oil labels.
Not only did Plaintiffs’ evidence fail to establish Wesson Oil consumers’
understanding of the natural label, but it failed to link any such understanding with
actual purchasing decisions. A misrepresentation is only material “‘if a reasonable
man would attach importance to its existence or nonexistence in determining his
choice of action in the transaction in question.’” Stearns, 655 F.3d at 1022
(citation omitted) (emphasis added). See also Pizza Hut, Inc. v. Papa John’s Int’l,
227 F.3d 489, 502 (5th Cir. 2000) (requiring evidence that the misleading
statements “had a tendency to influence the purchasing decisions of . . . the
consumers to which they were directed”). However, the third-party surveys the
Case: 15-55727, 09/21/2015, ID: 9691076, DktEntry: 13, Page 49 of 75
Ascertainability ■ Spivey ■ 5537
district court relied on to link consumers’ understanding of the word natural to the
absence of GMOs do not purport to tie consumers’ understandings to purchase
decisions but merely show “that consumers believe ‘natural’ means, among other
things, no GMOs.” ER00251.
For example, the “substantial majority” in the HealthFocus study the district
court cited was asked, “To what extent does [has not been genetically modified]
mak[e] you feel that a food or beverage is natural?”. ER03947. Exhibiting the
disconnect between abstract consumer preferences and their purchasing decisions,
that same study also found that less than 5% of consumers found “natural” to be
“most important when shopping.” ER00251, n.244; ER03952. In another survey
the district court cited, less than 40% of consumers found natural to be important to
their purchase decision of shelf-stable products. ER00251, n.242; ER03902. In
sum, the district court did not rely on evidence as to whether consumers might
change their purchase decisions if they were aware that products labeled as
‘natural’ contained GMOs – the critical question in this case.
It is not a difficult burden for Plaintiffs to isolate their theory of materiality
by creating admissible surveys asking Wesson Oil consumers whether they believe
Wesson Oils are GMO-free due to the natural label, and if so, whether that was a
material factor in their purchasing decisions. Indeed, Plaintiffs tried to do so (but
the district court rejected the Kozup survey) and ConAgra actually did so (but the
Case: 15-55727, 09/21/2015, ID: 9691076, DktEntry: 13, Page 50 of 75
56 ■ Class Actions ■ July 201738
district court did not address the Hanssens survey). It is an insufficient substitute
to cobble together surveys or data targeting divergent sets of consumers, divergent
products, and divergent time periods, some of which show (at most) that Wesson
consumers like products labeled “natural,” and others of which show that some
consumers of food products generally associate “natural” with “no GMOs.” Thus,
the district court erred in permitting a classwide inference of materiality despite not
having a single method tailored to the classes’ theory of liability.
b. The district court erred in finding it appropriate to apply a classwide presumption of materiality when at least one-third of a proposed class does not find the label claim material.
To find materiality, the district court (1) noted that Plaintiffs’ surveys show
that between 59% and 65% of consumers like natural products, ER00250, and (2)
referenced surveys which show that somewhere between 46% and slightly over
two-thirds of consumers associate a natural claim with the absence of
GMOs. ER00250, nn.240-244; ER03689 (64%); ER03862 (61%); ER03947;
ER04952-53 (46%).7 Even assuming these two sets of surveys can reliably be
combined to create one “majority,” this means that approximately a third of
7 The second Hartman Group survey the district court referenced showed that 61% of 2010 consumers associate natural with no GMOs but that this fell to 46% in 2012. ER04953. Without explanation, the district court cherry-picked and mentioned only the less recent 61% figure. Compare id. with ER00250, nn.240 & 243.
Case: 15-55727, 09/21/2015, ID: 9691076, DktEntry: 13, Page 51 of 75
Ascertainability ■ Spivey ■ 5739
consumers (at least) do not associate a natural claim with the absence of GMOs,
and that this number may actually be more than half of consumers. Finding that
materiality predominated among class members in the face of this evidence was
error.
This Court has recognized that “[i]f the misrepresentation or omission is not
material as to all class members, the issue of reliance ‘would vary from consumer
to consumer’ and the class should not be certified.” Stearns, 655 F.3d at 1022-23
(emphasis added) (citation omitted). To meet the “reasonable consumer” standard
under California law, a plaintiff is required to show that “‘a significant portion’ of
targeted consumers, acting reasonably in the circumstances, could be misled.’” In
re Celexa & Lexapro Mktg. & Sales Practices Litig., 291 F.R.D. 13, 20 (D. Mass.
2013) (citation omitted). Indeed, showing that more than “a handful of class
members” or that “numerous class members” would not find the alleged
misrepresentation material can mean that the element of materiality is not subject
to common proof on a classwide basis.8 In re Vioxx Class Cases, 180 Cal. App.
8 The district court cited a single district court decision for the proposition that even 24% would be sufficient to show materiality. ER00252 (citing Oshana v. Coca-Cola Co., No. 04 C 3596, 2005 WL 1661999 (N.D. Ill. July 13, 2005)). But Oshana noted that such a survey would be insufficient to rebut evidence of a single plaintiff’s own conduct to the contrary—it had nothing to say about proving materiality for a class of consumers. Id. at *8-9.
Case: 15-55727, 09/21/2015, ID: 9691076, DktEntry: 13, Page 52 of 75
58 ■ Class Actions ■ July 201740
4th 116, 129 (2009); In re Countrywide Fin. Corp. Mortg. Mktg. & Sales Practices
Litig., Nos. 08md1988-DMS, 2011 WL 6325877, at *10 (S.D. Cal. Dec. 16, 2011).
A bare majority would not qualify as “classwide proof” under many, if not
all, of the state laws at issue. See, e.g., Fairbanks v. Farmers New World Life Ins.
Co., 197 Cal. App. 4th 544, 565-66 (2011) (no common proof where contrary
evidence as to “roughly half” of the putative class); Strawn v. Farmers Ins. Co. of
Oregon, 258 P.3d 1199, 1213 (Or. 2011) (to obtain a classwide inference, “the
misrepresentation must be of a nature that the class members logically would have
had a common understanding of the misrepresentation, and naturally would have
relied on it to the same degree and in the same way”); Henry Schein, Inc. v.
Stromboe, 102 S.W.3d 675, 694 (Tex. 2003) (no classwide proof where “there is
no evidence that purchasers actually did rely on [the challenged] statements so
uniformly that common issues of reliance predominate”).
The district court acknowledged that Plaintiffs had to show that “ConAgra’s
conduct is deceptive and misleads reasonable consumers and/or class members.”
ER00272 (emphasis added). Yet Plaintiffs’ own evidence shows that the
challenged statement would not be material for far more than a “handful” of
consumers. Vioxx, 180 Cal. App. 4th at 129. It can be said unequivocally that,
regardless of which third-party survey offered by Plaintiffs one considers,
“numerous class members” do not believe that natural means an absence of
Case: 15-55727, 09/21/2015, ID: 9691076, DktEntry: 13, Page 53 of 75
Ascertainability ■ Spivey ■ 5941
GMOs. See Countrywide Fin. Corp. Mortg. Mktg. & Sales, 2011 WL 6325877 at
*10. Rule 23’s predominance requirement cannot be satisfied when a third to a
half of the proposed class would not find a label claim to be material. Thus, the
court erred in relying on these surveys to establish materiality for this reason as
well.9
3. The district court did not consider ConAgra’s own superior rebuttal evidence which shows that the term ‘natural’ holds so many different meanings for different consumers that materiality cannot be shown.
Not only did Plaintiffs’ own evidence point to a lack of predominance, but
ConAgra presented its own, and far superior, evidence that common
understandings did not exist in this case, but the district court did not address it in
its discussion of materiality. ER00250-256. Even if Plaintiffs raise an inference of
materiality, ConAgra is entitled to offer evidence to rebut it. E.g., Caro, 18 Cal.
App. 4th at 667 n.20; Johnson v. Harley-Davidson Motor Co. Group, LLC, 285
F.R.D. 573, 581 (E.D. Cal. 2012) (finding materiality not subject to common proof
where defendants offer persuasive evidence that there are numerous individualized
9 The district court also erred in relying on the surveys without adequately considering their methodologies. It could not have done so – these surveys included very limited information about their methodologies. The most Plaintiffs’ expert Dr. Howlett could say is, with little explanation, that they “appear[ed] valid and reliable.” ER05117.
Case: 15-55727, 09/21/2015, ID: 9691076, DktEntry: 13, Page 54 of 75
60 ■ Class Actions ■ July 201742
issues as to whether the reasonable consumer would find the misconduct
complained of material). This is exactly what ConAgra did.
ConAgra’s survey suffered from none of the infirmities that plagued
Plaintiffs’ surveys. Dr. Hanssens’ consumer survey confirms that “100% Natural”
had no statistically measurable difference in purchase intent or consumer beliefs
about the presence of GMO ingredients between actual and potential Wesson Oil
consumers who were shown Wesson Oil labels with and without the natural
label. See supra Part III. There is no doubt that Dr. Hanssens’ survey was far
superior to the few generalized third-party surveys that Plaintiffs brought forth,
which did not measure the labels on Wesson Oil or even cooking oils
generally. When actual evidence exists as to consumers’ beliefs related to the
products at issue and how the alleged misrepresentation affects their purchase
decision, “the Court does not need to look to the hypothetical reasonable consumer
. . . and does not need to infer reliance given the evidence presented.” Algarin, 300
F.R.D. at 453; Astiana v. Kashi Co., 291 F.R.D. 493, 508 (S.D. Cal. 2013). The
district court found Hanssens’ survey reliable, ER00069; ER00140-141, and had
no apparent reason to ignore it in the predominance analysis.
The only contradictory evidence that the district court considered was that
the U.S. Food and Drug Administration (“FDA”) has never promulgated a formal
definition of the word “natural,” and has declined to require disclosure of a food
Case: 15-55727, 09/21/2015, ID: 9691076, DktEntry: 13, Page 55 of 75
Ascertainability ■ Spivey ■ 6143
product’s use of bioengineered ingredients. ER05382-383; ER00253. Based in
part on consumer focus groups, the FDA has found that such information “is not a
material fact.” See FDA, DRAFT Guidance for Industry: Voluntary Labeling
Indicating Whether Foods Have or Have Not Been Developed Using
Bioengineering (Jan. 2001), available at
http://www.fda.gov/Food/GuidanceRegulation/GuidanceDocumentsRegulatoryInf
ormation/LabelingNutrition/ucm059098.htm (last visited Sept. 18, 2015).
The district court found this evidence irrelevant, stating it did not have
anything to say about how a “reasonable consumer” would have understood the
natural label. ER00253. Numerous other courts have disagreed, holding that such
agency guidance is “relevant and informative” as it is based on extensive studies
and consumer data. Rubio v. Capital One Bank, 613 F.3d 1195, 1201-02 (9th Cir.
2010); see also Jones, 2014 WL 2702726, at *15; Rojas v. General Mills, Inc., No.
12-cv-05099-WHO, 2013 WL 5568389, at *4 (N.D. Cal. Oct. 9, 2013); Ivie v.
Kraft Foods Global, Inc., No. C-12-02554-RMW, 2013 WL 685372, at *12 (N.D.
Cal. Feb. 25, 2013).
The existence of this substantial rebuttal evidence is unsurprising.
Difficulties in pinning down the consumers’ common understanding of “all-
natural” or “100% Natural” labels have plagued food labeling class actions in this
circuit and across the country. In considering similar claims, district courts often
Case: 15-55727, 09/21/2015, ID: 9691076, DktEntry: 13, Page 56 of 75
62 ■ Class Actions ■ July 201744
hold that the term “natural” carries so many different meanings for different
consumers that predominance cannot be shown. For example, in Jones, which is
currently pending on appeal, the district court concluded that there was a lack of
the requisite cohesion among class members because “there is no fixed meaning
for the word ‘natural.’” 2014 WL 2702726, at *14. Citing other district court
cases from this circuit as well as guidance from the FDA, Jones explained that
“there is no single, controlling definition of the word ‘natural,’” and that therefore
different customers will weigh the “natural” label in different ways in their
purchasing decisions. Id. at *15. See also, e.g., Allen v. Hyland’s, Inc., 300 F.R.D.
643, 668 (C.D. Cal. 2014); Kashi, 291 F.R.D. at 504-05; Thurston v. Bear Naked,
Inc., No. 3:11-cv-02890-H (BGS), 2013 WL 5664985, at *8 (S.D. Cal. July 30,
2013).
Similarly, in Randolph v. J.M. Smucker Co., a case involving an “all natural”
label on cooking oils, the court held there was no predominance where the plaintiff
had not shown that a reasonable consumer would agree that “all natural” meant a
product did not contain GMOs. 303 F.R.D. 679, 695-96 (S.D. Fla. 2014). The
facts of Randolph are almost identical to those here, and even involve a Florida
class, like this case. Yet the district court there denied certification based on
ascertainability and predominance. Id.
* * *
Case: 15-55727, 09/21/2015, ID: 9691076, DktEntry: 13, Page 57 of 75
Ascertainability ■ Spivey ■ 6345
For all of the reasons above, the district court abused its discretion by
finding that Plaintiffs could show they were entitled to a classwide inference of
materiality. This holding was based on several findings that were “‘illogical,
implausible, or without support in inferences that may be drawn from facts in the
record,’” Leyva v. Medline Indus. Inc., 716 F.3d 510, 513 (9th Cir. 2013) (citation
omitted), and the district court should have adhered to its initial decision denying
certification. It did not conduct the required rigorous analysis, and its decision to
certify reduces the predominance requirement to a nullity.
B. Individual issues predominate with respect to damages.
The district court also failed to hold Plaintiffs to their burden of showing that
damages are calculable on a classwide basis. Rather, it allowed Plaintiffs to rely
upon little more than the mere assurances of an expert that damages could be
calculated. This expert proposed a facially suspect “hybrid damages model” that
has never been performed by others in the field and has never been used before by
the expert who proposed it. While Plaintiffs need not actually calculate damages at
the certification stage, they must at least proffer a damages model that is both
reliable and persuasive. The district court failed to conduct the rigorous analysis
necessary to ensure that Plaintiffs’ underdeveloped and last-minute model met
these criteria. Accordingly, it did not identify any single, tested model
demonstrably capable of linking Plaintiffs’ damages to their theory of liability.
Case: 15-55727, 09/21/2015, ID: 9691076, DktEntry: 13, Page 58 of 75
64 ■ Class Actions ■ July 201746
1. Comcast requires a damages model linked to Plaintiffs’ theory of liability.
In Comcast Corp. v. Behrend, the Supreme Court emphasized that Rule
23(b)(3) requires plaintiffs, as part of predominance, to “establish[] that damages
are capable of measurement on a classwide basis.” 133 S. Ct. at 1433. Without
such a showing, “[q]uestions of individual damage calculations will inevitably
overwhelm questions common to the class.” Id. Importantly, “a model purporting
to serve as evidence of damages in [a] class action must measure only those
damages attributable to th[e] theory” of liability underlying the class claims. Id.
Simply put, at the class-certification stage, Plaintiffs’ proffered model for
calculating damages “must be consistent with its liability case,” id. (citation
omitted), and “courts must conduct a ‘rigorous analysis’ to determine whether that
is so.” Id. (quoting Dukes, 131 S. Ct. at 2551-52).
Because Plaintiffs’ theory of liability is tied specifically to the believed
absence of GMOs, the district court correctly held that Comcast requires Plaintiffs
to present a damages model capable of calculating not simply the injury resulting
from the natural label, but more specifically, the injury resulting from the portion
of the natural label attributable to the presence of GMOs. ER00257-258. As the
district court put it, Plaintiffs must “segregate the price premium attributable to a
consumer’s understanding that ‘100% Natural’ means the cooking oils contain no
genetically modified organisms.” ER00258.
Case: 15-55727, 09/21/2015, ID: 9691076, DktEntry: 13, Page 59 of 75
Ascertainability ■ Spivey ■ 6547
While the district court correctly required Plaintiffs to proffer a damages
model tied to the value of the absence of GMOs, it erred in determining that
Plaintiffs’ proposed “model” met the requisite standard.
2. The district court twice rejected one of the models that formed the basis of the hybrid model.
In their initial class certification motion, Plaintiffs did not even attempt to
calculate the GMO “price premium.” Instead, they relied predominantly on their
expert witness, Dr. Colin Weir, who proposed a “hedonic regression” model to
calculate only the price premium attributable to the natural label, even though he
acknowledged that “100% Natural” and “non-GMO” are “not equivalent.”
ER00050, ER00100. In its order denying class certification, the district court
concluded that Dr. Weir’s model did not satisfy Comcast and therefore, that
Plaintiffs had not met their burden of showing the predominance of common issues
as to damages. ER00101. However, it gave Plaintiffs a second chance to present a
damages model that would be able to isolate GMO-specific damages.
Incredibly, once given this second opportunity, Plaintiffs’ primary argument
was that the district court had been incorrect that Comcast required them to isolate
the additional price premium attributable to GMOs. See, e.g., ER05251
(“Plaintiffs maintain that under Comcast . . . the appropriate measure of damages is
the percentage of Wesson Oils’ retail price attributable to the ‘100% Natural’
label.”). Perhaps recognizing the significant weaknesses in their proffered models,
Case: 15-55727, 09/21/2015, ID: 9691076, DktEntry: 13, Page 60 of 75
66 ■ Class Actions ■ July 201748
Plaintiffs devoted the vast majority of their briefing on this issue to arguing that
they should be allowed to follow their original approach rather than on proffering
compliant models. See generally ER05302-5314. Even more tellingly, however,
Plaintiffs’ damages experts did the same. Dr. Weir again presented the same
hedonic regression model that calculated the price premium for all possible
meanings of the natural label. He opined, for instance, that the “economic harm
suffered by Plaintiffs and all other members of the proposed Class is the amount of
additional money they paid for Wesson Oil because of the presence of the ‘100%
Natural’ claim,” and testified that the district court’s analysis was incorrect “from
an economic perspective.” ER04996, ¶ 7 (emphasis added); ER07928-29.
Appropriately, the district court again rejected his model as insufficient to satisfy
Comcast.10 ER00259.
3. The district court erroneously accepted a “hybrid” model that was incapable of reliably or persuasively calculating classwide damages.
However, the district court erroneously proceeded to accept Plaintiffs’ last-
ditch attempt to calculate classwide damages, which took the form of a new
declaration by another expert, Dr. Elizabeth Howlett. She proposed multiplying
10 Dr. Weir’s regression model is also insufficient for other reasons identified by other courts that have recently rejected regression models almost identical to Dr. Weir’s. See, e.g., Brazil v. Dole Packaged Foods, LLC, No. 12-CV-01831-LHK, 2014 WL 5794873, at *11-14 (N.D. Cal. Nov. 6, 2014).
Case: 15-55727, 09/21/2015, ID: 9691076, DktEntry: 13, Page 61 of 75
Ascertainability ■ Spivey ■ 6749
the number produced by Dr. Weir’s rejected model by the number produced by her
own model, producing a hybrid model that the district court found to satisfy
Comcast.
To explain why the district court erred, a brief explanation of Dr. Weir’s
hedonic regression model and Dr. Howlett’s conjoint analysis model is necessary.
As explained by Dr. Weir, regression analysis is a tool used by economists that
“identifies and quantifies the relationship between two or more variables.”
ER05010. It does so by comparing the values of a dependent variable—here, the
price of cooking oil—with the values of various independent variables—here, the
natural label, and other factors that might influence price. ER05010, ER05014.
Hedonic regression is a specific application of this analysis that attempts to
determine how much consumers pay for each feature of a product. ER05011. It
uses historic data—that is, information from past sales and pricing. ER05013.
Using this model, Dr. Weir said he could calculate a “percentage price premium”
that he concluded represented the portion of the price of oil attributable to the
natural label. ER05023 & n.53.
Dr. Howlett’s primary proposal was a conjoint analysis technique used in
marketing research to ascertain how much consumers prefer different features of a
product. ER05129. Unlike hedonic regression, which uses historic pricing data,
conjoint analysis conducts surveys of consumers’ present preferences. ER05024.
Case: 15-55727, 09/21/2015, ID: 9691076, DktEntry: 13, Page 62 of 75
68 ■ Class Actions ■ July 201750
Survey participants are asked to choose between multiple iterations of a given
hypothetical product, each with different features. Repeating this process allows a
calculation of the “relative importance” of the tested features. ER05130-31. Dr.
Howlett proposed using six features that consumers might associate with the
natural label, one of which would be the absence of GMOs. ER05133. After
conducting the surveys, each feature would be assigned a percentage ranking.
ER05137.
Much of her declaration, like Dr. Weir’s, proposed calculating damages
resulting from the natural label as a whole, but Dr. Howlett further asserted that
she could simply layer her proposed “conjoint analysis” model onto Dr. Weir’s
hedonic regression to satisfy Comcast. She “propos[ed] to use consumer surveys
to segregate the percentage of the price premium specifically attributable to a
consumer’s belief that ‘100% Natural’ means ‘no GMOs’” and to multiply that
percentage by Dr. Weir’s figures. ER00259-260. This “hybrid damages model” is
what the district court exclusively and erroneously relied upon to find that Comcast
was satisfied. ER00259, ER00261.
The district court appeared to consider only the reliability of Dr. Weir’s and
Dr. Howlett’s models, which it addressed both in the portion of its opinion denying
ConAgra’s motion to strike the experts’ declarations under Federal Rule of
Evidence 702 and Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579
Case: 15-55727, 09/21/2015, ID: 9691076, DktEntry: 13, Page 63 of 75
Ascertainability ■ Spivey ■ 6951
(1993), and in the portion of its opinion relating to Rule 23 predominance.
ER00145-152, ER00158-164, ER00257-270. However, the district court was
required to decide not only whether Plaintiffs’ experts were sufficiently reliable,
but also whether they were sufficiently persuasive. As this Court explained in
Ellis, the focus of Daubert and Rule 702 is whether an expert’s testimony is
sufficiently reliable to be admissible. 657 F.3d at 982. Under Rule 23, however,
the district court must additionally conduct a “rigorous analysis” to determine
whether that testimony is also persuasive. “[T]o the extent the district court
limited its analysis of whether there was [predominance] to a determination of
whether Plaintiffs’ evidence on that point was admissible, it did so in error.” Id.
The Third Circuit has similarly emphasized that a district court cannot
merely conclude that Rule 23 is satisfied because an expert’s opinion is sufficiently
reliable to be admissible.
Opinion testimony should not be uncritically accepted as establishing a Rule 23 requirement merely because the court holds the testimony should not be excluded, under Daubert or for any other reason. Under Rule 23 the district court must be ‘satisfied’ . . . or ‘persuaded’ . . . that each requirement is met before certifying a class. Like any evidence admissible expert opinion may persuade its audience, or it may not.
In re Blood Reagents Antitrust Litig., 783 F.3d 183, 188 n.10 (3d Cir. 2015)
(citations and quotation marks omitted). “[I]n order to certify a class, the damages
methodology must be ‘sound.’” Ludlow v. BP, P.L.C., --- F.3d ---, 2015 WL
Case: 15-55727, 09/21/2015, ID: 9691076, DktEntry: 13, Page 64 of 75
70 ■ Class Actions ■ July 201752
5235010, at *5 (5th Cir. Sept. 8, 2015) (citation omitted). Dr. Howlett’s cursory
hybrid proposal was neither sound nor persuasive, for three reasons.
First, although Plaintiffs and their experts made much of the fact that
hedonic regression and conjoint analysis are each familiar and accepted methods
that are commonly used in their respective fields, ER00146, ER00261, Dr. Howlett
provided no evidence that it was common or accepted to combine the two analyses
in the manner she proposed. ER08057 (citing ER07989-90) (“Dr. Howlett . . . is
unaware of anyone else in a peer-reviewed article that has ever combined the
results of a conjoint analysis and a hedonic regression model in order to assign a
price premium to a sub-feature.”); ER05478 (“[Dr. Howlett is] unaware of anyone
that has exactly done this same thing.”). See Lust v. Merrell Dow Pharm., Inc., 89
F.3d 594, 597 (9th Cir. 1996) (terming it a “‘very significant fact’” that an expert
“failed to subject his method to peer-review and to develop his opinion outside the
litigation” (citation omitted)).
That Plaintiffs could not show that anyone besides Dr. Howlett has thought
to combine two methods that have apparently each been used for decades raises
inherent doubts about the viability of Dr. Howlett’s hybrid damages model, a
model developed expressly for this litigation. Perhaps because of this lack of
precedent in the relevant academic and research communities, this is also the first
time, to ConAgra’s knowledge, that a district court in a food labeling class action
Case: 15-55727, 09/21/2015, ID: 9691076, DktEntry: 13, Page 65 of 75
Ascertainability ■ Spivey ■ 7153
has permitted this particular kind of hybrid damages model as a valid means of
measuring damages.
Not only did Plaintiffs fail to show that this hybrid damages method was
commonly used by others in the field, but perhaps more importantly, Dr. Howlett
freely admitted that she herself had never attempted or used this method, either.
ER00159; ER08116 (“I have not previously used choice-based conjoint analysis to
separate the elements of a price premium calculated with hedonic regression . . .
.”). Moreover, Dr. Howlett also explicitly acknowledged that she was not an
expert as to Dr. Weir’s hedonic regression, a critical component of her hybrid
model. ER08113 (“I readily admit that I am not an expert in hedonic regression
analysis.”). Courts regularly deny certification where plaintiffs offer nothing more
than mere assurances that their expert may be able to satisfy Rule 23, Comcast,
133 S. Ct. at 1434, or present hybrid models that have never before been
attempted. See Prism Techs. LLC v. AT&T Mobility, LLC, Case Nos. 8:12CV122-
126, LLC, 2014 U.S. Dist. LEXIS 132619, at *22 (D. Neb. Sept. 22, 2014).
Second, and as explained by ConAgra’s expert, Dr. Keith Ugone, there are
self-evident incompatibilities between conjoint analysis and hedonic regression
that also should have triggered the district court to conduct a more probing analysis
as to the two methods’ compatibility. The models combine apples and oranges:
hedonic regression measures a historical “price premium,” and conjoint analysis
Case: 15-55727, 09/21/2015, ID: 9691076, DktEntry: 13, Page 66 of 75
72 ■ Class Actions ■ July 201754
measures the relative importance of current consumer preferences. Relative
importance is not synonymous with price premium, as courts have noted.11 In re
NJOY, Inc. Consumer Class Action Litigation, No. CV-14-00428-MMM, 2015 WL
4881091, at *42 (C.D. Cal. Aug. 14, 2015); Oracle Am., Inc. v. Google Inc., No.
10-CV-03561, 2012 WL 850705, at *9-11 (N.D. Cal. Mar. 13, 2012).12 These
cases reflect that consumers’ likings do not necessarily translate into increased
willingness to pay. For example one of Plaintiffs’ materiality surveys found that
well over half of the consumers surveyed state that “natural” or some variation is
an important statement on labels when shopping for foods and beverages.
ER03953. However, the same survey found that only about one third of consumers
11 Even the district court appeared to agree. At the hearing, it made a telling remark that goes to the heart of Comcast: when ConAgra’s counsel pointed out what the model needed to measure to satisfy Comcast but did not, the district court did not disagree but wondered “[g]ee, is there any way of doing that?” and asked if a plaintiff could “ever show that.” ER00117. This suggests that the district court believed Plaintiffs could not meet the Comcast standard here but excused the model’s deficiencies to allow certification nonetheless.
12 In concluding that the hybrid damages model satisfied Comcast, the district court incorrectly relied on Guido v. L’Oreal USA, Inc., Nos. 2:11-CV-01067 & -05465-CAS, 2014 WL 6603730 (C.D. Cal. Jul. 24, 2014) and Khoday v. Symantec Corp.,Civil No. 11-180 (JRT/TNL), 2014 WL 1281600 (D. Minn. Mar. 13, 2014).ER00262-263. Among other differences, neither case involved an attempt to segregate the price of only one of many potential meanings of a label claim. In Guido, unlike here, the damages expert testified that the proposed method in that case was well-accepted in his field, had been subject to peer review and was not developed primarily for the purposes of that litigation. 2014 WL 6603730, at *6.
Case: 15-55727, 09/21/2015, ID: 9691076, DktEntry: 13, Page 67 of 75
Ascertainability ■ Spivey ■ 7355
would be willing to pay more for natural at all – concrete evidence that relative
importance does not translate to price premium. ER03939.
Additionally, Dr. Weir’s hedonic regression utilizes historic data on pricing,
while Dr. Howlett’s conjoint analysis calculates the value to consumers at the time
of the surveys. ER00148, ER00264. Combining these two datasets is problematic
given that consumers’ understanding and valuation of natural claims and the
importance of GMOs are not static over time. See, e.g., ER05449-50 (natural
products are priced 3.76% more than non-natural products in 2009, but 0.25%
more in 2010); ER04953 (natural signifies no GMOs to 46% of consumers in 2012
and 61% in 2010). The hybrid damages model would always yield a fictional
result that could not be accurate as to any one class member. Simply put,
multiplying the relative importance of Dr. Howlett’s “absence of GMOs”
interpretation by Dr. Weir’s price premium value for the natural label does not
magically result in the necessary price premium.
Third, Dr. Howlett’s explanation of the hybrid damages model was cursory
at best. As described above, the district court accepted the hybrid damages model
despite the undisputed fact that it had not been used by other researchers and that
she had never performed the method before. Even if these deficiencies were not
alone enough to render her declaration unpersuasive, they should have at least
alerted the district court to require a thorough explanation of why the untested
Case: 15-55727, 09/21/2015, ID: 9691076, DktEntry: 13, Page 68 of 75
74 ■ Class Actions ■ July 201756
hybrid model would work. See, e.g., Clausen v. M/V New Carissa, 339 F.3d 1049,
1056 (9th Cir. 2003) (“‘Where peer review and publication are absent, the experts
must explain precisely how they went about reaching their conclusions . . . .’”
(citation omitted)).
While Dr. Howlett provided a detailed explanation of conjoint analysis, her
explanation of the hybrid damages model amounted to essentially just a paragraph
of her declaration. ER05138, ¶ 139. Her explanation was conclusory and
supported by little explanation or evidence showing why it was acceptable to
combine conjoint analysis and hedonic regression. ER05138. The district court
should have heeded its conclusion in its first order denying certification, where it
rejected Dr. Weir’s damages model because it was “left with only [the expert’s]
assurance that [they] can build a model to calculate damages.” ER00052-53. See
Randolph, 303 F.R.D. at 697 (“Plaintiff has simply stated, in a conclusory fashion,
that hedonic regression will be able to calculate the premium included in
Defendant's products . . . .”).
* * *
In sum, the district court entirely failed to conduct the requisite rigorous
analysis to determine that Dr. Howlett’s hybrid damages model—the only method
it considered to satisfy Comcast—was reliable and persuasive. The hybrid model
is not an accurate reflection of the market price, rendering it incapable of
Case: 15-55727, 09/21/2015, ID: 9691076, DktEntry: 13, Page 69 of 75
Ascertainability ■ Spivey ■ 7557
calculating a GMO price premium. Worse, the model was untested, either by
general researchers or by Dr. Howlett herself, and insufficiently explained or
supported. Accordingly, the district court abused its discretion in relying upon this
model to show that damages could be calculated on a classwide basis, and it should
not have found that predominance was satisfied with respect to damages.
V. A class action is not a superior vehicle for adjudicating Plaintiffs’ claims because it is unmanageable.
In addition to predominance, Rule 23(b)(3) requires Plaintiffs to demonstrate
the superiority of the class action mechanism. To show superiority, Plaintiffs must
address several factors, one of which is “the likely difficulties in managing a class
action.” Fed. R. Civ. P. 23(b)(3)(D). They must not only identify the difficulties
likely to be encountered in the management of the action, but must also present the
court with some explanation as to how these difficulties will be addressed. Zinser
v. Accufix Research Inst., Inc., 253 F.3d 1180, 1189 (9th Cir. 2001). Here,
Plaintiffs’ explanations do not pass muster. In its initial decision denying
certification, the district court noted that “it ha[d] concerns about the manageability
of any trial proceeding” despite Plaintiffs’ various proposals. ER00103 & n.143.
Indeed, “if a class action…involve[s] the application of multiple states’ laws—
even if only on one claim—it is likely to be too unmanageable under Rule
23(b)(3)(D).” Gartin v. S&M NuTec LLC, 245 F.R.D. 429, 441 (C.D. Cal. 2007).
That is precisely the case here.
Case: 15-55727, 09/21/2015, ID: 9691076, DktEntry: 13, Page 70 of 75
76 ■ Class Actions ■ July 201758
As the district court noted, the certification of eleven separate state classes
will “require proof of different elements” for the various state consumer protection
laws. ER00272. Even if some of these elements may “fall into consistent
patterns,” id., this does not alleviate the fact that different evidence, defenses, jury
instructions and verdict forms will be necessary for these claims. See, e.g.,
Gianino v. Alacer Corp., 846 F. Supp. 2d 1096, 1099 (C.D. Cal. 2012). The
district court’s proposed solution to this problem was to “sever the classes
following certification for separate adjudication of the claims of the state classes.”
ER00273. Yet eleven separate trials undermines the primary purposes of class
actions – to conserve judicial resources and prevent multiple trials on the same
liability issues. Gunnells v. Healthplan Servs., Inc., 348 F.3d 417, 446 (4th Cir.
2003).
The same reasons that make the classes unascertainable, supra Part II, also
make it unmanageable. ER05364, n.13. That Plaintiffs have failed to identify an
administratively feasible method for identifying class members demonstrates that
there are significant and unresolved manageability problems. Even courts that
(wrongly) reject the proposition that Plaintiffs must demonstrate that their class
definition is administratively feasible as part of the ascertainability analysis
consider it relevant to manageability. See, e.g., Mullins, 795 F.3d at 664.
Accordingly, even if this Court declines to consider administrative feasibility under
Case: 15-55727, 09/21/2015, ID: 9691076, DktEntry: 13, Page 71 of 75
Ascertainability ■ Spivey ■ 7759
the rubric of ascertainability, it should hold that the class mechanism would be
unmanageable and therefore not superior.
* * *
If this Court concludes, as it should, that the district court abused its
discretion as to either ascertainability, typicality, predominance, or superiority, it
should reverse the decision below and order the district court to deny certification
of the putative classes.13 Plaintiffs have had numerous opportunities to file
complaints and class certification motions that would satisfy the strictures of Rule
23. They have failed to do so, and should not be granted yet another opportunity to
prolong this litigation.
CONCLUSION
For the foregoing reasons, this Court should reverse the decision of the
district court and remand with instructions to deny Plaintiffs’ amended motion for
class certification.
STATEMENT OF RELATED CASES
Under Ninth Circuit Rule 28-2.6, ConAgra is aware of three related cases
pending in this Court that come within the requirements of the Rule: Brazil v.
13 A ruling on materiality alone would affect only nine of the eleven classes. Seesupra note 5.
Case: 15-55727, 09/21/2015, ID: 9691076, DktEntry: 13, Page 72 of 75
78 ■ Class Actions ■ July 201760
Dole Packaged Foods, LLC, No. 14-17480, Jones v. ConAgra Foods, No. 14-
16327, and Bruton v. Gerber Products Co., No. 15-15174.
Dated: September 21, 2015 Respectfully submitted,
MCGUIREWOODS LLP
/s/ A. Brooks Gresham A. Brooks Gresham Angela M. Spivey R. Trent Taylor Laura E. Coombe E. Rebecca Gantt
Counsel for Defendant-Appellant ConAgra Foods, Inc.
Case: 15-55727, 09/21/2015, ID: 9691076, DktEntry: 13, Page 73 of 75
Ascertainability ■ Spivey ■ 7961
CERTIFICATE OF COMPLIANCE
1. This brief complies with the type-volume limitation of Federal Rule of
Appellate Procedure 32(a)(7)(B) because it contains 13,881 words,
excluding the parts of the brief exempted by Federal Rule of Appellate
Procedure 32(a)(7)(B)(iii).
2. This brief complies with the typeface and type-style requirements of Federal
Rule of Appellate Procedure 32(a)(5) and Federal Rule of Appellate
Procedure 32(a)(6) because it has been prepared in proportionally-spaced
typeface using Microsoft Word, in 14-point size.
Dated: September 21, 2015 /s/ A. Brooks Gresham A. Brooks Gresham
Counsel for Defendant-Appellant ConAgra Foods, Inc.
Case: 15-55727, 09/21/2015, ID: 9691076, DktEntry: 13, Page 74 of 75
80 ■ Class Actions ■ July 201762
CERTIFICATE OF SERVICE
I hereby certify that on September 21, 2015, I electronically filed the
foregoing with the Clerk of the Court for the United States Court of Appeals for
the Ninth Circuit using the appellate CM/ECF system. All participants in the
case are registered CM/ECF users and will be served by the system.
Dated: September 21, 2015 /s/ A. Brooks GreshamA. Brooks Gresham
Counsel for Defendant-Appellant ConAgra Foods, Inc.
Case: 15-55727, 09/21/2015, ID: 9691076, DktEntry: 13, Page 75 of 75
Ascertainability ■ Spivey ■ 81
III. Briseño v. Conagra Foods, Inc., U.S. Court of Appeals for the Ninth Circuit. Answering Brief of Plaintiffs-AppelleesNo. 15-55727
In the
UNITED STATES COURT OF APPEALS FOR THE NINTH CIRCUIT
ROBERT BRISEÑO, Plaintiff-Appellee,
v.
CONAGRA FOODS, INC., Defendant-Appellant.
ON APPEAL FROM THE UNITED STATES DISTRICT COURT FOR THE CENTRAL DISTRICT OF CALIFORNIA
(The Honorable Margaret M. Morrow) Case No. 2:11-cv-05379
ANSWERING BRIEF OF PLAINTIFFS-APPELLEES David E. Azar MILBERG LLP 10866 Wilshire Blvd, Suite 600 Los Angeles, California 90024 Telephone: (213) 617-1200 Ariana J. Tadler Henry J. Kelston Meagan Keenan Carey Alexander MILBERG LLP One Pennsylvania Plaza New York, New York 11030 Telephone: (212) 594-5300
Adam J. Levitt Edmund S. Aronowitz GRANT & EISENHOFER P.A. 30 North LaSalle Street Suite 2350 Chicago, Illinois 60605 Telephone: (312) 214-0000 Mary S. Thomas GRANT & EISENHOFER P.A. 123 Justison Street 7th Floor Wilmington, Delaware 19801 Telephone: (302) 622-7000
Class Counsel for Plaintiffs-Appellees Robert Briseño, et al.
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 1 of 90
82 ■ Class Actions ■ July 2017 i
TABLE OF CONTENTS
TABLE OF CONTENTS .............................................................................. i
TABLE OF AUTHORITIES ...................................................................... iv
STATEMENT WITH RESPECT TO ORAL ARGUMENT ..................... xii
INTRODUCTION ....................................................................................... 1
JURISDICTIONAL STATEMENT ............................................................ 3
STATEMENT OF THE ISSUES ................................................................ 3
STATEMENT OF THE CASE ................................................................... 3
SUMMARY OF ARGUMENT .................................................................... 6
ARGUMENT ............................................................................................... 9
I. The District Court’s class certification decision is subject to an abuse of discretion standard of review. ............................................ 9
II. The District Court did not abuse its discretion in holding that the eleven certified classes are ascertainable. ....................................... 9
A. The District Court applied the correct legal standard in evaluating the “definiteness” or “ascertainability” of the classes certified. ...................................................................... 11
B. Where the challenged misrepresentation appeared on every unit sold, class member self-identification satisfies ascertainability. ...................................................................... 14
C. Imposing a heightened ascertainability standard would not protect defendants’ due process rights but would imperil the viability of low-dollar value consumer class actions. ..... 21
III. The District Court correctly determined that Plaintiffs satisfied Rule 23(a)(3)’s typicality requirement and Rule 23(b)(3)’s predominance and superiority requirements. ................................ 31
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 2 of 90
Ascertainability ■ Spivey ■ 83ii
A. The class representatives’ claims are typical of other class members’ claims because they all arise from ConAgra’s common course of conduct. ..................................................... 32
1. Like the class representatives, all class members purchased Wesson Oils and paid more for them as a result of the “100% Natural” label. .............................. 33
2. Varying understandings of the term “100% Natural” among class members are irrelevant to typicality. ..... 34
3. That some class representatives might consider purchasing Wesson Oils in the future if ConAgra makes true disclosures is irrelevant to typicality. ...... 36
B. Plaintiffs’ claims are susceptible to classwide proof, satisfying Rule 23’s predominance requirement. ................. 38
1. Whether the “100% Natural” claim on Wesson Oils was false, misleading, and material to a reasonable consumer are inherently common, classwide questions. ....................................................................... 40
2. ConAgra’s demand that this Court reweigh common evidence and find Plaintiffs failed to demonstrate materiality goes far beyond the probing of merits allowed at the class certification stage. ....................... 45
3. Classwide evidence suggests ConAgra’s “100% Natural” claim on Wesson Oils is material to a reasonable consumer. .................................................... 48
C. The class action mechanism is superior to any other available method for adjudicating Plaintiffs’ claims concerning a uniformly labeled, low-dollar value consumer product. .................................................................. 58
1. Without the class action device, no individual purchaser of Wesson Oils could reasonably afford to seek a legal remedy. ...................................................... 59
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 3 of 90
84 ■ Class Actions ■ July 2017 iii
2. There is no reason that the factual issues in this case could not be adjudicated in a single trial. .................... 60
3. Eleven trials are superior to separate individual trials for each of the tens of millions of Wesson Oils purchasers. .................................................................... 60
IV. Plaintiffs’ damages methodology satisfies Comcast. ....................... 62
A. Plaintiffs’ proposed hedonic regression methodology reliably isolates the percentage “price premium” specifically and solely attributable to the presence of a “natural” label claim on cooking oils. .................................... 68
B. Analysis of choice-based conjoint survey results reliably isolates the percentage “relative importance” of the “no-GMO” portion of a “natural” claim. ....................................... 69
C. Combining the percentage yielded by hedonic regression with the percentage yielded by conjoint analysis is proper. ..................................................................................... 71
CONCLUSION .......................................................................................... 75
STATEMENT OF RELATED CASES ..................................................... 75
CERTIFICATE OF COMPLIANCE ......................................................... 76
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 4 of 90
Ascertainability ■ Spivey ■ 85iv
TABLE OF AUTHORITIES
Page(s) CASES
Alcantar v. Hobart Serv., 800 F.3d 1047 (9th Cir. 2015) .............................................................. 45
Algarin v. Maybelline, 300 F.R.D. 444 (S.D. Cal. 2014) ........................................................... 19
Allapattah Servs. v. Exxon Corp., 333 F.3d 1248 (11th Cir. 2003) ............................................................ 67
Amchem Products, Inc. v. Windsor, 521 U.S. 591 (1997) ........................................................................ 29, 59
Amgen Inc. v. Conn. Ret. Plans & Trust Funds, 133 S. Ct. 1184 (2013) .................................................................. passim
Astiana v. Ben & Jerry’s Homemade, Inc., No. C 10-4387, 2014 WL 60097 (N.D. Cal. Jan. 7, 2014) ................... 17
Astiana v. Kashi Co., 291 F.R.D. 493 (S.D. Cal. 2013) ............................................... 15, 29, 37
Blackie v. Barrack, 524 F.2d 891 (9th Cir. 1975) ................................................................ 47
Brazil v. Dole Packaged Foods, LLC, No. 12-CV-01831, 2014 WL 2466559 (N.D. Cal. May 30, 2014) ........ 26
Brazil v. Dole Packaged Foods, LLC, No. 12-CV-01831, 2014 WL 5794873 (N.D. Cal. Nov. 6, 2014) .......... 68
Brecher v. Republic of Argentina, 806 F.3d 22 (2d Cir. 2015) ................................................................... 21
Bruton v. Gerber Products Co., No. 12-CV-02412, 2014 WL 2860995 (N.D. Cal. June 23, 2014) ....... 17
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 5 of 90
86 ■ Class Actions ■ July 2017 v
Busey v. Macon County Greyhound Park, Inc., 562 F. App’x 782 (11th Cir. 2014) ........................................................ 22
Butler v. Sears, Roebuck & Co., 727 F.3d 796 (7th Cir. 2013) ................................................................ 38
Byrd v. Aaron’s Inc., 784 F.3d 154 (3d Cir. 2015) ........................................................... 21, 25
Carnegie v. Household Int’l, Inc., 376 F.3d 656 (7th Cir. 2004) ................................................................ 59
Carrera v. Bayer Corp., 727 F.3d 300 (3d Cir. 2013) ......................................................... passim
Comcast Corp. v. Behrend, 133 S. Ct. 1426 (2013) .................................................................. passim
Daar v. Yellow Cab Co., 433 P.2d 732 (Cal. 1967) ................................................................ 12, 28
Daubert v. Merrell Dow Pharm., Inc., 509 U.S. 579 (1993) .............................................................................. 66
Doninger v. Pac. Nw. Bell, Inc., 564 F.2d 1304 (9th Cir. 1977) .............................................................. 13
Edwards v. First Am. Corp., 798 F.3d 1172 (9th Cir. 2015) .............................................................. 38
Edwards v. Ford Motor Co., 603 F. App’x 538 (9th Cir. 2015) .......................................................... 41
Ellis v. Costco Wholesale Corp., 657 F.3d 970 (9th Cir. 2011) ................................................................ 46
EQT Prod. Co. v. Adair, 764 F.3d 347 (4th Cir. 2014) .......................................................... 22, 53
Erica P. John Fund, Inc. v. Halliburton Co., 563 U.S. 804, 131 S. Ct. 2179 (2011) ................................................... 39
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 6 of 90
Ascertainability ■ Spivey ■ 87vi
Estrella v. Freedom Fin. Network, LLC, No. C 09-03156, 2010 WL 2231790 (N.D. Cal. June 2, 2010) ............ 35
Exxon Corp. v. Texas Motor Exch. of Houston, Inc., 628 F.2d 500 (5th Cir. 1980) ................................................................ 54
Fairbanks v. Farmers New World Life Ins. Co., 197 Cal. App. 4th 544 (2011) ............................................................... 56
Fancaster, Inc. v. Comcast Corp., 832 F. Supp. 2d 380 (D.N.J. 2011) ...................................................... 54
Forcellati v. Hyland’s, Inc., No. CV 12–1983, 2014 WL 1410264 (C.D. Cal. Apr. 9, 2014) 15, 26, 30
Forrand v. Federal Exp. Corp., No. CV 08-1360, 2013 WL 1793951 (C.D. Cal. Apr. 15, 2013) ........... 53
Fractus, S.A. v. Samsung, No. 6:09-cv-203, 2011 WL 7563820 (E.D. Tex. Apr. 29, 2011) ........... 55
Frey v. First Nat’l Bank Sw., 602 F. App’x 164 (5th Cir. 2015) .......................................................... 22
Guido v. L’Oreal, USA, Inc., Nos. CV CV 11–1067, CV 11–5465, 2013 WL 3353857 (C.D. Cal. July 1, 2013) ................................................................................... 15, 70
Halperin v. Nichols, Safina, Lerner & Co., No. 94 C 6960, 1996 WL 634037 (N.D. Ill. Oct. 29, 1996) .................. 35
Hanlon v. Chrysler Corp., 150 F.3d 1011 (9th Cir. 1998) ........................................................ 32, 38
Hanon v. Dataproducts Corp., 976 F.2d 497 (9th Cir. 1992) ................................................................ 32
Harold’s Stores, Inc. v. Dillard Dep’t Stores, Inc., 82 F.3d 1533 (10th Cir. 1996) .............................................................. 54
Henry Schein, Inc. v. Stromboe, 102 S.W.3d 675 (Tex. 2003) ................................................................. 57
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 7 of 90
88 ■ Class Actions ■ July 2017 vii
Hilao v. Estate of Marcos, 103 F.3d 767 (9th Cir. 1996) ................................................................ 30
Hinojos v. Kohl’s Corp., 718 F.3d 1098 (9th Cir.2013) ............................................................... 41
In re Celexa v. Lexapro Mktg. & Sales Practices Litig., 291 F.R.D. 13 (D. Mass. 2013) ............................................................. 56
In re Clorox Consumer Litig., 301 F.R.D. 436 (N.D. Cal. 2014) .......................................................... 18
In re Countrywide Fin. Corp. Mortg. & Sales Practices Litig., Nos. 08md1988, 10cv0257, 2011 WL 6325877 (S.D. Cal. Dec. 16, 2011) ...................................................................................................... 56
In re Nexium Antitrust Litig., 777 F.3d 9 (1st Cir. 2015) .................................................................... 21
In re NJOY Consumer Class Action Litig., No. CV 14-00428, 2015 WL 4881091 (C.D. Cal. Aug. 14, 2015) (Morrow, J.) .......................................................................................... 26
In re POM Wonderful, LLC, No. ML10-02199, 2014 WL 1225184 (C.D. Cal. Mar. 25, 2014) ........ 18
In re Prempro Prods. Liab. Litig., 514 F.3d 825 (8th Cir.2008) ................................................................. 72
In re Tobacco II Cases, 207 P.3d 20 (Cal. 2009) ........................................................................ 42
In re Vioxx Class Cases, 180 Cal. App. 4th 116 (2009) ............................................................... 56
In re Visa Check/Mastermoney Antitrust Litig. v. Visa, U.S.A., Inc., 280 F.3d 124 (2d Cir. 2001) ................................................................. 60
In re Wells Fargo Home Mortg. Overtime Pay Litig., 571 F.3d 953 (9th Cir. 2009) .................................................................. 9
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 8 of 90
Ascertainability ■ Spivey ■ 89viii
Int’l Union of Bricklayers & Allied Craftsman Local Union No. 20, AFL-CIO v. Martin Jaska, Inc., 752 F.2d 1401 (9th Cir. 1985) .............................................................. 43
Jones v. ConAgra Foods, Inc., No. C 12-01633, 2014 WL 2702726 (N.D. Cal. June 13, 2014) .............. ................................................................................................... 17, 19, 58
Khoday v. Symantec Corp., 93 F. Supp. 3d 1067, 1082 (D. Minn. 2015) ........................................ 70
Klay v. Humana, Inc., 382 F.3d 1241 (11th Cir. 2004) ............................................................ 61
Kosta v. Del Monte Foods, Inc., 308 F.R.D. 217 (N.D. Cal. 2015) .................................................... 52, 53
Krueger v. Wyeth, Inc., No. 03CV2496, 2015 WL 5839197 (S.D. Cal. Oct. 7, 2015) .......... 15, 26
Kwikset Corp. v. Superior Court, 246 P.3d 877 (Cal. 2011) ...................................................................... 74
Lilly v. Jamba Juice Co., 308 F.R.D. 231 (N.D. Cal. 2014) (Tigar, J.) ................................... 26, 29
Mailloux v. Arrow Fin. Servs., LLC, 204 F.R.D. 38 (E.D.N.Y. 2001) ............................................................. 35
Marsu, B.V. v Walt Disney Co., 185 F.3d 932 (9th Cir. 1999) ................................................................ 67
Martin v. Pacific Parking Sys. Inc., 583 F. App’x 803 (9th Cir. 2014) .......................................................... 16
McCrary v. Elations Co., LLC, No. EDCV 13-00242, 2014 WL 1779243 (C.D. Cal. Jan. 13, 2014) .......................................................................................... 15, 24, 29
Moore v. Ulta Salon, Cosmetics & Fragrance, Inc., No. CV 12-3224, 2015 WL 7422597 (C.D. Cal. Nov. 16, 2015) .......... 11
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 9 of 90
90 ■ Class Actions ■ July 2017 ix
Mullins v. Direct Digital, LLC, 795 F.3d 654 (7th Cir. 2015) ........................................................ passim
O’Connor v. Boeing N. Am., Inc., 184 F.R.D. 311 (C.D. Cal. 1998) .......................................................... 11
O’shana v. Coca-Cola Co., No. 04 C 3596, 2005 WL 1661999 (N.D. Ill. July 13, 20) ................... 56
Pizza Hut, Inc. v. Papa John’s Int’l, 227 F.3d 489 (5th Cir. 2000) ................................................................ 55
Pulaski & Middleman, LLC v. Google, Inc., 802 F.3d 979 (9th Cir. 2015) .......................................................... 41, 67
Rahman v. Mott’s LLP, No. 13-cv-03482, 2014 WL 6815779 (N.D. Cal. Dec. 3, 2014) ............ 26
Red v. Kraft Foods, Inc., No. CV 10-1028, 2012 WL 8019257 (C.D. Cal. Apr. 12, 2012) ........... 37
Ries v. AriZona Beverages USA LLC, 287 F.R.D. 523 (N.D. Cal. 2012) .......................................................... 15
Rikos v. Procter & Gamble Co., 799 F.3d 497 (6th Cir. 2015) .......................................................... 22, 25
Sanchez-Knutson v. Ford Motor Co., No. 14-61344, 2015 WL 6395040 (S.D. Fla. Oct. 6, 2015) .................. 70
Sethavanish v. ZonePerfect Nutrition Co., No. 12-2907, 2014 WL 580696 (N.D. Cal. Feb. 13, 2014) ............. 18, 26
Stearns v. Ticketmaster Corp., 655 F.3d 1013 (9th Cir. 2011) ...................................................... passim
Story Parchment Co. v. Paterson Parchment Paper Co., 282 U.S. 555 (1931) .............................................................................. 67
Strawn v. Farmers Ins. Co. of Oregon, 258 P.3d 1199 (Or. 2011) ..................................................................... 57
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 10 of 90
Ascertainability ■ Spivey ■ 91x
THOIP v. Walt Disney Co., 690 F. Supp. 2d 218 (S.D.N.Y. 2010) ................................................... 54
Universal City Studios, Inc. v. Nintendo Co., Ltd., 746 F.2d 112 (2d Cir. 1984) ................................................................. 54
Water Pik, Inc. v. Med-Systems, Inc., 726 F.3d 1136 (10th Cir. 2013) ............................................................ 54
Wells Fargo & Co. v. WhenU.com, Inc., 293 F. Supp. 2d 734 (E.D. Mich. 2003) ............................................... 55
Wolin v. Jaguar Land Rover N. Am., LLC, 617 F.3d 1168 (9th Cir. 2010) ................................................ 9, 32
Xavier v. Philip Morris USA Inc., 787 F. Supp. 2d 1075 (N.D. Cal. 2011) ................................................ 19
Yokoyama v. Midland Nat’l Life Ins. Co., 594 F.3d 1087 (9th Cir. 2010) .............................................................. 42
Young v. Nationwide Mut. Ins. Co., 693 F.3d 532 (6th Cir. 2012) ................................................................ 22
RULES
Central District of California L.R. 7-6 ..................................................... 27
Federal Rule of Civil Procedure 23 .................................................. passim
OTHER AUTHORITIES
Charles Alan Wright et al., 7A Fed. Prac. & Proc. Civ. § 1760 (3d ed. 2015) ................................................................................................ 23
Geoffrey C. Shaw, Note, Class Ascertainability, 124 Yale L.J. 2354 (2015) .................................................................................................... 28
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 11 of 90
92 ■ Class Actions ■ July 2017 xi
Jerold S. Solovy et al., 5-23 Moore’s Federal Practice - Civil § 23.21 (2015) ....................................................................................... 23
Joseph M. McLaughlin, 1 McLaughlin on Class Actions § 4:2 (12th ed. 2015) ................................................................................................ 23
MANUAL FOR COMPLEX LITIGATION (FOURTH) § 21.222 (2004) ................ 24
Shari S. Diamond, Reference Guide on Survey Research, in REFERENCE MANUAL ON SCIENTIFIC EVIDENCE (Federal Judicial Center 3d ed. 2011) .............................................................................. 51
William B. Rubenstein, Newberg on Class Actions § 3:1 (5th ed. 2015) ...................................................................................................... 23
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 12 of 90
Ascertainability ■ Spivey ■ 93xii
STATEMENT WITH RESPECT TO ORAL ARGUMENT
Pursuant to Federal Rule of Appellate Procedure 34(a), Plaintiffs-
Appellees state that oral argument should be heard in this case because
of the importance of this Court’s decision to the ability of consumers to
bring class actions involving low-priced consumer goods.
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 13 of 90
94 ■ Class Actions ■ July 2017 1
INTRODUCTION
The District Court properly exercised its discretion in certifying
eleven statewide classes and 22 class claims under Rule 23(b)(3).
This is a straightforward consumer case based on a uniform “100%
Natural” misrepresentation on the label of all Wesson brand cooking
oils (“Wesson Oils”). Class certification under Rule 23(b)(3) is
particularly appropriate in this type of case, and Appellees Robert
Briseño, Michele Andrade, Jill Crouch, Julie Palmer, Pauline Michael,
Cheri Shafstall, Dee Hopper-Kercheval, Kelly McFadden, Necla Musat,
Maureen Towey, Erika Heins, Rona Johnston, and Anita Willman
(“Plaintiffs”) satisfied all certification requirements. Appellant
ConAgra Foods, Inc. (“ConAgra”) incorrectly contends that the District
Court erred in its analysis of ascertainability, typicality, predominance,
superiority, and Plaintiffs’ damages methodology.
The classes are ascertainable because they are defined by
objective criteria so that class members can self-identify through
declarations or claim forms under the facts of this case. Plaintiffs’
claims are typical of, and, indeed, identical to, the claims of the classes
they represent. Common liability questions (including the falsity and
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 14 of 90
Ascertainability ■ Spivey ■ 952
materiality of the common “100% Natural” representation, and its
ability to cause economic harm) predominate over any individualized
questions. Class adjudication is the superior method for fairly and
efficiently adjudicating the claims of the millions of purchasers of
Wesson Oils. Finally, Plaintiffs’ proposed damages methodology (a
combination of hedonic regression and conjoint analysis) translates
ConAgra’s act of misrepresenting Wesson Oils as “100% Natural” into
its economic impact on the classes.
For all of these reasons, the decision below should be affirmed.
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 15 of 90
96 ■ Class Actions ■ July 2017 3
JURISDICTIONAL STATEMENT
ConAgra’s jurisdictional statement is accurate.
STATEMENT OF THE ISSUES
1. Whether the District Court properly exercised its discretion in holding ascertainability is satisfied because the class definition is based on objective criteria and because self-identification through claim forms is a valid method for determining individual class members’ identities under the facts of this case.
2. Whether the District Court properly exercised its discretion in certifying classes of purchasers of Wesson brand cooking oils in eleven states with respect to 22 specifically enumerated state law claims under Federal Rule of Civil Procedure 23(b)(3).
3. Whether the District Court properly exercised its discretion in
holding Plaintiffs’ proposed methodology for calculating classwide damages translates their theory of liability into its economic impact in satisfaction of Comcast Corp. v. Behrend, 133 S. Ct. 1426 (2013).
STATEMENT OF THE CASE
ConAgra represented that every container of Wesson Oils was
“100% Natural” on the front label every unit sold throughout the class
period. (ER00138.) There are no consumer “varieties” of Wesson Oil on
which the “100% Natural” claim did not appear. (ER00149.)
When Plaintiffs purchased Wesson Oils, they reasonably believed
that Wesson Oils were made from natural ingredients because ConAgra
had labeled them “100% Natural.” (ER003452-ER003485.) Plaintiffs
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 16 of 90
Ascertainability ■ Spivey ■ 974
later learned that ConAgra made Wesson Oils from genetically modified
organisms (“GMOs”), specifically GMO corn, soy, and canola. (Id.)
Plaintiffs alleged ConAgra’s “100% Natural” label is fraudulent,
deceptive, and misleading to reasonable consumers because products
derived from GMOs, like Wesson Oils, cannot honestly be labeled as
“natural.” (ER05866-ER05867; see also ER06044-ER06094.) Plaintiffs
alleged ConAgra’s “100% Natural” label harmed all purchasers of
Wesson Oils because that label created a retail price premium they all
paid. (See, e.g., ER05899.) Plaintiffs asserted state law claims for
breach of consumer protection statutes, breach of express or implied
warranty, and unjust enrichment. (ER05895-ER05906.)
On July 14, 2014, Plaintiffs moved for certification of twelve
statewide classes. (ER05932-ER06020.)
On August 1, 2014, the District Court denied class certification,
without prejudice, for three reasons. First, the District Court struck the
opening expert report of economist Colin Weir because the Court
considered some of Mr. Weir’s supporting information to have been
untimely provided. (ER00047-ER00053; ER00098-99.) Second, the
District Court found that Plaintiffs (having presented state-specific
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 17 of 90
98 ■ Class Actions ■ July 2017 5
information in an appendix) had not shown whether the substantive
laws of states other than California required “individualized proof of
reliance and/or causation.” (ER00096-ER00097.) Third, the District
Court held that Plaintiffs’ proposed damages methodology did “not
suffice under Comcast” because Plaintiffs intended to calculate “the
price premium attributable to use of the term ‘100% Natural’ and all of
the meanings consumers ascribe to it,” without isolating the price
premium associated with the “no-GMO” meaning of “natural.”
(ER00101.)
On November 14, 2014, Plaintiffs filed their amended motion for
certification of eleven statewide classes and 34 state law claims.
(ER07766-ER07778.) In support, Plaintiffs submitted an amended
expert report from Mr. Weir that now included the necessary
supporting information and an expert report from Dr. Elizabeth
Howlett that described how a choice-based conjoint analysis survey
could isolate the relative value of a “no-GMO” meaning of the “natural”
claim on Wesson Oils. (ER04922-ER05227.) Plaintiffs also submitted
various internal ConAgra marketing studies and third-party surveys
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 18 of 90
Ascertainability ■ Spivey ■ 996
regarding factors that motivated purchase of Wesson Oils and consumer
understandings of natural claims. (ER00250-ER00251.)
On February 23, 2015, the District Court granted, in part,
Plaintiffs’ amended class certification motion. The District Court
admitted the majority of Plaintiffs’ expert testimony, including Mr.
Weir’s hedonic regression report and Dr. Howlett’s conjoint analysis
report. The District Court analyzed the elements of all 34 claims.
(ER00201-ER00250.) The District Court weighed the evidence and held
Plaintiffs “made a sufficient showing for the purposes of class
certification that the ‘100% Natural’ claim is material and that
consumers generally understand it, inter alia, as a representation that
Wesson Oils do not contain GMOs.” (ER00252.) The District Court
certified all eleven statewide classes and the 22 claims in which
Plaintiffs could establish elements of reliance and/or causation from
classwide evidence. (ER00274-ER00275.)
SUMMARY OF ARGUMENT
First, the District Court properly found that Plaintiffs had
satisfied ascertainability because the classes are defined by objective
criteria (purchase of Wesson Oils during the class period) so that class
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 19 of 90
100 ■ Class Actions ■ July 2017 7
members can self-identify in a claims process. Self-identification is
reliable here because all units of Wesson Oils had the same “100%
Natural” claim on the label throughout the class period. Certification
does not deprive ConAgra of the ability to raise any defenses, nor is
there a danger of classwide judgment failing to achieve preclusive res
judicata effect, as final judgment will resolve any future claims
premised on the “100% Natural” label on Wesson Oils. Adopting the
view of ascertainability advanced by ConAgra—that sworn claims forms
attesting to purchase could not satisfy ascertainability here—would
effectively eliminate the entire category of low-dollar value consumer
class actions.
Second, the District Court correctly held that Plaintiffs satisfied
all Rule 23(a) prerequisites and the Rule 23(b)(3) predominance and
superiority requirements because (i) Plaintiffs’ certified consumer
protection, breach of warranty, and unjust enrichment claims are
identical to the claims of the other class members they seek to
represent, (ii) the answers to common classwide questions regarding
falsity, materiality, and damages predominate over any individualized
issues, and (iii) classwide adjudication of claims regarding the labeling
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 20 of 90
Ascertainability ■ Spivey ■ 1018
of Wesson Oils is superior to either millions of individual adjudications,
or, as would be more likely, no adjudication at all.
Third, Plaintiffs’ methodology for calculating classwide damages—
a combination of (i) the results of a hedonic regression of historical sales
data to determine the “natural” label price premium percentage on
cooking oils, and (ii) the results of a conjoint analysis of survey
responses to determine the relative value consumers assign to a “no
GMO” interpretation of a “natural” claim—aligns with Plaintiffs’
liability theory. Plaintiff’s liability theory is that ConAgra misled
reasonable consumers by labeling GMO-derived Wesson Oils as 100%
Natural. This 100% Natural label created a price premium all Wesson
Oils purchasers had to pay. Plaintiffs damages methodology measures
this “100% Natural” price premium and then subdivides it by the
relative importance of “no-GMO” meaning of “natural.” Thus, Plaintiffs’
damages methodology satisfies Comcast, because it translates
Plaintiffs’ liability theory into its economic impact.
For all these reasons, the District Court’s certification order
should be affirmed.
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 21 of 90
102 ■ Class Actions ■ July 2017 9
ARGUMENT
I. The District Court’s class certification decision is subject to an abuse of discretion standard of review.
This Court reviews class certification decisions for abuse of
discretion. In re Wells Fargo Home Mortg. Overtime Pay Litig., 571
F.3d 953, 957 (9th Cir. 2009). “When reviewing a grant of class
certification, we accord the district court noticeably more deference than
when we review a denial of class certification.” Wolin v. Jaguar Land
Rover N. Am., LLC, 617 F.3d 1168, 1171 (9th Cir. 2010) (citation and
internal quotation omitted).
II. The District Court did not abuse its discretion in holding that the eleven certified classes are ascertainable.
Accepting Plaintiffs’ proposed definitions, the District Court
defined the eleven statewide classes as “[a]ll persons who reside in the
[class states] who have purchased Wesson Oils within the applicable
statute of limitations periods established by the laws of their state of
residence (the ‘Class Period’) through the final disposition of this and
any and all related actions.” (ER00139; ER00274-ER00275.) In both its
August 1, 2014 and February 2, 2015 Orders, the District Court
analyzed whether these classes were sufficiently definite and thus
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 22 of 90
Ascertainability ■ Spivey ■ 10310
ascertainable, and held that they were. (ER00075-ER00081; ER00181-
ER00185.)
The District Court did not err. (Cf. Appellant’s Br. 12-23.) First,
the District Court applied the correct legal standard in evaluating the
“definiteness” or “ascertainability” of the classes. Second, the District
Court appropriately exercised its discretion in holding that, where the
alleged misrepresentation consistently appeared on the label of every
unit of the product throughout the class period, class member self-
identification through declarations or claim forms is a manageable and
reliable method for determining the identities of individual class
members. Third, allowing this case to proceed as a class action for
damages and equitable monetary relief does not imperil ConAgra’s due
process right to present its defenses. Adopting ConAgra’s
“ascertainability” interpretation—that claims forms are insufficient as a
matter of law—“would effectively prohibit class actions involving low
priced consumer goods (the very type of claims that would not be filed
individually) thereby upending ‘[t]he policy at the very core of the class
action mechanism.’” (ER00184 (citation omitted).)
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 23 of 90
104 ■ Class Actions ■ July 2017 11
A. The District Court applied the correct legal standard in evaluating the “definiteness” or “ascertainability” of the classes certified.
“Ascertainability” is not explicitly mentioned in Rule 23. See Fed.
R. Civ. P. 23. Instead, “ascertainability” is a court-constructed doctrine
that confirms “a class does in fact exist” and “the description of the class
is definite enough so that it is administratively feasible for the court to
ascertain whether an individual is a member.” O’Connor v. Boeing N.
Am., Inc., 184 F.R.D. 311, 319 (C.D. Cal. 1998) (emphasis added); see
also Moore v. Ulta Salon, Cosmetics & Fragrance, Inc., No. CV 12-3224,
2015 WL 7422597, at *15 (C.D. Cal. Nov. 16, 2015) (“[C]ourts should
‘take notice’ of whether the class definition is ‘precise, objective, and
presently ascertainable[,]’ but they need not address such issues
directly ‘unless they pose a problem in the context of the particular case
at hand.’”)(citations omitted).
The ability of the class proponent or the court to identify every
person who falls within the class definition with perfect accuracy is not
a proper component of the ascertainability analysis, nor is it a
prerequisite to certification. See Mullins v. Direct Digital, LLC, 795
F.3d 654, 657-658 (7th Cir. 2015) (noting some “courts have imposed a
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 24 of 90
Ascertainability ■ Spivey ■ 10512
new requirement that plaintiffs prove at the certification stage that
there is a ‘reliable and administratively feasible’ way to identify all who
fall within the class definition” and rejecting it because “[n]othing in
Rule 23 mentions or implies this heightened requirement under Rule
23(b)(3), which has the effect of skewing the balance that district courts
must strike when deciding whether to certify classes”); see also Daar v.
Yellow Cab Co., 433 P.2d 732, 740 (Cal. 1967) (“Defendant apparently
fails to distinguish between the necessity of establishing the existence
of an ascertainable class and the necessity of identifying the individual
members of such class as a prerequisite to a class suit.”). Thus,
ConAgra’s contention that the administrative feasibility of individually
identifying class members in an “objectively verifiable” manner is a
standalone requirement of the ascertainability doctrine is incorrect.
(Appellant’s Br. 14-23.)
Here, the District Court held that (i) Plaintiffs’ proposed class
definition “identifies putative class members by objective
characteristics; this is the mark of an ascertainable class,” and (ii)
under the circumstances here, class member self-identification through
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 25 of 90
106 ■ Class Actions ■ July 2017 13
a claim form or declaration is a sufficiently reliable method for
determining class membership at the appropriate time, explaining:
While it is true that identifying class members may well require the creation of a claim form or declaration that those asserting membership in the class must submit (likely under penalty of perjury), courts have concluded that such a procedure makes the class ascertainable, at least where the alleged mislabeling occurred throughout the class period, and on a single product or narrow group of products.
(ER00078-ER00079 (citations omitted); see also ER00185.) The district
court’s factual determination regarding ascertainability is entitled to
deference, as ConAgra itself acknowledges in its answering brief in
Jones v. ConAgra. See Jones v. ConAgra Foods, Inc., No. 14-16327,
Appellee’s Answering Brief at 25 (9th Cir. Jan. 21, 2015) (“the district
court’s purely factual (not to mention discretionary) determination” of
the labeling and ingredient facts that relate to ascertainability
“commands this Court’s deference.”); see also Doninger v. Pac. Nw. Bell,
Inc., 564 F.2d 1304, 1309 (9th Cir. 1977) (“[T]he judgment of the trial
court should be given the greatest respect and the broadest discretion
. . . because the district court is in the best position to consider the most
fair and efficient procedure for conducting any given litigation.”)
(quotation and citation omitted).
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 26 of 90
Ascertainability ■ Spivey ■ 10714
Thus, the District Court did not fail to apply the correct legal
standard or fail to consider “administrative feasibility” at all. (Cf.
Appellant’s Br. 3, 8, 12, 18-20, and 23.) The record also shows Plaintiffs
proposed self-identification as a method for identifying class members
at the appropriate time, ConAgra’s contention to the contrary
notwithstanding. (Compare Appellant’s Br. 8, 12, 18, and 23 with
ER06297-ER06298 (Plaintiffs’ reply brief proposing self-identification).)
B. Where the challenged misrepresentation appeared on every unit sold, class member self-identification satisfies ascertainability.
Self-identification is sufficiently reliable (and administratively
feasible) in this case because the “100% Natural” claim appears on the
front label of every unit of Wesson Oils sold during the class period.
The class period is from 2007 through final disposition of this action,
plus or minus 2-3 years to account for differences in applicable statutes
of limitations.
All class members were exposed to the same “100% Natural” label
claim regardless of the size of the bottle, the variety of oil, or where they
made their Wesson Oils purchase. (ER00138.) A person need only
remember and attest that they purchased a “Wesson” branded cooking
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 27 of 90
108 ■ Class Actions ■ July 2017 15
oil since 2007 in one of eleven states to accurately self-identify as a class
member. (ER00184.) Under such circumstances, self-identification
satisfies ascertainability. See Krueger v. Wyeth, Inc., No. 03CV2496,
2015 WL 5839197, at *6 (S.D. Cal. Oct. 7, 2015) (“[C]ourts in this circuit
have found proposed classes ascertainable even when the only way to
determine class membership is with self-identification through
affidavits.”) (emphasis in original, citations omitted); Forcellati v.
Hyland’s, Inc., No. CV 12–1983, 2014 WL 1410264, at *5 (C.D. Cal. Apr.
9, 2014) (holding self-identification proper even when “there are no
records that could confirm class membership”); McCrary v. Elations Co.,
LLC, No. EDCV 13-00242, 2014 WL 1779243, at *8 (C.D. Cal. Jan. 13,
2014) (holding class ascertainable where “class definition clearly
define[d] the characteristics of a class member by providing a
description of the allegedly offending product and the eligible dates of
purchase”); Astiana v. Kashi Co., 291 F.R.D. 493, 500 (S.D. Cal. 2013)
(same); Guido v. L’Oreal, USA, Inc., Nos. CV CV 11–1067, CV 11–5465,
2013 WL 3353857, at *18 (C.D. Cal. July 1, 2013) (same); Ries v.
AriZona Beverages USA LLC, 287 F.R.D. 523, 535 (N.D. Cal. 2012)
(same).
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 28 of 90
Ascertainability ■ Spivey ■ 10916
That all Wesson Oils purchasers were exposed to the same “100%
Natural” claim and need only recall that they purchased Wesson Oils to
self-identify as a class member distinguishes this case from Martin v.
Pacific Parking Sys. Inc., 583 F. App’x 803 (9th Cir. 2014)
(unpublished), on which ConAgra relies. (Appellant’s Br. 15, 18, 21.)
The class definition in Martin featured multiple criteria that could
confound the memory of potential class members. Specifically, the
proposed class in Martin included only individuals who (i) used their
personal, not business, credit cards to pay for parking at defendant’s lot,
(ii) received parking permits with expiration dates printed on them, and
(iii) did not suffer identity theft. The Court noted, “Martin presented
little persuasive, admissible evidence that all of the machines at issue
in this case printed expiration dates uniformly from 2008 to 2011.”
Martin, 583 F. App’x. at 804. In the present case, by contrast,
“[t]hroughout the proposed class period every bottle of Wesson Oil
carried a front label stating that the product was ‘100% Natural’” and
“ConAgra’s conduct was indisputably uniform with respect to all
members of the putative class.” (ER00138; ER00227.) In addition,
here, all Plaintiffs and other class members were harmed because they
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 29 of 90
110 ■ Class Actions ■ July 2017 17
all paid a “price premium” for Wesson Oils caused by ConAgra’s
uniform “100% Natural” labeling claim. (See ER00179-ER00181;
ER00259-ER00260.)
In the cases on which ConAgra relies, variations in the array of
products at issue, variations in labeling during the class period, or the
inability of class members to remember whether they even bought the
product in question—all circumstances that are absent in this case—
rendered self-identification by class members so unreliable as to be
administratively infeasible. See Bruton v. Gerber Products Co., No. 12-
CV-02412, 2014 WL 2860995, at *9 (N.D. Cal. June 23, 2014) (“the fact
that the same products were sold with and without the challenged label
statements simultaneously make Plaintiff’s proposed class
identification method administratively unfeasible”); Jones v. ConAgra
Foods, Inc., No. C 12-01633, 2014 WL 2702726, at *10 (N.D. Cal. June
13, 2014) (defendant sold multiple versions of each of the contested
product labels during the class period, some bearing the allegedly
misleading statements and others not); Astiana v. Ben & Jerry’s
Homemade, Inc., No. C 10-4387, 2014 WL 60097, at *3 (N.D. Cal. Jan.
7, 2014) (no method to determine which ice creams were made with
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 30 of 90
Ascertainability ■ Spivey ■ 11118
synthetic alkali versus natural alkali); In re Clorox Consumer Litig.,
301 F.R.D. 436, 441 (N.D. Cal. 2014) (record demonstrated consumers
could not “necessarily remember when they bought cat litter, or which
sizes, types, or even brands of cat litter they purchased”); see also In re
POM Wonderful, LLC, No. ML10-02199, 2014 WL 1225184, at *6 (C.D.
Cal. Mar. 25, 2014) (noting that “[n]o bottle, label, or package included
any of the alleged misrepresentations”).
In one outlier case on which ConAgra relies, a district court denied
certification on ascertainability grounds because the plaintiff did not
“present any method for determining class membership, let alone an
administratively feasible method,” even though all the nutrition bars at
issue were labeled the same way throughout the class period. See
Sethavanish v. ZonePerfect Nutrition Co., No. 12-2907, 2014 WL
580696, at *6 (N.D. Cal. Feb. 13, 2014). That case depended on
adoption of a “heightened” ascertainability doctrine, which, as explained
below, this Court should not impose.
In another case on which ConAgra relies, the court noted that it
would be difficult for class members to self-identify even when there
was no evidence of labeling variation but also held that “a lack of
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 31 of 90
112 ■ Class Actions ■ July 2017 19
ascertainability alone will not defeat class certification” and that the
challenges entailed in administration were not so burdensome as to
defeat certification. See Algarin v. Maybelline, 300 F.R.D. 444, 456
(S.D. Cal. 2014) (citations omitted).
Here, because every unit of Wesson branded cooking oils sold
contained the misleading “100% Natural” label, the product array,
labeling variation, and memory issues as to specific products that
inhibited ascertainability in other cases are not at issue, thus the class,
defined as all persons who bought Wesson Oils, satisfies
ascertainability. See Jones, 2014 WL 2702726, at *10 (“[T]his Court
might be persuaded that a class of ‘all people who bought Twinkies,’ for
example, during a certain period, could be ascertained—one would at
least have more confidence in class members' ability to accurately self-
identify[.]”).
Additionally, this is not a case where the class definition depends
on an individual’s state of mind; an individual simply must have
purchased Wesson Oils during the multi-year class period. See Xavier
v. Philip Morris USA Inc., 787 F. Supp. 2d 1075, 1089 (N.D. Cal. 2011)
(denying certification where “[t]he question” of class membership
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 32 of 90
Ascertainability ■ Spivey ■ 11320
“would come down to the state of mind of the putative class member,
and it would be easy to fade in or out of the class depending on the
outcome.”). In addition, “ConAgra may also be able to test an
individual’s claim that he or she is a class member by comparing
information about the individual’s purchase with information it
maintains concerning the retailers that sold its products during the
class period or other similar information.” (ER00079.)
Moreover, ConAgra’s own marketing expert relied on the ability of
consumers to self-identify as Wesson Oils purchasers by answering the
following online screening question—“Which of the following brands of
vegetable oil have you personally purchased in the past 12 months?”—
as part of his attempt to generate survey results that would aid
ConAgra in this case. (See ER00792-ER00793 and ER00850 (emphasis
in original).) ConAgra cannot legitimately contend self-identification is
only reliable when it serves ConAgra’s own interests.
Given these facts, the District Court did not abuse its discretion in
holding self-identification sufficient here.
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 33 of 90
114 ■ Class Actions ■ July 2017 21
C. Imposing a heightened ascertainability standard would not protect defendants’ due process rights but would imperil the viability of low-dollar value consumer class actions.
Every Circuit that has addressed the basic contours of the
ascertainability doctrine agrees that “objective criteria” or
“definiteness” is the overarching standard. See In re Nexium Antitrust
Litig., 777 F.3d 9, 19 (1st Cir. 2015) (“[T]he definition of the class must
be ‘definite,’ that is, the standards must allow the class members to be
ascertainable . . . The class definition here satisfies these standards by
being defined in terms of purchasers of Nexium during the class
period”) (citations omitted); Brecher v. Republic of Argentina, 806 F.3d
22, 24-25 (2d Cir. 2015) (“A class is ascertainable when defined by
objective criteria that are administratively feasible and when
identifying its members would not require a mini-hearing on the merits
of each case.”) (citation and quotation omitted); Byrd v. Aaron’s Inc.,
784 F.3d 154, 163 (3d Cir. 2015) (“The ascertainability inquiry is two-
fold, requiring a plaintiff to show that: (1) the class is ‘defined with
reference to objective criteria’; and (2) there is ‘a reliable and
administratively feasible mechanism for determining whether putative
class members fall within the class definition.’”) (internal citations
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 34 of 90
Ascertainability ■ Spivey ■ 11522
omitted); EQT Prod. Co. v. Adair, 764 F.3d 347, 358 (4th Cir. 2014)
(“However phrased, the [ascertainability] requirement is the same.”)
(citations omitted); Frey v. First Nat’l Bank Sw., 602 F. App’x 164, 168
(5th Cir. 2015) (“We have stated that ‘in order to maintain a class
action, the class sought to be represented must be adequately defined
and clearly ascertainable.’”) (citation omitted); Rikos v. Procter &
Gamble Co., 799 F.3d 497, 525 (6th Cir. 2015) (“‘For a class to be
sufficiently defined, the court must be able to resolve the question of
whether class members are included or excluded from the class by
reference to objective criteria.’”) (quoting Young v. Nationwide Mut. Ins.
Co., 693 F.3d 532, 538 (6th Cir. 2012)); Mullins, 795 F.3d at 657 (“We
and other courts have long recognized an implicit requirement under
Rule 23 that a class must be defined clearly and that membership be
defined by objective criteria rather than by, for example, a class
member’s state of mind.”); Busey v. Macon County Greyhound Park,
Inc., 562 F. App’x 782, 787 (11th Cir. 2014) (“An identifiable class exists
if its members can be ascertained by reference to objective criteria. The
analysis of the objective criteria also should be administratively
feasible.”) (citations and quotations omitted).
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 35 of 90
116 ■ Class Actions ■ July 2017 23
Leading treatises and other secondary sources are in accord. See
William B. Rubenstein, Newberg on Class Actions § 3:1 (5th ed. 2015)
(“Although formulations of this doctrine vary, all essentially require a
putative class to be ascertainable with reference to objective criteria.”)
(footnotes omitted); Joseph M. McLaughlin, 1 McLaughlin on Class
Actions § 4:2 (12th ed. 2015) (“[A] long-implicit requirement of Rule 23
and due process is that plaintiffs propose a class that is presently
ascertainable based on objective criteria that do not require the court to
delve into the merits of the claims.”); Charles Alan Wright et al., 7A
Fed. Prac. & Proc. Civ. § 1760 (3d ed. 2015) (“[T]he class does not have
to be so ascertainable that every potential member can be identified at
the commencement of the action . . . [but] the class description [must be]
sufficiently definite so that it is administratively feasible for the court
to determine whether a particular individual is a member.”); Jerold S.
Solovy et al., 5-23 Moore’s Federal Practice - Civil § 23.21 (2015) (“For a
class to be sufficiently defined, the court must be able to resolve the
question of whether class members are included or excluded from the
class by reference to objective criteria.”); MANUAL FOR COMPLEX
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 36 of 90
Ascertainability ■ Spivey ■ 11724
LITIGATION (FOURTH) § 21.222 (2004) (“The definition must be precise,
objective, and presently ascertainable.”).
Where courts diverge is on the narrow question of whether a class
defined as all purchasers of a particular low-value consumer product
within a defined period of time is “ascertainable” when it is unlikely
that there are written records that identify the purchasers, leaving self-
identification through claim forms or declarations as the most practical
way of determining the identities of individual class members.
Under the so-called “Carrera rule,” self-identification in such
circumstances does not satisfy ascertainability. See Carrera v. Bayer
Corp., 727 F.3d 300, 306-308 (3d Cir. 2013); see also McCrary, 2014 WL
1779243, at *8 (“It appears that pursuant to Carrera in any case where
the consumer does not have a verifiable record of its purchase, such as a
receipt, and the manufacturer or seller does not keep a record of buyers,
Carrera prohibits certification of the class.”).
However, the “Carrera rule” is contrary to better-reasoned
opinions that hold self-identification, under appropriate circumstances,
satisfies ascertainability. See Mullins, 795 F.3d at 658 (“The policy
concerns motivating the heightened ascertainability requirement are
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 37 of 90
118 ■ Class Actions ■ July 2017 25
better addressed by applying carefully the explicit requirements of Rule
23(a) and especially (b)(3). These existing requirements already
address the balance of interests that Rule 23 is designed to protect.”);
Rikos, 799 F.3d at 525 (holding there is “no reason to follow Carrera,
particularly given the strong criticism it has attracted from other
courts”); see also Byrd, 784 F.3d at 172 (Rendell, J., concurring) (“[T]he
time has come to do away with this newly created aspect of Rule 23 in
the Third Circuit. Our heightened ascertainability requirement defies
clarification. Additionally, it narrows the availability of class actions in
a way that the drafters of Rule 23 could not have intended.”).
Indeed, the ultimate resolution of Carrera itself effectively
repudiated the Third Circuit’s holding: the district certified a
settlement class under Rule 23(b)(3), defined as all purchasers of the
product in the State of Florida, in which class members could receive a
cash payment from the settlement fund by submitting a claim form
with no additional proof of purchase. See Carrera v. Bayer, No. 2:08-cv-
04716, Final Approval Order and Judgment (Document 147) (D.N.J.
Apr. 27, 2015).
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 38 of 90
Ascertainability ■ Spivey ■ 11926
The overwhelming majority of district judges within this Circuit
that have confronted the issue have rejected the Carrera rule. See, e.g.,
Krueger, 2015 WL 5839197, at *7 (Houston, J.); In re NJOY Consumer
Class Action Litig., No. CV 14-00428, 2015 WL 4881091, at * 24 (C.D.
Cal. Aug. 14, 2015) (Morrow, J.); Rahman v. Mott’s LLP, No. 13-cv-
03482, 2014 WL 6815779, at *10 (N.D. Cal. Dec. 3, 2014) (Illston, J.)
(“In light of the precedent set by many other district courts in this
Circuit, the Court declines to follow Carrera.”); Lilly v. Jamba Juice Co.,
308 F.R.D. 231, 240 (N.D. Cal. 2014) (Tigar, J.); Forcellati, 2014 WL
1410264, at *5 (King, J.); Brazil v. Dole Packaged Foods, LLC, No. 12-
CV-01831, 2014 WL 2466559, at *6 (N.D. Cal. May 30, 2014) (Koh, J.);
but see Sethavanish, 2014 WL 580696, at *6 (Conti, J.).
The reasons for rejecting the Carrera rule are sound. As the
Seventh Circuit explained in Mullins, none of the policy concerns
described in Carrera warrant imposing a heightened bar to class
actions. Mullins, 795 F.3d at 663-672.
First, “the due process question is not whether the identity of class
members can be ascertained with perfect accuracy at the certification
stage but whether the defendant will receive a fair opportunity to
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 39 of 90
120 ■ Class Actions ■ July 2017 27
present its defenses when putative class members actually come
forward.” Mullins, 795 F.3d at 670. While a defendant may argue that
challenging the claims of individual class members can be cost-
prohibitive, “[a] district court can tailor fair verification procedures to
the particular case, and a defendant may need to decide how much it
wants to invest in litigating individual claims.” Id. at 670.
Moreover, there is no bona fide reason to deem class member
affidavits as insufficient evidentiary proof of product purchase given
their regular use as sometimes the only proof in other significant court
proceedings:
If not disputed, self-serving affidavits can support a defendant’s motion for summary judgment, for example, and defendants surely will be entitled to a fair opportunity to challenge self-serving affidavits from plaintiffs. We are aware of only one type of case in American law where the testimony of one witness is legally insufficient to prove a fact. See U.S. Const., Art. III, § 3 (“No person shall be convicted of treason unless on the testimony of two witnesses to the same overt act, or on confession in open court.”). There is no good reason to extend that rule to consumer class actions.
Mullins, 795 F.3d at 669; see also Central District of California L.R. 7-6
(establishing default rule that “declarations” “alone” are used to
establish “[f]actual contentions involved in any motion and opposition to
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 40 of 90
Ascertainability ■ Spivey ■ 12128
motions”). ConAgra’s demand that it must be able to cost-effectively
cross-examine each individual who claims class membership to ask
them whether they really purchased Wesson Oils turns the entire
purpose of class litigation on its head.
ConAgra’s additional argument that its due process rights are
threatened by the use of affidavits in this case because “the preclusive
effect of final judgment would be easy to evade” is baseless.
(Appellant’s Br. 14.) Because the class here is defined by objective
criteria (the purchase of Wesson Oils in certain states during a certain
time period), claim preclusion can be determined by reference to the
facts alleged in the complaint in any future action. If some hypothetical
future complaint challenges a “natural” claim on the label of a Wesson
cooking oil, purchased during the class period and in one of the eleven
states, that claim would be precluded by a class judgment in this case.
It does not matter whether that future plaintiff was specifically
identified as a class member in this litigation earlier. See Geoffrey C.
Shaw, Note, Class Ascertainability, 124 Yale L.J. 2354, 2375-78 (2015);
see also Daar, 433 P.2d at 740 (“If the existence of an ascertainable
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 41 of 90
122 ■ Class Actions ■ July 2017 29
class has been shown, there is no need to identify its individual
members in order to bind all members by the judgment.”).
The Carrera rule also has the “effect of barring class actions
where class treatment is often most needed: in cases involving
relatively low-cost goods or services, where consumers are unlikely to
have documentary proof of purchase.” Mullins, 795 F.3d at 658
(quoting Amchem Products, Inc. v. Windsor, 521 U.S. 591, 617 (1997)).
“These are cases where the class device is often essential ‘to overcome
the problem that small recoveries do not provide the incentive for any
individual to bring a solo action prosecuting his or her rights.’” Id.
(citation omitted); see also Astiana v. Kashi Co., 291 F.R.D. at 500 (“If
class actions could be defeated because membership was difficult to
ascertain at the class certification stage, there would be no such thing
as a consumer class action.”) (citation and internal quotation omitted);
McCrary, 2014 WL 1779243, at *8 (holding that the effect of Carrera is
to “eviscerate[] low purchase price consumer class actions.”); Lilly, 308
F.R.D. at 238 (“Adopting the Carrera approach would have significant
negative ramifications for the ability to obtain redress for consumer
injuries.”).
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 42 of 90
Ascertainability ■ Spivey ■ 12330
“Given the significant harm caused by immunizing corporate
misconduct,” the Seventh Circuit reasoned that “a district judge has
discretion to allow class members to identify themselves with their own
testimony and to establish mechanisms to test those affidavits as
needed.” Mullins, 795 F.3d at 669.
Additionally, ConAgra cannot (and does not) claim it has a due
process interest in how damages are distributed amongst class
members or that the number of class members who self-identify makes
a difference to ConAgra’s aggregate liability (two policy arguments set
forth in Carrera), as neither argument is valid. See Hilao v. Estate of
Marcos, 103 F.3d 767, 786 (9th Cir. 1996) (noting that a defendant’s
interest is “only in the total amount of damages for which it will be
liable,” not “the identities of those receiving damage awards”);
Forcellati, 2014 WL 1410264, at *6 (“because Defendants’ liability will
be determined in the aggregate, and they will have no claim to any
leftover damages, whether any given individual is or is not a rightful
class member is entirely immaterial to Defendants’ monetary liability
in this case.”); cf. Carrera, 727 F.3d at 309-310.
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 43 of 90
124 ■ Class Actions ■ July 2017 31
The “no receipt, no recovery” effect of the so-called “Carrera rule”
does not protect any legitimate due process right, is fatal to low-dollar
value consumer class actions, and should be rejected.
III. The District Court correctly determined that Plaintiffs satisfied Rule 23(a)(3)’s typicality requirement and Rule 23(b)(3)’s predominance and superiority requirements.
The District Court did not abuse its discretion in concluding that
Plaintiffs satisfied Rule 23’s typicality, predominance, and superiority
requirements. (Cf. Appellant’s Br. 8-10.)
First, Plaintiffs satisfied the typicality requirement by presenting
evidence that the class representatives’ claims against ConAgra arise
out of the exact same course of conduct as the claims of the other class
members they seek to represent. (ER00188-ER00191.)
Second, Plaintiffs satisfied Rule 23(b)(3)’s predominance
requirement by establishing that the main liability issues for all claims
certified, including materiality and proximate cause, can be determined
using the classwide evidence reviewed by the District Court. (ER00199-
ER00270.)
Third, Plaintiffs satisfied Rule 23(b)(3)’s superiority requirement
by demonstrating that class adjudication is superior to the other
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 44 of 90
Ascertainability ■ Spivey ■ 12532
available alternatives of either no litigation at all, or millions of
individual adjudications in small-claims courts around the country.
(ER00270-ER00273.)
A. The class representatives’ claims are typical of other class members’ claims because they all arise from ConAgra’s common course of conduct.
Rule 23(a)(3) requires that “the claims or defenses of the
representative parties are typical of the claims or defenses of the class.”
Fed R. Civ. P. 23(a)(3). Typicality is a “permissive” requirement,
Hanlon v. Chrysler Corp., 150 F.3d 1011, 1020 (9th Cir. 1998), designed
to “assure that the interest of the named representative aligns with the
interests of the class.” Wolin, 617 F.3d at 1174-75 (quoting Hanon v.
Dataproducts Corp., 976 F.2d 497, 508 (9th Cir. 1992). Accordingly, the
typicality inquiry asks courts to consider whether class members “have
the same or similar injury, whether the action is based on conduct
which is not unique to the named plaintiffs, and whether other class
members have been injured by the same course of conduct.” Id. at
1175. That is the case here.
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 45 of 90
126 ■ Class Actions ■ July 2017 33
1. Like the class representatives, all class members purchased Wesson Oils and paid more for them as a result of the “100% Natural” label.
Here, the class representatives’ claims are identical to the claims
of the other class members. All of these consumer protection, breach of
warranty, and unjust enrichment claims arise from ConAgra’s false and
deceptive labeling of Wesson Oils as “100% Natural,” even though they
are made from GMO ingredients. The class representatives and the
other class members were all injured in the same way: they all paid
more for Wesson Oils because ConAgra’s “100% Natural” label created a
retail price premium.
Plaintiffs established typicality through evidence. Each bottle of
Wesson Oil “carried a front label stating that the product was ‘100%
Natural.’” (ER00138.) The material inside each bottle of Wesson Oil is
made from GMO crops. (ER05921; ER06026-ER06027; ER06086.)
Common evidence shows whether “there is a price premium associated
with the ‘100% Natural’ label on Wesson Oils.” (ER00268-ER00269.)
Under these circumstances, the District Court correctly concluded
that the class representatives’ claims and injuries are typical of the
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 46 of 90
Ascertainability ■ Spivey ■ 12734
claims and injuries of the other class members they seek to represent.
(ER00084-ER00089; ER00188-ER00191.)
2. Varying understandings of the term “100% Natural” among class members are irrelevant to typicality.
ConAgra argues that the District Court abused its discretion by
purportedly ignoring “substantial evidence” that some class members
“will not share the class representatives’ experiences” of being “deceived
by the “100% Natural” as meaning the product was not made from
GMOs. (Appellant’s Br. 24, 25.)
ConAgra is wrong. ConAgra’s “substantial evidence” on this issue
was a survey conducted for this litigation by Dr. Dominique Hanssens,
purporting to show that the “100% Natural” claim has no statistically
measurable impact on consumers purchasing Wesson Oil. (Appellant’s
Br. 24.) The District Court did not “ignore” the Hanssens survey, but
instead, considered it within the total mix of evidence, including Dr.
Howlett’s direct criticism of it, contrary third-party surveys, and
ConAgra’s own marketing research documents “demonstrat[ing] that
pure and natural claims play a significant role in consumer purchasing
decisions.” (ER00068-ER00069; ER00086; ER00190.)
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 47 of 90
128 ■ Class Actions ■ July 2017 35
That some consumers might not have interpreted the “100%
Natural” claim to mean “no-GMOs” does not destroy typicality for the
specific claims certified by the District Court, as Rule 23 typicality does
not depend on whether the class representatives and class members
shared the same subjective experience. See, e.g., Estrella v. Freedom
Fin. Network, LLC, No. C 09-03156, 2010 WL 2231790, at *10 (N.D.
Cal. June 2, 2010) (“[B]ecause plaintiffs’ allegations are amenable to
proving reliance on a classwide basis, and because [defendant] does not
argue that any of its customers received disclosure statements that
were materially different from those received by the named plaintiffs,
[the class representatives] are typical of the proposed class.”); Mailloux
v. Arrow Fin. Servs., LLC, 204 F.R.D. 38, 42 (E.D.N.Y. 2001) (“Since the
‘least sophisticated consumer’ standard is an objective standard, . . . the
court will not have to make an individual determination of each
potential class member’s subjective understanding of the letter in
question. . . . Consequently, the typicality requirement has been
satisfied.”); Halperin v. Nichols, Safina, Lerner & Co., No. 94 C 6960,
1996 WL 634037, at *6 (N.D. Ill. Oct. 29, 1996) (holding
representative’s “subjective understanding” had “nothing whatever to
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 48 of 90
Ascertainability ■ Spivey ■ 12936
do with the typicality of his claim.”). Typicality is not destroyed simply
because people are different from one another in legally irrelevant
ways.
3. That some class representatives might consider purchasing Wesson Oils in the future if ConAgra makes true disclosures is irrelevant to typicality.
Nine of the class representatives submitted declarations stating
that if ConAgra stopped falsely claiming Wesson Oils were “100%
Natural,” then they would consider purchasing Wesson Oils in the
future. (ER00167; ER07869-ER07910.) The class representatives also
continue to buy all manner of food products other than Wesson Oils.
ConAgra contends that some of these other products were labeled
“natural” but may also have been made from GMO ingredients.
(Appellant’s Br. 26.) ConAgra contends that these two facts render the
class representatives atypical and subject to unique defenses. (Id.)
Again, typicality only requires that the class representatives
suffered similar injuries (here, payment of a price premium) from the
same course of conduct (here, ConAgra’s false “100% Natural” labels on
Wesson Oils) as the other class members. Whether the class
representatives purchased other foods containing GMOs, or would
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 49 of 90
130 ■ Class Actions ■ July 2017 37
consider purchasing Wesson Oils should ConAgra change its conduct, is
not relevant to Rule 23 typicality. See Astiana v. Kashi, 291 F.R.D. at
503 (noting representative’s purchase of “unhealthy or otherwise
artificial foods says nothing about whether they purchased Kashi
products specifically because they were supposedly healthy and
natural.”) (citation omitted); Red v. Kraft Foods, Inc., No. CV 10-1028,
2012 WL 8019257, at *12 n.17 (C.D. Cal. Apr. 12, 2012) (holding that
whether plaintiffs “purchase and consume other foods with the same
ingredients they allege to be unhealthy in this suit . . . is wholly
irrelevant.”).
Neither do the class representatives’ eating habits impact
materiality. The relevant materiality inquiry is whether ConAgra’s
claim that Wesson Oils are “100% Natural” was material to the
reasonable consumer, not to particular individuals. As the District
Court explained in its extensive analysis of Stearns v. Ticketmaster
Corp., 655 F.3d 1013 (9th Cir. 2011), the class representatives will
establish reliance and causation through classwide evidence, not
individualized evidence. Id. at 1022 (citation omitted). (ER00254-
ER00257.)
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 50 of 90
Ascertainability ■ Spivey ■ 13138
Accordingly, the class representatives are not subject to unique
defenses and satisfy typicality.
B. Plaintiffs’ claims are susceptible to classwide proof, satisfying Rule 23’s predominance requirement.
Rule 23(b)(3) requires that “questions of law or fact common to
class members predominate over any questions affecting only individual
members.” Fed. R. Civ. P. 23(b)(3). “Common issues predominate over
individual issues when the common issues ‘represent a significant
aspect of the case and they can be resolved for all members of the class
in a single adjudication.’” Edwards v. First Am. Corp., 798 F.3d 1172,
1182 (9th Cir. 2015) (citation omitted); see also Hanlon, 150 F.3d at
1022 (“In contrast to Rule 23(a)(2), Rule 23(b)(3) focuses on the
relationship between the common and individual issues.”). The mere
existence of individualized questions does not preclude a predominance
finding. See Edwards, 798 F.3d at 1183-84; Amgen Inc. v. Conn. Ret.
Plans & Trust Funds, 133 S. Ct. 1184, 1196 (2013); Butler v. Sears,
Roebuck & Co., 727 F.3d 796, 801 (7th Cir. 2013), cert denied, 134 S. Ct.
1277 (2014) (predominance is not determined “simply by counting
noses: that is, determining whether there are more common issues or
more individual issues”).
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 51 of 90
132 ■ Class Actions ■ July 2017 39
“Considering whether ‘questions of law or fact common to class
members predominate’ begins, of course, with the elements of the
underlying cause[s] of action.” Erica P. John Fund, Inc. v. Halliburton
Co., 563 U.S. 804, 131 S. Ct. 2179, 2184 (2011). Plaintiffs here moved to
certify eleven single-state classes asserting 34 separate causes of action.
(ER07770-ER07771.) The District Court held that for all 22 causes of
action certified, common questions susceptible to common proof
predominated over any individualized questions. Those common
questions include: (i) whether the “100% Natural” claim was false or
misleading to a reasonable person; (ii) whether the claim is material to
a reasonable consumer, which would entitle Plaintiffs to a classwide
inference of reliance and causation; and (iii) whether the claim created
a price premium all class members had to pay, thus establishing injury
and damages. (ER00201-00250; see also ER00254 (“Because plaintiffs
have adduced sufficient evidence that the ‘100% Natural’ label is
material to a reasonable consumer and that the consumer would
understand it to mean, inter alia, that a product labeled in this fashion
contains no GMOs, the court concludes that materiality can be proved
on a classwide basis.” (footnote omitted).) Accordingly, the District
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 52 of 90
Ascertainability ■ Spivey ■ 13340
Court was within its discretion holding that common questions of law
and fact predominate.
1. Whether the “100% Natural” claim on Wesson Oils was false, misleading, and material to a reasonable consumer are inherently common, classwide questions.
Here, the overarching predominant liability question for all of the
certified claims is whether ConAgra’s “100% Natural” claim on a food
product made from GMOs is objectively true, as ConAgra claims, or
false, deceptive, or misleading, as Plaintiffs claim, and the District
Court would have been within its discretion to certify classes in order to
determine the answer to this “falsity” question on its own. See Amgen,
133 S.Ct. at 1196 (holding Rule 23(b)(3) “does not require a plaintiff
seeking class certification to prove that each element of her claim is
susceptible to classwide proof.”) (emphasis in original).
ConAgra contends that, regardless of this common “falsity”
question, individualized questions would overwhelm because proof of
individual reliance and/or causation is required, and thus the certified
claims turn on whether each and every person who purchased Wesson
Oils (i.e., each individual member of the classes certified by the District
Court) was individually “misled by the natural label in their individual
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 53 of 90
134 ■ Class Actions ■ July 2017 41
purchasing decisions.” (Appellant’s Br. 28.) ConAgra’s statement of the
underlying law is incorrect. As this Court recently confirmed:
To state a claim under the UCL or the FAL “based on false advertising or promotional practices, it is necessary only to show that members of the public are likely to be deceived.” This inquiry does not require “individualized proof of deception, reliance and injury.” “[I]n effect, California has created what amounts to a conclusive presumption that when a defendant puts out tainted bait and a person sees it and bites, the defendant has caused an injury; restitution is the remedy.”
Pulaski & Middleman, LLC v. Google, Inc., 802 F.3d 979, 985-86 (9th
Cir. 2015) (citations and footnotes omitted).
ConAgra’s attempt to graft an individualized inquiry onto the
materiality determination is contrary to law. See Edwards v. Ford
Motor Co., 603 F. App’x 538, 541 (9th Cir. 2015) (“Because materiality
is governed by an objective ‘reasonable person’ standard under
California law, an inquiry that is the same for every class member, a
finding that the defendant has failed to disclose information that would
have been material to a reasonable person who purchased the
defendant’s product gives rise to a rebuttable inference of reliance as to
the class.”) (citation omitted); Hinojos v. Kohl’s Corp., 718 F.3d 1098,
1107 (9th Cir.2013) (“representation is ‘material’ . . . if a reasonable
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 54 of 90
Ascertainability ■ Spivey ■ 13542
consumer would attach importance to it or if ‘the maker of the
representation knows or has reason to know that its recipient regards
or is likely to regard the matter as important in determining his choice
of action.’”) (citation omitted); Stearns, 655 F.3d at 1022 (holding that
materiality is established “if a reasonable man would attach importance
to its existence or nonexistence in determining his choice of action in
the transaction in question”) (citation omitted); Yokoyama v. Midland
Nat’l Life Ins. Co., 594 F.3d 1087, 1092-93 (9th Cir. 2010) (concluding
that the district court erred in denying class certification when it held
Hawaii statute required individualized showings of reliance because
Hawaii “look[ed] to a reasonable consumer, not the particular
consumer” and thus the liability portion would be uniform, as it “will
focus on the standardized written material given to all plaintiffs to
determine whether those materials are likely to mislead consumers
acting reasonably under the circumstances.” ); In re Tobacco II Cases,
207 P.3d 20, 29-30 (Cal. 2009) (“to state a claim under either the UCL
or the false advertising law . . . it is necessary only to show that
members of the public are likely to be deceived.”).
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 55 of 90
136 ■ Class Actions ■ July 2017 43
ConAgra incorrectly asserts that the District Court “treated
materiality as a uniform issue for all nine classes without addressing
any state law differences[.]” (Appellant’s Br. 29 n.5.) The District
Court, however, analyzed the elements of each particular state law
claim and only certified claims where the “reasonable consumer”
standard applied. (ER00201-ER00250.) Furthermore, ConAgra failed
to argue in its opening brief how differences between the certified
California consumer protection claims and any of the other certified
claims made a difference in the predominance analysis, thereby waiving
any argument that differences between the claims certified make a
difference in this appeal. See Int’l Union of Bricklayers & Allied
Craftsman Local Union No. 20, AFL-CIO v. Martin Jaska, Inc., 752
F.2d 1401, 1404 (9th Cir. 1985) (holding that the Court “will not
ordinarily consider matters on appeal that are not specifically and
distinctly raised and argued in appellant’s opening
brief”) (citation omitted).
ConAgra’s argument that materiality is an individualized inquiry
is premised on a flawed reading of this Court’s decision in Stearns.
Most fundamentally, ConAgra ignores that, in Stearns, this Court
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 56 of 90
Ascertainability ■ Spivey ■ 13744
remanded the California UCL claim to the district court for
reconsideration because “individualized proof of reliance and causation”
was not required. Stearns, 655 F.3d. at 1020 (“[R]elief under the UCL
is available without individualized proof of deception, reliance and
injury. Thus, the district court’s concerns about reliance and causation
were not well taken.”) (internal citations and quotations omitted).
As the District Court recognized, Stearns also turned on the
plaintiffs’ failure to adduce any evidence regarding materiality. Here,
by contrast, Plaintiffs did adduce evidence (including third-party
surveys and ConAgra’s internal marketing materials) sufficient to
demonstrate that Plaintiffs can prove materiality on a classwide basis.
(ER00254-ER00257; see also ER00086 (“Plaintiffs proffer documents
detailing the results of ConAgra’s marketing research; they contend
that this research demonstrates that pure and natural claims play a
significant role in consumer purchasing decisions. Because the
documents were filed under seal, the court does not detail the findings
here. It concurs, however, in plaintiffs’ description of the documents.”)
(footnotes omitted).)
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 57 of 90
138 ■ Class Actions ■ July 2017 45
Thus, as the District Court properly held, “Plaintiffs need not
prove at this stage that every ConAgra customer would find the ‘100%
Natural’ claim material or would believe that it meant the products
contained no GMOs.” (ER00252.) “Rather, they need only demonstrate
that a reasonable consumer would understand it that way and find it
material.” (Id.)
2. ConAgra’s demand that this Court reweigh common evidence and find Plaintiffs failed to demonstrate materiality goes far beyond the probing of merits allowed at the class certification stage.
A district court’s inquiry on class certification should not devolve
into a mini-trial on the merits of the case. Amgen, 133 S. Ct. at 1194-95
(Rule 23 “grants courts no license to engage in free-ranging merits
inquiries at the certification stage,” instead “[m]erits questions may be
considered … only to the extent … relevant to determining whether the
Rule 23 prerequisites for class certification are satisfied.”).
As this Court recently confirmed, it is reversible error when a
district court “evaluate[s] the merits rather than focusing on whether
the questions presented—meritorious or not—were common to the
class.” Alcantar v. Hobart Serv., 800 F.3d 1047, 1053 (9th Cir. 2015);
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 58 of 90
Ascertainability ■ Spivey ■ 13946
see also Ellis v. Costco Wholesale Corp., 657 F.3d 970, 983 n.8 (9th Cir.
2011).
Thus, “whether class members could actually prevail on the merits
of their claims” is not a proper inquiry in determining the preliminary
question “whether common questions exist.” Ellis, 657 F.3d at 983 n.8.
“To hold otherwise would turn class certification into a mini-trial,” id.,
when the purpose of class certification is merely “to select the ‘metho[d]’
best suited to adjudication of the controversy ‘fairly and efficiently’.”
Amgen, 133 S. Ct. at 1191 (alteration in original).
Despite these clear guidelines, ConAgra asks this Court to second-
guess the District Court and conduct exactly the type of mini-trial that
is inappropriate at the class certification stage. (See Appellant’s Br. 32-
45.) In doing so, ConAgra misstates both the proper scope of the class
certification inquiry and the substance of the District Court’s
materiality findings.
The District Court did not make “a finding of classwide
materiality” or “h[o]ld that Plaintiffs were entitled to the classwide
inference of materiality.” (Appellant’s Br. 31.) Rather, the District
Court held that Plaintiffs adduced sufficient evidence to show that
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 59 of 90
140 ■ Class Actions ■ July 2017 47
materiality is susceptible to classwide proof. (See, e.g., ER00236 (“The
court considers below whether plaintiffs have adduced sufficient
evidence that the “100% Natural” claim was material such that it is
appropriate to certify the class because plaintiffs may be able to prove
that ‘a reasonable consumer could have been misled’ by the label
claim.”); ER00247 (“[T]he court concludes that plaintiffs may be able to
show on a classwide basis that the ‘100% Natural’ label had a common
meaning that was material to members of the putative class.”).)
Whether the evidence Plaintiffs adduced is sufficient to ultimately
prevail on the materiality issue is irrelevant at the class certification
stage. The law has long been clear that “neither the possibility that a
plaintiff will be unable to prove his allegations, nor the possibility that
the later course of the suit might unforeseeably prove the original
decision to certify the class wrong, is a basis for declining to certify a
class which apparently satisfies [Rule 23].” Blackie v. Barrack, 524
F.2d 891, 901 (9th Cir. 1975). Similarly:
Rule 23(b)(3) requires a showing that questions common to the class predominate, not that those questions will be answered, on the merits, in favor of the class. Because materiality is judged according to an objective standard, the materiality of Amgen’s alleged misrepresentations and omissions is a question common to all members of the class
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 60 of 90
Ascertainability ■ Spivey ■ 14148
… As to materiality, therefore, the class is entirely cohesive: It will prevail or fail in unison. In no event will the individual circumstances of particular class members bear on the inquiry … [T]he office of a Rule 23(b)(3) certification ruling is not to adjudicate the case[.]
Amgen, 133 S. Ct. at 1191 (citations omitted); see also id. at 1204 (“just
as a plaintiff class’s inability to prove materiality creates no risk that
individual questions will predominate, so even a definitive rebuttal on
the issue of materiality would not undermine the predominance of
questions common to the class.”).
Here, the District Court properly avoided transforming class
certification into a mini-trial on whether “100% Natural” is material to
a reasonable consumer and did not abuse its discretion in finding that
the evidence adduced was sufficient to show that materiality is
susceptible of classwide proof.
3. Classwide evidence suggests ConAgra’s “100% Natural” claim on Wesson Oils is material to a reasonable consumer.
Even if the ultimate materiality of the “100% Natural” claim to a
reasonable consumer were a relevant class certification question—
which, as a merits question, it is not—ConAgra distorts the evidence on
this issue.
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 61 of 90
142 ■ Class Actions ■ July 2017 49
Classwide proof demonstrating that the “100% Natural” claim on
Wesson Oils is material to a reasonable consumer, and that reasonable
consumers interpret that “natural” claim to mean “no-GMOs,” is
actually available and the District Court considered it in rendering its
certification decision. This common proof includes:
• Dr. Charles Benbrook’s expert opinion testimony that “foods
that contain genetically engineered ingredients are not
natural” (ER00057);
• ConAgra’s internal marketing research showing that
“consumers exposed to a ‘100% Natural’ or ‘Natural’ claim on
ConAgra product labels generally consider the
representation a significant factor in their purchasing
decisions.” (ER00250; see also ER00086, ER00190, and
ER05257-ER5260); and,
• Third-party surveys that “tend to show that, however they
interpret it, consumers find the “100% Natural claim
material to their purchasing decisions” and that “consumers
generally understand it, inter alia, as a representation that
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 62 of 90
Ascertainability ■ Spivey ■ 14350
Wesson Oils do not contain GMOs.” (ER00251-ER00252; see
also ER05261-ER05262.)
ConAgra’s attack on this common evidence has nothing to do with
whether common questions predominate over individualized questions.
ConAgra contends that this evidence is not sufficiently connected to
Wesson Oils, but, instead, only has to do with unrelated products.
(Appellant’s Br. 32-38.) ConAgra is wrong.
The third-party surveys specifically pertained to natural claims on
food products and the relationship between natural claims and GMOs.
(ER00250-ER00251; see also ER00256 (“[P]laintiffs have adduced
substantial evidence that a ‘100% Natural’ claim on a food product is
material to consumers; the industry studies and surveys they proffer
indicate that a majority of consumers consider the claim material to
their purchasing decision, and that consumers of Wesson Oils
understand that “100% Natural” means, inter alia, that Wesson Oils do
not contain GMOs.”).) The internal ConAgra documents concerned the
materiality of “natural” claims to consumers of Wesson Oils specifically.
(ER00250.)
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 63 of 90
144 ■ Class Actions ■ July 2017 51
ConAgra’s contention that Plaintiffs failed to provide any survey
evidence “tied to consumers of Wesson Oil” because the District Court
excluded an additional survey (the “Kozup survey”) on procedural
grounds is also incorrect. (See Appellant’s Br. 2, 9, 25, 28, 33-39.) The
purpose of the Kozup survey was not to establish materiality; rather, it
was to establish actual falsity of the “100% Natural” claim on the
merits. (ER05262-ER05263.) Further, Plaintiffs submitted the Kozup
survey in response to ConAgra’s survey (the “Hanssens survey”), which
purported to show that consumers were not misled by the “100%
Natural” claim, notwithstanding Plaintiffs’ assertion, then and now,
that the Hanssens survey is not relevant to class certification. Indeed,
that ConAgra seeks to rely on survey evidence confirms that the key
issues in this case can be proven or disproven by classwide evidence.
See Shari S. Diamond, Reference Guide on Survey Research, in
REFERENCE MANUAL ON SCIENTIFIC EVIDENCE 361 (Federal Judicial
Center 3d ed. 2011), available at 2011 WL 7724258, at * 2 (“When
properly designed, executed, and described, surveys (1) economically
present the characteristics of a large group of respondents or other
units and (2) permit an assessment of the extent to which the measured
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 64 of 90
Ascertainability ■ Spivey ■ 14552
respondents or other units are likely to adequately represent a relevant
group of individuals or other units.”).
ConAgra’s contention that the District Court ignored the
Hanssens survey is also inaccurate. (Appellant Br. 41-45.) As already
noted, the District Court specifically considered the Hanssens survey,
as well as Dr. Howlett’s criticisms of it, third-party surveys, and
ConAgra’s own contradictory internal market research showing that the
“100% Natural” claim on Wesson Oils is a significant driver of sales.
(ER00189-190; see also ER00286-ER00289; ER05258-ER05260.)
ConAgra’s claim that the District Court abused its discretion by failing
to consider the Hanssens survey is simply ConAgra’s demand that this
Court review and reweigh the factual evidence, disregard Plaintiffs’
evidence, and overrule the District Court’s factual determination that
materiality can be determined by classwide evidence.
The cases on which ConAgra relies provide no support for its
extraordinary request. For example, in Kosta v. Del Monte Foods, Inc.,
class certification was denied in a food labeling case where the expert
opined on the materiality of food labels generally, with no focus on the
label claims at issue in the case. See Kosta v. Del Monte Foods, Inc.,
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 65 of 90
146 ■ Class Actions ■ July 2017 53
308 F.R.D. 217, 229-30 (N.D. Cal. 2015) (expert offered opinion on
materiality of food labels generally without any focus on the particular
“antioxidant,” “natural source,” “no artificial flavors and preservatives,”
“must be refrigerated,” and “fresh” label claims at issue). Here, in
contrast, the evidence deals directly with “natural” claims, “natural”
claims on ConAgra products, and whether consumers interpret
“natural” claims to mean “no-GMOs.”
Other cases on which ConAgra relies are wholly unrelated to
consumer claims or the relevance of survey evidence. See EQT Prod.
Co. v. Adair, 764 F.3d at 366-67 (remanding to determine whether
defendants’ common practices were relevant to alleged coalbed methane
gas royalty underpayments); Forrand v. Federal Exp. Corp., No. CV 08-
1360, 2013 WL 1793951, at *3 (C.D. Cal. Apr. 15, 2013) (denying
certification of claims that required individualized worker-by-worker
inquiries to determine whether workers were in fact unpaid for on-the-
clock work).
The trademark, patent, and Lanham Act cases on which ConAgra
relies are even farther afield, having nothing to do with either the type
of consumer survey evidence at issue here (regarding labeling terms
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 66 of 90
Ascertainability ■ Spivey ■ 14754
and purchase intent) or the relevant class certification predominance
inquiry of whether common evidence can answer whether the “100%
Natural” claim is material to the reasonable consumer. See Universal
City Studios, Inc. v. Nintendo Co., Ltd., 746 F.2d 112, 118-19 (2d Cir.
1984) (holding survey failed to demonstrate arcade owner confusion
between King Kong and Donkey Kong due to poor design); Harold’s
Stores, Inc. v. Dillard Dep’t Stores, Inc., 82 F.3d 1533, 1546 (10th Cir.
1996) (affirming admission of survey regarding “whether and to what
degree Dillard’s infringement and sale of Harold’s copyrighted fabric
designs damaged Harold’s future clothing sales”); Exxon Corp. v. Texas
Motor Exch. of Houston, Inc., 628 F.2d 500, 507 (5th Cir. 1980)
(trademark case involving survey evidence regarding confusion between
competing marks); Fancaster, Inc. v. Comcast Corp., 832 F. Supp. 2d
380 (D.N.J. 2011) (same); THOIP v. Walt Disney Co., 690 F. Supp. 2d
218, 236-37 (S.D.N.Y. 2010) (excluding survey regarding consumer
confusion over mark on t-shirts where survey failed to replicate manner
in which consumers encountered products in the marketplace); Water
Pik, Inc. v. Med-Systems, Inc., 726 F.3d 1136, 1145 (10th Cir. 2013)
(affirming exclusion of survey regarding likelihood of confusion between
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 67 of 90
148 ■ Class Actions ■ July 2017 55
marks on sinus irrigation products); Fractus, S.A. v. Samsung, No. 6:09-
cv-203, 2011 WL 7563820, at *1 (E.D. Tex. Apr. 29, 2011) (excluding
surveys that “do not measure the value of Plaintiff’s technology, but
merely measure the perceived consumer value of cell phones with any
internal antennas”); Wells Fargo & Co. v. WhenU.com, Inc., 293 F.
Supp. 2d 734, 752-53 (E.D. Mich. 2003) (excluding survey concerning
likelihood of confusion over different pop-ads from prior cases involving
completely different claims and parties); Pizza Hut, Inc. v. Papa John’s
Int’l, 227 F.3d 489, 504 (5th Cir. 2000) (reversing denial of summary
judgment for defendant where plaintiff failed to introduce any evidence
demonstrating that the “Better Ingredients. Better Pizza” slogan “had
the tendency to deceive consumers”).
Plaintiffs’ survey evidence demonstrates that between two-thirds
and one half of all consumers consider “natural” claims on food labels
“material” to their purchasing decisions and believe, among other
things, that “natural” means “no-GMOs.” (Appellant’s Br. 40-41.) As
the District Court acknowledged, other courts have found statements
material to the reasonable consumer even when they deceive as few as
20% of survey respondents. (ER00252 (citing Oshana v. Coca-Cola Co.,
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 68 of 90
Ascertainability ■ Spivey ■ 14956
No. 04 C 3596, 2005 WL 1661999, at *9 (N.D. Ill. July 13, 2005) (“Coca-
Cola provides no authority that a misrepresentation is immaterial if
only 24% of consumers would behave differently . . . [T]here is sufficient
evidence to raise a genuine issue of fact as to whether the alleged
misrepresentations are material to a reasonable consumer.”)); see also
ER05261.)
None of the cases ConAgra cites require 100% agreement in
surveys in order for a misleading claim to be legally material to the
reasonable consumer and each of those cases involves a factual scenario
distinguishable from the present case. In re Celexa v. Lexapro Mktg. &
Sales Practices Litig., 291 F.R.D. 13, 20 (D. Mass. 2013) (denying
certification where “plaintiff-specific inquiries would need to be made to
satisfy the exposure requirement”); In re Vioxx Class Cases, 180 Cal.
App. 4th 116, 134 (2009) (materiality individualized because “patient-
specific factors are part of the prescribing decision”); In re Countrywide
Fin. Corp. Mortg. & Sales Practices Litig., Nos. 08md1988, 10cv0257,
2011 WL 6325877, at *2 (S.D. Cal. Dec. 16, 2011) (denying certification
where challenged transactions involved individualized transactions);
Fairbanks v. Farmers New World Life Ins. Co., 197 Cal. App. 4th 544,
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 69 of 90
150 ■ Class Actions ■ July 2017 57
564 (2011) (affirming denial of certification “in the absence of a common
marketing scheme”); Strawn v. Farmers Ins. Co. of Oregon, 258 P.3d
1199, 1215 (Or. 2011) (affirming denial of defendant’s motion for
directed verdict in class trial because “a jury could infer from evidence
common to the class that the individual class members relied on
[defendant’s] misrepresentation”); Henry Schein, Inc. v. Stromboe, 102
S.W.3d 675, 684 (Tex. 2003) (no classwide proof because “there was no
evidence that all purchasers relied on the same or even similar
representations” by defendant).
Finally, ConAgra argues that the “natural” label claim is
susceptible to “so many different meanings” as to preclude
predominance. (Appellant’s Br. 44.) The District Court previously held
that there is “a common sense definition of ‘natural’—i.e., existing in
nature, and nothing artificial or synthetic.” (ER00057.) Furthermore,
Plaintiffs submitted evidence showing how the bioengineering process
and the GMOs resulting therefrom are not natural, let alone “100%
Natural.” (See ER00061-ER00063; ER05253; see also ER07799-
ER07830 (expert declaration describing how and why genetically
engineered crops, and the foods made from them, cannot be considered
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 70 of 90
Ascertainability ■ Spivey ■ 15158
natural).) The District Court further found that Plaintiffs “proffered
sufficient evidence that reasonable consumers associate the claim
[“100% Natural”], with the fact that the products contain no GMOs,”
and explicitly distinguished this case from precedent on which ConAgra
relies. (ER00254 (distinguishing Jones v. ConAgra, 2014 WL 2702726,
on basis of evidence presented by Plaintiffs here).) Thus, ConAgra’s
arguments do not defeat predominance but rather go to the ultimate
merits issue in this case concerning whether Wesson Oils are, or are
not, “100% Natural” because they are made from GMOs.
For these reasons, the District Court acted within its discretion
when it held the materiality of a “natural” claim to a reasonable
consumer predominates and is not overwhelmed by any individualized
questions.
C. The class action mechanism is superior to any other available method for adjudicating Plaintiffs’ claims concerning a uniformly labeled, low-dollar value consumer product.
The District Court properly held that a class action is the superior
method for adjudicating this controversy. (ER00273.)
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 71 of 90
152 ■ Class Actions ■ July 2017 59
1. Without the class action device, no individual purchaser of Wesson Oils could reasonably afford to seek a legal remedy.
The Supreme Court has declared that the “policy at the very core
of the class action mechanism is to overcome the problem that small
recoveries do not provide the incentive for any individual to bring a solo
action prosecuting his or her rights.” Amchem, 521 U.S. at 617 (citation
omitted). But for class actions, groups of harmed individuals “would be
without effective strength to bring their opponents into court at all.” Id.
Given the “low average price of a bottle of Wesson Oil,” each class
member’s individual maximum recovery in this case is relatively small.
(ER00271; see also ER00102 (showing average retail cooking oil prices
generally in the $3-5 range).) As courts have recognized, the “realistic
alternative to a class action is not [millions of] individual suits, but zero
individual suits, as only a lunatic or a fanatic sues for $30.” Carnegie v.
Household Int’l, Inc., 376 F.3d 656, 661 (7th Cir. 2004). Accordingly, “a
class action has to be unwieldy indeed before it can be pronounced an
inferior alternative—no matter how massive the fraud or other
wrongdoing that will go unpunished if class treatment is denied—to no
litigation at all.” Id.
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 72 of 90
Ascertainability ■ Spivey ■ 15360
2. There is no reason that the factual issues in this case could not be adjudicated in a single trial.
ConAgra identifies no reason the District Court could not
adjudicate the disputed factual issues in a single trial for all eleven
certified classes. Whether ConAgra’s claim was false, misleading, and
material to the reasonable consumer are questions that will be resolved
by reference to survey evidence, expert testimony, and ConAgra’s own
records. All of this evidence is common to all class members. Similarly,
each class’s damages are based on the same damages methodology.
There is no reason these issues need to be separately determined eleven
times.
3. Eleven trials are superior to separate individual trials for each of the tens of millions of Wesson Oils purchasers.
The District Court found, and ConAgra does not now dispute, that
three of the four Rule 23(b)(3) factors favor a superiority finding.
(ER00270-ER00271.) ConAgra contends the District Court erred in
assessing the “likely difficulties in managing a class action,” because
each certified class needs its own merits trial. (Appellant’s Br. 58.)
District Courts have “a number of management tools” at their
disposal to address manageability concerns. In re Visa
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 73 of 90
154 ■ Class Actions ■ July 2017 61
Check/Mastermoney Antitrust Litig. v. Visa, U.S.A., Inc., 280 F.3d 124,
140-42 (2d Cir. 2001) (enumerating five tools and noting that
manageability determinations are “particularly within” a trial court’s
discretion). Courts must consider whether a class action will create
“relatively more management problems than any of the alternatives
(including, most notably, [hundreds of thousands of] separate lawsuits
by the class members).” Klay v. Humana, Inc., 382 F.3d 1241, 1272-76
(11th Cir. 2004) (“[W]here a court has already made a finding that
common issues predominate over individualized issues, we would be
hard pressed to conclude that a class action is less manageable than
individual actions.”).
Even if eleven trials or a bellwether approach were necessary,
which is not the case, the specter of eleven trials does not threaten to
consume judicial resources, especially when weighed against the
alternative of millions of lawsuits brought by individual consumers.
Indeed, class actions were designed for a case like this, where a
defendant, through a uniform course of conduct, harmed millions of
individual consumers in relatively small amounts. Because the damage
inflicted on any individual consumer does not justify the cost of an
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 74 of 90
Ascertainability ■ Spivey ■ 15562
individual lawsuit, the group’s collective rights will be vindicated only if
consumers can band together through a class action. On these facts,
the District Court properly determined that a class action is superior to
ConAgra’s demand for a free pass for its bad behavior.
IV. Plaintiffs’ damages methodology satisfies Comcast.
In Comcast, the Supreme Court held that district courts must
determine whether a class plaintiff’s damages model can translate the
“legal theory of the harmful event into an analysis of the economic
impact of that event.” Comcast, 133 S. Ct. at 1435 (citation and
emphasis omitted). The Court reasoned that “a model purporting to
serve as evidence of damages in [a] class action must measure only
those damages attributable to that theory.” Id. at 1433. Here, the
District Court properly concluded that Plaintiffs’ damages methodology
met this test.
Plaintiffs’ liability theory is that ConAgra misled the class
representatives specifically, and reasonable consumers generally, into
believing that Wesson Oils are “100% Natural,” when, in fact, Wesson
Oils are not “100% Natural” because they are made from GMO
ingredients. (See, e.g., ER00260, ER05249-ER05252, ER05305,
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 75 of 90
156 ■ Class Actions ■ July 2017 63
ER05866-ER05867.) ConAgra harmed the class representatives and
the other class members by causing them to pay more than they
otherwise would have had to pay for Wesson Oils (i.e., charging them a
“price premium”) and by failing to provide them with the benefit of their
bargain: a cooking oil that was, in fact, “100% Natural.” (ER00180-
ER00181; ER00307-ER00311; ER05251-ER05252; ER05309-ER05314.)
Plaintiffs proposed, and the District Court accepted, a damages
methodology that reliably measures the amount of price premium
specifically attributable to ConAgra misleading class members into
believing Wesson Oils were not made from GMOs by using the “100%
Natural” claim. (ER00270.) Under this methodology, the ultimate
amount of money Plaintiffs would ask a jury to award can be calculated
using three different numbers:
(a) The total amount the class members paid at retail,
in dollars, for Wesson Oils (an amount well in excess of $1
billion over the course of the class period (ER05253));
(b) The percentage “price premium” caused by the “100%
Natural” label claim on Wesson Oils as determined by
hedonic regression analysis of actual historical retail prices
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 76 of 90
Ascertainability ■ Spivey ■ 15764
and actual product attributes (including “natural” label
claims) for thousands of different shortening/oil products
sold in the United States (including Wesson Oils), that
Plaintiffs preliminarily calculated as 2.28% (ER00147-
ER00149, ER05020-ER05024); and
(c) The percentage relative value (sometimes referred to as
a “partworth”) of the “no-GMO” understanding of a
“natural” label claim on cooking oils, to be determined by
an analysis of choice-based conjoint survey responses.
(ER05129-ER05138.)
(See also ER05314 (summarizing Plaintiffs’ methodology).) Stated as a
formula, Plaintiffs’ damages methodology is:
(a) * (b) * (c) = Total $ amount of damages or restitution1
1 Certified Indiana, New York, and Oregon consumer protection claims also support an award of statutory damages on a per-violation basis and no specific methodology is required to calculate such damages. At trial, Plaintiffs intend to seek all applicable statutory damages.
In addition, Plaintiffs argued in the District Court, and preserve for appeal following final judgment, that ConAgra is liable for the entirety of the “100% Natural” price premium. (ER05306-ER5309.) The District Court disagreed with Plaintiffs’ arguments, holding that the “natural” price premium was not sufficiently tied to Plaintiffs’ liability theory under Comcast because it did “not isolate the price premium
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 77 of 90
158 ■ Class Actions ■ July 2017 65
ConAgra contends that Plaintiffs’ damages methodology fails to
comply with Comcast because “it is unable to isolate the price premium
associated with the believed absence of GMOs.” (Appellant’s Br. 9.)
ConAgra further contends that the District Court “failed to conduct the
rigorous analysis necessary[.]” (Appellant’s Br. 45.) ConAgra is wrong
on both counts. Plaintiffs’ damages methodology isolates the amount of
economic harm attributable to the false “no-GMO” interpretation of
ConAgra’s “100% Natural” claim and the District Court rigorously
analyzed Plaintiffs’ proposed methodology. (ER00145-ER00164 and
ER00257-ER00270.)
While ConAgra quibbles with certain minor aspects of Plaintiffs’
proposed hedonic regression and conjoint analysis methodologies,
ConAgra’s real concern appears to be that the two methods have not
previously been combined “in order to assign a price premium to a sub-
feature.” (Appellant’s Br. 52.) However, none of the “natural” labeling
decisions since Comcast required a damages model to isolate that
attributable to consumers’ belief that ConAgra’s products did not contain GMOs,” but when the “natural” price premium was subdivided by multiplying it with the “no-GMO” relative importance of the natural claim, the damages model did sufficiently isolate the price premium attributable to consumers’ no-GMO belief. (ER00259.)
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 78 of 90
Ascertainability ■ Spivey ■ 15966
portion of the “natural” label price premium solely attributable to the
ingredient or process that made the natural representation false, so the
lack of case law on the subject is unsurprising. (See ER05306-ER5309.)
Moreover, that a methodology has not previously been used is not a
reason to reject it. See Daubert v. Merrell Dow Pharm., Inc., 509 U.S.
579, 593 (1993) (holding austere “general acceptance” standard
incompatible with permissive backdrop of Federal Rules of Evidence
and noting that publication “is not the sine qua non of admissibility; it
does not necessarily correlate with reliability, and in some instances
well-grounded but innovative theories will not have been published”)
(internal citations omitted). Multiplying three numbers—a dollar
value, a “natural claim” price premium percentage, and a “no-GMO”
interpretation relative value percentage—to output a final classwide
dollar damages figure, as Plaintiffs and their experts proposed, is a
proper damages methodology for this case, especially considering the
lack of substantive challenge to any one of these three figures on its
own.
Furthermore, despite ConAgra’s contention that Plaintiffs’
damages methodology “would always yield a fictional result that could
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 79 of 90
160 ■ Class Actions ■ July 2017 67
not be accurate as to any one class member,” (Appellant’s Br. 55), it is
well established in this Circuit—and in other Circuits across the United
States—that “individualized damages calculations alone cannot defeat
class certification,” Pulaski & Middleman, LLC, 802 F.3d at 987; see
also Mullins, 795 F.3d at 671 (“the need for individual damages
determinations at this later stage of the litigation does not itself justify
the denial of certification.”); Allapattah Servs. v. Exxon Corp., 333 F.3d
1248, 1261 (11th Cir. 2003) (“numerous courts have recognized that the
presence of individualized damages issues does not prevent a finding
that the common issues in the case predominate”), aff'd, 545 U.S. 546
(2005). A valid methodology for calculating damages need not yield a
result that is precise to the penny. See Story Parchment Co. v.
Paterson Parchment Paper Co., 282 U.S. 555, 563 (1931); see also
Marsu, B.V. v Walt Disney Co., 185 F.3d 932, 939 (9th Cir. 1999) (“The
fact that the amount of damage may not be susceptible of exact proof or
may be uncertain, contingent, or difficult of ascertainment does not bar
recovery.”).
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 80 of 90
Ascertainability ■ Spivey ■ 16168
Because Plaintiffs’ damages methodology translates their theory
of liability into its economic impact, the District Court properly
concluded Plaintiffs’ methodology satisfies the Comcast standard.
A. Plaintiffs’ proposed hedonic regression methodology reliably isolates the percentage “price premium” specifically and solely attributable to the presence of a “natural” label claim on cooking oils.
As ConAgra itself acknowledges, hedonic regression and conjoint
analysis are both well-accepted methodologies used in economics and
marketing research. (Appellant’s Br. 49-50.) ConAgra further concedes
that Mr. Weir’s hedonic regression methodology calculates the
“percentage price premium” of cooking oil “attributable to the natural
label” from actual, historical price and attribute data. (Appellant’s Br.
49.)
ConAgra suggests in a footnote that Mr. Weir’s regression model
is insufficient for “other reasons identified by other courts” that have
rejected other regression models ConAgra claims to be “almost
identical” to Mr. Weir’s, citing Brazil v. Dole Packaged Foods, LLC, No.
12-CV-01831, 2014 WL 5794873, at *11-14 (N.D. Cal. Nov. 6, 2014).
ConAgra is incorrect. Mr. Weir’s regression model is not “almost
identical” to the rejected regression proposed by Dr. Oral Capps in
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 81 of 90
162 ■ Class Actions ■ July 2017 69
Brazil. Unlike Dr. Capps, Mr. Weir controlled for price and the
presence or absence of various label claims (including the “natural”
label claim at issue here) by collecting data from Nielsen and IRI data
sets obtained through discovery and inputting that data into his
hedonic regression model. (ER05004-ER05010.) Also, unlike Dr.
Capps, Mr. Weir did not depend on a single instance of Internet
research to determine which products contained a “natural” label claim
and which did not; rather, Mr. Weir relied on Nielsen’s comprehensive
collection of data. (ER05008-ER05009.) ConAgra’s vague suggestion
that Mr. Weir’s regression is “insufficient for other reasons” is no basis
for reversing the District Court’s careful determination of the propriety
of Mr. Weir’s analysis and methodology. (ER00145-ER00152.)
B. Analysis of choice-based conjoint survey results reliably isolates the percentage “relative importance” of the “no-GMO” portion of a “natural” claim.
ConAgra also concedes that Dr. Howlett’s conjoint analysis is
capable of calculating a percentage representing the “relative
importance” of the “absence of GMO” portion of a “natural” claim and
that Dr. Howlett “provided a detailed explanation of conjoint analysis”
to the District Court. (Appellant’s Br. 50, 56.)
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 82 of 90
Ascertainability ■ Spivey ■ 16370
ConAgra nonetheless contends that it “is problematic” that any
conjoint analysis performed today would calculate the value of a “no-
GMO” interpretation of a natural claim “at the time of the surveys” and
that “consumers’ understanding and valuation of natural claims and
the importance of GMOs are not static over time.” (Appellant’s Br. 55.)
But conjoint analysis is routinely accepted as a proper methodology for
calculating past harm. See Sanchez-Knutson v. Ford Motor Co., No. 14-
61344, 2015 WL 6395040, at *7 (S.D. Fla. Oct. 6, 2015) (rejecting as
‘unfounded” argument “that conjoint analysis, an analytic survey
method used to measure customer preferences for specific features of
products, is an improper damages theory post-Comcast”); Guido, 2014
WL 6603730, *4-14 (holding that plaintiff's expert’s proposed conjoint
analysis damages theory was not junk science, could be applied on a
classwide basis for predominance purposes under Comcast, and was
consistent with plaintiff’s theory of liability); Khoday v. Symantec
Corp., 93 F. Supp. 3d 1067, 1082 (D. Minn. 2015) (“The Court finds that
Gaskin’s conjoint analysis is generally a permissible method for
calculating damages.”). Moreover, Mr. Weir, who is also an expert in
conjoint analysis, rebutted ConAgra’s arguments on this point, and
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 83 of 90
164 ■ Class Actions ■ July 2017 71
explained how survey data from the present is routinely used to assess
the past. (ER05655-ER05656.)
C. Combining the percentage yielded by hedonic regression with the percentage yielded by conjoint analysis is proper.
Unable to show how or why either hedonic regression or conjoint
analysis is insufficient on its own, ConAgra instead contends that the
District Court did not rigorously analyze whether it could multiply the
two resulting percentages together with the total paid by consumers for
Wesson Oils to support its holding that “[s]uch a calculation would
necessarily produce a damage figure attributable solely to ConAgra’s
alleged misconduct – i.e., misleading consumers to believe that Wesson
Oils contain no GMOs by placing a ‘100% Natural’ label on the
products.” (ER00260.)
As Plaintiffs acknowledge, “natural” on a food product means not
only “no GMOs,” but also “no artificial flavors,” “no artificial colors,” and
“no artificial preservatives.” (ER05132-ER05133.) As Plaintiffs further
acknowledge, their case theory is that Wesson Oils are not “100%
Natural” because they are made from GMO ingredients. (ER00013-
ER00017.) Accordingly, complying with the directives set forth in the
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 84 of 90
Ascertainability ■ Spivey ■ 16572
District Court’s August 1, 2014 Order, Plaintiffs proposed a conjoint
analysis methodology that would segregate the relative value of the “no-
GMO” interpretation of “natural” from the relative value of all the other
possible interpretations of what “natural” means on a bottle of cooking
oil. (ER05129-ER05138; ER05312-ER05314.) As ConAgra concedes,
conjoint analysis is well suited to this task. (Appellant Br. 50, 56.)
ConAgra argues that “Dr. Howlett provided no evidence that it
was common or accepted to combine” the output of the hedonic
regression analysis (a percentage price premium for a natural claim)
and the output of the conjoint analysis (a percentage relative
importance of a “no-GMO” understanding of a natural claim) in the
manner proposed, that is, through multiplication. (Appellant’s Br. 52.)
But there was and is no need to present extensive proof that two
percentages of the type at issue can be combined through multiplication
in order to factor both results into a logical conclusion. See In re
Prempro Prods. Liab. Litig., 514 F.3d 825, 831 (8th Cir.2008) (allowing
expert testimony that was “an exercise in basic math using simple
deductive reasoning”). Dr. Howlett opined how this combination
isolates the specific portion of the “natural” price premium on Wesson
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 85 of 90
166 ■ Class Actions ■ July 2017 73
Oils specifically attributable to consumer understanding that “natural”
means “no-GMO.” (See ER05138.) As Mr. Weir explained, “Dr.
Howlett’s proposed conjoint methodology, which is a standard
application of basic, time-tested conjoint methodology, would function
no differently if the starting price premium were calculated using
hedonic regression, another conjoint methodology, any other empirical
technique, or a value not calculated in any way, such as a market price
premium agreed upon by the parties, or prescribed by the court.”
(ER05655.)
Thus, combining the hedonic regression result and the conjoint
analysis result does not mix “apples and oranges,” as ConAgra
contends, (Appellant’s Br. 53), or improperly “translate” relative value
into price premium. (Appellant’s Br. 54-55.) Instead, because these
results are bare numbers with no physical dimension, combining them
with multiplication is mathematically proper. Furthermore, the
hedonic regression price premium and conjoint analysis relative value
are both determined specifically with respect to Wesson Oils and the
issues in this case. Thus, Plaintiffs’ combination does not use “magic[]”
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 86 of 90
Ascertainability ■ Spivey ■ 16774
to arrive at the necessary price premium, (Appellant’s Br. 55), it uses
arithmetic, and use of arithmetic is not an abuse of discretion.
To the extent ConAgra argues that the District Court abused its
discretion by failing to “rigorously analyze” the combination of the
hedonic regression and conjoint analysis methodologies in order to
determine whether it accords with Comcast, the District Court’s multi-
page analysis of exactly this topic, in which it carefully considered all of
ConAgra’s substantive arguments and case law belies this argument.
(ER00260-ER0270.)
Where plaintiffs are “deceived by misrepresentations into making
a purchase, the economic harm is the same: the consumer has
purchased a product that he or she paid more for than he or she
otherwise might have been willing to pay if the product had been
labeled accurately.” Kwikset Corp. v. Superior Court, 246 P.3d 877, 890
(Cal. 2011) (emphasis in original). By combining the results of the
hedonic regression and conjoint analysis, Plaintiffs present a damages
methodology that measures only those damages attributable to
ConAgra’s misleading labeling of GMO-containing Wesson Oils as
“100% Natural.
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 87 of 90
168 ■ Class Actions ■ July 2017 75
CONCLUSION
For the foregoing reasons, the District Court’s order granting in
part Plaintiffs’ amended class certification motion, certifying eleven
state classes of purchasers of Wesson brand cooking oils, and certifying
22 claims for classwide resolution, should be affirmed.
STATEMENT OF RELATED CASES
Pursuant to Fed. R. App. P. 28 and Ninth Circuit Rule 28-2.6,
Respondents are not aware of any related cases pending in this Court
that have not already been identified by Appellant.
Dated: December 18, 2015
Respecftully submitted, MILBERG LLP /s/ Ariana J. Tadler Ariana J. Tadler David E. Azar Henry J. Kelston Meagan Keenan Carey Alexander
Respectfully submitted, GRANT & EISENHOFER P.A. /s/ Adam J. Levitt Adam J. Levitt Mary S. Thomas Edmund S. Aronowitz
Class Counsel for Plaintiffs-Appellees Robert Briseño et al.
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 88 of 90
Ascertainability ■ Spivey ■ 16976
CERTIFICATE OF COMPLIANCE
1. This brief complies with the type-volume limitation of Federal
Rule of Appellate Procedure 32(a)(7)(B) because it contains 13,885
words, excluding the parts of the brief exempted by Federal Rule of
Appellate Procedure 32(a)(7)(B)(iii).
2. This brief complies with the typeface and type styles limitations
of Federal Rules of Appellate Procedure 32(a)(5) and 32(a)(6) because it
has been prepared in proportionally-spaced typeface using 14-point
Century font in Microsoft Word.
Dated: December 18, 2015 /s/ Edmund S. Aronowitz Edmund S. Aronowitz Counsel for Plaintiffs-Appellees
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 89 of 90
170 ■ Class Actions ■ July 2017 77
CERTIFICATE OF SERVICE
I hereby certify that on December 18, 2015, I electronically filed
the foregoing with the Clerk of the Court for the United States Court of
Appeals for the Ninth Circuit using the appellate CM/ECF system. All
participants in the case are registered CM/ECF users and will be served
by the system.
Dated: December 18, 2015 /s/ Edmund S. Aronowitz Edmund S. Aronowitz Counsel for Plaintiffs-Appellees
Case: 15-55727, 12/18/2015, ID: 9799073, DktEntry: 24, Page 90 of 90
Ascertainability ■ Spivey ■ 171
IV. Briseño v. Conagra Foods, Inc., U.S. Court of Appeals for the Ninth Circuit. Appellant’s Reply BriefNo. 15-55727
In the
United States Court of Appeals For the Ninth Circuit
__________________
ROBERT BRISEÑO, Plaintiff-Appellee,
v.
CONAGRA FOODS, INC., Defendant-Appellant._________________
ON APPEAL FROM THE UNITED STATES DISTRICT COURT FOR THE CENTRAL DISTRICT OF CALIFORNIA
(The Honorable Margaret M. Morrow) Case No. 2:11-cv-05379
__________________
APPELLANT’S REPLY BRIEF__________________
A. Brooks GreshamLaura E. Coombe MCGUIREWOODS LLP 1800 Century Park East, 8th Floor Los Angeles, CA 90067 Telephone: (310) 315-8291
Angela M. SpiveyMCGUIREWOODS LLP 1230 Peachtree St, NE, Suite 2100 Atlanta, GA 30309 Telephone: (404) 443-5720
R. Trent Taylor MCGUIREWOODS LLP Gateway Plaza 800 E. Canal St. Richmond, VA 23219 Telephone: (804) 775-1182
E. Rebecca Gantt MCGUIREWOODS LLP World Trade Center 101 W. Main St., Suite 9000 Norfolk, VA 23510 Telephone: (757) 640-3731
Counsel for Defendant-Appellant ConAgra Foods, Inc.
Case: 15-55727, 03/02/2016, ID: 9887081, DktEntry: 34, Page 1 of 38
172 ■ Class Actions ■ July 2017i
TABLE OF CONTENTS
Page
TABLE OF AUTHORITIES ................................................................................... iii
INTRODUCTION ..................................................................................................... 1
ARGUMENT ............................................................................................................. 2
I. Plaintiffs’ claim that they need not demonstrate administrative feasibility is wrong and belied by the authorities they cite. ................................................. 2
A. A class is not ascertainable if potential class members cannot be identified in an administratively feasible fashion. ................................ 3
B. Plaintiffs fail to rebut ConAgra’s argument that they have not proposed an administratively feasible plan for identifying potential class members. ....................................................................................... 4
C. Plaintiffs’ discussion of the Carrera rule and the ability to self-identify through affidavits is a red herring. ........................................... 7
II. The district court erred in holding that common questions predominate over individual ones. ................................................................................................ 9
A. Individual issues predominate with respect to materiality because Plaintiffs did not establish that it is susceptible to classwide proof. ..... 9
1. Plaintiffs and the district court rely on a patchwork of unpersuasive and ultimately inadmissible evidence that lacks a sufficient nexus to Plaintiffs’ theory of liability while overlooking contrary, probative, and admissible evidence. ..... 11
2. The district court and Plaintiffs ignored probative, admissible evidence demonstrating that materiality is not susceptible to classwide proof. ........................................................................ 14
3. The issue of materiality and reliance varies among consumers in this case because there is no uniform understanding of “natural” or “GMO.” ................................................................. 16
Case: 15-55727, 03/02/2016, ID: 9887081, DktEntry: 34, Page 2 of 38
Ascertainability ■ Spivey ■ 173ii
4. ConAgra does not ask this Court to unnecessarily delve into the merits of this case but to engage in the rigorous analysis required by Rule 23. .................................................................. 17
B. Individual issues predominate with respect to damages. .................... 18
1. Plaintiffs’ argument that its hybrid damages model meets Comcast because “use of arithmetic is not an abuse of discretion” is unsupported. ........................................................ 18
2. Plaintiffs’ insistence that the two damage models can be combined simply defies logic. .................................................. 23
3. Plaintiffs’ reliance on Guido and related cases is misplaced. ... 25
III. This Court should remand with instructions to dismiss Plaintiffs’ amended complaint with prejudice. .................................................... 27
CONCLUSION ........................................................................................................ 29
CERTIFICATE OF COMPLIANCE ....................................................................... 30
CERTIFICATE OF SERVICE ................................................................................ 31
Case: 15-55727, 03/02/2016, ID: 9887081, DktEntry: 34, Page 3 of 38
174 ■ Class Actions ■ July 2017 iii
TABLE OF AUTHORITIES Page(s)
Cases
Allen v. ConAgra Foods, Inc.,No. 13-cv-01279, ECF No. 150 at 1 (N.D. Cal. Jan. 9, 2015) ............................. 5
Apple, Inc. v. Samsung Electronics Co.,No. 11-cv-01846, 2014 WL 976898 (N.D. Cal. Mar. 6, 2014) .......................... 25
Brecher v. Republic of Arg.,806 F.3d 22 (2d Cir. 2015) ................................................................................... 3
Briseno v. ConAgra Foods, Inc.,No. 11-cv-05379, ECF No. 1 (C.D. Cal. June 28, 2011) ................................... 28
Byrd v. Aaron’s Inc.,784 F.3d 154 (3d Cir. 2015) ................................................................................. 3
Carijano v. Occidental Petroleum Corp.,643 F.3d 1216 (9th Cir. 2011) ............................................................................ 23
Chapman v. Maytag Corp.,297 F.3d 682 (7th Cir. 2002) .............................................................................. 22
Clausen v. M/V New Carissa,339 F.3d 1049 (9th Cir. 2003) ................................................................ 20, 21, 23
In re Clorox Consumer Litig.,301 F.R.D. 436 (N.D. Cal. 2014) .......................................................................... 4
Comcast v. Behrend,133 S. Ct. 1426 (2013) .................................................................................passim
Destfino v. Reiswig,630 F.3d 952 (9th Cir. 2011) ........................................................................ 28, 29
In re Dial Complete Mktg. and Sales Practice Litig.,312 F.R.D. 36 (D.N.H. 2015) ............................................................................. 21
Domingo v. T.K.,289 F.3d 600 (9th Cir. 2002) .............................................................................. 19
Case: 15-55727, 03/02/2016, ID: 9887081, DktEntry: 34, Page 4 of 38
Ascertainability ■ Spivey ■ 175iv
EQT Prod. Co. v. Adair,764 F.3d 347 (4th Cir. 2014) ................................................................................ 3
Faulk v. Sears Roebuck & Co.,No. 11-cv-02159, 2013 WL 1703378 (N.D. Cal. Apr. 19, 2013) ...................... 10
Guido v. L’Oreal USA, Inc.,No. 11-cv-01067, 2014 WL 6603730 (C.D. Cal. Jul. 1, 2014) .............. 25, 26, 27
Hawkins v. Comparet-Cassani,251 F.3d 1230 (9th Cir. 2001) .......................................................................... 5, 6
Hayes v. Wal-Mart Stores, Inc.,725 F.3d 349 (3d Cir. 2013) ................................................................................. 4
Housing Works, Inc. v. Turner,362 F. Supp. 2d 434 (S.D.N.Y. 2005) ................................................................ 24
Karhu v. Vital Pharm., Inc.,621 F. App’x 945 (11th Cir. 2015) ................................................................... 3, 4
Kosta v. Del Monte Foods, Inc.,308 F.R.D. 217 (N.D. Cal. 2015) ........................................................................ 10
Larin v. Bank of Am., NA,617 F. App’x 651 (9th Cir. 2015) ....................................................................... 28
Lilly v. Jamba Juice Co.,308 F.R.D. 231 (N.D. Cal. 2014) ...................................................................... 4, 5
Lust v. Merrell Dow Pharm., Inc.,89 F.3d 594 (9th Cir. 1996) ................................................................................ 20
Marcus v. BMW of N. Am., LLC,687 F.3d 583 (3d Cir. 2012) ................................................................................. 8
Martin v. Pac. Parking Sys. Inc.,583 F. App’x 803 (9th Cir. 2014) ................................................................. 3, 4, 5
McCanic v. City of Long Beach,993 F.2d 883, 1993 WL 171470 (9th Cir. 1993) ................................................ 28
Case: 15-55727, 03/02/2016, ID: 9887081, DktEntry: 34, Page 5 of 38
176 ■ Class Actions ■ July 2017 v
Meadows v. Anchor Longwall & Rebuild, Inc.,306 F. App’x 781 (3d Cir. 2009) ........................................................................ 19
Mike’s Train House, Inc. v. Lionel, LLC,472 F.3d 398 (6th Cir. 2006) .............................................................................. 22
Miller v. Fuhu, Inc.,No. 14-cv-06119, 2015 WL 7776794 (C.D. Cal. Dec. 1, 2015) ........................ 21
Mullins v. Direct Digital, LLC,795 F.3d 654 (7th Cir. 2015) ............................................................................ 3, 8
In re Mushroom Direct Purchaser Antitrust Litig.,No. 06-cv-00620, 2015 WL 5766929 (E.D. Pa. Aug. 27, 2015) ........................ 19
In re Nexium Antitrust Litig.,777 F.3d 9 (1st Cir. 2015) ..................................................................................... 3
In re NJOY, Inc. Consumer Class Action Litig.,No. 14-cv-00428, 2015 WL 4881091 (C.D. Cal. Aug. 14, 2015) .......... 25, 26, 27
Prism Techs. LLC v. AT&T Mobility, LLC,No. 12-cv-12201, 2014 U.S. Dist. LEXIS 132619 (D. Neb. Sept. 22, 2014) ............................................................................................................. 20
Pulaski & Middleman, LLC v. Google, Inc.,802 F.3d 979 (9th Cir. 2015) .......................................................................... 9, 10
Randolph v. J.M. Smucker Co.,303 F.R.D. 679 (S.D. Fla. 2014) ............................................................. 12, 13, 21
Rikos v. Procter & Gamble Co.,799 F.3d 497 (6th Cir. 2015) ................................................................................ 3
Saavedra v. Eli Lilly & Co.,No. 12-cv-09366, 2014 WL 7338930 (C.D. Cal. Dec. 18, 2014) ...................... 25
Sethavanish v. ZonePerfect Nutrition Co.,No. 12-cv-02907, 2014 WL 580696 (N.D. Cal. Feb. 13, 2014) ....................... 4, 5
Stearns v. Ticketmaster Corp.,655 F.3d 1013 (9th Cir. 2011) ............................................................................ 15
Case: 15-55727, 03/02/2016, ID: 9887081, DktEntry: 34, Page 6 of 38
Ascertainability ■ Spivey ■ 177vi
Steroid Hormone Prod. Cases, 104 Cal. Rptr. 3d 329 (Cal. Ct. App. 2010) ......................................................... 9
Ventress v. Japan Airlines,747 F.3d 716 (9th Cir. 2014) ................................................................................ 8
In re Vioxx Class Cases,180 Cal. App. 4th 116 (2009) ....................................................................... 10, 12
Wal-Mart Stores, Inc. v. Dukes,131 S. Ct. 2541 (2011) .......................................................................................... 2
Young v. Nationwide Mut. Ins. Co.,693 F.3d 532 (6th Cir. 2012) ................................................................................ 3
Zinser v. Accufix Research Inst., Inc.,253 F.3d 1180 (9th Cir. 2001) .............................................................................. 5
Rules
FED. R. CIV. P. 15(a) ................................................................................................. 27
FED. R. CIV. P. 23 ..............................................................................................passim
Other Authorities
Use of the Term “Natural” in the Labeling of Human Food Products: Request for Information and Comments, 80 Fed. Reg. 69,905 (Nov. 12, 2015) .................................................................................................................. 16
U.S. Food & Drug Administration, Guidance for Industry: Voluntary Labeling Indicating Whether Foods Have or Have Not Been Derived from Genetically Engineered Plants (Nov. 2015), available at http://ow.ly/Yxg4Z ............................................................................................. 16, 17
Case: 15-55727, 03/02/2016, ID: 9887081, DktEntry: 34, Page 7 of 38
178 ■ Class Actions ■ July 2017 1
INTRODUCTION
In their Answering Brief, Plaintiffs ask to be exempted from no fewer than
four separate, well-established class-certification requirements. They essentially
request that their multiple attempts to meet these requirements be graded on a
curve and that they be rewarded with certification for their efforts alone. Both they
and the district court apparently believe that the class certification requirements
that have been forged by the U.S. Supreme Court and lower courts over the course
of a number of years are simply too difficult to be met and that the bar should be
lowered for Plaintiffs.
But neither Rule 23 nor the extensive case law interpreting it permit courts
to relax standards on an ad hoc basis. Thus, this Court should reject Plaintiffs’
attempt to end-run around the ascertainability requirement’s mandate that they
submit an administratively feasible plan for identifying potential class members.
Similarly, Plaintiffs cannot show that common issues predominate as to materiality
with generalized survey evidence that is not tied to their precise legal theory.
Moreover, there are not one but two fatal flaws with Plaintiffs’ hybrid damages
model: (1) it fails to meet the requirement that it be scientifically reliable, and (2)
it fails to meet the requirement that it measure “damages attributable to th[e]
theory” of liability underlying the class claims. Comcast v. Behrend, 133 S. Ct.
1426, 1433 (2013). Each of these flaws on its own is sufficient to require reversal.
Case: 15-55727, 03/02/2016, ID: 9887081, DktEntry: 34, Page 8 of 38
Ascertainability ■ Spivey ■ 1792
Nonetheless, the district court succumbed to Plaintiffs’ repeated requests to
relax the class certification standards, reversing course after initially holding
Plaintiffs to these requirements. Indeed, the district court even openly asked at the
hearing “is there any way of doing that” and could a plaintiff “ever show that” in
response to ConAgra counsel’s assertion of what Comcast required Plaintiffs to
show. ER00117.
This Court should not make the same mistake. If the district court’s order is
allowed to stand, it would dramatically lower the bar for class certification in the
Ninth Circuit. Reversal of the district court’s class certification decision is
necessary to ensure that Rule 23’s requirements do not become “mere pleading
standard[s].” Wal-Mart Stores, Inc. v. Dukes, 131 S. Ct. 2541, 2551 (2011).
ARGUMENT
I. Plaintiffs’ claim that they need not demonstrate administrative feasibility is wrong and belied by the authorities they cite.
Plaintiffs argue that a class is ascertainable even if there is no
administratively feasible plan to identify potential class members. Answering Br.
11–12. They also claim that even if administrative feasibility is required, they
have met their burden because potential class members can self-identify through
affidavits. Id. at 14. Plaintiffs are wrong on both counts.
Case: 15-55727, 03/02/2016, ID: 9887081, DktEntry: 34, Page 9 of 38
180 ■ Class Actions ■ July 2017 3
A. A class is not ascertainable if potential class members cannot be identified in an administratively feasible fashion.
Plaintiffs suggest that few courts consider administrative feasibility to be
part of deciding whether a class is ascertainable. Id. at 11–12. That is not true –
the overwhelming majority of federal appellate courts have held that a class is not
ascertainable if members cannot be identified in an administratively feasible
manner. See Byrd v. Aaron’s Inc., 784 F.3d 154, 163 (3d Cir. 2015); Brecher v.
Republic of Arg., 806 F.3d 22, 24 (2d Cir. 2015); Karhu v. Vital Pharm., Inc., 621
F. App’x 945, 947 (11th Cir. 2015); In re Nexium Antitrust Litig., 777 F.3d 9, 19–
20 (1st Cir. 2015); EQT Prod. Co. v. Adair, 764 F.3d 347, 358 (4th Cir. 2014).1
Strangely, Plaintiffs cite these cases in arguing that administrative feasibility is not
part of deciding whether a class is ascertainable. Answering Br. 21–22. That
claim is belied by the fact that all of these courts explicitly referenced
administrative feasibility when deciding whether the class was ascertainable.
This Court has already addressed whether class representatives must submit
an administratively feasible plan for identifying potential class members. See
Martin v. Pac. Parking Sys. Inc., 583 F. App’x 803 (9th Cir. 2014). In Martin, this
1 The Seventh Circuit has taken the contrary position. See Mullins v. Direct Digital, LLC, 795 F.3d 654, 657–58 (7th Cir. 2015). There is also a split in the Sixth Circuit as to whether administrative feasibility is required to show that a class is ascertainable. Compare Rikos v. Procter & Gamble Co., 799 F.3d 497, 525 (6th Cir. 2015), with Young v. Nationwide Mut. Ins. Co., 693 F.3d 532, 537–38 (6th Cir. 2012).
Case: 15-55727, 03/02/2016, ID: 9887081, DktEntry: 34, Page 10 of 38
Ascertainability ■ Spivey ■ 1814
Court held that the class was not ascertainable because the plaintiff “has not
demonstrated that it would be administratively feasible to determine which
individuals used personal, and not business, credit cards to purchase parking.” Id.
at 804 (emphasis added). Plaintiffs have no answer to the point that this Court
explicitly referenced administrative feasibility as an element of deciding whether a
class was ascertainable.
Additionally, numerous district courts in this Circuit have also held that
administrative feasibility is part of deciding whether a class is ascertainable. The
district court in this case is one example. ER00075. Other examples abound. See,
e.g., Lilly v. Jamba Juice Co., 308 F.R.D. 231, 236–37 (N.D. Cal. 2014);
Sethavanish v. ZonePerfect Nutrition Co., No. 12-cv-02907, 2014 WL 580696, at
*4 (N.D. Cal. Feb. 13, 2014); In re Clorox Consumer Litig., 301 F.R.D. 436, 441–
42 (N.D. Cal. 2014); see also Opening Br. 16 & n.2.
B. Plaintiffs fail to rebut ConAgra’s argument that they have not proposed an administratively feasible plan for identifying potential class members.
Because ascertainability is required for class certification, Plaintiffs bear the
burden of establishing this element. Hayes v. Wal-Mart Stores, Inc., 725 F.3d 349,
356 (3d Cir. 2013); Karhu, 621 F. App’x at 948. Indeed, this Court has repeatedly
held that plaintiffs bear the burden of submitting plans or proposals to the district
court that demonstrate a class action is warranted. See Martin, 583 F. App’x at
Case: 15-55727, 03/02/2016, ID: 9887081, DktEntry: 34, Page 11 of 38
182 ■ Class Actions ■ July 2017 5
804; Hawkins v. Comparet-Cassani, 251 F.3d 1230, 1238 (9th Cir. 2001); Zinser v.
Accufix Research Inst., Inc., 253 F.3d 1180, 1189 (9th Cir. 2001). In Martin, this
Court held that the mere suggestion that class members could self-identify through
affidavits, without more, is not an administratively feasible plan. See Martin, 583
F. App’x at 804 (“[The plaintiff] has pointed this court to no place in the record
where he proposed even a general plan” for identifying potential class members
“beyond suggesting that individuals self-identify.”).
District courts in this Circuit have applied this requirement specifically in
the context of ascertainability. In one case, the court certified the proposed class in
part because the plaintiffs submitted a “plan to identify class members.” Lilly, 308
F.R.D. at 239. The court in Lilly thought it was especially important that the
plaintiffs’ plan was “detailed” and “aimed at providing notice to other potential
class members.” Id. at 238–39. By contrast, class certification has been denied in
cases when the class representatives have not presented an administratively
feasible plan for pinpointing class members. See, e.g., Allen v. ConAgra Foods,
Inc., No. 13-cv-01279, ECF No. 150 at 1 (N.D. Cal. Jan. 9, 2015) (holding that the
plaintiff “[did] not satisfy the ascertainability requirement” because she did not
“present a plan for how the class members will be identified”); Sethavanish, 2014
WL 580696, at *6 (similar).
Case: 15-55727, 03/02/2016, ID: 9887081, DktEntry: 34, Page 12 of 38
Ascertainability ■ Spivey ■ 1836
There is no dispute that Plaintiffs have never submitted a plan or proposal
for identifying potential class members. The district court twice noted that
Plaintiffs defended ascertainability solely on the ground that the class was
objectively defined. See ER00182; ER00076. Plaintiffs’ only response on appeal
is to note that the district court speculated about how class members could be
identified in the future. Answering Br. 13. True, the district court said that
“identifying class members may well require the creation of a claim form or
declaration.” ER00078 (emphasis added). But the court’s statement that such
forms may well be required highlights that there is nothing in the record at this
time that resembles an administratively feasible plan for identifying class members.
The Plaintiffs cannot satisfy their burden to submit a plan by relying on the district
court to create one. That would run headfirst into this Court’s holding that “[t]he
district court is not to bear the burden of … correcting Rule 23(a) problems; rather,
the burden is on Plaintiffs to submit proposals to the court.”2 Hawkins, 251 F.3d at
1238 (quotation marks omitted).
2 This reasoning also undermines the district court’s speculation that “ConAgra may [] be able to test an individual’s claim that he or she is a class member” by looking to retail information. ER00079. Although Plaintiffs reference this statement on appeal, Answering Br. 20, this comes from the district court’s original order denying class certification. Plaintiffs never proposed looking to retail records in the district court, and they cannot meet their burden to propose an administratively feasible plan by pointing to indeterminate proposals that the district court offered. See Opening Br. 21.
Case: 15-55727, 03/02/2016, ID: 9887081, DktEntry: 34, Page 13 of 38
184 ■ Class Actions ■ July 2017 7
The district court did not require the Plaintiffs to submit a plan because it
thought that doing so “would effectively prohibit class actions involving low
priced consumer goods.” ER00184. That is not true. See Opening Br. 20–21.
And regardless, concern over a subset of class actions is not a license to
unilaterally adjust burdens of proof. After all, “a party seeking to maintain a class
action must affirmatively demonstrate his compliance with Rule 23.” Comcast,
133 S. Ct. at 1432 (quotation marks omitted; emphasis added). The district court
committed legal error when it failed to hold the Plaintiffs to their burden and
instead lowered the bar for the ascertainability requirement.
C. Plaintiffs’ discussion of the Carrera rule and the ability to self-identify through affidavits is a red herring.
Plaintiffs use more than ten pages in their brief to explain why this Court
should not adopt the Carrera rule. See Answering Br. 21–31. But ConAgra does
not urge this Court to adopt that rule. Rather, ConAgra believes that the district
court erred by not requiring Plaintiffs to submit an administratively feasible plan
for identifying potential class members.
Plaintiffs also claim that the absence of a plan is not a problem because the
district court “can tailor fair verification procedures to the particular case[s].”
Answering Br. 27 (quotation marks omitted). Plaintiffs make this point for the first
time on appeal. Nevertheless, while the district court can create special procedures
for a certain case, it cannot do so until the Plaintiffs have submitted a plan for
Case: 15-55727, 03/02/2016, ID: 9887081, DktEntry: 34, Page 14 of 38
Ascertainability ■ Spivey ■ 1858
identifying potential class members. That principle follows ineluctably not only
from the burden of proof, but also from basic principles of our adversarial system.
Our legal process “rel[ies] on the parties to frame the issues for decision and
assign[s] to courts the role of neutral arbiter.” Ventress v. Japan Airlines, 747 F.3d
716, 724 (9th Cir. 2014). In this context, that means Plaintiffs should submit a
plan for identifying class members in an administratively feasible manner. Once
they do, ConAgra can contest their proposal and explain its flaws, if appropriate.
After both sides have presented their arguments, the district court can decide
whether to “tailor fair verification procedures to the particular case[].” Mullins,
795 F.3d at 670. But the use of tailoring after the district court has evaluated the
parties’ arguments is not a substitute for requiring the Plaintiffs to present a plan
that ConAgra can contest. Relying on the district court to tailor verification
procedures is an end-run around the administrative feasibility requirement.3
3 Plaintiffs also claim that because ConAgra relied on self-identification to conduct a survey in this case, it cannot object to using affidavits for identifying class members. Answering Br. 20. This argument is made for the first time on appeal. Nevertheless, it is a false equivalence. There are significant due process concerns with establishing class membership—and therefore the right to recovery—through affidavits alone. See Marcus v. BMW of N. Am., LLC, 687 F.3d 583, 594 (3d Cir. 2012). ConAgra’s request for individuals to self-identify in a survey does not pose similar problems.
Case: 15-55727, 03/02/2016, ID: 9887081, DktEntry: 34, Page 15 of 38
186 ■ Class Actions ■ July 2017 9
II. The district court erred in holding that common questions predominate over individual ones.
Plaintiffs claim that common questions predominate as to the materiality of
the “100% Natural” label and as to the class members’ damages. They are wrong
on both counts.
A. Individual issues predominate with respect to materiality because Plaintiffs did not establish that it is susceptible to classwide proof.
Contrary to Plaintiffs’ argument, ConAgra does not contend that
“individualized questions would overwhelm because proof of individual reliance
and/or causation is required” as a matter of law. Answering Br. 40. Rather,
ConAgra argues that absent classwide proof of materiality, individualized inquiries
regarding reliance and causation would be required, and therefore predominate.
See Opening Br. 29. It is undisputed that “a misrepresentation is deemed material
if a reasonable man would attach importance to its existence or nonexistence in
determining his choice of action in the transaction in question.” Steroid Hormone
Prod. Cases, 104 Cal. Rptr. 3d 329, 338 (Cal. Ct. App. 2010) (quotation marks
omitted); Answering Br. 41. Likewise, it is well-settled that in order to certify a
class and satisfy the predominance requirement, Plaintiffs must offer some way of
proving materiality and reliance by a reasonable consumer on a classwide basis.4
4 Plaintiffs’ reliance on Pulaski & Middleman, LLC v. Google, Inc., 802 F.3d 979 (9th Cir. 2015), does not negate this. See Answering Br. 41. In Pulaski, this Court stated that “restitution is [not] available on a classwide basis [until] the class
Case: 15-55727, 03/02/2016, ID: 9887081, DktEntry: 34, Page 16 of 38
Ascertainability ■ Spivey ■ 18710
See, e.g., In re Vioxx Class Cases, 180 Cal. App. 4th 116, 133 (2009); Faulk v.
Sears Roebuck & Co., No. 11-cv-02159, 2013 WL 1703378, at *9 (N.D. Cal. Apr.
19, 2013).
Plaintiffs’ theory of liability here “is that ConAgra misled reasonable
consumers by labeling GMO-derived Wesson Oils as 100% Natural.” Answering
Br. 8. Accordingly, as Plaintiffs acknowledge, they were required to point to
evidence demonstrating that a reasonable consumer of Wesson Oil not only
attached significance to the 100% Natural label in making his or her purchasing
decision but that he or she also understood 100% Natural to mean that the product
did not contain GMOs. See Answering Br. 45.
None of the evidence relied upon by the district court nor the evidence to
which Plaintiffs refer in their Answering Brief actually addresses this critical
question. See Kosta v. Del Monte Foods, Inc., 308 F.R.D. 217, 229–30 (N.D. Cal.
2015) (“While materiality and reliance for purposes of UCL, FAL and CLRA
claims can be subject to common proof on a classwide basis under some
circumstances, Plaintiffs here have offered no valid means by which such
classwide proof would be made.”).
representative makes the threshold showing of liability under the UCL and FAL.”Id. at 986. In other words, the class representative has to offer common evidence demonstrating that the public would likely be deceived because that is an elementof a UCL or FAL claim. Id. Plaintiffs fail in this regard.
Case: 15-55727, 03/02/2016, ID: 9887081, DktEntry: 34, Page 17 of 38
188 ■ Class Actions ■ July 2017 11
1. Plaintiffs and the district court rely on a patchwork of unpersuasive and ultimately inadmissible evidence that lacks a sufficient nexus to Plaintiffs’ theory of liability while overlooking contrary, probative, and admissible evidence.
Plaintiffs need not prove their case during class certification, and ConAgra
does not contend otherwise. But Plaintiffs must do more than present a patchwork
of unsponsored, inadmissible surveys from which they cobble together conclusions
that do not meet the standards of Rule 23. Notably, Plaintiffs offer no response to
the detailed breakdown in ConAgra’s Opening Brief of the deficiencies in the
survey evidence they submitted in support of materiality. See Opening Br. 32–38.
Rather, they continue to cite evidence that does not target their theory of liability.
See Answering Br. 48–52. First, they cite Dr. Charles Benbrook’s statement that
foods containing GMOs are not natural. Id. at 49. But that says nothing about
whether the “100% Natural” label on Wesson cooking oils was material to class
members or whether they equated the natural representation with no GMOs in
making their purchasing decisions. Likewise, Plaintiffs cite a bevy of studies
relied upon by the district court allegedly showing that consumers think natural
means GMO-free. Id. at 50; ER00251. Not one of those surveys relates
specifically to Wesson Oil or whether such a belief played a role in a consumer’s
purchasing decision.
Plaintiffs also relied on unsupported, inapplicable, and likely inadmissible
third-party surveys in the district court. They cited one survey where a majority of
Case: 15-55727, 03/02/2016, ID: 9887081, DktEntry: 34, Page 18 of 38
Ascertainability ■ Spivey ■ 18912
consumers said that they “look for a ‘natural’ claim when shopping.” ER00250;
ER03682. But looking for a label on an item is not the same as purchasing an item
because of the label. See In re Vioxx Class Cases, 180 Cal.App.4th at 129 (a
representation is material only “if it induced the consumer to alter his position to
his detriment”)(emphasis added). What’s more, that survey focused on “packaged
or processed foods” generally rather than Wesson cooking oil specifically.
ER03682–3703. To be meaningful, a survey must focus on the precise products
that form the basis of Plaintiffs’ legal theory.5 See Comcast, 133 S. Ct. at 1433.
Plaintiffs also relied on a survey that demonstrated a majority of respondents “were
‘somewhat interested’ or ‘very interested’ in purchasing natural products.”
ER00250; ER03726; ER03755. This survey is triply flawed: (1) it does not
indicate whether consumers think natural means GMO-free; (2) it does not indicate
whether natural is material to consumers; and (3) it is not limited to Wesson
cooking oils. This problem was endemic to Plaintiffs’ evidence of predominance.
In fact, at least two of these exact surveys have been previously rejected as
insufficient to support this precise theory of liability. See Randolph v. J.M.
5 ConAgra cited a panoply of cases holding that to be meaningful, surveys must rely upon responses by potential consumers of the products in question. Opening Br. 34–35. Plaintiffs respond by noting that the cases cited by ConAgra do not relate to “the type of consumer survey evidence at issue here” or the predominance inquiry. Answering Br. 53–55. But Plaintiffs miss the point – that general surveys not tied to the products in question are simply not probative evidence in anycontext.
Case: 15-55727, 03/02/2016, ID: 9887081, DktEntry: 34, Page 19 of 38
190 ■ Class Actions ■ July 2017 13
Smucker Co., 303 F.R.D. 679, 695–96 (S.D. Fla. 2014). In Randolph, the plaintiff
sought to certify a class based on the allegation that J.M. Smucker Co.
misrepresented “to consumers that Crisco oils are ‘All Natural,’” even though they
were made with GMOs. Id. at 682–83. The plaintiffs in that case sought to rely on
both the Consumer Reports National Research Center survey from 2014 and the
Mintel survey from 2010 to establish the materiality of the “no GMOs”
interpretation of the “natural” label. Randolph v. J.M. Smucker Co., 303 F.R.D.
679 (S.D. Fla. 2014), Reply Brief for Plaintiff at 4–6, No. 13-cv-80581, 2014 WL
5365765 (S.D. Fla. Sept. 29, 2014). These are two of the surveys on which
Plaintiffs here rely to establish materiality. ER03680–3818. The district court in
Smuckers concluded that the predominance requirement was not satisfied because
“Plaintiff has not [] established whether the use of the term ‘All Natural’ in this
context would deceive an objectively reasonably consumer.” 303 F.R.D. at 696.
Moreover, the court held that “Plaintiff’s evidence supports the assertion that the
use of GMOs is a widely disputed issue; however, this evidence, in and of itself,
demonstrates the uncertainty with this question, and unequivocally exposes the fact
that there is a lack of consensus on the use of such products.” Id. at 695 (emphasis
added).
Plaintiffs also try to bolster their case by pointing to internal ConAgra
marketing research. Answering Br. 49. This evidence, however, suffers from the
Case: 15-55727, 03/02/2016, ID: 9887081, DktEntry: 34, Page 20 of 38
Ascertainability ■ Spivey ■ 19114
same flaws as the surveys discussed above. The district court noted this point
explicitly: “None of the three [internal ConAgra] surveys plaintiffs cite directly
links consumers’ understanding of the ‘100% Natural’ label to the specific issue
raised in this case.” ER00251. Rather than address the flaws in their survey
evidence on appeal, Plaintiffs resort to simply parroting the district court’s
conclusion that they adduced sufficient evidence to show that materiality is
susceptible to classwide proof. Answering Br. 49–50. But ConAgra has
demonstrated above that this is not the case. At no point in the district court or on
appeal have the Plaintiffs tied their evidence of predominance to the precise theory
of liability in this case. The district court’s failure to enforce this requirement was
a legal error that constitutes a per se abuse of discretion.
2. The district court and Plaintiffs ignored probative, admissible evidence demonstrating that materiality is not susceptible to classwide proof.
Probative evidence also demonstrates that materiality is not susceptible to
classwide proof. For example, Dr. Hanssens’ survey focused on Plaintiffs’ theory
of liability and confirmed that “100% Natural” had no statistically measurable
difference in purchase intent or consumer beliefs about the presence of GMO
ingredients. ER05355–56. Additionally, Dr. Hanssens’ survey was limited to
actual and potential Wesson Oil consumers who were shown Wesson Oil labels
with and without the natural representation. Id. Plaintiffs offer no explanation on
Case: 15-55727, 03/02/2016, ID: 9887081, DktEntry: 34, Page 21 of 38
192 ■ Class Actions ■ July 2017 15
appeal for how it could possibly be proper for the district court to rely on
generalized survey evidence while ignoring Dr. Hanssens’ specific survey
evidence.
In addition, the district court and Plaintiffs ignored ConAgra’s marketing
research indicating that only 42% of respondents even view or notice the 100%
natural statement at all. ER01688, ER00735. Logically, if a consumer does not
even notice a claim, not only can it not be material to his purchasing decision, but
it necessarily follows that he cannot be misled by it.6
ConAgra is not arguing that the district court should delve into whether
every consumer was misled. Rather, ConAgra’s argument is that there cannot be
classwide proof of materiality when the best survey evidence Plaintiffs have
demonstrates that the issue of reliance “var[ies] from consumer to consumer.”
Stearns v. Ticketmaster Corp., 655 F.3d 1013, 1023 (9th Cir. 2011). While
Plaintiffs need not prove their case and rebut all of the contrary evidence, the
district court was required to consider the evidence before it, decide the issues
based upon the admissible evidence, and rule accordingly. Had the court done so,
6 Indeed, the results of this study should be taken into account when examining the third-party surveys relied on by Plaintiffs. Even if 61% of consumers associate a natural claim with the absence of GMOs, ER03862, but only 42% of those consumers even notice the claim, then at most 25% of consumers could possibly find it to be material.
Case: 15-55727, 03/02/2016, ID: 9887081, DktEntry: 34, Page 22 of 38
Ascertainability ■ Spivey ■ 19316
the renewed motion for class certification, like the original motion, would have
been found equally deficient.
3. The issue of materiality and reliance varies among consumers in this case because there is no uniform understanding of “natural” or “GMO.”
It is no surprise that Plaintiffs cannot show that a reasonable class member
bought Wesson cooking oil because they thought the “100% Natural” label meant
GMO-free. The Food and Drug Administration (FDA) recently reiterated that the
meaning of “natural” on food products is far from clear. In a notification of
request for comments, the FDA said that “the term ‘natural’ is used on a variety of
products to mean a variety of things.” Use of the Term “Natural” in the Labeling
of Human Food Products; Request for Information and Comments, 80 Fed. Reg.
69,905, 69,906 (Nov. 12, 2015). The “widespread use” of natural makes it difficult
to pin down exactly what it conveys to consumers. Id. Additionally, the agency
remarked that there is “evidence that consumers regard many uses of this term
[“natural”] as non-informative.” Id. (emphasis added).
Moreover, the FDA issued additional policy guidance regarding GMOs in
November of 2015. U.S. Food & Drug Administration, Guidance for Industry:
Voluntary Labeling Indicating Whether Foods Have or Have Not Been Derived
from Genetically Engineered Plants (Nov. 2015), available at http://ow.ly/Yxg4Z.
Significantly, the guidance notes that there is “potential confusion regarding the
Case: 15-55727, 03/02/2016, ID: 9887081, DktEntry: 34, Page 23 of 38
194 ■ Class Actions ■ July 2017 17
meaning of the acronym ‘GMO’” and “[b]ecause the term ‘genetically modified’
can encompass any alteration to the genetic composition of a plant, including
alterations achieved through traditional hybridization or breeding techniques, that
term could apply to most cultivated food crops since most food crops are the
product of selective breeding.” Id. In addition, the FDA reaffirmed its 1992 policy
which made two key points: (1) there is generally no difference between GMO
foods and non-GMO foods, and (2) whether a plant is genetically engineered “is
generally not material information.” Id.
These findings by the federal agency charged with administering and
enforcing food-label regulations are significant. They underscore that consumers
do not have any uniform understanding of “natural” or “GMO” and that their
beliefs on this issue are generally not material to their purchasing decisions. This
renders classwide proof of materiality and reliance even less likely.
4. ConAgra does not ask this Court to unnecessarily delve into the merits of this case but to engage in the rigorous analysis required by Rule 23.
When faced with arguments that the evidence they produced to support
classwide proof of materiality and reliance is flawed, Plaintiffs do not attempt to
address or rebut these flaws. Instead, they argue that ConAgra’s arguments go to
the merits and should therefore not be considered by the court. Answering Br. 45–
48. They further urge this Court not to “second guess” the district court’s decision,
Case: 15-55727, 03/02/2016, ID: 9887081, DktEntry: 34, Page 24 of 38
Ascertainability ■ Spivey ■ 19518
id. at 46, but in doing so they are asking the Court not to engage in the very
rigorous analysis that it is required to undertake. As Plaintiffs concede, the Court
should consider merits questions when they are relevant to determining whether
the Rule 23 prerequisites for class certification have been satisfied. Id. at 45. The
evidence or lack thereof that ConAgra addresses is the evidence and record on
which the district court based its decision. Certainly, if the district court
considered and relied on it to certify the class, it is fair game for ConAgra to
demonstrate that the conclusions reached by the district court cannot be drawn
from the evidence relied upon. See Comcast, 133 S. Ct. at 1432.
B. Individual issues predominate with respect to damages.
1. Plaintiffs’ argument that its hybrid damages model meets Comcast because “use of arithmetic is not an abuse of discretion” is unsupported.
ConAgra argued that Plaintiffs failed to adequately explain why the hybrid
damages model they proposed passes muster under both Comcast and Daubert.
Plaintiffs’ response on appeal is telling. Instead of relying on case law or their
expert reports, they flippantly assert that “because these results [from each of the
two models] are bare numbers with no physical dimension, combining them with
multiplication is mathematically proper.” Answering Br. 73. They further argue
that “it uses arithmetic, and use of arithmetic is not an abuse of discretion.” Id. at
74. To top it off, Plaintiffs summarily declare that “there was and is no need to
Case: 15-55727, 03/02/2016, ID: 9887081, DktEntry: 34, Page 25 of 38
196 ■ Class Actions ■ July 2017 19
present extensive proof that two percentages of the type at issue can be combined
through multiplication in order to factor both results into a logical conclusion.” Id.
at 72.
This is classic ipse dixit—essentially arguing that their experts can combine
these two models because they say they can. It is of course axiomatic that such
ipse dixit testimony is properly excluded and cannot form the basis for a crucial,
required underpinning for an eleven-state class action. See, e.g., In re Mushroom
Direct Purchaser Antitrust Litig., No. 06-cv-00620, 2015 WL 5766929, at *3–5
(E.D. Pa. Aug. 27, 2015); see also Meadows v. Anchor Longwall & Rebuild, Inc.,
306 F. App’x 781, 790 (3d Cir. 2009); Domingo v. T.K., 289 F.3d 600, 607 (9th
Cir. 2002). This is especially true when Plaintiffs concede that this methodology
has never before been used by anyone, including Plaintiffs’ expert Dr. Howlett.
Answering Br. 65–66. Nowhere—not in their Answering Brief, not in their
experts’ declarations, not in their multiple motions and replies in support of class
certification—do Plaintiffs cite a single source that permits the outcomes of these
two different models that measure two different things to be combined in the way
that they propose. Indeed, as ConAgra noted in its Opening Brief, Dr. Howlett’s
explanation of the hybrid damages model amounted to a single paragraph and
provided no supporting authorities. Opening Br. 56.
Case: 15-55727, 03/02/2016, ID: 9887081, DktEntry: 34, Page 26 of 38
Ascertainability ■ Spivey ■ 19720
It cannot be the case that a party can meet the “rigorous analysis” required
by Comcast for damage models with such flimsy support. If an expert can rely on
a brand-new hybrid model that no one has ever attempted before and do so without
any citation or barely any discussion at all of why it is appropriate to combine the
outcomes of two models—and justify it by backhandedly arguing that it is
“mathematically proper” (once again without any citation)—then there is nothing
left to Comcast. Any expert and any model could meet such a low threshold.
Plaintiffs attempt to defend the district court’s acceptance of this hybrid
model by asserting “that a methodology has not previously been used is not a
reason to reject it.” Answering Br. 66. While this is technically true, it does not
mean that an expert’s ipse dixit and completely new model should be accepted with
virtually no consideration of its reliability or whether it meets Comcast (and
Daubert). Indeed, the fact that it has never before been used logically suggests that
the scrutiny of the model should be more rigorous than it otherwise would be, not
less. Simply put, the cases are legion that hold that a damages model that is
untested and/or fails to adequately explain why or how it is able to reach its
conclusions cannot support class certification. See, e.g., Clausen v. M/V New
Carissa, 339 F.3d 1049, 1056 (9th Cir. 2003); Lust v. Merrell Dow Pharm., Inc.,
89 F.3d 594, 597–98 (9th Cir. 1996); Prism Techs. LLC v. AT&T Mobility, LLC,
No. 12-cv-12201, 2014 U.S. Dist. LEXIS 132619, at *21–22 (D. Neb. Sept. 22,
Case: 15-55727, 03/02/2016, ID: 9887081, DktEntry: 34, Page 27 of 38
198 ■ Class Actions ■ July 2017 21
2014); Randolph, 303 F.R.D. at 697–98; In re Dial Complete Mktg. and Sales
Practice Litig., 312 F.R.D. 36, 79–80 (D.N.H. 2015); Miller v. Fuhu, Inc., No. 14-
cv-06119, 2015 WL 7776794, at *21–22 (C.D. Cal. Dec. 1, 2015). ConAgra
discussed many of these cases in its Opening Brief, Opening Br. 52–56, but
Plaintiffs declined to rebut any of them.
This Court has previously confronted experts who have little experience
with their proposed models and who developed them solely for litigation. See
Clausen, 339 F.3d at1049. In Clausen, the Court noted that a “very significant fact
to be considered” is whether experts are testifying about “research they have
conducted independent of the litigation, or whether they have developed their
opinions expressly for purposes of testifying.” Id. at 1056. If the model has been
developed solely for litigation, then the expert must demonstrate that the “analysis
supporting the proffered conclusions have been subjected to normal scientific
scrutiny through peer review and publication.” Id. (emphasis added).
It is undisputed that Dr. Howlett did not combine hedonic regression and
conjoint analysis as part of genuine scientific research. Indeed, she combined
those models solely for purposes of this litigation. Consequently, she and the
Plaintiffs were obligated to present evidence demonstrating that her analysis had
been scrutinized through “peer review and publication.” Clausen, 339 F.3d at
1056. But Dr. Howlett confessed that this was not true—she was “unaware of
Case: 15-55727, 03/02/2016, ID: 9887081, DktEntry: 34, Page 28 of 38
Ascertainability ■ Spivey ■ 19922
anyone else in any peer-reviewed article that has ever combined the results of a
conjoint analysis and a hedonic regression model.” ER08057 (citing ER07989–
90). Plaintiffs cannot be excused from showing that a model is scientifically
reliable simply because it is novel. “[T]he novelty of a theory does not shield
an expert’s testimony from judicial scrutiny.” Mike’s Train House, Inc. v. Lionel,
LLC, 472 F.3d 398, 407 (6th Cir. 2006).
Chapman v. Maytag Corp., 297 F.3d 682 (7th Cir. 2002), provides an even
closer analogue for Dr. Howlett’s shortcomings in this case. There, the court had
to decide whether the analysis and conclusions of the plaintiff’s expert were
scientifically reliable. The court held they were not because the expert’s
“theory is novel and unsupported by any article, text, study, scientific literature or
scientific data produced by others in his field.” Id. at 688. Moreover, it was
particularly important to the court that “by his own admission, [the expert] has
not published any writings or studies concerning his [] theory.” Id. The absence of
scientific validation meant that the expert’s theory amounted to “[u]nsubstantiated
testimony.” Id.
Chapman’s reasoning applies directly to this case. Like the theory there, Dr.
Howlett’s novel combination of hedonic regression and conjoint analysis is not
supported in the scientific literature. Plaintiffs bore the burden to show that Dr.
Howlett’s model was “subjected to normal scientific scrutiny through peer review
Case: 15-55727, 03/02/2016, ID: 9887081, DktEntry: 34, Page 29 of 38
200 ■ Class Actions ■ July 2017 23
and publication.” Clausen, 339 F.3d at 1056. They did not meet that burden, and
the district court’s failure to hold them to it was an abuse of discretion. Carijano v.
Occidental Petroleum Corp., 643 F.3d 1216, 1236 (9th Cir. 2011).
2. Plaintiffs’ insistence that the two damage models can be combined simply defies logic.
Even if Plaintiffs here were permitted to pass the Comcast hurdle with the
vague and unsupported assertion that the hybrid model is “mathematically proper,”
this assertion is demonstrably incorrect – it is in no way “mathematically proper”
to multiply the outcomes of the regression model and the conjoint analysis model
together. This is because the two models measure two different things – a fact that
even Plaintiffs concede. Answering Br. 71–72. The regression model measures
the price premium, and the conjoint analysis purports to measure the relative
importance of a single trait of the product.
While no one disputes that two numbers in the abstract can be multiplied
together (as Plaintiffs apparently argue), what matters here is whether these two
particular numbers can be multiplied. It literally is mixing “apples and oranges”
as ConAgra argued in its Opening Brief because the two models do not share the
same unit of measurement. Plaintiffs’ argument here is the equivalent of
suggesting that if you add one child’s seven (7) apples with another child’s three
(3) oranges, then you get ten (10) apples. Plaintiffs ignore the fact that the second
child did not have apples at all—just oranges. Similarly, Plaintiffs gloss over the
Case: 15-55727, 03/02/2016, ID: 9887081, DktEntry: 34, Page 30 of 38
Ascertainability ■ Spivey ■ 20124
fact that the conjoint analysis does not measure a premium, but just the relative
importance of current consumer preferences, which is an altogether different
concept. Even the trial court recognized this distinction when it noted “[t]he
assertedly imperfect correlation between the relative importance of a product
feature to consumers and the price premium attributable to that feature.”
ER00263.
Plaintiffs incorrectly argue that there was a “lack of substantive challenge to
any one of the[] three figures” including the price premium, the dollar value, and
the relative value percentage. Answering Br. 66. ConAgra submitted an 81-page
declaration from its expert, Dr. Keith Ugone, as an exhibit to its Opposition to
Class Certification and cited and quoted from it extensively in the briefing both at
the trial court and this Court. See ER00889–01171; ER01172–01248; ER05408–
05578; Opening Br. 53. This declaration attacked Plaintiffs’ two models over the
course of multiple pages.
Accordingly, the district court erred when it ignored ConAgra’s detailed
criticisms and permitted such an illogical damages model to pass muster under
Comcast. See, e.g., Housing Works, Inc. v. Turner, 362 F. Supp. 2d 434, 447–48
(S.D.N.Y. 2005) (excluding illogical expert report that failed to address facts that
would, by common sense, dictate different conclusions from those reached by the
expert).
Case: 15-55727, 03/02/2016, ID: 9887081, DktEntry: 34, Page 31 of 38
202 ■ Class Actions ■ July 2017 25
3. Plaintiffs’ reliance on Guido and related cases is misplaced.
Plaintiffs rely on several cases that permitted conjoint analysis, such as
Guido v. L’Oreal USA, Inc, No. 11-cv-01067 et al., 2014 WL 6603730 (C.D. Cal.
July 1, 2014). Answering Br. 70. The district court in its order certifying eleven
classes also relied on these cases. ER00262–65. But there are important
distinctions between the circumstances here and the circumstances in Guido.
Indeed, the district court judge that certified the classes in this case rejected
a conjoint analysis model that is almost identical to the one here in In re NJOY,
Inc. Consumer Class Action Litig., No. 14-cv-00428, 2015 WL 4881091, at *42
(C.D. Cal. Aug. 14, 2015). There, the court held that the conjoint analysis “does
not satisfy Comcast” because “it does not permit the court to turn the relative
valuation into an absolute valuation to be awarded as damages.” Id. (alteration
omitted). The court explained that the model looked “only to the demand side of
the market equation, converting what is properly an objective evaluation of relative
fair market values into a seemingly subjective inquiry of what an average
consumer wants.” Id. (quotation marks and alterations omitted). Other courts have
also criticized conjoint analysis for this very same reason. See Saavedra v. Eli
Lilly & Co., No. 12-cv-09366, 2014 WL 7338930, at *4 (C.D. Cal. Dec. 18, 2014);
Apple, Inc. v. Samsung Elec. Co., No. 11-cv-01846, 2014 WL 976898, at *11
(N.D. Cal. Mar. 6, 2014). The In re NJOY court noted conjoint analysis was
Case: 15-55727, 03/02/2016, ID: 9887081, DktEntry: 34, Page 32 of 38
Ascertainability ■ Spivey ■ 20326
permitted in Guido because the defendant failed to raise concerns regarding market
supply there. 2015 WL 4881091, at *42–43. The In re NJOY court also
distinguished the result in this case, stating that the hybrid damage model’s use of
hedonic regression accounted for the supply and market factors, and thus could
properly be relied upon to permit certification. Id. at *44.
However, the result in In re NJOY highlights a critical flaw in the district
court’s reasoning in this case. The conjoint analysis here does not account for the
supply and market factors specific to Plaintiffs’ theory of liability – that “100%
Natural” means no GMOs. The In re NJOY decision makes clear that conjoint
analysis does not take into account market supply factors. Id. Though the
regression analysis here does take into account market supply factors, it only does
so for the “natural” label claim as a whole, not for the particular “no GMO”
portion of the “natural” label claim that Plaintiffs challenge in this lawsuit. In this
way, the hybrid damages model suffers from the same deficiency identified by the
district court in its decision on Plaintiffs’ first motion for class certification – it is
not tied to Plaintiffs’ theory of liability. ER00100–01.
If a conjoint analysis that isolates the relative valuation of a particular
feature does not satisfy Comcast by itself, multiplying it by the outcome of a
different model does not magically make it satisfy Comcast. In other words, in
Plaintiffs’ hybrid damages model here, there is no consideration at all of market
Case: 15-55727, 03/02/2016, ID: 9887081, DktEntry: 34, Page 33 of 38
204 ■ Class Actions ■ July 2017 27
supply factors specific to the “no GMO” interpretation of the “natural” label claim.
Multiplying it by the price premium—that allegedly did take into account the
supply and demand features—does not cure the original defect of the conjoint
model. This is why Plaintiffs’ hybrid damages model does not meet Comcast—it
does not measure only those damages attributable to the theory of liability.7 There
is simply no way to square the result in In re NJOY with the result here. The
absence of the market supply factors from the conjoint analysis should be fatal to
the hybrid damages model here. Thus, the circumstances of this case are more
similar to In re NJOY than to Guido and the related cases relied on by Plaintiffs
and the district court.8
III. This Court should remand with instructions to dismiss Plaintiffs’ amended complaint with prejudice.
Under the Federal Rules of Civil Procedure, the Plaintiffs “may amend
[their] pleading once as a matter of course.” FED. R. CIV. P. 15(a). After that, they
may file an amended complaint with the court’s leave “when justice so requires.”
Id. Justice does not demand that Plaintiffs get endless chances to make their case
after they have already filed multiple complaints and multiple motions for class
7 Even the district court below apparently agreed. As noted in ConAgra’s opening brief, the district court wondered whether the Plaintiffs could “ever” satisfy Comcast given their legal theory. Opening Br. 54 n.11. Plaintiffs fail to rebut this point in their answering brief, which underscores that the district court deviated from the appropriate standard under Rule 23(b)(3) and Comcast.8 ConAgra relies on the arguments it made in its Opening Brief regarding typicality and superiority.
Case: 15-55727, 03/02/2016, ID: 9887081, DktEntry: 34, Page 34 of 38
Ascertainability ■ Spivey ■ 20528
certification. This Court has recognized that it is appropriate to dismiss a “second
amended complaint with prejudice … in light of plaintiffs’ repeated failure to cure
the deficiencies in their pleadings.” Destfino v. Reiswig, 630 F.3d 952, 959 (9th
Cir. 2011) (alterations omitted; some emphasis added); see also Larin v. Bank of
Am., NA, 617 F. App’x 651, 652 (9th Cir. 2015) (similar); McCanic v. City of Long
Beach, 993 F.2d 883, 1993 WL 171470, at *3 (9th Cir. 1993) (Table) (similar).
Plaintiffs filed their original complaint on June 28, 2011. Briseno v.
ConAgra Foods, Inc., No. 11-cv-05379, ECF No. 1 (C.D. Cal. Jun. 28, 2011).
They subsequently filed two amended complaints. Id. ECF Nos. 80 & 143. After
that, the Plaintiffs filed two motions for class certification. Id. ECF Nos. 241 &
363. After three complaints and two motions for class certification, this Court will
decide whether the class was properly certified nearly five years after Plaintiffs
filed their original complaint.
This case has gone on long enough. The district court gave Plaintiffs several
opportunities to comply with Rule 23. If this Court were to hold that Plaintiffs’
class was improperly certified, it should not give them “yet another opportunity to
prolong this litigation.” Opening Br. 59. Dismissal with prejudice is appropriate
“in light of plaintiffs’ repeated failure to cure the deficiencies in their pleadings”
and their motions for class certification. Destfino, 630 F.3d at 959 (alterations
omitted).
Case: 15-55727, 03/02/2016, ID: 9887081, DktEntry: 34, Page 35 of 38
206 ■ Class Actions ■ July 2017 29
CONCLUSION
For the foregoing reasons, this Court should reverse the decision of the
district court and remand with instructions to deny Plaintiffs’ amended motion for
class certification.
Dated: March 2, 2016 Respectfully submitted,
MCGUIREWOODS LLP
/s/ A. Brooks Gresham A. Brooks Gresham Angela M. Spivey R. Trent Taylor Laura E. Coombe E. Rebecca Gantt
Counsel for Defendant-Appellant ConAgra Foods, Inc.
Case: 15-55727, 03/02/2016, ID: 9887081, DktEntry: 34, Page 36 of 38
Ascertainability ■ Spivey ■ 20730
CERTIFICATE OF COMPLIANCE
1. This brief complies with the type-volume limitation of Federal Rule of
Appellate Procedure 32(a)(7)(B) because it contains 6,981 words, excluding
the parts of the brief exempted by Federal Rule of Appellate Procedure
32(a)(7)(B)(iii).
2. This brief complies with the typeface and type-style requirements of Federal
Rule of Appellate Procedure 32(a)(5) and Federal Rule of Appellate
Procedure 32(a)(6) because it has been prepared in proportionally-spaced
typeface using Microsoft Word, in 14-point size.
Dated: March 2, 2016 /s/ A. Brooks Gresham A. Brooks Gresham
Counsel for Defendant-Appellant ConAgra Foods, Inc.
Case: 15-55727, 03/02/2016, ID: 9887081, DktEntry: 34, Page 37 of 38
208 ■ Class Actions ■ July 2017 31
CERTIFICATE OF SERVICE
I hereby certify that on March 2, 2016, I electronically filed the foregoing
with the Clerk of the Court for the United States Court of Appeals for the Ninth
Circuit using the appellate CM/ECF system. All participants in the case are
registered CM/ECF users and will be served by the system.
Dated: March 2, 2016 /s/ A. Brooks Gresham A. Brooks Gresham
Counsel for Defendant-Appellant ConAgra Foods, Inc.
Case: 15-55727, 03/02/2016, ID: 9887081, DktEntry: 34, Page 38 of 38
Ascertainability ■ Spivey ■ 209
V. Briseño v. Conagra Foods, Inc., U.S. Court of Appeals for the Ninth Circuit. Opinion
FOR PUBLICATION
UNITED STATES COURT OF APPEALSFOR THE NINTH CIRCUIT
ROBERT BRISENO, individually and on behalf of all others similarly situated,
Plaintiff-Appellee,
v.
CONAGRA FOODS, INC.,Defendant-Appellant.
No. 15-55727
D.C. No.2:11-cv-05379-
MMM-AGR
OPINION
Appeal from the United States District Courtfor the Central District of California
Margaret M. Morrow, District Judge, Presiding
Argued and Submitted September 12, 2016San Francisco, California
Filed January 3, 2017
Before: William A. Fletcher, Morgan B. Christen,and Michelle T. Friedland, Circuit Judges.
Opinion by Judge Friedland
210 ■ Class Actions ■ July 2017
2 BRISENO V. CONAGRA FOODS
SUMMARY*
Class Certification
The panel affirmed the district court’s class certification in putative class actions brought against ConAgra Foods in eleven states by consumers who purchased Wesson-brand cooking oil products labeled “100% Natural” during the relevant period.
Plaintiffs argued that the “100% Natural” label was false or misleading because Wesson oils are made from bioengineered ingredients that plaintiffs contend are “not natural.” ConAgra manufactures, markets, distributes, and sells Wesson products. Defendant urged reversal of the district court’s class certification because the district court did not require Plaintiff-Appellee Robert Briseno and the other named class representatives to proffer an administratively feasible way to identify members of the certified classes.
The panel held that the language of Federal Rule of Civil Procedure 23 neither provides nor implies that demonstrating an administratively feasible way to identify class members is a prerequisite to class certification. The panel therefore joined the Sixth, Seventh, and Eighth Circuits in declining to adopt an administrative feasibility requirement.
* This summary constitutes no part of the opinion of the court. It has been prepared by court staff for the convenience of the reader.
Ascertainability ■ Spivey ■ 211
BRISENO V. CONAGRA FOODS 3
COUNSEL
Angela Spivey (argued), McGuireWoods LLP, Atlanta, Georgia; R. Trent Taylor, McGuireWoods LLP, Richmond, Virginia; E. Rebecca Gantt, McGuireWoods LLP, Norfolk, Virginia; A. Brooks Gresham and Laura E. Coombe, McGuireWoods LLP, Los Angeles, California; for Defendant-Appellant.
Adam Levitt (argued) and Edmund S. Aronowitz, Grant & Eisenhofer P.A., Chicago, Illinois; Mary S. Thomas, Grant & Eisenhofer P.A., Wilmington, Delaware; Ariana J. Tadler, Henry J. Kelston, Meagan Keenan, and Carey Alexander, Milberg LLP, New York, New York; David E. Azar, Milberg LLP, Los Angeles, California; for Plaintiff-Appellee.
OPINION
FRIEDLAND, Circuit Judge:
This appeal requires us to decide whether, to obtain class certification under Federal Rule of Civil Procedure 23, class representatives must demonstrate that there is an “administratively feasible” means of identifying absent class members. Defendant-Appellant ConAgra Foods, Inc. (“ConAgra”) urges us to reverse class certification because the district court did not require Plaintiff-Appellee Robert Briseno and the other named class representatives (collectively, “Plaintiffs”) to proffer a reliable way toidentify members of the certified classes here—consumers
212 ■ Class Actions ■ July 2017
4 BRISENO V. CONAGRA FOODS
in eleven states who purchased Wesson-brand cooking oils labeled “100% Natural” during the relevant period.1
We have never interpreted Rule 23 to require such a showing, and, like the Sixth, Seventh, and Eighth Circuits, we decline to do so now. See Sandusky Wellness Ctr., LLC, v. Medtox Sci., Inc., 821 F.3d 992, 995–96 (8th Cir. 2016); Rikos v. Procter & Gamble Co., 799 F.3d 497, 525 (6th Cir. 2015); Mullins v. Direct Digital, LLC, 795 F.3d 654, 658 (7th Cir. 2015), cert. denied, 136 S. Ct. 1161 (2016). Aseparate administrative feasibility prerequisite to class certification is not compatible with the language of Rule 23. Further, Rule 23’s enumerated criteria already address thepolicy concerns that have motivated some courts to adopt a separate administrative feasibility requirement, and do so without undermining the balance of interests struck by the Supreme Court, Congress, and the other contributors to the Rule. We therefore affirm.
I
Plaintiffs are consumers who purchased Wesson-brand cooking oil products labeled “100% Natural.” The “100% Natural” label appeared on every bottle of Wesson-brand oil throughout the putative class periods (and continues to appear on those products). Plaintiffs argue that the “100% Natural” label is false or misleading because Wesson oils are made from bioengineered ingredients (genetically modified organisms, or GMOs) that Plaintiffs contend are “not
1 We address ConAgra’s other challenges to the district court’s class certification order in a concurrently filed memorandum disposition.
Ascertainability ■ Spivey ■ 213
BRISENO V. CONAGRA FOODS 5
natural.” ConAgra manufactures, markets, distributes, and sells Wesson products.
Plaintiffs filed putative class actions asserting state-law claims against ConAgra in eleven states, and those cases were consolidated in this action. Plaintiffs moved to certify eleven classes defined as follows:2
All persons who reside in the States of California, Colorado, Florida, Illinois, Indiana, Nebraska, New York, Ohio, Oregon, South Dakota, or Texas who have purchased Wesson Oils within the applicable statute of limitations periods established by the laws of their state of residence (the “Class Period”) through the final disposition of this and any and all related actions.
As relevant here, ConAgra opposed class certification on the ground that there would be no administratively feasible way to identify members of the proposed classes because consumers would not be able to reliably identify themselves as class members. As a result, ConAgra argued that the class was not eligible for certification.3
2 We refer to Plaintiffs’ amended motion for class certification, which is the subject of this appeal.
3 ConAgra called this a failure of “ascertainability.” We refrain from referring to “ascertainability” in this opinion because courts ascribe widely varied meanings to that term. For example, some courts use the word “ascertainability” to deny certification of classes that are not clearly or objectively defined. See, e.g., Brecher v. Republic of Argentina, 806 F.3d 22, 24–26 (2d Cir. 2015) (holding that a class
214 ■ Class Actions ■ July 2017
6 BRISENO V. CONAGRA FOODS
The district court acknowledged that the Third Circuit and some district courts have refused certification in similar circumstances, but it declined to join in their reasoning. Instead, the district court held that, at the certification stage, it was sufficient that the class was defined by an objective criterion: whether class members purchased Wesson oil during the class period.
The district court ultimately granted Plaintiffs’ motion in part and certified eleven statewide classes to pursue certain claims for damages under Federal Rule of Civil Procedure 23(b)(3). ConAgra timely sought and obtained permission to appeal pursuant to Rule 23(f).
II
Federal Rule of Civil Procedure 23 governs the maintenance of class actions in federal court. Parties seeking class certification must satisfy each of the four requirements of Rule 23(a)—numerosity, commonality, typicality, and adequacy—and at least one of the requirements of Rule 23(b). Ellis v. Costco Wholesale Corp., 657 F.3d 970, 979–80 (9th Cir. 2011).
defined as all owners of beneficial interests in a particular bond series, without reference to the time owned, was too indefinite); DeBremaecker v. Short, 433 F.2d 733, 734 (5th Cir. 1970) (affirming denial of class certification because a class composed of state residents “active in the ‘peace movement’” was uncertain and overbroad). Others have used the term in referring to classes defined in terms of success on the merits. See, e.g., EQT Prod. Co. v. Adair, 764 F.3d 347, 360 n.9 (4th Cir. 2014) (remanding and instructing the district court to consider, “as part of its class-definition analysis,” inter alia, whether the proposed classes could be defined without creating a fail-safe class). Our court does not have its own definition. See infra note 4.
Ascertainability ■ Spivey ■ 215
BRISENO V. CONAGRA FOODS 7
ConAgra argues that, in addition to satisfying these enumerated criteria, class proponents must also demonstrate that there is an administratively feasible way to determine who is in the class.4 ConAgra claims that Plaintiffs did not propose any way to identify class members and cannot prove that an administratively feasible method exists because consumers do not generally save grocery receipts and are unlikely to remember details about individual purchases of a low-cost product like cooking oil. We have not previously interpreted Rule 23 to require such a demonstration, and, for the reasons that follow, we do not do so now.
A
We employ the “traditional tools of statutory construction” to interpret the Federal Rules of Civil
4 On appeal, ConAgra continues to present administrative feasibility as part of a threshold “ascertainability” prerequisite to certification. ConAgra relies on a footnote in Berger v. Home Depot USA, Inc.,741 F.3d 1061 (9th Cir. 2014), to argue that our court has recognized such a requirement. But in that footnote we explicitly declined to decide whether the district court abused its discretion by denying certification based on a “threshold ascertainability test.” Id. at 1071 n.3. ConAgra cites no other precedent to support the notion that our court has adopted an “ascertainability” requirement. This is not surprising because we have not. Instead, we have addressed the types of alleged definitional deficiencies other courts have referred to as “ascertainability” issues, see supra note 3, through analysis of Rule 23’s enumerated requirements. See, e.g., Torres v. Mercer Canyons Inc., 835 F.3d 1125, 1136–39 (9th Cir. 2016) (addressing claim that class definition was overbroad—and thus arguably contained some members who were not injured—as a Rule 23(b)(3) predominance issue); Probe v. State Teachers’ Ret. Sys.,780 F.2d 776, 780 (9th Cir. 1986) (recognizing that a class must not be vaguely defined and must be “sufficiently definite to conform to Rule 23”). Although the parties here use the word “ascertainability,” they dispute only whether a class proponent must proffer an administratively feasible way to identify class members. That is therefore the only issue we decide.
216 ■ Class Actions ■ July 2017
8 BRISENO V. CONAGRA FOODS
Procedure. Republic of Ecuador v. Mackay, 742 F.3d 860, 864 (9th Cir. 2014) (quoting United States v. Petri, 731 F.3d 833, 839 (9th Cir. 2013)). In construing what Rule 23 requires, our “‘first step’” is thus “‘determin[ing] whether the language at issue has a plain meaning.’” Id. (quoting McDonald v. Sun Oil Co., 548 F.3d 774, 780 (9th Cir. 2008)); see also Beech Aircraft Corp. v. Rainey, 488 U.S. 153, 163 (1988) (noting that interpretation of the federal rules “begin[s] with the language of the Rule itself”). “When interpreting [the Rule], words and phrases must not be read in isolation, but with an eye toward the ‘purpose and context of the statute.’” Petri, 731 F.3d at 839 (quoting Dolan v. U.S. Postal Serv., 546 U.S. 481, 486 (2006)). “An interpretation that gives effect to every clause is generally preferable to one that does not.” Mackay, 742 F.3d at 864.
Beginning then with the plain language, Rule 23(a) is titled “Prerequisites” and provides:
One or more members of a class may sue or be sued as representative parties on behalf of all members only if:
(1) the class is so numerous that joinder of all members is impracticable;
(2) there are questions of law or fact common to the class;
(3) the claims or defenses of the representative parties are typical of the claims or defenses of the class; and
(4) the representative parties will fairly and adequately protect the interests of the class.
Ascertainability ■ Spivey ■ 217
BRISENO V. CONAGRA FOODS 9
FED. R. CIV. P. 23(a). This provision identifies the prerequisites to maintaining a class action in federal court. It does not mention “administrative feasibility.”
Traditional canons of statutory construction suggest that this omission was meaningful. Because the drafters specifically enumerated “[p]rerequisites,” we may conclude that Rule 23(a) constitutes an exhaustive list. See Silvers v. Sony Pictures Entm’t, Inc., 402 F.3d 881, 885 (9th Cir. 2005) (explaining that, under the doctrine of expressio unius estexclusio alterius, the enumeration of certain criteria to the exclusion of others should be interpreted as an intentional omission). We also take guidance from language used in other provisions of the Rule. In contrast to Rule 23(a), Rule 23(b)(3) provides, “The matters pertinent to these findings include,” followed by four listed considerations. FED. R.CIV. P. 23(b)(3) (emphasis added). If the Rules Advisory Committee had intended to create a non-exhaustive list in Rule 23(a), it would have used similar language. See Russello v. United States, 464 U.S. 16, 23 (1983) (“‘[W]here Congress includes particular language in one section of a statute but omits it in another section of the same Act, it is generally presumed that Congress acts intentionally and purposely in the disparate inclusion or exclusion.’” (alteration in original) (quoting United States v. Wong Kim Bo, 472 F.2d 720, 722 (5th Cir. 1972) (per curiam))). Moreover, Rule 23(b)(3) requires a court certifying a class under that section to consider “the likely difficulties in managing a class action.” FED. R. CIV. P. 23(b)(3)(D). Imposing a separate administrative feasibility requirement would render that manageability criterion largely superfluous, a result that contravenes the familiar preceptthat a rule should be interpreted to “give[] effect to every clause.” Mackay, 742 F.3d at 864.
218 ■ Class Actions ■ July 2017
10 BRISENO V. CONAGRA FOODS
Supreme Court precedent also counsels in favor of hewing closely to the text of Rule 23. In Amchem Products, Inc. v. Windsor, 521 U.S. 591 (1997), the Court considered whether a settlement-only class could be certified without satisfying the requirements of Rule 23. In holding that it could not,5 the Court underscored that the Federal Rules of Civil Procedure result from “an extensive deliberative process involving . . . a Rules Advisory Committee, public commenters, the Judicial Conference, [the Supreme] Court, [and] Congress.” Id. at 620. The Court warned that “[t]he text of a rule thus proposed and reviewed limits judicial inventiveness” and admonished that “[c]ourts are not free to amend a rule outside the process Congress ordered.” Id. The lesson of Amchem Products is plain: “Federal courts . . .lack authority to substitute for Rule 23’s certification criteria a standard never adopted.” Id. at 622.
In sum, the language of Rule 23 does not impose a freestanding administrative feasibility prerequisite to class certification. Mindful of the Supreme Court’s guidance, we decline to interpose an additional hurdle into the class certification process delineated in the enacted Rule. SeeSandusky Wellness Ctr., LLC, v. Medtox Sci., Inc., 821 F.3d 992, 996 (8th Cir. 2016) (declining to recognize a “separate, preliminary” requirement and, instead, “adher[ing] to a rigorous analysis of the Rule 23 requirements”).
5 The Court recognized, however, that a settlement-only class—which by definition will not proceed to trial—can be certified without consideration of potential trial-management challenges under Rule 23(b)(3)(D). See Amchem Prods., 521 U.S. at 620.
Ascertainability ■ Spivey ■ 219
BRISENO V. CONAGRA FOODS 11
B
We recognize that the Third Circuit does require putative class representatives to demonstrate “administrative feasibility” as a prerequisite to class certification.6 See Byrd
6 Other circuits have cited the Third Circuit’s administrative feasibility standard but have not actually imposed the standard in the same manner as has the Third Circuit. The First Circuit has cited Carrera v. Bayer Corp., 727 F.3d 300 (3d Cir. 2013), for the proposition that at the class certification stage, it must be anticipated that, by the time a case reaches the liability and claims administration stages, there will be an administratively feasible way to distinguish injured from uninjured class members. See In re Nexium Antitrust Litig., 777 F.3d 9, 19–20 (1st Cir. 2015). Requiring plaintiffs to propose a mechanism for eventually determining whether a given class member is entitled to damages is different from requiring plaintiffs to demonstrate an administratively feasible way to identify all class members at the certification stage. In Brecher v. Republic of Argentina, 806 F.3d 22 (2d Cir. 2015), the Second Circuit mentioned administrative feasibility and cited Marcus v. BMW of N. Am., LLC, 687 F.3d 583 (3d Cir. 2012), but administrative feasibility played no role in the court’s decision, which instead turned on the principle that a class definition must be objective and definite. Brecher,806 F.3d at 24–26. The Fourth Circuit has reversed class certification based in part on potential “administrative barrier[s]” to ascertaining class members and cited the Third Circuit in doing so. See EQT Prod. Co. v. Adair, 764 F.3d 347, 358–60 (4th Cir. 2014). But the “administrative barriers” identified by the court in EQT sounded in definitional deficiencies, numerosity questions, predominance problems, and management difficulties, see id.—issues that all implicate other class certification criteria. It is thus far from clear that the Fourth Circuit requires an affirmative demonstration of administrative feasibility as a separate prerequisite to class certification. Even the Third Circuit has cabined its administrative feasibility rule in recent cases. See In re Cmty. Bank of N. Va. Mortg. Lending Practices Litig., 795 F.3d 380, 396–97(3d Cir. 2015) (distinguishing Carrera as addressing particular “evidentiary problems”), cert. denied sub nom. PNC Bank v. Brian W.,136 S. Ct. 1167 (2016); Byrd, 784 F.3d at 164 (clarifying that Carreradid not create a “records requirement” at the class certification stage and instead “only requires the plaintiff to show that class members can be identified” (quoting Carrera, 727 F.3d at 308 n.2 (emphasis added))).
220 ■ Class Actions ■ July 2017
12 BRISENO V. CONAGRA FOODS
v. Aaron’s Inc., 784 F.3d 154, 162–63 (3d Cir. 2015); Carrera v. Bayer Corp., 727 F.3d 300, 306–08 (3d Cir. 2013). The Third Circuit justifies its administrative feasibility requirement not through the text of Rule 23 but rather as a necessary tool to ensure that the “class will actually function as a class.” Byrd, 784 F.3d at 162. The Third Circuit suggests that its administrative feasibility prerequisite achieves this goal by (1) mitigating administrative burdens; (2) safeguarding the interests of absent and bona fide class members; and (3) protecting the due process rights of defendants. See Carrera, 727 F.3d at 307, 310. The Seventh Circuit soundly rejected those justifications in Mullins v. Direct Digital, LLC, 795 F.3d 654 (7th Cir. 2015), and the Sixth Circuit followed suit, see Rikos v. Procter & Gamble Co., 799 F.3d 497, 525 (6th Cir. 2015) (citing Mullins in declining to follow Carrera). We likewise conclude that Rule 23’s enumerated criteria already address the interests that motivated the Third Circuit and, therefore, that an independent administrative feasibility requirement is unnecessary.
1
One rationale the Third Circuit has given for imposing an administrative feasibility requirement is the need to mitigate the administrative burdens of trying a Rule 23(b)(3) class action. Courts adjudicating such actions must provide notice that a class has been certified and an opportunity for absent class members to withdraw from the class. See Wal-Mart Stores, Inc. v. Dukes, 564 U.S. 338, 362 (2011); accordFED. R. CIV. P. 23(c)(2)(B). The Third Circuit largely justifies its administrative feasibility prerequisite as necessary to ensure that compliance with this procedural requirement does not compromise the efficiencies Rule
Ascertainability ■ Spivey ■ 221
BRISENO V. CONAGRA FOODS 13
23(b)(3) was designed to achieve.7 See Shelton v. Bledsoe,775 F.3d 554, 562 (3d Cir. 2015); Carrera, 727 F.3d at 307.
But Rule 23(b)(3) already contains a specific, enumerated mechanism to achieve that goal: the manageability criterion of the superiority requirement. Rule 23(b)(3) requires that a class action be “superior to other available methods for fairly and efficiently adjudicating the controversy,” and it specifically mandates that courts consider “the likely difficulties in managing a class action.” FED. R. CIV. P. 23(b)(3)(D).
Moreover, as the Seventh Circuit has observed, requiring class proponents to satisfy an administrative feasibility prerequisite “conflicts with the well-settled presumption that courts should not refuse to certify a class merely on the basis of manageability concerns.” Mullins, 795 F.3d at 663; see also In re Visa Check/MasterMoney Antitrust Litig.,280 F.3d 124, 140 (2d Cir. 2001) (Sotomayor, J.) (holding that refusal to certify a class “on the sole ground that it would be unmanageable is disfavored and ‘should be the exception rather than the rule’” (quoting In re S. Cent. States Bakery Prods. Antitrust Litig., 86 F.R.D. 407, 423 (M.D. La. 1980))), overruled on other grounds by In re IPO Sec. Litig.,471 F.3d 24 (2d Cir. 2006), and superseded by statute on other grounds as stated in Attenborough v. Constr. & Gen. Bldg. Laborers’ Local 79, 238 F.R.D. 82, 100 (S.D.N.Y. 2006). This presumption makes ample sense given the variety of procedural tools courts can use to manage the
7 Because the notice requirement is mandatory only for Rule 23(b)(3) classes, the Third Circuit has declined to extend its “ascertainability” prerequisite, which includes its administrative feasibility requirement, to Rule 23(b)(2) classes. See Shelton, 775 F.3d at 562–63. We understand ConAgra’s arguments here to be similarly limited to Rule 23(b)(3) class actions.
222 ■ Class Actions ■ July 2017
14 BRISENO V. CONAGRA FOODS
administrative burdens of class litigation. For example, Rule 23(c) enables district courts to divide classes into subclasses or certify a class as to only particular issues. FED. R. CIV. P.23(c)(4), (5); see also In re Visa Check/MasterMoney,280 F.3d at 141 (listing “management tools available to” district courts).
Adopting a freestanding administrative feasibility requirement instead of assessing manageability as one component of the superiority inquiry would also have practical consequences inconsistent with the policies embodied in Rule 23. Rule 23(b)(3) calls for a comparative assessment of the costs and benefits of class adjudication, including the availability of “other methods” for resolving the controversy. By contrast, as the Seventh Circuit has emphasized, a standalone administrative feasibility requirement would invite courts to consider the administrative burdens of class litigation “in a vacuum.” See Mullins, 795 F.3d at 663. That difference in approach would often be outcome determinative for cases like this one, in which administrative feasibility would be difficult to demonstrate but in which there may be no realistic alternative to class treatment. See id. at 663–64. Class actions involving inexpensive consumer goods in particular would likely fail at the outset if administrative feasibility were a freestanding prerequisite to certification.
The authors of Rule 23 opted not to make the potential administrative burdens of a class action dispositive and instead directed courts to balance the benefits of class adjudication against its costs. We lack authority to substitute our judgment for theirs. See Amchem Prods., 521 U.S. at 620 (“[T]he Rule as now composed sets the requirements [courts] are bound to enforce.”).
Ascertainability ■ Spivey ■ 223
BRISENO V. CONAGRA FOODS 15
2
The Third Circuit has also justified its administrative feasibility requirement as necessary to protect absent class members and to shield bona fide claimants from fraudulent claims.
A
With respect to absent class members, the Third Circuit has expressed concern about whether courts would be able to ensure individual notice without a method for reliably identifying class members. See Byrd, 784 F.3d at 165; Carrera, 727 F.3d at 307. We believe that concern is unfounded, because neither Rule 23 nor the Due Process Clause requires actual notice to each individual class member.
Rule 23 requires only the “best notice that is practicable under the circumstances, including individual notice to all members who can be identified through reasonable effort.” FED. R. CIV. P. 23(c)(2)(B) (emphasis added). In other words, “[t]he rule does not insist on actual notice to all class members in all cases” and “recognizes it might be impossibleto identify some class members for purposes of actual notice.” Mullins, 795 F.3d at 665. And courts have long employed cy pres remedies when some or even all potential claimants cannot be identified. See Six (6) Mexican Workers v. Ariz. Citrus Growers, 904 F.2d 1301, 1306 (9th Cir. 1990) (“In a majority of class actions at least some unclaimed damages or unlocated class members remain. The existence of a large unclaimed damage fund, while relevant to the manageability determination, does not necessarily make a class action ‘unmanageable.’” (citation omitted)). The notion that an inability to identify all class members
224 ■ Class Actions ■ July 2017
16 BRISENO V. CONAGRA FOODS
precludes class certification cannot be reconciled with our court’s longstanding cy pres jurisprudence. See id.
Likewise, the Due Process Clause does not require actual, individual notice in all cases. See Silber v. Mabon,18 F.3d 1449, 1453–54 (9th Cir. 1994); see also Mullins,795 F.3d at 665 (explaining that when individual notice by mail is “not possible, courts may use alternative means such as notice through third parties, paid advertising, and/or posting in places frequented by class members, all without offending due process”). Courts have routinely held that notice by publication in a periodical, on a website, or even at an appropriate physical location is sufficient to satisfy due process. See, e.g., Hughes v. Kore of Ind. Enter., Inc.,731 F.3d 672, 676–77 (7th Cir. 2013) (holding that sticker notices on two allegedly offending ATMs, as well as publication in the state’s principal newspaper and on a website, provided adequate notice to class members in an action challenging ATM fees); Juris v. Inamed Corp.,685 F.3d 1294, 1319 (11th Cir. 2012) (holding that notice to unidentified class members by periodical and website satisfied due process).
Moreover, the lack-of-notice concern presumes that some harm will inure to absent class members who do not receive actual notice. In theory, inadequate notice might deny an absent class member the opportunity to opt out and pursue individual litigation. But in reality that risk is virtually nonexistent in the very cases in which satisfying an administrative feasibility requirement would prove most difficult—low-value consumer class actions. Such cases typically involve low-cost products and, as a result, recoveries too small to incentivize individual litigation. At the same time, an administrative feasibility requirement like that imposed by the Third Circuit would likely bar such actions because consumers generally do not keep receipts or
Ascertainability ■ Spivey ■ 225
BRISENO V. CONAGRA FOODS 17
other records of low-cost purchases. Practically speaking, a separate administrative feasibility requirement would protect a purely theoretical interest of absent class members at the expense of any possible recovery for all class members—in precisely those cases that depend most on the class mechanism. Justifying an administrative feasibility requirement as a means of ensuring perfect recovery at the expense of any recovery would undermine the very purpose of Rule 23(b)(3)—“vindication of ‘the rights of groups of people who individually would be without effective strength to bring their opponents into court at all.’” Amchem Prods.,521 U.S. at 617 (quoting Benjamin Kaplan, A Prefatory Note, 10 B.C. INDUS. & COM. L. REV. 497, 497 (1969)).
B
The Third Circuit has also expressed concern that without an administrative feasibility requirement, individuals will submit illegitimate claims and thereby dilute the recovery of legitimate claimants. See Carrera, 727 F.3d at 310.
The fraud concern may be valid in theory, but “in practice, the risk of dilution based on fraudulent or mistaken claims seems low, perhaps to the point of being negligible.” Mullins, 795 F.3d at 667. This is especially true in class actions involving low-cost consumer goods. Why would a consumer risk perjury charges and spend the time and effort to submit a false claim for a de minimis monetary recovery? And even if consumers might do so, courts “can rely, as they have for decades, on claim administrators, various auditing processes, sampling for fraud detection, follow-up notices to explain the claims process, and other techniques tailored by the parties and the court” to avoid or minimize fraudulent claims. Id.
226 ■ Class Actions ■ July 2017
18 BRISENO V. CONAGRA FOODS
As to the dilution concern specifically, consistently low participation rates in consumer class actions make it very unlikely that non-deserving claimants would diminish the recovery of participating, bona fide class members.8 See id.“It is not unusual for only 10 or 15% of the class membersto bother filing claims.” Christopher R. Leslie, The Significance of Silence: Collective Action Problems and Class Action Settlements, 59 FLA. L. REV. 71, 119 (2007). Moreover, if certification is denied to prevent dilution, deserving class members “will receive nothing, for they would not have brought suit individually in the first place.” Mullins, 795 F.3d at 668. As the Seventh Circuit put it, “[w]hen it comes to protecting the interests of absent class members, courts should not let the perfect become the enemy of the good.” Id. at 666.
3
Finally, the Third Circuit has characterized its administrative feasibility requirement as necessary to protect the due process rights of defendants “to raise individual challenges and defenses to claims.” Carrera, 727 F.3d at 307. The gravamen of this due process concern seems to be
8 Theoretically, if there were non-legitimate claimants, they would dilute a cy pres fund. But that outcome would not impact bona fide claimants, who would have already received distributions. See Nachshin v. AOL, LLC, 663 F.3d 1034, 1036 (9th Cir. 2011) (explaining that, after distributions have been made to any claimants, “[t]he cy pres doctrine allows a court to distribute unclaimed or non-distributable portions of a class action settlement fund to the ‘next best’ class of beneficiaries” (emphasis added)). Nor would it affect the defendant, whose liability will already have been determined. See Six (6) Mexican Workers,904 F.2d at 1307 (“The use of cy pres or fluid recovery to distribute unclaimed funds may be considered only after a valid judgment for damages has been rendered against the defendant.”).
Ascertainability ■ Spivey ■ 227
BRISENO V. CONAGRA FOODS 19
that defendants must have an opportunity to dispute whether class members really bought the product or used the service at issue.9 See id. (stating that a defendant has a “due process right to challenge the proof used to demonstrate class membership”); Marcus v. BMW of N. Am., LLC, 687 F.3d 583, 594 (3d Cir. 2012) (“Forcing [defendants] to accept as true absent persons’ declarations that they are members of the class, without further indicia of reliability, would have serious due process implications.”).
As an initial matter, defendants plainly can mount such challenges as to the named class representatives. Class representatives must establish standing by, for example, showing that they bought the product or used the service at issue. See Mazza v. Am. Honda Motor Co., Inc., 666 F.3d 581, 595 (9th Cir. 2012) (holding that class representatives who allegedly paid more for or purchased a product due to a defendant’s deceptive conduct have suffered an “injury in fact” that establishes Article III standing); Bates v. United Parcel Serv., Inc., 511 F.3d 974, 985 (9th Cir. 2007) (stating that “[t]he plaintiff class bears the burden of showing” that “at least one named plaintiff” meets the Article III standing requirements). At the class certification stage, the class
9 Relatedly, ConAgra argues that an administrative feasibility requirement would protect its ability to meaningfully assert a res judicata defense in future actions asserting the same claims. But determining whether a plaintiff in that future action was a member of this class precluded from relitigating would be possible so long as the class definition in this action was clear (and ConAgra does not dispute that it is). If a future plaintiff were to assert a claim challenging the “100% Natural” label on Wesson oil purchased during the class period in one of the eleven states at issue, that would show that she was a member of theclass bound by the judgment. This would be so regardless of how “administratively feasible” it was to prove the entirety of the membership at the class certification stage in this action. See Geoffrey C. Shaw, Note, Class Ascertainability, 124 YALE L.J. 2354, 2374–78(2015).
228 ■ Class Actions ■ July 2017
20 BRISENO V. CONAGRA FOODS
representatives bear the burden of demonstrating compliance with Rule 23. See Wal-Mart Stores, 564 U.S. at 350 (“A party seeking class certification must affirmatively demonstrate his compliance with the Rule.”). And if the case proceeds past the certification stage, the plaintiff class must carry the burden of proving every element of its claims to prevail on the merits. See id. at 351 n.6 (observing that, in a securities fraud class action, “plaintiffs seeking 23(b)(3) certification must prove that their shares were traded on an efficient market, an issue that they will surely have to prove again at trial in order to make out their case on the merits” (citation omitted)); id. at 367 (“[T]he Rules Enabling Act forbids interpreting Rule 23 to ‘abridge, enlarge or modify any substantive right.’” (quoting 28 U.S.C. § 2072(b))); Shady Grove Orthopedic Assocs., P.A., v. Allstate Ins. Co.,559 U.S. 393, 408 (2010) (“A class action. . . . leaves the parties’ legal rights and duties intact and the rules of decision unchanged.”). Defendants can oppose the class representatives’ showings at every stage. Indeed, in litigating class certification, ConAgra took discovery of the class representatives, challenged whether they bought Wesson oil products, attacked their credibility, and disputed whether they relied on the label at issue. As the case proceeds, ConAgra will have further opportunities to contest every aspect of Plaintiffs’ case.
Defendants will have similar opportunities to individually challenge the claims of absent class members if and when they file claims for damages. At the claims administration stage, parties have long relied on “claim administrators, various auditing processes, sampling for fraud detection, follow-up notices to explain the claims process, and other techniques tailored by the parties and the court” to validate claims. Mullins, 795 F.3d at 667. Rule 23 specifically contemplates the need for such individualized
Ascertainability ■ Spivey ■ 229
BRISENO V. CONAGRA FOODS 21
claim determinations after a finding of liability. See FED. R.CIV. P. 23 advisory committee’s note to 1966 amendment (explaining that certification may be proper “despite the need, if liability is found, for separate determinations of the damages suffered by individuals within the class”); see also Levya v. Medline Indus. Inc., 716 F.3d 510, 513–14 (9th Cir. 2013) (reaffirming, after Comcast Corp. v. Behrend,133 S.Ct. 1426 (2013), that the need for individualized damages determinations after liability has been adjudicated does not preclude class certification). ConAgra does not explain why such procedures are insufficient to safeguard its due process rights.10
Given these existing opportunities to challenge Plaintiffs’ case, it is not clear why requiring an administratively feasible way to identify all class members at the certification stage is necessary to protect ConAgra’s due process rights. As the Seventh Circuit put it, “[t]he due process question is not whether the identity of class members can be ascertained with perfect accuracy at the certification stage but whether the defendant will receive a fair opportunity to present its defenses when putative class members actually come forward.” Mullins, 795 F.3d at 670. ConAgra may prefer to terminate this litigation in one fell swoop at class certification rather than later challenging each individual class member’s claim to recovery, but there is no due process right to “a cost-effective procedure for challenging every individual claim to class membership.” Id. at 669.
10 District courts also have discretion to allow limited discovery from absent class members if the particular circumstances of a specific case justify it. See WILLIAM B. RUBENSTEIN, NEWBERG ON CLASS ACTIONS § 9:13 (5th ed. 2013) (“[C]ertain forms of limited discovery from absent class members may be permitted in certain circumstances.”).
230 ■ Class Actions ■ July 2017
22 BRISENO V. CONAGRA FOODS
If the concern is that claimants in cases like this will eventually offer only a “self-serving affidavit” as proof of class membership, it is again unclear why that issue must be resolved at the class certification stage to protect a defendant’s due process rights. If a Wesson oil consumer were to pursue an individual lawsuit instead of a class action, an affidavit describing her purchases would create a genuine issue if ConAgra disputed the affidavit, and would prevent summary judgment against the consumer. See Mullins,795 F.3d at 669; accord FED. R. CIV. P. 56(c)(1)(A). Given that a consumer’s affidavit could force a liability determination at trial without offending the Due Process Clause, we see no reason to refuse class certification simply because that same consumer will present her affidavit in a claims administration process after a liability determination has already been made.
Moreover, identification of class members will not affect a defendant’s liability in every case. For example, in this case, Plaintiffs propose to determine ConAgra’s aggregate liability by (1) calculating the price premium attributable to the allegedly false statement that appeared on every unit sold during the class period, and (2) multiplying that premium by the total number of units sold during the class period. We agree with the Seventh Circuit that, in cases in which aggregate liability can be calculated in such a manner, “the identity of particular class members does not implicate the defendant’s due process interest at all” because “[t]he addition or subtraction of individual class members affects neither the defendant’s liability nor the total amount of damages it owes to the class.” Mullins, 795 F.3d at 670; see also Six (6) Mexican Workers, 904 F.2d at 1307 (“Where the only question is how to distribute damages, the interests affected are not the defendant’s but rather those of the silent class members.”). The defendant will generally know how
Ascertainability ■ Spivey ■ 231
BRISENO V. CONAGRA FOODS 23
many units of a product it sold in the geographic area in question, and if the defendant is ultimately found to have charged, for example, 10 cents more per unit than it could have without the challenged sales practice, the aggregate amount of liability will be determinable even if the identity of all class members is not. The Third Circuit recognized as much in Carrera. See Carrera, 727 F.3d at 310 (acknowledging but not addressing the argument that “[the defendant’s] total liability” would not be “affected by unreliable affidavits”).
For these reasons, protecting a defendant’s due process rights does not necessitate an independent administrative feasibility requirement.
C
In summary, the language of Rule 23 neither provides nor implies that demonstrating an administratively feasible way to identify class members is a prerequisite to class certification, and the policy concerns that have motivated the Third Circuit to adopt a separately articulated requirement are already addressed by the Rule. We therefore join the Sixth, Seventh, and Eighth Circuits in declining to adopt an administrative feasibility requirement. See Sandusky Wellness Ctr., LLC, v. Medtox Sci., Inc., 821 F.3d 992, 995–96 (8th Cir. 2016) (recognizing that some courts have imposed an administrative feasibility requirement, but declining to do so); Rikos v. Procter & Gamble Co., 799 F.3d 497, 525 (6th Cir. 2015) (“We see no reason to follow Carrera.”); Mullins v. Direct Digital, LLC, 795 F.3d 654, 658 (7th Cir. 2015) (rejecting the administrative feasibility requirement as incompatible with Rule 23 and “the balance of interests that Rule 23 is designed to protect”).
232 ■ Class Actions ■ July 2017
24 BRISENO V. CONAGRA FOODS
III
For the forgoing reasons, the district court did not err in declining to condition class certification on Plaintiffs’ proffer of an administratively feasible way to identify putative class members.
AFFIRMED.